• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information
Logo of nihpaAbout Author manuscriptsSubmit a manuscriptNIH Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
Mol Cancer Ther. Author manuscript; available in PMC Feb 1, 2012.
Published in final edited form as:
PMCID: PMC3061393

Knockdown of Oncogenic KRAS in Non-Small Cell Lung Cancers Suppresses Tumor Growth and Sensitizes Tumor Cells to Targeted Therapy


Oncogenic KRAS is found in >25% of lung adenocarcinomas, the major histologic subtype of non-small cell lung cancer (NSCLC), and is an important target for drug development. To this end, we generated four NSCLC lines with stable knockdown selective for oncogenic KRAS. As expected, stable knockdown of oncogenic KRAS led to inhibition of in vitro and in vivo tumor growth in the KRAS mutant NSCLC cells, but not in NSCLC cells that have wild-type KRAS (but mutant NRAS). Surprisingly, we did not see large-scale induction of cell death and the growth inhibitory effect was not complete. To further understand the ability of NSCLCs to grow despite selective removal of mutant KRAS expression, we performed microarray expression profiling of NSCLC cell lines with or without mutant KRAS knockdown and isogenic human bronchial epithelial cell lines (HBECs) with and without oncogenic KRAS. We found that while the MAPK pathway is significantly down-regulated after mutant KRAS knockdown, these NSCLCs showed increased levels of phospho-STAT3 and phospho-EGFR, and variable changes in phospho-Akt. In addition, mutant KRAS knockdown sensitized the NSCLCs to p38 and EGFR inhibitors. Our findings suggest that targeting oncogenic KRAS by itself will not be sufficient treatment but may offer possibilities of combining anti-KRAS strategies with other targeted drugs.


The development of NSCLC is a multi-step process involving a number of genetic and epigenetic abnormalities acquired over time (1). Based on the recent successes of molecularly targeted therapy in some cancers, many of these genetic and epigenetic lesions may represent potential therapeutic targets for NSCLC (2, 3). The proto-oncogene KRAS is one of the most important of these potential targets because its mutation is common in many cancers and it sits at the apex of multiple growth regulatory cascades (4, 5). KRAS encodes a small GTP-binding protein that is involved in many cellular processes including proliferation, differentiation, and apoptosis (4). Wild-type KRAS has intrinsic GTPase activity, which catalyzes the hydrolysis of bound GTP to GDP thereby inactivating the RAS growth-promoting signal, while oncogenic KRAS is locked into the GTP-bound state, leading to constitutive RAS signaling. KRAS mutations are detected in >25% of lung adenocarcinomas (6), 85% of which affect codon 12 (1), and these mutations are associated with poor prognosis in NSCLC patients (7). Thus, oncogenic mutations of KRAS play an important role in the development of NSCLC.

Although several strategies to inhibit KRAS including farnesyltransferase inhibitors have been explored, these approaches tend not to be specific for the mutant form of KRAS and so lead to inhibition of wild-type KRAS activity, which is essential for normal growth and development (8, 9). The relative failure of KRAS inhibitors in cancer clinical trials likely derives from specificity issues as well as from differences in how mutant KRAS controls its downstream effectors in different individual tumors. In this study, we use an RNA interference (RNAi) based approach specifically targeting mutant KRAS mRNA to investigate how loss of oncogenic KRAS signaling affects the malignant phenotype of NSCLCs. Recent studies have demonstrated the potential of gene expression profiling analysis along with RNAi technology to uncover oncogenic KRAS-specific gene signatures in lung cancers and other types of cancers (10, 11). In this study, we used a new approach to uncover oncogenic KRAS-specific gene signatures by microarray gene expression profiling with RNAi-mediated mutant-specific KRAS knockdown in NSCLC cells and mutant KRAS-transformed bronchial epithelial cells. Overall, our results show that KRAS mutation has canonical MAPK-dependent effects on cell proliferation and the malignant phenotype in NSCLC, but that among different NSCLCs mutant KRAS can lead to different outputs in cellular signaling that affect cell survival. Our findings suggest that by itself, oncogenic KRAS is not an ‘Achilles heal’ of NSCLC and that treatment of NSCLCs with KRAS mutations will require knowledge of other tumor molecular abnormalities, which in turn provide additional targeted therapy opportunities for KRAS mutation-positive NSCLC patients.

Materials and Methods

Cell lines

Five NSCLC cell lines, NCI-H23, H1792, H358, H441 and H1299 and four HBEC lines were obtained from the Hamon Center collection (University of Texas Southwestern Medical Center). All lines were genotyped by STR analysis in accordance with AACR best practices. HBEC3 cells were established by retroviral-transfection with CDK4 and the catalytic component of telomerase (hTERT) and four variants of an isogenic series of HBECs were used in this study: HBEC3, HBEC3/mutant KRAS (HBEC3K), HBEC3/shRNA targeting p53 (HBEC3p53), and HBEC3/mutant KRAS/shRNA targeting p53 (HBEC3K53) (12). Characteristics of these cell lines are summarized in Table 1 (1318). Cancer cells were cultured with RPMI 1640 medium supplemented with 5% fetal bovine serum. HBEC3 and its derivatives were cultured with Keratinocyte-SFM (Invitrogen, Carlsbad, CA) medium with 50 μg/ml bovine pituitary extract (Invitrogen) and 5 ng/ml EGF (Invitrogen).

Table 1
Characteristics of non-small cell lung cancer cell lines and immortalized human bronchial epithelial cell lines.

Construction and use of retroviral vectors

To provide specific oncogenic KRAS knockdown, retroviral vectors producing shRNA against mutant KRAS were constructed by inserting annealed 64-mer sense and antisense oligos into pSUPER.retro (pRS) (OligoEngine, Seattle, WA) as described (19). The 64-mer oligos were as follows: pRS-KRAS-V12, 5′-gatccccGTTGGAGCTGTTGGCGTAGttcaagagaCTACGCCAACAGCTCCAACtttttggaaa-3′ (sense) and 5′-agcttttccaaaaaGTTGGAGCTGTTGGCGTAGtctcttgaaCTACGCCAACAGCTCCAACggg-3′ (anti-sense); pRS-KRAS-C12, 5′-gatccccGTTGGAGCTTGTGGCGTAGttcaagagaCTACGCCACAAGCTCCAACtttttggaaa-3′ (sense) and 5′-agcttttccaaaaaGTTGGAGCTTGTGGCGTAGtctcttgaaCTACGCCACAAGCTCCAACggg-3′ (anti-sense). The sequences targeting the mutation in KRAS are indicated in capitals in the oligonucleotide sequences. Cells were infected with retroviral vectors (12). Briefly, the pRS vector was co-transfected with pVPack-GP and pVPackVSV-G vectors (Stratagene, La Jolla, CA) into 293T cells by using FuGENE 6 transfection reagent (Roche, Indianapolis, IN) according to the manufacturers’ protocol. After 48 h, the culture medium was filtered through a 0.45 μm filter to obtain retroviral supernatants. Cells were then infected with the retroviral supernatants and 4 μg/ml polybrene, and after 10 h, supernatants were removed and cells were grown with complete growth medium for an additional 24 h. Infected cells were then selected with 1 μg/ml puromycin. After seven days of puromycin selection, cells were used for analyses.

Transfection of synthetic small interfering RNA (siRNA)

siRNAs targeting against the KRAS mutations were designed and purchased from Dharmacon (Lafayette, CO). The siRNA sequences were 5′-GUUGGAGCUUGUGGCGUAGTT-3′ (sense) and 5′-CUACGCCACCAGCUCCAACTT-3′ (anti-sense) for the KRAS G12C mutation, and 5′-GUUGGAGCUGUUGGCGUAGTT-3′ (sense) and 5′-CUACGCCAACAGCUCCAACTT-3′ (antisense) for the KRAS G12V mutation. siRNAs were transfected into cells using Oligofectamine transfection reagent (Invitrogen, Carlsbad, CA) as described (20). Three days after the transfection, cells were harvested for analysis.

Western Blot analysis

Western Blot was performed using whole cell lysates, separated on SDS/polyacrylamine gel, and electro-blotted to nitrocellulose membranes (Schleicher & Schuell, Keene, NH) as described (20). The membranes were incubated with mouse monoclonal anti-KRAS (Santa Cruz, Santa Cruz, CA), mouse monoclonal anti-Actin (Sigma, St. Louis, MO), rabbit polyclonal anti-MEK1/2 (Cell Signaling, Beverly, MA), rabbit polyclonal anti-phospho-MEK1/2 (Cell Signaling), rabbit polyclonal anti–extracellular signal-regulated kinase (ERK) 1/2 (Cell Signaling), rabbit polyclonal anti-phospho-ERK1/2 (Cell Signaling), rabbit polyclonal anti-Akt (Cell Signaling), rabbit polyclonal anti-phospho-Akt (Thr308; Cell Signaling), rabbit polyclonal anti-signal transducers and activators of transcription (STAT) 3 (Cell Signaling), rabbit polyclonal anti-phospho-STAT3 (Tyr705; Cell Signaling), rabbit polyclonal epidermal growth factor receptor (EGFR; Cell Signaling) and rabbit polyclonal anti-phospho-EGFR (Tyr1068; Cell Signaling) antibodies. The membranes were developed with peroxidase-labeled anti-mouse or anti-rabbit IgG (Amersham Pharmacia, Piscataway, NJ) by Super Signal chemiluminescence substrate (Pierce, Rockford, IL). Actin protein levels were used as a control for adequacy of equal protein loading. Protein expression levels were quantified by densitometry analysis.

RT-PCR and Restriction fragment length polymorphism (RFLP) analysis

To detect transcripts of wild-type or mutant KRAS, PCR-RFLP method was performed as described (12, 21). Total RNA was extracted from cells using RNeasy Mini Kit (QIAGEN, Valencia, CA) and cDNA was synthesized using 2 μg of total RNA with the SuperScript II First-Strand Synthesis using oligo (dT) primer System (Invitrogen). Aliquots of the reaction mixture were used for the subsequent PCR amplification. The primer sequences for KRAS amplification were 5′-GACTGAATATAAACTTGTGGTAGTTGGACCT-3′ (sense) and 5′-TCCTCTTGACCTGCTGTGTCG-3′ (antisense). The sense primer was designed to introduce a base substitution that created a BstNI recognition site for the wild-type codon 12 (GGT) but not for the codon 12 with the KRAS mutation. PCR conditions were as follows: a reaction volume of 40 μl for 15 min at 95°C for initial denaturation, followed by 25 cycles of 30 sec at 95°C, 60 sec at 55 °C, 30 sec at 72°C, and a final extension at 72 °C for 10 min. PCR products (15ml) were digested with 30 units of BstNI (New England Biolabs, Beverly, MA) at 60 °C for 3 h, and were visualized on 3% agarose gels stained with ethidium bromide.

KRAS copy number analysis

KRAS copy number was analyzed as described (22). Quantitative real-time PCR (qPCR) was performed using the Chromo4 PCR System (Bio-Rad Laboratories, Hercules, CA). We evaluated KRAS copy number in each sample by comparing the KRAS locus to the reference LINE-1, a repetitive element for which copy numbers per haploid genome are similar among all of the human normal and neoplastic cells (23).

In vivo xenograft growth

In vivo tumor growth was examined by injecting cells into nu/nu nude mice. 5 × 106 cells infected with pRS-KRAS-C12 or pRS control vector were injected into nude mice subcutaneously. Five mice were injected for each treatment. Tumor volume was measured for 70 days and 21 days for H358 cells and H1299 cells, respectively.

RNA quality and microarray analysis

The quality of total RNA was analyzed by formaldehyde gel and/or by capillary electrophoresis on the Experion System (Bio-Rad). Total RNA was labeled and amplified by our genomics core facility according to manufacturer’s instructions (24). cRNA was re-analyzed after labeling to ensure optimal amplification. cRNA was hybridized to U133 Plus 2.0 (~47000 transcripts) (Affymetrix, Santa Clara, CA), and scanned by our microarray core facility (25). Array analysis was performed as described (26). Briefly, after scanning, arrays were checked for quality using GCOS and then normalized using RMA. After normalization all the data were compiled using R and BRB ArrayTools (developed by Dr. Richard Simon and BRB-ArrayTools Development Team). Class comparisons were made using BRB ArrayTools. Bioinformatics were performed using NIH-DAVID (27) and BRB ArrayTools.

Quantitative real-time RT-PCR

The expression of the DUSP6 and NT5E genes was examined by quantitative real-time RT-PCR as described (28). Primers and probes for each gene were purchased from Applied Biosystems. For the quantitative analysis, the TBP gene was used as an internal reference gene to normalize input cDNA. PCR was performed in a reaction volume of 20 μl, including 2 μl cDNA using the Gene Amp 7700 Sequence Detection System and software (Applied Biosystems). The comparative Ct method was used to compute relative expression values.

Statistical analysis

Statistical analyses were performed using GraphPad Prism version 5.0 software program for Windows (GraphPad Software, San Diego, CA). Correlations between groups were analyzed by Pearson’s correlation coefficient. Differences between groups were analyzed by ANOVA with Bonferroni post-hoc test. Differences in tumor growth in nude mice were analyzed by repeated measures ANOVA. P<0.05 was considered to be significant.


Stable knockdown of oncogenic KRAS by retrovirus-mediated shRNA in NSCLC cell lines

We used four NSCLC cell lines H23, H1792, H358, and H441, which harbor heterozygous KRAS mutations at codon 12, and the H1299 NSCLC cell line with wild-type KRAS (but mutant NRAS) as a control (Table 1, Supplementary Fig. 1). We used the H1299 cell line, which has an NRAS mutation at codon 61 (15) as a control for the phenotypic effects of mutant KRAS knockdown. The KRAS mutation-positive NSCLC lines exhibited different levels of KRAS protein expression (Fig. 1A), which were positively correlated with KRAS copy number in the NSCLC lines (Fig. 1B).

Fig. 1Fig. 1
(A) Immunoblot of KRAS protein expression in H23, H1792, H358, H441, H1299 NSCLC cell lines. Thirty μg of whole cell lysate was loaded per lane and Western Blot was performed. (B) Positive correlation between KRAS copy number and the protein expression ...

To examine the effects of loss of oncogenic KRAS on NSCLC, we used a retroviral shRNA strategy to obtain stable and specific knockdown of mutant KRAS by pRS-KRAS-C12 targeting the KRAS G12C mutation and pRS-KRAS-V12 targeting the KRAS G12V mutation. This system has the advantage that the shRNA vectors can silence expression of mutant KRAS but retain expression of wild-type KRAS, which is indispensable for viability of normal cells (8, 9), in the four NSCLC cell lines with heterozygous KRAS mutation. H23, H1792 and H358 cells were infected with the pRS-KRAS-C12 vector and H441 cells were infected with the pRS-KRAS-V12 vector, and selected with puromycin to generate mutant KRAS-knockdown derivatives of these cell lines. H1299 cells with wild-type KRAS were infected with pRS-KRAS-C12 or pRS-KRAS-V12 to generate experimental controls. All cell lines were also infected with the pRS non-targeting control vector to generate pRS-infected control cells. In the KRAS mutant cell lines, KRAS protein expression was reduced in pRS-KRAS-infected cells compared with parental or pRS control vector-infected cells, whereas the infection with pRS, pRS-KRAS-C12 or pRS-KRAS-V12 did not affect KRAS protein expression in H1299 cells with wild-type KRAS (Fig. 1C). Thus, using the mutant-specific shRNA vector systems, we were able to isolate clones with stable loss of oncogenic KRAS expression in these NSCLC cell lines.

Since there are no commercially available mutant specific KRAS antibodies, we used PCR-RFLP analysis in order to verify that our RNAi strategy was specific for only mutant KRAS. In KRAS mutant NSCLC cells (H23, H1792, H358 and H441), the mutant 186-bp band was detected with higher intensity compared to the wild-type 156-bp cDNA fragment (Fig. 1D), demonstrating that cells with KRAS mutations express more mutant KRAS transcripts than wild-type, as previously reported by us (12) and others (29). In the KRAS-disrupted clones of H23, H1792, H358 and H441 cells, the intensity of the uncut bands was reduced compared to those in parental cells or in the control vector-infected cells (Fig. 1D). Thus, in our hands, retroviral-mediated RNAi substantially depleted expression of mutant KRAS mRNA but not wild-type KRAS mRNA (19).

shRNA-mediated knockdown of mutant KRAS suppresses in vitro and in vivo growth of NSCLC cell lines

Next we assessed the effect of shRNA-mediated stable knockdown of mutant KRAS expression on cell proliferation by the MTT assay. shRNA-mediated knockdown of mutant KRAS inhibited cell proliferation in all of the KRAS mutant NSCLC cell lines but not in H1299, whereas cell proliferation was not affected by the control vector in all cell lines (Supplementary Fig. 2A). The effect of KRAS knockdown on anchorage-independent growth was also assessed by soft-agar colony formation assay. KRAS knockdown led to a marked decrease (≥75%) in colony numbers in the KRAS mutant NSCLC cell lines but not in H1299 compared to the colony numbers of the parental cells (Supplementary Fig. 2B). Thus, shRNA-mediated knockdown of mutant KRAS significantly inhibited in vitro cell growth of KRAS mutation-positive NSCLC cells.

