Download The Composition of Low-Density Lipoprotein and

CIinicaI Science (1979) 57,83-88
The composition of low-density lipoprotein and very-lowdensity lipoprotein subfractions in primary hypothyroidism
and the effect of hormone-replacement therapy
F I O N A C . B A L L A N T Y N E , A. A. E P E N E T O S * , M U R I E L C A S L A K E ,
S. F O R S Y T H E I A N D D. B A L L A N T Y N E *
Department of Biochemistry, Royal Infirmary. Glasgow and 'Division of CIinical Medicine, Victoria Infimary.
Glasgow, Scotland, U.K.
(Received 2 October 1978; accepted 9 March 1979)
Summary
1. The lipid and protein composition of subfractions of plasma low-density lipoprotein (LDL) has
been determined in nine patients with primary
hypothyroidism before and after 3 months of
thyroxine therapy. Analyses were also made of
subfractions of very-low-density lipoprotein
(VLDL) in four of the patients.
2. Before therapy seven of the patients had
elevated LDL-cholesterol and two had increased
VLDL-cholesterol concentrations. On thyroxine
replacement the mean LDL-cholesterol fell to
normal. No significant change occurred in the
concentration of cholesterol in VLDL or in highdensity lipoprotein (HDL).
3. The concentrations of cholesterol, triglyceride
and apolipoprotein B (apoB) were increased in the
LDL subfraction of S, 10-4-20, which corresponds
mainly to intermediate-density lipoprotein. This
subfraction showed a marked fall on therapy. The
cholesterol and apoB concentrations in the major
LDL fraction of S, 5.7-12 also decreased on
therapy, but the fall in the subfraction of S, 3.56-5 did not reach statistical significance.
4. Only the VLDL subfraction of smallest size
(S, 20-60) had any abnormality before therapy,
with an increased concentration of cholesterol. On
thyroxine the concentration of triglyceride rose in
the VLDL subfractions.
5. These data suggest that thyroxine exerts its
major effect on lipoprotein metabolism by promoting the conversion into LDL of intermediatedensity lipoprotein, formed by catabolism of
VLDL.
Key words: hypothyroidism, low-density lipoprotein, thyroxine, very-low-density lipoprotein.
Abbreviations: apoB, apolipoprotein B; HDL,
LDL and VLDL, high-density, low-density and
very-low-density lipoprotein; T3, tri-iodothyronine;
T4, thyroxine; TSH, thyroid-stimulating hormone.
Introduction
It has long been known (Peters & Man, 1950) that
elevated plasma concentrations of cholesterol are
commonly found in hypothyroidism. The increase
in cholesterol correlates with that in thyroidstimulating hormone (Kutty, Bryant & Farid,
1978). Triglyceride concentrations may be normal
or increased (Kutty et al., 1978). Although most of
the hyperlipoproteinaemias (Beaumont, Carlson,
Cooper, Fejfar, Fredrickson & Strasser, 1970) can
occur in hypothyroidism, the most common finding
is an increase in the main cholesterol-transporting
protein, low-density lipoprotein (LDL) (Rossner &
Rosenqvist, 1974). Elevated lipid and lipoprotein
concentrations usually return to normal on
hormone-substitution therapy (Walton, Campbell
& Tonks, 1965a; Rossner & Rosenqvist, 1974),
Correspondence: Dr F. C. Ballantyne, Department of
Biochemistry, Royal Infirmary, Glasgow G4 OSF,
Scotland, U.K.
83
F. C. Ballantyne et al.
84
All had the typical clinical features of the disorder
and the diagnosis was confirmed by estimation of
serum concentrations of thyroxine (T4) and triiodothyronine (T3) (Challand, R a t c u e &
Ratcliffe, 1975) and of thyroid-stimulating hormone (TSH) (Hall, Amos & Ormston, 1971)
(Table 1). None had been previously investigated
for disordered thyroid function. None showed overt
clinical or laboratory evidence of other disorders,
e.g. liver or renal disease. In particular, no patient
had evidence of disorders which are commonly
associated with hyperlipidaemia, such as diabetes
or excess intake of alcohol, or had a family history
of hyperlipidaemia or premature vascular disease.
After an overnight fast 50 ml of blood was
removed. A portion (10 ml) was allowed to clot and
serum separated for estimation of T3, T4 and TSH.
