Atypical Protein Kinase Cι Is an Oncogene in Human Non–Small Cell Lung Cancer

Introduction

Lung cancer is the leading cause of cancer death in the United States accounting for an estimated 160,440 deaths in 2004 ( 1). The 5-year survival rate is only 14%, underscoring the need for more effective modalities for prevention, diagnosis, prognosis, and treatment. Most lung cancer (∼80%) is classified as non–small cell lung cancer (NSCLC), with most remaining cases (∼18%) being SCLC ( 2). Early-stage NSCLC tumors are often treated with surgery and radiotherapy, whereas advanced-stage metastatic disease often receives combination chemotherapy ( 2). Unfortunately, despite surgical removal and adjuvant therapy, many early-stage NSCLCs relapse and become fatal.

Many genetic and epigenetic changes conspire to produce a malignant, invasive NSCLC ( 3). Common molecular changes include oncogenic mutation of K-ras (∼50%), up-regulation of c-myc (20-30%), mutation of p53 (∼50%), deletion of p16^Ink4A/Arf, and mutation of fragile histidine triad ( 4, 5). NSCLCs often overexpress the epidermal growth factor receptor (EGFR), ErbB2 (HER-2/neu), and/or the hepatocyte growth factor/scatter factor receptor (Met). Cytogenetic analysis reveals frequent chromosomal losses at 3p, 6q, 8p, 9p, 13q, 17p, and 19q and gains at 1q, 3q, 5p, and 8q ( 6).

The atypical protein kinase C (aPKC) isozymes PKCι and PKCζ are a structurally and functionally distinct subclass of PKCs. aPKCs exhibit diacylglycerol-, calcium-, and phosphatidylserine-independent catalytic activity due to a unique NH₂-terminal regulatory domain ( 7). aPKC activity is regulated by phosphoinositide-dependent kinase (PDK1) phosphorylation ( 8) and by interactions with upstream effectors including Ras ( 9). aPKCs function in the establishment of cell polarity ( 10) and in cell survival ( 9, 11, 12) through direct interactions with downstream adaptor molecules involving a PB1 domain within the NH₂-terminal regulatory region of aPKC ( 13). aPKCs establish epithelial cell polarity through regulation of Rac1 and/or cdc42 ( 13) and promote cell survival through the canonical nuclear factor-κB (NF-κB) pathway ( 14) to activate transcription of the antiapoptotic survival genes Bcl-XL ( 15) and c-IAP1 and c-IAP2 ( 16).

We recently showed that PKCι is overexpressed in many established human NSCLC cell lines ( 17). Furthermore, disruption of PKCι signaling blocks the transformed growth of A549 lung adenocarcinoma cells in vitro and tumorigenicity in vivo ( 17). Molecular dissection of signaling downstream of PKCι revealed that PKCι is necessary for transformed growth by activating a Rac1/Pak/mitogen-activated protein kinase kinase (MEK)/extracellular signal-related kinase (ERK) signaling axis, whereas PKCι is not tightly coupled to NF-κB activity in A549 cells ( 17). Herein, we provide compelling evidence that PKCι is a major cancer gene in human NSCLC. We find that PKCι is overexpressed in the vast majority of primary NSCLC tumors, that the PKCι gene is a target for NSCLC tumor-specific amplification, that gene amplification is an important mechanism by which PKCι expression is regulated in NSCLC, and that PKCι expression is a prognostic indicator of poor survival in NSCLC patients independent of tumor stage. Mechanistically, we show that disruption of PKCι signaling blocks the transformed growth of human NSCLC cells harboring PKCι gene amplification. We conclude that PKCι is a critical oncogene involved in human lung tumorigenesis.

Materials and Methods

Reagents. Antibodies were from the following sources and were used at the indicated concentrations: anti-PKCζ (Santa Cruz Biotechnology, Santa Cruz, CA; 1:100), PKCι (BD PharMingen, San Diego, CA; 1:4,000), actin (Santa Cruz Biotechnology, 1:2,000), and the FLAG epitope (Sigma, St. Louis, MO; 1:2,000). Recombinant human PKCι and PKCζ protein were from Upstate Biochemical (Charlottesville, VA).

