Association of Clinicopathological Factors With MMP13 (rs2252070) Gene Polymorphism in Swedish Patients With Colorectal Cancer

Discussion

Mucinous cancer is an aggressive subtype of CRC and the factors involved in the development of mucinous CRC are not yet known (16).

PNI is defined as tumor growth in, around and through nerves, and has been identified as an important pathological feature of a number of malignancies, including tumors of the pancreas, colon, rectum, prostate and stomach (17). PNI has been shown to be a marker of poor prognosis and reduced survival, including in CRC (18). The molecular mechanisms of PNI have not been successfully identified but the migration of tumors into the nerve sheath requires degradation of the ECM, which can be facilitated by MMPs (17).

The lymphatic system is an important metastatic pathway for CRC. LVI is defined as the presence of tumor cells in the lymphatic system and vascular structures and has gained acceptance as a predictor of poor outcome in patients affected by CRC (19, 20).

Accumulating evidence has established that SNPs of MMPs are associated with gastrointestinal cancer, including CRC (11, 13). Several SNPs in genes encoding MMPs influence MMP gene expression, protein production and thereby the enzymatic activity (11).

In the present study, we examined the association of MMP13 rs2252070 gene polymorphism with various clinical features and long-term survival in Swedish patients with CRC. Our data showed an association with mucinous cancer, PNI and LVI, where we found that carrying a G allele has a protective role against these clinical features. Yoon et al. (14) concluded that the A allele in MMP13 rs2252070 is associated with two-fold higher transcriptional activity of the gene compared with that of the G allele. One could speculate that our data regarding mucinous cancer, PNI and LVI reflect the transcriptional activity and therefore the level and activity of MMP13. In that case, however, our data contradict a study by Foda et al. (21) in which the protein expression of MMP13 was found to be significantly higher in non-mucinous carcinomas compared to mucinous carcinomas.

The promoter of MMP13 contains several binding sites for transcription factors e.g., activating protein-1 (AP1), inhibitor of growth family member 2 (ING2) and polyomavirus enhancer activator 3 (PEA3) (10). Besides the regulation of expression by transcription, the activity of MMP13 is controlled by presence of tissue inhibitor of metalloproteinases in the TME (8). Moreover, intrinsic mechanisms, such as changes of expression of oncogenes and tumor suppressors, seem to directly activate MMP13 expression (10, 22). Thus, the activity of MMP13 can be modulated in different ways.

According to the AJCC classification system (15), T stage, i.e., the depth of tumor invasion, is used rather than tumor size defined as the maximum diameter of the horizontal tumor extension. One study suggests that in patients with tumors at the same T stage, the prognosis of those with a larger tumor size is worse than that of patients with a smaller tumor size, which may be due to the larger tumor burden and higher possibility of invading vascular and lymphatic channels by larger tumors (23). The tumor size has been reported to be prognostic in CRC and has an impact on survival (23-25), but it remains controversial.

Previous studies have investigated the relationship between MMP13 and CRC and proposed that MMP13 is involved in tumor progression [reviewed in (10)] and indicate that the MMP13 expression in tumor tissue is significantly higher than in the corresponding normal mucosa (26). However, no association with tumor size has been found (26). In our study, in terms of genotypes of MMP13 rs2252070, we established that the tumor size ≥4 cm was more common in both G allele carriers regarding rectal cancer. To the best of our knowledge, this study is the first to show an association between MMP13 gene polymorphism and tumor size in rectal cancer. The resulting difference here between the colon and rectum may be the obvious difference in molecular carcinogenesis, pathology, and expression of distinguishable genes for these locations (27). The underlying molecular mechanism involved in this process is unclear and requires detailed analyses. Regardless of the mechanism, our findings suggest a role for MMP13 gene polymorphism in modulating tumor size.

CEA is a large glycoprotein involved in cell adhesion and is secreted by a variety of tumors, including CRC (28). Both preoperative and postoperative serum levels of CEA are used in clinical practice as a prognostic factor and to monitor recurrence for patients with CRC. However, controversy exists, and previous studies have noted different effectiveness regarding its associations with stage and prognosis (28-30). In this study, we found that individuals carrying the G/G genotype in MMP13 rs2252070 had a higher preoperative level (≥5 ng/ml) of CEA in serum. Furthermore, we noted that the carriers of the G/G genotype were more likely to have poorly differentiated cancer. One research group demonstrated that preoperative serum levels of CEA did not correlate with tumor differentiation, but multivariate analysis revealed that the preoperative serum level of CEA was a significant independent prognostic factor (31). It remains to be clarified whether our results regarding the G/G genotype are related to the preoperative serum level of CEA and poorly differentiated cancer.

Lymph node involvement is an important variable of the AJCC system (15). The classification of lymph node status, N0-N2, represents tumor aggressiveness and progression of CRC, and is a predictor of worse outcome (32). In a previous report, a positive correlation between MMP13 expression and lymph node status was found (33). In our study, the genotypes of MMP13 rs2252070 were evaluated in a CRC cohort to clarify their role in lymph node status. We observed that the lymph node status by genotype distribution was statistically different, and that the G/G genotype was markedly more common in patients with N2 tumors. The underlying mechanisms should be further investigated.

In the present study, we examined the relationship between MMP13 rs2252070 polymorphism and clinicopathological factors in patients with CRC and demonstrated that the presence of G/G was significantly associated with poor differentiation, higher serum level of CEA and increased lymph node status. Moreover, the presence of a G allele was significantly related to larger tumor size in rectal cancer but had a protective role against mucinous cancer, PNI and LVI. For other parameters, such as age, sex, depth of tumor, TNM stage, tumor location and recurrence, no relation to MMP13 rs2252070 gene polymorphism were identified.

MMP13 rs2252070 (−70A/G) is associated with altered transcriptional activity, with twice as much transcriptional activity with the A allele compared to the G allele in the same position (14). It is possible that the gene expression of MMP13 is concomitant with other gene–gene or gene–environmental interactions. The effects of the G/G genotype may not be attributed to a reduced level of MMP13 per se, but a possible change in the balance between different MMPs. MMP13 has as a central role in the MMP activation cascade by the activation of other MMPs, such as MMP2 and MMP9, which has been suggested to be a pivotal event in the development of CRC (8, 10). However, much remains unknown about the possible coordination of MMP13 action with other MMP family members during cancer pathogenesis. Tumor necrosis factor-α is an inflammatory mediator, one of several present at high levels in solid tumors and the serum of patients with CRC and is implicated in the progression and metastatic development of CRC (8, 10). A previous study has reported the regulation of MMP13 through the activation of extracellular signal-regulated kinase–nuclear factor B–MMP13 pathway promotes by the chemokine receptor CCR4 resulting in up-regulation of tumor necrosis factor-α in CRC (34).

In CRC development, intrinsic mechanisms include changes of expression of oncogenes and tumor-suppressor genes (2, 3). Suppression of tumor-suppressor gene p53 increases MMP13 expression in squamous cell carcinomas of the head and neck (22). Recently, a study indicated that MMP13 plays a key role in the context of the TWIST family bHLH transcription factor 1–CD44–MMP13 axis in tumor aggressiveness in esophageal squamous cell carcinoma and in epithelial–mesenchymal transition driven tumor progression (35). Further studies may confirm whether the results can be translated to CRC.

Some limitations of this study are worth noting. This study was of an exploratory nature. Factors influencing carcinogenesis such as environmental and lifestyle factors were not considered. The patients were selected from one hospital and therefore the generalizability of our results is limited, and they should be validated in another cohort. Moreover, further studies are required in which a larger number of samples must be examined.

In conclusion, as described in previous studies, there is an involvement of MMP13 in CRC progression and metastasis. The exact mechanisms of its regulation and actions are still not fully understood. The current study is, to our knowledge the first to investigate the relationship between MMP13 rs2252070 and a large number of clinicopathological factors in Swedish patients with CRC.

We noted that MMP13 rs2252070 is a useful predictor of differentiation, serum level of CEA, lymph node status and tumor size in rectal cancer, and the presence of a G allele has a protective role against mucinous cancer, PNI and LVI in CRC. Considering the multifaceted role of MMP13 in CRC, more in-depth research is needed for further understanding of the significance of MMP13 rs2252070 in CRC.