miR-200b, ZEB2 and PTPN13 Are Downregulated in Colorectal Carcinoma with Serosal Invasion

Serosal invasion is an independent negative prognostic factor in certain cancers, including CRC. However, the mechanisms behind serosal invasion are poorly understood. We therefore assumed that epithelial-mesenchymal transition (EMT) might be involved. Our study included 34 patients with CRC, 3 stage pT2, 14 stage pT3 and 17 showing serosal invasion (stage pT4a according to TNM staging system). RNA isolated from formalin-fixed paraffin-embedded tissue samples was analysed for expression of the miR-200 family and their target genes CDKN1B, ONECUT2, PTPN13, RND3, SOX2, TGFB2 and ZEB2 using real-time PCR. We found upregulation of miR-200b and ONECUT2 in CRC pT3 and pT4a compared to normal mucosa, and downregulation of CDKN1B in CRC pT3. Moreover, we observed, downregulation of miR-200b, PTPN13 and ZEB2 in CRC with serosal invasion (pT4a) compared to pT3. Our results suggest the involvement of partial EMT in serosal invasion of CRC. In addition, PTPN13 seems to be one of the important regulators involved in serosal invasion, and ONECUT2 in tumour growth.


Introduction
Colorectal adenocarcinoma (CRC) is one of the most common cancers and the second leading cause of cancer-related deaths in Europe. Its burden is expected to increase as the population ages [1]. Based on the American Joint Committee on Cancer (AJCC) tumournode-metastasis (TNM) staging system [2] for CRC, in patients who are negative for distant metastases, the stage depends on the degree of invasion into the intestinal wall (pT stage [1][2][3][4] and the number of regional lymph nodes positive for metastases [3]. The final substage of the invasion in the intestinal wall is peritoneal involvement or serosal invasion and is labelled stage pT4. It is an independent prognostic factor in CRC and is associated with increased risk of intraperitoneal spread, recurrence and decreased overall survival [4][5][6][7][8]. While several studies have investigated mechanisms leading to peritoneal carcinomatosis [9,10], serosal invasion is still poorly understood. Due to its negative impact on prognosis, it is important to understand the mechanisms or regulators of direct cancer invasion of the serosa. Several pathogenetic mechanisms, including epithelial-mesenchymal transition (EMT), could be involved.
EMT is a process by which immotile epithelial cells undergo a deep molecular reprogramming, gradually losing their epithelial features and acquiring a migratory and invasive mesenchymal phenotype [11][12][13]. Despite extensive research, the exact role of EMT in CRC remains controversial. Previous studies have indicated that EMT plays an important role in CRC, in both its development and progression [14][15][16][17][18]. Furthermore, EMT is believed to be one of the key processes in the development of metastases in CRC [17,[19][20][21]. Our past results revealed that EMT plays a role only in its partial "form" in the development, progression and metastasis of CRC [16,17]. However, there are limited data on the involvement of EMT in invasion through different layers of the bowel wall and finally through the peritoneal membranes, i.e., serosal invasion.
The aim of our study was to investigate the involvement of EMT in local progression of CRC with an emphasis on serosal invasion. Our approach is based on the results of our previous studies showing that downregulation of all five members of the miR-200 family is a key feature and one of the most useful molecular markers of EMT [16,17,22]. We therefore analysed the expression of the miR-200 family and their target genes in CRC in low pT stages (pT2-3) and compared it with advanced CRC, showing serosal invasion (pT4a).

Tissue Samples
Tissue samples from 34 patients with CRC were included in the study. Patients treated by radiotherapy, chemotherapy or biologic drugs prior to surgery were excluded from the study. Samples of CRC and corresponding normal mucosa were collected retrospectively from the archives of the Institute of Pathology, Faculty of Medicine, University of Ljubljana. Tissue samples were fixed in 10% buffered formalin and embedded in paraffin (FFPE) for the purpose of routine histopathologic examination. On the basis of histopathological features, the samples were divided into two groups: patients without serosal invasion (CRC pT2-T3) and patients with serosal invasion (CRC pT4a). Serosal invasion was defined according to the International Collaboration on Cancer Reporting (ICCR), as a tumour infiltrating the bowel wall and breaching the visceral peritoneum, with tumour cells either visible on the visceral peritoneal surface or separated from the peritoneal surface by inflammatory cells only. A peritoneal or omental deposit that was discontinuous from the primary mass was defined as metastatic disease (M1c) and therefore not included in the study [23,24].

