Immunohistochemical expression of hypoxia-inducible factor-1α in stromal cells of vaginal tissue in post-menopausal women with pelvic organ prolapse.

There were no significant differences in body mass index, age, duration of menopause, parity or number of abortions between women with and without POP ( Table ). Women in both the groups went through menopause at least four years earlier. Immunohistochemical analysis showed a significant increase in the number of HIF-1α-positive stromal cell nuclei counted per 100 stromal cells in the vaginal wall of the women with prolapse [7.33±1.26, median=7, interquartile range (IR)=2] in comparison to the control group (2.12±0.99, median=2, IR=2, P < 0.001) (Figs 1 and 2 ). For the variable HIF-1α, the post hoc compute achieved power was calculated, and it was 1 (effect size d=4.598123). Post hoc achieved power for other variables ranged between 0.07 and 0.17. Characteristics of the study participants (n=120) Vagina, haematoxylin-eosin staining, ×400. E, stratified squamous epithelium; LP, lamina propria. Vagina, hypoxia-inducible factor-1α protein detection by immunohistochemistry - diaminobenzidine chromogen and haematoxylin counterstained, ×400. Nuclear staining within positive stromal cells (arrow).

Patient characteristics and tissue collection : Vaginal wall tissue samples were obtained during surgeries from 120 post-menopausal women. There were 60 women each in POP and control groups. Control group included women with other benign gynaecological diseases such as ovarian cysts and uterine myomas. Biopsy samples of 1 cm × 1 cm of the anterior vaginal wall tissue were taken from the same part of the wall in each case i.e . from the part right next to the connection between the anterior vaginal wall and the cervix. POP was assessed using POP-quantification (POP-Q) score 18 . In this study, women with the POP-Q 4 (complete eversion) were included in the POP group, while women with the POP-Q 0 were included in the control group. All women included in the study were multiparous. Women with any additional diseases, such as diabetes, malignant diseases, pelvic inflammatory disease and endometriosis were excluded from the study. None of the women has ever used hormone replacement therapy or has been smoking cigarettes. Biopsy samples were obtained during vaginal hysterectomy for POP group and abdominal hysterectomy for control group at the department of Gynecology and Obstetrics, University Hospital in Split, Croatia. Each patient has signed written informed consent to participate in the study. Convenience sample criterion of selection was used when choosing our POP and the control groups. The RECORD (REporting of studies Conducted using Observational Routinely- collected health Data) statement guidelines were followed during the study 19 . Power of the study and effect size were calculated using G*Power statistical power analysis program (version 3.1.9.2., Heinrich Heine University Düsseldorf, Germany). Power of the study was calculated for all variables and effect size for the variables where null hypothesis was rejected. The study was approved by the Research Ethics Committee of University Hospital Center, Split. This study was a part of a larger project which included a whole series of researches pertaining to pelvic floor damage. The study was conducted between 2009 and 2014. Immunohistochemistry : Tissue samples were fixed in 10 per cent buffered formalin and processed through standard processes in the automatic tissue processor (Shandon Excelsior, Thermo Fisher Scientific, United Kingdom), embedded in paraffin, cut at 4 μm and placed on positive charged slides (Superfrost Plus Adhesion Slides, Thermo Scientific). Immunohistochemistry was performed using the BenchMark ULTRA Automated IHC/ISH slide staining system (Ventana Medical Systems, Inc., USA), using horseradish peroxidase detection system. After tissue deparaffinization for 10 min at 72°C, slides were pretreated with Tris-based buffer for 52 min at 95°C and incubated with 3 per cent H 2 O 2 for four minutes at 36°C to inactivate endogenous peroxidase. Slides were incubated with primary antibody for HIF-1α (SC-10790, Santa Cruz Biotechnology, Dallas, USA) for 92 min at 37°C. Reaction was visualized with a Olympus BX41 Microscope using diaminobenzidine (DAB) chromogen and counterstained with haematoxylin (ULTRAVIEW Universal DAB Detection Kit, Ventana Medical Systems, Inc.). All washes between the various steps were done with phosphate-buffered saline solution. The same immunohistochemical protocol was followed for the negative controls, with the omission of the primary antibodies. Human colon cancer tissues were used as positive controls. HIF-1α expression was scored by counting 100 stromal cells nuclei in the representative slides of the vaginal wall using Olympus Image Analyzer (magnification ×200). Counting was performed at the hot spots in the fields with the most prominent immunohistochemical reaction. Data were expressed in the form of the number of positive stromal cells nuclei/total number of stromal cell nuclei in the area scored. All immunohistochemical slides were evaluated by two investigators who were blinded to the study group. Statistical analysis : Data were tested for normality of distribution using the Kolmogorov-Smirnov test. As data were not normally distributed (data not shown), non-parametric Mann–Whitney test was used. All analyses were conducted using SPSS (version 18; SPSS Inc., Chicago, IL, USA), with the significance level set at P <0.05.

