C LIN I CA L R ES E AR CH A RT IC LE

Progesterone Receptors and Proliferation of the

Endometrium in Obese Women With Polycystic Ovary
Syndrome—A Lifestyle Intervention Study

Mariana Paulson,1,2,3 Lena Sahlin,2 and Angelica Lindén Hirschberg1,3

Downloaded from https://academic.oup.com/jcem/article-abstract/102/4/1244/3109069 by guest on 27 March 2019

1
Division of Obstetrics and Gynecology, and 2Pediatric Endocrinology Unit, Department of Women’s and
Children’s Health, Karolinska Institutet, Stockholm 17177, Sweden; and 3Department of Obstetrics and
Gynecology, Karolinska University Hospital, Stockholm 17176, Sweden

Context: Polycystic ovary syndrome (PCOS) is associated with increased risk of endometrial cancer.
This is usually explained by chronic anovulation and deficient progesterone activity. However, the
role of progesterone receptors (PRs) in endometrial proliferation is unclear.

Objective: To evaluate PRs in relation to endometrial proliferation in women with PCOS.

Design: Cross-sectional study and lifestyle intervention.

Setting: Clinical and laboratory research unit at a university hospital.

Participants: Twenty obese women with PCOS and 10 age- and body mass index–matched regularly
menstruating controls.

Intervention: Dietary management and physical exercise.

Main Outcome Measures: Endometrial messenger RNA (mRNA) levels and immunostaining of the
nuclear PRs A (PRA) and B (PRB), nongenomic progesterone receptor membrane component
1 (PGRMC1) and 2 (PGRMC2), and proliferation marker Ki67.

Results: Before lifestyle intervention, mRNA expression of PRAB was lower while PRB was higher in
proliferative endometrium of obese women with PCOS compared with controls (P , 0.05). After
lifestyle intervention and weight loss, mRNA expression of PRAB was still low but PRB mRNA de-
creased and was not different to controls in proliferative endometrium (P , 0.01). The subgroup of
PCOS women who remained anovulatory displayed higher protein levels of PRB, PGRMC1,
PGRMC2 and of the proliferative marker Ki67 on cycle days 21 to 23 than controls (P , 0.05). In
contrast, the subgroup of PCOS women with confirmed ovulation showed immunostaining, in-
cluding Ki67, in secretory endometrium that was not different to controls, except for higher PRA
(P , 0.05).

Conclusions: Lifestyle intervention improves, but not fully restores PR expression and decreases prolif-
eration in secretory endometrium of obese PCOS women. (J Clin Endocrinol Metab 102: 1244–1253, 2017)

olycystic ovary syndrome (PCOS) is a heterogeneous anovulation, hyperandrogenism, and polycystic ovaries. In the
P endocrine metabolic disorder (1, 2). It affects about
10% of fertile age women and is characterized by chronic
long term, PCOS is associated with an increased incidence
of endometrial hyperplasia and cancer (3). The endometrial

ISSN Print 0021-972X ISSN Online 1945-7197 Abbreviations: E%, percentage of energy; FSH, follicle-stimulating hormone; LC-MS/MS,
Printed in USA liquid chromatography/tandem mass spectrometry; LH, luteinizing hormone; mRNA,
Copyright © 2017 Endocrine Society messenger RNA; OB-C, obese women without polycystic ovary syndrome (controls); OB-
Received 1 September 2016. Accepted 16 December 2016. PCOS, obese women with polycystic ovarian syndrome; PCOS, polycystic ovarian syn-
First Published Online 20 December 2016 drome; PCR, polymerase chain reaction; PGRMC1, progesterone receptor membrane
component 1; PGRMC2, progesterone receptor membrane component 2; PR, pro-
gesterone receptor; PRA, nuclear progesterone receptor A; PRB, nuclear progesterone
receptor B; SHBG, sex hormone-binding globulin.

1244 https://academic.oup.com/jcem J Clin Endocrinol Metab, April 2017, 102(4):1244–1253 doi: 10.1210/jc.2016-3155
doi: 10.1210/jc.2016-3155 https://academic.oup.com/jcem 1245

abnormalities have been attributed to chronic anovulation, estrogen and androgen receptors in women with PCOS
a lack of progesterone, and, therefore, unopposed estrogen (15). The effect of lifestyle changes on the endometrial
stimulation of the tissue (3). expression of PRs and proliferation is not known. The
Progesterone protects the endometrium from estrogen- aim of the present study was to investigate genomic and
stimulated proliferation via genomic pathways by the nongenomic PRs in relation to proliferation, as de-
nuclear progesterone receptors (PRs) A and B (PRA and termined by the proliferation marker Ki67, in the en-
PRB) and via more rapid, nongenomic pathways such as dometrium of women with PCOS.
progesterone receptor membrane component (PGRMC)
1 and PGRMC2 (4, 5). The expression of both PRA and Materials and Methods
PRB in epithelial cells of the endometrium increases

