Search tips
Search criteria 


Logo of nihpaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
Fertil Steril. Author manuscript; available in PMC 2010 August 1.
Published in final edited form as:
PMCID: PMC2735259


Ashim Kumar, M.D.,a,c Denis Magoffin, PhD.,a,b,c Iqbal Munir, M.D., PhD.,a,b and Ricardo Azziz, M.D., M.P.H., M.B.A.a,b,c,d



To determine changes in AA production, and transcription of DHEA sulfotransferase (SULT2A1) in the NCI-H295R human adrenocortical cell line in response to insulin and testosterone, an environment mimicking the PCOS state.


In vitro experiment using NCI-H295R adrenocortical cell lines.


Academic medical center.


NCI-H295R human adrenocortical cell line


The transcriptional activity of SULT2A1 and adrenal steroid production was quantified after exposure to various treatments (e.g. forskolin, insulin, testosterone, and combinations thereof).

Main Outcome Measures

Quantification of mRNA for DHEA sulfotransferase (SULT2A1) by real-time reverse transcription-polymerase chain reaction and measurement of steroid production by radioimmunoassay.


Testosterone decreased DHEAS and cortisol, and increased DHEA, secretion by H295R cells; the inhibitory effects of testosterone on DHEAS and cortisol production were augmented by insulin. There was a trend towards an increase in the transcription of SULT2A1 by insulin and testosterone.


Testosterone and insulin appear to be modulators of AA production in this human adrenocortical cell model. These results suggest that testosterone may augment DHEA secretion in the human adrenal, although they do not support the role of this sex steroid or insulin on the elevated DHEAS levels frequently observed in PCOS.

Keywords: insulin, testosterone, sulfotransferase, NCI-H295R, adrenal, steroidogenesis


Polycystic ovary syndrome (PCOS) is a hyperandrogenic disorder, affecting at least ~7% of unselected reproductive-aged women, and exacts a high financial cost to society (15). In addition to ovarian hyperandrogenism, Adrenal androgen (AA) secretion appears to be upregulated in PCOS. Principal AAs are dehydroepiandrosterone (DHEA) and its sulfated form, DHEAS; approximately 25% of patients demonstrate basal levels of DHEAS above the upper normal age-adjusted limit (6).

Regulation of AA production and its role in the pathophysiology of PCOS remain elusive (4, 7). The lack of a feedback loop between AAs and their main secretagogue, adrenocorticotropic hormone (ACTH), implies loose central regulatory control. Thus, extra-adrenal factors have been implicated in the modulation of AA production (8). These could include testosterone, as well as insulin, which are elevated in ~65% of PCOS subjects (9).

Previous studies have provided conflicting data regarding the in vivo effects of insulin and testosterone on AA synthesis (1018). Study populations have ranged from normal males and/or females to women diagnosed with PCOS. The association between insulin or testosterone and AA production was studied in the basal state and in response to alterations in insulin and testosterone levels. The conclusions are discordant across the various stratagem used in the in vivo studies.

Previous in vitro experiments focusing on the effect of insulin and testosterone on human adrenocortical tissue suggested a regulatory role in AA production, although the end effects were variable from one adrenal to another (19). Published reports on studies conducted using human adrenals have generally disregarded the inherent adrenal to adrenal variability (20). The constancy of the NCI-H295R adrenocortical cell line facilitates a close examination of the steroidogenic pathway to determine the effect of insulin, testosterone and the combination thereof on androgen biosynthesis. By decreasing 3β-hydroxysteroid dehydrogenase (3β-HSD) activity, testosterone shifts steroidogenesis towards the delta-5 pathway (21). An increase in CYP17 expression induced by insulin further directs steroidogenesis toward the delta-5 pathway (2223). The end result of the combined effect would be increase in DHEA production. DHEAS levels, however, depend upon the activity of SULT2A1.

