Interplay of Cortisol, Testosterone, and Abdominal Fat Mass in Normal-weight Women With Polycystic Ovary Syndrome

Daniel A. Dumesic; Adina F. Turcu; Haiping Liu; Tristan R. Grogan; David H. Abbott; Gwyneth Lu; Devyani Dharanipragada; Gregorio D. Chazenbalk

Disclosures

J Endo Soc. 2023;7(8)

Abstract and Introduction

Abstract

Context: Ovarian and adrenal steroidogenesis underlie endocrine-metabolic dysfunction in polycystic ovary syndrome (PCOS). Adipocytes express aldo-keto reductase 1C3 and type 1 11β-hydroxysteroid dehydrogenase, which modulate peripheral androgen and cortisol production.

Objectives: To compare serum adrenal steroids, including 11-oxygenated androgens (11-oxyandrogens), cortisol, and cortisone between normal-weight women with PCOS and body mass index- and age-matched ovulatory women with normal-androgenic profiles (controls), and assess whether adrenal steroids associate with abdominal adipose deposition.

Design: Prospective, cross-sectional, cohort study.

Setting: Academic medical center.

Patients: Twenty normal-weight women with PCOS and 20 body mass index-/age-matched controls.

Intervention(s): Blood sampling, IV glucose tolerance testing, and total-body dual-energy x-ray absorptiometry.

Main Outcome Measure(s): Clinical characteristics, hormonal concentrations, and body fat distribution.

Results: Women with PCOS had higher serum total/free testosterone (T) and androstenedione (A4) levels and a greater android/gynoid fat mass than controls (androgens P < .001; android/gynoid fat mass ratio, P = .026). Serum total/free T and A4 levels correlated positively with android/gynoid fat mass ratio in all women combined (P < .025, all values). Serum 11ß-hydroxyA4, 11-ketoA4, 11ß-hydroxyT, 11-ketoT, cortisol, and cortisone levels were comparable between female types and unrelated to body fat distribution. Serum 11-oxyandrogens correlated negatively with % total body fat, but lost significance adjusting for cortisol. Serum cortisol levels, however, correlated inversely with android fat mass (P = .021), with a trend toward reduced serum cortisol to cortisone ratio in women with PCOS vs controls (P = .075), suggesting diminished 11β-hydroxysteroid dehydrogenase activity.

Conclusion: Reduced cortisol may protect against preferential abdominal fat mass in normal-weight PCOS women with normal serum 11-oxyandrogens.

Introduction

Ovarian and adrenal steroidogenesis are interwoven into the endocrine-metabolic pathophysiology of polycystic ovary syndrome (PCOS).^[1–5] Within the adrenal, cytochrome P450 11ß-hydroxylase catalyzes the hydroxylation of androstenedione (A4) and testosterone (T) to 11ß-hydroxyandrostendione (11OHA4) and 11ß-hydroxytestosterone (11OHT), respectively.^[4–6] 11β-hydroxysteroid dehydrogenase type 2 then converts 11OHA4 and 11OHT to their respective ketosteroids, 11-ketoandrostendione (11KA4) and 11-ketotestosterone (11KT), with 11KT also originating from reduction of 11KA4 via aldo-keto reductase 1C3 (AKR1C3; also known as 17β-hydroxysteroid dehydrogenase type 5). Although 11OHA4 is quantitatively the dominant 11-oxygenated androgen (11-oxyandrogen), its bioactivity is negligible, but it serves as substrate for 11KT, a potent androgen with bioactivity comparable to that of T.^[4–6]

Elevated serum 11OHA4 levels have been reported in one-half of National Institutes of Health (NIH)-defined women with PCOS who are overweight.^[1] More recently, serum 11-oxyandrogen levels have been shown to be elevated in 40% of PCOS women by Rotterdam criteria, and to positively correlate with body mass index (BMI), fasting circulating insulin levels, and insulin resistance in some,^[7,8] but not all,^[9] studies.

We have previously shown that normal-weight hyperandrogenic women with PCOS exhibit low-normal insulin sensitivity and metabolic alterations that accompany preferential abdominal fat accumulation.^[10–14] It is unclear, however, whether adrenal steroidogenesis interacts with an elevated androgens pathway in normal-weight women with PCOS to further affect endocrine-metabolic pathophysiology related to altered body fat distribution.^[15] It is also unclear whether adipocyte expression of AKR1C3 and types 1 or 2 11βHSD in these women with PCOS modulate peripheral androgen and cortisol activation/inactivation. Therefore, the present study compares serum adrenal steroids, including 11-oxyandrogens, cortisol, and cortisone, between normal-weight women with PCOS and BMI- and age-matched ovulatory women with normal-androgenic profiles (controls), and assesses whether adrenal steroids associate with preferential abdominal adipose accumulation.

1 2 3 4

Next Section

Abstract and Introduction
Materials and Methods
Results
Discussion
References

Table 1. Patient characteristics and serum hormone and metabolic levels in normal-weight control vs women with PCOS ^a,b
Patient characteristics	Controls (N = 18)	PCOS (N = 18)	P value
Age (y)	27.7 ± 4.8	24.8 ± 4.4	.069
BMI (kg/m²)	21.7 ± 1.6	22.1 ± 1.9	.495
Waist (cm)	75.2 ± 5.1	76.1 ± 4.9	.572
Hip (cm)	89.3 ± 6.0	87.9 ± 5.5	.483
Total body mass (kg)	60.7 ± 7.8	60.9 ± 7.2	.945
Total body fat (kg)	19.3 ± 3.2	19.9 ± 3.2	.620
Percent body fat (%)	31.8 ± 2.9	32.8 ± 4.1	.429
Android fat (kg)	1.1 ± 0.3	1.2 ± 0.4	.126
Percent android fat (%)	5.5 ± 0.7	6.2 ± 1.3	.062
Gynoid fat (kg)	4.1 ± 0.8	4.0 ± 0.6	.663
Percent gynoid fat (%)	21.4 ± 1.6	20.3 ± 1.0	.026
Android/gynoid fat ratio	0.26 ± 0.04	0.31 ± 0.07	.026
Hormone/metabolic levels	Controls (N = 20)	PCOS (N = 20)	P value
Log LH (mIU/mL)	0.89 ± 0.2	1.1 ± 0.2	.004
FSH (mIU/mL)	6.0 ± 2.3	5.3 ± 1.4	.257
E1 (pg/mL)	63.0 ± 32.4	67.4 ± 28.1	.655
E2 (pg/mL)	100.4 ± 106.7	72.0 ± 56.1	.301
Total T (ng/dL)^c	27.4 ± 6.6	53.2 ± 20.4	<.001
Free T (pg/mL)^c	3.4 ± 1.4	8.0 ± 3.2	<.001
A4 (ng/dL)^c	106.9 ± 34.9	208.4 ± 73.3	<.001
11KT (ng/dL)^c	30.3 ± 13.0	35.2 ± 18.5	.333
11OHT (ng/dL)^c	11.9 ± 7.9	12.3 ± 5.5	.866
11KA4 (ng/dL)^c	17.4 ± 7.4	17.5 ± 7.5	.960
11OHA4 (ng/dL)^c	107.9 ± 61.8	111.3 ± 59.1	.858
17OHP4 (ng/dL)^c	35.8 ± 15.3	62.7 ± 34.2	.004
DHEAS (μg/dL)	181.4 ± 97.0	221.8 ± 69.6	.139
Cortisol (ng/dL)^c	11199.7 ± 5189.7	9740.4 ± 3781.9	.314
Cortisone (ng/dL)^c	2122.5 ± 619.9	2278.4 ± 652.0	.440
Fasting glucose (mg/dL)^d	85.2 ± 6.0	85.8 ± 6.5	.796
Fasting insulin (μU/mL)^d	4.6 ± 1.9	5.8 ± 2.1	.075
Si (×10⁻⁴/min/μU/mL)^d	6.0 ± 5.1	4.2 ± 2.0	.143
Log AIRg (μU/mL)^d	2.4 ± 0.2	2.5 ± 0.2	.150
SHBG (nmol/L)	68.2 ± 33.2	49.2 ± 26.8	.054
Log adipose-IR^d	1.2 ± 0.2	1.4 ± 0.2	.017
Log triglyceride (mg/dL)	1.7 ± 0.1	1.8 ± 0.2	.035
HDL-C (mg/dL)	64.8 ± 11.9	62.2 ± 12.5	.513
Non-HDL-C (mg/dL)	89.2 ± 25.6	93.6 ± 26.6	.592
LDL-C (mg/dL)	78.6 ± 23.7	78.2 ± 24.9	.960
Total cholesterol (mg/dL)	154.8 ± 29.4	155.9 ± 29.0	.913

Mean ± SD. Boldface values represent significant differences between female groups.
Conversion to SI Units: T (× 0.0347 nmol/L), free T (× 3.47 pmol/L), A4 (× 0.0349 nmol/L), KT (× 0.0331 nmol/L), OHT (× 0.0329 nmol/L), KA4 (× 0.0322 nmol/L), OHA4 (× 0.0331 nmol/L), 17OHP4 (× 0.0303 nmol/L), DHEAS (× 0.0271 μmol/L), cortisol (× 0.0276 nmol/L), cortisone (× 0.0277 nmol/L), E1 (× 3.699 pmol/L), E2 (× 3.67 pmol/L), LH (× 1.0 IU/L), FSH (× 1.0 IU/L), glucose (× 0.0555 mmol/L), insulin (× 7.175 pmol/L, total cholesterol (× 0.0259 mmol/L), HDL-cholesterol (× 0.0259 mmol/L), LDL-cholesterol (× 0.0259 mmol/L), non-HDL-cholesterol (× 0.0259 mmol/L), triglycerides (× 0.0113 mmol/L).
Abbreviations: 11KA4, 11-ketoA4; 11KT, 11-ketoT; 11OHA4, 11ß-hydroxyA4; 11OHT, 11ß-hydroxyT; A4, androstenedione; AIRg, acute response to glucose; BMI, body mass index; DHEAS, dehydroepiandrosterone sulfate; E1, estrone; E2, estradiol; HDL-C, high-density lipoprotein cholesterol; LDL-C, low-density lipoprotein cholesterol; PCOS, polycystic ovary syndrome; Si, insulin sensitivity index; T, testosterone.
^aModified from references previously reported for these subjects [10-13, 16].
^bTotal body dual-energy x-ray absorptiometry studies (controls = 18, PCOS = 18).
^cNew measurements as determined by liquid chromatography-tandem mass spectrometry at the University of Michigan, Ann Arbor [21, 22].
^dFrequently sampled IV glucose tolerance testing, and fasting glucose/insulin as well as log adipose insulin resistance studies (controls = 19, PCOS = 20).

Table 2. Steroid ratios as indirect enzymatic markers of AKR1C3, CYP11ß1, HSD11ß2, and HSD11ß1 in normal-weight women with PCOS vs controls
Ratio	Controls (N = 20)	PCOS (N = 20)	P value
AKR1C3
T/A4	0.265 ± 0.067	0.255 ± 0.051	.599
11KT/11KA4	1.835 ± 0.520	2.100 ± 0.736	.198
CYP11ß1
11OHA4/A4	1.040 ± 0.480	0.560 ± 0.284	<.001
11OHT/T	0.460 ± 0.300	0.240 ± 0.147	.006
HSD11ß2
11KA4/11OHA4	0.170 ± 0.057	0.170 ± 0.066	.999
11KT/11OHT	2.810 ± 0.818	2.965 ± 0.835	.559
HSD11ß1
Cortisol/cortisone	5.245 ± 1.938	4.295 ± 1.251	.075

Mean ± SD. Boldface values represent significant differences between female groups.
Abbreviations: 11KA4, 11-ketoandrostenedione; 11KT, 11-ketotestosterone; 11OHA4, 11ß-hydroxyandrostenedione; 11OHT, 11ß-hydroxytestosterone; A4, androstenedione; AKR1C3, aldo-keto reductase family 1 member C3 (also known as 17ß-hydroxysteroid dehydrogenase type 5); CYP11ß1, adrenal-specific cytochrome P450 11ß-hydroxylase; HSD11ß1, 11ß-hydroxysteroid dehydrogenase type 1; HSD11ß2, 11ß-hydroxysteroid dehydrogenase type 2; PCOS, polycystic ovary syndrome; T, testosterone.

Table 3. Correlations of serum steroid levels with clinical, hormonal, and metabolic measurements
Parameter	11-OHT^a R, P	11-KT^a R, P	11-OHA4^a R, P	11-KA4^a R, P	Cortisol^a R, P	Cortisone^a R, P	Total T^a R, P	Free T^a R, P	A4^a R, P	E1, ^b R, P	E2, ^b R, P
BMI	R = +0.36 P = .023	R = +0.49 P = .001	R = +0.43 P = .058	R = +0.34 P = .034	R = −0.01 P = .973	R = +0.15 P = 0.353	R = +0.42 P = .007	R = +0.33 P = 0.051	R = +0.34 P = .029	R = +0.12 P = .456	R = +0.06 P = .702
WHR	R = +0.22 P = .181	R = +0.11 P = .501	R = +0.15 P = .347	R = −0.02 P = .908	R = −0.03 P = .838	R = +0.15 P = 0.344	R = +0.28 P = .075	R = +0.44 P = .005	R = +0.33 P = .035	R = +0.10 P = .539	R = −0.03 P = .836
Total body mass	R = +0.25 P = .148	R = +0.19 P = .276	R = +0.12 P = .481	R = +0.10 P = .575	R = +0.04 P = .836	R = +0.04 P = .797	R = +0.18 P = .281	R = +0.08 P = .634	R = +0.10 P = .572	R = +0.07 P = .688	R = +0.06 P = .732
Total body fat	R = −0.08 P = .660	R = −0.13 P = .447	R = −0.21 P = .219	R = −0.15 P = .382	R = −0.30 P = .079	R = −0.33 P = .052	R = +0.14 P = .409	R = +0.09 P = .604	R = +0.02 P = .912	R = +0.09 P = .591	R = −0.07 P = .699
% total body fat	R = −0.36 P = .031	R = −0.37 P = .027	R = −0.42 P = .011	R = −0.34 P = .046	R = −0.45 P = .006	R = −0.49 P = .002	R = +0.02 P = .887	R = +0.05 P = 0.758	R = −0.07 P = .693	R = +0.07 P = .690	R = −0.18 P = .305
Android fat	R = −0.14 P = .426	R = −0.10 P = .553	R = −0.24 P = .161	R = −0.14 P = .402	R = −0.38 P = .021	R = −0.35 P = .035	R = +0.34 P = .044	R = +0.35 P = .038	R = +0.28 P = .095	R = +0.26 P = .128	R = −0.17 P = .318
Gynoid fat	R = −0.12 P = .497	R = −0.18 P = .290	R = −0.26 P = .133	R = −0.17 P = .329	R = −0.24 P = .151	R = −0.29 P = .087	R = +0.04 P = 0.807	R = −0.03 P = .858	R = −0.10 P = .548	R = 0.00 P = .997	R = +0.07 P = .674
Android/gynoid fat ratio	R = −0.09 P = .589	R = −0.02 P = .908	R = −0.14 P = .401	R = −0.09 P = .601	R = −0.30 P = .075	R = −0.23 P = .174	R = +0.41 P = .013	R = +0.49 P = .003	R = +0.45 P = .006	R = +0.32 P = .058	R = −0.21 P = .209
Log adipose-IR	R = +0.05 P = .748	R = +0.25 P = .120	R = +0.16 P = 0.325	R = +0.24 P = .140	R = +0.02 P = .918	R = +0.22 P = 0.179	R = +0.34 P = .037	R = +0.40 P = .011	R = +0.39 P = .013	R = +0.03 P = .852	R = +0.05 P = .747
