HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

J. Biol. Chem., Vol. 277, Issue 16, 14127-14134, April 19, 2002

This Article Abstract Full Text (PDF) All Versions of this Article: 277/16/14127    most recent M109939200v1 Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Degenhardt, K. Articles by White, E. Search for Related Content PubMed PubMed Citation Articles by Degenhardt, K. Articles by White, E. Social Bookmarking                      What's this?

Bax and Bak Independently Promote Cytochrome c Release from Mitochondria^*

Kurt Degenhardt^§, Ramya Sundararajan, Tullia Lindsten^¶, Craig Thompson^¶, and Eileen White^§**

From the  Howard Hughes Medical Institute,  Center for Advanced Biotechnology and Medicine, ^§ Department of Molecular Biology and Biochemistry, the ^** Cancer Institute of New Jersey, Rutgers University, Piscataway, New Jersey 08854, and ^¶ University of Pennsylvania, Departments of Medicine, Cancer Biology, and Pathology and Laboratory Medicine, Abramson Family Cancer Research Institute, Philadelphia, Pennsylvania 19104-6160

Received for publication, October 15, 2001, and in revised form, February 6, 2002

	ABSTRACT

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Pro-apoptotic Bax and Bak have been implicated in the regulation of p53-dependent apoptosis. We assessed the ability of primary baby mouse kidney (BMK) epithelial cells from bax^/, bak^/, and bax^/ bak^/ mice to be transformed by E1A alone or in conjunction with dominant-negative p53 (p53DD). Although E1A alone transformed BMK cells from p53-deficient mice, E1A alone did not transform BMK cells from bax^/, bak^/, or bax^/ bak^/ mice. Thus, the loss of both Bax and Bak was not sufficient to relieve p53-dependent suppression of transformation in epithelial cells. To test the requirement for Bax and Bak in other death signaling pathways, stable E1A plus p53DD-transformed BMK cell lines were derived from the bax^/, bak^/, and bax^/ bak^/ mice and characterized for their response to tumor necrosis factor- (TNF-)-mediated apoptosis. The loss of both Bax and Bak severely impaired TNF--mediated apoptosis, but the presence of either Bax or Bak alone was sufficient for cell death. Cytochrome c was released from mitochondria, and caspase-9 was activated in Bax- or Bak-deficient cells in response to TNF- but not in cells deficient in both. Thus, either Bax or Bak is required for death signaling through mitochondria in response to TNF-, but both are dispensable for p53-dependent transformation inhibition.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Apoptosis can be initiated in transformed cells by an intrinsic mechanism when deregulation of the cell cycle initiates an apoptotic response mediated by the tumor suppressor p53. Apoptosis can also be initiated by an extrinsic mechanism when TNF-¹ or Fas ligand initiates an apoptotic response mediated by death receptors. When the adenovirus E1A oncogene stimulates proliferation during transformation, the cellular response is apoptosis mediated by p53 (1, 2). Activation of p53 results in altered transcription of a wide variety of genes that are involved in many facets of cell metabolism, cell cycle regulation, and apoptosis (3, 4). Genes transcriptionally up-regulated by p53 that have been implicated in promoting apoptosis include the Bcl-2 family members bax, bak, puma, and noxa (5-8). Evidence suggests that Bax and Bak function is required for the release of cytochrome c from the mitochondria to the cytosol during apoptosis (9, 10). Cytochrome c release from the mitochondria occurs in many apoptotic signaling pathways including those implemented by p53 and TNF- (11-13). In many cases, this event is pivotal in the regulation of apoptosis, because cytochrome c in the cytosol complexes with APAF-1 and, in turn, promotes caspase-9 activation (14). This caspase activation initiates a caspase cascade that is required for p53-dependent apoptosis (12, 15) and results in DNA fragmentation, cleavage of cellular proteins such as poly(ADP-ribose) polymerase and nuclear lamins, and cell death by apoptosis (16).

Although there is up-regulation of bax, bak, puma, and noxa by p53, transcriptional up-regulation of at least Bax is not sufficient for p53-mediated apoptosis, because a mutant of p53 that up-regulates bax is not able to induce apoptosis (17). Evidence suggests that Bax and Bak undergo changes in protein conformation that have been linked to their pro-apoptotic function (11, 13, 18, 19). Thus, merely an increase in the amount of Bax or Bak may not result in cell death. Furthermore, other Bcl-2 family members Bcl-2, Bcl-x_L, and Bid, also regulate cytochrome c release, in some cases by modulating the function of Bax and/or Bak. Finally, caspase activation can also be regulated downstream of mitochondria by the inhibitors of apoptosis proteins (20). Which of these events is essential for cell death in particular pathways has not always been clear.

Cytokines such as TNF- and Fas initiate apoptosis through separate, yet convergent, pathways. Upon receptor ligand interaction, the death receptor will recruit proteins into a death-inducing signaling complex (21). Among the proteins recruited to the receptor are FLICE-associated death domain-containing protein/caspase-8 heterodimers, and the complex formed will result in the activation of caspase-8 (22, 23). Active caspase-8 cleaves Bid-generating tBid (24) which promotes conformational changes in Bak (25) and Bax (13). Evidence suggests that Bax and Bak oligomerize in mitochondrial membranes, the effect of which is to release cytochrome c from the inter-membrane space thereby inducing APAF-1-dependent caspase-9 activation (11, 14, 18, 25-27). Whether Bax and/or Bak function in this way in specific apoptotic pathways, and the mechanism by which they effect release of proteins from mitochondria, remains to be addressed.

Recently, mice deficient for Bax, Bak, or both have been characterized (28, 29). The bak^/ mice are developmentally normal, and the bax^/ mice have limited abnormalities including lymphoid hyperplasia and male sterility (28, 29). Mice deficient for both Bax and Bak die perinatally and have multiple developmental defects including webbing between the digits, imperforated vaginal canal, and accumulation of excess cells in the hematopoietic and central nervous systems (29). Mouse embryo fibroblasts (MEF) from these mice show that Bax and Bak are necessary for apoptosis induced by staurosporin, UV radiation, etoposide, thapsigargin, and tunicamycin in the short term (30). The release of cytochrome c in response to the overexpression of tBid is also inhibited in bax^/ bak^/ cells (30), suggesting that these proteins are essential regulators of apoptotic mitochondrial function in response multiple stimuli. Whether these findings can be extrapolated to other cell types and additional death stimuli was not known.

Here we determined that Bax and/or Bak were not essential for p53 to suppress oncogenic transformation which is likely mediated in all or in part by apoptosis. E1A expression in primary kidney epithelial cells induces p53-dependent apoptosis that must be inhibited for transformation to occur (1, 31, 32). We show that E1A alone transformed primary baby mouse kidney epithelial cells (BMK) from p53^/ mice but not from wild-type, bax^/, bak^/, or bax^/ bak^/ mice. However, wild-type, bax^/, bak^/, and bax^/ bak^/ BMK cells were efficiently transformed by E1A and dominant-negative p53 (p53DD). Thus, the loss of Bax, Bak, or both did not abrogate the requirement for loss of p53 function during E1A-induced epithelial cell transformation. Stable E1A plus p53DD-transformed BMK cell lines were derived from the foci from bax^/, bak^/, and bax^/ bak^/ mice in the BMK cell transformation assay and were characterized for their ability to undergo apoptosis by death receptor signaling pathways. The loss of either Bax or Bak did not abrogate TNF--induced apoptosis; however, the loss of both Bax and Bak conferred resistance to apoptosis. The absence of either Bax or Bak did not affect the release of cytochrome c from mitochondria to the cytosol in response to TNF-; however, the loss of both dramatically prevented the release of cytochrome c and caspase-9 activation. These findings indicate that Bax and Bak function redundantly to release cytochrome c from the mitochondria to implement apoptosis in response to death receptor signaling, but both are dispensable for p53-mediated suppression of oncogenic transformation.

	EXPERIMENTAL PROCEDURES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Transformation Assay-- Three matings were performed as follows: mice were bred by crossing mice that were bax^+/ bak^/ with bax^+/ bak^/; bax^+/ bak^+/ with bax^+/ bak^/; and bax^+/ bak^+/+ with bax^+/ bak^+/+. Newly born litters of pups (less than 48 h) were collected and numbered. Tail snips from each pup were collected and processed for DNA preparation, and PCR genotyping was performed as described for Bak (29) and Bax (33). Preparation of BMK cells from primary kidneys from murine pups was based on the BRK preparation protocol (34) with the following modifications. Kidneys from each pup were removed and processed separately under sterile conditions. Each pair of kidneys was washed with PBS, placed into 10 ml of PBS containing 2.5 mg/ml dispase II (Roche Molecular Biochemicals) and 2.5 µg/ml collagenase A (Roche Molecular Biochemicals), mechanically disrupted, and then stirred at 37 °C for 30 min. Following the incubation, 5 ml of DMEM plus 5% FBS was added and mixed by pipetting, and clumps were allowed to settle for 5 min, and the BMK were collected from the supernatant by centrifugation at 400 × g for 5 min. The BMK cells were resuspended in 550 µl of DMEM plus 5% FBS, divided into two portions of 250 µl each, and electroporated in the presence of 10 µg of linearized pCMVE1A plasmid DNA (35) and 110 µg of salmon sperm carrier DNA or 10 µg of linearized pCMVE1A plasmid DNA and 10 µg of linearized p53DD plasmid DNA (36) and 100 µg of salmon sperm carrier DNA. Post-transfection each condition was plated into three 6-cm tissue culture dishes. The culture medium was changed 1-2 times per week for 6 weeks to allow colonies to form. Colonies were ring-cloned from two plates to ensure that clones derived would be independent using standard tissue culture techniques. Giemsa stain (Sigma) was used according to the manufacturer's instructions to visualize colonies on culture plates.

