| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Prince Henrys Institute of Medical Research (K.N.H., K.P., M.E.E.J., E.R.S.), Clayton, Victoria 3168, Australia; and Department of Biochemistry and Molecular Biology (K.N.H., E.R.S.), Monash University, Clayton, Victoria 3800, Australia
Address all correspondence and requests for reprints to: Kylie Hewitt, Prince Henrys Institute of Medical Research, P.O. Box 5152, Clayton, Victoria 3168, Australia. E-mail: kylie.hewitt{at}phimr.monash.edu.au.
| Abstract |
|---|
|
|
|---|
mRNA expression showed no significant changes. Expression of transcripts encoding adipocyte differentiated regulatory protein, a fatty acid transporter, was significantly elevated in estrogen-deficient males, and 17ß-estradiol replacement significantly reduced these levels. Scavenger receptor class b type 1 showed no significantly changes. This study reveals that the previously reported disruption to triglyceride homeostasis in estrogen-deficient males can be reversed with 17ß-estradiol treatment, indicating an important role for estrogen in maintaining triglyceride and fatty acid homeostasis in males. | Introduction |
|---|
|
|
|---|
Mouse models of estrogen deficiency also present with dyslipidemia. Specifically, the aromatase knockout (ArKO) mouse (4) presents with central obesity, hypercholesterolemia, hyperinsulinemia, hyperleptinemia, and hypertriglyceridemia (5), and importantly the male mice have hepatic steatosis (Ref. 5 and Hewitt, K. N., W. C. Boon, Y. Murata, M. E. Jones, and E. R. Simpson, submitted manuscript). Estrogen receptor (ER)
-knockout (
ERKO) mice and the ER
and -ß double-knockout (
ßERKO) mice (6, 7) also display an obese phenotype similar to the ArKO mice (5), but the hepatic phenotype of these animals has not been reported. The ERß-knockout (ßERKO) mouse does not have a lipid phenotype (7), thereby suggesting ER
is the more important ER in terms of lipid homeostasis.
We have reported that the hepatic steatosis present in ArKO males is due to an accumulation of hepatic triglycerides. Molecular characterization of this phenotype revealed that estrogen deficiency in males led to elevated fatty acid synthase (FAS) and acetyl CoA carboxylase
(ACC
) expression. There was also an increase in expression of adipocyte differentiated related protein (ADRP), a fatty acid transporter present in the liver (Hewitt, K. N., W. C. Boon, Y. Murata, M. E. Jones, and E. R. Simpson, submitted manuscript). There was no change in apolipoprotein E expression, suggesting there was no compensatory increase in very low-density lipoprotein secretion, which would allow an increase in hepatic triglyceride clearance, thus further exacerbating the phenotype (Hewitt, K. N., W. C. Boon, Y. Murata, M. E. Jones, and E. R. Simpson, submitted manuscript). Additionally, we failed to observe an increase in fatty acid ß-oxidation. As well as elevated hepatic triglycerides in estrogen-deficient male mice, previously we have reported that male ArKO mice also have elevated levels of hepatic cholesterol (8). In addition to our model, Toda et al. (9) generated another ArKO mouse model by disrupting exon 9 of the Cyp19 gene. They reported the presence of hepatic steatosis due to a disruption in ß-oxidation, shown at the level of gene expression as well as the catalytic activity of these enzymes. This phenotype was reversed by 17ß-estradiol treatment (10); however, they did not indicate whether this was sexually dimorphic.
Human models of aromatase deficiency have also been identified (11, 12, 13, 14, 15), and, like the ArKO mouse, they also present with dyslipidemia. In one male patient, hepatic steatosis that was reversed with 17ß-estradiol treatment was described (14).
Therefore, the aim of this present study was to attempt to rescue the fatty liver phenotype in the male ArKO mouse with 17ß-estradiol treatment.
| Materials and Methods |
|---|
|
|
|---|
Diets
Soy-free mouse chow (Glen Forest stock feeders, Perth, Australia) is the diet used to feed the mice; it contains no soy products, which are found in regular mouse chow, because soy is known to have estrogenic effects (17). This diet contains 15% of calories as fat (0.02% cholesterol), 20% calories as protein, and 65% of calories as carbohydrate.
