Protein Kinase C (PKC) βII Induces Cell Invasion through a Ras/Mek-, PKCι/Rac 1-dependent Signaling Pathway*

Protein kinase C βII (PKCβII) promotes colon carcinogenesis. Expression of PKCβII in the colon of transgenic mice induces hyperproliferation and increased susceptibility to colon cancer. To determine molecular mechanisms by which PKCβII promotes colon cancer, we established rat intestinal epithelial (RIE) cells stably expressing PKCβII. Here we show that RIE/PKCβII cells acquire an invasive phenotype that is blocked by the PKCβ inhibitor LY379196. Invasion is not observed in RIE cells expressing a kinase-deficient PKCβII, indicating that PKCβII activity is required for the invasive phenotype. PKCβII induces activation of K-Ras and the Ras effector, Rac1, in RIE/PKCβII cells. PKCβII-mediated invasion is blocked by the Mek inhibitor, U0126, and by expression of either dominant negative Rac1 or kinase-deficient atypical PKCι. Expression of constitutively active Rac1 induces Mek activation and invasion in RIE cells, indicating that Rac1 is the critical downstream effector of PKCβII-mediated invasion. Taken together, our results define a novel PKCβII → Ras → PKCι /Rac1 → Mek signaling pathway that induces invasion in intestinal epithelial cells. This pathway provides a plausible mechanism by which PKCβII promotes colon carcinogenesis.


Protein kinase C ␤II (PKC␤II) promotes colon carcinogenesis. Expression of PKC␤II in the colon of transgenic mice induces hyperproliferation and increased susceptibility to colon cancer. To determine molecular mechanisms by which PKC␤II promotes colon cancer, we established rat intestinal epithelial (RIE) cells stably expressing PKC␤II. Here we show that RIE/PKC␤II cells acquire an invasive phenotype that is blocked by the PKC␤ inhibitor LY379196. Invasion is not observed in RIE cells expressing a kinase-deficient PKC␤II, indicating that PKC␤II activity is required for the invasive phenotype. PKC␤II induces activation of K-Ras and the Ras effector, Rac1, in RIE/PKC␤II cells. PKC␤II-mediated invasion is blocked by the Mek inhibitor, U0126, and by expression of either dominant negative Rac1 or kinase-deficient atypical PKC. Expression of constitutively active Rac1 induces Mek activation and invasion in RIE cells, indicating that Rac1 is the critical downstream effector of PKC␤II-mediated invasion. Taken together, our results define a novel PKC␤II 3 Ras 3 PKC /Rac1 3 Mek signaling pathway that induces invasion in intestinal epithelial cells. This pathway provides a plausible mechanism by which PKC␤II promotes colon carcinogenesis.
Colorectal cancer is the second leading cause of cancer death in the United States (1). Colon carcinogenesis is a complex multistep process involving progressive disruption of intestinal epithelial cell proliferation, differentiation, and survival mechanisms (2). Colon carcinogenesis is driven by environmental factors that modulate cell signaling pathways, and by genetic mutation of transforming oncogenes, and deletion or mutation of DNA repair enzymes and tumor suppressor genes (2). An important challenge is to understand how environmental and genetic factors interact to define colon cancer risk.
Protein kinase C (PKC) 1 is a family of 12 lipid-dependent serine/threonine kinases involved in proliferation, differentiation, and survival (3,4). Specific, reproducible changes in PKC isozyme expression patterns occur during carcinogen-induced colon carcinogenesis in rodents. Normal rat colonic epithelium contains multiple PKC isozymes, including PKC ␣, ␦, and (5,6). In azoxymethane-induced rat colonic tumors, the levels of PKC ␣, ␦, and are reduced and the level of PKC␤II is increased when compared with normal colonic epithelium (5,6). We recently demonstrated reduced PKC␣ expression and increased PKC␤II and PKC expression in azoxymethane-induced mouse colon tumors (7,8). Our subsequent studies provided direct evidence that both PKC␤II and PKC play critical but distinct roles in colon carcinogenesis (7)(8)(9)(10).
