Skip to main content
NCBI home page
The Journal of Clinical Endocrinology and Metabolism logoLink to The Journal of Clinical Endocrinology and Metabolism
. 2013 Nov 27;99(4):1384–1392. doi: 10.1210/jc.2013-2398

Evidence That Increased Ovarian Aromatase Activity and Expression Account for Higher Estradiol Levels in African American Compared With Caucasian Women

N D Shaw 1,*, S S Srouji 1,*, C K Welt 1, K H Cox 1, J H Fox 1, J M Adams 1, P M Sluss 1, J E Hall 1,✉
PMCID: PMC3973772  PMID: 24285681

Abstract

Context:

Serum estradiol levels are significantly higher across the menstrual cycle in African American (AAW) compared with Caucasian women (CW) in the presence of similar FSH levels, yet the mechanism underlying this disparity is unknown.

Objective:

The objective of the study was to determine whether higher estradiol levels in AAW are due to increased granulosa cell aromatase mRNA expression and activity.

Design:

The design of the study included daily blood sampling and dominant follicle aspirations at an academic medical center during a natural menstrual cycle.

Subjects:

Healthy, normal cycling AAW (n = 15) and CW (n = 14) aged 19–34 years participated in the study.

Main Outcome Measures:

Hormone levels in peripheral blood and follicular fluid (FF) aspirates and aromatase and FSH receptor mRNA expression in granulosa cells were measured.

Results:

AAW had higher FF estradiol [1713.0 (1144.5–2032.5) vs 994.5 (647.3–1426.5) ng/mL; median (interquartile range); P < .001] and estrone [76.9 (36.6–173.4) vs 28.8 (22.5–42.1) ng/mL; P < .001] levels than CW, independent of follicle size. AAW also had lower FF androstenedione to estrone (7 ± 1.8 vs 15.8 ± 4.1; mean ± SE; P = .04) and T to estradiol (0.01 ± 0.002 vs 0.02 ± 0.005; P = .03) ratios, indicating enhanced ovarian aromatase activity. There was a 5-fold increase in granulosa cell aromatase mRNA expression in AAW compared with CW (P < .001) with no difference in expression of FSH receptor. FSH, inhibin A, inhibin B, and AMH levels were not different in AAW and CW.

Conclusions:

Increased ovarian aromatase mRNA expression, higher FF estradiol levels, and decreased FF androgen to estrogen ratios in AAW compared with CW provide compelling evidence that racial differences in ovarian aromatase activity contribute to higher levels of estradiol in AAW across the menstrual cycle. The absence of differences in FSH, FSH receptor expression, and AMH suggest that population-specific genetic variation in CYP19, the gene encoding aromatase, or in factors affecting its expression should be sought.

The recent finding of significant racial disparities in the age of onset of puberty among American youth, with African American girls demonstrating thelarche and menarche earlier than Caucasian girls (1), has reignited interest in the question of the influence of racial background on reproductive hormone dynamics. In the context of epidemiological studies showing increased rates of breast cancer (2) and leiomyomas (3) and decreased rates of osteoporosis in African American women (AAW) compared with Caucasian women (CW) (4), these observations suggest a pattern of increased lifetime estrogen exposure in AAW.

We have recently demonstrated that estradiol (E2) levels are higher across the menstrual cycle in age- and body mass index (BMI)-matched regularly cycling AAW compared with CW (5), consistent with the results of previous, less detailed investigations (6–9). Serum estrogen is derived from the ovary, the adrenal, and peripheral aromatization. Although an ovarian source of the higher E2 in AAW seems most likely because serum E2 is primarily derived from ovarian E2 synthesis and secretion in premenopausal women with a small contribution made from aromatization in peripheral tissues (10, 11), this has yet to be shown.

In previous studies, we found no racial differences in gonadotropin stimulation of the ovary or in the number or predicted size of the dominant follicle at ovulation, suggesting a racial difference in the E2 synthetic pathway within the dominant follicle. Our observation of decreased ratios of serum androstenedione (AD) to E2 and AD to estrone (E1) in AAW compared with CW further suggested that the racial variation may lie in ovarian aromatase activity. Thus, the present study was designed to determine whether the ovary is responsible for the higher serum E2 levels in AAW compared with CW and specifically to determine whether ovarian aromatase activity is increased in AAW compared with CW.

To investigate this hypothesis, we measured reproductive hormones and granulosa cell aromatase mRNA expression in follicular fluid (FF) aspirates from the dominant follicle obtained during natural menstrual cycles in AAW and CW. The current study provides evidence for increased aromatase activity in AAW and further shows that the increase in aromatase in AAW is not due to changes in FSH or FSH receptor (FSHR) expression.

Materials and Methods

Subjects

Healthy AAW (19–34 y; n = 15) and CW (22–34 y; n = 14) were studied. As in our previous study, subjects were classified as either African American or Caucasian if they, their parents, and both sets of grandparents self-identified as such (5). All subjects had a history of regular menstrual cycles (cycle length 25–35 d) with an ovulatory cycle preceding participation in the study confirmed by a luteal phase progesterone (P) greater than 3 ng/mL (9.5 nmol/L). Subjects had no clinical evidence of androgen excess, had normal thyroid function and prolactin levels, did not exercise excessively [<20 miles of running or equivalent per week (12)], were nonsmokers, and were not taking any medications or over-the-counter supplements known to interact with the reproductive axis. In addition, subjects had a normal hemoglobin, platelet count, prothrombin time, and partial thromboplastin time and a transvaginal ultrasound documenting normal ovarian position.

The study was approved by the Human Research Committee of the Massachusetts General Hospital, and signed informed consent was obtained from each subject before participation.

