Expression of thioredoxin 1 and peroxiredoxins in squamous cervical carcinoma and its predictive role in NACT

All tissue samples were collected from the Second Affiliated Hospital of Wenzhou Medical University after informed written patient consent. This study was approved by the ethical committee of the Second Affiliated Hospital of Wenzhou Medical University and conducted according to the Helsinki declaration. At first, a total number of 13 squamous cervical cancer tissues and adjacent normal tissues were collected between January 2014 and April 2015 for the expression analysis of Trx 1, Prx 1 and Prx 2 by western-blotting and immunohistochemistry. The median age of the 13 patients is 49y (range, 36~74y). These patients are in stage IB1 (10 of 13 patients, 76.9%) and IIA1 (3 of 13 patients, 23.1%) with the median diameter of 2.13 ± 0.12. Ten cases were grade I, 2 cases were grade II and 1 case was grade III. Stage and grading were according to the International Federation of Gynecology and Obstetrics (FIGO) system. Paired tumor cervical cancer tissues and adjacent normal tissues from each patient were obtained after radical hysterectomy.

Another cohort (cohort 2) consisted of 35 patients with IB2 or IIA2 (bulky, primary tumor > 4 cm in diameter) squamous cervical cancer treated with NACT at the Second Affiliated Hospital of Wenzhou Medical University between January 2007 and August 2014. Clinicopathological variables were recorded, including age at diagnosis, stage, tumor grade and size. Details of treatment was documented for retrospective analysis. Clinicopathologic data was detailed in Table 1. None of these patients had received chemotherapy, immunotherapy, hormonal therapy or radiotherapy before the specimen collection. Paired tumor samples from each patient were obtained during cervical biopsy (pre-chemotherapy) or surgery (post-chemotherapy). Hematoxylin and eosin-stained slides of the cervical tumor from each patient were reviewed for confirmation of pathological features, as well as to select suitable tissue blocks for immunohistochemistry analysis.

All eligible patients received one or two courses of cisplatin-based NACT, as previous described [19]: cisplatin, 60 mg/m² on day 1; 5-fluorouracil, 750 mg/m² on day 1; mitomycin (8 mg/m²) on day 1 for one or two courses, every 28 days. All of these chemotherapeutics were administrated via uterine artery injection. The chemotherapy response was evaluated by measuring the tumor’s two dimensions (the longest diameter and its perpendicular diameter) with magnetic resonance imaging or other radiographic means (CT scan and ultrasound). Eighteen cases (51.4%) were measured by MRI, 4 cases (11.4%) were measured by CT scan and 13 cases (37.2%) were measured by ultrasound. The chemotherapy response was determined two weeks after completion of neoadjuvant chemotherapy, according to the WHO criteria [20]. Complete response (CR) was defined as the complete remission of the tumor. Partial response (PR) was defined as least a 50% decrease in the tumor volume. Stable disease (SD) meant a steady state or a response less than 50%, and progressive disease (PD) was defined as an unequivocal increase of at least 25% in the tumor volume. The patients with CR or PR were defined as chemotherapy responders, while the patients with SD or PD were deemed as chemotherapy non-responders. We evaluated the response after the first cycle chemotherapy, if the result showed response, we then did the second one. Totally, there are 18 cases under-going one cycle and 17 cases under-going two cycles. All patients were treated with radial hysterectomy and bilateral pelvic lymphadenectomy two to three weeks after completion of the NACT regimen as described previously [19].

Samples were homogenized and lysed in Laemmli buffer with a cocktail of protease inhibitors. The total protein concentrations were quantified by the bicinchoninic acid protein assay (Thermo Scientific, Rockford, IL). Equal amounts of total protein were resolved by sodium dodecyl sulfate- polyacrylamide gel electrophoresis, transferred to a nitrocellulose membrane under constant voltage and blocked with tris buffered saline with tween (TBST) containing 5% non-fat dried milk. Primary antibodies and secondary antibodies were diluted in TBST and applied with a washing step in between. Proteins were detected using the Amersham ECL western blotting detection kit (GE Healthcare, Piscataway, NJ). Primary antibodies used including: anti-Trx 1 (CST, USA; 1:1000), anti-Prx 1(Abcam, USA; 1:2000), anti- Prx 2 (Abcam, USA; 1:1000), anti- β-actin (Abcam, USA; 1:5000). ImageJ software was used to quantify western blot bands in this study. ImageJ is a Java-based image analysis package widely used in quantitating visual results such as bands on gels or photomicrographs.

Formalin-fixed paraffin embedded tissues were cut into sections of 4 μm and then mounted onto poly-L-lysine-coated slides. In briefly, after de-paraffinized in xylene and rehydrated through graded ethanol solutions, sections were boiled in a 10 μmol/L citric buffer suolution (pH 6.0) in a microwave oven for 10 min, followed by incubation with 3% hydrogen peroxide in methanol to suppress the endogenous peroxidase activity and overnight incubation with primary antibodies (Trx1, CST, USA; 1:200; Prx1, Abcam, USA; 1:200, Prx 2, Abcam, USA; 1:200) at 4 °C. Subsequently, the sections were incubated with pre-diluted secondary antibody (Santa Cruz, USA) for 2 h at room temperature, followed by further incubation with 3,3-diaminobenzidine tetrahydrochloride. Finally, the slides were counterstained with hematoxylin. Appropriate positive and negative controls were stained in parallel. For negative controls, primary antibodies were replaced with phosphate-buffered saline solution. Human cervical cancer tissues were used as a positive control.

