Abstract
Objective:
Prognostic markers for pancreatic ductal adenocarcinoma (PDA) have failed to accurately predict patient prognosis. Recently, interest has developed in the accuracy of integrin-associated PINCH protein expression in human cancers as a predictive marker of tumour status. The goal of this study was to define the expression of PINCH protein in PDA.
Methods:
Human PDA samples and orthotopic tumours from a murine model were analysed by immunohistochemistry for PINCH expression. In the animal model, PINCH expression was compared between primary and metastatic tumours. In the human samples, PINCH expression was correlated with stage, nodal involvement, margin status and overall survival.
Results:
In the murine model, there was greater PINCH expression in metastatic tumours than in primary tumours. In the human PDA samples, greater staining for PINCH in the tumour cells was correlated with higher T status. Additionally, high PINCH expression in the stroma was associated with decreased overall survival.
Conclusions:
Findings of increased PINCH protein in more advanced stages of human PDA, as well as in metastatic tumours in the animal model, support the hypothesis that PINCH is an important controller of cell survival and migration. Additionally, the importance of the differential expression of PINCH in the human tumour and stroma warrants further evaluation.
Introduction
Pancreatic ductal adenocarcinoma (PDA) remains one of the most poorly controlled solid organ malignancies in the USA and is the fourth leading cause of cancer-related deaths.1 Surgical resection remains the primary treatment. Yet, nearly 80% of surgically resected patients with localized pancreatic cancer die as a result of recurrent or metastatic disease.1 Additionally, the pathologic positive margin rate at resection remains 50–75%. These outcomes imply a particularly invasive tumour phenotype.
Recent studies have suggested an important role for the interaction of tumour cells and tumour-associated stroma cells in the regulation of cell–cell adhesion and invasion.2,3 The adhesion of cells to the extracellular matrix (ECM) is essential for cell migration, division and cell–cell signalling.4,5 Variations in this interaction between the tumour and surrounding tissues may facilitate the survival, malignant transformation and invasion of tumour cells.3,6–9 Integrin-associated proteins facilitate this abnormal signalling and control.4,5,9 PINCH, a particularly interesting new cysteine-histidine-rich integrin-associated protein, is an important component of the focal adhesion complexes that link the ECM to the support structures within the cell and is expressed in many human tissues. PINCH expression has been shown to be upregulated in many malignant tissues, including breast, colon, prostate, lung and oral squamous cell carcinomas.10–12 In these tumours PINCH expression localizes to the peri-tumoral stroma cells, particularly at the tumour's invasive edges. Additionally, PINCH expression is correlated with poorer patient prognosis in colon and oral epithelial cancers.10,12 The aggressive nature and strong desmoplastic response in PDA implies that PINCH expression may be an important outcome marker for PDA and may identify which patients are more likely to experience recurrence.
The purpose of this study was to investigate the expression of PINCH in PDA and to determine the association between PINCH expression and tumour aggressiveness or patient prognosis. We initially compared PINCH expression between primary and metastatic tumours in an orthotopic PDA animal model. We subsequently correlated PINCH expression in primary human paraffin-embedded PDA tumours, within both the tumour and the peri-tumoral stroma cells, with disease stage and patient outcomes. We hypothesized that PINCH protein expression would be greater in the peri-tumoral stroma compared with tumour cells and that the degree of PINCH protein expression would correlate with a more invasive tumour or poorer patient prognosis.
Materials and methods
Cell culture
The human pancreatic cancer cell line AsPc1 was obtained from the American Type Culture Collection (Rockville, MD, USA). Cells were maintained in DMEM supplemented with 10% heat-inactivated foetal bovine serum (Gibco, Inc., Grand Island, NY, USA). Cells were cultured at 37 °C in a 5% CO2 incubator. The cell line was transfected with and stably expressed red fluorescent protein (RFP), as previously described.13
Orthotopic tumour production
Male nude mice (NCR-nu/nu) aged 4–6 weeks were utilized for the orthotopic tumour model. All studies were conducted with the approval and guidance of the University of Utah Institutional Animal Care and Use Committee.