The remaining blood was collected into potassium
EDTA (5 mmol/l final concentration) and plasma
separated for estimation of lipoproteins. Samples
were stored at 4OC and analysed within 24 h of
removal.
The patients were treated with L-thyroxine
initially in a dose of 50 pgfday and then increasing
by 50 pglday at monthly intervals until each
patient was judged to be euthyroid by normal TSH
concentration and by clinical criteria (Table 1).
Total LDL, VLDL and H D L and also LDL
subfractions from all patients were analysed both
before and after 3 months of therapy. VLDL
subfractions from four of the patients (nos. 1-4)
were also analysed at both times.
confirming that the abnormalities result from
thyroid insufficiency.
The major lipoprotein classes are heterogeneous
and the technique of density-gradient ultracentrifugation (Lindgren, Jensen & Hatch, 1972)
can be used to isolate subfractions from very-lowdensity lipoproteins (VLDL) and LDL. There is
good evidence (Eisenberg, Bilheimer, Levy &
Lmdgren, 1973; Sigurdsson, Nicoll & Lewis, 1975;
Ballantyne, Ballantyne, Olsson, Rossner &
Carlson, 1977) that the large VLDL particles (S, >
100) are metabolized to progressively smaller
VLDL particles under the action of lipoprotein
lipase. This leads to formation of particles intermediate in density between VLDL and LDL, which
correspond mainly to LDL of S, 10.4-20. These
are then converted into the major LDL component
of S, 5.7-12 (Eisenberg et al., 1973; Eisenberg &
Levy, 1975). It is generally assumed that in turn
these LDL particles are converted into small LDL
particles (S,3.5-6.5).
The present study was therefore undertaken to
determine the protein and lipid composition of
LDL subfractions of S, 10.4-20, 5.7-12 and 3.56.5 in nine patients with primary hypothyroidism
both before and after 3 months of hormonereplacement therapy to determine which subfraction was most affected by the action of thyroid
hormone. Analysis was also made of VLDL
subfractions of S, > 100, 60-100 and 20-60 in
four of the patients.
Materials and methods
Methods
Patients
Analysis of VLDL, LDL and HDL. Plasma was
centrifuged for 16 h at density 1.006 kg/l in a
Prepspin 50 Ultracentrifuge (MSE Ltd, Crawley,
Nine consecutive new patients (eight female, one
male) with primary hypothyroidism were studied.
TABLE1. Clinical data for patients with hypothyroidism before and afler 3 months' thyroxine therapy
Patients nos. 1-8 were female; no. 9 was male. U.D., undetectable.
Patient
no.
Age
(years)
Height
(cm)
Weight
(kg)
1
76
56
57
60
71
53
46
51
74
155
145
I60
157
152
157
154
147
175
62.3
52.3
77.5
46.1
57.2
82.1
56.2
48. I
69.0
2
3
4
5
6
7
8
9
Reference
range
-
T4 (nmol/l)
T3 (nmol/l)
TSH (munitdl)
Before
ARer
Before
ARer
Before
After
<I7
(17
18
49
40
54
<I7
<I7
(17
126
103
2.5
0.7
<0.5
0-5
1.9
1.5
2.0
1.8
1.9
1.6
2.1
2.0
1.2
2.2
I .4
>50
>50
> 50
3.6
5.6
8.0
45
3.5
8.0
100
I20
86
100
93
138
60
55-144
0.8
1.2
0.6
0.9-2.8
>50
26
40
> 50
> 50
5.6
5.0
8.0
9.2
U.D.-8.0
Low-density lipoproteins in hypothyroidism
Sussex, U.K.). The top fraction (VLDL) was
separated from the bottom fraction (LDL + HDL
and other plasma proteins) by tube-slicing and the
LDL was then precipitated with heparin/MnCl,
from a portion of the bottom fraction (Carlson,
1973). The concentrations of cholesterol in total
plasma and in all isolated fractions and of
triglyceride in plasma were determined after extraction in the presence of Zeolite mixture into isopropanol on Auto-Analyzers I1 by methods SE40016 FH4 and SE4-0023FE5 respectively
(Technicon Instrument Co. Ltd, London).
