Endometriosis Triggers Excessive Activation of Primordial Follicles via PI3K-PTEN-Akt-Foxo3 Pathway
Arisa Takeuchi,1 Kaori Koga,1 Erina Satake,1 Tomoko Makabe,1 Ayumi Taguchi,1 Mariko Miyashita,1 Masashi Takamura,1 Miyuki Harada,1 Tetsuya Hirata,1
Yasushi Hirota,1 Osamu Yoshino,2 Osamu Wada-Hiraike,1 Tomoyuki Fujii,1 and Yutaka Osuga1
1Obstetrics and Gynecology, University of Tokyo, Hongo Bunkyo Tokyo 113-8655, Japan; and 2Obstetrics
and Gynecology, Kitasato University School of Medicine, Kitasato, Minami-ku, Sagamihara-shi, Kanagawa 252-0374, Japan
ORCiD numbers: 0000-0002-5057-4120 (K. Koga).
Abstract
Context: The ovarian reserve is reduced in patients with endometriosis. We hypothesize that the phosphatidylinositol 3-kinase (PI3K)-phosphatase and tensin homolog deleted on chromosome 10 (PTEN) Akt-Forkhead box O (Foxo3) pathway is involved in reducing the ovarian reserve.
Objective: To elucidate the signaling mechanism by which endometriosis decreases ovarian reserve.
Design: Studies were conducted by using a mouse model for endometriosis and human ovaries. The endometriosis mouse model was established and ammonium trichloro (dioxoethylene-o,o0) tel- lurate (AS101), an inhibitor of PI3K-PTEN-Akt pathway, was administered to experimental mice. Human ovaries were collected during surgery from patients with endometrioma or from patients with no ovarian pathology (control ovaries). The number of follicles and expression of Foxo3, PTEN, phosphorylated mammalian target of rapamycin and phosphorylated Akt by oocytes in primordial follicles in mouse and human ovaries were detected by immunohistochemical staining and evaluated.
Results: In the endometriosis mouse model, the proportion of primordial follicles was diminished, and the proportion of primary, secondary, antral, and growing follicles was increased in comparison with controls. In both mouse and human ovaries, the PI3K-PTEN-Akt-Foxo3 pathway was activated in samples from endometriosis. Administration of AS101 restored the proportion of primordial follicles in endometriotic mice ovaries to control levels.
Conclusions: The current study describes the excessive activation of primordial follicles and the role of the PI3K-PTEN-Akt-Foxo3 pathway in the reduction of ovarian reserve associated with endometriosis. Our results suggest that a PI3K-PTEN-Akt inhibitor should be considered for further investigation as promising medicines for the prevention of the ovarian reserve reduction in patients with endometriosis. (J Clin Endocrinol Metab 104: 5547–5554, 2019)
Introduction
Endometriosis causes infertility via various mecha- nisms, such as tubal adhesion, reduction in ovarian reserve, and implantation failure (1). Among these, re- duction in ovarian reserve has been emphasized because it worsens with age and cannot be restored with assisted reproductive technology (1). Recently, a systematic re- view and meta-analysis showed that ovarian reserve evaluated with antimu¨ llerian hormone (AMH) is reduced in patients with ovarian endometriomas compared with patients with other benign ovarian cysts and patients with healthy ovaries (2). A recent prospective longitu- dinal study also demonstrated that serum AMH levels underwent an accelerated progressive decline in women with endometrioma in comparison with AMH levels in healthy women (3). This decline is observed not only in endometrioma but also in endometriosis without endo- metrioma (4). Therefore, understanding the mechanism by which endometriosis decreases ovarian reserve is crucial for developing new strategies to prevent and treat endometriosis-associated infertility.
The ovarian reserve consists of dormant primordial follicles, which are maintained by a continuous repression of follicle activation. Excessive activation of primordial follicles causes premature exhaustion of primordial folli- cles numbers, which results in diminished ovarian reserve, and is commonly observed in patients of idiopathic or chemotherapy-induced premature ovarian insufficiency (5). In endometriosis, Kitajima et al. (6) previously showed that the proportion of primordial follicles is low, whereas growing follicles are high in women with ovarian endo- metrioma, and they suggested that a similar mechanism with premature ovarian insufficiency was likely to be responsible for ovarian reserve decline in women with endometriomas. However, the underlying mechanism causing the reduction has not been clarified.
