Download Liver-Derived IGF-I Regulates GH Secretion at the Pituitary Level in

0013-7227/01/$03.00/0
Printed in U.S.A.
Endocrinology 142(11):4762– 4770
Copyright © 2001 by The Endocrine Society
Liver-Derived IGF-I Regulates GH Secretion at the
Pituitary Level in Mice
KRISTINA WALLENIUS, KLARA SJÖGREN, XIAO-DING PENG, SEUNGJOON PARK,
VILLE WALLENIUS, JUN-LI LIU, MIA UMAERUS, HÅKAN WENNBO, OLLE ISAKSSON,
LAWRENCE FROHMAN, RHONDA KINEMAN, CLAES OHLSSON, AND JOHN-OLOV JANSSON
Research Centre for Endocrinology and Metabolism (K.W., K.S., V.W., O.I., C.O., J.-O.J.), Sahlgrenska University Hospital,
Göteborg SE-413 45, Sweden; Department of Medicine (X.-D.P., S.P., L.F., R.K.), Section of Endocrinology and Metabolism,
University of Illinois, Chicago, Illinois 60612; Department of Medicine (J.-L.L.), McGill University, Montréal QCH3A1A1,
Canada; AstraZeneca R&D (M.U., H.W.), SE-43183 Mölndal, Sweden
We have reported that liver-specific deletion of IGF-I in mice
(LI-IGF-Iⴚ/ⴚ) results in decreased circulating IGF-I and increased GH levels. In the present study, we determined how
elimination of hepatic IGF-I modifies the hypothalamic-pituitary GH axis to enhance GH secretion. The pituitary mRNA
levels of GH releasing factor (GHRF) receptor and GH secretagogue (GHS) receptor were increased in LI-IGF-Iⴚ/ⴚ mice,
and in line with this, their GH response to ip injections of
GHRF and GHS was increased. Expression of mRNA for pituitary somatostatin receptors, hypothalamic GHRF, somatostatin, and neuropeptide Y was not altered in LI-IGF-Iⴚ/ⴚ
G
mice, whereas hypothalamic IGF-I expression was increased.
Changes in hepatic expression of major urinary protein and
the PRL receptor in male LI-IGF-Iⴚ/ⴚ mice indicated an altered GH release pattern most consistent with enhanced GH
trough levels. Liver weight was enhanced in LI-IGF-Iⴚ/ⴚ mice
of both genders. In conclusion, loss of liver-derived IGF-I enhances GH release by increasing expression of pituitary
GHRF and GHS receptors. The enhanced GH release in turn
affects several liver parameters, in line with the existence of
a pituitary-liver axis. (Endocrinology 142: 4762– 4770, 2001)
H SECRETION IN rodents is sexually dimorphic and
pulsatile (1). Male rats have episodic bursts of GH
secretion and low GH levels between the pulses, and female
rats have higher basal interpulse GH levels and more frequent but lower amplitude pulses (2). Pituitary GH secretion
is regulated by hypothalamic GH releasing factor (GHRF)
and somatostatin. Increased hypothalamic GHRF secretion is
followed by GH pulses, and somatostatin secretion increases
during GH troughs (3, 4).
Several studies have shown that pharmacological treatment with high doses of IGF can inhibit GH secretion in both
man and rodents (5– 8). The mechanisms mediating the inhibitory effect of IGF-I on GH secretion have been studied in
GH-deficient animals treated with IGF-I or in vitro using
pituitary cell cultures. There are, however, no studies on the
mechanisms mediating IGF-I feedback in animals with intact
GH secretion.
In primary pituitary cell cultures, IGF-I has been shown to
suppress both basal and GHRF-stimulated GH release and
synthesis. Thus, it has been suggested that IGF-I can inhibit
the stimulatory effect of GHRF on GH release directly at the
pituitary level (9, 10). In GH-deficient rodents, IGF-I treatment suppresses the increased GHRF receptor expression in
these animals (11). On the other hand, central administration
of IGF-I to GH-deficient rats decreases GHRF and increases
somatostatin expression, suggesting that IGF-I can also act at
the hypothalamic level (12). In line with this, it has been shown
that IGF-I is locally produced in the hypothalamus (13).
IGF-I could also inhibit GH secretion by regulating the
expression of the GH secretagogue receptor (GHS-R). Activation of the GHS-R with synthetic GH secretagogues (GHS),
or the endogenous ligand Ghrelin, induces GH secretion (14,
15). GHS-R activation stimulates GH release directly at the
pituitary level (16) and increases hypothalamic GHRF release
and may inhibit hypothalamic somatostatin release (14, 17).
It is unclear, however, whether endogenous GHS-R ligands
contribute to the regulation of GH pulsatility.
The secretory pattern of GH regulates several sexually dimorphic liver functions in rodents, such as expression of major
urinary protein (MUP) and the PRL receptor (PRL-R) (18 –21).
MUP is expressed at about three times higher levels in livers of
male, compared with female rodents, and PRL-Rs are expressed
at higher levels in females (18). Furthermore, continuous treatment of male mice with GH leads to suppression of MUP and
induction of PRL-R expression (18). Therefore, MUP and PRL-R
expression are markers of GH trough levels.
Mice with liver-specific IGF-I knockout have 80% decreased serum IGF-I levels and increased circulating GH
levels (22, 23). In the present study, we have investigated
how elimination of hepatic IGF-I modifies the hypothalamicpituitary GH axis to enhance GH secretion.
Abbreviations: GAPDH, Gyceraldehyde-3-phosphate dehydrogenase; GHRF, GH releasing factor; GHRF-R, GHRF receptor; GHS, GH
secretagogues; GHS-R, GH secretagogue receptor; LI-IGF-I⫺/⫺, liverspecific IGF-I knockout; MUP, major urinary protein; PRL-R, PRL receptor; RPA, ribonuclease protection assay; sst, somatostatin receptor.
