Download Thyroid Hormone, Metabolism and the Brain

Thyroid Hormone, Metabolism
and the Brain
Lars P. Klieverik
proefschrift Klieverik.indb 1
4-8-2009 15:22:17
Colofon
Thyroid hormone, metabolism and the brain.
Proefschrift, Universiteit van Amsterdam, Nederland
ISBN:
978-90-9024541-6
© 2009, Lars Peter Klieverik, Amsterdam
Cover en lay-out: Chris Bor, Medische fotografie en illustratie,
Academisch Medisch Centrum, Amsterdam
Drukkerij:
Buijten & Schipperheijn, Amsterdam
De uitgave van dit proefschrift werd mede mogelijk gemaakt door steun van:
Diabetes Fonds, Ferring pharmaceuticals, GlaxoSmithKline, Goodlife health care, Ipsen Farmaceutica,
Novartis Pharma, Novo Nordisk, Pfizer, Sandoz, Sanovi aventis en de Universiteit van Amsterdam.
proefschrift Klieverik.indb 2
4-8-2009 15:22:17
Thyroid Hormone, Metabolism
and the Brain
Academisch proefschrift
Ter verkrijging van de graad van doctor
aan de Universiteit van Amsterdam
op gezag van de Rector Magnificus
Prof. Dr. D.C. van den Boom
ten overstaan van een door het college voor promoties ingestelde commissie,
in het openbaar te verdedigen verdedigen in de Agnietenkapel
op 17 september 2009, te 10:00 uur
door
Lars Peter Klieverik
Geboren te Hengelo (Overijssel), Nederland
proefschrift Klieverik.indb 3
4-8-2009 15:22:17
Promotiecommissie
Promotores:
Prof. dr. E. Fliers
Prof. dr. H.P. Sauerwein
Co-promotores:
Dr. A. Kalsbeek
Dr. ir. M.T. Ackermans
Overige leden:
Prof. dr. R.M. Buijs
Prof. dr. M.M. Levi
Prof. dr. R.P.J. Oude Elferink
Prof. dr. J.A. Romijn
Prof. dr. ir. T.J.Visser
Prof. dr. W.M. Wiersinga
Faculteit der Geneeskunde
proefschrift Klieverik.indb 4
4-8-2009 15:22:17
voor mijn ouders
proefschrift Klieverik.indb 5
4-8-2009 15:22:17
Table of contents
Chapter 1
General Introduction
9
1.1 Thyroid Hormone (TH)
11
1.2 TH and the central nervous system
11
1.3 a Metabolic alterations during thyrotoxicosis
13
b Dynamic measurement of metabolic fluxes
14
1.4 Hypothalamic regulation of glucose metabolism
15
1.5 Thyronamines; metabolic effects of rapidly acting TH analogues
18
1.6 General hypothesis
18
1.7 Thesis outline
19
Chapter 2
Thyroid hormone effects on whole body energy homeostasis and tissuespecific fatty acid uptake in vivo.
Submitted
27
Chapter 3
Effects of thyrotoxicosis and selective hepatic autonomic denervation on
glucose metabolism in rats.
American Journal of Physiology; Endocrinology and Metabolism 2008:
294, E513-520.
47
Chapter 4
Thyroid hormone modulates glucose production via a sympathetic
pathway from the hypothalamic paraventricular nucleus to the liver.
Proceedings of the National Academy of Sciences of the United States
of America 2009: 106(14), 5966-5971.
63
Chapter 5
Effects of systemic and intracerebroventricular administration of
3-iodothyronamine (T1AM) and thyronamine (T0AM) on glucose
metabolism in rats.
Journal of Endocrinology 2009: 201(3), 377-386
79
Chapter 6
Endocrine determinants of changes in energy expenditure and body
weight after cessation of anti-thyroid drugs in euthyroid patients treated
for Graves’ hyperthyroidism.
Submitted
97
Chapter 7
General discussion
109
7.1 Historical perspective
7.2 Biological relevance
7.3 A new role for the hypothalamus in sensing hormonal signals and
modulating autonomic output; involvement of TH
7.4 Potential clinical relevance
proefschrift Klieverik.indb 6
4-8-2009 15:22:18
Chapter 8
proefschrift Klieverik.indb 7
English summary
123
Nederlandse samenvatting
129
Author affiliations
133
Dankwoord
135
Biography
139
4-8-2009 15:22:18
proefschrift Klieverik.indb 8
4-8-2009 15:22:21
1
proefschrift Klieverik.indb 9
General Introduction
4-8-2009 15:22:24
proefschrift Klieverik.indb 10
4-8-2009 15:22:24
1.1 Thyroid Hormone
Chapter 1
11
General Introduction
First isolated in 1914 by Kendall (1), and first synthesized in 1925 by Harrington (2) thyroxine (T4)
is a classic hormone that is used worldwide to treat millions of patients with thyroid disorders.
During the past decades much progress has been made in the understanding of thyroid hormone
(TH) physiology and a substantial part of TH biology has been elucidated. T4 is the main secretory
product of the thyroid gland. In humans, it comprises ~80% of the THs secreted, the remaining
~20% being secreted as triiodothyronine (T3). T4 has only limited affinity for the nuclear thyroid
hormone receptors (TRs) as compared with T3, which is regarded the primary biologically active
form. In order to become bio-active, T4 has to be converted to T3 by outer-ring deiodination.
Furthermore, both T4 and T3 can be inactivated by inner-ring deiodination. These reactions are
catalysed by the iodothyronine deiodinases type 1, 2 and 3 (D1, D2 and D3), that are expressed
in a multitude of peripheral tissues, each deiodinase with its specific tissue distribution. Outer
ring deiodination, i.e. the activating pathway, is catalysed by D1 and D2. Inner ring deiodination
of T4 and T3 to lower iodothyronines that have no affinity for the TRs, i.e. the inactivating
pathway, is catalysed by both D1 and D3 (3).
Thyroidal TH secretion is regulated via a classical central negative feedback mechanism.
Thyrotropin-releasing hormone (TRH) is synthesized by neurons in the paraventricular nucleus
(PVN) of the hypothalamus and reaches the anterior pituitary via the median eminence and
the portal system. In the anterior pituitary TRH stimulates the secretion of thyroid stimulating
hormone (TSH or thyrotropin) from the thyrotropes. TSH is the major factor stimulating thyroid
hormone synthesis and secretion by the thyroid gland. In turn, both circulating T3 and T4 exert
an inhibitory effect on the release of TRH and TSH from the hypothalamus and the pituitary,
respectively.
The nuclear thyroid hormone receptors (TR) are products of the TRα and TRβ genes. These
receptors are members of the ligand-dependent transcription modulator family. This implies that
upon intra-nuclear binding of T3 to a TR and via interaction with several co-factors, the complex
binds to a thyroid hormone responsive element (TRE) in the promoter region of a TH-responsive
gene, ultimately affecting gene transcription (4). Many actions of TH can be explained by
this transcriptional mechanism of action. TH transport across the cell membrane is required,
since both deiodinases and TRs are located intracellularly. In recent years, several specific TH
transporters have emerged, such as the monocarboxylate transporters (MCT) and the organic
anion transporting polypeptide (OATP) family. Of the latter family, OATP1C1 appears to be
critical for transport of T4 across the blood-brain-barrier (5).
1.2 Thyroid Hormone and the central nervous
system
Although a pivotal role of thyroid hormones in the developing mammalian brain has been long
established and extensively documented (for review see (6)), the adult central nervous system
(CNS) was generally assumed to be a thyroid hormone-insensitive organ until the 1980s (7). A
number of findings in more recent literature have casted doubt upon this assumption and have
gradually led to the general acceptance of the notion that the adult brain is a highly TH-sensitive
proefschrift Klieverik.indb 11
4-8-2009 15:22:24
12
organ. This notion is in line with the well-known psychomotor and cognitive dysfunction often
observed in adult-onset thyroid disorders, especially hypothyroidism (8).
By use of radio-labeled iodothyronines, the group of dr. M.B. Dratman showed for the first time
that THs can be transported across the blood-brain-barrier and choroid plexus in rats (9;10). Further
studies showed uptake and concentration of radio-labeled T3 in nerve terminals (11-13), pointing
to a new role for THs as (precursors for) amino-acid neurotransmitters or neuro-modulators
(13;14). A molecular basis for an active transport mechanism for THs in the brain was provided
by the identification of TH-specific transporters both in neurons and in capillaries of the choroid
plexus (15;16). The existence of active thyroid hormone transport in the human brain may explain
why concentrations of free T4 (FT4) and free T3 (FT3) are within the same range, or even higher,
in the cerebrospinal fluid (CSF) as compared with plasma (17). THs and several TH derivatives are
widely distributed in the CNS. For instance, both T4 and T3 concentrations have been reported in
the pmol/g range in rat hypothalamus (18). These tissue concentrations are similar to T4 and T3
concentrations in rat liver, which is considered to be a major TH target tissue.
Importantly, also the thyroid hormone receptors TRα and TRβ, as well as the deiodinating enzymes
D2 and D3, are widely distributed both in the rat and human CNS, including the hypothalamus
(19-23). It has been estimated that more than 75% of neuronal T3 is derived from conversion of T4
to T3 by D2, underlining the importance of D2 in regulating T3 bio-availability in the CNS (24).
A phenomenon pointing to an important functional role of T3 in the CNS is the efficient homeostatic
mechanism in the brain ensuring notably stable local T3 -tissue concentrations in the face of
pathological changes in systemic TH status. For example, when rats are rendered hypothyroid by
thyroidectomy, D2 activity in the cerebral cortex rapidly increases, promoting the local conversion
of T4 into T3, and this can be prevented by the administration of systemic TH replacement at the
same time (25). Similar adaptive mechanisms have been reported in the hypothalamus, where
hypothyroidism elevates D2 mRNA expression and activity, whereas thyrotoxicosis decreases
local D2 levels (26). In addition, the TH inactivating enzyme D3 is highly T3-responsive throughout
the CNS, as evidenced by a dose-dependent induction during thyrotoxicosis (27;28). These TH
dependent adaptations in local deiodinase enzyme expression result in remarkably stable tissue
T3 concentrations in the CNS over a wide range of systemic TH conditions (29).
Furthermore, under non-pathological conditions, hypothalamic deiodinase levels also show
marked fluctuations in association with physiologic processes and stimuli such as day-night
rhythmicity (30), seasonal (i.e. photoperiod) changes (31) and food availability (32). However,
in most of these processes it remains unclear if deiodinase regulated TH bio-availability in the
hypothalamus may play a regulatory role in the (metabolic) adaptation of the organism to these
conditions, or if these represent epiphenomena (for review see (33)).
Recently, with the advances in functional neuro-imaging techniques, there have been several
clinical reports describing metabolic changes in the CNS associated with altered TH status in
patients (34-36), possibly explaining the neuro-cognitive symptoms in these patients. Interestingly,
even in subclinical hypothyroidism, which may be regarded as a subtle thyroid hormone deficit,
fMRI revealed malfunction of brain areas critical for working memory, which could be ameliorated
by thyroxine supplementation (37).
proefschrift Klieverik.indb 12
4-8-2009 15:22:24
1.3a Metabolic alterations during thyrotoxicosis
13
General Introduction
proefschrift Klieverik.indb 13
Chapter 1
The role of TH in the regulation of lipid and glucose metabolism has been the subject of study
ever since the recognition of the link between thyroid function and body weight. The effects of
TH on metabolism are among the foremost actions of TH in vertebrates. This is illustrated by
the profound alterations in (energy) metabolism during hyper- and hypothyroidism, and also by
the metabolic adaptations of the HPT-axis to physiologic stressors such as food deprivation and
critical illness (38).
Hypermetabolism is one of the hallmarks of thyrotoxicosis, reflected in an increase of resting
energy expenditure (REE). This increased REE can be measured in humans by indirect calorimetry,
assessing whole body O2 consumption and CO2 production. In fact, this was an important tool
in the diagnosis of thyrotoxicosis before the development of sensitive radioimmunoassays for T4
and TSH (39;40).
To ensure replenishment of macronutrients, appetite is simultaneously stimulated during
thyrotoxicosis, which is obviously advantageous. There are recent data to suggest that this
hyperphagia represents a direct TH effect in the hypothalamic ventromedial nucleus, which
is involved in appetite regulation (41). In addition, TH may affect appetite indirectly via the
hypothalamic neuropeptide Y (NPY)/ Agouti related peptide (AGRP) system that is critical for
appetite regulation (42).
The alterations in glucose metabolism during thyrotoxicosis have been extensively studied.
Whole body glucose utilization is increased during chronic human thyrotoxicosis (43). To provide
the substrates needed, endogenous glucose production (EGP) is increased, paralleled by a mild
increase of plasma glucose concentration (44). The EGP increase during thyrotoxicosis is facilitated
by the increased activity of relevant hepatic gluconeogenic enzymes such as phosphoenolpyruvate
carboxykinase (PEPCK) and pyruvate carboxylase (45;46) and by increased hepatic expression of
the glucose transporter GLUT2 (47;48). The mechanism by which T3 stimulates of these genes
is complex as illustrated by T3-mediated modulation of PEPCK expression. T3 directly modulates
transcription of these genes via he nuclear TRs binding to their thyroid hormone responsive
elements. In addition, T3 appears to act on a pre-translational level, amplifying cyclic-AMPmediated induction of transcription. Thereby, it induces the sensitivity to other (cyclic-AMP
regulated) hormones such as catecholamines and glucagon (49).
Although evidence is scarce, thyroid hormone also seems to interfere with hepatic sensitivity to
insulin in humans, the most important hormone inhibiting EGP (50). This may explain the often
reported glucose intolerance in hyperthyroid patients (44;51;52), and is likely to contribute to
the T3-induced EGP increase during thyrotoxicosis. Data on insulin sensitivity on the level of
peripheral glucose uptake during thyrotoxicosis are conflicting as decreased as well as unaltered
insulin sensitivities have been reported (44).
Fatty acids (FA) are an additional important substrate fuelling the increase in REE during thyrotoxicosis
(53). In the acute phase, i.e. in the first days after the development of thyrotoxicosis, increased fat
oxidation appears to be the sole mechanism by which the increase in energy demand is fulfilled
(54). The FAs needed for this, can be provided by several mechanisms. First, lipolysis, i.e. hydrolysis
of triglycerides into FAs and glycerol, is stimulated. Increased sensitivity to catecholamines, as
4-8-2009 15:22:24
14
mentioned above, seems to be a principal mechanism by which THs promote lipolysis. Second,
hepatic de novo lipogenesis is promoted by T3 via induction of lipogenic enzymes such as fatty
acid synthase (55). Third, TH increases the amount of available nutrients by increasing food intake,
including FAs, which are to be absorbed from the gastrointestinal tract.
Finally, increased quantities of amino acids, provided by an increase in muscle proteolysis during
thyrotoxicosis, are available to enter the pathway of gluconeogenesis (56;57). Increased striated
muscle proteolysis may well be responsible for muscle wasting and myopathies associated with
severe thyrotoxicosis (57).
In summary, thyrotoxicosis is characterized by complex metabolic alterations that point to a
generalized catabolic state. Both glucose, lipid as well as protein metabolism are involved in the
hypermetabolism of thyrotoxicosis (43).
1.3b Dynamic measurement of metabolic fluxes
Plasma concentrations of metabolically relevant substances represent the final result of (often
reciprocal) metabolic fluxes of the respective substance and therefore provide very limited
information about kinetics. For example, plasma glucose concentration is determined by glucose
production, in non-fasting conditions mainly accounted for by the liver, as well as glucose
disposal by tissues such as muscle, white adipose tissue and the brain. In order to quantify
these processes, the technique of stable isotope dilution is the tool of choice. This technique
permits dynamic measurement of metabolic fluxes in vivo. For the experiments described in
chapters 3, 4 and 5 of this thesis, we adapted the stable isotope (6,6-2H2-glucose) dilution
technique, previously used in our department for measuring glucose fluxes in patients, for use
in laboratory rats. Similarly, to permit measurement of tissue-specific fatty acid (FA) uptake, and
more specifically to differentiate between uptake of FAs from different sources, we adapted a
dual-isotope technique originally developed in mice (58) for use in rats, as described in chapter
2 of the present thesis.
The principle of the stable isotope dilution technique used in this thesis to dynamically measure
EGP in vivo can be summarized as follows. In the fasted state, when the amount of nutrients
entering the circulation from the gastro-intestinal tract is negligible, a continuous intravenous
infusion of 6,6-2H2-glucose (i.e. the glucose tracer) is started. After equilibration time (which is
shortened by applying a prime bolus), steady state will ensue. This means that if metabolic fluxes
involved in glucose homeostasis remain constant, an equilibrium will be reached between the
glucose tracer and unlabelled (endogenously produced) glucose. This is reflected in a constant
ratio of 6,6-2H2-glucose to unlabelled glucose in plasma, termed 6,6-2H2-glucose enrichment,
which can be accurately measured by gas chromatography coupled to mass spectrometry (GC/
MS). Endogenous glucose production (EGP) can now be calculated from plasma 6,6-2H2-glucose
enrichment by using Steele’s equation for steady state conditions (59). Measuring EGP by
using the isotope dilution technique is based on the principle that whereas 6,6-2H2-glucose
enrichment is not affected by glucose disposal (i.e. uptake), the 6,6-2H2-glucose pool is diluted
by the production of endogenous (unlabelled) glucose. Evidently, an important assumption is
that metabolic processing of labelled and unlabelled glucose is identical, i.e. the process of
proefschrift Klieverik.indb 14
4-8-2009 15:22:24
glucose uptake does not discriminate between labelled and unlabelled glucose, which appears
reasonable (60).
proefschrift Klieverik.indb 15
15
General Introduction
The brain is a major energy consuming organ, and depends almost entirely on glucose as a
substrate. It is therefore not surprising that the plasma glucose concentration is tightly controlled
and that evolution has provided us with an efficient but complex system for maintenance of energy
and glucose homeostasis. This regulatory system is concentrated in the hypothalamus, and there
is substantial evidence pointing to the arcuate nucleus (ARC) as one of the main hypothalamic
nuclei involved. The ARC is localized at the base of the hypothalamus/third ventricle where the
blood-brain-barrier for large molecules is absent. It expresses a variety of hormone receptors,
including TRs. This makes sense, as in order to respond adequately to circulating nutrient and
hormonal signals conveying peripheral information regarding the energy status of the body,
neurons should ideally be in direct contact with these signals. After sensing these nutrient and
hormonal signals in the ARC, the information is conveyed to a number of nuclei including the
hypothalamic paraventricular nucleus (PVN) via well established neural circuits (61). The PVN
subsequently integrates this information with signals from other brain regions, and in turn uses
several output pathways for the regulation of peripheral metabolism. One of these pathways is
the neuro-endocrine system sending humoral signals via the median eminence to the anterior
pituitary gland, in turn regulating a variety of endocrine glands like the adrenal or thyroid gland
(hypothalamus-pituitary-adrenal (HPA) and hypothalamus-pituitary-thyroid (HPT) axis, respectively).
Another pathway arising from the PVN sends neural information from its pre-autonomic neurons
to the autonomic nervous system (ANS). PVN pre-autonomic neurons project to motor-neurons
of both branches of the ANS, i.e. the sympathetic and the parasympathetic nervous system.
Motorneurons of the sympathetic nervous system are located in the intermediolateral column
in the spinal cord (IML), whereas the parasympathetic motorneurons can be found in the
dorsal motor nucleus of the vagus nerve (DMV) in the brainstem. From the DMV, pre-synaptic
parasympathetic projections run with the vagus nerve, synapsing with postsynaptic neurons in
ganglions close to the organs or tissues of projection. Pre-synaptic projections of the sympathetic
motorneurons are generally much shorter, synapsing in ganglia closer to the spinal cord. From
there onwards, post-synaptic fibers (forming the cervical sympathetic chain) run to their target
tissues (62-64) (Fig 1).
Via these ANS pathways, the hypothalamus can modulate metabolism in organs like the liver,
heart and adipose tissue. In addition, it can affect hormone secretion via autonomic innervation of
endocrine glands, for example glucagon and insulin secretion by the pancreas. In a similar way, the
hypothalamus may affect TH secretion via autonomic innervation of the thyroid gland (65;66).
The classical view on the brain’s involvement in the hormonal regulation of metabolism was
dominated by two one-way roads. Hormones were thought to be the main factor via which the
brain could control metabolic organs such as the liver. On the other hand, feedback information
from the periphery was thought to be conveyed to the brain mainly via the sensory, i.e. afferent,
Chapter 1
1.4 Hypothalamic regulation of glucose
metabolism
4-8-2009 15:22:25
16
Figure 1 Schematic representation of the autonomic projections from the hypothalamus that regulate hepatic
metabolism. Circulating hormones and nutrients, providing information regarding the metabolic status of the
body, are sensed within the hypothalamic arcuate nucleus (ARC), where the blood-brain-barrier (BBB) is largely
absent. The information is then is conveyed to pre-autonomic neurons in the paraventricular nucleus (PVN)
where it is integrated with information from other brain regions. The autonomic nervous system is used as
an efferent pathway to regulate hepatic metabolism. Pre-autonomic neurons project to the intermediolateral
column (IML) of the spinal cord and the dorsomedial nucleus (DMV) of the brain stem, where they synapse with
sympathetic an parasympathetic efferent neurons, respectively, projecting to the liver.
fibers of the ANS. However, this view has recently been extended by the concept that a multitude
of hormonal signals are sensed directly by the brain within the hypothalamus (ARC), and that
the hypothalamus in turn regulates metabolism in peripheral tissues via its connections with the
ANS, independently of its neuro-endocrine output. Thus, hormones not only affect metabolic
processes directly by hormone receptor-mediated actions in target tissues, but also indirectly by
actions in hypothalamic nuclei. The hypothalamus in turn uses effector-pathways such as the
ANS to control metabolic processes in target tissues. This can be accomplished by “direct” ANSmediated modulation of metabolism in target tissues, but also by ANS mediated modulation of
the sensitivity to hormones in these target tissues (67;68).
This concept is nicely illustrated by insulin, which has among other important anabolic effects, a
major plasma glucose-lowering influence, when secreted by the pancreas in response to a meal.
This plasma glucose-lowering effect is accomplished by facilitating glucose uptake in muscle and
adipose tissue, as well as by inhibiting glucose production by the liver. The latter effect is mediated
by insulin actions on several levels of hepatic glucose metabolism (i.e. both glycogenolysis and
proefschrift Klieverik.indb 16
4-8-2009 15:22:25
Chapter 1
17
General Introduction
gluconeogenesis) and is well established (69). Recently, it has become clear that in rodents, EGP
can also be inhibited by low dose intracerebroventricular (icv) infusion of insulin, independently
of circulating insulin concentrations. The pathways that are responsible for this central effect of
insulin on hepatic glucose metabolism have been unravelled to a large extent, and involve insulin
receptors in the ARC, hypothalamic potassium-dependent ATP channels, NPY projections from
the ARC and autonomic output from the hypothalamus to the liver (70;71). In addition, insulin
modulates the release of neuropeptide Y (NPY), agouti related peptide (AGRP) and α-melanocytestimulating hormone (αMSH), which are among the main hypothalamic neuropeptides involved
in the regulation of energy and glucose homeostasis (72;73).
There are convincing data that underline the physiological relevance of these central insulin
actions. First, a reduction of ~50% in the expression insulin receptors in the ARC, local
obstruction of insulin binding to its receptor or pharmacological blockade of downstream insulin
signalling (PI3K) in the ARC all markedly decrease the ability of insulin to inhibit EGP (70;74). In
addition, selective hepatic parasympathectomy leads to a ~50% loss of the inhibitory effect of
hyperinsulinemia on EGP, as measured under hyperinsulinemic clamp conditions (75). Overall,
these data strongly support the notion that under physiological conditions, insulin’s actions in
the hypothalamus transmitted to the liver via the autonomic nervous system, are responsible for
a significant part of its EGP-lowering effects.
There are convincing data showing that besides suppressing EGP via parasympathetic signalling,
the hypothalamus can also stimulate sympathetic projections to the liver in order to increase
EGP (76). Thus, the hypothalamus can reciprocally modulate EGP by using its sympathetic and
parasympathetic autonomic outputs to the liver. Other hormones/nutrients that are sensed
within the hypothalamus and in turn evoke regulatory metabolic responses in peripheral organs
include leptin (73), glucocorticoids (77), estrogen (78) and fatty acids (79).
Figure 2 Structural formulas of the iodothyronines thyroxine
(T4), 3-iodothyronine (T3) and 3-iodothyronamine (T1AM). T1AM,
like other thyronamines, is a structural homologue of T3 and T4.
Note that a molecule of T1AM can be obtained by decarboxylation
(removal of CO2H group), and deiodination of T3 or T4.
proefschrift Klieverik.indb 17
4-8-2009 15:22:25
1.5 Thyronamines; metabolic effects of rapid
acting Thyroid Hormone analogues
18
The recently discovered thyronamines, such as 3-iodothyronamine (T1AM) and the fully deiodinated
thyronamine (T0AM) are decarboxylated and (to a certain extent) deiodinated analogues of TH
(fig 2). These compounds are agonists of the G-protein coupled trace amino acid associated
receptor 1 (TAAR1), a member of a large family of (former) orphan receptors. T1AM and T0AM
have been reported to occur in a multitude of rodent tissues among which liver, brain and
blood (80). Interestingly, the TAAR1 is expressed in the ARC (81). Upon systemic administration
to rodents, thyronamines induce profound physiological effects. These effects occur on a very
rapid timescale (seconds to minutes), and interestingly, many of them are opposite in direction
to the effects of THs. For example, thyronamines induce profound hypothermia, bradycardia
and a decrease of cardiac output. In addition, a number of marked metabolic effects have
been described. The metabolic phenotype that follows systemic thyronamine infusion resembles
metabolic adaptations to fasting in several ways, i.e. ketogenesis, hypoinsulinemia and a shift
to fat oxidation at the cost of glucose oxidation. In addition, thyronamines induce marked
hyperglycemia. The rapid timescale on which these effects occur, may be in line with a neural
mechanism of action. At present it is unknown if thyronamines act in the CNS to modulate
metabolism in peripheral organs like the liver.
1.6 General hypothesis
It has been long noted that there is a striking similarity between many of the symptoms of
thyrotoxicosis on the one hand and the effects of sympathetic nervous system (SNS) stimulation
on the other. This holds true for, e.g., tachycardia, tremor, increased perspiration and nervousness.
Already by the end of the 19th century this led to the surgical treatment of thyrotoxicosis by
resection of the cervical sympathetic chain (82) and later by high spinal anaesthesia or adrenal
denervation. The latter procedures were reported in 135 thyrotoxic patients with a contraindication for thyroidectomy by Dr. George Crile in Cleveland (US), one of the pioneers in the field
(83). Although these practices were gradually abandoned, it is still common practice nowadays
to start treatment of severe thyrotoxicosis with beta-adrenergic blockers. These drugs may induce
a rapid symptomatic relief enabling anti-thyroid drugs such as methimazole to gradually reduce
thyroid hormone synthesis. Interestingly, also many of the metabolic consequences of thyrotoxicosis
as outlined above (i.e. increased EGP, lipolysis, decreased insulin sensitivity) are similar to effects of
increased sympathetic tone. Along these lines, a study in hyperthyroid patients showed enhanced
noradrenalin (NE) secretion in subcutaneous white adipose tissue (WAT) without changes in
circulating catecholamines, probably causing higher rates of WAT lipolysis in these patients (84).
In addition, during thyrotoxicosis autonomic input to the heart shifts to increased sympathetic
and decreased parasympathetic tone (85-87). Collectively, these data raise the notion of increased
sympathetic tone during thyrotoxicosis.
Our group has reported the functional neuro-anatomy of sympathetic and parasymathetic output
from the hypothalamic PVN to the liver in rats (64) and more recently a major role for this autonomic
output in the regulation of hepatic glucose metabolism has emerged (70;71;76). Interestingly, in
proefschrift Klieverik.indb 18
4-8-2009 15:22:25
the PVN and ARC, as well as additional hypothalamic nuclei, abundant expression of all major
thyroid hormone receptor isoforms has been reported in both humans and rodents (19;22).
In this thesis, we describe our efforts to test the hypothesis that thyroid hormones (or TH
derivatives) modulate metabolism on the level of peripheral organs (i.e. the liver) via actions in
the brain, and more specifically, the hypothalamus.
First, we studied the effects of thyroid status on whole body energy metabolism and tissuespecific metabolism, in particular on fatty acid kinetics. For this, we performed experiments with
hypothyroid, euthyroid and thyrotoxic rats using metabolic cages and a dual isotope infusion
technique. The results are described in chapter 2.
In chapter 3, we aimed to elucidate a possible role of the ANS projections to the liver in the
alterations of hepatic glucose metabolism induced by thyrotoxicosis. In order to do this, we
assessed hepatic glucose production and its sensitivity to insulin by combining stable isotope
dilution and hyperinsulinemic euglycemic clamping in euthyroid and thyrotoxic rats that
underwent selective hepatic autonomic (i.e. either sympathetic or parasympathetic) denervation
or a sham operation.
As a next step, we administered T3 both intracerebroventricularly and selectively to the PVN
by microdialysis in euthyroid rats, while assessing EGP with stable isotope infusion. Moreover,
we combined hypothalamic T3 administration with selective sympathetic denervation of the
liver to study the involvement of the sympathetic hepatic projections in the effects on hepatic
glucose production induced by hypothalamic T3. These experiments are described and discussed
in chapter 4.
The thyronamines T1AM and T0AM, analogues of TH, exhibit neurotransmitter-like properties,
and the physiologic profile that evolves upon administration of these compounds in rodents may
well fit with involvement of the hypothalamus in these actions. In the experiments described
in chapter 5, we studied for the first time if the effects of T1AM and T0AM on hepatic glucose
metabolism and glucoregulatory hormones can be explained by actions of thyronamines in the
CNS.
Finally, to explore the interrelationship between THs and energy metabolism in the clinical setting,
we performed a study in patients with Graves disease rendered euthyroid by pharmacotherapy
aimed at blocking thyroid hormone synthesis by an antithyroid drug and restoring plasma
thyroid hormone concentrations by exogenous substitution of thyroxine (so called ‘block and
replacement therapy” or BRT). We studied these patients on 2 occasions, i.e. during BRT and 12
proefschrift Klieverik.indb 19
19
General Introduction
1.7 Thesis outline
Chapter 1
In the present thesis we explore the hypothesis that a significant part of the metabolic alterations
during thyrotoxicosis are mediated via effects of thyroid hormones in the central nervous system.
More specifically, we hypothesize that the effects of thyroid hormone on hepatic glucose
metabolism in vivo are mediated to a significant extent via (pre-autonomic) neurons in the
hypothalamus that contact the liver via autonomic projections.
4-8-2009 15:22:25
weeks after BRT cessation, providing circumstances with subtle differences in plasma free (F)T4
and FT3 concentrations. The data are described and discussed in chapter 6.
In chapter 7, we aim to place our findings into a wider perspective, with a special emphasis on
biological as well as potential clinical relevance.
Reference List
20
1. Kendall EC 1919 Isolation of the Iodine Compound Which Occurs in the Thyroid . J.Biol.Chem. 39,
125-147
2. Harington CR 1926 Chemistry of Thyroxine: Constitution and Synthesis of Desiodo-Thyroxine. Biochem J
20:300-313
3. Bianco AC, Salvatore D, Gereben B, Berry MJ, Larsen PR 2002 Biochemistry, cellular and molecular
biology, and physiological roles of the iodothyronine selenodeiodinases. Endocr Rev 23:38-89
4. Yen PM 2001 Physiological and molecular basis of thyroid hormone action. Physiol Rev 81:1097-1142
5. Visser WE, Friesema EC, Jansen J, Visser TJ 2008 Thyroid hormone transport in and out of cells. Trends
Endocrinol Metab 19:50-56
6. Bernal J 2007 Thyroid hormone receptors in brain development and function. Nat Clin Pract Endocrinol
Metab 3:249-259
7. Sourkes T. In: Basic Neurochemistry, Siegel GJ, Alberts RW, Katzman R, Agranoff BW (Eds.) 1976, p728
Little Brown, Boston.
8. Dugbartey AT 1998 Neurocognitive aspects of hypothyroidism. Arch Intern Med 158:1413-1418
9. Cheng LY, Outterbridge LV, Covatta ND, Martens DA, Gordon JT, Dratman MB 1994 Film autoradiography
identifies unique features of [125I]3,3’5’-(reverse) triiodothyronine transport from blood to brain. J
Neurophysiol 72:380-391
10. Dratman MB, Crutchfield FL, Schoenhoff MB 1991 Transport of iodothyronines from bloodstream to
brain: contributions by blood:brain and choroid plexus:cerebrospinal fluid barriers. Brain Res 554:229236
11. Dratman MB, Crutchfield FL, Axelrod J, Colburn RW, Thoa N 1976 Localization of triiodothyronine in
nerve ending fractions of rat brain. Proc Natl Acad Sci U S A 73:941-944
12. Dratman MB, Crutchfield FL 1978 Synaptosomal [125I]triiodothyronine after intravenous [125I]thyroxine.
Am J Physiol 235:E638-E647
13. Dratman MB, Futaesaku Y, Crutchfield FL, Berman N, Payne B, Sar M, Stumpf WE 1982 Iodine-125-labeled
triiodothyronine in rat brain: evidence for localization in discrete neural systems. Science 215:309-312
14. Dratman MB, Gordon JT 1996 Thyroid hormones as neurotransmitters. Thyroid 6:639-647
15. Friesema EC, Ganguly S, Abdalla A, Manning Fox JE, Halestrap AP, Visser TJ 2003 Identification of
monocarboxylate transporter 8 as a specific thyroid hormone transporter. J Biol Chem 278:4012840135
16. Pizzagalli F, Hagenbuch B, Stieger B, Klenk U, Folkers G, Meier PJ 2002 Identification of a novel human
organic anion transporting polypeptide as a high affinity thyroxine transporter. Mol Endocrinol 16:22832296
17. Kirkegaard C, Faber J 1991 Free thyroxine and 3,3’,5’-triiodothyronine levels in cerebrospinal fluid in
patients with endogenous depression. Acta Endocrinol (Copenh) 124:166-172
18. Pinna G, Brodel O, Visser T, Jeitner A, Grau H, Eravci M, Meinhold H, Baumgartner A 2002 Concentrations
of seven iodothyronine metabolites in brain regions and the liver of the adult rat. Endocrinology
143:1789-1800
proefschrift Klieverik.indb 20
4-8-2009 15:22:25
19. Alkemade A, Vuijst CL, Unmehopa UA, Bakker O, Vennstrom B, Wiersinga WM, Swaab DF, Fliers E
2005 Thyroid hormone receptor expression in the human hypothalamus and anterior pituitary. J Clin
Endocrinol Metab 90:904-912
21. Alkemade A, Friesema EC, Unmehopa UA, Fabriek BO, Kuiper GG, Leonard JL, Wiersinga WM, Swaab
DF, Visser TJ, Fliers E 2005 Neuroanatomical pathways for thyroid hormone feedback in the human
hypothalamus. J Clin Endocrinol Metab 90(7):4322-4334
23. Fliers E, Alkemade A, Wiersinga WM, Swaab DF 2006 Hypothalamic thyroid hormone feedback in health
and disease. Prog Brain Res 153:189-207
24. Crantz FR, Silva JE, Larsen PR 1982 An analysis of the sources and quantity of 3,5,3’-triiodothyronine
specifically bound to nuclear receptors in rat cerebral cortex and cerebellum. Endocrinology 110:367375
21
General Introduction
22. Lechan RM, Qi Y, Berrodin TJ, Davis KD, Schwartz HL, Strait KA, Oppenheimer JH, Lazar MA 1993
Immunocytochemical delineation of thyroid hormone receptor beta 2-like immunoreactivity in the rat
central nervous system. Endocrinology 132:2461-2469
Chapter 1
20. Lechan RM, Qi Y, Jackson IM, Mahdavi V 1994 Identification of thyroid hormone receptor isoforms
in thyrotropin-releasing hormone neurons of the hypothalamic paraventricular nucleus. Endocrinology
135:92-100
25. Leonard JL, Kaplan MM, Visser TJ, Silva JE, Larsen PR 1981 Cerebral cortex responds rapidly to thyroid
hormones. Science 214:571-573
26. Tu HM, Kim SW, Salvatore D, Bartha T, Legradi G, Larsen PR, Lechan RM 1997 Regional distribution
of type 2 thyroxine deiodinase messenger ribonucleic acid in rat hypothalamus and pituitary and its
regulation by thyroid hormone. Endocrinology 138:3359-3368
27. Tu HM, Legradi G, Bartha T, Salvatore D, Lechan RM, Larsen PR 1999 Regional expression of the type 3
iodothyronine deiodinase messenger ribonucleic acid in the rat central nervous system and its regulation
by thyroid hormone. Endocrinology 140:784-790
28. Escobar-Morreale HF, Obregon MJ, Hernandez A, Escobar dR, Morreale dE 1997 Regulation of
iodothyronine deiodinase activity as studied in thyroidectomized rats infused with thyroxine or
triiodothyronine. Endocrinology 138:2559-2568
29. Escobar-Morreale HF, Obregon MJ, Escobar del RF, Morreale de EG 1995 Replacement therapy for
hypothyroidism with thyroxine alone does not ensure euthyroidism in all tissues, as studied in
thyroidectomized rats. J Clin Invest 96:2828-2838
30. Kalsbeek A, Buijs RM, van SR, Kaptein E, Visser TJ, Doulabi BZ, Fliers E 2005 Daily variations in type II
iodothyronine deiodinase activity in the rat brain as controlled by the biological clock. Endocrinology
146:1418-1427
31. Yoshimura T, Yasuo S, Watanabe M, Iigo M, Yamamura T, Hirunagi K, Ebihara S 2003 Light-induced
hormone conversion of T4 to T3 regulates photoperiodic response of gonads in birds. Nature 426:178181
32. Diano S, Naftolin F, Goglia F, Horvath TL 1998 Fasting-induced increase in type II iodothyronine deiodinase
activity and messenger ribonucleic acid levels is not reversed by thyroxine in the rat hypothalamus.
Endocrinology 139:2879-2884
33. Lechan RM, Fekete C 2005 Role of thyroid hormone deiodination in the hypothalamus. Thyroid 15:883897
34. Constant EL, de Volder AG, Ivanoiu A, Bol A, Labar D, Seghers A, Cosnard G, Melin J, Daumerie C 2001
Cerebral blood flow and glucose metabolism in hypothyroidism: a positron emission tomography study.
J Clin Endocrinol Metab 86:3864-3870
35. Schreckenberger MF, Egle UT, Drecker S, Buchholz HG, Weber MM, Bartenstein P, Kahaly GJ 2006
Positron emission tomography reveals correlations between brain metabolism and mood changes in
hyperthyroidism. J Clin Endocrinol Metab 91:4786-4791
proefschrift Klieverik.indb 21
4-8-2009 15:22:25
36. Smith CD, Ain KB 1995 Brain metabolism in hypothyroidism studied with 31P magnetic-resonance
spectroscopy. Lancet 345:619-620
37. Zhu DF, Wang ZX, Zhang DR, Pan ZL, He S, Hu XP, Chen XC, Zhou JN 2006 fMRI revealed neural
substrate for reversible working memory dysfunction in subclinical hypothyroidism. Brain 129:29232930
38. Lechan RM, Fekete C 2006 The TRH neuron: a hypothalamic integrator of energy metabolism. Prog Brain
Res 153:209-235
22
39. Wiersinga WM, Chopra IJ 1982 Radioimmunoassay of thyroxine (T4), 3,5,3’-triiodothyronine (T3),
3,3’,5’-triiodothyronine (reverse T3, rT3), and 3,3’-diiodothyronine (T2). Methods Enzymol 84:272-303
40. Baron DN 1959 Estimation of the basal metabolic rate in the diagnosis of thyroid disease. Proc R Soc
Med 52:523-525
41. Kong WM, Martin NM, Smith KL, Gardiner JV, Connoley IP, Stephens DA, Dhillo WS, Ghatei MA, Small
CJ, Bloom SR 2004 Triiodothyronine stimulates food intake via the hypothalamic ventromedial nucleus
independent of changes in energy expenditure. Endocrinology 145:5252-5258
42. Ishii S, Kamegai J, Tamura H, Shimizu T, Sugihara H, Oikawa S 2003 Hypothalamic neuropeptide Y/Y1
receptor pathway activated by a reduction in circulating leptin, but not by an increase in circulating ghrelin,
contributes to hyperphagia associated with triiodothyronine-induced thyrotoxicosis. Neuroendocrinology
78:321-330
43. Moller N, Nielsen S, Nyholm B, Porksen N, Alberti KG, Weeke J 1996 Glucose turnover, fuel oxidation
and forearm substrate exchange in patients with thyrotoxicosis before and after medical treatment. Clin
Endocrinol (Oxf) 44:453-459
44. Dimitriadis GD, Raptis SA 2001 Thyroid hormone excess and glucose intolerance. Exp Clin Endocrinol
Diabetes 109 Suppl 2:S225-S239
45. Loose DS, Cameron DK, Short HP, Hanson RW 1985 Thyroid hormone regulates transcription of the
gene for cytosolic phosphoenolpyruvate carboxykinase (GTP) in rat liver. Biochemistry 24:4509-4512
46. Weinberg MB, Utter MF 1979 Effect of thyroid hormone on the turnover of rat liver pyruvate carboxylase
and pyruvate dehydrogenase. J Biol Chem 254:9492-9499
47. Mokuno T, Uchimura K, Hayashi R, Hayakawa N, Makino M, Nagata M, Kakizawa H, Sawai Y, Kotake
M, Oda N, Nakai A, Nagasaka A, Itoh M 1999 Glucose transporter 2 concentrations in hyper- and
hypothyroid rat livers. J Endocrinol 160:285-289
48. Weinstein SP, O’Boyle E, Fisher M, Haber RS 1994 Regulation of GLUT2 glucose transporter expression
in liver by thyroid hormone: evidence for hormonal regulation of the hepatic glucose transport system.
Endocrinology 135:649-654
49. Hoppner W, Sussmuth W, Seitz HJ 1985 Effect of thyroid state on cyclic AMP-mediated induction of
hepatic phosphoenolpyruvate carboxykinase. Biochem J 226:67-73
50. Cavallo-Perin P, Bruno A, Boine L, Cassader M, Lenti G, Pagano G 1988 Insulin resistance in Graves’
disease: a quantitative in-vivo evaluation. Eur J Clin Invest 18:607-613
51. Holness MJ, Sugden MC 1987 Hepatic carbon flux after re-feeding. Hyperthyroidism blocks glycogen
synthesis and the suppression of glucose output observed in response to carbohydrate re-feeding.