Hormone preparation
17ß-estradiol (E2) (Sigma, St Louis, MO) was dissolved in methylene chloride (Unilab, Ackland, New Zealand) and added to peanut oil (methylene chloride/oil, 1:2 vol/vol). The methylene chloride was evaporated by bubbling with nitrogen and warming the solution to 37 C. Vehicle oil injections were identically prepared, omitting E2.
Experimental design
ArKO males received daily sc injections of E2 (3 µg/ kg·d) from 18 wk of age for 6 wk. WT controls did not receive any treatment and ArKO controls received vehicle oil injections.
Tissue collection and histology
Animals were weighed and then killed at 24 wk of age by CO2 asphyxiation. Blood was removed by cardiac puncture and stored at -20 C. The liver was removed, weighed, and part of it snap frozen in liquid nitrogen and stored at -80 C for gene and lipid analysis. Part of the liver was immersion fixed in Bouins fixative then stored in 70% ethanol. Fixed samples were embedded in a random orientation in paraffin and sliced into 7-µm sections. Sections were then stained with hematoxylin, counterstained with eosin, and coverslipped with DPX (BDH, Poole, UK). Gonadal, visceral, and brown adipose tissue (BAT) pad were also removed and weighed.
Measurement of hepatic triglycerides and cholesterol
Hepatic triglycerides and cholesterol were extracted from liver by homogenizing 0.2 g of tissue in 10 ml chloroform/methanol (2:1 vol/vol) (18). Samples were centrifuged for 20 min at 800 g, the lipid phase removed, and the chloroform evaporated. Triglyceride and cholesterol were quantified using Infinity triglyceride lipid stable reagent (Thermo Trace, Melbourne, Australia) and cholesterol lipid incorporating dynamic stabilization technology (Thermo Trace), respectively. Triglyceride and cholesterol calibrators were used as references (Sigma).
Gene analysis
RNA was extracted from liver using the phenol-chloroform method (Ultraspec RNA, Fisher Biotech, Australia) and quantified spectrophotometrically. RNA quality was assessed on an ethidium bromide (Sigma) agarose (Promega Life Sciences, Madison, WI) gel. Two-step RT-PCR was performed using random primers (Roche, Mannheim, Germany) and AMV reverse transcriptase enzyme (Promega). A Lightcycler (Roche) was used to quantitate mouse transcripts using specific primer pairs (Table 1
). Primer pairs were shown to be specific through single peak melting curves, and a single product was seen on an ethidium bromide (Sigma) agarose (Promega) gel corresponding to the appropriate product size as measured by a 1-kb ladder (Promega). All products were sequenced to confirm identity.
|
Statistical analysis
All graphs were expressed as means ± SEM, and statistics were performed using ANOVA (SPSS for Windows, version 10.0, Chicago, IL). A minimum of five animals were used per group.
| Results |
|---|
|
|
|---|
|
Fat pad weight data
ArKO male control mice had significantly heavier gonadal fat pad weight, compared with WT control mice (P = 0.038, Fig. 1C
). E2 replacement did not significantly reduce gonadal fat pad weight in the ArKO mice (P = 0.111, Fig. 1C
).
ArKO control mice also had significantly heavier visceral fat pad weight, compared with WT control mice (P = 0.027, Fig. 1D
), and E2 treatment significantly reduced visceral fat pad weight in the treated ArKO mice (P = 0.04, Fig. 1D
).
There were no significant differences between ArKO and WT male mice in BAT weight (P = 0.098, Fig. 1E
), although there was an apparent increase in weight in the ArKO mice; however, E2 replacement in the ArKO mice led to a significant reduction in BAT weight, compared with ArKO control mice (P = 0.008, Fig. 1E
).
