Impact of PINCH expression on survival in colorectal cancer patients

PINCH, particularly interesting new cystein-histidine-rich protein, was first identified in 1994 as an evolutionary conserved protein belonging to the LIM family, consisting of five LIM domains [1]. A LIM domain mediates protein interactions and consists of a protein-binding motif with a specific three-dimensional structure comprising a double zinc finger [2]. The PINCH gene is located to chromosome 2q12.2 and encodes a protein that functions as an adaptor protein [3]. PINCH is known to directly associate with two proteins: integrin-linked kinase (ILK) [4] and Nck-2 [5]. ILK is an intracellular serine/threonine protein kinase that is a constituent of integrin-mediated cell-matrix focal adhesions, structures that mediate cell adhesion and signal transduction between the extracellular matrix and the intracellular compartment [6]. Nck-2 is an adaptor protein capable of recognizing several key components of growth factor receptor kinase-signalling pathways [5]. PINCH binds Nck-2 and ILK by means of two separate LIM domains, LIM1 for ILK [4] and LIM4 for Nck-2 [5], and forms a multiprotein complex with these two proteins [4]. Thus, PINCH could provide a connection between the growth factor receptor and integrin-signalling pathways by mediating the interaction between Nck-2 and ILK.

The tumour-associated stroma is important in facilitating cancer growth and invasion, and PINCH expression has been shown to be up-regulated in tumour-associated stroma of several common cancer types, especially at the tumour invasive margin [7]. This indicates that PINCH could be involved in tumour progression. Further, PINCH functions in the integrin and growth factor signalling pathways, both important mediators of the tumour-stromal interaction. In concordance with the theory of PINCH promoting tumour progression, it has been shown that a high stromal expression of PINCH at the tumour invasive margin is related to worse prognosis in colorectal cancer [8].

In this study we further investigated the relationship of PINCH expression with survival and clinicopathological variables in colorectal cancer patients and found that PINCH expression at the tumour invasive margin or adjacent normal mucosa is independently related to prognosis. We also studied the effect of PINCH expression on outcome of adjuvant chemotherapy, and found that high PINCH expression could be related to treatment outcome. However, we did not find it to be an independent marker.

Predictive and prognostic markers in colorectal cancer patients have been the subject of intense research. The determination of prognosis predominantly relies on the histopathological examination, although there are certainly other factors influencing survival. Approaches are being made to improve prognostic methods, including analyzing additional histopathological factors and molecular and genetical markers. Although these markers are promising they are not yet routinely used. Potential markers include, for example, allelic imbalances, chromosomal instability, expression of oncogenes, loss of tumour suppressor genes, markers of proliferation, angiogenesis, inflammation and cell adhesion as well as genes involved in the response to chemo- and radiotherapy.

In this study we have shown that the intensity of stromal immunohistochemical staining for PINCH was increased from normal mucosa to primary tumour and from primary tumour to lymph node metastasis in colorectal cancer. We have also shown that strong staining for PINCH in normal adjacent mucosa was related to worse survival. Furthermore, in poorly differentiated tumours, PINCH staining at the tumour invasive margin was significantly related to survival, while in better differentiated tumours it was not. The pattern of PINCH staining increasing from normal tissue to tumour and from tumour to metastasis has been seen in previous studies [8] and indicates that PINCH is involved in tumour progression and invasion. PINCH up-regulation in the stroma of oral squamos cell carcinoma also predicts lymph node metastasis, further implicating PINCH in invasion and metastasis [9]. It has previously been shown that PINCH staining at the tumour invasive margin is an independent prognostic marker in colorectal cancer [8], which we have also seen a tendency towards in the current study. This further supports the idea that PINCH, through involvement in the tumour-stromal interaction, promotes tumour invasiveness. The finding that the intensity of PINCH staining in poorly differentiated tumours was significantly related to survival while in better differentiated tumours it was not suggests that the effect of PINCH expression on the aggressiveness of the tumour could be dependent on differentiation status. It seems as the impact of PINCH expression on survival is limited in well differentiated tumours. As it has previously been seen [8], we also found strong PINCH expression to be related to better differentiation. Poor differentiation is correlated to worse survival [10], nevertheless strong expression of PINCH at the tumour invasive margin seems to be related to a worse prognosis. Further, PINCH staining in the adjacent normal mucosa was found to be related to survival, which has not been reported previously. Although histologically the mucosa adjacent to a tumour appears normal, it is from colorectal cancer patients and may differ biologically from the normal mucosa of a healthy individual. Because of the proximity of the adjacent normal mucosa with the invasive margin of the tumour, it is possible that this seemingly normal tissue is already involved in signalling and interactions with the tumour. The finding that PINCH expression in the adjacent normal mucosa is related to survival implicates PINCH in the biological changes occurring in the mucosa around a tumour.

