| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Metabolism Unit, Center for Metabolism and Endocrinology, Department of Medicine, and Molecular Nutrition Unit, Center for Nutrition and Toxicology, Novum, Karolinska Institute at Huddinge University Hospital, S-141 86, Stockholm, Sweden
1Correspondence: CME, M63, Huddinge University Hospital, S-141 86 Stockholm, Sweden. E-mail: mats.rudling{at}cnt.ki.se
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
-hydroxylase (C7OH), and for
the maintenance of resistance to dietary cholesterol. The present study
aimed to determine whether GH has beneficial effects on plasma lipids
and hepatic cholesterol metabolism in mice devoid of LDLR. Compared
with wild-type controls, LDLR-deficient mice had 
250% elevated
plasma total cholesterol and 50% increased hepatic cholesterol
levels; hepatic HMG CoA reductase activity was reduced by 70%, whereas
C7OH activity was increased by 40%. In LDLR mice, GH infusion
reduced plasma cholesterol and triglycerides up to 40%, whereas HMG
CoA reductase and C7OH activities were stimulated by 50%
and 110% respectively. GH also stimulated HMG CoA reductase and
C7OH activities in control mice, whereas hepatic LDLR and plasma
lipoproteins were unchanged. The effects of cholestyramine and
atorvastatin on C7OH in LDLR-deficient mice were potentiated by GH,
and this was associated with a further reduction in plasma cholesterol.
GH treatment reduces plasma cholesterol and triglycerides and
stimulates C7OH activity in mice devoid of LDLR, particularly in
combination with resin or statin treatment. The potential of GH therapy
in patients with homozygous familial hypercholesterolemia should be
evaluated.—Rudling, M., Angelin, B. Growth hormone reduces plasma
cholesterol in LDL receptor-deficient mice.
Key Words: hypercholesterolemia • familial • cholesterol 7-monooxygenase • statins • resins
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
2
3
4
5
6
7
8
9)
. Thus, by stimulating adipocyte
lipolysis, GH mobilizes free fatty acids, which upon entering the liver
promote triglyceride synthesis. The production of apolipoprotein
(apo) E and apoB-48 is simultaneously enhanced, resulting in a more
rapid catabolism of secreted very low density lipoproteins (VLDL). GH
also has a stimulatory effect on hepatic LDL receptor (LDLR) expression
(4
, 10)
and is critically important in maintaining the
normal resistance to dietary cholesterol in rodents (7)
.
Recently GH has been identified as a key stimulator of the enzymatic
activity of rat hepatic cholesterol 7-hydroxylase (C7OH)
(9)
, the rate-limiting enzyme in the metabolism of
cholesterol to bile acids. GH stimulates both C7OH activity and
fecal bile acid excretion not only in hypophysectomized (GH-deficient)
animals, but also in normal young rats (10)
. In normal
rats, the infusion of GH reduces plasma cholesterol more or less
independent of changes in hepatic LDLR expression (10)
.
The question therefore arises as to what extent GH can reduce plasma
cholesterol also when LDLRs are absent, a situation that occurs
clinically in the therapy-resistant homozygous form of familial
hypercholesterolemia (11)
.
Two animal models are available to study this question: the naturally
occurring rabbit strain Watanabe heritable hyperlipidemic (WHHL) rabbit
with defective LDLR (12)
and the LDLR knockout mouse
strain (LDLR-KO) recently developed by Ishibashi et al.
(13)
. Previous studies indicate that C7OH activity and
bile acid synthesis are suppressed in WHHL rabbits (14)
,
whereas no data are available for LDLR-KO mice. Transient gene therapy
resulting in overexpression of C7OH activity (14)
or
lipoprotein lipase (16)
, as well as increased apoB-48
production (17)
, has recently been shown to reduce plasma
cholesterol in LDLR-deficient mice. In the current investigation we
wanted to test whether LDLR are obligate for the cholesterol-lowering
effects of GH and whether GH could have beneficial effects on plasma
lipids in a situation where LDL receptors are absent. Our approach was
to monitor plasma lipoproteins and hepatic cholesterol metabolism in
normal and LDLR-deficient mice during the continuous infusion of GH.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
and 25 male
wild-type C57 BL/6J mice, purchased from Bommice (Denmark), were used.