To further evaluate the effects of reduced mutant KRAS expression on the tumorigenic phenotype, and in particular its contribution to in vivo tumor growth, we used the H358 cell line since this cell line consistently produces tumors in nude mice (30) and the effects of KRAS knockdown on in vitro growth inhibition was most prominent among four KRAS mutant NSCLC lines (Supplementary Fig. 2A, 2B), indicating that H358 cells appears to be dependent on oncogenic KRAS signaling. There was a significant difference in in vivo tumor growth between H358 cells with pRS-KRAS-C12 and those with the control vectors (Supplementary Fig. 2C). H1299 cells infected with pRS-KRAS-C12or the control vector were also injected into mice to verifywhether there were any off-target effects of the pRS-KRAS-C12 infection that might inadvertently suppress tumor growth. In fact, the pRS-KRAS-C12 positive H1299 cells grew better than the vector controls, although this difference was not significant (Supplementary Fig. 2C). These results demonstrate that in vivo tumor growth was partially inhibited by shRNA-mediated knockdown of mutant KRAS expression in KRAS mutant NSCLC cells.

Effects of mutant KRAS knockdown on expression profiles in NSCLC cell lines

KRAS sits at the hub of multiple signaling cascades and activating mutations of KRAS can lead to diverse aberrant signal transduction events. As a first step to elucidate how oncogenic KRAS affects downstream pathways and to discover what might account for the residual tumorigenicity observed after KRAS knockdown, we performed microarray expression profiling (Affymetrix HG-U133-Plus2 array) of NSCLCs with and without oncogenic KRAS, and compared the expression profiles. As a control, we profiled H1299 cells (wild-type KRAS) after infection with the two vectors. Importantly, there were almost no differences in gene expression in H1299 with either vector (Supplementary Fig 3A). Within the mutant KRAS NSCLC cell lines, microarray analysis revealed considerable heterogeneity in genes whose expression changed more than two-fold with knockdown of oncogenic KRAS (Supplementary Fig. 3B). Thus, while knockdown of oncogenic KRAS has a clear and consistent effect on cell growth in NSCLC, there is significant diversity in its effect at the level of gene transcription.

Recently, Sweet-Cordero et al. attempted to identify an oncogenic KRAS-specific gene expression signature in NSCLC using cell lines and primary tumors (11). However, due to the heterogeneity of the KRAS expression phenotype in NSCLC, they employed a KRAS mouse model to identify a robust signature. In our study, we used an isogenic series of HBECs (HBEC3, HBEC3-K, HBEC3-p53 and HBEC3-p53K) that contain various oncogenic changes including forced expression of mutant KRAS-V12 (K) and shRNA dependent knockdown of p53 (p53) for similar purposes (Supplementary Fig. 4A) (12).

When NSCLC cell lines and HBECs were grouped according to whether they expressed mutant KRAS (H23-vector, H1792-vector, H358-vector, H441-vector, HBEC3-K and HBEC3-p53K) or not (H23-KRAS knockdown [KD], H1792-KRAS [KD], H358-KRAS [KD] H441-KRAS [KD], HBEC3, and HBEC3-p53) (Table 1), we identified 53 unique (several of which were poorly annotated) genes that distinguished these groups in a statistically significant manner (Supplementary Fig. 4B; Supplementary Table 1). Among these genes, two genes, ecto-5′-nucleotidase (NT5E, also known as CD73) and dual specificity phosphatase6 (DUSP6, also known as MKP-3), were identified as significantly down-regulated genes. While relationship between Ras and NT5E remains unknown, recent studies by us and others have elucidated that DUSP6 is up-regulated through ERK- MAPK signaling pathway upon oncogenic activation of EGFR (12, 31) or Ras (32), and controls the pathway as a negative feedback mediator (31). The reproducibility of transcriptional regulation of DUSP6 and NT5E was validated bytransient knockdown of mutant KRAS by synthetic siRNAs. siRNAs targeting against mutant KRAS were transfected into H23, H1792, H358 and H441 cells and expression levels of NT5E and DUSP6 were examined at three days post-transfection by quantitative RT-PCR analysis. Log ratios of mRNA expression changes by siRNA mediated KRAS knockdown were significantly correlated with those by the retroviral shRNA mediated-knockdown in the four NSCLC lines (Supplementary Fig. 4C), confirming that the expression of DUSP6 or NT5E expression was down-regulated by mutant KRAS knockdown.

Functional annotation analysis of the 53 genes revealed that many of these genes fell into the MAPK cascade as both positive and negative regulators of cell proliferation and survival (Supplementary Table 2). Importantly, we found that three dual specificity phosphatase activities (DUSP-4, -5, and -6) were down-regulated, and TGFβ an important activating ligand for SMAD/p38 signaling was up-regulated by oncogenic KRAS knockdown (Supplementary Table 3). These data suggest that the cells responded to loss of oncogenic KRAS signaling by reducing levels of negative feedback MAPK regulators and up-regulating autocrine signals to establish compensatory pro-growth stimuli through alternative, but related signal transduction cascades. In addition, a comparison of the gene expression changes between HBEC cells with ectopic oncogenic KRAS and the NSCLC knockdown experiments suggest that the early effectors of mutant-RAS dependent signaling are similar between cell lines (underscored by the fact that nearly all members of each group either over-expressed on under expressed each of the 53 genes).