Biochem J 247:627-634
52. Holness MJ, Sugden MC 1987 Continued glucose output after re-feeding contributes to glucose
intolerance in hyperthyroidism. Biochem J 247:801-804
53. Bech K, Damsbo P, Eldrup E, Beck-Nielsen H, Roder ME, Hartling SG, Volund A, Madsbad S 1996 beta-cell
function and glucose and lipid oxidation in Graves’ disease. Clin Endocrinol (Oxf) 44:59-66
54. Oppenheimer JH, Schwartz HL, Lane JT, Thompson MP 1991 Functional relationship of thyroid hormoneinduced lipogenesis, lipolysis, and thermogenesis in the rat. J Clin Invest 87:125-132
proefschrift Klieverik.indb 22
4-8-2009 15:22:25
55. Radenne A, Akpa M, Martel C, Sawadogo S, Mauvoisin D, Mounier C 2008 Hepatic regulation of
fatty acid synthase by insulin and T3: evidence for T3 genomic and nongenomic actions. Am J Physiol
Endocrinol Metab 295:E884-E894
57. Riis AL, Jorgensen JO, Gjedde S, Norrelund H, Jurik AG, Nair KS, Ivarsen P, Weeke J, Moller N 2005
Whole body and forearm substrate metabolism in hyperthyroidism: evidence of increased basal muscle
protein breakdown. Am J Physiol Endocrinol Metab 288:E1067-E1073
59. Steele R 1959 Influences of glucose loading and of injected insulin on hepatic glucose output. Ann N Y
Acad Sci 82:420-430
60. Wolfe RR. Calculations of substrate kinetics: single pool model. Radioactive and stable isotope tracers
in biomedicine. Principles and practice of kinetic analysis. 2004: 215-258. Wiley-Liss, New York.
61. Seeley RJ, Drazen DL, Clegg DJ 2004 The critical role of the melanocortin system in the control of energy
balance. Annu Rev Nutr 24:133-149
23
General Introduction
58. Teusink B, Voshol PJ, Dahlmans VE, Rensen PC, Pijl H, Romijn JA, Havekes LM 2003 Contribution of fatty
acids released from lipolysis of plasma triglycerides to total plasma fatty acid flux and tissue-specific fatty
acid uptake. Diabetes 52:614-620
Chapter 1
56. Carter WJ, Van Der Weijden Benjamin WS, Faas FH 1981 Effect of experimental hyperthyroidism on
skeletal-muscle proteolysis. Biochem J 194:685-690
62. Kreier F, Fliers E, Voshol PJ, van Eden CG, Havekes LM, Kalsbeek A, Van Heijningen CL, Sluiter AA,
Mettenleiter TC, Romijn JA, Sauerwein HP, Buijs RM 2002 Selective parasympathetic innervation of
subcutaneous and intra-abdominal fat--functional implications. J Clin Invest 110:1243-1250
63. Kreier F, Kap YS, Mettenleiter TC, van Heijningen C, van d, V, Kalsbeek A, Sauerwein HP, Fliers E, Romijn
JA, Buijs RM 2005 Tracing from fat tissue, liver and pancreas: A neuroanatomical framework for the role
of the brain in type 2 diabetes. Endocrinology 147(3):1140-1147
64. La Fleur SE, Kalsbeek A, Wortel J, Buijs RM 2000 Polysynaptic neural pathways between the hypothalamus,
including the suprachiasmatic nucleus, and the liver. Brain Res 871:50-56
65. Kalsbeek A, Fliers E, Franke AN, Wortel J, Buijs RM 2000 Functional connections between the
suprachiasmatic nucleus and the thyroid gland as revealed by lesioning and viral tracing techniques in
the rat. Endocrinology 141:3832-3841
66. Klieverik LP, Kalsbeek A, Fliers E. Autonomic innervation of the thyroid gland and it’s functional
implications. www.hotthyroidology.com (online-journal of the European Thyroid Association) issue dec
2005.
67. Buijs RM, Wortel J, Van Heerikhuize JJ, Feenstra MG, Ter Horst GJ, Romijn HJ, Kalsbeek A 1999
Anatomical and functional demonstration of a multisynaptic suprachiasmatic nucleus adrenal (cortex)
pathway. Eur J Neurosci 11:1535-1544
68. Buijs RM, Kalsbeek A 2001 Hypothalamic integration of central and peripheral clocks. Nat Rev Neurosci
2:521-526
69. Barthel A, Schmoll D 2003 Novel concepts in insulin regulation of hepatic gluconeogenesis. Am J Physiol
Endocrinol Metab 285:E685-E692
70. Obici S, Zhang BB, Karkanias G, Rossetti L 2002 Hypothalamic insulin signaling is required for inhibition
of glucose production. Nat Med 8:1376-1382
71. van den Hoek AM, van Heijningen C, Schroder-van der Elst JP, Ouwens DM, Havekes LM, Romijn JA,
Kalsbeek A, Pijl H 2008 Intracerebroventricular administration of neuropeptide Y induces hepatic insulin
resistance via sympathetic innervation. Diabetes 57:2304-2310
72. Seeley RJ, Drazen DL, Clegg DJ 2004 The critical role of the melanocortin system in the control of energy
balance. Annu Rev Nutr 24:133-149
73. Sandoval D, Cota D, Seeley RJ 2008 The integrative role of CNS fuel-sensing mechanisms in energy
balance and glucose regulation. Annu Rev Physiol 70:513-535
proefschrift Klieverik.indb 23
4-8-2009 15:22:25
74. Obici S, Feng Z, Karkanias G, Baskin DG, Rossetti L 2002 Decreasing hypothalamic insulin receptors
causes hyperphagia and insulin resistance in rats. Nat Neurosci 5:566-572
75. Pocai A, Lam TK, Gutierrez-Juarez R, Obici S, Schwartz GJ, Bryan J, guilar-Bryan L, Rossetti L 2005
Hypothalamic K(ATP) channels control hepatic glucose production. Nature 434:1026-1031
76. Kalsbeek A, La FS, Van HC, Buijs RM 2004 Suprachiasmatic GABAergic inputs to the paraventricular
nucleus control plasma glucose concentrations in the rat via sympathetic innervation of the liver. J
Neurosci 24:7604-7613
24
77. Cusin I, Rouru J, Rohner-Jeanrenaud F 2001 Intracerebroventricular glucocorticoid infusion in normal
rats: induction of parasympathetic-mediated obesity and insulin resistance. Obes Res 9:401-406
78. Clegg DJ, Brown LM, Woods SC, Benoit SC 2006 Gonadal hormones determine sensitivity to central
leptin and insulin. Diabetes 55:978-987
79. Lam TK, Pocai A, Gutierrez-Juarez R, Obici S, Bryan J, Aguilar-Bryan L, Schwartz GJ, Rossetti L 2005
Hypothalamic sensing of circulating fatty acids is required for glucose homeostasis. Nat Med 11:320327
80. Scanlan TS, Suchland KL, Hart ME, Chiellini G, Huang Y, Kruzich PJ, Frascarelli S, Crossley DA, Bunzow
JR, Ronca-Testoni S, Lin ET, Hatton D, Zucchi R, Grandy DK 2004 3-Iodothyronamine is an endogenous
and rapid-acting derivative of thyroid hormone. Nat Med 10:638-642
81. Borowsky B, Adham N, Jones KA, Raddatz R, Artymyshyn R, Ogozalek KL, Durkin MM, Lakhlani PP,
Bonini JA, Pathirana S, Boyle N, Pu X, Kouranova E, Lichtblau H, Ochoa FY, Branchek TA, Gerald C 2001
Trace amines: identification of a family of mammalian G protein-coupled receptors. Proc Natl Acad Sci U
S A 98:8966-8971
82. Poncet MA 1897 Le traitement chirurgical des goitre exophthalmique par la section ou la résection due
sympathique cervical. Bull Acad Med 38:121
83. Crile GW 1929 The interdependence of the thyroid, adrenals and nervous system. Amer.J.Surg. 6:616
84. Haluzik M, Nedvidkova J, Bartak V, Dostalova I, Vlcek P, Racek P, Taus M, Svacina S, Alesci S, Pacak
K 2003 Effects of hypo- and hyperthyroidism on noradrenergic activity and glycerol concentrations in
human subcutaneous abdominal adipose tissue assessed with microdialysis. J Clin Endocrinol Metab
88:5605-5608
85. Burggraaf J, Tulen JH, Lalezari S, Schoemaker RC, De Meyer PH, Meinders AE, Cohen AF, Pijl H 2001
Sympathovagal imbalance in hyperthyroidism. Am J Physiol Endocrinol Metab 281:E190-E195
86. Cacciatori V, Bellavere F, Pezzarossa A, Dellera A, Gemma ML, Thomaseth K, Castello R, Moghetti P,
Muggeo M 1996 Power spectral analysis of heart rate in hyperthyroidism. J Clin Endocrinol Metab
81:2828-2835
87. Chen JL, Chiu HW, Tseng YJ, Chu WC 2006 Hyperthyroidism is characterized by both increased
sympathetic and decreased vagal modulation of heart rate: evidence from spectral analysis of heart rate
variability. Clin Endocrinol (Oxf) 64:611-616
proefschrift Klieverik.indb 24
4-8-2009 15:22:25
proefschrift Klieverik.indb 25
4-8-2009 15:22:25
proefschrift Klieverik.indb 26
4-8-2009 15:22:27
2
Thyroid hormone effects
on whole body energy
homeostasis and tissue-specific
fatty acid uptake in vivo
Lars P. Klieverik
Claudia P. Coomans
Erik Endert
Hans P. Sauerwein
Louis M. Havekes
Peter J. Voshol
Patrick C.N. Rensen
Johannes A. Romijn
Andries Kalsbeek
Eric Fliers
proefschrift Klieverik.indb 27
4-8-2009 15:22:28
Abstract
28
The effects of thyroid hormone (TH) status on energy metabolism and tissue-specific substrate
supply in vivo are incompletely understood at present. To study the effects of TH status on
energy metabolism and tissue-specific fatty acid (FA) fluxes, we used metabolic cages as well as
14C-labelled FA and 3H-labeled triglyceride (TG) infusion in rats treated with methimazole and
either 0 (hypothyroidism), 1.5 (euthyroidism) or 16.0 (thyrotoxicosis) μg/100g*day of thyroxine
for 11 days.
Thyrotoxicosis increased total energy expenditure (TEE) by 38% (p=0.02), resting energy
expenditure (REE) by 61% (p=0.002) and food intake by 18% (p=0.004). Hypothyroidism
tended to decrease TEE (10%; p=0.064), and REE (12%; p=0.025), but did not affect food
intake. TH status did not affect spontaneous physical activity (SPA). Thyrotoxicosis increased fat
oxidation (p=0.006), whereas hypothyroidism decreased glucose oxidation (p=0.035). Plasma FA
concentration was increased in thyrotoxic, but not in hypothyroid rats. Thyrotoxicosis increased
albumin-bound FA uptake in muscle and white adipose tissue (WAT), whereas hypothyroidism
had no effect in any tissue studied, suggesting mass-driven albumin-bound FA uptake. During
thyrotoxicosis, TG-derived FA uptake was increased in muscle and heart, unaffected in WAT, and
decreased in brown adipose tissue. Conversely, during hypothyroidism TG-derived FA uptake
was increased in WAT in association with increased lipoprotein lipase activity, but unaffected in
oxidative tissues and decreased in liver.
In conclusion, TH status determines EE independently of SPA. The changes in whole body lipid
metabolism are accompanied by tissue-specific changes in TG-derived FA uptake in accordance with
hyper- and hypometabolic states induced by thyrotoxicosis and hypothyroidism, respectively.
proefschrift Klieverik.indb 28
4-8-2009 15:22:28
Introduction
Chapter 2
29
Effects of thyroid status on energy homeostasis and fatty acid uptake
Thyroid hormone (TH) is a primary denominator of energy homeostasis, reflected by the strong
association between hyperthyroidism and increased energy expenditure (EE) in man. This
is exemplified by the widespread clinical use of calorimetry in addition to the determination
of protein bound iodine in diagnosing thyrotoxicosis (1;2) before sensitive thyroxine (T4) and
triiodothyronine (T3) RIAs became available in the 1970s (3). Whereas modulation of resting
EE by thyroid hormone status is well established in humans and rodents, it has been difficult
to assess TH effects on total EE (TEE) in vivo. In addition, the mechanism of the increased
EE induced by THs has remained incompletely understood (4). For example, few studies have
addressed how THs influence spontaneous physical activity (SPA) and if changes in SPA may
contribute to the alterations in EE associated with thyrotoxicosis and hypothyroidism in freely
moving organisms (5;6).
We have previously studied glucose metabolism during thyrotoxicosis in rats and found increased
endogenous glucose production and hepatic insulin resistance (7). Furthermore, thyrotoxicosis is
associated with major changes in lipid metabolism. Fatty acids (FA) are a preferential fuel source
during thyrotoxicosis (8;9), especially during the first days after the induction of thyrotoxicosis
(10). These FAs are provided to tissues mostly by hydrolysis (i.e. lipolysis) of circulating triglyceride
(TG)-rich lipoprotein particles by the enzyme lipoprotein lipase (LPL), located in the capillary
lumen. In addition, albumin-bound FAs are provided to the tissues from he plasma, a process
which is independent of LPL. Although there is evidence suggesting that LPL is regulated by TH
(11), it is unknown at present how FA fluxes via these two pathways are modulated by TH in
metabolically relevant tissues in vivo .
The aim of the present study was to examine the effects of thyrotoxicosis and hypothyroidism on
whole body energy metabolism and SPA in rats. To delineate how the effects of thyroid status
on whole body energy homeostasis are reflected in substrate (i.e. lipid) supply on the tissue level
in vivo, we additionally studied the rates of disappearance and tissue-specific partitioning of both
TG-derived and albumin-bound FAs. We report effects of thyroid hormone status on total energy
expenditure (TEE), resting energy expenditure (REE), SPA and substrate (i.e. glucose and lipid)
oxidation, that are paralleled by complex and tissue-specific effects of TH on FA uptake.
Materials and Methods
Two separate experiments were performed
In experiment #1, 3 groups of rats were studied, i.e. hypothyroid (Hypo, n=7), euthyroid (Eu,
n=7) and thyrotoxic rats (Tox, n=7). All groups were treated with methimazol (MMI) in drinking
water. After 7 days, all groups were implanted with subcutaneous osmotic mini-pumps (day (D)
0), delivering either vehicle (Hypo group), or thyroxine (T4) at a dose of 1.5 (Eu group) or 16.0
μg/100g*day (Tox group) (7). Rats were subsequently placed in metabolic cages for determining
TEE, food intake, respiratory exchange ratio (RER), and fat and glucose oxidation during a 48h
period (D9 and D10).
In experiment #2 (D11), hypothyroid (Hypo, n=9), euthyroid (Eu, n=8) and thyrotoxic rats (Tox,
n=7) were i.v. infused with albumin-bound 14C-oleate (FA) and VLDL-like emulsion-incorporated
proefschrift Klieverik.indb 29
4-8-2009 15:22:28
glycerol tri[3H]oleate (TG). This method enables measurement of FA turnover, tissue-specific FA
partitioning and differentiation between albumin-bound and TG-derived FA uptake on the tissue
level (12).
Animals
30
Twenty-nine male Wistar rats (Harlan, Horst, the Netherlands) were submitted to both
experimental procedures described below (exp #1 and exp #2). Animals were housed under
constant conditions of temperature (21±1 °C) and humidity (60±2%) with a 12-h light, 12-h
dark (L/D) schedule (lights on 7.00 h am). Animals were allowed to adapt for 6 d before the
first experimental manipulations. During adaptation, animals were housed in groups of 8 per
cage. Body weight (BW) was between 320 and 360 g. Food and drinking water was available ad
libitum. All of the following experiments were conducted with the approval of the Animal Care
Committee of the Leiden University Medical Center.
Hormonal treatment; Block and Replacement
At D0 of the protocol animals were placed in individual cages and treated with methimazole
0.025% (MMI, Sigma, the Netherlands) in drinking water containing 0.3% saccharin. At D7,
osmotic minipumps (OMP, Alzet 2ml2, Durect Corp., Cupertino, USA) loaded with L-thyroxine
(T4, Sigma, the Netherlands) solved in 6.5 mM NaOH and 50% propylene glycol, were implanted
under the dorsal skin during the surgical procedure. OMPs delivered either vehicle (hypothyroid
rats), or T4 at a dose of 1.5 µg (replacement dose; euthyroid group) or 16.0 µg (thyrotoxic group)
/100 g BW*day, as described previously (7).
Surgery
At D7, animals were anaesthetized using a mixture of Hypnorm (Janssen; 0.05 mL/100 g BW,
i.m.) and Dormicum (Roche, the Netherlands; 0.04 mL/100 g BW, s.c.).
In all animals an intra-atrial silicone cannula was implanted through the right jugular vein for
infusion and sampling (13). The cannula was tunnelled to the head subcutaneously, fixed with
dental cement to 4 stainless-steel screws inserted into the skull. A mixture of 60% Amoxicillin,
20% heparin and 20% saline in polyvinylpyruvidon (Sigma, the Netherlands) was used to fill the
cannula and prevent inflammation and occlusion.
Experiment #1
Energy expenditure, fat oxidation, spontaneous physical activity and food intake At D7,
animals were placed into an 8-cage combined, open circuit indirect calorimetry system (LabMaster
system, TSE Systems, Bad Homburg, Germany, for the remainder of this manuscript referred to
as “metabolic cages”), measuring food and water intake, O2 uptake and CO2 production, as
well as SPA. Although the cages including bedding were identical to the cages in which the
rats were housed the first 7 days (only the cover of the metabolic cage differs), animals were
adapted to this environment before the start of the actual measuring periods (D9 and D10)
for approximately 48 h. EE, RER and fat oxidation were calculated from the O2 uptake and
CO2 production relative to individual body weights (14). O2 uptake and CO2 production were
measured with 10 min intervals. Food and water intake and physical activity were measured
continuously. Activity monitoring and detection of animal location was performed with infrared
proefschrift Klieverik.indb 30
4-8-2009 15:22:28
Experiment #2
proefschrift Klieverik.indb 31
31
Effects of thyroid status on energy homeostasis and fatty acid uptake
Radiolabeled FA infusion At D11, rats were restrained from access to food from 5h prior to
the labeled lipid infusion (i.e. from 9.00 am onwards). Rats were connected to a metal collar
attached to polyethylene tubing (for blood-sampling and isotope infusion) which was kept out of
reach of the animals by a counterbalanced beam. This allowed all subsequent manipulations to
be performed outside the cages without handling the freely moving animals. After obtaining a
blood sample for measurement of plasma TH, FA, and TG concentrations (800 µL), rats received
a primed (500 µl in 5 min), continuous (500 µl/h) infusion of albumin-bound 14C-oleate (FA)
and VLDL-like emulsion-incorporated glycerol tri[3H]oleate (TG) i.v. for 2h. At the end of the 2-h
infusion period we obtained another blood sample (800 µL) for measurement of plasma [3H]-FA
and [14C]-FA radioactivity. Rats were sacrificed and striated muscle (M triceps brachii), heart,
liver, three white adipose tissue (WAT) depots (gonadal (epididymal), subcutaneous, visceral)
and infra-scapular brown adipose tissue (BAT) were harvested, snap frozen and stored at -20°C
for subsequent analysis. In experiment #2, data from 24 rats were analyzed. Five rats had to be
excluded from the final analysis due to jugular catheter occlusion.
Preparation of radiolabeled emulsion particles Protein-free VLDL-like TG-rich emulsion particles were prepared from 100 mg total lipid at a weight ratio of triolein (Sigma, St. Louis, MA,
US): egg yolk phosphatidylcholine (Lipoid, Ludwigshafen, Germany): lyso phosphatidylcholine
(Sigma, St. Louis, MA, US): cholesteryl oleate (Janssen, Beersse, Belgium): cholesterol (Sigma,
St. Louis, MA, US) of 70: 22.7: 2.3: 3.0: 2.0 in the presence of 800 µCi of glycerol tri[9, 10(n)3H]oleate ([3H]TG) (Amersham, Little Chalfont, UK), as reported previously (15). Lipids were
hydrated in 10 mL of 2.4M NaCl, 10 mM Hepes, 1 mM EDTA, pH 7.4, and sonicated for 30
min at 10 µm output using a Soniprep 150 (MSE Scientific Instruments, UK) equipped with a
water bath for temperature (54°C) maintenance. The emulsion was separated into fractions with
a different average size by density gradient ultracentrifugation. Intermediate (80 nm) [3H]-TG
particles were mixed with a trace amount of [14C]-oleic acid (Amersham, Little Chalfont, UK)
complexed to bovine serum albumin (BSA).
Tissue uptake analysis Tissues were dissolved in 5 mol/L KOH in 50% (vol/vol) ethanol. After
overnight saponification, protein content was determined in the various organs using BCA kit
(BCA Protein Assay Kit, Thermo Scientific). Radioactivity was measured in the saponified organs
and corrected for the corresponding protein concentration and plasma specific activities of [3H]FA and [14C]-FA. Calculations of tissue FA uptake and rate of disappearance were performed as
described previously (12).
Analysis of lipoprotein lipase (LPL) and hepatic lipase (HL) activity Striated muscle, heart,
liver and three WAT depots were cut into small pieces and put in 1 mL 2% BSA-containing DMEM
Chapter 2
sensor pairs arranged in strips for horizontal (X level) and vertical (Z level) activity, detecting every
ambulatory movement. Spontaneous physical activity (SPA or XA), high-frequent activity (XF;
equivalent of breathing activity), total activity (XT=XA+XF) and rearing (Z) were monitored. The
infrared sensors for detection of movement allowed continuous recording in both light and dark
phases. In experiment #1, data from 21 rats were analyzed. Eight rats had to be excluded from
the final analysis due to incomplete calorimetry measurements.
4-8-2009 15:22:28
medium. Heparin (2 units) was added and samples were incubated at 37°C for 60 minutes.
After centrifugation (10 min at 13.000 rpm), the supernatants were taken and snap-frozen
until analysis. Total LPL and HL activity was determined as modified from Zechner et al. (16).
In short, the lipolytic activity of tissue supernatant was assessed by determination of [3H]oleate
production upon incubation of tissue supernatant with a mix containing an excess of both [3H]
triolein, heat-inactivated human plasma as sources of the LPL coactivator apoC2 and FA-free BSA
as FFA acceptor.
32
Plasma analysis
Plasma concentrations of the thyroid hormones T3 and T4 were determined by an in-house RIA, with
inter- and intra-assay CV of 7–8% and 3–4% (T3), and 3–6 and 2–4% (T4), respectively. Detection
limits for T3 and T4 were 0.3 nmol/L and 5 nmol/L, respectively. Plasma TSH concentrations
were determined by a chemiluminescent immunoassay (Immulite 2000, Diagnostic Products
Corp., Los Angeles, CA), using a rat-specific standard (17). The inter- and intra-assay CV for
TSH were less than 4% and 2% at ±3.5 mU/L, respectively, and the detection limit was 0.2
mU/L. Blood samples were kept in chilled paraoxon-coated Eppendorf tubes to prevent ex vivo
lipolysis. The tubes were placed on ice and immediately centrifuged at 4ºC. Plasma levels of
TG and FA were determined using commercially available kits and standards according to the
manufacturers instructions (Instruchemie, Delfzijl, The Netherlands). Lipids were extracted from
plasma according to Bligh and Dyer (18). The lipid fraction was dried under nitrogen, dissolved
into chloroform/methanol (5:1 [vol/vol]) and subjected to TLC (LK5D gel 150; Whatman) using
hexane:diethylether:acetic acid (83:16:1) [vol/vol/vol]) as mobile phase. Standards for FA and
TG were included during the TLC procedure to locate spots of these lipids. Spots were scraped,
lipids dissolved in hexane and radioactivity measured.
Statistics
Both energy homeostasis and FA uptake data were analyzed by non-parametric Kruskall-Wallis
(KW) test, and a Mann Whitney U post hoc test was performed if KW revealed significance to
determine which experimental groups differed from each other. Cosinor analysis was performed
on the metabolic cage data of individual animals (48h). Curve fitting was performed using
constrained nonlinear regression analysis (SPSS 16.0). Subsequently, only if the significance level
(P value) of the fitted curve was less than 0.05, data were used to calculate mesor, amplitude
and acrophase of the individual curve. Significance was defined at p≤0.05. Data are presented
as mean ± SEM.
Results
Experiment #1: Effects of thyroid status on energy homeostasis
Body weight and eating behaviour At the time of starting hormonal (T4) treatment (osmotic
mini-pump implantation; D0), there were no differences in bodyweight (BW) between groups
(Hypo 339±13, Eu 340±5, Tox 345±4 g, p=0.63). At the time of placement in the metabolic cages
(D7), BW was decreased by 13±3 g in thyrotoxic rats, compared with an increase of 2±5 and 13±9
g in euthyroid and hypothyroid rats, respectively (Hypo vs Eu p=0.383, Eu vs Tox p=0.017, Hypo
proefschrift Klieverik.indb 32
4-8-2009 15:22:28
Chapter 2
33
Effects of thyroid status on energy homeostasis and fatty acid uptake
vs Tox p=0.053). However, during the whole period in the metabolic cages (D7-10), BW increased
to a similar extent in all groups (Hypo 12±1, Eu 14±3, Tox 12±1 g, p=0.194).
After placement in the metabolic cages animals were allowed to adapt to this new environment
for 48h (D7-8). Subsequently, we gathered energy homeostasis data for 48h (D9-10). During this
time period, thyrotoxic rats showed increased cumulative food intake by 18% as compared with
euthyroid rats. Hypothyroid rats ate less than euthyroid rats, although this did not reach statistical
significance (Hypo 44±2 g, Eu 48±2 g, Tox 57±2 g, Eu vs Tox p=0.004, Hypo vs Eu p=0.128).
Plasma thyroid hormones Plasma concentrations of T3, T4 and TSH following the 48 h
measurement of energy homeostasis in experiment #1 are given in Table 1. Plasma T3 and
T4 concentrations were 163% and 30% higher, respectively, in thyrotoxic rats as compared
with euthyroid rats. In hypothyroid rats, plasma T3 and T4 concentrations were decreased to
44% and 15%, respectively, of euthyroid levels. Plasma TSH was 12.9±2.2 mU/L in hypothyroid
rats, and showed similar values in euthyroid and thyrotoxic rats (0.3±0.1 and 0.2±0.0 mU/L,
respectively).
Total EE, physical activity and resting EE Total EE (TEE) showed a clear diurnal rhythm in
all treatment groups, with a rise in the dark (i.e. active) period (fig 1a). As expected, this was
paralleled by a similar rhythm in SPA in all groups (fig 1b). There was a marked, 37% increase
in mean TEE/kg in thyrotoxic relative to euthyroid rats (p=0.02). This increase persisted when
mean TEE was not corrected for BW (p=0.04, data not shown). Hypothyroid rats showed a
trend (p=0.064) towards decreased (-10%) mean TEE relative to euthyroid rats. Cosinor analysis
revealed similar changes in the mesor of the fitted curves. In addition, there was a decrease in
the amplitude of the rhythm in EE by 46% in hypothyroid relative to euthyroid rats (p=0.017)
as well as a ~1 h phase-advance of the acrophase relative to Eu and Tox rats (p=0.017). There
were no differences in the mean levels of SPA between groups (Kruskall-Wallis p=0.901). Also
mesor, amplitude and acrophase of the fitted curves of SPA exhibited no differences between
Tox, Hypo and Eu groups (Table 2). Likewise, there were no differences in high-frequency activity
(equivalent of breathing; XF), rearing (Z) or total activity (XT) between groups, nor in mesor,
amplitude or acrophase of the fitted curves (data not shown).
In each animal we determined the total number of 10 min intervals in which the activity sensors
did not detect any activity (activity units (AU) =0). Hypothyroid rats tended to spend more time
inactive than euthyroid rats (533±72 vs 359±40 min, p=0.064), whereas thyrotoxic rats showed
a trend towards less inactivity time compared to euthyroid rats (256±33 min, p=0.073). During
these periods of inactivity, mean EE, termed resting EE (REE), was markedly higher in thyrotoxic
Table 1: Experiment #1: plasma thyroid hormone concentrations after 48h measurement of energy homeostasis
(day 11) in hypothyroid (Hypo), euthyroid (Eu) and thyrotoxic (Tox) rats.
Hypo n = 7
Eu n = 7
Tox n = 7
T3 (nmol/L)
0.50 ± 0.11 *
1.14 ± 0.07
3.00± 0.27 *
T4 (nmol/L)
20 ± 2 *
TSH (mU/L)
12.9 ± 2.2 *
136 ± 8
177± 8 *
0.3 ± 0.1
0.2± 0.0
* p≤0.01 vs Eu
proefschrift Klieverik.indb 33
4-8-2009 15:22:28
34
Fig 1a Forty-eight-hour total energy expenditure (TEE) in hypothyroid (Hypo), euthyroid (Eu) and thyrotoxic
(Tox) rats entrained to a regular 12/12 L/D cycle. Horizontal black bars indicate the dark phase of the L/D
cycle. Data are mean of 7 animals per group at each time point and the interval between time points was 10
min. Cosinor data and statistical analysis are given in table 2. b Forty-eight-hour spontaneous physical activity
(SPA) in hypothyroid (Hypo), euthyroid (Eu) and thyrotoxic (Tox) rats entrained to a regular 12/12 L/D cycle.
Horizontal black bars indicate the dark phase of the L/D cycle. Data are mean of 7 animals per group at each
time point and the interval between time points was 10 min. Cosinor data and statistical analysis are given in
table 2.c Resting energy expenditure (REE, defined as the mean energy expenditure during 10 min intervals of
inactivity (see text) in each individual animal) in hypothyroid (Hypo), euthyroid (Eu) and thyrotoxic (Tox) rats.
Note the increase of REE in Tox vs Eu rats (**p<0.01) that was more pronounced than the decrease of REE in
Hypo vs Eu rats (*p<0.05). Data are mean ± SEM of 7 animals per group.
proefschrift Klieverik.indb 34
4-8-2009 15:22:29
Energy Expenditure (kcal/h*kg)
Mesor
Amplitude
Acrophase (h)
n
6.52 ± 0.26 “ ^
0.64 ± 0.15 * ^
23.15 ± 0.16 * ^
7
Eu
7.24 ± 0.21
1.19 ± 0.09
00.23 ± 0.18
7
Tox
10.46 ± 0.67 *
1.70 ± 0.32
00.39 ± 0.22
7
p=0.001
p=0.009
Hypo
p = 0.015
Spontaneous Physical Activity (AU)
Mesor
Amplitude
Hypo
63 ± 6
Eu
62 ± 5
Tox
Acrophase (h)
n
37 ± 8
00.26 ± 0.15
7
38 ± 5
01.09 ± 0.35
7
60 ± 7
33 ± 6
00.13 ± 2.25
7
p = 0.901
p = 0.780
p = 0.248
35
Effects of thyroid status on energy homeostasis and fatty acid uptake
Table 2: Data derived from cosinor curve-fit of all individual animals with statistical analysis.
Chapter 2
as compared with euthyroid rats (p=0.002), and lower in hypothyroid rats (p=0.025 vs Eu, fig
1c). REE/TEE ratios showed no differences between groups (Hypo 0.83±0.02, Eu 0.85±0.02, Tox
0.82±0.01, KW p=0.657).
RER and substrate oxidation Euthyroid rats showed a diurnal rhythm in RER (fig 2a), although
less evident than the rhythm in TEE and SPA. The nocturnal acrophase fits with a relative increase
in glucose oxidation in the dark (i.e. feeding) period. Both Tox and -although to a lesser extentHypo rats showed a decrease in mean RER levels relative to euthyroid rats (Tox: 94% of Eu
levels, p=0.006, Hypo: 96% of Eu levels, p=0.041). In addition, cosinor analysis revealed that the
amplitude of the day-night rhythm in RER was markedly increased in Tox rats (155%, p=0.007).
Although the RER phase difference between the Eu and Tox groups did not reach statistical
Respiratory Exchange Ratio
Acrophase (h)
n
Hypo
0.94 ± 0.01 *
0.0123 ± 0.0014 ^
20.44 ± 0.30 * ^
7
Eu
0.97 ± 0.01
0.0120 ± 0.0010
22.27 ± 0.24
7
Tox
0.91 ± 0.01 *
0.0306 ± 0.0110 *
03.35 ± 2.48
7
Mesor
p = 0.014
Amplitude
p = 0.032
p = 0.018
* p≤0.05 vs Eu, “ p=0.073 vs Eu, ^p<0.05 vs Tox
significance (p=0.128), there appeared to be an inverse rhythm in Tox relative to hypothyroid
rats (p=0.017). Hypothyroid animals showed a RER increase in the early part of the dark period,
whereas Tox rats showed a pronounced trough in RER during the feeding periods at the beginning
and end of the dark period. Cosinor analysis confirmed that hypothyroid rats exhibit a decrease
(p=0.053) in the mesor of their RER day-night rhythm relative to Eu rats, but less pronounced
than in Tox animals (p=0.008). The acrophase of the RER rhythm in hypothyroid animals was ~2-h
phase-advanced relative to Eu (p=0.037), and almost 7-h relative to Tox animals (p=0.017).
proefschrift Klieverik.indb 35
4-8-2009 15:22:29
36
Fig 2a Forty-eight-hour respiratory exchange ratio (RER) in hypothyroid (Hypo), euthyroid (Eu) and thyrotoxic
(Tox) rats entrained to a regular 12/12 L/D cycle. Horizontal black bars indicate the dark phase of the L/D cycle.
Data are mean of 7 animals per group at each time point, and the interval between time points was 10 min.
Cosinor data and statistical analysis are given in table 2. b Mean 48 h glucose oxidation in hypothyroid (Hypo),
euthyroid (Eu) and thyrotoxic (Tox) rats. Note that Hypo rats exhibit decreased glucose oxidation as compared
with Eu animals (*p<0.05). c Mean 48 h fat oxidation in hypothyroid (Hypo), euthyroid (Eu) and thyrotoxic (Tox)
rats. Note that Tox rats exhibit increased fat oxidation as compared with Eu animals (**p<0.01).
Substrate oxidation is depicted in figs 2b and 2c. In euthyroid animals, mean levels of glucose
oxidation were ~30-fold higher than fat oxidation, in line with ad libitum access to carbohydraterich chow. Tox animals showed no difference in glucose oxidation relative to euthyroid rats,
proefschrift Klieverik.indb 36
4-8-2009 15:22:29
Experiment #2: Effects of thyroid status on lipid turnover, uptake and partitioning
Table 3: Experiment #2: plasma thyroid hormone concentrations before radio-labeled FA infusion (day 11) in
hypothyroid (Hypo), euthyroid (Eu) and thyrotoxic (Tox) rats.
Hypo n = 9
Eu n = 8
Tox n = 7
T3 (nmol/L)
0.41 ± 0.08 **
1.21 ± 0.09
2.78 ± 0.27 **
T4 (nmol/L)
19 ± 2 **
TSH (mU/L)
10.9 ± 1.8 **
139 ± 7
187 ± 5 **
0.2 ± 0.0
0.2 ± 0.0
37
Effects of thyroid status on energy homeostasis and fatty acid uptake
In order to determine how whole body alterations in fat oxidation induced by thyrotoxicosis
and hypothyroidism translated into substrate (i.e. FA) uptake at the tissue level, we applied a
dual FA-isotope infusion technique that permits differentiation between plasma TG-derived and
plasma albumin-bound FA uptake.
Plasma thyroid hormones Plasma T3, T4 and TSH concentrations (table 3) showed differences
between groups very similar to the T3, T4 and TSH differences in experiment #1.
FA and TG plasma concentrations and rate of disappearance Plasma FA concentrations
were 118% higher in thyrotoxic relative to euthyroid rats (p=0.004). Plasma TG concentrations
tended to increase in thyrotoxic (p=0.059) rats, and showed a significant decrease in hypothyroid
(p=0.046) compared with euthyroid rats (Table 4).
Rate of disappearance (Rd) of 14C-FA was 59% increased in thyrotoxic relative to euthyroid rats
(p=0.054). There were no differences in Rd of 3H-TG between groups (Table 4).
Tissue-specific TG-derived FA uptake, lipoprotein lipase (LPL), hepatic lipase (HL) activity
and albumin-bound FA uptake Thyrotoxicosis induced an increase of TG-derived FA uptake
Chapter 2
whereas Hypo rats showed a mild decrease in mean level of glucose oxidation relative to Eu
(19%, p=0.035) and Tox animals (23%, p=0.026, fig 2b). Mean levels of fat oxidation were
markedly (479%) increased in Tox relative to Eu rats (p=0.006), but there was no difference in
fat oxidation between Hypo and Eu rats (p=0.110, fig 2c). Thus, RER showed a decrease in both
Tox and -to a lesser extent- Hypo animals relative to Eu rats. However, the mechanism of this
decrease was different between groups, i.e. a decrease of glucose oxidation in Hypo animals,
and a pronounced increase in fat oxidation in Tox animals.
**p<0.0001 vs Eu
in striated muscle (58%, p=0.040, fig 3a), and tended to increase TG-derived FA uptake in
heart (78%, p=0.059) relative to euthyroid rats, but did not induce alterations in muscle or
heart LPL activity (fig 3b). Thyrotoxicosis induced a pronounced decrease in TG-derived FA
uptake to 21% of euthyroid levels in BAT (p<0.0001). It should be noted that FA uptake was
approximately 30-fold higher in BAT as compared to oxidative striated muscle, in line with the
high mitochondrial density and high FA oxidative capacity of brown adipocytes. Thyrotoxicosis
had no effect on TG-derived FA uptake in any of the WAT depots, although it induced a modest
decrease in LPL activity in gonadal WAT (48%, p=0.043). In contrast, hypothyroid rats showed a
pronounced increase of TG-derived FA uptake in gonadal and visceral WAT (184%, p=0.002 and
75%, p=0.036, respectively, fig 3a), associated with an increase in LPL activity both in gonadal
proefschrift Klieverik.indb 37
4-8-2009 15:22:29
Table 4: Experiment #2: plasma fatty acids (FA) and triglycerides (TG) before radio-labeled FA infusion
(day 11) and Rate of disappearance (Rd) of labeled FA and TG in hypothyroid (Hypo), euthyroid (Eu) and
thyrotoxic (Tox) rats.
Hypo n = 9
Eu n=8
Tox n=7
Plasma concentration
38
FA (mmol/L)
0.50 ± 0.10
0.44 ± 0.03
0.96 ± 0.33 *
TG (mmol/L)
0.45 ± 0.08 *
0.71 ± 0.07
1.30 ± 0.31 ^ “
6.95 ± 1,75
6,64 ± 1,69
10,56 ± 1,19 *
6,89 ± 2,67
7,05 ± 2,63
6,48 ± 0,76
Rd (μmol/kg*min)
[14C]-FA
[3H]-TG
* p≤0.05, ^ p=0.059 vs Eu, “ p<0.05 vs Hypo
and visceral WAT (234%, p=0.001 and 306%, p=0.012, respectively, fig 3b). In liver, hypothyroid
rats showed decreased TG-derived FA uptake relative to euthyroid rats (37%, p=0.046) but no
change in HL activity. Conversely, thyrotoxic rats showed no effect on TG-derived FA uptake but
an increase in HL activity (52%, p=0.009).
Thyrotoxicosis induced an increase in albumin-bound FA uptake in striated muscle (71%,
p=0.054, fig 3c), a 97% increase of albumin-bound FA uptake in gonadal WAT (p=0.043), and it
similarly tended to increase albumin-bound FA uptake in subcutaneous and visceral WAT (71%,
p=0.059 and 129%, p=0.059, respectively). There was no difference in albumin-bound FA uptake
between hypothyroid and euthyroid rats in any of the tissues studied.
Discussion
We studied the changes in whole body energy metabolism associated with thyroid hormone
status and we delineated how these changes translate into substrate (i.e. FA) uptake at the
tissue level. Our main findings are that thyrotoxicosis induces a hypermetabolic phenotype
(increased TEE, REE, and fat oxidation) as well as increased food intake favouring substrate
replenishment. Interestingly, thyrotoxicosis did not increase SPA, indicating that changes in SPA
do not contribute to increased TEE. Moreover, hypermetabolism was associated with increased
TG-derived FA uptake in most oxidative tissues, whereas TG-derived FA uptake was unaltered in
WAT. Conversely, hypothyroidism induced a hypometabolic phenotype with a mild decrease in
REE, a trend towards decreased TEE, and a decrease of glucose oxidation. In addition, TG-derived
FA uptake was increased in lipid storing WAT, concomitantly with increased LPL activity. However,
during hypothyroidism TG-derived FA uptake in oxidative tissues was unaltered. These alterations
in TG-derived FA uptake during thyrotoxicosis and hypothyroidism indicate that FA uptake from
TG-rich lipoproteins is differentially regulated by thyroid hormones in a tissue-specific manner.
The mechanism of the increase of TEE induced by thyrotoxicosis is incompletely understood. It
has been known for many years that REE is highly responsive to thyroid hormones (1). In addition,
many thyrotoxic patients show a characteristic resting tremor and self-reported increased physical
activity. Conversely, many hypothyroid patients complain of slowness (19). However, there is
little experimental evidence indicating that thyroid status modulates locomotor behaviour or
proefschrift Klieverik.indb 38
4-8-2009 15:22:30
Chapter 2
39
Effects of thyroid status on energy homeostasis and fatty acid uptake
a Triglyceride (TG)-derived fatty acid (FA) uptake in striated muscle, heart, liver, three white adipose tissue
(WAT) depots (gonadal, subcutaneous, visceral) and infra-scapular brown adipose tissue (BAT) of hypothyroid
(Hypo), euthyroid (Eu) and thyrotoxic (Tox) rats. b Lipoprotein lipase activity in striated muscle, heart, and
three white adipose tissue (WAT) depots (gonadal, subcutaneous, visceral) and hepatic lipase activity in liver
of hypothyroid (Hypo), euthyroid (Eu) and thyrotoxic (Tox) rats. c Albumin-bound FA uptake in striated muscle,
heart, liver, three WAT depots (gonadal, subcutaneous, visceral) and BAT of hypothyroid (Hypo), euthyroid (Eu)
and thyrotoxic (Tox) rats.. 0.05<p<0.10, *p≤0.05, **p<0.01 vs Eu.
proefschrift Klieverik.indb 39
4-8-2009 15:22:30
40
SPA, and it is unclear how this relates to the alterations in energy homeostasis induced by
hypothyroidism and thyrotoxicosis. This is of particular interest, since accumulating evidence
suggests that EE associated with SPA, termed non-exercise activity thermogenesis (NEAT), is an
independent (negative) determinant of (development of) obesity in humans and rodents. As
thyroid hormone is a principal regulator of energy metabolism, it may also be involved in the
regulation of NEAT. The present study shows that although moderate hyperthyroidism increases
TEE by 37%, it induces no alterations in SPA. In contrast, resting energy expenditure (REE),
defined as the energy expended during time periods when no activity was detected, is increased
by 61% in thyrotoxic rats. Moreover, hypothyroidism induces a significant 12% decrease of REE,
but it does not affect SPA either, and REE/TEE ratios are unaffected by both hypothyroidism and
thyrotoxicosis. Together, our data strongly suggest that the increased energy requirements of
SPA are not determined by thyroid hormone status and do not explain increased TEE associated
with thyrotoxicosis.
Our data are in contrast with those of Levine et al. (5) who reported increased SPA during
thyrotoxicosis in rats, suggesting that NEAT was a significant component of the increase in TEE.
This discrepancy is most likely explained by the pharmacological dose of T3 used by Levine et al.
to induce thyrotoxicosis, resulting in a ~13-fold increase in plasma T3. In the present study serum
T3 was increased only 2.5-fold, in keeping with the range of plasma T3 often found in patients
with thyrotoxicosis. Interestingly, this was paralleled by a relatively mild, 30% increase in plasma
T4 concentrations. In thyrotoxic patients, a relative overproduction of T3 giving rise to increased
plasma T3/T4 ratios may be observed (20). Deiodinase type 1 (D1), which is mainly expressed
in liver and kidney, is positively regulated by T3 (21). Therefore, D1-mediated T3 production is
thought to be a major source of extra-thyroidal T3 during hyperthyroidism (22). Indeed, the
increased T3/T4 ratio in our rat model of thyrotoxicosis is paralleled by an induction of hepatic D1
expression (7), which may underlie the relatively mild increase of plasma T4 relative to T3.