LDLR-KO animals had been backbred into C57BL/6J mice for >5
generations. The absence of LDL receptors in LDLR-KO animals was
confirmed by ligand blot of hepatic membranes obtained from pooled
livers from each animal group, using ß-VLDL as ligand (not shown).
Animals were housed under standardized conditions in groups of 10. The
mice had free access to water and chow; the light cycle hours were
between 6 AM and 6 PM. The studies were
approved by the Institutional Animal Care and Use Committee.
At the start of the experiment, osmotic mini pumps (model 2001, Alzet
Corp., Cupertino, CA.) containing human recombinant GH (purchased from
Pharmacia and Upjohn, Uppsala, Sweden) were implanted subcutaneously
under ether anesthesia. Non-GH-treated animals were sham-operated.
Previous control experiments have shown that the plasma lipoprotein
pattern is not altered by the presence of such minipumps
(10)
. GH was infused at a rate of 1.0 mg/kg per day if not
otherwise stated. After 6 days of infusion, animals were anesthetized
with ether and blood was collected from the eye. The mice were then
killed by cervical dislocation between 10 and 12 AM. The
livers were removed and 1 g of liver was taken for subsequent
preparation of microsomes for assay of enzymatic activities as
described below. The remaining liver was immediately frozen in liquid
nitrogen and stored at -70°C.
Cholesterol in the liver was determined after extraction with chloroform/methanol (2:1,v/v) and subsequent drying under nitrogen. Cholesterol in plasma and fast performance liquid chromatography (FPLC) fractions was assayed with the Boehringer Mannheim cholesterol assay kit, using a 5.2 mmol/l cholesterol standard from Merck (Darmstadt, Germany; cat. no. 14164).
Quantitation of mRNA
Hepatic total RNA was isolated by ultracentrifugation on CsCl
after homogenization of tissues in guanidium isothiocyanate. mRNA
levels for the LDLR, HMG CoA reductase, and C7OH were quantitated by
a solution hybridization titration assay using mouse cRNA-probes
(18)
. The mRNA abundance was expressed as copies of mRNA
molecules/cell, assuming 15 pg of RNA/cell; this is not an absolute
quantification.
Enzymatic activities of HMG CoA reductase and C7OH
Hepatic microsomes were prepared by differential
ultracentrifugation of liver homogenates in the absence of fluoride as
described previously (19)
. Microsomal HMG CoA reductase
was assayed from the conversion of [14C] HMG
CoA to mevalonate (19)
and expressed as picomoles formed
per milligram protein per minute. The activity of C7OH was
determined from the formation of 7-hydroxycholesterol (pmol/mg
protein/min) from endogenous microsomal cholesterol by the use of
isotope dilution mass spectrometry as described in detail elsewhere
(20)
. All enzyme assays were carried out in duplicate.
Ligand blot assay of LDL receptors was performed using
125I-labeled rabbit ß-migrating VLDL, as
described (4)
. Membrane proteins were separated by
SDS-PAGE (6% polyacrylamide). After electrotransfer of proteins to
nitrocellulose filters and subsequent incubation with
125I-labeled rabbit ß-VLDL, filters were
exposed onto DuPont X-ray film at -70°C.
Size-fractionation of lipoproteins by FPLC
Equal volumes of plasma from each animal were pooled (1–2 ml)
and the density was adjusted to 1 g/ml with KBr. After
ultracentrifugation at 105 g for
48 h, the supernatant was removed and adjusted with 0.15 M NaCl,
0.01% EDTA, 0.02% sodium azide, pH 7.3. An aliquot of this solution
was injected on a 54 x 1.8 cm Superose 6B column after filtration
through a Millipore 0.45 µm HA filter; 2 ml fractions were selected
(21)
.