Experimental protocol

Phenotypic studies

Subjects underwent a dual-energy x-ray absorptiometry (DEXA) scan (Hologic Discovery A and APEX software) to assess body fat distribution and bone mineral density (BMD). Daily blood samples were obtained, beginning on the first day of menses and continuing until the day of follicle aspiration. Serum LH, FSH, E2, E1, and AD were measured daily. Inhibin B and antimullerian hormone (AMH) were measured on cycle day 3 and on the day of aspiration, and inhibin A, P, and T were measured on the day of aspiration. Serial transvaginal ultrasounds were performed to monitor follicle size and to determine ovarian morphology, determined by the Rotterdam criteria (13).

Follicle aspirations

The dominant follicle was aspirated when it reached a diameter of 13 mm or greater. Diameter was calculated as the mean of the follicular length and width at the time of aspiration. Retrospective evidence of an LH surge was determined by a 2-fold increase in LH above the mean of the previous three values followed by a decrease. All aspirations were performed in the in vitro fertilization suite of the Brigham and Women's Hospital using standard oocyte retrieval techniques, as previously described (14). In brief, the largest follicle was identified by transvaginal ultrasound and aspirated into a sterile needle and tubing, using 2 mL of DMEM. FF samples were centrifuged (3000 rpm × 15 min) and the supernatants recovered and stored at −80°C for subsequent hormone analyses. The pellets were separated, mixed with 250 μL Trizol, and frozen at −80°C until mRNA extraction.

RNA extraction, cDNA synthesis, and quantitative real-time PCR

Total RNA was extracted from granulosa cells using the RNeasy minikit (QIAGEN). Reverse transcription was performed with the Affinity Script QPCR cDNA synthesis kit (Agilent Technologies) using oligo(deoxythymidine) primers. Quantitative real-time PCR was performed for the expression of CYP19 and FSHR with glyceraldehyde-3-phosphate dehydrogenase (GAPDH) as an endogenous control using Fast SYBR Green master mix (Applied Biosystems). Primers were designed for the aromatase splice variant found in human ovaries (15), whereas previously published and validated primers were selected for GAPDH expression. Both primer pairs were validated to have 90%–100% efficiency in standard curve reactions (16). Primer sequences were as follows: CYP19 forward, 5′-GCA ACA GGA GCT ATA GAT GAA C-3′, aromatase reverse, 5′-AGG CAC GAT GCT GGT GAT G-3′; and GAPDH forward, 5′-ACC CAC TCC TCC ACC TTT G-3′, GAPDH reverse, 5′-CTC TTG TGC TCT TGC TGG G-3′. To assess FSHR expression, we used Taqman gene expression assays (Life Technologies) with probes for FSHR (Hs00174865_m1) and GAPDH (Hs02758991_g1) and Taqman Fast Advanced master mix (Applied Biosystems). All reactions were run with an Applied Biosystems StepOnePlus real-time PCR system (Life Technologies) with no RT controls. Relative quantification was determined using the 2−ΔΔCT method (17) and melt curves were examined to verify amplification of pure mRNA samples. PCR amplication was successful in 13 samples from AAW and in 11 samples from CW for CYP19 expression and in nine AAW and 10 CW samples for FSHR expression.

Hormone assays

Hormone and peptide measurements were performed using the same assay in serum and FF with the exception of E2. All FF measurements were adjusted for the 2-mL DMEM volume used during aspiration. Serum E2 was measured using the AxSYM immunoassay (Abbott Laboratories), as previously described (18), which has a functional sensitivity of 20 pg/mL (73.4 pmol/L) and an interassay coefficient of variation (CV) of 6.5%–10% within the range of reported samples. Cross-reactivity with E1 was 0.1%. FF E2 was measured using the Architect chemiluminescent immunoassay (Abbott Laboratories), which has a functional sensitivity of 25 pg/mL (91.8 pmol/L) and an interassay CV of 10%–15%. Cross-reactivity with E1 was 0.07%. E1 was measured using an RIA (Diagnostic Systems Laboratories) with a functional sensitivity of 12 pg/mL and an interassay CV of 4.1% within the range of reported samples, as previously described (19).

P was measured using enzyme-amplified chemiluminescence (Immulite 1000; Siemens), which has a sensitivity of 0.2 ng/mL (0.6 nmol/L) and an interassay CV of 10.6%–14%. AD was measured by RIA (Diagnostic Product Corp) (20), which has an assay sensitivity of 0.04 ng/mL (1.4 pmol/L) and a CV of 6%–8%. T was measured using the Coat-A-Count RIA kit (Diagnostic Product Corp), with an interassay CV of less than 10% and a functional sensitivity of 4 ng/dL (138.7 pmol/L) (21). LH and FSH were measured using a two-site monoclonal nonisotopic system (AxSYM; Abbott Laboratories) as previously described (22) with assay sensitivities of 0.3 and 0.7 IU/L for LH and FSH, respectively, and CVs of less than 4%. LH and FSH levels are expressed in international units per liter as equivalents of the Second International Pituitary Reference Preparation 80/552 for LH and the Second International Pituitary Reference Preparation 78/549 for FSH.

Inhibin A was measured by ELISA (Serotec), as previously described (23) using a lyophilized human FF calibrator standardized as equivalents of the World Health Organization recombinant human inhibin A preparation 91/624. The interassay CV was less than 10% and the functional sensitivity was 8 pg/mL. Inhibin B was measured by ELISA (Serotec), as previously described (24), with a CV of 15%–18% and a limit of detection of 15 pg/mL. AMH was measured using a two-site ELISA (Diagnostic Systems Laboratory, Beckman-Coulter), which has a detection limit of 0.02 ng/mL (0.14 pmol/L) and an interassay CV of less than 11% for quality control serum containing 2.65 ng/mL (18.9 pmol/L).