RNA Isolation from FFPE Tissue Slides
Tissue samples were cut at 10 µm from FFPE tissue blocks, andfour 10 µm sections were used for the isolation procedure. For cases with peritoneal involvement (pT4a), the representative tissue block containing clear peritoneal invasion was selected.
AllPrep DNA/RNA FFPE kit (Qiagen, Hilden, Germany) was used for total RNA isolation according to the manufacturer's instructions. Concentration was measured using the spectrophotometer ND-1000 (Nanodrop, Thermo Fisher Scientific, Waltham, MA, USA) at the wavelengths 260, 280 and 230 nm.
Reverse transcription (as described under Section 2.4.1) followed by amplification of the GAPDH, a housekeeping gene (amplicon length is 100 bp), using qPCR and Sybr Green technology, was performed as quality control. All of the samples included in the current study had passed amplification of GAPDH, and those that did not amplify were not included in the study.

Expression of microRNAs Family miR-200 Using qPCR
Expression of miR-200 was analysed using TaqMan probes (Thermo Fisher). Prior to qPCR, two separated pools of RNA samples were created. One pool was obtained from tumour samples and the other was obtained from corresponding normal mucosa. After reverse transcription, the cDNA was diluted from 4-fold dilution to 1024-fold dilution (5 steps), and the probes were tested for qPCR efficiency. All the reactions for the efficiency of qPCR for each of the specific probe were performed in triplicate on RotorGene Q (Qiagen, Hilden, Germany). 2.3.2. Quantitative Real-Time PCR (qPCR) of miRNAs miRNAs expression was assessed in 10 µL qPCR master mix, containing 5.0 µL TaqMan 2x FastStart Essential DNA Probe Master (Roche, Basel, Switzerland), 0.5 µL TaqMan assay and 4.5 µL RT products diluted 100-fold. All qPCR reactions were performed in duplicates as follows: initial denaturation at 95 • C for 10 min, and 40 cycles for 15 s at 95 • C (denaturation) and for 60 s at 60 • C (primers annealing and elongation). qPCR was performed using RotorGene Q (Qiagen, Hilden, Germany) with signal collected at the endpoint of every cycle.

Expression of Target Genes of miR-200 Family Using qPCR
qPCR was used for the analysis of mRNA expression of protein-coding genes. A pre-designed mixture of probes and primers based on the TaqMan methodology (Thermo Fisher) was used for target mRNA expression. Accordingly, for selected genes, we chose TaqMan primers and probes that amplify and detect PCR products less than 100 bp long. Again, the same two pools of RNA as described above (Section 2.3), obtained from normal mucosa and from CRC, were used to test efficiency. After RT (Section 2.4.1) and PreAmp (Section 2.4.2), the pre-amplified cDNA was diluted, ranging from 5-fold dilution to 625-fold dilution (4 steps), and all of the used probes were tested for qPCR efficiency. All the reactions were performed in triplicate on RotorGene Q (Qiagen, Hilden, Germany). The mRNAs CDKN1B, ONECUT2, PTPN13, RND3, SOX2, TGFB2 and ZEB2, which are proven targets of the miR-200 family in CRC and were expressed in our previous study (Rankovic, 2019), were analysed relatively to the geometric mean of IPO8 and B2M.

Quantitative Real-Time PCR (qPCR) of mRNAs
The resulting PreAmp reaction was diluted 5-fold. The 10 µL reaction volume of qPCR contained 5.0 µL of 2× FastStart Essential DNA Probe Master Mix (Roche, Basel, Switzerland) and TaqMan probe, 0.5 µL of specific TaqMan probe and 4.5 µL of diluted PreAmp reaction. Thermal conditions were as follows: 50 • C for 2 min, initial denaturation at 95 • C for 10 min and 40 cycles of denaturation at 95 • C for 15 s and annealing at 60 • C for 1 min. All qPCR analyses were performed in duplicates on Rotor Gene Q (Qiagen, Hilden, Germany). The signal was collected at the endpoint of each cycle.

