Association of the immunohistochemical detection of gamma-glutamyl transferase expression with clinicopathological findings in postmenopausal women with endometrioid adenocarcinoma of the uterus.

Gamma-glutamyltransferase (GGT) is a transmembrane protein that plays an important role in glutathione (GSH) salvage and homeostasis, particularly at low cysteine concentrations (1, 2). GSH acts as an antioxidant and protects cells against oxidative stress, especially by detoxifying peroxides and free radicals. It has an important role in maintaining the intracellular redox balance. GGT is upregulated after acute oxidative stress and during pro-oxidant periods. The significant determinants of GGT expression and activity are reactive oxygen species (ROS) generated from the nicotinamide adenine dinucleotide phosphate (NADPH) oxidase system (3). GGTis expressed in many human tissues, predominantly on the luminal surface of secretory epithelial cells, particularly of the hepatobiliary tract, pancreas and kidneys (4). The uterus contains a mixture of GGT -positive and -negative glands in both the secretory and proliferative phase of the menstrual cycle. The fluid within the GGT-positive secretory glands is also GGT -positive (4). Germ cells, surface epithelium, and most stromal cells in the ovary are negative for GGT, while cilia on the epithelium of the Fallopian tubes stain positive for GGT. Cancer cells are characterised by higher endogenous reactive oxygen species production than normal, untransformed cells. They can compensate and benefit from such increased oxidative stress situations. Many human tumours express high levels of GGT. However, the distribution and concentration of GGT in JOURNAL OF PHYSIOLOGY AND PHARMACOLOGY 2016, 67, 3, 395-402 www.jpp.krakow.pl K. PITYNSKI1, T. OZIMEK1, N. GALUSZKA1, T. BANAS1, K. MILIAN-CIESIELSKA2, M. PIETRUS1, K. OKON2, M. MIKOS3, G. JUSZCZYK4, A. SINCZAK-KUTA2, A. STOJ2 ASSOCIATION OF THE IMMUNOHISTOCHEMICAL DETECTION OF GAMMA-GLUTAMYL TRANSFERASE EXPRESSION WITH CLINICOPATHOLOGICAL FINDINGS IN POSTMENOPAUSAL WOMEN WITH ENDOMETRIOID ADENOCARCINOMA OF THE UTERUS 1Department of Gynecology and Oncology, Jagiellonian University Medical College, Cracow, Poland; 2Department of Pathomorphology, Jagiellonian University Medical College, Cracow, Poland; 3Dietl Specialistic Hospital, Cracow, Poland; 4Department of Public Health, Medical University of Warsaw, Warsaw, Poland Gamma-glutamyl transferase (GGT) is a membrane enzyme present not only in the liver but also in healthy endometrial epithelium. Its overexpression has been demonstrated in numerous malignancies, where it exerts an anti-apoptotic effect and causes drug resistance in response to oxidation stress. Aim of the study was investigation of GGT expression in postmenopausal patients with endometrioid adenocarcinoma of the uterus (EAC). The material comprised 98 paraf fin- embedded post-operative tumour samples of EAC from postmenopausal patients and a control group of 60 normal human postmenopausal endometrium samples. For immunohistochemical specimen staining, polyclonal IgG anti-GGT was used; for GGT expression measurement, a semi-quantitative method was applied. In EAC patients, 16 (16.33%) were diagnosed as stage IA, 46 (46.93%) as stage IB, 14 (14.29%) as stage II, and 22 (22.45%) as stage IIIA-C, according to the International Federation of Gynaecology and Obstetrics (FIGO) classification. Fifty-six (57.14%) patients were diagnosed with low- or moderate-grade (G1-2) disease, and 42 (42.86%) were diagnosed with high-grade (G3) disease. Cytoplasmic GGT staining was confirmed in all samples, while apical membrane GGT staining was observed only in G1-2 EAC specimens and the control group. In G3 EAC specimens, GGT cytoplasmic staining and high nuclear polymorphism areas were predominantly shown. Comparable high GGT median apical expression was confirmed in healthy endometrium (2.0, S.E.M. = 0.28) and in G1-2 EAC (2.0, S.E.M. = 0.27); however, in G3 tumours, GGT expression was significantly lower (0.0, S.E.M. = 0.07) than in healthy endometrium (P < 0.001 and P < 0.001, respectively). After stratification of the cancer cases according to FIGO staging, the lowest median apical GGT expression levels were in II EAC (0.0, S.E.M. = 0.64) tumours compared with IA (4.0, S.E.M. = 0.47) tumours, specimen and normal endometrium (2.0, S.E.M. = 2.8) (P < 0001). Stage IB EAC and IIIA-C EAC (1.0, S.E.M. = 0.16) cases showed only moderate median apical expression of GGT (1.0, S.E.M. = 0.24). We concluded that impaired GGT expression has the potential to become a valuable tool for stratifying EEC patients’ prognosis and treatment planning. Key words: endometrioid adenocar cinoma, uter us, gamma-glutamyl transferase, immunohistochemistr y, postmenopausal women, cancer grade human tumours present several