Processing and analysis of human lung cancer tissues. H&E-stained sections of matched normal and lung tumor tissues were analyzed by a pathologist to confirm initial diagnosis, staging, and overall integrity of the tissue samples. Seventy-four cases of NSCLC [37 lung adenocarcinoma (LAC) and 37 squamous cell carcinoma (SCC)] and matched normal lung tissues were chosen for extraction of DNA, RNA, and protein. Ten 10-μm-thick slices were cut from each frozen tissue block. DNA was isolated in phenol/chloroform, total RNA isolated using RNAqueous 4PCR kit (Ambion, Austin, TX), and protein isolated by direct solubilization in SDS-PAGE sample buffer.

Analysis of PKCι protein expression in human lung tissue. Protein from human tumor samples was quantified using nitric acid mediated nitration of tyrosine ( 18). Equal amounts of protein (∼30 μg) from each sample was resolved in 12% SDS-PAGE gels (Invitrogen, Carlsbad, CA), transferred to polyvinylidene difluoride (PVDF) membrane (Immobilin-P, Millipore, Billerica, MA), and subjected to immunoblot analysis using the appropriate antibodies and Enhanced Chemiluminescence (ECL) Plus detection (Amersham, Piscataway, NJ) as described previously ( 19, 20). Images were obtained on X-omat AR film and antigens quantified by fluorescence detection using a Typhoon 9410 Variable Mode Imager. The fluorescent signal was analyzed using ImageQuant 5.2 software (Amersham).

Immunohistochemistry was done on paraffin-embedded sections of primary tumor and normal lung tissues. The tissue was deparaffinized by placing slides into three changes of xylene and rehydrated in a graded ethanol series. The rehydrated tissue samples were rinsed in water and subjected to antigen retrieval in citrate buffer (pH 6.0) as described by the manufacturer (DAKO, Carpinteria, CA). Slides were treated with 3% H₂O₂ for 5 minutes to reduce endogenous peroxidase activity and washed with PBS containing 0.5% (w/v) Tween 20. PKCι was detected using PKCι antibody at a 1:100 dilution in PBS/Tween and visualized using the Envision Plus Dual Labeled Polymer Kit following the manufacturer's instructions (DAKO). Images were captured and analyzed using ImagePro software.

Statistical and survival analysis. Cancer-specific survival was estimated using the Kaplan-Meier method. The duration of follow-up was calculated from the sample date to the date of death or last follow-up. The associations of the clinical and pathologic features studied with death from NSCLC were assessed using Cox proportional hazards regression models and summarized with risk ratios (RR) and 95% confidence intervals (95% CI). Natural logarithmic transformations were explored if the distributions of continuously scaled variables were not approximately normal. In addition, the relationships between continuously scaled variables and death from NSCLC were investigated using martingale residuals from the Cox model ( 21). Statistical analyses were done using the SAS software package (SAS Institute, Cary, NC) and Ps ≤ 0.05 were considered statistically significant.

Real-time PCR analysis for PKCι gene amplification. Genomic DNA from each sample or cell line was analyzed for amplification of PKCι using Taqman technology on an Applied Biosystems (Foster City, CA) 7900HT sequence detection system. The human RNaseP1 gene was used as a DNA template control and for normalization of results to total DNA. The primer/probe set for the human PKCι gene was forward primer, 5′-GGCTGCATTCTTGCTTTCAGA-3′; reverse primer, 5′-CCAAAAATATGAAGCCCAGTAATCA-3′; and probe, 5′-CAATCTTACCTGCTTTCT-3′. The primer/probe set for the RNaseP1 gene was designed and provided by ABI Assay on Demand. A tumor cell line or primary tumor was scored positive for PKCι gene amplification when analysis revealed a minimum of one additional PKCι allele when compared with the matched normal sample.