We have developed and analyzed transgenic mice that express elevated PKC␤II in the colonic epithelium (7,9,10). Transgenic PKC␤II mice exhibit hyperproliferation of the colonic epithelium and are prone to carcinogen-induced colon cancer (9,10). This cancer-prone phenotype results, at least in part, from the imposition of a PKC␤II-mediated hyperproliferative phenotype (9,10). To assess the molecular mechanisms by which PKC␤II promotes colon cancer, we established nontransformed rat intestinal epithelial (RIE) cell lines that overexpress PKC␤II (RIE/PKC␤II cells) (10,11). Genomic analysis of RIE/PKC␤II cells revealed that PKC␤II induces the expression of the Cox-2 enzyme and suppresses the expression of the transforming growth factor ␤ receptor type II (TGF␤RII) (11). As a result, RIE/PKC␤II cells have lost the ability to respond to the growth inhibitory effects of TGF-␤, and this loss is dependent upon the activity of Cox-2 (11). Therefore, PKC␤II establishes a novel, pro-carcinogenic PKC␤II 3 Cox-2 3 TGF␤RII signaling pathway that confers resistance to TGF-␤ (11). This pathway is activated in transgenic PKC␤II mice and accounts, at least in part, for the hyperproliferative phenotype exhibited by these mice (11). Carcinogens induce this pathway, whereas chemopreventive -3 fatty acids inhibit PKC␤II activity, suppress PKC␤II-mediated hyperproliferation, and attenuate the cancer-prone phenotype exhibited by transgenic PKC␤II mice (10,11).
Here, we demonstrate that PKC␤II induces an invasive phenotype in RIE cells through a signaling pathway that is distinct from the PKC␤II 3 Cox-2 3 TGF␤RII pathway responsible for TGF-␤ resistance. PKC␤II-mediated invasion is dependent upon Ras, PKC, Rac1, and Mek. Our findings define a novel proinvasive PKC␤II 3 Ras 3 PKC /Rac1 3 Mek signaling pathway. Our data indicate that PKC␤II promotes colon cancer through activation of at least two distinct signaling pathways, one that confers TGF-␤ resistance, and a second that induces invasion, in intestinal epithelial cells.

EXPERIMENTAL PROCEDURES
Reagents and Antibodies-The PKC␤-selective inhibitor LY379196 was a kind gift of Dr. James R. Gillig, Lilly Pharmaceutical Co. Puromycin dihydrochloride was purchased from Calbiochem. Polybrene and crystal violet were from Sigma. Anti-PKC␤II, anti-actin, and horseradish peroxidase-conjugated donkey anti-goat secondary antibodies were purchased from Santa Cruz Biotechnology, Inc. The PhosphoPlus Mek 1/2 antibody was from Cell Signaling Technology. Anti-FLAG epitope antibody was from Sigma. Anti-Rac1 mouse monoclonal antibody was * The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
Construction of a Kinase-deficient Mutant of Human PKC␤II-A kinase-deficient mutant of human PKC␤II (kdPKC␤II) was generated by PCR-based site-directed mutagenesis using the QuikChange XL site-directed mutagenesis kit (Stratagene). Full-length wild type human PKC␤II was used as a template, and the mutagenic primer, 5Ј-GGC ACA GAT GAG CTC TAT GCT GTG TGG ATC CTG AAG AAG GAC G-3Ј, was used to replace lysine 372, a residue essential for ATP binding and activity, with a tryptophan residue (12). The kdPKC␤II mutant was inserted into the pBABE/FLAG/puro retroviral expression vector, and virus stocks were produced using Phoenix-E cells.
Retrovirus Production and Transient Transfection-Recombinant retroviruses containing Myc-tagged dominant negative Rac1 (RacN17) or constitutively active Rac1 (RacV12) were generated by excising the Myc-tagged Rac1 constructs from pEXV/Rac vectors (13) with EcoRI and ligating them into the EcoRI site of the LZRS-GFP retrovirus. Each construct was confirmed by sequence analysis. LZRS-GFP-Rac1 retroviruses were used to infect RIE cells and derivative cell lines as described above.
