Key words
anti-Mullerian hormone - insulin resistance - apoptosis - pancreatic islet cells
Introduction
Polycystic ovary syndrome (PCOS) is the most common endocrine disorder in
reproductive-age women, with a prevalence of 15% depending on the diagnostic
criteria used and the population studied [1].
Its clinical manifestations may include menstrual irregularities, signs of
hyperandrogenism, and obesity. Insulin resistance (IR) is one of the most common
features of PCOS and is associated with an increased risk of type 2 diabetes and
cardiovascular events, with a consequent negative effect on long-term health [2]. Although IR is not required for a diagnosis
of PCOS, it is clear that intrinsic IR is present in many women with PCOS [3]. However, the underlying etiology and
pathogenesis of PCOS remain poorly understood, and there is no effective method of
preventing the disease.
As a member of the transforming growth factor-β (TGF-β) family,
anti-Müllerian hormone (AMH) is expressed by granulosa cells (GCs) of early
developing follicles in the ovary during the reproductive age [4]
[5].
AMH is considered a reliable of ovarian reserve in humans and strongly correlate
with the number of antral follicles [6]
[7]. In the ovary, the main physiologic role of
AMH seems to be limited to the inhibition of the early stages of follicular
development [8] and preventing the recruitment
of non-dominant follicles [9]. Although AMH is
mainly thought to play a local role in ovarian follicle recruitment and development
[7], it is also available in the
peripheral circulation.
IR results from impaired insulin signaling in target tissues that leads to increased
levels of insulin required to control plasma glucose levels. IR and pancreatic
β-cell dysfunction are two major pathophysiological characteristics of type
2 diabetes mellitus (T2DM), thus pancreatic β-cells play a
crucial role in T2DM pathogenesis [10]
[11]. In the early stage of
T2DM development, the pancreas first increases insulin release by
improving the function of the β-cells or increasing their number when IR
occurs in peripheral tissues. The β-cell apoptosis caused by high-glucose
toxicity results in decreased β-cell function and insulin sensitivity,
further reducing insulin secretion in diabetic patients with long-term
hyperglycemia.
It is well known that obesity is always related to IR in women with PCOS, while
whether the increased AMH levels in PCOS is involved in the pathological process of
IR in reproductive-age women with PCOS and whether the increased AMH has an effect
on pancreatic β-cells still remains unclear. Hence, the aim of the present
study was to investigate the underlying relationship between serum AMH levels and
the HOMA-IR, and also with the insulin levels after glucose stimulation.
Furthermore, we conducted an in vitro model to examine the effects of high AMH in
association with high leptin levels on insulin secretion and the expression of
pancreatic β-cell pro-apoptosic genes in the presence of high glucose
concentrations.
Subjects and Methods
Samples collection
The study was approved by the ethical committee of School of Medicine, Zhejiang
University, Hangzhou, China. Total of 345 women were recruited from the
Reproductive Center of Hangzhou Women’s Hospital between 2013 and 2015.
Fifty women whose age was more than 37 years and/or with hypertension
and other medical diseases were excluded, 295 women including 114 women with
PCOS and 181 women without PCOS were enrolled. After stratified by BMI, PCOS
women were subgrouped into normal BMI (18.5–24.9
kg/m2, n=73) and high BMI groups (≥25
kg/m2, n=41). Among PCOS women with normal BMI,
35 women with normal AMH level and 38 women with high AMH level were analyzed.
And among obese PCOS women, 22 women with normal AMH and 19 women with high AMH
level were analyzed. The diagnosis of PCOS was made according to the Rotterdam
Consensus [12]. Women with
hyperprolactinemia, hypothyroidism, androgen-secreting tumors, Cushing syndrome,
congenital adrenal hyperplasia, or diabetes were also excluded.
Analysis of serum hormone and biochemical indicators analysis
Basal hormone and metabolic profiles of patients were examined on days
2–5 of menstruation after fasting for 10–12 hours, in subjects
who did not take any medications, including hormones, in the past 3 months. The
concentrations of basal sex hormone and insulin levels were measured at the
clinical chemistry laboratory of Hangzhou Women’s Hospital (DX8001;
Beckman Instrument). Serum glucose levels were determined using an automated
hexokinase method (AU5821; Beckman Instrument), the triglyceride contents were
measured by a standard colorimetric assay (AU5821; Beckman Instrument) and the
cholesterol levels were measured by enzyme assays (AU5821; Beckman Instrument).