Our experimental approach does not allow for measurement of other components of TEE, such
as diet-induced thermogenesis (absorption, digestion and metabolism of food) and facultative
thermogenesis (energy expended to maintain body temperature during cold exposure in
homeothermic species). However, it is reasonable to assume that the 18% increase in 48 h
cumulative food intake led to an increase of diet induced thermogenesis in thyrotoxic rats,
although this component generally comprises only a minor part (~10-15%) of TEE. Facultative
thermogenesis is unlikely to have played a role in our study, as it is generally negligible under
thermo-neutral circumstances.
TH is known to play a role in regulating seasonal adaptations in several species, for example
reproduction and maintenance of body weight (23;24), but its possible involvement in modulating
rhythms of shorter phase, i.e. circadian rhythms, has received less attention. This possibility is
theoretically supported by thyroid hormone receptor α1 mRNA expression in the region of
the main circadian oscillator, i.e. the suprachiasmatic nuclei (SCN) (25), although this has not
been confirmed at the protein-level (26;27). Previous studies have shown lack of an effect of
hypothyroidism on rhythms of locomotor (wheel running) activity (28). Therefore, the subtle
effects of hypothyroidism on the acrophase of the daily TEE and RER rhythms we did observe,
are most likely occurring downstream of the SCN. Euthyroid rats exhibited a through in RER
proefschrift Klieverik.indb 40
4-8-2009 15:22:30
41
Effects of thyroid status on energy homeostasis and fatty acid uptake
proefschrift Klieverik.indb 41
Chapter 2
during the light period, fitting with relatively high fat oxidation during the inactive period when
little food is consumed. The shift in acrophase of RER in thyrotoxic rats appears to be mainly due
to increased fat oxidation during the nightly feeding periods (data not shown), suggesting that
during thyrotoxicosis, high energy demands require mobilization of energy stores on top of the
nutrients supplied by increased food intake during the active period.
Lipoprotein lipase (LPL) is the key enzyme regulating tissue-specific FA disposal by hydrolyzing
triglycerides (TG) in circulating TG-rich lipoprotein particles. LPL has been proposed as a metabolic
“gatekeeper” (29), directing substrate to tissues dependent on the body’s metabolic status
(30;31). In the present study, we explored TH effects on tissue FA uptake, and we were able to
differentiate between TH effects on TG-derived (i.e. LPL-dependent) and albumin-bound (i.e. LPLindependent) FA uptake on the tissue level. In addition, we measured tissue-specific LPL activity.
In keeping with the observed hypermetabolic state associated with thyrotoxicosis, we found that
thyrotoxicosis increases TG-derived FA uptake in major oxidative tissues such as striated muscle
and the heart, without affecting TG-derived FA uptake in lipid-storing WAT depots. This increase
in TG-derived FA uptake in oxidative tissues was not paralleled by increased local LPL activity. It
has been previously reported that the linear relationship between muscle TG-derived FA uptake
and LPL activity in euthyroid animals is lost after experimental alterations in thyroid status (32).
This may be explained by TH effects on additional determinants of the process of TG-derived
FA-uptake. In addition, TH induced stimulation of local blood flow (33;34) may have interfered
with FA-uptake, independently of LPL activity.
Conversely, hypothyroidism increased TG-derived FA uptake in WAT. Indeed, earlier studies in
rats have also reported increased LPL activity in WAT during hypothyroidism (35) that could
be reversed by tri-iodothyronine (T3) administration (36;37). However, hypothyroidism had
no effect on TG-derived FA uptake in oxidative tissues. In the liver, hypothyroidism decreased
TG-derived FA uptake but not hepatic lipase (HL) activity, whereas thyrotoxicosis increased HL
activity, but not TG-derived FA uptake. Taken together, the present evidence suggests that
during thyrotoxicosis, hypermetabolism and increased FA oxidation are facilitated by preferential
shuttling of TG-derived FA’s to oxidative tissues. Conversely, during hypothyroidism TG-derived
FA’s are shuttled to lipid storing WAT, away from the liver and oxidative tissues, via increased
tissue-specific LPL activity.
Circulating FAs that are not incorporated in TG-rich lipoprotein particles are bound to albumin
in plasma. We found that thyrotoxicosis increases albumin-bound FA uptake in muscle as well
as in the WAT depots, whereas hypothyroidism had no effect on albumin-bound FA uptake
in any of the tissues studied. Plasma FA concentration were increased in thyrotoxic, but not
in hypothyroid relative to euthyroid animals. This is in line with the notion that tissue uptake
of albumin-bound FA is mainly driven by the concentration gradient between the capillary
lumen and the intracellular space (30). Interestingly, in contrast to the increase in FA-uptake in
oxidative tissues like striated muscle and heart, thyrotoxicosis induced a pronounced decrease of
TG-derived FA uptake in BAT. BAT is the main site for adaptive thermogenesis in rodents. During
cold exposure, sympathetic stimulation of BAT induces local conversion of T4 to T3, thereby
generating heat via induction of mitochondrial uncoupling (38). Simultaneously, LPL is markedly
induced via a β-adrenergic mechanism, enabling replenishment of the FA used for mitochondrial
4-8-2009 15:22:30
42
oxidation (39). During thyrotoxicosis, increased thermogenesis has been proposed to evoke a
compensatory decrease in sympathetic tone to BAT (40;41). We now speculate that such a
decrease in sympathetic tone may explain the marked decrease in TG-derived FA uptake in the
present study, possibly via decreased LPL activity.
In conclusion, our data indicate that FA uptake from TG-rich lipoproteins is regulated by TH in
a tissue-specific manner. Thyrotoxicosis increases TG-derived FA uptake in all oxidative tissues
except BAT, whereas hypothyroidism increases TG-derived FA uptake in lipid storing WAT via
increased LPL activity, and decreases uptake in liver. In contrast, albumin bound FA uptake
during hypothyroidism and thyrotoxicosis appears to be merely mass, i.e. concentration gradient,
driven.
Acknowledgements
The Ludgardine Bouwman-foundation and T.I. Pharma are kindly acknowledged for financial
support. We thank E. Johannesma-Brian and M.J. Geerlings for analytical support, and S.A.A. van
den Berg for excellent technical assistance.
Reference List
1. Baron DN 1959 Estimation of the basal metabolic rate in the diagnosis of thyroid disease. Proc R Soc
Med 52:523-525
2. Luddecke HF 1958 Basal metabolic rate, protein-bound iodine and radioactive iodine uptake: a
comparative study. Ann Intern Med 49:305-309
3. Wiersinga WM, Chopra IJ 1982 Radioimmunoassay of thyroxine (T4), 3,5,3’-triiodothyronine (T3),
3,3’,5’-triiodothyronine (reverse T3, rT3), and 3,3’-diiodothyronine (T2). Methods Enzymol 84:272-303
4. Kim B 2008 Thyroid hormone as a determinant of energy expenditure and the basal metabolic rate.
Thyroid 18:141-144
5. Levine JA, Nygren J, Short KR, Nair KS 2003 Effect of hyperthyroidism on spontaneous physical activity
and energy expenditure in rats. J Appl Physiol 94:165-170
6. Jacobsen R, Lundsgaard C, Lorenzen J, Toubro S, Perrild H, Krog-Mikkelsen I, Astrup A 2006 Subnormal
energy expenditure: a putative causal factor in the weight gain induced by treatment of hyperthyroidism.
Diabetes Obes Metab 8:220-227
7. Klieverik LP, Sauerwein HP, Ackermans MT, Boelen A, Kalsbeek A, Fliers E 2008 Effects of thyrotoxicosis
and selective hepatic autonomic denervation on hepatic glucose metabolism in rats. Am J Physiol
Endocrinol Metab 294:E513-E520
8. Moller N, Nielsen S, Nyholm B, Porksen N, Alberti KG, Weeke J 1996 Glucose turnover, fuel oxidation
and forearm substrate exchange in patients with thyrotoxicosis before and after medical treatment. Clin
Endocrinol (Oxf) 44:453-459
9. Randin JP, Scazziga B, Jequier E, Felber JP 1985 Study of glucose and lipid metabolism by continuous
indirect calorimetry in Graves’ disease: effect of an oral glucose load. J Clin Endocrinol Metab 61:11651171
10. Oppenheimer JH, Schwartz HL, Lane JT, Thompson MP 1991 Functional relationship of thyroid hormoneinduced lipogenesis, lipolysis, and thermogenesis in the rat. J Clin Invest 87:125-132
11. Saffari B, Ong JM, Kern PA 1992 Regulation of adipose tissue lipoprotein lipase gene expression by
thyroid hormone in rats. J Lipid Res 33:241-249
proefschrift Klieverik.indb 42
4-8-2009 15:22:30
12. Teusink B, Voshol PJ, Dahlmans VE, Rensen PC, Pijl H, Romijn JA, Havekes LM 2003 Contribution of fatty
acids released from lipolysis of plasma triglycerides to total plasma fatty acid flux and tissue-specific fatty
acid uptake. Diabetes 52:614-620
14. McLean JA, Tobin G 1987 In: Animal and Human Calorimetry. Cambridge University Press; 100-112.
15. Rensen PC, Herijgers N, Netscher MH, Meskers SC, van Eck M, van Berkel TJ 1997 Particle size determines
the specificity of apolipoprotein E-containing triglyceride-rich emulsions for the LDL receptor versus
hepatic remnant receptor in vivo. J Lipid Res 38:1070-1084
17. Kalsbeek A, Fliers E, Franke AN, Wortel J, Buijs RM 2000 Functional connections between the
suprachiasmatic nucleus and the thyroid gland as revealed by lesioning and viral tracing techniques in
the rat. Endocrinology 141:3832-3841
18. Bligh EG, Dyer WJ 1959 A rapid method of total lipid extraction and purification. Can J Biochem Physiol
37:911-917
19. Braverman LE, Utiger RD. Introduction to hypothyroidism. The Thyroid, A Fundamental and Clinical Text
[9th edition], 679-700. 1-1-2009. Lippincott Williams & Wilkins.
20. Abuid J, Larsen PR 1974 Triiodothyronine and thyroxine in hyperthyroidism. Comparison of the acute
changes during therapy with antithyroid agents. J Clin Invest 54:201-208
21. Zavacki AM, Ying H, Christoffolete MA, Aerts G, So E, Harney JW, Cheng SY, Larsen PR, Bianco AC
2005 Type 1 iodothyronine deiodinase is a sensitive marker of peripheral thyroid status in the mouse.
Endocrinology 146:1568-1575
22. Bianco AC, Kim BW 2006 Deiodinases: implications of the local control of thyroid hormone action. J Clin
Invest 116:2571-2579
23. Barrett P, Ebling FJ, Schuhler S, Wilson D, Ross AW, Warner A, Jethwa P, Boelen A, Visser TJ, Ozanne DM,
Archer ZA, Mercer JG, Morgan PJ 2007 Hypothalamic thyroid hormone catabolism acts as a gatekeeper
for the seasonal control of body weight and reproduction. Endocrinology 148:3608-3617
43
Effects of thyroid status on energy homeostasis and fatty acid uptake
16. Zechner R 1990 Rapid and simple isolation procedure for lipoprotein lipase from human milk. Biochim
Biophys Acta 1044:20-25
Chapter 2
13. Steffens AB. 1955 A method for frequent sampling blood and continuous infusion of fluids in the rat
without disturbing the animal. Physiol Behav 4, 833-836.
24. Yoshimura T, Yasuo S, Watanabe M, Iigo M, Yamamura T, Hirunagi K, Ebihara S 2003 Light-induced
hormone conversion of T4 to T3 regulates photoperiodic response of gonads in birds. Nature 426:178181
25. Bradley DJ, Young WS, III, Weinberger C 1989 Differential expression of alpha and beta thyroid hormone
receptor genes in rat brain and pituitary. Proc Natl Acad Sci U S A 86:7250-7254
26. Alkemade A, Vuijst CL, Unmehopa UA, Bakker O, Vennstrom B, Wiersinga WM, Swaab DF, Fliers E
2005 Thyroid hormone receptor expression in the human hypothalamus and anterior pituitary. J Clin
Endocrinol Metab 90:904-912
27. Lechan RM, Qi Y, Jackson IM, Mahdavi V 1994 Identification of thyroid hormone receptor isoforms
in thyrotropin-releasing hormone neurons of the hypothalamic paraventricular nucleus. Endocrinology
135:92-100
28. Morin LP 1988 Propylthiouracil, but not other antithyroid treatments, lengthens hamster circadian
period. Am J Physiol 255:R1-R5
29. Greenwood MR 1985 The relationship of enzyme activity to feeding behavior in rats: lipoprotein lipase
as the metabolic gatekeeper. Int J Obes 9 Suppl 1:67-70
30. Frayn KN, Arner P, Yki-Jarvinen H 2006 Fatty acid metabolism in adipose tissue, muscle and liver in health
and disease. Essays Biochem 42:89-103
proefschrift Klieverik.indb 43
4-8-2009 15:22:30
31. Goldberg IG, Eckel RH, Abumrad NA 2008 Regulation of fatty acid uptake into tissues: lipoprotein lipaseand CD36-mediated pathways. J Lipid Res
32. Kaciuba-Uscilko H, Dudley GA, Terjung RL 1980 Influence of thyroid status on skeletal muscle LPL activity
and TG uptake. Am J Physiol 238:E518-E523
33. Dimitriadis G, Mitrou P, Lambadiari V, Boutati E, Maratou E, Koukkou E, Panagiotakos D, Tountas N,
Economopoulos T, Raptis SA 2008 Insulin-stimulated rates of glucose uptake in muscle in hyperthyroidism:
the importance of blood flow. J Clin Endocrinol Metab 93:2413-2415
44
34. McAllister RM, Sansone JC, Jr., Laughlin MH 1995 Effects of hyperthyroidism on muscle blood flow
during exercise in rats. Am J Physiol 268:H330-H335
35. Gavin LA, McMahon F, Moeller M 1985 Modulation of adipose lipoprotein lipase by thyroid hormone
and diabetes. The significance of the low T3 state. Diabetes 34:1266-1271
36. Saffari B, Ong JM, Kern PA 1992 Regulation of adipose tissue lipoprotein lipase gene expression by
thyroid hormone in rats. J Lipid Res 33:241-249
37. Gavin LA, Cavalieri RR, Moeller M, McMahon FA, Castle JN, Gulli R 1987 Brain lipoprotein lipase is
responsive to nutritional and hormonal modulation. Metabolism 36:919-924
38. Silva JE 2006 Thermogenic mechanisms and their hormonal regulation. Physiol Rev 86:435-464
39. Carneheim C, Nedergaard J, Cannon B 1984 Beta-adrenergic stimulation of lipoprotein lipase in rat
brown adipose tissue during acclimation to cold. Am J Physiol 246:E327-E333
40. Silva JE 2003 The thermogenic effect of thyroid hormone and its clinical implications. Ann Intern Med
139:205-213
41. Silva JE. 2005 Thermogenesis and the sympathoadrenal system in thyrotoxicosis. In: The Thyroid, A
Fundamental and Clinical Text [9th edition], 607-620. Lippincott Wiliiams &Wilkins.
proefschrift Klieverik.indb 44
4-8-2009 15:22:30
proefschrift Klieverik.indb 45
4-8-2009 15:22:30
proefschrift Klieverik.indb 46
4-8-2009 15:22:33
3
Effects of thyrotoxicosis and
selective hepatic autonomic
denervation on hepatic
glucose metabolism in rats
Lars P. Klieverik
Hans P. Sauerwein
Mariëtte T. Ackermans
Anita Boelen
Andries Kalsbeek
Eric Fliers
American Journal of Physiology: Endocrinology and
Metabolism 2008:294, E513-520
proefschrift Klieverik.indb 47
4-8-2009 15:22:37
Abstract
48
Thyrotoxicosis is known to induce a broad range of changes in carbohydrate metabolism. Recent
studies have identified the sympathetic and parasympathetic nervous system as major regulators
of hepatic glucose metabolism.
The present study aimed to investigate the pathogenesis of altered endogenous glucose
production (EGP) in rats with mild thyrotoxicosis. Rats were treated with methimazole in drinking
water and L-thyroxine (T4) from osmotic minipumps to either reinstate euthyroidism or induce
thyrotoxicosis. Euthyroid and thyrotoxic rats underwent either a sham operation, or a selective
hepatic sympathetic (Sx) or parasympathetic denervation (Px). After 10 days of T4 administration,
all animals were submitted to a hyperinsulinemic euglycemic clamp combined with stable isotope
dilution, to measure EGP.
Plasma tri-iodothyronine (T3) showed a fourfold increase in thyrotoxic as compared with euthyroid
animals. EGP was increased by 45% in thyrotoxic as compared with euthyroid rats and correlated
significantly with plasma T3. In thyrotoxic rats, hepatic PEPCK mRNA expression was increased
3,5-fold. Relative suppression of EGP during hyperinsulinemia was 34% less in thyrotoxic than
in euthyroid rats, indicating hepatic insulin resistance. During thyrotoxicosis, Sx attenuated the
increase in EGP, while Px resulted in increased plasma insulin with unaltered EGP as compared
with intact animals, compatible with a further decrease in hepatic insulin sensitivity.
We conclude that chronic, mild thyrotoxicosis in rats increases EGP, while it decreases hepatic
insulin sensitivity. Sympathetic hepatic innervation contributes only to a limited extent to increased
EGP during thyrotoxicosis, while parasympathetic hepatic innervation may function to restrain
EGP in this condition.
proefschrift Klieverik.indb 48
4-8-2009 15:22:37
Introduction
49
Effects of thyrotoxicosis and selective hepatic autonomic denervation on hepatic glucose metabolism in rats
Materials and Methods
Chapter 3
Thyrotoxicosis is associated with a broad range of alterations in metabolism and energy
homeostasis. Endogenous glucose production (EGP), lipolysis and proteolysis are increased
during thyrotoxicosis, providing the substrates needed for the concomitant increase in energy
expenditure (1). Thyrotoxic patients exhibit increased EGP and hepatic insulin resistance, i.e.,
hampered suppressive action of insulin on EGP (1;2). It is widely assumed that the metabolic
alterations during thyrotoxicosis represent direct effects of thyroid hormone (TH) on the
expression of TH-responsive genes, mediated by binding of tri-iodothyronine (T3) to thyroid
hormone receptors in peripheral organs (3). Indeed, one of the main target organs for metabolic
effects of TH is the liver, which has a key role in maintaining glucose homeostasis.
Until quite recently, the role of the central nervous system (CNS) as a focus of TH action was
thought to be largely confined to development. However, evidence is accumulating that TH has
many functions in the adult CNS as well, explaining for example the neuro-cognitive symptoms
of hypothyroidism and thyrotoxicosis (4-6). Recent studies have identified the autonomic
nervous system (ANS), controlled by hypothalamic centers, as a key player in the regulation
of glucose metabolism (7). We have previously demonstrated polysynaptic sympathetic and
parasympathetic pathways between the hypothalamic paraventricular nucleus (PVN), known to
control the autonomic efferent nerves, and the liver (8). Furthermore, we have shown that the
hypothalamus can stimulate EGP via sympathetic input to the liver (9). The functional relevance
of the parasympathetic input for hepatic glucose metabolism is evident from central effects
of insulin and fatty acids on EGP that could be completely abolished by transection of the
hepatic branch of the vagal nerve (10;11). Collectively, these observations indicate an important
physiological role for both sympathetic and parasympathetic efferent branches of the ANS in
regulating EGP.
It is unknown at present if indirect effects of TH via the efferent branches of the ANS contribute
to the changes in metabolism during thyrotoxicosis. We hypothesized that part of the changes
in glucose metabolism during thyrotoxicosis are mediated via sympathetic or parasympathetic
input to the liver. To test this hypothesis, we examined the effects of chronic, mild thyrotoxicosis
on EGP, as well as hepatic insulin sensitivity, using euglycemic hyperinsulinemic clamps and
stable isotope dilution in rats. We combined this with selective microsurgical sympathetic and
parasympathetic hepatic denervations to study the role of these efferent ANS branches in the
pathogenesis of thyrotoxicosis–induced changes in hepatic glucose metabolism.
Animals
Male Wistar rats (Harlan, Horst, the Netherlands), housed under constant conditions of
temperature (21 ± 1 °C) and humidity (60 ± 2%) with a 12-h light, 12-h dark (L/D) schedule
(lights on at 7.00 h am), were used for all experiments. Animals were allowed to adapt to
the new environment for at least 6 days before the first experimental manipulations. During
adaptation, animals were housed in groups of 4 per cage. Bodyweight (BW) was between 325
and 375 g. Food and drinking water were available ad libitum. All of the following experiments
proefschrift Klieverik.indb 49
4-8-2009 15:22:37
were conducted with the approval of the Animal Care Committee of the Royal Netherlands
Academy of Art and Sciences.
Hormonal treatment; Block and Replacement
50
At day 0 of the protocol animals were placed in individual cages (25 x 25 x 35 cm) and treated
with methimazole 0.025% (MMI, Sigma, the Netherlands) in drinking water containing 0.3%
saccharin. At day 7, osmotic minipumps (OMP, flow rate 5 µl/hr, Alzet 2ml2, Durect Corp.,
Cupertino, USA) loaded with L-thyroxine (T4, Sigma, the Netherlands) solved in NaOH 6.5 mM
and propylene glycol 50%, were implanted under the dorsal skin during the surgical procedure
(see below). OMPs delivered either 1.75 µg (replacement dose; euthyroid groups) or 16 µg
(thyrotoxic groups) T4/ 100 g BW per day.
Surgery
General procedure At day 7, animals were anaesthetized using a mixture of Hypnorm (Janssen;
0.05 mL/100 g BW, i.m.) and Dormicum (Roche, the Netherlands; 0.04 mL/100 g BW, s.c.). During
abdominal surgery, the abdominal cavity was bathed regularly with saline to prevent drying of
the viscera. The abdominal wall and skin were closed separately with sutures (5–0 Perma-Hand
Said Ethicon). After surgery the animals were placed in an incubator (temperature 30 °C) until
awakening and saline (10 mL) was injected subcutaneously to compensate for deficient fluid
intake during recovery. Post-operative care was provided by subcutaneous injection of Temgesic
(Schering-Plough; Utrecht importer), 0.01 mL/100 g BW on the morning after surgery.
Jugular vein and carotid artery cannulation In all animals an intra-atrial silicone cannula
was implanted through the right jugular vein for infusion (12), and a second silicone cannula
was placed in the left carotid artery for blood sampling. Both cannulas were tunnelled to the
head subcutaneously, fixed with dental cement to 4 stainless-steel screws insered into the skull.
A mixture of Amoxicillin 60%, heparin 20% and saline 20% in polyvinylpyruvidon (Sigma, the
Netherlands) was used to fill the cannulas and prevent inflammation and occlusion. In the 10 days
between surgery and the hyperinsulinemic clamps, this mixture was replaced at least 3 times.
Hepatic denervations A laparotomy was performed in the midline. The liver lobes were gently
pushed up and the ligaments around the liver lobes severed. For hepatic sympathectomy (Sx),
the bile duct and the portal vein complex were visualized using an operating microscope (25×
magnification). The bile duct was isolated from the portal vein complex and all tissue running
allong the bile duct was transected using microsurgical instruments. At the level of the hepatic
portal vein, the hepatic artery devides into the hepatic artery proper and the gastro-duodenal
artery. This division occurs on the ventral surface of the portal vein. At this point, the arteries
were separated via blunt dissection from the portal vein. All nerve bundles running along the
hepatic artery proper were removed. Any connective tissue attachments between the hepatic
artery and the portal vein were cut, eliminating any possible nerve crossings. The sympathetic
denervation involves an impairment of both efferent and afferent nerves, but this procedure
does not impair the parasympathetic vagal input to the liver, as shown previously (9).
For hepatic parasympathectomy (Px), the fascia containing the common hepatic vagal branch
(HV) was stretched by gently moving the stomach, revealing the HV as it separates from the
left vagal trunk. The neural tissue was transected between the ventral vagal trunk and the liver.
proefschrift Klieverik.indb 50
4-8-2009 15:22:37
The fascia between the stomach/esophagus and the liver were also transected to remove any
additional small branches (9).
Sham-operated rats, also referred to as intact (Int) rats underwent all procedures as described
above, except for transection of the neural tissue.
51
Effects of thyrotoxicosis and selective hepatic autonomic denervation on hepatic glucose metabolism in rats
At day 17, a hyperinsulinemic euglycemic clamp combined with stable isotope dilution was
performed in all animals. Each experiment consisted of a tracer equilibration period (t=-75 to
t=-15 min), a basal period (t=-15 to t=0 min) and a hyperinsulinemic period (t=0 to t=160 min).
The protocol was based on a validated hyperinsulinemic euglycemic clamping protocol for mice
(13), adjusted for the use of stable isotope tracers and for BW.
In the afternoon on the day before the hyperinsulinemic clamp, rats were connected to a metal
collar attached to polyethylene tubing (for blood-sampling and infusion) which was kept out of
reach of the animals by a counterbalanced beam. This allowed all subsequent manipulations
to be performed outside the cages without handling the animals. At 16.00 pm a venous blood
sample was obtained for determination of plasma concentrations of thyroid stimulating hormone
(TSH), T3 and T4.
On the hyperinsulinemic clamp day, food was removed from the cages 5 h before the first
basal measurements. At 11.00 am, a primed (8.0 μmol in 5 min) continuous (16.6 μmol/h)
infusion of the stable isotope tracer [6,6-2H2]-glucose (>99% enriched; Cambridge Isotope
Laboratories, Cambridge, USA) was started using an infusion pump (Harvard Apparatus,
Holliston, Massachusetts, USA). Before this, a blood sample for determination of background
isotopic enrichment was taken (200 μL, t=-75 min). After 60 min of equilibration time, blood
samples (200 μL) were obtained for measurement of glucose concentration, isotopic enrichment
(t=-15, -5 and 0 min), and plasma insulin concentration (t= -15 min).
Subsequently, a primed (10 mU in 4 min) followed by continuous (62.5 mU/h) infusion of human
recombinant insulin (Actrapid 100 IU/mL, Novo Nordisk, Alphen aan de Rijn, the Netherlands)
was started. To maintain euglycemia at 5.5 mmol/L (clamping period), glucose 25% was infused
at a variable rate. Blood glucose concentrations were measured every 10 min and consequently
glucose infusion rate was adjusted if needed. The 25% glucose solution was 1% enriched with
[6,6-2H2]-glucose to approximate the values for enrichment reached in plasma, thereby minimizing
changes in isotopic enrichment due to variable infusion rates of exogenous glucose. At t=100
min, blood samples (200 μL) were obtained for measurement of plasma glucose concentration,
isotope enrichment (t=100, 115, 130, 145 and 160 min) and plasma insulin concentration (t=100
min). After the clamp, rats were sacrificed and liver tissue was snap frozen and stored at -80°C
for subsequent analysis. Endogenous glucose production (EGP) and rate of disappearance (Rd)
were calculated using modified forms of Steele equations (14;15).
Chapter 3
Hyperinsulinemic euglycemic clamp and stable isotope dilution
Plasma measurements
Plasma glucose concentrations were determined in blood spots (<5 μL) using a glucose meter
(FreestyleTM, Abbott, the Netherlands) with inter- and intra-assay coefficients of variation (CV) of
less than 6% and 4%, respectively. Plasma concentrations of the thyroid hormones T3 and T4 were
determined by an in-house RIA (16), with inter- and intra-assay CV of 7–8% and 3–4% (T3), and
proefschrift Klieverik.indb 51
4-8-2009 15:22:37
52
3–6 and 2–4% (T4), respectively. Detection limits for T3 and T4 were 0.3 nmol/L and 5 nmol/L,
respectively. Plasma TSH concentrations were determined by a chemiluminescent immunoassay
(Immulite 2000, Diagnostic Products Corp., Los Angeles, CA), using a rat-specific standard. The
inter- and intra-assay CV for TSH were less than 4% and 2% at ±3.5 mU/L, respectively, and
the detection limit was 0.60 mU/L. Plasma insulin was measured by a commercially available
Elisa (Mercodia, Uppsala, Sweden). The inter and intra-assay CV were 4% and 2%, detection
limit 13 pmol/L. Glucose enrichment was measured as described earlier (17). The [6,6-2H2]glucose enrichment (tracer/ tracee ratio) inter-assay CV was 1%, the intra-assay CV 1%, and the
detection limit 0.04%.
High performance liquid chromatography electrochemical measurements
To check the effectiveness of the hepatic sympathectomy, norepinephrine (NE) content in the
liver was measured. Liver tissue samples of 50 mg were homogenized in 1 ml ice-cold NH4Cl
buffer (0.2 M, containing 12 nmol/L α-methylnorepinephrine and 1 g/L EDTA, pH 7.0), and
centrifuged twice (14000 rpm) for 15 min at 4°C. NE was determined with an in-house HPLC
method. Essentially, NE was selectively isolated by liquid-liquid extraction (18) and derivatized with
the fluorescent 1,2-diphenylethylenediamine (19). The fluorescent derivatives were separated by
reversed phase liquid chromatography and detected by scanning fluorescence detection (510
pump, 717plus auto-sampler, 474 Scanning fluorescence detector, Waters Chromatography, the
Netherlands). Separation of NE from other endogenous compounds was achieved with a Waters
Xterra RP18 column (5 μm 3.9 x 150 mm). As an internal standard, α-methylnorepinephrine was
used. Intra and inter-assay CV were 3% and 20%, respectively. The detection limit for NE was
0.05 nmol/L.
We have previously evidenced the effectiveness of our method for selective hepatic
parasympathectomy by using retrograde viral tracing (9). In these studies, the success rate of
hepatic parasympathectomy was over 90%.
RNA isolation and Real Time PCR
mRNA was isolated from 10 mg liver tissue using a Magna Pure apparatus and a Magna Pure LC
mRNA isolation kit II (tissue) (Roche Molecular Biochemicals, Mannheim, Germany) according to
the manufacturer’s protocol. cDNA synthesis was performed with the 1st Strand cDNA synthesis
kit for RT-PCR (AMV) (Roche Molecular Biochemicals). Previously published primer pairs were
used to amplify HPRT (hypoxanthine phosphoribosyl transferase, a housekeeping gene) (20). We
designed primer pairs for phosphoenolpyruvate carboxikinase (PEPCK) and type 1 iodothyronine
deiodinase (D1) with the following sequences; PEPCK forward TGCCCTCTCCCCTTAAAAAAG
and reverse CGCTTCCGAAGGAGATGATCT, D1 forward GAAGTGCAACGTCTGGGATT and
reverse CTGCCGAAGTTCAACACCA. Real Time PCR was performed using the LightCycler (Roche
Molecular Biochemicals, Mannheim, Germany) as described earlier (21). PCR programs were as
follows: pre-denaturation 10 min 95°C, amplification for 45 cycles which consists of denaturation
for 10 sec 95°C, annealing at various temperatures for 10 sec and elongation for 15 sec at 72°C
(annealing temperature: HPRT 54°C, PEPCK 55°C and D1 55°C). For quantification, standard
curves were generated of a sequence-specific PCR product ranging from 0.01 fg/µL until 100 fg/
µL. Samples were corrected as to their mRNA content using HPRT mRNA and plotted as relative
proefschrift Klieverik.indb 52
4-8-2009 15:22:37
expression. Samples were individually checked for their PCR-efficiency (22). The median of the
efficiency was calculated for each assay and samples with a greater than 0.05 difference of the
efficiency median value were excluded from the analysis.
Results
Six groups of rats were studied, i.e. 3 euthyroid and 3 thyrotoxic groups. Euthyroid rats were
either sham operated (Eu Int, n=10), or underwent a selective hepatic sympathetic (Eu Sx, n=7)
or parasympathetic denervation (Eu Px, n=9). The same applied to thyrotoxic rats (Tox Int, n=10;
Tox Sx, n=7; Tox Px, n=10). Selective denervation of the sympathetic input to the liver resulted in
a significant 72% reduction in hepatic NE content (13.6 ± 2.3 vs. 49.1 ± 4.5 ng/g liver, p< 0.01,
Sx (n=14) vs. Int (n=20), respectively). On the other hand, the hepatic NE content of animals
with a selective parasympathetic denervation did not differ from intact animals (47.1 ± 8.6 ng/g
liver, Px (n=19)).
Body weight and eating behaviour
At the time of surgery, there was no difference in bodyweight (BW) between groups (Eu Int
348±5, Eu Sx 344±4, Eu Px 358±4, Tox Int 353±8, Tox Sx 350±7; Tox Px 362±5 g, ns).
The changes in BW after surgery and the overnight food intake before the clamp for each
experimental group are depicted in Fig 1. As expected, all thyrotoxic groups lost weight between
surgery and the hyperinsulinemic clamp, in contrast to euthyroid groups. Nevertheless, food
intake was higher (p< 0.01) in all thyrotoxic groups as compared with Eu Int rats, which is in
accordance with increased energy expenditure during thyrotoxicosis. It is important to note that
BW increased in all groups during the 3 days preceding the clamp, indicating full recovery from
surgery and a positive energy balance.
Plasma thyroid hormones
Plasma thyroid hormone concentrations at the time of the hyperinsulinemic clamps after 10
days of T4 treatment are depicted in Table 1.
In Eu Int rats, biochemical euthyroidism was evidenced by similar TSH concentrations as reported
earlier in control rats without hormonal treatment (16;23). Plasma TSH concentrations were
between 2.00 and 3.00 mU/L in all euthyroid groups. TSH was suppressed to levels below the limit
of detection in all thyrotoxic animals. Plasma T3 and T4 concentrations were significantly higher
in Tox Int as compared with Eu Int rats (4.3-fold, p< 0.01 and 2.0-fold, p< 0.01, respectively).
In euthyroid rats hepatic autonomic denervation did not affect plasma T3 and T4. However, in
proefschrift Klieverik.indb 53
53
Effects of thyrotoxicosis and selective hepatic autonomic denervation on hepatic glucose metabolism in rats
Data were analyzed by mixed model analysis of variance (ANOVA), with nature of denervation
(Int, Sx, Px) and thyroid hormonal status (euthyroid, thyrotoxic) as fixed effects. Significance
was defined at p<0.05. A LSD post hoc test was performed if ANOVA revealed significance to
determine which experimental groups differed from each other. Student one sample t-test was
used to determine statistical differences from zero. Mann Whitney U test was used to analyse
the PCR data. Pearson correlation was used to test for associations between factors. Data are
presented as mean ± SE.
Chapter 3
Statistics
4-8-2009 15:22:37
1a
1b
30
^^
^^
0
^^
∆ BW (g)
Food (g)
- 10
20
10
- 20
#
#
- 30
54
#
0
10
Tox Px
7
Tox Sx
10
Tox Int
9
Eu Px
7
Eu Sx
n= 10
Eu Int
- 40
Fig 1 a Food consumed by euthyroid (left) and thyrotoxic (right) groups in the night before the hyperinsulinemic
clamp. ANOVA was significant (p< 0.0001) for factor thyroid hormone status. ^^ p< 0.01 vs. Eu Int as revealed
by post hoc LSD.
b BW difference (ΔBW) between time of surgery and time of the hyperinsulinemic clamp in experimental
groups. Note that despite the increase in food consumption (1a), none of the thyrotoxic groups regained their
time of surgery BW at time of the hyperinsulinemic clamps. # p< 0.01; H0: ΔBW=0, as revealed by student one
sample t-test. Data are mean ± SE. The number of animals per experimental group is depicted under the bars
of figure 1a. See legend with table 1 for definition of group abbreviations.
Table 1 T3, T4 and TSH plasma concentration at time of the hyperinsulinemic clamps in experimental animals.
Eu Int
n = 10
T3 (nmol/L)
0.86 ± 0.08
T4 (nmol/L)
130 ± 6
TSH (mU/L)
2.53 ± 0.94
Eu Sx
n=7
Eu Px
n=9
Tox Int
n = 10
Tox Sx
n=7
0.88 ± 0.05 0.73 ± 0.04 3.66 ± 0.31 ^^2.78 ± 0.33 ^^*
111 ± 9
135 ± 5
2.13 ± 1.02 2.85 ± 0.83
Tox Px
n = 10
2.81 ± 0.18 ^^*
255 ± 18 ^^
169 ± 15 ^*
209 ± 17 ^^*
< 0.60 ^
< 0.60 ^
< 0.60 ^
Eu Int = euthyroid sham liver denervated, Eu Sx = euthyroid hepatic sympatectomized, Eu Px = euthyroid hepatic
parasympatectomized, Tox Int = thyrotoxic sham liver denervated, Tox Sx = thyrotoxic hepatic sympatectomized,
Tox Px = thyrotoxic hepatic parasympatectomized. Data are mean ± SE. ^ p< 0.05, ^^ p< 0.01 vs. Eu Int, * p<
0.01 vs. Tox Int as revealed by post hoc LSD, only when ANOVA indicated p< 0.05 for the respective factor.
thyrotoxic rats both sympathetic and parasympathetic hepatic denervation lowered plasma T3
and T4, but not the T3/ T4 ratio, as compared with intact animals. In Tox Int rats, the T3/ T4 ratio
was significantly increased as compared with Eu Int animals.
Glucose, insulin, glucose kinetics and hepatic mRNA expression
Basal state Tox Int rats exhibited increased basal plasma glucose concentration as compared
with Eu Int rats (Fig 2a). Basal insulin concentration was not affected by thyrotoxicosis per
se (Fig 2b). In accordance with the increase in basal plasma glucose concentration, basal EGP
was increased by 45% in Tox Int as compared with Eu Int rats (Fig 2c). In line with this, mRNA
expression of PEPCK, the rate limiting enzyme of gluconeogenesis, was increased 3,5-fold in
thyrotoxic animals (Fig 3a). In addition, mRNA expression of the T3 responsive gene deiodinase
type 1 was upregulated 4-fold in the liver of thyrotoxic animals supporting the thyrotoxic state on
the level of the hepatocyte (Fig 3b).
proefschrift Klieverik.indb 54
4-8-2009 15:22:37
2a
7
^
5
4
3
n= 10
9
10
7
600
*
500
Insulin (pmol/L)
10
400
300
200
100
0
EGP Basal(µmol/kg*min)
2c
100
^^
^
^^
75
50
25
Tox Px
Tox Sx
Tox Int
Eu Px
Eu Sx
Eu Int
0
Fig 2 a Mean basal plasma glucose concentrations.
ANOVA indicated p< 0.003 for factor thyroid hormonal
status. ^ p< 0.05 vs. Eu Int as revealed by post hoc
LSD.
b Basal plasma insulin concentrations. Note the
elevated insulin concentration in thyrotoxic parasympathectomized rats. ANOVA indicated p< 0.001 for
factor denervation status and p< 0.05 for interaction.
* p≤ 0.01 vs. Tox Int as revealed by post hoc LSD.
c Endogenous glucose production (EGP) in the basal
condition of experimental groups. Note increased
EGP all thyrotoxic groups relative to euthyroid intact
rats, except for Tox Sx animals, in which the relative
increase in EGP is lower. ANOVA was significant (p<
0.0001) for factor thyroid hormonal status. ^ p≤
0.05, ^^ p≤ 0.01 vs. Eu Int as revealed by post hoc
LSD. Data are mean ± SE. The number of animals
per experimental group is depicted under the bars of
figure 2a. See legend with table 1 for definition of
group abbreviations.
proefschrift Klieverik.indb 55
55
Effects of thyrotoxicosis and selective hepatic autonomic denervation on hepatic glucose metabolism in rats
2b
7
Chapter 3
Glucose (mmol/L)
^
6
In euthyroid rats, selective hepatic sympathetic
or parasympathetic denervation did not affect
glucose concentration, insulin concentration
or EGP. In thyrotoxic rats, there was no
statistically significant effect of sympathetic
denervation on basal EGP in thyrotoxic
animals. However, the relative increase of
EGP in Tox Sx rats as compared with Eu Int
rats was smaller than in the other thyrotoxic
groups (Fig 2c). In line with this, Tox Sx animals
exhibited unaltered glucose concentration as
compared with Eu Int rats, in contrast to the
increased glucose concentration in Tox Int
and Tox Px rats (Fig 2a). During thyrotoxicosis,
selective hepatic parasympatectomy induced a
marked increase in basal insulin concentration
(Fig 2b). This increase in insulin concentration
was accompanied by unaltered glucose
concentration and EGP in Tox Px as compared
with Tox Int animals, indicating hepatic insulin
resistance. There was no effect of hepatic
denervation in both euthyroid and thyrotoxic
rats on hepatic mRNA expression of PEPCK
and D1 (data not shown).
Hyperinsulinemic state After 100 min
of insulin infusion, insulin concentrations
increased in all groups (p<0.05) as compared
to the basal state (mean increment 229±25
pmol/L). There was no difference in plasma
insulin concentration between groups.
As expected, EGP decreased in all groups
in response to hyperinsulinemia. During
the clamp, EGP was 122% higher in Tox
Int as compared with Eu Int rats. There
was no effect of sympathetic denervation
on hyperinsulinemic EGP in euthyroid or
thyrotoxic rats. Parasympathetic denervation
slightly increased hyperinsulinemic EGP in
euthyroid and thyrotoxic rats, although this
effect missed statistical significance (Fig 4a).
In Tox Int rats, the relative suppression of
EGP in the hyperinsulinemic as compared
4-8-2009 15:22:38
3b
5
^
4
3
2
1
0
Eu
n=9
^^
5
Relative D1 mRNA expression
56
Relative PEPCK mRNA expression
3a
4
3
2
1
0
Tox
n=9
Eu
n=8
Tox
n=9
Fig 3 Hepatic mRNA expression of PEPCK (a) and D1 (b), relative to HPRT, a housekeeping gene. Note the 3,5fold increase in hepatic PEPCK and the 4-fold increase in D1 expression in thyrotoxic intact relative to euthyroid
intact rats. Statistical differences are depicted by symbols; ^ p< 0.05, ^^ p< 0.01 vs. Eu. Data are mean ± SE.
Number of animals per experimental group is depicted under the bars of both figures. Eu = euthyroid intact
animals, Tox = thyrotoxic intact animals.
4a
4b
75
^^
^^
^^
50
25
75
^
50
^
^
25
Tox Px
10
Tox Sx
7
Tox Int
10
Eu Px
9
Eu Sx
7
Eu Int
0
0
n= 10
Suppression EGP (%)
EGP Clamp(µmol/kg*min)
100
Fig 4 a Endogenous glucose production (EGP) during the hyperinsulinemic clamp of experimental groups. Note
increased EGP in all thyrotoxic groups relative to euthyroid intact rats. ANOVA was significant (p< 0.0001) for
factor thyroid hormonal status. ^^ p≤ 0.01 vs. Eu Int as revealed by post hoc LSD.
b Percent suppression of EGP in the hyperinsulinemic state relative to the basal state of experimental groups.