Log LH	R = −0.15 P = .361	R = −0.04 P = .818	R = +0.05 P = .765	R = −0.12 P = .470	R = +0.06 P = .719	R = +0.01 P = 0.958	R = +0.55 P < .001	R = +0.62 P < .001	R = +0.61 P < .001	R = +0.40 P = .010	R = +0.15 P = .356
Si	R = −0.02 P = .913	R = −0.05 P = .766	R = −0.04 P = .826	R = −0.02 P = .916	R = +0.16 P = .338	R = +0.01 P = .960	R = −0.12 P = .481	R = −0.27 P = .098	R = −0.21 P = .201	R = +0.04 P = .808	R = +0.03 P = .841
Log AIRg	R = −0.05 P = .781	R = −0.09 P = .591	R = −0.09 P = .583	R = +0.10 P = .562	R = −0.27 P = .091	R = −0.17 P = .308	R = +0.16 P = .328	R = +0.39 P = .015	R = +0.26 P = .105	R = −0.07 P = .683	R = +0.23 P = .161
Log triglyceride	R = −0.02 P = .886	R = +0.02 P = .924	R = +0.06 P = .721	R = +0.10 P = .540	R = −0.06 P = .701	R = −0.02 P = .280	R = +0.23 P = .144	R = +0.35 P = .025	R = +0.31 P = .054	R = +0.09 P = .601	R = +0.14 P = .404
Fasting glucose	R = +0.27 P = .091	R = +0.23 P = .152	R = +0.25 P = .126	R = +0.23 P = .161	R = +0.25 P = .119	R = +0.19 P = 0.251	R = +0.00 P = .990	R = +0.05 P = 0.747	R = +0.07 P = .679	R = +0.02 P = .886	R = +0.08 P = .587
Fasting insulin	R = +0.16 P = .318	R = +0.21 P = .193	R = +0.13 P = .420	R = +0.33 P = .041	R = −0.05 P = 0.766	R = +0.27 P = .095	R = +0.25 P = .130	R = +0.44 P = .005	R = +0.37 P = .020	R = −0.15 P = 0.376	R = +0.08 P = .619
SHBG	R = −0.32 P = .047	R = −0.20 P = .210	R = −0.27 P = .088	R = −0.27 P = .088	R = −0.02 P = .890	R = −0.10 P = .533	R = −0.17 P = .303	R = −0.57 P < .0001	R = −0.35 P = .025	R = −0.05 P = .743	R = 0.00 P = .987
DHEAS	R = +0.47 P = .002	R = +0.36 P = .021	R = +0.47 P = .002	R = +0.33 P = .038	R = +0.33 P = .040	R = +0.31 P = .050	R = +0.14 P = .388	R = +0.29 P = .073	R = +0.34 P = .031	R = +0.04 P = .804	R = −0.22 P = .182

Total body dual-energy x-ray absorptiometry studies (controls = 18; PCOS = 18); frequently sampled IV glucose tolerance testing, and fasting glucose/insulin as well as log adipose insulin resistance (adipose-IR) studies (controls = 19; PCOS = 20). Boldface values represent significant correlations.
Abbreviations: 11OHA4, 11ß-hydroxyandrostendione; 11OHT, 11ß-hydroxytestosterone; 11KA4, 11-ketoandrostendione; 11KT, 11-ketotestosterone; A4, androstenedione; AIRg, acute response to glucose; BMI, body mass index; DHEAS, dehydroepiandrosterone sulfate; E1, estrone; E2, estradiol; Si, insulin sensitivity; T, testosterone; WHR, waist to hip ratio.
^aNew measurements as determined by liquid chromatography-tandem mass spectrometry at the University of Michigan, Ann Arbor [21, 22].
^bE1 and E2 modified from references previously reported for these subjects [10–13, 16].