Postnatal Testosterone Exposure Results in Insulin Resistance ...

0013-7227/07/$15.00/0 Printed in U.S.A.

Endocrinology 148(11):5369 –5376 Copyright © 2007 by The Endocrine Society doi: 10.1210/en.2007-0305

Postnatal Testosterone Exposure Results in Insulin Resistance, Enlarged Mesenteric Adipocytes, and an Atherogenic Lipid Profile in Adult Female Rats: Comparisons with Estradiol and Dihydrotestosterone Camilla Alexanderson, Elias Eriksson, Elisabet Stener-Victorin, Theodore Lystig, Britt Gabrielsson, Malin Lo¨nn, and Agneta Holma¨ng Departments of Physiology (C.A., E.S.-V., A.H.) and Pharmacology (E.E.), Institute of Neuroscience and Physiology and Department of Molecular and Clinical Medicine, Institute of Medicine (T.L., B.G., M.L.), The Sahlgrenska Academy, Go¨teborg University, 405 30 Go¨teborg, Sweden Postnatal events contribute to features of the metabolic syndrome in adulthood. In this study, postnatally administered testosterone reduced insulin sensitivity and increased the mesenteric fat depot, the size of mesenteric adipocytes, serum levels of total cholesterol, low-density lipoprotein cholesterol, and triglycerides, and the atherogenic index in adult female rats. To assess the involvement of estrogen and androgen receptors in these programming effects, we compared testosterone-exposed rats to rats exposed to estradiol or dihydrotestosterone (DHT). Estradiol-treated rats had lower insulin sensitivity than testosterone-treated rats and, like those rats, had enlarged mesenteric adipocytes and increased triglyceride levels. DHT also reduced insulin sensitivity but did not mimic the other metabolic effects of testosterone. All treated

I

N UTERO AND EARLY-LIFE events can contribute to metabolic disturbances such as type 2 diabetes, cardiovascular disease, and hyperlipidemia in adulthood (1–3). In addition to effects of nutrition, it is becoming clear that other factors that influence fetal development may contribute to the onset of disease or abnormal physiological function in adulthood. Fetal exposure to excessive levels of steroid hormones (e.g. glucocorticoids) or sex hormones can permanently alter the adult female phenotype (4, 5). The sexual differentiation of the fetus reflects a complex interplay between genetic and hormonal factors (6, 7). A key player is testosterone, which virilizes the male fetus during certain critical periods before and, in some species such as rodents, shortly after birth. Female fetuses exposed to excessive levels of testosterone during these periods may display permanent abnormalities with respect to, e.g. the urogenital tract, brain morphology, and behavior in adulthood (6, 8, 9). In humans, this phenomenon is illustrated by the characteristics of girls with congenital adrenal hyperplasia First Published Online July 26, 2007 Abbreviations: AR, Androgen receptor; DHT, dihydrotestosterone; ER, estrogen receptor; HDL-C, high-density lipoprotein cholesterol; LDL-C, low-density lipoprotein cholesterol; TC, total cholesterol; TG, triglycerides. Endocrinology is published monthly by The Endocrine Society (http:// www.endo-society.org), the foremost professional society serving the endocrine community.

rats were probably anovulatory, but only those treated with testosterone had reduced testosterone levels. This study confirms our previous finding that postnatal administration of testosterone reduces insulin sensitivity in adult female rats and shows that this effect is accompanied by unfavorable changes in mesenteric fat tissue and in serum lipid levels. The findings in the estradiol and DHT groups suggest that estrogen receptors exert stronger metabolic programming effects than androgen receptors. Thus, insults such as sex hormone exposure in early life may have long-lasting effects, thereby creating a predisposition to disturbances in insulin sensitivity, adipose tissue, and lipid profile in adulthood. (Endocrinology 148: 5369 –5376, 2007)

(10). Testosterone affects the urogenital tract mainly through its metabolite dihydrotestosterone (DHT), acting through androgen receptors (ARs) (11); however, the masculinization of the brain may be mediated either by DHT through the AR or by the other major testosterone metabolite, estradiol, acting through ␣- and ␤-estrogen receptors (ERs). The relative roles of these receptors in this context are species dependent. For example, the AR plays a larger role in the rhesus monkey (9), whereas the ER appears to be more important in rodents (12). In line with the well-established influence of sex steroids on adult metabolism, the prevalence of metabolic disorders differs between men and women, as does normal fat distribution (13). Because prenatal and neonatal factors may influence the susceptibility to metabolic disturbances in adulthood, and because steroids play an important role in the early programming of the organism, the possibility that early exposure to sex steroids may influence metabolism later on in life is worth exploring. We have previously provided support for the existence of such an influence by showing that administration of one dose of testosterone to female pups, corresponding to the level of androgen secretion in newborn male rats, triggers insulin resistance and centralization of body fat in adulthood (14). Similarly, prenatal androgenization has been reported to influence various metabolic indices, including insulin sensitivity, in female rhesus monkeys and sheep (15–23). Moreover, congenital adrenal

5369

5370

Endocrinology, November 2007, 148(11):5369 –5376

hyperplasia in humans is associated with enhanced body mass index and reduced insulin sensitivity, although it is unclear whether these aberrations are due to the disorder per se or to the glucocorticoids used to treat these subjects (10). This study had two aims. First, we wanted to confirm and extend our previous observation that early exposure of testosterone (3 h after birth) may influence metabolism in the adult female rat by assessing to what extent this treatment influences not only insulin sensitivity and body fat distribution, as previously shown, but also, e.g. lipid profile and adipocyte size. And second, we wanted to assess whether this early influence of testosterone is mediated by AR, ER, or both AR and ER by comparing the effects of testosterone with those of estradiol and DHT, respectively. Materials and Methods Animals Nulliparous time-mated Wistar female rats (Scanbur BK AB, Sollentuna, Sweden) were maintained under controlled noise-free conditions (light from 0700 –1700 h; temperature 21 ⫾ 2 C; humidity 55– 65%) with one rat per cage until parturition. Pups were raised with a lactating mother until 21 d of age. The rats were then housed four to five per cage and fed standard rat pellets ad libitum. Standard principles of laboratory animal care were followed. All procedures were approved by the animal ethics committee of Go¨teborg University.

Study procedures Within 3 h after birth, female pups were weighed and sc injected with 1 mg testosterone propionate (Apoteksbolaget, Stockholm, Sweden) (n ⫽ 10), 1 mg DHT propionate (Steraloids, Newport, RI) (n ⫽ 10), or 0.5 mg estradiol benzoate (Apoteksbolaget) dissolved in vehicle (n ⫽ 5). Controls (n ⫽ 12) received vehicle only. Within 1 wk after delivery, pups were redistributed so that each lactating mother had six to seven pups from different experimental groups. Fasting plasma samples were collected for cholesterol and triglyceride (TG) analyses at 7 wk of age. At 9 wk, tail blood samples were collected for analyses of serum testosterone and estradiol. At 15 wk, insulin sensitivity was measured with a euglycemic-hyperinsulinemic clamp. At the end of the study, rats were decapitated, and mesenteric and inguinal adipose tissues were removed for measurements of depot weight and adipocyte size.

Vaginal smears The estrous status was determined from vaginal smears taken daily at 8 –9 wk of age. The rat estrous cycle (estrus, diestrus 1, diestrus 2, and proestrus) usually lasts about 4 d (24). Cycles of 4 –5 d with a characteristically clear ovulation and a rich amount of epithelial cells without leukocytes in the smears were considered normal. Sampling for sex hormone analyses and clamp measurements in controls were performed in the estrous phase.

Euglycemic-hyperinsulinemic clamp At 15 wk of age, the rats underwent a euglycemic-hyperinsulinemic clamp as described (25). Briefly, the rats were anesthetized with thiobutabarbital sodium (Inactin; Sigma, St. Louis, MO; 130 mg/kg body weight). Catheters were inserted into the left carotid artery for blood sampling and into the right jugular vein for infusion of glucose and insulin. Body temperature was maintained at 37 C with a heating blanket. After a bolus injection, insulin (100 U/ml, human Actrapid; Novo Nordisk Pharma, Copenhagen, Denmark) was continuously infused at a rate of 8 mU/min䡠kg. To maintain plasma glucose concentration at 6 mmol/liter, a 20% glucose solution was infused at a rate guided by the glucose concentration in 10-␮l blood samples obtained every 5 min for 40 min and then every 10 min. The mean glucose infusion rate was calculated from values during the last 90 –120 min. At 110 and 120 min of infusion, 50-␮l blood samples were taken for determination of insulin

Alexanderson et al. • Postnatal Sex Steroid Programming Alters Metabolism

concentration. A total of less than 1.5 ml blood was used for the determinations; this was compensated for by the infusion volumes.

Analytical methods Plasma insulin was analyzed with human insulin ELISA kits (Mercodia, Uppsala, Sweden). Serum testosterone and estradiol were determined with RIA kits (testosterone RIA kit, DSL-4100; third-generation estradiol RIA, DSL-39100; Diagnostic Systems Laboratories, Webster, TX). Serum levels of total cholesterol (TC), high-density lipoprotein cholesterol (HDL-C), and TG were determined enzymatically with Konelab autoanalyzer version 2.0. Low-density lipoprotein cholesterol (LDL-C) was calculated as TC ⫺ (HDL-C ⫹ TG)/2, and the atherogenic index was calculated as (TC ⫺ HDL-C)/HDL-C, as described (26).

Adipocyte size Approximately 0.4 g adipose tissue was cut into small pieces and treated with 1.05 mg/ml collagenase (type A; Roche, Mannheim, Germany) in MEM (Invitrogen, Carlsbad, CA) containing 5.5 mm glucose, 25 mm HEPES, 4% bovine albumin (fraction V; Sigma), and 0.15 ␮m adenosine (pH 7.4) for 60 min at 37 C as described (27). After filtration through a 250-␮m nylon mesh, the adipocytes were washed three times and suspended in fresh medium. The mean adipocyte size and the size distribution of the cell population were determined by computerized image analysis (KS400 software; Carl Zeiss, Oberkochen, Germany) (28). In brief, the cell suspension was placed between a siliconized glass slide and a coverslip and transferred to the microscope stage. Nine random visual fields were photographed with a CCD camera (Axiocam; Carl Zeiss). The surface of the relevant areas was measured automatically, and the diameter of the corresponding circles was calculated. Uniform microspheres 98.00 ␮m in diameter (Bangs Laboratories, Fishers, IN) were used as a reference.

Statistical analysis Results are expressed as mean ⫾ sem. Unpaired t tests were used for pairwise comparisons and ANOVA and Fisher’s test for multiple comparisons. Adipocyte size distributions were compared by using twosample Kolmogorov-Smirnov (KS) statistics (29). An exact P value for the comparison of two groups A and B was calculated through permutations. For n subjects in group A and m subjects in group B, KS statistics were calculated for all possible ways of dividing n ⫹ m subjects into two groups of sizes n and m. The observed KS statistic was then ranked against the KS statistics from all of the possible permutations. The permutation P value is the percentage of possible KS statistics that are at least as extreme as the KS statistic from the original data. For these comparisons, statistical calculations were performed with the R language (http://www.R-project.org). P ⬍ 0.05 was considered significant; all tests are two sided.

Results Body weight

At the end of the study, estradiol-exposed rats were significantly heavier than controls (316 ⫾ 15 vs. 270 ⫾ 9 g, P ⬍ 0.05). Body weight was similar in controls and in rats treated with testosterone (289 ⫾ 6 g) or DHT (278 ⫾ 7 g). Vaginal smears

Because the vaginal openings were absent in all the hormone-treated groups, estrous cyclicity could not be assessed by vaginal smears. All control rats had a normal estrous cycle of 4 –5 d. Fat depots

The mesenteric adipose tissue was heavier in testosteroneexposed rats than in controls. However, the inguinal adipose tissue weight was similar in the testosterone group and con-


trols. The weights of mesenteric and inguinal fat depots in the other treatment groups did not differ from those of controls (Table 1). Adipocyte size

In rats treated with testosterone or estradiol, but not in those treated with DHT, the size distribution curve of mesenteric adipocytes was shifted to the right, and the mean mesenteric adipocyte size was greater than in controls (Fig. 1 and Table 1). In the inguinal fat depot, adipocyte size and size distribution were similar in treated rats and controls (Fig. 2 and Table 1). In each adipocyte sample, 446-4835 adipocytes (mean 1777 ⫾ 159) were analyzed. The mean diameter of the reference microspheres was 98.0 ⫾ 0.06 ␮m (range 97.6 –98.4 ␮m; n ⫽ 12). Glucose infusion rate

During euglycemic-hyperinsulinemic clamp studies, the glucose infusion rate was significantly lower in all treatment groups than in controls, indicating peripheral insulin resistance, and was significantly lower in estradiol-treated rats than in those treated with testosterone or DHT (Fig. 3). At 60 –120 min (steady state), the plasma glucose level was approximately 6 mmol/liter, and the plasma insulin concentrations did not differ significantly (controls, 310.7 ⫾ 29.8 mU/liter; testosterone, 296.6 ⫾ 26.5 mU/liter; DHT, 280.3 ⫾ 33.9 mU/liter; estradiol, 403.6 ⫾ 28.6 mU/liter).


5371

Discussion

This study shows that a single neonatal dose of testosterone, which presumably activates both AR and ER, causes insulin resistance, increase in intraabdominal adipose tissue and adipocyte size, and dyslipidemia in adult female rats. Rats treated with estradiol, which activates ER only, displayed the same aberrations, except for the increase in visceral adipose tissue. Although insulin resistance was even more pronounced in the estradiol-treated animals than in those given testosterone, the dyslipidemia was less marked. Rats exposed to DHT, which activates only AR, had reduced insulin sensitivity but none of the other aberrations observed in testosterone- and estradiol-treated rats. The data suggest that postnatal activation of the ER exerts a stronger influence than postnatal activation of the AR on the adipose tissue of the adult animal but that activation of either the AR or the ER shortly after birth may cause insulin resistance at adulthood. It should be noted that the rats given testosterone, estradiol, or DHT probably all had in common that they were anovulatory, as indicated by postmortem morphological examinations of their ovaries revealing absence of corpora lutea (data not shown). The fact that we observed clear-cut differences between the three hormone-treated groups with respect to various metabolic indices, however, indicate that the influence of early hormone administration on metabolism could not merely be secondary to the absence of cyclic variations in female sex steroids. Adipose tissue, adipocyte size, and size distribution

Lipid metabolism

Testosterone-treated rats had increased concentrations of TC, LDL-C, and TG and a higher atherogenic index. In estradiol-treated rats, only the TG concentrations were higher than in controls. The lipid profile was unaltered in the DHT group (Table 2). Testosterone and estradiol concentrations

Testosterone concentrations were lower in testosteronetreated rats than in controls (0.9 ⫾ 0.1 vs. 2.6 ⫾ 0.4 nmol/liter, P ⬍ 0.01), but no statistically significant differences were observed in rats treated with DHT (4.5 ⫾ 0.9 nmol/liter) or estradiol (1.5 ⫾ 0.5 nmol/liter) vs. controls. Estradiol concentrations did not differ significantly between groups (controls, 0.07 ⫾ 0.005 nmol/liter; testosterone, 0.09 ⫾ 0.013 nmol/liter; DHT, 0.08 ⫾ 0.014 nmol/liter; estradiol, 0.06 ⫾ 0.002 nmol/liter).