The amount of AMH was quantified with automated chemiluminescent immunoassay
method (Roche Diagnostics; Basel, Switzerland), measuring a range of 0.01
– 23 ng/ml, the detection sensitivity of 0.3 ng/ml and
intra- and inter-assay coefficients of variation<10%. HOMA-IR
index was calculated according to the formula:
HOMA-IR+=+(fasting glucose)
(mmol/l)+×+(fasting insulin)
(μU/ml)/22.5) [13].
Pancreatic islet isolation and insulin secretion
In anesthetized 8-week-old rats, 2 mg/ml collagenase (Worthington
Biochemical, Lakewood, NJ) was injected into the bile duct as previously
described [14]
[15]. The pancreas was isolated, with static
incubation at 38°C for 20 minutes followed by shaking incubation at
38°C for 7 minutes. After ficoll gradient separation, freshly isolated
islets of similar size were handpicked under a stereomicroscope. The islets (20
per well) were preincubated in RPMI-1640 (Sigma-Aldrich, St. Louis, MO, USA)
washed, and incubated in 500 μl fresh RPMI-1640 containing 11.1
mmol/l glucose, supplemented with 10% fetal calf serum that
contained high glucose (25 mmol/l) and this was incubated with different
concentration of recombinant rat leptin (L5073, Sigma-Aldrich) and/or
r-AMH (CSB-EP001666RA, CUSABIO) for 30 minutes (37°C). Fifty microliters
of the medium were removed for insulin analysis and the islets were extracted
and stored (–80°C) for protein assay and western blotting. After
incubation for 24 hours, cells were treated with different concentration of
leptin (0, Tris-HCl, 25, 50, 100, 200 ng/ml).
Western blotting
Western blotting was performed as described previously [16]. The islets samples were lysed and
centrifuged before measuring the concentration of protein. In brief, the
extracted protein was transferred to polyvinylidene difluoride membranes and
incubated overnight at 4°C with rabbit polyclonal
anti-α-caspase-3 (1:1000; Cell Signaling Technology, Boston, CA, USA),
anti-α-caspase-8 (1:1000; Cell Signaling Technology, Boston, CA, USA),
anti-α-caspase-9 (1:1000; Cell Signaling Technology, Boston, CA, USA),
anti-Bax (1:1000; Abcam, Cambridge, MA, USA) and human polyclonal
anti-β-actin (1:10 000; Abcam). The samples were then incubated
with fluorescence-labeled anti-mouse IgG or anti-rabbit IgG antibody (1:5000,
Dylight 680 or 800, KPL) for 1 h at room temperature, and, the protein bands
were obtained with enhanced chemiluminescence (Millipore). Image of WB was
captured by a CCD camera and analyzed with an Odyssey Imager (Li-cor; Odyssey,
NE, USA). By using this software, the intensity of the bands was translated into
a numerical value – the brighter the band, the higher the number. The
results of western blots were assessed visually by making comparisons between
bands in different lanes.
Statistical analyses
All data are presented as mean±SD or SEM. An independent-sample
t-test, the nonparametric test was used to evaluate the statistical
significance between two groups. Associations between parameters were assessed
using the Pearson correlation coefficient. Multiple linear regression analysis,
using HOMR-IR, fasting insulin and insulin levels after oral administration of
glucose as dependent variables, and, BMI, serum leptin levels, AMH levels, basal
hormone levels, and lipids levels as independent variables, were calculated.
SPSS (version 19.0 for Windows) was used for the statistical analysis. p-Values
of<0.05 were considered to be statistically significant.
Results
PCOS exhibit hyperinsulinemia and IR
The flow chart of our prospective cohort study was presented in Fig. S1.
Among 345 women of reproductive age, despite 50 women with advanced age
and/or diabetes, hypertension or other medical diseases, 295 women (114
with PCOS and 181 healthy women) were recruited (87.41%).
