A high dose of total recombinant FSH suppresses granulosa cell apoptosis and maintains oocyte quality in endometriosis : A cross-sectional study

Endometriosis is one of the most common conditions causing Background: infertility and an indication to undergo fertilization (IVF). High apoptosis in vitro rate and oxidative stress in patients with endometriosis are believed to negatively affect the IVF success rate. However, there have been conflicting results on the effect of endometriosis on IVF success, and there have been limited studies that directly assess endometriosis and its effect on oocyte quality. This study was performed to explore the correlation between mRNA BAX/BCL-2 expression and oocyte quality in endometriosis compared to non-endometriosis subjects. This was a cross-sectional study. 15 endometriosis and 15 Methods: non-endometriosis subjects were recruited through convenience sampling at Cipto Mangunkusumo Hospital, Jakarta. All subjects underwent follicle stimulation with recombinant follicle-stimulating hormone (FSH). Granulosa cells were collected and tested for BAX and BCL-2 expression and the results were compared to the oocyte quality and fertilization rate of the patients. The total dose of recombinant FSH received by the endometriosis Results: group was significantly higher compared with that of the non-endometriosis group (p = 0.005). There was a difference in BAX level (p = 0.029) and BCL-2 level (p<0.001) between groups. However, the BAX/BCL-2 ratio did not differ significantly (p = 0.787) between groups. No significant correlation was found between the BAX/BCL-2 ratio and any of the oocyte quality parameters measured. We found that there is a significantly higher dose in total dose Conclusion: recombinant FSH received by the endometriosis group compared with the non-endometriosis group. We also found that there was no significant 1-3 1 3 3

difference in BAX/BCL-2 ratio between the endometriosis and non-endometriosis groups.

Keywords
Apoptosis, BAX/BCL-2 ratio, Endometriosis, Oocyte Quality, r-FSH Financial support research grant from University of Indonesia-Dr Cipto Mangunkusumo General Hospital, Jakarta and Grant information: Indonesia.

Introduction
Endometriosis is a condition in which endometrial tissues and glands are found outside the uterine cavity. This condition might cause pelvic pain and infertility. Ectopic endometrial tissue causes chronic inflammation, widespread fibrotic change, and adhesion 1,2 . Endometriosis occurs in 25-40% of women with infertility, and 30-50% of women with endometriosis also have infertility 3,4 . Endometriosis has been found to cause folliculogenesis and oocyte maturation disturbances, an increase in oxidative stress, and an imbalance in inflammatory cytokines that result in infertility 5-8 . In vitro fertilization (IVF) is one of the treatment options for patients with endometriosis. In 2016, in Indonesia, there were 8152 IVF cycles were performed, 7.03% of them were in women with endometriosis 9 .
Studies on endometriosis affecting IVF success rate have shown conflicting results 10 . Some meta-analyses report that endometriosis affects the rate of pregnancy, miscarriage, and live birth during IVF. However, there has not been any adequate study that evaluates oocyte quality as an outcome of endometriosis. Oocyte quality is defined as the oocyte's ability to undergo maturation and fertilization 11 . Some studies that evaluated oocytes quality from endometriosis patients found that oocyte quality were poorer regardless of sperm quality and uterine cavity condition 12,13 . Increased apoptosis rates have been found in endometriosis ovaries, and this has been linked to decreased oocyte quality 14 . Studies on apoptosis show a difference between endometriosis ovaries and non-endometriosis ovaries due to increased oxidative stress 15 . This condition triggers apoptosis through intrinsic and extrinsic pathways as well as meiotic spindle dysfunction and direct destruction through lipid peroxidase 16,17 . Apoptosis is regulated by specific genes that code protein and initiate the whole process. B-cell lymphoma/leukemia 2 (BCL-2) plays a crucial role in regulating apoptosis. This protein family consists of two categories: apoptosis inhibitors, BCL-2 and BCL-2-like 1 (BCL-2-LI or BCL-XL); as well as an apoptosis trigger, BCL-2-associated X protein (BAX) 18 . This study aimed to measure the gene expression of these apoptosis regulators, BCL-2 and BAX, and evaluate their correlation with oocyte quality in endometriosis and non-endometriosis IVF patients. The proposed conclusion was that there was significant difference in BAX/BCL-2 mRNA expression in patients with endometriosis and non-endometriosis.