Liver morphology
ArKO male mice have an accumulation of lipids in their livers (Fig. 2B
), which are not apparent in the WT male mice (Fig. 2A
). After E2 replacement there was a reversal of the fatty liver phenotype (Fig. 2C
).
|
|
Expression of transcripts encoding enzymes involved in de novo synthesis of fatty acids
To explain the changes in body, fat, and liver weight, hepatic triglyceride levels after E2 replacement, liver samples were analyzed for expression of various genes, which have previously been shown to the altered in ArKO male mice.
ArKO control mice showed a significant up-regulation of FAS expression, compared with WT mice (P = 0.004). With E2 treatment there was a decrease in FAS expression in the ArKO mice (P = 0.096, Fig. 4A
), although this did not reach significance.
|
, another key enzyme involved in de novo synthesis of fatty acids, showed similar changes; however, these did not reach the level of significance among any of the groups measured (Fig. 4B
Gene expression of fatty acids transporters
ADRP, a fatty acid transporter, was significantly up-regulated in the ArKO male mice, compared with WT control mice (P = 0.005); additionally, there was also a significant decrease in its expression after the E2 replacement in ArKO mice (P = 0.003, Fig. 4C
).
Scavenger receptor class b type 1 is also involved in uptake of fatty acids and cholesterol uptake but showed no significant differences among any of the groups (Fig. 4D
).
| Discussion |
|---|
|
|
|---|
|
Estrogen deficiency in male mice leads to hepatic steatosis, probably due in part to an increase in lipogenesis indicated by increases in FAS and ACC
expression; E2 replacement lowered the expression of FAS. Despite these findings, other studies have shown estrogen to be a stimulator of lipogenesis in chick livers (21, 22), the rat liver (23), and male Xenopus laevis (24), although in ewes E2 treatment inhibited lipogenesis (25). This was, however, in adipose tissue rather than in the liver.
Long-chain fatty acid uptake may also be elevated in the state of estrogen deficiency as seen by increased levels of ADRP expression. This is reversed with E2 replacement. The mechanism whereby estrogen regulates ADRP expression is not yet understood; nevertheless, another transporter, FAT/CD36, is also affected by sex steroids (26). These studies reveal a possible involvement of estrogen in the regulation of FFA transport.
Adipose tissue is believed to have a buffering action by suppressing the release of FFAs into the circulation and increasing triglyceride clearance (27). In the circumstance of obesity, adipose tissue function is altered so that buffering is less effective and the adipocytes are filled, hence resisting further fat storage (27). Conversely, rodents who have lipodystrophy also present with IR and diabetes (28, 29, 30). This lack of adipose tissue prevents any buffering of FFAs, also contributing to the accumulation of triglycerides in the liver as well as skeletal muscle and pancreas (29, 31). Previously we have established that in the state of estrogen deficiency in both males and females, obesity is associated with an increased adipocyte volume (5) and estrogen replacement reverses the obese phenotype by causing a decrease in adipocyte volume, whereas there was very little change in adipocyte number (5, 32). The present study demonstrated that E2 replacement in males decreases obesity by reducing gonadal, visceral, and BAT adipose depots. It is possible that the large adipocyte volumes observed in the estrogen-deficient mice (5, 32) are unable to buffer the effects of FFAs; therefore, the liver, another important site of triglyceride buffering (33), stores trigylcerides as a means of protecting other sites within the body. E2 acts on the adipose depots to reduce their size and hence obviate the need for the liver to assist in the buffering of triglycerides.
The presence of central obesity is not necessarily a predictor of hepatic steatosis; as shown in estrogen-deficient males and females, both are obese but only the males present with hepatic steatosis. In this case centrally mediated factors may be playing an important role; estrogen-deficient males show loss of neurones in the arcuate nucleus and medial preoptic area regions of the hypothalamus (Hill, R., M. E. Jones, S. Pompolo, E. R. Simpson, and W. C. Boon, submitted manuscript). The arcuate nucleus and medial eminence regions contain the highest concentration of leptin receptors in the brain (34). Leptin is a peptide hormone that is secreted from adipose tissue, which acts in a feedback fashion on its receptors that are present in the hypothalamus. Its functions are to regulate food intake and spontaneous physical activity (1), and more recently it has been shown to inhibit lipogenesis and cholesterol synthesis and stimulate fatty acid oxidation (1, 34). Leptin-deficient models present with hepatic steatosis; hence, there may be a reduction in leptin signaling in the ArKO males, which is due to damage to the hypothalamus that is contributing to the hepatic steatosis.