Abnormal insulin signaling and excessive ovarian steroidogenesis are the leading etiologic hypotheses in PCOS. The resulting elevated testosterone and insulin levels may be the primary mediators of a wide array of sequelae, including elevated AAs, associated with PCOS. To test this hypothesis we evaluated the effects of insulin and testosterone on AA production and on SULT2A1 transcription using the NCI-H295R adrenocortical cell line.


Experimental Model

The NCI-H295 adrenocortical cell line was established by Gazdar and colleagues, and has been used extensively to study human adrenocortical function (2428). The cells have been documented to express the panel of steroidogenic enzymes of interest in androgen biosynthesis (24, 27). However, it is a transformed cell line and thus does not exactly mimic the adrenocortical cells as evidenced by the NCI-H295R cells’ decreased expression of ACTH receptors necessitating forskolin to simulate ACTH stimulation (28). IRB approval was not required because of the use of the NCI-H295R adrenocortical cell line.

Cell Culture

The NCI-H295R cells were obtained from the American Type Culture Collection (ATCC CRL-2128, Manassas, VA), and were propagated and subcultured per ATCC guidelines. The cells were plated in 24-well plates at a density of 1 × 105 cells per well, and cells were grown in the medium outlined by ATCC for 1 to 2 days until sub-confluent. The medium was aspirated and replaced with medium without Nu-serum (BD Biosciences, San Jose, CA) and ITS+ Premix (BD Biosciences), and the various treatments stated below. The cells were then incubated for 24 hours at which time the medium was aspirated from the wells, and both the cells and medium were frozen at −80° C. Our time course experiments (4, 8, 24, 48, 72 hr) revealed near maximal steroidogenesis and mRNA at 24 hours (data not shown) as noted by other investigators (2930).


The treatments included control (medium only), forskolin (15 µmol/L), insulin (10 ng/mL), testosterone (1 mmol/L), insulin+testosterone, forskolin+insulin, forskolin+testosterone, and forskolin+insulin+testosterone were performed in triplicate. The dose of forskolin was chosen after evaluation of dose-response incubations and is similar to published reports (data not shown) (31). Previous experiments indicated that 10 ng/mL is the physiologic dose of insulin consistent with upper physiologic post-prandial levels (22). A supra-physiologic testosterone concentration was used to mimic the hyperandrogenic adrenal environment. Forskolin was used in lieu of ACTH as the NCI-H295R cells have been reported to have an attenuated response to ACTH, perhaps due to a decrease in ACTH receptors (28).

Radioimmunoassay (RIA)

The aspirated medium was stored at −80° C until thawed for evaluation by RIA for DHEA, DHEAS, and cortisol. Double antibody RIA kits were used for the assays (Diagnostic Systems Laboratories, Webster, TX). The medium was diluted as necessary so that the unknowns fell within the linear portion of the standard curve. The results were normalized to the controls.

Real time Reverse Transcription-Polymerase Chain Reaction (real time RT-PCR)

The mRNA for SULT2A1 was measured by real time RT-PCR. The cells were homogenized using a QiaShredder (Qiagen, Valencia, CA) and total RNA was extracted using the RNeasy Mini Kit (Qiagen). First strand cDNA was generated using the Omniscript reverse transcription kit (Qiagen) with oligo(dT)12–18 primer (Invitrogen, Carlsbad, California). PCR was performed using MyiQ Single-Color Real-Time PCR Detection System coupled to the iCycler (Bio-Rad Laboratories, Hercules, CA) using QuantiTect SYBR PCR reagents (Qiagen). A 50 µL reaction mix consisted of 25 µL 2X QuantiTect SYBR PCR Master Mix, 5 µL first strand cDNA, and 200 nmol/L of the sense and antisense primers (4 µL), and 16 µL of H2O.

