- Short communication
- Open Access
PRL-3 overexpression in epithelial cells is induced by surrounding stromal fibroblasts
Molecular Cancervolume 8, Article number: 46 (2009)
We isolate and culture carcinoma-associated fibroblasts (CAFs) from primary tumour (CAFpt), CAFs from corresponding synchronous liver metastasis (CAFlm) as well as normal colonic fibroblasts (NCF) from the same patient. From these cultures, conditioned media (CM) was obtained. Culture of a wide panel of colorectal and pancreatic cell lines in CM from CAFlm resulted in overexpression of mRNA PRL-3 and higher overexpression in CAFs than in non-activated fibroblasts. Moreover PRL-3 mRNA expression correlates with expression of α-SMA and deposition of collagen fibrils in the stroma. We demonstrate that products secreted by CAFs trigger PRL-3 overexpression in cancer cells. Identification of these factors may contribute to new stroma-targeted therapies for desmoplastic tumours.
In many epithelial tumours inclouding colorectal, the relevance of PRL-3 phosphatase to the tumorigenic process of distant dissemination is well documented [1–5]. While some manuscripts adress the downstream pathways in which PRL-3 could influence the processes of migration and invasion [6, 7], there is a complete lack of knowledge of the events upstream PRL-3.
In colorectal tumorigenesis, PRL-3 was identified as a molecule consistently overexpressed in hepatic metastases. These tumours are characterised by a strong desmoplastic reaction, consisting predominantly of carcinoma-associated fibroblasts (CAFs) and collagen I fibrils, which are also a hallmark of colorectal adenocarcinomas (CRC). In the case of primary tumours, the desmoplastic reaction contributes to the spreading of tumoral epithelial cells. But for liver metastasis, the biological significance of desmoplasia is still unclear. Some authors consider desmoplasia as a wall to keep tumour cells under control. If desmoplasia is a mechanism of walling off or is a reaction to promote spreading of epithelial cells will depend probably on the composition of extracellular matrix (ECM) and cells responsible for the deposition of ECM components. Therefore, cross-talk between epithelial and mesenchymal cells could influence the behaviour of the tumour population.
In this study we report modulation of PRL-3 expression in CRC cell lines dependant on soluble factors released into fibroblast conditioned media (CM). Moreover, we describe an association between PRL-3 overexpression in epithelial cells and desmoplastic reaction in human samples.
PRL-3 is overexpressed in different colorectal and pancreatic cell lines when cultured in conditioned media derived from carcinoma-associated fibroblasts from liver metastasis
We cultured a wide panel of colorectal and pancreatic cell lines, derived from primary tumours and secondary sites for 24 hours in their standard culture medium (DMEMF12 + 10%FBS) or in conditioned medium obtained from CAFs derived from a CRC liver metastasis. CM was produced as follows: CAFs were grown to confluence and CM was harvested after 48 hours of culture, centrifuged to remove cellular debris and filtered. PRL-3 mRNA expression was examined by real-time Q-PCR in nine colorectal and two pancreatic cell lines. All colorectal cell lines exhibited low levels of mRNA PRL-3 expression when cultured in standard DMEMF12 medium plus 10%FBS. On the other hand, pancreatic cell lines displayed moderate levels. But interestingly, as shown in figure 1A, both cell lineages increased PRL-3 mRNA expression when cultured in CM derived from CAFs, with the highest fold change observed in DLD-1 cells (6.479 fold). We hypothesised that products released by the stroma interact with receptors in the epithelial compartment to induce PRL-3 overexpression. Since our colorectal cell panel include almost all genetic backgrounds, this cross-talk may be a common event in CRC.
CM from various fibroblasts lineages produces differential mRNA PRL-3 expression
We cultured DLD-1 cells for 24 h in CM, produced as described above, obtained from Foreskin fibroblasts (FF) and Normal colonic fibroblasts (NCF), metastasic primary tumour CAFs (CAFpt) and liver metastasis CAFs (CAFlm) from the same patient (matched samples). As shown in figure 1B, all CM induced significant (P < 0.05) overexpression of mRNA PRL-3 measured by Q-PCR compared with DMEMF12 control. Analysis of statistical significance used the ANOVA test and Tukey post hoc. Moreover, it is possible to put expression values in two groups according to activated (CAFpt and CAFlm; approximately 7 fold change in relation to control DMEMF12 + 10% FBS) or non-activated fibroblasts (FF and NCF; approximately 3.5 fold change). Comparing non activated fibroblasts, normal colonic fibroblasts (from a patient with colorectal cancer) and foreskin fibroblasts exerts the same effect on mRNA PRL-3 expression (P = 0.983). Same trend is observed between CAFpt and CAFlm (P = 0.514). Interestingly, comparing results from matched samples, conditioned media from NCF induced significantly (P = 0.036) lower levels of mRNA PRL-3 than CM from activated fibroblasts of the same patient. Accordingly, we conclude that CM obtained from CAFs more efficiently induce the overexpression of phosphatase PRL-3 than non-activated fibroblasts. The high expression values obtained with CAFpt are in agreement with a previous report  demonstrating high mRNA PRL-3 expression in primary Stage III tumours that developed distant metastasis. As shown in figure 2, this results were confirmed at the protein level in DLD-1 cells stimulated with CM from CAFpt and CAflm, reaching the highest intensity in cells cultured with CM CAFlm (figure 2E).
mRNA PRL-3 expression levels correlates with the degree of desmoplasia
These results, in combination with prior reports outlining PRL-3 overexpression in hepatic metastases and metastasic primary tumours [8, 9] led to the hypothesis that PRL-3 expression is induced by products secreted by surrounding activated fibroblasts adjacent to the tumour compartment. We then evaluated the expression of smooth muscle actin (α-SMA), and collagen deposition (Masson staining) in a subset of 43 metastasic primary tumours (22 stage IIIB and 21 stage IIIC) for which we had paraffin-embedded tissue, from the entire 80 samples series where we previously assessed PRL-3 mRNA expression . The evaluation was done by two blinded independent investigators (XS and LP). In samples with high mRNA PRL-3 expression, 85.71% (18/21) of tumours displayed high degree of infiltration of CAFs evaluated as the proportion of mesenchymal cells stained with α-SMA antibody (Chi square test, P = 0.029; Figure 3A). Moreover, the evaluation of collagen deposition revealed a high degree of fibrils in tumours with high mRNA PRL-3 expression, accounting for 71.4% with intense deposition (Figure 3B), 19.04% for moderate and 9.52% for slight or absence (Figure 3C)(Chi square test, P = 0.009). In addition, binary logistic regression showed that both high expression of α-SMA and intense and moderate Masson staining associate with tumour aggressiveness and were predictors of distant metachronous metastasis in univariate analysis (P = 0.05, OR 4.35, CI 0.99–19.12 for α-SMA; P = 0.024, OR 6.92, CI 1.29–37.05 for collagen deposition). As shown in figure 4, high expression of α-SMA and intense deposition of collagen fibrils trend with worse prognosis, although not reaching statistic significance probably due to sample size.
Hence, tumours with high mRNA PRL-3 have a more aggressive behaviour and high probability to relapse. As high primary tumour PRL-3 expression is induced by stromal factors, the abundance of myofibroblasts in cancer-associated stroma may be a useful indicator of disease recurrence after curative colorectal cancer surgery as suggested by Tsujino T et al . Notwithstanding, for prognosis purposes is mandatory to validate the α-SMA expression and collagen deposition in larger and consecutive series in the adjuvant setting.
Crosstalk between tumour microenvironment and transformed epithelial cells plays a central role in the fate of malignant tumour cells by locally modifying signalling pathways at the primary tumour site, enhancing the migration and invasion of metastatic cells and, preparing secondary organ sites for metastatic growth . More interestingly, genetic alterations in CAFs  and deregulation of key pathways  may enhance the migration and invasion through those crosstalk mechanisms. PRL-3 is key molecule involved in those processes of migration and invasion, and has been involved in the natural outgrowth of metastatic capabilities of desmoplastic tumours like colon, pancreas and breast. Here, we demonstrate that elevated PRL-3 expression, observed in colorectal primary tumours and distant secondary sites, is induced by products secreted by the surrounding activated stroma. Identification of those CAF-derived factors may inform on new stroma-targeted therapies for desmoplastic tumours.
carcinoma-associated fibroblast from primary tumour
carcinoma-associated fibroblast from liver metastasis
normal colonic fibroblast
alpha smouth muscle actin.
Saha S, Bardelli A, Buckhaults P, Velculescu VE, Rago C, St Croix B, Romans KE, Choti MA, Lengauer C, Kinzler KW, Vogelstein B: A phosphatase associated with metastasis of colorectal cancer. Science. 2001, 294: 1343-6. 10.1126/science.1065817
Peng L, Ning J, Meng L, Shou C: The association of the expression level of protein tyrosine phosphatase PRL-3 protein with liver metastasis and prognosis of patients with colorectal cancer. J Cancer Res Clin Oncol. 2004, 130: 521-6. 10.1007/s00432-004-0563-x
Polato F, Codegoni A, Fruscio R, Perego P, Mangioni C, Saha S, Bardelli A, Broggini M: PRL-3 phosphatase is implicated in ovarian cancer growth. Clin Cancer Res. 2005, 11 (19 Pt 1): 6835-9. 10.1158/1078-0432.CCR-04-2357
Radke I, Gotte M, Kersting C, Mattsson B, Kiesel L, Wulfing P: Expression and prognostic impact of the protein tyrosine phosphatases PRL-1, PRL-2, and PRL-3 in breast cancer. Br J Cancer. 2006, 95: 347-54. 10.1038/sj.bjc.6603261
Miskad UA, Semba S, Kato H, Matsukawa Y, Kodama Y, Mizuuchi E, Maeda N, Yanagihara K, Yokozaki H: High PRL-3 expression in human gastric cancer is a marker of metastasis and grades of malignancies: an in situ hybridization study. Virchows Arch. 2007, 450: 303-10. 10.1007/s00428-006-0361-8
Bardelli A, Saha S, Sager JA, Romans KE, Xin B, Markowitz SD, Lengauer C, Velculescu VE, Kinzler KW, Vogelstein B: PRL-3 expression in metastatic cancers. Clin Cancer Res. 2003, 9: 5607-15.
Liang F, Wang WQ, Sun JP, Udho E, Zhang ZY: PRL3 promotes cell invasion and proliferation by down-regulation of Csk leading to Src activation. J Biol Chem. 2007, 282: 5413-5419. 10.1074/jbc.M608940200
Molleví DG, Aytes A, Padullés L, Martínez-Iniesta M, Baixeras N, Salazar R, Ramos E, Figueras J, Capella G, Villanueva A: PRL-3 is essentially overexpressed in primary colorectal tumours and associates with tumour aggressiveness. Brit J Cancer. 2008, 99: 1718-1725. 10.1038/sj.bjc.6604747
Kato H, Semba S, Miskad UA, Seo Y, Kasuga M, Yokozaki H: High expression of PRL-3 promotes cancer cell motility and liver metastasis in human colorectal cancer: a predictive molecular marker of metachronous liver and lung metastases. Clin Cancer Res. 2004, 10: 7318-28. 10.1158/1078-0432.CCR-04-0485
Tsujino T, Seshimo I, Yamamoto H, Ngan CY, Ezumi K, Takemasa I, Ikeda M, Sekimoto M, Matsuura N, Monden M: Stromal Myofibroblasts Predict Disease Recurrence for Colorectal Cancer. Clin Cancer Res. 2007, 13: 2082-2090. 10.1158/1078-0432.CCR-06-2191
Psaila B, Lyden D: The metastatic niche: adapting the foreign soil. Nat Rev Cancer. 2009, 9 (4): 285-293. 10.1038/nrc2621
Kiaris H, Chatzistamou I, Trimis G, Frangou-Plemmenou M, Pafiti-Kondi A, Kalofoutis A: Evidence for nonautonomous effect of p53 tumor suppressor in carcinogenesis. Cancer Res. 2005, 65 (5): 1627-30. 10.1158/0008-5472.CAN-04-3791
Trimis G, Chatzistamou I, Politi K, Kiaris H, Papavassiliou AG: Expression of p21waf1/Cip1 in stromal fibroblasts of primary breast tumors. Hum Mol Genet. 2008, 17 (22): 3596-600. 10.1093/hmg/ddn252
The authors declare that they have no competing interests.
DGM drafted the manuscript. Moreover carried out cellular cultures, statistical analysis and coordinated the study. AA performed primary fibroblasts cultures. MB carried out PCR quantification. LP performed immunohystochemical staining of α SMA. MMI participated in tha Masson staining. XSJ provided fresh colorectal specimen and evaluated the stainings. RS carried out the management of clinical data from the colorectal database. AV provided financial support and helped to draft the manuscript. All authors read and approved the final manuscript.