For the tumour induction surgery, mice were anesthetized with isoflurane and the tail of the pancreas was exposed through a left subcostal 4-mm incision into the peritoneal cavity.13 Mice received a single sub-capsular injection of 1.5 × 106 RFP-labelled AsPc1 cells suspended in 150 µL of serum-free media (DMEM). The abdomen was closed using two interrupted 6–0 silk sutures closing skin and muscle simultaneously. All procedures were carried out utilizing a 12× Universal S3B microscope (Carl Zeiss, Inc., Thornwood, NY, USA). Tumours were allowed to progress for 8 weeks. At 8 weeks the animals were killed, the tumours removed, sites of metastases noted, and metastatic tumours removed. Primary and metastatic tumours were fixed in 10% formalin, dehydrated, and embedded in paraffin.
Patients
Utilizing our institution's clinical cancer research database and tissue bank, we randomly identified 20 PDA patients' paraffin-embedded block specimens and two normal controls. Each patient had been diagnosed with pancreatic head adenocarcinoma and had undergone an en bloc resection with a pancreaticoduodenectomy. The study pathologist analysed the tissue blocks to determine which blocks represented the primary tumours and tumour margins. New tissue slides for immunohistochemistry were made from these identified sections. With the approval of the institutional review board, each patient's chart was reviewed for tumour staging, nodal status and margin status at surgical resection, recurrence, and overall and disease-free survival.
Immunohistochemistry
Immunohistochemical staining was performed on 4-µm thick sections of formalin-fixed, paraffin-embedded tissue. Prior to staining, heat-induced epitope retrieval (HIER) was performed in citrate buffer, pH 6.0, in an electric pressure cooker (DC2000; BioCare Medical LLC, Walnut Creek, CA, USA). Samples were then stained for PINCH (BD Biosciences, Inc., San Jose, CA, USA). Following incubation with the primary antibody, samples were processed using a commercially available Alkaline Phosphatase Kit (Vector Laboratories, Inc., Burlingame, CA, USA). A pathologist (author LE) analysed the samples to determine the percentage of cells positively stained for PINCH, as well as the intensity of the staining. The pathologist was blinded to tumour location for the primary and metastatic tumours from the animal PDA model. The index for the proportion of cells stained for PINCH was defined by positive staining for PINCH of: (1) 0–25%; (2) 26–50%; (3) 51–75%, and (4) 76–100%. Intensity of staining for PINCH was scored as: (0) no staining; (1) mild staining; (2) moderate staining, and (3) strong staining.
Statistics
In the analysis of animal tumours, a t-test was utilized to determine if there were differences in the proportion and intensity indices for staining for PINCH in the primary orthotopic tumour and peri-tumoral stroma compared with metastatic tumours. A t-test was also utilized to determine if there were any differences in the proportion and intensity indices for staining for PINCH between the tumour and stroma cells within the tumour.
In the analysis of human tumours, a t-test was utilized to determine if there were differences in the proportion and intensity indices for staining for PINCH in the tumour cells compared with the peri-tumoral stroma cells. A non-parametric Spearman's rank correlation was performed to determine whether staining correlated with T and N tumour status. The final goal was to determine if staining for PINCH in the tumour and peri-tumoral stroma correlated with overall survival. The data were divided into groups near the median and a likelihood ratio test using the Cox proportional hazards model was performed. The estimated hazard ratio (Cox model) and the median survival in each group (estimated from the Kaplan–Meier curve) were also determined. Statistical significance was set at P < 0.05.
Results
PINCH expression in animal orthotopic PDA tumours
Nine of the 10 animals developed a primary orthotopic tumour. All of these nine animals developed distant metastases (n= 19 metastases). The location of the metastases included the diaphragm (n= 6/19), mesentery (n= 5/19), liver (n= 4/19) and peritoneal cavity (n= 4/19). There was constituent PINCH expression in the mouse pancreas islet cells and acinar cells (Fig. 1). The orthotopic tumour did not show significantly more staining than the normal surrounding pancreas. The mouse model tumours showed equivalent staining when the peri-tumoral stroma cells were compared with the tumour cells (Fig. 2).
Figure 1.
Light microscope images. (A) Mouse primary tumour immunohistochemically stained for PINCH, demonstrating strong constitutive expression in the pancreas acinar cells (P). Both the tumour and peri-tumoral stroma cells of the primary tumour stain positive for PINCH. (B) Mouse metastatic tumour cells stain to a greater degree than primary tumour cells. (Original magnification 200× in both images)
Figure 2.
Comparisons between tumour cells and tumour stroma cells in metastatic and primary tumours for (A) the proportion of cells positively stained for PINCH and (B) staining intensity. The proportion of cells stained for PINCH and staining intensity were statistically greater in tumour stroma compared with tumour cells. *P < 0.05; data are reported as mean ± standard error
Staining for PINCH, both by proportion and intensity of staining, was higher in the metastatic tumour cells compared with the primary orthotopic tumour cells (Figs 1 and 2). By contrast with the tumour cells, there were no differences in staining for PINCH within the peri-tumoral stoma cells of the primary tumours compared with the metastatic peri-tumoral stoma cells. There were no differences in staining for PINCH between the various metastatic sites, but the small sample size at each location limits the significance of this finding.
PINCH expression in human PDA tumours
PINCH protein was constitutively expressed in the control pancreas, with greater staining in the islet cells than in the pancreatic acinar cells (Fig. 3). Staining for PINCH was greater according to both the proportion of stained cells and staining intensity in the PDA peri-tumoral stroma cells compared with the tumour cells (Fig. 4). All the patient samples demonstrated PINCH expression in tumour and peri-tumoral stroma cells. In 70% (14/20) of cases, >50% of the peri-tumoral stroma cells stained positive for PINCH. By contrast, only 15% (3/20) of the tumour samples showed positive staining for PINCH on >50% of the tumour cells. The differences in staining between the tumour and peri-tumoral stroma cells were also qualitatively evident by light microscopy (Fig. 3).
Figure 3.
Light microscope images. (A) Human control pancreas in which the islet cells (I) of the pancreas stain with a higher intensity compared with the acinar cells (A). (B) Human pancreatic ductal adenocarcinoma tissue demonstrating staining for PINCH was stronger in the peri-tumoral stroma cells (S) compared with the tumour cells (T). (Original magnification 400× in both images)
Figure 4.
Comparisons between tumour stroma cells and tumour cells for (A) the proportion of cells positively stained for PINCH and (B) staining intensity. The proportion of cells stained for PINCH and staining intensity were statistically greater in tumour stroma compared with tumour cells. *P < 0.05; data are reported as mean ± standard error
PINCH expression correlated with disease stage and patient outcomes
Patients
Demographic data for the 20 patients are presented in Table 1, including gender, age, T stage, N stage, margin status and survival. Patients included 12 women with an average age of 68 ± 14 years and eight men with an average age of 62 ± 13 years. There were no correlations between patient gender or age and staining for PINCH. The majority of our patients had T3 tumours (n= 13). According to American Joint Committee on Cancer (AJCC) staging, one patient had stage I cancer, 10 had stage II cancer, and nine had stage III cancer. Increased staining for PINCH in the tumour cells, by both proportion and intensity of staining, was correlated with higher T status (P < 0.05) (Table 2). PINCH expression in the tumour cells was not statistically correlated with N status. Additionally, staining for PINCH in the peri-tumoral stroma was not statistically correlated with either T or N status. Staining for PINCH and margin status were not statistically correlated (P > 0.6; data not presented).
Table 1.
Patient demographic information
| Patient | Gender | Age at diagnosis, years | T status | N status | Positive pancreatic margin | Positive retro-peritoneal margin | Survival, days |
|---|---|---|---|---|---|---|---|
| 1 | F | 77 | 2 | 1 | No | No | 995 |
| 2 | M | 52 | 3 | 1 | No | No | 267 |
| 3 | M | 65 | 1 | 0 | No | No | 545 |
| 4 | F | 60 | 3 | 0 | No | No | 1086 |
| 5 | F | 67 | 4 | 1 | No | Yes | 157 |
| 6 | F | 35 | 3 | 1 | No | No | 221 |
| 7 | F | 73 | 3 | 0 | No | No | 457 |
| 8 | M | 77 | 3 | 1 | No | Yes | 207 |
| 9 | F | 71 | 1 | 1 | No | Yes | 921 |
| 10 | M | 37 | 3 | 1 | No | Yes | 337 |
| 11 | F | 82 | 3 | 0 | No | No | 2250 |
| 12 | M | 71 | 3 | 1 | No | No | 399 |
| 13 | F | 73 | 3 | 0 | No | No | 424 |
| 14 | M | 56 | 4 | 0 | No | No | 884 |
| 15 | F | 80 | 3 | 0 | No | Yes | 284 |
| 16 | M | 72 | 4 | 1 | Yes | Yes | 222 |
| 17 | F | 73 | 3 | 1 | No | No | 1700 |
| 18 | M | 66 | 1 | Yes | No | 1923 | |
| 19 | F | 82 | 3 | 1 | Yes | Yes | 310 |
| 20 | F | 48 | 3 | 0 | Yes | No | 314 |
F, female; M, male
Table 2.