Statistical Analysis
First, all Cqs were corrected for PCR efficiencies. The results were presented as relative gene expression. The Cq of the gene of interest (GOI, Ct GOI ) was calculated relative to the geometric mean of the Cqs of reference genes (RGs, Ct RG ), and named ∆Cq. Statistical analysis of data was performed using SPSS version 27 (SPSS Inc., Chicago, IL, USA). All analysed differences were considered significant at p < 0.05. The Wilcoxon rank test and ∆Cq were used to calculate statistically significant expression differences of mRNAs and miRNAs between tumours and adjacent normal tissue. For statistical comparison of relative expression of mRNA and miRNA between independent groups of samples, the Mann-Whitney test was used with either ∆Cq or ∆∆Cq. The Spearman correlation coefficient was calculated to determine whether there was any association between status T and expression of selected miRNAs and mRNAs.

Patients and Tissue Samples
There were 34 patients with different stages of CRC included in the study. The pTNM stage was determined after resection according to international guidelines [25]. In 16 patients, serosal invasion was confirmed histologically (Figure 1). The clinicopathologic characteristics of the patients are presented in Table 1.

Expression of miR-200 Family and Its Target Genes in Carcinoma with and without Serosal Invasion Compared to Normal Mucosa
We observed significant upregulation of miR-200b (p = 0.003) and ONECUT2 (p = 0.018) in the tumour tissue of CRC pT3 compared to the corresponding normal mucosa, as well as downregulation of CDKN1B (p = 0.013). Similarly, we observed significant upregulation of miR-200b (p = 0.002) and ONECUT2 (p = 0.028) in CRC pT4a tumour tissue  1 According to pathologic tumour-node-metastasis classification [25]. Abbreviations: RC, right colon; LC, left colon; RSC, rectosigmoid colon.

Expression of miR-200 Family and Its Target Genes in Carcinoma with and without Serosal Invasion Compared to Normal Mucosa
We observed significant upregulation of miR-200b (p = 0.003) and ONECUT2 (p = 0.018) in the tumour tissue of CRC pT3 compared to the corresponding normal mucosa, as well as downregulation of CDKN1B (p = 0.013). Similarly, we observed significant upregulation of miR-200b (p = 0.002) and ONECUT2 (p = 0.028) in CRC pT4a tumour tissue with serosal invasion compared to the corresponding normal mucosa. The results are summarized in Figure 2. No other miRNA or mRNA was differentially expressed.

Comparison of Expression of miRNAs and Its Target Genes in Carcinoma with Serosal Invasion Compared to Carcinoma without Serosal Invasion
To evaluate the contribution of EMT in the invasion of cancer cells through serosal membranes, we further compared the expression of the analysed transcripts between CRC pT3 and CRC pT4a. We observed all members of the miR-200 family to be downregulated in pT4a; however, only miR-200b reached statistically significant downregulation (p = 0.003). Furthermore, we also observed significant downregulation for PTPN13 (p = 0.037) and ZEB2 (p = 0.017) in CRC pT4a compared to CRC pT3. The results are summarized in Figure 3. We further compared the expression of the analysed transcripts in all normal mucosa samples (n = 34) with CRC T3 (n = 15) and CRC pT4a (n = 16). Due to the small number of patients with CRC pT2 (n = 3), we did not perform statistical analysis in those samples since it would have given us unreliable results. We observed statistically significant upregulation for miR-200b (p < 0.001), miR-429 (p = 0.021), ONECUT2 (p < 0.001) and PTPN13 (p = 0.030) in CRC pT3. Comparing all normal samples to CRC pT4a, upregulation of ONECUT2 (p < 0.001) and downregulation of miR-200a (p = 0.005), CDKN1B (p = 0.028) and SOX2 (p = 0.022) were observed.
These data suggest that miR-200b and ONECUT2 are important contributors in local CRC spread.

Comparison of Expression of miRNAs and Its Target Genes in Carcinoma with Serosal Invasion Compared to Carcinoma without Serosal Invasion
To evaluate the contribution of EMT in the invasion of cancer cells through serosal membranes, we further compared the expression of the analysed transcripts between CRC pT3 and CRC pT4a. We observed all members of the miR-200 family to be downregulated in pT4a; however, only miR-200b reached statistically significant downregulation (p = 0.003). Furthermore, we also observed significant downregulation for PTPN13 (p = 0.037) and ZEB2 (p = 0.017) in CRC pT4a compared to CRC pT3. The results are summarized in Figure 3.

