The PTEN phosphatase controls intestinal epithelial cell polarity and barrier function: role in colorectal cancer progression

Abstract

Background: The PTEN phosphatase acts on phosphatidylinositol 3,4,5-triphosphates resulting from phosphatidylinositol 3-kinase (PI3K) activation. PTEN expression has been shown to be decreased in colorectal cancer. Little is known however as to the specific cellular role of PTEN in human intestinal epithelial cells. The aim of this study was to investigate the role of PTEN in human colorectal cancer cells.

Methodology/principal findings: Caco-2/15, HCT116 and CT26 cells were infected with recombinant lentiviruses expressing a shRNA specifically designed to knock-down PTEN. The impact of PTEN downregulation was analyzed on cell polarization and differentiation, intercellular junction integrity (expression of cell-cell adhesion proteins, barrier function), migration (wound assay), invasion (matrigel-coated transwells) and on tumor and metastasis formation in mice. Electron microscopy analysis showed that lentiviral infection of PTEN shRNA significantly inhibited Caco-2/15 cell polarization, functional differentiation and brush border development. A strong reduction in claudin 1, 3, 4 and 8 was also observed as well as a decrease in transepithelial resistance. Loss of PTEN expression increased the spreading, migration and invasion capacities of colorectal cancer cells in vitro. PTEN downregulation also increased tumor size following subcutaneous injection of colorectal cancer cells in nude mice. Finally, loss of PTEN expression in HCT116 and CT26, but not in Caco-2/15, led to an increase in their metastatic potential following tail-vein injections in mice.

Conclusions/significance: Altogether, these results indicate that PTEN controls cellular polarity, establishment of cell-cell junctions, paracellular permeability, migration and tumorigenic/metastatic potential of human colorectal cancer cells.

Figure 1. PTEN down-regulation impairs intestinal epithelial cell polarization and differentiation.

Caco-2/15 cells were infected with either recombinant lentiviruses encoding a shRNA which specifically knocks-down PTEN (shPTEN) or negative control shRNAs (shGFP or shMUT). A and C. After selection of infected cells, Caco-2/15 were lysed at −2, 0, 3, 6, 9 and 12 days post-confluence. Proteins were then analyzed by Western blot. B. Left panels: Newly confluent Caco-2/15 cells expressing shGFP or shPTEN were observed by optical microscopy. Scale bars  = 100 µm. Middle panels: Newly confluent Caco-2/15 cells expressing shGFP or shPTEN were fixed and observed by transmission electron microscopy. Scale bars  = 2 µm. Right panels: Newly confluent shGFP or shPTEN expressing-cells were observed by scanning electron microscopy. Scale bars  = 1 µm.

Figure 2. PTEN regulates tight junction integrity and function.

A. Caco-2/15 cells expressing shMUT or shPTEN were fixed after 3 days post-confluence and observed by transmission electron microscopy. Scale bars  = 500 nm. Arrows indicate adherens junctions while brackets indicate tight junctions. B. Caco-2/15 cells were cultivated on porous membranes after which transepithelial resistance was measured in triplicate 3 and 9 days after the cells reached confluence. *** Significantly different at p≤0.005 (Student's t-test). C. Caco-2/15 cells expressing shGFP or shPTEN were lysed at -2, 0, 3, 6, 9 and 12 days post-confluence followed by Western blot analysis of junctional proteins.

Figure 3. PTEN down-regulation in Caco-2/15 increases migration/invasion capacity and tumorigenic potential.

A. Left panels: Cell morphology was analyzed by phase contrast microscopy at subconfluence. Right panels: Newly confluent cells stably expressing shMUT or shPTEN were wounded and treated with 2 mM hydroxyurea. After 48 h, movement of the coherent sheet across the linear wound margin (white line) was evaluated by phase contrast microscopy. The graph on the right illustrates the relative area covered by migrating cells as assessed on 5 different wound experiments per condition using Image J software. Scale bars  = 100 µm. B. Activated levels of GTP-bound Rac or Cdc42 were analyzed with G-LISA activation assay biochemistry kits on subconfluent (SC) and newly confluent (C) Caco-2/15 cells. C. Invasion of cells was studied using Matrigel-coated Transwells. After 48 h, invading cells were fixed, stained with crystal violet 1% and counted. D. Subconfluent shMUT and shPTEN-expressing Caco-2/15 cells were lysed and total RNA isolated for gene expression analyzed by Q-PCRs. The relative level of each RNA was calculated using the standard curve method and normalized to the corresponding PDGB RNA level. E. 2×10⁶ proliferating Caco-2/15 cells expressing shMUT or shPTEN were injected subcutaneously in 5 nude mice per condition. Tumor weight was evaluated 42 days after injection. * p≤0.05, *** p≤0.005, statistical differences determined using Student's t-test.

Figure 4. PTEN down-regulation in HCT116 increases migration/invasion capacity and tumorigenic potential.

A. HCT116 cells were infected either with recombinant lentiviruses encoding shMUT or shPTEN. After selection of infected cells, proliferating cells were lysed and protein expression was analyzed by Western blot. B. Left panels: Cell morphology was analyzed by phase contrast microscopy at subconfluence. Right panels: Newly confluent cells were wounded and treated with 2 mM hydroxyurea. After 48 h, movement of the coherent sheet across the linear wound margin (white line) was evaluated by phase contrast microscopy. The graph on the right illustrates the relative area covered by migrating cells as evaluated on 5 different wound experiments per condition using Image J software. Scale bars  = 100 µm. C. Invasion of cells was studied using Matrigel-coated Transwells. After 48 h, invading cells were fixed, stained with crystal violet 1% and counted. D. 2×10⁶ proliferating HCT116 cells expressing shMUT or shPTEN were injected subcutaneously in 5 nude mice per condition. Tumor weight was evaluated 30 days after injection. * Significantly different at p≤0.05.

Figure 5. PTEN down-regulation increases metastatic potential of HCT116 and CT26 cells in vivo.

A. Proliferating HCT116 cells expressing shMUT or shPTEN were injected in the lateral tail-vein of 12 mice per condition. Mice injected with HCT116-expressing shPTEN developed tumors in (various, multiple) locations such as the heart and dorsal muscle 50–75 days after tail-vein injection. T: tumor. B. CT26 cells were infected either with recombinant lentiviruses encoding a shRNA which specifically knocked-down the murine form of PTEN (shPten) or with a negative control shRNA (shTGFP). After selection of infected cells, proliferating cells were lysed and protein expression was analyzed by Western blot. Proliferating CT26 cells expressing shTGFP or shPten were also injected in the lateral tail-vein of 6 mice per condition. Total number of detectable pulmonary metastasis was evaluated 14 days after tail-vein injection. Significantly different at *** p≤0.005.

Figure 6. PTEN expression is decreased in colorectal cancers.

PTEN protein expression was investigated by Western blot in 53 paired samples of colon cancers (resection margins and primary tumors). Expression levels were normalized to the intensity of Ponceau red staining and to a reference sample, resulting in a dimensionless value (arbitrary units-AU). Amounts of PTEN protein in tumor tissues relative to their matched normal samples were analyzed by paired t-test. *** Significantly different at p≤0.005.