The interaction between tumour and stroma has been recognized as an important factor influencing tumour growth and progression [11, 12]. PINCH is involved in several signalling pathways important to the tumour-stromal interaction by functioning as an adaptor protein in the integrin- and growth factor signalling pathways [5, 13]. Loss of PINCH in c. elegans results in a phenotype identical to integrin null mutants [3], indicating that PINCH is required for integrin signalling, one of the key components in the cell-stromal interaction. Integrin-mediated adhesion to the extracellular matrix results in the stimulation of various intracellular pathways important in the regulation of cell attachment, migration, proliferation [14, 15], survival and apoptosis [16]. PINCH directly associates with the proteins integrin-linked kinase (ILK) and Nck-2, and assembly of the multi-protein complex containing PINCH, ILK and Nck-2 prevents the proteolytic degradation of the proteins that are part of the complex [13]. Therefore, an increase in PINCH expression will increase the stability of the other proteins in the complex, thus increasing ILK and Nck-2 signalling. Possibly, the aggressive behaviour of tumours with a high stromal expression of PINCH can be explained by an up-regulation of the signalling pathways that are dependent on the adaptor function of PINCH. It is not clear which factors during tumour progression that influence the expression of PINCH itself. Studies of epithelial-to-mesenchymal transition (EMT) in renal tubuli have shown PINCH mRNA expression to be increased in response to transforming growth factor (TGF)-β1 [17], a known inducer of EMT in several biological systems, including renal tubuli [18]. TGF-β1 has been implicated in the progression of colorectal cancer [19]; therefore it is possible that TGF-β1 could be involved in the regulation of PINCH in colorectal tumours.

ILK is a serine/threonine protein kinase that associates with the cytoplasmic domain of the integrin β1 and β3 subunits, thereby regulating integrin mediated signal transduction [6]. PINCH and ILK have been shown to be indispensable for proper control of cell shape change, motility and survival [13]. ILK is activated through the PI3K-signalling pathway and functions in cell survival by regulating the PKB/Akt signalling pathway [20]. Up-regulation of ILK in tumour cells is observed in several types of cancer, and is associated with tumour stage, metastasis and worse prognosis [21‐23]. ILK overexpression promotes an invasive phenotype, induces in vivo tumourigenesis [24], anchorage independent cell growth [6] and anchorage independent cell cycle progression [25]. Clearly, ILK plays an oncogenic role when overexpressed in epithelial cells. However, in our study we found PINCH to be overexpressed in the stroma, therefore ILK signalling may be increased in stromal cells. The question is therefore how increased ILK signalling in stromal cells can influence the aggressiveness of the tumour.

The fibroblasts of the tumour associated stroma can affect tumour development by secreting soluble factors such as vascular endothelial growth factor (VEGF) [26] and matrix metalloproteinases (MMP:s) [27]. ILK contributes to tumour progression by increasing the expression of VEGF via the activation of PKB/Akt and HIF-1α, thereby stimulating angiogenesis [28]. Angiogenesis is a prerequisite for tumour growth and progression [29], and the levels of angiogenetic factors such as VEGF are related to prognosis in several types of cancer, including colorectal cancer [30]. Angiogenesis is speculated to contribute to metastasis by increasing the number of leaky vessels into which the tumour cells can intravasate [31] and the VEGF expression has been shown to be related to tumour stage [32], lymph node metastasis [33, 34], distant metastasis [34] and depth of tumour invasion [32]. Furthermore, at our laboratory, a recent study (unpublished data) of rectal cancer patients showed that PINCH expression was related to blood- and lymph vessel density, implicating PINCH as a regulator of angiogenesis.

