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

This Article Abstract Full Text (PDF) 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 Benedict, M. A. Articles by Núñez, G. Search for Related Content PubMed PubMed Citation Articles by Benedict, M. A. Articles by Núñez, G. Social Bookmarking                      What's this?

J Biol Chem, Vol. 275, Issue 12, 8461-8468, March 24, 2000

Expression and Functional Analysis of Apaf-1 Isoforms
EXTRA WD-40 REPEAT IS REQUIRED FOR CYTOCHROME c BINDING AND REGULATED ACTIVATION OF PROCASPASE-9^*

Mary A. Benedict, Yuanming Hu^§, Naohiro Inohara, and Gabriel Núñez^¶

From the Department of Pathology, Comprehensive Cancer Center and Cellular and Molecular Biology Program, University of Michigan Medical School, Ann Arbor, Michigan 48109

	ABSTRACT

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS AND DISCUSSION REFERENCES

Apaf-1 is an important apoptotic signaling molecule that can activate procaspase-9 in a cytochrome c/dATP-dependent fashion. Alternative splicing can create an NH₂-terminal 11-amino acid insert between the caspase recruitment domain and ATPase domains or an additional COOH-terminal WD-40 repeat. Recently, several Apaf-1 isoforms have been identified in tumor cell lines, but their expression in tissues and ability to activate procaspase-9 remain poorly characterized. We performed analysis of normal tissue mRNAs to examine the relative expression of the Apaf-1 forms and identified Apaf-1XL, containing both the NH₂-terminal and COOH-terminal inserts, as the major RNA form expressed in all tissues tested. We also identified another expressed isoform, Apaf-1LN, containing the NH₂-terminal insert, but lacking the additional WD-40 repeat. Functional analysis of all identified Apaf-1 isoforms demonstrated that only those with the additional WD-40 repeat activated procaspase 9 in vitro in response to cytochrome c and dATP, while the NH₂-terminal insert was not required for this activity. Consistent with this result, in vitro binding assays demonstrated that the additional WD-40 repeat was also required for binding of cytochrome c, subsequent Apaf-1 self-association, binding to procaspase-9, and formation of active Apaf-1 oligomers. These experiments demonstrate the expression of multiple Apaf-1 isoforms and show that only those containing the additional WD-40 repeat bind and activate procaspase-9 in response to cytochrome c and dATP.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS AND DISCUSSION REFERENCES

Programmed cell death, or apoptosis is an evolutionarily conserved mechanism of cellular demise that is critical for embryonic development and homeostasis in adult tissues (1, 2). Genetic studies in Caenorhabditis elegans have identified two genes, ced-3 and ced-4, that are required for programmed cell death (3). Once the protein product of ced-3 was determined to be a cysteine protease (4), a family of multiple related cysteine proteases (designated caspases) was identified and found to function as the executionary arm of the apoptotic program (5, 6). This executionary arm consists of a proteolytic cascade in which upstream regulatory caspases, such as caspase-9, activate downstream effector caspases, such as caspases-3 and -7 (7). In vivo, this process ultimately results in the cleavage of target proteins and the orderly demise and removal of the cell (5, 8).

Apaf-1 was identified as a mammalian homologue of CED-4 involved in the cytochrome c-dependent activation of caspase-3 through capase-9 (9), and also as an activity that activates caspases in non-transformed cell extracts (10). The NH₂ terminus of Apaf-1 is highly homologous to CED-4 and contains a caspase recruitment domain (CARD)¹ followed by an ATPase domain (9). This structural similarity is consistent with their roles as activators of apoptosis. A critical role for Apaf-1 in the regulation of apoptosis was confirmed by the analysis of Apaf-1-deficient mice in which abnormalities were observed in several tissues, particularly the brain, and characterized by the lack of developmental cell death (11, 12). Cells derived from these mice were also resistant to a wide variety of apoptotic stimuli, including chemotherapy, dexamethasone, and -irradiation, but not Fas or tumor necrosis factor (12). In contrast to the NH₂ terminus of Apaf-1, the COOH terminus lacks homology with CED-4 and is comprised of either 12 or 13 WD-40 repeats (WDRs) (9, 13-15).

In the presence of both cytochrome c and dATP, Apaf-1 is thought to undergo a conformational change such that it binds procaspase-9 (7). The activation of caspase-9 is thought to be due to the induced proximity of procaspase-9 molecules, which leads to autoprocessing and enzymatic activation (16, 17). It has been proposed that this assembly of procaspase-9 molecules is mediated by the oligomerization of multiple Apaf-1 molecules (13-17), and that this oligomerization can be inhibited by the WDR region (16-18). Cytochrome c binds partially purified Apaf-1 and is clearly required for Apaf-1 mediated activation of procaspase-9 (9). However, the mechanism by which cytochrome c functions and its binding site remain unknown.

Recently, several investigators have described the existence of multiple Apaf-1 splice variants (13-15). In tumor cell lines, alternative splicing can create an NH₂-terminal 11-amino acid insert between the CARD and ATPase domains or an additional COOH-terminal WDR between the fifth and sixth WDRs. However, the relative expression of these forms and their ability to activate procaspase-9 remain unknown. In the present studies, we used RT-PCR to demonstrate that the Apaf-1 isoform containing both the NH₂-terminal and COOH-terminal inserts (termed here Apaf-1XL) is the major form expressed in all human tissues examined. A form containing the NH₂-terminal insert but lacking the extra WDR was also expressed. Comparative analysis of all the Apaf-1 isoforms isolated to date demonstrated that the NH₂-terminal 11-amino acid insert of Apaf-1XL was not required for cytochrome c binding or cytochrome c/dATP promotion of procaspase-9 activation. However, only Apaf-1 isoforms containing the additional WDR were able to bind cytochrome c, self-associate, and bind and activate procaspase-9 in a cytochrome c/dATP-dependent fashion.