Effects of mutant KRAS knockdown on phosphoproteins in EGFR and RAS pathways in KRAS mutant NSCLC cell lines

The principal finding of our array analysis was that genes involved in MAPK regulation were transcriptionally altered by oncogenic KRAS. While this is not surprising, we decided to further explore these changes at the protein level since many components of this pathway regulate protein phosphorylation. Before knockdown, it should be noted that the levels of KRAS protein expression varied among the KRAS mutation-positive NSCLC cell lines (Fig. 1A). The effects of KRAS knockdown on phosphorylation levels of MEK, ERK, AKT and EGFR protein are summarized (Fig. 2A, 2B). Consistent with our array analysis, KRAS knockdown reduced the levels of pMEK and pERK in all cell lines. Conversely, pEGFR levels were up-regulated in all cell lines after KRAS knockdown. Of note, the effects of KRAS knockdown on pAKT levels were different among these four lines. We also examined the effects of KRAS knockdown on the levels of phosphorylated STAT3 since it is an important downstream mediator of the EGFR signaling. STAT3 is activated by tyrosine phosphorylation, acting as a transcription factor to regulate gene expression (33), and the phosphorylation is mediated by oncogenic activation of EGFR in NSCLC cells (34). We found that pSTAT3 levels were up-regulated by KRAS knockdown in all cell lines, which appears to be a result from EGFR phosphorylation induced by KRAS knockdown. These results indicate that oncogenic KRAS positively regulates the MEK-ERK activation but negatively regulates activation of STAT3 and EGFR. On the other hand, AKT activation is differentially regulated by oncogenic KRAS among NSCLC cells carrying KRAS mutations. Our data, along with those recently reported by Singh et al., reflect the complexities of oncogenic KRAS-induced regulation of downstream pathways and led us to consider the possibility that oncogenic KRAS by itself does not explain all of the activity driving oncogenesis through the MAPK pathway in established tumors (10).

Fig. 2Fig. 2
(A) The effects of KRAS knockdown on the phosphorylation of MEK, ERK, Akt, STAT3 and EGFR in the KRAS mutation-positive NSCLC cell lines. Thirty μg of whole cell lysate was loaded per lane for western blot analysis. Mock: pRS vector-infected cells, ...

The synergistic effect of p38 and EGFR inhibitors with KRAS knockdown on cell growth of NSCLC

To be able to tolerate the “stress” of oncogenic KRAS expression, lung epithelial cells likely require secondary changes to adapt to their altered signaling homeostasis. Based on our finding that different NSCLCs responded differently to knockdown of oncogenic KRAS in terms of mRNA expression profile and MAPK-related signal transduction, we hypothesized that each of these lines had acquired different secondary changes potentially in KRAS controlled signaling pathways and/or, through some feedback mechanism to compensate for loss of oncogenic KRAS. Based on these considerations and the western blot data, we considered whether NSCLCs in which oncogenic KRAS was targeted (in our case with stable knockdown) acquired sensitivity to other targeted therapies. To test this hypothesis, we examined the effect of inhibitors to MEK1/2 (U0126), p38 (p38 V), PI3K (LY294002), AKT 1/2 (Akt VIII), EGFR tyrosine kinase (gefitinib) and an anti-EGFR antibody (cetuximab) on cell growth of KRAS mutant NSCLC cells. The treatment of p38 V in combination with KRAS knockdown resulted in significant growth inhibition in H23 at ≥2.5 μM, in H1792 at ≥0.025 μM, and in H441 at 2.5 μM compared to control treatments of p38 in the parental cells without KRAS knockdown (Fig. 3A). Thus, KRAS knockdown enhanced sensitivity to the p38 inhibitor markedly in H1792 and less so in H23 and H441. The treatment of gefitinib with KRAS knockdown resulted in significant growth inhibition in H23 at 10 μM, in H1792 at 0.1 to 10 μM, and in H358 at 0.01 to 1 μM compared to the treatment of gefitinib without KRAS knockdown (Fig. 3B). Also, the treatment of cetuximab with KRAS knockdown resulted in significant growth inhibition in H1792 at ≥0.5 μg/ml (Fig. 3C). Thus, KRAS knockdown significantly enhanced sensitivity to gefitinib in H23, H1792 and H358 and cetuximab in H1792, although the effects of sensitization were modest in some cases. In contrast, KRAS knockdown did not significantly affect the sensitivity to U1026, LY294002 and Akt VIII in these NSCLC lines (data not shown). These results suggest that inhibitors of EGFR and p38 could be effective for treatment of KRAS mutation-positive NSCLC in combination with KRAS knockdown.

Fig. 3Fig. 3Fig. 3
The effects of (A) p38 V, (B) gefitinib, and (C) cetuximab on colony formation in H23, H1792, H358 and H441 cells with KRAS knockdown (red line) or without KRAS knockdown (blue line) at various concentrations. Number of colonies in untreated cells was ...


In this report, we used shRNA methodology to stably silence the expression of oncogenic KRAS in four NSCLC cell lines. We found that stable knockdown of oncogenic KRAS led to reduced proliferation rates and anchorage independent growth in vitro, as well as decreased in vivo tumorigenicity. However, we found that knockdown of oncogenic KRAS did not lead to apoptosis or complete loss of tumorigenicity, which is in contrast to previously reported results in the pancreatic cancer cell line CAPAN-1 carrying homozygous KRAS mutations. (19). One possible explanation on differences between their study and ours is that NSCLC cell lines used in our study carry both mutant and wild-type KRAS alleles. Hence, the mutant-specific KRAS shRNA vectors will not affect wild-type KRAS expression, which appears to be indispensable for cell viability under normal conditions. While it is possible that the lack of complete loss of cell viability by mutant KRAS knockdown would not be significant due to the remaining wild-type KRAS, it is also possible that retained expression of wild-type KRAS in the oncogenic KRAS-disrupted clone permits survival of the NSCLC cells through normal regulation of signal transduction. Of note, Brummelkamp et al. also reported that oncogenic KRAS knockdown had no significant effect on the ability of CAPAN-1 cells to proliferate adherently. Thus, it is also possible that knocking down oncogenic KRAS does not have striking effects on cell proliferation in contrast to the effects on anchorage-independent growth.