The PCR protocol consisted of HotStarTaq DNA polymerase activation at 95°C for 15 min and then 40 cycles of amplification (95°C for 15 sec, 56°C for 30 sec, and 72°C for 15 sec). The oligonucleotide primers 5’TCGTGATAAGGGATGAAGATGTAATAA3’ (sense) and 5’TGCATCAGGCAGAGAATCTCA (antisense) were used to amplify SULT2A1 (32). The sequence was then cloned into TA vector (Topo TA Cloning Kit, Invitrogen). The plasmid was linearized. The CT values/crossing points were plotted against the log of a series of known DNA concentrations. The cDNA was diluted as necessary so that the quantitative results were within the linear range of the standard curve to allow for interpolation of the amount of mRNA in each sample. RT-PCR was performed for 18s RNA for each sample in order to standardize the results.

Statistical Analysis

The Wilcoxon signed-rank test was used to determine significance. All values are expressed as mean ± standard error. A P-value of <0.05 was used to determine statistical significance.


In order to asses the modulatory effect of insulin and testosterone on AA production, the DHEA, DHEAS, and cortisol levels in the growth medium of the NCI-H295R adrenocortical cells were measured by RIA after 24 hours of exposure to the treatment(s) (Figure 1). Real time RT-PCR was performed for a quantitative analysis of SULT2A1 RNA.

Figure 1Figure 1Figure 1
DHEA (Figure 1a), DHEAS (Figure 1b), and cortisol (Figure 1c) levels after 24 hours of incubation with forskolin alone, insulin + forskolin, testosterone + forskolin, and insulin + testosterone + forskolin. The y-axis is in relative units. The Y-bar indicates ...

DHEAS production after 24 hours of incubation with the various modulators is depicted in Figure 1a. The DHEAS levels in the wells treated with insulin+forskolin was similar to the wells treated with forskolin alone. The addition of testosterone resulted in a significant decrease in DHEAS production at 24 hours, compared to forskolin alone (P<0.05). The combination of insulin and testosterone suppressed DHEAS further than testosterone alone (P<0.05).

The levels of DHEA in response to the various treatments are depicted in Figure 1b. As for DHEAS, the DHEA level in the wells treated with insulin+forskolin was similar to the level in the wells with forskolin alone. In contrast to what was observed with DHEAS, the addition of testosterone (testosterone+forskolin) significantly increased DHEA production (P<0.05). The addition of insulin to testosterone did not further augment DHEA production.

Cortisol production behaved similar to that of DHEAS (Figure 1c). The cortisol levels in the wells treated with insulin+forskolin was similar to the wells treated with forskolin alone, the addition of testosterone resulted in a significant decrease in production at 24 hours (P<0.05), and the combination of insulin and testosterone suppressed cortisol even further (P<0.05).

In order to determine whether the effect of insulin and testosterone on DHEAS and DHEA were due to changes in sulfotransferase activity, we determined the effect of these hormones, alone or in combination on SULT2A1 (Figure 2). However, neither insulin nor testosterone alone appeared to affect transcription of SULT2A1, compared to forskolin alone. Although not statistically significant, there was a trend towards an increase in the transcription SULT2A1 with the combination of insulin and testosterone, as compared to for skolin alone, contrary to expected (P=0.11).

Figure 2
SULT2A1 mRNA after 24 hours of incubation with forskolin alone, insulin + forskolin, testosterone + forskolin, and insulin + testosterone + forskolin. The y-axis is in relative units. The Y-bar indicates standard error of the mean.


Our results indicate that extra-adrenal modulators, specifically insulin and testosterone, may influence adrenocortical steroidogenesis. Testosterone appears to be a stimulator of DHEA production in the NCI-H295R cells. Alternatively, testosterone suppressed DHEAS and cortisol production; an effect augmented by the addition of insulin. These data suggest that extra-adrenal (e.g. ovarian) testosterone may augment DHEA secretion in the human adrenal, as is observed in PCOS. Alternatively, they do not support the role of this sex steroid or insulin on the elevated DHEAS levels frequently observed in PCOS.