Women’s age was comparable. Serum metabolic parameters in women with or
without PCOS were measured. There were no significant differences in height,
basal FSH, PRL, and serum triglyceride (TG) in both groups. However, body
weight, body mass index (BMI), serum LH, E2, total testosterone (TT),
Total cholesterol (TC), LDL-c levels were higher and HDL-c was lower in PCOS
group ([Table 1]). Furthermore, levels of
serum leptin AMH, insulin, fasting glucose, and HOMA-IR were significantly
higher in women with PCOS.
Table 1 Serum metabolic parameters between women with and
without PCOS.
|
Non-PCOS
|
PCOS
|
p-Value
|
|
No. of subjects
|
181
|
114
|
|
|
Age (years)
|
27.71±4.17
|
27.53±4.14
|
0.372
|
|
Height (cm)
|
157.80±4.79
|
159.98±4.96
|
0.095
|
|
Weight (kg)
|
54.61±6.37
|
62.16±12.45
|
<0.0001
|
|
BMI (kg/m2)
|
21.22±2.32
|
23.82±4.65
|
<0.0001
|
|
Basal LH level (mIU/ml)
|
5.81±3.89
|
10.95±6.65
|
<0.0001
|
|
Basal FSH level (mIU/ml)
|
5.84±1.84
|
5.93±1.46
|
0.380
|
|
Basal E2 level (pmol/l)
|
82.63±70.28
|
176.55±90.66
|
<0.0001
|
|
TT level (nmol/l)
|
0.63±0.44
|
2.21±0.91
|
<0.0001
|
|
Basal PRL level (ng/ml)
|
13.54±10.14
|
15.21±8.64
|
0.143
|
|
Serum triglyceride (mM)
|
1.52±1.39
|
1.77±1.23
|
0.079
|
|
Serum Cholesterol (mM)
|
4.73±0.81
|
4.98±1.08
|
0.022
|
|
Serum HDL-c (mM)
|
1.58±0.39
|
1.43±0.45
|
0.004
|
|
Serum LDL-c (mM)
|
2.55±0.63
|
2.97±0.90
|
<0.0001
|
|
Serum leptin level (ng/ml)
|
5.42±5.31a
|
11.84±11.28b
|
<0.0001
|
|
Serum AMH level (ng/ml)
|
3.60±1.71
|
7.66±6.48
|
<0.0001
|
|
Fasting glucose (mM)
|
4.46±0.59
|
4.81±0.84
|
<0.0001
|
|
Fasting insulin (μU/ml)
|
6.10±3.77
|
15.17±8.84
|
<0.0001
|
|
HOMA-IR
|
1.21±0.77
|
2.63±1.50
|
<0.0001
|
Data=mean±SD, p-values were calculated by t-test.
BMI: Body mass index; LH: Luteinizing hormone; FSH: Follicle-stimulating
hormone; E2: Estradiol; TT: Total testosterone; PRL: Prolactin; HDL-c:
High-density lipoprotein-cholesterol; LDL-c: Low-density
lipoprotein-cholesterol; HOMA-IR: Homeostatic Model Assessment for
Insulin Resistance. a n=47, b
n=91.
Possible correlations between insulin resistance, insulin levels after
glucose stimulation, and hormone and lipids parameters in PCOS women
To investigate the risk factors for IR and insulin secretion after glucose
stimulation in PCOS patients, we analyzed the correlation coefficients between
HOMA-IR, and metabolic and hormone parameters. HOMA-IR was positively correlated
to BMI, leptin, AMH, and LDL-c levels ([Table
2]). Glucose treatment for 1st and 2nd hours of fasting markedly
increased BMI, insulin and leptin levels. AMH showed a significant positive
trend in correlation with HOMA-IR and insulin concentrations ([Table 2]). Our present data suggest that
high BMI and leptin levels were the main risk factor of IR, while the elevated
AMH levels were associated with insulin secretion on the background of glucose
stimulation.