dif ferences from what is observed in normal tissues (5, 6). Moreover , the heterogeneous expression of GGT in different tumour types, and even different tumours of the same type, is observed (6). The enzyme has been suggested to protect against apoptosis and is upregulated after acute oxidative stress via Ras and several downstream signalling pathways (7, 8). There is widespread interest in the role of this enzyme in tumour formation, progression, invasion, and drug resistance. High serum levels of GGT were shown to be associated with inferior prognosis in many human cancers. Epidemiologic studies have indicated that elevated levels of γ-glutamyltransferase (GGT), a key enzyme of glutathione metabolism, might be associated with increased cancer risk. Recent experimental models have further elucidated the ability of cellular GGT to modulate crucial redox-sensitive functions, such as antioxidant/antitoxic defences and cellular proliferative/apoptotic balance, and its role in tumour progression, invasion, and drug resistance has been proposed (6, 9-11). The actions of GGT favouring tumour growth may be twofold: it acts as a source of essential amino acids both for protein synthesis and for the maintenance of intracellular GSH levels. Under specific conditions, the metabolism of GSH by GGT can exert pro-oxidant ef fects, with modulatory ef fects on several redox-sensitive processes at the membrane surface and in the extracellular microenvironment (6, 9). This suggests that the pro-oxidant reactions stimulated by GGT serve as an additional source of endogenous ROS in cancer cells, possibly contributing to the ‘persistent oxidative stress’ that has been described as a factor in genomic instability and carcinogenesis (12). A recent interesting observation showed that GGT -rich exosomes are released from human cancer cells. In the resistant and invasive phenotype of malignant cells, secreted GGT may play roles similar to those described for Helicobacter infection, leading to the establishment of cancer metastases (13). Endometrial cancer is presently the most common cause of morbidity related to gynaecological cancers both in well- developed European countries and the USA (14). Many women who enter menopause face mood and sleep disorders that can increase appetite and lead to obesity , which is consider to be a major risk factor of endometrial cancer . Fortunately recently developed combined fluoxetine with melatonin therapy was proved to be ef fective in treatment of mood sleep and appetite disorders in postmenopausal patients indirectly reducing the risk of endometrial cancer development (15). Endometrioid adenocarcinoma of the uterus (EAC) accounts for approximately 80% of all endometrial cancer cases (16). Although the prognosis is favourable compared with other female malignancies (e.g. ovarian cancer), doubts regarding the extent of treatment of EAC patients are still present. The discussion has focused on finding the proper balance that would avoid both under treatment resulting in an increased risk of recurrence and overtreatment that could potentially lead to unnecessary and bothersome complications (17, 18). Thus far, there have been some established histopathological risk factors, such as FIGO (International Federation of Gynaecology and Obstetrics) stage, depth of myometrial invasion, histological grading, or presence of lymphovascular invasion that help predict the group of patients at high risk for nodal disease that would require lymphadenectomy or adjuvant treatment (19, 20). Additionally higher intratumoral micro-vessels density was confirmed in endometrial cancer compared to benign endometrial changes and normal human endometrium and it is considered as an independent, poor prognostic factor of overall survival and recurrence-free survival (21). Recent studies have focused on the identification of novel biomarkers (e.g., stem cell-associated nuclear transcription factors) that could be helpful in the identification of patients diagnosed with low-stage EAC who might be at high risk of metastases and disease recurrence (22). New histological markers would be helpful to gain a better understanding of the disease, stratify patient risk and plan tailored treatments. The connection between GGT expression and endometrial cancer pathogenesis has not been explored extensively thus far . Seebacher et al. proved that the elevated pre-therapeutic GGT serum level in endometrial cancer patients correlates with shorter progression-free survival and can be used as an independent prognostic factor (23). Since only Hanigan and Frierson have investigated GGT expression in EC specimens previously (their study was limited to only four EAC cases) (4), our study sought to assess GGT activity in EAC tissue as a possible novel histopathological prognostic factor in the future.