Downloaded from https://academic.oup.com/jcem/article-abstract/102/4/1244/3109069 by guest on 27 March 2019

during the proliferative phase of the normal menstrual Subjects
cycle and then gradually decreases during the secretory Women of reproductive age (18 to 40 years) fulfilling the 3
Rotterdam criteria for PCOS (i.e., oligo- or anovulation,
phase (4). However, in stromal cells, the expression of
hyperandrogenism, and polycystic ovaries) (16) and age- and
PRA remains unchanged and PRB decreases during the body mass index (BMI)–matched controls were recruited from
secretory phase of the menstrual cycle (4). The 2 nuclear 2008 to 2012 at the Women’s Health Research Unit at the
PRs seem to mediate different actions of progesterone in Karolinska University Hospital (Stockholm, Sweden) (14, 15).
the uterus. From studies of receptor knockout mice, it is All the women were nonsmoking and were free of any hormonal
treatment for $3 months before beginning the study. The ex-
known that PRB promotes the hyperplasia of epithelial
clusion criteria were pregnancy or lactation during the pre-
cells and PRA suppresses this effect (6). ceding 12 months, a current disease or endocrine disorder other
PGRMC1 and PGRMC2 are localized predominantly than PCOS, eating disorders, and regular medication use, in-
in the cytoplasm and are found in the stroma and luminal cluding insulin-sensitizing drugs.
and glandular epithelium of the primate endometrium Two groups were defined according to the presence of PCOS:
(7). Expression of endometrial PGRMC1 and PGRMC2 women with PCOS and a BMI .27 kg/m2 (OB-PCOS; n = 20)
and healthy overweight or obese women with a BMI .27 kg/m2
was found to be differently regulated by estrogen and
without PCOS (OB-C; n = 10). In the OB-PCOS group, all the
progesterone during the menstrual cycle (5). Although women were initially anovulatory, presenting with either
PGRMC1 expression is greater during the proliferative amenorrhea (no bleeding for the previous 3 months) or oligo-
phase and then gradually decreases during the secretory menorrhea (5 to 9 periods during the previous year with intervals
phase, the expression of PGRMC2 is greater in the se- of .6 weeks). The women in the OB-C group had regular
menstrual cycles.
cretory phase than in the proliferative phase (5).
In women with PCOS, the endometrial expression of
Ethical approval
PRs, the distribution of PRs, and their functional role
The local ethics committee approved the present study
remain unclear. Data have suggested that the endome- (approval no. 2008/865-32), and all participants provided
trium in PCOS women displays a reduced responsiveness written informed consent.
to progesterone, which could explain the increased risk of
hyperplasia and cancer (8). In support, progesterone- Intervention
regulated genes, including S100P, were found to be The women in the OB-PCOS group, closely supervised by a
substantially lower in the endometrium of PCOS women dietician, underwent 3 months of an individually adapted
lifestyle intervention program aimed at weight reduction (14,
despite treatment with clomiphene citrate or pro-
15). The participants were recommended a diet high in protein
gesterone compared with controls (9). Few studies have and low in carbohydrates [40% of energy (E%) carbohydrates,
been reported on the expression of classic nuclear PRs in 30E% fat, and 30E% proteins] (11, 17). A strict schedule of 3
the endometrium of women with PCOS. In the study by main meals and 2 or 3 snacks daily was introduced. Food intake
Quezada et al. (10), immunostaining of PRs in secretory was self-reported and corrected, if needed.
All women in the OB-PCOS group received a membership
endometrium was shown to be greater in PCOS women
to a local gym network. Furthermore, recommendations con-
than in healthy controls. Furthermore, data are lacking cerning the frequency, duration, and type of training exercise to
regarding the expression of PGRMC1 and PGRMC2 increase physical activity were determined from each in-
receptors and their actions in the endometrium of women dividual’s condition, interest, and experience. An average
with PCOS. physical activity included aerobic activity for 45 minutes 2 or 3
A change in lifestyle is a first-line therapy for obese times per week, with participation recorded by the gym staff.
women with PCOS (11). We have previously demon-
Experimental design
strated that weight loss and exercise improves re-
Before and after 3 months of the intervention, the women in
productive function (12, 13), upregulates the endometrial the OB-PCOS group were examined on cycle days 6 to 8 and 21 to
gene and protein levels of molecules involved in insu- 23 of the menstrual cycle, which was determined by spontaneous
lin signaling (14), and alters endometrial expression of menstruation or after induced bleeding (medroxyprogesterone
1246 Paulson et al Progesterone Receptors in Women With PCOS J Clin Endocrinol Metab, April 2017, 102(4):1244–1253

acetate, 10 mg/d for 7 days). The women in the OB-C group computer with an image analysis system (Leica Imaging System
were examined once, during a single menstrual cycle on cycle Ltd., Cambridge, UK) were used. Cytosolic immunostaining
days 6 to 8 and 21 to 23. A fasting blood sample was collected was evaluated using the manual scoring, which was performed
at the same time from a peripheral vein and stored at 270°C by 2 of us who were blinded to the group and order of sampling.
for analysis of hormones and binding proteins. A gyneco- For manual scoring, a 4-point grading scale was used: 2, neg-
logical examination, including transvaginal ultrasonography, ative; +, faint; ++, moderate; and +++, strong immunostaining.
was performed by the same investigator (A.L.H.) using
Sonoline SL-250 equipment (Siemens Healthcare Diagnostics, Serum analyses
Erlangen, Germany). After local anesthesia, biopsy specimens Serum concentrations of testosterone were analyzed using
were collected using an endometrial suction curette (Pipet liquid chromatography/tandem mass spectrometry (LC-MS/MS)
Curet; CooperSurgical, Trumbull, CT). (19). Free testosterone was calculated from the serum con-
During the lifestyle intervention, the menstrual pattern of