Size fractionation of lipoproteins by miniaturized on-line
FPLC
This was performed using a micro-FPLC column (30x0.32 cm
Superose 6B) coupled to a system for on-line lipoprotein separation and
subsequent detection of cholesterol. Ten microliters of plasma were
injected from every animal, and the cholesterol content in lipoproteins
was determined on-line using the commercially available Boehringer
Mannheim cholesterol assay kit (MPR 2 1 442 350), which was
continuously mixed with the separated lipoproteins at a flow rate of 40
+ 40 µl/min. Absorbance was continuously measured at 500 nm, and data
were collected every 10 s using EZ ChromTM software (Scientific
Software, San Ramon, CA).
Statistical analysis
Data are expressed as means ±SE. The significance
of differences between groups was tested by one-way ANOVA, followed by
planned comparison or post hoc comparisons of group means according to
the LSD methods. (Statistica Software, Stat Soft, Tulsa, OK).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Compared with C57BL/6J mice, total cholesterol and triglycerides in
plasma were increased by 250 and 80% respectively
(P<0.001) in untreated LDLR-KO animals (Fig. 1A
). Total hepatic cholesterol was increased by 45%
(P<0.001) in LDLR-KO animals. Treatment with GH reduced the
total plasma cholesterol by more than 40% in LDLR-KO animals
(P<0.001), whereas no significant effect was observed in
C57BL/6J mice. There was a slight reduction of total plasma
triglycerides in LDLR-KO animals (P<0.05); hepatic
cholesterol decreased somewhat in these mice when treated with GH. FPLC
separation of plasma lipoproteins showed that the reduction of
cholesterol in LDLR-KO animals occurred within all size fractions,
although it was most pronounced within LDL and VLDL fractions (Fig. 1B
). In LDLR-KO animals, triglycerides within VLDL were
reduced after GH treatment, whereas GH increased VLDL (and somewhat
reduced HDL) triglycerides in C57BL7/6J mice (Fig. 1C
). The
expression of LDL receptors in C57BL/6J mice was not altered by GH
treatment (not shown).
|
To confirm and further evaluate the unexpected finding of lowered
plasma total and LDL cholesterol levels after infusion of GH in
LDLR-deficient mice, a repeated study with identical design was
performed. In this experiment, we also determined the enzymatic
activities of HMG CoA reductase and C7OH, representing important
rate-limiting steps in the synthesis of cholesterol and bile acids,
respectively. After 6 days of GH infusion, similar effects were
observed on plasma lipids and hepatic cholesterol, although the
reduction of total plasma cholesterol was somewhat less in this
experiment (Fig. 2A
) compared with the previous experiment. Separation of
plasma lipoproteins by FPLC confirmed that GH reduced plasma
lipoprotein cholesterol and triglycerides within LDL and particularly
VLDL particles in the LDLR-KO animals (Fig. 2B
, C
). Again,
the response in C57/BL6J mice was different. Assay of C7OH activity
showed a 60% higher activity (P<0.05) in untreated LDLR-KO
animals as compared with C57/BL6J animals (Fig. 3A
). Treatment with GH increased the activities in both animal
groups (by 48% in LDLR-KO, P<0.01; and by 260% in
C57/BL6J mice, P<0.001). The activity of HMG CoA reductase
was strongly suppressed in LDLR-KO animals, averaging 30% of that
found in control C57/BL6J mice (Fig. 3B
;
P<0.001). GH treatment increased the activity of HMG CoA
reductase by 113 and 53%, respectively, in LDLR-KO
(P<0.01) and C57/BL6J mice (P<0.001).
|
|
Quantitation of the mRNA levels for C7OH by solution hybridization
showed a 60% lower abundance (P<0.001) in LDLR-KO animals
compared with C57/BL6J animals (Fig. 3C
). GH treatment
increased the C7OH mRNA level by 170% (P<0.001) in
LDLR-KO mice, whereas there was a 50% reduction (P<0.001)
in C57/BL6J animals. The mRNA levels for HMG CoA reductase were reduced
by 30% in untreated LDLR-KO compared with C57/BL6J mice. The infusion
of GH normalized this abnormality in the LDLR-KO animals, whereas no
effect was observed in the normal mice.