In our study, we used a new computer-based method that can detect subtle variations in the size distribution of adipocytes isolated by collagenase digestion (28). Exposure to either testosterone or estradiol, but not to DHT, increased the size of mesenteric adipocytes and shifted the size distribution curve to the right. In the testosterone-treated rats, this was accompanied by a significant increase in the weight of the mesenteric adipose tissue (whereas no effect on the inguinal adipose tissue was observed). In the estradiol-treated animals, the weight of both the mesenteric and the inguinal adipose tissue trended higher, but neither of these effects reached the level of statistical significance. This may, however, reflect low statistical power because there were fewer rats in the estradiol group than in the other groups. Therefore, our data on the possible effects of estradiol treatment on fat tissue weight should be interpreted with caution. The lack of effect of DHT on adipocyte size and on the

TABLE 1. Absolute weight, relative weight, and adipocyte size of mesenteric and inguinal adipose tissues Mesenteric adipose tissue

Inguinal adipose tissue

Group

Weight (g)

Weight (g/kg BW)

Adipocyte size (␮m)

Weight (g)

Weight (g/kg BW)

Adipocyte size (␮m)

Controls Testosterone DHT Estradiol

4.2 ⫾ 0.4 5.4 ⫾ 0.4a 4.5 ⫾ 0.4 5.2 ⫾ 0.5

15.1 ⫾ 1.1 18.5 ⫾ 0.9a 16.2 ⫾ 1.2 16.2 ⫾ 0.7

69.0 ⫾ 3.4 83.1 ⫾ 3.0a 73.7 ⫾ 3.2 85.6 ⫾ 2.6b

2.3 ⫾ 0.3 2.2 ⫾ 0.1 2.5 ⫾ 0.3 3.1 ⫾ 0.3

8.1 ⫾ 0.8 7.7 ⫾ 0.3 9.0 ⫾ 1.0 9.7 ⫾ 1.1

65.2 ⫾ 4.6 71.1 ⫾ 3.6 72.1 ⫾ 3.0 67.0 ⫾ 3.7

Values are mean ⫾ SEM (n ⫽ 4 –10 rats per group). BW, Body weight. a P ⬍ 0.05 vs. controls (unpaired t test). b P ⬍ 0.01 vs. controls (unpaired t test).

5372



B

A 1.0

1.0

Dihydrotestosterone Mesenteric Control Mesenteric (n=16289)

Testosterone Mesenteric (n=9332) Control Mesenteric (n=16289)

0.8

0.6

P=0.017 0.4

Proportion < = x

0.8

Proportion < = x

(n=14015)

0.6

P= 0.76 0.4

0.2

0.2

0.0

0.0

50

150

100

50

100

150

Adipocyte diameter (µm)


C 1.0

Estrogen Mesenteric (n=6832) Control Mesenteric (n=16289)

Proportion < = x

0.8

P=0.016

0.6

0.4

0.2

0.0

50

100

150

Adipocyte diameter (µm) FIG. 1. Size distribution of mesenteric adipocytes in controls (n ⫽ 5) vs. rats treated with testosterone (A) (n ⫽ 6), DHT (B) (n ⫽ 5), or estradiol (C) (n ⫽ 4). The data were analyzed with the KS test in combination with permutation analysis. The vertical line represents the KS distance.

weight of the adipose tissue suggests that the effect of testosterone on these parameters is mediated by ER rather than AR. Thus, it is reasonable to conclude that early ER activation leads to enhanced size of the mesenteric adipocytes. Such an effect might be expected to lead to an increase in the weight of the mesenteric tissue; however, the effect of estradiol on the weight of the mesenteric adipose tissue did not reach statistical significance. This may, as mentioned, be due to low power, but it is also possible that activation of ER without simultaneous activation of AR may lead to an increase in adipocyte size but to an accompanying reduction in adipocyte number, the latter effect being counteracted by the parallel activation of AR obtained in the testosterone-treated but not in the estradiol-treated animals. The possibility that early ER activation indeed leads to a reduction in adipocyte number gains support by a previous study in which estrogen administered 2–12 d postnatally reduced the number of fat cells (30); however, in this study, where adipocyte size was not measured, the adipose mass was reduced. To shed further light on this issue, studies assessing both the size and the number of adipocytes are clearly warranted. Several observations support the notion that sex steroids influence the size as well as the number of adipocytes. Ev-

idence from experiments in knockout mice hence suggests that absence of either the ER␣ or the AR causes adipocyte hypertrophy and hyperplasia (as well as impaired insulin sensitivity) (31–33); moreover, aromatase knockout mice, which cannot synthesize estrogens and have high testosterone levels, also are obese and have enlarged adipocytes (34). The fact that the change in sex steroid activity induced by genetic manipulation in these different knockout strains is present throughout development and adult life, however, makes it hazardous to compare these effects with those observed in our study in rats, where only one dose of sex steroids was given. Adult women with high androgen levels are often characterized by an accumulation of visceral fat (35), suggesting that androgens promote abdominal obesity in females. This conclusion, which is supported by the effects of testosterone in female-to-male transsexuals (36), may be regarded as contradictory to our study, which shows that rats given neonatal testosterone have increased mesenteric adipose tissue but reduced testosterone levels. If the influence of early testosterone administration on adipose tissue is interpreted in terms of masculinization, early administration of testosterone must be expected to enhance the responsiveness of fat



B

A 1.0

1.0

Dihydrotestosterone Inguinal Control Inguinal (n=5435)

Testosterone Inguinal (n=8110) Control Inguinal (n=5435)

(n=7377)

0.8

0.6

Proportion < = x

0.8

Proportion < = x

5373

P=0.74

0.4

0.4

0.2

0.2

0.0

0.0 50

200

150

100

P=0.64

0.6

50

100

150

200



C 1.0

Estrogen Inguinal (n=4638) Control Inguinal (n=5435)