Forkhead box O3 (Foxo3) is the most studied molecule in the oocyte that regulates the activation of primordial follicles. In Foxo3-null mice, all dormant follicles were prematurely activated in the ovary at puberty (7). When Foxo3 is constitutively active in mouse oocytes, the de- velopment of oocytes and follicles was retarded and the ovarian reserve was preserved (8, 9). These results imply that a main function of Foxo3 is to maintain the dormant state of primordial follicles in the ovary. As for the regulation of Foxo3, phosphatidylinositol 3-kinase (PI3K) – phosphatase and tensin homolog deleted on chromosome 10(PTEN)-Akt signaling is the principal pathway (10, 11), and these findings encouraged us to hypothesize that the PI3K-PTEN- Akt-Foxo3 pathway is involved in the reduction of ovarian reserve asso- ciated with endometriosis. To test this hypothesis, we conducted a study us- protocols approved by the Animal Care and Use Committee of the University of Tokyo. Five-week-old female BALB/c mice were obtained from Japan SLC, Inc. The mice were fed on a mouse diet and water and maintained on a light/dark cycle (12 hours/12 hours) under controlled living conditions.
The endometriosis model mice were induced according to the method Takamura et al. reported (12) (Fig. 1). This model was established based on the retrograde shed endometrium theory. Uterine tissues taken from homologous donor mice were chopped into small fragments and dispersed into the peritoneal cavity in recipient mice (day 0) to mimic the men- strual shedding of human endometrial tissue. Donor and re- cipient mice received subcutaneous 100-mg/kg injections of estradiol valerate (Cayman Chemical, Ann Arbor, MI) every 7 days commencing 14 days (day 214) before endometrium dispersion (day 0). Fourteen days after dispersion (day 14) the mice were euthanized, and the ovaries and endometriotic le- sions (Fig. 2) were collected and weighed. The positive control mice for follicle activation were prepared according to the method given by Kalich-Philosoph et al. (13). Briefly, cyclo- phosphamide (Tokyo Chemical Institute, Tokyo, Japan) at 75 mg/kg was given once on day 7 before euthanasia 7 days later (day 14). The negative control mice received normal saline solution instead of endometrial tissue or cyclophosphamide. In a preliminary study, we have confirmed that the estradiol injections used for the endometriosis model mice do not affect the proportion of each stage of follicles; however, to exclude the possible effect of estradiol, both cyclophosphamide-treated and control mice received estradiol injections in the same manner and time point as the endometriosis model mice. AS101 [am- monium trichloro (dioxoethylene-o,o0) tellurate], a PI3K-PTEN- Akt pathway modulator, was purchased from Sigma Aldrich (St. Louis, MO) and injected intraperitoneally at a dose of 10 mg per mouse [equivalent to intravenous treatment (3 mg/m2) admin- istered to human in previous studies (13)] from day 27 every second day until euthanasia (day 14).
Follicle counts in mouse ovaries
Ovaries fixed in paraformaldehyde were paraffin-embedded and serially sectioned (20-mm sections). Follicle counts were ing a mouse endometriosis model and ovarian samples from patients with endometriosis.
Materials and Methods
Generation of the mouse model
All procedures involving experimental animals were conducted in accordance with
Figure 1. Preparation of control, endometriosis model and cyclophosphamide-treated mice (positive control for primordial follicle activation). All mice received a subcutaneous injection of 100 mg/kg estradiol valerate every week for 4 wk. For the endometriosis model mouse, uterine tissues taken from homologous donor mice were chopped into small fragments.
These fragments, which mimic shed human endometrium in women, were dispersed into the peritoneal cavity in recipient mice (day 0). After 14 d (day 14), mice were euthanized and ovaries and endometriosis lesions were collected. The positive control mice (for primordial follicle activation) were prepared by administering cyclophosphamide at 75 mg/kg once on day 7. Negative control mice received normal saline solution instead of uterine fragments or cyclophosphamide. AS101, a PI3K-PTEN-Akt pathway modulator, was injected intraperitoneally at 10 mg per mouse every second day from 7 d before the dispssersion of endometrial tissue (day 7) until euthanasia (day 14).