Data analysis

One AAW was excluded from analysis because review of daily serum hormone levels indicated that her follicle was aspirated after an LH surge. Data not normally distributed were log transformed for analysis. AD to E1 and T to E2 ratios were calculated as indices of aromatase activity and E1 to E2 and AD to T ratios as indices of type 1 17β-hydroxysteroid dehydrogenase activity. All FF hormone levels were compared between AAW and CW using analysis of covariance to control for age and follicle size, but age was removed from the final models because it did not influence any of the variables studied. Daily serum hormone (12 d) was compared between AAW and CW using a repeated-measures ANOVA after centering on the day of aspiration. Pearson's correlation was used to determine the relationship between serum and FF hormone levels. Aromatase and FSHR mRNA were compared between AAW and CW using an unpaired t test after controlling for GAPDH expression.

Data are expressed as the mean ± SEM unless otherwise indicated, and a value of P ≤ .05 was considered statistically significant.

Results

Baseline characteristics

AAW and CW were of similar age (Table 1). Although BMI was slightly higher in AAW, the total percentage body fat and body fat distribution, determined by DEXA, were not different between the two groups. In line with previous studies (4), AAW had significantly greater BMD. The proportion of women with polycystic ovarian morphology was not different between the two groups. All subjects had regular menstrual cycles (25–35 d), and there was no racial difference in the lengths of the three menstrual cycles immediately preceding the cycle of follicle aspiration (CW 27.4 ± 1.0 d vs AAW 28.3 ± 0.8 d, P = .6).

Table 1.

Baseline Characteristics and Cycle Day 3 Reproductive Hormone and Peptide Levels in AAW and CW

AAW CW P Value
n 15 14
Age, y 24.8 ± 1.2 26.4 ± 1.0 .3
BMI, kg/m2 25.6 ± 1.1 22.7 ± 0.6 .03
Body fat, % 31.4 ± 2.0 30.1 ± 1.6 .6
Android fat, % 30.7 ± 3.0 29.0 ± 2.0 .7
Gynoid fat, % 38.3 ± 2.0 39.9 ± 1.6 .5
BMD T-score 1.7 ± 0.2 0.2 ± 0.3 <.001
PCOM, %a 69.0 78.6 .7
Cycle day 3 serum levels
    FSH, IU/L 6.6 ± 0.4 6.5 ± 0.3 .6
    E2, pg/mL 52.1 ± 6.7 54.3 ± 3.8 1.0
    Inhibin B, pg/mL 86.1 ± 7.8 95.6 ± 7.2 .5
    AMH, ng/mL 2.3 ± 0.3 2.6 ± 0.4 .6
Open in a new tab

Abbreviation: PCOM, polycystic ovarian morphology. To convert to SI units, multiply by 7.14 for AMH (picomoles per liter), 3.67 for E2 (picomoles per liter), and 1 for inhibin B (nanograms per liter). Bold values denote significant P value.

a

Polycystic ovarian morphology is based on Rotterdam criteria (13).

Serum reproductive hormones and peptides prior to follicle aspiration

FSH, inhibin B, and AMH, measured on cycle day 3, were not different between AAW and CW (Table 1). Consistent with our previous study (5), the patterns of LH and FSH measured in daily blood samples up to and including the day of aspiration were not different between the two groups. In addition, there were no racial differences in mean serum E2 [AAW 84.9 ± 7.2 vs CW 87.7 ± 5.8 pg/mL (311.5 ± 26.4 vs 321.9 ± 21.3 pmol/L)], AD [AAW 2.4 ± 0.2 vs CW 2.2 ± 1.1 ng/mL (83.8 ± 7.0 vs 76.8 ± 38.4 pmol/L)], inhibin A (AAW 14.8 ± 1.3 vs CW 14.0 ± 1.0 pg/mL), or inhibin B (AAW 91.6 ± 8.7 vs CW 100.4 ± 4.8 pg/mL) in the 12 days preceding and including the day of follicle aspiration; however, mean serum E1 levels were significantly higher in AAW than CW throughout this period [74.5 ± 6.5 vs 56.5 ± 4.1 pg/mL (275.5 ± 24.0 vs 208.9 ± 15.2 pmol/L); P = .02]. There was a progressive increase in E2 in the days immediately preceding the day of aspiration with levels peaking at 171.1 ± 13.8 pg/mL (627.9 ± 50.6 pmol/L).

Serum E2 (r = 0.6, P = .001), E1 (r = 0.5, P = .009), inhibin A (r = 0.5, P = .003), and inhibin B (r = 0.4, P = .02) levels on the day of aspiration were directly related to dominant follicle size, whereas the A to E1 ratio was inversely related to follicle size (r = −0.4, P = .04). On the day of aspiration, there were no racial differences in serum androgens, P, inhibins AD and B, or AMH on the day of aspiration (Table 2). Serum levels of E2 and E1 tended to be higher in AAW compared with CW, but differences did not reach significance. These results were unchanged after controlling for differences in BMI, percentage body fat, or the size of the dominant follicle on the day of aspiration.

Table 2.

Effects of Race on Serum (Day of Follicle Aspiration) and FF Hormone and Peptide Levels in AAW and CW

Serum
 P Value** Follicular Fluid
 P Value**
AAW CW AAW CW
E2 176.0 ± 18.5 pg/mL 163.9 ± 21.2 pg/mL .1 1713.0 (1144.5–2032.5) ng/mLa 994.5 (647.3–1426.5) ng/mLa <.001
AD 2.3 ± 0.2 ng/mL 2.2 ± 0.2 ng/mL .5 333.1 (269.3–470.3) ng/mLa 504.7 (151–8 = 647.5) ng/mLa .7
E1 113.2 ± 15.1 pg/mL 88.6 ± 8.5 pg/mL .1 76.9 (36.6–173.4) ng/mLa 28.8 (22.5–42.1) ng/mLa <.001
T 34.0 ± 3.7 ng/dL 28.8 ± 2.1 ng/dL .3 18.6 ± 1.6 ng/mL 19.2 ± 2.7 ng/mL .6
T to E2 ratio 0.2 ± 0.03 0.2 ± 0.03 .4 0.01 ± 0.002 0.02 ± 0.005 .03
AD to E1 ratio 0.02 ± 0.002 0.03 ± 0.003 .3 7.0 ± 1.8 15.8 ± 4.1 .04
E1 to E2 ratio 0.7 ± 0.1 0.6 ± 0.1 .9 0.07 ± 0.01 0.04 ± 0.008 .03
Inhibin A 23.2 ± 3.2 pg/mL 19.8 ± 2.6 pg/mL .2 27.8 ± 5.1 ng/mL 21.7 ± 2.7 ng/mL .1
Inhibin B 61.4 ± 7.7 pg/mL 59.8 ± 5.3 pg/mL .9 74.9 (59.1–148.4) ng/mLa 70.2 (53.0–83.6) ng/mLa .1
P 0.5 ± 0.05 ng/mL 0.4 ± 0.05 ng/mL .7 753.0 (313.5–977.5) ng/mLa 506.0 (273.5–916) ng/mLa .3
AMH 3.0 ± 0.3 ng/mL 3.1 ± 0.5 ng/mL .5 3.2 (2.0–5.1) ng/mLa 3.8 (3.2–5.9) ng/mLa .9
Open in a new tab