Thus, in accordance with the findings in normal rats (10)
,
there was no reduction of plasma total cholesterol in normal mice after
infusion with GH. Furthermore, in line with previous results in
Sprague-Dawley rats (9)
, GH infusion in mice stimulated
the activity of C7OH in both normal and LRLR-KO animals. However,
the activity of HMG CoA reductase was suppressed in LDLR-KO animals,
and GH treatment to LDLR-KO mice increased the activity to the same
level as found in normal C57BL/6J animals.
Since GH actually stimulated the HMG CoA reductase activity >100% in
LDLR-KO animals, we wanted to determine the effect of combining GH with
a potent inhibitor of HMG CoA reductase or a potent bile acid binding
resin. Eight groups of 5 LDLR-KO mice each were treated with
atorvastatin, cholestyramine, and GH alone and in combinations. After 4
days of drug treatment, the animals were bled and killed as described
in Materials and Methods. Individual plasma samples were analyzed for
cholesterol and triglycerides, and 10 µl of plasma was subjected to
micro-FPLC analysis. Livers from all animals were frozen for assay of
cholesterol and enzymatic activities of C7OH and of HMG CoA
reductase.
Plasma total cholesterol was reduced by 20% after treatment with
atorvastatin or GH (P<0.05 for both), whereas there was no
significant reduction after cholestyramine. The reduction in plasma
triglycerides was more pronounced, averaging 28, 21, and 20% during
atorvastatin (P<0.001), GH (P<0.01), and
cholestyramine (P<0.05), respectively. The effects were
potentiated when GH was combined with atorvastatin or cholestyramine
treatment, with reductions of cholesterol by 28 and 34%, respectively
(P<0.001), and of triglycerides by 33 and 40%,
respectively (P<0.001). The combination of GH and
cholestyramine or of all three drugs resulted in reductions of
cholesterol by 33–34% (P<0.001) and of triglycerides by
40–46% (P<0.001). To study the changes in plasma
lipoproteins in more detail, we determined the plasma lipoprotein
pattern in all individual samples by micro-FPLC. The means of these
chromatograms are shown in Fig. 4
along with the lipoprotein pattern in normal C57BL/6J mice. Single
therapy with cholestyramine slightly reduced cholesterol in VLDL and
LDL, whereas HDL cholesterol was unaltered (Fig. 4A
). Single
therapy with atorvastatin resulted in a reduction of cholesterol in LDL
and HDL, whereas VLDL cholesterol was unchanged (Fig. 4B
).
Treatment with GH alone reduced cholesterol within VLDL and LDL, and to
some extent in HDL fractions. Combination treatment with GH and
atorvastatin or cholestyramine reduced cholesterol within LDL, HDL, and
particularly within VLDL (Fig. 4C
). Combined treatment with
all three drugs resulted in an even more pronounced reduction (Fig. 4C
). When GH was removed from the triple drug combination
treatment, LDL and particularly VLDL cholesterol increased, whereas HDL
cholesterol was reduced (Fig. 4D
).
|
We then assayed the enzymatic activity of HMG CoA reductase in all
samples (Fig. 5A
). Cholestyramine increased the activity by 134% whereas GH
increased the activity somewhat less (by 52%). When atorvastatin was
used alone, the measured activity of the enzyme was increased 24-fold.
However, when atorvastatin was combined with GH or cholestyramine or
with both, the stimulatory response to atorvastatin was attenuated.
|
The results from the assay of C7OH activity are shown in Fig. 5B
. As expected, cholestyramine increased the activity
(150%, P<0.01), whereas the effects of atorvastatin or
GH were modest and not statistically significant. However, when GH was
added to cholestyramine treatment, the activity increased by almost
threefold (P<0.001). When GH was added to atorvastatin
treatment there was a clear increase (150%, P<0.01).