Animals
Materials and Methods
Transgenic mice were bred and recombination was induced by interferon treatment at 4 wk of age, as described earlier (22). Interferontreated siblings, homozygous for loxP but lacking Mx-Cre, were used as
4762
Wallenius et al. • Liver-Pituitary IGF-I Feedback
Endocrinology, November 2001, 142(11):4762– 4770 4763
controls. The mice were killed by cervical dislocation at 2.5–3 months of
age and organs were collected and weighed and then immediately snap
frozen in liquid nitrogen. All experiments were conducted in accordance
with institutional guidelines and were approved by the local committee
for animal care.
RIA of IGF-I and GH
Plasma was obtained by centrifuging heparinized capillaries with
blood obtained from the tip of the tail of unanesthetized mice at different
times throughout the day. Plasma IGF-I levels were measured 3 wk after
interferon treatment by a double-antibody IGF binding protein-blocked
RIA according to Blum and Breier (24). Mouse GH levels were measured
by RIA (RPA 551, purchased before November 1999; Amersham Pharmacia Biotech, Little Chalfont, UK), according to the manufacturer’s
instructions, with a detection range of 1.3–100 ng/ml. Mouse GH was
also measured (see Fig. 5B) as described previously (25) using reagents
kindly supplied by NHPP, NIDDK, and Dr. Parlow.
Measurement of MUP
One microliter of urine from mice was run on 10% NuPage Bis-Tris
gels in MES-buffer (Novex, San Diego, CA) (18). MUP, with a size of
about 20 kDa, was the dominant protein on gels. Gels were Coomassie
stained and MUP was quantified by scanning and densitometric analysis
using ImageQuant software (Molecular Dynamics, Inc., Sunnyvale, CA).
Because an overwhelming proportion of all protein in urine was MUP,
measurement of total protein with the Dc protein assay kit (Bio-Rad
Laboratories, Inc., Hercules, CA) based on the Lowry method was also
used to quantitate MUP levels. BSA (fraction V) was used as a relative
standard (Sigma, St. Louis, MO).
Real-time RT-PCR of IGF-I mRNA in hypothalamus
and liver
First-strand cDNA was synthesized from 1 ␮g of total RNA from liver
and hypothalamus using Superscript II RT (Life Technologies, Inc., St.
Louis, MO) with random hexamers according to the manufacturer’s
instructions. Taqman-PCR was performed with the ABI Prism 7700
sequence detection system (Applied Biosystems, Foster City, CA) using
VIC-labeled fluorogenic probes specific for either the IGF-I transcript or
the internal standard M36B4. Oligo primers and probes (Table 1) were
chosen using the Primer Express software (Applied Biosystems). The
PCR was performed using Taqman Universal PCR Mastermix (Applied
Biosystems) to which primers and probes were added (final concentrations 400 nm and 200 nm, respectively). Each run included reactions for
the specific gene, IGF-I, the internal standard, and negative controls for
both primer sets. All samples were run in triplicate in 96-well plates in
the ABI Prism 7700 sequence detector according to the manufacturer’s
standard protocol. For both primer sets, serial dilutions were conducted
with different cDNA preparations to confirm the kinetics of the PCR.
These analyses verified that the efficiencies of amplification were equal
for both primer sets and thereby allowing quantification by the comparative CT method (user bulletin #2, Applied Biosystems).
Multiplex RT-PCR for pituitary GHRF-R, GHS-R, and
somatostatin receptor (sst) 1–5 mRNA
The relative levels of pituitary receptor mRNA were measured by
multiplex RT-PCR using glyceraldehyde-3-phosphate dehydrogenase
(GAPDH) as an internal control. One microgram of total RNA was
reverse transcribed using the Superscript preamplification system for
first-strand synthesis (Life Technologies, Inc.) with random hexamer
priming in 20 ␮l volume. The resultant cDNA was used in two separate
PCR mixtures containing specific primers for GHRF-R, GHS-R, and
GAPDH (reaction #1) or sst1-5 and GAPDH (reaction #2). Primer sequences used in each reaction are shown in Table 2. Reaction conditions
and validation procedures for the multiplex RT-PCR have been previously described (26). There was no significant difference in GAPDH
mRNA levels between experimental groups. Therefore, signal intensity
for each of the pituitary receptor subtypes was adjusted by that of
GAPDH to control for variability in the amount of total RNA used in the
reverse transcription reaction and the efficiency of conversion of RNA
to cDNA.
Riboprobes
The mouse PRL-R cDNA probe identifies the intracellular PRL-R
sequence (27). The plasmid was linearized using XhoI, and the antisense
32
[ P]CTP-labeled PRL-R RNA probe was synthesized using T3 RNA
polymerase. Mouse hypothalamic cDNA was amplified by PCR using
primers for mouse GHRF, somatostatin, NPY, or ␤-actin, and pituitary
cDNA was amplified using primers for mouse GH to generate riboprobes. The primer sequences and probe sizes for GHRF, somatostatin,
NPY, and ␤-actin have been previously reported (26). Primer sequences
used to generate riboprobes for mouse GH were sense: 5⬘-CTGGCTGCTGACACCTACAAA-3⬘ and antisense: taatacgactcactatagggagagttcaagctggtccaCAGGAGAGCAGCCCATAGTTT (capital letters indicated mouse GH gene-specific sequences, GenBank Accession x02891
and K03232). The antisense primers were modified to contain a 17-base
T7 RNA polymerase recognition sequence (5⬘-TAATACGACTCACTATA-3⬘), a 6-base transcription initiation sequence, and 15 or 20 bases
of nonspecific sequence appended at the 5⬘ end. The nonspecific sequence was added to the primer to allow for the differentiation of
protected and unprotected probe following RNase digestion. The amplified PCR products were used as templates for in vitro transcription
performed using the MAXIscript kit (Ambion, Inc., Austin, TX) in the
presence of [␣-32P]CTP. Radiolabeled riboprobes were gel purified before use.