	EXPERIMENTAL PROCEDURES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS AND DISCUSSION REFERENCES

RT-PCR Analysis of Cell Lines, and Normal Human Tissues-- Five µg of RNA from HeLa cells, human embryonic kidney 293T cells, or a panel of normal human tissues (CLONTECH, Palo Alto, CA) were used to generate first strand cDNA using a commercially available kit (Life Technologies, Inc., Gaithersburg, MD). Full-length Apaf-1 cDNAs were obtained by PCR from 293T cDNA, a HeLa cDNA library, or normal human tissue cDNAs using the specific primers: (5'-GATGGATCCACCCTAGGACCATGGATGCAAAAGCTCGAAATTG-3' and 5'-CTAGCTAGCTTACTCGAGTTCTAAAGTCTGTAAAATATATAAAATAC-3'). PCR conditions were as follows: 30 cycles of 30 s denaturation at 94 °C, 30 s annealing at 62 °C, and 5 min extension at 72 °C, followed by a 10-min extension at 72 °C. The relative amounts of Apaf-1 cDNAs with or without the NH₂-terminal 11-amino acid insert were determined using the specific primers: N1, 5'-AAGAGGAAAAAGTAAG-3' and N2, 5'-TACTCCACCTTCACACAG-3' (see Fig. 1), and the following PCR conditions: 25 cycles of 30 s denaturation at 94 °C, 30 s annealing at 52 °C, and 3-min extension at 72 °C, followed by a 10-min extension at 72 °C. The relative amounts of Apaf-1 cDNAs with or without the additional COOH-terminal WDR were determined using the specific primers: C1, 5'-CAGCTGATGGAACCTTAAAGC-3' and C2, 5'-GTCTGGTCATCAGAAGATGTC-3' (see Fig. 1) and the following PCR conditions: 25 cycles of 30 s denaturation at 94 °C, 30 s annealing at 62 °C, and 3 min extension at 72 °C, followed by a 10-min extension at 72 °C. Positive controls for these PCR reactions included as templates, 10-pg samples containing the indicated ratios of gel purified insert DNA from the two Apaf-1 plasmids, Apaf-1XL and Apaf-1S. Specific amplification of Apaf-1 fragments was confirmed using the following negative controls: PCR reactions with no template and PCR reactions performed with first strand cDNA control reactions were made without reverse transcriptase. PCR products were run on 0.8, 1, or 2.5% agarose gels and analyzed by staining with ethidium bromide.

Plasmid Constructions-- Full-length Apaf-1 PCR products were digested with BamHI and XhoI and cloned in-frame into a pcDNA3 vector engineered to encode a COOH-terminal Myc epitope tag (19). Plasmids were prepared in Escherichia coli strains XL-10 (Stratagene, La Jolla, CA) or STBL2 (Life Technologies, Inc.) grown at 30 °C to avoid spontaneous mutations. Inserts were sequenced in their entirety. The HeLa Apaf-1 cDNA cloned (termed here Apaf-1S) was identical to that previously described (9) (Fig. 1). The Apaf-1 cDNAs cloned from 293T cells, however, contained an additional 11-amino acid NH₂-terminal insert and an extra WD repeat (termed here Apaf-1XL; see Fig. 1). Two other full-length Apaf-1 isoforms identified by us (Figs. 1, B, C, and D) and others (13, 14), with either the NH₂-terminal insert (termed here as Apaf-1LN) or the COOH-terminal WD-40 insert (termed here as Apaf-1LC; see Fig. 1), were constructed by exchanging the BamHI/EcoRV fragments of Apaf-1S and Apaf-1XL. The Myc epitope-tagged Apaf-1 NH₂-terminal deletion mutant (Apaf-1S 1-559) referred to herein as "N," has been previously described (16). The Myc epitope-tagged "N+" deletion mutant (Apaf-1XL 1-570) containing the 11-amino acid insert was constructed from the N construct by replacing the BamHI/EcoRV fragment with that from Apaf-1XL (Fig. 1). The Myc epitope-tagged Apaf-1S deletion mutant (468-1194), referred to herein as "C," has been previously described (16). The "C+" deletion mutant (Apaf-1XL 479-1248) containing the additional WDR was generated from the C construct, by replacing the EcoRI fragment with that from pcDNA3 Apaf-1XL (Fig. 1). All hemagglutinin (HA) epitope-tagged Apaf-1 constructs were generated by transferring the sequenced BamHI/XhoI cDNA inserts from the pcDNA3-Myc vector to the pcDNA3-HA vector (19). An expression plasmid containing the untagged Apaf-1S cDNA first described in Ref. 9 was obtained from Dr. X. Wang (University of Texas Southwestern Medical Center, Dallas, TX). Both the pcDNA3 HA-procaspase-9 (C287S) mutant and the pcDNA3 procaspase-9 used for the in vitro caspase-9 assay have been previously described (20)

Transfection, Immunoprecipitation, and Immunoblotting-- 2.5 × 10⁶ human embryonic kidney 293T cells were transfected by the calcium phosphate method with 1-5 µg each of the indicated plasmids, as reported (19). 24 h after transfection, cells were lysed in hypotonic Buffer A (7) containing 250 mM sucrose and disrupted using a 30-gauge needle. Following centrifugation at 17,000 × g at 4 °C, cytosolic extracts were collected and used for either in vitro binding assays or in vitro caspase-9 assays. In some experiments, transfected cells were lysed with 0.2% Nonidet P-40 buffer, as described previously (19) prior to immunoprecipitation. Protein immunoprecipitation and immunoblotting with relevent antibodies were performed as described (20). Rabbit anti-Myc, mouse anti-Myc, and rabbit anti-HA antibodies were obtained from Santa Cruz Biotech (Santa Cruz, CA). Mouse anti-HA antibody was obtained from Roche Molecular Biochemicals (Indianapolis, IN) and mouse anti-cytochrome c antibody was obtained from Pharmingen (San Diego, CA). Proteins were detected using the enhanced chemiluminescence (ECL) system (Amersham Pharmacia Biotech). For anti-Apaf-1 immunoblotting, Nonidet P-40 lysates were made as described above, using the following human cell lines: embryonic kidney 293T, breast cancer T47D, cervical cancer C33A, erythroleukemia K562, and monocytic leukemia U937. Apaf-1 protein was detected with two different polyclonal anti-Apaf-1 antibodies obtained from Dr. X. Wang (University of Texas Southwestern Medical School) and Cayman Chemical Co. (Ann Arbor, MI).