Proportion < = x

0.8

P=0.55

0.6

0.4

0.2

0.0 50

100

150


FIG. 2. Size distribution of inguinal adipocytes in controls (n ⫽ 5) vs. rats treated with testosterone (A) (n ⫽ 6), DHT (B) (n ⫽ 5), or estradiol (C) (n ⫽ 4). The data were analyzed with the KS test in combination with permutation analysis. The vertical line represents the KS distance.

tissue to androgens in the adult organism. In line with this possibility, it has been suggested that one effect of the early influence of androgens on the brain is to enhance its responsiveness to testosterone in the adult organism (37). To what extent early androgen exposure may sensitize also the fat tissue to the influence of testosterone should be the subject of forthcoming studies. In this context, it should, however, also be considered that some studies indeed suggest that lowering androgen levels may lead to enhanced body weight in hyperandrogenic women (38). Lipid metabolism

Our analysis of serum lipids showed increases in the atherogenic index and in TC, LDL, and TG levels in testosterone-exposed rats, increased TG levels in the estradioltreated rats, but no abnormalities in the DHT group. The apparent effects of testosterone and estradiol on serum lipids may be explained by the influence of estrogen on the tightly regulated hepatic enzymes involved in cholesterol synthesis, uptake, and clearance. Studies in aromatase knockout mice, for example, which display age-progressive obesity, hepatic steatosis, and hypercholesterolemia (39), have shown that estrogen participates in the regulation of hepatic cholesterol metabolism only in females, indicating that this important

homeostatic pathway is sexually dimorphic (40). Notably, studies in rats have shown that neonatally administered testosterone exerts a programming influence on the activity of hepatic enzymes involved in the metabolism of androgens (41, 42). It would therefore be of interest to investigate the programming effects of sex steroids on hepatic lipid metabolism (e.g. the transcriptional regulation of cholesterol synthesis enzymes). Insulin sensitivity

In our euglycemic-hyperinsulinemic clamp studies, insulin sensitivity was reduced in all three groups of rats. The effects of a single postnatal dose of testosterone on insulin sensitivity are in agreement with our previous findings (14). Because the clamp technique mainly measures the insulin sensitivity of muscle (43), the testosterone injection might have irreversibly changed mechanisms involved in muscle tissue insulin sensitivity. DHT-treated animals displayed a reduction in insulin sensitivity of comparable magnitude as seen in testosterone-treated animals; in the estradiol-exposed group, this effect was even more pronounced. This hence seems to be an effect that may be elicited by activation of either ER or AR in the postnatal phase. That both AR and ER␣

5374



Glucose Infusion Rate (mg/kg/min)

25

*

20

predictor of type 2 diabetes, is associated with insulin resistance and may represent a failure of the adipose tissue mass to expand and thus to accommodate an increased energy influx (47, 48). In addition, hypertrophic adipocytes display adipokine gene expression likely to enhance the progression of insulin resistance (49, 50). There are indications that the size of visceral adipocytes is a strong marker for impaired insulin action, whereas the sc adipocyte size appears to have a weaker impact (51–53). Consequently, the reduced insulin sensitivity in rats exposed to estradiol or testosterone, but not in those exposed to DHT, may reflect, at least in part, the increased size of mesenteric adipocytes.

*

†

†

15

***

10

5

0

Control

T

DHT

E

Conclusions

FIG. 3. Glucose infusion rate in control rats and rats treated with testosterone (T), DHT, or estradiol (E). The infusion rate was lower in all groups than in controls and was higher in the DHT and testosterone groups than in the estradiol group. Values are mean ⫾ SEM (n ⫽ 5–10). *, P ⬍ 0.05; ***, P ⬍ 0.001 vs. controls; †, P ⬍ 0.05 vs. estradiol group (ANOVA).

(but not ␤) may influence insulin sensitivity gains support also from previous mouse knockout studies (33, 44). Interestingly, administration of testosterone to adult female rats led to insulin resistance in muscle after 48 h (45) and abdominal obesity after 12 wk of treatment (25), the same phenotype we found in our study. These findings suggest that androgens are directly involved in the insulin signaling cascade, an involvement that seems to be independent of the effect on visceral obesity (which may further reduce the insulin sensitivity) (25, 45). As discussed above, the observation that testosterone-treated rats had low testosterone levels, but displayed metabolic and anthropometric changes normally associated with high levels of testosterone in females, may seem contradictory but could tentatively be due to an altered responsiveness to this hormone. It is noteworthy that all hormone-treated groups displayed insulin resistance, despite considerable differences in their testosterone levels, which were reduced in the testosterone-treated rats but not in the other groups. Prenatally androgenized female rhesus monkeys have been shown to display a metabolic phenotype similar to that observed in the testosterone-treated animals in our study, with insulin resistance, increased abdominal fat, and dyslipidemia (46). Unfortunately the possible effects of estradiol or DHT have not been assessed in this model. Enlargement of abdominal adipocytes, an independent