Patients and materials The material included in the study came from 158 postmenopausal patients: 98 cases of EAC and 60 women with pelvic or gan prolapse (control). Before sur gery, every patient was weighed, and their heights were ensured. Body mass index (BMI) was calculated as the body mass divided by the square of the body height and is expressed as kg/m 2 units. Formalin-fixed paraffin-embedded (FFPE) tissue sections were taken from the files of our hospitals’ tissue archives. Histopathological classification and grading were performed on haematoxylin and eosin-stained slides according to standard histopathological procedures. Detection of gamma-glutamyl transferase (GGT) Immunohistochemistry was performed on 4- to 5-µm-thick FFPE tissue sections. Briefly , slides were deparaf finised and dehydrated in 100% ethanol, washed in distilled water and microwaved (600 W for 10 min and 5 min) in antigen retrieval solution (EDTA, pH 8.0). Next, they were washed in distilled water, cooled at room temperature (R T) for 20 – 30 min and immersed in 3% H 2O2 to block endogenous peroxidase. After washing in distilled water (5 min) and in Wash Buffer (Tris-HCl; DakoCytomation, S3006) twice for 5 min, 100 µl of the gamma- glutamyltransferase 1 precursor antibody (GGT1; polyclonal; Thermo Scientific P A5-21344) was applied to each tissue section. The antibody was diluted 1:400 in Dako Antibody Diluent with Background Reducing Components (S3022). Next, the slides were incubated for 30 min at R T. After washing in Wash Buffer, the secondary antibody was applied and incubated for 30 min at R T (Dako REAL EnVision HRP Rabbit/Mouse, K5007). The enzymatic reaction was performed with DAB (incubation for 10 – 30 s). Tissue sections were counterstained with haematoxylin and coverslipped. Scoring system GGT staining was evaluated in each tumour specimen and normal postmenopausal endometrium. Every tumour was scored according to the intensity of staining and number of stained cells (0, expression in 76% of cells). Positive staining was observed in the cell membrane, particularly in the apical part of glandular cells (staining scored as 1 - weak, 2 - moderate or 3 - strong). In cases where staining was classified as 3 (strong), the apical part of the cytoplasm also exhibited a positive reaction for GGT . The final immunoreactivity score was determined by multiplying the 396 intensity scores by the extent of the positivity scores of stained cells, providing a score range of 0 – 9. Each sample was assessed blindly by three independent observers (K.M-C., N.G ., and T.O.). In the case of any discrepancy between the observations, samples were verified once again together to achieve a consensus. Statistical analysis The Shapiro-Wilk test was used to examine the distribution of variables in the patients. The clinical features of the study group and control group were compared using the parametric Student’s t-test and non-parametric Mann-Whitney U-test or χ2 test where appropriate. One-way analysis of variance (ANOVA) and Kruskal-W illis test were used to evaluate more than two groups of variables followed by Fisher ’s post-hoc test where appropriate. Multivariate analysis of variance (MANOV A) was used to identify factors that may influence GGT expression. Gamma correlation or multivariate regression was used to evaluate the relationship between GTT expression and clinical features expressed as continuous variables. Clinical features of the study patients are presented as median values and the standard error of the median (S.E.M.) or number of cases and percentage. GGT immunoreactivity is presented using an arbitrary relative scale (points) as the median and interquartile range (IQR). P = 0.05 was accepted as statistically significant. All calculations were carried out using S tatistica software v. 10 (StatSoft, USA, 2011).

The average age of the patients with EAC was 59.31 ± 9.76 years, and they did not dif fer significantly from the average age of women in the control group (59.80 ± 11.56 years) (P = 0.779). Patients with EAC had a significantly higher mean BMI, significantly lower mean age of the first period, and a lower median number of pregnancies and deliveries. There were no significant differences between the mean age at menopause and menstrual cycle characteristics; however , the women from the study group had a significantly lower median number of gestations and deliveries compared with the controls (1.5, S.E.M. = 0.28, vs. 3.0, S.E.M. = 2.45, P= 0.028, and 0.5, S.E.M. = 0.67, vs. 3.0, S.E.M. = 1.17, P = 0.016, respectively). Detailed clinical features are presented in Table 1. From the 98 patients with diagnosed EAC, 16 (16.33%) were in stage IA, 46 (46.93%) in stage IB, 14 (14.29%) in stage II, and 22 (22.45%) in stage III, according to FIGO. Fifty-six (57.14%) patients were diagnosed with low- or moderate-grade (G1-2) disease, and 42 (42.86%) were diagnosed with high-grade (G3) disease. High- grade cases showed significantly more advanced clinical FIGO staging than G1-2 samples (P < 0.001) (Table 2). Expression of gamma-glutamyl transferase in tumour tissues and control specimens In all EAC samples and control group specimens, GGT membrane expression was predominant compared with 397 Patients with endometrial adenocarcionoma (n=98) Healthy controls (n=60) P Mean age [years] (S.D.