In Vitro Caspase-9 Assay-- Cytosolic extracts were prepared as described above. The in vitro caspase-9 assay was performed as described previously (15). Reactions were stopped with 5 × SDS loading buffer, boiled, and loaded onto a 15% polyacrylamide/SDS gel. Gels were dried and exposed for autoradiography.

In Vitro Binding Assay-- 293T cells were transiently transfected with the indicated Myc- or HA-tagged Apaf-1 plasmid. Cytosolic extracts (see above) of the indicated plasmids were combined with or without 8 µg/ml cytochrome c (Sigma), 1 mM dATP (Roche Molecular Biochemicals), and 5 mM ATPS (Sigma) for 2 h at 4 or 23 °C, in the presence of polyclonal anti-Myc antibody and protein A/G-agarose beads. Immunoprecipitation and anti-Myc or -HA immunoblotting was performed as described above.

Fractionation of Apaf-1 Isoforms by Gel Filtration-- 4 × 10⁷ 293T cells were transfected with Myc-tagged pcDNA3, pcDNA3-Apaf-1XL, or pcDNA3-Apaf-1LN. 24 h post-transfection, S-100 cytosolic extracts were prepared as described (20) and incubated at 30 °C for 30 min in the presence or absence of 10 µg/ml cytochrome c and 1 mM dATP. Then 300 µl of lysates were loaded on a Superdex-200 HR gel filtration column (Amersham Pharmacia Biotech) pre-equilibrated with buffer A (20 mM Hepes-KOH, pH 7.5, 10 mM KCl, 1.5 mM MgCl₂, 1 mM EDTA, 1 mM EGTA, 1 mM dithiothreitol and 0.1 mM phenylmethylsulfonyl fluoride) at a flow rate of 0.5 ml/min using a Bio-Rad BioLogic HR Workstation. The column was calibrated with an Amersham Pharmacia Biotech HMW gel filtration protein standards kit plus carbonic anhydrase and cytochrome c (thyroglobulin, M_r = 669,000; ferritin, M_r = 440,000; catalase, M_r = 232,000; bovine serum albumin, M_r = 66,000; carbonic anhydrase, M_r = 29,000; cytochrome c, M_r = 12,400). After discarding the majority of the void volume, fractions of 400 µl were collected. Aliquots of 50 µl from each fraction were run on a SDS-polyacrylamide electrophoresis gel followed by immunoblotting with anti-Myc polyclonal antibody. Aliquots of 50 µl from each fraction were also incubated with 100 µM DEVD-AMC to measure DEVDase activity.

Fluorimetric Assay of Caspase Activity-- Assays of DEVD-AMC cleaving activity were carried out as described (22) using synthetic fluorogenic substrate Ac-Asp-Glu-Val-Asp-7-amino-4-methylcoumarin (Ac-DEVD-AMC) (Alexis Biochemicals, San Diego, CA). 50 µl from each column fraction were assayed in 100 µl of caspase assay buffer (20 mM Pipes, pH 7.2, 100 mM NaCl, 10 mM dithiothreitol, 1 mM EDTA, 0.1% Chaps, 10% sucrose). The reaction was started with addition of 100 µM DEVD-AMC and AMC released was measured at various times following the start of the reaction. The DEVD-AMC cleaving activity was expressed as normalized fluorescence produced after 3 h incubation at 37 °C.

	RESULTS AND DISCUSSION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS AND DISCUSSION REFERENCES

Identification of Apaf-1 Splice Variants-- Previous studies in our laboratory analyzed the mechanism by which the carboxyl terminus of Apaf-1 binds and inhibits the NH₂ terminus, preventing oligomerization and caspase-9 activation (16). Cytochrome c was reported to bind to purified Apaf-1 and be required as a co-factor for Apaf-1-mediated procaspase-9 activation (7, 9). We hypothesized that the inhibitory effect of the COOH-terminal WDRs might be relieved by binding to cytochrome c. However, we were unable to demonstrate cytochrome c binding (Fig. 3A) to the Apaf-1S isoform originally described by Zou et al. (9) and also cloned by us from HeLa cDNA (Fig. 1). We also noticed that endogenous 293T Apaf-1 protein appeared to migrate somewhat slower than transfected Apaf-1S (Fig. 1E), and in our hands we were unable to demonstrate cytochrome c/dATP-dependent in vitro activation of procaspase-9 by the Apaf-1S isoform (Fig. 2). We, therefore, used RT-PCR to clone other potential full-length Apaf-1 cDNAs from 293T cells. Two full-length Apaf-1 cDNAs were cloned from 293T cells and were identical to the Apaf-1S isoform, except that they contained an 11-amino acid insert (GKDSVSGITSY) at position 98 between the CARD and ATPase domain and a 43-amino acid WDR inserted between the fifth and sixth existing WDRs of Apaf-1S (Fig. 1A) (15). The presence of the NH₂-terminal insert is consistent with the utilization of an alternative exon donor site in exon 3 and a single acceptor site in exon 4 (GenBank accession numbers AF098871 and AF098873, respectively). The presence of the additional COOH-terminal WDR is consistent with the utilization of an additional exon 17a (GenBank accession numbers AF117658 and AF117659). Recently, Zou et al. (13) have also reported the cloning of this Apaf-1 cDNA from HeLa cells. For consistency and clarity in this paper, we have termed this isoform Apaf-1XL (15). This was done to distinguish it from two other alternative human Apaf-1 cDNA splice variants (13, 14) (and identified in Fig. 1, B, C, and D) that are also longer than the originally identified Apaf-1S (Fig. 1A). We constructed these two alternative Apaf-1 cDNAs using the Apaf-1S and Apaf-1XL cDNAs as described under "Experimental Procedures." For clarity in this paper, we have termed them Apaf-1LC (long COOH terminus: containing the additional WDR, but lacking the NH₂-terminal insert) and Apaf-1LN (long NH₂ terminus: containing the NH₂-terminal insert, but lacking the additional WDR) (Fig. 1A).

