androgens pathway) and its expression is increased in SAT when compared to VAT. Therefore, in the eugonadal male, the VAT adipogenesis is inhibited, given that DHT is poorly inactivated. Conversely, in obese men ARK1C2 expression is increased in VAT when compared to lean subjects and, moreover, visceral adipocytes are characterized by higher density of AR when compared to subcutaneous. Thus, hypogonadism, which occurs in men with severe obesity, facilitates fat deposition in VAT [15, 47]. On the other hand, in obese women as well as in those with PCOS, AKR1C3 converts A into T (activation androgens pathway) through its 17β-hydroxysteroid-dehydrogenase activity [48]. In vitro (adipose cellular supernatant) and in vivo (adipose tissue microdialysis) studies showed that AKR1C3 expression in the adipose tissue of PCOS is increased by insulin resistance [49].
Another key enzyme in androgen metabolism is 5α-reductase, in particular the type 1 isoform: it is expressed in different tissues, including the adipose one of both sexes, and converts T into DHT. Indeed, its inhibition by means of dutasteride (an inhibitor of both isoforms of 5α-reductase) facilitates fat deposition, in particular, in the liver of adult men, while its induction leads to hirsutism in obese women with PCOS [50–53].
A third class of enzymes is represented by UDP-glucuronosyl transferase (mainly UGT2B17 and UGT2B15) which is involved in the glucuronidation of 5α-end metabolites such as ADG and ADTG (inactivation androgens pathway). Since 5α-reductase activity is difficult to quantify in clinical practice, the glucoronate metabolites above quoted have been proposed as its indirect parameter. Indeed, both ADG and ADTG positively correlate with BMI in overweight adult men, whereas this correlation is negative in obese women without hirsutism and in severely obese hypogonadal men (BMI >40 kg/m2), indicating that 5α-reductase activity is very sensitive either to the circulating androgens or to those locally produced in the target tissues [20, 51, 54].
Adipose tissue is regarded as the major source of endogenous oestrogens after the gonads in both sexes, thanks to the aromatase activity, having C19 steroids as substrate [12, 55]. Given that its contribution increases with age and considering the expression of ER in adipose tissue, regional differences in aromatase expression (higher in SAT of thighs and buttocks than in VAT) have been associated with distribution of FM (i.e., gynoid fat distribution) [56, 57].
In conclusion, beside the thought of adipose tissue as a steroid reservoir, it should be considered an organ with intense enzyme activity by androgen synthesis, inactivation and conversion to oestrogens, co-regulating their metabolic clearance and production rates, eventually [58–60].
Androgen Deficiency and Body Composition: Clinical Studies
Several clinical conditions can be taken into account when describing the interaction between androgen deficiency and body composition. As specific clinical examples, obesity, Klinefelter syndrome (KS), prostate cancer and MtoF in men and hypopituitarism in women are reviewed in the present paragraph.
Obesity is the epidemic public health problem of the century: it is estimated that 1.9 billion adults are overweight and 600 million are obese worldwide [61]. In men, obesity and androgen levels reciprocally influence each other, as described in the “hypogonadal-obesity cycle”: obese subjects have higher VAT; VAT is characterized by aromatase activity, converting T to oestradiol as already stated; oestradiol suppress LH release, worsening T deficiency; low T favours differentiation of FM, instead of FFM [20, 62]. Moreover, visceral adiposity is associated with hyperinsulinism and insulin resistance, leading to decreased synthesis of sex hormone binding globulin (SHBG) in the liver and T in Leydig cells [19, 21, 63, 64]. Finally, obesity and metabolic syndrome have been associated with low-grade inflammation in the hypothalamus and impairment of the GnRH via the kisspeptin pathway induced by tumour necrosis factor-α [65, 66]. Current guidelines on obesity management recommend weight loss in overweight and obese men by lifestyle changes, pharmacotherapy and weight-loss procedures, including bariatric surgery [67]. Independently from the selected strategy, an inverse relation between BMI and T was also confirmed in this setting. In the European Male Ageing Study, a longitudinal survey on 2,736 community-dwelling men aged 40–79 years at baseline followed for a mean of 4.4 ± 0.3 years; a weight decrease of at least 10% (mean decrease of 13.7 kg) was associated with a statistically significant increase in total T (2.9 nmol/L) and SHBG (13.6 nmol/L); in particular, total and free T were associated with a cubic relationship with the percentage of weight loss, whereas this relationship was linear for SHBG [68]. In 58 men with BMI of 36.1 ± 3.8 kg/m2, a 9-week of very low calorie diet followed by 12-month maintenance period was associated with a weight loss of 14.3 ± 9.1 kg and an increase in total T, free T and SHBG. In 22 men with BMI 44.9 ± 1.0 kg/m2, Roux-en-Y gastric bypass surgery was associated with a BMI reduction of 16.6 ± 1.2 kg/m2 and an increase in total T, free T, and SHBG after 2 years of follow-up; a decreased oestradiol was also reported [19]. A weight loss of at least 5–10% is needed for a significant increase of T, as recommended for other strong outcomes in obesity management [67]. As expected, bariatric surgery is more significantly effective in comparison with low-calorie diet, both on weight loss and on androgens [69]. Another option for weight loss is represented by T replacement therapy in hypogonadal subjects. Current guidelines on androgen deficiency syndromes recommend an active assessment of hypogonadism in all subjects with increased body fat and BMI by history and physical examination; if clinically suspected, total T and SHBG should be requested for the laboratory confirmation of the diagnosis [67, 70]. In these patients, T showed to be effective on weight loss and waist circumference reduction. In 362 men with obesity grades I, II and III under T undecanoate for up to 6 years, the mean change in BMI from baseline was –3.99 ± 0.14, –6.58 ± 0.16, and –8.79 ± 0.23 kg/m2, respectively; the mean change in waist circumference from baseline was –9.24 ± 0.3, –12.29 ± 0.33, and –12.44 ± 0.36 cm, respectively [71]. In a meta-analysis of observational studies, T replacement therapy has been associated with a weight loss of –3.50