Values are reported in picograms per milliliter for E2, E1, inhibin A, and inhibin B; nanograms per milliliter for AD, P, and AMH; and nanograms per deciliter for T. To convert serum values to SI units, multiple by 3.67 for E2 and E1 (picomoles per liter); 35 for AD and T (picomoles per liter); 3.2 for P (nanomoles per liter); 1 for inhibin A and B (nanograms per liter); and 7.14 for AMH (picomoles per liter). All FF values are reported in nanograms per milliliter.

a

Values are reported as median (IQR) due to nonnormality.

**

P values reflect analyses controlled for follicle size.

FF and granulosa cell analyses

Serial transvaginal ultrasounds prior to follicle aspiration demonstrated that each subject grew a single dominant follicle and that the growth rate of the dominant follicle was not different in AAW and CW (1.7 ± 0.2 vs 1.5 ± 0.2 mm/d; P = .5) (Figure 1). Aspiration was performed on cycle day 13.1 [interquartile range (IQR) 11.3–14.8] at a dominant follicle size of 17.1 mm (IQR 15.4–18.7) in CW and on cycle day 11.8 (IQR 9.5–13.0) and a dominant follicle size of 16.7 mm (IQR 15.8–17.2) in AAW, with no difference in cycle day or follicle size between AAW and CW. Follicular fluid E2, E1, P, and inhibin A levels increased linearly with follicle size (E2: r = 0.6, P < .001; E1: r = 0.4, P < .01; P: r = 0.6, P < .01; inhibin A: r = 0.5, P < .01), whereas FF T to E2 ratio (r = −0.4, P = .01) and AD to E1 ratio (r = −0.3, P = .03) were inversely related to follicle size, consistent with the expected increase in aromatase activity in larger dominant follicles (25).

Figure 1.

Figure 1.

Open in a new tab

Follicle size vs cycle day in individual CW and AAW, indicating the absence of a difference in growth rate of the dominant follicle between the two racial groups.

The AAW had higher FF E2 (P = .02) and E1 levels (P < .005) than CW. These racial differences became more apparent after controlling for follicle size (Table 2). The FF concentrations of AD and T were similar in AAW and CW, resulting in lower ratios of AD to E2, T to E2, and AD to E1 in AAW, consistent with enhanced ovarian aromatase activity in the dominant follicle in AAW (Figure 2). Granulosa cell aromatase mRNA expression as determined by quantitative RT-PCR was 5-fold higher in AAW compared with CW (P < .001) (Figure 3A). FSHR mRNA expression in these cells (P = .83) was not different between AAW and CW (Figure 3B).

Figure 2.

Figure 2.

Open in a new tab

Mean (±SEM) FF levels of E2 and E1 were significantly higher in AAW compared with CW, which in the presence of similar T and AD levels, resulted in lower T to E2 and AD to E1 ratios in AAW. **, P < .001; *, P < .05. All hormone levels are reported in nanograms per milliliter.

Figure 3.

Figure 3.

Open in a new tab

Mean (±SEM) relative quantification (RQ) of aromatase gene expression (top panel; AAW, n = 13; CW, n = 11) and FSHR gene expression (bottom panel; AAW, n = 9; CW, n = 10) in aspirated granulosa cells, demonstrating a 5-fold higher expression of aromatase in AAW compared with CW with no difference in FSHR expression in AAW compared with CW. **, P < .001.

There were also no racial differences in FF P, inhibin A, inhibin B, or AMH levels, either as absolute values or adjusted for follicle size (Table 2). The FF E1 to E2 ratio was higher in AAW, suggesting reduced type 1 17β-hydroxysteroid dehydrogenase activity in AAW relative to CW, although the type 1 isoenzyme can also act on AD (26), and there was no difference in the AD to T ratio in AAW and CW.

Correlation between FF and serum hormone levels

Significant correlations were found between FF and serum E2 (r = 0.75, P < .001), E1 (r = 0.4, P = .03), and inhibin A (r = 0.81, P < .001) levels on the day of aspiration, whereas FF and serum P, AD, or AMH levels were not significantly correlated.

Discussion

We have previously shown that serum E2 levels are higher across the menstrual cycle in healthy AAW compared with their Caucasian counterparts (5). We hypothesized that this racial disparity is due to increased ovarian aromatase activity in AAW compared with CW, leading to increased E2 synthesis and secretion. Through direct sampling of granulosa cells from dominant follicles aspirated in the mid- to late follicular phase, we have now demonstrated significantly higher ovarian aromatase mRNA expression in age-matched AAW compared with CW. Consistent with this finding, we have also shown that FF E2 levels are higher and aromatase substrate to product ratios (AD to E1 and T to E2) are lower in AAW compared with CW independent of follicle size, indicating that increased aromatase mRNA expression translates to increased aromatase activity in AAW. In the current study, peripheral androgen levels were similar between AAW and CW, arguing against a difference in adrenal androgen substrate availability for peripheral aromatization. In addition, fat mass and distribution were also similar in AAW and CW, suggesting that differences in peripheral aromatization of adrenal or ovarian androgens is unlikely to play a significant role in the elevated serum levels of E1 and E2 in AAW observed in this and our previous study (5).