Trx 1, Prx 1 and Prx2 protein were detected primarily in cytoplasm and partially in cytomembrane. Assessment of expression of Trx1, Prx-1 and Prx2 staining was determined as previously described [19]. In brief, expression of the these markers was determined by an individual labeling score considering percent and staining intensity of positive cells by two independent assessors blinded to the study end points. [21] Intensity of stained cells was graded semi-quantitatively into four levels as following: 0 (no staining); 1 point (weak staining:  light yellow); 2 points (moderate staining: yellow brown) and 3 points (strong staining: brown). The percentage was scored as following: 0 (0 to 5%), 1 point (6 to 24%), 2 points (25 to 49%), 3 points (50 to 74%), and 4 points (75 to 100%). Intensity and fraction of positive cell scores were multiplied for each marker and thus got the immunoreactive score.

The first cohort consisted 13 bulky squamous cervical cancer patients. The median age was 49y (range 36~74y), and stage IB1 and stage IIA1 was 76.9 and 23.1%, respectively.

The second cohort consisted 35 patients with bulky stage IB2~IIA2 squamous cervical cancer treated with NACT. The median age was 44y (range 25~63y). Nineteen tumors were stage IB2 (54.3%), and sixteen cases were stage IIA2 (45.7%). Of the 35 patients identified, 5 patients (14.3%) showed a complete response and 12 patients (34.3%) showed a partial response, while 13 patients (37.1%) had a stable disease and other five patients (14.3%) showed progressive disease. Thus, 17 cases were chemo-sensitive (48.6%) and 18 were chemo-resistant (51.4%). There were no significant differences in patients or tumor characteristics between the responder and non-responder cohorts (Table 1). Clinicopathologic characteristics are shown in Table 1.

Initially, the expression of Trx 1, Prx 1 and Prx 2 in 13 cases of squamous cervical cancer and adjacent normal tissue was investigated by Western blotting. As shown in Fig. 1, all of these three indicated proteins were over-expressed in squamous cervical cancer tissues compared to adjacent normal tissues. The expression and localization of Trx 1, Prx 1 and Prx 2 in squamous cervical cancer and adjacent normal tissue were further examined by immunohistochemistry. A strong cytoplasmic or membranous expression of Trx 1 was observed in all 13 squamous cervical cancer samples analyzed, and generally limited to tumor cells (Fig. 2). The expression pattern of Prx 1 and Prx 2 in squamous cervical cancers was similar to Trx 1. Interestingly, with regard to adjacent normal tissue, moderate staining of Trx 1 was observed in superficial layer cells and intermediate layer cells of cervical epithelial among all the 13 samples, while only three cases showed positive staining in basal layer cells, in which two cases showed weak staining and the third case showed moderate staining. Similarly, all of the adjacent normal tissues (13/13, 100%) showed weak or moderate staining for Prx 1 and Prx 2 in superficial layer and intermediate layer of cervix epithelia, only 3 cases (23.1%) expressed weak or moderate staining in basal layer of cervix. Since, cervical epithelial cells in basal cells may be the source of malignant transformed cells, we compared the different expression of these three proteins between cervical cancers samples and the basal layer cells of adjacent normal tissues. Quantification of Trx 1, Prx 1 and Prx 2 staining showed that Trx 1, Prx 1 and Prx 2 immunoreactive scores were much higher in squamous cervical cancer compared to basal layer cells of adjacent normal tissues (P = 0.007, P = 0.01, P = 0.01, respectively), (Table 2). Figure 2 illustrates the immunostaining for all markers studied in squamous cervical cancer tissues and adjacent normal tissues.

To investigate the effect of chemotherapy on the expression of Trx 1, Prx 1 and Prx 2 in squamous cervical cancer samples, the expression of Trx 1, Prx 1 and Prx 2 was performed in 35 matched primary squamous cervical cancer biopsies before and after cisplatin-based NACT using immunohistochemistry. Pre-chemotherapy cervical cancer tissues consistently showed weak or moderate positive staining of Trx 1, Prx 1 and Prx 2, while post-chemotherapy tissue consistently showed moderate or intense positive staining (Fig. 3). Using a Wilcoxon test, the expression of Trx 1, Prx 1 and Prx 2 were significantly stronger in post-chemotherapy tissues compared to pre-chemotherapy tissues, (P = 0.000, P = 0.001, P = 0.010, respectively) (Table 3), suggesting Trx 1, Prx 1 and Prx 2 protein were implicated in squamous cervical tumor responses to chemotherapy.

To explore the role of Trx 1, Prx 1 and Prx 2 in the development of chemo-resistance, the expression of Trx 1, Prx 1 and Prx 2 in pre-chemotherapy cervical cancer tissues between chemotherapy responders and non-responders was examined by immunohistochemistry. As shown in Table 4, the expression of Trx 1, Prx 1 and Prx 2 in pre-chemotherapy cervical cancer tissues was markedly stronger in chemotherapy non-responders than that in responders, suggesting patients with high levels of Trx 1, Prx1 and Prx2 were more resistant to cisplatin-based NACT than those with low protein expression (P < 0.05, respectively).

The correlation between Trx1 and Prx1/2 in 13 cervical squamous cancer samples without chemotherapy and 35 pre-chemotherapy cervical squamous cancer samples were further explored. Interestingly, our results showed that there was a positive correlation between Trx 1 and Prx 2, but not between Trx 1 and Prx 1, suggesting Trx 1 and Prx 2 may function together in cervix carcinogenesis and chemoresistance (Table 5).