There are no reports in the literature evaluating AA production in the NCI-H295R adrenocortical cells in response to testosterone. In one study, expression of the androgen receptor was observed in the NCI-H295R cell line and in human adrenal tissue (30). In NCI-H295R cells dihydrotestosterone resulted in a significant reduction in cellular proliferation; however, these investigators did not evaluate the change in AA production in response to testosterone itself (30). In mouse adrenal cells, Stalvey found that testosterone resulted in decreased transcription of 3β-HSD (21). This effect could potentially decrease cortisol and increase delta-5 steroid (e.g. DHEA) production as observed in our data, although we did not assess transcription of the 3β-HSD gene (i.e. HSD3B). Our in vivo studies, using exogenous testosterone administered for one month to oophorectomized healthy women (12), or ovarian suppression with a long-acting GnRH-a in hyperandrogenic premenopausal women (32), indicate that while ovarian factors, primarily androgens, may affect the rate of DHEA sulfation (and consequently the circulating levels of DHEAS), the adrenocortical response to ACTH remains relatively unchanged. Although, both our in vivo and in vitro data suggest a potential modulatory role for testosterone in the regulation of AA production and DHEAS levels, the mechanisms are variant and the end-effects observed inconsistent.

Kelly and colleagues evaluated the cortisol:androstendione ratio of the NCI-H295R adrenocortical cells in response to insulin and reported no change in this ratio with insulin+forskolin as compared to forskolin alone (23). We found similar results with insulin alone. Alternatively, the addition of insulin to H295R cells decreased cortisol and DHEAS production only in the presence of testosterone (insulin + testosterone + forskolin). These data do not support our hypothesis of a stimulatory effect of insulin on DHEAS, with an apparent negative relationship between insulin and DHEAS production, at least in this model of the adrenal cortex. In vivo data also does not support an active role of insulin on AA production. We were unable to detect a significant association between circulating DHEAS and fasting insulin levels in PCOS women and age and race-matched healthy eumenorrheic controls (6), and Kauffman and colleagues failed to detect an association between DHEAS and insulin in White and Mexican- American women with PCOS (33).

Preliminarily, our data does not suggest that changes in SULT2A1 accounts for the decrease in DHEAS observed with testosterone or insulin + testosterone, as the combination of testosterone and insulin actually appeared to increase the transcription of SULT2A1. While this observation remains to be confirmed, other mechanisms may be responsible for this discrepancy including increase degradation of DHEAS and post-translational modification (e.g. glycosylation) of the SULT2A1 mRNA. Although the adrenocortical cell line does provide a modality for isolating the effects of the potential secretagogues, the use of a transformed cell line limits the interpretability of the data to the in vivo setting. Another limitation of the study includes the use of a supra-physiologic dose of testosterone.

The data from this study suggests that insulin and testosterone may play a role, albeit limited, in the regulation of AA production. While testosterone did result in an increase in DHEA production, contrary to our initial hypothesis testosterone alone or insulin and testosterone acted to decrease cortisol and DHEAS production by adrenocortical cells. These data imply that the adrenal hyperandrogenemia or the excess DHEAS found in some PCOS patients is not a result of their hormonal milieu, but may represent an intrinsic (e.g. inherited) abnormality. Finally, we should note that genetic association studies may point towards areas for future exploration.


Supported by National Institutes of Health grants RO1-HD29364, K24-HD01346, and the Helping Hand of Los Angeles, Inc. (to RA)


Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.