Cell culture, plasmids, transfections, and drug treatments. Human ChaGo, H520, H1299, and SK-MES-1 NSCLC cell lines as well as the nontransformed HBE4 lung epithelial cell line were obtained from American Type Culture Collection (Manassas, VA) and maintained as suggested by the supplier. The cells were maintained in a humidified tissue culture incubator at 37°C in 5% CO₂. H1299 and ChaGo cells were stably transfected with recombinant pBabe retroviruses containing Flag-tagged human full-length wild-type PKCι (wtPKCι), kinase dead PKCι (kdPKCι), or empty vector pBabe as described previously ( 19). Expression of FLAG-epitope-tagged PKCι and total PKCι was analyzed by immunoblot analysis as described previously ( 19).

Adherent growth kinetics of H1299 cells transfected with empty pBabe, pbabe/kdPKCι, and pbabe/wtPKCι were determined by plating cells (1 × 10⁴ cells per well) into multiwell culture dishes and monitoring cell growth daily over a 7-day period. Each day, cells from triplicate wells were trypsinized and counted using a hemocytometer.

Immunoblot analysis of non–small cell lung cancer cell lines. Cells were harvested by washing with PBS and scraping off the plate. The cell pellet was lysed in SDS sample buffer. Protein lysates were quantitated by using the nitration of tyrosine in nitric acid ( 18). Equal amounts of protein (∼20 μg) were loaded for each sample, resolved in 12% SDS-PAGE gels (Invitrogen), and transferred to PVDF membrane (Millipore Immobilin-P). A solution of 5% milk and PBS/Tween 20 was used for blocking. Western blot analysis was done with appropriate antibodies and detected using ECL Plus (Amersham).

Soft agar growth assays. Anchorage-independent growth was assayed by the ability of H1299 and ChaGo cell transfectants to form colonies in soft agar. The bottom agar consisted of growth medium containing 10% fetal bovine serum (FBS) and 0.75% agarose in 60-mm tissue culture dishes. Nine hundred cells were resuspended in growth medium containing 10% FBS and 0.75% agarose and plated on top of the bottom agar. The cells were incubated at 37°C in 5% CO₂. Cell colonies were visualized and quantified under a dissecting microscope (Olympus, Melville, NY) after 4 to 6 weeks in culture.

Results

PKCι is overexpressed in non–small cell lung cancer. We have implicated PKCι in cellular transformation in animal models ( 19, 20) and more recently in human NSCLC cells ( 17). To determine whether PKCι is relevant to human disease, we assessed PKCι expression in primary NSCLC tumors. Seventy-four cases of NSCLC equally representing the two major subtypes of NSCLC (37 LAC and 37 SCC) and matched normal lung tissues were selected for analysis using the following inclusion criteria: (a) patients had received no cancer therapy before tissue collection; (b) staging and diagnosis was verified by pathologic examination; (c) tumor samples were highly enriched in tumor cells (>90%) with no necrosis; and (d) sufficient tumor and matched normal tissue was available for analysis. The clinical and pathologic features of the NSCLC cases analyzed are given in Table 1 .

We began our analysis by assessing PKCι protein expression of matched normal and tumor samples by immunoblot analysis. Results from 10 representative cases analyzed by immunoblot analysis for PKCι, PKCζ, and actin are shown ( Fig. 1A ). As can be seen, all NSCLC cases exhibited demonstrable overexpression of PKCι when compared with matched normal lung tissue. Immunoblot analysis of all 74 cases showed that PKCι overexpression (defined as >2-fold increase in PKCι in tumor compared with matched normal) was evident in 51 of 74 (69%) cases. Quantitative analysis of the data showed a statistically significant increase in PKCι protein expression in NSCLC compared with normal lung tissue (P < 0.001; Fig. 1B). No tumors exhibited a significant decrease in PKCι compared with matched normal tissue. Interestingly, immunoblot analysis for the related aPKC isozyme, PKCζ, revealed that none of the tumors or matched normal lung tissues expressed detectable levels of PKCζ, indicating that PKCι is the predominant aPKC in normal and malignant human lung tissue. To determine whether the increase in PKCι protein expression in human NSCLC cases was accompanied by an increase in PKCι mRNA levels, quantitative real-time reverse transcriptase-PCR analysis for PKCι mRNA was done on both tumor and matched normal lung tissue (n = 39; Fig. 1C). This analysis showed that PKCι mRNA levels were significantly higher in NSCLC tumors when compared with matched normal lung tissue (P = 0.003). PKCι mRNA was routinely >10-fold higher than PKCζ mRNA in these tissues, confirming the predominance of PKCι in the human lung (data not shown).

• Download figure
• Open in new tab
• Download powerpoint

Figure 1.

PKCι is overexpressed in NSCLC. A, immunoblot analysis of 10 representative primary NSCLCs and matched normal lung tissue for PKCι, PKCζ, and actin. Recombinant human PKCζ was included to assure antibody specificity. B, quantitative analysis of PKCι expression in primary NSCLCs (n = 74). NSCLCs exhibit a statistically significant increase in PKCι protein expression. *, P < 0.001, compared with matched normal lung epithelium. C, quantitative analysis of PKCι mRNA abundance in primary NSCLCs by quantitative reverse transcriptase-PCR (n = 39). *, P < 0.008, compared with matched normal lung epithelium.

We next assessed whether the increase in PKCι expression represented an increase in PKCι expression in NSCLC tumor cells. Immunohistochemistry of paraffin sections of representative normal and NSCLC cases revealed light PKCι staining in normal lung epithelial cells with much more intense staining in the vast majority of tumor cells within the samples ( Fig. 2A ). Little or no staining was observed in tumor-associated stroma, confirming that PKCι overexpression is predominantly confined to lung cancer cells. Higher magnification revealed PKCι staining consistent with localization to the cytoplasm, plasma membrane, and nucleus of normal lung epithelial and tumor cells ( Fig. 2B).

• Download figure
• Open in new tab
• Download powerpoint

Figure 2.

Overexpression of PKCι is confined to lung cancer cells and correlates with poor survival in NSCLC patients. A, immunohistochemistry of normal lung, LAC, and SCC for PKCι. Bar, 100 μm. B, intracellular PKCι staining pattern in normal lung epithelial cells and lung adenocarcinoma cells. Bar, 10 μm. C, Kaplan-Meier survival curves. NSCLC patients were divided into two groups based on PKCι expression as described in Materials and Methods. High PKCι expression correlates with poor survival.

We next assessed whether PKCι expression in NSCLC tumors correlates with tumor stage and/or cancer-specific death in NSCLC patients. The average PKCι protein expression for the sample set was 3.1 (±1.9 SD) with a median of 2.4 and a range of 1.0 to 10.3. The RR for the association of PKCι expression with death from disease was 1.28 (95% CI, 1.06-1.54; P = 0.011), indicating that each 1-unit increase in PKCι expression was associated with a 28% increase in the risk of death from disease. Thus, PKCι protein expression is associated with increased risk of cancer-specific death.

We next subjected the NSCLC cases to martingale residual analysis to determine a useful cut point to assess the relationship between protein expression and death from NSCLC. Based on this analysis, the NSCLC cases were divided into two groups based on the level of PKCι expression (see Materials and Methods). Patients in the high PKCι expression group (PKCι >5, n = 12) were 2.6 times more likely to die from NSCLC than patients in the low PKCι expression group (RR, 2.58; 95% CI, 0.99-6.74; P = 0.05; Fig. 2C). The prognostic value of PKCι expression was comparable to tumor stage, which is currently one of the best prognostic indicators in NSCLC ( 22). Within our data set, patients with late-stage disease (stages III and IV) were ∼2.4 times more likely to die than those with early-stage disease (stages I and II; RR, 2.35; 95% CI, 1.03-5.36; P = 0.043). Interestingly, patients with early-stage disease had PKCι levels comparable with patients with late-stage disease because the median PKCι expression in low-stage tumors (stages I and II) is 3.1 (n = 27) compared with 3.2 for high-stage tumors (P = 0.688, n = 9).