Antibodies-- The following antibodies were used for indirect immunofluorescence, immunoprecipitation, and Western blotting analysis; the mouse monoclonal antibodies directed against the native and denatured forms of the cytochrome c protein were purchased from PharMingen; the rabbit N-20 polyclonal antibody directed against amino acids 11-30 of human Bax was purchased from Santa Cruz Biotechnology (Santa Cruz, CA); the rabbit polyclonal antibody (NT) against amino acids 1-21 of human Bax, which recognizes a conformation-specific form of Bax, and the rabbit polyclonal antibody (NT) raised against amino acids 23-37 of human Bak were purchased from Upstate Biotechnology, Inc. (Lake Placid, NY); the mouse monoclonal antibodies directed against p53 (pAB421), the anti-adenovirus 2 E1A (M73), and actin were purchased from Oncogene Research Products (Boston, MA); the mouse monoclonal antibody directed against caspase-9 was purchased from StressGen Biotechnologies (Victoria, British Columbia, Canada); and the rhodamine-conjugated, goat-anti-mouse antibody was purchased from Jackson ImmunoResearch Laboratories, Inc. (West Grove, PA).

Immunoprecipitation-- BMK cells were prepared for immunoprecipitation as described previously (18). Briefly, the immunoprecipitation was carried out in 2% CHAPS lysis buffer (20 mM Tris (pH 7.4), 137 mM NaCl, 2 mM EDTA, 10% glycerol, and 2% CHAPS). The Sepharose was washed three times in a 0.5% CHAPS wash buffer (20 mM Tris (pH 7.4), 137 mM NaCl, 2 mM EDTA, 10% glycerol, and 0.5% CHAPS). Cell lysates and immunoprecipitates were resolved by SDS-PAGE and then analyzed by Western blotting with the Bax NT antibody.

TNF- Apoptosis Induction Assay-- BMK cell lines were untreated, treated with CHX alone (0.05 µg/ml CHX) (Sigma), or treated with both TNF- (100 units/ml mTNF-) (Roche Molecular Biochemicals) and CHX (0.05 µg/ml). After 16 h of treatment cells were harvested by trypsinization, centrifuged, and resuspended in PBS. Cells were diluted 1:100 in 0.25% trypan blue solution (Invitrogen) and counted in a hemocytometer to assess the number of dead blue cells from the total number of cells counted. For fluorescence-activated cytometry BMK cells were treated with TNF/CHX or CHX alone for 16 h. After treatment cells were harvested by trypsinization, centrifuged, and resuspended in PBS. The cells were fixed with 70% ethanol and stained with propidium iodide (10 µg/ml) and RNase A (50 µg/ml) and incubated overnight. The cells were analyzed on a Becton Dickinson FACSCalibur system (San Jose, CA).

Clonogenic Survival Assay-- Wild-type and Bax- and Bak-deficient BMK cells were harvested, counted, and plated at specific dilutions, as indicated in the figure. Twenty four hours after plating the cells were treated with or without TNF/CHX for 16 h, as described above. Following treatment with TNF/CHX, the plates were washed once with PBS and normal growth medium (DMEM + 5% FBS) was restored. After 4 days the plates were Giemsa-stained as described above, and colonies of cells were counted.

Western Blotting-- Cell extracts were analyzed by SDS-PAGE and blotted semidry as described previously (13). Proteins were detected by antibody as indicated and visualized by enhanced chemiluminescence according to the manufacturer's specifications (Amersham Biosciences).

Subcellular Fractionation-- Cells were treated with TNF/CHX for 0 and 4 h. Subcellular fractionation was performed as described (37). Cells were harvested, washed in PBS, and resuspended in ice-cold lysis buffer (10 mM HEPES, pH 7.4, 42 mM KCl, 5 mM MgCl₂, 2 mM EDTA) containing protease inhibitors (1 mM phenylmethylsulfonyl fluoride, 20 µg/ml leupeptin, 10 µg/ml aprotinin, and 10 µg/ml pepstatin A). Cells were incubated in lysis buffer for 20 min on ice. Sucrose (2 M in lysis buffer) was added to adjust to a final concentration of 0.32 M sucrose and passed through a 30-gauge syringed needle 10 times. The cell suspension was centrifuged at 1000 × g for 10 min to eliminate nuclei and unbroken cells. The supernatant was further centrifuged at 10,000 × g to remove the mitochondrial pellet and then at 100,000 × g to yield the cytosolic soluble fraction (S-100). The S-100 was subjected to Western blot analysis by probing with a monoclonal antibody that recognizes the denatured form of the cytochrome c.

