P63 rabbit monoclonal antibodies were purchased from Abcam (Cambridge, UK). GAPDH rabbit polyclonal antibody was purchased from ProteinTech Group, Inc. (Chicago, IL, USA).

A total of 38 pairs of formalin-fixed, paraffin-embedded (FFPE) sections of CRC tissues and non-cancerous tissues (NCTs) were obtained from the Pathology Department of the Third XiangYa Hospital of Central South University (Hunan, China) between October 2013 and June 2014. None of these patients had received radiotherapy or chemotherapy prior to surgery. The histopathological type and the stage of CRC were determined according to the criteria of the World Health Organization classification. All of the patients were staged using the US National Comprehensive Cancer Network (NCCN) Clinical Guidelines 2014. The present study was approved by the Ethics Committee of the Third XiangYa Hospital of Central South University. Patient consent was obtained both from the patients and the families of the patients.

RT-qPCR was performed as previously described (16). Briefly, total RNA was isolated using an E.Z.N.A. Total RNA Kit II (Omega Bio-Tek Inc., Norcross, GA, USA). miRNA was isolated using an E.Z.N.A. PF miRNA Isolation kit (Omega Bio-Tek Inc.). RT-qPCR was performed using Real-Time Quantitative PCR SYBR-Green detection reagent (CoWin Biotech Co., Ltd., Beijing, China). miRNA RT-qPCR was performed using an All-in-One™ miRNA qRT-PCR Detection kit (GeneCopoeia, Rockville, MD, USA). Primers and the reaction conditions have been previously described (16). The relative expression of TAp63 was normalized using the 2^−ΔΔCt method relative to GAPDH. The relative expression of miR-133b was normalized using the 2^−ΔΔCt method relative to U6-snRNA. All PCR reactions were run in triplicate. Each sample was amplified in triplicate.

Western blot analysis was performed as previously described using primary antibodies (16). Briefly, cells were lysed in a lysis buffer and centrifuged at 14,000 × g at 4°C for 10 min. The supernatants were collected and a BCA protein assay was performed. Total protein (100 mg) was separated on a 10% polyacrylamide gel and transferred to polyvinylidene difluoride (PVDF) membranes (Invitrogen, Carlsbad, CA, USA). The membranes were blocked for 2 h, and were then incubated with a primary antibody at 4°C overnight. After being washed with PBST, the membranes were incubated with an HRP-conjugated goat anti-rabbit IgG secondary antibody for 60 min at room temperature. The images were obtained using Kodak film (Shanghai, China). All experiments were performed in triplicate.

Immunohistochemistry (IHC) was carried out using Dako EnVision System (Dako Diagnostics AG, Zug, Switzerland) following the manufacturers protocol. For TAp63 protein, staining localized in the nuclei was considered positive. Images were captured using an Olympus microscope. Image-Pro Plus version 6.0 software (Media Cybernetics, Inc., Rockville, MD, USA) was used to assess the area and density of the dyed region, and the integrated optical density (IOD) value of the IHC section. The mean densitometry of the digital image (magnification, ×400) was designated as representative TAp63 staining intensity (indicating the relative TAp63 expression level). The signal density of the tissue areas from five randomly selected fields were counted in a blinded manner and subjected to statistical analysis.

ISH was performed on CRC tissues. A locked nucleic acid (LNA) detection probe for miR-133b, an LNA U6-positive and an LNA-negative control probe with a scramble sequence were used in this analysis. All probes were purchased from Auragene Bioscience Corporation, Inc. (Changsha, China). All probes were labeled with digoxigenin. After being washed with phosphate-buffered saline (PBS) and pre-hybridization, the slides were incubated with DIG-labeled LNA miR-133b, DIG-labeled LNA U6 probes (positive control) and DIG-labeled LNA scrambled sequence (negative control) at 49°C overnight. Washes were carried out at 49°C, and then the slides were incubated with blocking solution. After PBS washing, the images were captured using an Olympus microscope. Image-Pro Plus version 6.0 software was used to assess the area and density of the dyed region and the IOD value of the ISH section.

The secondary structure of 446 bp pre-miR-133b sequences was predicted using RNAfold web server (http://nhjy.hzau.edu.cn/kech/swxxx/jakj/dianzi/Bioinf4/miRNA/miRNA1.htm).