We next assessed whether PKCι expression in early-stage disease was a useful predictor of outcome. In this subset of patients, those with high PKCι expression (>5) were almost 11 times more likely to die from their disease than patients with low PKCι expression (RR, 10.87; 95% CI, 0.96-122.66; P = 0.05). Therefore, PKCι expression in NSCLC patients predicts poor survival independent of tumor stage, indicating that PKCι expression is useful in identifying patients with early-stage disease at elevated risk of relapse.

PKCι gene amplification regulates PKCι expression in non–small cell lung cancer. Given the prevalence of PKCι overexpression in NSCLC tumors, we wished to investigate the mechanism by which tumor-specific overexpression is achieved. Of immediate interest to us was the fact that the PKCι gene resides at chromosome 3q26. Amplification of chromosome 3q26 is among the most frequent cytogenetic changes observed in NSCLC, occurring in ∼20% of all NSCLC ( 6, 23) predominantly in SCC. Because the PKCι gene resides at 3q26, it is possible that gene amplification is a mechanism by which PKCι is overexpressed in NSCLC tumors. To assess this possibility, we conducted quantitative real-time PCR to assess PKCι gene copy number in four established human lung squamous cell carcinoma cell lines Ski-Mes1, H520, H1299, and ChaGo. Our analysis revealed PKCι amplification in three of the four established human SCC cell lines tested (H520, H1299, and ChaGo cells) but not in Sk-Mes1 or nontransformed HBE4 lung epithelial cells ( Fig. 3A ). The presence of PKCι gene amplification correlated with the presence of chromosome 3q26 amplification in these cell lines ( 24), indicating that PKCι is part of the 3q26 amplicon. Furthermore, there is a positive correlation between PKCι gene copy number, PKCι mRNA abundance, and PKCι protein expression in these four cell lines, showing that gene amplification is an important mechanism driving PKCι expression in SCC cells ( Fig. 3A).

• Download figure
• Open in new tab
• Download powerpoint

Figure 3.

The PKCι gene is frequently amplified in NSCLC cell lines and primary tumors. A, analysis of the human NSCLC Sk-Mes1, H520, H1299, and ChaGo cell lines and the nontransformed HBE4 lung epithelial cell line for PKCι gene copy number, mRNA abundance, and protein expression. PKCι gene amplification correlates with PKCι mRNA abundance and PKCι protein expression. B, primary lung adenocarcinomas rarely exhibit PKCι gene amplification. PKCι gene copy number was determined by quantitative real-time PCR as described in Materials and Methods. PKCι gene copy values were normalized to the value of the single copy gene RNaseP in the same samples. Gene amplification was defined as a PKCι gene copy number that is more than 1 SD above the PKCι gene copy number determined in patient-matched normal lung tissues (n = 74). Horizontal line, cutoff for defining gene amplification based on this criterion. C, primary squamous cell carcinomas frequently harbor PKCι gene amplification. Gene amplification was defined as stated in (B).

We next assessed whether PKCι gene amplification also occurs in primary NSCLC tumors. For this purpose, quantitative real-time PCR analysis for PKCι gene copy number was conducted on all 37 LAC ( Fig. 3B) and all 37 SCC tumors ( Fig. 3C). We observed PKCι gene amplification (defined as a tumor-specific gain of one or more PKCι alleles) in 27 of 74 (36.5%) of NSCLC tumors. The majority of cases exhibiting PKCι gene amplification were SCCs (26 of 27 or 96%) consistent with the distribution of chromosome 3q26 amplification in NSCLC ( 6, 23).

To determine whether PKCι gene amplification drives PKCι overexpression in SCC, we analyzed primary SCC tumors for PKCι protein expression. Immunohistochemical analysis showed that SCCs harboring PKCι gene amplification expressed higher levels of PKCι protein than did SCC tumors without PKCι amplification ( Fig. 4A ). Quantitative analysis of PKCι expression of all SCC tumor samples (n = 37) showed that tumors harboring PKCι gene amplification expressed significantly higher levels of PKCι protein than SCC tumors without PKCι gene amplification ( Fig. 4B). In addition, there is a positive correlation between PKCι gene copy number and PKCι protein expression in SCC tumors ( Fig. 4C). Taken together, these data show that gene amplification drives PKCι expression in a significant subset of SCC tumors. Interestingly, there was no statistically significant difference in PKCι protein expression in LAC and SCC tumors (mean PKCι expression of 3.0 in LAC versus 2.9 in SCC; P = 0.53) despite the fact that PKCι gene amplification is largely confined to SCC tumors. These data suggest that PKCι expression may be regulated by distinct mechanisms in these tumor types.

• Download figure
• Open in new tab
• Download powerpoint

Figure 4.

PKCι gene amplification drives PKCι protein expression in SCCs. A, immunohistochemistry of representative primary SCC tumors that either do or do not harbor PKCι gene amplification. B, SCC tumors that harbor PKCι gene amplification express higher PKCι protein levels than tumors without PKCι gene amplification (n = 74). C, PKCι gene copy number correlates with PKCι protein expression in SCC (n = 37).

PKCι is required for the transformed phenotype of squamous cell carcinoma cells harboring PKCι gene amplification. We recently showed that PKCι is overexpressed in many human lung cancer cell lines and that PKCι plays a critical role in the transformed growth of A549 lung adenocarcinoma cells, a cell line that does not harbor PKCι gene amplification ( 17). Because the PKCι gene is within the chromosome 3q26 amplicon, it is possible that other genes within the amplicon are the critical targets of amplification and that PKCι is dispensable for the transformed phenotype of SCC cells harboring chromosome 3q26 amplification. Several candidate oncogenes reside in the chromosome 3q26 region including the Ski-like gene SnoN ( 25), the catalytic subunit of phosphatidylinositol-3 kinase (PI3Kα; ref. 26), the Evi1 oncogene ( 25), and the RNA component of human telomerase ( 24). However, the importance of these genes in NSCLC formation has not been systematically evaluated and it is likely that multiple genes within the 3q26 amplicon are functionally targeted in NSCLC tumors. To assess the role of PKCι in the transformed growth of SCC cells harboring PKCι gene amplification, we established H1299 cells that stably express either wild-type human PKCι (wtPKCι) or a kinase deficient, dominant-negative PKCι allele (kdPKCι; Fig. 5A ). Interestingly, expression of either wtPKCι or kdPKCι had no effect on the growth of H1299 cells in adherent culture ( Fig. 5B). In contrast, expression of kdPKCι significantly blocked the ability of H1299 cells to grow as anchorage-independent colonies in soft agar, whereas expression of wtPKCι had no effect on soft agar growth ( Fig. 5C). Similar results were obtained in ChaGo cells expressing kdPKCι, which exhibited significant inhibition of soft agar growth compared with control vector expressing cells ( Fig. 5D). These results are similar to those obtained in A549 cells ( 17), indicating that PKCι is required for transformed growth of human lung cancer cells regardless of whether they harbor PKCι gene amplification or not. Our data show that PKCι is a critical target of the chromosome 3q26 amplicon.

• Download figure
• Open in new tab
• Download powerpoint

Figure 5.

PKCι is required for transformation of H1299 and ChaGo lung cancer cells that harbor PKCι gene amplification. A, immunoblot analysis of H1299 cells stably transfected with pBabe, wtPKCι, or kdPKCι for Flag, PKCι, and actin. B, growth kinetics of H1299 cell transfectants in adherent culture in the presence of 10% serum. C, anchorage-independent growth of H1299 cell transfectants in soft agar. D, immunoblot analysis of ChaGo cells stably transfected with empty pBabe or kdPKCι for Flag, PKCι, and actin. Analysis of anchorage-independent growth of ChaGo cell transfectants in soft agar.

Discussion

Since the original discovery of PKCs nearly 30 years ago, these enzymes have been implicated in the regulation of cellular proliferation, differentiation, and survival. The identification of PKC as a major cellular receptor for tumor-promoting phorbol esters strongly suggested a functional link between PKC and cancer. However, despite intensive study, a direct link between a PKC isozyme and human cancer has remained elusive. In this report, we provide compelling evidence that the aPKCι isozyme is a critical cancer gene in NSCLC. Our evidence for this conclusion is multifold. First, PKCι is overexpressed in NSCLC cell lines and primary tumors. The prevalence of PKCι overexpression in NSCLC is similar to that of other oncogenes implicated in NSCLC development such as Met, EGFR, PI3K, and Raf ( 27).

Second, PKCι expression predicts poor survival in NSCLC patients. PKCι expression is a useful prognostic indicator of poor survival independent of tumor stage, currently the most reliable prognostic marker for NSCLC ( 22). Therefore, PKCι expression holds considerable promise as a novel prognostic tool to identify early-stage NSCLC patients at elevated risk of progression or relapse. Elevated risk patients could be candidates for aggressive adjuvant therapy and diligent surveillance after surgical removal of their primary tumor.

Third, like many other oncogenes, PKCι is a target for frequent, tumor-specific somatic genetic alteration by gene amplification. The PKCι gene resides at chromosome 3q26, the most frequent site of chromosomal gain in NSCLC ( 6, 23). PKCι gene amplification occurs frequently and correlates with PKCι expression in both established human NSCLC cell lines and primary NSCLC tumors. Therefore, PKCι gene amplification is an important mechanism driving PKCι overexpression in NSCLC. Because PKCι is required for NSCLC cell growth and tumorigenicity in vivo, our data provide compelling evidence that PKCι is a critical target for gene amplification within the 3q26 amplicon that promotes NSCLC carcinogenesis. Our results take on even broader significance given the fact that chromosome 3q26 amplification also occurs frequently in SCC of the head and neck ( 28), esophagus ( 25, 29), cervix ( 30), and ovary ( 6, 31). It is very likely that the PKCι gene is also amplified in these tumors.

Fourth, we show that PKCι is required for the transformed domain of NSCLC cells harboring PKCι gene amplification. Expression of a dominant-negative, kinase-deficient PKCι allele blocks the transformed growth of both H1299 and ChaGo cells in vitro. These results are consistent with our recent finding that PKCι is required for the transformed growth and tumorigenicity of human A549 lung adenocarcinoma cells ( 17), indicating that PKCι is required for transformation of many human NSCLC cells. We recently identified Rac1 as a critical downstream effector of PKCι-dependent transformation and elucidate a PKCι → Rac1 → Pak → MEK1,2 → ERK1,2 signaling pathway that is necessary for A549 cell tumor growth in vivo ( 17). These results are interesting in light of those obtained in other cell systems. In chronic myelogenous leukemia cells, we showed that PKCι is required for Bcr-Abl-mediated transformation and identified NF-κB as a requisite downstream effector of PKCι-dependent cell survival ( 11, 12, 32). In contrast, in rat intestinal epithelial cells, we identified the small molecular weight GTPase Rac1 as a critical downstream effector of oncogenic Ras-mediated, PKCι-dependent transformation ( 19). Thus, it seems that PKCι can contribute to transformation through activation of at least two different signaling pathways depending upon cellular context. It will be of interest to determine whether the same or different PKCι signaling pathways are operative in other human tumor types.

In conclusion, this is the first report since the discovery of PKC almost 30 years ago to conclusively show that a PKC isozyme possesses the properties of a human oncogene. PKCι is overexpressed in NSCLC tumors. PKCι expression is predictive of poor clinical outcome. The PKCι gene is functionally targeted for genetic alteration in a significant subset of NSCLC tumors. PKCι activity is required for the transformed phenotype of NSCLC cells. Finally, PKCι activates a critical signaling pathway that is required for NSCLC tumor growth. Our finding that inhibition of PKCι signaling has a profound inhibitory effect on transformed growth, while having little or no effect on adherent cell growth, suggests that PKCι may be a particularly attractive target for the development of novel, mechanism-based therapies for the treatment of NSCLC. In this regard, we predict that NSCLC tumors harboring PKCι gene amplification may be particularly responsive to PKCι-directed therapies.