Figure 2. Representative findings of mouse endometriosis lesion. [(a-1)–(a-3)] Macroscopic findings of endometriosis lesion. Scale indicates 1 cm. [(b-1) and (b-2)] Microscopic findings of endometriosis lesion. Original magnification is 3 40 (b-1) and 3 100 (b-2). The bar indicates 200 mm (b-1) and 80 mm (b-2).
was diagnosed on the basis of pathological findings. Control ovaries were collected during surgery from patients with other gynecologic diseases, such as cervical can- cer, but no evidence of endometriosis or any other ovarian pathology (n 5 16). Before surgery, all patients in both groups had experienced regular menstrual cycles and had not received hormonal therapy for at least 3 months before surgery. The ovarian tissues were fixed in Bouin fluid for 24 hours. Following fixation, the tissues were dehydrated in 70% ethanol and em- bedded in paraffin wax, and 5-mm serial sections were prepared. Paraffin-embedded sections were deparaffinized and then rehydrated in decreasing concentrations of ethanol.
Immunohistochemical studies
Immunohistochemical studies were conducted for key proteins in the PI3K- PTEN-Akt-Foxo3 pathway. The expres- sion of Foxo3, PTEN, phosphorylated mammalian target of rapamycin (pmTOR), conducted by observing hematoxylin and eosin–stained specimens of the whole ovary. Each follicle stage (primordial, primary, secondary, antral, and growing follicle) was morphologically classified by two independent researchers according to the method of Myers et al. (14).
Collection of human ovarian tissues and preparation of tissue sections
The institutional review board of the University of Tokyo approved the experimental procedures (registration number 0324-4), and all patients provided written informed consent. Clinical characteristics of the patients are shown in Table 1. Ovaries were obtained from patients with endometrioma that had undergone laparoscopic surgery (n 5 12). Endometriosis and phosphorylated Akt (pAkt) in oocytes of primordial fol- licles was analyzed. The following antibodies were used: monoclonal rabbit Foxo3a antibody (catalog no. 12829) at a 1:500 dilution; monoclonal rabbit PTEN antibody (catalog no. 9188) at a 1:100 dilution; monoclonal rabbit pmTOR antibody (catalog no. 2976) at a 1:100 dilution; and monoclonal rabbit pAkt antibody (catalog no. 4060) at a 1:100 dilution (all from Cell Signaling Technology, Danvers, MA). Rehydrated sections were treated with 0.3% hydrogen peroxide for 5 minutes to neutralize endogenous peroxidases and rinsed for 5 minutes with distilled water. Antigen retrieval was performed by using Target Retrieval (Dako Cytomation, Carpentaria, CA). After washing with PBS, slides were incubated with antibodies in a moist chamber with primary antibody at 4°C overnight. All sections were visualized by using DAB (Dako Cytomation) for
Table 1. Characteristics of Controls and Patients With Endometriosis
Characteristic Control (n 5 16) Endometriosis (n 5 12) P Value
Age, y 36.8 (31–40)a 39.0 (29–40)a NS
BMI, kg/m2 20.8 (18–24)a 22.7 (18–25)a NS
Endometriosis rASRM stage, n
I/II 0 0
III 0 2
IV 0 10
Endometrioma size, cm NA 5.8 (3–10)a
Disease, n
Cervical cancer 10 0
Endometrial cancer 4 0
Other gynecologic disease 2 0
Preoperative (.3 mo) hormonal treatment
Oral contraceptives 1 3
GnRH agonist 1 2
Abbreviations: BMI, body mass index; NA, not applicable; NS, not significant; rASRM, revised American Society for Reproductive Medicine.
aAverage (range).
substrate, followed by hematoxylin counterstaining and ana- lyzed under a light microscope.
Statistical analysis
Statistical analysis was conducted with the use of JMP software, version 11.0 (SAS Institute Inc. Cary, NC). The difference in number and proportion of stained primordial follicles was calculated by using nonparametric analysis (Mann- Whitney U test). P values , 0.05 were considered to indicate statistically significant differences.