Analysis of VLDL and L D L subfractions. These
lipoproteins were each separated into three subfractions (VLDL of s, > 100, 60-100 and 20-60
and LDL of S, 104-20, 5.7-12 and 3.5-6.5) by
cumulative flotation through NaCl density
gradients (Lindgren et al., 1972) in a 6 x 14 ml
titanium swing-out rotor (MSE Ltd) in either a
Prepspin 50 or a Superspeed 75 ultracentrifuge.
The cholesterol and triglyceride concentrations of
all subfractions were estimated by fully automated
enzymatic procedures (Boehringer Corporation
Ltd, London) on Auto-Analyzer I1 equipment. The
total apolipoprotein concentrations of the fractions
were determined (Lowry, Rosebrough, Farr &
Randall, 195 1) after delipidation with chloroform/
methanol (1 : 1, v/v) (L. A. Carlson, unpublished
work). After precipitation of apolipoprotein B
(apoB) with tetramethylurea (Kane, 1973; Kane,
Sata, Hamilton & Havel, 1975) the concentration
of protein soluble in tetramethylurea was estimated
by using human serum albumin (Hoechst Ltd,
85
London) as standard. The apoB protein concentration was then calculated by difference.
Statistical analyses. Comparisons were made by
pair difference t-test, except when variances were
not homogeneous as determined by the F-test. In
that situation the Wilcoxon signed rank test was
used.
Results
Initially, six of the patients had type IIa hyperlipoproteinaemia, one had type IIb hyperlipoproteinaemia and two had a normal pattern (Table 2).
Type 111 hyperlipoproteinaemia,which can occur in
association with hypothyroidism (Lasser, Burns &
Solar, 1974), was excluded by a normal ratio of
VLDL-cholesterol to total triglyceride and by the
absence of a 'floating beta' band on electrophoresis
of plasma or the top fraction isolated after ultracentrifugation at density 1.006 kg/l. After 3
months three patients had type IIa hyperlipoproteinaemia and six had normal lipoproteincholesterol concentrations. Both the elevated mean
total and LDLcholesterol concentrations showed
significant falls on thyroxine replacement (Table 2).
The changes did not correlate with the rise in serum
thyroxine concentration. The fall in total triglyceride can probably be ascribed mainly to a fall
in LDL-triglyceride. VLDL- and HDL-cholesterol
concentrations were initially normal and did not
change on therapy.
Table 3 gives the lipid and protein composition
of the three LDL subfractions before and after
TABLE2. Concentrations of the major lipoprotein classes before and sfter 3 months' thyroxine therapy (mean dose
200 PgldaY)
All results are in mmol/l. *0.05 > P > 0.01 (compared with pretreatment values by pair difference r-test or Wilcoxon
signed rank test).
~~
Subject
Total triglyceride
Total cholesterol
VLDL-cholesterol
LDLcholesterol
HDL-cholesterol
no.
Before
1
2
3
4
5
6
7
8
9
Mean f SEM
Reference
range
After
2.3
1.6
2.1
1.2
2.0
2.2
1.9
2.8
1.2
1.9 f
0.2
<2.0
1 .o
1.8
1.2
1.2
2.0
1.8
0.9
1.4
1.2
1.4+
0.1.
Before
15.8
6.5
10.0
5.3
9.0
9.5
10.8
19.2
7.1
10.4+
1.5
4.0-7.0
After
6.9
5.1
5.9
5.3
8.3
9.1
6.2
7.8
4.8
6.6+
0.6"
Before
After
Before
0.4
0.5
0.2
0.5
0.4
0.5
13.2
4.6
8.1
3.6
7.1
0.2
0.5
0.6
0.6
0.9
0.9
1.5
0.4
0.7 f
0.1
0-6
7-6
0.1
0.5
0.3
8.1
17.0
5.7
8.3 f
1.4
0.149
0.4 f
0.1
After
2.M.9
4.9
3.2
3.7
3.8
6.5
7.6
4.0
5.3
3.2
4.8k
0.6.