Correlations between miRNAs, Their Target Genes and Status T
Association analysis showed a correlation between the increasing T stage (including adjacent normal mucosa as T0) and the expression of

Discussion
Serosal invasion in CRC is one of the well-recognized adverse prognostic features associated with an increased risk for tumour spread and recurrence; however, its mechanisms are still not completely elucidated. In the present study, we analysed the expression of the miR-200 family and their target genes to investigate the possible role of EMT in local tumour spread through different layers of the bowel wall, focusing on serosal invasion. Our previous studies support the hypothesis that partial EMT is induced during the progression from normal colon mucosa to adenoma and carcinoma and that partial MET is involved in CRC metastases (12,13). We therefore hypothesized that partial EMT might also be involved in local tumour spread. There are two main findings in the present study. First, we observed downregulation of miR-200b, PTPN13 and ZEB2 in CRC with serosal invasion (pT4a) compared to pT3 tumours. Second, we observed upregulation of miR-200b and ONECUT2 in CRC compared to normal mucosa in both pT3 and pT4a CRC, downregulation of CDKN1B in CRC pT3 and upregulation of ONECUT2 with increasing status T.
Expression of miR-200b was downregulated in CRC with serosal invasion compared to locally less advanced CRC. Indeed, several studies have shown that downregulation of miR-200b is associated with the progression of CRC. In a large series of 1141 CRC cases, Slattey et al. found miR-200b to be decreased in advanced stages of CRC [26]. The expression of miR-200b was significantly downregulated at the invasive front of tumours compared to normal and core tumour tissues [14,15,17], suggesting that downregulation of miR-200b and the subsequent upregulation of their target genes promotes a switch in cell phenotype, resulting in the increased invasiveness and metastatic potential of tumour cells.
However, miR-200b was upregulated in all the tumour samples investigated compared to normal mucosa. Similar to our results, Diaz et al. and Pan et al. demonstrated miR-200b upregulation in tumour tissue versus normal adjacent tissue [27,28]. It has been shown that miR-200b stimulates tumour growth by targeting and negatively regulating CDKN1B, a negative regulator of the cell cycle, in TGFBR2-null CRC cell lines [29]. This is in accordance with our study that showed downregulation of CDKN1B in CRC compared to normal mucosa. Fu et al. showed that both CDKN1B and RND3 are the direct targets of miR-200b. While miR-200b-induced CDKN1B degradation promotes cancer cell proliferation, reduced expression of miR-200b was accompanied by an increased expression of RND3, an important regulator of cell migration [29,30]. It is possible that during the early stages of carcinogenesis, upregulation of miR-200b has a role in proliferation by inhibiting CDKN1B, while later, during the progression of CRC from pT3 to pT4a, downregulation of miR-200b promotes invasiveness and migration, i.e., serosal membrane invasion by targeting other yet unidentified genes. It has been shown that in addition to EMT, miR-200b can also regulate tumour proliferation, apoptosis, invasion, migration and chemotherapy resistance through the WNT/β-catenin pathway, MAPK signalling pathway, PI3K/Akt pathway and others [31].
We observed downregulation of PTPN13 and ZEB2 in CRC with serosal invasion (pT4a) compared to CRC pT3. ZEB2, a target of all miRNAs of the miR-200 family, is an EMT-inducing transcription factor, previously shown to be associated with the progression of CRC and the risk of distant but not local recurrence [32,33]. Our previous study showed ZEB2 to be downregulated in cancers with nodal metastases compared with those without [17]. Our present study could not confirm the involvement of ZEB2 in local CRC spreading, including serosal invasion. Therefore, other EMT transcription factors might be involved in the progression of CRC. The answer might be in downregulation of PTPN13.
Our present study showed upregulation of PTPN13 in CRC; however, it was lower in samples with serosal invasion (pT4a) than in those without (pT3). PTPN13 (also called FAP-1) is a non-receptor protein tyrosine phosphatase that acts in a number of important growth and apoptosis pathways [34]. PTPN13 negatively regulates EMT by inhibiting Slug and Snail, two main EMT transcription factors and also through cell junction stabilization via its positive role in desmosome formation [35,36]. Regarding breast cancer animal model and cell lines, Hamyeh et al. suggested that PTPN13 inhibits tumour aggressiveness through a SNAIL-and ZEB-independent MET-like transition and cell junction stabilization [36]. Our results thus indicate a tumour suppressor role of PTPN13 in the early stages of CRC growth, while in serosal invasion it may act as a tumour promoter via activation of Slug and Snail and inhibition of FAS-induced apoptosis [37][38][39].
Compared with healthy tissues, PTPN13 expression is decreased in different types of cancer [34]. Accordingly, its expression is linked to less aggressive tumours and better patient survival [34]. Similar to our results, a recent study on the diagnostic and prognostic role of PTPN family members in digestive tract cancers using Oncomine, Ualcan and TCGA data found the expression of PTPN13 in CRC to be upregulated compared to normal samples. The expression did not correlate with TNM stage and overall survival [40]. We found no correlation between PTPN13 and its regulator miR-200c. In addition to posttranscriptional regulation of PTPN13 in CRC by miRNAs [41,42], its expression might be regulated through promoter hypermethylation or loss of heterozygosity [43][44][45][46].
We observed another interesting result. Although not differentially expressed between pT3 and pT4a, expression of ONECUT2 increased with the pT stage and might be thus indirectly related to local tumour spread. ONECUT2 is a relatively new member of the ONE-CUT transcription factor family that can widely regulate protein expression associated with cell proliferation, migration, adhesion, differentiation and cell material metabolism [47]. The role of ONECUT2 in cancer progression has been found to be elevated in different types of cancer, including CRC [16,[48][49][50][51][52]. Chen et al. showed that the expression of ONECUT2 correlated with poor prognosis in patients with gastric cancer. In addition, similar to our results in CRC, they noted a correlation of ONECUT2 with tumour stage [50]. In contrast, levels of ONECUT2 were found to be lower in breast cancer, and patients with low expression levels showed worse survival than those with high expression levels of ONECUT2 [53]. The studies involving ONECUT2 expression and its regulation in CRC are limited. Using CRC cell lines and tumour tissues, Sun et al. have shown that miR-429 inhibits cell growth and invasion by targeting ONECUT2 [52]. However, we were not able to confirm a correlation between miR-429 and ONECUT2 expression. Recent studies described novel ONECUT2 regulatory mechanisms. In CRC, circular RNA (circ_0084615) serves as a sponge for miR-599, which directly binds to ONECUT2 [54]. Furthermore, studies on CRC cells showed that by silencing circ_0084615, EMT-related markers ZEB2 and vimentin are reduced and E-cadherin is elevated. It has been suggested that in addition to miRNA and circRNAs, the expression of ONECUT2 can be regulated by DNA hypermethylation [55] and even hypomehylation [56]. Recently, a single-nucleotide polymorphism in ONECUT2 gene was identified as strong predictive marker for CRC risk [57]. However, while being upregulated in all stages of the cancer progression through the bowel wall, ONECUT2 does not seem to contribute significantly to the invasion through the serosal membranes, but rather is responsible for the continuous growth of tumours.
There are several advantages and limitations of our study. As the purpose of the study was to compare tumours in different pT stages, we used the whole tissue sections of the resected tumours. In samples with serosal invasion, in addition to EMT, mesothelial-tomesenchymal transition contributes to gene expression results [58]. This limitation could be resolved using laser-captured microdissection to include only representative tumour cells in the analysis. Another limitation of our study is the small number of patients, particularly CRC pT2. Nucleic acids from FFPE tissue blocks are difficult to analyse, and specialized methodologies are often necessary. One of the reasons for the relatively small sample size is that after initial RNA isolation, only samples that passed the quality control were included in the study. Finally, we analysed only two EMT transcription factors, ZEB1 and ZEB2, ZEB1 being below detection limit or even not expressed. However, although the current study has several limitations, there are limited data on molecular analyses in CRC with serosal invasion compared to those without. We therefore believe that PTPN13, ZEB2 and ONECUT2 might be promising targets for further research.

Conclusions
Our results suggest that decreased expression of miR-200b, ZEB2 and PTPN13 is involved in local tumour spread of CRC, particularly in serosal invasion (pT4a). We also observed that increasing expression of ONECUT2 is associated with tumour growth and its spread through the bowel wall. Our findings further support that at least certain components of EMT might be involved in serosal invasion in CRC.  Informed Consent Statement: Patient consent was waived due to the following reason. As stated in the approval document, the study is retrospective, observational, performed on tissue samples that were obtained during routine diagnostic/therapeutic procedures and consisted of either excision or resection. Therefore, enough tissue was available for routine analysis and research. Moreover, tissue is still available for any additional analysis in the future. Our State Ethical Committee does not require informed consent from patients in such studies. However, informed consent was obtained before the routine procedure.

Data Availability Statement:
The datasets used and analysed during the current study are available from the corresponding author on reasonable request.

Conflicts of Interest:
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.