Analysis of the transcriptional changes associated with knockdown of oncogenic KRAS suggest that the cells responded by down-regulating negative regulators of MAPK, in particular those of ERK 1/2 and JNK, as well as by up-regulating Ras related GTPases and ligands for EGFR and SMAD. The biological relevance of down-regulating phosphatase activity and up-regulating autocrine ligands demonstrates the adaptabilityof NSCLCs of finding alternative signal transduction cascades – through MAPK-related pathways – to re-establish oncogenic signaling. Most importantly, we show that mutant KRAS knockdown in these NSCLCs led to sensitization to molecularly targeted therapeutics: inhibitors of EGFR and p38, especially for H1792 cells, although the effect of KRAS knockdown was modest and required high concentrations of these inhibitors in some cases. To our knowledge, this is the first report showing that loss of oncogenic KRAS activity leads to up-regulation of phospho-EGFR and sensitizes formerly resistant NSCLCs to inhibitors of EGFR.

In NSCLC, mutations in the EGFR and KRAS genes are major molecular abnormalities that appear to be mutually exclusive (5, 35). Thus, it will be essential to develop therapeutic modalities for NSCLC carrying either EGFR or KRAS mutations. While EGFR-TKI therapy is effective in EGFR mutation-positive patients (3639), they provide no benefit for KRAS mutant lung cancer (40). Currently, there are no effective treatment strategies for KRAS mutant NSCLC. In our study, shRNA-mediated knockdown of mutant KRAS in NSCLCs led to up-regulation of pEGFR and significantly increased sensitivity to gefitinib. Our findings may explain, at least in part, why NSCLCs with KRAS mutations are resistant to EGFR-TKIs: oncogenic activation of KRAS obviates the need for signaling via ligand-dependent activation and phosphorylation of EGFR. However, our data also demonstrate that NSCLCs, which probably evolved through a KRAS–dependent process, retain the capacity to up-regulate other signals, which feed through MAPK. This conclusion is supported by a recent study showing that exogenous expression of mutant KRAS in human bronchial epithelial cells reduced the phosphotyrosine immunoprecipitates of EGFR in the phosphoproteomic studies (41).

The present study indicates, as expected, that activation of the MAPK pathway is a critical event in the initiation of KRAS mutant NSCLC while additional molecular abnormalities resulting in deregulation of other signaling pathways such as the PI3K-Akt pathway could play a role in the maintenance of the oncogenic phenotype. These findings suggest that multiple molecular targets exist for combination therapy in NSCLC patients carrying KRAS mutations. Our results show that inhibiting oncogenic KRAS activity alone is not sufficient to completely inhibit tumor growth of KRAS mutant NSCLC. By applying the results from four NSCLC cell lines carrying additional distinct molecular abnormalities to a broader genetic analysis of KRAS mutation-positive patients, it might be possible to identify the critical molecules to target in combination with mutant KRAS knockdown to improve treatment for patients.

We found that mutant KRAS knockdown enhanced sensitivity to the p38 inhibitor p38V in NSCLC cells. This unexpected result might be due to autocrine up-regulation of TGFβ, which was significantly up-regulated after knockdown of oncogenic KRAS (Supplementary Table 3; Supplementary Fig. 4B), as well as crosstalk between ERK and p38. p38 is involved in many cellular processes including cell proliferation and apoptosis in response to stress stimuli and cytokines. Although the role of p38 in tumor development is not as well-defined as ERK, recent evidence suggests that the two pathways, ERK and p38, play opposing roles in tumor cells (42). It is possible that p38 signaling substitutes for deactivation of ERK signaling caused by loss of mutant KRAS; thus dual inhibition of KRAS and p38 activities could be effective for the treatment of KRAS mutation-positive NSCLC.

Recent advances in molecular biology have made considerable progress in molecularly targeted cancer therapy. A number of molecularly targeted drugs are clinically available or being tested for clinical use in lung cancer based on the growing knowledge of the molecular abnormalities relevant to lung carcinogenesis (3). Considering that KRAS mutations are observed >25% of lung adenocarcinomas (6), therapeutic strategies targeting oncogenic KRAS and its signaling pathway would be very valuable in treating NSCLCs with activating KRAS mutations. The present study suggests that targeting oncogenic KRAS alone does not seem to be ‘Achilles heel’ of mutant KRAS NSCLC. Recently, several studies using genome-wide RNAi screen with human cancer cell lines have identified potential molecular targets that are synthetically lethal with oncogenic KRAS, including PLK1 (43), TBK1 (44) and STK33 (45). While these molecules are potential targets in NSCLC cells with KRAS mutations, the synthetic lethal interactions are influenced by their oncogenic KRAS dependency; thus, the extent of the “synthetic lethality” appears to be cell type-dependent. For instance, Singh et al. found that only some KRAS mutant cells retained a requirement for this oncogene (10), which led the authors to stratify KRAS mutant tumors into two broad subgroups: dependent or independent of sustained mutant KRAS signaling. Added to this, other studies reported dependent but not independent KRAS mutant tumors are sensitive to suppression of STK33 (45) and TBK1 (44). Our findings, together with other recent reports, do point to the fact that mutant KRAS oncogenic signaling is likely more complex than previously perceived. Tumor cells might be more robust than we thought in that they can have versatile mechanisms to overcome “targeted inhibition” and that even knocking down mutant KRAS itself does not seem to be an effective Achilles heel. The observance of these diverse subsets of KRAS mutant NSCLCs – with distinct molecular profiles, signaling pathway activation, drug sensitivity, epithelial-to-mesenchymal status, and molecular aberrations – supports what we envision as a more effective treatment strategy for KRAS mutant NSCLCs: the combined targeting of oncogenic KRAS in addition with other therapeutic agents specific to the molecular profile of the tumor. Thus, the treatment of KRAS mutation-positive NSCLC patients will require a priori knowledge of other molecular abnormalities in the tumors in order to achieve more effective therapies for NSCLC. Further studies focusing on the cells’ compensatory mechanisms could also be quite fruitful.

Supplementary Material



Funding for this work came from SPORE (P50CA70907); DOD PROSPECT, Texas Higher Education Coordinating Board Advanced Technology Program (grant #01001901392003); Gillson Longenbaugh Foundation; NASA Specialized Center of Research (grant #NNJ05HD36G). NS was supported by Grant 18790532 (to N.S.) from the Ministry of Education, Culture, Sports, Science and Technology, Japan. DSS was supported by IASLC post-doctoral fellowship. We thank Drs. Yoshio Tomizawa and Tamotsu Ishizuka of the Department of Medicine and Molecular Science, Gunma University Graduate School of Medicine, Gunma, Japan for technical support.


Conflict of interest statement

All authors have no financial or personal relationships with other people or organizations that could inappropriately influence our work.


1. Sekido Y, Fong KM, Minna JD. Molecular genetics of lung cancer. Annu Rev Med. 2003;54:73–87. [PubMed]
2. Sato M, Shames DS, Gazdar AF, Minna JD. A translational view of the molecular pathogenesis of lung cancer. J Thorac Oncol. 2007 Apr;2(4):327–43. [PubMed]
3. Sun S, Schiller JH, Spinola M, Minna JD. New molecularly targeted therapies for lung cancer. J Clin Invest. 2007 Oct;117(10):2740–50. [PMC free article] [PubMed]
4. Downward J. Targeting RAS signalling pathways in cancer therapy. Nat Rev Cancer. 2003 Jan;3(1):11–22. [PubMed]
5. Suda K, Tomizawa K, Mitsudomi T. Biological and clinical significance of KRAS mutations in lung cancer: an oncogenic driver that contrasts withEGFR mutation. Cancer Metastasis Rev. 2010 Mar;29(1):49–60. [PubMed]
6. Buttitta F, Barassi F, Fresu G, et al. Mutational analysis of the HER2 gene in lung tumors from Caucasian patients: mutations are mainly present in adenocarcinomas with bronchioloalveolar features. Int J Cancer. 2006 Dec 1;119(11):2586–91. [PubMed]
7. Mascaux C, Iannino N, Martin B, et al. The role of RAS oncogene in survival of patients with lung cancer: a systematic review of the literature with meta-analysis. Br J Cancer. 2005 Jan 17;92(1):131–9. [PMC free article] [PubMed]
8. Johnson L, Greenbaum D, Cichowski K, et al. K-ras is an essential gene in the mouse with partial functional overlap with N-ras. Genes Dev. 1997 Oct 1;11(19):2468–81. [PMC free article] [PubMed]
9. Koera K, Nakamura K, Nakao K, et al. K-ras is essential for the development of the mouse embryo. Oncogene. 1997 Sep 4;15(10):1151–9. [PubMed]
10. Singh A, Greninger P, Rhodes D, et al. A gene expression signature associated with “K-Ras addiction” reveals regulators of EMT and tumor cell survival. Cancer Cell. 2009 Jun 2;15(6):489–500. [PMC free article] [PubMed]
11. Sweet-Cordero A, Mukherjee S, Subramanian A, et al. An oncogenic KRAS2 expression signature identified by cross-species gene-expression analysis. Nat Genet. 2005 Jan;37(1):48–55. [PubMed]
12. Sato M, Vaughan MB, Girard L, et al. Multiple oncogenic changes (K-RAS(V12), p53 knockdown, mutant EGFRs, p16 bypass, telomerase) are not sufficient to confer a full malignant phenotype on human bronchial epithelial cells. Cancer Res. 2006 Feb 15;66(4):2116–28. [PubMed]
13. Forgacs E, Biesterveld EJ, Sekido Y, et al. Mutation analysis of the PTEN/MMAC1 gene in lung cancer. Oncogene. 1998 Sep 24;17(12):1557–65. [PubMed]
14. Ma PC, Schaefer E, Christensen JG, Salgia R. A selective small molecule c-MET Inhibitor, PHA665752, cooperates with rapamycin. Clin Cancer Res. 2005 Mar 15;11(6):2312–9. [PubMed]
15. Phelps RM, Johnson BE, Ihde DC, et al. NCI-Navy Medical Oncology Branch cell line data base. J Cell Biochem Suppl. 1996;24:32–91. [PubMed]
16. Sanchez-Cespedes M, Parrella P, Esteller M, et al. Inactivation of LKB1/STK11 is a common event in adenocarcinomas of the lung. Cancer Res. 2002 Jul 1;62(13):3659–62. [PubMed]
17. Shigemitsu K, Sekido Y, Usami N, et al. Genetic alteration of the beta-catenin gene (CTNNB1) in human lung cancer and malignant mesothelioma and identification of a new 3p21.3 homozygous deletion. Oncogene. 2001 Jul 12;20(31):4249–57. [PubMed]
18. Sithanandam G, Smith GT, Masuda A, Takahashi T, Anderson LM, Fornwald LW. Cell cycle activation in lung adenocarcinoma cells by the ErbB3/phosphatidylinositol 3-kinase/Akt pathway. Carcinogenesis. 2003 Oct;24(10):1581–92. [PubMed]
19. Brummelkamp TR, Bernards R, Agami R. Stable suppression of tumorigenicity by virus-mediated RNA interference. Cancer Cell. 2002 Sep;2(3):243–7. [PubMed]
20. Sunaga N, Miyajima K, Suzuki M, et al. Different roles for caveolin-1 in the development of non-small cell lung cancer versus small cell lung cancer. Cancer Res. 2004 Jun 15;64(12):4277–85. [PubMed]
21. Nishikawa T, Maemura K, Hirata I, et al. A simple method of detecting K-ras point mutations in stool samples for colorectal cancer screening using one-step polymerase chain reaction/restriction fragment length polymorphism analysis. Clin Chim Acta. 2002 Apr;318(1–2):107–12. [PubMed]
22. Soh J, Okumura N, Lockwood WW, et al. Oncogene mutations, copy number gains and mutant allele specific imbalance (MASI) frequently occur together in tumor cells. PLoS One. 2009;4(10):e7464. [PMC free article] [PubMed]