View larger version (58K): [in this window] [in a new window]   Fig. 1.   Expression of Apaf-1 isoforms. A, schematic representation of Apaf-1 isoforms examined in this study. The CARD, ATPase domain, and WDRs are shown, as are the presence or absence of the 11-amino acid NH2-terminal insert following the CARD and the 43-amino acid COOH-terminal insert between the fifth and sixth WDRs. N1/N2 and C1/C2 represent the primers used to amplify the regions flanking the NH2-terminal and COOH-terminal inserts, respectively. The deletion mutant Apaf-1S (1-559) has been termed N, while the deletion mutant Apaf-1XL (1-570) has been termed N+. C refers to the deletion mutant Apaf-1S (468-1194), while C+ refers to the deletion mutant Apaf-1XL (479-1248). B, RT-PCR analysis of the expression of full-length Apaf-1 forms in human tissue RNAs. Primers used were identical to those used to amplify full-length Apaf-1 cDNAs (see "Experimental Procedures"). The last two lanes are positive control reactions using either 10 pg of gel purified Apaf-1XL or Apaf-1S inserts as templates. C, RT-PCR analysis of the same human tissue RNAs as above, using primers N1 and N2. The last four lanes are positive control reactions in which the templates were 10 pg of Apf-1 XL and Apaf-1S, mixed at the indicated ratios. D, RT-PCR analysis of the human tissue RNAs using primers C1 and C2. The last six lanes are positive controls, with Apaf-XL and Apaf-1S mixed at the indicated ratios. E, anti-Apaf-1 immunoblot analysis of 200 µg of cell lysate from various tumor cell lines. The last four lanes are positive control lysates from 293T cells transiently transfected with the indicated Apaf-1 plasmid. In addition to the transfected Apaf-1 isoforms, an endogenous Apaf-1 protein is observed in 293T cells.

View larger version (19K): [in this window] [in a new window]   Fig. 2.   Cytochrome c-dependent in vitro activation of procaspase-9 requires the additional WD-40 repeat. Ten µg cytosolic extracts of 293T cells transiently transfected with either the vector control or the indicated Myc-tagged or untagged Apaf-1 isoforms were incubated with in vitro translated [35S]methionine-labeled procaspase-9, with or without 1 mM dATP, 8 µg/ml cytochrome c, or 5 mM ATPS at 30 °C for 30 min. Anti-Apaf-1 (X. Wang) immunoblot analysis of the cytosolic extracts used to measure in vitro procaspase-9 activation, is shown in the lower panel. Endogenous Apaf-1 protein, although present, is not detected in this exposure.

Expression Apaf-1 Isoforms in Tissues and Cell Lines-- All of the human Apaf-1 cDNAs described have been isolated from tumor cell lines (9, 13-15). To determine if multiple Apaf-1 cDNAs are present in normal human tissues, we performed full-length Apaf-1 PCR analysis on cDNAs generated from normal human tissue RNAs (Fig. 1B). This analysis demonstrated the existence of at least two Apaf-1 cDNA forms. The larger form co-migrated with the cloned Apaf-1XL fragment, while the smaller form migrated slightly above that of Apaf-1S (Fig. 1B). Restriction mapping of each of these gel purified full-length Apaf-1 PCR products confirmed their identities as Apaf-1 cDNAs.² Because of limited gel resolution, minor amounts of other Apaf-1 cDNAs may have been present but not detectable. To better examine the relative amounts of the different Apaf-1 forms, we performed PCR analysis of the human tissue cDNAs using two sets of primers that flank the two different insertions. Primers N1 and N2 flank the NH₂-terminal 11-amino acid insertion, while primers C1 and C2 flank the additional WDR (Fig. 1A). PCR analysis using primers N1 and N2 showed that in all tissues the great majority of the products (>80%) contained the 11-amino acid NH₂-terminal insertion, as determined by comparison with control reactions containing various ratios of Apaf-1XL and Apaf-1S DNAs (Fig. 1C). PCR analysis using primers C1 and C2 showed that in all tissues both types of products are represented, although the relative amounts of the two types varied among the tissues (Fig. 1D). A compilation of the PCR results from Fig. 1, C and D, suggests that the major full-length Apaf-1 cDNAs observed in most of these tissues appears to be Apaf-1XL (containing both NH₂-terminal and COOH-terminal insertions). At the level of mRNA expression, tissues such as bone marrow, colon, and spleen appear to have roughly equal amounts of Apaf-1XL and Apaf-1LN (containing just the NH₂-terminal insertion), while tissues such as brain, kidney, stomach, and skeletal muscle express more Apaf-1XL.

To determine whether multiple Apaf-1 isoforms are also expressed at the protein level, we examined multiple cell lines by immunoblotting using a polyclonal anti-Apaf-1 antibody. Lysates of 293T cells transiently transfected with the different Apaf-1 forms (described in Fig. 1A) were run as controls. The major immunoreactive band in each of the cell lines co-migrated with the Apaf-1XL form (Fig. 1E), which is consistent with the data from the mRNA analysis of human tissues identifying Apaf-1XL as the major form expressed (Fig. 1, B-D). As in the tissue mRNAs, multiple Apaf-1 protein isoforms were also expressed in these cell lines (Fig. 1E). These other bands appear to co-migrate with the Apaf-1LN and Apaf-1S forms, however, exact identification would require either protein sequencing or isoform-specific antibodies. Immunoblotting of these lysates with another polyclonal anti-Apaf-1 antibody (Cayman Chemical Co.) confirmed these results.²

Cytochrome c/dATP-dependent in Vitro Activation of Procaspase-9 requires the additional WD-40 repeat-- Purified Apaf-1 has been reported to activate procaspase-9 in a cytochrome c and dATP-dependent fashion (7). To determine if the newly identified Apaf-1 cDNAs also share this activity, and to determine the role of the NH₂-terminal and COOH-terminal insertions, the four full-length Myc-tagged Apaf-1 constructs (Fig. 1) and the originally described untagged Apaf-1S were expressed in 293T cells. Cytosolic extracts of these cells were prepared and anti-Apaf-1 immunoblotting confirmed comparable expression of each of the transfected Apaf-1 forms (Fig. 2). Ten µg of the fresh cytosolic extracts were also analyzed for their ability to activate procaspase-9 in vitro. Extracts were diluted so that endogenous Apaf-1 activity could not be detected. As shown in Fig. 2, only Apaf-1XL and Apaf-1LC containing the extra WDR, but not those isoforms lacking it, were able to activate procaspase-9 in a cytochrome c and dATP-dependent fashion as indicated by the appearance of the intermediate p35 proteolytic fragments. These same results were obtained using 1, 3, and 30 µg of cytosolic extracts or when the extracts were incubated for 30, 60, or 90 min at 30 °C with cytochrome c and dATP.² Due to the presence of low levels of dATP or ATP in the cytosolic extracts, we confirmed the requirement for ATP by the addition of the non-hydrolyzable ATP analogue, ATPS, which almost completely inhibited the cytochrome c/dATP-dependent activation of procaspase-9 (Fig. 2). Although the original untagged Apaf-1S was previously reported to activate procaspase-9 in a cytochrome c/dATP-dependent fashion (9), we were unable to detect such activity under our assay conditions (Fig. 2).

The additional WD-40 Repeat of Apaf-1 Is Necessary but Not Sufficient for the Binding of Cytochrome c-- The apparent requirement of the extra WDR for cytochrome c/dATP-dependent activation of procaspase-9 prompted us to examine the ability of the different Apaf-1 forms to bind cytochrome c. As shown in Fig. 3A, following anti-Myc antibody immunoprecipitation in the presence or absence of cytochrome c, dATP, or ATPS and immunoblotting with mouse anti-cytochrome c antibody, only the Apaf-1 constructs Apaf-1XL and Apaf-1LC, containing the extra WDR, bound to cytochrome c. This cytochrome c binding also required ATP/dATP, as binding was greatly inhibited by the addition of the non-hydrolyzable ATP analogue, ATPS (Fig. 3A). The NH₂-terminal 11-amino acid insert was not required for cytochrome c binding, as the form Apaf-1LC, lacking the NH₂-terminal insert, bound cytochrome c (Fig. 3A). Cytochrome c binding to the COOH-terminal deletion mutant, C+ (Apaf-1XL 479-1248), containing the extra WDR was not detected, suggesting that this region, although necessary (Fig. 3A), is not sufficient to mediate cytochrome c binding (Fig. 3B). Cytochrome c also failed to bind to either the N+ (Apaf-1XL 1-570) or N (Apaf-1S 1-559) deletion mutants (Fig. 3B). This data is consistent with a model in which the COOH-terminal WDR region with the additional WDR contains a binding site for cytochrome c that may be unmasked only following a conformational change driven by ATP hydrolysis in the NH₂ terminus.

View larger version (20K): [in this window] [in a new window]   Fig. 3.   The additional WD-40 repeat of Apaf-1 is necessary but not sufficient for the binding of cytochrome c. Cytosolic extracts from 293T cells transiently transfected with either vector control or the indicated Myc-tagged Apaf-1 plasmids were incubated with or without 8 µg/ml cytochrome c, 1 mM dATP, and 5 mM ATPS in the presence of monoclonal anti-Myc antibody and protein A/G-agarose beads. After incubation for 2 h at 4 °C, immunoprecipitation was performed as described under "Experimental Procedures." Cytochrome c associated with Apaf-1 proteins was detected by immunoblotting. Similar results were obtained when cell extracts were incubated with cytochrome c and dATP for 2 h at 25 °C or for 20 h at 4 °C. A, immunoprecipitated full-length Myc-tagged Apaf-1 proteins are shown in the upper panel. Cytochrome c bound to Apaf-1 proteins is shown in the lower panel. B, immunoprecipitated Myc-Apaf-1XL and Myc-tagged Apaf-1 deletion mutants are shown in the upper panel. Bound cytochrome c is shown in the lower panel.

Cytochrome c/dATP-dependent Apaf-1 Self-association Requires the Additional WD-40 Repeat-- We and others have previously shown that Apaf-1 can self-associate to form homo-oligomers (13-18). As our previous studies utilized Apaf-1 protein in Nonidet P-40 cellular extracts, we first used Nonidet P-40 extracts to compare the four different full-length Apaf-1 constructs for their ability to form homo-oligomers. Myc- and HA-tagged isoforms were expressed in 293T cells, lysed in Nonidet P-40 buffer, immunoprecipitated with anti-Myc antibody, and immunoblotted with anti-HA or anti-Myc antibodies. As shown in Fig. 4A, all four Apaf-1 isoforms were able to self-associate. Anti-HA immunoblotting confirmed similar expression of the HA-Apaf-1 isoforms in the Nonidet P-40 extracts. Since these studies were performed in the presence of detergent which could affect protein conformation, we repeated these self-association experiments in the absence of detergent. Cytosolic extracts of Myc- and HA-Apaf-1 isoforms made without detergent were mixed in the presence or absence of cytochrome c, dATP, or ATPS. Under these conditions, only Apaf-1XL and Apaf-1LC, both containing the extra WDR, underwent cytochrome c/dATP-dependent self-association (Fig. 4B). Because Apaf-1LC lacks the NH₂-terminal 11-amino acid insert, this region is clearly not required for cytochrome c/dATP-dependent self-association. Anti-HA immunoblotting confirmed similar expression of the HA-Apaf-1 isoforms. This data is consistent with a model in which cytochrome c binds the Apaf-1 COOH terminus with the extra WDR, thus relieving the inhibition of the NH₂ terminus, and allowing Apaf-1 self-association (16). The requirement of the extra WDR for Apaf-1 self-association is also consistent with our observation that it is required for procaspase-9 activation (Fig. 2). Not surprisingly, the two constructs, Apaf-1XL and Apaf-1LC, with the extra WDRs, were capable of forming cytochrome c/dATP-dependent hetero-oligomers (Fig. 4B). However, hetero-oligomers between the two major cDNAs detected in human tissues, Apaf-1XL (with the extra WDR) and Apaf-1LN (lacking the extra WDR), were not observed.²

View larger version (31K): [in this window] [in a new window]   Fig. 4.   Cytochrome c/dATP-dependent Apaf-1 self-association requires the additional WD-40 repeat. A, in Nonidet P-40 extracts, all Apaf-1 isoforms can self-associate. Nonidet P-40 extracts of 293T cells transiently transfected with the indicated Myc- or HA-tagged Apaf-1 plasmids were immunoprecipitated with rabbit anti-Myc antibody and immune complexes were immunoblotted with either mouse anti-Myc or mouse anti-HA. Immunoprecipitated Myc-Apaf-1 proteins are shown in the upper panel, and self-associated HA-Apaf-1 proteins are shown in the middle panel. The lower panel depicts an anti-HA immunoblot of the Nonidet P-40 extracts to confirm equivalent expression of the HA-Apaf-1 isoforms used. B, in cytosolic extracts, cytochrome c/dATP-dependent Apaf-1 self-association requires the additional WD-40 repeat. 293T cells were transiently transfected with the indicated Myc- or HA-tagged Apaf-1 plasmid. Cytosolic extracts of the indicated plasmids were combined with or without 8 µg/ml cytochrome c, 1 mM dATP, and 5 mM ATPS, in the presence of polyclonal anti-Myc antibody and protein A/G-agarose beads. Immunoprecipitation and anti-Myc or -HA immunoblotting was performed as described above. Immunoprecipitated Myc-Apaf-1 isoforms are shown in the upper panel, and associated HA-Apaf-1 isoforms are shown in the middle panel. The lower panel depicts an anti-HA immunoblot of the cytosolic extracts to confirm equivalent expression of the HA-Apaf-1 isoforms used.

Cytochrome c/dATP-dependent Formation of Active Apaf-1 Oligomers Requires the Extra WD-40 Repeat-- To further characterize the role of the extra WD-40 repeat in Apaf-1 oligomerization, we used size exclusion chromatography to fractionate extracts from 293T cells transiently transfected with plasmids producing Myc-tagged Apaf-1XL and Apaf-1LN which only differ in the presence of an extra WD-40 repeat (Fig. 1). Cell extracts were prepared and incubated in vitro with cytochrome c and dATP to induce Apaf-1 oligomerization or similarly treated in the absence of cytochrome c and dATP as a control. Fractions separated by gel filtration chromatography were evaluated for Apaf-1 by immunoblotting with anti-Myc antibody. In the absence of cytochrome c and dATP, Apaf-1XL eluted predominantly as a ~200-kDa protein which is consistent with a monomeric form of this protein (Fig. 5A). Preincubation of Apaf-1XL extracts with cytochrome c and dATP resulted in a significant shift in the Apaf-1 elution profile such that the majority of the Apaf-1XL eluted around fraction 5 which corresponds to ~700 kDa (Fig. 5A). These results agree with recent work by several laboratories that showed that endogenous Apaf-1 in cells or purified Apaf-1 forms containing the extra WD-40 repeat form a large multimeric complex upon addition of cytochrome c and dATP (13, 14, 23). The elution profile of Apaf-1LN was different from that of Apaf-1XL in that Apaf-1LN eluted in multiple consecutive fractions, corresponding to 200 to 700 kDa (Fig. 5B). This elution profile could potentially be the result of specific association of Apaf-1LN with another protein(s), altered protein folding, and/or formation of Apaf-1LN complexes. However, in contrast to Apaf-1XL, preincubation of Apaf-1LN extracts with cytochrome c and dATP did not change its elution profile (Fig. 5B). The multimeric Apaf-1 complex that is formed upon incubation with cytochrome c and dATP has been shown to include the processed forms of caspase-9 and caspase-3 (23). To determine the caspase activity associated with Apaf-1 protein complexes, the different protein fractions prepared from Apaf-1XL, Apaf-1LN, or control extracts were evaluated for DEVDase activity. We detected DEVDase activity in Apaf-1XL extracts in the absence and presence of cytochrome c and dATP, although the activity was clearly increased after addition of cytochrome c and dATP (Fig. 5C). The major peak of DEVDase activity in Apaf-1XL extracts was found around fraction 5 which corresponds to oligomeric Apaf-1XL (Fig. 5C). The DEVDase activity detected in the absence of cytochrome c and dATP may be explained by low level Apaf-1XL oligomerization that is detected in extracts from cells transiently transfected with the Apaf-1XL construct (Fig. 5A). Other peaks of DEVDase activity were found around fractions 13 and 18 (Fig. 5C). The latter was most prominent in Apaf-1XL extracts preincubated with cytochrome c and dATP and corresponds to a size of ~60 kDa. The DEVDase activity associated with fraction 18 most likely corresponds to a dimeric form of active caspase-3, as we found that purified processed caspase-3 also eluted in fraction 18.³ Active caspase-3 not bound to the Apaf-1 complex appears to represent caspase-3 that is processed by the oligomeric Apaf-1-caspase-9 complex and subsequently released from the complex (14, 23). Significantly, extracts from cells transfected with Apaf-1LN were devoid of DEVDase activity even after preincubation of the extracts with cytochrome c and dATP, when compared with extracts prepared from cells transfected with control plasmid (Fig. 5C).

