Abstract
We have reported previously that codon 169 of the proapoptotic gene BAX is a mutational hot spot in gastrointestinal cancer. Two different mutations were found in this codon, replacing the wild-type threonine by alanine or methionine. To compare the proapoptotic activity of these Bax mutants with wild-type Bax, we established an ecdysone (muristerone A)-inducible system in cultured human embryonal kidney 293 cells. Addition of muristerone A induced a dose-dependent decrease in the viability of cells transfected with wild-type BAX, but this loss of viability was inhibited in cells transfected with BAX mutants. Furthermore, muristerone A induced morphological changes characteristic of apoptosis, including cell shrinkage, rounding, formation of apoptotic bodies, detachment and nuclear condensation and fragmentation, in cells transfected with wild-type BAX. These hallmarks of apoptosis were clearly diminished in cells transfected with BAX mutants. Mutation of threonine 169 did not affect the binding of Bax to Bax, Bcl-2, or Bcl-XL. These results demonstrate that missense mutations at codon 169 of BAX are functional because they inhibit its apoptotic activity. This is the first report of the functional significance of missense mutations in BAX, or any other proapoptotic member of the Bcl-2 family, in primary human tumors.
Introduction
Homeostasis of tissues is regulated by a balance between cell proliferation and apoptosis. Therefore, dysregulation of apoptosis may be involved in the initiation and progression of human cancer (1) . The Bcl-2 family of proteins plays a pivotal role in the regulation of apoptosis (2) . Some member of this family, such as Bcl-2 and Bcl-XL, function as cell death suppressors, whereas others such as Bax induce apoptosis. The Bcl-2 oncogene was identified at t(14:18) chromosomal translocation breakpoints in B-cell neoplasms (3) , representing the first genetic alteration in human cancer of a gene encoding for a protein involved in the control of apoptosis.
Several lines of evidence indicate that the proapoptotic protein Bax (4) plays a tumor suppressor role. Mice deficient in bax grow normally but eventually develop lymphoid hyperplasia (5) . The tumor suppressor p53 is a direct transcriptional activator of BAX (6) . Ablation of BAX in mice reportedly decreased apoptosis induced by a transgene expressing a truncated T antigen that selectively binds Rb and induces p53-dependent apoptosis (7) . In addition, chemotherapeutic agents induce apoptosis in embryonic fibroblasts in a p53-dependent manner, and elimination of BAX prevents approximately half of those chemotherapy-induced deaths (8) . Reduced Bax expression is associated with poor responses to chemotherapy and shorter survival in women with breast adenocarcinoma (9) . Overexpression of BAX in breast cancer cells restores sensitivity to a variety of apoptotic stimuli and reduces tumor growth in nude mice (10) .
The finding of somatic frameshift mutations in the BAX gene in colon cancers of the MMP 3 provided genetic evidence for the role of BAX inactivation in human tumor progression (11) . Later studies have found frameshift mutations of BAX in other MMP+ tumors (12 , 13) . BAX mutations have also been found in cell lines derived from hematopoietic malignancies (14 , 15) .
The existence of other types of nonframeshift somatic mutations in BAX added further support to the concept that this gene functions as a tumor suppressor (12, 13, 14 ,, 16 , 17) . Interestingly, some of these mutations introduce single amino acid substitutions (missense mutations). Analysis of these spontaneous mutations thus could contribute to an improved understanding of the mechanisms by which the Bax protein induces apoptosis.
We found a mutational hot spot at codon 169 in MMP+ gastrointestinal cancers (12 , 13) . Two different mutations occurred in this codon, replacing the wild-type threonine by alanine or methionine. This residue lies just proximal of the COOH-terminal hydrophobic region implicated in insertion of Bax into membranes (18, 19, 20) . Here we demonstrate that mutation of threonine 169 either to alanine or methionine decreases the proapoptotic activity of Bax.
Materials and Methods
Plasmid Constructions.
A ∼0.6 Kb human BAX-cDNA subcloned into the EcoRI site of pcDNA3 (Invitrogen, Carlsbad, CA) was liberated by EcoRI digestion, and the fragment was subcloned into the EcoRI site of pIND (Invitrogen), creating pIND-BAX. Mutant BAX plasmids were created using the QuickChange site-directed mutagenesis kit (Stratagene, La Jolla, CA) and pIND-BAX as the template with the following mutagenic primers: pIND-BAXT169A, 5′ -CTTTGGGACGCCCGCGTGGCAGACCG-3′ and 5′ -CGGTCTGCCACGCGGGCGTCCCAAAG-3′ and pIND-BAXT169M 5′-CAC-GGTCTGCCACATGGGCGTCCCAAAG-3′ and 5′ -CACGGTCTGCCACA-TGGGCGTCCCAAAG-3. Clones with the wild or mutant BAX cDNA inserted in sense orientation were identified by restriction digestion and confirmed by DNA sequencing using the ABI PRISM dye terminator cycle sequencing kit (Perkin-Elmer, Branchburg, NJ).
Generation of Stable Transfectants with Inducible Bax Expression.
To generate stable transfectants with inducible Bax expression, we used human transformed primary embryonal kidney cells (HEK-293) transfected with pVgRXR (293VgRXR; Invitrogen). pVgRXR encodes for a heterodimer of the ecdysone receptor and the retinoid X receptor, which binds to the ecdysone response element (encoded by pIND) in the presence of muristerone A (21) . 293VgRXR cells were transfected with pIND-BAX, pIND-BAX-T169A, and pIND-BAX-T169M by a calcium phosphate precipitation method, and stable transfectants were selected with 1 mg/ml Geneticin (Life Technologies, Inc., Gaithersburg, MD). After 3 weeks, antibiotic-resistant cells were maintained in DMEM supplemented with 10% FCS, 2 mm l-glutamine, 100 units/ml penicillin, 100 μg/ml streptomycin, 0.5 mg/ml zeocin, and 0.5 mg/ml Geneticin in a humidified atmosphere of 5% CO2 and 37°C. For induction of Bax expression, muristerone A was added at various concentrations for specified periods of time.
RT-PCR.
After 12 and 24 h of incubation with 5 μm muristerone A, poly(A)+ RNA was isolated from 293 transfectants using mRNA separator (Clontech laboratories, Inc., Palo Alto, CA), according to the manufacturer’s protocol. cDNA was synthesized using SuperScript II RNase H− reverse transcriptase (Life Technologies, Inc.) and oligo(dT) primer. PCR was then performed for 1 cycle of 94°C for 4 min followed by 30 cycles of 94°C for 30 s, 55°C for 30 s, and 72°C for 30 s using primers corresponding to the transcribed region of pIND upstream of BAX (ecdysone forward primer) 5′ -CTCTGAATACTTTCAACAAGTTAC-3′ and the coding region of BAX 5′-GCCCATCTTCTTCCAGATGG-3′. As a control, a glyceraldehyde-3-phosphate dehydrogenase fragment was amplified using primers 5′-CAGCCGAGCCACATCG-3′ and 5′ -TGAGGCTGTTGTCATACTTCTC-3′.
Cell Viability Assay.
Cell viability was determined by the MTT assay (22) . Cells were incubated in 24-well plates in the presence of increasing concentrations of muristerone A. After 48 h, the medium was removed and replaced with medium containing MTT (0.5 mg/ml), and cultures were incubated for an additional 2 h. The blue MTT formazan that precipitated was dissolved in 100 μl of isopropanol:1 m HCI (24:1), and the absorbance values at 590 nm were determined using a multiwell plate reader.
DAPI Staining.
Cells were washed with PBS and fixed with 2% paraformaldehyde-PBS for 10 minutes. After another wash with PBS, cells were stained with a 5 μg/ml solution of DAPI in PBS. Cells were washed three times with PBS, allowed to dry, mounted with Mowiol, and examined with a fluorescence microscope.
GST Protein Production.
pGEX4T-Bax, pGEX4T-Bcl2, pGEX4T-Bcl-XL, and pGEX4T-ΔDN TRAF3 plasmids were described previously (23, 24, 25) . GST-fusion proteins were produced as described (26) .
Protein Binding Experiments.
In vitroGST-protein binding assays were performed as described (26 , 27) . Briefly, the coding region of BAX and the mutants BAX-T169A, BAX-T169M, and TRAF6were subcloned into pcDNA3 plasmid and in vitro translated and labeled with [35S]methionine using the TnT coupled retyculocyte system (Promega, Inc., Madison, WI). Equal amounts of each labeled protein (2 μl of lysate) were then diluted with 250 μl of binding buffer [142 mm KCl, 5 mm MgCl2, 10 mm HEPES (pH 7.4), 0.2% NP40, 0.5 mm DTT, 1 mm EGTA, 0.5 mm phenylmethylsulfonyl fluoride, and a mixture of other protease inhibitors] and incubated with 0.25 μg of the GST-fusion proteins immobilized on glutathione-Sepharose at 4°C for 2 h. The resins were then extensively washed with binding buffer, and the GST-protein binding complexes were eluted with buffer containing 50 mm Tris-HCl (pH 8.0), 1 mm DTT, and 100 mm glutathione and analyzed by SDS-PAGE and fluorography.
Statistical Analysis.
Levels of significance between samples were determined using the t test for paired samples.
Results
Inducible Expression of Bax in 293 Transfectants.
To compare the proapoptotic activity of Bax mutants with wild-type Bax, we established an ecdysone (muristerone A)-inducible system in 293 cells. To avoid clonal bias, pools of clones were used for all experiments. Transcription of the BAX transgene was analyzed by reverse transcription-PCR. Before treatment with muristerone A, BAX PCR products were not detected in any 293 transfectants. After 12 h of incubation with 5 μm muristerone A, similar levels of induction of BAX mRNA were detected in 293-BAX, 293-BAX-T169A, and 293-BAX-T169M but not in 293-VgRXR control cells (Fig. 1) ⇓ . Similar results were obtained after 24 h of incubation (data not shown). Sequencing of each PCR product confirmed the presence of either wild-type or mutant BAX cDNAs as expected (data not shown).
Generation of stable transfectants with inducible Bax expression. Stably transfected cells were incubated for 12 h with 5 μm muristerone A, and poly (A)+ RNA was obtained. pIND/BAX mRNA was determined by RT-PCR using primers corresponding to the transcribed region of pIND upstream of BAX and the coding region of BAX. Ct, control cells; GAPDH, glyceraldehyde-3-phosphate dehydrogenase.
Newly synthesized Bax protein was detected by in vitro labeling with [35S]methionine for a 2.5-5h period after induction with 5 μm muristerone A for 16 h. Immunoprecipitation experiments demonstrated that addition of muristerone A induced Bax protein production in BAX-transfected cells but not in control cells that had been transfected only with the ecdysone receptor vector (data not shown).
Functional Analysis of Bax Mutants.
The effect of the induction of Bax, Bax-T169A, and Bax-T169M on the viability of the transfected cells was studied. Cells were incubated in the presence of 5 μm muristerone A for 48 h, and viability was determined using the MTT assay (Fig. 2A) ⇓ . This concentration of muristerone A induced a 60% decrease in the viability of 293-Bax cells but only a 30 and a 20% decrease in 293-Bax-T169A and 293-Bax-T169M cells, respectively. The effect was dose dependent, with 0.1 μ m muristerone A inducing a 50% decrease in the viability of 293-Bax cells but without any effect on 293-Bax-T169A or 293-Bax-T169M cells (Fig. 2B) ⇓ . Thus, substitution of threonine 169 appears to reduce the cell death-promoting activity of Bax.
Effect of induction of Bax, Bax-T169A, and Bax-T169M expression on viability of 293 cells. A, cells were incubated with 5 μm muristerone A for 48 h, and cell viability was determined by the MTT assay. Results correspond to the average of eight independent experiments performed in triplicate (*, P < 0.001 versus Bax); bars, SD. Ct, control cells. B, cells were incubated with increasing concentrations of muristerone A for 48 h, and cell viability was determined by the MTT assay. Results correspond to one representative experiment performed in triplicate.
To demonstrate that induction of Bax with muristerone A induced apoptosis, the morphology of DAPI-stained cells was analyzed by microscopy (Fig. 3) ⇓ . Addition of muristerone A (5 μm) to cultures of 293-Bax cells induced morphological changes characteristic of apo-ptosis, including cell shrinkage, rounding, formation of apoptotic bodies, and cell detachment. Chromatin condensation and nuclear fragmentation were also induced. These features were clearly less frequent in cells expressing Bax mutants. Thus, mutations substituting threonine 169 impair the proapoptotic activity of Bax.
Effect of induction of Bax, Bax-T169A, and Bax-T169M expression on apoptosis of 293 cells. Cells were incubated in the absence (−) or in the presence (+) of 5 μm muristerone A for 48 h and examined by phase contrast microscopy or stained with DAPI and examined at 24 h by fluorescence microscopy. Ct, control cells.
Mutation of Threonine 169 Does Not Affect in Vitro Bax Dimerization Capabilities.
Missense mutations of BAX found in cell lines derived from hematopoietic malignancies reportedly result in alterations of the dimerization capabilities of Bax and a decrease of its proapoptotic function (14) . To explore the effects of threonine 169 missense mutations on the dimerization of Bax, the binding of in vitro translated Bax, Bax-T169A, Bax-T169M, and TRAF6 as a negative control to GST-Bax, GST-Bcl-2, and GST-Bcl-XL was analyzed (Fig. 4) ⇓ . As a negative control, we used a chimeric protein containing GST and TRAF domain of TRAF3, which does not bind to Bax. Mutation of threonine 169 did not affect the binding of Bax to any of the Bcl-2 family proteins analyzed.
Binding of Bax and Bax mutants to Bcl-2 family proteins. In vitro binding analyses were performed using GST-Bax, GST-Bcl-2, GST-Bcl-XL, or GST-ΔN TRAF3 fusion proteins and in vitro translated [35S] methionine-labeled Bax, Bax-T169A, Bax-T169M, and TRAF6 as a negative control.
Discussion
Our results demonstrate that mutation of threonine 169 of Bax decreases its proapoptotic activity. This is the first report to analyze the functional significance of missense mutations in BAX that have been found in primary human tumors.
We have reported previously other missense mutations of BAX in gastrointestinal cancers (12 , 13) , but the functional significance of these mutations has not yet been studied. Two other missense mutations with functional significance have been found in cell lines derived from hematological malignancies, G108V in Daudi Burkitt’s lymphoma cells and G67R in HPB-acute lymphoblastic leukemia cells (14) . However, these mutations have not been found in primary tumors (28) , and it is unclear whether they were selected during the establishment of the cell lines.
The G108V and G67R mutations reside within the BH1 and BH3 domains, respectively. These mutations resulted in alteration of the dimerization capabilities of Bax and decreased its proapoptotic function (14) . We found that mutation of threonine 169 does not affect its binding to Bax, Bcl-2, or Bcl-XL. The position of threonine 169 just before the COOH-terminal membrane-anchoring domain of Bax makes it very unlikely that this mutation affects binding to other members of the Bcl-2 family.
Targeting from cytosol to mitochondrial membranes appears to be essential for the cytotoxic function of Bax (18, 19, 20) . Bax is present predominantly in the cytosol, and induction of apoptosis shifts the subcellular localization of Bax from soluble to membrane bound (19) . Removal of the COOH-terminal hydrophobic domain from Bax inhibited redistribution during apoptosis and inhibited the death-promoting activity of Bax (18 , 20) . A NH2-terminal domain represses the transmembrane signal anchor function of the carboxyl hydrophobic region of Bax (29) . It remains unclear how Bax insertion into membranes is controlled, but a phosphorylation-regulated mechanism is a reasonable candidate. Phosphorylation of Bax has been reported in vitro, but the specific sites for this phosphorylation are still unknown (30) . It is remarkable that two independent mutations occurred at codon 169, replacing the wild-type threonine by two different nonphosphorylatable amino acids, alanine and methionine. Importantly, these mutations decreased the proapoptotic activity of Bax, suggesting that threonine 169 is necessary for full proapoptotic activity of Bax. These results make threonine 169 a suitable candidate to be a phosphorylation site in Bax. Interestingly, threonine 167 has also been found mutated in a colorectal cancer (16) . In addition, two mutations in the COOH-terminal hydrophobic domain of Bax have been reported in colon cancer (17) . We speculate that these mutations could affect targeting of Bax to membranes.
Translocation of Bax to mitochondria induces dissipation of the mitochondrial inner transmembrane potential and the release of cytochrome c through the outer mitochondrial membrane (31, 32, 33, 34) . These actions may be mediated by the binding of Bax to components of the permeability transition pore complex (35) . Moreover, Bax has been reported to form ion channels in synthetic membranes in vitro (36) . Threonine 169 or its phosphorylation could be important for binding to permeability transition pore complex proteins or for the channel-forming capability of Bax. Alternatively, threonine 169 could be important for the interaction of Bax with other unidentified proteins required for its proapoptotic function.
In conclusion, we have provided evidence for the functional significance of BAX mutations found in gastrointestinal cancers. In addition to the impairment of the apoptotic activity of Bax by the missense mutations at threonine 169, the common BAX frameshift mutations found in gastrointestinal and endometrial cancer of the MMP (11, 12, 13) ,4 obviously affect its activity. In this context, these frameshift mutations provide selective advantage to the tumor cells in in vivo tumorigenicity assays. 5 In contrast with the common frameshift mutations at the mononucleotide (G)8 tract of BAX (11, 12, 13) , the hot spot for threonine 169 missense mutations is not restricted to tumors of the MMP (12) . Therefore, mutational inactivation of the proapoptotic activity of Bax appears to be under positive selective pressure during tumor progression, regardless of the molecular pathway followed by the tumor, either the mutator pathway for (pseudo)diploid cancer of the MMP or the suppressor pathway for aneuploid cancer without enhanced microsatellite instability (37 , 38) .
Footnotes
-
↵1 Supported by NIH Grants CA63585 and CA38579 (to M. P.). J. G. was supported by a fellowship from Ministerio de Educacaion y Cultura (Spain). H. Y. was supported by a Postdoctoral Fellowship for Research Abroad from the Japan Society for the Promotion of Science. J. M. Z. was supported by the Breast Cancer Research Program of the University of California.
-
↵2 To whom requests for reprints should be addressed, at The Burnham Institute, 10901 North Torrey Pines Road, La Jolla, CA 92037. Phone: (619) 646-3112; Fax: (619) 646-3190; E-mail: mperucholjcrf.edu.
-
↵3 The abbreviations used are: MMP, microsatellite mutator phenotype; MTT, 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide; DAPI, 4′,6-diamidino-2-phenylindole.
-
↵4 Schwartz, H. Yamamoto, M. Navarro, M. Maestro, J. Reventos, and M. Perucho. Frameshift mutations at mononucleotide repeats in caspase-5 and other target genes, in endometrial and gastrointestinal cancer of the microsatellite mutator phenotype, submitted for publication.
-
↵4 Y. Inov, H. Yamamoto, S. Krajewsky, J. C. Reed, and M. Perucho. Bax mutational inactivation is under strong selection in vitro and in vivoin tumor cells of the microsatellite mutator phenotype, submitted for publication.
- Received December 29, 1998.
- Accepted March 18, 1999.
- ©1999 American Association for Cancer Research.
References
Volume 59, Issue 9