Indirect Immunofluorescence-- Cells were untreated or treated with TNF/CHX for 4 h. Indirect immunofluorescence was performed essentially as described (38). Cells were fixed at room temperature with 4.0% paraformaldehyde for 15 min and permeabilized with ice-cold PBS containing 0.2% Triton X-100 for 5 min. Cells were labeled with the antibodies that recognize native cytochrome c and visualized with rhodamine-conjugated goat anti-mouse antibody. Digital photography was performed using a Nikon FXA microscope equipped with epifluorescence optics (Nikon Inc., Garden City, NY).

	RESULTS

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Loss of Bax and Bak Was Not Sufficient to Relieve p53-dependent Suppression of Transformation-- The p53 tumor suppressor protein is crucial in preventing E1A-mediated transformation in primary baby rat kidney (BRK) epithelial cells (1, 2). BRK cells transfected with E1A alone fail to form foci due to the onset of apoptosis. However, BRK cells transfected with E1A and dominant-negative p53 efficiently form transformed foci and have been cloned into stable cell lines (1, 31, 32). The utility of this strategy was expanded by adapting it to primary mouse cells where many apoptotic and tumor suppressor genes have been targeted for gene disruption. To explore the requirement of Bax and Bak in p53-mediated apoptosis, we transfected primary BMK cells with E1A or E1A plus p53DD and assayed for transformation. Wild-type BMK cells were not transformed by E1A alone (Fig. 1). Transformation by E1A requires inactivation of p53 with a dominant-negative p53 mutant in wild-type cells or the use of BMK cells from p53-deficient mice to abolish p53 function (Fig. 1). Expression of E1A alone does not promote transformation of BMK cells from mice lacking Bax, Bak, or both Bax and Bak, indicating that both Bax and Bak may be dispensable for p53-dependent apoptosis (Fig. 1). No transformants were obtained with E1A alone (Fig. 1), and all E1A plus p53DD-derived cell lines expressed E1A and high levels of the mutant p53 (see below). Interestingly, overexpression of anti-apoptotic Bcl-2 or the adenovirus Bcl-2 homologue E1B 19K inhibits p53-mediated apoptosis (39-41) and enables transformation by E1A (42, 43). Thus, the loss of Bax and Bak function may not be equivalent to the gain of an anti-apoptotic Bcl-2 function.

View larger version (38K): [in this window] [in a new window]   Fig. 1.   The loss of both Bax and Bak is insufficient to relieve E1A-induced p53-mediated apoptosis. Transformation of primary BMK cells. BMK cells from wild-type (WT), p53/, bax/, bak/, and bax/ bak/ mice (genotype is indicated on the left) were transfected with E1A or E1A plus p53DD as indicated at the top. Following ~7 weeks the transformed foci were fixed and stained with Giemsa.

The formation of transformed foci by E1A plus p53DD in the wild-type, Bax, Bak, and Bax/Bak-deficient BMK cells allowed us to establish stable transformed cell lines, with defined deficiency in these apoptotic regulators. Although these cell lines are not suitable for studies on p53-mediated apoptosis due to the constitutive inactivation of endogenous p53, they can be used to define the roles of Bax and Bak in other apoptotic pathways.

Generation of Stable BMK Cell Lines from bax^/, bak^/, and bax^/ bak^/ Mice-- In order to study molecular events of apoptosis affected by Bax and Bak expression, stable BMK cell lines, transformed with E1A plus p53DD, were derived from the wild-type (W), bax^/ (X), bak^/ (K), and bax^/ bak^/ (D) mice. Individual foci from the transformation assay, as in Fig. 1, were cloned and grown into cell lines. Multiple independently cloned cell lines from each genotype were analyzed (W1-3, X1-3, K1-3, and D1-3), and expression of E1A, p53DD, Bax, and Bak was assessed by immunoblotting (Fig. 2A). The expression of E1A was similar in all of the cell lines tested, as were the levels of p53DD (Fig. 2A). Because all of the cell lines examined expressed both E1A and p53DD, this indicated that expression of both was required for transformation. The immunoblots for Bax and Bak matched genotyping of the parental mouse; however, the wild-type and Bak-deficient cell lines were derived from mice heterozygous for Bax (Bax^/+), and two Bax-deficient cell lines (X2, X3) were derived from mice heterozygous for Bak (Bak^/+) (Fig. 2B). The generation of these cell lines provides a renewable source of BMK epithelial cells with specific gene deficiencies in bax and/or bak. This strategy can be utilized to obtain epithelial cell lines deficient in nearly any gene for which a mouse with a targeted gene disruption has been developed and survives to birth.