Preparation of Recombinant PAK and Rhotekin GST Fusion Proteins-GST⅐PAK and GST⅐Rhotekin were used to assay for Rac1 and RhoA activity, respectively. Escherichia coli transformed with the GST⅐PAK or GST⅐Rhotekin constructs were grown at 37°C to an absorbance of 0.6 -0.7. Expression of the fusion proteins was induced with isopropyl-␤-D-thiogalactopyranoside (1 mM) for 3 h at 30°C. Bacteria were harvested by centrifugation at 5000 ϫ g for 10 min. The supernatant was discarded, and the bacterial pellet was suspended in 2 ml of ice-cold Nonidet P-40 lysis buffer (1ϫ phosphate-buffered saline buffer containing 0.2% Nonidet P-40, 25 g/ml aprotinin, 20 g/ml leupeptin, and 1 mM phenylmethylsulfonyl fluoride), sonicated, and centrifuged at 14,000 ϫ g for 10 min at 4°C. The supernatant was collected and incubated with glutathione-Sepharose beads (Amersham Biosciences) for 1 h at 4°C with gentle rocking. The GST⅐PAK or GST⅐Rhotekin beads were washed twice in Nonidet P-40 lysis buffer and twice in 1ϫ phosphate-buffered saline buffer.
Ras, Rac1, and RhoA Activity Assays-Ras, Rac1, and RhoA activities were assessed by affinity isolation of GTP-bound Ras, Rac1, or RhoA using binding domains of Raf-1, Pak, and Rhotekin, respectively, essentially as described (14,15). Briefly, cells were lysed into 500 l of lysis buffer (20 mM HEPES, pH 7.4, 100 mM NaCl, 0.5% Nonidet P-40, 0.2% deoxycholic acid, 5 mM MgCl 2, 2.5 mM NaF, 1 mM sodium vanadate, 25 g/ml aprotinin, 20 g/ml leupeptin, 1 mM phenylmethylsulfonyl fluoride) at 4°C for 5 min. Cellular debris was removed by centrifugation at 20,000 ϫ g for 5 min, and supernatants were transferred to new tubes containing 10 l of Ras-binding domain of Raf-1 (Raf-1⅐RBD) coupled to agarose beads (Cytoskeleton, Inc.), GST⅐Rho-binding domain of PAK (PAK⅐RBD), or GST⅐Rhotekin⅐RBD immobilized on glutathione-Sepharose beads. An aliquot of each supernatant was reserved to determine total Ras, Rac1, RhoA, and actin expression by immunoblot analysis. Following a 45-min incubation at 4°C, the agarose beads were collected by centrifugation, the supernatant was discarded, and the agarose beads were washed twice in wash buffer (20 mM HEPES, pH 7.4, 100 mM NaCl, 0.1% Nonidet P-40, 5 mM MgCl 2, 2.5 mM NaF, 1 mM sodium vanadate, 25 g/ml aprotinin, 20 g/ml leupeptin, 1 mM phenylmethylsulfonyl fluoride). Bound proteins were solubilized by the addition of 25 l of SDS sample buffer, resolved by SDS-PAGE, and subjected to immunoblot analysis using Ras, Rac1, or RhoA antibody as appropriate. Cells were incubated for 24 h in serum-free medium and stimulated with 5% fetal bovine serum for the times indicated in the figure legends prior to assay. Ras and Rac1 activity was quantitated by densitometry.
Invasion Assay-Cell invasion was assayed using Matrigel-coated Transwell cell culture chambers (6.5-mm diameter, 8-m pore size; BD Biosciences). Cells maintained for 24 h in serum-free-medium were trypsinized, resuspended in serum-free DME medium, and placed in the upper chamber of the Transwell insert (5 ϫ 10 4 cells/well), and DME medium containing 10% fetal bovine serum was added to the lower chamber. After 22 h at 37°C, 5% CO 2 , non-invasive cells in the upper chamber were removed, and invasive cells were fixed in 100% methanol and stained with 0.5% crystal violet in 2% ethanol. Five random microscopic fields at ϫ400 magnification were counted in each filter using a calibrated ocular grid. Experiments were carried out in triplicate. The data are expressed as the average number of cells/field ϮS.E. In some cases, the data are presented as -fold change from RIE cells, with the value for RIE cells set as 1. In some cases, cells were treated with LY379196 (50 nM), U0126 (10 M), or celecoxib (25 M) for 24 h prior to being subjected to the invasion assay using medium containing the appropriate inhibitor at the concentration indicated above.

PKC␤II Induces an Invasive Phenotype in RIE Cells-We
previously characterized transgenic mice expressing elevated levels of PKC␤II in the colonic epithelium (7,9,10). These mice exhibit colonic hyperproliferation and enhanced colon carcinogenesis (9). To elucidate the molecular mechanisms by which PKC␤II promotes colon carcinogenesis, we established a cell model system to dissect PKC␤II-mediated signaling pathways (10,11). RIE cells are immortalized but not transformed and, like most cells within the intestinal epithelium, express little or no PKC␤II (Fig. 1A). Therefore, we established RIE cells stably expressing human PKC␤II by transgenesis (Fig. 1A). Surprisingly, RIE/PKC␤II cells acquire a highly invasive phenotype (Fig. 1B). Treatment of RIE and RIE/PKC␤II cells with the selective PKC␤ inhibitor LY379196 (16) blocked invasion of RIE/PKC␤II cells while having no effect on RIE cells (Fig. 1C). Unlike RIE/PKC␤II cells, RIE cells expressing a kinase-defi- cient PKC␤II mutant (RIE/kdPKC␤II) do not exhibit enhanced invasion (Fig. 1D). Taken together, these results demonstrate that PKC␤II induces an invasive phenotype that is dependent upon PKC␤II activity.
PKC␤II Activates K-Ras in RIE Cells-There is evidence for reciprocal cross-talk between PKC and Ras (17)(18)(19)(20). Many cellular effects of the PKC activator phorbol myristate acetate are dependent upon Ras, and PKC activates cellular Ras in response to many extracellular stimuli (19). Therefore, we measured Ras expression and activity in RIE and RIE/PKC␤II cells (Fig. 2). RIE and RIE/PKC␤II cells express abundant K-Ras but very little H-Ras ( Fig. 2A), consistent with the epithelial nature of RIE cells. Expression of PKC␤II in RIE cells has no demonstrable effect on the level of K-Ras expression and does not induce H-Ras expression. We next assessed the level of K-Ras activity in RIE and RIE/PKC␤II cells in the presence and absence of serum (Fig. 2B). Although RIE cells exhibit little or no detectable active Ras in the absence of serum, RIE/PKC␤II cells show a significant amount of active Ras under serum-free conditions. Treatment of cells with 5% serum leads to transient activation of Ras in both RIE and RIE/ PKC␤II cells. However, both the duration and the magnitude of serum-stimulated Ras activation are higher in RIE/PKC␤II cells than in RIE cells.
Expression of oncogenic Ras in RIE cells induces invasion (21). To assess whether PKC␤II-mediated invasion is Ras-dependent, we initially measured, and then inhibited, the activity of the Ras effector, Mek. RIE/PKC␤II cells exhibit an elevated level of phosphorylated, active Mek when compared with RIE cells (Fig. 2C). The increase in phospho-Mek in RIE/PKC␤II cells reflects increased Mek activity since it is not accompanied by a corresponding increase in total Mek levels. Treatment of RIE and RIE/PKC␤II cells with the selective Mek 1 and 2 inhibitor U0126 (22) blocked invasion of RIE/PKC␤II cells while having no effect on RIE cells, indicating that PKC␤IImediated invasion is dependent upon Mek activity (Fig. 2D).