*) 59.31 (± 9.76) 59. 80 (± 11.56) NS # Mean BMI [kg/m 2] (±S.D.*) 28.72 (± 3.81) 23.41 (± 2.03) NS # Grading - Low/moderate grade (G 1–2 ) - High grade (G 3) 57 (58.16%) 42 (41.84%) NA ### NA ### Mean age of first menstrual period [years] (±SD*) 9.52 (± 1.52) 11.45 (± 1.81) NS # Mean age of menopause [years] 53.21 (± 2.21) 52.45 (± 2.31) NS # Mean duration of menstrual cycle [days] (± S.D. *) 27.50 (± 2.5) 28.00 (± 2.0) NS # Menstrual cycles - regular - irregular 86 (87.80%) / 12 (12.20%) 46 (81.67) / 14 (18.33%) NS # Menstrual cycles - painful - painless 38 (48.72 / 40 (51.28%) 32 (21.48) / 117 (78.52%) NS # Mean duration of menstruation [days] 4.25 (± 1.52) 4.32 (± 0.97) NS # Type of menstrual bleeding - scant - normal - heavy 6 (6.12%) 75 (76.53%) 17 (17.35%) 22 (3.33%) 49 (81.67%) 9 (15.00%) NS # Median number of gestations (S.E.M.**) 1.5 (0.28) 3 (2.45) 0.028 ## Median number of deliveries (S.E.M.**) 0.5 (0.67) 3 (1.17) 0.016 ## *S.D., standard deviation; **S.E.M., standard error of median; #NS, statistically not significant; ## statistically significant value; ###NA, data not available. Table 1. Clinical characteristics of the patients with endometrioid adenocarcinoma of the uterus compared with healthy controls. cytoplasmic staining in both glandular and stromal cells. In glandular areas, GGT-positive cells were mostly localised at the upper parts of glands, which was described as apical staining. No area-related pattern was observed in GGT-positive stromal cells. The homogeneous cytoplasmic type of GGT staining was present in all assessed samples from all groups and was not further analysed. Apical staining was observed in both the control group and G1-2 EAC tumours ( Figs. 1 and 2); however , in G3 EAC specimens, cytoplasmic staining and high nuclear polymorphism areas were predominant. The majority of G1-2 EAC tumours (52/56) and control specimens (55/60) showed apical staining for GGT, while significantly fewer G3 EAC samples (10/42) were GGT-positive (46/69) for apical expression (P < 0.001). Comparable high GGT median apical expression was confirmed in normal postmenopausal endometrium (2.0, S.E.M. = 0.28) and in G1-2 EAC (2.0, S.E.M. = 0.26) tumours. In G3 tumours, GGT expression was significantly lower (0.0, S.E.M. = 0.07) than in normal postmenopausal endometrium (P= 0.02) (Fig. 3). After the stratification of cancer cases according to FIGO stringing, the highest median apical GGT expression was observed in IA EAC tumours (4.0, S.E.M. = 0.47), a finding that was similar to that in normal postmenopausal endometrium derived from the control group (2.0, S.E.M. = 0.28). The lowest median apical GGT expression was in I EAC (0.0, S.E.M. = 0.24) tumours. In IB EAC and IIIA-C EAC specimens, only moderate median apical GGT expression (1.0, S.E.M. = 0.24, and 1.0, S.E.M. = 0.16) was observed. The median apical expression of GGT in the control group and IA EAC tumours were significantly higher than that in IB – IIIC EAC specimens (P < 0.001) (Fig. 4). While analysing sections, we also noticed that, in the majority of G1-2 EAC cases, both the adjacent endometrial stromal cells and stroma between them were noticeably GGT - positive compared with the control group, where significantly fewer cells presented this staining pattern. Indeed, in 75% (21 out of 28) of G1-2 specimens, all observers noticed stromal cell staining; however , in the control group, it was observed in 51.67% (31 out of 60) of cases (P = 0.038). No stromal staining was noticed in G3 EAC specimens. Multivariate regression analysis confirmed a significant association between GGT apical expression and tumour grading (P < 0.001) and FIGO staging (P = 0.002), while there was no relationship between GGT cytoplasmic expression and tumour grading, staging, age of EAC diagnosis parity, or age of the first period and BMI. The age of EAC diagnosis, parity, or age of the first period and BMI also had no ef fect on the median apical GGT expression in EAC samples. 398 Tumor grading FIGO staging G1-2 G 3 P# IA 16 (16.33%) 0 (0.00%) IB 30 (30.61%) 16 (16.33%) II 6 (6.12%) 8 (8.16%) III A–C 4 (4.08%) 18 (18.37%) < 0.001 # Statistically significant at the level P < 0.05. Table 2 . Association between the International Federation of Gynaecology and Obstetrics staging and grade in patients with endometrioid adenocarcinoma of the uterus. Fig. 1. Gamma-glutamyltransferase (GGT) apical staining in a low-grade (G1) endometrioid adenocarcinoma of the uterus (EAC) in postmenopausal patients: A, weak staining (+); B, moderate staining ( ++); C, strong staining ( +++). Fig. 2. Gamma-glutamyltransferase (GGT) apical staining in a high-grade (G3) endometrioid adenocarcinoma of the uterus (EAC) in postmenopausal patients: A, weak staining (+); B, moderate staining (++); C, strong staining (+++). The arrow shows positive membrane staining for GGT.