This study confirms our previous observation that postnatal administration of testosterone leads to insulin resistance in adult female rats and extends this finding by showing that this effect is accompanied by changes in the mesenteric adipose tissue and in serum lipid levels. The outcome in rats administered estradiol or DHT suggests that ER activation exerts stronger programming effects on metabolic indices than AR activation but that AR activation without concomitant ER activation is sufficient to cause insulin resistance. Female rats administered testosterone, estradiol, or DHT are in all likelihood all anovulatory; the observation that these different treatments lead to different effects on metabolism, however, indicates that the influences on metabolism are not merely the result of the animals being noncycling. Likewise, there was no apparent relationship between testosterone levels in the adult rats and metabolic indices. Thus, the effects of early sex steroid administration on metabolism are clearly not merely secondary to an influence on serum levels of sex steroids in the adult animal. By administering a dose of testosterone yielding hormone levels corresponding to those normally observed postnatally in male rats, which are much higher than those normally observed in females, we aimed to shed light on mechanisms underlying normal sexual dimorphism with respect to metabolism. The fact that sex steroids are capable of exerting an early programming influence on the metabolism on the adult organism, however, suggests that also more modest aberrations in the hormonal environment of the female fetus may influence the susceptibility to metabolic aberrations and somatic illness in adulthood.

TABLE 2. Serum concentrations of TC, LDL-C, and TG (mmol/liter) and atherogenic index in 7-wk-old controls and testosterone-, DHT-, and estradiol-treated rats Group

Control Testosterone DHT Estradiol

Serum lipid level (mmol/liter) TC

LDL-C

TG

2.7 ⫾ 0.15 3.6 ⫾ 0.15c 2.5 ⫾ 0.13 2.9 ⫾ 0.16

0.2 ⫾ 0.06 0.6 ⫾ 0.09b 0.2 ⫾ 0.07 0.2 ⫾ 0.06

0.9 ⫾ 0.05 1.3 ⫾ 0.10b 1.0 ⫾ 0.13 1.2 ⫾ 0.12a

Values are mean ⫾ SEM (n ⫽ 5–11 rats per group). a P ⬍ 0.05 vs. controls (unpaired t test). b P ⬍ 0.01 vs. controls (unpaired t test). c P ⬍ 0.001 vs. controls (unpaired t test).

Atherogenic index

0.22 ⫾ 0.02 0.34 ⫾ 0.02c 0.25 ⫾ 0.03 0.26 ⫾ 0.02


Acknowledgments We gratefully acknowledge the excellent technical assistance provided by Britt-Mari Larsson, Birgitta Odeń, and Lisa Svedbom. Received March 6, 2007. Accepted July 17, 2007. Address all correspondence and requests for reprints to: Camilla Alexanderson, Institute of Neuroscience and Physiology, Department of Physiology/Endocrinology, Sahlgrenska Academy, Go¨teborg University, Box 434, 405 30 Go¨teborg, Sweden. E-mail: [email protected]. This work was supported by grants from Novo Nordisk Foundation, the Royal Society of Arts and Sciences in Go¨teborg, The Go¨teborg Medical Society, the Swedish Medical Research Council (Project Nos. 12206, 14735, 14736, 6399, 2005-72VP-15445-01A, and 2005-72VX-15276-01A), Petrus and Augusta Hedlund’s Stiftelse, the Swedish Diabetes Association Research Foundation, the National Board of Health and Welfare, Wilhelm and Martina Lundgren Foundation, Adlerbertska Research Foundation, the Swedish federal government under the agreement between the government and the county councils concerning economic support for providing an infrastructure for research and education of doctors, and Tore Nilssons Stiftelse fo¨r Medicinsk Forskning. Disclosure Statement: The authors have nothing to disclose.