Downloaded from https://academic.oup.com/jcem/article-abstract/102/4/1244/3109069 by guest on 27 March 2019

centrations of total testosterone and sex hormone-binding
the women in the OB-PCOS group was recorded, and ovulation globulin (SHBG), using a fixed albumin concentration of 40
was confirmed by an elevated serum level of progesterone g/L and successive computerized approximation involving a
(.17 nmol/L) in the luteal phase of the menstrual cycle. The system of equation derived from the law of mass action (20).
menstrual pattern was considered to be improved when a shift Serum concentrations of luteinizing hormone (LH), follicle-
had occurred from amenorrhea to oligomenorrhea or regular stimulating hormone (FSH), insulin, and SHBG were deter-
menstruation or from oligomenorrhea to regular menstruation. mined using chemiluminescent enzyme immunometric assays
and estradiol using a sensitive radioimmunoassay, as previously
Specimens described (13). The detection limits and coefficients of variation
The collected endometrial samples were divided into 2 within and between the assays were 0.1 nmol/L and 9.2% for
pieces. One piece was preserved in RNAlater (Thermo Fisher testosterone, 0.7 U/L and 6% and 9% for LH, 0.1 U/L and 8%
Scientific Life Sciences, Waltham, MA) for further RNA iso- and 8% for FSH, 2.7 mIU/L and 3.0% and 4.0% for insulin,
lation, and the second was fixed in 4% phosphate-buffered 0.2 nmol/L and 6.5% and 8.7% for SHBG, and 5 pmol/L and
formaldehyde for 9 hours and then transferred to 70% ethanol 3% and 6% for estradiol, respectively.
for future embedding in paraffin.
Statistical analysis
Real-time polymerase chain reaction analysis All values are expressed as mean 6 standard deviation or
median and quartile range (25th to 75th percentile), depending
Real-time polymerase chain reaction (PCR) was performed
on the distribution. Differences within groups were analyzed
in an iCycler iQ Real Time PCR System (Bio-Rad Laboratories,
using the paired t test or Wilcoxon matched pairs test,
Inc., Hercules, CA). A reaction volume of 25 mL was used,
depending on the distribution. Differences between .2 groups
containing complementary DNA corresponding to 50 ng RNA
were evaluated using the Kruskal-Wallis test followed by
for PRAB, PGRMC1, PGRMC2, and S100P or 100 ng for PRB,
Dunn’s test, and differences between 2 groups were evaluated
12.5 mL of iQ SYBR Green Supermix (Bio-Rad) and 0.3 mM of
using the Mann-Whitney U test. Correlations between the
each oligonucleotide primer. The oligonucleotide primers for
variables were evaluated using Spearman’s rank order cor-
PRAB, PRB, PGRMC1, PGRMC2, S100P, and RPL13A
relation test. The significance level was set at P , 0.05.
(housekeeping gene) are presented in Supplemental Table 1. All
measurements were performed in duplicate. An RNA sample
without reverse transcription was used as a negative control. Results
The melting curve was used to control the purity of the PCR
products (data not shown). Before lifestyle intervention
Clinical characteristics on cycle days 6 to 8
Immunohistochemistry
Of the 20 obese women with PCOS, 18 completed the
The protein expression of PRA, PRB, PGRMC1, PGRMC2,
and Ki67 was evaluated. Immunostaining was performed on lifestyle intervention. The age and endocrinological pa-
5-mm-thick sections using a standard immunohistochemical rameters for the obese women with PCOS and their BMI-
technique (avidin-biotin-peroxidase), as previously described matched controls are listed in Table 1. As previously
(18). All the slides were blocked with normal horse serum di- reported, the OB-PCOS group had significantly higher
luted in phosphate-buffered saline. After blocking, the primary serum levels of total and free testosterone and lower levels
and secondary antibodies were applied. The primary and sec-
of FSH and SHBG on cycle days 6 to 8 than the OB-C
ondary antibodies and their dilutions and incubation times are
listed in Supplemental Table 2. Negative controls were obtained group (14, 15). The endometrial thickness in the OB-PCOS
by replacing the primary antibody with the respective con- women was not significantly different from that in the
centration of nonimmune immunoglobulin G from the same controls (Table 1).
species.
Luminal epithelium, glandular epithelium, and stroma were Gene and protein expression on cycle days 6 to 8
evaluated separately. Depending on the protein localization,
Messenger RNA (mRNA) expression of PRAB and
either image analysis or manual scoring was used. Nuclear
immunostaining was evaluated by image analysis. For image PRAB/B (i.e., representing PRA) was lower and PRB
analysis, a Leica microscope (Leica, Wetzlar, Germany) and expression was higher in the proliferative endometrium
Sony video camera (Sony, Park Ridge, NJ) connected to a of OB-PCOS women compared with their BMI-matched
doi: 10.1210/jc.2016-3155 https://academic.oup.com/jcem 1247