1. Azziz R, Carmina E, Dewailly D, Diamanti-Kandarakis E, Escobar-Morreale HF, Futterweit W, et al. Positions statement: criteria for defining polycystic ovary syndrome as a predominantly hyperandrogenic syndrome: an Androgen Excess Society guideline. J Clin Endocrinol Metab. 2006;91:4237–4245. Epub 2006 Aug 29. [PubMed]
2. Diamanti-Kandarakis E, Kouli CR, Bergiele AT, Filandra FA, Tsianateli TC, Spina GG, et al. A survey of the polycystic ovary syndrome in the Greek island of Lesbos: hormonal and metabolic profile. J Clin Endocrinol Metab. 1999;84:4006–4011. [PubMed]
3. Asunción M, Calvo RM, San Millán JL, Sancho J, Avila S, Escobar-Morreale HF. A prospective study of the prevalence of the polycystic ovary syndrome in unselected Caucasian women from Spain. J Clin Endocrinol Metab. 2000;85:2434–2438. [PubMed]
4. Azziz R, Woods KS, Reyna R, Key TJ, Knochenhauer ES, Yildiz BO. The prevalence and features of the polycystic ovary syndrome in an unselected population. J Clin Endocrinol Metab. 2004;89:2745–2749. [PubMed]
5. Azziz R, Marin C, Hoq L, Badamgarav E, Song P. Health care-related economic burden of the polycystic ovary syndrome during the reproductive life span. J Clin Endocrinol Metab. 2005;90:4650–4658. [PubMed]
6. Kumar A, Woods KS, Bartolucci AA, Azziz R. Prevalence of adrenal androgen excess in patients with the polycystic ovary syndrome (PCOS) Clin Endocrinol (Oxf) 2005;62:644–649. [PubMed]
7. Knochenhauer ES, Key TJ, Kahsar-Miller M, Waggoner W, Boots LR, Azziz R. Prevalence of the polycystic ovary syndrome in unselected black and white women of the southeastern United States: a prospective study. J Clin Endocrinol Metab. 1998;83:3078–3082. [PubMed]
8. McKenna TJ, Fearon U, Clarke D, Cunningham SK. A critical review of the origin and control of adrenal androgens. Baillieres Clin Obstet Gynaecol. 1997;11:229–248. [PubMed]
9. DeUgarte CM, Bartolucci AA, Azziz R. Prevalence of insulin resistance in the polycystic ovary syndrome using the homeostasis model assessment. Fertil Steril. 2005;83:1454–1460. [PubMed]
10. Phillips GB. Relationship between serum dehydroepiandrosterone sulfate, androstenedione, and sex hormones in men and women. Eur J Endocrinol. 1996;134:201–206. [PubMed]
11. De Leo V, la Marca A, Lanzetta D, Cariello PL, D’Antone D, Morgante G. Effects of flutamide on pituitary and adrenal responsiveness to corticotrophin releasing hormone (CRF) Clin Endocrinol (Oxf) 1998;49:85–89. [PubMed]
12. Azziz R, Gay FL, Potter SR, Bradley E, Jr, Boots LR. The effects of prolonged hypertestosteronemia on adrenocortical biosynthesis in oopherectomized women. J Clin Endocrinol Metab. 1991;72:1025–1030. [PubMed]
13. Vermesh M, Silva PD, Rosen GF, Vijod AG, Lobo RA. Effect of androgen on adrenal steroidogenesis in normal women. J Clin Endocrinol Metab. 1988;66:128–130. [PubMed]
14. Gonzalez F, Chang L, Horab T, Lobo RA. Evidence for heterogeneous etiologies of adrenal dysfunction in polycystic ovary syndrome. Fertil Steril. 1996;66:354–361. [PubMed]
15. Kauffman RP, Baker VM, DiMarino P, Castracane VD. Hyperinsulinemia and circulating dehydroepiandrosterone sulfate in white and Mexican American women with polycystic ovary syndrome. Fert Stert. 2006;85:1010–1016. [PubMed]
16. Falcone T, Finegood DT, Fantus IG, Morris D. Androgen response to endogenous insulin secretion during the frequently sampled intravenous glucose tolerance test in normal and hyperandrogenic women. J Clin Endocrinol Metab. 1990;71:1653–1657. [PubMed]