Sample selection
Subjects were selected through convenience sampling. In total, 30 women were recruited for the study, who were undergoing IVF treatment at Yasmin Kencana Clinic, Cipto Mangunkusumo Hospital, Jakarta. Sample recruitment was done in June 2016 to August 2017.
All subjects found to positively have endometriosis, determined from a clinical examination and imaging procedures done by an obstetrician/gynecologist, were included in this study as a case group. However, subjects found to have other ovarian disturbances, history of ovarian removal, smoking, alcohol consumption, and sperm factor were excluded. For control group, we included all subjects with ovarian disturbance other than endometriosis and excluded subjects with male infertility factor.
Granulosa cell sampling was conducted in the IVF Laboratory, while RNA isolation and quantitative real-time polymerase chain reaction (PCR) were done in the Integrated Laboratory Faculty of Medicine, Universitas Indonesia, between August and October 2017.
This study was conducted according to ethical standards in the Declaration of Helsinki and approved by the Faculty of Medicine Universitas Indonesia-Dr. Cipto Mangunkusumo General Hospital Research Ethical Committee with approval number LB.02.01/X.2/871/2017. All subjects in this study had the research clearly explained to them and have signed informed consent.
Data collection RNA and DNA extraction from granulosa cells. Follicular fluid and granulosa cells were taken during ovum pickup, as part of routine IVF treatment. Granulosa cells were obtained during follicular aspiration and separated from the oocyte. These cells were then kept in a tube containing 500 uL RNA Later solution and stored at a temperature of -80°C.
RNA was isolated from granulosa cells using the QIAamp RNA Blood Mini Kit (QIAGEN) according to the modified QIAamp RNA Blood Mini Handbook 19 . Buffer RLT and betamercaptoethanol were set with a 600 uL:6 uL ratio for each ample. Briefly, the granulosa cell samples in RNA Later solution were thawed and centrifuged at 20°C for 5 minutes (8000 × g). The supernatant was discarded, and RLT buffer mixed with beta-mercaptoethanol was added and homogenized. The sample was moved to a QIAshredder spin column and centrifuged for 8 minutes at 20°C (8000 × g). 600 uL Ethanol 70% was added to flow-through liquid, homogenized, and moved into QIAmp spin column. The QIAamp spin column was then centrifuged for 15 seconds (16.000 × g). The flow-through was then discarded and the column was washed using 600 uL RW1 buffer, centrifuged for 15 seconds (16.000 × g), and the flow-through was discarded. 500 uL RPE buffer was then added into the column for final washing, centrifuged for 15 seconds (16.000 × g), and the flow-through was discarded. This step was repeated twice. 25 uL RNase-free water was then added into the column as an elution solution to bind the RNA and the column was incubated for 10 minutes. The column then underwent centrifuge for 1 minute (16.000 × g) and the RNA collected at the collection tube under QIAamp spin column. The RNA concentration was measured with Nano Drop and stored at -80°C. cDNA was synthesized from RNA using the Qiagen Quantitect Reverse Transcription Kit according to the modified Quantitect Reverse Transcription Mini Handbook protocol 20 . The RNA template was mixed with RNase-free water with certain ratio according to RNA concentration obtained to gain equal cDNA concentration for all samples. This RNA and RNase-free water mix will obtain total volume of 14 uL. gDNA Wipeout buffer was then added into the mix and then incubated at 42°C for 10 minutes. Into each sample was then added Reverse-transcription master mix 1 uL, Quantiscript RT buffer 4 uL, and RT Primer mix 1 uL. the sample was then incubate at 42°C for 15 minutes and 95°C for 3 minutes. cDNA samples were then stored at -20°C until ready to be used.
Real-time PCR for BAX and BCL-2 expression Quantitative real-time PCR was conducted using Qiagen Quantitect SYBR Green PCR Kit according to the modified Quantitect SYBR Green PCR Mini Handbook protocol 21 . qPCR was done by using absolute quantification, which measures sample by referring to standard curve. Standard curve was constructed using gBlocks oligonucleotide which made from amplicon product from each BAX and BCL-2 primer. The quantitative real-time PCR was performed using primers in Table 1. PCR master mix was firstly prepared and contain of 12.5 uL SYBR Green PCR mix, 0.25 uL forward primer, 0.25 uL reverse primer, and 10 uL nuclease-free water for each sample. After homogenized, the cDNA samples were then added into tube containing PCR mix and placed in real-time PCR instrument (Qiagen Rotor Gene-Q real-time PCR). Thermal profile of the instrument was set as shown in Table 2, while annealing temperature was set according do melting temperature in Table 1 for each gene, 58°C for BAX and 60°C for BCL-2.
Construction of standard curve was performed using the same PCR master mix composition, except for the cDNA samples. The cDNA was replaced by gBlocks oligonucleotide which has known concentration and diluted 5 times. The concentration of each gBlocks used were 1 until 1 × 10-8 ng/uL. Each gBlocks concentration was run duplicate to obtain standard curve with coefficient of determination value close to 1.
The standard curve was then used to measure each sample.
Oocyte quality assessment. Oocyte quality was measured by assessing morphology, maturation index, and fertilization rate. This assessment was performed under an Inverted Microscope (Olympus). Oocyte morphology was classified with Xia criteria, and mean score was calculated from the total number of oocytes obtained per subject 22 . The maturation index was defined as percentage of oocytes in metaphase II, whereas the fertilization rate was defined as percentage of metaphase II oocytes which successfully fertilized.