Table 1. Clinical Characteristics, Endocrinological Variables, and Relative mRNA Levels of Progesterone
Receptors and Protein Expression of Ki67 on Cycle Days 6 to 8 in Obese Women with PCOS and BMI-Matched
Controls
Variable OB-PCOSa (n = 18) OB-C (n = 10)
Patient characteristics
Age, yrb 28.2 6 4.9 33.5 6 3.4
BMI, kg/m2b 37.4 6 5.4 34.0 6 5.1
Endometrial thickness, mmb 5.6 6 2.2 5.2 6 1.3
Endocrinological variables
LH, IU/Lb 4.6 6 1.6 4.9 6 2.0
FSH, IU/Lb 5.1 6 1.1c 7.3 6 2.9

Downloaded from https://academic.oup.com/jcem/article-abstract/102/4/1244/3109069 by guest on 27 March 2019

Total testosterone, nmol/Lb 1.3 6 0.6c 0.7 6 0.2
SHBG, nmol/Lb 27.4 6 11.3d 41.9 6 18.6
Free testosterone, pmol/Lb 31.4 6 18.3e 12.4 6 5.1
Estradiol, pmol/Lb 148 6 66 171 6 70
Insulin, mIU/Lb 75.6 6 81.2 31.5 6 15.6
Relative mRNA expression
PRAB 1.1 (0.5–1.6)d 1.9 (1.2–2.1)
PRB 2.2 (1.7–3.3)d 1.5 (1.2–1.9)
PRAB/B 0.4 (0.3–0.6)d 1.3 (0.7–1.7)
PGRMC1 1.5 (1.0–1.7) 1.5 (0.8–2.1)
PGRMC2 0.8 (0.5–1.0) 1.0 (0.7–1.2)
Ki67 protein expression
Stroma 3.7 (2.2–17.9) 3.2 (2.0–5.4)
Glandular epithelium 94.9 (48.6–97.9) 86.7 (23.9–93.9)
Luminal epithelium 62.3 (44.9–79.2) 81.3 (66.0–90.1)
Data presented as mean 6 standard deviation or median (interquartile range [25th to 75th quartile]).
Abbreviation: mRNA, messenger RNA.
a
Before the intervention.
b
Data previously reported.
c
Statistically significant change, P , 0.01.
d
Statistically significant change, P , 0.05.
e
Statistically significant change, P , 0.001.

controls (Table 1). No statistically significant differences expected, the serum progesterone levels were higher in the
were found in either PGRMC1 mRNA or PGRMC2 OB-PCOS women with a confirmed luteal phase than in
mRNA levels between groups. Furthermore, we found no the OB-PCOS women with no luteal phase (Supplemental
substantial differences in immunostaining between the Table 3) (14, 15).
groups for any of the proteins examined (data not shown),
including the proliferative marker Ki67 (Table 1). Gene expression on cycle days 6 to 8 and 21 to 23
No substantial changes were found in PR expression in
the whole group of PCOS women on cycle days 6 to 8
After lifestyle intervention
after the lifestyle intervention (Supplemental Table 4).
Clinical characteristics Moreover, mRNA expression of PRAB and PRAB/B
After the lifestyle intervention, body weight was sig- in proliferative endometrium (cycle days 6 to 8) was still
nificantly reduced in the OB-PCOS women (average lower in the women with PCOS after the lifestyle intervention
of 5%), 6 OB-PCOS women had confirmed ovulation (P , 0.01 and P , 0.001, respectively). However, expression
and luteal phase at cycle days 21 to 23, and 12 of the 18 of PRB mRNA had changed and was no longer signif-
OB-PCOS women were still anovulatory. As previously icantly different from that of the controls on cycle days
reported (14, 15), OB-PCOS women with no luteal phase 6 to 8.
had higher serum levels of LH and FSH and lower levels Figure 1 shows the gene expression of PRs on cycle
of SHBG, estradiol, and progesterone on cycle days 21 to days 6 to 8 and 21 to 23 in the endometrium of the
23 compared with the OB-PCOS and OB-C groups with a OB-PCOS women with no luteal phase (n = 12), OB-
confirmed luteal phase (Supplemental Table 3) (14, 15). PCOS women with a confirmed luteal phase (n = 6), and
However, the serum levels of total testosterone were still the control group. In the control group (OB-C), endo-
higher in both subgroups of OB-PCOS women compared metrial mRNA expression of PRAB, PRB, and PGRMC1
with the OB-C group (Supplemental Table 3) (14, 15). As was lower on cycle days 21 to 23 compared with days 6 to 8
1248 Paulson et al Progesterone Receptors in Women With PCOS J Clin Endocrinol Metab, April 2017, 102(4):1244–1253

Figure 1. Relative mRNA expression of the nuclear receptors PRAB, PRB, PRAB/B ratio and the nongenomic receptors PGRMC1 and PGRMC2 after Downloaded from https://academic.oup.com/jcem/article-abstract/102/4/1244/3109069 by guest on 27 March 2019
lifestyle intervention on cycle days 6 to 8 and 21 to 23 in the endometrium of obese women with PCOS and no luteal phase (OB-PCOS, no luteal phase,
n = 12), obese women with PCOS and a confirmed luteal phase (OB-PCOS, confirmed luteal phase, n = 6), and age- and BMI-matched controls (OB-C,
confirmed luteal phase, n = 10). Box and whisker plots represent the median and interquartile range (25th to 75th percentile). *P , 0.05, **P , 0.01,
***P , 0.001 indicating differences within groups; boxes with different letters indicate differences between groups (P , 0.05).