View larger version (61K): [in this window] [in a new window]   Fig. 2.   Protein expression levels in stably transformed BMK cell lines. A, whole cell lysates from stably transformed BMK cell lines were analyzed by SDS-PAGE and immunoblotted with antibodies specific for E1A, p53, Bak, Bax, and actin. Three representative cell lines from each genotype are indicated as follows: Wild-type (W1,2,3); bax/ (X1,2,3); bak/ (K1,2,3); and bax/ bak/ (D1,2,3). B, BMK genotype. Bax and Bak genotypes for each of the three representative cell lines are shown.

The Loss of Both Bax and Bak Is Required for Resistance to Apoptosis Induced by TNF--- To test the requirement for Bax and Bak in extrinsic death signaling pathways, the Bax, Bak, and Bax plus Bak-deficient BMK cell lines were characterized for their apoptotic response induced by TNF-. The BMK cell lines were untreated, treated with cycloheximide alone (CHX), or TNF- and cycloheximide (TNF/CHX) for 16 h, photographed (Fig. 3A), and analyzed by trypan blue exclusion (Fig. 3B). Cell viability following cycloheximide treatment of all the cell lines was very similar to the untreated control (Fig. 3B). When treated with TNF/CHX the viability of the wild-type, bax^/, and bak^/ BMK cells dropped dramatically (Fig. 3B), indicating that the presence of either Bax or Bak is sufficient for TNF/CHX to exert its apoptotic effect. In contrast, the absence of both Bax and Bak caused the BMK cells to be resistant to TNF/CHX-mediated killing (Fig. 3B). Viability was also assessed by fluorescence-activated cytometry for DNA content, and these results show the same trends as trypan blue exclusion with the exception that the relative resistance of D2 was slightly greater than D1 (data not shown). Taken together, these data indicate that the presence of either Bax or Bak is sufficient for TNF/CHX to mediate apoptosis. Because the absence of both Bax and Bak prevents TNF--mediated killing, this suggests that Bax and Bak are functionally redundant.

View larger version (70K): [in this window] [in a new window]   Fig. 3.   The loss of both Bax and Bak is required for resistance to apoptosis induced by TNF-. A, phase contrast photographs of BMK cell lines from wild-type (WT, W3), bax/ (X1), bak/ (K1), and bax/ bak/ (D3) mice treated with CHX or TNF/CHX as indicated. B, BMK viability was assessed by trypan blue exclusion. Three representative cell lines from each genotype were untreated (blue bars), CHX-treated (green bars), or TNF/CHX-treated (yellow bars) for 16 h, and viability was assessed by trypan blue exclusion and expressed as a percent of total cells.

Bax and Bak Deficiency Confers Clonogenic Survival to TNF--- To address whether the protection to TNF--mediated death signaling conferred by deficiency of both Bax and Bak in short term survival assays was sufficient to produce long term clonogenic survival, wild-type (bax^+/ bak^+/+; W3) and Bax plus Bak-deficient (bax^/bak^/; D3) BMK cell lines were untreated or treated with TNF/CHX for 16 h and assessed for colony formation ability 4-7 days later. When plated at a density of 5 × 10⁵ cells per dish, the wild-type cell line displayed a marked reduction in colony formation with only 83 colonies formed following TNF/CHX treatment compared with the expected near monolayer of the untreated wild-type control (Fig. 4) consistent with the efficient cell death observed in short term assays (Fig. 3). In contrast, the bax^/ bak^/ cell line plated and treated with TNF/CHX at the same cell density formed colonies that were too numerous to count (Fig. 4). The bax^/ bak^/ cell line plated at a lower density and treated with TNF/CHX yielded countable colonies (1800) indicating that the enhancement of clonogenic survival was at least 43-fold greater in the bax^/ bak^/ cell line compared with the wild-type cell line (Fig. 4). The other wild-type (W1 and W2) cell lines behaved similarly to the W3 cell line, whereas the other bax^/ bak^/ cell lines (D1 and D2) behaved similarly to the D3 cell line (data not shown). As the enhanced short term survival to TNF/CHX conferred by deficiency of Bax and Bak was also reflected in a long term survival assay, this indicates that apoptosis is not merely delayed but that inhibition of apoptosis was sufficient to generate sustained survival and a cell growth advantage.

View larger version (71K): [in this window] [in a new window]   Fig. 4.   Bax and Bak deficiency confers enhanced clonogenic survival in response to TNF--mediated cell death. Wild-type (W3) and bax/ bak/ (D3) cell lines were plated at a density of 5 × 105 cells per dish and treated with TNF/CHX as indicated and as described under "Experimental Procedures." Colony formation was assessed by restoration of normal growth conditions followed by Giemsa staining. The bax/ bak/ cell line was also plated at a lower density to generate countable numbers of colonies.

TNF- Induced a Bak-independent Conformational Change in Bax-- TNF/CHX treatment of HeLa cells results in the exposure of an otherwise buried epitope in the amino terminus of Bax (11, 13, 18, 19). This is consistent with a TNF--mediated death signaling stimulating a conformational change in the Bax protein. In order to determine whether Bak was involved in mediating the amino terminal conformational change in Bax, a conformation specific antibody to the amino terminus of Bax was used for immunoprecipitation of wild-type (bax^+/ bak^+/+; W2), Bax-deficient (bax^/ bak^+/+; X1), Bak-deficient (bax^+/ bak^/; K1), and Bax plus Bak-deficient (bax^/ bak^/; D2) BMK cell lines. Immunoblot analysis showed that equal amounts of Bax were available for immunoprecipitation in both CHX and TNF/CHX-treated cell lysates (Fig. 5A). The cell lines were treated with TNF/CHX and CHX and subjected to immunoprecipitation. Bax was immunoprecitated from lysates of wild-type and bak^/ cells but only upon treatment with TNF/CHX (Fig. 5A). These data suggested that Bax undergoes a TNF--mediated amino terminal conformation change and that this is independent of Bak expression.

View larger version (39K): [in this window] [in a new window]   Fig. 5.   Bak is not required for an amino-terminal conformational change in Bax, and loss of both Bax and Bak is required to prevent cytochrome c release from mitochondria. A, exposure of Bax amino terminus is Bak-independent. One representative cell line from each genotype (W2, X1, K1, and D2) was treated with CHX or TNF/CHX for 6 h and subjected to immunoprecipitation (IP) with an amino-terminal conformation-specific anti-Bax NT antibody. Immunoprecipitates and corresponding cell lysates were resolved by SDS-PAGE and immunoblotted with an anti-Bax antibody. B, Bax or Bak is required for cytochrome c release from mitochondria. One representative cell line from each genotype (W3, X1, K1, and D2) was treated with CHX or TNF/CHX for 4 h and subjected to indirect immunofluorescence with antibody directed against native cytochrome c. Arrows indicate cells with cytochrome c released from the mitochondria. Yellow numbers represent the percentage of flat cells with released cytochrome c. C, subcellular fractionation of cytochrome c from TNF/CHX-treated BMK cell lines. One representative cell line from each genotype (W3, X1, K1, and D2) was treated with CHX or TNF/CHX for 6 h and subjected to cellular fractionation. Cell fractions and corresponding whole cell lysates (WCL) were resolved by SDS-PAGE and immunoblotted with an anti-cytochrome c or anti-cytochrome oxidase subunit IV antibody.

The Absence of Both Bax and Bak Prevents Cytochrome c Release from Mitochondria-- Cytochrome c is in mitochondria upon release into the cytosol, and it forms a cytochrome c-APAF-1 complex that activates caspase-9 (44). By using indirect immunofluorescence the subcellular localization of cytochrome c was determined in CHX and TNF/CHX-treated BMK cell lines. All cell lines treated with CHX alone displayed mitochondrial cytochrome c localization (Fig. 5B). TNF/CHX treatment of wild-type, bax^/, and bak^/ cell lines caused translocation of cytochrome c to the cytosol, whereas the bax^/ bak^/ cell line treated with TNF/CHX resulted in the dramatic retention of cytochrome c in mitochondria (Fig. 5B). Quantitative evaluation of immunofluorescence data revealed that there were fewer cells with released cytochrome c in the bak^/ cell line compared with wild-type or bax^/ cell line treated with TNF/CHX (Fig. 5B). Strikingly, however, almost no bax^/ bak^/ cells were observed with released cytochrome c upon treatment with TNF/CHX (Fig. 5B). Cellular fractionation showed an accumulation of cytochrome c in the cytosol (S100) of wild-type, bax^/, and bak^/ cell lines but not in the bax^/ bak^/ cell line in response to TNF/CHX (Fig. 5C). Immunoblotting for cytochrome oxidase subunit IV, an integral membrane protein of mitochondria, showed that the S100 fraction was not contaminated with mitochondria (Fig. 5C). Taken together, the data indicated that the presence of either Bax or Bak is necessary for cytochrome c to be released from mitochondria following TNF/CHX treatment. Loss of both Bax and Bak prevented cytochrome c release, and the apoptotic pathway may be blocked at this point which caused the bax^/ bak^/ cells to remain viable.