Before
After
2.2
1.7
1.4
1.1
1.3
1 .o
1.7
1 .o
1.5
1.7
1.7
1.1
1.3
1.5
2.1
1.5
1.3
1.5
0.1
1.o
1.4f
0.1
1.2-1.8
F. C . Ballantyne et al.
86
TABLE3. Concentrations of lipids and proteins in LDL subfractions of all nine patients before and after 3 months’
thyroxine therapy
Mean values f SEM are shown. The mean dose of thyroxine was 200 pg/day. P < 0.05, ** (0.01 (compared with
pretreatment values by pair difference t-test or Wilcoxon signed rank test).
Cholesterol
(plnol/l)
104-20
5.7- I2
3.5-6.5
Before
After
Before
After
Before
After
Triglyceride
@mol/l)
Cholesterol:
triglyceride ratio
Total protein
(mg/O
9.4f 1.1
11.9 f 2.8
15.7 f 4.0
23.2 f 5.7
15.4 f 1 . 1
20.2 f 2.6
244 f 55
60 f 12**
506 f 71
271 f 40.
250 f 82
162f 43
205 f 81
48 f 12..
468 f 191
I34 f 72
75 f 30
36 f 8
1360 f 323
482 f 135..
3232 f 761
I371 f 278.
1038 f 352
694 f 146
ApoB
(mg/l)
221 f 5 1
49 f 10.
494 f 70
257 f 40.
241 f 79
155 f 44
TABLE4. Concentrations of lipids and proteins in VLDL subfractions of four patients before and afler 3 months’
thyroxine therapy
Mean values & SEM are shown.
s,
> 100
6&100
2&60
Before
After
Before
After
Before
After
Cholesterol
(plnol/l)
Triglyceride
@mol/l)
Cholesterol:
triglyceride ratio
16 f 6
55 k 13
4 4 f 16
78 f 6
419 f 43
263 f 21
76 f 8
129 f 4 9 t
95 f 25
166 f 24.
171 f 50
342 f 43**
0.18 f 0.08
0.82 f 0 . 5 3 t
0.45 f 0.17
0.51f0.11
3.44 f 1.14
0.80f0.10t
Total protein
(mdl)
5f 1
6+2
9f2
15 f 3.
78 f 20
81f8
ApoB
(mg/O
<4
(4
8? I
7f I
6 4 f 21
4 2 5 13
P < 0.05, ** P < 0.01 compared with pretreatment values by pair difference t-test or Wilcoxon signed rank test.
t Insufficient data for Wilcoxon signed rank test.
therapy. The incomplete recovery of cholesterol
can be at least partly ascribed to differences in the
methods used to estimate cholesterol in the subfractions and in the whole LDL class. Thyroxine
therapy exerted its main effect o n the LDL subfraction of largest size (S,10.4-20) with highly
significant falls in both lipids and proteins. The
decreases in the major LDL subfraction (S,5.712) were not so marked and in LDL of S,3.5-6.5
they were not statistically significant.
VLDL subfractions were isolated in four of the
nine patients. The only major change on therapy
was an increase in the concentration of triglyceride
in the subfractions (Table 4).
Discussion
In this study the patients acted as their own
‘controls’, since they were studied both before and
after hormone replacement. Some comparison can
also be made with results obtained (Stromberg,
Ballantyne, Ballantyne, Third & Bedford, 1977) on
a group of normal subjects with the reservations
that there are differences in age, sex and body
weight between the two groups. In the untreated
patients with hypothyroidism LDL of S, 10.4-20
and S, 5.7- 12 contained increased concentrations
of cholesterol, triglyceride and apoB, whereas only
cholesterol was increased in LDL of S, 3.5-6.5.
The composition of VLDL subfractions was less
affected and only in the subfraction of smallest size
(S, 20-60) was the cholesterol concentration
increased. After 3 months’ thyroxine therapy the
composition of both LDL and VLDL subfractions
was similar to that of normal subjects.