23. Wang TL, Maierhofer C, Speicher MR, et al. Digital karyotyping. Proc Natl Acad Sci U S A. 2002 Dec 10;99(25):16156–61. [PMC free article] [PubMed]
24. Affymetrix – Home [Internet] Affymetrix, Inc; Available from: http://www.affymetrix.com/
25. UTSW Microarray Core [Internet] The Microarray Core Facility, University of Texas Southwestern Medical Center; Available from: http://microarray.swmed.edu/
26. Shames DS, Girard L, Gao B, et al. A genome-wide screen for promoter methylation in lung cancer identifies novel methylation markers for multiple malignancies. PLoS Med. 2006 Dec;3:e486. [PMC free article] [PubMed]
27. National Institute of Allergy and Infectious Diseases (NIAID), NIH; DAVID Functional Annotation Bioinformatics Microarray Analysis [Internet]: DAVID Bioinformatics Resources 6.7. Available from: http://david.abcc.ncifcrf.gov/
28. Suzuki M, Sunaga N, Shames DS, Toyooka S, Gazdar AF, Minna JD. RNA interference-mediated knockdown of DNA methyltransferase 1 leads to promoter demethylation and gene re-expression in human lung and breast cancer cells. Cancer Res. 2004 May 1;64(9):3137–43. [PubMed]
29. Uchiyama M, Usami N, Kondo M, et al. Loss of heterozygosity of chromosome 12p does not correlate with KRAS mutation in non-small cell lung cancer. Int J Cancer. 2003 Dec 20;107(6):962–9. [PubMed]
30. Fujiwara T, Grimm EA, Mukhopadhyay T, Zhang WW, Owen-Schaub LB, Roth JA. Induction of chemosensitivity in human lung cancer cells in vivo by adenovirus-mediated transfer of the wild-type p53 gene. Cancer Res. 1994 May 1;54(9):2287–91. [PubMed]
31. Zhang Z, Kobayashi S, Borczuk AC, et al. Dual specificity phosphatase 6 (DUSP6) is an ETS-regulated negative feedback mediator of oncogenic ERK signaling in lung cancer cells. Carcinogenesis. 2010 Apr;31(4):577–86. [PMC free article] [PubMed]
32. Bluthgen N, Legewie S, Kielbasa SM, et al. A systems biological approach suggests that transcriptional feedback regulation by dual-specificity phosphatase 6 shapes extracellular signal-related kinase activity in RAS-transformed fibroblasts. Febs J. 2009 Feb;276(4):1024–35. [PubMed]
33. Levy DE, Lee CK. What does Stat3 do? J Clin Invest. 2002 May;109(9):1143–8. [PMC free article] [PubMed]
34. Alvarez JV, Greulich H, Sellers WR, Meyerson M, Frank DA. Signal transducer and activator of transcription 3 is required for the oncogenic effects of non-small-cell lung cancer-associated mutations of the epidermal growth factor receptor. Cancer Res. 2006 Mar 15;66(6):3162–8. [PubMed]
35. Gazdar AF, Minna JD. Multifocal lung cancers--clonality vs field cancerization and does it matter? J Natl Cancer Inst. 2009 Apr 15;101(8):541–3. [PMC free article] [PubMed]
36. Mok TS, Wu YL, Thongprasert S, et al. Gefitinib or carboplatin-paclitaxel in pulmonary adenocarcinoma. N Engl J Med. 2009 Sep 3;361(10):947–57. [PubMed]
37. Morita S, Okamoto I, Kobayashi K, et al. Combined survival analysis of prospective clinical trials of gefitinib for non-small cell lung cancer with EGFR mutations. Clin Cancer Res. 2009 Jul 1;15(13):4493–8. [PubMed]
38. Maemondo M, Inoue A, Kobayashi K, et al. Gefitinib or chemotherapy for non-small-cell lung cancer with mutated EGFR. N Engl J Med. 2010 Jun 24;362(25):2380–8. [PubMed]
39. Mitsudomi T, Morita S, Yatabe Y, et al. Gefitinib versus cisplatin plus docetaxel in patients with non-small-cell lung cancer harbouring mutations of the epidermal growth factor receptor (WJTOG3405): an open label, randomised phase 3 trial. Lancet Oncol. 2010 Feb;11(2):121–8. [PubMed]
40. Linardou H, Dahabreh IJ, Kanaloupiti D, et al. Assessment of somatic k-RAS mutations as a mechanism associated with resistance to EGFR-targeted agents: a systematic review and meta-analysis of studies in advanced non-small-cell lung cancer and metastatic colorectal cancer. Lancet Oncol. 2008 Oct;9(10):962–72. [PubMed]
41. Guha U, Chaerkady R, Marimuthu A, et al. Comparisons of tyrosine phosphorylated proteins in cells expressing lung cancer-specific alleles of EGFR and KRAS. Proc Natl Acad Sci U S A. 2008 Sep 16;105(37):14112–7. [PMC free article] [PubMed]
42. Ranganathan AC, Adam AP, Aguirre-Ghiso JA. Opposing roles of mitogenic and stress signaling pathways in the induction of cancer dormancy. Cell Cycle. 2006 Aug;5(16):1799–807. [PMC free article] [PubMed]
43. Luo J, Emanuele MJ, Li D, et al. A genome-wide RNAi screen identifies multiple synthetic lethal interactions with the Ras oncogene. Cell. 2009 May 29;137(5):835–48. [PMC free article] [PubMed]
44. Barbie DA, Tamayo P, Boehm JS, et al. Systematic RNA interference reveals that oncogenic KRAS-driven cancers require TBK1. Nature. 2009 Nov 5;462(7269):108–12. [PMC free article] [PubMed]
45. Scholl C, Frohling S, Dunn IF, et al. Synthetic lethal interaction between oncogenic KRAS dependency and STK33 suppression in human cancer cells. Cell. 2009 May 29;137(5):821–34. [PubMed]
PubReader format: click here to try


Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...


Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...