The Loss of Both Bax and Bak Prevents Caspase-9 Activation-- To demonstrate the functional consequence of cytochrome c release in TNF--mediated apoptosis, we assessed the role of caspase-9. Immunoblot analysis using an anti-caspase-9 antibody showed the appearance of p35, the processed active form of caspase-9, in extracts from wild-type, bak^/, and bax^/ cell lines treated with TNF/CHX for 6 h. Processed caspase-9 did not appear in extracts from the bax^/ bak^/ cell line treated with TNF/CHX (Fig. 6). This indicated that caspase-9 was not activated in the bax^/ bak^/ BMK cells in response to TNF/CHX treatment. Thus, activation of caspase-9 relies on the presence of either Bax or Bak.

View larger version (45K): [in this window] [in a new window]   Fig. 6.   Loss of both Bax and Bak prevents caspase-9 activation. One representative cell line from each genotype (W3, X1, K1, and D2) was treated with CHX or TNF/CHX for 6 h, and cell extracts were made. Cell extracts were resolved by SDS-PAGE and immunoblotted with anti-caspase-9 antibody or anti-actin antibody. The unprocessed pro-form (pro-caspase-9) and the processed p35 of active enzyme are indicated.

	DISCUSSION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

We have developed a system that utilizes primary kidney epithelial cells to study the role of oncogenes and tumor suppressor genes in transformation. As many of the mechanisms implicated in transformation of epithelial cells are also mechanisms involved in oncogenesis, and tumor cells are most often of epithelial origin, this model system may provide essential clues to the molecular pathways involved in cancer development. Surprisingly, we found that Bax and Bak do not play an essential role in p53-mediated inhibition of transformation but are required for apoptosis mediated by death receptor signaling in epithelial cells. Thus p53 may suppress transformation by means other than apoptosis induction or p53 may induce apoptosis by a Bax- and Bak-independent pathway. Attempts to transform BMK cells with E1A and a temperature-sensitive p53 mutant have not been successful, necessitating an alternative approach to discriminate between these two possibilities.

Although Bax and Bak may play a role in p53-dependent apoptosis, their presence is not required for p53 to inhibit transformation, indicating that deficiency of both Bax and Bak is not functionally equivalent to loss of p53. Indeed, bax^/ bak^/ mice have not been reported to be tumor-prone as have p53^/ mice. This finding suggests a role for an alternative pathway controlling p53-mediated transformation inhibition. p53 transcriptionally regulates many genes that could prevent transformation in the absence of Bax and Bak (3, 4). Another well characterized function of p53 that could limit transformation is the induction of cell cycle arrest. In the case of the failure of E1A alone to transform bax^/ bak^/ cells, this is probably not the case because small colonies appeared and then regressed which is the hallmark of p53-mediated apoptosis. These results underscore the importance of recent efforts to identify p53-regulated genes that may be vital to understanding these molecular mechanisms essential for tumor suppression.

Transformation studies in MEFs using E1A and activated Ras show that both these oncogenes are required and sufficient to transform wild-type MEFs, but in the absence of Bax and Bak there is a greater transformation efficiency similar to the transformation efficiency in the absence of p53 (45). Our data showed that both E1A and p53DD are required to transform BMK cells regardless of the functional status of Bax and/or Bak, whereas E1A alone efficiently transformed p53^/ BMK cells, suggesting that in epithelial cells a Bax- and Bak-independent p53-mediated pathway prevents transformation. Taken together, this suggests that a p53-mediated Bax- and Bak-independent pathway may exist that could be inhibited by activated Ras.

To explore the roles of the mitochondrial proteins Bax and Bak in other death signaling pathways, we examined TNF--mediated apoptosis. For these studies we derived transformed epithelial cells that stably express E1A and dominant-negative p53 from mice with defined deficiencies in Bax, Bak, or both. Viability data showed that treatment with TNF- induced apoptosis in wild-type, bax^/, and bak^/ cell lines. However, cell lines lacking both Bax and Bak were resistant to TNF--mediated apoptosis in both short term and long term clonogenic survival assays. These results suggested that Bak can compensate for loss of Bax and vice versa, but the loss of both prevents cell death and that control of mitochondrial function by one of these proteins is essential for apoptosis induced by TNF-. However, the enhanced short term and long term survival conferred by Bax and Bak deficiency in response to TNF--mediated death signaling may not necessarily translate into a long term survival advantage in response to cytotoxic stimuli. In contrast to cytotoxic agents, TNF- directly signals cell death through caspase-8 activation, and removal of the death stimulus, which presumably inhibits the generation of active caspase-8, is known to permit clonogenic survival of cells protected by E1B 19K expression (43). Addressing the issue of long term cell survival through the treatment of the BMK cell lines with cytotoxic chemotherapeutic agents, where differences conferred by Bax and Bak deficiency would have clinical relevance, will be interesting.