View larger version (36K): [in this window] [in a new window]   Fig. 5.   The extra WD-40 repeat is required for cytochrome c and dATP-dependent oligomerization of Apaf-1. A and B, 293T lysates containing Myc-tagged Apaf-1XL (A) or Apaf-1LN (B) were incubated with or without dATP and cytochrome c and fractionated on a Superdex 200 HR column. Equal amounts (50 µl) of each fraction were separated on a SDS-polyacrylamide electrophoresis gel, and Apaf-1XL (A) or Apaf-1LN (B) were detected by immunoblotting with anti-Myc polyclonal antibody. The elution profiles of selected size exclusion standards are indicated by arrowheads on top in kilodaltons. C, the elution profile of DEVD-AMC cleaving activity of control (circles), Apaf-1XL (squares), or Apaf-1LN (triangles) lysates incubated without (left panel) or with (right panel) dATP and cytochrome c. Equal amounts (300 µl) of lysates were incubated with or without cytochrome c and dATP and fractionated on a Superdex 200 HR column as in A and B. The DEVD-AMC cleaving activity was measured as described under "Experimental Procedures," and expressed as arbitrary units (normalized fluorescence at 460 nM).

Cytochrome c/dATP-dependent Binding of Apaf-1 to Procaspase-9 Requires the Extra WD-40 Repeat-- We and others have previously demonstrated the binding between the originally described Apaf-1S and procaspase-9 (20, 21). We, therefore, compared the four full-length Apaf-1 isoforms for their ability to associate with procaspase-9 in the presence of Nonidet P-40. Each Myc-tagged Apaf-1 construct and HA-procaspase-9 (C287S) were expressed in 293T cells, lysed in Nonidet P-40 buffer, immunoprecipitated with anti-Myc antibody, and immunoblotted with anti-HA or anti-Myc antibodies. As shown in Fig. 6A, all Apaf-1 isoforms tested bound to procaspase-9. Anti-HA immunoblotting confirmed similar expression of procaspase-9 in each extract. However, when binding was analyzed in cytosolic extracts lacking detergent, in the presence or absence of cytochrome c, dATP or ATPS, only Apaf-1XL and Apaf-1LC containing the extra WDR bound to procaspase-9 in a cytochrome c and dATP-dependent fashion (Fig. 6B). Because Apaf-1LC lacks the NH₂-terminal 11-amino acid insert, this region is clearly not required for cytochrome c/dATP-dependent procaspase-9 binding. The requirement of the extra WDR for procaspase-9 binding is in complete agreement with our observation that this region is also required for procaspase-9 activation (Fig. 2). Previous data from our laboratory suggests that the COOH-terminal WDRs of Apaf-1 can bind and inhibit the ability of the NH₂ terminus to self-associate and activate procaspase-9 in vitro (16). Taken together, the data presented herein suggest a model in which cytochrome c, in the presence of ATP/dATP, binds to the COOH termini of only those Apaf-1 isoforms containing the extra WDR, thus relieving the inhibition of the NH₂ terminus, and allowing Apaf-1 self-association, procaspase-9 binding, and the activation of procaspase-9.

View larger version (39K): [in this window] [in a new window]   Fig. 6.   Cytochrome c-dependent binding of Apaf-1 to procaspase-9 requires the extra WD-40 repeat. A, in Nonidet P-40 extracts, all Apaf-1 isoforms can bind to procaspase-9. Nonidet P-40 extracts of 293T cells transiently co-transfected with the indicated Myc-Apaf-1 plasmids and HA-mt procaspase-9(C287S) were immunoprecipitated with rabbit anti-Myc antibody and immune complexes were immunoblotted with either mouse anti-Myc or mouse anti-HA. Immunoprecipitated Myc Apaf-1 isoforms are shown in the upper panel, and associated HA-mt procaspase-9 is shown in the middle panel. The lower panel depicts an anti-HA immunoblot of the Nonidet P-40 extracts to confirm similar expression of HA-mt procaspase-9 in each extract. B, in cytosolic extracts, cytochrome c/dATP-dependent binding of Apaf-1 to procaspase-9 requires the additional WDR. 293T cells were transiently transfected with the indicated Myc-Apaf-1 plasmid or HA-mt procaspase-9. Cytosolic extracts of the indicated plasmids were combined with or without 8 µg/ml cytochrome c, 1 mM dATP, and 5 mM ATPS, in the presence of polyclonal anti-Myc antibody and protein A/G-agarose beads. Immunoprecipitation and anti-Myc or -HA immunoblotting was performed as described above. Immunoprecipitated Myc-Apaf-1 isoforms are shown in the upper panel and associated HA-mt procaspase-9 is shown in the bottom panel. The asterisk denotes nonspecific IgG heavy chain bands.

The Apaf-1XL form, with both the NH₂-terminal and COOH-terminal inserts, appears to be the major Apaf-1 RNA expressed in most tissues and likely represents the cytochrome c/dATP-dependent activator of procaspase-9 originally described (7, 9, 10). It is interesting to note that at the mRNA level, some tissues such as bone marrow, spleen, and colon express significant amounts of Apaf-1LN, (which lacks the extra WDR and fails to bind and activate procaspase-9 in response to cytochrome c and dATP). In addition, anti-Apaf-1 immunoblotting of some tumor cell lines revealed bands that co-migrate with Apaf-1LN, suggesting that this Apaf-1 form is expressed at the protein level. It is tempting to speculate as to a possible function for Apaf-1LN. Given that in the presence of detergent the Apaf-1LN can both self-associate and bind procaspase-9, it might function as either an activator or an inhibitor of procaspase-9 activation, depending on whether it formed a functional or non-functional apoptosome complex. In our hands, Apaf-1LN does not inhibit in vitro cytochrome c/dATP-dependent procaspase-9 activation by the Apaf-1XL form,² suggesting that it does not function as a cytochrome c/dATP-dependent inhibitor. However, this is not unexpected, because of its inability to bind cytochrome c. In order for the Apaf-1LN form to function as an activator or inhibitor of procaspase-9 activation, one might hypothesize the existence of a cellular signal or co-factor other than cytochrome c that would bind specifically to the COOH terminus of Apaf-1 forms lacking the additional WDR. This co-factor might function similarly to cytochrome c and act to relieve the inhibitory action of the COOH terminus, allowing Apaf-1 self-association and procaspase-9 binding. Such a modification in binding specificity due to a change in the number of structural repeats is not without precedent. In some plants, the specificity of disease-resistance genes can be altered simply by a change in the number of leucine-rich repeats (24). In the case of Apaf-1 isoforms lacking the extra WDR, the specific binding of some unknown co-factor could potentially result in a non-cytochrome c-dependent pathway of procaspase-9 activation or inhibition.

	ACKNOWLEDGEMENT

We are grateful to Dr. Xiaodong Wang for the generous gift of reagents.

	FOOTNOTES

^* This work was supported in part by National Institutes of Health Grant CA-64556 (to G. N.) and U. S. Army Medical Research Command Grant DAMD196-609.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.

Supported by a pre-doctoral fellowship from the U. S. Army Medical Research Command.

^§ Supported by National Institutes of Health Postdoctoral Training Grant 2T32HL07517.

^¶ Recipient of Research Career and Development Award CA-64421 from the National Institutes of Health. To whom correspondence should be addressed. Tel.: 734-764-8514; Fax: 734-647-9654; E-mail: Gabriel.Nunez@umich.edu.

² M. Benedict and G. Núñez, unpublished results.

³ Y. Hu and G. Núñez, unpublished results.

	ABBREVIATIONS

The abbreviations used are: CARD, caspase recruitment domain; ATPS, adenosine 5'-O-(thiotriphosphate); HA, hemagglutinin; PCR, polymerase chain reaction; RT, reverse transcriptase; WDR, WD-40 repeat region; Pipes, 1,4-piperazinediethanesulfonic acid; Chaps, 3-[(3-cholamidopropyl)dimethylammonio]-1-propanesulfonic acid; DEVD-AMC, Asp-Glu-Val-Asp-7-amino-4-methylcoumarin.

	REFERENCES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS AND DISCUSSION REFERENCES

CiteULike

Complore

Connotea

Del.icio.us

Digg

Reddit

Technorati

This article has been cited by other articles:

				A. K. Kanungo, Z. Hao, A. J. Elia, T. W. Mak, and J. T. Henderson Inhibition of Apoptosome Activation Protects Injured Motor Neurons from Cell Death J. Biol. Chem., August 8, 2008; 283(32): 22105 - 22112. [Abstract] [Full Text] [PDF]

				E. J. Vallender and B. T. Lahn A primate-specific acceleration in the evolution of the caspase-dependent apoptosis pathway Hum. Mol. Genet., October 15, 2006; 15(20): 3034 - 3040. [Abstract] [Full Text] [PDF]

				A. Holleman, M. L. den Boer, R. X. de Menezes, M. H. Cheok, C. Cheng, K. M. Kazemier, G. E. Janka-Schaub, U. Gobel, U. B. Graubner, W. E. Evans, et al. The expression of 70 apoptosis genes in relation to lineage, genetic subtype, cellular drug resistance, and outcome in childhood acute lymphoblastic leukemia Blood, January 15, 2006; 107(2): 769 - 776. [Abstract] [Full Text] [PDF]

				C. McDonald, F. F. Chen, V. Ollendorff, Y. Ogura, S. Marchetto, P. Lecine, J.-P. Borg, and G. Nunez A Role for Erbin in the Regulation of Nod2-dependent NF-{kappa}B Signaling J. Biol. Chem., December 2, 2005; 280(48): 40301 - 40309. [Abstract] [Full Text] [PDF]

				Y. Sun, S. Orrenius, S. Pervaiz, and B. Fadeel Plasma membrane sequestration of apoptotic protease-activating factor-1 in human B-lymphoma cells: a novel mechanism of chemoresistance Blood, May 15, 2005; 105(10): 4070 - 4077. [Abstract] [Full Text] [PDF]

				S. Kornbluth and K. White Apoptosis in Drosophila: neither fish nor fowl (nor man, nor worm) J. Cell Sci., May 1, 2005; 118(9): 1779 - 1787. [Abstract] [Full Text] [PDF]

				M. Zanon, A. Piris, I. Bersani, C. Vegetti, A. Molla, A. Scarito, and A. Anichini Apoptosis Protease Activator Protein-1 Expression Is Dispensable for Response of Human Melanoma Cells to Distinct Proapoptotic Agents Cancer Res., October 15, 2004; 64(20): 7386 - 7394. [Abstract] [Full Text] [PDF]

				D. Twiddy, D. G. Brown, C. Adrain, R. Jukes, S. J. Martin, G. M. Cohen, M. MacFarlane, and K. Cain Pro-apoptotic Proteins Released from the Mitochondria Regulate the Protein Composition and Caspase-processing Activity of the Native Apaf-1/Caspase-9 Apoptosome Complex J. Biol. Chem., May 7, 2004; 279(19): 19665 - 19682. [Abstract] [Full Text] [PDF]

				C. A. Belmokhtar, J. Hillion, C. Dudognon, S. Fiorentino, M. Flexor, M. Lanotte, and E. Segal-Bendirdjian Apoptosome-independent Pathway for Apoptosis: BIOCHEMICAL ANALYSIS OF APAF-1 DEFECTS AND BIOLOGICAL OUTCOMES J. Biol. Chem., August 8, 2003; 278(32): 29571 - 29580. [Abstract] [Full Text] [PDF]

				J. Masumoto, W. Zhou, F. F. Chen, F. Su, J. Y. Kuwada, E. Hidaka, T. Katsuyama, J. Sagara, S.'i. Taniguchi, P. Ngo-Hazelett, et al. Caspy, a Zebrafish Caspase, Activated by ASC Oligomerization Is Required for Pharyngeal Arch Development J. Biol. Chem., January 31, 2003; 278(6): 4268 - 4276. [Abstract] [Full Text] [PDF]

				A. G. Martin and H. O. Fearnhead Apocytochrome c Blocks Caspase-9 Activation and Bax-induced Apoptosis J. Biol. Chem., December 20, 2002; 277(52): 50834 - 50841. [Abstract] [Full Text] [PDF]