Spearman's rank correlation coefficients (rs) for PINCH expression vs. tumour status (T) and nodal status (N)
| Staining for PINCH | T status | N status |
|---|---|---|
| Tumour cells, proportion | 0.493a | 0.180 |
| Tumour cells, intensity | 0.477a | 0.148 |
| Tumour stroma, proportion | −0.022 | 0.372 |
| Tumour stroma, intensity | 0.179 | 0.241 |
Statistically positive correlation between PINCH expression and T status (P < 0.05)
The data were also analysed to determine if there were any correlations between survival and staining for PINCH. There was a trend toward significance between the proportion of cells positively stained for PINCH (P= 0.07) and staining intensity (P= 0.06) in the peri-tumoral stroma cells and survival, with stronger staining associated with poorer survival. The median survival was 884 days in subjects with low PINCH expression (n= 7) compared with 337 days in subjects with high PINCH expression (n= 13), with a hazard ratio of 2.3 (Fig. 5).
Figure 5.
Kaplan–Meier curve demonstrating the correlation between PINCH expression in the stroma and survival. Individuals with lower PINCH expression on average lived longer (P= 0.068). No patients were lost to follow-up and thus no censor points are indicated on the graph. The median survival was 884 days in subjects with low PINCH (n= 7) compared with 337 days in subjects with high PINCH (n= 13), with a hazard ratio of 2.3
Discussion
The data from the present investigation support the hypothesis that PINCH plays an important role in the progression of cancer. Specifically, in the orthotopic animal model of PDA, we observed greater PINCH expression in the invasive metastatic tumours compared with the primary tumours. Secondly, within the human samples analysed, higher T status and poorer survival correlated with greater PINCH expression. Thus, in both the animal model and patient population, PINCH expression correlated with a more invasive and aggressive tumour phenotype.
Another interesting observation concerns the constituent PINCH expression within the normal pancreas seen in both the mouse PDA model and the human control samples. Although this was not quantitatively determined, there did not appear to be greater PINCH expression within the tumours compared with the normal pancreas. This is by contrast with other cancers. Wang-Rodriguez et al. confirmed increased expression of PINCH protein in breast, prostate, lung and colorectal cancers relative to healthy tissue.11 More recently, it has been shown that PINCH expression is greater in oesophageal and oral squamous cell carcinoma, colorectal cancer and gliomas relative to normal tissue.12,14–16 Taken together, the observation by others of greater PINCH expression in the tumour, as well as our demonstration of a correlation between PINCH expression and poorer patient prognosis, suggest that PINCH expression may be an important component of cancer progression in general.
Another important finding of the present investigation is that PINCH expression appears to be stronger in the stroma cells adjacent to the cancer cells in human PDA. PINCH expression in the peri-tumoral stroma cells is roughly double that of the cancer cells, in terms of both the proportion of cells stained and the intensity of staining. PINCH expression is also demonstrated to a greater degree in the peri-tumoral stroma tissue of other cancers compared with the adjacent cancer cells.11,12,14–16 Evidence demonstrating that the interactions between peri-tumoral stroma cells and cancer cells may play an important part in the progression and metastases of cancer is beginning to accumulate.2,3,6–8 This may be even more important in PDA, as it is associated not only with earlier incidences of metastases, but also with a much greater desmoplastic response. Pancreatic stoma cells have been shown to activate the ERK and AKT pathways and thus are strong promoters of growth and invasion.2,3,8 These same pathways are also activated by PINCH.4,5,17 Thus PINCH in the peri-tumoral stroma cells may be an important regulator of the growth and migration of these cells, as well as adjacent cancer cells, and therefore a good marker for determining the invasiveness of a tumour.