When the present study was undertaken, it was
expected that abnormalities would be found in the
VLDL spectrum in hypothyroidism. Rossner &
Rosenqvist (1974) found an increase in the
concentration of VLDL-cholesterol but not triglyceride in three of seven patients and the mean
concentration of cholesterol fell on therapy. From
studies on four patients on high and low fat diets
O’Hara, Porte & Williams (1966) presented
evidence that thyroid hormone deficiency appeared
to affect the whole VLDL spectrum and ascribed
this to an influence on lipoprotein lipase (Porte,
O’Hara & Williams, 1966). However, our series of
Lo w-density lipoproteins in hypothyroidism
patients with hypothyroidism show relatively little
derangement in VLDL. Increased concentrations
of VLDL-cholesterol were found in only one of the
present nine subjects and in one of a further four
patients studied before but not on therapy. Thyroxine replacement did not significantly affect the
concentration of VLDL-cholesterol. In the four
patients in whom subfractions of VLDL were
prepared the only abnormality in the untreated
state was an increased concentration of cholesterol
in VLDL of S, 20-60. The fall in this on therapy
was not significant, but a significant rise did occur
in the concentration of triglyceride leading to a
normal cholesterol :triglyceride ratio.
Before therapy the most marked increase was
found in the LDL subfraction of largest size (S,
10.4-20), and this was corrected by thyroxine
replacement. However, no change occurred in the
ratio of cholesterol :triglyceride on therapy, suggesting that thyroxine decreases the number of
particles rather than alters the composition of
individual particles. Most lipoprotein of intermediate density, formed by sequential catabolism
of VLDL by lipoprotein lipase, is found in the LDL
subfraction of S, 10.4-20, although some may also
separate in the VLDL subfraction of S, 12-20
(Eisenberg et al., 1973). This intermediate-density
lipoprotein is relatively resistant to subsequent
catabolism by extrahepatic lipoprotein lipase and it
is probably converted into the main LDL component (S, 5-7-12) by a mechanism different from
that involved in its formation (Eisenberg et al.,
1973; Eisenberg 8c Levy, 1975), possibly under the
action of a hepatic lipase. The findings of the
present study provide good evidence that thyroxine
plays an important role in mediating the conversion
of intermediatedensity lipoprotein into LDL, perhaps by an effect on the converting enzyme.
Abnormalities were also found in the composition of LDL of S, 5.7-12, the major transport
medium for cholesterol and apoB, the concentrations of which fell after thyroxine replacement. This
is consistent with and extends reports of abnormal
cholesterol and apoB metabolism in hypothyroidism. In a definitive study, Miettinen (1968)
followed the faecal excretion of end products of
cholesterol metabolism after intravenous injection
of [3Hlcholesterolin three patients before and after
therapy. The fall in serum cholesterol induced by
thyroid hormone was associated with a marked
increase in excretion of neutral steroids derived
from circulating cholesterol, and with a smaller
increase in the output of bile acids. The metabolism
of the protein component of LDL, which is mainly
87
apoB, was studied in vivo by Walton, Scott, Dykes
& Davies (1965b) after intravenous injection of
1311-labelled LDL (probably contaminated with
some VLDL). They found that the rate of LDLprotein catabolism was impaired, but returned to
normal on therapy.
It is concluded that in our series of patients with
hypothyroidism, the major defect in lipoprotein
metabolism is impaired conversion of IDL into
LDL, although other abnormalities also occur in
the metabolism of VLDL and of LDL. The defects
can be rectified by thyroid hormone replacement.
Acknowledgments
We acknowledge the co-operation of the Radioimmunoassay Unit, Department of Biochemistry,
Glasgow Royal Infirmary, in performing the T3,
T4 and TSH analyses. We also thank Dr J. G.
Ratcliffe and Dr W. A. Ratcliffe for helpful
discussions on endocrine assessments.
References
BALLANTYNE,
F.C., BALL-,
D., OLSSON,A.G., R~SSNER,
S. & CARLSON,L.A. (1977) Metabolism of very low density
lipoprotein of Sf 100-400 in type V hyperlipoproteinaemia
Acta Medica Scandinavica, 202.153-161.
BEAUMONT,
J.L., CARLSON,L.A., COOPER,G.R., FEJPAR,Z.,
FREDRICKSON,
D.S. & STRASSER,
T. (1970) Classificationof
hyperlipidaemias and hyperlipoproteinaemias. Bulletin of the
World Health Organisation, 43,89 1-9 15.
CARLSON,K. (1973) Lipoprotein fractionation. Journal of
Clinical Pathology, 26 (Suppl. 5), 32-37.
CHALLAND,
G.S., RATCLIPPE,
W.A. & RATCLIFFE,
J.G.(1975)
Semi-automated radioimmunoassays for total serum thyroxine and triiodothyronine. Clinica Chimica Acra, 60,25-32.