Ribonuclease protection assays (RPAs)
Liver RNA for PRL-R was prepared from frozen liver according to
Chomczynski and Sacchi (28). Mouse PRL-R mRNA levels in the liver
were measured by the RPA II kit (Ambion, Inc.). The assay was
performed according to the manufacturer’s instructions using 40-␮g
liver RNA per sample, with 18S as an internal standard (Ambion,
Inc.). RPA for hypothalamic GHRF, somatostatin, and NPY mRNA
and for pituitary GH mRNA was performed using HybSpeed RPA kit
(Ambion, Inc.) following the manufacturer’s instructions with minor
modifications. The riboprobes were mixed in two reactions: reaction
#1: GHRF [2 ⫻ 104 cpm; specific activity, 1 ⫻ 109 cpm/␮g], somatostatin [1 ⫻ 104 cpm; specific activity, 3 ⫻ 108 cpm/␮g], NPY [2 ⫻
104 cpm; specific activity, 9 ⫻ 108 cpm/␮g], and ␤-actin [4 ⫻ 103 cpm;
specific activity, 8 ⫻ 107 cpm/␮g; reaction #2: GH [5 ⫻ 103 cpm;
specific activity, 4 ⫻ 107 cpm/␮g] and ␤-actin [1 ⫻ 104 cpm; specific
activity, 3 ⫻ 108 cpm/␮g]. The mixture was incubated for 20 min at
68 C in 10 ␮l of HybSpeed hybridization buffer containing 50% of the
total RNA isolated from a single hypothalamus (reaction #1), 1 ␮g of
mouse pituitary RNA (reaction #2) or 50 ␮g of yeast RNA (negative
TABLE 1. Primer sequences for Taqman RT-PCR for mouse IGF-I mRNA levels
Forward primers
IGF-I
M36B4a
Probes
IGF-I
M36B4a
Reverse primers
IGF-I
M36B4a
a
Internal standard.
5⬘-GCTCTTCAGTTCGTGTGTGGAC-3⬘
5⬘-GAGGAATCAGATGAGGATATGGGA-3⬘
5⬘-VIC-TTCAACAAGCCCACAGGCTAT-TAMRA-3⬘
5⬘-VIC-TCGGTCTCTTCGACTAATCCCGCCAA-TAMRA-3⬘
5⬘-CATCTCCAGTCTCCTCAGATC-3⬘
5⬘-AAGCAGGCTGACTTGGTTGC-3⬘
4764
Endocrinology, November 2001, 142(11):4762– 4770
Wallenius et al. • Liver-Pituitary IGF-I Feedback
TABLE 2. Primer sequences, cycling number, and annealing temperature of multiplex RT-PCR for mouse pituitary receptor mRNA levels
Primer sets and
concentrations
Reaction 1
GHRF-R (0.08 mM)
GHS-R (0.28 mM)
GAPDHa (0.06 mM)
Reaction 2
sst1 (0.6 mM)
sst2 (0.6 mM)
sst3 (0.6 mM)
sst4 (0.6 mM)
sst5 (0.6 mM)
GAPDHb (0.04 mM)
Genbank accession
no. and
nucleotide positions
L07379
nt 438 – 459
nt 1013–992
See a
nt 122–142
nt 441– 422
M32599
nt 204 –225
nt 670 – 649
M81831
nt 514 –535
nt 1023–1005
AF008914
nt 2471–2492
nt 2728 –2707
M91000
nt 1126 –1144
nt 1522–1504
U26176
nt 744 –763
nt 962–942
AF030441
nt 1829 –1850
nt 1994 –1976
M32599
nt 756 –775
nt 1063–1042
Primer sequences:
sense
antisense
5⬘-TCTCCATTGTAGCCCTCTGCG
5⬘-GACTTGGAAAGCCGCCAGTAC
No. of cycles and
annealing temperature
28 cycles
65 C
5⬘-CTGCTCTGCAAACTCTTCCAG
5⬘-AAGACGCTCGACACCCATAC
5⬘-AATTCAACGGCACAGTCAAGGC
5⬘-GGATGCAGGGATGATGTTCTGG
5⬘-ACCAGCATCTACTGTCTGACTG
5⬘-GCCCAGGATGACAGACAAC
25 cycles
54 C
5⬘-CACCAGTATCTTCTGCTTGACG
5⬘-AAGGCGTAGATAATGAAACCTG
5⬘-CCTCTCCTACCGCTTCAAG
5⬘-TCACTGCATGTGGGTGTTC
5⬘-TTCTGGCCATCGGATTATGC
5⬘-GGGACACATGGTTGACAGTGG
5⬘-CAGTATCTTCTGCCTGATGGTC
5⬘-GACATCCGCAAAGACCAAG
5⬘-ATGTGTCCGTCGTGGATCTG
5⬘-GTGGTCCAGGGTTTCTTACTCC
a
Primer sequences for the mouse GHS-R were selected using a partial mouse GHS-R sequence generated by PCR sequencing of a cDNA
amplified from mouse pituitary RNA using rat GHS-R-specific primers under low stringency conditions; for details refer to Genbank accession
no. AF332997.
control). Unhybridized probes for all RPAs were digested by treating
the reactions with RNase A/T1 mix (1.0 ␮g/20 U) for 1 h at 37 C.
Protected fragments were separated by electrophoresis through a 5%
polyacrylamide/8 M urea gel. Gels were dried on chromatography
paper and exposed to a PhosphoImager screen. Band intensity was
evaluated using a PhosphoImager and ImageQuant software (Molecular Dynamics, Inc.).
mice (22, 23). Hypothalamic IGF-I mRNA levels were increased by 31% in the female LI-IGF-I⫺/⫺ mice (P ⬍ 0.01;
Fig. 1C). There was a similar tendency in the male LI-IGFI⫺/⫺ mice, but this effect was not statistically significant
(P ⫽ 0.05; Fig. 1C).