This study is the first detailed investigation of GGTexpression in EAC tissue. Excluding the study of Hanigan et al. that was limited to only four endometrial cancer cases, no one has investigated GGTexpression in EAC specimens (5). Based on the variable clinicopathological parameters that low- and moderate- grade EAC are regarded as type I endometrial cancer while high- 399 Fig. 3. Expression of gamma-glutamyltransferase (GGT) according to tumour grade in patients with endometrioid adenocarcinoma of the uterus. * GGT-IHC, gamma-glutamyltransferase immunohistochemistry staining. Fig. 4. Expression of gamma-glutamyltransferase (GGT) according to tumour International Federation of Gynaecology and Obstetrics FIGO staging in patients with endometrioid adenocarcinoma of uterus. * GGT-IHC, gamma-glutamyltransferase immunohistochemistry staining. grade tumours can be categorised as type I and type II disease, we divided patients with EAC into G1-2 and G3 subgroups for proper data analysis. Among the investigation concerning the relationship between GGTand endometrial cancer, Seebacher et al. proved that an elevated pre-therapeutic GGTserum level in endometrial cancer patients correlates with shorter progression-free survival and can be used as independent prognostic factor (23). We carefully analysed the results presented by Seebacher et al., who reported impaired 5-year survival in women with elevated serum GGT levels. However, they also found no stage- or grade-dependent differences in the GGT serum concentration. According to their detailed findings, women diagnosed with G3 endometrial cancer (of both endometrioid and non-endometrioid origin) had lower GGT serum levels than those of patients with G1 tumours, but the differences were not significant (29.5 ± 31.4 ng/dl vs. 33.1 ± 52.9 mg/dl; P = 0.600). These data are in contrast to the findings of Polterauer et al., who investigated pre-therapeutic serum GGT levels in patients with cervical cancer and found significant associations between serum GGT levels and FIGO stage (P < 0.0001) and age (P < 0.0001), which were not reported by Seebacher et al . for endometrial cancer (23, 24). Similarly to women with endometrial cancer, patients with cervical cancer and elevated GGTserum levels were associated with poor disease-free and overall survival in univariate analyses, although these associations were not confirmed in a multivariate Cox-regression model (24). As the role of GGTin gynaecological malignancies is not clear due to non-unequivocal results, we evaluated GGT expression in normal endometrium and in EAC specimens. At first glance, our results seem to contrast strongly with those of Seebacher et al ., as we showed significantly decreased GGT expression in G3 EAC. However, we must emphasise that these GGT serum and tissue expression levels cannot be compared directly (23). First, the serum GGT level depends not only on malignant tissue secretion but also on its involvement in metabolic reactions, including oxidative stress, kidney excretion, and liver metabolism. Consequently, the GGT serum level may not be a direct function of apical GGT expression in endometrial tissues. Second, we investigated only the endometrioid type of endometrial cancer, while Seebacher et al . evaluated GGT serum concentrations in women with endometrioid and non-endometrioid endometrial cancers. Our study focused on postmenopausal patients, while Seebacher et al . included every woman with endometrial cancer irrespective of menopausal status. Seebacher et al. found no association between endometrial cancer grade or stage and the GGT serum level; therefore, they considered the serum GGT concentration as an independent predictor of 5-year survival, while we found a strong negative association between GGTapical expression and grade (23). Our results are seemingly opposite those of Seebacher et al., and we believe that further studies should be undertaken to elucidate the role of GGTin endometrial cancer and evaluate its clinical significance. We found that GGTexpression in EAC tissue differs with that of healthy endometrium. The presented differences in the assessed features correlated with both tumour biology (histological grading) as well as the stage of EAC. In summary , the best expression of membrane GGTwas present in the control group of healthy endometrium and well- to moderate-differentiated (G1-2) EAC. In G3, GGT expression of the EAC samples was much poorer, and over 75% of samples were GGT -negative. Surprisingly, apical expression of GGT was more intense in specimens with better cell dif ferentiation and