Note that all thyrotoxic groups exhibit decreased % suppression of EGP as compared with euthyroid intact
rats, indicating hepatic insulin resistance during thyrotoxicosis. ANOVA indicated p< 0.009 for factor thyroid
hormonal status, p= 0.216 for factor denervation. ^ p≤ 0.05 vs. Eu Int as revealed by post hoc LSD. Data are
mean ± SE. Number of animals per experimental group is depicted under the bars of figure 3a. See legend with
table 1 for definition of group abbreviations.
with the basal state (% suppression of EGP) was decreased relative to Eu Int rats (Fig 4b). There
was a highly significant positive correlation between plasma T3 concentration and EGP both
in the basal and hyperinsulinemic state (Fig 5). Rd was similar between the groups (data not
shown).
proefschrift Klieverik.indb 56
4-8-2009 15:22:38
5b
5a
p< 0.0001
r 2= 0.41
EGP Clamp (µmol/kg*min)
100
75
50
25
0
Eu Int
Eu Sx
Eu Px
Tox Int
Tox Sx
Tox Px
p< 0.0001
r 2= 0.27
125
100
75
Chapter 3
EGP Basal(µmol/kg*min)
125
50
25
57
0
1
2
3
T3 (nmol/L)
4
5
6
0
1
2
3
4
5
6
T3 (nmol/L)
Fig 5 Relation between basal EGP (a) or hyperinsulinemic EGP (b) and plasma T3 concentration in all experimental
animals. Pearson correlation coefficient and p values are depicted in the lower right corner of both figures.
Discussion
The primary findings of this study are that chronic, mild thyrotoxicosis in rats increases EGP, while
it decreases relative suppression of EGP during hyperinsulinemic clamps, indicating hepatic insulin
resistance. This is supported by a highly significant, positive correlation between plasma T3 and
EGP. The increased EGP during thyrotoxicosis can be attenuated by selective hepatic Sx. Selective
hepatic Px increases plasma insulin in thyrotoxic rats without a change in EGP, indicating hepatic
insulin resistance. These combined findings indicate that T3 is an important direct determinant of
EGP in thyrotoxicosis with a small contribution via the sympathetic nervous system. Furthermore,
parasympathetic innervation of the liver may function to restrain EGP during mild thyrotoxicosis,
as after Px more insulin is needed to keep EGP at the level found in mild thyrotoxicosis.
The striking resemblance between many of the effects of thyrotoxicosis and sympathetic nervous
stimulation has been long noted. Because of this similarity, the syndrome of goiter, exophthalmus
and tachycardia as described by Von Basedow (i.e. the Merseberg triad), has been regarded a
disease of the sympathetic nervous system by physiologists at the time (24). By the end of the
19th century, this led to the surgical treatment of severe thyrotoxicosis by cervical sympathetic
chain resection (25) and later by high spinal anaesthesia or adrenal denervation (26). These
practices were gradually abandoned with increasing knowledge of the thyroid gland and of TH.
However, it is still common practice nowadays to start treatment of severe thyrotoxicosis with
beta-adrenergic blockers until a clinical effect of anti-thyroid drugs is reached.
In literature, the idea of increased sympathetic tone during hyperthyroidism has gradually moved
to the background. However, as more accurate techniques for measuring sympathetic tone
have become available, evidence is building up for increased sympathetic neural output to white
adipose tissue in hyperthyroid patients (27). In addition, hyperthyroid patients exhibit increased
sympathetic and decreased parasympathetic output to the heart as revealed by heart rate
spectral analysis (28;29).
To discriminate between peripheral and central effects of TH, an experimental model is needed
in which central and peripheral manipulations can be performed without interfering with the
systemic thyroid hormone milieu. For this, we have used “block and replacement” treatment in
proefschrift Klieverik.indb 57
Effects of thyrotoxicosis and selective hepatic autonomic denervation on hepatic glucose metabolism in rats
0
4-8-2009 15:22:38
58
rats. This means that rats are treated with the thyreostatic MMI in drinking water to inhibit TH
synthesis and simultaneously, T4 is administered by use of osmotic minipumps. In this way, T4
is released continuously, mimicking the release of TH by the thyroid gland. We used two doses
of T4, i.e., a replacement dose giving rise to sustained euthyroidism and an 8-fold higher dose
inducing mild thyrotoxicosis. The duration of T4 administration was 10 days, resulting in a chronic
state of mild thyrotoxicosis in rats treated with the highest T4 dose, as evidenced by a 4.3-fold
and 2.0-fold increase in plasma T3 and T4, respectively, and a decrease of plasma TSH. In line with
this, the hepatic mRNA expression of the D1, a T3 responsive gene which is a sensitive marker of
peripheral thyroid status (30), showed a 4-fold increase in thyrotoxic animals.
In the present study, we have shown for the first time that mild thyrotoxicosis induces hepatic
insulin resistance in freely moving, conscious rats. A hyperinsulinemic euglycemic clamp combined
with isotope dilution is the gold standard for measuring hepatic insulin sensitivity (31). Although
there have been reports of altered EGP during thyroid hormone excess in vivo (32;33), to our
knowledge up until now hyperinsulinemic euglycemic clamps combined with isotope dilution
have never been used to study alterations in EGP and its sensitivity to insulin in thyrotoxic rats.
In the literature, data on the effect of hepatic Px on EGP and hepatic insulin resistance are
discordant. Studies combining manipulation of fatty acid metabolism and hepatic vagal
denervation in rats, point to an important role of the hepatic vagal nerve in mediating the effects
of central lipid sensing on EGP (10;34). Likewise, the repressive effect of icv infused insulin on
EGP can be completely abolished by hepatic vagal denervation (11). In these and other studies
(35), hepatic Px in itself does not affect EGP. In the present study, during thyrotoxicosis but
not during euthyroidism, selective hepatic Px induced insulin resistance. The notion arises that
a consistent effect of vagal hepatic denervation becomes manifest only in combination with an
additional stimulus such as manipulation of central lipid sensing, icv insulin infusion or systemic
thyroid hormone excess.
We observed a decrease in plasma concentrations of T3 and T4 by both Sx and Px in thyrotoxic
rats, although equal doses of T4 were administered via osmotic minipumps in thyrotoxic intact
and thyrotoxic denervated animals. Hepatic denervation did not result in significant changes
in hepatic D1 mRNA expression, in accordance with the unaffected T3/ T4 ratio. Thus, altered
synthesis or clearance of TH binding proteins such as transthyretin in the liver, or altered hepatic
clearance of TH resulting from hepatic autonomic denervation during thyrotoxicosis are more
plausible explanations.
The present study shows that the alterations in glucose metabolism induced by thyrotoxicosis
are slightly modulated by selective hepatic autonomic denervation. Thus, the efferent autonomic
nerves may be responsible for part of these changes. At this stage, it remains unknown which CNS
areas control the ANS efferent nerves affecting hepatic glucose metabolism during thyrotoxicosis.
The hypothalamus is a key central site of thyroid hormone action, and both the human (36) and
rat (37) hypothalamus abundantly express thyroid hormone receptors. This specifically applies to
the paraventricular nucleus, where the (pre-) autonomic neurons that control sympathetic and
parasympathetic motor-neurons are located, and the arcuate nucleus, where the blood brain
barrier is absent. This suggests that the hypothalamus is perfectly equipped to sense and process
TH signals, not only via the neuro-endocrine route but also via its connections with the ANS, and
proefschrift Klieverik.indb 58
4-8-2009 15:22:38
We wish to thank Mr. J. van der Vliet for his excellent help with animal surgery, and we are
indebted to Ms. E.M. Johannesma, Ms. B.C.E. Voermans and Ms. A.F.C. Ruiter for performing
the hormone and isotope analyses, and Ms. R. van der Spek for performing RNA isolation and
real time PCR.
Reference List
1. Dimitriadis GD, Raptis SA 2001 Thyroid hormone excess and glucose intolerance. Exp Clin Endocrinol
Diabetes 109 Suppl 2:S225-S239
2. Cavallo-Perin P, Bruno A, Boine L, Cassader M, Lenti G, Pagano G 1988 Insulin resistance in Graves’
disease: a quantitative in-vivo evaluation. Eur J Clin Invest 18:607-613
3. Yen PM 2001 Physiological and molecular basis of thyroid hormone action. Physiol Rev 81:1097-1142
4. Constant EL, de Volder AG, Ivanoiu A, Bol A, Labar D, Seghers A, Cosnard G, Melin J, Daumerie C 2001
Cerebral blood flow and glucose metabolism in hypothyroidism: a positron emission tomography study.
J Clin Endocrinol Metab 86:3864-3870
5. Schreckenberger MF, Egle UT, Drecker S, Buchholz HG, Weber MM, Bartenstein P, Kahaly GJ 2006
Positron emission tomography reveals correlations between brain metabolism and mood changes in
hyperthyroidism. J Clin Endocrinol Metab 91:4786-4791
6. Smith CD, Ain KB 1995 Brain metabolism in hypothyroidism studied with 31P magnetic-resonance
spectroscopy. Lancet 345:619-620
7. Schwartz MW, Porte D, Jr. 2005 Diabetes, obesity, and the brain. Science 307:375-379
proefschrift Klieverik.indb 59
59
Effects of thyrotoxicosis and selective hepatic autonomic denervation on hepatic glucose metabolism in rats
Acknowledgements
Chapter 3
to regulate hepatic glucose metabolism. The observations that other hormones such as insulin
(11), estrogen (38) and glucocorticoids (39) affect metabolism via central (hypothalamic) sites of
action independently of the peripheral hormonal milieu support the possibility that TH may affect
autonomic outflow from the hypothalamus to the liver. However, we are aware of only one
study addressing peripheral physiological effects of centrally administered TH. In this study (40),
an intracerebroventricular (icv) bolus infusion of T3 in surgically thyroidectomized, hypothyroid
rats increased heart rate. The same dose showed no effect after intravenous infusion, suggesting
T3-responsive central nervous system regulation of heart rate. Taken together, these data support
the notion that TH may affect hepatic glucose metabolism via central (hypothalamic) actions,
which will be the subject of further studies.
In conclusion, we have shown that chronic, mild thyrotoxicosis in rats increases EGP and induces
hepatic insulin resistance. The increase in EGP is slightly attenuated by selective sympathetic liver
denervation. T3 is an important direct determinant of EGP in thyrotoxicosis with a small indirect
contribution via the sympathetic nervous system. Selective parasympathetic denervation during
thyrotoxicosis aggravates hepatic insulin resistance. By inference, parasympathetic innervation of
the liver may function to restrain EGP during mild thyrotoxicosis. During systemic thyrotoxicosis,
the peripheral effects of TH on EGP apparently outweigh the hypothesized central effects. This
does not exclude a role for central TH action in fine-tuning of EGP. Further studies will be needed
to reveal if thyroid hormone affects EGP upon central administration in physiological, euthyroid
conditions.
4-8-2009 15:22:39
8. La Fleur SE, Kalsbeek A, Wortel J, Buijs RM 2000 Polysynaptic neural pathways between the hypothalamus,
including the suprachiasmatic nucleus, and the liver. Brain Res 871:50-56
9. Kalsbeek A, La FS, Van HC, Buijs RM 2004 Suprachiasmatic GABAergic inputs to the paraventricular
nucleus control plasma glucose concentrations in the rat via sympathetic innervation of the liver. J
Neurosci 24:7604-7613
10. Lam TK, Pocai A, Gutierrez-Juarez R, Obici S, Bryan J, Aguilar-Bryan L, Schwartz GJ, Rossetti L 2005
Hypothalamic sensing of circulating fatty acids is required for glucose homeostasis. Nat Med 11:320327
60
11. Obici S, Zhang BB, Karkanias G, Rossetti L 2002 Hypothalamic insulin signaling is required for inhibition
of glucose production. Nat Med 8:1376-1382
12. Steffens AB 1959 A method for frequent sampling blood and continuous infusion of fluids in the rat
without disturbing the animal. Physiol Behav 4, 833-836
13. Van den Hoek AM, Heijboer AC, Corssmit EP, Voshol PJ, Romijn JA, Havekes LM, Pijl H 2004 PYY3-36
reinforces insulin action on glucose disposal in mice fed a high-fat diet. Diabetes 53:1949-1952
14. Finegood DT, Bergman RN, Vranic M 1987 Estimation of endogenous glucose production during
hyperinsulinemic-euglycemic glucose clamps. Comparison of unlabeled and labeled exogenous glucose
infusates. Diabetes 36:914-924
15. Steele R 1959 Influences of glucose loading and of injected insulin on hepatic glucose output. Ann N Y
Acad Sci 82:420-430
16. Kalsbeek A, Fliers E, Franke AN, Wortel J, Buijs RM 2000 Functional connections between the
suprachiasmatic nucleus and the thyroid gland as revealed by lesioning and viral tracing techniques in
the rat. Endocrinology 141:3832-3841
17. Ackermans MT, Pereira Arias AM, Bisschop PH, Endert E, Sauerwein HP, Romijn JA 2001 The quantification
of gluconeogenesis in healthy men by (2)H2O and [2-(13)C]glycerol yields different results: rates of
gluconeogenesis in healthy men measured with (2)H2O are higher than those measured with [2-(13)C]
glycerol. J Clin Endocrinol Metab 86:2220-2226
18. Smedes F, Kraak JC, Poppe H 1982 Simple and fast solvent extraction system for selective and quantitative
isolation of adrenaline, noradrenaline and dopamine from plasma and urine. J Chromatogr 231:25-39
19. van der Hoorn FA, Boomsma F, Man in ‘t Veld AJ, Schalekamp MA 1989 Determination of catecholamines
in human plasma by high-performance liquid chromatography: comparison between a new method
with fluorescence detection and an established method with electrochemical detection. J Chromatogr
487:17-28
20. Sweet MJ, Leung BP, Kang D, Sogaard M, Schulz K, Trajkovic V, Campbell CC, Xu D, Liew FY 2001 A
novel pathway regulating lipopolysaccharide-induced shock by ST2/T1 via inhibition of Toll-like receptor
4 expression. J Immunol 166:6633-6639
21. Boelen A, Kwakkel J, Thijssen-Timmer DC, Alkemade A, Fliers E, Wiersinga WM 2004 Simultaneous changes
in central and peripheral components of the hypothalamus-pituitary-thyroid axis in lipopolysaccharideinduced acute illness in mice. J Endocrinol 182:315-323
22. Ramakers C, Ruijter JM, Deprez RH, Moorman AF 2003 Assumption-free analysis of quantitative realtime polymerase chain reaction (PCR) data. Neurosci Lett 339:62-66
23. Kalsbeek A, Buijs RM, van SR, Kaptein E, Visser TJ, Doulabi BZ, Fliers E 2005 Daily variations in type II
iodothyronine deiodinase activity in the rat brain as controlled by the biological clock. Endocrinology
146:1418-1427
24. Leak D 1970 The thyroid and the autonomic nervous system. 1-5. William Heinemann Medical Books,
London.
25. Poncet MA 1897 Le traitement chirurgical des goitre exophthalmique par la section ou la résection due
sympathique cervical. Bull Acad Med 38:121
proefschrift Klieverik.indb 60
4-8-2009 15:22:39
26. Crile GW 1929 The interdependence of the thyroid, adrenals and nervous system. Amer.J.Surg. 6:616.
29. Chen JL, Chiu HW, Tseng YJ, Chu WC 2006 Hyperthyroidism is characterized by both increased
sympathetic and decreased vagal modulation of heart rate: evidence from spectral analysis of heart rate
variability. Clin Endocrinol (Oxf) 64:611-616
61
30. Zavacki AM, Ying H, Christoffolete MA, Aerts G, So E, Harney JW, Cheng SY, Larsen PR, Bianco AC
2005 Type 1 iodothyronine deiodinase is a sensitive marker of peripheral thyroid status in the mouse.
Endocrinology 146:1568-1575
31. Ferrannini E, Mari A 1998 How to measure insulin sensitivity. J Hypertens 16:895-906
32. Jin ES, Burgess SC, Merritt ME, Sherry AD, Malloy CR 2005 Differing mechanisms of hepatic glucose
overproduction in triiodothyronine-treated rats vs. Zucker diabetic fatty rats by NMR analysis of plasma
glucose. Am J Physiol Endocrinol Metab 288:E654-E662
33. Okajima F, Ui M 1979 Metabolism of glucose in hyper- and hypo-thyroid rats in vivo. Glucose-turnover
values and futile-cycle activities obtained with 14C- and 3H-labelled glucose. Biochem J 182:565-575
34. Pocai A, Lam TK, Gutierrez-Juarez R, Obici S, Schwartz GJ, Bryan J, guilar-Bryan L, Rossetti L 2005
Hypothalamic K(ATP) channels control hepatic glucose production. Nature 434:1026-1031
35. Bernal-Mizrachi C, Xiaozhong L, Yin L, Knutsen RH, Howard MJ, Arends JJ, Desantis P, Coleman T,
Semenkovich CF 2007 An afferent vagal nerve pathway links hepatic PPARalpha activation to
glucocorticoid-induced insulin resistance and hypertension. Cell Metab 5:91-102
36. Alkemade A, Vuijst CL, Unmehopa UA, Bakker O, Vennstrom B, Wiersinga WM, Swaab DF, Fliers E
2005 Thyroid hormone receptor expression in the human hypothalamus and anterior pituitary. J Clin
Endocrinol Metab 90:904-912
37. Lechan RM, Qi Y, Jackson IM, Mahdavi V 1994 Identification of thyroid hormone receptor isoforms
in thyrotropin-releasing hormone neurons of the hypothalamic paraventricular nucleus. Endocrinology
135:92-100
38. Clegg DJ, Brown LM, Woods SC, Benoit SC 2006 Gonadal hormones determine sensitivity to central
leptin and insulin. Diabetes 55:978-987
39. Cusin I, Rouru J, Rohner-Jeanrenaud F 2001 Intracerebroventricular glucocorticoid infusion in normal
rats: induction of parasympathetic-mediated obesity and insulin resistance. Obes Res 9:401-406
40. Goldman M, Dratman MB, Crutchfield FL, Jennings AS, Maruniak JA, Gibbons R 1985 Intrathecal
triiodothyronine administration causes greater heart rate stimulation in hypothyroid rats than
intravenously delivered hormone. Evidence for a central nervous system site of thyroid hormone action.
J Clin Invest 76:1622-1625
proefschrift Klieverik.indb 61
Effects of thyrotoxicosis and selective hepatic autonomic denervation on hepatic glucose metabolism in rats
28. Burggraaf J, Tulen JH, Lalezari S, Schoemaker RC, De Meyer PH, Meinders AE, Cohen AF, Pijl H 2001
Sympathovagal imbalance in hyperthyroidism. Am J Physiol Endocrinol Metab 281:E190-E195
Chapter 3
27. Haluzik M, Nedvidkova J, Bartak V, Dostalova I, Vlcek P, Racek P, Taus M, Svacina S, Alesci S, Pacak
K 2003 Effects of hypo- and hyperthyroidism on noradrenergic activity and glycerol concentrations in
human subcutaneous abdominal adipose tissue assessed with microdialysis. J Clin Endocrinol Metab
88:5605-5608
4-8-2009 15:22:39
proefschrift Klieverik.indb 62
4-8-2009 15:22:42
4
Thyroid hormone modulates
glucose production via
a sympathetic pathway
from the hypothalamic
paraventricular nucleus
to the liver
Lars P. Klieverik
Sarah F. Janssen
Annelieke van Riel
Ewout Foppen
Peter H. Bisschop
Mireille J. Serlie
Anita Boelen
Mariëtte Ackermans
Hans P. Sauerwein
Eric Fliers
Andries Kalsbeek
Proceedings of the National Academy of Sciences of
the United States of America 2009: 106(14), 59665971.
proefschrift Klieverik.indb 63
4-8-2009 15:22:45
Abstract
64
Thyrotoxicosis increases endogenous glucose production (EGP) and induces hepatic insulin
resistance. We have recently shown that these alterations can be modulated by selective hepatic
sympathetic and parasympathetic denervation, pointing to neurally mediated effects of thyroid
hormone on glucose metabolism. Here, we investigated the effects of central triiodothyronine
(T3) administration on EGP.
We used stable isotope dilution to measure EGP before and after intracerebroventricular (icv)
bolus infusion of T3 or vehicle in euthyroid rats. To study the role of hypothalamic pre-autonomic
neurons, bilateral T3 microdialysis in the paraventricular nucleus (PVN) was performed during 2 h.
Finally, we combined T3 microdialysis in the PVN with selective hepatic sympathetic denervation
to delineate the involvement of the sympathetic nervous system in the observed metabolic
alterations.
T3 microdialysis in the PVN increased EGP by 11±4% (p=0.020) while EGP decreased by 5±8%
(ns) in vehicle treated rats (T3 vs Veh p=0.030). Plasma glucose increased by 29±5% (p=0.0001)
after T3 microdialysis versus 8±3% in vehicle treated rats (T3 vs Veh p=0.003). Similar effects
were observed after icv T3 administration. Effects of PVN T3 microdialysis were independent
of plasma T3, insulin, glucagon and corticosterone. However, selective hepatic sympathectomy
completely prevented the effect of T3 microdialysis on EGP.
We conclude that stimulation of T3-sensitive neurons in the PVN of euthyroid rats increases EGP
via sympathetic projections to the liver, independently of circulating glucoregulatory hormones.
This represents a novel central pathway for modulation of hepatic glucose metabolism by thyroid
hormone.
proefschrift Klieverik.indb 64
4-8-2009 15:22:45
Introduction
Chapter 4
65
Hypothalamic T3 modulates hepatic glucose production
Thyroid hormones are crucial regulators of metabolism, as illustrated by the profound metabolic
derangements in patients with thyrotoxicosis or hypothyroidism (1). Thyrotoxicosis is associated
with an increase in endogenous glucose production (EGP), hepatic insulin resistance and
concomitant hyperglycemia (1;2). We have recently shown that selective hepatic sympathetic
denervation attenuates the hyperglycemia and increased EGP during thyrotoxicosis, while selective
hepatic parasympathetic denervation aggravates hepatic insulin resistance in thyrotoxic rats. By
inference, the increase in EGP during thyrotoxicosis may be mediated in part by sympathetic
input to the liver, while parasympathetic hepatic input may function to restrain insulin resistance
during thyrotoxicosis (3).
The central nervous system is emerging as an important target for several endocrine and humoral
factors in regulating metabolism. Hormones like insulin (4), estrogen (5) and corticosteroids (6)
appear to use dual mechanisms to affect metabolism, i.e. by direct actions in the respective target
tissue and by indirect actions via the hypothalamus, in turn affecting target tissues via autonomic
nervous system (ANS) projections. For example, it has been convincingly shown that the
suppression of EGP by central, i.e., hypothalamic, insulin administration can be largely abolished
by selective hepatic vagal denervation (7;8). The hypothalamus also can stimulate sympathetic
efferent nerves in order to increase hepatic glucose production (9). Thyroid hormone receptors
(TR) are expressed in both the human and rat hypothalamus, showing abundant expression in
the paraventricular (PVN) and arcuate nuclei (10;11) These nuclei are both key players in the
regulation of glucose metabolism via ANS connections with the liver.
We hypothesized that T3 may increase EGP via a neural route from the hypothalamus to the
liver. To explore this hypothesis we investigated whether the increased EGP and hyperglycemia
observed earlier during systemic thyrotoxicosis could be established by inducing “central
thyrotoxicosis” in peripherally euthyroid animals. In addition, we studied the possible involvement
of the hypothalamic PVN and the sympathetic outflow to the liver in the metabolic effects of
central T3. We demonstrate for the first time that upon selective administration to the PVN, T3
increases EGP and plasma glucose, and that these hypothalamic T3 effects are mediated via
sympathetic projections to the liver.
Materials and Methods
Animals
Male Wistar rats (Harlan, Horst, the Netherlands), housed under constant conditions of
temperature (21 ± 1 °C) and humidity (60 ± 2%) with a 12-h light, 12-h dark schedule (lights on
at 7.00 h am) were used for all experiments. Body weight was between 350 and 375 g. Food
and drinking water were available ad libitum. All of the following experiments were conducted
with the approval of the Royal Netherlands Academy of Arts and Sciences.
Experimental groups
Experiment #1 In the first experiment rats treated with methimazole and thyroxine were equipped
with unilateral cannulas aimed at the left lateral cerebral ventricle to receive an icv bolus infusion
proefschrift Klieverik.indb 65
4-8-2009 15:22:45
66
of T3 or vehicle. At t=0 and at t=24h isotope dilution and blood sampling were performed for
measurement of EGP, plasma glucose and (glucoregulatory) hormone concentrations.
Experiment #2 In the second experiment, rats were equipped with bilateral microdialysis (MD)
probes aimed at the hypothalamic PVN. After a basal EGP measurement at t=0, isotope infusion
was continued and continuous T3 or vehicle MD was started. After 90 min, blood samples
were obtained for measurement of EGP, plasma glucose and (glucoregulatory) hormone
concentrations.
Experiment #3 In the third experiment, T3 MD in the PVN (see experiment-2) was performed
in surgically hepatic sympatectomized animals (T3 MD HSx, n=8) and sham denervated animals
(T3 MD Sham, n=6). In all PVN MD experiments, to avoid inclusion of animals that were not
systemically euthyroid after 2h of MD (see results experiment 1), we excluded rats with plasma T3
levels above the upper limit of the reference range (1.8 nmol/L) from the final analysis. In order
to minimize bias, we excluded rats with basal insulin concentrations above the upper limit of the
reference range (>655 pmol/L) from the final analysis. Reference ranges were determined as
mean ± 2 SD from basal samples of 26 intact rats of the same age with no hormonal treatment.
Moreover, we carefully checked MD probe placement. Only animals with bilateral probes that
were positioned within or at the border of the PVN were included in the final analysis.
Hormonal treatment
In experiment #1 we pre-treated rats with methimazole 0.025% and 0.3% saccharin in drinking
water starting 7 days prior to surgery, and administered T4 (1.75 µg/100 g/day) using osmotic
minipumps starting at time of surgery to reinstate euthyroidism (block and replacement), as
reported previously (3).
Surgery
Animals were anaesthetized using Hypnorm (Janssen; 0.05 mL/100 g BW, i.m.) and Dormicum
(Roche, the Netherlands; 0.04 mL/100 g BW, s.c.). In all animals an intra-atrial silicone cannula
was implanted through the right jugular vein and a second silicone cannula was placed in the
left carotid artery for isotope infusion and blood sampling. Both cannulas were tunnelled to the
head subcutaneously (9). Stainless steel icv probes were implanted in the left cerebral ventricle
using the following stereotaxic coordinates: anteroposterior: -0.8 mm, lateral: +2.0 mm, ventral:
-3.2 mm, with the toothbar set at -3.4 mm. The U-shaped tip of the MD probe was 1.5 mm long,
0.7 mm wide, and 0.2 mm thick (9). Bilateral MD probes were stereotaxically implanted, directly
lateral to the PVN, using the following stereotaxic coordinates: anteroposterior: -1.8 mm, lateral:
2.0 mm, ventral: -8.1 mm, with the toothbar set at -3.4 mm. Hepatic sympathetic denervation
(HSx) was performed as described previously (3;9). It involves an impairment of both efferent
and afferent nerves, but this procedure does not impair the parasympathetic vagal input to the
liver (9). Sham-operated rats underwent the same surgical procedures as HSx animals, except
for transection of the neural tissue. To confirm succesfull sympathetic denervation, HPLC for
noradrenaline (NA) was performed on liver homogenates, as described earlier (3).
Stable isotope dilution and central T3 administration
General procedure 10 days after surgery, stable isotope dilution was performed combined with
central administration of T3. In the afternoon on the day before the central T3 experiments,
proefschrift Klieverik.indb 66
4-8-2009 15:22:45
Chapter 4
67
Hypothalamic T3 modulates hepatic glucose production
rats were connected to a metal collar attached to polyethylene tubing (for blood sampling and
infusion) which was kept out of reach of the animals by a counterbalanced beam. This allowed
all subsequent manipulations to be performed outside the cages without handling the animals.
At 14.00 pm a blood sample was obtained for determination of basal plasma thyroid hormones
concentrations. On the day of the central T3 experiments, (basal) EGP was determined using the
stable isotope tracer [6,6-2H2]-glucose, as described previously (3).
Experiment #1: bolus T3 infusion After the last basal blood sample, the isotope infusion
pump was stopped. Animals received an icv bolus infusion of either 1,5 nmol/100 g BW T3
(Sigma, the Netherlands) in 0,05 M NaOH (T3 icv group) or 0,05 M NaOH (Vehicle group) in 4 μL
over 160 sec. This dose and the 24h time-interval were adopted from Goldman et al, showing
positive chronotropic effects of icv T3 in hypothyroid rats (12). After the bolus infusion, food was
placed back in the cages. Five h after the icv bolus infusion, a blood sample was obtained for
measurement of plasma T3. The next day, the infusion of [6,6-2H2]-glucose was started again
with subsequent blood sampling for measurement of glucose concentration, hormones and
isotopic enrichment. All experimental manipulations on the second day were performed in the
same way and at the same time points as on the day before.
Experiment #2 and #3: T3 MD in the hypothalamic PVN Recovery of the MD probes for T3
was 0.24%, as established by in vitro experiments. A solution of 155 μg/ml T3 dissolved in 2 mM
NaOH in PBS (pH 9), was infused through the MD probe-inlet equivalent to 100 pmol/h T3 (T3
MD group). Vehicle MD rats were microdialysed with 2 mM NaOH in PBS (pH 9). The dose of 100
pmol/h T3 was chosen based on the study by Kong et al (26), which is - to our knowledge - the
only study to date reporting local brain infusion of T3. Ringer dialysis (3 μL/min) was performed
from 60 min before the start of isotope infusion and continued until after the last basal blood
sample (t=0 min), when the Ringer was replaced by either T3 or vehicle. Ninety min after the start
of the T3 vehicle administration (with continued isotope infusion) blood samples (200 μL) were
obtained for measurement of glucose concentration, glucoregulatory hormones (t= 90 min), T3
and T4 (t= 120 min) and isotopic enrichment (t=90, 100, 110 and 120 min).
After the central infusion experiments, rats were sacrificed and whole brains were frozen for
subsequent analysis of MD probe placement. Hypothalamic (PVN) placement of bilateral probes
was evaluated blindly in each experimental animal by an experienced neuro-anatomist and scored
on the basis of anteroposteriority, laterality and dorsoventrality. Endogenous glucose production
(EGP) was calculated from isotope enrichment using adapted Steele equations (13).
Plasma analyses
Plasma glucose concentrations were determined in blood spots using a glucose meter (FreestyleTM,
Abbott, the Netherlands) with inter- and intra-assay coefficients of variation (CV) of less than 6%
and 4%, respectively. Plasma concentrations of the thyroid hormones T3 and T4 were determined
by in-house RIA (14). Plasma TSH concentrations were determined by a chemiluminescent
immunoassay, using a rat-specific standard and plasma insulin, glucagon and corticosterone
concentrations were measured using commercially available kits (see supplementary information).
[6,6-2H2]-glucose enrichment was measured as described earlier (15).
proefschrift Klieverik.indb 67
4-8-2009 15:22:45
Statistics
68
Data were analyzed by analysis of variance (ANOVA) with repeated measures, with treatment
group (T3 or Veh) as between-animal factor and time (basal or after) as within-animal factor.
Paired sample and two-sample Student’s t-test were used as post hoc tests to determine where
time points within treatment groups and between treatment groups differed from each other,
respectively. Post hoc tests were performed if ANOVA revealed significance. Mann Whitney U
tests were used for analysis of Δ in time (before–after intervention) between groups. Spearman
correlation was used to test for associations between factors. Significance was defined at p≤0.05.
Data are presented as mean ± SEM.
Results
In Experiment #1, we infused euthyroid rats treated with methimazole and T4 from an osmotic
minipump (so-called block and replacement treatment) with either icv T3 (n=8) or vehicle (Veh,
n=7). In Experiment #2, we administered T3 or vehicle in the hypothalamic PVN via bilateral
microdialysis (MD, i.e. retro-dialysis) (Veh MD, n=7 vs T3 MD, n=9). In Experiment #3 we
performed PVN T3 MD in surgically hepatic sympatectomized animals (T3 MD HSx, n=8) and
sham-denervated animals (T3 MD Sham, n=6).
At time of central T3 administration, animals weighed between 320-360 g. In all experimental
groups, body weight increased during the last 3 days preceding central T3 administration,
indicating adequate recovery from surgery and a positive energy balance. There was no difference
in mean body weight of the treatment groups at time of central T3 administration in any of the
experiments described.
Experiment #1: icv T3 infusion
Icv T3 infused animals consumed an equal amount of food as compared with icv Veh infused rats
during the 24h following icv infusion (14.0±1.8 vs 13.6±1.2 g, respectively). Nevertheless, icv T3
infused animals lost weight in this time period as compared with icv Veh treated rats (-3.0±0.6
vs -1.1±0.4 % of body weight, respectively, p= 0.028).
Glucose, glucose kinetics, glucoregulatory hormones Mean basal glucose concentrations
were not significantly different between Veh icv and T3 icv groups (p=0.148, fig 1a). Basal EGP
was also similar between Veh and T3 icv treated groups (p=0.301, fig 1b). At t=24h after icv T3
infusion, there was a significant (p=0.027) increase in plasma glucose (28±8%), compared with a
non-significant 5±4% increase in Veh icv treated rats (Fig 1a). ANOVA revealed a trend (p=0.062)
for an increase in EGP in time, but no time*group effect. When analyzed separately, the EGP
increase 24h after an T3 icv bolus infusion almost reached significance as compared to the basal
state (p=0.057), but not so in Veh icv treated rats (p=0.482, fig 1b).
There were no differences in basal plasma insulin and corticosterone between the groups. In
both groups, plasma insulin and corticosterone 24h after icv infusion were not different from
the basal values.
proefschrift Klieverik.indb 68
4-8-2009 15:22:46
a
EGP (mmol/kg*min)
80
6.5
6.0
5.5
5.0
70
50
40
69
60
4.5
Bas
24h
T3 icv
(n=8)
Bas
24h
Veh icv
(n=7)
Bas
24h
T3 icv
(n=8)
Figure 1 a Mean plasma glucose concentration in icv vehicle (Veh) and T3 treated animals, before (basal) and
after (24h) bolus icv infusion. Note the marked, 22% plasma glucose increase in icv T3 treated rats, whereas in
Veh treated rats there was no significant effect on plasma glucose. ANOVA indicated p=0.003 for factor time
and p= 0.032 for factor time*group. *p< 0.01.
b Endogenous glucose production (EGP) before (Bas) and after (24h) icv T3 or Veh bolus infusion. EGP tended
to increase after icv T3 bolus infusion treated rats (^p=0.057), but not in icv Veh treated animals (p=0.482).
ANOVA indicated p= 0.062 for factor time.
Plasma thyroid hormones Basal plasma T3, thyroxine (T4), T3/T4 ratios and TSH concentrations
did not differ between the two treatment groups (Table 1). Surprisingly, the plasma T3
concentration in animals treated with methimazole and T4 was 18±5% higher 24h after icv
T3 infusion as compared with basal values (p=0.005). Vehicle treated animals showed a nonsignificant decrease in plasma T3. By contrast, plasma T4 concentrations showed a significant
28±5% decrease in icv T3 treated rats (p=0.002), and a non-significant 15±9% decrease in vehicle
treated animals. The plasma T3/T4 ratio increased by 65±7% 24h after icv T3 whereas it did not
change in Veh-treated rats (icv T3 vs Veh p=0.008). Plasma TSH did not differ between groups
and did not change in time. Five hours after the icv T3 infusion, there was an increase in plasma
T3 to values above the reference range for euthyroid animals (icv T3 4.45±0.48 nmol/L vs icv Veh
Hypothalamic T3 modulates hepatic glucose production
Bas
24h
Veh icv
(n=7)
^
Chapter 4
Glucose (mmol/L)
b
*
7.0
Table 1: Plasma hormone concentrations before (Basal) and after (24h) icv vehicle and T3 infusion
Veh icv n = 7
T3 icv n=8
Basal
24h
Basal
24h
T3 (nmol/L)
1,25 ± 0,19
0,92 ± 0,04
1,21 ± 0,12
1,40 ± 0.12 *
T4 (nmol/L)
154 ± 13
128 ± 12
149 ± 11
106 ± 8 *
T3/T4 (%)
0,87 ± 0,18
0,77 ± 0,10
0,82 ± 0,07
1,35 ± 0,12 *
TSH (mU/L)
0,37 ± 0,10
0,26 ± 0,02
0,29 ± 0,06
0,24 ± 0,03
Insulin (pmol/L)
290 ± 47
351 ± 55
295 ± 46
392 ± 67
78 ± 39
150 ± 60
181 ± 45
162 ± 49
Corticosterone (ng/mL)
* p< 0.05 vs Basal value within the same group
proefschrift Klieverik.indb 69
4-8-2009 15:22:46
1.32±0.36 nmol/L, p< 0.0001). In order to replicate the effects of centrally administered T3 on
glucose metabolism using a refined approach and to identify the brain area where T3 exerts its
effect on glucose metabolism we applied T3 locally in the hypothalamus by MD in Experiment #2.
Experiment #2: T3 MD in the hypothalamic PVN
70
Glucose, glucose kinetics and glucoregulatory hormones Mean basal glucose was 5.5±0.2
mmol/L in Veh MD and 5.0±0.1 mmol/L in T3 MD groups (p=0.030). Mean basal EGP was not
different between Veh and T3 MD groups. T3 MD induced a pronounced increase in plasma
glucose concentration (fig 2a), which was significantly larger than that in Veh MD rats (p=0.004,
fig 3a). After 2h of Veh MD, there was a 5.1±7.7% decrease in EGP. In contrast, after 2h
a
*
6.5
6.0
5.5
5.0
4.5
Bas
2h
Veh MD
(n=7)
Bas
T 3 MD
(n=9)
"
60
*
EGP (mmol/kg*min)
Glucose (mmol/L)
b
**
7.0
2h
*
55
50
45
40
Bas
2h
Veh MD
(n=7)
Bas
2h
T3 MD
(n=9)
Figure 2a Mean plasma glucose concentration in intra-hypothalamic vehicle (Veh MD) and T3 microdialysis (T3
MD) treated rats, before (Bas) and after (2h) microdialysis. Note the pronounced increase in T3 MD treated
rats, as compared to the mild increase in Veh MD treated rats. ANOVA indicated p<0.0001 for factor time and
p=0.005 for factor time*group. *p<0.05, **p<0.0001.
b Endogenous glucose production (EGP) before (Bas) and after (2h) T3 or vehicle microdialysis in the
hypothalamic PVN. Note the EGP increase after 2h of T3 microdialysis, in contrast to the EGP decrease in Veh
microdialysis treated animals, with no difference in basal EGP between groups. ANOVA revealed p= 0.029 for
factor time*group. “p<0.01, *p< 0.05.
of T3 MD there was a significant 10.7±3.7% increase in EGP relative to basal values (ANOVA
(Time*Group) p=0.029, fig 2b). The basal glucose concentration was no determinant of the
plasma glucose or EGP response to T3 MD (Spearman correlation p=0.546 and p=0.406 for basal
glucose concentration vs relative plasma glucose and EGP increase, respectively). In addition,
when the animals in the T3 MD group that had lower basal glucose concentrations than the Vehtreated animal with the lowest basal glucose value were excluded from the analysis, the relative
increase in plasma glucose was still significantly higher in T3 MD as compared with Veh MD
animals (n=5 vs n=7,respectively, p=0.004). Plasma glucagon showed a trend towards a decrease
in veh treated rats (Veh basal vs after -13±5%, p=0.058), whereas it showed a non-significant
9±6% increase in veh T3 treated rats (ANOVA (Time*Group) p=0.023). However, the glucagon
proefschrift Klieverik.indb 70
4-8-2009 15:22:46
Table 2: Plasma hormone concentrations before (Basal) and after (2h) vehicle and T3 microdialysis (MD)
T3 MD n=9
Veh MD n = 7
T4 (nmol/L)
2h
1,11 ± 0,05
1,01 ± 0,10
Basal
1,13 ± 0,06
2h
1,14 ± 0,14
54 ± 6 *
60 ± 4
35 ± 4 * “
1,60 ± 0,20
2,17 ± 0,49
1,92 ± 0,13
3,35 ± 0,50 *
Insulin (pmol/L)
291 ± 70
271 ± 65
247 ± 19
277 ± 28
92 ± 7
99 ± 8
174 ± 46
107 ± 18
Glucagon (pg/mL)
Corticosterone (ng/mL)
97 ± 7
126 ± 71
85 ± 9 ^
120 ± 29
* p<0.05 vs Basal value within the same group, “ p<0.05 vs Veh 2h,
^ p=0.058 vs Bas (ANOVA factor Time*Group p=0.023).
changes were not a determinant of EGP changes in either group (Veh MD r=-0.39, p=0.40, T3 MD
r=0.37, p=0.34). There were no differences in basal plasma glucagon, insulin and corticosterone
between groups. In both groups, plasma insulin and corticosterone after 2h of MD did not differ
from the basal values (table 2).
Plasma thyroid hormones Plasma T3 and T4 concentrations and T3/T4 ratios are depicted in
table 2. There were no differences in basal plasma T3 concentrations between Veh and T3 MD
groups, and T3 concentrations did not change after MD. Of note, there was no increase in
plasma T3 concentration after 2h of T3 MD. Plasma T4 was lower in T3 treated rats as compared
with vehicle after 2h of MD. Basal plasma T4 concentrations were not significantly different
between T3 MD and Veh MD rats. The plasma T3/T4 ratio increased after 2h of T3 MD (p=0.021),
but did not alter after Veh MD.
71
Hypothalamic T3 modulates hepatic glucose production
74 ± 6
T3/T4 (%)
Chapter 4
T3 (nmol/L)
Basal
Experiment #3: PVN T3 MD in selective hepatic sympathectomized and shamdenervated rats
To study the role of sympathetic projections from the hypothalamus to the liver in the observed
effects on EGP, we performed T3 PVN MD in animals that had undergone either a selective
surgical sympathetic denervation (T3 MD HSx, n=8) or a sham denervation (T3 MD Sham, n=6)
of the liver. Hepatically sympatectomized animals showed a significant 85.5% reduction in NA
content as compared to sham denervated rats, with no overlap between the groups (T3 MD
Sham 38.9±5.9 ng/g, T3 MD HSx 5.6±0.9 ng/g, p=0.002). This decrease in NA content was
similar in previous reports by our group involving selective hepatic sympathectomy (3).
Glucose, glucose kinetics, plasma thyroid hormones and glucoregulatory hormones Basal
plasma glucose and basal EGP were not different between both groups, which is in line with
our previous data showing no effect of sympathectomy on (basal) EGP in euthyroid rats (3). In
sham-denervated rats, T3 MD induced a 19±3% (p<0.0001) increase in plasma glucose. In HSx
rats, plasma glucose increased by 18±3% (p=0.002) after 2h of T3 MD (ANOVA (Time) p<0,0001,
(Time*Group) p=0,889, fig 4a). EGP increased by 9,9±5,0% in sham-denervated rats after 2h
T3 MD, very similar to earlier results in intact hypothalamic T3-treated animals in Experiment #2
(Fig 3b). In contrast, HSx animals showed an EGP decrease of 5,6±7,2% upon hypothalamic T3
proefschrift Klieverik.indb 71
4-8-2009 15:22:46
a
40
*
35
*
∆ Glucose (%)
30
72
25
20
15
10
5
0
Veh
T3
Veh
ICV
∆ EGP (%)
b
55
50
45
40
35
30
25
20
15
10
5
0
-5
-10
-15
-20
T3
MD
HSx
T3
Sham
MD
*
*
Veh
ICV
T3
Veh
T3
MD
HSx
Figure 3a Relative difference between
basal plasma glucose and plasma
glucose after (Δ Glucose (%)) (i) in icv
Veh or T3 treated rats, (ii) in rats treated
with vehicle (Veh MD) or T3 microdialysis
(T3 MD) in the hypothalamic PVN, (iii)
hypothalamic T3 microdialysis in selective
hepatic sympathectomized (T3 MD HSx)
or Sham-denervated (T3 MD Sham) rats.
Note the significant increase of plasma
glucose in icv T3 and T3 microdialysis
treated rats relative to their respective
Veh controls. Hepatic sympathectomy
did not abolish the plasma glucose
increase upon T3 microdialysis. *p<0.05
vs vehicle control group.
b Relative difference between basal EGP
and EGP after (Δ EGP (%)) (i) in icv Veh
or T3 treated rats, (ii) in rats treated with
vehicle (Veh MD) or T3 microdialysis
(T3 MD) in the hypothalamic PVN,
(iii) hypothalamic T3 microdialysis in
selective hepatic sympathectomized
(T3 MD HSx) or Sham-denervated (T3
MD Sham) rats. Note that the increase
of EGP in response to T3 microdialysis
relative to Veh treated rats, replicated
by T3 microdialysis in Sham-denervated
animals, is totally prevented by selective
hepatic sympathectomy. *p≤0.05 Veh
MD vs T3 MD and T3 MD Sham vs T3
MD HSx.
Sham
T3 MD
MD, similar to the EGP decrease following vehicle MD in Experiment #2. The increase in the
T3-treated intact (Exp 2) or T3-treated sham-denervated (Exp 3) animals differed significantly
from the decrease seen in the vehicle-treated intact (Exp 2) or T3-treated HSx (Exp 3) animals,
respectively ( p≤0.05, fig 3b).
There were no differences in basal plasma insulin and glucagon between the Sham-denervated
and HSx groups, and plasma insulin and glucagon did not change after 2h of T3 MD (table 3).
Plasma thyroid hormones Plasma T3 decreased significantly by 31±3% in HSx animals after
T3 MD (p<0.0001), while T3 MD had no effect on plasma T3 in Sham denervated rats (table 3).
Plasma T4 decreased to a similar extent in both groups after T3 MD (-36.4±4.7% T3 MD Sham vs
-45.9±4.2% T3 MD HSx). The T3/T4 ratio was higher after T3 MD compared with basal values in
Sham-denervated (p=0.012), but not in HSx animals (table 3).
chapter 4 nieuw.indd 72
6-8-2009 11:05:48
Table 3: Plasma hormone concentrations before (Basal) and after (2h) T3 microdialysis in sham-denervated
(T3 MD Sham) and hepatic sympathectomized rats (T3 MD HSx)
T3 MD Sham n = 8
Basal
1,17 ± 0,08
T4 (nmol/L)
79 ± 5
T3/T4 (%)
Insulin (pmol/L)
Basal
2h
1,25 ± 0,04
0,87 ± 0,08 *
50 ± 4 *
76 ± 6
41 ± 4 *
1,48 ± 0,07
2,24 ± 0,26 *
1,71 ± 0,12
2,26 ± 0,31
181 ± 20
211 ± 40
203 ± 31
189 ± 37
60 ± 5
70 ± 8
69 ± 9
57 ± 9
* p<0.05 vs Basal value within the same group
Discussion
The principal finding of this study is that T3 administered to the hypothalamic PVN in euthyroid
rats rapidly increases EGP, with a concomitant increase in plasma glucose concentration. An
intact sympathetic input to the liver is essential for the hypothalamic effect of T3 on EGP to
occur. Moreover, the T3-induced effects occur independently of plasma glucoregulatory hormone
concentrations.
The first indication that the thyrotoxicosis-associated increase in EGP and concomitant
hyperglycemia can be mimicked by central T3 administration in euthyroid rats came from our
experiments involving icv T3 infusion. However, these data were not conclusive as 5 h after
central T3 infusion plasma T3 concentrations increased above the euthyroid reference range.
Thus, a causal relation between the plasma T3 increase after 5 h and the metabolic alterations
after 24 h could not be excluded, in spite of the fact that plasma T3 had almost returned to basal
values after 24 hours. We decided to use bilateral microdialysis (MD), which enables precise
local administration within the hypothalamus and thereby offers detailed neuro-anatomical
information, to confirm our hypothesis that T3 can modulate hepatic glucose production via
actions in the hypothalamic PVN. The hypothalamic PVN not only harbours hypophysiotropic
neurons projecting to the median eminence, but also contains pre-autonomic neurons
controlling autonomic projections to the liver (16). The increase in EGP and plasma glucose
upon administration of T3 in the PVN was independent of plasma T3, insulin and corticosterone
concentrations. Plasma glucagon showed a small increase in response to hypothalamic T3 relative
to vehicle treatment. This effect on plasma glucagon may point to an effect of hypothalamic T3
on the endocrine pancreas. However, its small magnitude and the lack of correlation between
the glucagon and EGP changes exclude that the glucagon changes are responsible to a significant
extent for the observed EGP increase. Taken together, the observations are compatible with a
neural (autonomic) modulation of hepatic glucose metabolism by hypothalamic T3. Indeed, we
confirmed our hypothesis that hypothalamic T3 modulates EGP via sympathetic projections to the
liver by demonstrating that the hypothalamic T3-induced EGP increase can be totally prevented by
prior surgical selective hepatic sympathetic denervation. In addition, this denervation experiment
confirmed that the T3-induced changes in glucagon release are not the main determinant of the
changes in EGP.
proefschrift Klieverik.indb 73
73
Hypothalamic T3 modulates hepatic glucose production
Glucagon (pg/mL)
2h
1,08 ± 0,1
Chapter 4
T3 (nmol/L)
T3 MD HSx n=6
4-8-2009 15:22:47
74
The hypothalamic PVN contains many hypophysiotropic TRH neurons, projecting to the median
eminence and regulating the hypothalamo-pituitary-thyroid (HPT) axis. Hypothalamic T3 treatment
may cause a down-regulation of TRH gene expression in these neurons, in turn inducing decreased
thyroidal T4 and T3 secretion as a reflection of central hypothyroidism (17). Our MD experiments
lasted for 2h, which may be too rapid for modulation of TRH gene transcription, pituitary TSH
release, and thyroid hormone secretion. In addition, central hypothyroidism induced by central
T3 administration would be expected to cause opposite changes in glucose metabolism, i.e.
decreased EGP and glucose concentration (18).
It has been documented extensively that during cold stress, sympathetic stimulation of brown
adipose tissue increases local T3 availability via activation of deiodinase type 2 (D2) (19).
Deiodinase type 1 (D1) is the principal hepatic thyroid hormone deiodinating enzyme and is a
major contributor to T3 production in the rat (20). β-adrenergic blockers such as propanolol are
widely used in the initial clinical management of hyperthyroid patients, in part because these drugs
inhibit T4 to T3 conversion on the hepatic level (21). However, it is unknown if hepatic D1 activity
is neurally regulated. Interestingly, in the present study icv T3 administration decreased plasma
T4, whereas plasma T3 was elevated after 24h. Given that these experiments were performed
in rats treated with methimazole and thyroxine, these changes occurred independently from
thyroidal TH secretion. This raises the interesting possibility of a central T3 effect on hepatic
deiodinating activity. Moreover, hypothalamic T3 administration for 2h increased the plasma T3/
T4 ratio as compared with Veh treatment, which was also the case after hypothalamic T3 in shamdenervated rats, but not in rats that underwent prior selective hepatic sympathetic denervation.
Collectively, these findings are compatible with the concept of sympathetic stimulation of T4
to T3 conversion by hepatic D1. By inference, we might speculate that sympathetic stimulation
of hepatic T4 to T3 conversion could be partly responsible for the increase in EGP following
hypothalamic T3 administration, which will be the subject of further study.
Although the observed weight loss in icv T3 treated rats in the 24h following icv infusion may
be compatible with increased energy expenditure by T3, we were surprised to find that icv
T3 administration did not affect food intake in the 24h following icv infusion as compared
with vehicle treated rats. Recent studies by Kong et al. (22) involving local intrahypothalamic
T3 administration provided evidence that the hypothalamic ventromedial nucleus (VMN) is a
key nucleus for the orexigenic effects of T3. Although it is known that thyroid hormone bioavailability in the CNS is strongly regulated by deiodinases (in particular D2) (23), little is known
about thyroid hormone transport mechanisms between the ventricular system and specific
hypothalamic nuclei (24). Consequently, the effect of icv T3 bolus infusion on local T3 tissue
concentrations in the VMN or in other hypothalamic nuclei (and, thereby, on eating behavior) is
very difficult to predict at present.
The rapid time scale of the effects of intra-hypothalamic T3 administration on glucose metabolism
in itself fits with neural signalling from the hypothalamus to the liver via autonomic (sympathetic)
efferents, whereas at first sight it may be hard to reconcile with thyroid hormone receptor (TR)mediated effects on gene transcription and translation (25). Recently, an increasing number
of rapid, so called “non genomic” thyroid hormone effects have been reported. These may be
mediated by TRs, for example via interaction of TR subtype α1 (TRα1) with the phosphatidylinositol
proefschrift Klieverik.indb 74
4-8-2009 15:22:47
Chapter 4
75
Hypothalamic T3 modulates hepatic glucose production
3-kinase/protein kinase Akt (PI3K/Akt) pathway (26), which is a critical downstream target
of insulin signal-transduction in hypothalamic neurons regulating EGP (7;27). Alternatively,
membrane-bound receptors have emerged as high-affinity T3 binding sites that could mediate
these rapid effects via non-transcriptional mechanisms (28).
In the present study, we demonstrate that the EGP increase induced by hypothalamic T3
administration is mediated via altered sympathetic outflow to the liver. Recent studies in mice
have shown that suppression of TRα1 signalling via a mutation causing a 10-fold lower affinity for
T3 enhances basal metabolism. This appeared to be mediated via increased sympathetic tone to
brown adipose tissue, overriding the peripheral actions of the receptor (29). These observations
suggested an important role for TRα1 in regulating sympathetic outflow from the hypothalamus.
In contrast, the notion of increased sympathetic tone during thyrotoxicosis is not supported by
experiments in β-adrenergic knockout mice focussing on cardiac physiology and metabolic rate
(30). However, recent studies in patients with hyperthyroidism did show increased sympathetic
tone in subcutaneous adipose tissue (31), increased sympathetic and decreased parasympathetic
tone to the heart (32;33) as well as increased urinary catecholamine excretion (32;34), pointing
to increased sympathetic activity during thyrotoxicosis in humans. Finally, the present findings
are in line with previously reported observations from our group that the thyrotoxicosis-induced
changes in (hepatic) glucose metabolism can be differentially modulated by either selective
sympathetic or parasympathetic denervation of the liver (3).
Our finding that hepatic sympathectomy prevents the EGP increase, but not the plasma glucose
increase induced by hypothalamic T3 points to effects on glucose metabolism other than via EGP
in sympathectomized animals. Decreased peripheral glucose uptake is one of the possibilities,
perhaps mediated via autonomic input to major glucose disposing tissues like striated muscle and
white adipose tissue (35).
We conclude that stimulation of T3-sensitive neurons in the PVN of euthyroid rats increases EGP
via sympathetic projections to the liver, independently of circulating glucoregulatory hormone
concentrations. Thus, we report a novel central pathway for modulation of hepatic glucose
production by T3 involving the hypothalamic PVN and the sympathetic nervous system.
Acknowledgements
We wish to thank E.M. Johannesma-Brian and A.F.C. Ruiter for performing the hormone and
isotope analyses. The Ludgardine Bouwman-foundation is kindly acknowledged for financial
support.
Reference List
1. Franklyn JA. 2000 Metabolic changes in thyrotoxicosis. In: The Thyroid, a clinical and fundamental text.
8:667-672. Lippincott Williams & Wilkins, Philadelphia.
2. Dimitriadis GD, Raptis SA 2001 Thyroid hormone excess and glucose intolerance. Exp Clin Endocrinol
Diabetes 109 Suppl 2:S225-S239
3. Klieverik L, Sauerwein HP, Ackermans M, Boelen A, Kalsbeek A, Fliers E 2008 Effects of Thyrotoxicosis
and Selective Hepatic Autonomic Denervation on Hepatic Glucose Metabolism in Rats. Am J Physiol
Endocrinol Metab 294:E513-520
proefschrift Klieverik.indb 75
4-8-2009 15:22:47
76
4. Prodi E, Obici S 2006 Minireview: the brain as a molecular target for diabetic therapy. Endocrinology
147:2664-2669
5. Clegg DJ, Brown LM, Woods SC, Benoit SC 2006 Gonadal hormones determine sensitivity to central
leptin and insulin. Diabetes 55:978-987
6. Cusin I, Rouru J, Rohner-Jeanrenaud F 2001 Intracerebroventricular glucocorticoid infusion in normal
rats: induction of parasympathetic-mediated obesity and insulin resistance. Obes Res 9:401-406
7. Obici S, Zhang BB, Karkanias G, Rossetti L 2002 Hypothalamic insulin signaling is required for inhibition
of glucose production. Nat Med 8:1376-1382
8. Obici S, Feng Z, Karkanias G, Baskin DG, Rossetti L 2002 Decreasing hypothalamic insulin receptors
causes hyperphagia and insulin resistance in rats. Nat Neurosci 5:566-572
9. Kalsbeek A, la Fleur FS, van Heijningen HC, Buijs RM 2004 Suprachiasmatic GABAergic inputs to the
paraventricular nucleus control plasma glucose concentrations in the rat via sympathetic innervation of
the liver. J Neurosci 24:7604-7613
10. Alkemade A, Vuijst CL, Unmehopa UA, Bakker O, Vennstrom B, Wiersinga WM, Swaab DF, Fliers E
2005 Thyroid hormone receptor expression in the human hypothalamus and anterior pituitary. J Clin
Endocrinol Metab 90:904-912
11. Lechan RM, Qi Y, Jackson IM, Mahdavi V 1994 Identification of thyroid hormone receptor isoforms
in thyrotropin-releasing hormone neurons of the hypothalamic paraventricular nucleus. Endocrinology
135:92-100
12. Goldman M, Dratman MB, Crutchfield FL, Jennings AS, Maruniak JA, Gibbons R 1985 Intrathecal
triiodothyronine administration causes greater heart rate stimulation in hypothyroid rats than
intravenously delivered hormone. Evidence for a central nervous system site of thyroid hormone action.
J Clin Invest 76:1622-1625
13. Steele R 1959 Influences of glucose loading and of injected insulin on hepatic glucose output. Ann N Y
Acad Sci 82:420-430
14. Kalsbeek A, Fliers E, Franke AN, Wortel J, Buijs RM 2000 Functional connections between the
suprachiasmatic nucleus and the thyroid gland as revealed by lesioning and viral tracing techniques in
the rat. Endocrinology 141:3832-3841
15. Ackermans MT, Pereira Arias AM, Bisschop PH, Endert E, Sauerwein HP, Romijn JA 2001 The quantification
of gluconeogenesis in healthy men by (2)H2O and [2-(13)C]glycerol yields different results: rates of
gluconeogenesis in healthy men measured with (2)H2O are higher than those measured with [2-(13)C]
glycerol. J Clin Endocrinol Metab 86:2220-2226
16. La Fleur SE, Kalsbeek A, Wortel J, Buijs RM 2000 Polysynaptic neural pathways between the hypothalamus,
including the suprachiasmatic nucleus, and the liver. Brain Res 871:50-56
17. Segerson TP, Kauer J, Wolfe HC, Mobtaker H, Wu P, Jackson IM, Lechan RM 1987 Thyroid hormone
regulates TRH biosynthesis in the paraventricular nucleus of the rat hypothalamus. Science 238:78-80
18. Okajima F, Ui M 1979 Metabolism of glucose in hyper- and hypo-thyroid rats in vivo. Glucose-turnover
values and futile-cycle activities obtained with 14C- and 3H-labelled glucose. Biochem J 182:565-575
19. Silva JE 2006 Thermogenic mechanisms and their hormonal regulation. Physiol Rev 86:435-464
20. Nguyen TT, Chapa F, DiStefano JJ, III 1998 Direct measurement of the contributions of type I and type
II 5’-deiodinases to whole body steady state 3,5,3’-triiodothyronine production from thyroxine in the rat.
Endocrinology 139:4626-4633
21. Wiersinga WM 1991 Propranolol and thyroid hormone metabolism. Thyroid 1:273-277
22. Kong WM, Martin NM, Smith KL, Gardiner JV, Connoley IP, Stephens DA, Dhillo WS, Ghatei MA, Small
CJ, Bloom SR 2004 Triiodothyronine stimulates food intake via the hypothalamic ventromedial nucleus
independent of changes in energy expenditure. Endocrinology 145:5252-5258
proefschrift Klieverik.indb 76
4-8-2009 15:22:47
23. Bernal J 2002 Action of thyroid hormone in brain. J Endocrinol Invest 25:268-288
24. Alkemade A, Friesema EC, Unmehopa UA, Fabriek BO, Kuiper GG, Leonard JL, Wiersinga WM, Swaab
DF, Visser TJ, Fliers E 2005 Neuroanatomical pathways for thyroid hormone feedback in the human
hypothalamus. J Clin Endocrinol Metab 90:4322-4334
26. Hiroi Y, Kim HH, Ying H, Furuya F, Huang Z, Simoncini T, Noma K, Ueki K, Nguyen NH, Scanlan TS,
Moskowitz MA, Cheng SY, Liao JK 2006 Rapid nongenomic actions of thyroid hormone. Proc Natl Acad
Sci USA 103:14104-14109
28. Davis PJ, Leonard JL, Davis FB 2008 Mechanisms of nongenomic actions of thyroid hormone. Front
Neuroendocrinol 29:211-218
29. Sjogren M, Alkemade A, Mittag J, Nordstrom K, Katz A, Rozell B, Westerblad H, Arner A, Vennstrom B
2007 Hypermetabolism in mice caused by the central action of an unliganded thyroid hormone receptor
alpha1. EMBO J 145:2767-2774
30. Bachman ES, Hampton TG, Dhillon H, Amende I, Wang J, Morgan JP, Hollenberg AN 2004 The metabolic
and cardiovascular effects of hyperthyroidism are largely independent of beta-adrenergic stimulation.
Endocrinology 145:2767-2774
31. Haluzik M, Nedvidkova J, Bartak V, Dostalova I, Vlcek P, Racek P, Taus M, Svacina S, Alesci S, Pacak
K 2003 Effects of hypo- and hyperthyroidism on noradrenergic activity and glycerol concentrations in
human subcutaneous abdominal adipose tissue assessed with microdialysis. J Clin Endocrinol Metab
88:5605-5608
32. Burggraaf J, Tulen JH, Lalezari S, Schoemaker RC, De Meyer PH, Meinders AE, Cohen AF, Pijl H 2001
Sympathovagal imbalance in hyperthyroidism. Am J Physiol Endocrinol Metab 281:E190-E195
77
Hypothalamic T3 modulates hepatic glucose production
27. Niswender KD, Morrison CD, Clegg DJ, Olson R, Baskin DG, Myers MG, Jr., Seeley RJ, Schwartz MW
2003 Insulin activation of phosphatidylinositol 3-kinase in the hypothalamic arcuate nucleus: a key
mediator of insulin-induced anorexia. Diabetes 52:227-231
Chapter 4
25. Yen PM 2001 Physiological and molecular basis of thyroid hormone action. Physiol Rev 81:1097-1142
33. Cacciatori V, Bellavere F, Pezzarossa A, Dellera A, Gemma ML, Thomaseth K, Castello R, Moghetti P,
Muggeo M 1996 Power spectral analysis of heart rate in hyperthyroidism. J Clin Endocrinol Metab
81:2828-2835
34. Eustatia-Rutten CF, Corssmit EP, Heemstra KA, Smit JW, Schoemaker RC, Romijn JA, Burggraaf J 2008
Autonomic nervous system function in chronic exogenous subclinical thyrotoxicosis and the effect of
restoring euthyroidism. J Clin Endocrinol Metab 93:2835-2841
35. Kreier F, Kap YS, Mettenleiter TC, van Heijningen C, van d, V, Kalsbeek A, Sauerwein HP, Fliers E, Romijn
JA, Buijs RM 2005 Tracing from fat tissue, liver and pancreas: A neuroanatomical framework for the role
of the brain in type 2 diabetes. Endocrinology 147:1140-1147
proefschrift Klieverik.indb 77
4-8-2009 15:22:47
proefschrift Klieverik.indb 78
4-8-2009 15:22:50
5
Central effects of
thyronamines on glucose
metabolism in rats
Lars P. Klieverik
Ewout Foppen
Mariëtte T. Ackermans
Mireille J. Serlie
Hans P. Sauerwein
Thomas S. Scanlan
David K. Grandy
Eric Fliers
Andries Kalsbeek
Journal of Endocrinology 2009: 201(3), 377-386
proefschrift Klieverik.indb 79
4-8-2009 15:22:53
Abstract
80
Thyronamines are naturally occurring, chemical relatives of thyroid hormone. Systemic
administration of synthetic 3-iodothyronamine (T1AM) and - to a lesser extent - thyronamine
(T0AM), leads to acute bradycardia, hypothermia, decreased metabolic rate and hyperglycemia.
This profile led us to hypothesize that the central nervous system is among the principal targets
of thyronamines.
We investigated whether a low dose intracerebroventricular (icv) infusion of synthetic thyronamines
recapitulates the changes in glucose metabolism that occur following intraperitoneal (ip)
thyronamine administration.
Plasma glucose, glucoregulatory hormones and endogenous glucose production (EGP) using
stable isotope dilution were monitored in rats before and 120 min after an ip (50 mg/kg) or icv
(0.5 mg/kg) bolus infusion of T1AM, T0AM, or vehicle. To identify peripheral effects of centrally
administered thyronamines, drug-naïve rats were also infused intravenously with low dose (0.5
mg/kg) thyronamines.
Systemic T1AM rapidly increased EGP and plasma glucose, increased plasma glucagon, and
corticosterone, but failed to change plasma insulin. Compared to ip-administered T1AM, a
100-fold lower dose administered centrally induced a more pronounced acute EGP increase
and hyperglucagonemia while plasma insulin tended to decrease. Both systemic and central
infusions of T0AM caused smaller increases in EGP, plasma glucose and glucagon compared with
T1AM. Neither T1AM nor T0AM influenced any of these parameters upon low dose intravenous
administration.
We conclude that central administration of low dose thyronamines suffices to induce the
acute alterations in glucoregulatory hormones and glucose metabolism following systemic
thyronamine infusion. Our data indicate that thyronamines can act centrally to modulate glucose
metabolism.
proefschrift Klieverik.indb 80
4-8-2009 15:22:54
Introduction
Chapter 5
81
Central thyronamines and glucose metabolism
Thyronamines are a group of naturally occurring, chemical relatives of thyroid hormone (TH)
with pronounced and rapid physiologic effects (1). Two representatives of the thyronamines,
3-iodothyronamine (T1AM) and thyronamine (T0AM) have been extracted from rat and mouse
brain, heart, liver and blood. T1AM and T0AM can theoretically be derived from iodothyronines
thyroxine (T4), 3,3’,5-triiodothyronine (T3) and/or 3,3’,5’-triiodothyronine (reverse T3) by removal
of the carboxylate group on the β-alanine side chain in addition to deiodination. Indeed,
thyronamines have recently been identified as iso-enzyme specific substrates of the iodothyronine
deiodinases type 1, 2 and 3 (2). T1AM and, to a lesser extent, T0AM are potent in vitro agonists
of the trace amine associated receptor type 1 (TAAR1) (1;3), a Gs protein-coupled membrane
receptor with a broad expression profile (4). In rodents and humans, high levels of TAAR1
expression are found in liver, kidney, gastrointestinal tract, pancreas, heart and many areas of
the brain (5;6). Moreover, T1AM has the potential to act as an adrenergic receptor α2 (ARα2)
agonist in the mouse, explaining in part the decrease in insulin secretion by pancreatic beta cells
exposed to thyronamines (7).
When administered to rodents, T1AM and T0AM have striking effects on physiology. Within
minutes after systemic administration, profound hypothermia, bradycardia and decreased cardiac
output occur. In addition, thyronamines rapidly induce metabolic alterations such as decreased
metabolic rate and a dramatic shift to preferential lipid fuelling at the cost of carbohydrate
oxidation (1;8). These apparently non-genomic effects are thought to occur via binding to and
activating membrane-bound G protein-coupled receptors (GPCRs) such as TAAR1 and ARα2
(1;9). Furthermore, it has been proposed but not yet demonstrated THs can be converted to
thyronamines by enzymatic deiodination and decarboxylation. Since most actions of T1AM
and T0AM are opposite in direction to the bio-active thyroid hormone T3, thyronamines have
been hypothesized to play a role in fine-tuning and/or antagonizing T3 actions on a moment-tomoment timescale (9;10).
The brain, in particular the hypothalamus, regulates most of the processes affected by
thyronamines (body temperature, cardiac function, energy metabolism). Moreover, a principal
role in regulating hepatic glucose metabolism has recently emerged for the hypothalamus (1113). As T1AM and T0AM are present in rat brain, we hypothesized these novel compounds could
affect glucose metabolism via actions in the central nervous system (CNS).
In the present study we tested the hypothesis that thyronamines act centrally to induce changes
in glucose metabolism using stable isotope dilution and 3 different routes of administration:
systemic (ip), central (intracerebroventricular; icv), and intravenous (iv) in rats. Our results are
consistent with the interpretation T1AM and T0AM can act centrally to recapitulate the changes
in glucose metabolism that occur following systemic thyronamine administration.
Materials and Methods
Animals
Male Wistar rats (Harlan, Horst, the Netherlands) between 350 and 400 g bodyweight (BW),
housed under constant conditions of temperature (21 ± 1 °C) and humidity (60 ± 2%) with a
proefschrift Klieverik.indb 81
4-8-2009 15:22:54
12-h light, 12-h dark (L/D) schedule (lights on at 7.00 h am), were used in all experiments. Food
and drinking water were available ad libitum. All of the following experiments were conducted
with the approval of the Animal Care Committee of the Royal Netherlands Academy of Arts and
Sciences.
T1AM and T0AM
82
T1AM-HCl (391 g/mol) and T0AM-HCl (264 g/mol) were synthesized as previously reported (3)
and dissolved in 20% DMSO and 80% saline (vehicle) at a concentration of 40 mg/mL.
Experimental groups
Two independent studies were performed For the first study, permanent jugular vein and
carotid artery cannulae were placed in rats (n=22) under anesthesia (see below). Animals were
allowed to recover from the surgery for 8 days prior to any further manipulations. Each rat
thus cannulated received an intraperitoneal (ip) bolus infusion of 50 mg/kg of T1AM (n=7),
50 mg/kg of T0AM (n=8), or an equal volume (500 μL) of vehicle (n=7). For the second study
rats (n=31) were equipped with a guide cannula placed into the left lateral cerebral ventricle
in addition to the carotid artery and jugular vein cannulae. Rats thus cannulated received an
intracerebroventricular (icv) 100-fold lower dose (0.5 mg/kg) of either T1AM (n=9), T0AM (n=8),
or DMSO-saline vehicle (n=8) in a volume of 4 μL. To control for the possibility any observed effect
of the icv-infused thyronamines was somehow due to spill-over into the circulation an additional
group of cannulated rats was infused intravenously (iv) with 0.5 mg/kg T1AM (n=3) and T0AM
(n=3) in a volume of 500 μL. In both of these experiments, before and 120 min after ip or icv
bolus infusion, isotope dilution and blood sampling were conducted to permit measurement
of endogenous glucose production (EGP), and the concentration of plasma glucose, insulin,
glucagon, corticosterone, thyroid stimulating hormone (TSH), T3 and T4 concentrations.
Surgery
Jugular, Carotid and icv cannulae Animals were anaesthetized using a mixture of Hypnorm
(Janssen; 0.05 mL/100 g BW, i.m.) and Dormicum (Roche, the Netherlands; 0.04 mL/100 g BW,
s.c.). Vascular and icv cannulae were fixed with dental cement to 4 stainless-steel screws inserted
into the skull. Post-operative care was provided by subcutaneous injection of 0.01 mL/100 g BW
of Temgesic (Schering-Plough; Utrecht, the Netherlands). In all animals an intra-atrial silicone
cannula was implanted through the right jugular vein and a second silicone cannula was placed
in the left carotid artery for isotope infusion and blood sampling as described previously (14). For
the second study, stainless steel icv cannulae were implanted into the left cerebral ventricle using
the following stereotaxic coordinates: anteroposterior: -0,8 mm, lateral: +2,0 mm, ventral: -3,2
mm, with the toothbar set at -3,4 mm. Guide cannula placement was confirmed by dye (4 μL of
ethylene blue) injection and inspection post-mortem. Only animals that showed staining of the
left lateral cerebral ventricle and third cerebral ventricle were included in the final analysis.
Stable isotope dilution and systemic vs central thyronamine
administration
General procedure Eight days post-surgery, stable isotope dilution was performed in combination
with the administration of synthetic thyronamines. Animals weighed between 335-380 g. Body
proefschrift Klieverik.indb 82
4-8-2009 15:22:54
After the t=0 bloodsample, in study 1, animals received an ip bolus of either T1AM, T0AM (50
mg/kg in 500 μL) or vehicle. In study 2, again after the t=0 bloodsample, animals received an
icv bolus infusion of either T1AM, T0AM (0.5 mg/kg in 4 μL) or vehicle delivered through the icv
cannula in 105 sec using a Hamilton syringe. After ip or icv bolus infusion, blood samples were
obtained for measurement of glucose concentration, isotopic enrichment (5, 10, 20, 30, 45,
60, 75, 90 and 120 min), plasma corticosterone (t=10, 20, 30, 60 and 120 min), plasma insulin,
glucagon (t= 10, 60 and 120 min) and plasma TSH, T3 and T4 concentrations (t= 120 min).
83
Central thyronamines and glucose metabolism
Bolus infusion of synthetic thyronamines
Chapter 5
weight (BW) increased in all groups during the 3 days preceding the experimental infusions,
indicating recovery from surgery and a positive energy balance. One day before the experimental
infusions, rats were connected to a metal collar to which polyethylene tubing (for blood-sampling
and infusion) was attached and kept out of reach of the animals by a counterbalanced beam. This
permitted all subsequent manipulations to be performed outside the cages without handling the
animals (14). For determining basal plasma concentrations of TSH, T3 and T4 a blood sample was
obtained at 14.00 PM. On the day of thyronamine administration, food was removed from the
cages 4 h (~8.30 AM) before the first basal measurements. At ~11.00 AM a blood sample was
taken (200 μL, t=-110 min) for determination of background isotopic enrichment. Subsequently,
a primed (8.0 μmol in 5 min) continuous (16.6 μmol/h) infusion of the stable isotope tracer
[6,6-2H2]-glucose (>99% enriched; Cambridge Isotope Laboratories, Cambridge, MA) was started
using an infusion pump (Harvard Apparatus, Holliston, MA). After an equilibration period of 90
min additional blood samples (200 μL) were obtained for the determination of basal plasma
glucose, isotopic enrichment (t= -20, -10 and 0 min), plasma corticosterone (t= -20 and 0 min),
insulin, and glucagon (t= 0 min) concentrations.
Plasma hormone and isotope analyses
Plasma glucose concentration was determined in triplicate by a glucose oxidase method
(Boehringer Mannheim, Germany). Plasma glucagon and corticosterone were measured using
a commercially available radioimmunoassay (RIA, LINCO research, st.Charles, MO and ICN
biomedicals, Costa Mesa, CA, respectively). Plasma concentrations of T3 and T4 were determined
by an in-house RIA (15), with inter- and intra-assay variation coefficient (CV) of 7–8% and 3–4%
(T3), and 3–6 and 2–4% (T4), respectively. Detection limits for T3 and T4 were 0.3 nmol/L and
5 nmol/L, respectively. Plasma TSH concentrations were determined by a chemiluminescent
immunoassay (Immulite 2000, Diagnostic Products Corp., Los Angeles, CA) using a rat-specific
standard. The inter- and intra-assay CV’s for TSH were less than 4% and 2% at ±3.5 mU/L,
respectively, with a detection limit of 0.40 mU/L. Plasma insulin was measured by a commercially
available Elisa (Mercodia, Uppsala, Sweden) (16). The inter- and intra-assay CV’s were 4% and
2%, detection limit 13 pmol/L. All samples were measured in duplicate, e.g. 2 tubes were
analysed per plasma sample. Glucose enrichment was measured as previously described (17).
The [6,6-2H2]-glucose enrichment (tracer/ tracee ratio) inter-assay CV was 1%, the intra-assay CV
1%, and the detection limit 0.04%.
proefschrift Klieverik.indb 83
4-8-2009 15:22:54
Calculations and statistical analysis
84
EGP was calculated from isotope enrichment and plasma glucose concentration using modified
forms of steady state (basal) and non-steady state (after thyronamine infusion) Steele equations
(18). Data were analyzed by two-way analysis of variance (ANOVA) with repeated measurements,
with treatment group (T1AM, T0AM or Veh) and time as dependent factors. Significance was
defined at p<0.05 using paired t-tests (i.e. within treatment groups) and independent t-tests (i.e.
between treatment groups) to identify experimental groups that differed significantly. The SPSS
statistical software program version 16.0 (SPSS Inc., Chicago, Illnois) was used for statistical
analysis. Data are presented as mean ± standard error of the mean (SEM).
Results
In two independent studies, 8 groups of rats were investigated. In the first study rats received an
intraperitoneal (ip) bolus infusion of either T1AM (50 mg/kg, n=7), T0AM (50 mg/kg, n=8) or vehicle
(n=7). In the second study, rats were intracerebroventricularly (icv) infused with a 100-fold lower
dose (i.e. 0.5 mg/kg) of either T1AM (n=9), T0AM (n=8) or vehicle (n=8). To address the possibility
that any physiologic response observed following icv-infusion of the thyronamines was due to the
peripheral action of drug that spilled over into the circulation, two additional groups of animals were
intravenously (iv) infused with 0.5 mg/kg of T1AM (n=3) or 0.5 mg/kg of T0AM (n=3).
Study #1: Systemic thyronamine infusion
Rats injected ip with 50 mg/kg T1AM or 50 mg/kg T0AM exhibited a behavioral phenotype as
described previously (1). Interestingly, the animals injected with T1AM displayed the more robust
phenotype even though the dose (on a per mole basis) was approximately half that of T0AM.
Glucose homeostasis – plasma concentration and endogenous production Systemic
infusion of either T1AM or T0AM by the ip route of administration induced a rapid and significant
increase in plasma glucose concentration (fig 1a). The onset and magnitude of this effect was
25.0
22.5
20.0
17.5
15.0
12.5
10.0
7.5
5.0
2.5
0.0
b
EGP (umol/kg*min)
Glucose (mmol/L)
a
Bolus ip
ANOVA p< 0.0001
-20 -10 0 10 20 30
45
60
Time (min)
75
90 105 120
110
100
T1 AM 50 mg/kg ip (n=7)
T0 AM 50 mg/kg ip (n=8)
Vehicle ip (n=7)
Bolus ip
90
80
70
60
50
40
30
ANOVA p< 0.001
-10 0 10 20 30
45
60
75
90 105 120
Time (min)
Fig 1a Plasma glucose concentrations before and after ip bolus infusion of T1AM, T0AM or vehicle. Note that
from t=10 min, glucose concentration is higher in T1AM and T0AM treated animals as compared with vehicle
rats (p<0.05). From t=60, glucose concentration is higher in T1AM as compared with T0AM treated rats (p<0.05).
ANOVA RM factor time p<0.0001, time*group p<0.0001, group p<0.0001.
b Endogenous glucose production (EGP) before and after ip bolus infusion of T1AM, T0AM or vehicle. From t=10
and t=20, EGP is higher in T1AM and T0AM infused rats, respectively, as compared with vehicle (p<0.05). From
t=60, EGP is lower in T0AM relative to T1AM treated animals (p<0.05). From t=90, EGP in T0AM treated rats is not
different from vehicle rats. ANOVA RM factor time p<0.0001, time*group p<0.0001, group p=0.001.
proefschrift Klieverik.indb 84
4-8-2009 15:22:54
Insulin (pmol/L)
400
Bolus ip
300
200
100
0
T1 AM 50 mg/kg ip (n=7)
T0 AM 50 mg/kg ip (n=8)
Vehicle ip (n=7)
0 10
60
b
Glucagon (pg/mL)
200
*
Bolus ip
*
*
150
100
50
0
0 10
60
c
Bolus ip
400
120
**
500
Corticosterone (ng/mL)
120
^
300
250
ANOVA ns
Fig 2a Plasma insulin concentration before (t=0) and
after intraperitoneal (ip) injection of T1AM, T0AM or
vehicle. ANOVA indicates no effects of time and no
differences in insulin concentration between groups
(ANOVA RM factor time p=0.556, time*group p=0.735,
group p=0.213).
b Plasma glucagon concentration before (t=0) and
after ip injection of T1AM, T0AM or vehicle. From
t=10, glucagon is higher in both T1AM and T0AM
treated rats relative to vehicle rats (*p<0.05 vs Veh).
At t=60, glucagon concentration is higher in T1AM as
compared with T0AM treated rats (^p=0.05 T0AM vs
T1AM). ANOVA RM factor time p<0.0001, time*group
p<0.0001, group p=0.001.
c Plasma corticosterone concentration before (t=-20
and t=0 min) and after (t=5 – t=120 min) ip injection
of T1AM, T0AM or vehicle. T0AM treated animals have
higher plasma corticosterone from t=30 and T1AM
treated animals only on t=60 as compared with vehicle
(*p<0.05 vs Veh). Note that at no time-point plasma
corticosterone between T1AM and T0AM injected
animals differs. ANOVA RM factor time p<0.0001,
time*group p=0.050, group p=0.030.
85
Central thyronamines and glucose metabolism
a
Chapter 5
similar for the two compounds until 45 min post infusion when the effect of T0AM apparently
plateaued while the T1AM-induced hyperglycemia continued to develop eventually reaching a
maximum 371±27% of basal values 120 min after infusion.
Within 10 minutes of receiving an ip bolus of T1AM the EGP increased to 143±3% of basal
values (p=0.001 vs Veh, fig 1b), that was sustained for the duration of the experiment. Similarly,
ip administration of T0AM rapidly increased EGP, reaching a maximum of 158±16% of basal
values after 20 min (p=0.032 vs Veh). Forty-five minutes after injection with T0AM EGP gradually
returned to basal values by t=120 minutes.
Glucoregulatory Hormones Given the profound effect of T1AM and T0AM on plasma glucose
and EGP we characterized the status of three glucoregulatory hormones: insulin, glucagon,
and corticosterone (fig 2). Surprisingly, even though T1AM and T0AM (50 mg/kg, ip) produced
*
*
300
200
100
0
proefschrift Klieverik.indb 85
-20 0 102030
60
Time (min)
120
4-8-2009 15:22:55
86
hyperglycemia and elevated EGP, plasma insulin concentrations were unchanged relative to
plasma from vehicle-injected rats (fig 2a). In contrast, plasma glucagon concentrations were
significantly increased within 10 minutes of either T1AM or T0AM administration (fig 2b).
However, by 60 minutes post injection, the time-effect profiles of the two compounds had begun
to diverge with T0AM’s effect reaching a plateau at 240% (p=0.007, T0AM vs Veh, t=60) of basal
levels while T1AM’s effect continued to develop for the duration of the experiment (447±44%
of basal values at 120 min, p<0.0001 T1AM vs Veh, fig 2b). Of note, the time-course profiles of
plasma glucagon (fig 2b) and plasma glucose (fig 1a) in response to ip T1AM and T0AM, were
essentially superimposable.
Plasma corticosterone displayed a significant increase in response to both T1AM and T0AM
injected ip, compared to vehicle-injected rats (fig 2c). T1AM infusion induced a maximal increase
at t=60 min (482±106% vs 179±71% of basal levels at t=60 min, T1AM vs Veh, p=0.022). T0AM
infusion increased plasma corticosterone to a similar extent (393±59% vs 172±102% of basal
levels at t=120 min, T0AM vs Veh, p=0.022). At no time point was there a difference in the
corticosterone response between T1AM and T0AM infused groups.
Plasma T3, T4 and TSH concentrations before and 120 min after ip T1AM, T0AM and vehicle infusion
are depicted in table 1. Within 120 min both T4 and TSH levels were significantly decreased in
response to ip T1AM or, to a larger extent, ip T0AM (50 mg/kg). Intriguingly, T3 levels were also
decreased 120 min after ip T0AM (50 mg/kg) as compared with vehicle injected rats.
Study #2: Central thyronamine infusion
Icv infusion of 0.5 mg/kg T1AM or T0AM did not induce any of the phenotypical alterations
observed after systemic thyronamine administration.
Glucose homeostasis – plasma concentration and endogenous production With an icv
bolus infusion of 0.5 mg/kg T1AM plasma glucose concentration began to increase immediately
(fig 3a) until it reached a maximum 199±13% of basal levels 30 min after infusion (p<0.0001 vs
Veh). During the next 90 min plasma glucose decreased slightly stabilizing at approximately 163%
of basal values. T0AM (0.5 mg/kg, icv) significantly elevated plasma glucose as well, but to a lesser
degree than T1AM (maximum 134±6% at t=45, p<0.0001 vs Veh, fig 3a).
T1AM (0.5 mg/kg, icv) induced a rapid and significant increase in EGP (fig 3b) by 10 minutes
post infusion reaching a maximum 178±16% of basal values at 30 min (p<0.0001 vs Veh) that
gradually decreased with time to 113±5% when the experiment was terminated at t=120 min.
Table 1: Plasma thyroid hormone concentrations before (Basal) and after (2h) intraperitoneal (ip) vehicle, T1AM
and T0AM infusion
T1AM ip n=7
Veh ip n = 7
Basal
2h
Basal
T3 (nmol/L)
0,82 ± 0,05
0,78 ± 0,05
0,88 ± 0,04
T4 (nmol/L)
86 ± 3
70 ± 4*
91 ± 4
48 ± 3* “
1,46 ± 0,27
2,03 ± 0,31
1,79 ± 0,25
1,09 ± 0.29”
TSH (mU/L)
2h
0,83 ± 0,05
* p<0.05 vs Basal value within the same group, ** p<0.05 vs Veh Basal, “ p<0.05 vs Veh 2h, ^ p<0.05 vs T1AM 2h
proefschrift Klieverik.indb 86
4-8-2009 15:22:55
a
b
EGP (umol/kg*min)
Bolus icv
12
10
8
6
4
2
T1AM 0.5 mg/kg icv
T0 AM 0.5 mg/kg icv
Vehicle icv
110
ANOVA p<0.0001
-20 -10 0 10 20 30
45
60
75
90
105 120
Bolus icv
100
90
80
Chapter 5
Glucose (mmol/L)
14
70
60
50
40
ANOVA p<0.0001
30
-20 -10 0 10 20 30
60
75
90
105 120
Time (min)
Fig 3a Plasma glucose concentration before and after intracerebroventricular (icv) bolus infusion of T1AM,
T0AM or vehicle. From t=5 and t=10, glucose concentration in T1AM and T0AM infused rats, respectively, is
higher compared with vehicle rats (p<0.05). From t=10, glucose concentration in T1AM infused rats is higher
relative to T0AM infused rats (p<0.05). ANOVA RM factor time p<0.0001, time*group p<0.0001, group
p<0.0001.
b Endogenous glucose production (EGP) before and after icv bolus infusion of T1AM, T0AM or vehicle. There is
no significant difference between basal samples of any group. From t=10, EGP in T1AM infused rats is higher
compared with vehicle rats (p<0.05). In T0AM rats, EGP at t=30, 45, 90, 105 and 120 min is higher relative to
vehicle infused rats (p<0.05). From t=20 to t=90, EGP is higher in T1AM relative to T0AM infused rats (p<0.05).
ANOVA RM factor time p<0.0001, time*group p<0.0001, group p=0.006.
Although T0AM (0.5 mg/kg, icv) significantly increased EGP above basal levels (p<0.0001 t=0
vs t=10), its maximum effect (20±9% increase at t=20, p=0.069 vs Veh) was approximately one
third of T1AM’s maximal effect (fig 3b).
Importantly, when T1AM or T0AM were infused intravenously (iv) at the dose that was used in
the icv infusion experiments (0.5 mg/kg, T1AM n=3, T0AM n=3)) neither thyronamine had any
effect on plasma glucose concentrations, EGP, plasma insulin or glucagon at any time point as
compared to basal values (data not shown).
When the absolute changes in plasma glucose concentration produced by 0.5 mg/kg T1AM, icv
and 50 mg/kg T1AM ip are compared over time, their profiles are practically super-imposable for
the first 30 min of exposure (fig 4a). Thereafter they diverged as the systemic effect of T1AM
continued to develop. Plotting the absolute values for EGP in response to 0.5 mg/kg T1AM icv
and 50 mg/kg T1AM ip revealed both routes of administration produced identical profiles during
T0AM ip n=8
ANOVA RM
Basal
2h
p
p
p
1,12 ± 0.07**
0,64 ± 0.07* ^
0,000
0,000
0,515
42 ± 2* “
0,000
0,002
0,000
0,028
0,004
0,014
80 ± 3
1,68 ± 0,16
proefschrift Klieverik.indb 87
87
Central thyronamines and glucose metabolism
Time (min)
45
0,40 ± 0.04* “ ^
Time Time*Group Group
4-8-2009 15:22:55
∆ Glucose (mmol/L)
88
b
T1AM 0.50 mg/kg icv (n=9)
T1AM 50 mg/kg ip (n=7)
Vehicle ip (n=7)
Vehicle icv (n=8)
20
15
*
*
Bolus
10
*
*
*
*
5
*
0
60
∆ EGP (µmol/kg*min)
a
50
40
*
Bolus
*
30
*
*
*
20
10
0
-10
-20
-5
-20 -10 0 10 20 30
45
60
75
-30
90 105 120
Time (min)
-20 -10 0 10 20 30
45
60
75
90 105 120
Time (min)
Fig 4a Changes in plasma glucose concentration (delta) before and after intraperitoneal (ip) vs intra
cerebroventricular (icv) bolus infusion of T1AM (T1AM ip, T1AM icv) or vehicle (Veh ip, Veh icv). From t=5,
glucose concentration in T1AM ip and T1AM icv groups is increased relative to the respective vehicle group.
Note that on t=20, 30 and 45 min, glucose concentration in T1AM ip and T1AM icv rats is not different. The
changes in plasma glucose concentration are expressed as the difference compared to the plasma glucose
concentration at t=0 min for each individual animal. ANOVA RM factor time p<0.0001, time*group p<0.0001,
group p<0.0001. *p<0.05; T1AM ip vs T1AM icv.
b Changes in endogenous glucose production (EGP) before and after intraperitoneal (ip) vs intracerebroventricular (icv) bolus infusion of T1AM (T1AM ip, T1AM icv) or vehicle (Veh ip, Veh icv). From t=5 and t=10 min,
EGP in T1AM icv and T1AM ip infused rats, respectively, is increased as compared with respective vehicle
groups (p<0.05). Note that on t=30, EGP is higher in T1AM icv than in T1AM ip treated rats (*p<0.05). The
changes in EGP are expressed as the difference compared to EGP at t=0 min for each individual animal.
ANOVA RM factor time p<0.0001, time*group p<0.0001, group p<0.0001.
the initial 20 min post exposure (fig 4b). However, at 30 min thereafter the magnitude of the
EGP effect elicited by T1AM icv was significantly greater than the effect of T1AM ip.(fig 4b).
Glucoregulatory Hormones Similar to T1AM ip (fig 2a), neither T1AM or T0AM administered
icv induced a significant change in plasma insulin content (fig 5a). Although there was a trend
for insulin to decrease 10 min after icv infusion of T1AM this response failed to achieve statistical
significance (p=0.063, fig 5a).
Plasma glucagon increased by 155% (from 69±10 to 176±20 pg/mL) 10 min after icv T1AM
infusion (p<0.0001 vs Veh). During the same time period T0AM icv also significantly increased
plasma glucagon but only by 58% (p=0.004 vs Veh). Interestingly the magnitude of T1AM ‘s
impact on circulating glucagon levels was dependent on the route of administration with icv
infusion producing a greater effect in the first 10 min than ip administration (108±21 vs 50±11
pg/mL, respectively, p=0.044). Unlike the sustained elevation that followed ip administration of
Table 2: Plasma thyroid hormone concentrations before (Basal) and after (2h) intracerebroventricular (icv)
vehicle, T1AM and T0AM infusion
T1AM icv n=9
Veh icv n = 8
Basal
2h
Basal
2h
T3 (nmol/L)
1,27 ± 0,08
0,77 ± 0,05*
1,39 ± 0,15
0,86 ± 0,09*
T4 (nmol/L)
81 ± 7
53 ± 4*
79 ± 7
64 ± 7*
TSH (mU/L)
1,64 ± 0,33
1,36 ± 0,26
0,64 ± 0,19
0,77 ± 0,16*
* p≤0.05 vs Basal value within the same group
proefschrift Klieverik.indb 88
4-8-2009 15:22:56
a
Insulin (pmol/L)
400
300
200
"
100
ANOVA ns
0
0 10
60
Glucagon (pg/mL)
Bolus icv
250
^
*
200
150
*
100
50
0
0 10
60
120
c
Corticosteron (ng/mL)
500
Bolus icv
**
400
*
300
**
**
89
Central thyronamines and glucose metabolism
T1 AM 0.5 mg/kg icv (n=9)
T0 AM 0.5 mg/kg icv (n=8)
Vehicle icv (n=8)
b
300
120
Chapter 5
Fig 5a Plasma insulin concentrations before (t=0) and
after icv bolus infusion of T1AM, T0AM or vehicle.
Note that at t=10 min, there is a trend for insulin
concentrations to be depressed in animals receiving
0.5 mg/kg T1AM icv compared to vehicle treated rats
(“p=0.063). ANOVA RM factor time p=0.01, time*group
p=0.161, group p=0.758.
b Plasma glucagon concentrations before (t=0) and
after icv bolus infusion of T1AM, T0AM or vehicle. At
t=10 min glucagon is significantly higher in T1AM icv
and, to a lesser extent, T0AM icv as compared with
vehicle icv treated rats. * p<0.01 vs vehicle icv, ^ p<0.01
T1AM vs T0AM icv. ANOVA RM factor time p<0.0001,
time*group p<0.0001, group p<0.0001.
c Plasma corticosterone concentration before (t= -20 and
0 min) and after (t=5 to t=120 min) icv bolus infusion
of T1AM, T0AM or vehicle. Circulating corticosterone
levels rapidly increase following infusion of T1AM or
T0AM at t=20 and t=10 min as compared with vehicle,
respectively. Note that T1AM and T0AM icv infused
groups do not differ at any time point, except for t=5
min.
*p<0.05 vs vehicle icv, ^ p<0.01 T1AM vs T0AM icv.
ANOVA RM factor time p<0.0001, time*group p=0.002,
group p=0.001.
Bolus icv
500
**
^
200
100
0
-20
0 102030
60
Time (min)
120
either T1AM or T0AM, plasma glucagon returned to basal levels within 60 min of infusing either
compound icv (fig 5b).
Plasma corticosterone concentrations were significantly increased following icv infusion of either
0.5 mg/kg T1AM or T0AM (fig 5c) with both treatments producing nearly equivalent maximum
ANOVA RM
T0AM icv n=8
Basal
2h
Time Time*Group Group
p
p
p
1,00 ± 0,05
0,73 ± 0,05*
0,000
0,198
0,056
87 ± 4
62 ± 4*
0,000
0,184
0.578
0,000
0,665
0,734
1,65 ± 0,25
proefschrift Klieverik.indb 89
0,58 ± 0,10 *
4-8-2009 15:22:56
effects by t=20 min post injection (delta corticosterone t=20 vs t=0; 6±44 ng/mL Veh icv, 206±58
ng/mL T1AM icv , 296±49 ng/mL T0AM icv, p=0.02 T1AM vs Veh, p<0.0001 T0AM vs Veh).
Plasma T3, T4 and TSH concentrations before and after central (icv) infusion of either T1AM, T0AM,
or vehicle (saline-DMSO) infusion are shown in table 2. Although plasma T3, T4 and TSH were found
to decrease in all treatment groups by 120 min post icv bolus infusion, neither icv T1AM nor T0AM
had a statistically significant effect on plasma T3, T4 or TSH compared to vehicle icv.
90
Discussion
In an effort to determine if the thyronamines T1AM and T0AM can affect glucose homeostasis by
acting directly on the brain we compared their physiologic consequences following systemic and
central administration. The major finding of our study is that central administration of low dose
(i.e. 1% of the systemic dose) T1AM acutely increases EGP and plasma glucose concentration to
a similar -or even greater- extent compared with systemic T1AM, concomitant with an increase
of plasma glucagon and corticosterone concentrations. Similar effects were observed following
central T0AM infusion, albeit to a lesser extent. When administered intravenously, the same low
dose of T1AM and T0AM that was effective centrally had no detectable effect on plasma glucose
or EGP, thus excluding the possibility that the observed responses were the result of leakage of
the centrally administered compound into the circulation and acting peripherally.
Rats infused intraperitoneally with T1AM, and to a lesser extent T0AM, exhibited a behavioural
phenotype within minutes of administration, which was remarkably similar to the fully reversible
behavioural changes reported earlier in mice (1;19). In short, animals exhibited a decrease in
overall locomotor activity and responsiveness to external stimuli (visual, auditory) while reflexes
were preserved. Furthermore, the hyperglycemia that develops in mice (7) following thyronamine
exposure also is seen in rats (fig 1a). Moreover, we show for the first time that the hyperglycemia
induced by the thyronamines T1AM and (to a lesser extent) T0AM occurs simultaneously with
a rapid (i.e. within 10 min), approximately 50% increase in EGP, that was maintained for the
duration of the experiment (fig 1b).
With respect to the systemic administration of thyronamines there are several mechanisms that
may contribute to the alterations in glucose metabolism we observed. First, plasma glucagon
increases rapidly in response to systemic T1AM and T0AM administration, concomitant with the
increase in plasma glucose and EGP. It was expected that the thyronamine-induced hyperglycemia
(up to 22 mmol/L; fig 1a) would provoke a considerable insulin response. However, plasma
insulin did not change in spite of the overt hyperglycemia produced by either thyronamine. The
plasma glucagon increase together with this inadequate insulin response are likely to be causal
factors in the increase in plasma glucose and EGP induced by thyronamines. These effects on
plasma glucagon and insulin might be explained by direct actions of T1AM and T0AM on the
pancreatic alpha and beta cells, supposedly by binding to GPCRs such as TAAR1 or ARα2 (1;9).
Indeed, pharmacological stimulation of ARα2 has been shown to induce hyperglycemia and
inhibit insulin release (20). Secondly, given the rapid onset of thyronamine-induced changes it
is well possible that T1AM and T0AM activation of GPCRs expressed in hepatocytes underlies
proefschrift Klieverik.indb 90
4-8-2009 15:22:56
91
Central thyronamines and glucose metabolism
proefschrift Klieverik.indb 91
Chapter 5
the stimulation of EGP we observed, analogous to the stimulation of β-adrenergic receptors by
norepinephrine.
With regard to the possible mechanisms underlying the effects of centrally administered
thyronamines on glucose metabolism it is interesting that concomitant with the rapid increase
of EGP (fig 3b), plasma insulin levels tended to decrease acutely after central administration of
0.5 mg/kg T1AM (fig 5a) in contrast to systemically (i.e. ip) administered drug. In addition, after
icv infusion of 0.5 mg/kg T1AM there was a rapid increase of plasma glucagon (fig 5b), which
was more pronounced than the early glucagon increase after systemic thyronamine infusion. No
change in EGP, plasma insulin, and glucagon levels was observed after intravenous infusion of 0.5
mg/kg T1AM, confirming that T1AM-imposed actions on the CNS are causal to these phenomena.
This dependence of thyronamine effects on the route of administration point to neural or (neuro)
transmitter-type, rather than humoral-type of actions. Indeed, it has been demonstrated that T1AM
modulates synaptosomal transport of neurotransmitters such as dopamine and noradrenalin (21),
supposedly by behaving as endogenous monoamine re-uptake inhibitors (10). In addition, low
dose T1AM administration in the lateral cerebral ventricles and in the arcuate nucleus was recently
reported to rapidly increase food intake (22). The effects of thyronamines on plasma insulin and
glucagon in the present study may be explained by increased sympathetic tone in the pancreas,
mediated via central thyronamine actions. In addition, centrally administered thyronamines might
stimulate autonomic outflow from the hypothalamus to the liver thereby elevating EGP. In support
of this conjecture is accumulating evidence demonstrating the brain’s important role, particularly
the hypothalamus, in regulating hepatic glucose metabolism via sympathetic and parasympathetic
projections to the liver (12-15).
As thyronamines have been hypothesized to constitute a novel aspect of thyroid hormone biology
(1;9), it was of interest to assess how thyroid-related parameters in euthyroid animals responded
to synthetic thyronamines. Systemic infusion of these compounds, in particular T0AM, depressed
plasma TSH, T4 and T3 levels whereas central administration had no such effect. These responses
could represent a state reminiscent of the non-thyroidal illness syndrome (23-25). Although it is
conceivable the thyronamines altered TH secretion by decreasing TSH release from the pituitary,
the observation that central thyronamine administration does not induce plasma thyroid hormone
alterations relative to vehicle, argues against this possibility. Finally, an effect of thyronamines on
plasma concentrations of iodothyronines via interaction with the deiodinase enzymes seems less
likely as T1AM does not interfere with D1-mediated iodothyronine deiodination in vitro (2).
Systemic thyronamine administration produced a significant increases in plasma corticosterone
levels that were similar in T1AM and T0AM infused rats (fig 2c), and could be recapitulated by
low dose (0.5 mg/kg) icv infusion of synthetic T1AM or T0AM (fig 5c), suggesting that these
represent central effects of thyronamines on the hypothalamus-pituitary-adrenal (HPA)-axis. The
elevated corticosterone could contribute in a limited way to the hyperglycaemic state but, more
importantly, because in both experiments the hyperglycemia was much more pronounced in
T1AM as compared to T0AM infused animals, it is unlikely to account for the major glucose
increase induced by systemic and central T1AM.
There are several mechanisms by which circulating thyronamines might exert their actions in the
CNS. First, there could be passive or active transport of circulating thyronamines across the blood
4-8-2009 15:22:56
92
brain barrier, the latter analogous to iodothyronines (26). Second, circulating thyronamines might
bind cell-surface receptors in the plasma membrane of neurons located in circumventricular nuclei
such as the arcuate nucleus where the blood brain barrier is absent. The arcuate nucleus is known
to mediate central actions of the peptide hormones like insulin and leptin via locally expressed
leptin and insulin receptors (27) and TAAR1 is expressed in the arcuate nucleus (5). However, our
finding that the EGP increase is not as robust following central administration of thyronamines
as it is after systemic administration suggests their central actions alone are insufficient to
account for the persistent EGP increase and hyperglycemia. Consistent with this interpretation
are the results from a recent study in which mice pre-treatment with 6-hydroxydopamine still
developed hyperglycemia and hypoinsulinemia following ip administration of T1AM, suggesting
that these T1AM-induced alterations can occur in the absence of sympathetic signalling (7).
Another possibility is that thyronamines impose both peripheral and central actions on glucose
metabolism, occurring independently. In this intriguing scenario, central actions could be
mediated by thyronamines formed locally in the brain, by conversion from iodothyronines such
as T4, T3 and/or rT3.
The rapid and pronounced metabolic effects produced by central administration T1AM and
T0AM suggest that one or more receptors mediate their actions. Indeed T1AM and T0AM dosedependently activate the Gs protein–coupled TAAR1 receptor (4-6;28). TAAR1 belongs to a
large family of related receptors (4;5) and this receptor’s mRNA is expressed in a wide variety
of tissues including many areas of the brain (5;6). The fact that the rank-order of potency as a
TAAR1 agonist in vitro, T1AM being more potent that T0AM (1), is also reflected in the metabolic
responses described in the present study, fits with the notion that TAAR1 may mediate some
actions of T1AM and T0AM. In addition, Regard and colleagues (7) recently showed that whereas
T1AM induces hyperglycemia after systemic administration in wild-type mice, this effect is lost
in adrenergic receptor alpha 2 (ARα2) deficient mice as well as mice pre-treated with the ARα2
antagonist yohimbine. Moreover, by using a transgenic approach, they provided strong evidence
that the hyperglycemia and concurrent hypoinsulinemia following T1AM infusion in mice was
dependent upon pancreatic Gi protein coupled receptor expression. Collectively, these data
suggest that, at least for the effects of systemically administered T1AM on glucose metabolism,
ARα2 is important. Interestingly, ARα2 are highly expressed in the hypothalamus as well and
contribute to the hypothalamic regulation of sympathetic outflow (29), supporting their possible
involvement in mediating effects of centrally administered thyronamines .
To date every published metabolic and physiologic study involving thyronamines, including this
one, has relied on the administration of synthetic material (7;8;19;30). Therefore, the biological
significance of endogenous thyronamines remains to be addressed. In this context it will be
important to establish how and where these compounds are synthesized. Although there is
currently no direct evidence in the literature for in vivo conversion of thyronamines from precursor
iodothyronines (i.e. T3, T4, rT3), the enzymes indispensable for such conversion such as aromatic
amino adic decarboxylase (AADC) for decarboxylation and iodothyronine deiodinases type 2
an 3 for deiodination are widely distributed in the CNS, and indeed within the hypothalamus
(31-34). An important remaining question is whether the metabolic effects in the present and
other studies represent physiological or pharmacological effects of thyronamines. The systemic
proefschrift Klieverik.indb 92
4-8-2009 15:22:56
We wish to thank A. van Riel, E.M. Johannesma-Brian and A.F.C. Ruiter for excellent technical
assistance.
Reference List
1. Scanlan TS, Suchland KL, Hart ME, Chiellini G, Huang Y, Kruzich PJ, Frascarelli S, Crossley DA, Bunzow
JR, Ronca-Testoni S, Lin ET, Hatton D, Zucchi R, Grandy DK 2004 3-Iodothyronamine is an endogenous
and rapid-acting derivative of thyroid hormone. Nat Med 10:638-642
2. Piehl S, Heberer T, Balizs G, Scanlan TS, Smits R, Koksch B, Kohrle J 2008 Thyronamines are isozymespecific substrates of deiodinases. Endocrinology 149:3037-3045
3. Hart ME, Suchland KL, Miyakawa M, Bunzow JR, Grandy DK, Scanlan TS 2006 Trace amine-associated
receptor agonists: synthesis and evaluation of thyronamines and related analogues. J Med Chem
49:1101-1112
4. Grandy DK 2007 Trace amine-associated receptor 1-Family archetype or iconoclast? Pharmacol Ther
116:355-390
5. Borowsky B, Adham N, Jones KA, Raddatz R, Artymyshyn R, Ogozalek KL, Durkin MM, Lakhlani PP,
Bonini JA, Pathirana S, Boyle N, Pu X, Kouranova E, Lichtblau H, Ochoa FY, Branchek TA, Gerald C 2001
Trace amines: identification of a family of mammalian G protein-coupled receptors. Proc Natl Acad Sci
U S A 98:8966-8971
6. Bunzow JR, Sonders MS, Arttamangkul S, Harrison LM, Zhang G, Quigley DI, Darland T, Suchland
KL, Pasumamula S, Kennedy JL, Olson SB, Magenis RE, Amara SG, Grandy DK 2001 Amphetamine,
3,4-methylenedioxymethamphetamine, lysergic acid diethylamide, and metabolites of the catecholamine
neurotransmitters are agonists of a rat trace amine receptor. Mol Pharmacol 60:1181-1188
7. Regard JB, Kataoka H, Cano DA, Camerer E, Yin L, Zheng YW, Scanlan TS, Hebrok M, Coughlin SR 2007
Probing cell type-specific functions of Gi in vivo identifies GPCR regulators of insulin secretion. J Clin
Invest 117:4034-4043
8. Braulke LJ, Klingenspor M, DeBarber A, Tobias SC, Grandy DK, Scanlan TS, Heldmaier G 2008
3-Iodothyronamine: a novel hormone controlling the balance between glucose and lipid utilisation. J
Comp Physiol B 178:167-177
9. Liggett SB 2004 The two-timing thyroid. Nat Med 10:582-583
93
Central thyronamines and glucose metabolism
Acknowledgements
Chapter 5
dose of T1AM used in our study has been shown to induce a 10-fold increase in plasma T1AM
concentration within 3 h after infusion in Siberian hamsters (8). However, there are currently no
data on the pharmacokinetic characteristics (distribution volume, clearance, binding to plasma
proteins) of thyronamines, and at present it is unknown how thyronamine tissue concentrations
during experimental manipulations compare to their concentrations under more physiologic
conditions.
We conclude that central administration of a low dose of either T1AM or T0AM can acutely induce
increased EGP and hyperglycemia, concomitant with increased plasma glucagon, corticosterone
and a deficient insulin response. These changes are very similar to the acute changes observed
after systemic T1AM and T0AM administration. Our data indicate that thyronamines can act
centrally in order to modulate glucose metabolism.
10. Weatherman RV 2007 A triple play for thyroid hormone. ACS Chem Biol 2:377-379
proefschrift Klieverik.indb 93
4-8-2009 15:22:56
11. Kalsbeek A, La Fleur S, Van Heijningen C, Buijs RM 2004 Suprachiasmatic GABAergic inputs to the
paraventricular nucleus control plasma glucose concentrations in the rat via sympathetic innervation of
the liver. J Neurosci 24:7604-7613
12. Obici S, Feng Z, Karkanias G, Baskin DG, Rossetti L 2002 Decreasing hypothalamic insulin receptors
causes hyperphagia and insulin resistance in rats. Nat Neurosci 5:566-572
13. Pocai A, Lam TK, Gutierrez-Juarez R, Obici S, Schwartz GJ, Bryan J, Aguilar-Bryan L, Rossetti L 2005
Hypothalamic K(ATP) channels control hepatic glucose production. Nature 434:1026-1031
94
14. Klieverik LP, Sauerwein HP, Ackermans MT, Boelen A, Kalsbeek A, Fliers E 2008 Effects of thyrotoxicosis
and selective hepatic autonomic denervation on hepatic glucose metabolism in rats. Am J Physiol
Endocrinol Metab 294:E513-E520
15. Kalsbeek A, Fliers E, Franke AN, Wortel J, Buijs RM 2000 Functional connections between the
suprachiasmatic nucleus and the thyroid gland as revealed by lesioning and viral tracing techniques in
the rat. Endocrinology 141:3832-3841
16. Ackermans MT, Klieverik LP, Endert E, Sauerwein HP, Kalsbeek A, Fliers E 2008 Plasma insulin
concentrations during a hyperinsulinaemic clamp: what do we measure? Ann Clin Biochem 45:429-430
17. Ackermans MT, Pereira Arias AM, Bisschop PH, Endert E, Sauerwein HP, Romijn JA 2001 The quantification
of gluconeogenesis in healthy men by (2)H2O and [2-(13)C]glycerol yields different results: rates of
gluconeogenesis in healthy men measured with (2)H2O are higher than those measured with [2-(13)C]
glycerol. J Clin Endocrinol Metab 86:2220-2226
18. Steele R 1959 Influences of glucose loading and of injected insulin on hepatic glucose output. Ann N Y
Acad Sci 82:420-430
19. Doyle KP, Suchland KL, Ciesielski TM, Lessov NS, Grandy DK, Scanlan TS, Stenzel-Poore MP 2007 Novel
thyroxine derivatives, thyronamine and 3-iodothyronamine, induce transient hypothermia and marked
neuroprotection against stroke injury. Stroke 38:2569-2576
20. Angel I, Niddam R, Langer SZ 1990 Involvement of alpha-2 adrenergic receptor subtypes in hyperglycemia.
J Pharmacol Exp Ther 254:877-882
21. Snead AN, Santos MS, Seal RP, Miyakawa M, Edwards RH, Scanlan TS 2007 Thyronamines inhibit plasma
membrane and vesicular monoamine transport. ACS Chem Biol 2:390-398
22. Dhillo WS, Bewick GA, White NE, Gardiner JV, Thompson EL, Bataveljic A, Murphy KG, Roy D, Patel NA,
Scutt JN, Armstrong A, Ghatei MA, Bloom SR 2009 The thyroid hormone derivative 3-iodothyronamine
increases food intake in rodents. Diabetes Obes Metab 11:251-260
23. Adler SM, Wartofsky L 2007 The nonthyroidal illness syndrome. Endocrinol Metab Clin North Am 36:65772
24. Boelen A, Kwakkel J, Thijssen-Timmer DC, Alkemade A, Fliers E, Wiersinga WM 2004 Simultaneous changes
in central and peripheral components of the hypothalamus-pituitary-thyroid axis in lipopolysaccharideinduced acute illness in mice. J Endocrinol 182:315-323
25. Fliers E, Guldenaar SE, Wiersinga WM, Swaab DF 1997 Decreased hypothalamic thyrotropin-releasing
hormone gene expression in patients with nonthyroidal illness. J Clin Endocrinol Metab 82:4032-4036
26. Dratman MB, Crutchfield FL, Schoenhoff MB 1991 Transport of iodothyronines from bloodstream to
brain: contributions by blood:brain and choroid plexus:cerebrospinal fluid barriers. Brain Res 554:229236
27. Niswender KD, Schwartz MW 2003 Insulin and leptin revisited: adiposity signals with overlapping
physiological and intracellular signaling capabilities. Front Neuroendocrinol 24:1-10
28. Wainscott DB, Little SP, Yin T, Tu Y, Rocco VP, He JX, Nelson DL 2007 Pharmacologic characterization of
the cloned human trace amine-associated receptor1 (TAAR1) and evidence for species differences with
the rat TAAR1. J Pharmacol Exp Ther 320:475-485
proefschrift Klieverik.indb 94
4-8-2009 15:22:56
29. Li DP, Atnip LM, Chen SR, Pan HL 2005 Regulation of synaptic inputs to paraventricular-spinal output
neurons by alpha2 adrenergic receptors. J Neurophysiol 93:393-402
31. Alkemade A, Friesema EC, Unmehopa UA, Fabriek BO, Kuiper GG, Leonard JL, Wiersinga WM, Swaab
DF, Visser TJ, Fliers E 2005 Neuroanatomical pathways for thyroid hormone feedback in the human
hypothalamus. J Clin Endocrinol Metab 90:4322-4334
33. Tu HM, Kim SW, Salvatore D, Bartha T, Legradi G, Larsen PR, Lechan RM 1997 Regional distribution
of type 2 thyroxine deiodinase messenger ribonucleic acid in rat hypothalamus and pituitary and its
regulation by thyroid hormone. Endocrinology 138:3359-3368
34. Zhu MY, Juorio AV 1995 Aromatic L-amino acid decarboxylase: biological characterization and functional
role. Gen Pharmacol 26:681-696
proefschrift Klieverik.indb 95
95
Central thyronamines and glucose metabolism
32. Lechan RM, Fekete C 2005 Role of thyroid hormone deiodination in the hypothalamus. Thyroid 15:883897
Chapter 5
30. Chiellini G, Frascarelli S, Ghelardoni S, Carnicelli V, Tobias SC, DeBarber A, Brogioni S, Ronca-Testoni S,
Cerbai E, Grandy DK, Scanlan TS, Zucchi R 2007 Cardiac effects of 3-iodothyronamine: a new aminergic
system modulating cardiac function. FASEB J 21:1597-1608
4-8-2009 15:22:57
proefschrift Klieverik.indb 96
4-8-2009 15:23:00
6
Energy homeostasis before
and after cessation of block
and replacement therapy
in euthyroid patients with
Graves’ disease
Lars P. Klieverik
Andries Kalsbeek
Mariëtte T. Ackermans
Hans P. Sauerwein
Wilmar M. Wiersinga
Eric Fliers1
proefschrift Klieverik.indb 97
4-8-2009 15:23:03
Abstract
98
Context: Patients with Graves’ hyperthyroidism who are treated with a combination of a
thyrostatic drugs and thyroxine (T4), i.e., block and replacement therapy (BRT), often report
excessive body weight (BW) gain.
Objective: The aim of the present study was to investigate changes in BW and resting energy
expenditure (REE) upon cessation of BRT in euthyroid patients with Graves disease, and to identify
determinants of BW and energy metabolism in this setting.
Design and patients: We studied 22 euthyroid patients with Graves disease who had been
treated with BRT for 13.5 [9.5 – 48.0] months on two separate occasions, i.e. (i) during BRT, and
(ii) 12 weeks after cessation of BRT. Patients were biochemically euthyroid on both occasions. At
both visits, we assessed BW and body composition, energy metabolism using indirect calorimetry,
and serum hormone concentrations.
Results: There were no differences in BW or REE between the two visits. At visit 1, serum FT4
correlated positively with resting energy expenditure (REE, r=0.433, p=0.044) and negatively
with body fat % (r=-0.450, p=0.035), while serum free triiodothyronine (FT3) tended to correlate
with REE (r=0.390, p=0.066). Plasma FT3 as well as the FT3/FT4 ratio showed an increase 12 w
after cessation of BRT (by 20%, p<0.0001 and by 16%, p=0.007, respectively). Moreover, the %
change in FT3/FT4 ratio showed a significant and positive correlation with the % change in REE
between the 2 visits (r=0.465, p=0.029).
Conclusions: We conclude that serum FT4 is a determinant of REE in euthyroid patients treated
with BRT for Graves’ hyperthyroidism. Twelve weeks after cessation of BRT, BW and energy
homeostasis are unaltered. However, as the serum FT3/FT4 ratio increases after cessation of BRT
and as this change in FT3/FT4 ratio is a positive determinant of changes in REE, a longer term
decrease in BW is likely to occur.
proefschrift Klieverik.indb 98
4-8-2009 15:23:03
Introduction
Chapter 6
99
Energy homeostasis and block and replacement therapy
Patients with a first episode of Graves’ hyperthyroidism are often treated initially with
pharmacological therapy. Many clinicians use a therapeutic regimen commonly referred to as
“block and replacement therapy” (BRT) to attain biochemical euthyroidism. This involves the
administration of a thyroid hormone (TH) synthesis-blocking agent such as methimazole (MMI)
or propylthiouracil (PTU) to which L-thyroxine (T4) is added once serum TH concentrations reach
the euthyroid range.
Hyperthyroidism is associated with profound changes in energy homeostasis that usually resolve
upon treatment with BRT. As a result, the weight loss experienced during hyperthyroidism is
usually regained during treatment. Some older studies have reported no difference between
premorbid body weight (BW) and BW after one year of BRT (1;2), fitting with the notion of tight
BW set-point regulation. However, a more recent study in 162 patients reported a continuing BW
gain after treatment for hyperthyroidism. BW had increased by ~4 kg after 1 year, and increased
by ~10 kg after four years after of treatment (3). In line with this observation, 79% of patients
report weight gain exceeding their pre-morbid BW following treatment of hyperthyroidism (4).
Although the etiology of this excessive weight gain is incompletely understood at present, some
authors have proposed that it may result from subnormal energy expenditure due to iatrogenic
suppression of TH concentrations to the lower end of the normal range (5;6).
It is remarkable that studies addressing this issue to date have focussed on the comparison
between BW and energy homeostasis in untreated hyperthyroid patients with the euthyroid
situation during BRT. Approximately 50% of patients with Graves hyperthyroidism remain
euthyroid following cessation of BRT after at least one year of treatment, reflecting the
remission rate of autoimmune hyperthyroidism (7). In our Outpatient Clinic of Endocrinology,
BRT is discontinued in patients with Graves hyperthyroidism after approximately one year of
treatment. This regimen offers a good opportunity to investigate changes in BW upon cessation
of BRT in euthyroid patients with Graves disease and to identify determinants of BW and energy
metabolism in this setting.
Subjects and Methods
Subjects
Patients were recruited from the Outpatient Clinics of Endocrinology and Metabolism at
the Academic Medical Center of the University of Amsterdam. Inclusion criteria were: (i) a
minimum of six months of MMI or PTU in combination with T4 pharmacotherapy for Graves’
hyperthyroidism, diagnosed on the basis of standard biochemical and scintigraphic criteria (8), (ii)
biochemical euthyroidism (defined as serum FT4 between 9 – 23 pmol/L, serum T3 between 1.32.7 nmol/L and serum TSH<5 mU/L) on both study occasions, and (iii) age between 20 and 60
years. As low or even suppressed serum TSH values can be found in euthyroid patients treated
for Graves hyperthyroidism, probably resulting from binding of circulating TSH receptor (TSH-R)
auto-antibodies (TBII) to the TSH-R on the thyrotrophs (9), patients with serum TSH values below
the lower limit of the reference range could participate in the study.
proefschrift Klieverik.indb 99
4-8-2009 15:23:03
Exclusion criteria were (i) pregnancy (ii) abnormal liver function as apparent from serum alaninaminotransferase >45 U/L, aspartate-aminotransferase > 40 U/L or gamma-glutamyl-transferase
>60 U/L, and (iii) abnormal kidney function (serum kreatinine>95 μmol/L) at inclusion.
The study was approved by the Medical Ethical Committee of the Academic Medical Center of
the University of Amsterdam and accordingly, written informed consent was obtained from all
subjects prior to inclusion.
100
Protocol
In this observational study, the participants were studied in the morning after an overnight fast
on two occasions, i.e. (i) during biochemical euthyroidism while on BRT (visit 1) and (ii) during
biochemical euthyroidism12 w after discontinuation of BRT (indicating remission of Graves’
hyperthyroidism) (visit 2). At visit 1, a medical history was obtained regarding premorbid BW,
initial presentation of Graves hyperthyroidism. Data on BW progression during the course of
treatment was obtained from clinic records for every individual patient. On both occasions,
BW, resting energy expenditure (REE), body composition and serum T3, T4, FT3, FT4, rT3, TSH,
insulin, glucose, adrenalin and noradrenalin were assessed. All measurements were performed
in the morning between 8.00 and 12.00 AM . Twenty-seven patients used no medication other
than BRT. One patient used a β-adrenergic blocker (metoprolol), and another patient used a
benzodiazepine (diazepam). Both patients were instructed not to take any co-medication within
at least 24h prior to study measurements. All patients were instructed not to take thyroxine in
the morning of visit 1, and smoking was not allowed within 12 h prior to measurements. Patients
were instructed to prevent physical exercise at least 3 days before both study visits, to eat 3
meals per day and not to change their eating habits between the visits.
Body composition and indirect calorimetry
Body composition was measured using bioelectrical impedance analysis (Maltron BF-906,
Rayleigh, UK). Oxygen consumption (VO2) and CO2 production (VCO2) were measured with 20
sec intervals for a total time of 30 min by indirect calorimetry using a ventilated hood system
(Sensormedicsmodel 2900; Sensormedics, Anaheim, USA). Patients were studied in the morning
after an overnight fast. During and 30 min prior to the calorimetry measurements, subject were
instructed to rest in the supine position in a temperature-controlled room (23°C). REE, lipid
and glucose oxidation were calculated from VO2 and VCO2 values using algorhythms previously
reported by Frayn et al (10). Respiratory quotient was calculated as VCO2 / VO2. The final 25 min
of each calorimetry measurement, during which stable VO2 and VCO2 values are reached, were
averaged for further analysis in every individual patient.
Hormones
Serum (total) T3 and (total) T4 and rT3 were measured with in-house radioimmunoassay’s (RIA)
(11). Serum FT4 , FT3 and TSH were measured by time-resolved fluoroimmunoassay (Delfia
,Wallac Oy, Turku, Finland). For FT4 the intra-assay variation was 4-5%, the inter-assay variation
6-7%, and the detection limit 2 pmol/L. For FT3 the intra-assay variation was ±6%, the inter-assay
variation ±9%, and the detection limit 1 pmol/L. For TSH the intra-assay variation was 3-4%,
the inter-assay variation 4-5%, and the detection limit 0.01 mU/L. Insulin was measured with a
chemiluminescent immunometric assay (Immulite 2000 system, Diagnostic Products Corporation,
proefschrift Klieverik.indb 100
4-8-2009 15:23:03
Statistics
Results
Thirty-nine biochemically euthyroid patients were initially included and studied before cessation
of BRT. Of these patients, nine had developed a relapse of hyperthyroidism while two patients
who had undergone radio-iodine (I131) ablative therapy had developed hypothyroidism at visit 2.
Six patients were excluded from the final analysis on the basis of serum FT4 or T3 values outside
the reference range, or serum TSH values above the upper limit of the reference range. Thus, the
final analysis was performed in 22 patients who were biochemically euthyroid at both visits.
Anthropometric characteristics
At visit 1, the patients (18 women and 4 men) were 45.5 [24–56] years of age. Body mass index
was 23.8 [19.9–35.8] kg/m2, body height 168 [158–182] cm, and BW was 67.5 [49.4–106.6]
kg. Lean body mass and fat mass were 69.6 [52.0–80.7]%, and 30.4 [19.3–48.0] % of total
body weight, respectively.
101
Energy homeostasis and block and replacement therapy
All data were analyzed using non-parametric tests. We performed comparisons between study
occasions with the Wilcoxon Signed Rank test, and expressed correlations as Spearman’s rank
correlation coefficient (r). SPSS statistical software version 12.0.1 (SPSS Inc, Chicago, IL) was
used for statistical analysis. Data are presented as median [minimum-maximum].
Chapter 6
Los Angeles, USA) with an intra-assay variation of 3-6%, an inter-assay variation of 4-6%, and a
detection limit of 15 pmol/L. Serum TBII was quantitatively determined by a second generation
luminescence receptor assay (DYNOtest TRAK human assay, B.R.A.H.M.S.). Noradrenalin and
adrenalin were determined with an in-house HPLC method. Intra-assay variation noradrenalin:
2%; adrenalin 9%; inter-assay variation noradrenalin: 10%; adrenalin: 14-18%; detection limit:
0.05 nmol/L for both hormones.
Body weight and energy homeostasis
Patients reported that before the onset of the symptoms -later attributed to hyperthyroidismthat had urged them to seek medical advice, BW, i.e., premorbid BW, was 64.0 [47.5 – 97.0] kg.
Table 1 Body weight (BW)
median
min - max
kg
p
kg
Pre-morbid BW (self reported)
64.0
a
47.5 - 97.0
BW at diagnosis GD
62.0
b
45.0 - 92.0
<0.0001 (a vs b)
BW visit 1 (during BRT)
68.1
c
49.4 - 106.6
<0.0001 (b vs c)
BW visit 2 (12w after BRT cessation)
67.1
d
48.0 - 106.5
0.889 (c vs d)
median
kg
%
Δ BW (visit 1 - pre-morbid)
2.4
3.7
-17.1 - 14.6
Δ BW (visit 1 - visit 2)
0.0
0.0
-5.4 - 3.2
proefschrift Klieverik.indb 101
4-8-2009 15:23:03
Table 2 REE, substrate oxidation and body composition
Visit 2
Visit 1
median
min - max
median
min - max
19.7
14.3 - 24.3
20.3
13.2 - 26.2
0.249
Body fat mass (% of total BW)
30.4
19.3 - 48.0
29.1
19.6 - 49.8
0.102
RQ
0.81
0.74 - 0.92
0.81
0.72 - 0.91
0.910
Glucose oxidation (mg/min*kg)
1.12
0.17 - 2.23
1.26
0.00 - 2.22
0.745
Lipid oxidation (mg/min*kg)
0.75
0.15 - 1.23
0.71
0.20 - 1.11
0.858
REE (kCal/kg*24h)
102
p
Eighty-two percent (18/22) of patients reported that decreased BW was among the symptoms
of hyperthyroidism. At the time of diagnosis of hyperthyroidism, the patients had lost 4.8 [0.0 –
25.0] kg of self-reported premorbid BW. Nineteen patients were started on methimazole and 3
patients were started on PTU, to which thyroxine was added as soon as biochemical euthyroidism
was reached. At visit 1, when BRT was discontinued, patients had been treated with BRT for
13.5 [9.5 – 48.0] months. During BRT patients gained 6.9 [-0.1 – 14.6] kg. The difference
between premorbid BW and BW at visit 1 (i.e., presumed weight gain) was 2.4 [-17.1 – 14.6] kg
(premorbid BW vs visit 1 BW, p=0.001).
There was no significant difference in median BW between visit 1 and visit 2, 12 w after cessation
of BRT (table 1, p=0.899). The median difference in BW between visit 1 and visit 2 calculated per
individual patient was 0.0 [-5.4 – 3.2] kg. Furthermore, there was no difference in REE, fat and
glucose oxidation, RQ or body fat mass between visit 1 and visit 2 (table 2).
Hormones
Serum T4 and FT4 were not different between visit 1 and 2 (p=0.362 and p=0.676, respectively,
table 3). Serum T3 tended to increase at visit 2 as compared with visit 1 (3%, p=0.069), whereas
FT3 showed a statistically significant increase at visit 2 (20%, p=0.005). Serum TSH levels decreased
by 59% and serum TBII concentrations by 15% at visit 2 compared with visit 1 (p=0.001 and
p=0.007, respectively). Accordingly, the serum T3/T4 ratio increased by 10% from 1.79% (visit 1)
to 1.97% (visit 2, p=0.033), and the serum FT3/FT4 ratio increased by 16% from 31.8% (visit 1)
to 37.0% (visit 2, p=0.007, see fig 1).
At visit 1, there was a positive correlation between serum FT4 and REE (p=0.044, fig 2a), and
a negative correlation between serum FT4 and % body fat mass (p=0.035, fig 2b). In addition,
0.6
*
FT3 / FT4 ratio
0.5
0.4
Fig 1 Serum FT3/FT4 ratios of 22 euthyroid Graves
patients during BRT (visit 1; grey boxplots) and 12 weeks
after cessation of BRT (visit 2, open boxplots). *p=0.007
visit 1 vs visit 2.
0.3
0.2
0.1
0.0
proefschrift Klieverik.indb 102
*p=0.007
Visit 1
Visit 2
4-8-2009 15:23:04
a
b
22.5
20.0
17.5
15.0
8
10 12 14 16 18 20 22
FT4 (pmol/L)
r= - 0.450 p=0.035
8
10 12 14 16 18 20 22
FT4 (pmol/L)
d
27.5
25.0
55
50
45
40
35
30
25
20
15
10
5
0
Body fat mass (%)
REE (kCal/kg*24u)
r=0.433 p=0.044
22.5
20.0
17.5
15.0
r=0.390 p=0.066
12.5
2
3
4
5
FT3 (pmol/L)
6
7
r= - 0.237 p=0.288
2
3
4
5
6
FT3 (pmol/L)
7
Fig 2 Correlations of serum FT4 with body fat % (fig 2a) and resting energy expenditure (REE, fig 2b), and
serum FT3 with body fat % (fig 2c) and REE (fig 2d). Spearman’s correlation coefficients (r) and p values are
plotted under the horizontal axis of each graph.
103
Energy homeostasis and block and replacement therapy
c
Body fat mass (%)
25.0
12.5
55
50
45
40
35
30
25
20
15
10
5
0
Chapter 6
REE (kCal/kg*24h)
27.5
Table 3 Plasma hormone concentrations at visit 1 and visit 2
Visit 2
Visit 1
median
min - max
median
p
Reference
range
min - max
T4 (nmol/L)
95
60 - 170
90
70 - 130
0.362
70 - 150
FT4 (pmol/L)
13.7
10.0 - 19.3
14.0
10.5 - 21.3
0.676
10.0 - 23.0
T3 (nmol/L)
1.68
1.2 - 2.3
1.73
1.3 - 2.7
0.069
1.3 - 2.7
FT3 (pmol/L)
4.4
3.0 - 6.2
5.3
3.9 - 9.3
0.005
3.3 - 8.2
TSH (mU/L)
1.70
0.04 - 4.50
0.70
0.01 - 2.03
0.001
0.5 - 5.00
TBII (U/L)
1.3
0.5 - 8.9
1.1
0.5 - 7.6
0.007
Insulin (pmol/L)
30
15 - 131
32
15 - 136
0.695
34 - 172
Adrenalin (nmol/L)
0.08
0.05 - 0.48
0.09
0.05 - 0.51
0.872
0.00 - 0.55
Noradrenalin (nmol/l)
1.05
0.43 - 3.69
1.38
0.53 - 4.18
0.833
0.00 - 3.25
proefschrift Klieverik.indb 103
4-8-2009 15:23:04
Fig 3 Correlation of the difference in REE corrected
for BW between visit 1 and visit 2 (Δ REE (%)) and the
difference in serum FT3/FT4 ratios between visit 1 and
visit 2 (Δ FT3/FT4 (%)). Spearman’s correlation coefficient
(r)= 0.465, p= 0.029.
20
15
∆ REE (%)
10
5
0
-5
-10
-15
104
-20
-50
-25
0
25
50
75
∆ FT3 / FT 4 (%)
100
125
there was a highly significant negative correlation between REE and % body fat mass (r= -0.549,
p=0.008). Serum FT3 tended to correlate positively with REE (p=0.066, fig 2c), but showed no
correlation with % body fat mass (p=0.288, fig 2d).
However, at visit 2, these significant correlations were absent (FT4 vs body fat % r=-0.247,
p=0.270, FT4 vs REE r=0.320, p=0.172, FT3 vs body fat % r=-0.109, p=0.630, FT3 vs REE r=0.135,
p=0.550). Intriguingly, the difference in (Δ) FT3/FT4 ratio and ΔREE between visit 1 and visit 2
showed a positive correlation (R=-0.465, p=0.029, see Fig 3).
There were no differences in serum glucose, insulin, adrenalin or noradrenalin between the two
visits (table 3).
Discussion
Patients with Graves disease who are treated with a combination of a thyreostatic drug and
thyroxine replacement, i.e. block and replacement therapy (BRT), often experience a marked
gain in body weight (BW) (3), that has been reported to exceed self-reported premorbid BW
(4). However, there are some inconsistencies in the published data on this issue. In the present
study, we found a median weight excess of 2.4 kg compared with self-reported premorbid BW
in a group of 22 euthyroid patients with Graves hyperthyroidism who had been treated with
BRT for a median of 13.5 months. To further investigate this phenomenon we assessed energy
homeostasis and serum hormone concentrations in these patients both before and 12w after
cessation of BRT. Somewhat unexpectedly, patients showed no change in BW or REE at 12 w
after BRT cessation.
With respect to the 2.4 kg BW excess we are aware of the uncertainty regarding the reliability
of self-reported premorbid BW (12). Symptoms such as a decrease in BW are slowly progressive
and patients may find it hard to remember the precise onset of symptoms. Therefore, we
cannot exclude that underestimation of the self-reported premorbid BWs may have led to an
overestimation of the BW excess during BRT in our patients. In any case, the present study does
not support the clinical impression of a marked and excessive BW increase after ~1 year of BRT
following hyperthyroidism.
Hyperthyroidism is associated with a marked increase in REE (13;14). The simultaneous increase
in appetite and caloric intake does not completely prevent weight loss in most hyperthyroid
patients (15). By inference, BW changes appear to be primarily determined by EE in these
proefschrift Klieverik.indb 104
4-8-2009 15:23:04
105
Energy homeostasis and block and replacement therapy
proefschrift Klieverik.indb 105
Chapter 6
patients. In addition, it has been shown previously that caloric intake decreases rapidly to
normal within 3 months of initiation of anti-thyroid treatment, whereas in the same period BW
increases to premorbid levels (1). Therefore, increased caloric intake is not a likely candidate to
explain the BW gain after initiation of thyreostatic treatment. Our finding that REE and BW are
similar during treatment with BRT compared with 12 weeks after BRT cessation, suggests that
BRT does not induce a major change in energy homeostasis and, in turn, BW gain in excess of
the premorbid situation.
The patients in our study were euthyroid on both occasions, following the design of the study.
This enabled us to investigate the relationship between subtle changes in serum TH values within
the euthyroid range on the one hand, and measures of energy homeostasis on the other hand.
As anticipated, serum FT3/FT4 ratios significantly (16%) increased after cessation of BRT. This
can probably be attributed to reinstated thyroidal T3 secretion, whereas enzymatic conversion
from exogenous T4 is probably the major source of serum T3 during BRT. Interestingly, there
was a significant and positive correlation between the difference in serum FT3/FT4 ratio and the
difference in REE between the 2 study occasions, suggesting that the serum FT3/FT4 ratio is a
positive determinant of REE under these circumstances (fig 3). In addition, serum FT4 levels within
the euthyroid range were a positive determinant of REE. Also FT3 tended to positively determine
REE. This illustrates the sensitivity of REE to even small fluctuations in circulating thyroid hormone
concentrations, which is supported by previous reports (16;17). As anticipated, REE and % body
fat mass showed a highly significant negative correlation. Therefore, it is likely that the negative
correlation between serum FT4 and % body fat mass can be explained by the observation that
FT4 positively determines REE. It appears remarkable that these correlations reached statistical
significance despite the relatively small number and heterogeneity (i.e. regarding age, sex, BMI,
FFM) of patients. Why this was only evident during BRT treatment and not at visit 2 remains
unclear. It may be that after cessation of BRT, the pathophysiological changes in the HPT-axis
associated with Graves disease (e.g., by circulating TBII) prevent the detection of clear-cut
associations between circulating thyroid hormones and REE. Another explanation may be that
a period of 12 weeks is not sufficient to reinstate physiological thyroid hormone synthesis and
release from the thyroid gland.
Although REE and BW were unaltered 12 w after BRT cessation, the data presented do raise a
number of interesting possibilities. Our finding that changes in serum FT3/FT4 ratios positively
determine REE in euthyroid patients (fig 3), suggests that tissues capable of modulating energy
metabolism sense even small changes in circulating THs and alter REE accordingly. Candidates
include tissues significantly contributing to REE and expressing TH receptors such as the heart
and CNS, while striated muscle and adipose tissue may contribute to a lesser extent (18).
Another possibility to explain these findings may be TH-dependent regulation of REE via the
brain, as the hypothalamus plays a major role in the regulation of energy metabolism (19;20).
Interestingly, there is a high density of TH receptors in the rat and human hypothalamic arcuate
and paraventricular nuclei (21;22), and these nuclei are both key players in BW regulation
(16). Recently, we have shown that hypothalamic T3 administration stimulates hepatic glucose
production via a neural pathway involving the hypothalamic paraventricular nucleus and the
sympathetic nervous system (23). Thus, the hypothalamus is able to sense THs and in turn
4-8-2009 15:23:04
106
modulate hepatic glucose metabolism by means of altered sympathetic outflow to the liver. This
raises the possibility that it may similarly modulate energy metabolism by autonomic projections
to other target tissues such as skeletal muscle and adipose tissue in response to TH (24;25).
Serum TSH concentration was lower at visit 2 as compared with visit 1, although it remained
within reference intervals. This may be explained either by subtle T4 under-treatment during BRT
despite serum T4 and FT4 levels within the reference range . Alternatively, the decrease in serum
TSH may point to a tendency to develop a relapse of hyperthyroidism in some patients. However,
this latter possibility appears to be less likely given the decrease of serum TBII, 12 w after BRT
cessation.
In conclusion, we observed no changes in BW and resting energy metabolism 12 weeks after
cessation of BRT in euthyroid Graves patients, so the present findings do not support the notion
that BRT per se induces excessive BW gain. However, patients who stopped BRT showed an
increase in the serum FT3/FT4 ratio and these changes in FT3/FT4 ratio were a positive determinant
of changes in REE. Hence, it is well possible that the changes in circulating thyroid hormones
after cessation of BRT require a longer observational period in order to translate into clear-cut
alterations in BW.
Acknowledgements
We wish to thank E.M. Johannesma-Brian for excellent analytical assistance.
Reference List
1. Abid M, Billington CJ, Nuttall FQ. Thyroid function and energy intake during weight gain following
treatment of hyperthyroidism. J Am Coll Nutr 1999; 18(2):189-193.
2. Lonn L, Stenlof K, Ottosson M, Lindroos AK, Nystrom E, Sjostrom L. Body weight and body composition
changes after treatment of hyperthyroidism. J Clin Endocrinol Metab 1998; 83(12):4269-4273.
3. Dale J, Daykin J, Holder R, Sheppard MC, Franklyn JA. Weight gain following treatment of hyperthyroidism.
Clin Endocrinol (Oxf) 2001; 55(2):233-239.
4. Jansson S, Berg G, Lindstedt G, Michanek A, Nystrom E. Overweight--a common problem among women
treated for hyperthyroidism. Postgrad Med J 1993; 69(808):107-111.
5. Jacobsen R, Lundsgaard C, Lorenzen J et al. Subnormal energy expenditure: a putative causal factor in
the weight gain induced by treatment of hyperthyroidism. Diabetes Obes Metab 2006; 8(2):220-227.
6. Abid M, Billington CJ, Nuttall FQ. Thyroid function and energy intake during weight gain following
treatment of hyperthyroidism. J Am Coll Nutr 1999; 18(2):189-193.
7. Abraham P, Avenell A, Watson WA, Park CM, Bevan JS. Antithyroid drug regimen for treating Graves’
hyperthyroidism. Cochrane Database Syst Rev 2005;(2):CD003420.
8. Vos XG, Smit N, Endert E, Tijssen JG, Wiersinga WM. Frequency and characteristics of TBII-seronegative
patients in a population with untreated Graves‘ hyperthyroidism: a prospective study. Clin Endocrinol
(Oxf) 2008; 69(2):311-317.
9. Brokken LJ, Wiersinga WM, Prummel MF. Thyrotropin receptor autoantibodies are associated with
continued thyrotropin suppression in treated euthyroid Graves‘ disease patients. J Clin Endocrinol Metab
2003; 88(9):4135-4138.
10. Frayn KN. Calculation of substrate oxidation rates in vivo from gaseous exchange. J Appl Physiol 1983;
55(2):628-634.
proefschrift Klieverik.indb 106
4-8-2009 15:23:04
11. Wiersinga WM, Chopra IJ. Radioimmunoassay of thyroxine (T4), 3,5,3‘-triiodothyronine (T3),
3,3‘,5‘-triiodothyronine (reverse T3, rT3), and 3,3‘-diiodothyronine (T2). Methods Enzymol 1982; 84:272303.
14. Moller N, Nielsen S, Nyholm B, Porksen N, Alberti KG, Weeke J. Glucose turnover, fuel oxidation and
forearm substrate exchange in patients with thyrotoxicosis before and after medical treatment. Clin
Endocrinol (Oxf) 1996; 44(4):453-459.
107
15. Hoogwerf BJ, Nuttall FQ. Long-term weight regulation in treated hyperthyroid and hypothyroid subjects.
Am J Med 1984; 76(6):963-970.
16. al Adsani H, Hoffer LJ, Silva JE. Resting energy expenditure is sensitive to small dose changes in patients
on chronic thyroid hormone replacement. J Clin Endocrinol Metab 1997; 82(4):1118-1125.
17. Ortega E, Pannacciulli N, Bogardus C, Krakoff J. Plasma concentrations of free triiodothyronine predict
weight change in euthyroid persons. Am J Clin Nutr 2007; 85(2):440-445.
18. Muller MJ, Bosy-Westphal A, Kutzner D, Heller M. Metabolically active components of fat-free mass
and resting energy expenditure in humans: recent lessons from imaging technologies. Obes Rev 2002;
3(2):113-122.
19. Sandoval D, Cota D, Seeley RJ. The integrative role of CNS fuel-sensing mechanisms in energy balance
and glucose regulation. Annu Rev Physiol 2008; 70:513-535.
20. Seeley RJ, Drazen DL, Clegg DJ. The critical role of the melanocortin system in the control of energy
balance. Annu Rev Nutr 2004; 24:133-149.
21. Alkemade A, Vuijst CL, Unmehopa UA et al. Thyroid hormone receptor expression in the human
hypothalamus and anterior pituitary. J Clin Endocrinol Metab 2005; 90(2):904-912.
Energy homeostasis and block and replacement therapy
13. Bech K, Damsbo P, Eldrup E et al. beta-cell function and glucose and lipid oxidation in Graves‘ disease.
Clin Endocrinol (Oxf) 1996; 44(1):59-66.
Chapter 6
12. Gorber SC, Tremblay M, Moher D, Gorber B. A comparison of direct vs. self-report measures for assessing
height, weight and body mass index: a systematic review. Obes Rev 2007; 8(4):307-326.
22. Lechan RM, Fekete C. The TRH neuron: a hypothalamic integrator of energy metabolism. Prog Brain Res
2006; 153:209-235.
23. Klieverik LP, Janssen SF, van Riel A et al. Thyroid hormone modulates glucose production via a sympathetic
pathway from the hypothalamic paraventricular nucleus to the liver. Proc Natl Acad Sci U S A 2009;
106(14):5966-5971.
24. Kreier F, Fliers E, Voshol PJ et al. Selective parasympathetic innervation of subcutaneous and intraabdominal fat--functional implications. J Clin Invest 2002; 110(9):1243-1250.
25. Kalsbeek A, La FS, Van HC, Buijs RM. Suprachiasmatic GABAergic inputs to the paraventricular nucleus
control plasma glucose concentrations in the rat via sympathetic innervation of the liver. J Neurosci
2004; 24(35):7604-7613.
proefschrift Klieverik.indb 107
4-8-2009 15:23:04
proefschrift Klieverik.indb 108
4-8-2009 15:23:08
7
proefschrift Klieverik.indb 109
General discussion
4-8-2009 15:23:11
proefschrift Klieverik.indb 110
4-8-2009 15:23:11
7.1 Historical perspective
111
General discussion
proefschrift Klieverik.indb 111
Chapter 7
Already by the end of the 19th century the similarity between the syndrome characterized by
goiter, exophthalmos and palpitations (i.e., the so-called “Merseburger triad”) that we now
know as Graves’ disease, and the effects of activation of the sympathetic nervous system
(SNS) was recognized. This led to the widespread theory - decades before the discovery and
isolation thyroxine (T4)- that increased SNS activity played an important pathophysiological
role in this syndrome (1). Clinical application of this notion came with the treatment of severe
thyrotoxicosis by resection of the cervical sympathetic chain in the late 19th century (2), and by
high spinal anaesthesia or adrenal demedullation (3) as a surgical alternative to thyroidectomy
up until the 1930s. After the successful isolation and synthesis of thyroid hormone (TH) and the
subsequent development of antithyroid drugs, these practices were gradually abandoned. Later
on, β-adrenergic blockers came in use for the initial management of severe thyrotoxicosis, a
practice that is still successfully employed today. When it appeared that plasma catecholamine
concentrations during thyrotoxicosis are typically low to normal (4;5), the theory of increased
SNS activity during thyrotoxicosis moved to the background.
With the technical advances in biomedical science from the 1960s onwards, much knowledge
was gathered concerning the link between TH and the SNS. A considerable amount of evidence
was obtained supporting the concept of increased
sensitivity to catecholamines during thyrotoxicosis. Both on the receptor level (β-adrenergic
receptor expression) and on the post-receptor level (i.e., expression of G-proteins, adenylate
cyclase), β-adrenergic signal transduction turned out to be more responsive during thyrotoxicosis
and, conversely, less responsive during hypothyroidism. Many of these studies focussed on the
effects of THs in white adipose tissue (WAT), e.g., lipolysis, the heart and in brown adipose tissue
(BAT). BAT, which is critical for non-shivering thermogenesis in rodents and other small mammals,
even became exemplary for the synergism between TH and the SNS. In response to cold, increased
sympathetic input from the hypothalamic “thermostat” on the one hand, and increased local T3
tissue concentrations -by catecholamine mediated stimulation of local deiodinase type 2 (D2)
activity- on the other, appeared to act synergistically to stimulate thermogenesis (6). Collectively,
these data provided a theoretical explanation for some of the so-called “hyperadrenergic”
symptoms of thyrotoxicosis, although increased β-adrenergic responsiveness at the molecular
level does not always translate into increased physiological sensitivity to catecholamines in vivo
(7;8).
On closer examination, data on alterations of SNS activity during thyrotoxicosis in vivo are often
conflicting. One plausible explanation for this phenomenon may be that adequately measuring
sympathetic neural activity in vivo is notoriously difficult. Over the years, several techniques have
been in use. Plasma catecholamine concentrations appeared to be a poor measure of sympathetic
activity, as it shows considerable fluctuation and mainly provides information regarding
catecholamine secretion in the body compartment from which the plasma sample was obtained.
The 24h urinary excretion of catecholamines, that has been reported to be increased during
hyperthyroidism (9;10), reflects total catecholamine production over 24h. However, it still offers
no differentiation between catecholamines produced by the adrenal medulla and catecholamines
4-8-2009 15:23:11
112
produced by sympathetic nerve endings. In addition, it does not distinguish between sympathetic
activity in specific tissues. For many years, tissue efflux rate of tritiated noradrenaline (3H-NA)
after systemic 3H-NA infusion in experimental animals was regarded the best available method for
measuring tissue-specific SNS activity. This technique somewhat unexpectedly showed increased
3H-NA efflux from the heart and adrenal glands of hypothyroid instead of hyperthyroid rats (11).
Finally, muscle neurography was used to record tibial nerve sympathetic activity, and pointed to
increased activity in hypothyroid, and decreased activity in thyrotoxic patients (12). Although
these techniques potentially provide more accurate information on tissue-specific sympathetic
tone, the data are in conflict with more recent data demonstrating alterations in heart rate
variability in thyrotoxic patients indicating a shift to increased sympathetic and decreased vagal
input to the heart (13-15). Additionally, microdialysis techniques have become available for in
vivo measurement of compounds such as NA in the interstitial space in human subcutaneous
adipose tissue. This technique revealed that in hyperthyroid patients glycerol –reflecting the rate
of lipolysis- and NA concentrations in subcutaneous WAT are increased whereas the opposite
is the case in hypothyroid patients (16). This suggests increased WAT lipolysis due to increased
sympathetic tone in hyperthyroid, and the opposite in hypothyroid patients.
As can be concluded from the various studies mentioned, different techniques have been used
to estimate sympathetic activity, focussing either on whole body sympathetic activity, or more
selectively on SNS activity in specific organs or tissues. From a neuro-anatomical perspective,
autonomic projections to different body compartments and organs show a distinct differentiation
up to the level of the hypothalamus. For example, different sets of hypothalamic pre-autonomic
neurons project to intra-abdominal organs and subcutaneous WAT (17). This suggests that
the neural activity of sympathetic outflow to several body compartments can be differentially
regulated by the hypothalamus. It may well be that there is no such thing as a generalized
change in sympathetic tone during thyrotoxicosis, but rather that the effect of thyrotoxicosis on
sympathetic signalling differs between tissues. This, in turn, may partly explain the inconsistency
of data regarding SNS activity during thyrotoxicosis.
7.2 Biological relevance
The data mentioned above on “peripheral” sympathetic tone during thyrotoxicosis share the
limitation that they provide very limited mechanistic information. It has remained unclear what the
functional meaning of the supposed alterations in autonomic tonus might be. These alterations
may be viewed as adaptive, i.e. secondary, as proposed by some authors (18;19), or rather
causal to the physiological changes during thyrotoxicosis or hypothyroidism. The recent discovery
that hormones (e.g., insulin (20), glucocorticoids (21), estrogens (22)) can modulate peripheral
metabolism by neural actions mediated via the brain, provides a new perspective for a causal link
between alterations in autonomic output and metabolic alterations during thyrotoxicosis.
Along these lines, we hypothesized that part of the metabolic alterations during thyrotoxicosis
are mediated by TH actions in the hypothalamus and the ANS.
Indeed, in chapter 3 we showed that hyperglycemia and increased glucose production during
thyrotoxicosis can be attenuated by selective hepatic sympathectomy, and that the hepatic insulin
proefschrift Klieverik.indb 112
4-8-2009 15:23:11
Chapter 7
113
General discussion
resistance during thyrotoxicosis can be aggravated by selective hepatic parasympathectomy.
In other words, the changes in hepatic glucose metabolism during thyrotoxicosis can be
differentially modulated by either sympathetic or parasympathetic selective denervation of
the liver. In addition, our findings presented in chapter 4 show that T3 administered locally
in the hypothalamic PVN, which harbours the pre-autonomic neurons that control the ANS,
stimulates hepatic glucose production via sympathetic projections to the liver on a relatively
short timescale. Needless to say, while these data provide strong evidence for the involvement
of the hypothalamus in the changes in hepatic glucose metabolism during thyrotoxicosis, they
cannot be easily extrapolated to alterations in other metabolic processes in the liver, e.g., lipid
metabolism. Neither can they be extrapolated to metabolic processes in other peripheral organs
such as lipolysis in adipose tissue during thyrotoxicosis. On the other hand, this latter possibility
is supported by the earlier observation of autonomic connections between the hypothalamus
and WAT (23) and of ANS-mediated modulation of WAT lipolysis (24;25). These interesting
possibilities remain to be investigated.
Our data show that the changes in hepatic glucose metabolism during thyrotoxicosis can be
partly explained by “indirect” effects of TH via the hypothalamus and sympathetic projections to
the liver, in addition to “direct” TH effects on the level of the hepatocytes.
Figure 1 Schematic representation of “direct” (hepatic) and “indirect” (central) effects of thyroid hormones
(TH) on the liver. Circulating THs can bind to thyroid hormone receptors (TRs) in hepatocytes, directly affecting
hepatocyte function. In addition, THs may bind to TRs in the hypothalamus. TH may act via TRs in the
paraventricular nucleus (PVN). TH may also act via TRs in the arcuate nucleus (ARC), where the blood-brainbarrier is largely absent, from where information is then conveyed to the PVN. In turn, THs may modulate
hepatic metabolism by altering autonomic signalling from the hypothalamus to the liver. This pathway involves
pre-autonomic neurons in the PVN, sympathetic efferent neurons in the intermediolateral column (IML) and/or
parasympathetic efferent neurons in the dorsomedial nucleus (DMV).
proefschrift Klieverik.indb 113
4-8-2009 15:23:11
114
An important question is what could be the purpose of this dual mechanism of TH action.
As outlined in the classical concept of endocrinology, proposed by Starling in 1915, hormones
are produced by endocrine glands and transported via the circulation to their target organs and
target tissues. Hence, all tissues would be expected to “see” similar concentrations of circulating
hormones. For thyroid hormone, the existence of tissue-specific expressions and activities of
deiodinase enzymes and TH receptor (TR) isoforms offers a certain degree of tissue specific
modulation of TH availability and action. Our finding that the hypothalamus is responsive to T3
and accordingly modulates its sympathetic signalling to the liver, appears to add an additional
level of complexity to this picture. The brain, and more specifically the hypothalamus, has a
unique feature in that it is capable of altering metabolism in virtually all peripheral tissues
and organs differentially via its connections with the ANS. It follows that the hypothalamus
would be an excellent candidate to orchestrate fine-tuning of TH action in peripheral tissues
differentially via its ANS projections, potentially adding a new level of complexity to the tissue
specific regulation of TH action. We may speculate that this differential sympathetic regulation
of tissue-specific metabolism is mediated by differential expression of TR (isoforms) on distinct
sets of pre-autonomic neurons in the hypothalamus projecting to specific organs/tissues. In other
words, the TR-isoform expression profile on pre-autonomic neurons projecting to, e.g., striated
muscle may be different from pre-autonomic neurons projecting to liver or WAT, giving rise to
differential autonomic (sympathetic) output upon binding of T3. This exciting possibility will be
the subject of further studies.
Another factor that may be relevant for the question as to what may be the purpose of the dual
mechanism of TH action, is the factor time. TH is known to elicit a wide variety of physiological
actions of miscellaneous nature. Many of these TH actions can be explained by transcriptional
effects, mediated via interaction with nuclear TRs. In short, TH enters the cell and binds to the
ligand-binding domain of a TR in the nucleus. The receptor then undergoes major conformational
changes that ultimately cause binding of the TR to a thyroid responsive element (TRE) in the
promoter region of a TH responsive gene. Binding to these TREs can induce either positive
or negative regulation of gene transcription, which is ultimately reflected in the translation of
mRNA into protein (26). Examples of such transcriptional effects include many of TH actions in
the highly coordinated and programmed process of human brain development, or the striking
TH effects on amphibian metamorphosis, exemplifying the wide variety of morphological and
biochemical changes in response to TH (27). As mentioned in the introduction (section 1.3a),
many of the metabolic effects of TH are mediated via TR-mediated transcriptional regulation
of TH-responsive genes. A common denominator of these “transcriptional” TH actions is that
they occur on a relatively long timescale (i.e., hours), which seems logical given the many steps
involved in the complex transcriptional mechanism. However, a growing number of TH effects
have been reported that occur at a much shorter timescale (i.e., within minutes), and are therefore
hard to reconcile with a transcriptional mechanism (28;29). Indeed, a molecular basis for these
“non-transcriptional” TH effects is emerging, e.g., by the recent discovery that the membrane
bound receptor intergrin αVβ3 has high affinity for THs (30;31). Upon binding to this receptor,
several intracellular signalling pathways such as the mitogen-activated protein kinase (MAPKERK) pathway can be initiated to elicit a variety of molecular effects including modulation of
proefschrift Klieverik.indb 114
4-8-2009 15:23:11
115
General discussion
7.2 A new role for the hypothalamus in sensing
hormonal signals and modulating autonomic
output; involvement of Thyroid Hormone
Chapter 7
Na/H proton exchangers in the plasma membrane (for review see (32)). In addition, TRs located
in the cytoplasm have been shown to signal rapidly via non-transcriptional mechanisms, for
example by interaction with the phosphatidylinositol 3-kinase (PI3K) (33) and the downstream
Akt kinase (Akt) / mammalian target of rapamycin (mTOR) signalling cascade (34).
Hypothalamic actions of TH conveyed to peripheral organs by autonomic nerves as described in
this thesis appear to occur within the timeframe of these “non-transcriptional” TH actions. We
have not investigated in the present thesis if the metabolic effects of autonomic stimulation of the
liver occur via transcriptional or non-transcriptional mechanisms, or both. One could, however,
image that it might be efficient for TH to dispose of multiple mechanisms of action covering a
(time) spectrum from fast to relatively slow. This may help the organism to adequately respond
to physiological stressors that require metabolic adaptation both acutely and more chronically.
We chose the pathological condition of thyrotoxicosis to study hypothalamic TH effects on
metabolism, but the finding that hypothalamic T3 is able to modulate hepatic glucose metabolism
via sympathetic projections to the liver may also prove to have physiological implications. It
has become clear that the hypothalamus is able to directly sense circulating hormones and
nutrients that provide the brain with information regarding the metabolic status of the body,
i.e., the status of peripheral tissues like the liver and adipose tissue. A well recognized example
is the hormone leptin, which is secreted by adipose tissue (i.e. high after a meal, suppressed
during starvation) and binds to specific leptin receptors in the hypothalamic arcuate nucleus
(ARC). This provides the hypothalamus with valuable information regarding the nutritional status
of the body. This information is then conveyed to the PVN via a set of functionally reciprocal
neurons containing either alpha melamocotin-stimulating hormone (αMSH) or co-expressing
neuropeptide Y and agouti-related peptide (NPY/AGRP) where it is integrated with other
information. This information may consist of other hormonal (nutritional) signals sensed via the
ARC (e.g., insulin, gut hormones like ghrelin or glugagon-like-peptide-1 (35)), but it also includes
neural inputs from other brain regions such as the brain stem and the limbic system. In the case
of food deprivation, the PVN coordinates several regulatory actions in response to these signals,
adjusting (i) appetite (stimulating food intake), (ii) the HPT-axis (decreasing TRH expression in
hypophysiotropic neurons, contributing to decreased circulating thyroid hormone levels (36))
and (iii) EGP (changing autonomic signalling to the liver to increase EGP (37)). All of these appear
to be useful adaptive responses to reduced food availability.
In addition, during starvation expression of D2 by glial cells in the ARC increases, thereby
increasing local T3 production (38). Interestingly, elevated hypothalamic T3 appears to contribute
to the decreased expression of TRH and thereby decreased circulating TH concentrations during
starvation, and this phenomenon is not secondary to the decreased circulating TH levels (39). The
fasting-induced increase of D2-derived T3 in the ARC also appears to be critical for an appropriate
proefschrift Klieverik.indb 115
4-8-2009 15:23:11
116
orexigenic (appetite stimulating) response of NPY neurons, via a mechanism involving uncoupling
protein type 2 and increased mitochondrial density in these neurons (40). Finally, light induced
hypothalamic D2-mediated T4 to T3 conversion is critical for the timing of reproduction in
response to seasonal (photo-period) changes in birds (41), involving a neuro-endocrine response
in the hypothalamus-pituitary-gonadal axis.
Collectively, these data show that under physiological conditions, hypothalamic T3 levels are
subject to extensive regulation and play an important role in eliciting adequate adaptive
responses. Our data show that upon stimulation by T3, the hypothalamus modulates sympathetic
input to the liver to alter local glucose metabolism. This may provide another pathway by which
physiological alterations in hypothalamic T3 induced by a variety of physiological stimuli may
mediate regulatory actions on the level of peripheral organs.
The exact mechanism by which T3 stimulates PVN neurons to elicit sympathetic signals to the liver
remains to be established. One of the interesting questions is if the PVN pre-autonomic express
TRs, and -along the same lines- if T3 has a direct stimulating effect on these pre-autonomic
neurons. An alternative possibility is that not T3 itself, but rather TH-derivatives are responsible
for the hypothalamic actions of T3 on these neurons. Candidates for this role include the
thyronamines, which are TH-derivatives that can be formed by deiodination and decarboxilation
of TH. In this scenario, T3 would have to be converted to, e.g., 3-iodothyronamine (T1AM) in the
hypothalamus. Although there is no direct evidence for in vivo conversion of iodothyronines into
thyronamines at present, the enzymes indispensable for such conversion (i.e. deiodinases and
decarboxylating enzymes such as amino acid decarboxylase) are abundantly expressed in the
hypothalamus (42). Furthermore, the trace amine associated receptor type 1 (TAAR1), a G-protein
coupled membrane receptor that is activated by T1AM, is expressed in the hypothalamic ARC
(43). In addition, we have shown that T1AM and thyronamine (T0AM) modulate hepatic glucose
production upon central administration (chapter 5).
7.4 Potential clinical relevance
Both hypo- and hyperthyroidism are associated with mood disorders such as anxiety and
depression as well as with mild cognitive impairment, mainly in the domains of working memory
and executive function (44). In addition, even subclinical hypothyroidism, which is regarded as
a mild form of hypothyroidism, has been shown to impair short term working memory (45),
reflected in altered neuronal activity in brain areas crucial for these functions as shown by
functional MRI (fMRI) (46).
Hypothyroidism has been treated with T4 replacement therapy ever since the isolation of T4 and
its production in pharmacological quantities. The clinical impression that a substantial number
of patients with adequately treated hypothyroidism (i.e., serum TSH within the reference range)
keep having mild impairments has been confirmed by studies showing reduced psychological
well-being and neuro-cognitive functioning in these patients (47;48). Rat studies have shown that
it is impossible to normalize T3 and T4 levels in all tissues at the same time by any replacement
dose of T4 alone (49). This raises the theoretical possibility that a subtle dysbalance in TH
homeostasis occurs in certain tissues during T4 replacement therapy in patients as well and
proefschrift Klieverik.indb 116
4-8-2009 15:23:11
117
General discussion
proefschrift Klieverik.indb 117
Chapter 7
that a complete restoration of thyroid hormone homeostasis in all tissues cannot be reached
(50). It should be noted, however, that in the animal experimental studies cerebral cortex T3
content was shown to remain within normal limits over a relatively wide range of plasma T4
concentrations. This observation supports the efficiency of auto-regulation of T3 bio-avalability
involving local deiodinase expression in cerebral cortex. Although no human data are available to
confirm these animal data, the supposed dysbalance in TH homeostasis in certain tissues during
T4 replacement therapy may relate to the impaired well-being in T4-treated patients. Additional
factors have been reported that could be causally related to the impaired well-being in T4-treated
patients. For example, we have shown earlier that polymorphisms in the T4 transporter organic
anion transporting polypeptide 1C1 (OATP1C1), which is expressed at the bloodbrain-barrier,
are associated with symptoms of fatigue and depression in patients with adequately treated
hypothyroidism (51). This may imply that suboptimal TH transport to the brain relates to the
impaired wellbeing in these patients.
Our finding that the hypothalamic–ANS–liver pathway is sensitive to TH may add another level of
complexity to this issue. In chapter 6, we have shown that in patients with Graves disease treated
with the thyrostatic methimazole and T4 replacement, i.e., block and replacement therapy or
BRT, plasma FT3/FT4 ratios are ~20% lower as compared to FT3/FT4 ratios in he same patients
in the untreated situation. This probably relates to the fact that in the treated situation, these
patients lack thyroidal T3 production, and T3 is derived from on peripheral enzymatic conversion
of exogenous T4.
Systemic treatment of rats with a dose of T4 inducing a 115% and 210% increase of circulating
T4 and T3 concentrations, respectively, was shown to translate into ~330% and 140% increases
in hypothalamic T4 and T3 content, respectively (52). Of note, these hypothalamic alterations
occurred in spite of local auto-regulation involving local D2 and D3. However, it is impossible at
this stage to infer from these data how the alterations in plasma TH concentrations in patients
on BRT are translated into hypothalamic bioavailability of TH at the level of the PVN or ARC.
Although this remains uncertain, the possibility exists that subtle alterations in circulating THs are
sensed within the hypothalamus, resulting in altered ANS output to peripheral organs such as
the liver, and possibly to other organs as well. One might think of alterations in heart rate, sleepwake rhythm, or temperature regulation. Whether a TH–driven hypothalamus-ANS response
plays a role in the impairment of well being in these patients, will be a fascinating subject of
further research.
A first step to explore this issue will be to analyze untreated and treated patients with overt or
subclinical thyroid function disorders using functional imaging. Imaging techniques like functional
MRI, SPECT and PET have shown spectacular technical advances in recent years. It is now possible
to analyze brain metabolism in, e.g., hypothalamic and brain stem areas.
In clinical thyroidology, the hypothalamus is mainly known for its well established role in feedback
regulation of the HPTaxis. In this thesis, we report metabolic effects of thyroid hormone elicited
by TH actions in the hypothalamus and mediated via the autonomic nervous system. Hopefully,
this may mark the beginning of further exploring this new role for THs in the regulation of
hypothalamic neural output to peripheral organs and, in more general terms, add to our
understanding of the neural actions of thyroid hormone.
4-8-2009 15:23:12
Reference List
118
1. Leak D. 1970 The thyroid and the autonomic nervous system. William Heinemann Medical Books,
London.
2. Poncet MA 1897 Le traitement chirurgical des goitre exophthalmique par la section ou la résection due
sympathique cervical. Bull Acad Med 38:121
3. Crile GW 1929 The interdependence of the thyroid, adrenals and nervous system. Amer.J.Surg. 6:616
4. Coulombe P, Dussault JH, Walker P 1976 Plasma catecholamine concentrations in hyperthyroidism and
hypothyroidism. Metabolism 25:973-979
5. Coulombe P, Dussault JH, Letarte J, Simmard SJ 1976 Catecholamines metabolism in thyroid diseases. I.
Epinephrine secretion rate in hyperthyroidism and hypothyroidism. J Clin Endocrinol Metab 42:125-131
6. Silva JE 2006 Thermogenic mechanisms and their hormonal regulation. Physiol Rev 86:435-464
7. Liggett SB, Shah SD, Cryer PE 1989 Increased fat and skeletal muscle beta-adrenergic receptors but
unaltered metabolic and hemodynamic sensitivity to epinephrine in vivo in experimental human
thyrotoxicosis. J Clin Invest 83:803-809
8. Levey GS, Klein I 1990 Catecholamine-thyroid hormone interactions and the cardiovascular manifestations
of hyperthyroidism. Am J Med 88:642-646
9. Pijl H, de Meijer PH, Langius J, Coenegracht CI, van den Berk AH, Chandie Shaw PK, Boom H, Schoemaker
RC, Cohen AF, Burggraaf J, Meinders AE 2001 Food choice in hyperthyroidism: potential influence of the
autonomic nervous system and brain serotonin precursor availability. J Clin Endocrinol Metab 86:58485853
10. Eustatia-Rutten CF, Corssmit EP, Heemstra KA, Smit JW, Schoemaker RC, Romijn JA, Burggraaf J 2008
Autonomic nervous system function in chronic exogenous subclinical thyrotoxicosis and the effect of
restoring euthyroidism. J Clin Endocrinol Metab 93(7):2835-2841
11. Tu T, Nash CW 1975 The influence of prolonged hyper- and hypothyroid states on the noradrenaline
content of rat tissues and on the accumulation and efflux rates of tritiated noradrenaline. Can J Physiol
Pharmacol 53:74-80
12. Matsukawa T, Mano T, Gotoh E, Minamisawa K, Ishii M 1993 Altered muscle sympathetic nerve activity
in hyperthyroidism and hypothyroidism. J Auton Nerv Syst 42:171-175
13. Burggraaf J, Tulen JH, Lalezari S, Schoemaker RC, De Meyer PH, Meinders AE, Cohen AF, Pijl H 2001
Sympathovagal imbalance in hyperthyroidism. Am J Physiol Endocrinol Metab 281:E190-E195
14. Cacciatori V, Bellavere F, Pezzarossa A, Dellera A, Gemma ML, Thomaseth K, Castello R, Moghetti P,
Muggeo M 1996 Power spectral analysis of heart rate in hyperthyroidism. J Clin Endocrinol Metab
81:2828-2835
15. Chen JL, Chiu HW, Tseng YJ, Chu WC 2006 Hyperthyroidism is characterized by both increased
sympathetic and decreased vagal modulation of heart rate: evidence from spectral analysis of heart rate
variability. Clin Endocrinol (Oxf) 64:611-616
16. Haluzik M, Nedvidkova J, Bartak V, Dostalova I, Vlcek P, Racek P, Taus M, Svacina S, Alesci S, Pacak
K 2003 Effects of hypo- and hyperthyroidism on noradrenergic activity and glycerol concentrations in
human subcutaneous abdominal adipose tissue assessed with microdialysis. J Clin Endocrinol Metab
88:5605-5608
17. Kreier F, Kap YS, Mettenleiter TC, van Heijningen C, van d, V, Kalsbeek A, Sauerwein HP, Fliers E, Romijn
JA, Buijs RM 2006 Tracing from fat tissue, liver, and pancreas: a neuroanatomical framework for the role
of the brain in type 2 diabetes. Endocrinology 147:1140-1147
18. Silva JE. 2005 Thermogenesis and the sympathoadrenal system in thyrotoxicosis. In: The Thyroid, A
Fundamental and Clinical Text, 9th edition: 607-620. Lippincott Williams &Wilkins.
proefschrift Klieverik.indb 118
4-8-2009 15:23:12
19. Silva JE 2003 The thermogenic effect of thyroid hormone and its clinical implications. Ann Intern Med
139:205-213
20. Obici S, Zhang BB, Karkanias G, Rossetti L 2002 Hypothalamic insulin signaling is required for inhibition
of glucose production. Nat Med 8:1376-1382
22. Clegg DJ, Brown LM, Woods SC, Benoit SC 2006 Gonadal hormones determine sensitivity to central
leptin and insulin. Diabetes 55:978-987
24. Fliers E, Kreier F, Voshol PJ, Havekes LM, Sauerwein HP, Kalsbeek A, Buijs RM, Romijn JA 2003 White
adipose tissue: getting nervous. J Neuroendocrinol 15:1005-1010
25. Romijn JA, Fliers E 2005 Sympathetic and parasympathetic innervation of adipose tissue: metabolic
implications. Curr Opin Clin Nutr Metab Care 8:440-444
119
General discussion
23. Kreier F, Fliers E, Voshol PJ, van Eden CG, Havekes LM, Kalsbeek A, Van Heijningen CL, Sluiter AA,
Mettenleiter TC, Romijn JA, Sauerwein HP, Buijs RM 2002 Selective parasympathetic innervation of
subcutaneous and intra-abdominal fat--functional implications. J Clin Invest 110:1243-1250
Chapter 7
21. Cusin I, Rouru J, Rohner-Jeanrenaud F 2001 Intracerebroventricular glucocorticoid infusion in normal
rats: induction of parasympathetic-mediated obesity and insulin resistance. Obes Res 9:401-406
26. Yen PM 2001 Physiological and molecular basis of thyroid hormone action. Physiol Rev 81:1097-1142
27. Tata JR 2006 Amphibian metamorphosis as a model for the developmental actions of thyroid hormone.
Mol Cell Endocrinol 246:10-20
28. Davis PJ, Leonard JL, Davis FB 2008 Mechanisms of nongenomic actions of thyroid hormone. Front
Neuroendocrinol 29:211-218
29. Wu SY, Green WL, Huang WS, Hays MT, Chopra IJ 2005 Alternate pathways of thyroid hormone
metabolism. Thyroid 15:943-958
30. Bergh JJ, Lin HY, Lansing L, Mohamed SN, Davis FB, Mousa S, Davis PJ 2005 Integrin alphaVbeta3
contains a cell surface receptor site for thyroid hormone that is linked to activation of mitogen-activated
protein kinase and induction of angiogenesis. Endocrinology 146:2864-2871
31. Davis PJ, Davis FB, Cody V 2005 Membrane receptors mediating thyroid hormone action. Trends
Endocrinol Metab 16:429-435
32. Davis PJ, Leonard JL, Davis FB 2008 Mechanisms of nongenomic actions of thyroid hormone. Front
Neuroendocrinol 29:211-218
33. Storey NM, Gentile S, Ullah H, Russo A, Muessel M, Erxleben C, Armstrong DL 2006 Rapid signaling at
the plasma membrane by a nuclear receptor for thyroid hormone. Proc Natl Acad Sci U S A 103:51975201
34. Kenessey A, Ojamaa K 2006 Thyroid hormone stimulates protein synthesis in the cardiomyocyte by
activating the Akt-mTOR and p70S6K pathways. J Biol Chem 281:20666-20672
35. Murphy KG, Bloom SR 2006 Gut hormones and the regulation of energy homeostasis. Nature 444:854859
36. Lechan RM, Fekete C 2006 The TRH neuron: a hypothalamic integrator of energy metabolism. Prog Brain
Res 153:209-235
37. Kalsbeek A, La FS, Van HC, Buijs RM 2004 Suprachiasmatic GABAergic inputs to the paraventricular
nucleus control plasma glucose concentrations in the rat via sympathetic innervation of the liver. J
Neurosci 24:7604-7613
38. Diano S, Naftolin F, Goglia F, Horvath TL 1998 Fasting-induced increase in type II iodothyronine deiodinase
activity and messenger ribonucleic acid levels is not reversed by thyroxine in the rat hypothalamus.