The results obtained showed a significant difference in expression of HIF-1α transcription factor in response to oxidative stress between women with POP and those in the control group. This has led us to assumption that in the control group, stromal cells of the vaginal tissue are exposed to normoxic conditions. It is evident that HIF-1α, as it binds to the promoter region, can activate a number of genes and thus starts a series of important cell events. Moreover, oxidative stress can inhibit protein synthesis and contribute to muscle atrophy 20 . Thurmond et al 21 have reported that there are structural changes in the prostate of older men in response to hypoxia. Tehrani et al 22 have compiled the screening of POP without a physical examination. Their questionnaire-based study examined urinary incontinence following laughing, sneezing or coughing, urinary urgency, feeling pain during defecation and feeling or seeing a bulge in the vagina. Analyzing pelvic floor distress inventory questionnaire, Barber et al 23 have singled out the question: ‘Do you usually have a bulge or something that you can see or feel falling out in your vaginal area?’ as the most significant indicator of the disorder. Tan et al 24 composed specific questions related to prolapse, which included the questions on urinary splinting, digital assistance for defecation and bulge per vagina. Other investigators 25 26 have also created their questionnaires for POP screening. All these questionnaires have been of significance in pelvic floor evaluation of healthy patients without clinical examination. Those at risk would be the women who will develop POP and those with an increased risk for developing a pelvic floor defect in the future. Patients with an increased risk could be recognized on time and could undergo prevention or therapy. The wide spectrum of cell processes affected by HIF-1α suggests that it could be clinically significant. Evidence suggests that crosstalk between HIF1-α and aryl hydrocarbon receptor (AhR) modulates the immune response to different signals such as immunological, metabolic and environmental 27 . For example, tetrachlorodibenzo-p-dioxin (TCDD)-AhR binding acts as a trigger for signalling pathways which lead to impairment of endometrial function 28 . Although complex, interference between the xenobiotic- and hypoxia-sensing pathways has been elucidated in the study of biphasic role of nuclear receptor coactivator 2 between AhR and hypoxic conditions 29 . Various therapeutic processes reducing oxidative stress in cells could significantly contribute to tissue regeneration. One could argue that such preventive processes in cells could improve the quality of the vaginal cavity, thus reducing the risk of POP. Our study had certain limitations. The parts of vaginal floor from where the samples were taken could be a limitation of this study, given the fact that in the control group vaginal floor was healthy, while in women with POP, there were notable defects. The expression of HIF-1α could vary in different parts of vaginal floor in women with POP, depending on the length of hypoxia in that exact part. The strength of our study was the strict criteria used for the inclusion of patients. Considering the power of the study, although the sample in this study was convenience sample, the only significant variable in the study, HIF-1α showed both strong effect size and high power. The other variables showed no significant difference, thus showing that the POP was associated with expression of HIF-1α. The results of this study showed that HIF-1α expression was related to POP, and in the future studies the other variables influencing the prolapse would also be examined.