Further contributing to an invasive phenotype, ILK overexpression stimulates the expression of matrix metalloproteinase -9 (MMP-9) [35]. Metalloproteinases are zinc-dependent endopeptidases that degrade components of the extracellular matrix [36], a process that is necessary for angiogenesis, tumour invasion and metastasis to occur [37]. Increased expression of MMP:s is associated with tumour invasion [38, 39] and metastasis [40, 41]. In particular, MMP-9 plays a key role in the degradation of several components of the ECM, including type IV, V and XI collagen, gelatin and laminin [42], and has been found to be over-expressed in several types of cancer and to be associated with a worse prognosis [43, 44].

Remodelling of the extracellular matrix is a necessary process in order for a tumour to grow and progress [45]. The remodelling consists not only of the breakdown of ECM, but also the neosynthesis of certain ECM components such as fibronectin [46]. Assembly of fibronectin is regulated by ILK in a process requiring PINCH [24]. Therefore, an increase in PINCH expression at the tumour invasive margin could be associated with enhanced assembly of fibronectin matrix. Since the fibronectin matrix has a major impact on cell adhesion, migration, and cell growth [47, 48], an increased assembly could stimulate the migratory and invasive capacity of the tumour cells. A high expression of fibronectin in tumour stroma has been found to be correlated to lymph node metastasis, proliferation and worse survival [49]. The interaction between fibronectin and tumour cells activates various signalling pathways involved in tumour progression, leading for example to the increased expression of metalloproteinases MMP-2 and -9 [50, 51].

ILK is connected to the growth factor signalling pathways through PINCH, since PINCH forms a multiprotein complex with ILK and the adaptor protein Nck-2 [5]. Nck-2 recognizes several key components of growth factor receptor kinase-signalling pathways, including EGF receptors, PDGF receptor-β and insulin receptor substrate-1(IRS-1) [5]. Thus, PINCH provides a physical connection between the growth factor receptor-signalling pathways and the integrin-mediated pathways by connecting ILK and the integrins with EGF-and PDGF receptors and IRS-1 [4]. Through the complex formation with Nck-2, increased PINCH expression may be associated with an up-regulation of the growth factor signalling pathways. Growth factors are important regulators of the tumour-stromal interaction, and for some carcinomas an increase in growth factor receptors in stromal cells is thought to be an essential part in tumour to stroma signalling and hence tumour growth and progression [52, 53].

Since PINCH expression is related to tumour progression and prognosis, we were interested in investigating whether PINCH was also related to response to treatment. In our group of patients, 27 patients received adjuvant chemotherapy. In the univariate analysis we found chemotherapy to be significantly related to survival in patients with weak stromal staining for PINCH at the invasive margin, while in patients with strong staining there was no relationship between chemotherapy and survival. This indicates that PINCH could be one factor influencing the outcome of adjuvant chemotherapy. However, in the multivariate analysis we found no significant correlation between chemotherapy outcome and PINCH expression. Possibly, this could be due to the relationship of PINCH expression with inflammatory infiltration and differentiation grade, known prognostic factors in colorectal cancer [10, 54]. Further, the relatively low number of patients may contribute to the lack of significant results. More results would require a prospective setting; as with many studies of prognostic or predictive markers the major weakness of this study is the use of retrospective, non-randomized material as well as the low number of the patients.

Stromal immunohistochemical staining for PINCH in normal mucosa adjacent to a tumour was found to be related to survival in colorectal cancer patients. Although appearing histologically normal, biological changes may occur in the mucosa near a tumour, possibly affecting the behaviour of the tumour. Further, PINCH staining at the tumour invasive margin was related to survival in poorly differentiated tumours but not in better differentiated tumours, indicating that the impact of PINCH on prognosis is dependent on differentiation status. Regarding treatment, adjuvant chemotherapy was significantly related to survival in patients with weak stromal staining for PINCH, but not in patients with strong staining. Thus, patients with weak PINCH staining seem to benefit more from adjuvant chemotherapy than patients with strong staining. However, these results could not be confirmed in a multivariate analysis. Taken together, our results indicate that PINCH could be one factor influencing the prognosis in colorectal cancer patients.