GH secretion is also centrally mediated, and GH-deficient patients have also presented with hepatic steatosis (35, 36), and reversal has been shown in one patient on GH replacement (36). GH is known to act on liver to regulate the expression of certain cytochrome P450 isoforms in a sexually dimorphic fashion (37, 38). Whether its action extends to triglyceride balance remains to be determined.
Previously we reported that estrogen deficiency in males resulted in elevated hepatic cholesterol levels, which were further increased by a high-cholesterol diet (8). This present study has shown that although estrogen deficiency led to elevated hepatic cholesterol levels, E2 replacement in the males did not reverse this phenotype. Thus, hepatic cholesterol levels may be regulated independently of obesity.
The emergence of estrogen as an important regulator of lipid homeostasis is becoming increasingly clear. This present study adds further weight to this concept and highlights the role of estrogen in regulating lipid homeostasis in males, particularly hepatic triglyceride homeostasis.
| Footnotes |
|---|
Abbreviations: ACC
, Acetyl CoA carboxylase
; ADRP, adipocyte differentiated related protein; ArKO, aromatase knockout; BAT, brown adipose tissue; E2, 17ß-estradiol; ER, estrogen receptor; ERKO, ER knockout; FAS, fatty acid synthase; FFA, free fatty acid; IR, insulin resistance; WT, wild-type.
Received October 14, 2003.
Accepted for publication December 8, 2003.
| References |
|---|
|
|
|---|
knockout mice. Proc Natl Acad Sci USA 97:1272912734
-deficient male mice. Biochem Biophys Res Commun 278:640645[CrossRef][Medline]
. Cell 93:693704[CrossRef][Medline]
This article has been cited by other articles:
![]() |
M. Penza, C. Montani, A. Romani, P. Vignolini, B. Pampaloni, A. Tanini, M. L. Brandi, P. Alonso-Magdalena, A. Nadal, L. Ottobrini, et al. Genistein Affects Adipose Tissue Deposition in a Dose-Dependent and Gender-Specific Manner Endocrinology, December 1, 2006; 147(12): 5740 - 5751. [Abstract] [Full Text] [PDF] |
||||
![]() |
F. Magkos, B. W. Patterson, and B. Mittendorfer No effect of menstrual cycle phase on basal very-low-density lipoprotein triglyceride and apolipoprotein B-100 kinetics Am J Physiol Endocrinol Metab, December 1, 2006; 291(6): E1243 - E1249. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. Hamaguchi, T. Kojima, N. Takeda, T. Nakagawa, H. Taniguchi, K. Fujii, T. Omatsu, T. Nakajima, H. Sarui, M. Shimazaki, et al. The Metabolic Syndrome as a Predictor of Nonalcoholic Fatty Liver Disease Ann Intern Med, November 15, 2005; 143(10): 722 - 728. [Abstract] [Full Text] [PDF] |
||||
![]() |
E. R. Simpson, M. Misso, K. N. Hewitt, R. A. Hill, W. C. Boon, M. E. Jones, A. Kovacic, J. Zhou, and C. D. Clyne Estrogen--the Good, the Bad, and the Unexpected Endocr. Rev., May 1, 2005; 26(3): 322 - 330. [Full Text] [PDF] |
||||
![]() |
P. S. Cooke and A. Naaz Role of Estrogens in Adipocyte Development and Function Experimental Biology and Medicine, December 1, 2004; 229(11): 1127 - 1135. [Abstract] [Full Text] [PDF] |
||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |