AIM: To investigate the prognostic significance of S100A4 expression in colorectal cancer and its correlation with expression of E-cadherin and p53.

METHODS: A cohort of archival formalin-fixed paraffin-embedded specimens was selected from 127 patients with colorectal cancer who underwent surgical resection between April 2000 and March 2004 at the Department of Surgery, Korea University Guro Hospital. The expression of protein S100A4 was evaluated according to the proportion of positively stained cancer cells. In each case, three core biopsies with a diameter of 2 mm were punched out and positioned in a recipient paraffin array block. Four-μm sections of these tissue array blocks were used for immunohistochemical analysis of protein S100A4, E-cadherin, and p53. Clinicopathological data were based on the original histopathologic reports and clinical records of patients.

RESULTS: In normal colorectal mucosa, protein S100A4 immunoreactivity was clearly absent in both cytoplasm and nucleus. However, positive immunoreactivity of protein S100A4 was detected in 45 (35.4%) of the tumor cases. There was no significant association between positive immunoreactivity of protein S100A4 and clinicopathological parameters such as tumor differentiation or TNM stage, and also no correlation between the reactivity and E-cadherin or p53 expression. However, positive immunoreactivity of protein S100A4 was found to be associated with tumor recurrence (P = 0.004), and was also associated with significantly worse overall survival in the Kaplan-Meyer survival analysis (P = 0.044). After adjustment for tumor differentiation, tumor depth and nodal status, however, it failed to achieve statistical significance (P = 0.067).

CONCLUSION: The expression of protein S100A4 is associated with tumor recurrence and poor overall survival in patients with colorectal cancer.

Although the last decade has brought significant improvements in the disease-free and overall survival of colorectal cancer (CRC) patients; achieved largely by more accurate staging of disease, an improved and expanded role of surgery and increased number of available chemotherapeutic options, approximately 20% of advanced CRC patients still die of recurrence of the disease[1]. Invasion and metastasis, which are the most life-threatening properties of malignant tumors, result from the interaction between tumor cells and the surrounding tissues. The invasion and metastasis processes themselves consist of pathogenic sequential steps, such as proliferation and detachment of neoplastic cells, invasion to extracellular matrix, angiogenesis, vascular dissemination, lodging in a distant vascular bed, extravasation into the target organ, and proliferation. The activation of many genes and the expression of their products have been involved in this progression[2]. Current conventional staging has a significant impact on survival of CRC patients. However, there is marked variability in outcome that exists within each stage, and certain populations of patients with early recurrence, resistance to chemotherapy and decreased survival cannot be predicted using conventional histopathologic staging. Thus, the identification of molecular factors that have prognostic significance in CRC is essential to improve treatment and outcome[3-5].

Over the past few years, the S100 family of proteins has emerged as an important group with the capacity to promote invasiveness and metastasis of many human neoplasms. In particular, recent studies have established the mechanisms of action of protein S100A4, and indicate its possible prognostic role in human neoplasia[6-8]. However, studies regarding protein S100A4 have mainly been limited to research laboratories. Moreover, the mechanism of action of protein S100A4 in tumors is not fully understood. Therefore, it would be of great interest to find out whether the detection of protein S100A4 has any predictive value, and also whether it may help select patients who require more extensive diagnostic evaluation to rule out metastatic disease and/or more aggressive treatments.

The aims of this study were to investigate immunohistochemically the prognostic significance of protein S100A4 expression in CRC, compared with clinicopathologic parameters and overall survival, and to investigate the correlation between protein S100A4 expression and E-cadherin and p53, which have been suggested as possible targets of protein S100A4.

Calcium binding proteins form a large family involved in numerous functions ranging from the control of cell-cycle progression and cell differentiation to enzyme activation and regulation of muscle contraction[10,11]. The S100 proteins represent one of the largest subfamilies of the calcium binding proteins with at least 19 different members; the degree of homology ranges from 25% to 65%. They were initially characterized as low-molecular weight acidic proteins and named by their solubility in 100% ammonium sulfate (“S100”). S100A4, also known as p9Ka, CAPL, or calvasculin, is a member of the S100 family consisting of 101 amino acids and with a molecular weight of about 11.6 kDa. The corresponding gene, cloned by different groups, is known as mts1 (metastasin), pEL98, 18A2, 42A, and fsp (fibroblast-specific protein)[10-13].

The biologic functions of several S100 proteins in carcinogenesis have not been fully elucidated to date. Recently, however, much interest has focused on S100A4 and some other S100 family members, such as S100A2, S100A6, and S100B, for their potential roles in invasive growth and metastasis of neoplastic diseases. S100A4 or its corresponding mRNA are found at higher levels in metastatic relative to non-metastatic rat[14] and mouse[15] tumor cell lines. Transfection experiments further showed that rodent or human S100A4 can induce a metastatic phenotype in previously non-metastatic rat mammary cells[16,17]. Conversely, antisense S100A4 RNA or anti-S100A4 ribozyme suppressed the metastatic potential of highly metastatic cell lines[18,19]. Moreover, in pilot studies of human colorectal adenocarcinoma specimens, elevated levels of immunohistochemically detected S100A4 are associated with the more malignant carcinomatous regions of the primary tumors and with liver metastases[20]. The tight association between S100A4 expression and metastasis observed in these laboratory analyses has led to a number of studies examining the utility of S100A4 expression as a prognostic marker in human cancers. Protein S100A4 has been shown to be a prognostic marker in a number of human cancers, including breast cancer[21], esophageal-squamous cancer[22], non-small cell lung cancer[23], gastric cancer[24], malignant melanoma[25], prostate cancer[26], and pancreatic cancer[27]. The universality of S100A4 expression in a variety of cancers illustrates the potential use of S100A4 as a marker for tumor metastasis and disease progression.

The purpose of this investigation was to establish clinical significance of the calcium-binding protein, S100A4, in CRC. It was found that 35.4% of CRC specimens were stained strongly by the polyclonal antibodies against protein S100A4, in concordance with earlier reports[28,29]. The staining in specimens is not restricted to only carcinoma cells, because highly expressed levels are also detected in normal tissues, in particular, smooth muscle cells, endothelial cells of both arteries and veins, and some reactive fibroblast-like cells and lymphocytes[30]. However, the present study was undertaken on only carcinoma cells.

In this study, positive expression of protein S100A4 was associated with tumor recurrence, in accordance with previous study[28]. This result suggests that the protein S100A4 may play a role in predicting a patient subgroup which would show more unfavorable outcome, thus leading us to substage-oriented tailored therapy with more intensive treatment and more strict follow-up surveillance.

This study showed that the overall survival of CRC patients who had immunohistochemically detectable levels of protein S100A4 was significantly worse than those CRC patients with negative expression of protein S100A4 according to univariate analysis. Because S100A4 was first discovered as a metastasis-inducing protein in experimental models[14-19], and metastasis is the major event responsible for death in patients of CRC, it is quite possible that protein S100A4 causes earlier deaths by its ability to induce metastasis in human CRC. Although it failed to achieve statistical significance in multivariate analysis, the present result suggests a need for further and larger studies to investigate the role of protein S100A4 expression in CRC.

As a typical member of the S100 family, S100A4 exerts dual functions, both intracellular and extracellular. Intracellularly, it interacts with and functionally modifies the tumor suppressor protein p53, non-muscle myosin II, and liprin β1[12,13]. S100A4 interacts with the C terminus of p53 and inhibits protein kinase C (PKC) phosphorylation of the tumor suppressor in vitro. Likewise, the interaction between p53 and S100A4 inhibits p53 from binding to its consensus DNA-binding sequence[31]; thus it was expected that S100A4 would be a general inhibitor of p53 function. It has been suggested that a complex of S100A4 with p53 and the sequestration of p53 may result in stimulation of the cells to enter the S phase by abrogating the control functions of p53 at the G1-S checkpoint[8,31,32]. However, as shown in this study, this possibility was difficult to prove by immunohistochemical analysis of these two proteins in CRC. An examination of p53-regulated genes in S100A4-expressing cells indicates that the expression of several genes are up-regulated (e.g. bax); other genes are down-regulated initially and then later up-regulated (e.g. mdm2), and some genes are inhibited (e.g. p21, thrombospondin-1)[31]. These opposite effects of S100A4 expression on p53-regulated genes could explain why there was no correlation between S100A4 and p53, notwithstanding the potential interaction of these two proteins.

Another possible mechanism of action of S100A4 in carcinogenesis is cytoskeletal dysregulation by down-regulation of E-cadherin induced by protein S100A4. E-cadherin is a member of the large cadherin superfamily. It is the predominant intercellular adhesion molecule expressed by intestinal epithelial cells, and functions to mediate epithelial cell-cell adhesion and maintain the integrity of the epithelium[33-35]. The expressions of E-cadherin and protein S100A4 in two mouse tumor cell lines were found to be inversely regulated, and transfection experiments showed a reciprocal down-regulation of both molecules, suggesting that the invasiveness of tumors expressing protein S100A4 may be at least partially induced by the abrogation of E-cadherin expression[36]. A similar mechanism has also been postulated in humans, on the basis of immunohistochemical analysis of both proteins in a series of non-small cell lung cancer[23] and gastric cancer[24]; an inverse correlation of E-cadherin and protein S100A4 expression was demonstrated. In this study, we attempted to immunohistochemically establish an inverse correlation between the expression of protein S100A4 and E-cadherin in CRC; however, data failed to prove the relationship (Kendall’s Tau-b correlation coefficient = -0.035, P = 0.681). Nevertheless, it is quite possible that different antibodies against protein S100A4, different cancer tissue, and small numbers enrolled in this study might have contributed to this difference.

In conclusion, in the present retrospective study, positive immunoreactivity of protein S100A4 is closely associated with cancer recurrence. However, there is no correlation between the expression of protein S100A4 and E-cadherin or p53. The overall survival for patients with CRC expressing immunohistochemically detectable levels of protein S100A4 is significantly worse than for those patients with CRC considered negative for S100A4. Furthermore, protein S100A4 shows borderline tendency toward being a prognostic marker in multivariate regression analysis. Although these results suggest that protein S100A4 could be a useful biologic predictor of cancer recurrence and poor outcome, subsequent prospective, large-scale studies are required to confirm its importance and relevance in the invasive potential of human CRC.