Treatment with GHRF and the GHS, ipamorelin
Plasma GH, pituitary GH mRNA levels, and liver weight
Mice were anesthetized with a mixture of ketamine and medetomidine just before the first blood sample was taken. They were then
immediately injected ip with GHRF (40 ␮g/kg) or the GHS ipamorelin
(500 ␮g/kg) (29). Blood samples were collected 15, 30, and 60 min after
injection.
Plasma GH levels were analyzed by a ␹2 test dividing the
data into two groups, above or below 1.3 ng/ml. A significantly greater proportion of GH values were above 1.3
ng/ml in male LI-IGF-I⫺/⫺ mice, compared with control
mice (53% vs. 24%, ␹2 ⫽ 6.31, P ⬍ 0.02, n ⫽ 34 –36), and in
female LI-IGF-I⫺/⫺ mice, compared with control mice
(62% vs. 34%, ␹2 ⫽ 8.15, P ⬍ 0.01, n ⫽ 47–59). These data
are in line with earlier results obtained from pooled male
and female data (22, 23). Pituitary GH mRNA levels did
not differ between male (14.9 ⫾ 1.0 vs. 14.8 ⫾ 0.7 arbitrary
densitometric units, n ⫽ 5) and female (15.2 ⫾ 1.0 vs. 14.2 ⫾
1.1 arbitrary densitometric units, n ⫽ 5– 6) LI-IGF-I⫺/⫺
and control mice. As previously reported (22), liverspecific elimination of IGF-I did not significantly affect
body weight at this age (data not shown). However, relative liver weight (percent liver weight/body weight) was
significantly higher in both male and female LI-IGF-I⫺/⫺
mice, compared with control mice (male: 6.4 ⫾ 0.1% vs.
5.7 ⫾ 0.1%, P ⬍ 0.01 and female: 7.0 ⫾ 0.2% vs. 6.1 ⫾ 0.1%,
Statistical analysis
Differences between groups were compared by t test with the exception of circulating GH data, in which the ␹2-test was used. Logarithmic transformation was used where appropriate. Values are given as
means and sem. P values of ⬍0.05 were considered significant.
Results
Plasma IGF-1 and liver and hypothalamic IGF-I
mRNA levels
Serum levels of IGF-I were decreased by 80%, and liver
mRNA levels were decreased by 90% in both male and
female liver-specific IGF-I knockout (LI-IGF-I⫺/⫺) mice,
compared with control mice (Fig. 1, A and B). These data
are in line with previous pooled data from male and female
Wallenius et al. • Liver-Pituitary IGF-I Feedback
Endocrinology, November 2001, 142(11):4762– 4770 4765
FIG. 2. MUP in urine from male and female, LI-IGF-I⫺/⫺ mice, compared with corresponding control mice. A, Representative gel showing
urine analyzed by gel electrophoresis followed by Coomassie staining.
B, The relative protein content in urine from male and female LIIGF-I⫺/⫺ and control mice was quantified by the Lowry method.
There were 10 –15 mice in each group. **, P ⬍ 0.01 vs. corresponding
control mice.
FIG. 1. IGF-I levels in serum and IGF-I mRNA in liver and hypothalamus in LI-IGF-I⫺/⫺ mice, compared with corresponding control
mice. A, IGF-I levels in serum were measured when the mice were 7
wk old by a RIA. IGF-I mRNA levels in liver (B) and hypothalamus
(C) were measured in 2.5-month-old mice by the Taqman real-time
PCR. There were five to six mice in each group. **, P ⬍ 0.01 vs.
corresponding control mice.
measured by the Lowry method and were markedly lower
in male LI-IGF-I⫺/⫺ mice, compared with male control mice
(Fig. 2B).
Expression of PRL-R mRNA in the liver was measured in
LI-IGF-I⫺/⫺ mice and control mice using RPA (Fig. 3A).
Densitometric scanning showed that the PRL-R mRNA levels were significantly higher in the livers of male LI-IGFI⫺/⫺ mice, compared with male control mice (Fig. 3B).
There was no significant difference in hepatic PRL-R mRNA
levels between female LI-IGF-I⫺/⫺ and female control mice
(Fig. 3B). Taken together, the results of Figs. 2 and 3 indicate
that these GH-regulated hepatic functions are altered in male
LI-IGF-I⫺/⫺ mice in a manner consistent with these mice
having increased GH trough levels (18).
P ⬍ 0.01), as previously shown with pooled male and
female data (22).
Expression of hypothalamic neuropeptides and pituitary
receptors that regulate GH secretion
MUP and hepatic PRL-R mRNA levels
MUP levels were analyzed by gel electrophoresis and Coomassie staining of urine samples from LI-IGF-I⫺/⫺ and
control mice (Fig. 2A). Densitometric scanning of gels
showed that the MUP levels were three times higher in urine
from control males, compared with control females, confirming earlier results by Nordstedt and Palmiter (18). The
MUP levels were decreased by 28% in male LI-IGF-I ⫺/⫺
mice, compared with control males (11.2 ⫾ 1.0 vs. 15.7 ⫾ 1.0
ODu*mm2, P ⬍ 0.02, n ⫽ 5– 6). There was no difference in the
MUP levels between female LI-IGF-I⫺/⫺ mice and controls.
