1. Introduction

Despite considerable research, the relationships between obesity and metabolic disorders have yet to be fully understood. Recent evidence has revealed that fat depots, rather than the volume of fat, are essential in determining systemic insulin sensitivity. Adipose tissue is classified into visceral adipose tissue, including epididymal, mesenteric and perirenal fat, and subcutaneous adipose tissue according to its anatomical location. Increases in visceral adipose tissue are considered to be linked to insulin resistance [1, 2]. Especially, mesenteric fat is postulated to relate more closely to metabolic disorders, as mesenteric fat secretes free fatty acids and other substances directly into the portal vein [3]. Although the mechanisms regulating fat distribution remain obscure, sex hormones are unquestionably one of the determinants.

Since men tend to accumulate much more visceral fat than women, androgens have been postulated to promote insulin resistance. In practice, low serum testosterone levels promote obesity. Numerous studies have demonstrated that androgen deprivation therapy (ADT) increases the risk of obesity, metabolic syndrome, type 2 diabetes and cardiovascular disease in patients with prostate cancer [4-8]. Basaria et al, pointed out that high fat mass, as well as low bone density and anemia, was observed in men with prostate cancer treated with ADT compared with ones treated without it. They concluded that patients receiving ADT are at enhanced risk for insulin resistance and cardiovascular disease. Katznelson et al, reported that percent body fat was greater in acquired hypogonadal men compared with eugonadal controls, which was improved by testosterone replacement therapy [9].

Recently, a high prevalence of hypogonadism in men with obesity, metabolic syndrome and type 2 diabetes has been recognized. Dhindsa et al, reported that total testosterone and free testosterone inversely relate to BMI and fat mass [10] in type 2 diabetic men. Kapoor et al, in a cross-sectional study of 355 type 2 diabetic subjects, found overt and borderline hypogonadism in 42%, with 42 of these men having free testosterone levels <0.255 nM [11]. Although the Framingham Heart Study concluded that sex hormone-binding globulin (SHBG), but not testosterone, is significantly associated with metabolic syndrome [12], both low SHBG and low free testosterone may contribute to low serum total testosterone level in obese and diabetic men [13]. Another issue currently of interest is whether low testosterone is a cause or result [13, 14]. Weight loss induced by diet or surgery has been demonstrated to increase testosterone level and sexual function [15-17]. Probably, low testosterone and metabolic disorders worsen each other. The results of clinical studies of testosterone replacement therapy were reviewed by Grossmann [13]. RTC was performed in 10 trials in obese men with borderline low testosterone levels. Although reduced fat mass was commonly observed, improved insulin sensitivity was detected in only 2 trials. Six RCTs in diabetic patients similarly demonstrated beneficial changes in body composition. However, reduction of HOMA-R was detected in 3 trials, and decreased HbA1c in one. These data suggested the limited efficacy of testosterone replacement. Although only one meta-analysis noted that combined prostate events including prostate cancer, elevated PSA and prostate biopsies were more frequent in testosterone-treated men [18], there is no clear evidence that testosterone replacement increases the incidence of prostate cancer. However, the possibility remains that the study was too small to detect significant results. In contrast, a significantly increased risk of cardiovascular events has been associated with testosterone therapy [19, 20], emphasizing that its potential risks should not be ignored.

Dehydroepiandrosterone (DHEA) and its sulfate ester, dehydroepiandrostrone-sulphate (DHEA-S) are referred as a weak androgen produced in adrenal gland (90%) and testis (10%) in men [21]. DHEA is an intermediate product, which is synthesized from pregnenolone, and converted to testosterone and estrogen. DHEA is one of the most abundantly secreted steroids, although its precise physiological roles remain uncertain. DHEA exerts 0.1-2% of the activity of testosterone on the genital organs [22], and 42% on bone formation in mice [23]. Since no specific nuclear receptor for DHEA or DHEA-S has been identified, these hormones are regarded as precursors of more active androgens, such as testosterone and 5α-dihydrotestosterone (DHT), or estrogens. In addition, DHEA and DHEA-S can be converted to more active forms subcellularly in target tissues, the underlying mechanism of which was referred to as “intracine” by Labrie [24].

Both serum DHEA and testosterone levels decline during the aging process [25, 26]. Hence, low serum DHEA level has been assumed to be involved in the development of age-related diseases and shortening of the life span. Such studies suggest an association between high serum DHEA-S level and longevity. However, numerous studies have reported that serum DHEA(-S) exhibits positive, negative or no relation to adiposity, cardiovascular disease and mortality in men and women [27]. Recent longitudinal and cross sectional studies support the favorable effects of DHEA-S on cardiovascular disease and all-cause mortality in both sexes [28-30].

Like the case of testosterone, inconsistent results of DHEA replacement have been published. DHEA replacement decreased fat mass and elevated bone mineral density (BMD) [31], whereas, opposite results were obtained [32] in elderly men and women with DHEA deficiency. Recently, Corona et al, conducted a meta-analysis study of 25 RTC trials of DHEA supplementation in elderly men. They observed no significant effects on the levels of glucose, insulin, total cholesterol or BMD with DHEA, while a small but significant reduction of fat mass was detected in the supplemented group [33].

Production of testosterone in the testis is regulated by gonadotropin, while that of DHEA in the adrenal gland is by ACTH. Low free testosterone is correlated positively with LH in diabetic men, and therefore, hypogonadotropic hypogonadism is common in these patients [34]. However, the pathogenesis of low DHEA level has been unclear. Both serum testosterone and DHEA levels decline with aging. Although some studies have published data on testosterone and DHEA in elderly persons [35, 36], to our knowledge, no research has focused on individual relationships among testosterone, DHEA and metabolic disorders. Theoretically, low testosterone level might be compensated for by DHEA via an intracrine mechanism in men having low testosterone and normal DHEA level. The opposite can also be supposed. Therefore, we speculate that severe metabolic impairment might be observed in men with low testosterone and low DHEA levels. Further study is necessary to help clarify this issue.

In animal studies, extensive research has elucidated the physiological and pharmacological roles of androgens. However, few papers have compared testosterone and DHEA. The hormonal actions of testosterone and DHEA are mediated via the androgen receptor (AR), and so the difference in biological activity between these hormones may be caused by the efficacy of steroid converting enzymes mediated by an intracrine mechanism. In addition, numerous cell surface receptors for testosterone and DHEA have been identified [37, 38]. Differences in the biological responses to testosterone and DHEA may be derived from these membrane receptors. Anagnostopoulou et al, reported opposite effects of DHEA and testosterone on the apoptosis of prostate and colon cancer [39]. They concluded that the differential effects of these hormones on nerve growth factor receptors in cancer cells accounted for these results. Piñeiro et al, showed that DHT, DHEA-S, stanozolol (non-aromatizable androgen), and androstenedion, but not testosterone, suppressed leptin secretion in cultured adipocytes sampled from female omental adipose tissue [40]. The authors presumed that aromatization of testosterone might result in effects opposite to those of other androgens. Sato et al, considered that testosterone increased the expression level of Glut4 more potently than DHEA in cultured skeletal muscle, which was abrogated by a DHT inhibitor [41]. These results suggested that DHT, a metabolite of testosterone and DHEA, finally acts as an androgen in skeletal muscle. In this article, we outline our research investigating the impact of androgens, testosterone and DHEA, on adiposity and glucose metabolism, and the results of our recent study.