Table 2 Correlation coefficients between insulin
resistance, insulin levels after glucose stimulation, and hormone
and lipids parameters in PCOS women.
|
Parameters
|
HOMA-IR (n=74)
|
Fasting insulin (n=74, mmol/l)
|
1 hour insulin (n=42, μU/ml)
|
2 hours insulin (n=42, μU/ml)
|
|
BMI (kg/m2)
|
0.227
#
|
0.157
|
0.341
#
|
0.326
#
|
|
Leptin levels (ng/ml)
|
0.326*
|
0.192
|
0.315
#
|
0.239
|
|
AMH levels (ng/ml)
|
0.222
#
|
0.211
#
|
0.506†
|
0.468*
|
|
Basal LH level (mIU/ml)
|
0.107
|
–0.127
|
0.082
|
–0.321
#
|
|
Basal FSH level (mIU/ml)
|
0.014
|
0.010
|
0.287
#
|
0.150
|
|
TT (nmol/l)
|
–0.058
|
0.094
|
0.060
|
0.217
|
|
Serum TG (mmol/l)
|
0.069
|
0.102
|
–0.039
|
–0.106
|
|
Serum TC (mmol/l)
|
0.048
|
–0.001
|
0.173
|
0.234
|
|
Serum HDL (mmol/l)
|
–0.242
#
|
–0.284
#
|
–0.161
|
–0.335
#
|
|
Serum LDL (mmol/l)
|
0.248
#
|
0.139
|
0.133
|
0.287
#
|
HDL: High-density lipoprotein; HOMA: Homeostatic Model Assessment; and
LDL: Low-density lipoprotein. †
p<0.001; * p<0.01;
#
p<0.05.
In order to determine the interaction between obesity and elevated AMH levels on
insulin resistance, we further analyzed the metabolic and hormone parameters
among PCOS women after stratified by BMI (Fig. S1 and [Table 3]). No statistic correlation
between leptin and AMH levels was found (data not shown). Surprising, PCOS women
in high AMH level group showed an increased HOMA-IR when compared to normal AMH
level group among obese PCOS women (3.96±1.27 vs. 3.02±1.42).
The HOMA-IR in obese PCOS women with high AMH was the highest among the four
subgroups ([Table 3]). However, among
PCOS women with normal BMI, women with high AMH presented an obviously elevated
fasting insulin levels when compared to normal AMH group (20.14±11.96
vs. 10.71±4.98). But there was no difference in HOMA-IR between normal
and high AMH groups among PCOS women with normal BMI ([Table 3]).
Table 3 Serum metabolic parameters between PCOS women with
normal and high AMH levels stratified by BMI.
|
Normal BMI (18.5–24.9 Kg/m2)
|
p-Value
|
High BMI (≥25 Kg/m2)
|
p-Value
|
|
Normal AMH
|
High AMH
|
Normal AMH
|
High AMH
|
|
No. of subjects
|
35
|
38
|
|
22
|
19
|
|
|
Age (years)
|
26.94±4.38
|
27.73±4.22
|
0.229
|
26.95±5.54
|
26.71±4.86
|
0.442
|
|
Height (cm)
|
159.50±5.22
|
160.26±4.47
|
0.274
|
159.43±5.12
|
161.31±5.27
|
0.141
|
|
Weight (kg)
|
55.41±8.60
|
55.12±6.28
|
0.433
|
75.40±9.63
|
74.03±10.17
|
0.331
|
|
BMI (kg/m2)
|
20.94±3.15
|
21.20±2.20
|
0.345
|
29.22±3.34
|
28.42±2.80
|
0.209
|
|
Serum AMH level (ng/ml)
|
3.18±1.57
|
10.95±4.69
|
<0.001
|
3.20±1.92
|
14.48±8.75
|
<0.001
|
|
Basal LH level (mIU/ml)
|
12.51±7.82
|
12.05±7.25
|
0.427
|
9.33±4.60
|
8.21±4.71
|
0.261
|
|
Basal FSH level (mIU/ml)
|
6.07±1.51
|
5.92±1.11
|
0.375
|
5.66±1.44
|
6.04±1.89
|
0.266
|
|
Basal E2 level (pmol/l)
|
180.22±93.14
|
189.14±76.16
|
0.381
|
170.65±100.26
|
162.62±94.05
|
0.414
|
|
TT level (nmol/l)
|
2.01±0.87
|
2.79±0.83
|
0.003
|
1.96±0.81
|
2.47±1.06
|
0.700
|
|
Basal PRL level (ng/ml)
|
14.83±8.57
|
15.14±10.21
|
0.456
|
14.97±7.56
|
16.83±9.89
|
0.283
|
|
Serum triglyceride (mM)
|
1.77±0.99
|
1.53±0.98
|
0.174
|
1.86±1.16
|
2.08±2.01
|
0.331
|
|
Serum Cholesterol (mM)
|
5.08±1.11
|
4.72±1.57
|
0.091
|
5.04±1.15
|
5.17±1.04
|
0.352
|
|
Serum HDL-c (mM)
|
1.58±0.39
|
1.41±0.35
|
0.066
|
1.40±0.36
|
1.20±0.35
|
0.041
|
|
Serum LDL-c (mM)
|
3.17±0.77
|
2.61±0.94
|
0.006
|
3.03±0.85
|
3.16±1.02
|
0.335
|
|
Fasting glucose (mM)
|
4.94±0.66
|
4.63±0.94
|
0.059
|
4.99±0.83
|
4.77±0.94
|
0.211
|
|
Fasting insulin (μU/ml)
|
10.71±4.98
|
20.14±11.96
|
<0.001
|
11.98±6.95
|
13.32±9.40
|
0.302
|
|
HOMA-IR
|
2.53±1.19
|
2.46±1.95
|
0.471
|
3.02±1.42
|
3.96±1.27
|
0.028
|
Data=mean±SD; p-values were calculated by t-test.
BMI: Body mass index; LH: Luteinizing hormone; FSH: Follicle-stimulating
hormone; E2: Estradiol; TT: Total testosterone; PRL: Prolactin; HDL-c:
High-density lipoprotein-cholesterol; LDL-c: Low-density
lipoprotein-cholesterol; HOMA-IR: Homeostatic Model Assessment for
Insulin Resistance.
Co-treatment of high levels of AMH plus leptin increases insulin secretion in
vitro
Insulin itself appears to be a proximate and important quantitative contributor
to insulin resistance [17]. In order to
verify the direct effect of AMH on insulin secretion, isolated pancreatic
(identified by incubating with dithizone, [Fig.
1a]) were cultured for 2 hours in a high glucose medium (25
mmol/l) and treated with different AMH (0.25 ng/ml and 1
ng/ml) concentrations, or with AMH (1 ng/ml) plus leptin (200
ng/ml) concentrations. We found a significant increase in levels of
insulin secretion in the group treated with high levels of AMH plus leptin,
compared to the group treated with AMH only ([Fig. 1b]). The results indicate that a slight increase in AMH
concentrations may accelerate pancreatic islets insulin secretion.
Fig. 1 Co-treatment of high levels of AMH plus leptin increased
insulin secretion. a: Adult islets were identified by incubating
with dithizone (DTZ). Scale bar, 200 mm. b: Insulin secretion in
the groups treated with different concentration of leptin or AMH levels
and high levels of AMH plus leptin.
High AMH upregulated the expression of pro-apoptotic proteins in pancreatic
islet cells in association with high leptin levels
To further examine whether cell death may happen in in vitro model, we tested the
expression of pro-apoptotic genes in pancreatic islets treated with leptin (200
ng/ml) or leptin plus high concentration of AMH (1 ng/ml) for 2 hours was
developed in a medium containing a high level of glucose. Our results showed
that isolated islet cells exposure to high AMH (1 ng/ml) plus leptin levels (200
ng/ml) increased the expression levels of pro-apoptotic proteins, such as Bax,
caspase-3, and caspase-8, compared to the group treated with leptin only ([Fig. 2a–e]).
Fig. 2 High AMH promotes apoptosis in pancreatic islet cells in
association with high leptin levels. a: Protein expression of
BAX, caspase-3, caspase-8 and caspase-9 in pancreatic islet cells
treated with leptin (200 ng/ml) or leptin plus high concentration of AMH
(1 ng/ml) for 2 hours (n=4, respectively). This figure shows one
representative experiment among three separate experiments. b:
Relative protein levels of BAX normalized to β-actin. c:
Relative protein levels of caspase-3 normalized to β-actin.
d: Relative protein levels of caspase-8 normalized to
β-actin. e: Relative protein levels of caspase-9
normalized to β-actin. The values are expressed as
mean±SEM.*p<0.05,**p<0.01.
Discussion
Our results illustrate that the possible average value for basal hormonal and lipids
levels, HOMA-IR as well as insulin levels for both fasting and after glucose
stimulation are significantly higher in PCOS patients compared to the controls.
Multiple linear regression analysis showed that HOMA-IR was positively correlated
to
BMI and leptin levels, which indicated that PCOS women with high leptin level were
susceptible to IR. Furthermore, elevated AMH levels were related to fasting insulin
level and continuously insulin release after glucose administration, and, HOMA-IR
was the highest in obese PCOS women with high AMH levels, which implied that high
AMH might promote the intolerance of glucose and pathological process of IR.
IR is highly prevalent among PCOS patients, independent of obesity [18]. Hyperinsulinemia seems to be associated
with hyperandrogenism, since insulin acts in the theca interna cells, potentiating
the effects of luteinizing hormone (LH) on steroidogenesis and reducing hepatic
production of sex-hormone-binding globulin, resulting in higher concentrations of
free androgens [19]. In our daily clinical
practice, AMH is widely used to observe ovarian response and oocyte yield. PCOS
patients had significantly higher AMH levels compared to healthy fertile women [20]. Serum AMH level increased with antral
follicle count (AFC), LH, T, and E2 in PCOS women and decreased with BMI,
age, and FSH in infertile control women [21].
Circulating AMH was well known to be play important role in ovarian follicle
development and was considered to be a marker of reproductive ageing. Recent
epidemiological investigation was surveyed in 3293 female participants and it was
observed that lower age-specific AMH levels were associated with a higher risk of
type 2 diabetes in women [22]. However,
whether AMH is involved in part of the pathological process of IR remains unclear.
It has been reported that there was a significant positive correlation between AMH
levels and IR in non-obese adolescent females with PCOS [23]. But most studies have shown that women
with PCOS and IR presented high AMH levels [24]
[25]. Hyperinsulinemia resulting
from IR appears to increase the premature differentiation of the granulosa cells,
suggesting that IR plays some role in AMH secretion by these cells. Caglar et al.
[26] designed a study to evaluate the
correlation between AMH levels and IR in normal-weight PCOS patients. Although the
PCOS patients were not subdivided according to different phenotypes, no correlation
was found between AMH levels and IR. Later on, Tian et al. [27] conducted a more comprehensive study
involving more patients, and according to the results, no correlation was observed
between AMH and indices of IR among all phenotypes of PCOS cases. In contrast, La
Marca et al. [28] found a significant positive
association between AMH and IR, as well as an association between androgens and AMH
in women with PCOS. Nevertheless, in relation to IR and PCOS, a direct correlation
has been found between antral follicle count, ovarian volume and hyperinsulinemia
[24]
[29]. So, it is reasonable to suggest that a high AMH level is a risk
factor of IR. In our study, we found that women with high serum AMH levels
positively correlated to fasting insulin and insulin levels of glucose tolerance,
and the patients with high AMH levels presented higher insulin levels as well.
Considering that excess body fat is the primary predictor of insulin resistance in
PCOS subjects [30], we then analyzed the
insulin levels and HOMA-IR between PCOS women with normal or high AMH levels
stratified by BMI. Indeed, women with BMI≥25 Kg/m2 showed
higher HOMA-IR then women with normal BMI. Importantly, PCOS women with
BMI≥25 Kg/m2 and high AMH presented a higher HOMA-IR as
well. But in PCOS women with normal BMI, women with high AMH showed an increased
fasting insulin level, which indicated that AMH might promote the insulin secretion
and played a role in the initial compensatory phase of diabetes, and once women
underwent overweight or obesity, high AMH would play a very important role in the
process of developing insulin resistance.
In addition, there are some indications that AMH may be directly involved in the
physiology of target organs and increased the risk of disease [31]
[32].
In vitro studies have confirmed that elevated AMH can affect the primary-cultured
human luteinizing granulosa cells to attenuate FSH-induced estradiol and
progesterone production [33]. High AMH level
had an inhibitory effect on early human ovarian follicular development and
suppressed the initiation of primordial follicle growth [34]. Taking these studies into account, we
speculate that high peripheral AMH may exert an influence on pancreatic islets
function and may increase the risk of IR and diabetes later in life, especially in
obese PCOS women.
Using in vitro model, we demonstrated that r-leptin and r-AMH exert effects on the
primary islets. Isolated islet cells treatment with leptin (200 ng/ml) or
leptin plus AMH (1 ng/ml) significantly promoted insulin secretion.
Interestingly, however, high levels of AMH non-significantly increase insulin
secretion, compared to a group co-treated leptin plus AMH. Based on these results,
we may suggest that high AMH in association with leptin might have a role in
aggravating the impairment of pancreatic islet function. In order to overcome
insulin resistance, the pancreatic islet β-cells releases more insulin
initially to prevent hyperglycemia [35].
However, with the elevation of glucose continues to rise, this hyperactivity causes
β-cell dysfunction that results in cell death [36]. For further analysis, we subjected the
pancreatic islets to r-leptin or r-leptin plus r-AMH to study the effect of high AMH
on apoptosis of islets under the condition of high glucose. Caspases
(cysteine-aspartic proteases, cysteine aspartases or cysteine-dependent
aspartate-directed proteases) are a family of protease enzymes playing essential
roles in programmed cell death (including apoptosis, pyroptosis, and necroptosis)
and inflammation. Caspase-3 is a converging point of the apoptotic pathway [37], and its peptide inhibitors have been shown
to prevent islet apoptosis [38]
[39]. Conversely, over-activation of some
caspases such as caspase-3 can lead to excessive programmed cell death.
As another pro-apoptotic protein, BAX plays a critical role in promoting apoptosis
[40]. Bax can translocate from the cytosol
to mitochondria in response to pro-apoptotic stimuli, resulting in mitochondrial
cytochrome-c release and activation of caspase-9 and caspase-3 [41]. In addition, activated caspase-8 can
cleave Bid, which in turn triggers the release of cytochrome-c from mitochondria and
the activation of caspase-9 and caspase-3 [42]. Hyperglycemia-induced β-cell apoptosis has been implicated and
has been studied mainly in T2DM [36]. We found that co-treatment of leptin plus AMH significantly
increased the expression of Bax, caspase-3, and caspase-8 in the pancreatic islets
isolated from rats.
The strength of our study is its prospective nature. All of the Rotterdam criteria
were met for each PCOS diagnosis. To our knowledge, the study was designed for the
first time to evaluate the association between AMH and IR in reproductive PCOS
women. However, there are a few limitations that need to be addressed. First, the
number of subjects was small in this study and thus do not wholly represent the
general population. Second, more detail information, such as ovarian size and
follicle count were not determined. Third, we only found co-treatment of high levels
of AMH plus leptin increases insulin secretion and upregulated the expression of
pro-apoptotic proteins when continuous incubating with high glucose level. The in
vitro experiment is insufficient to state the relation of the pathogenesis of
insulin resistance (IR) to high AMH. Therefore, a well-designed case-control or
cohort study needs to be performed and more experiments for investigating the
relationship between AMH and apoptosis were needed in the future.
Conclusions
In obese PCOS patients, high levels of AMH may present an increased risk for IR. In
vitro study showed that increasing AMH levels in pancreatic β-cells may
trigger the secretion of insulin, and high AMH level in association with leptin
promoted hyperglycemia-induced β-cell pro-apoptotic proteins expression.
However, further study needs to fully clarify the molecular mechanism of AMH-induced
activation of apoptosis.
Author Contributions
X.H.L and H.F.H conceived, designed, and supervised the study as well as critically
revising the manuscript; X.J.L and H.W collected the blood samples and wrote the
manuscript; T.T.Y and K.U carried out the in vitro experiments; X.Y.S and H.Y.S did
the hormone, lipids, and other blood tests; X.Y.F and Z.Y.L helped for the follow-up
study and discussions.