The regulation of mitochondrial function in BMK cells is in accordance with type II cells where TNF--mediated apoptosis is dependent on mitochondrial amplification of the death signal. This observation holds true unless the BMK cell lines are treated with very high doses of TNF- (10,000 units/ml), and under this condition even BMK cell lines lacking both Bax and Bak are susceptible to TNF--mediated apoptosis (data not shown). This could be explained under a cell type I model where high levels of TNF- would cause caspase-8 to directly activate caspase-3 and/or caspase-7, thereby circumventing the requirement for mitochondrial amplification of the death signal. Indeed, TNF/CHX treatment induces caspase-8 to cleave Bid into tBid initiating Bax/Bak-mediated cytochrome c release, but tBid did not induce death in bax^/ bak^/ MEFs (46). However, MEFs lacking both Bax and Bak are not resistant to TNF--mediated apoptosis (30), which implies that MEFs may resemble type I cells with respect to TNF- signaling, so the loss of Bax and Bak does not prevent TNF--mediated apoptosis. Alternatively, one obvious difference between the MEFs and the BMK cell lines is the expression of E1A and dominant-negative p53 in the BMK cell lines. Although E1A expression does sensitize cells to TNF- (47, 48),² this issue cannot be addressed using BMK cell lines because they all express E1A and dominant-negative p53. This apparent difference between MEFs and BMKs emphasizes the significance of understanding apoptotic events occurring in multiple cell types.

Here we have dissected the molecular mechanisms involved in the control of TNF--mediated apoptosis. TNF- promotes the relocalization of cytochrome c from mitochondria to the cytosol where it can complex with APAF-1 in the apoptosome (44). How Bax and Bak cause the release of cytochrome c from mitochondria has not been resolved. The structures of Bax, Bcl-2, Bid, and Bcl-x_L are all similar to bacterial toxins that act as pore-forming proteins which suggests that Bcl-2 family members may act by forming pores in mitochondrial membranes (50-55). In vitro studies utilizing lipid bilayers show that Bcl-2 family members can form pores (56-58). These studies as well as others (11, 18, 19, 25, 26, 59) suggest that pores may be formed upon conformational change and oligomerization of Bax and/or Bak. Because Bax and Bak are likely to be structurally similar, they may form pores composed of either homo- or hetero-oligomers. Because bax^/ cells can release cytochrome c as readily as wild-type cells, as demonstrated by indirect immunofluorescence and cellular fractionation, then Bak-containing cells do not rely on the formation of a Bax pore to release cytochrome c. Similarly, bak^/ cells also effectively release cytochrome c; thus, Bax containing cells do not rely on Bak to either form a pore or initiate a Bax amino-terminal conformational change, as suggested by immunoprecipitation. Cytochrome c relocalization is prevented when neither Bax nor Bak are present, supporting the proposal that these proteins are both functioning to regulate cytochrome c release from mitochondria. Apoptosis is an essential cellular response to insults, such as deregulated cell proliferation and viral infection. Indeed, either Bax or Bak is sufficient to mediate apoptosis in response to E1A expression during adenovirus infection where apoptosis serves to limit virus replication (49). Thus, the redundant function of Bax and Bak ensures preservation of proper apoptotic function and demonstrates the selective pressure for maintaining apoptotic propensity.

	FOOTNOTES

^* The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

To whom correspondence should be addressed. Tel.: 732-235-5329; Fax: 732-235-5795; E-mail: ewhite@cabm.rutgers.edu.

Published, JBC Papers in Press, February 8, 2002, DOI 10.1074/jbc.M109939200

² D. Perez and E. White, submitted for publication.

	ABBREVIATIONS

The abbreviations used are: TNF-, tumor necrosis factor-; CHAPS, N,N-dimethyl-N-(3-sulfopropyl)-3-[[(3,5,7,12)-3,7,12 trihydroxy-24-oxocholan-24-yl]-amino]-1-propanaminium inner salt; PBS, phosphate-buffered saline; FBS, fetal bovine serum; CHX, cycloheximide; DMEM, Dulbecco's modified Eagle's medium; BMK, baby mouse kidney; MEF, mouse embryo fibroblast; BRK, baby rat kidney.

	REFERENCES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES