NF-κB is required for H-ras oncogene induced abnormal cell proliferation and tumorigenesis

Article metrics


Oncogenic mutations in ras lead to constitutive activation of downstream signaling pathways that modulate the activities of transcription factors. In turn, these factors control the expression of a subset of genes responsible for neoplastic cell transformation. Recent studies suggest that transcription factor NF-κB contributes to cell transformation by inhibiting the cell death signal activated by oncogenic Ras. In this study, inhibition of NF-κB activity by forced expression of a super-repressor form of IκBα, the major inhibitor of NF-κB, markedly decreased the growth rate, saturation density and tumorigenicity of oncogenic H-Ras transformed rat embryo fibroblasts. Such clonally isolated cells overexpressing IκBα super-repressor not only were viable but also exhibited no sign of spontaneous apoptosis. Inhibition of NF-κB in these cells was functionally demonstrated by both the loss of cytokine induced DNA binding activity and a profoundly increased sensitivity to cell death in response to TNF-α treatment. In contrast, inhibition of NF-κB activity in non-transformed fibroblasts had minimal effect on growth, but rendered the cells resistant to a subsequent transformation by H-ras oncogene. Similar results were also obtained with rat intestinal epithelial cells harboring an inducible ras oncogene. Taken together, these findings suggest that NF-κB activity is essential for abnormal cell proliferation and tumorigenicity activated by the ras oncogene and highlight an alternative functional role for NF-κB in oncogenic Ras-mediated cell transformation that is distinct from its anti-apoptotic activity.


The Ras proteins are important signaling molecules that regulate various cellular processes including growth, differentiation, survival, and senescence (Downward, 1998; Kauffman-Zeh et al., 1997; Khwaja et al., 1997; Lin et al., 1998; Lowy et al., 1993; Zhu et al., 1998). In response to external stimuli such as growth factors, Ras activates multiple downstream effectors that initiate signaling cascades via activation of protein kinases (Lowy et al., 1993; White et al., 1995). Oncogenic mutations in ras have been linked to many types of human cancer (Kiaris and Spandidos, 1995). These mutations lock Ras proteins into a permanently activated state leading to constitutive stimulation of downstream signaling pathways. In turn, these pathways modulate the activities of transcription factors that regulate numerous genes responsible for cell transformation and tumorigenesis (Barbacid, 1987; Treisman, 1994). Consistent with this notion, members of the AP-1 and Ets transcription factor families have been shown to be required for cell transformation and tumorigenicity mediated by oncogenic Ras protein (Johnson et al., 1996; Langer et al., 1992.

NF-κB was originally identified as one of the nuclear factors that bind to immunoglobin κ light chain enhancer sequence (Sen and Baltimore, 1986). Subsequent cloning of the genes encoding these factors revealed a family of transcription factors with a well-conserved Rel homology domain. This domain is important for DNA binding, dimerization, and nuclear localization (Baldwin, 1996; Verma et al., 1995). In most cell types, NF-κB is stored in the cytoplasm by specific inhibitors, IκBs, which bind to NF-κB and mask its nuclear localization sequence (Baldwin, 1996; Verma et al., 1995). Upon stimulation by cytokines such as TNF-α and IL-1, IκBs become rapidly phosphorylated which triggers ubiquitination-mediated proteolysis of the inhibitors, resulting in nuclear translocation and activation of NF-κB (Baldwin, 1996; Verma et al., 1995; Baeurle and Baltimore, 1996).

Recently, it was reported that ectopic expression of a super-repressor form of IκBα attenuates Ras-induced focus formation, suggesting that NF-κB activity contributes to Ras-induced cellular transformation (Finco et al., 1997). In addition, it has been shown that NF-κB activity is required for the protection of oncogenic Ras-induced cell death, providing a potential mechanism which explains how NF-κB contributes to Ras-induced cellular transformation (Mayo et al., 1997).

Here, using a retroviral gene transfer system, we evaluated the functional relationship between H-Ras and NF-κB activity during cell transformation. In reciprocal experiments where either oncogenic H-ras or a super-repressor of NF-κB was first introduced into the rat embryo fibroblasts, we found that NF-κB was essential for abnormal cell proliferation and tumorigenesis activated by mutant Ras. Moreover, blocking NF-κB activity in Ras transformed cells failed to induce cell death, but rather rendered the cell with a retarded growth rate. These findings were further substantiated in a set of experiments with rat intestinal epithelial cells in which Ras promotes cell survival. Inactivation of NF-κB in these cells led to growth suppression rather than cell death upon H-Ras induction. These results provide an alternative functional role for NF-κB in Ras-mediated cell transformation that is distinct from its anti-apoptotic activity.


Differential effect of NF-κB inhibition on normal and H-Ras transformed cells

To determine whether NF-κB activation is necessary for oncogenic H-Ras-mediated cell transformation, the N-terminal FLAG-tagged wild-type or a super-repressor form of human IκBα, ΔN-IκBα, was introduced into Rat-1 and its H-Ras transformed derivative, Rat-1(Ras) cell lines (Liang et al., 1994), by a retroviral gene delivery system (Johnson et al., 1996). ΔN-IκBα lacks the N-terminal 36 amino acids encompassing two serine residues (S32, 36) required for signal-induced phosphorylation and degradation of this inhibitor (Brockman et al., 1995). Five days following viral infection and drug selection, pools of infected cells were counted. In comparison with pBabe vector control, both wild-type and ΔN-IκBα expressing viruses led to a significant reduction in cell number, 44 and 70%, respectively, when infected into transformed Rat-1(Ras) cells (Figure 1a). In contrast, all three viral stocks gave similar numbers of infected cells for the parental Rat-1 cells. Expression of human IκBα and ΔN-IκBα protein in pools of infected cells was confirmed by Western blot and immunoprecipitation with the agarose-conjugated anti-FLAG antibody followed by Western blot analysis using C-terminal specific IκBα antibody (Figure 1b).

Figure 1

Differential effect of NF-κB inhibition in normal and oncogenic H-Ras transformed cells. (a) Control empty virus, pBabe(puro), and a recombinant retrovirus directing expression of the wild-type human IκBα or ΔN-IκBα were infected into the Rat-1 and Rat-1(Ras) cells. Three days after viral infection and selection for puromycin, the surviving cells were counted. Each data point represented an average of duplicate samples. (b) Expression of the recombinant IκBα and ΔN-IκBα. Western blot analysis of cellular proteins from the Rat-1 and Rat-1(Ras) cells infected with the indicated viruses (upper panel). (Con); non-infected, (Vec); empty virus, (WT); wild-type IκBα, (DN); ΔN-IκBα. Cellular proteins from Rat-1(Ras) cells infected with empty, wild-type or ΔN-IκBα viruses were immunoprecipitated with anti-FLAG antibodies followed by Western blot analysis using an antibody specific for C-terminus of IκBα (lower panel)

Inactivation of NF-κB leads to growth inhibition and reduced tumorigenic potential of H-Ras transformed Rat-1 fibroblasts

The differential effect of ΔN-IκBα expression in parental and H-Ras transformed Rat-1 cells suggests that NF-κB activity may be required for either the abnormal cell proliferation or survival of the transformed cells. To further differentiate between these two possibilities, we compared the growth rate and tumorigenicity of pooled ΔN-IκBα retrovirus infected Rat-1(Ras) cells with that of control cells infected with an empty virus. Surprisingly, no significant difference was observed between two pools of cells (Table 1 and data not shown). One speculation was that the apparent discrepancy between the viral infection (Figure 1a) and the growth property of infected cells grown up in culture might be due to the heterogeneity in the level of ΔN-IκBα expression in infected cells. If inhibtion of NF-κB activity is growth inhibitory or apoptotic to H-Ras transformed cells, one would expect that the infected cells with a low level of expression of ΔN-IκBα, thus incomplete inhibition of NF-κB, should have a significant growth advantage over those with a higher level of expression. The former would be preferentially selected over time in culture, resulting in a population of cells no different from the control cells. To determine the validity of this hypothesis, it was necessary to isolate clonally purified cells expressing high level of ΔN-IκBα, thus more efficient blocking NF-κB activity. Such analysis from earlier pools of Rat-1(Ras) cells infected with ΔN-IκBα virus indeed allowed the isolation of two types of clonally purified cell lines, one with a low level of ΔN-IκBα expression, designated as RR/ΔN-IκBα-L, the other with a high level of ΔN-IκBα expression designated as RR/ΔN-IκBα-H (Figure 2a). The cell growth measurement revealed that RR/ΔN-IκBα-L was marginally different from the parental cell, approximately 10.8 h per doubling, whereas RR/ΔN-IκBα-H showed a much slower growth rate, approximately 17.2 h per doubling and nearly eightfold reduction in saturation density (Figure 2b). Compared to the parental and pooled ΔN-IκBα infected cells, RR/ΔN-IκBα-H cells manifested a marked reduction in tumorigenic potential in nude mice (Table 1). These results suggest that NF-κB activiy is required for the abnormal proliferation of H-Ras transformed cells, which contributes to tumorigenic potential of these cells. Gel mobility shift assay revealed that TNF-α-induced nuclear translocation of NF-κB was blocked in RR/ΔN-IκBα-H cells (Figure 3a). In addition, consistent with recent observations that NF-κB activity is required for TNF-α-induced cell death in various cell types (Antwerp et al., 1996; Beg and Baltimore, 1996; Liu et al., 1996; Wang et al., 1996), TNF-α treatment in RR/ΔN-IκBα-H cells led to a massive cell death within 18 h, whereas parental Rat-1(Ras) and RR/ΔN-IκBα-L cells were little affected (Figure 3b). The genomic DNA isolated from RR/ΔN-IκBα-H cells after TNF-α treatment also exhibited a characteristic DNA fragmentation and corresponding cytoplasmic extract was shown to be capable of activating pro-caspase-3 (Figure 3c). These results confirmed that NF-κB was functionally inactivated in RR/ΔN-IκBα-H cells.

Table 1 Inhibition of H-ras mediated tumorigenicity by ΔN-IκBα
Figure 2

Inactivation of NF-κB leads to growth inhibition and reduced tumorigenic potential of H-Ras transformed Rat-1 fibroblasts. (a) Northern blot analysis of ΔN-IκBα mRNA. Twenty μg of total RNAs of clonally isolated Rat-1(Ras) cells expressing ΔN-IκBα were analysed. (b) Overexpression of ΔN-IκBα led to a decreased growth rate and saturation density of the H-Ras transformed Rat-1 cells. The growth property of RR/ΔN-IκBα-H cells was compared with Rat-1(Ras), Rat-1(Ras)/pBabe, and RR/DN-IκBα-L cells

Figure 3

Functional inactivation of NF-κB in RR/ΔN-IκBα-H cells. (a) Representative electrophoretic mobility shift assay (EMSA) for NF-κB activity. EMSA was performed with nuclear extracts from Rat-1(Ras), Rat-1(Ras)/pBabe, RR/ΔN-IκBα-H1, and RR/ΔN-IκBα-L1 cells treated with (+) or without (−) TNF-α (20 ηg/ml) for 30 min. The NF-κB complex containing p65 subunit was indicated by the arrows. (b) TNF-α induced apoptosis in RR/ΔN-IκBα-H cells. Rat-1(Ras)/pBabe and RR/ΔN-IκBα-H1 were treated with or without rat TNF-α (50 ng/ml) and photographs were taken 18 h following the TNF-α treatment. (c) TNF-α induced apoptotic DNA fragmentation and pro-caspase-3 activation. Genomic DNAs and cytoplasmic extracts of Rat-1(Ras), Rat-1(Ras)/pBabe, RR/ΔN-IκBα-H1, and RR/ΔN-IκBα-L1 cells treated with or without TNF-α were isolated and analysed for DNA fragmentation (upper panel) and pro-caspase-3 activation (lower panel)

NF-κB inactivation leads to growth suppression in rat intestinal epithelial cells in which Ras promotes cell survival instead of apoptosis

To further determine if NF-κB activity is required for Ras-induced abnormal cell growth, we utilized rat intestinal epithelial (RIE) (Ko et al., 1995) cells containing an IPTG inducible oncogenic H-Ras. Induction of oncogenic H-Ras in RIE(iRas) cells led to a marked increase in NF-κB transcriptional activity with concomitant increase in cell growth rate over that of non-induced counterpart (Figure 4a,b). However, oncogenic H-Ras-induced abnormal cell growth was largely abolished in RIE(iRas)/ΔN-IκBα cells (Figure 4b). Expression levels of endogenous Ras and oncogenic H-Ras proteins were shown to be similar in both cells, ruling out the possibility that the resulting difference was caused by variations in level of H-Ras protein induction (Figure 4b). In contrast to a previous report (Mayo et al., 1997), induction of oncogenic H-Ras protein had little effect on cell viability of RIE(iRas)/ΔN-IκBα cells. Moreover, consistent with the findings that Ras promotes cell survival rather than cell death in epithelial cells (Cardone et al., 1998; Downward, 1998; Kauffman-Zeh et al., 1997; Khwaja et al., 1997; Romashkova and Makarov, 1999), H-Ras induction significantly prevented cell death of RIE(iRas)/ΔN-IκBα cells upon TNF-α treatment as determined by colonigenic assay (Figure 4c). These observations suggest that reduction in cell number of RIE(iRas)/ΔN-IκBα cells over that of control cells upon Ras induction is due to growth suppression rather than increased cell death.

Figure 4

NF-κB inactivation leads to growth suppression in rat intestinal epithelial cells in which Ras promotes cell survival instead of apoptosis. (a) Ras activation leads to an increased NF-κB transcriptional activity in RIE(iRas) cells. NF-κB-dependent luciferase activity was measured at indicated time points after IPTG induction. Values were normalized with β-galactosidase activity. The corresponding cellular extracts were also analysed for Ras protein expressed by Western blot (inset). F indicates farnesyl transferase inhibitor (25 μM). (b) Ras attenuates TNF-α-induced cell death in RIE(iRas)/ΔN-IκBα cells. Colonigenic assay was performed (see Materials and methods for detail) with RIE(iRas)/ΔN-IκBα cells in the presence or absence of IPTG. Data represent one of the two independent experiments in quadruplicate. FTI indicates farnesyl transferase inhibitor (25 μm). (c) The rat intestinal epithelial cells containing IPTG inducible oncogenic H-Ras, RIE(iRas), were infected with an empty virus or a recombinant virus expressing ΔN-IκBα. Three days after puromycin selection, the same numbers of RIE(iRas)/pBabe and RIE(iRas)/ΔN-IκBα cells were plated and cell numbers were counted at indicated time points in the presence or absence of IPTG (4 mM) (upper panel). Expression of oncogenic Ras proteins in RIE(iRas)/pBabe and RIE(iRas)/ΔN-IκBα cells was determined by Western blot analysis after 5 days following IPTG induction (lower panel)

NF-κB is not required for normal cell growth, but is essential for cell transformation by oncogenic Ras

In contrast to its Ras transformed counterpart, non-transformed Rat-1 fibroblasts were little affected by ectopic expression of wild-type or ΔN-IκBα (Figure 1a). To confirm this finding, we isolated Rat-1 clones expressing high level of ΔN-IκBα, designated as Rat-1/ΔN-IκBα-H cells. These cells were capable of blocking the nuclear translocation of NF-κB upon TNF-α treatment but exhibited little difference in growth rate compared to parental Rat-1 and Rat-1/pBabe cells (Figure 5a,b). This result suggests that NF-κB activity is not required for normal cell proliferation, which is consistent with the observation that NF-κB is inactivated in most cell types (Baldwin, 1996; Verma et al., 1995). Thus, assuming that NF-κB activity is critical for Ras-mediated cell transformation, non-transformed cells with an impaired NF-κB activity should be resistant to subsequent cell transformation by oncogenic Ras. The availability of such cell lines, Rat-1/ΔN-IκBα-H, enabled us to perform such a reciprocal experiment, which can directly test this hypothesis. To this end, Rat-1, Rat-1/pBabe, and Rat-1/ΔN-IκBα-H cells were infected with a recombinant retrovirus expressing an oncogenic form of H-Ras(V12), pBabe-H-Ras(hyg). After viral infection and selection for hygromycin-resistance, the infected cells were counted and analysed for H-Ras protein expression. Infection of H-Ras(V12) expressing virus into Rat-1/ΔN-IκBα-H cells led to approximately eightfold reduction in cell numbers compared to that of Rat-1 and Rat-1/pBabe cells, while the vector control virus resulted in similar numbers of cells (Figure 6a,b). In addition, Rat-1/ΔN-IκBα-H cells infected with H-Ras(V12) expressing virus formed much smaller and less transformed colonies compared to Rat-1/pBabe cells which after the infection formed colonies with multilayers of cells, resulting in bigger and darker stains with Giemsa (Figure 6a). Western blot analysis showed comparable level of H-Ras(V12) protein expression in both Rat-1/pBabe and Rat-1/ΔN-IκBα-H cells infected by H-Ras expressing virus (Figure 6c).

Figure 5

NF-κB activity is not required for the proliferation of non-transformed Rat-1 fibroblasts. (a) Inactivation of NF-κB activity in Rat-1/ΔN-IκBα-H cells. EMSA was performed with nuclear extracts from Rat-1, Rat-1/pBabe, and Rat-1/ΔN-IκBα-H cells treated with or without TNF-α (20 ηg/ml) for 30 min. The NF-κB complex containing p65 subunit was indicated by the arrows. (b) The growth rate of Rat-1 cells is not affected by NF-κB inhibition. 1×104 cells of Rat-1, Rat-1/pBabe, and Rat-1/ΔN-IκBα-H cells were plated in 60 mm culture plates and cell numbers were counted every day. The data represent an average of duplicate plates of each cell type

Figure 6

NF-κB activity is required for H-Ras-induced cell transformation. (a) Representative Giemsa staining of Rat-1/pBabe and Rat-1/ΔN-IκBα-H cells after retroviral infection. 1.5×105 cells of Rat-1/pBabe and Rat-1/ΔN-IκBα-H cells were infected with an empty virus, pBabe(hyg), or a recombinant virus expressing oncogenic H-Ras, pBabe-H-Ras(hyg), plates were stained with Giemsa blue (see Materials and methods) 5 days after hygromycin selection (400 μg/ml). (b) Rat-1, Rat-1/pBabe, and Rat-1/ΔN-IκBα-H cells were infected with pBabe(hyg) or pBabe-H-Ras(hyg) virus as in (a), and hygromycin resistant cells were counted 5 days thereafter. The values represent the relative ratio of cell numbers of Rat-1, Rat-1/pBabe, and Rat-1/ΔN-IκBα-H infected with pBabe-H-Ras(hyg) virus after normalization of the differences in pBabe(hyg) virus infection. Similar results were obtained in at least three independent experiments. (c) Western blot analysis of oncogenic Ras proteins from Rat-1/pBabe and Rat-1/ΔN-IκBα-H cells infected with the indicated viruses. V and R indicate pBabe(hyg) and pBabe-H-Ras(hyg) viruses, respectively


In this study, we have demonstrated that inhibition of NF-κB by overexpression of a super-repressor form of IκBα blocks several major phenotypic changes associated with H-ras oncogene-mediated cell transformation including cell growth rate, saturation density, and tumorigenicity. In contrast, a similar inhibition of NF-κB in either non-transformed cells or cells with an inducible H-Ras had little effect on their growth, but renders them resistant to a subsequent cell transformation by H-ras oncogene expression. These results suggest that NF-κB activity is not necessary for normal cell growth, but required for Ras-activated abnormal cell proliferation, which contributes to Ras-mediated cell transformation.

Earlier studies suggest that NF-κB activation may contribute to cell transformation and tumorigenicity of Ras transformed cells by regulating the expression of genes encoding cell adhesion molecules and by inhibiting transformation-associated cell death (Mayo et al., 1997; Higgins et al., 1993). It has been demonstrated that p65 antisense oligonucleotide inhibited growth and tumorigenicity of variously transformed cell lines including Ras transformed 3T3 cell, presumably by inhibiting cell adhesion processes (Higgins et al., 1993; Narayanan et al., 1993). Since a number of cell adhesion genes have been shown to contain consensus NF-κB binding sites (Ahmad et al., 1995; Ledebur and Parks, 1995; Lewis et al., 1994; Takeuchi and Baichwal, 1995) and cell adhesion molecules play important roles in tumorigenesis and metastasis, inhibition of gene expression of these molecules appears to partly contribute to anti-tumorigenic activity of this oligonucleotide. However, it is not known which of these cell adhesion molecules were affected. In addition, this treatment also resulted in similar effects on immortalized fibroblasts including Rat-1 cell, suggesting that the observed anti-tumorigenic effect of p65 oligonucleotide might not be specific for transformed cells. Moreover, we found no apparent defects in cell adhesion processes in Rat-1(Ras) cells with an impaired NF-κB activity and growth property of Rat-1 cell was not affected by NF-κB inhibition. A similar result has been reported in other cell types (Bargou et al., 1997). One possibility of this discrepancy might be that antisense oligonucleotide targets at p65 subunit whereas IκBα inhibits multiple subunits of NF-κB family proteins. Different subunits of NF-κB family proteins may play different roles in regulating genes encoding cell adhesion molecules (Narayanan et al., 1993).

Oncogenic transformation of the cell leads to an increased susceptibility to apoptotic stimuli. It has been shown that inhibition of RNA polymerase II resulted in apoptosis in oncogenically transformed cells, but growth arrest in non-transformed cells. It was suggested that oncogenic proteins, by modulating the activity of downstream transcription factors, may activate a set of genes which inhibit transformation-associated apoptosis (Koumenis and Giaccia, 1997). NF-κB has been shown to play an important role in protection from cell death, presumably by regulating the expression of anti-apoptotic genes (Antwerp et al., 1996; Beg et al., 1996; Liu et al., 1996; Wang et al., 1996; Chu et al., 1997; Yamit-Hezi and Dikstein, 1998; You et al., 1997). In light of these observations, it has been reported that Ras activates NF-κB to prevent transformation-mediated cell death and this anti-apoptotic activity of NF-κB may be important for Ras-mediated cell transformation and tumorigenesis (Mayo et al., 1997). However, several lines of evidence in this study also suggest an alternative possibility. First, it was observed that inhibition of NF-κB activity led to a decrease in growth rate for Ras transformed cells, but had little effect on non-transformed parental cells. Second, the Ras transformed cell lines with NF-κB activity blocked could be obtained and these cell lines exhibited no obvious spontaneous cell death, but underwent rapid apoptosis in response to TNF-α treatment. Third, inactivation of NF-κB activity in cells where Ras promotes cell survival rather than cell death led to inhibition of Ras-induced abnormal cell growth. Taken together, these results suggest that the activation of NF-κB is required for the abnormal cell proliferation induced by oncogenic Ras. However, it is possible that the discrepancy between the finding of this study and that of the previous report (Mayo et al., 1997) could be due to the differences in the level of super-repressor IκBα expression. It is conceivable that differences in inhibitor expression level may result in the inhibition of different members of NF-κB family members which could function downstream of oncogenic Ras as either a growth promoter or death protector. Alternatively, since Ras proteins can activate multiple downstream effectors (Vojtek and Der, 1998), the differences in Ras signaling intensity could be attributed to this discrepancy.

NF-κB has been implicated in growth control (Baldwin, 1996). It has been shown that inactivation of NF-κB led to an impaired cell proliferation of transformed cells, indicating that NF-κB play roles in abnormal cell proliferation during tumorigenesis (Bargou et al., 1997; Kaltschmidt et al., 1999). It has been reported that Ras transformation results in cyclin D1 expression which accelerates G1 progression of NIH3T3 cells (Liu et al., 1995). More recent studies showed that cyclin D1 is transcriptionally regulated by NF-κB (Guttridge et al., 1999; Hinz et al., 1999; Joyce et al., 1999). Thus, it might be possible that Ras activates NF-κB to elevate cyclin D1 level, which then lead to an accelerated cell cycle progression of Ras transformed cells. However, since p21waf/cip CDK inhibitor is also elevated by oncogenic Ras (Lin et al., 1998; Zhu et al., 1998), the overall growth control by Ras appears to be more complicated (Downward, 1997). In fact, the initial analysis of cell cycle regulatory proteins in RR/ΔN-IκBα-H cells revealed that both cyclin D1 and p21waf/cip were elevated compared to RR-pBabe cells (unpublished observations). One possible explanation for this observation might be that the elevated level of cyclin D1 and p21waf/cip is due to cell cycle arrest in these cells. It would be interesting to determine which stage of cell cycle was affected in these cells.

In summary, this study suggests a new role for NF-κB in Ras-mediated cell transformation which is distinct from its anti-apoptotic activity. Further characterizations of Ras transformed cells with an impaired NF-κB activity would provide new insights into the role of NF-κB in tumorigenesis.

Materials and methods

Cell culture

All cell lines were maintained in Dulbecco's modified Eagle medium (Life Technologies, Inc., Grand Island, NY, USA) containing 10% bovine calf serum (HyClone, Logan, UT, USA) and 1% penicillin-streptomycin (Life Technologies, Inc., Grand Island, NY, USA) at 37°C with 10% CO2.

Construction of retroviral vectors and cell transfection

Recombinant retroviral vectors directing expression of N-terminal FLAG-tagged human IκBα or ΔN-IκBα were provided by Dr C Aiken (Department of microbiology and immunology, Vanderbilt University). Coding regions were cloned into BamHI and SalI sites of pBabe(puro) vector. A cDNA fragment encoding human H-Ras(V12) was amplified by PCR and cloned into pCR-TRAP vector (GenHunter, Nashville, TN, USA) then BglII fragment was excised and cloned into a BamHI digested pBabe(hyg) retroviral vector. Recombinant retroviral constructs were transfected into Φ2 packaging cell line by standard calcium phosphate precipitation method. After selection for antibiotics resistance, 3 days for puromycin (2 μg/ml) and 5 days for hygromycin B (300 μg/ml), cells were pooled and propagated. Approximately 1×107 cells of these viral producers were seeded onto a 150 mm culture plate and were grown for 24–30 h in the absence of antibiotics. The virus-containing medium was harvested and filtered through 0.22 μm filter membrane (Costar, Costar Corp., MA, USA). The resulting medium was used directly or stored at −80°C until use.

Retroviral infection and cell growth measurement

The target cells were seeded in duplicate at a density of 1.5×104 cells in each well of six-well tissue culture dishes and were grown for 24 h before infection. Cells were infected by incubating for 6 h with 2–4 ml of the virus-containing medium in the presence of 8 μg/ml polybrene (Sigma). An equal volume of fresh medium was then added to dilute polybrene to 4 μg/ml and incubated for 12–18 h. Cells were washed twice with PBS and were grown in the fresh medium for additional 24 h. The infected cells were then trypsinized and selected for antibiotics resistance in 100 mm culture plates along with uninfected control cells. After incubation for 3 days in puromycin (5 μg/ml) and 5 days in hygromycin (400 μg/ml) containing medium, the infected cells were washed twice in PBS and cell numbers were counted. Under these conditions, the uninfected control cells were completely eliminated. For cell growth rate measurement, each cell line was seeded in duplicate in six-well tissue culture dishes (1.0×104 cells/well). Cell numbers were counted every day or every other day. For Giemsa staining, cells were washed three times in PBS and were fixed for 30 min in methanol:acetic acid (3 : 1 solution). After washing in PBS, cells were stained for 30 min in 0.15% Giemsa stain (Sigma) and 10% Glycerol. The cells were then washed in water for three times and photographed after drying.

Immunoprecipitation and Western blot analysis

Total cellular proteins were prepared from exponentially growing cells in a lysis buffer (50 mM Tris-HCl, pH 7.5, 150 mM NaCl, 0.5% NP-40) supplemented with 1 mM PMSF and protease inhibitor cocktails (Boehringer Mannheim). For Western blot analysis, 10–20 μg of proteins were analysed for each sample on a 12–15% of SDS–PAGE. For immunoprecipitation, total cellular proteins were incubated for 2 h at 4°C with the agarose-conjugated anti-FLAG (M2) beads. Immune complexes were washed three times in lysis buffer containing 0.1% of SDS at room temperature prior to SDS–PAGE. Immunoblottings were carried out using the ECL kit from Amersham according to manufacturer's instruction. IκBα antibody (C-20) was purchased from Santa Cruz Biotechnologies, Inc. (Santa Cruz, CA, USA), Ras antibody (Ab-3) was from Oncogene Science, and anti-FLAG M2 affinity gel from Kodak.

Electrophoretic mobility shift assay, luciferase reporter assay, and apoptosis analysis

Preparation of nuclear extracts and gel mobility assay were carried out as previously described (Brockman et al., 1995; Schreiber et al., 1989). α32P-dATP-labeled double stranded NF-κB consensus oligonucleotides, 5’-CAACGGCAGGGGAATTCCCCTCTCCTT-3’ (Brockman et al., 1995) was incubated with 7 μg of nuclear extract. For luciferase reporter assay, 7×106 cells were plated in 65 mm culture plate 24 h prior to transfection. NF-κB-dependent luciferase reporter plasmid along with pCMV-β-galactosidase plasmid were transfected using Fugene-6 according to manufacturer's instruction (Boehringer Mannheim). Twenty-four hours after transfection, cells were trypsinized and equal numbers of cells were plated onto 6 well culture plates. IPTG was added 24 h after re-plating and cells were harvested at indicated time periods for luciferase assay. For colonigenic assay, 200 cells were plated in quadruplicate onto 6 well plate. After 24 h, IPTG (1 mM final) was added for 10 h, then washed twice with medium. Cells were treated with TNF-α (25 ηg/ml) for 24 h and were cultured for 5 days and colony numbers were counted after staining with Giemsa. The analysis of apoptotic DNA fragmentation and pro-caspase activation assay were performed essentially as described previously (Liu et al., 1996). A plasmid expressing caspase-3 was provided by Xiaodong Wang (University of Texas Southwestern Medical School, TX, USA).

Tumorigenicity assay

NIH nu/nu mice, 6–8 week old males, were purchased from Toconic. 4×105 cells in PBS from each cell line were injected subcutaneously into the left or right hips of the nude mice. Tumor development was scored 3 weeks thereafter when H-Ras transformed cells invariably formed tumors ranging from 1–3 cm in diameter.


  1. Ahmad M, Marui N, Alexander RW and Medford RM. . 1995 J. Biol. Chem. 270: 8976–8983.

  2. Antwerp DJV, Seamus JM, Kafri T, Green DR and Verma IM. . 1996 Science 274: 787–789.

  3. Baeuerle PA and Baltimore D. . 1996 Cell 87: 13–20.

  4. Baldwin Jr AS. . 1996 Ann. Rev. Immunol. 14: 649–681.

  5. Barbacid M. . 1987 Ann. Rev. Biochem. 56: 779–827.

  6. Bargou RC, Emmerich F, Krappman D, Bommert K, Mapara MY, Arnold W, Royer HD, Grinstein E, Greiner A, Scheidereit C and Dörken B. . 1997 J. Clin. Inv. 100: 2961–2969.

  7. Beg A and Baltimore D. . 1996 Science 274: 782–784.

  8. Brockman JA, Scherer DC, McKinsey TA, Hall SM, Qi X, Lee WY and Ballard DW. . 1995 Mol. Cell. Biol. 15: 2809–2818.

  9. Cardone MH, Roy N, Stennicke HR, Salvesen GS, Franke TF, Stanbridge E, Frisch S and Reed JC. . 1998 Science 282: 1318–1321.

  10. Chu ZL, McKinsey TA, Liu L, Gentry JJ, Malim MH and Ballard DW. . 1997 Proc. Natl. Acad. Sci. USA 94: 10057–10062.

  11. Downward J. . 1997 Curr. Biol. 7: R258–260.

  12. Downward J. . 1998 Curr. Opin. Genet. Dev. 8: 49–54.

  13. Finco TS, Westwick JK, Norris JL, Beg AA, Der CJ and Baldwin AS. . 1997 J. Biol. Chem. 272: 24113–24116.

  14. Guttridge DC, Albanese C, Reuther JY, Pestell RG and Baldwin Jr AS. . 1999 Mol. Cell. Biol. 19: 5785–5799.

  15. Higgins AK, Perez JR, Coleman TA, Dorshkind K, McComas WA, Sarmiento UM, Rosen CA and Narayanan R. . 1993 Proc. Natl. Acad. Sci. USA 90: 9901–9905.

  16. Hinz M, Krappmann D, Eichten A, Heder A, Scheidereit C and Strauss M. . 1999 Mol. Cell. Biol. 19: 2690–2698.

  17. Johnson R, Spiegelman B, Hanahan D and Wisdom R. . 1996 Mol. Cell. Biol. 16: 4504–4511.

  18. Joyce D, Bouzahzah B, Fu M, Albanese C, D'Amico M, Steer J, Klein JU, Lee RJ, Segall JE, Westwick JK, Der CJ and Pestell RG. . 1999 J. Biol. Chem. 274: 25245–25249.

  19. Kaltschmidt B, Kaltschmidt C, Hehner SP, Droge W and Schmitz ML. . 1999 Oncogene 18: 3213–3225.

  20. Kauffman-Zeh A, Rodriguez-Viciana P, Ulrich E, Gilbert C, Coffer P, Downward J and Evan G. . 1997 Nature 385: 544–548.

  21. Khwaja A, Rodriguez-Viciana P, Wennstrom S, Warne PH and Downward J. . 1997 EMBO J. 16: 2783–2793.

  22. Kiaris H and Spandidos DA. . 1995 Intl. J. Oncol. 7: 413.

  23. Ko TC, Sheng HM, Reisman D, Thompson EA and Beauchamp RD. . 1995 Oncogene 10: 177–184.

  24. Koumenis C and Giaccia A. . 1997 Mol. Cell. Biol. 17: 7306–7316.

  25. Langer SJ, Bortner DM, Roussel MF, Sherr CJ and Ostrowski MC. . 1992 Mol. Cell. Biol. 12: 5355–5362.

  26. Ledebur HC and Parks TP. . 1995 J. Biol. Chem. 270: 933–943.

  27. Lewis H, Kaszubska W, DeLamarter JF and Whelan J. . 1994 Mol. Cell. Biol. 14: 5701–5709.

  28. Liang P, Averboukh L, Zhu W and Pardee AB. . 1994 Proc. Natl. Acad. Sci. USA 91: 12515–12519.

  29. Lin AW, Barradas M, Stone JC, van Aelst L, Serrano M and Lowe SW. . 1998 Genes Dev. 12: 3008–3019.

  30. Liu JJ, Chao JR, Jiang MC, Ng SY, Yen JJ and Yang-Yen HF. . 1995 Mol. Cell. Biol. 15: 3654–3663.

  31. Liu Z, Hsu H, Goeddel DV and Karin M. . 1996 Cell 87: 565–576.

  32. Lowy DR and Willumsen BM. . 1993 Ann. Rev. Biochem. 62: 851–891.

  33. Mayo MW, Wang CY, Cogswell PC, Rogers-Graham KS, Lowe SW, Der CJ and Baldwin AS. . 1997 Science 278: 1812–1815.

  34. Narayanan R, Higgins KA, Perez JR, Coleman TA and Rosen CA. . 1993 Mol. Cell. Biol. 13: 3802–3810.

  35. Romashkova JA and Makarov SS. . 1999 Nature 401: 86–90.

  36. Schreiber E, Matthias P, Muller M and Schaffner W. . 1989 Nucleic Acids Res. 17: 6419.

  37. Sen R and Baltimore D. . 1986 Cell 46: 705–716.

  38. Takeuchi M and Baichwal VR. . 1995 Proc. Natl. Acad. Sci. USA 92: 3561–3565.

  39. Treisman R. . 1994 Curr. Opin. Genet. Dev. 4: 96–101.

  40. Verma IM, Stevenson JK, Schwarz EM, Antwerp DV and Miyamoto S. . 1995 Genes Dev. 9: 2723–2735.

  41. Vojtek AB and Der CJ. . 1998 J. Biol. Chem. 273: 19925–19928.

  42. Wang C, Mayo MW and Baldwin AS. . 1996 Science 274: 784–787.

  43. White MA, Nicolette C, Minden A, Polverino A, Aelst LV, Karin M and Wigler MH. . 1995 Cell 80: 533–541.

  44. Yamit-Hezi A and Dikstein R. . 1998 EMBO J. 17: 5161–5169.

  45. You M, Ku PT, Hrdlickova R and Bose Jr HR. . 1997 Mol. Cell. Biol. 17: 7328–7341.

  46. Zhu J, Woods D, McMahon M and Bishop JM. . 1998 Genes Dev. 12: 2997–3007.

Download references


We thank R Wisdom, C Aiken, X Wang for kindly providing the retroviral vectors and plasmids; L Marnett and T Daniel for critical reading of the manuscript. This work was supported in part by grants, CA76960 and CA74067 from the National Institute of Health (P Liang), AI33839 from the National Institute of Allergy and Infectious Diseases (DW Ballard) and funds from the Vanderbilt Cancer Center (P Liang) and the Howard Hughes Medical Institute (DW Ballard).

Author information

Correspondence to Peng Liang.

Rights and permissions

Reprints and Permissions

About this article


  • Ras
  • NF-κB
  • abnormal cell proliferation
  • growth inhibition

Further reading