Endocrinology 139:2879-2884
proefschrift Klieverik.indb 119
4-8-2009 15:23:12
39. Coppola A, Hughes J, Esposito E, Schiavo L, Meli R, Diano S 2005 Suppression of hypothalamic deiodinase
type II activity blunts TRH mRNA decline during fasting. FEBS Lett 579:4654-4658
40. Coppola A, Liu ZW, Andrews ZB, Paradis E, Roy MC, Friedman JM, Ricquier D, Richard D, Horvath
TL, Gao XB, Diano S 2007 A central thermogenic-like mechanism in feeding regulation: an interplay
between arcuate nucleus T3 and UCP2. Cell Metab 5:21-33
41. Yoshimura T, Yasuo S, Watanabe M, Iigo M, Yamamura T, Hirunagi K, Ebihara S 2003 Light-induced
hormone conversion of T4 to T3 regulates photoperiodic response of gonads in birds. Nature 426:178181
120
42. Zhu MY, Juorio AV 1995 Aromatic L-amino acid decarboxylase: biological characterization and functional
role. Gen Pharmacol 26:681-696
43. Borowsky B, Adham N, Jones KA, Raddatz R, Artymyshyn R, Ogozalek KL, Durkin MM, Lakhlani PP,
Bonini JA, Pathirana S, Boyle N, Pu X, Kouranova E, Lichtblau H, Ochoa FY, Branchek TA, Gerald C 2001
Trace amines: identification of a family of mammalian G protein-coupled receptors. Proc Natl Acad Sci U
S A 98:8966-8971
44. Samuels MH 2008 Cognitive function in untreated hypothyroidism and hyperthyroidism. Curr Opin
Endocrinol Diabetes Obes 15:429-433
45. Samuels MH, Schuff KG, Carlson NE, Carello P, Janowsky JS 2007 Health status, mood, and cognition in
experimentally induced subclinical hypothyroidism. J Clin Endocrinol Metab 92:2545-2551
46. Zhu DF, Wang ZX, Zhang DR, Pan ZL, He S, Hu XP, Chen XC, Zhou JN 2006 fMRI revealed neural
substrate for reversible working memory dysfunction in subclinical hypothyroidism. Brain 129:29232930
47. Saravanan P, Chau WF, Roberts N, Vedhara K, Greenwood R, Dayan CM 2002 Psychological well-being in
patients on ‘adequate’ doses of l-thyroxine: results of a large, controlled community-based questionnaire
study. Clin Endocrinol (Oxf) 57:577-585
48. Wekking EM, Appelhof BC, Fliers E, Schene AH, Huyser J, Tijssen JG, Wiersinga WM 2005 Cognitive
functioning and well-being in euthyroid patients on thyroxine replacement therapy for primary
hypothyroidism. Eur J Endocrinol 153:747-753
49. Escobar-Morreale HF, Obregon MJ, Escobar del RF, Morreale de EG 1995 Replacement therapy for
hypothyroidism with thyroxine alone does not ensure euthyroidism in all tissues, as studied in
thyroidectomized rats. J Clin Invest 96:2828-2838
50. Romijn JA, Smit JW, Lamberts SW 2003 Intrinsic imperfections of endocrine replacement therapy. Eur J
Endocrinol 149:91-97
51. van der Deure WM, Appelhof BC, Peeters RP, Wiersinga WM, Wekking EM, Huyser J, Schene AH,
Tijssen JG, Hoogendijk WJ, Visser TJ, Fliers E 2008 Polymorphisms in the brain-specific thyroid hormone
transporter OATP1C1 are associated with fatigue and depression in hypothyroid patients. Clin Endocrinol
(Oxf) 69:804-811
52. Broedel O, Eravci M, Fuxius S, Smolarz T, Jeitner A, Grau H, Stoltenburg-Didinger G, Plueckhan H, Meinhold
H, Baumgartner A 2003 Effects of hyper- and hypothyroidism on thyroid hormone concentrations in
regions of the rat brain. Am J Physiol Endocrinol Metab 285:E470-E480
proefschrift Klieverik.indb 120
4-8-2009 15:23:12
proefschrift Klieverik.indb 121
4-8-2009 15:23:12
proefschrift Klieverik.indb 122
4-8-2009 15:23:15
8
proefschrift Klieverik.indb 123
Summary
Nederlandse samenvatting
Author Affiliations
Dankwoord
Biografie
4-8-2009 15:23:18
proefschrift Klieverik.indb 124
4-8-2009 15:23:18
Summary
Chapter 8
125
Summary
Thyrotoxicosis is associated with a broad spectrum of clinical symptoms and metabolic alterations.
Many of these symptoms including the metabolic changes show a remarkable similarity with the
effects of sympathetic nervous system stimulation. This similarity was noted by physiologists
long before the discovery and isolation of thyroid hormone (TH), and is still reflected in the initial
treatment of severe thyrotoxicosis with β-adrenergic blockers. In recent years, it has become
increasingly clear that peripheral energy metabolism is controlled not only locally via hormonal
actions, but also by the central nervous system (CNS) via its autonomic projections to a multitude
of peripheral organs, such as the liver. The most important brain area in the CNS for the control
of energy homeostasis is the hypothalamus, i.e., the neuroendocrine center of the brain.
In this thesis, we studied the metabolic alterations during thyrotoxicosis, and the possible
involvement of TH actions at the level of the hypothalamus and the autonomic nervous system
(ANS) in these metabolic alterations. We hypothesised that THs (and TH derivatives) alter
autonomic signalling via actions in the brain, or more specifically, in the hypothalamus, and
thereby modulate metabolism at the level of peripheral organs such as the liver.
Chapter 1 provides an introduction to thyroid hormone metabolism and the interplay between
thyroid hormone and the brain. Furthermore, the metabolic alterations induced by thyrotoxicosis
are reviewed with a special emphasis on the isotope techniques used in the present thesis to
assess metabolic fluxes. The hypothalamus is a major regulator of metabolism via its connections
with the autonomic nervous system (ANS). The functional anatomy of the hypothalamus and
its autonomic outflow, as well as the role of the hypothalamus in the regulation of glucose
metabolism, are discussed. Chapter 1 includes an introduction to the thyronamines, which are
recently reported thyroid hormone analogues that evoke a broad range of rapid metabolic
actions in vivo. Finally, the general hypothesis and thesis outline are presented.
In chapter 2, we studied the effects of thyroid status on whole body energy metabolism. We
found reciprocal alterations in thyrotoxic and hypothyroid rats. Thyrotoxic rats exhibit markedly
increased energy expenditure (EE) and lipid oxidation without alterations in spontaneous physical
activity. Hypothyroid rats show a mild decrease in EE and glucose oxidation. In addition, we
studied how the marked alterations in lipid oxidation are associated with tissue-specific fatty
acid (FA) uptake. The data show that the hypermetabolic phenotype during thyrotoxicosis is
facilitated by increased uptake of fatty acids derived from triglycerides (TG-FA) in oxidative
tissues, but that TG-FA uptake is unaltered in lipid-storing white adipose tissue (WAT). Conversely,
during hypothyroidism TG-FA uptake is unaltered in oxidative tissues, but increased in WAT. This
upregulation in WAT is associated with increased activity of the enzyme lipoprotein lipase, which
regulates tissue-specific FA disposal by hydrolizing TGs. Albumin-bound FA uptake, which is
quantitatively less important, appears to be mainly determined by the plasma FA concentration,
and is apparently not regulated in a tissue-specific manner.
proefschrift Klieverik.indb 125
4-8-2009 15:23:18
126
In chapter 3, we studied the role of the ANS projections to the liver in the alterations of hepatic
glucose metabolism induced by thyrotoxicosis. We assessed endogenous glucose production
(EGP) and its sensitivity to insulin by combining stable isotope dilution and hyperinsulinemic
euglycemic clamping in euthyroid and thyrotoxic rats that underwent prior selective hepatic
autonomic (i.e., either sympathetic or parasympathetic) denervation or a sham denervation. The
data show that the alterations in hepatic glucose metabolism can be differentially modulated
by either selective sympathetic or parasympathetic hepatic denervation. More specifically, the
increase in plasma glucose concentration and, to a lesser extent, EGP induced by thyrotoxicosis
can be partly prevented by prior hepatic sympathetic denervation. This suggests that sympathetic
innervation contributes to the increased plasma glucose concentration and higher EGP during
thyrotoxicosis. Parasympathetic denervation of the liver increases plasma insulin concentration
but not EGP during thyrotoxicosis. Hence, thyrotoxicosis-induced hepatic insulin resistance is
aggravated by selective parasympathetic denervation. By inference, parasympathetic hepatic
innervation may function to restrain EGP during thyrotoxicosis.
As a next step, in chapter 4 we hypothesized that the thyrotoxicosis-induced increase in EGP
can be mimicked by infusing TH directly into the hypothalamus of euthyroid rats. To address this
hypothesis we infused the bio-active thyroid hormone triiodothyronine (T3) into the hypothalamus
of euthyroid rats, while assessing EGP with stable isotope infusion. We administered T3 selectively
to the hypothalamic paraventricular nucleus (PVN), where “pre-autonomic” neurons regulating the
autonomic projections to the liver are localized. The data show that T3 in the hypothalamic PVN
increases EGP on a short timescale, independently of plasma thyroid hormone and glucoregulatory
hormone concentrations. Moreover, we combined hypothalamic T3 administration with selective
sympathetic denervation of the liver showing that intact sympathetic projections to the liver are
crucial for the rapid EGP-stimulating effects of hypothalamic T3. Together, these data reveal a
novel central pathway for modulation of EGP by T3 involving the hypothalamic PVN and the
sympathetic nervous system.
Thyronamines are TH derivatives that exhibit neurotransmitter-like properties, and the phenotype that evolves upon administration of thyronamines in rodents points to involvement
of the hypothalamus in these actions. This is supported by hypothalamic expression of trace
amino associated receptors (TAAR), that selectively bind thyronamines. In chapter 5, we
hypothesized that the effects of 3-iodothyronamine (T1AM) and thyronamine (T0AM) on hepatic
glucose metabolism and circulating glucoregulatory hormones can be explained by actions of
thyronamines in the CNS. We show that systemic T1AM infusion rapidly increases EGP and
plasma glucose, plasma glucagon and corticosterone, while it does not change plasma insulin
concentrations. Compared to systemic administered T1AM, a 100-fold lower dose administered
centrally induces a more pronounced acute EGP increase and hyperglucagonemia with a trend
towards lower plasma insulin. Both systemic and central infusions of T0AM cause similar, but
smaller alterations compared with T1AM. Neither T1AM nor T0AM influences any of these
parameters upon low dose systemic administration. These data show that central administration
of low dose thyronamines suffices to induce the acute alterations in glucose metabolism and
proefschrift Klieverik.indb 126
4-8-2009 15:23:18
glucoregulatory hormones following systemic thyronamine infusion. Thus, thyronamines can act
centrally to modulate glucose metabolism.
Chapter 8
127
Summary
In chapter 6, we explored the interrelationship between THs and energy metabolism in the
clinical setting. We performed a study in patients with Graves disease rendered euthyroid by
blocking thyroid hormone synthesis with an anti-thyroid drug while restoring plasma thyroid
hormone concentrations by exogenous substitution of thyroxine (the so called “block and
replacement therapy” or BRT). We studied these patients on 2 occasions, i.e., after ~13.5
months of BRT and 12 weeks after BRT cessation. We show that circulating free thyroxine (FT4)
is a determinant of resting energy expenditure (REE) in euthyroid patients treated with BRT for
Graves’ hyperthyroidism. A majority of patients report weight gain compared to their supposed
pre-morbid body weight (BW) following treatment of hyperthyroidism, which was also the
case in our study. We anticipated that if this weight gain was due to BRT, weight loss should
occur after cessation of BRT. However, 12 weeks after cessation of BRT, body weight (BW) and
energy homeostasis were unaltered. Nevertheless, there were subtle differences in serum TH
concentrations, albeit in the euthyroid range, between the 2 study occasions, i.e., serum FT3 as
well as the FT3/FT4 ratio showed an increase 12 w after cessation of BRT. As the serum FT3/
FT4 ratio is a positive determinant of changes in REE, a longer term decrease in BW is likely to
occur.
In chapter 7 we aim to place our findings into a wider perspective. We describe the views on the
interrelationship between thyroid hormone and the sympathetic nervous system in a historical
context. Subsequently, we try and adapt these views, fitting in our findings of hypothalamic
thyroid hormone modulation of hepatic glucose metabolism via sympathetic hepatic projections.
We discuss the potential biological relevance of these novel hypothalamic thyroid hormone
actions in terms of both thyrotoxicosis and more physiological conditions. We propose that
modulation of autonomic signalling by hypothalamic T3 may provide a new level of complexity
in the (rapid) regulation of peripheral metabolism by thyroid hormone. Finally, potential clinical
implications and future directions of research are discussed.
proefschrift Klieverik.indb 127
4-8-2009 15:23:19
proefschrift Klieverik.indb 128
4-8-2009 15:23:19
Nederlandse samenvatting
Chapter 8
129
Nederlandse samenvatting
Hyperthyreoïdie wordt gekenmerkt door een breed spectrum van klinische symptomen
en metabole veranderingen. Veel van deze symptomen en veranderingen tonen een
grote overeenkomst met veranderingen die optreden als gevolg van activatie van het
sympathische zenuwstelstel. Deze overeenkomst werd al lang voor de ontdekking en isolatie
van schildklierhormoon opgemerkt, en wordt ook tegenwoordig nog gereflecteerd in de
behandeling van ernstige hyperthyreoïdie met medicamenteuze remming van de sympatische
signaaloverdracht door middel van β-blokkers. De laatste jaren is steeds meer duidelijk geworden
dat de hypothalamus in belangrijke mate bijdraagt aan de regulatie van de stofwisseling via
autonome projecties naar een groot aantal perifere organen, zoals de lever.
In dit proefschrift hebben we de metabole veranderingen tijdens hyperthyreoïdie bestudeerd en
daarnaast bekeken in hoeverre schildklierhormoon effecten op het niveau van de hypothalamus
en het autonome zenuwstelsel betrokken zijn in deze metabole veranderingen. Onze hypothese
was dat schildklierhormoon (en schildklierhormoon-derivaten) de activiteit van het autonome
zenuwstelsel beïnvloeden via effecten in het brein, of meer specifiek via effecten in de hypothalamus,
en zodoende de stofwisseling moduleren op het niveau van perifere organen zoals de lever.
Hoofdstuk 1 geeft een overzicht van het schildklierhormoon metabolisme en de wisselwerking
tussen schildklierhormoon en het brein. Tevens worden huidige inzichten over de metabole
veranderingen tijdens hyperthyreoïdie besproken, met nadruk op isotooptechnieken die we in
dit proefschrift gebruikten om deze metabole veranderingen te bestuderen. De hypothalamus
is een belangrijke regulator van het metabolisme via verbindingen met het autonome
zenuwstelsel. De functionele anatomie van de hypothalamus en het autonome zenuwstelsel
en hun rol in de regulatie van het glucose metabolisme worden uiteengezet. Verder bevat
hoofdstuk 1 een korte introductie over thyronamines, schildklierhormoon-analogen met snelle
metabole effecten in vivo en ten slotte worden de algemene hypothese en inhoud van het
proefschrift gepresenteerd.
In hoofdstuk 2 bestudeerden we de effecten van hyper- en hypothyreoïdie op de energiestofwisseling. We vonden reciproke veranderingen in hyperthyreote en hypothyreote, vergeleken
met euthyreote ratten. Hyperthyreote ratten tonen een verhoogd energieverbruik en vet
oxidatie zonder veranderingen in spontane fysieke activiteit. Hypothyreote ratten tonen een
geringe afname in energieverbruik en glucose oxidatie. Daarnaast hebben we bestudeerd
hoe de veranderingen in vet oxidatie gepaard gaan met weefsel-specifieke vet opname. De
resultaten tonen dat het hypermetabole fenotype tijdens hyperthyreoïdie gefaciliteerd wordt
door een toegenomen opname van vetzuren afkomstig uit triglyceriden in oxidatieve weefsels,
terwijl de opname van dezelfde vetzuren in vetweefsel onveranderd is. Omgekeerd is tijdens
hypothyreoïdie de opname van vetzuren afkomstig van triglyceriden onveranderd in oxidatieve
weefsels, maar juist toegenomen in vetweefsel. Daarbij is er een lokaal toegenomen activiteit
van het enzym lipoproteïne lipase, dat het vrijmaken van vetzuren uit triglyceriden reguleert.
Opname van albumine-gebonden vetzuren lijkt niet onderhevig aan weefsel-specifieke regulatie,
proefschrift Klieverik.indb 129
4-8-2009 15:23:19
maar voornamelijk bepaald door de concentratie van vetzuren in het plasma, welke verhoogd
is tijdens hyperthyreoïdie.
130
In hoofdstuk 3, hebben we de rol van de autonome projecties naar de lever in veranderingen
van het hepatische glucose metabolisme tijdens hyperthyreoïdie bestudeerd. We bepaalden de
endogene glucose productie en gevoeligheid voor onderdrukking door insuline (hepatische insuline
gevoeligheid) in euthyreote en hyperthyreote ratten die selectieve autonome (sympathische of
parasympathische) lever-denervatie of een sham-operatie ondergingen. De resultaten tonen dat
de veranderingen in het hepatische glucose metabolisme tijdens hyperthyreoidie differentieel
gemoduleerd kunnen worden door sympathische dan wel parasympatische lever-denervatie.
Meer specifiek kunnen de toename in plasma glucose concentratie en (in mindere mate) de
toename in endogene glucose productie geïnduceerd door hyperthyreoïdie, deels worden
voorkomen door voorafgaande sympathische lever denervatie. Dit suggereert dat sympathische
innervatie bijdraagt aan de toegenomen plasma glucose concentratie en endogene glucose
productie tijdens hyperthyreoïdie. Parasympathische lever denervatie verhoogt de plasma
insuline concentratie, maar niet de endogene glucose productie tijdens hyperthyreoïdie. De door
hyperthyreoïdie geïnduceerde hepatische insuline-resistentie wordt dus verergerd door selectie
parasympathische denervatie. Dit suggereert dat parasympathische innervatie van de lever de
endogene glucose productie tijdens hyperthyreoïdie beteugelt.
Als volgende stap, was in hoofdstuk 4 onze hypothese dat de door hyperthyreoïdie
geïnduceerde toename van endogene glucose productie kan worden gereproduceerd
door schildklierhormoon toe te dienen in de hypothalamus van euthyreote ratten. Om deze
hypothese te testen dienden we het bio-actieve schildklierhormoon triiodothyronine (T3) toe
in de hypothalamus van euthyreote ratten, terwijl we de endogene glucose productie maten
met behulp van stabiele isotoop dilutie. We dienden T3 selectief toe in de paraventriculaire
nucleus (PVN) in de hypothalamus. Deze kern herbergt de “pre-autonome” neuronen die de
sympatische en parasympatische projecties naar de lever aansturen. De resultaten tonen dat T3
in de PVN de endogene glucose productie in korte tijd doet toenemen, onafhankelijk van plasma
concentraties van schildklierhormonen en andere glucoregulatoire hormonen. Daarnaast blijkt
uit experimenten met hypothalame T3 toediening in combinatie met selectieve sympathische
lever denervatie dat intacte sympathische projecties naar de lever essentieel zijn voor de snelle
stimulatie van endogene glucose productie door hypothalaam T3. Concluderend onthullen deze
data een nieuw centraal mechanisme voor modulatie van de endogene glucose productie door T3,
waarbij de hypothalame paraventriculaire kern en het sympathische zenuwstelsel betrokken zijn.
Thyronamines zijn schildklierhormoon-derivaten met neurotransmitter-achtige eigenschappen en
het fenotype dat snel na toediening van thyronamines ontstaat, wijst op mogelijke betrokkenheid
van de hypothalamus in deze effecten. In hoofdstuk 5 testten we onze hypothese dat de
effecten van 3-iodothyronamine (T1AM) en thyronamine (T0AM) op het hepatisch glucose
metabolisme en circulerende glucoregulatoire hormonen kunnen worden verklaard door
werking van thyronamines in het centraal zenuwstelsel. We tonen dat systemische toediening
proefschrift Klieverik.indb 130
4-8-2009 15:23:19
131
Nederlandse samenvatting
In hoofdstuk 6 hebben we de relatie tussen schildklierhormoon en het energie-metabolisme
uitgediept in de klinische setting. We hebben een studie verricht in patiënten met de ziekte van
Graves die euthyreoot zijn na behandeling met zogenaamde ‘block and replacement therapie’
(BRT). Dit betekent dat de schilklierhormoon productie medicamenteus geblokkeerd wordt, en
tegelijkertijd suppletie van schildklierhormoon tot euthyreote concentraties plaatsvindt door
middel van exogene toediening van thyroxine. We hebben deze patiënten op 2 gelegenheden
onderzocht: na ~13,5 maand BRT, en 12 weken na stoppen van deze behandeling. We tonen
dat de serum concentratie ongebonden thyroxine (“vrij” T4 ofwel FT4) een determinant is van het
energieverbruik in rust in met BRT behandelde, en hierdoor euthyreote Graves patiënten. Een
meerderheid van de patiënten rapporteert een toename in lichaamsgewicht na het starten met
de behandeling van hyperthyreoïdie (met BRT) tot boven het lichaamsgewicht van voor de ziekte.
Dit bleek ook het geval in onze studie. We veronderstelden dat indien deze gewichtstoename
het gevolg zou zijn van (factoren gerelateerd aan) BRT, afname van lichaamsgewicht op zou
moeten treden na stoppen van BRT. Tegen de verwachting in waren er 12 weken na het stoppen
van BRT geen veranderingen in de energie stofwisseling of het lichaamsgewicht waarneembaar
bij onze patiënten. Wel waren er subtiele verschillen in schildklierhormoonconcentraties in serum
tussen de twee studiegelegenheden binnen het euthyreote spectrum, i.e. zowel het serum vrij
T3 (FT3) als de serum FT3/FT4 ratio namen toe 12 weken na het stoppen van BRT. Omdat de
serum FT3/FT4 ratio een positieve determinant is van veranderingen in energieverbruik in rust, is
te verwachten dat een gewichtsafname op langere termijn op zal treden.
Chapter 8
van T1AM wordt gevolgd door een snelle toename in endogene glucose productie, plasma
glucose, glucagon en corticosteron concentraties, zonder veranderingen in plasma insuline
concentratie. Vergeleken met systemische toediening van T1AM, induceert een honderd maal
lagere dosis intracerebroventriculair T1AM een meer uitgesproken stijging van de endogene
glucose productie en een toename in plasma glucagon, terwijl de insuline concentratie in het
plasma neigt te dalen. Zowel systemische als centrale infusie van T0AM induceert dezelfde
veranderingen als T1AM, zij het in mindere mate. Systemische toediening van T1AM noch T0AM
in lage dosering sorteert verandering in de genoemde parameters. Deze data tonen dat centrale
toediening van thryonamines in lage dosis volstaat om de acute veranderingen in het glucose
metabolisme en glucoregulatoire hormonen die het gevolg zijn van systemische thyronamine
infusie, te repliceren. Thyronamines zijn dus in staat het glucose metabolisme te beïnvloeden via
effecten in het centraal zenuwstelsel.
In hoofdstuk 7 plaatsen we onze bevindingen in een breder perspectief. We beschrijven
verschillende ideeën over de relatie tussen schildklierhormoon en het sympathische zenuwstelsel
in historische context. We beschrijven hoe deze ideeën op basis van onze bevindingen over
modulatie van hepatisch glucose metabolisme door hypothalaam T3 met betrokkenheid van
sympathische projecties zouden moeten worden aangepast. Verder gaan we in op de mogelijke
biologische relevantie van de beschreven schildklierhormoon effecten op hypothalaam niveau met
betrekking tot hyperthyreoïdie en andere, meer fysiologische condities. Modulatie van autonome
activiteit door hypothalaam T3 zou een nieuw niveau van complexiteit kunnen betekenen in de
proefschrift Klieverik.indb 131
4-8-2009 15:23:19
(snelle) regulatie van het metabolisme in perifere organen. Tenslotte worden mogelijke klinische
implicaties en richtingen voor toekomstig onderzoek besproken.
132
proefschrift Klieverik.indb 132
4-8-2009 15:23:19
Author affiliations
Anita Boelen
Department of Endocrinology and Metabolism
Academic medical center, University of Amsterdam, the Netherlands
[email protected]
133
Author Affiliations
Peter H. Bisschop
Department of Endocrinology and Metabolism
Academic medical center, University of Amsterdam, the Netherlands
[email protected]
Chapter 8
Mariëtte T. Ackermans
Department of Clinical Chemistry, Laboratory of Endocrinology and Radiochemistry,
Academic medical center, University of Amsterdam, the Netherlands
[email protected]
Claudia P. Coomans
Department of Endocrinology and Metabolic Diseases,
Leiden University Medical Center, Leiden, the Netherlands
[email protected]
Erik Endert
Department of Clinical Chemistry, Laboratory of Endocrinology and Radiochemistry,
Academic medical center, University of Amsterdam, the Netherlands
[email protected]
Eric Fliers
Department of Endocrinology and Metabolism
Academic medical center, University of Amsterdam, the Netherlands
[email protected]
Ewout Foppen
Netherlands Institute for Neuroscience, Amsterdam, the Netherlands
[email protected]
David K. Grandy
Department of Physiology and Pharmacology, School of Medicine
Oregon Health and Science University, Portland, Oregon, USA
[email protected]
Louis M. Havekes
Department of Endocrinology and Metabolic Diseases,
Leiden University Medical Center, Leiden, the Netherlands
[email protected]
Sarah F. Janssen
Department of Endocrinology and Metabolism
Academic medical center, University of Amsterdam, the Netherlands
[email protected]
proefschrift Klieverik.indb 133
4-8-2009 15:23:19
Andries Kalsbeek
Department of Endocrinology and Metabolism
Academic medical center, University of Amsterdam, the Netherlands
and
Netherlands Institute for Neuroscience, Amsterdam, the Netherlands
[email protected]
134
Lars P. Klieverik
Department of Endocrinology and Metabolism
Academic medical center, University of Amsterdam, the Netherlands
[email protected]
Patrick C.N. Rensen
Department of Endocrinology and Metabolic Diseases,
Leiden University Medical Center, Leiden, the Netherlands
[email protected]
Annelieke van Riel
Department of Endocrinology and Metabolism
Academic medical center, University of Amsterdam, the Netherlands
[email protected]
Johannes A. Romijn
Department of Endocrinology and Metabolic Diseases,
Leiden University Medical Center, Leiden, the Netherlands
[email protected]
Hans P. Sauerwein
Department of Endocrinology and Metabolism
Academic medical center, University of Amsterdam, the Netherlands
[email protected]
Thomas S. Scanlan
Department of Physiology and Pharmacology, School of Medicine
Oregon Health and Science University, Portland, Oregon, USA
[email protected]
Mireille J. Serlie
Department of Endocrinology and Metabolism
Academic medical center, University of Amsterdam, the Netherlands
[email protected]
Peter J. Voshol
Department of Endocrinology and Metabolic Diseases,
Leiden University Medical Center, Leiden, the Netherlands
[email protected]
Wilmar M. Wiersinga
Department of Endocrinology and Metabolism
Academic medical center, University of Amsterdam, the Netherlands
[email protected]
proefschrift Klieverik.indb 134
4-8-2009 15:23:19
Dankwoord
Prof. dr. E. Fliers. Beste Eric, toen ik als co-assistent voor het eerst bij je langs kwam om te
praten over het onderzoeksvoorstel dat uiteindelijk mijn promotietraject zou worden, wist ik
eigenlijk onmiddelijk dat het goed zat. Enerzijds door het onderwerp, maar zeker ook door de
enorme bevlogenheid waarmee je me meteen wist in te pakken. Hoewel ik toen niet goed kon
weten wat er komen zou, heb ik er nooit spijt van gekregen. Mijn project was tot het laatst
een “rollercoaster” waarbij de inzichten in korte tijd 180 graden konden keren. Regelmatig
was ik ook wanhopig of ronduit gefrustreerd na het mislukken van weer een eindeloze proef.
Maar je wist me te overtuigen dat ook uit een “mislukte” proef veel waardevolle informatie te
halen was. Zonder jouw optimisme, vastberadenheid en betrokkenheid was het voltooien van
dit proefschrift voor mij onmogelijk geweest. Je gaf me steeds het gevoel dat het niet “mijn”,
maar “ons” project was, bijvoorbeeld als je in witte jas, tijdens een proef mijn dierverblijf binnen
kwam rennen met de vraag “..of ik al iets wist?”. Daarnaast zijn jouw nuance, fijnzinnigheid en
overtuigingskracht voor mij echt een voorbeeld. Ik heb het enorm gewaardeerd dat ik altijd bij je
kon binnenlopen voor het bespreken van persoonlijke beslommeringen, recente ontwikkelingen
en data, die we meestal ook in die volgorde bespraken. Dank voor deze mooie tijd. Ik kan me
geen betere mentor wensen, en hoop dat we in de toekomst nog veel samen mogen doen!
135
Dankwoord
Mijn promotores en co-promotores
Chapter 8
Een goede vriend vroeg me het laatste jaar regelmatig “wanneer dat opstel nou eens klaar zou
zijn..”. Niet helemaal uit de lucht gegrepen, want zonder de hulp van een groot aantal collega’s
en vrienden was dit boekje het predikaat “opstel” waarschijnlijk nooit ontstegen. En wanneer
krijg je nou de kans om in het openbaar deze mensen te bedanken, je bewondering uit te
spreken en te benoemen waarom ze zo belangrijk waren? Daarom hier mijn woord van dank
aan allen die zo betrokken waren.
Dr. A. Kalsbeek. Beste Dries, hoewel het begin niet zonder problemen was, heb je me altijd met
engelengeduld bijgestaan en gesteund. Je bent betrokken, recht-door-zee, nuchter (iets wat
Friezen en Tukkers gemeen schijnen te hebben..), en bij jou volgt op een oprechte vraag altijd
een eerlijk antwoord. Ik heb enorme bewondering voor je vasthoudendheid in het stap voor stap
ontrafelen van extreem ingewikkelde neurobiologische processen, en voor je vermogen uit een
ogenschijnlijk onsamenhangende berg data een logisch verhaal te destilleren. De combinatie van
jou en Eric is voor mij een heel goede geweest, en ik ben blijkbaar niet de enige, getuige je komst
naar het AMC. Jij hebt een heel groot aandeel gehad in dit boekje. Dank voor alles, ik ben heel
trots dat je mijn co-promotor bent!
Prof. dr. H.P. Sauerwein. Beste Hans, als AIO op een project dat zich uitstrekt over meerdere
onderzoeksgebieden, heb ik mogen putten uit de expertise van 3 onderzoeksgroepen. Ik heb me,
hoewel ergens toch een buitenstaander, vanaf het begin deel gevoeld van de “metabole groep”.
Mijn manuscripten hebben zeer geprofiteerd van jouw frisse commentaar en aanvullingen,
waarvoor ik je erg dankbaar ben. Je bent scherp en stellig, maar staat altijd open voor goede
tegenargumenten (hoewel ze heel goed moeten zijn..). Jij bent iemand waarover iedereen een
proefschrift Klieverik.indb 135
4-8-2009 15:23:19
mening lijkt te hebben, en dat tekent je persoonlijkheid. Eigenzinnig, aanwezig en bevlogen.
Weet dat je gemist wordt.
136
Dr. ir. M.T. Ackermans. Beste Mariëtte, jij bent de absolute spil geweest in het analytische deel
van dit project, maar niet alleen daarom onmisbaar. De manier waarop je mij en andere AIO’s
steeds weer (de beginselen van) de isotopenleer en andere assays uitlegde, is lovenswaardig.
Ondanks dat we de champagne niet hebben kunnen ontkurken (doen we de 17e september
alsnog!), vond ik het mooi en spannend betrokken te zijn bij het opzetten van de thyronamineMS. Ik heb je tot wanhoop gedreven met mijn on(na)volgbare samplecoderingen. Dank voor je
geduld en tolerantie. Nu er enkel nog ordelijke vrouwen op het endolab werken, kan het alleen
maar beter worden!
Hoofdstuk 6 was er nooit gekomen zonder de vrijwillige deelname van de betrokken patiënten
die geheel onbaatzuchtig hun tijd en moeite gaven, waarvoor ik hen erg dankbaar ben.
Graag dank ik prof. dr. R.M. Buijs, prof. dr. M.M. Levi, prof. dr. R.P.J. Oude Elferink, prof. dr. J.A.
Romijn, prof. dr. ir. T.J. Visser en prof. dr. W.M. Wiersinga voor het kritisch beoordelen van mijn
manuscript en hun bereidheid zitting te nemen in mijn promotiecommissie.
Fijne collega’s
Ondanks het feit dat ik lange tijd eigenlijk part-time bewoner was, heb ik me op afdeling F5
altijd erg thuis gevoeld. Marga, Marlies en Birgit, hartelijk dank voor alle steun, vooral ook bij de
laatste loodjes, en alle gezelligheid!
Peter en Mireille, veel dank voor jullie steun en kritische blik bij mijn stukken in wording. Peter,
jij bent de onbetwiste winnaar van het BRS inclusie classement; erg bedankt! Anke, Nadia,
Maarten, Regje, Saskia, Hidde, Mirjam, Gabor, Martine, Nicolette en Myrte; met jullie heb ik de
afgelopen jaren heel veel mogen delen. Dank voor alle gezelligheid, het lotgenotencontact, het
aanhoren van mijn verhalen en beleefd lachen om mijn slechte grappen!
Xander Vos. Beste Xander, mijn eerste herinnering aan jou is een bankje in de broeierige
binnenstad van Napoli waar we de Napolitaanse cultuur, flora en fauna opsnoven. Sindsdien is er
extreem veel gebeurd. Wij zijn in korte tijd goede vrienden geworden en ik hoop dat dit, ondanks
jouw verhuizing naar de Zaanse steppes, zo mag blijven. Leve de Polderbeukers!
Prof. W.M. Wiersinga. Beste Wilmar, het is uniek om als AIO te beginnen in een groep geleid
door een mondiaal endocrinologisch zwaargewicht. Ik herinner me Graves ophthalmopathie
sessies op de Endocrine Society congressen in Amerika waar vragenstellers vertwijfeld rondkeken
als ze eens iemand anders dan jij op de voorzittersstoel troffen, om vervolgens de werkelijke
voorzitter volledig te negeren en gewoon met jou, de èchte expert, in discussie te gaan. Enorm
bedankt voor alle steun en warme belangstelling. Het is een eer dat je zitting gaat nemen in mijn
promotiecommissie.
Chapter 8.indd 136
6-8-2009 11:07:23
Chapter 8
137
Dankwoord
Collega’s in het Nederlands instituut voor neurowetenschappen:
Beste Ewout, als dit boekje een tweede auteur mocht hebben, was jij dat. Hoe vaak hebben wij
samen geopereerd, ben je even ingesprongen voor een “paar” monsters, en heb je glucose en
corticosteron voor me gemeten. Ik ben je hiervoor enorm dankbaar. Temeer omdat je me altijd
hebt ontzien in je voorkeur voor snoeiharde Oostblok punk!
Superstudenten: Sarah, Annelieke en Rianne: Wat hebben jullie een monnikenwerk verricht voor
de experimenten in hoofdstuk 3 en 4. Jullie drive en doorzettingsvermogen zijn indrukwekkend.
Het was mooi, ik ben heel blij dat ik jullie getroffen heb!
Chun Xia, Marieke, Cathy, Corbert, Ajda, Valerie and Evelien, fellow-slaves of neuroscience! Your
presence has been essential for me to complete this project. Many times, especially in the first
years, I was desperate after another experiment gone bad, but a cup of tea while hearing I was
not the only one with such misery always did the job. Thank you for that, I hope to raise a glass
with you all on the 17th of September!
Felix, het waren jouw mooie stukken waardoor ik enthousiast werd voor dit onderwerp, en
uiteindelijk bij Eric aanklopte. Ik hoop dat ik ooit, op een mooie dag, de meester tijdens een
retro-peritoneale vet denervatie mag aanschouwen! Jan en Carolien, dank voor jullie analytische
hulp in de eerste jaren. Jillis, Chris en Nanneke: veel dank voor het scheppen van optimale
omstandigheden voor het doen van goede dierexperimenten.
Collega’s van het laboratorium voor Endocrinologie op F2:
Mieke, Ivo, Clementine, Marianne, Anneke en Joan; dank voor alle plezier en gezelligheid. Maar
wanneer mag die poster nou weg? Anneke; dank voor de fijne samenwerking en de fietsles in
San Francisco, ik hoop dat ons werk nog een staartje krijgt! Joan; dank voor je deskundigheid en
engelengeduld bij mijn PCR-bezigheden. An, Barbara, Marjo, Marianne en Els; met veel nadruk
wil ik jullie bedanken voor het meten van vele honderden monsters. Ik weet niet wat ik had
gemoeten zonder jullie; heel hartelijk dank! Eric Endert, veel dank voor je steun en de prioriteit
die je mijn project altijd hebt gegeven.
Dr. A. Boelen. Lieve Anita, er zijn van die mensen met wie het meteen klikt, en die eigenlijk nooit
iets fout kunnen doen. Ik heb erg genoten van alle gezelligheid in Boston, Toronto, Napels, San
Francisco en Washington, de vele theesessies over alles behalve onderzoek en de tochten op
natuurijs deze winter. Het is fantastisch te zien hoe jij in rap tempo aan je eigen onderzoekslijn
bouwt. Alle geluk!
A special thank to dr. Tom Scanlan and dr. David Grandy for an extremely exciting collaboration
on thyronamines.
In de laatste fase van mijn project ben ik een aantal maanden geadopteerd door fijne collega’s van
de afdeling Endocrinologie en Metabole Ziekten van het Leids Universitair Medisch Centrum.
Prof. dr. J.A. Romijn. Beste Hans, vanaf de eerste dag heb ik me meer dan welkom gevoeld
in het Leidse. Jouw hartelijke voorkomen en warme belangstelling hebben daar enorm aan
bijgedragen. Zeer veel dank voor je steun op meerdere vlakken. Het is een voorrecht dat je plaats
zal nemen in mijn promotiecommissie op 17 september.
proefschrift Klieverik.indb 137
4-8-2009 15:23:19
Beste Claudia, wij hebben samen echt heel mooie proeven gedaan. Dank voor de fijne
samenwerking en alle gezelligheid! Patrick, Marieke, Sjoerd en Peter, heel veel dank voor het
op gang helpen van de metabole kooi experimenten, de technische ondersteuning en jullie
expertise.
Vrienden en familie
138
Koen de Heer. Beste Koen, ik vind het geweldig dat jij me tijdens mijn verdediging bijstaat als
paranimf. Als polderbeukers-on-tour hebben we veel meegemaakt. Ik noem jouw lijdensweg naar
Alpe d’huez, die je als anti-klimmer (zoals het een echte polderbeuker betaamt), zonder dat colainfuus achteraf niet had overleefd, en de nimmer bevestigde afdaling zonder bril, achtervolgd
door een dolle stier in het centraal massief. We bewandelen tegengestelde paden; nu ik bijna
klaar ben met mijn promotie en aan mijn opleiding begin, rondt jij je opleiding binnenkort af en
begint je promotie-traject. Ik hoop dat we over een aantal jaar weer samen in het AMC mogen
werken!
Ferring pharmaceuticals, dank voor de (financiële) injecties; zonder jullie waren de Polderbeukers
een groepje welwillende wieler-amateurs gebleven.
Bas, Jurgen, Robert en Vincent, wij kennen elkaar al sinds onze vroege jeugd en hoewel de
meesten zijn uitgewaaierd, is onze vriendschap altijd gebleven. De laatste tijd is er veel gebeurd,
en ik vind het heel fijn dat het voor mij geen vraag is wie mijn trouwste vrienden zijn.
Lieve Mirthe, mijn kleine zusje. Ik ben heel blij dat jij zo’n prominente plaats hebt in mijn leven,
en hoop dat we nog heel veel mooie dingen gaan beleven samen. Ik ben trots dat je me met al
je wijsheid bijstaat op 17 september (..parawattes?!).
Lieve mam en pap, dit boekje begon met jullie, en niet voor niets. Zolang ik me kan herinneren
hebben jullie me gestimuleerd mijn hart te volgen in alles wat ik deed en zijn jullie er voor
me geweest. Jullie trots is onbetaalbaar. Liefste mam, iedereen die bij ons kwam werd
altijd ondergedompeld in jouw warmte en enthousiasme, en er is niets veranderd. Ik hou
onvoorwaardelijk van je. Lieve pap, ik ben bang dat wij meer op elkaar lijken dan jij soms denkt.
Hoewel de onderzoekswereld niet de jouwe is, ben je altijd vol belangstelling en kan ik eindeloos
met je discussiëren over wat dan ook. Wanneer gaan we samen weer het hooggebergte in? Dit
boekje is voor jullie.
Als laatste de voor mij belangrijkste.
Lieve Madeleine, als ik dit schrijf is Stan pas 2 dagen bij ons, en het lijkt allemaal niet op te
kunnen. Wil nog eindeloos veel meer van dit...
Ik hou zielsveel van jou.
Chapter 8.indd 138
6-8-2009 11:08:00
Biografie
139
Biografie
proefschrift Klieverik.indb 139
Chapter 8
Lars Peter Klieverik werd op 17 maart 1978 geboren, en groeide op in het Twentse Borne.
Na volgen van het basisonderwijs aan de plaatselijke Flora school, behaalde hij in 1996 zijn
Atheneum diploma aan het Twickel college in Hengelo (ov). In hetzelfde jaar begon hij aan een
studie medische biologie aan de Universiteit van Amsterdam, waarvan hij in 1997 het propedeuse
examen behaalde. Kort daarop werd hij ingeloot voor zijn eerste keus Geneeskunde aan dezelfde
universiteit. Tijdens zijn studie werkte hij lang als barman in een Amsterdams grachthotel en was
drummer in meerdere pop-rock bands. Daarnaast speelde hij jarenlang in het eerste herenteam
van Volleybal Vereniging Amsterdam en was hij actief als student-assistent in een grote studie
naar het gebruik van automatische defibrillatoren door de Amsterdamse brandweer en politie
(afdeling cardiologie, Academisch Medisch Centrum (AMC)). Zijn wetenschappelijke stage
volgde hij in het laboratorium voor Celbiologie en Histologie van prof. C.J. van Noorden en dr.
W.M. Frederiks in het AMC, waar hij onderzoek deed naar de effecten van een visolie dieet
op de metastasering van colonkanker. Uit dit onderzoeksproject kwam een tweetal publicaties
voort. In 2004 behaalde hij zijn artsexamen, waarna hij startte met zijn promotie-onderzoek als
beleids-AIO op de afdeling Endocrinologie en Metabolisme van het AMC (prof. dr. E. Fliers en
prof. dr. H.P. Sauerwein), in nauwe samenwerking met de vakgroep hypothalame integratie
mechanismen van het Nederlands instituut voor neurowetenschappen (dr. A. Kalsbeek). Op 1
april 2009 is hij begonnen met de opleiding tot internist in het opleidingscluster AMC (opleider
Prof. dr. P. Speelman), startend in het Flevoziekenhuis te Almere (opleider dr. S. Peters). Op 7 mei
2009 trouwde hij Madeleine Sombekke en samen kregen zij op 29 juli 2009 een zoon: Stan.
4-8-2009 15:23:19