Results
Number and proportion of follicles at each stage in control, endometriosis model, and cyclophosphamide-treated mice Initially, we evaluated the impact of endometriosis on the ovarian follicles by counting the follicles at each stage. The number of primordial follicles per ovary in the en- dometriosis model (mean 6 SEM, 704 6 70.1) and
cyclophosphamide-treated mice (472.5 6 48.9) was sig- nificantly lower than that in control mice (1103 6 100.4; P , 0.05 [Fig. 3(a)]. The proportion of primordial follicle numbers per total number of follicles in the endometriosis model (56.5% 6 1.6%) and cyclophosphamide-treated mice (45.6% 6 1.8%) was significantly lower in com- parison with that in control mice (73.3% 6 1.4%; P , 0.05) [Fig. 3(b)]. The proportion of primary, secondary, antral, and growing follicle numbers in the endometriosis model (17.9% 6 1.3%, 12.4% 6 1.6%, 11.9% 6 0.8%, and 1.2% 6 0.2%, respectively) and cyclophosphamide-treated mice (20.8% 6 1.4%, 17.1% 6 1.2%, 15.3% 6 0.9%, and 1.2% 6 0.2%) were higher in comparison with control mice (12.5% 6 0.8%, 6.7% 6 0.6%, 6.9% 6 0.6%, and 0.4% 6 0.1%) [Fig. 3(b)].
Correlation between number of primordial follicles and weight of endometriosis lesions in mouse model
We then clarified the correlation between the number of primordial follicles and the severity of endometriosis. As shown in Fig. 4, a statistically significant negative correlation (R2 5 0.2247; P , 0.005) was observed between the number of primordial follicles and the weight of endometriosis lesions in the mouse model.
Immunohistochemical studies for PI3K-PTEN-Akt-Foxo3 proteins in primordial follicles in mice
As shown in Fig. 5, key proteins in the PI3K-PTEN-Akt-Foxo3 pathway (Foxo3, PTEN, pmTOR, and pAkt) were assessed in endometriosis model mice and cyclophosphamide- treated mice (positive control). The proportion of Foxo3 and PTEN-positive primordial follicles was significantly lower in the endometriosis model (33.3% 6 0.58% and 29.7% 6 3.2) and cyclophosphamide-treated mice (24.2% 6 2.8% and 16.7% 6 2.1%) in comparison with control mice (82.0% 60.89% and 52.4% 6 6.2%; P , 0.05). The proportion of pmTOR- and pAkt-positive primordial follicles in endome- triosis model (47.6% 6 2.5% and 36.1% 6 4.3%) and cyclophosphamide-treated mice (50.0% 6 1.5% and 37.9% 6 5.8) was significantly higher (P , 0.05) in comparison with control mice (21.7%% 6 4.9% and 30.8% 6 2.9%; P , 0.05) (Fig. 6).
Figure 3. Number and percentage of primordial follicles in control, endometriosis model mice, and cyclophosphamide-treated mice. (a) The number of primordial follicles per ovary in the endometriosis group and cyclophosphamide group was significantly lower (P , 0.05) in comparison with the control. Bars and whiskers represent the mean and SEM. n 5 8 in each group. (b) The percentage of primordial follicle numbers per total follicle numbers in the endometriosis group and cyclophosphamide group was significantly lower (P , 0.05) in comparison with the control. Percentages of primary, secondary, antral, and growing follicles in the endometriosis group and cyclophosphamide group were higher in comparison with the control. Bars and whiskers represent the mean and SEM. *P , 0.05; n 5 8 in each group.
Immunohistochemical studies for PI3K-PTEN-Akt-Foxo3 proteins in primordial follicles in human ovaries
The activation of primordial follicles in human ovaries was analyzed by using immunohistochemical staining of key proteins in the PI3K-PTEN-Akt-Foxo3 pathway and counting of stained fol- licles. Figure 7(a)–7(d) depicts representative microscopic pictures of positively or negatively stained key proteins in the PI3K-PTEN-Akt-Foxo3 pathway in the nucleus of the oocytes in pri- mordial follicles in the human ovary. The expression of proteins in the ovaries of patients with endometriosis was compared with ovarian samples from women without endometriosis. As shown in Fig. 8, the percentage of
Figure 4. Correlation between number of primordial follicles and weight of endometriosis lesions in mice. A statistically significant negative correlation (R2 5 0.2247; P , 0.005) was noted in the relationship between the number of primordial follicles and the weight of endometriosis lesions in the mouse model. n 5 16.