We recently demonstrated that PKC␤II induces Cox-2 and suppresses TGF␤RII in RIE/PKC␤II cells and that Cox-2 activity is required for PKC␤II-mediated resistance to TGF-␤ and suppression of TGF␤RII expression (11). However, treatment of RIE and RIE/PKC␤II cells with the selective Cox-2 inhibitor celecoxib had no effect on the invasive behavior of either RIE or RIE/PKC␤II cells, demonstrating that PKC␤II-mediated invasion does not require Cox-2 activity (Fig. 2D).
PKC ␤II Activates Rac1, but Not RhoA, in RIE Cells-Oncogenic Ras-mediated invasion requires the GTPase activity of Rac1 (13,23). Interestingly, RIE/PKC␤II cells exhibit elevated levels of GTP-bound, active Rac1 when compared with RIE cells (Fig. 3A). Densitometric analysis of multiple experiments demonstrated that RIE/PKC␤II cells contain 5-fold higher Rac1 activity than RIE cells without an increase in the total levels of Rac1 (Fig. 3B). RIE and RIE/PKC␤II cells exhibit a similar, low level of RhoA activity in the absence of serum (Fig.  3C). Furthermore, treatment with serum leads to activation of RhoA to a similar extent in RIE and RIE/PKC␤II cells. Therefore, PKC␤II induces activation of Rac1 but not of the related GTPase RhoA in RIE/PKC␤II cells. LY379196 reduces Rac1 activity in RIE/PKC␤II cells to levels comparable with those observed in RIE cells while having no effect on Rac1 activity in RIE cells (Fig. 4A). Furthermore, although RIE/PKC␤II cells exhibit elevated Rac1 activity, RIE/kdPKC␤II cells exhibit Rac1 activity indistinguishable from RIE cells (Fig. 4B). Taken together, these results demonstrate that PKC␤II-mediated Rac1 activation, like cell invasion, requires PKC␤II activity.
PKC␤II-mediated Invasion Requires Rac1 and the Atypical PKC Isozyme, PKC-We next assessed whether Rac1 activity is required for PKC␤II-mediated invasion. Expression of a dominant negative Rac1 mutant, RacN17, blocks invasion of RIE/PKC␤II cells while having no effect on RIE cells (Fig. 4C), indicating that Rac1 activity is required for PKC␤II-mediated invasion. The atypical PKC isozyme PKC resides in a complex with the cell polarity protein Par6 and Rac1, where its activity may regulate signaling to Rac1 (24 -27). We recently demonstrated that oncogenic Ras-mediated invasion requires both Rac1 and PKC, indicating the involvement of this complex in Ras-mediated invasion (8). Therefore, we assessed the involvement of PKC in PKC␤II-mediated invasion. Genomic and immunoblot analysis demonstrated that PKC␤II does not induce A, RIE and RIE/PKC␤II cells were subjected to immunoblot analysis for K-Ras, H-Ras, and actin as described under "Experimental Procedures." B, RIE and RIE/ PKC␤II cells were incubated in the absence of serum for 24 h to induce quiescence and then stimulated with 5% fetal bovine serum (FBS) for the indicated times. Cells were lysed and assayed for Ras activity using the Ras-binding domain of Raf-1 as an affinity ligand. Ras-GTP and total Ras were detected by immunoblot analysis. Data are representative of three independent experiments. C, RIE and RIE/ PKC␤II cells were incubated in the absence of serum for 24 h as described in A or maintained in serum-containing medium. Cell lysates were subjected to immunoblot analysis for phospho-Mek, total Mek, and actin. D, RIE and RIE/PKC␤II cells were incubated in the presence of the Mek inhibitor U0126, the Cox-2 inhibitor celecoxib, or diluent control and assayed for invasion as significant changes in PKC mRNA or protein expression (data not shown). However, expression of a kinase-deficient mutant of PKC (kdPKC), which acts in dominant negative fashion (28), inhibits invasion in RIE/PKC␤II cells while having no effect on RIE cells (Fig. 4D). Thus, both oncogenic Ras-and PKC␤II-mediated invasion are dependent upon PKC. We recently demonstrated that Rac1 is downstream of PKC in cellular invasion and Ras-mediated transformation (8).
RIE cells transfected with a constitutively active Rac1 mutant (RIE/RacV12 cells) exhibit an invasive phenotype similar to that observed in RIE/PKC␤II cells (Fig. 5A). Furthermore, although invasion of RIE/PKC␤II cells is blocked by LY379196, RIE/PKC␤II cells expressing RacV12 exhibit an invasive phenotype that is not blocked by LY379196 (Fig. 5B). Thus, Rac1 functions downstream of PKC␤II and is sufficient to induce invasion. RIE/RacV12 cells exhibit elevated levels of phospho-Mek (Fig. 5C), consistent with reports that Rac1 can activate Mek (29,30). Furthermore, invasion of RIE/RacV12 cells is blocked by U0126, indicating that active Mek is required for RacV12-mediated invasion (Fig. 5D). Taken together, these results demonstrate that Rac1 is the relevant downstream target for PKC␤II-mediated invasion and that active Rac1 is both necessary and sufficient to induce an invasive phenotype in RIE cells through activation of Mek. DISCUSSION PKC␤II plays a critical promotive role in colon carcinogenesis. PKC␤II is elevated in both mouse and human colon tumors and in aberrant crypt foci in azoxymethane-treated mice, indicating that induction of PKC␤II is an early event in colon carcinogenesis (7). Expression of PKC␤II in the colonic epithelium of transgenic mice induces hyperproliferation and increased susceptibility to carcinogen-induced aberrant crypt foci and tumor formation (9), demonstrating that PKC␤II promotes colon carcinogenesis.
We have used rat intestinal epithelial cells as a model of the normal colonic epithelium to investigate the molecular mechanisms by which PKC␤II promotes colon carcinogenesis (10,11). Expression of PKC␤II in RIE cells induces several aspects of the transformed phenotype, including the acquisition of TGF-␤ resistance and an invasive phenotype (Fig. 6). RIE cells, like normal colonic epithelial cells, are growth-inhibited by TGF-␤ (9). However, RIE/PKC␤II cells no longer respond to TGF-␤ (9). We have demonstrated that PKC␤II induces a TGF-␤-resistant state in RIE cells through activation of a novel PKC␤II 3 Cox-2 3 TGF␤RII signaling axis (11). We also demonstrated that this pathway operates in the colonic epithelium of transgenic PKC␤II mice in vivo and that dietary -3 fatty acids exert their chemopreventive effects on colon carcinogenesis, at least in part, through inhibition of this pathway (11). Cox-2 is frequently overexpressed in colon cancers (31) and colon cancer cell lines (32,33) and has been implicated in multiple functions critical to colon carcinogenesis including angiogenesis (34), tumor cell proliferation (35,36), invasiveness (37), and metastatic potential (38 -41). Our data are consistent with the role of Cox-2 in the loss of TGF-␤ responsiveness, which is observed in the vast majority of colon cancers and colon cancer cell lines (42,43). PKC␤II-mediated induction of Cox-2 provides a plausible mechanism by which Cox-2 expression and TGF-␤ resistance is induced in the early stages of colon carcinogenesis. PKC␤II may also be responsible for maintaining Cox-2 expression and TGF-␤ resistance in established colon cancer cell lines and colon cancers.
Here, we demonstrate that PKC␤II induces an invasive phenotype in RIE cells. Activated K-Ras is important for PKC␤IImediated invasion since the MEK1-selective inhibitor U0126 blocks invasion of RIE/PKC␤II cells. Interestingly, although Cox-2 is critical for PKC␤II-mediated suppression of TGF␤RII (10), it does not appear to be involved in PKC␤II-mediated invasion. Thus, PKC␤II induces TGF-␤ resistance and invasion through two distinct pathways.