MUP levels (reflected by total protein in urine) were also
To investigate how liver-derived IGF-I affects GH secretion, the expression of hypothalamic GHRF, somatostatin, and NPY and pituitary receptors for GHRF, GHS, and
somatostatin was measured in LI-IGF-I⫺/⫺ and control
mice. GHRF, somatostatin, and NPY mRNA levels in the
hypothalamus of LI-IGF-I⫺/⫺ mice were not significantly
altered (Fig. 4, A and B). In contrast, the expression of the
receptors for GHRF and GHS were increased in both male
and female LI-IGF-I⫺/⫺ mice, compared with control
mice (Fig. 4, C and D). GHRF-R levels were increased by
26% in male and by 70% in female LI-IGF-I⫺/⫺ mice, and
the GHS-R levels were increased by 74% in male and by
112% in female LI-IGF-I⫺/⫺ mice. The mRNA levels of the
4766
Endocrinology, November 2001, 142(11):4762– 4770
Wallenius et al. • Liver-Pituitary IGF-I Feedback
had increased responsiveness to GHRF and GHS, which is
consistent with their increased expression of the receptors
for GHRF and GHS.
Discussion
FIG. 3. PRL-R mRNA levels in the liver of male and female LI-IGFI⫺/⫺ mice, compared with corresponding control mice. A, Two representative gel electrophoresis lanes from each experiment group. B,
Results of densitometric quantification of two gels. There were four
to six mice in each group. **, P ⬍ 0.01 vs. corresponding control mice.
different forms of the sst1–5 in the pituitary were not
significantly affected in these mice (Figs. 4, E and F). These
results demonstrate that circulating liver-derived IGF-I
regulates GH release at the pituitary level rather than at
the hypothalamic level.
Response to GHRF and GHS
To determine whether the increase in expression of
GHRF and GHS-Rs in LI-IGF-I⫺/⫺ mice results in an
enhanced sensitivity to GHRF and GHS, we compared the
GH response to these substances. Male and female LIIGF-I⫺/⫺ mice showed a rapid and pronounced response
to GHRF with peak levels of GH 15 min after treatment,
but the control mice did not respond to this dose of GHRF
(Fig. 5, A and B). Both male LI-IGF-I⫺/⫺ and control mice
responded to GHS treatment with peak levels of GH at 15
min after treatment (Fig. 5C). However, at 30 and 60 min
after treatment, the GH levels were significantly higher in
LI-IGF-I⫺/⫺ mice, compared with control mice (Fig. 5C).
Treatment of female LI-IGF-I⫺/⫺ mice with GHS increased GH levels to nearly 800 ng/ml at 15 min, and GH
remained elevated at 30 and 60 min, but this dose of GHS
did not increase GH levels at all in the control females (Fig.
5D). These results demonstrate that the LI-IGF-I⫺/⫺ mice
By using a unique mouse model for inducible and liverspecific IGF-I depletion, we demonstrated that liverderived IGF-I exerts a tonic inhibitory effect on GH-secretion in mice. GH levels were increased in these mice
showing that GH cannot normalize its own secretion via
short-loop feedback in the absence of liver-derived IGF-I.
Investigation of a patient with complete depletion of IGF-I
showed that both pulse height and the basal GH levels
were increased (30). The increased GH levels in this patient
could be reversed by IGF-I replacement (31). Therefore, it
is clear that endogenous IGF-I suppresses GH secretion in
humans as well as in mice. The present results extend those
reported by Woods et al. (30) on the IGF-I-deficient patient
by showing that GH secretion is modulated mainly by
liver-derived IGF-I that constitutes 80% of circulating
IGF-I (present results, 22, 23).
In humans and some other species, GH levels are increased
during fasting (32, 33). A physiological implication of the
negative feedback effect by IGF-I in humans has been demonstrated by Hartman et al. (32), who showed that the enhanced GH secretion in humans during fasting is caused by
a decrease in circulating IGF-I levels, presumably owing to
decreased hepatic IGF-I production. The increase in GH production could in turn be of importance for the lipolysis and
insulin antagonism during fasting. This theory is supported
by recent findings by us and others that the mice with liverspecific IGF-I depletion had decreased fat mass and decreased insulin sensitivity and that these effects may be mediated by the increased GH secretion (34, 35).
On the basis of earlier studies with exogenous IGF-I treatment, the negative feedback effect of IGF-I on GH secretion
could be exerted either in the hypothalamus [e.g., via suppressed GHRF or enhanced somatostatin release (7–9, 36)], or
directly at the pituitary level (9, 10, 36). Our data support the
latter hypothesis. Reduction of circulating IGF-I by 80% increased GHRF-R and GHS-R mRNA levels in pituitaries
from LI-IGF-I⫺/⫺ mice. These data are consistent with the
finding that GH receptor–null mice, which have a decrease
in both direct effects of GH and serum IGF-I levels (37), also
have increased GHRF-R and GHS-R mRNA levels (26). Sugihara et al. (38) demonstrated that IGF-I decreased GHRF-R
mRNA levels in primary rat pituitary cell cultures. An inhibitory effect of IGF-I on GHRF-R mRNA levels has also
been reported in vivo using IGF-I replacement in the GHdeficient spontaneous dwarf rat (11). In this same model,
IGF-I treatment had no effect on pituitary GHS-R expression
though GH treatment did suppress GHS-R expression (39).
One possible explanation for our present data in conjunction
with those of Kamegai et al. (39) is that IGF-I can suppress
GHS-R expression in the presence, but not in the absence, of
an IGF-I independent, direct GH action.
There was no effect of depletion of liver-derived IGF-I on
expression of the hypothalamic neuropeptides GHRF, somatostatin, and NPY, all of which participate in regulation
Wallenius et al. • Liver-Pituitary IGF-I Feedback
Endocrinology, November 2001, 142(11):4762– 4770 4767
FIG. 4. Expression of GHRF, somatostatin, and NPY in the hypothalamus and GHRF-R, GHS-R, and ssts in the pituitary of LI-IGF-I⫺/⫺ male
and female mice compared with control mice. GHRF, somatostatin, and NPY mRNA levels were measured by RPA (A, males; B, females).