Figure 6. Percentage of PI3K-PTEN-Akt-Foxo3 pathway protein–positive primordial follicles in control, endometriosis model mice, and cyclophosphamide-treated mice. The percentage
of Foxo3 and PTEN-positive primordial follicles was significantly lower (P , 0.05) in endometriosis model and cyclophosphamide- treated mice in comparison with the control. The percentage of pmTOR- and pAkt-positive primordial follicles in endometriosis model and cyclophosphamide-treated mice was significantly higher (P , 0.05) in comparison with the control. Bars and whiskers represent the mean and SEM. *P , 0.05; n 5 8 in each group.
Foxo3-positive primordial follicles was significantly lower in patient vs control samples (14.9% 6 5.1% vs 57.9% 6 4.7%; P , 0.05). There was no significant difference in PTEN-, pmTOR-, and pAkt-positive primordial follicles between the two groups (53.5% 6 13.1%, 29.5% 6 9.0%, and 45.2% 6 11.9% vs 67.7% 6 11.3%, 39.5% 6 10.7%, and 45.2% 6 11.9%, respectively).
Effect of AS101 on proportion of primordial follicles in control, endometriosis model, and cyclophosphamide-treated mice
AS101 is known to inhibit the PI3K-PTEN-Akt-Foxo3 pathway. The administration of AS101 to con- trol mice did not change the proportion of primordial follicles per total follicles (66.8% 6 1.2% and 60.4% 6 3.6%, control and AS101, respectively); however, AS101 significantly increased the proportion of primordial follicles in endometriosis model mice (54.9% 6 1.7% and 61.1% 6 1.2%) and cyclophosphamide-treated mice (41.7% 6 2.6% and 57.8% 6 1.4%) (P , 0.05) (Fig. 9).
Effect of AS101 on proportion of Foxo3-positive primordial follicles in endometriosis model mice
We then tested whether AS101 reduces the activation of primordial follicles in the endometriosis mouse model. The administration of AS101 to en- dometriosis model mice significantly increased the proportion of Foxo3-positive primordial follicles (35.5% 6 16.6% and 57.6% 6 8.8%, control and AS101, respectively; P , 0.05 (Fig. 10).
Figure 5. Immunohistochemical studies of primordial follicles in mouse ovary. [(a)–(d)] Representative microscopic pictures of positively or negatively stained key proteins in the PI3K-PTEN-Akt-Foxo3 pathway (Foxo3, PTEN, pmTOR, p-Akt) in the nucleus of the primordial follicle in the mouse ovary. In primordial follicles, phosphorylation of Akt and mTOR initiates activation of dormant follicles, whereas the induction of PTEN and activation (nuclear expression) of Foxo3 maintains dormancy.
Discussion
We found that in the mouse model of endometriosis used here, the number of primordial follicles was diminished but primary, secondary, antral, and growing follicle numbers increased. We then conducted immunohisto- chemical studies and discovered that in both mouse and human ovaries the PI3K-PTEN-Akt-Foxo3 pathway was activated in endometriosis samples. We further demonstrated that by inhibiting this pathway in our mouse model we were able to restore the
Figure 7. Immunohistochemical studies of primordial follicles in human ovary. [(a)–(d)] Representative microscopic pictures of positively or negatively stained key proteins in the PI3K-PTEN-Akt-Foxo3 pathway (Foxo3, PTEN, pmTOR, pAkt) in the nucleus of the primordial follicle in the human ovary. In primordial follicles, phosphorylation of Akt and mTOR initiates activation of dormant follicles, whereas the induction of PTEN and activation (nuclear expression) of Foxo3 maintains dormancy.
antral, and growing follicle numbers were greater, and this trend was similar to that seen in mice treated with cyclophospha- mide (positive control). These findings align with results whereby human ovaries with endometriomas contained signifi- cantly more morphologically atretic early follicles than controls (6). Taken together, these findings indicate it is plausible that the activation of dormant primordial follicles is accelerated in ovaries in en- dometriosis, resulting in the reduction of ovarian reserve. Interestingly, we also found that the weight of endometriosis lesions negatively correlated with the number of primordial follicles, suggesting that disease severity correlates with proportion of primordial follicle to control levels. These findings indicate that the endometriosis-related reduction of ovarian reserve is at least partially caused by the excessive activation of primordial follicles and this activation is mediated by the PI3K-PTEN-Akt- Foxo3 pathway.
Initially, we demonstrated that in the endometriosis mouse model, the number of primordial follicles was significantly diminished, whereas primary, secondary, accelerated activation of dormant follicles. In this context, it is important to note that the localization of endometriosis lesions in our mouse model was not in the ovaries (endometrioma) but rather in the peritoneum, a finding that contrasts with the above- mentioned data in patients with endometrioma. To- gether with recent data showing that endometriosis that has developed in sites other than the ovary can also reduce the ovarian reserve (4), endometriosis that develops in any location in the peritoneal cavity may have the same effect on the ovary.
We then analyzed the activity status of the PI3K-PTEN-Akt-Foxo3 path- way in primordial follicles in the ovary from endometriosis model mice. In primordial follicles, phosphorylation of Akt and mTOR initiates activation of dormant follicles, whereas the in- duction of PTEN and activation (nu- clear expression) of Foxo3 maintains dormancy (10, 11). Further, cyclo- phosphamide activates the quiescent primordial follicle in mice via activa- tion of the PI3K-PTEN-Akt-Foxo3 pathway, resulting in a reduction in ovarian reserve (13, 15). In the cur- rent study, we found that, similar to cyclophosphamide-treated mice, ova- ries from endometriosis model mice
Figure 8. Percentage of PI3K-PTEN-Akt-Foxo3 pathway protein–positive primordial follicles in control women (Ct) and patients with endometriosis (Pt). The percentage of Foxo3-positive primordial follicles was significantly lower (P , 0.05) in samples from patients with endometriosis compared with samples from control women. There was no significant difference in PTEN-, pmTOR-, and pAkt-positive primordial follicles between the two groups. Boxes represent the first (25%) and third (75%) quartiles, and horizontal lines in the boxes represent medians. Whiskers represent the 10th and 90th percentiles. *P , 0.05; n 5 16 Ct and 12 Pt.
had a significantly lower proportion of Foxo3 and PTEN-positive primordial follicles and a higher proportion of pmTOR- and pAkt-positive primordial follicles in comparison with controls. These findings indicate that activation
Figure 9. Effect of AS101 (AS) on percentage of primordial follicles in control, endometriosis model, and cyclophosphamide-treated mice. The administration of AS101 to control mice did not change the proportion of primordial follicles per total follicles, but the percentage of primordial follicles significantly increased (P , 0.05) in samples from endometriosis model and cyclophosphamide- treated mice. Bars and whiskers represent the mean and SEM.
of dormant primordial follicles in endometriosis model mice is caused by activation of the PI3K-PTEN-Akt- Foxo3 pathway.
We further evaluated the activity status of the PI3K- PTEN-Akt-Foxo3 pathway in primordial follicles in ovaries from patients with endometriosis. Our study analyzed the status of PI3K-PTEN-Akt-Foxo3 signaling using immunohistochemistry in human ovarian tissues. As shown in Fig. 6, the immunohistochemical study clearly distinguishes between Foxo3- and PTEN-, pmTOR-, and pAkt-positive and -negative expression in oocytes of human primordial follicles. In corroboration with analysis of our endometriosis model mice, the proportion of Foxo3-positive primordial follicles was significantly lower in ovaries from patients with endo- metriosis. In contrast, the percentage of PTEN-, pmTOR-,
Figure 10. Effect of AS101 on the percentage of Foxo3-positive primordial follicles in endometriosis model mice. The administration of AS101 to endometriosis model mice significantly increased the percentage of Foxo3-positive primordial follicles (P , 0.05). Bars and whiskers represent the mean and SEM. *P , 0.05; n 5 8 in each group.
and pAkt-positive primordial follicles did not significantly differ between groups. The lack of difference may be due to the small sample size; the activity of alternative reg- ulatory mechanisms, such as AMP-activated protein kinase/SIRT1 (16); or autophagy (17), which involves expression of Foxo3. The technical limitations in the immunohistochemical study for detecting PTEN, pmTOR, and pAkt expression may also affect the accuracy of the results.
Finally, we demonstrated that by inhibiting the PI3K- PTEN-Akt pathway, we restored the proportion of pri- mordial follicles in endometriosis model mice to control levels. Previously, Kalich-Philosoph et al. (13) showed that AS101, a PI3K-PTEN-Akt modulator, prevents follicle loss and preserves fertility in cyclophosphamide- treated mice. In the current study, we found that AS101 exerts similar effects in endometriosis model mice. In this context, it is interesting to note that the PI3K/PTEN/Akt pathway is also known to be involved in the pathogenesis of endometriosis per se (18) and that inhibitors of these signaling pathways have potential as effective treatments for endometriosis (19–21). Taken together, it is rea- sonable to propose that agents that attenuate the path- way, such as AS101, can be used to treat patients with endometriosis with the aim to control the disease and to preserve fertility. Up to this time, neither AS101 nor inhibitors of PI3K, Akt, and mTOR have been tested in human endometriosis or infertility; however, some of these agents have been tested in clinical trials or have been approved for clinical use, mainly for the treatment of patients with cancer (22). Further studies, in specific clinical studies, are warranted to test whether these agents can be clinically applied for the management of patients with endometriosis or its associated ovarian insufficiency.
The current study has strengths and the weaknesses. One of the strengths is that the phenotype of the ovarian pathology in our endometriosis mouse model very closely mimics that of patients with endometriosis, and therefore the model can be used for further experiments, such as drug discovery. In contrast, the endometriosis lesion in our mouse model does not perfectly mimic the human endometriosis: The main lesion is peritoneal and ovarian and deep infiltrating lesions are rare; this is the weakness of this model. Further studies, such as the analysis of ovaries from patients with peritoneal or deep infiltrating endometriosis, are warranted to evaluate the effect of endometriosis outside the ovary on the ovarian reserve. In summary, the current study highlights the role of the PI3K-PTEN-Akt-Foxo3 pathway in the excessive activation of primordial follicles and thereby the reduction of the ovarian reserve associated with endometriosis. This study indicates that the PI3K-PTEN-Akt-Foxo3 pathway inhibitors can prevent ovarian reserve decline and should be considered for further investigation as promising medicines for the prevention of the ovarian reserve reduction in patients with endometriosis.
Acknowledgments
The authors thank medical colleagues at the University of Tokyo Hospital for collecting clinical samples and Dr. Kate Hale for editing the manuscript. Financial Support: This work was supported by grants from the Japan Agency for Medical Research and Development (K.K.); the Ministry of Health, Labour, and Welfare; the Ministry of Education, Culture, Sports, Science, and Tech- nology; the Yamaguchi Endocrine Research Foundation (K.K.); the Shiseido Female Researcher Science Grant; the Society for Women’s Health Science Research (K.K.); the Kanzawa Medical Research Foundation (K.K.); and the Japan Society for the Promotion of Science (K.K.), and Shiseido Group (K.K.).
Additional Information
Correspondence and Reprint Requests: Kaori Koga, MD, PhD, Obstetrics and Gynecology, University of Tokyo, 7-3-1 Hongo Bunkyo Tokyo 113-8655, Japan. E-mail: kawotan-tky@ umin.ac.jp.
Disclosure Statement: The authors have nothing to disclose.
Data Availability: All data generated or analyzed during this study are included in this published article or in the data repositories listed in References.
References and Notes
1. Zondervan KT, Becker CM, Koga K, Missmer SA, Taylor RN, Vigano` P. Endometriosis. Nat Rev Dis Primers. 2018;4(1):9.
2. Muzii L, Di Tucci C, Di Feliciantonio M, Galati G, Di Donato V, Musella A, Palaia I, Panici PB. Antimu¨ llerian hormone is reduced in the presence of ovarian endometriomas: a systematic review and meta-analysis. Fertil Steril. 2018;110(5):932–940.e1.
3. Kasapoglu I, Ata B, Uyaniklar O, Seyhan A, Orhan A, Yildiz Oguz S, Uncu G. Endometrioma-related reduction in ovarian reserve (ERROR): a prospective longitudinal study. Fertil Steril. 2018; 110(1):122–127.
4. Ashrafi M, Arabipoor A, Hemat M, Salman-Yazdi R. The impact of the localisation of endometriosis lesions on ovarian reserve and assisted reproduction techniques outcomes. J Obstet Gynaecol. 201939(1):91-797.
5. Dolmans MM, Martinez-Madrid B, Gadisseux E, Guiot Y, Yuan WY, Torre A, Camboni A, Van Langendonckt A, Donnez J. Short- term transplantation of isolated human ovarian follicles and cor- tical tissue into nude mice. Reproduction. 2007;134(2):253–262.
6. Kitajima M, Dolmans MM, Donnez O, Masuzaki H, Soares M, Donnez J. Enhanced follicular recruitment and atresia in cortex derived from ovaries with endometriomas. Fertil Steril. 2014; 101(4):1031–1037.
7. Castrillon DH, Miao L, Kollipara R, Horner JW, DePinho RA. Suppression of ovarian follicle activation in mice by the tran- scription factor Foxo3a. Science. 2003;301(5630):215–218.
8. Liu L, Rajareddy S, Reddy P, Du C, Jagarlamudi K, Shen Y, Gunnarsson D, Selstam G, Boman K, Liu K. Infertility caused by retardation of follicular development in mice with oocyte-specific expression of Foxo3a. Development. 2007;134(1):199–209.
9. Pelosi E, Omari S, Michel M, Ding J, Amano T, Forabosco A, Schlessinger D, Ottolenghi C. Constitutively active Foxo3 in oo- cytes preserves ovarian reserve in mice. Nat Commun. 2013;4(1): 1843.
10. Adhikari D, Liu K. mTOR signaling in the control of activation of primordial follicles. Cell Cycle. 2010;9(9):1673–1674.
11. Jagarlamudi K, Rajkovic A. Oogenesis: transcriptional regulators and mouse models. Mol Cell Endocrinol. 2012;356(1-2):31–39.
12. Takamura M, Koga K, Izumi G, Urata Y, Nagai M, Hasegawa A, Harada M, Hirata T, Hirota Y, Wada-Hiraike O, Fujii T, Osuga Y. Neutrophil depletion reduces endometriotic lesion formation in mice. Am J Reprod Immunol. 2016;76(3):193–198.
13. Kalich-Philosoph L, Roness H, Carmely A, Fishel-Bartal M, Ligumsky H, Paglin S, Wolf I, Kanety H, Sredni B, Meirow D. Cyclophosphamide triggers follicle activation and “burnout”; AS101 prevents follicle loss and preserves fertility. Sci Transl Med. 2013;5(185):185ra62.
14. Myers M, Britt KL, Wreford NG, Ebling FJ, Kerr JB. Methods for quantifying follicular numbers within the mouse ovary. Repro- duction. 2004;127(5):569–580.
15. Chang EM, Lim E, Yoon S, Jeong K, Bae S, Lee DR, Yoon TK, ChoiY, Lee WS. Cisplatin Induces Overactivation of the Dormant Primordial Follicle through PTEN/AKT/FOXO3a Pathway which Leads to Loss of Ovarian Reserve in Mice. PLoS One. 2015;10(12): e0144245.
16. Canto´ C, Gerhart-Hines Z, Feige JN, Lagouge M, Noriega L, Milne JC, Elliott PJ, Puigserver P, Auwerx J. AMPK regulates energy expenditure by modulating NAD1 metabolism and SIRT1 activity. Nature. 2009;458(7241):1056–1060.
17. Sonigo C, Beau I, Grynberg M, Binart N. AMH prevents primordial ovarian follicle loss and fertility alteration in cyclophosphamide- treated mice. FASEB J. 2019;33(1):1278–1287.
18. McKinnon BD, Kocbek V, Nirgianakis K, Bersinger NA, Mueller MD. Kinase signalling pathways in endometriosis: potential targets for non-hormonal therapeutics. Hum Reprod Update. 2016;22(3): 382–403.
19. Leconte M, Nicco C, Ngoˆ C, Che´reau C, Chouzenoux S, Marut W, Guibourdenche J, Arkwright S, Weill B, Chapron C, Dousset B, Batteux F. The mTOR/AKT inhibitor temsirolimus prevents deep infiltrating endometriosis in mice. Am J Pathol. 2011;179(2):880–889.
20. Matsuzaki S, Pouly JL, Canis M. In vitro and in vivo effects of MK2206 and chloroquine combination therapy on endometriosis: autophagy may be required for regrowth of endometriosis. Br J Pharmacol. 2018;175(10):1637–1653.
21. Barra F, Ferro Desideri L, Ferrero S. Inhibition of PI3K/AKT/mTOR pathway for the treatment of endometriosis. Br J Phar- macol. 2018;175(17):3626–3627.
22. Esposito A, Viale G, Curigliano G. Safety, tolerability, and management of toxic effects of phosphatidylinositol 3-kinase inhibitor treatment in patients with cancer: a review [published online ahead of print 28 March 2019]. JAMA Oncol. doi: 10.1001/jamaoncol. 2019.0034.