One interesting finding was that this differential expression of PINCH between stroma and cancer cells did not exist in the animal model. One possible explanation for this might be that the cancer cells in the animal model were of human origin, whereas the stromal cells were of mouse origin. Thus, expression differences may depend on species origin. Despite these differences, PINCH expression still appeared to correlate with tumour progression in both the human samples and the animal model.
There are other pathways by which PINCH could influence the transformation and progression of PDA. Specifically, PINCH has a binding site for Nck-2 and integrin-linked kinase (ILK).4,5 Nck-2 is involved in growth factor mediating signalling, as well as actin cytoskeleton remodelling.18 The interaction of PINCH and Nck-2 directly links PINCH to the actin polymerization process and therefore it is likely that PINCH plays a role in cytoskeleton motility at cell–ECM adhesions. The binding of PINCH is also a key regulator of ILK expression and is important for the localization of ILK to the cell membrane.4,5 Disruption of the binding of these two proteins results in alterations to cell motility, as well as survival. In recognition of the role of PINCH protein signalling in malignant transformation, many investigators have hypothesized that PINCH protein expression would be increased in more aggressive tumours. This has been documented in colorectal cancers, gliomas and oral epithelial cancers.12,14–16 Specifically, PINCH expression is greater in higher-grade gliomas independent of patient age, gender and tumour size.14 Similar findings have been observed in oral epithelial tumours, with greater PINCH expression in nodal metastases relative to primary tumours, and in colorectal cancer, with greater PINCH expression correlating to poorer patient prognosis.10,15 We confirmed a similar pattern whereby increased invasiveness or tumour aggressiveness was correlated with increased PINCH expression in PDA. Several findings corroborate this conclusion: firstly, in the murine orthotopic model, PINCH expression was significantly higher in metastatic tumour tissue than in primary tumours; secondly, in the human PDA tissues, tumour T status correlated with increased PINCH expression, and, finally, a trend toward increased PINCH expression in PDA tumours with poorer patient survival is probably important. We feel that the trend in survival failed to reach statistical significance based on the small number of patients and the generally poor prognosis of all patients with PDA. However, the human data on T status and survival combined with the animal data on metastatic tumours suggest that PINCH plays an important role in the progression of PDA.
Conclusions
PINCH protein may function as a regulator or signal for ECM invasion and cellular migration. Findings of increased PINCH protein in more advanced stages of human PDA, as well as in metastatic tumours from the animal model, support this hypothesis. Additionally, the importance of the differential expression of PINCH in the human tumour and stroma warrants further evaluation.
Conflicts of interest
None declared.
References
- 1.Jemal A, Siegel R, Ward E, Murray T, Xu J, Thun MJ. Cancer statistics, 2007. CA Cancer J Clin. 2007;57:43–66. doi: 10.3322/canjclin.57.1.43. [DOI] [PubMed] [Google Scholar]
- 2.Hwang RF, Moore T, Arumugam T, Ramachandran V, Amos KD, Rivera A, et al. Cancer-associated stromal fibroblasts promote pancreatic tumor progression. Cancer Res. 2008;68:918–926. doi: 10.1158/0008-5472.CAN-07-5714. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Vonlaufen A, Joshi S, Qu C, Phillips PA, Xu Z, Parker NR, et al. Pancreatic stellate cells: partners in crime with pancreatic cancer cells. Cancer Res. 2008;68:2085–2093. doi: 10.1158/0008-5472.CAN-07-2477. [DOI] [PubMed] [Google Scholar]