EISENBERG,
S. & LEVY,R.I. (1975) Lipoprotein metabolism.
Advances in Lipid Research, 1 3 , 1 4 9 .
S., BILHEIMER,
D.W., LEVY,R.I. & LMDOREN,
EISENBERG,
F.T. (1973) On the metabolic conversion of human plasma
very low density lipoprotein to low density lipoprotein.
Biochimica et Biophysica Acta, 326,36 1-377.
HALL,R., AMOS,J. & OMSTON,B.J. (1971) Radioimmunoassay of human serum thyrotrophin. British Medical
Journal, i,582-585.
KANE, J.P. (1973) A rapid electrophoretic technique for
identification of subunit species of apoproteins in serum lipoproteins. A MIytical Biochemistry, 53,350-364.
KANE,J.P., SATA,T., HAMILTON,
R.L. & HAVEL, R.J. (1975)
Apoprotein composition of very low density lipoproteins of
human serum. Journal of Clinical Investigation, 515, 16221634.
K m , K.M., BRYANT,D.G. & FARID,N.R. (1978) Serum
lipids in hypothyroidism - a reevaluation. Journal of
Clinical Endocrinology and Metabolism, 6 5 5 - 6 0 .
LASSER,N.L., BURNS,J. & SOLAR,S. (1974) In:Atherosclerosis
111, pp. 621625. Ed. Schettler, G. & Weizel, A. Springer,
Berlin.
LIND~REN,
F.T., JENSEN,L.C. & HATCH,F.T. (1972) In: Blood
Lipids and Lipoproteins, pp. 181-274. Ed. Nelson, GJ.
Wiley, New York.
LOWRY,O.H., ROSEBROUGH,
N.J., FARR, A.L. & RANDALL,
R.J. (1951) Protein measurement with the Folin phenol
reagent. Journal of Biological Chemistry, 193,265-275.
88
F. C. Ballantyne et al.
METTINEN, T.A. (1966) Mechanism of serum cholesterol
reduction by thyroid hormones in hypothyroidism. Journal of
Laboratory and Clinical Medicine, 11,537-547.
OIHARA,D.D., PORTE,D. & WILLIAMS,
R.H. (1966) The effect
of diet and thyroxin on plasma lipids in myxoedema.
Metabolism, Clinical and Experimental, 15,123-134.
PETERS,J.P. & MAN, E.B. (1950) The significance of serum
cholesterol in thyroid disease. Journal of Clinical
Investigation, 29, 1- 11.
PORTE, D., O'HARA, D.D. & WILLIAMS,R.H. (1966) The
relation between postheparin lipolytic activity and plasma
triglyceride in myxedema. Metabolism, Clinical and Experimental, 15, 107-113.
R~SSNER
S., & ROSENQVIST,
U. (1974) Serum lipoproteins and
the intravenous fat tolerance test in hypothyroid patients
before and during substitution therapy. Atherosclerosis, 20,
365-381.
SIGURDSSON,
G., NICOLL, A. & LEWIS,B. (1975) Conversion of
very low density lipoprotein to low density lipoprotein. A
metabolic study of apolipoprotein B kinetics in human
subjects. Journal of Clinical Investigation, 56, 148 1-1490.
STR~MBERO,
P., BALLANTYNE,
D., BALLANTYNE,
F.C., THIRD,
J.L.H.C. & BEDFORD,D.K. (1977) The composition of low
(LDL) and very low (VLDL) density lipoprotein subfractions in type 111 hyperlipoproteinaemia: comparison with
normal subjects. Clinica Chimica Acta, 19, 163-172.
WALTON,K.W., C m ~ s e u ,D.A. & TONKS,E.L. (1965a)The
significance of alterations in serum lipids in thyroid dysfunction. I. The relation between serum lipoproteins, carotenoids
and vitamin A in hypothyroidism and thyrotoxicosis. Clinical
Science, 29, 199-2 15.
WALTON,K.W., Scorr, P.J., DYKES,P.W. & DAVIES,J.W.L.
(1965b) The significance of alterations in serum lipids in
thyroid dysfunction. 11. Alterations of the metabolism and
turnover of 1311-10wdensity lipoproteins in hypothyroidism
and thyrotoxicosis. Clinical Science, 29,2 17-236.