GHRF-R, GHS-R (C, males; D, females), and sst1–5 mRNA levels (E, males; F, females) were measured by multiplex RT-PCR. There were five
to six mice in each group. *, P ⬍ 0.05 vs. corresponding control mice.
of GH release (1, 3, 4, 36, 40, 41). This is consistent with
previous reports that systemic IGF-I treatment does not affect
the expression of GHRF or somatostatin in GH-deficient rats
(12). Hypothalamic IGF-I was significantly increased in female LI-IGF-I⫺/⫺ mice with a similar tendency in males.
This increase in hypothalamic IGF-I could be a response to
the increased GH levels (13, 42). The present results also
demonstrate that the increased expression of GHRF-R and
GHS-R by the absence of liver-derived IGF-I is not reversed
by the enhanced serum GH levels or the enhanced hypothalamic IGF-I expression.
The decrease in circulating IGF-I and increased expression
of pituitary GHRF- and GHS-Rs was accompanied by enhanced GHRF- and GHS-induced GH secretion in vivo.
Therefore, endogenous, liver-derived IGF-I exerts a GHRF
antagonistic effect similar to that originally shown in rat
pituitary cells in vitro (9). IGF-I infusion to humans leads to
decreased GH response to GHRF treatment in fed men, but
not women, in one study (5), but in another study, IGF-I
treatment did suppress both GHRF- and GHS-induced GH
secretion in fasted young women (43). In the present study,
the effect of liver IGF-I depletion on GHS responsiveness was
more pronounced in female than in male mice, although the
GHS-R expression was enhanced to a similar degree in both
sexes. These results suggest that mechanisms other than receptor expression may affect GHS responsiveness. Taken
together, the results of the present and previous data indicate
that liver-derived IGF-I exerts a feedback-regulation of GH
4768
Endocrinology, November 2001, 142(11):4762– 4770
Wallenius et al. • Liver-Pituitary IGF-I Feedback
FIG. 5. GH release after treatment of LI-IGF-I⫺/⫺ mice and control mice with GHRF and GHS. A, Male mice treated with GHRF (40 ␮g/kg
ip); B, female mice treated with GHRF (40 ␮g/kg ip); C, male mice treated with the GHS ipamorelin (500 ␮g/kg ip; pooled data from two
experiments); and D, female mice treated with the GHS ipamorelin (500 ␮g/kg ip). Blood samples were taken before treatment and then after
15, 30, and 60 min. There were four to nine mice in each group. *, P ⬍ 0.05; **, P ⬍ 0.01; ***, P ⬍ 0.001 vs. corresponding control mice.
secretion by suppression of GHRF-R and GHS-R expression
at the pituitary level. These enhanced receptor levels and
other, as yet unknown mechanisms may then decrease sensitivity to ligand stimulation.
It was suggested already in the 1980s that the masculinizing effects of the male GH secretion pattern could be
dependent on hepatic IGF-I production (18). It was shown
that continuous exposure to GH can feminize the expression of MUP and PRL-R in the livers of male mice (18). In
the present study, liver-specific IGF-I depletion indeed
caused a demasculinization of liver functions, and the
overall distribution of GH levels in LI-IGF-I⫺/⫺ mice was
changed from lower to higher values. A simple interpretation of these data combined is that the feminization of
hepatic functions is caused by an increase in the low basal
GH levels normally found in male rodents (2, 44). The
present data do not provide information on whether GH
pulse height was enhanced. Because serum IGF-I levels
were decreased by 80% in the LI-IGF-I⫺/⫺ mice, it thus
appears that the well-documented sexual dimorphism of
hepatic functions induced by the GH-secretion pattern (1,
20) can be influenced by a feedback signal from the liver.
Both male and female LI-IGF-I⫺/⫺ mice in this study had
increased relative liver weight, in line with earlier pooled
male and female data (22). It is reasonable to hypothesize
that the increased liver weight in the LI-IGF-I⫺/⫺ mice
also is due to the increased GH levels because GH can also
affect relative liver size (45, 46). Taken together, these
results are consistent with a pituitary-liver feedback axis
that is more important for regulation of liver functions
than it is for body growth.
In conclusion, loss of liver-derived IGF-I feedback on the
hypothalamic-pituitary system increases GH secretion in
both male and female mice (see proposed model in Fig. 6),
which, in turn, stimulates liver growth. Moreover, elevated
GH troughs in male mice with IGF-I knockout leads to feminization of GH-regulated sexually dimorphic liver functions. Our data show that depletion of liver-derived IGF-I
increases the expression and sensitivity of pituitary GHRF
and GHS receptors. Therefore, we conclude that the major
site of action of liver-derived IGF-I in the regulation of GH
secretion is at the pituitary rather than at the hypothalamic
level.
Acknowledgments
We thank Danielle Carmignac and Professor Iain Robinson for valuable help with GH measurements. We are grateful to Dr. Derek LeRoith
for providing the mice with loxP sequences flanking exon 4 of the IGF-I
gene, Dr. Ralph Kühn and Professor Claus Rajewsky for providing the
Mx-Cre mice, Professor Charles Weissmann for interferon-␣2/␣1 and Dr.
Ian Ahnfelt-Rönne and Dr. John Römer at Novo Nordisk A/S for providing Ipamorelin. We thank Maud Pettersson, Department of Clinical
Wallenius et al. • Liver-Pituitary IGF-I Feedback
FIG. 6. A proposed model of a liver-pituitary feedback axis and how
it is affected by depletion of liver-derived circulating IGF-I. Loss of
liver-derived IGF-I feedback on the hypothalamic-pituitary system
increases GH levels, including trough levels (and possibly also pulse
height). This in turn increases liver weight in both genders and feminizes liver functions (MUP and PRL-R expression) in male liverspecific IGF-I knockout mice. Our data indicate that increased expression of pituitary receptors for GHRF and GHS increases the
sensitivity to GHRF and GHS in the IGF-I-depleted mice. Therefore,
the major site of action of liver-derived IGF-I in the feedback regulation of GH-secretion is at the pituitary rather than at the hypothalamic level.
Pharmacology, for valuable technical assistance. The intracellular region
PRL receptor probe was a kind gift from Kåre Hultén.
Received June 6, 2001. Accepted July 19, 2001.
Address all correspondence and requests for reprints to: John-Olov
Jansson, Research Centre for Endocrinology and Metabolism, Gröna
stråket 8, SE-413 45 Göteborg, Sweden. E-mail: john-olov.jansson@
medic.gu.se.
This work was supported by the Swedish Medical Research Council
(0998), the European Union (Framework 5, QLRT-1999-02038), the
Swedish Foundation for Strategic research, the Bergvall foundation, the
Lundberg Foundation, the Nordic Insulin Pharma, the Swedish Medical
Society, the Göteborg Medical Society, Pharmacia-Upjohn, Novo Nordisk Foundation, the Swedish Association Against Rheumatic Disease,
the Adlerbertska Research Foundation, the Sahlgrenska University
Foundation, the Foundation of Ragnar and Torsten Söderberg, USPHS
Grant DK-30667 (to R.D.K.), and the Bane Foundation (to L.A.F.).
References
1. Jansson JO, Eden S, Isaksson O 1985 Sexual dimorphism in the control of
growth hormone secretion. Endocr Rev 6:128 –150
2. Eden S 1979 Age- and sex-related differences in episodic growth hormone
secretion in the rat. Endocrinology 105:555–560
3. Plotsky PM, Vale W 1985 Patterns of growth hormone-releasing factor and
somatostatin secretion into the hypophysial-portal circulation of the rat. Science 230:461– 463
Endocrinology, November 2001, 142(11):4762– 4770 4769
4. Tannenbaum GS, Ling N 1984 The interrelationship of growth hormone
(GH)-releasing factor and somatostatin in generation of the ultradian rhythm
of GH secretion. Endocrinology 115:1952–1957
5. Jaffe CA, Ocampo-Lim B, Guo W, et al. 1998 Regulatory mechanisms of
growth hormone secretion are sexually dimorphic. J Clin Invest 102:153–164
6. Abe H, Molitch ME, Van Wyk JJ, Underwood LE 1983 Human growth
hormone and somatomedin C suppress the spontaneous release of growth
hormone in unanesthetized rats. Endocrinology 113:1319 –1324
7. Tannenbaum GS, Guyda HJ, Posner BI 1983 Insulin-like growth factors: a role
in growth hormone negative feedback and body weight regulation via brain.
Science 220:77–79
8. Harel Z, Tannenbaum GS 1992 Synergistic interaction between insulin-like
growth factors-I and -II in central regulation of pulsatile growth hormone
secretion. Endocrinology 131:758 –764
9. Berelowitz M, Szabo M, Frohman LA, Firestone S, Chu L, Hintz RL 1981
Somatomedin-C mediates growth hormone negative feedback by effects on
both the hypothalamus and the pituitary. Science 212:1279 –1281
10. Yamashita S, Melmed S 1986 Insulin-like growth factor I action on rat anterior
pituitary cells: suppression of growth hormone secretion and messenger ribonucleic acid levels. Endocrinology 118:176 –182
11. Kamegai J, Unterman TG, Frohman LA, Kineman RD 1998 Hypothalamic/
pituitary-axis of the spontaneous dwarf rat: autofeedback regulation of growth
hormone (GH) includes suppression of GH releasing- hormone receptor messenger ribonucleic acid. Endocrinology 139:3554 –3560
12. Sato M, Frohman LA 1993 Differential effects of central and peripheral administration of growth hormone (GH) and insulin-like growth factor on hypothalamic GH-releasing hormone and somatostatin gene expression in GHdeficient dwarf rats. Endocrinology 133:793–799
13. Wood TL, Berelowitz M, Gelato MC, et al. 1991 Hormonal regulation of rat
hypothalamic neuropeptide mRNAs: effect of hypophysectomy and hormone
replacement on growth-hormone-releasing factor, somatostatin and the insulin-like growth factors. Neuroendocrinology 53:298 –305
14. Smith RG, Van der Ploeg LH, Howard AD, et al. 1997 Peptidomimetic
regulation of growth hormone secretion. Endocr Rev 18:621– 645
15. Kojima M, Hosoda H, Date Y, Nakazato M, Matsuo H, Kangawa K 1999
Ghrelin is a growth-hormone-releasing acylated peptide from stomach. Nature
402:656 – 660
16. Bowers CY, Momany FA, Reynolds GA, Hong A 1984 On the in vitro and in
vivo activity of a new synthetic hexapeptide that acts on the pituitary to
specifically release growth hormone. Endocrinology 114:1537–1545
17. Dickson SL, Bailey AR, Leng G 1999 Growth hormone (GH) secretagogues
and neuroendocrine regulation of GH secretion. Growth Horm IGF Res
(Suppl A)9:89 –91
18. Norstedt G, Palmiter R 1984 Secretory rhythm of growth hormone regulates
sexual differentiation of mouse liver. Cell 36:805– 812
19. Metcalf D, Greenhalgh CJ, Viney E, et al. 2000 Gigantism in mice lacking
suppressor of cytokine signalling-2. Nature 405:1069 –1073
20. Gustafsson JA, Eden S, Eneroth P, et al. 1983 Regulation of sexually dimorphic
hepatic steroid metabolism by the somatostatin-growth hormone axis. J Steroid
Biochem 19:691– 698
21. Waxman DJ, Pampori NA, Ram PA, Agrawal AK, Shapiro BH 1991 Interpulse
interval in circulating growth hormone patterns regulates sexually dimorphic
expression of hepatic cytochrome P450. Proc Natl Acad Sci USA 88:6868 – 6872
22. Sjogren K, Liu JL, Blad K, et al. 1999 Liver-derived insulin-like growth factor
I (IGF-I) is the principal source of IGF-I in blood but is not required for
postnatal body growth in mice. Proc Natl Acad Sci USA 96:7088 –7092
23. Yakar S, Liu JL, Stannard B, et al. 1999 Normal growth and development in
the absence of hepatic insulin-like growth factor I. Proc Natl Acad Sci USA
96:7324 –9329
24. Blum WF, Breier BH 1994 Radioimmunoassays for IGFs and IGFBPs. Growth
Regul (Suppl 1)4:11–19
25. Magoulas C, McGuinness L, Balthasar N, et al. 2000 A secreted fluorescent
reporter targeted to pituitary growth hormone cells in transgenic mice. Endocrinology 141:4681– 4689
26. Peng XD, Park S, Gadelha M, et al. 2001 The growth hormone (GH)-axis of
GH receptor/binding protein gene-disrupted and metallothionein-human
GH-releasing hormone transgenic mice: hypothalamic neuropeptide and pituitary receptor expression in the absence and presence of GH feedback.
Endocrinology 142:1117–1123
27. Ling C, Hellgren G, Gebre-Medhin M, et al. 2000 Prolactin (PRL) receptor
gene expression in mouse adipose tissue: increases during lactation and in
PRL-transgenic mice. Endocrinology 141:3564 –3572
28. Chomczynski P, Sacchi N 1987 Single-step method of RNA isolation by acid
guanidinium thiocyanate-phenol-chloroform extraction. Anal Biochem 162:
156 –159
29. Raun K, Hansen BS, Johansen NL, et al. 1998 Ipamorelin, the first selective
growth hormone secretagogue. Eur J Endocrinol 139:552–561
30. Woods KA, Camacho-Hubner C, Savage MO, Clark AJ 1996 Intrauterine
growth retardation and postnatal growth failure associated with deletion of the
insulin-like growth factor I gene. N Engl J Med 335:1363–1367
31. Camacho-Hubner C, Woods KA, Miraki-Moud F, et al. 1999 Effects of recombinant human insulin-like growth factor I (IGF-I) therapy on the growth
4770
32.
33.
34.
35.
36.
37.
38.
39.
Endocrinology, November 2001, 142(11):4762– 4770
hormone-IGF system of a patient with a partial IGF-I gene deletion. J Clin
Endocrinol Metab 84:1611–1616
Hartman ML, Clayton PE, Johnson ML, et al. 1993 A low dose euglycemic
infusion of recombinant human insulin-like growth factor I rapidly suppresses
fasting-enhanced pulsatile growth hormone secretion in humans. J Clin Invest
91:2453–2462
Nagatani S, Zeng Y, Keisler DH, Foster DL, Jaffe CA 2000 Leptin regulates
pulsatile luteinizing hormone and growth hormone secretion in the sheep.
Endocrinology 141:3965–3975
Sjogren K, Wallenius K, Liu JL 2001 Liver-derived IGF-I is of importance for
normal carbohydrate and lipid metabolism. Diabetes 50:1539 –1545
Yakar S, Liu JL, Fernandez AM, et al. 2001 Liver-specific IGF-1 gene deletion
leads to muscle insulin insensitivity. Diabetes 50:1110 –1118
Frohman LA, Jansson JO 1986 Growth hormone-releasing hormone. Endocr
Rev 7:223–253
Zhou Y, Xu BC, Maheshwari HG, et al. 1997 A mammalian model for Laron
syndrome produced by targeted disruption of the mouse growth hormone
receptor/binding protein gene (the Laron mouse). Proc Natl Acad Sci USA
94:13215–13220
Sugihara H, Emoto N, Tamura H, et al. 1999 Effect of insulin-like growth
factor-I on growth hormone-releasing factor receptor expression in primary rat
anterior pituitary cell culture. Neurosci Lett 276:87–90
Kamegai J, Wakabayashi I, Miyamoto K, Unterman TG, Kineman RD, Frohman LA 1998 Growth hormone-dependent regulation of pituitary GH secre-
Wallenius et al. • Liver-Pituitary IGF-I Feedback
40.
41.
42.
43.
44.
45.
46.
tagogue receptor (GHS-R) mRNA levels in the spontaneous dwarf Rat. Neuroendocrinology 68:312–318
Chan YY, Clifton DK, Steiner RA 1996 Role of NPY neurons in GH-dependent
feedback signalling to the brain. Horm Res 45:12–14
Minami S, Kamegai J, Sugihara H, Suzuki N, Wakabayashi I 1998 Growth
hormone inhibits its own secretion by acting on the hypothalamus through its
receptors on neuropeptide Y neurons in the arcuate nucleus and somatostatin
neurons in the periventricular nucleus. Endocr J 45(Suppl):S19 –S26
Hynes MA, Van Wyk JJ, Brooks PJ, D’Ercole AJ, Jansen M, Lund PK 1987
Growth hormone dependence of somatomedin-C/insulin-like growth factor-I
and insulin-like growth factor-II messenger ribonucleic acids. Mol Endocrinol
1:233–242
Ghigo E, Gianotti L, Arvat E, et al. 1999 Effects of recombinant human
insulin-like growth factor I administration on growth hormone (GH) secretion,
both spontaneous and stimulated by GH-releasing hormone or hexarelin, a
peptidyl GH secretagogue, in humans. J Clin Endocrinol Metab 84:285–290
Tannenbaum GS, Martin JB 1976 Evidence for an endogenous ultradian
rhythm governing growth hormone secretion in the rat. Endocrinology 98:
562–570
Quaife CJ, Mathews LS, Pinkert CA, Hammer RE, Brinster RL, Palmiter RD
1989 Histopathology associated with elevated levels of growth hormone and
insulin-like growth factor I in transgenic mice. Endocrinology 124:40 – 48
Sjogren K, Bohlooly YM, Olsson B, et al. 2000 Disproportional skeletal growth
and markedly decreased bone mineral content in growth hormone receptor
⫺/⫺ mice. Biochem Biophys Res Commun 267:603– 608