References 1. Barker DJ 1990 The fetal and infant origins of adult disease. BMJ 301:1111 2. Barker DJ, Hales CN, Fall CH, Osmond C, Phipps K, Clark PM 1993 Type 2 (non-insulin-dependent) diabetes mellitus, hypertension and hyperlipidaemia (syndrome X): relation to reduced fetal growth. Diabetologia 36:62– 67 3. Lucas A 1991 Programming by early nutrition in man. Ciba Found Symp 156:38 –50; discussion 50 –55 4. Kapoor A, Dunn E, Kostaki A, Andrews MH, Matthews SG 2006 Fetal programming of hypothalamo-pituitary-adrenal function: prenatal stress and glucocorticoids. J Physiol 572:31– 44 5. Robinson J 2006 Prenatal programming of the female reproductive neuroendocrine system by androgens. Reproduction 132:539 –547 6. Rey R, Picard JY 1998 Embryology and endocrinology of genital development. Baillieres Clin Endocrinol Metab 12:17–33 7. Warne GL, Kanumakala S 2002 Molecular endocrinology of sex differentiation. Semin Reprod Med 20:169 –180 8. Goy R, McEwen B 1980 Sexual differentiation of the brain. Cambridge, MA: MIT Press 9. Resko JA, Roselli CE 1997 Prenatal hormones organize sex differences of the neuroendocrine reproductive system: observations on guinea pigs and nonhuman primates. Cell Mol Neurobiol 17:627– 648 10. Ogilvie CM, Crouch NS, Rumsby G, Creighton SM, Liao LM, Way GS 2006 Congenital adrenal hyperplasia in adults: a review of medical, surgical and psychological issues. Clin Endocrinol (Oxf) 64:2–11 11. Bardin C, Catterall J 1981 Testosterone: a major determinant of extragenital sexual dimorphism. Science 211:1285–1294 12. Kudwa AE, Michopoulos V, Gatewood JD, Rissman EF 2006 Roles of estrogen receptors ␣ and ␤ in differentiation of mouse sexual behavior. Neuroscience 138:921–928 13. Haffner SM 2000 Sex hormones, obesity, fat distribution, type 2 diabetes and insulin resistance: epidemiological and clinical correlation. Int J Obes Relat Metab Disord 24(Suppl 2):S56 –S58 14. Nilsson C, Niklasson M, Eriksson E, Bjo¨rntorp P, Holma¨ng A 1998 Imprinting of female offspring with testosterone results in insulin resistance and changes in body fat distribution at adult age in rats. J Clin Invest 101:74 –78 15. Eisner JR, Dumesic DA, Kemnitz JW, Colman RJ, Abbott DH 2003 Increased adiposity in female rhesus monkeys exposed to androgen excess during early gestation. Obesity Res 11:279 –286 16. Eisner JR, Dumesic DA, Kemnitz JW, Abbott DH 2000 Timing of prenatal androgen excess determines differential impairment in insulin secretion and action in adult female rhesus monkeys. J Clin Endocrinol Metab 85:1206 –1210 17. Abbott DH, Eisner JR, Goodfriend TL, Medley RD, Peterson EJ, Colman RJ, Ricki J, Kemnitz JW, Dumesic DA, Leptin and total free fatty acids are elevated in the circulation of prenatally androgenized female rhesus monkeys. Program of the 84th Annual Meeting of The Endocrine Society, San Francisco, CA, 2002 (Abstract P2-329) 18. Abbott DH, Bruns CM, Barnett DK, Zhou R, Colman RJ, Kemnitz JW, Padmanabhan V, Goodfriend TL, Dumesic DA, Metabolic and reproductive consequences of prenatal testosterone exposure. Program of the 85th Annual Meeting of The Endocrine Society, Philadelphia, PA, 2003 (Abstract S34-341) 19. Recabarren SE, Padmanabhan V, Codner E, Lobos A, Durań C, Vidal M, Foster DL, Sir-Petermann T 2005 Postnatal developmental consequences of altered insulin sensitivity in female sheep treated prenatally with testosterone. Am J Physiol Endocrinol Metab 289:E801–E806


5375

20. Gill JW, Hosking BJ 1995 Acute prenatal androgen treatment increases birth weights and growth rates in lambs. J Anim Sci 73:2600 –2608 21. DeHaan KC, Berger LL, Bechtel PJ, Kesler DJ, McKeith FK, Thomas DL 1990 Effect of prenatal testosterone treatment on nitrogen utilization and endocrine status of ewe lambs. J Anim Sci 68:4100 – 4108 22. Hansen LR, Drackley JK, Berger LL, Grum DE 1995 Prenatal androgenization of lambs: alterations of growth, carcass characteristics, and metabolites in blood. J Anim Sci 73:1694 –1700 23. King A, Olivier N, MohanKumar P, Lee J, Padmanabhan V, Fink G 2007 Hypertension caused by prenatal testosterone excess in female sheep. Am J Physiol Endocrinol Metab 292:E1837–E1841 24. Marcondes FK, Bianchi FJ, Tanno AP 2002 Determination of the estrous cycle phases of rats: some helpful considerations. Braz J Biol 62:609 – 614 25. Holma¨ng A, Svedberg J, Jennische E, Bjo¨rntorp P 1990 Effects of testosterone on muscle insulin sensitivity and morphology in female rats. Am J Physiol Endocrinol Metab 259:E555–E560 26. Muruganandan S, Srinivasan K, Gupta S, Gupta PK, Lal J 2005 Effect of mangiferin on hyperglycemia and atherogenicity in streptozotocin diabetic rats. J Ethnopharmacol 97:497–501 27. Smith U, Sjo¨stro¨m L, Bjo¨rntorp P 1972 Comparison of two methods for determining human adipose cell size. J Lipid Res 13:822– 824 28. Bjo¨rnheden T, Jakubowicz B, Levin M, Odeń B, Edeń S, Sjo¨stro¨m L, Lo¨nn M 2004 Computerized determination of adipocyte size. Obes Res 12:95–105 29. Conover WJ 1971 Practical nonparametric statistics. New York: John Wiley, Sons 30. Cooke PS, Naaz A 2004 Role of estrogens in adipocyte development and function. Exp Biol Med 229:1127–1135 31. Heine PA, Taylor JA, Iwamoto GA, Lubahn DB, Cooke PS 2000 Increased adipose tissue in male and female estrogen receptor-␣ knockout mice. Proc Natl Acad Sci USA 97:12729 –12734 32. Sato T, Matsumoto T, Yamada T, Watanabe T, Kawano H, Kato S 2003 Late onset of obesity in male androgen receptor-deficient (AR KO) mice. Biochem Biophys Res Comm 300:167–171 33. Lin HY, Xu Q, Yeh S, Wang RS, Sparks JD, Chang C 2005 Insulin and leptin resistance with hyperleptinemia in mice lacking androgen receptor. Diabetes 54:1717–1725 34. Jones ME, Thorburn AW, Britt KL, Hewitt KN, Wreford NG, Proietto J, Oz OK, Leury BJ, Robertson KM, Yao S, Simpson ER 2000 Aromatase-deficient (ArKO) mice have a phenotype of increased adiposity. Proc Natl Acad Sci USA 97:12735–12740 35. Evans DJ, Barth JH, Burke CW 1988 Body fat topography in women with androgen excess. Int J Obes 12:157–162 36. Elbers JMH, Asscheman H, Seidell JC, Megens JAJ, Gooren LJG 1997 Longterm testosterone administration increases visceral fat in female to male transsexuals. J Clin Endocrinol Metab 82:2044 –2047 37. Meek LR, Romeo RD, Novak CM, Sisk CL 1997 Actions of testosterone in prepubertal and postpubertal male hamsters: dissociation of effects on reproductive behavior and brain androgen receptor immunoreactivity. Horm Behav 31:75– 88 38. Dumesic DA, Abbott DH, Eisner JR, Herrmann RR, Reed JE, Welch TJ, Jensen MD 1998 Pituitary desensitization to gonadotropin releasing hormone increases abdominal adiposity in hyperandrogenic anovulatory women. Fertil Steril 70:94 –101 39. Fisher CR, Graves KH, Parlow AF, Simpson ER 1998 Characterization of mice deficient in aromatase (ArKO) because of targeted disruption of the cyp19 gene. Proc Natl Acad Sci USA 95:6965– 6970 40. Hewitt KN, Boon WC, Murata Y, Jones MEE, Simpson ER 2003 The aromatase knockout mouse presents with a sexually dimorphic disruption to cholesterol homeostasis. Endocrinology 144:3895–3903 41. Einarsson K, Gustafsson JÅ, Stenberg Å 1973 Neonatal imprinting of liver microsomal hydroxylation and reduction of steroids. J Biol Chem 248:4987– 4997 42. Lamartiniere CA, Dieringer CS, Kita E, Lucier GW 1979 Altered sexual differentiation of hepatic uridine diphosphate glucuronyltransferase by neonatal hormone treatment in rats. Biochem J 180:313–318 43. DeFronzo RA 1988 Lilly lecture 1987. The triumvirate: ␤-cell, muscle, liver. A collusion responsible for NIDDM. Diabetes 37:667– 687 44. Bryzgalova G, Gao H, Ahren B, Zierath JR, Galuska D, Steiler TL, DahlmanWright K, Nilsson S, Gustafsson JA, Efendic S, Khan A 2006 Evidence that oestrogen receptor ␣ plays an important role in the regulation of glucose homeostasis in mice: Insulin sensitivity in the liver. Diabetologia 49:588 –597 45. Holma¨ng A, Larsson BM, Brzezinska Z, Bjo¨rntorp P 1992 Effects of short-term testosterone exposure on insulin sensitivity of muscles in female rats. Am J Physiol Endocrinol Metab 262:E851–E855 46. Abbott DH, Dumesic DA, Eisner JR, Colman RJ, Kemnitz JW 1998 Insights into the development of polycystic ovary syndrome (PCOS) from studies of prenatally androgenized female rhesus monkeys. Trends Endocrinol Metab 9:62– 67 47. Weyer C, Foley JE, Bogardus C, Tataranni PA, Pratley RE 2000 Enlarged subcutaneous abdominal adipocyte size, but not obesity itself, predicts type II diabetes independent of insulin resistance. Diabetologia 43:1498 –1506 48. Lelliott C, Vidal-Puig AJ 2004 Lipotoxicity, an imbalance between lipogenesis