(Fig. 1). In contrast, no substantial changes in these mRNA In the OB-PCOS group with no luteal phase, endo-
levels were evident between cycle days in the subgroups of metrial expression of PRAB, PRB, and PGRMC1 was
OB-PCOS women without or with a confirmed luteal higher and PRAB/B was lower on cycle days 21 to 23
phase. The only exception was PGRMC2 mRNA ex- compared with the controls (Fig. 1). In the OB-PCOS
pression, which was increased in the group of OB-PCOS group with a confirmed luteal phase, mRNA expression
women with a confirmed luteal phase and was higher than of PRAB and PRB was in between that of the other 2
that in the OB-PCOS women with no luteal phase. groups and not significantly different from that in either
However, PGRMC2 mRNA expression was not different the OB-PCOS women with no luteal phase or that in the
from that seen in the controls (Fig. 1). controls (Fig. 1). However, mRNA expression of PRAB/B
doi: 10.1210/jc.2016-3155 https://academic.oup.com/jcem 1249

was still lower and PGRMC1 expression still higher than

the expression in the controls (Fig. 1).
Furthermore, the expression of S100P mRNA on cycle
days 21 to 23 was lower in the group of OB-PCOS women
with no luteal phase than that in the controls (median,
0.006; quartile range, 0.003 to 0.02; vs median, 0.9;
quartile range, 0.03 to 4.0; P , 0.01). In contrast, the group
of OB-PCOS women with a luteal phase exhibited ex-
pression of S100P mRNA that was not different from that
of the controls (median, 0.01; quartile range, 0.04 to 0.7).

Downloaded from https://academic.oup.com/jcem/article-abstract/102/4/1244/3109069 by guest on 27 March 2019

Protein expression on cycle days 21 to 23
Strong nuclear immunostaining of PRA, PRB, and Ki67
and moderate to strong cytosolic immunostaining of
PGRMC1 and PGRMC2 was observed in all endometrial
compartments (Fig. 2). Because of a lack of tissue, 10 of 12
biopsy specimens in the group of OB-PCOS women with
no luteal phase were used for protein expression analysis.
In anovulatory OB-PCOS women, immunostaining of
PRB, PGRMC1, and PGRMC2 in the stroma and of
PGRMC1 in the luminal epithelium on cycle days 21 to 23
was significantly higher than in the controls (Figs. 2 and 3).
In the OB-PCOS group with a confirmed luteal phase, only
immunostaining of PRA was higher than that in the
controls, for all 3 compartments (Fig. 3).
In the control group, the proliferation marker Ki67
decreased in the epithelia but was increased in stroma in
the secretory phase compared with the proliferative phase
Figure 2. Representative images of PRA, PRB, PGRMC1, PGRMC2,
of the menstrual cycle (data not shown). In anovulatory and Ki67 immunoreactivity in the endometrium of OB-PCOS women
OB-PCOS women, immunostaining of Ki67 on cycle with no luteal phase (OB-PCOS, no luteal phase, n = 10) and obese
days 21 to 23 was significantly greater in all endometrial controls with a confirmed luteal phase (OB-C, confirmed luteal
phase, n = 10) on cycle days 21 to 23.
compartments than in the BMI-matched controls (Figs. 2
and 3). Immunostaining of Ki67 in the OB-PCOS group
with a confirmed luteal phase was not significantly dif- in the proliferative endometrium of OB-PCOS women
ferent from that of any of the other groups (Fig. 3). compared with their healthy BMI-matched controls. After
the lifestyle intervention, PRB mRNA expression decreased
Correlations on cycle days 21 to 23 and was not different from that of the controls on cycle days
In all subjects, on cycle days 21 to 23, the serum levels 6 to 8. In the subgroup of women with restored ovulation,
of progesterone correlated negatively with immuno- only PRA immunostaining differed and was higher on cycle
staining of the proliferation marker Ki67 in the stroma, days 21 to 23, but the protein levels of PRB, PGRMC1,
glandular epithelium, and luminal epithelium [Fig. 4(A)]. and PGRMC2 and the proliferative marker Ki67 were not
Moreover, in the same combined groups, Ki67 correlated different from those of the controls. OB-PCOS women who
positively with PRA and PRB in the stroma [Fig. 4(B)] and remained anovulatory had greater stromal immunostaining
PGRMC1 (r = 0.40) and PGRMC2 (r = 0.54). of PRB, PGRMC1, and PGRMC2 and the proliferation
marker Ki67 compared with the BMI-matched controls.
Discussion The incidence of endometrial hyperplasia and cancer
is increased in women with PCOS (21) and can be further
To the best of our knowledge, the present study is the first aggravated by obesity (22). Endometrial abnormalities in
to evaluate expression of endometrial PRs and proliferation, women with PCOS have mainly been attributed to
determined by the proliferation marker Ki67, in obese chronic anovulation and a lack of antiproliferative ac-
women with PCOS before and after a lifestyle intervention. tion of progesterone (3). During a normal menstrual
Before the intervention, the PRAB mRNA level and PRAB/B cycle, progesterone initiates endometrial differentiation
ratio were lower and the PRB mRNA expression was higher and prevents the endometrium from estrogen-driven
1250 Paulson et al Progesterone Receptors in Women With PCOS J Clin Endocrinol Metab, April 2017, 102(4):1244–1253