- 4.Fukuda T, Chen K, Shi X, Wu C. PINCH-1 is an obligate partner of integrin-linked kinase (ILK) functioning in cell shape modulation, motility, and survival. J Biol Chem. 2003;278:51324–51333. doi: 10.1074/jbc.M309122200. [DOI] [PubMed] [Google Scholar]
- 5.Xu Z, Fukuda T, Li Y, Zha X, Qin J, Wu C. Molecular dissection of PINCH-1 reveals a mechanism of coupling and uncoupling of cell shape modulation and survival. J Biol Chem. 2005;280:27631–27637. doi: 10.1074/jbc.M504189200. [DOI] [PubMed] [Google Scholar]
- 6.Apte MV, Park S, Phillips PA, Santucci N, Goldstein D, Kumar RK, et al. Desmoplastic reaction in pancreatic cancer: role of pancreatic stellate cells. Pancreas. 2004;29:179–187. doi: 10.1097/00006676-200410000-00002. [DOI] [PubMed] [Google Scholar]
- 7.Schneider E, Schmid-Kotsas A, Zhao J, Weidenbach H, Schmid RM, Menke A, et al. Identification of mediators stimulating proliferation and matrix synthesis of rat pancreatic stellate cells. Am J Physiol Cell Physiol. 2001;281:C532–543.. doi: 10.1152/ajpcell.2001.281.2.C532. [DOI] [PubMed] [Google Scholar]
- 8.McCarroll JA, Phillips PA, Kumar RK, Park S, Pirola RC, Wilson JS, et al. Pancreatic stellate cell migration: role of the phosphatidylinositol 3-kinase(PI3-kinase) pathway. Biochem Pharmacol. 2004;67:1215–1225. doi: 10.1016/j.bcp.2003.11.013. [DOI] [PubMed] [Google Scholar]
- 9.C Wu. PINCH, N(i)ck and the ILK: network wiring at cell-matrix adhesions. Trends Cell Biol. 2005;15:460–466. doi: 10.1016/j.tcb.2005.07.002. [DOI] [PubMed] [Google Scholar]
- 10.Gao J, Arbman G, Rearden A, Sun XF. Stromal staining for PINCH is an independent prognostic indicator in colorectal cancer. Neoplasia. 2004;6:796–801. doi: 10.1593/neo.04304. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Wang-Rodriguez J, Dreilinger AD, Alsharabi GM, Rearden A. The signaling adapter protein PINCH is up-regulated in the stroma of common cancers, notably at invasive edges. Cancer. 2002;95:1387–1395. doi: 10.1002/cncr.10878. [DOI] [PubMed] [Google Scholar]
- 12.Zhu Z, Yang Y, Zhang Y, Wang Z, Cui D, Zhang J, et al. PINCH expression and its significance in esophageal squamous cell carcinoma. Dis Markers. 2008;25:75–80. doi: 10.1155/2008/473860. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Scaife CL, Shea JE, Dai Q, Firpo MA, Prestwich GD, Mulvihill SJ. Synthetic extracellular matrix enhances tumor growth and metastasis in an orthotopic mouse model of pancreatic adenocarcinoma. J Gastrointest Surg. 2008;12:1074–1080. doi: 10.1007/s11605-007-0425-3. [DOI] [PubMed] [Google Scholar]
- 14.Wang MW, Gu P, Zhang ZY, Zhu ZL, Li YM, Zhao HX, et al. Expression of PINCH protein in gliomas and its clinicopathological significance. Oncology. 2007;72:343–346. doi: 10.1159/000113064. [DOI] [PubMed] [Google Scholar]
- 15.Zhang JT, Li QX, Wang D, Zhu ZL, Yang YH, Cui DS, et al. Up-regulation of PINCH in the stroma of oral squamous cell carcinoma predicts nodal metastasis. Oncol Rep. 2005;14:1519–1522. [PubMed] [Google Scholar]
- 16.Zhao ZR, Zhang ZY, Cui DS, Jiang L, Zhang HJ, Wang MW, et al. Particularly interesting new cysteine-histidine rich protein expression in colorectal adenocarcinomas. World J Gastroenterol. 2006;12:298–301. doi: 10.3748/wjg.v12.i2.298. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Chen K, Tu Y, Zhang Y, Blair HC, Zhang L, Wu C. PINCH-1 regulates the ERK-Bim pathway and contributes to apoptosis resistance in cancer cells. J Biol Chem. 2008;283:2508–2517. doi: 10.1074/jbc.M707307200. [DOI] [PubMed] [Google Scholar]
- 18.Vaynberg J, Fukuda T, Chen K, Vinogradova O, Velyvis A, Tu Y, et al. Structure of an ultraweak protein-protein complex and its crucial role in regulation of cell morphology and motility. Mol Cell. 2005;17:513–523. doi: 10.1016/j.molcel.2004.12.031. [DOI] [PubMed] [Google Scholar]