PKC␤II induces the activity of the Ras effector, Rac1, a critical downstream effector of Ras transformation (13). Rac1 is essential for Ras-mediated changes in the actin cytoskeleton that induce invasion (14) and is required for PKC␤II-mediated invasion. Expression of a constitutively active Rac1 mutant, RacV12, is sufficient to induce invasion in RIE cells in the absence of PKC␤II. Furthermore, invasion in RIE/PKC␤II cells transfected with RacV12 is no longer blocked by LY379196, demonstrating that active Rac1 is sufficient to induce PKC␤IIindependent invasion. Interestingly, RacV12 induces activation of Mek, the activity of which is required for RacV12induced invasion. Taken together, these data provide compelling evidence that both Rac1 and Mek are critical effectors of PKC␤II-induced invasion. The proinvasive activity of FIG. 3. PKC␤II activates Rac1 in RIE cells. A, RIE and RIE/ PKC␤II cells were grown in DME medium containing 5% fetal bovine serum for 2 days to a confluence of ϳ70%. Cells were lysed and assayed for Rac1 activity (top panel), total Rac1 (middle panel), and actin (lower panel) as described under "Experimental Procedures." B, Rac1 activity is plotted for RIE and RIE/PKC␤II cells relative to RIE cells. Data represent the mean Ϯ S.E. from four independent determinations. *, p Ͻ 0.002 versus RIE cells. C, RIE and RIE/PKC␤II cells were maintained in serum-free medium for 24 h to induce quiescence. Some cultures were then treated with 5% fetal bovine serum for 24 h. Lysates were assayed for RhoA activity (top panel), total RhoA (middle panel), and actin (lower panel). Data are representative of three independent experiments. PKC␤II can be explained through the activation of Ras, which is capable of activating both Rac1 and Mek. The fact that RacV12 can also activate Mek in the absence of PKC␤II suggests that Rac1 is the critical effector of invasion. The relative contribution of the canonical Ras/Raf pathway, and of Rac1, in the activation of Mek-dependent invasion will require further experimentation.
Our data are consistent with earlier reports that PKC isozymes are important regulators of cytoskeletal function (24, 44 -46). PKC␤II has been shown to interact with the actin cytoskeleton (44). Here we demonstrate that PKC␤II activates Rac1, and for the first time, demonstrate the functional importance of PKC-mediated activation of Rac1. We recently demonstrated that like Rac1, the atypical PKC isozyme, PKC, is required for Ras-mediated transformation (8). PKC resides between Ras and Rac1 in a pathway required for Ras-mediated invasion and anchorage-independent growth (8). PKC␤II-mediated invasion is also dependent upon PKC, providing further evidence for the involvement of Ras in PKC␤II-induced invasion. PKC resides in a complex with Rac1 and Par6 that regulates the polarity of epithelial cells (25)(26)(27). Our present data indicate that this complex is also involved in the process of cellular invasion through upstream input from PKC␤II and activated Ras. Future studies will be aimed at determining the role of PKC/Par6/Rac1 complexes in transformation and invasion.
Taken together, our data are consistent with a model in which PKC␤II induces invasion through activation of a PKC␤II 3 Ras 3 PKC/Rac1 3 Mek signaling pathway (Fig.  6). Our data have important implications for the role of PKC␤II  in colon carcinogenesis. We previously demonstrated that induction of PKC␤II expression occurs very early in colon carcinogenesis, prior to acquisition of oncogenic mutations such as activated Ras. PKC␤II activates cellular Ras, suggesting that PKC␤II activation represents a phenocopy of an activated ras allele. Thus, the pathway delineated in the present report could explain the critical role of PKC␤II in early carcinogenesis, prior to acquisition of a Ras mutation. An important question is whether PKC␤II is required to maintain the transformed phenotype of colon cancer cells harboring a Ras mutation or whether it becomes dispensable in the presence of oncogenic Ras. We have found that many human colon cancer cell lines express elevated levels of PKC␤II regardless of Ras mutational status, suggesting that PKC␤II plays an important function in the transformed phenotype of colon cancer cells even in the presence of oncogenic Ras. In this regard, PKC␤II may be required to maintain the TGF-␤-resistant state present in virtually all colon cancer cell lines and colon cancers or to maintain other aspects of the transformed phenotype not yet identified. Future studies will investigate the role of PKC␤II in transformed colon cancer cells.