Downloaded from https://academic.oup.com/jcem/article-abstract/102/4/1244/3109069 by guest on 27 March 2019

Figure 3. Immunohistochemistry scores of PRA, PRB, and Ki67, evaluated by image analyses, and PGRMC1 and PGRMC2, evaluated by manual
scoring, in the endometrial stroma, glandular epithelium (GE), and luminal epithelium (LE) on cycle days 21 to 23 of obese PCOS women with no
luteal phase (OB-PCOS, no luteal phase, n = 10), PCOS women with a confirmed luteal phase (OB-PCOS, confirmed luteal phase, n = 6), and
obese controls with a confirmed luteal phase (OB-C, confirmed luteal phase, n = 10). Box and whisker plots represent the median and interquartile
range (25th to 75th percentile). Different letters indicate differences between groups (P , 0.05).

proliferation (3). Because progesterone action could be impaired receptivity and contribute to the reduced fer-
mediated via both nuclear and membrane-bound PRs, tility, endometrial hyperplasia, and cancer in obese
altered expression of these receptors might indicate women with PCOS. However, little is known about the
doi: 10.1210/jc.2016-3155 https://academic.oup.com/jcem 1251

coexpression of PRs and the proliferation marker Ki67

in the endometrium of these women.
In our study, the subgroup of OB-PCOS women who
remained anovulatory after the lifestyle intervention
exhibited increased mRNA expression of PRAB and PRB
on cycle days 21 to 23 compared with the controls.
Furthermore, the same group of OB-PCOS women had
higher stromal immunostaining of PRB on the same cycle
days compared with the controls. These findings might be
explained by the absence of a regulatory effect of pro-

Downloaded from https://academic.oup.com/jcem/article-abstract/102/4/1244/3109069 by guest on 27 March 2019

gesterone on nuclear PRs in anovulatory women. It is well
documented that progesterone downregulates its own
receptors in preparation for embryo implantation (3).
Thus, a lack of progesterone might cause increased ex-
pression of PRs and endometrial dysfunction.
After the lifestyle intervention, ovulation was con-
firmed in one-third of the PCOS women. In contrast to
the anovulatory women, the ovulatory subgroup dis-
played mRNA expression of PRAB and PRB that was not
different from that of the controls. This finding is in
agreement with a previous report in which PR mRNA
expression in secretory endometrium of women with
PCOS was not significantly different from that of the
controls (10). However, we found that the protein levels
of PRA in the endometrium of OB-PCOS women with
restored ovulation were significantly higher compared
with the levels in the controls. The same finding of higher
protein expression of PR in the mid-secretory endome-
trium has been reported previously for women with
PCOS compared with controls (10). Increased expression
of PRs despite ovulation might indicate an impaired
endometrial response to progesterone (9). Although the
possible mechanism of endometrial progesterone re-
sistance remains to be clarified (23), alterations in PR
function and expression could lead to abnormal re-
ceptivity, poor reproductive function, and hyperplasia in
the endometrium of women with PCOS.
To our knowledge, no information has been pre-
viously reported regarding membrane-bound PRs in the
endometrium of PCOS women. We found that in the
control group, PGRMC1 mRNA expression was sig-
Figure 4. (A) Correlation between serum levels of progesterone nificantly downregulated from cycle days 6 to 8 to cycle
and immunoreactivity of the proliferative marker Ki67 on cycle days days 21 to 23. These results are in agreement with those
21 to 23 in endometrial stroma, glandular epithelium (GE), and
luminal epithelium (LE) in the combined 3 groups (obese women from previous studies showing lower mRNA expression
with PCOS and no luteal phase, obese women with PCOS and of PGRMC1 in the secretory phase compared with the
a confirmed luteal phase, and obese controls with a confirmed proliferative phase of the menstrual cycle (5). Although
luteal phase, n = 26). (B) Correlation between immunoreactivity of
Ki-67 vs PRA and PRB on cycle days 21 to 23 in the endometrial
both subgroups of OB-PCOS women exhibited higher
stroma in the combined 3 groups (obese women with PCOS and no gene expression of PGRMC1 in the endometrium on
luteal phase, obese women with PCOS and a confirmed luteal cycle days 21 to 23 compared with the controls, only the
phase, and obese controls with a confirmed luteal phase, n = 26).
subgroup of anovulatory women with PCOS displayed
protein levels of PGRMC1 and PGRMC2 significantly
greater than those of the controls. Higher expression of
endometrial PGRMC1 can be attributed to the lack of
1252 Paulson et al Progesterone Receptors in Women With PCOS J Clin Endocrinol Metab, April 2017, 102(4):1244–1253