The highest enzyme activity was observed when atorvastatin was used in
combination with cholestyramine (almost fivefold), whereas the addition
of GH in that situation did not further increase C7OH activity.
Determination of total and free cholesterol in liver homogenates and
microsomes (not shown) revealed that total liver cholesterol was
reduced, particularly in animals receiving cholestyramine alone
(-26%), atorvastatin + GH (-36%), and all three drugs in
combination (-25%). Smaller changes were seen in the other groups
(-18 to -8%). The changes in microsomal free cholesterol were
clearly of a smaller magnitude, speaking against the possibility that
the activity data for C7OH would be heavily influenced by substrate
(cholesterol) availability.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
250% higher in LDLR-KO as compared with
wild-type C57BL/6J mice and could be reduced by up to 40% after GH
infusion. GH reduced plasma triglycerides by 20–30%, whereas hepatic
cholesterol was reduced by 5–8%. It should be noted that these
effects were obtained using a relatively moderate dose of GH: 1–2
mg/kg/day for only 6 days. Long-term chronic studies with GH would
certainly be of interest, but such studies for animals would require
the use of homologous GH to avoid immune reactions against the hormone.
Although the stimulation of LDLR expression induced by GH is probably
of physiological significance (2)
, it is clear from the
present study that other mechanisms are important for the
lipid-lowering effect of this hormone.
How GH, statins and bile acid sequestrants may reduce plasma lipids in
the absence of functional LDL receptors is unclear. The level of LDL
cholesterol in blood reflects the balance between synthesis and
catabolism. In previous studies using transient overexpression of
C7OH in hamsters and LDLR-deficient mice, it was concluded that the
associated reduction of plasma LDL cholesterol was due to a reduced
entry (synthesis) of LDL into plasma (22)
. Evidence for
this was the finding that VLDL was markedly reduced and that LDL
catabolism was not induced. In the present study, cholesterol and
triglycerides in VLDL were decreased during GH infusion (Figs. 1
and 2B
, C
). This would support the concept that reduced hepatic
synthesis may at least partly explain the GH-induced reduction of
plasma lipids in LDLR-KO mice.
It was recently shown that one important effect of GH in
cholesterol metabolism is to stimulate the loss of fecal bile acids and
stimulate the enzymatic activity of C7OH as well as that of HMG CoA
reductase (9)
. Therefore, one would theoretically expect
beneficial lipid-lowering effects by combining these three types of
drugs. This was indeed found. The effect of GH on C7OH activity was
drastic when given in combination with cholestyramine (Fig. 5B
). As mentioned, one clear finding in the current study
was that GH, when given to LDLR-deficient animals, strongly reduced the
severe increase in VLDL and LDL cholesterol and triglycerides within
VLDL. When GH was given to normal mice, there was instead an increase
in VLDL triglycerides as previously observed in normal rats (5
, 10)
. In addition, GH did not reduce HDL cholesterol to the
extent seen in response to atorvastatin. It was found that atorvastatin
alone or when given together with GH, despite an obviously strong
inhibitory effect on the reductase in vivo, could actually induce the
activity of C7OH. It remains to be explored whether this effect of
atorvastatin may be important for the plasma lipid-lowering effect of
that drug in other situations as well.
Some of the effects of GH were obviously different in LDLR-KO mice and
their wild-type controls. The activity of C7OH was higher in LDLR-KO
mice than in normal C57BL/6J mice. This result is in contrast to that
previously described in another animal model of familial
hypercholesterolemia, homozygous LDLR-deficient Watanabe rabbits
(14)
, where the activity of C7OH was lower than in
controls. Despite an increased basal activity of C7OH in LDLR-KO
compared with C57BL/6J mice, the mRNA level for C7OH was strongly
reduced in LDLR-KO mice. Furthermore, after GH treatment, both types of
mice responded with an increased activity of the enzyme whereas the
mRNA levels were reduced in C57BL/6J mice in contrast to the LDLR-KO
mice, which displayed strongly increased C7OH mRNA levels. The
finding of reciprocal relations between mRNA and enzymatic activity was
previously shown to occur in hypophysectomized rats, indicating a
possible post-transcriptional regulation of C7OH activity in some
conditions (9)
. The current data would lend further
support for that concept, but the difference in response to GH in
normal and LDLR-KO mice requires further study.
In conclusion, the infusion of GH to LDLR-KO mice reduces plasma lipids, particularly when used in combination with established lipid-lowering drugs. The effects appear beneficial in that both cholesterol and triglycerides are reduced within VLDL and LDL, and less so in HDL. The feasibility of this strategy of treatment should be evaluated in patients with the homozygous form of familial hypercholesterolemia.
|
ACKNOWLEDGMENTS |
|---|
Received for publication October 11, 2000. Accepted for publication February 2, 2001.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
-hydroxylase in the rat. J. Clin. Invest. 99,2239-2245[Medline]
-hydroxylase by cholesterol in New Zealand white and Watanabe heritable hyperlipidemic rabbits. J. Clin. Invest. 95,1497-1504
-hydroxylase (CYP7a) in mice lacking the low density lipoprotein (LDL) receptor gene. J. Biol. Chem. 273,126-132This article has been cited by other articles:
|
|
 |
|
  C. Galman, M. Matasconi, L. Persson, P. Parini, B. Angelin, and M. Rudling Age-induced hypercholesterolemia in the rat relates to reduced elimination but not increased intestinal absorption of cholesterol Am J Physiol Endocrinol Metab, September 1, 2007; 293(3): E737 - E742. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. Johansson, M. Rudling, T. S. Scanlan, T. Lundasen, P. Webb, J. Baxter, B. Angelin, and P. Parini Selective thyroid receptor modulation by GC-1 reduces serum lipids and stimulates steps of reverse cholesterol transport in euthyroid mice PNAS, July 19, 2005; 102(29): 10297 - 10302. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Matasconi, B. Angelin, and M. Rudling Pituitary control of lipoprotein and bile acid metabolism in male rats: growth hormone effects are not mediated by prolactin Am J Physiol Endocrinol Metab, July 1, 2004; 287(1): E114 - E119. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Lind, M. Rudling, S. Ericsson, H. Olivecrona, M. Eriksson, B. Borgstrom, G. Eggertsen, L. Berglund, and B. Angelin Growth Hormone Induces Low-Density Lipoprotein Clearance but not Bile Acid Synthesis in Humans Arterioscler. Thromb. Vasc. Biol., February 1, 2004; 24(2): 349 - 356. [Abstract] [Full Text]
|
|
|
|
 |
|
 T. B. Twickler, M. J. M. Cramer, G. M. Dallinga-Thie, M. J. Chapman, D. W. Erkelens, and H. P. F. Koppeschaar Adult-Onset Growth Hormone Deficiency: Relation of Postprandial Dyslipidemia to Premature Atherosclerosis J. Clin. Endocrinol. Metab., June 1, 2003; 88(6): 2479 - 2488. [Full Text] [PDF]
|
|
|
|
 |
|
 C. Galman, I. Arvidsson, B. Angelin, and M. Rudling Monitoring hepatic cholesterol 7{alpha}-hydroxylase activity by assay of the stable bile acid intermediate 7{alpha}-hydroxy-4-cholesten-3-one in peripheral blood J. Lipid Res., April 1, 2003; 44(4): 859 - 866. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Gullberg, M. Rudling, C. Salto, D. Forrest, B. Angelin, and B. Vennstrom Requirement for Thyroid Hormone Receptor {beta} in T3 Regulation of Cholesterol Metabolism in Mice Mol. Endocrinol., August 1, 2002; 16(8): 1767 - 1777. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