more favourable grading. In G1-2 EAC patients, it was highly expressed compared with G3 patients, where most G3 cases showed no membrane staining at all. Carcinomas arising from some GGT -positive epithelium retained their GGT -positive phenotype. The more structural and functional dif ferences between normal and neoplastic cells, the more diverse the GGT expression. There is evidence that GGT is dysregulated in malignant cells by producing reactive oxygen species, causing tumour progression towards more aggressive phenotypes associated with a poorer prognosis (5, 25). Many human tumours express high levels of GGT. However, the distribution and concentration of GGT in human tumours present several differences from what is observed in normal tissues (5, 6). Moreover, the heterogeneous expression of GGT in different tumour types, and even different tumours of the same type, was observed (26). In a study of human GGT - transfected melanoma cells, higher levels of GGT activity were associated with greater levels of background DNA damage and oxidised bases (27); this activity was unrelated to dif ferences in cell cycle distribution and apoptotic rates. To identify a significant determinant for GGTexpression and activity in endometrial cancer cells, Ravuri et al . tested the Ishikawa cell line and established that endogenous production of reactive oxygen species by the NADPH oxidase complexes is a determinant of γ-glutamyltransferase expression (3). Neoplastic cells in many tumours are not polarised and, therefore, express GGT on their entire cell surface (5). Unlike normal cells in which GGT only has access to substrates in ductal fluids, the GGT on tumour cells can cleave glutathione (GSH) in interstitial fluid and blood. The expression of GGT provides tumour cells with an additional source of cysteine and cystine from the cleavage of extracellular GSH and oxidised glutathione (GSSG). In our investigation, in addition to its expression in tumour cells, GGT expression was also found in the stromal cells between them. It is interesting that GGT staining was present not only on the membrane but also within the cytoplasm. Both membrane and cytoplasmic cancer was observed in human prostate carcinoma. The cytoplasmic staining may reflect GGT protein that is being synthesised and processed within the cell (28). Because differences in GGT expression between the controls and G1-2 EAC were not observed, it would be interesting to understand the origin of the increased G1-2 endometrial stromal staining and elevated serum GGT levels in some EAC patients (20). It is possible that cancer tissue could be the source for stromal and serum GGT elevation. GGT expression in ovarian cancer tissue is reflected in GGTserum levels (2). The elevated serum level of GGT detected in several types of neoplasia may be due not only to its release from cancer cells, but might also be associated with systematic changes in the disease, such as example, inflammation (29). A positive correlation between greater advancement of the tumour and serum GGTlevel was observed in renal and cervical cancers (24, 30). To our knowledge, this is the first comprehensive study evaluating GGT expression in EAC and its correlation with clinicopathological features. All laboratory evaluations were performed in one setting by staf f highly experienced in immunohistochemistry, and were conducted during a short period of time to reduce research bias. The numerous limitations of our study need to be acknowledged when interpreting and applying the outcomes. Foremost, we did not evaluate serum GGT levels and oxidative stress markers in our study and are unable to draw reliable conclusions on the role of GGT in EAC. However , exploring GTT mechanisms in EAC was not an aim of our study. As the roles of GGT in physiological and pathological processes were broadly investigated and described in detail, even if no unequivocal results were obtained, as discussed earlier, we aimed only to evaluate the possible clinical utility of GTT. Therefore, IHC was chosen as an evaluation method because it is an essential pathological technique. A second limitation is that we failed to analyse whether GGT expression is directly associated with prediction of the final therapeutic result in patients with EAC. As women with low-grade EAC have very favourable progression- free and overall survival prognoses, patients need to be re- evaluated, at least at the 5-year follow-up, to evaluate the utility of GTT as a therapy predictor for women with EAC; such data will be 400 available in the future. We also acknowledge that the