Understanding that there is a positive correlation between follicle size and FF E2 levels (14), we initially explored the possibility that higher follicular phase serum E2 levels in AAW might reflect more rapid folliculogenesis, larger or more mature dominant follicles, or multiple folliculogenesis in AAW. Based on serial pelvic ultrasounds, however, we determined that there was no difference in the growth rate of the dominant follicle between AAW and CW and that all subjects grew a single dominant follicle. These findings are consistent with our previous studies, indicating a similar follicular phase length and estimated dominant follicle size at the time of ovulation in AAW and CW (5). We also demonstrated that FF E2 levels remained significantly higher in AAW after controlling for follicle size. The absence of differences in FF P and inhibin A, also controlled for follicle size, further suggests that follicles from AAW were not more mature at the time of aspiration that those from CW. Taken together, these data indicate that follicle size and maturity, growth rate, and the number of dominant follicles do not account for higher ovarian E2 production in AAW.

Estrogen production by the ovary is dependent on androgenic substrates and ovarian aromatase activity, which is controlled by a complex network of autocrine, paracrine, and endocrine activators and inhibitors. AD and T are synthesized in the theca cell and serve as the major substrates for granulosa cell aromatase (27). There were no differences in FF levels of AD and T between AAW and CW in the current study. Furthermore, previous in vitro studies in human granulosa cells have shown that these androgens are present in sufficient concentration in the FF of preovulatory follicles to maximize aromatase activity and therefore that substrate availability is unlikely to be a rate-limiting factor in E2 biosynthesis (28). Our findings of similar FF androgen levels in AAW and CW, decreased FF ratios of AD to E1 and T to E2 in AAW, and increased aromatase mRNA expression in AAW strongly suggest that increased serum E2 levels in AAW are due to the up-regulation of ovarian aromatase in AAW. Of note, a precedent for racial differences in aromatase expression comes from a recent study by Ishikawa et al (29), demonstrating that aromatase mRNA levels are higher in leiomyomas of AAW compared with Japanese and Caucasian women.

Lower aromatase substrate to product ratios in AAW compared with CW may reflect an increased amount of enzyme as suggested by increased aromatase expression. In addition, these findings may reflect increased enzyme activity in AAW due either to differences in hormones and/or peptides that modulate aromatase activity or to variation in the aromatase gene and/or its promoter. FSH is the major activator of granulosa cell aromatase activity (27). As in our previous study (5), FSH levels were not different throughout the follicular phase or on the day of follicle aspiration, and furthermore, expression of FSHR did not differ between AAW and CW. Taken together, these findings provide evidence that alterations in the FSH pathway do not account for the observed racial differences in aromatase activity. AMH reduces FSH-induced stimulation of aromatase mRNA in both rodent (30) and human granulosa cells from small antral follicles (31) but was not different between AAW and CW in the current study either in serum on day 3 or on the day of aspiration or in FF. Likewise, there were no differences in inhibin A or inhibin B, which are also major secretory products of the ovary. Studies investigating the effect of inhibin A on granulosa cell aromatase activity have been inconsistent, variably showing stimulation (32), inhibition (33, 34), or no effect (35, 36) on E2 synthesis. In rodent studies, T (37) and E2 (38) stimulate, whereas prolactin (39) and P (40) inhibit aromatase activity. In the current study, all women were normoprolactinemic, and there were no racial differences in androgens or P, although increased FF estrogen levels may potentiate the already increased aromatase activity in AAW. Although these data provide no evidence that ovarian aromatase activity in AAW is increased by hormones or ovarian peptides, the potential contribution of IGF-I and activin, both of which may stimulate aromatase activity (35, 41), could not be assessed.

A number of studies that have correlated race-specific variants in CYP19, which encodes aromatase, with serum reproductive hormone levels raise the possibility that variants in the aromatase gene account for the observed racial differences in FF and serum E2. Significant racial differences in the distributions of aromatase single-nucleotide polymorphism and haplotype frequencies were found at three of five CYP19 loci tested in women included in the multiethnic Study of Women's Health Across the Nation (43). This study further determined that at single-nucleotide polymorphism CYP19 rs936306, the T allele frequency was highest in AAW, and the TT haplotype was associated with increased aromatase activity in all racial groups with the strongest correlation observed in AAW. Although it is also possible that AAW harbor variants in the FSHR, that increase sensitivity to FSH-induced aromatase activity, this hypothesis is weakened by the conflicting results of studies attempting to correlate FSHR polymorphisms with ovarian responses to recombinant FSH in in vitro fertilization protocols (reviewed in Reference 44). Moreover, we did not note any significant difference in FSHR mRNA expression in AAW compared with CW.

We previously found an increase in E2 across the menstrual cycle in AAW compared with CW, with specific differences apparent in the late follicular phase and across all portions of the luteal phase (5). In the current study, serum E1, but not E2, levels were higher in AAW compared with CW during the mid- to late follicular phase prior to aspiration. Our inability to detect a difference in serum E2 levels may reflect the smaller sample size in the current study. The marked elevation in FF E2 levels in AAW after controlling for follicle size suggests that peripheral samples may lag behind FF as a reflection of intraovarian physiology.

It has been suggested that racial differences in ovarian function may contribute to disparities in live birth rates among AAW compared with CW undergoing assisted reproductive technologies (ARTs) after accounting for known prognostic factors such as age, parity, socioeconomic status, number of embryos transferred, and ART clinic volume (45, 46). In the current study, clinical markers of ovarian function measured in serum on cycle day 3 (FSH, inhibin B, and AMH) and in FF aspirates (inhibin B and AMH) were not different between AAW and CW. Several studies in late reproductive age and postmenopausal women have similarly found that inhibin B and AMH levels do not vary with race (47–50). Two studies that reported lower AMH and/or inhibin B levels in AAW did not exclude smokers (51) or women with irregular menstrual cycles or reproductive disorders (52), which may have introduced bias. Thus, most studies using standard clinical markers of granulosa cell function (53–55), including the current study, do not support the idea that racial differences in ovarian reserve contribute to worse ART outcomes in AAW.

In summary, this study provides compelling evidence that the higher serum estrogen levels observed in AAW reflect increased ovarian estrogen production in AAW. Increased expression of ovarian aromatase mRNA in the face of similar levels of FSH, FSHR, androgenic substrates, and AMH in AAW and CW strongly suggests that the amount of aromatase is greater in AAW, although genetic variation at the CYP19 locus resulting in increased enzyme activity may also contribute. Further studies of both ovarian aromatase mRNA and protein levels correlated with CYP19 genotypes will be necessary to better understand racial differences in ovarian estrogen production.

Acknowledgments

We thank Kara Klingman, Teresa Alati, Lacey Plummer, the Clinical Laboratory Research Core (Pathology Service), and the Clinical Research Center staff for their support in conducting these studies.

The study was registered with ClinicalTrials.gov with the number NCT00334971.

This work was supported by National Institutes of Health Grants R01HD 42708 and M01 RR01066. N.D.S. received support from the National Institutes of Health Grant 1K23 HD073304-01). This work was also supported by the Harvard Catalyst and the Harvard Clinical and Translational Science Center (National Center for Research Resources and the National Center for Advancing Translational Sciences, National Institutes of Health Award 8UL1TR000170-05, and financial contributions from Harvard University and its affiliated academic health care centers).

The content of this study is solely the responsibility of the authors and does not necessarily represent the official views of Harvard Catalyst, Harvard University and its affiliated academic health care centers, the National Center for Research Resources, or the National Institutes of Health.

Disclosure Summary: The authors have nothing to declare.

Footnotes

Abbreviations:
AAW
African American women
AD
androstenedione
AMH
antimullerian hormone
ART
assisted reproductive technology
BMD
bone mineral density
BMI
body mass index
CV
coefficient of variation
CW
Caucasian women
DEXA
dual-energy x-ray absorptiometry
E1
estrone
E2
estradiol
FF
follicular fluid
FSHR
FSH receptor
GAPDH
glyceraldehyde-3-phosphate dehydrogenase
IQR
interquartile range
P
progesterone.

References

  • 1. Herman-Giddens ME, Slora EJ, Wasserman RC, et al. Secondary sexual characteristics and menses in young girls seen in office practice: a study from the Pediatric Research in Office Settings network. Pediatrics. 1997;99(4):505–512 [DOI] [PubMed] [Google Scholar]
  • 2. American Cancer Society. Breast Cancer Facts, Figures, 2009–2010. Atlanta, GA: American Cancer Society; 2010 [Google Scholar]
  • 3. Faerstein E, Szklo M, Rosenshein N. Risk factors for uterine leiomyoma: a practice-based case-control study. I. African-American heritage, reproductive history, body size, and smoking. Am J Epidemiol. 2001;153(1):1–10 [DOI] [PubMed] [Google Scholar]
  • 4. Aloia JF. African Americans, 25-hydroxyvitamin D, and osteoporosis: a paradox. Am J Clin Nutr. 2008;88(2):545S–550S [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5. Marsh EE, Shaw ND, Klingman KM, et al. Estrogen levels are higher across the menstrual cycle in African-American women compared with Caucasian women. J Clin Endocrinol Metab. 2011;96(10):3199–3206 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6. Haiman CA, Pike MC, Bernstein L, et al. Ethnic differences in ovulatory function in nulliparous women. Br J Cancer. 2002;86(3):367–371 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7. Perry HM, III, Horowitz M, Morley JE, et al. Aging and bone metabolism in African American and Caucasian women. J Clin Endocrinol Metab. 1996;81(3):1108–1117 [DOI] [PubMed] [Google Scholar]
  • 8. Pinheiro SP, Holmes MD, Pollak MN, Barbieri RL, Hankinson SE. Racial differences in premenopausal endogenous hormones. Cancer Epidemiol Biomarkers Prev. 2005;14(9):2147–2153 [DOI] [PubMed] [Google Scholar]
  • 9. Woods MN, Barnett JB, Spiegelman D, et al. Hormone levels during dietary changes in premenopausal African-American women. J Natl Cancer Inst. 1996;88(19):1369–1374 [DOI] [PubMed] [Google Scholar]
  • 10. Bulun SE, Simpson ER. Competitive reverse transcription-polymerase chain reaction analysis indicates that levels of aromatase cytochrome P450 transcripts in adipose tissue of buttocks, thighs, and abdomen of women increase with advancing age. J Clin Endocrinol Metab. 1994;78(2):428–432 [DOI] [PubMed] [Google Scholar]
  • 11. Zumoff B. Influence of obesity and malnutrition on the metabolism of some cancer-related hormones. Cancer Res. 1981;41(9 Pt 2):3805–3807 [PubMed] [Google Scholar]
  • 12. Feicht CB, Johnson TS, Martin BJ, Sparkes KE, Wagner WW., Jr Secondary amenorrhoea in athletes. Lancet. 1978;2(8100):1145–1146 [DOI] [PubMed] [Google Scholar]
  • 13. Revised 2003 consensus on diagnostic criteria and long-term health risks related to polycystic ovary syndrome (PCOS). Hum Reprod. 2004;19(1):41–47 [DOI] [PubMed] [Google Scholar]
  • 14. Schneyer AL, Fujiwara T, Fox J, et al. Dynamic changes in the intrafollicular inhibin/activin/follistatin axis during human follicular development: relationship to circulating hormone concentrations. J Clin Endocrinol Metab. 2000;85(9):3319–3330 [DOI] [PubMed] [Google Scholar]
  • 15. Means GD, Kilgore MW, Mahendroo MS, Mendelson CR, Simpson ER. Tissue-specific promoters regulate aromatase cytochrome P450 gene expression in human ovary and fetal tissues. Mol Endocrinol. 1991;5(12):2005–2013 [DOI] [PubMed] [Google Scholar]
  • 16. Catteau-Jonard S, Jamin SP, Leclerc A, Gonzales J, Dewailly D, di CN. Anti-Mullerian hormone, its receptor, FSH receptor, and androgen receptor genes are overexpressed by granulosa cells from stimulated follicles in women with polycystic ovary syndrome. J Clin Endocrinol Metab. 2008;93(11):4456–4461 [DOI] [PubMed] [Google Scholar]
  • 17. Livak KJ, Schmittgen TD. Analysis of relative gene expression data using real-time quantitative PCR and the 2[-δ δ C(T)] method. Methods. 2001;25(4):402–408 [DOI] [PubMed] [Google Scholar]
  • 18. Welt CK, Falorni A, Taylor AE, Martin KA, Hall JE. Selective theca cell dysfunction in autoimmune oophoritis results in multifollicular development, decreased estradiol, and elevated inhibin B levels. J Clin Endocrinol Metab. 2005;90(5):3069–3076 [DOI] [PubMed] [Google Scholar]
  • 19. Welt CK, Jimenez Y, Sluss PM, Smith PC, Hall JE. Control of estradiol secretion in reproductive ageing. Hum Reprod. 2006;21(8):2189–2193 [DOI] [PubMed] [Google Scholar]
  • 20. Taylor AE, Khoury RH, Crowley WF., Jr A comparison of 13 different immunometric assay kits for gonadotropins: implications for clinical investigation. J Clin Endocrinol Metab. 1994;79(1):240–247 [DOI] [PubMed] [Google Scholar]
  • 21. Pagan YL, Srouji SS, Jimenez Y, Emerson A, Gill S, Hall JE. Inverse relationship between luteinizing hormone and body mass index in polycystic ovarian syndrome: investigation of hypothalamic and pituitary contributions. J Clin Endocrinol Metab. 2006;91(4):1309–1316 [DOI] [PubMed] [Google Scholar]
  • 22. Welt CK, Adams JM, Sluss PM, Hall JE. Inhibin A and inhibin B responses to gonadotropin withdrawal depends on stage of follicle development. J Clin Endocrinol Metab. 1999;84(6):2163–2169 [DOI] [PubMed] [Google Scholar]
  • 23. Lambert-Messerlian GM, Hall JE, Sluss PM, et al. Relatively low levels of dimeric inhibin circulate in men and women with polycystic ovarian syndrome using a specific two-site enzyme-linked immunosorbent assay. J Clin Endocrinol Metab. 1994;79(1):45–50 [DOI] [PubMed] [Google Scholar]
  • 24. Seminara SB, Boepple PA, Nachtigall LB, et al. Inhibin B in males with gonadotropin-releasing hormone (GnRH) deficiency: changes in serum concentration after short-term physiologic GnRH replacement—a clinical research center study. J Clin Endocrinol Metab. 1996;81(10):3692–3696 [DOI] [PubMed] [Google Scholar]
  • 25. Erickson GF, Hsueh AJ, Quigley ME, Rebar RW, Yen SS. Functional studies of aromatase activity in human granulosa cells from normal and polycystic ovaries. J Clin Endocrinol Metab. 1979;49(4):514–519 [DOI] [PubMed] [Google Scholar]
  • 26. Poutanen M, Miettinen M, Vihko R. Differential estrogen substrate specificities for transiently expressed human placental 17β-hydroxysteroid dehydrogenase and an endogenous enzyme expressed in cultured COS-m6 cells. Endocrinology. 1993;133(6):2639–2644 [DOI] [PubMed] [Google Scholar]
  • 27. Moon YS, Tsang BK, Simpson C, Armstrong DT. 17β-Estradiol biosynthesis in cultured granulosa and thecal cells of human ovarian follicles: stimulation by follicle-stimulating hormone. J Clin Endocrinol Metab. 1978;47(2):263–267 [DOI] [PubMed] [Google Scholar]
  • 28. Hillier SG, Reichert LE, Jr, van Hall EV. Control of preovulatory follicular estrogen biosynthesis in the human ovary. J Clin Endocrinol Metab. 1981;52(5):847–856 [DOI] [PubMed] [Google Scholar]
  • 29. Ishikawa H, Reierstad S, Demura M, et al. High aromatase expression in uterine leiomyoma tissues of African-American women. J Clin Endocrinol Metab. 2009;94(5):1752–1756 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30. di Clemente N, Goxe B, Remy JJ, et al. Inhibitory effect of AMH upon the expression of aromatase and LH receptors by cultured granulosa cells of rat and porcine immature ovaries. Endocrine. 1994;2:553–558 [Google Scholar]
  • 31. Pellatt L, Rice S, Dilaver N, et al. Anti-Mullerian hormone reduces follicle sensitivity to follicle-stimulating hormone in human granulosa cells. Fertil Steril. 2011;96(5):1246–1251 [DOI] [PubMed] [Google Scholar]
  • 32. Campbell BK, Baird DT. Inhibin A is a follicle stimulating hormone-responsive marker of granulosa cell differentiation, which has both autocrine and paracrine actions in sheep. J Endocrinol. 2001;169(2):333–345 [DOI] [PubMed] [Google Scholar]
  • 33. Jimenez-Krassel F, Winn ME, Burns D, Ireland JL, Ireland JJ. Evidence for a negative intrafollicular role for inhibin in regulation of estradiol production by granulosa cells. Endocrinology. 2003;144(5):1876–1886 [DOI] [PubMed] [Google Scholar]
  • 34. Ying SY, Becker A, Ling N, Ueno N, Guillemin R. Inhibin and β type transforming growth factor (TGF β) have opposite modulating effects on the follicle stimulating hormone (FSH)-induced aromatase activity of cultured rat granulosa cells. Biochem Biophys Res Commun. 1986;136(3):969–975 [DOI] [PubMed] [Google Scholar]
  • 35. Hillier SG, Miro F. Local regulation of primate granulosa cell aromatase activity. J Steroid Biochem Mol Biol. 1993;44(4–6):435–439 [DOI] [PubMed] [Google Scholar]
  • 36. Hutchinson LA, Findlay JK, de Vos FL, Robertson DM. Effects of bovine inhibin, transforming growth factor-β and bovine Activin-A on granulosa cell differentiation. Biochem Biophys Res Commun. 1987;146(3):1405–1412 [DOI] [PubMed] [Google Scholar]
  • 37. Wu YG, Bennett J, Talla D, Stocco C. Testosterone, not 5α-dihydrotestosterone, stimulates LRH-1 leading to FSH-independent expression of Cyp19 and P450scc in granulosa cells. Mol Endocrinol. 2011;25(4):656–668 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38. Adashi EY, Hsueh AJ. Estrogens augment the stimulation of ovarian aromatase activity by follicle-stimulating hormone in cultured rat granulosa cells. J Biol Chem. 1982;257(11):6077–6083 [PubMed] [Google Scholar]
  • 39. Dorrington J, Gore-Langton RE. Prolactin inhibits oestrogen synthesis in the ovary. Nature. 1981;290(5807):600–602 [DOI] [PubMed] [Google Scholar]
  • 40. Schreiber JR, Nakamura K, Erickson GF. Progestins inhibit FSH-stimulated granulosa estrogen production at a post-cAMP site. Mol Cell Endocrinol. 1981;21(2):161–170 [DOI] [PubMed] [Google Scholar]
  • 41. Erickson GF, Garzo VG, Magoffin DA. Insulin-like growth factor-I regulates aromatase activity in human granulosa and granulosa luteal cells. J Clin Endocrinol Metab. 1989;69(4):716–724 [DOI] [PubMed] [Google Scholar]
  • 42. Hillier SG, Miro F. Local regulation of primate granulosa cell aromatase activity. J Steroid Biochem Mol Biol. 1993;44(4–6):435–439 [DOI] [PubMed] [Google Scholar]
  • 43. Sowers MR, Wilson AL, Kardia SR, Chu J, Ferrell R. Aromatase gene (CYP 19) polymorphisms and endogenous androgen concentrations in a multiracial/multiethnic, multisite study of women at midlife. Am J Med. 2006;119(9 suppl 1):S23–S30 [DOI] [PubMed] [Google Scholar]
  • 44. Boudjenah R, Molina-Gomes D, Torre A, et al. Genetic polymorphisms influence the ovarian response to rFSH stimulation in patients undergoing in vitro fertilization programs with ICSI. PLoS One. 2012;7(6):e38700. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 45. Seifer DB, Frazier LM, Grainger DA. Disparity in assisted reproductive technologies outcomes in black women compared with white women. Fertil Steril. 2008;90(5):1701–1710 [DOI] [PubMed] [Google Scholar]
  • 46. Seifer DB, Zackula R, Grainger DA. Trends of racial disparities in assisted reproductive technology outcomes in black women compared with white women: Society for Assisted Reproductive Technology, 1999 and 2000 vs. 2004–2006. Fertil Steril. 2010;93(2):626–635 [DOI] [PubMed] [Google Scholar]
  • 47. Freeman EW, Sammel MD, Gracia CR, et al. Follicular phase hormone levels and menstrual bleeding status in the approach to menopause. Fertil Steril. 2005;83(2):383–392 [DOI] [PubMed] [Google Scholar]
  • 48. Freeman EW, Gracia CR, Sammel MD, Lin H, Lim LC, Strauss JF., III Association of anti-mullerian hormone levels with obesity in late reproductive-age women. Fertil Steril. 2007;87(1):101–106 [DOI] [PubMed] [Google Scholar]
  • 49. Gracia CR, Freeman EW, Sammel MD, Lin H, Nelson DB. The relationship between obesity and race on inhibin B during the menopause transition. Menopause. 2005;12(5):559–566 [DOI] [PubMed] [Google Scholar]
  • 50. Su HI, Sammel MD, Freeman EW, Lin H, DeBlasis T, Gracia CR. Body size affects measures of ovarian reserve in late reproductive age women. Menopause. 2008;15(5):857–861 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 51. Schuh-Huerta SM, Johnson NA, Rosen MP, Sternfeld B, Cedars MI, Reijo Pera RA. Genetic variants and environmental factors associated with hormonal markers of ovarian reserve in Caucasian and African American women. Hum Reprod. 2012;27(2):594–608 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 52. Seifer DB, Golub ET, Lambert-Messerlian G, et al. Variations in serum mullerian inhibiting substance between white, black, and Hispanic women. Fertil Steril. 2009;92(5):1674–1678 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 53. Missmer SA, Hankinson SE, Spiegelman D, Barbieri RL, Marshall LM, Hunter DJ. Incidence of laparoscopically confirmed endometriosis by demographic, anthropometric, and lifestyle factors. Am J Epidemiol. 2004;160(8):784–796 [DOI] [PubMed] [Google Scholar]
  • 54. Pellatt L, Hanna L, Brincat M, et al. Granulosa cell production of anti-Mullerian hormone is increased in polycystic ovaries. J Clin Endocrinol Metab. 2007;92(1):240–245 [DOI] [PubMed] [Google Scholar]
  • 55. Shebl O, Ebner T, Sommergruber M, Sir A, Tews G. Anti-muellerian hormone serum levels in women with endometriosis: a case-control study. Gynecol Endocrinol. 2009;25(11):713–716 [DOI] [PubMed] [Google Scholar]

Articles from The Journal of Clinical Endocrinology and Metabolism are provided here courtesy of The Endocrine Society

RESOURCES