We analyzed the data from February 2015. All statistical calculations were performed using GraphPad Prism (GraphPad Prism software, version 6.01; GraphPad, San Diego, CA, USA) and Statistical Package for the Social Sciences (SPSS) PASW Statistics software, version 20 (SPSS, Inc., Chicago, IL, USA). A Mann-Whitney U test was used to compare differences between two groups. The correlation between TAp63 and miR-133b was determined by the Spearmans rank correlation test. The sensitivity and specificity in CRC tissues were calculated for each cut-off value. The area under receiver-operating characteristic curve (AUC) was also estimated. The level of statistical significance was set at P<0.05. All experiments were repeated three times.

The clinical and histopathological features of the patients are summarized in Table I. In total 38 patients were included in the present study, which comprised 26 men and 12 women, with a combined age range of 55–83 years. Regarding the degree of cell differentiation, out of the 38 cases, 6 (15.79%) were well-differentiated adenocarcinomas, 27 (71.05%) were moderately differentiated and 5 (13.16%) cases were poorly differentiated adenocarcinomas.

Regarding the evolutionary stage of the studied CRC cases, the majority of the patients were in the late disease stages: 5 (11.16%) were diagnosed in stage I, 11 (28.95%) in stage II, 18 (47.37%) in stage III and 4 (10.53%) in stage IV.

To study the expression of TAp63, we measured the mRNA and protein expression levels of TAp63 in 38 paired human CRC tissues and NCT samples using qRT-PCR, western blotting and IHC methods, respectively. Compared with the NCTs, the TAp63 mRNA expression level was significantly lower in 21 (55.27%) tumor tissues (Fig. 1A, P<0.05). Consistent with the RT-qPCR results, a decrease of TAp63 protein expression was observed in the CRC tissues, compared with the NCTs (Fig. 1B, P<0.05). IHC also revealed that TAp63 was downregulated in the CRC tissues (Fig. 1C). Moreover, we evaluated the density of the staining signal of all 38 cases using the Image-Pro Plus 6.0 image analysis software. The mean density of cancerous tissues from all 38 cases was 0.005345±0.0005019, while that of the adjacent tissue was 0.01483±0.0003317. This result suggests a significantly lower level of TAp63 expression in CRC tissues compared to their NCTs (Fig. 1D, P<0.05).

To study the relationship between TAp63 and miR-133b in human CRC, we assessed the expression levels of miR-133b in the 38 paired CRC and NCT samples using RT-qPCR and ISH. Compared with the NCTs, the miR-133b expression level was significantly lower in 31 (81.58%) tumor tissues (Fig. 2A, P<0.05). ISH revealed that miR-133b was downregulated in the CRC tissues (Fig. 2B). Moreover, we measured the density of the staining signal in all 38 cases using Image-Pro Plus 6.0 image analysis software. The mean density of cancerous tissues from all 38 cases was 0.009005±0.000585, while that of the adjacent tissue was 0.01314±0.000909. This result suggests a significantly lower level of miR-133b expression in the CRC tissues compared to their NCTs (Fig. 2D, P<0.05). In addition, the expression of miR-133b was found to be positively correlated with TAp63 (Fig. 2C, P<0.05).

The association of TAp63 and miR-133b with various clinicopathological parameters of the CRC cases are listed in Table I. An inverse association was found between TAp63 expression and lymph node and distant metastases (P<0.05). miR-133b expression was also significantly correlated with lymph node and distant metastases. No statistically significant associations were found between the expression of TAp63 or miR-133b and other clinicopathological parameters (P>0.05).

By receiver operating characteristic (ROC) curve analysis, we proved that TAp63 and miR-133b are suitable predictors for CRC. The AUC of TAp63 for CRC was 0.623 [95% confidence interval (CI), 0.497–0.748; P=0.046], with 73.7% sensitivity and 50% specificity, respectively (Fig. 3A). The AUC of miR-133b for CRC was 0.857 (95% CI, 0.774–0.940, P<0.0001), with 78.9% sensitivity and 81.6% specificity, respectively (Fig. 3B). The combined AUC of TAp63 and miR-133b for CRC was 0.881 (95% CI, 0.805–0.956, P<0.0001), with 89.5% sensitivity and 71.1% specificity, respectively (Fig. 3C).

To uncover whether there are point mutations in miR-133b DNA, we scanned 446 bp of the pre-miR-133b coding region. The results showed that only one patient had a single nucleotide change, G to T, located 6 bp from the 3′ end of pre-miR-133b (Fig. 4A). To further explore the function of the mutation site, we compared the miR-133b predicted secondary structure between the gene of the patient and the GenBank gene. As shown in Fig. 4B, the nucleotide change did not cause an apparent change in the stem-loop structure.