number of patients with G3 EAC was low; however, the predominant EAC grades are 1 and 2, accounting for 80% of cases (18). Based on the above discussion, we conclude that our initial results regarding GTT expression in EAC need to be further validated in a lar ger cohort to gain more epidemiological and clinical impact. Only a prospective follow-up GGT expression analysis in the EAC specimen conducted in a larger population and with age-matched controls over a longer period of time will fully allow us to elucidate the clinical usefulness of GGTexpression. We believe that the most important unresolved issue to address for further investigation is the low rate of membrane staging of GGT in less-differentiated endometrial cancer cases compared with better-differentiated ones. The result conflicts with what was found in other cancer types, including ovarian and cervical cancer, and further investigation is essential to resolve this issue unequivocally. Although exploring GTTinvolvement in EAC cell metabolism was not the aim of our study, the decreased apical GTTexpression in high-grade EAC should be discussed in the context of its possible role in tumour development. First, it is known that immature high-grade tumours have significantly dif ferent metabolism and protein expression patterns compared with well- differentiated low-grade neoplasms and healthy tissues. Therefore, it is not surprising that we observed lower GTTexpression in G3 EAC. Altered GTT levels may render EAC cells more susceptible to oxidative stress. Normal cells and tissues have developed many mechanisms to reduce oxidative stress, as it directly impairs their function and may result in apoptotic death. Uterine endometrium is directly exposed to dif ferent infection agents that can lead to subclinical or clinical inflammation. In a bovine model an elevated expression of mRNAs of chemokines (CXCL1 and CXCL2), interleukins, prostaglandins and metallopeptidase was confirmed during the late puerperium preventing, according to authors, cows from persistent endometritis (31). In neoplastic cells however , inflammatory background resulting in ROS overproduction, lead to genomic instability and enhance mutation formation. As neoplasms are fast growing, cellular loss due to enhanced DNAmutations is not an issue. This genetic instability also increases the occurrence of ‘favourable’ mutations that may result in tumour resistance to chemotherapy or radiotherapy. Therefore, unlike healthy cells, GTT-impaired synthesis in neoplastic cells may be desired, as it leads to neoplastic growth and spread. Authors contribution: KP, TO, and KO designed the study . NG and MP selected the cases. MCK, NG , and TO performed histopathological assessment of selected cases. TB and MM participated in the study design and data analysis. KP , TB, and TO drafted the manuscript. AS-K and AS were responsible for specimen evaluation and the selection of eligible cases. All authors read and approved the final manuscript.

The cost of materials and reagents was covered with help from a grant from the S tudents’ Scientific Association of Jagiellonian University Medical College. Conflict of interest: None declared.

1. Hanigan MH, Ricketts WA. Extracellular glutathione is a source of cysteine for cells that express gamma-glutamyl transpeptidase. Biochemistry 1993; 32: 6302-6306. 2. Obrador E, Carretero J, Ortega A, et al . Gamma-glutamyl transpeptidase overexpression increases metastatic growth of B16 melanoma cells in the mouse liver. Hepatology 2002; 35: 74-81. 3. Ravuri C, Svineng G , Pankiv S, Huseby NE. Endogenous production of reactive oxygen species by the NADPH oxidase complexes is a determinant of γ-glutamyltransferase expression. Free Radic Res 2011; 45: 600-610. 4. Hanigan MH, Frierson HF Jr . Immunohistochemical detection of gamma-glutamyl transpeptidase in normal human tissue. J Histochem Cytochem 1996; 44: 1101-1108. 5. Hanigan MH, Frierson HF Jr , Swanson PE, De Young BR. Altered expression of gamma-glutamyl transpeptidase in human tumors. Hum Pathol 1999; 30: 300-305. 6. Pompella A, De Tata V, Paolicchi A, Zunino F. Expression of gamma-glutamyltransferase in cancer cells and its significance in drug resistance. Biochem Pharmacol 2006; 71: 231-238. 7. Pandur S, Pankiv S, Johannessen M, Moens U, Huseby NE. Gamma-glutamyltransferase is upregulated after oxidative stress through the Ras signal transduction pathway in rat colon carcinoma cells. Free Radic Res 2007; 41: 1376-1384. 8. Andreassen K, Mortensen B, Winberg JO, Huseby NE. Increased resistance towards oxidative stress accompanies enhancement of metastatic potential obtained by repeated in vivo passage of colon carcinoma cells in syngeneic rats. Clin Exp Metastasis 2002; 19: 623-629. 9. Pompella A, Corti A, Paolicchi A, Giommarelli C, Zunino F. γ-Glutamyltransferase, redox regulation and cancer drug resistance. Curr Opin Pharmacol 2007; 7: 360-366. 10. Franzini M, Corti A, Lorenzini E, et al. Modulation of cell growth and cisplatin sensitivity by membrane γ- glutamyltransferase in melanoma cells. Eur J Cancer 2006; 42: 2623-2630. 11. Dominici S, Valentini M, Maellaro E, et al . Redox modulation of cell surface protein thiols in U937 lymphoma cells: the role of γ-glutamyl transpeptidase-dependent H 2O2 production and S-thiolation. Free Radic Biol Med 1999; 27: 623-635. 12. Toyokuni S, Okamoto K, Yodoi J, Hiai H. Persistent oxidative stress in cancer. FEBS Lett 1995; 35: 1-3. 13. Franzini M, Corti A, Fierabracci V, Pompella A. Helicobacter , gamma-glutamyltransferase and cancer: further intriguing connections. World J Gastroenterol 2014; 20: 18057-18058. 14. Siegel R, Ma J, Zou Z, Jemal A. Cancer statistics, 2014. CA Cancer J Clin 2014; 64: 9-29. 15. Chojnacki C, Walecka-Kapica E, Klupinska G , Pawlowicz M, Blonska A, Chojnacki J. Ef fects of fluoxetine and melatonin on mood, sleep quality and body mass index in postmenopausal women. J Physiol Pharmacol 2015; 66: 665-671. 16. Felix AS, Weissfeld JL, Stone RA, et al. Factors associated with Type I and Type II endometrial cancer. Cancer Causes Control 2010; 21: 1851-1856. 17. Press JZ, Gotlieb WH. Controversies in the treatment of early stage endometrial carcinoma. Obstet Gynecol Int 2012; 2012: 578490. doi: 10.1155/2012/578490 18. Hsiao SM, Wei LH. Controversies in the adjuvant therapy of endometrial cancer . ISRN Obstet Gynecol 2011; 201 1: 724649. doi: 10.5402/2011/724649 19. SGO Clinical Practice Endometrial Cancer Working Group, Burke WM, Orr J, Leitao M, et al . Endometrial cancer: a review and current management strategies: part I. Gynecol Oncol 2014; 134: 385-392. 20. SGO Clinical Practice Endometrial Cancer Working Group, Burke WM, Orr J, Leitao M, et al . Endometrial cancer: a review and current management strategies: part II. Gynecol Oncol 2014; 134: 393-402. 21. Czekierdowski A, Czekierdowska S, Czuba B, et al . Microvessel density assessment in benign and malignant 401 endometrial changes. J Physiol Pharmacol 2008; 59 (Suppl. 4): 45-51. 22. Pitynski K, Banas T, Pietrus M, Milian-Ciesielska K, Ludwin A, Okon K. SOX-2, but not Oct4, is highly expressed in early-stage endometrial adenocarcinoma and is related to tumour grading. Int J Clin Exp Pathol 2015; 8: 8189-8198. 23. Seebacher V, Polterauer S, Grimm C, et al . Prognostic significance of gamma-glutamyltransferase in patients with endometrial cancer: a multi-centre trial. Br J Cancer 2012; 106: 1551-1555. 24. Polterauer S, Hofstetter G , Grimm C, et al . Relevance of gamma-glutamyltransferase - a marker for apoptotic balance - in predicting tumor stage and prognosis in cervical cancer . Gynecol Oncol 2011; 122: 590-594. 25. Dawson J, Smith GD, Boak J, Peters TJ. γ-Glutamyl transferase in human and mouse breast tumours. Clin Chim Acta 1979; 96: 37-42. 26. Corti A, Franzini M, Paolicchi A, Pompella A. Gamma- glutamyltransferase of cancer cells at the crossroads of tumor progression, drug resistance and drug tar geting. Anticancer Res 2010; 30: 1169-1182. 27. Corti A, Duarte TL, Giommarelli C, et al . Membrane gamma-glutamyl transferase activity promotes iron- dependent oxidative DNA damage in melanoma cells. Mutat Res 2009; 669: 112-121. 28. Hanigan MH, Gallagher BC, Townsend DM, Gabarra V . Gamma-glutamyl transpeptidase accelerates tumor growth and increases the resistance of tumors to cisplatin in vivo. Carcinogenesis 1999; 20: 553-559. 29. Grimm C, Hofstetter G , Aust S, et al . Association of gamma-glutamyltransferase with severity of disease at diagnosis and prognosis of ovarian cancer . Br J Cancer 2013; 109: 610-614. 30. Simic T, Dragicevic D, Savic-Radojevic A, Cimbaljevic S, Tulic C, Mimic-Oka J. Serum gamma glutamyl-transferase is a sensitive but unspecific marker of metastatic renal cell carcinoma. Int J Urol 2007; 14: 289-293. 31. Peter S, Michel G , Hahn A, et al . Puerperal influence of bovine uterine health status on the mRNA expression of pro-inflammatory factors. J Physiol Pharmacol 2015; 66: 449-462. Received: October 21, 2015 Accepted: May 25, 2016 Author ’s address: Assoc. Prof. Kazimierz Pitynski, Department of Gynaecology and Oncology , Jagiellonian University Medical College, 23 Kopernika S treet, 31-501 Cracow, Poland. E-mail: [email protected] 402