Data analysis
During the stimulation phase, subjects were exposed to recombinant FSH, and the total dose of FSH might influence the result of the study; therefore, the total dose of FSH provided to the participants was noted.
BAX/BCL-2 ratio was the independent variable, whereas the oocyte quality was the dependent variable. Statistical analysis was conducted with IBM SPSS (Statistical Package for Social Sciences) version 22, to test distribution and comparison between the endometriosis group and the control group for each variable. A correlation test was done to test any association between BAX/BCL-2 ratio and oocyte quality indicators in both groups. Variables with normally distributed data were tested with an independent t test, whereas other variables were tested with the Mann-Whitney test.

Results
A total of 30 subjects took part in this study: endometriosis group (n = 15) and control group (n = 15). Characteristics of these subjects are in Table 3. The patients' age in both groups showed a normal distribution, with a mean age of 33.27±4.448 for endometriosis group and 32.67±3.559 for non-endometriosis group. Total dose of recombinant FSH received was calculated and there was a significant difference between the groups; mean dose of FSH received by the endometriosis group was 3760.0±1054.15 while the non-endometriosis group was 2763.3±700.82 (p=0.005). The total number of oocytes obtained during ovum pickup was normally distributed in both groups ( Table 3). The mean number of oocytes was lower in the endometriosis group, although the difference was not significant.
Gene expression of BAX and BCL-2 were significantly different between the groups, with the concentrations of both genes being lower in the endometriosis group (Table 4). However, there was no statistically significant difference between the BAX/BCL-2 ratio between groups (p=0.787).
Oocyte quality was measured with three indicators: morphology, maturation index, and fertilization rate. We found no oocyte quality difference in the endometriosis group compared with the non-endometriosis group (Table 4).

Discussion
BAX and BCL-2 are two proteins that play a crucial role in cell apoptosis through the intrinsic pathway. BAX is an apoptosis initiator, whereas BCL-2 is anti-apoptotic. A previous study by Tommi et al. found that the ratio between BAX and BCL-2 is one of the important indicators of apoptosis activity in cells 23 .
The increases in BAX/BCL-2 ratio correspond to increased apoptosis activity and vice versa.
In this study, we found significant differences in BAX and BCL-2 mRNA concentrations in both groups. The expression of both genes was lower in the endometriosis group compared with the non-endometriosis group (p = 0.029 in BAX and p<0.001 in BCL-2). However, there was no difference in the BAX/BCL-2 mRNA ratio between groups (p = 0.787). This result showed that there is no difference in apoptosis activity in endometriosis granulosa cells compared with non-endometriosis cells. This finding is inconsistent with a previous study by Wiweko et al, which stated that BAX mRNA concentration in endometriosis granulosa cells is higher than in the control group 24 . However, the lower level of BAX mRNA does not imply less apoptotic activity. One interaction model of BAX and BCL-2 protein showed that while BAX induces apoptosis, BCL-2 inhibits the process 25 . In this case, the low level of BCL-2 in endometriosis might result in decreased anti-apoptotic activity. There was also a study that showed fewer follicles in rats with BCL-2 deficiency, and that higher BCL-2 expression would suppress apoptosis and atresia 26 .
Filali et al. reported the essential role of BCL-2 mRNA in granulosa cells in determining oocyte quality. BCL-2 mRNA expression was found to be higher in the granulosa cells of mature oocytes, whereas there was no difference in BAX mRNA in the same cells. Moreover, BCL-2 mRNA correlates with the fertilization rate due to the lower apoptosis activity 27 . BAX/BCL-2 ratio has a more important role as an apoptosis marker, as mentioned earlier.
However, in this study, we found no significant difference in apoptosis activity. This finding is inconsistent with a previous In the present study, the oocyte number in the endometriosis group was lower than in the non-endometriosis group, but was not significant (p = 0.336). This finding is in line with studies by Rossi et al. and Yang et al. that showed a smaller number of oocytes in endometriosis subjects 29,30 . Regarding oocyte quality, we found no differences in the mean oocyte score, maturation index, and fertilization rate between the groups. The mean oocyte score showed no significant difference (p = 0.611). The oocyte score was assessed by Xia criteria, which analyzes oocyte morphology based on perivitelline space, polar body II, and cytoplasm. This result is inconsistent with Shebl et al, who reported that fewer normal oocytes were observed in patients with endometriosis compared with control subjects 31 . Similarly, the maturation index and fertilization rate in both groups did not show a marked difference (p=0.225 and p=0.693, respectively). This is inconsistent with other related studies that showed maturation index and fertilization rate in endometriosis were poor; Rossi et al. In this study, there was no correlation between BAX/BCL-2 mRNA ratio and mean oocyte score, maturation index, and fertilization rate in both groups. This result suggests that similar apoptosis activity will result in similar oocyte quality. Although we found no difference in BAX/BCL-2 mRNA ratio and oocyte quality, there was a significant discrepancy in recombinant FSH dose given to each group (p=0.005). The mean FSH dose in the endomteriosis group was 3760 IU, whereas the control group 2763.3 received IU. Shen et al. reported that FSH might play a role in protecting granulosa cells from oxidative stress, although the mechanism is currently unknown. FSH also protects mitochondria in granulosa cells undergoing oxidative stress 34 . Our study shows that subjects with endometriosis may benefit from higher dose of recombinant FSH to produce oocytes with similar quality to those in the control group. Therefore, further studies on FSH and its correlation with apoptosis activity and oocyte quality during ovum pickup are needed.

Conclusions
Although we found lower levels of BAX mRNA and BCL-2 mRNA in patients with endometriosis, the ratio of these mRNAs was not statistically significant between groups. There was also no correlation between BAX/BCL-2 mRNA ratio and oocyte number, FSH dose, and oocyte quality in both groups. However, we noted a significantly higher dose of recombinant FSH given to the subjects with endometriosis compared with those in the non-endometriosis group. We suggest for future studies a similar protocol in collecting granulosa cells from size-matched follicles to minimize the discrepancy in BAX and BCL-2 mRNA concentrations. Moreover, examination and records on basal antral follicle and anti-müllerian hormone concentration should become a part of the standard IVF protocol.

Chii-ruey Tzeng
College of Medicine, Taipei Medical University Hospital, Taipei City, 110, Taiwan The manuscript entitled "A high dose of total recombinant FSH suppresses granulosa cell apoptosis and maintains oocyte quality in endometriosis: A cross-sectional study" -this study was conducted in 15 patients with endometriosis and 15 control patients without endometriosis. The draft is oversimplified with very preliminary results, and most data were published elsewhere. This study has a lack of novelty.
The sample size was too small to interpret the results with a proper power. The sample size estimation to get a proper power should be performed before conduction of the study.
Since the sample size is small, the baseline characteristics of patients with endometriosis should be described more clearly, such as BMI, duration of infertility, primary or secondary infertility, baseline FSH/LH/E2/P4, antral follicle count, AMH, or the revised American Fertility Society (rAFS) score, etc. Such information is basic and important. The stage of endometriosis could also be a confounding factor to affect the results.
The size of endometriosis of ovary, unilateral or bilateral should also be mentioned.
The title is interesting. However, their data didn't support their title. The gene expression of BAX and BCL-2 is only an association in patients with endometriosis. There was no ex-vivo functional assay, such as comparing the apoptosis activity in granulosa cells from patients with endometriosis under different doses of recombinant FSH, to support the title of manuscript. Thereafter, the title should be considered to be rewritten.
The conclusion section in the abstract should be rewritten. It was loosely organized and did not incorporate the title of this manuscript.
Whether this was a retrospective or prospective study according to the statement in the methods section could not be determined clearly. Please describe it clearly.
What are the diagnosis criteria of endometriosis in this manuscript, especially for clinical examination mentioned in the methods section? Besides, the grade of endometriosis should be illustrated.
I confirm that I have read this submission and believe that I have an appropriate level of expertise to state that I do not consider it to be of an acceptable scientific standard, for reasons outlined above.
The benefits of publishing with F1000Research: Your article is published within days, with no editorial bias You can publish traditional articles, null/negative results, case reports, data notes and more The peer review process is transparent and collaborative Your article is indexed in PubMed after passing peer review Dedicated customer support at every stage For pre-submission enquiries, contact research@f1000.com