17. Nestler JE, Clore JN, Strauss JF, 3rd, Blackard WG. The effects of hyperinsulinemia on serum testosterone, progesterone, dehydroepiandrosterone sulfate, and cortisol levels in normal women and in a woman with hyperandrogenism, insulin resistance, and acanthosis nigricans. J Clin Endocrinol Metab. 1987;64:180–184. [PubMed]
18. Nestler JE, McClanahan MA, Clore JN, Blackard WG. Insulin inhibits adrenal 17,20-lyase activity in man. J Clin Encocrinol Metab. 1992;74:362–367. [PubMed]
19. Hines GA, Smith ER, Azziz R. Influence of insulin and testosterone on adrenocortical steroidogenesis in vitro: preliminary studies. Fertil Steril. 2001;76:730–735. [PubMed]
20. Rainey WE, Carr BR, Wang ZN, Parker CR., Jr Gene profiling of human fetal and adult adrenals. J Endocrinol. 2001;171:209–215. [PubMed]
21. Stalvey JR. Inhibition of 3beta-hydroxyseroid dehydrogenase-isomerase activity in mouse adrenal cells: a direct effect of testosterone. Steroids. 2002;67:721–731. [PubMed]
22. Munir I, Yen HW, Geller DH, Torbati D, Bierden RM, Weitsman SR, et al. Insulin augmentation of 17alpha-hydroxylase activity is mediated by phosphatidyl inositol 3-kinase but not extracellular signal-regulated kinase-1/2 in human ovarian theca cells. Endocrinology. 2004 Jan;145(1):175–183. [PubMed]
23. Kelly SN, McKenna TJ, Young LS. Modulation of steroidogenic enzymes by orphan nuclear transcriptional regulation may control diverse production of cortisol and androgens in the human adrenal. J Endocrinol. 2004;181:355–365. [PubMed]
24. Gazdar AF, Oie HK, Shackleton CH, Chen TR, Triche TJ, Myers CE, et al. Establishment and characterization of a human adrenocortical carcinoma cell line that expresses multiple pathways of steroid biosynthesis. Cancer Res. 1990;50:5488–5496. [PubMed]
25. Staels B, Hum DW, Miller WL. Regulation of steroidogenesis in NCI-H295 cells: a cellular model of the human fetal adrenal. Mol Endocrinol. 1993;7:423–433. [PubMed]
26. Marx C, Bornstein SR, Wolkersdorfer GW. Cellular immune-endocrine interaction in adrenocortical tissues. Eur J Clin Invest. 2000;30S3:1–5. [PubMed]
27. Rainey WE, Bird IM, Mason JI. The NCI-H295 cell line: a pluripotent model for human adrenocortical studies. Mol Cell Endocrinol. 1994;100:45–50. [PubMed]
28. Rainey WE, Saner K, Schimmer BP. Adrenocortical cell lines. Mol Cell Endocrinol. 2004;228:23–38. [PubMed]
29. Zatelli MC, Rossi R, Degli Uberti EC. Androgen influences transforming growth factor-β1 gene expression in human adrenocortical cells. J Clin Endocrinol Metab. 2000;85:847–852. [PubMed]
30. Rossi R, Zatelli MC, Valentini A, Cavazzini P, Fallo F, del Senno L, degli Uberto EC. Evidence for androgen receptor gene expression and growth inhibitory effect of dihydrotestosterone on human adrenocortical cells. J Endocrinol. 1998;159:373–380. [PubMed]
31. Bird IM, Pasquarette MM, Rainey WE, Mason JI. Differential control of 17 alpha-hydroxylase and 3 beta-hydroxysteroid dehydrogenase expression in human adrenocortical H295R cells. J Clin Endocrinol Metab. 1996;81:2171–2178. [PubMed]
32. Azziz R, Rittmaster RS, Fox LM, Bradley EL, Jr, Potter HD, Boots LR. Role of the ovary in the adrenal androgen excess of hyperandrogenic women. Fertil Steril. 1998;69:851–859. [PubMed]
33. Kauffman RP, Baker VM, DiMarino P, Castracane VD. Hyperinsulinemia and circulating dehydroepiandrosterone sulfate in white and mexican american women with polycystic ovary syndrome. Fertil Steril. 2006;85:1010–1016. [PubMed]