progesterone production in this group of women (5). but does not fully restore, the expression of PRs in the
However, the regulation and role of PGRMC1 and endometrium of obese women with PCOS. We propose
PGRMC2 in the endometrium of women with PCOS that impaired progesterone action in the endometrium
remain to be determined. can lead to proliferative abnormalities and an increased
Activation of progesterone signaling affects the gene risk of cancer in women with PCOS.
expression pattern in the endometrium and its proliferative
activity (9). In our study, the anovulatory OB-PCOS Acknowledgments
women exhibited lower endometrial mRNA expression
We are grateful to Åsa Nybacka, Department of Clinical Nu-
of the progesterone regulated gene S100P on cycle days 21
trition and Dietetics, for supervising and supporting the women
to 23. In contrast, in the subgroup of PCOS women with to change their dietary habits. We thank Lotta Blomberg, Siv

Downloaded from https://academic.oup.com/jcem/article-abstract/102/4/1244/3109069 by guest on 27 March 2019

restored ovulation, the expression of S100P was not dif- Rödin Andersson, and Berit Legerstam at the Women’s Clinical
ferent from that of the controls. Moreover, expression of Health Research Unit and Britt Masironi at the Department of
the proliferation marker Ki67 was higher in all endome- Women’s and Children’s Health for skillful technical assistance.
trial compartments on cycle days 21 to 23 of the an-
ovulatory PCOS women compared with that in the Address all correspondence and requests for reprints to:
controls. In contrast, the subgroup of women with PCOS Mariana Paulson, PhD, Department of Women’s and Chil-
and restored ovulation had protein levels of Ki67 that were dren’s Health, Karolinska Institutet, Stockholm SE-17176,
Sweden. E-mail: mariana.paulson@ki.se.
not different from those of the controls. The role of
This study was supported financially by the Swedish Re-
progesterone in regulating endometrial proliferation ac-
search Council (ALH 20324), Karolinska Institutet, and the
tivity is well established (3). In our study, the progesterone regional agreement on medical training and clinical research
levels correlated negatively with immunostaining of (ALF) between Stockholm County Council and Karolinska
Ki67. Thus, it is likely that a progesterone-deficient Institutet (to A.L.H. and L.S.).
milieu could explain the abnormal gene profile and Disclosure Summary: The authors have nothing to disclose.
greater proliferative activity in anovulatory obese PCOS
women. Furthermore, a positive correlation between References
Ki67 and PRs in the stroma might suggest that higher
expression of PRs in obese PCOS women is a com- 1. Goodarzi MO, Dumesic DA, Chazenbalk G, Azziz R. Polycystic
ovary syndrome: etiology, pathogenesis and diagnosis. Nat Rev
pensatory response of the endometrium to increased
Endocrinol. 2011;7(4):219–231.
proliferation activity. 2. Rosenfield RL, Ehrmann DA. The pathogenesis of polycystic ovary
Certain limitations of the present study must be rec- syndrome (PCOS): the hypothesis of PCOS as functional ovarian
ognized, such as the small number of subjects and the hyperandrogenism revisited. Endocr Rev. 2016;37(5):467–520.
3. Giudice LC. Endometrium in PCOS: Implantation and pre-
short intervention period. The strengths were the par- disposition to endocrine CA. Best Pract Res Clin Endocrinol
ticipation of a homogenous group of obese PCOS women Metab. 2006;20(2):235–244.
fulfilling all 3 criteria of the Rotterdam consensus and 4. Critchley HO, Saunders PT. Hormone receptor dynamics in a
receptive human endometrium. Reprod Sci. 2009;16(2):191–199.
age- and BMI-matched healthy controls. Furthermore, 5. Pru JK, Clark NC. PGRMC1 and PGRMC2 in uterine physiology
endometrial biopsy specimens were collected on stan- and disease. Front Neurosci. 2013;7:168.
dardized cycle days. 6. Patel B, Elguero S, Thakore S, Dahoud W, Bedaiwy M, Mesiano S.
Role of nuclear progesterone receptor isoforms in uterine patho-
In conclusion, we have demonstrated that expression
physiology. Hum Reprod Update. 2015;21(2):155–173.
of PRAB mRNA and the PRAB/B ratio are lower in the 7. Keator CS, Mah K, Slayden OD. Alterations in progesterone re-
proliferative endometrium of obese women with PCOS ceptor membrane component 2 (PGRMC2) in the endometrium of
compared with expression in the controls. After a lifestyle macaques afflicted with advanced endometriosis. Mol Hum
Reprod. 2012;18(6):308–319.
intervention, PRAB mRNA expression was still low but 8. Li X, Feng Y, Lin JF, Billig H, Shao R. Endometrial progesterone
expression of PRB mRNA had decreased and was not resistance and PCOS. J Biomed Sci. 2014;21:2.
different from that of the controls in the proliferative 9. Savaris RF, Groll JM, Young SL, DeMayo FJ, Jeong JW, Hamilton
AE, Giudice LC, Lessey BA. Progesterone resistance in PCOS en-
endometrium. This change might indicate improved PR dometrium: a microarray analysis in clomiphene citrate-treated and
action and thereby endometrial function in women with artificial menstrual cycles. J Clin Endocrinol Metab. 2011;96(6):
PCOS after intervention. In contrast to the PCOS women 1737–1746.
10. Quezada S, Avellaira C, Johnson MC, Gabler F, Fuentes A, Vega
who remained anovulatory, the group of women with
M. Evaluation of steroid receptors, coregulators, and molecules
PCOS and confirmed ovulation displayed protein ex- associated with uterine receptivity in secretory endometria from
pression of PRB, PGRMC1, and PGRMC2 and the untreated women with polycystic ovary syndrome. Fertil Steril.
proliferation marker Ki67 on cycle days 21 to 23 that was 2006;85(4):1017–1026.
11. Moran LJ, Hutchison SK, Norman RJ, Teede HJ. Lifestyle changes
not different from that of the controls. Our findings might in women with polycystic ovary syndrome. Cochrane Database
indicate that ovulation after lifestyle intervention alters, Syst Rev. 2011;CD007506.
doi: 10.1210/jc.2016-3155 https://academic.oup.com/jcem 1253

12. Nybacka Å, Carlström K, Fabri F, Hellström PM, Hirschberg AL. and Genes (Diogenes) Project. Diets with high or low protein
Serum antimüllerian hormone in response to dietary management content and glycemic index for weight-loss maintenance. N Engl J
and/or physical exercise in overweight/obese women with poly- Med. 2010;363(22):2102–2113.
cystic ovary syndrome: secondary analysis of a randomized con- 18. Hulchiy M, Zhang H, Cline JM, Hirschberg AL, Sahlin L. Effects of
trolled trial. Fertil Steril. 2013;100(4):1096–1102. long-term tibolone treatment on nuclear sex steroid hormone re-
13. Nybacka Å, Carlström K, Ståhle A, Nyrén S, Hellström PM, ceptors and G-protein-coupled estrogen receptor-1 expression in
Hirschberg AL. Randomized comparison of the influence of dietary the macaque uterus. Menopause. 2012;19(3):332–338.
management and/or physical exercise on ovarian function and 19. Ke Y, Bertin J, Gonthier R, Simard JN, Labrie F. A sensitive, simple
metabolic parameters in overweight women with polycystic ovary and robust LC-MS/MS method for the simultaneous quantification
syndrome. Fertil Steril. 2011;96(6):1508–1513. of seven androgen- and estrogen-related steroids in postmenopausal
14. Ujvari D, Hulchiy M, Calaby A, Nybacka Å, Byström B, Hirschberg serum. J Steroid Biochem Mol Biol 2014;144(Pt B):523–534.
AL. Lifestyle intervention up-regulates gene and protein levels of 20. Södergård R, Bäckström T, Shanbhag V, Carstensen H. Calculation
molecules involved in insulin signaling in the endometrium of of free and bound fractions of testosterone and estradiol-17 beta to

Downloaded from https://academic.oup.com/jcem/article-abstract/102/4/1244/3109069 by guest on 27 March 2019

overweight/obese women with polycystic ovary syndrome. Hum human plasma proteins at body temperature. J Steroid Biochem.
Reprod. 2014;29(7):1526–1535. 1982;16(6):801–810.
15. Hulchiy M, Nybacka Å, Sahlin L, Hirschberg AL. Endometrial 21. Barry JA, Azizia MM, Hardiman PJ. Risk of endometrial, ovarian
expression of estrogen receptors and the androgen receptor in and breast cancer in women with polycystic ovary syndrome:
women with polycystic ovary syndrome: a lifestyle intervention a systematic review and meta-analysis. Hum Reprod Update. 2014;
study. J Clin Endocrinol Metab. 2016;101(2):561–571. 20(5):748–758.
16. Rotterdam ESHRE/ASRM-Sponsored PCOS Consensus Work- 22. Wise MR, Jordan V, Lagas A, Showell M, Wong N, Lensen S,
shop Group. Revised 2003 consensus on diagnostic criteria and Farquhar CM. Obesity and endometrial hyperplasia and cancer in
long-term health risks related to polycystic ovary syndrome. Fertil premenopausal women: a systematic review. Am J Obstet Gynecol.
Steril. 2004;81(1):19–25. 2016;214(6):689.e1–689.e17.
17. Larsen TM, Dalskov SM, van Baak M, Jebb SA, Papadaki A, 23. Al-Sabbagh M, Lam EW, Brosens JJ. Mechanisms of endometrial
Pfeiffer AF, Martinez JA, Handjieva-Darlenska T, Kunešová M, progesterone resistance. Mol Cell Endocrinol. 2012;358(2):
Pihlsgård M, Stender S, Holst C, Saris WH, Astrup A; Diet, Obesity, 208–215.