5376


de novo and fatty acid oxidation. Int J Obes Relat Metab Disord 28(Suppl 4):S22–S28 49. Jernås M, Palming J, Sjo¨holm K, Jennische E, Svensson PA, Gabrielsson BG, Levin M, Sjo¨gren A, Rudemo M, Lystig TC, Carlsson, B, Carlsson LM, Lo¨nn M 2006 Separation of human adipocytes by size: hypertrophic fat cells display distinct gene expression. FASEB J 20:E832–E839 50. Sopasakis VR, Sandqvist M, Gustafson B, Hammarstedt A, Schmelz M, Yang X, Jansson PA, Smith U 2004 High local concentrations and effects on differentiation implicate interleukin-6 as a paracrine regulator. Obes Res 12:454 – 460


51. Lundgren M, Bureń J, Ruge T, Myrna¨s T, Eriksson JW 2004 Glucocorticoids down-regulate glucose uptake capacity and insulin-signaling proteins in omental but not subcutaneous human adipocytes. J Clin Endocrinol Metab 89:2989 –2997 52. Giorgino F, Laviola L, Eriksson JW 2005 Regional differences of insulin action in adipose tissue: insights from in vivo and in vitro studies. Acta Physiol Scand 183:13–30 53. Lundgren M, Svensson M, Lindmark S, Renstro¨m F, Ruge T, Eriksson JW 2007 Fat cell enlargement is an independent marker of insulin resistance and ‘hyperleptinaemia.’ Diabetologia 50:625– 633

Endocrinology is published monthly by The Endocrine Society (http://www.endo-society.org), the foremost professional society serving the endocrine community.

The Organization for the Study of Sex Differences (www.ossdweb.org) is convening its second annual scientific meeting from June 4-6, 2008, in New Orleans, LA. The program will consist of opening and closing keynote speakers, seven independent symposia, and two poster sessions. Symposia topics include sex differences in vascular and renal disease, sleep, drug abuse, cardiovascular disease, stress, immunity and infection, and x-linked disorders. The deadline for poster abstract submissions is March 15, 2008. Visit http://www.ossdweb.org/meeting_2008.html to register or contact Viviana Simon, Ph.D., (viviana@ ossdweb.org) for more information.