Portal de Periódicos da CAPES

The Secreted Protease Factor CPAF Is Responsible for Degrading Pro-apoptotic BH3-only Proteins in Chlamydia trachomatis-infected Cells

2006; Elsevier BV; Volume: 281; Issue: 42 Linguagem: Inglês

10.1016/s0021-9258(19)84062-0

1083-351X

Mustak Pirbhai, Feng Dong, Youmin Zhong, Kelvin Z. Pan, Guangming Zhong,

Amoebic Infections and Treatments

Chlamydia trachomatis has evolved a profound anti-apoptotic activity that may aid in chlamydial evasion of host defense. The C. trachomatis anti-apoptotic activity has been correlated with blockade of mitochondrial cytochrome c release, inhibition of Bax and Bak activation, and degradation of BH3-only proteins. This study presents evidence that a chlamydia-secreted protease factor designated CPAF is both necessary and sufficient for degrading the BH3-only proteins. When the C. trachomatis-infected cell cytosolic extracts were fractionated by column chromatography, both the CPAF protein and activity elution peaks overlapped with the BH3-only protein degradation activity peak. Depletion of CPAF with a CPAF-specific antibody removed the BH3-only protein degradation activity from the infected cell cytosolic extracts, whereas depletion with control antibodies failed to do so. Notably, recombinant CPAF expressed in bacteria was able to degrade the BH3-only proteins, whereas CPAF mutants similarly prepared from bacteria failed to do so. Finally, bacterium-expressed CPAF also degraded the human BH3-only protein Pumaα purified from bacteria. These results demonstrate that CPAF contributes to the chlamydial anti-apoptotic activity by degrading the pro-apoptotic BH3-only Bcl-2 subfamily members. Chlamydia trachomatis has evolved a profound anti-apoptotic activity that may aid in chlamydial evasion of host defense. The C. trachomatis anti-apoptotic activity has been correlated with blockade of mitochondrial cytochrome c release, inhibition of Bax and Bak activation, and degradation of BH3-only proteins. This study presents evidence that a chlamydia-secreted protease factor designated CPAF is both necessary and sufficient for degrading the BH3-only proteins. When the C. trachomatis-infected cell cytosolic extracts were fractionated by column chromatography, both the CPAF protein and activity elution peaks overlapped with the BH3-only protein degradation activity peak. Depletion of CPAF with a CPAF-specific antibody removed the BH3-only protein degradation activity from the infected cell cytosolic extracts, whereas depletion with control antibodies failed to do so. Notably, recombinant CPAF expressed in bacteria was able to degrade the BH3-only proteins, whereas CPAF mutants similarly prepared from bacteria failed to do so. Finally, bacterium-expressed CPAF also degraded the human BH3-only protein Pumaα purified from bacteria. These results demonstrate that CPAF contributes to the chlamydial anti-apoptotic activity by degrading the pro-apoptotic BH3-only Bcl-2 subfamily members. Chlamydia trachomatis infection is a leading cause of both preventable blindness in developing countries and sexually transmitted diseases worldwide. Urogenital infection with C. trachomatis can lead to complications such as pelvic inflammatory diseases, ectopic pregnancy, and infertility in women and reactive arthritis in men. The precise pathogenic mechanisms of C. trachomatis-induced diseases are still unclear despite decades of extensive research efforts. It is thought that the inflammatory responses triggered by C. trachomatis-infected cells can significantly contribute to the chlamydia-induced pathologies in humans (1Rasmussen S.J. Eckmann L. Quayle A.J. Shen L. Zhang Y.X. Anderson D.J. Fierer J. Stephens R.S. Kagnoff M.F. J. Clin. Investig. 1997; 99: 77-87Crossref PubMed Scopus (432) Google Scholar, 2Stephens R.S. Trends Microbiol. 2003; 11: 44-51Abstract Full Text Full Text PDF PubMed Scopus (271) Google Scholar, 3Morrison R.P. Belland R.J. Lyng K. Caldwell H.D. J. Exp. Med. 1989; 170: 1271-1283Crossref PubMed Scopus (224) Google Scholar), suggesting that the root cause of C. trachomatis pathogenicity may lie in the chlamydial ability to invade host cells and to maintain long-term survival in the infected hosts. C. trachomatis has evolved a unique intracellular biphasic life cycle (4Hackstadt T. Fischer E.R. Scidmore M.A. Rockey D.D. Heinzen R.A. Trends Microbiol. 1997; 5: 288-293Abstract Full Text PDF PubMed Scopus (158) Google Scholar) and is able to both productively replicate and persistently survive within a cytoplasmic vacuole of eukaryotic cells (5Wyrick P.B. Cell. Microbiol. 2000; 2: 275-282Crossref PubMed Scopus (135) Google Scholar, 6Beatty W.L. Byrne G.I. Morrison R.P. Trends Microbiol. 1994; 2: 94-98Abstract Full Text PDF PubMed Scopus (176) Google Scholar, 7Hackstadt T. Curr. Opin. Microbiol. 1998; 1: 82-87Crossref PubMed Scopus (62) Google Scholar). A typical chlamydial infection starts with entry of an infectious elementary body into host cells via endocytosis. An elementary body differentiates into a noninfectious but metabolically active reticulate body, which multiplies and differentiates back to elementary bodies. The mature elementary bodies are then released for spreading to the adjacent host cells. The entire growth cycle takes ∼48–72 h to complete in vitro during productive infection, whereas a persistent infection can last for a long period of time in vivo. To ensure its long-term survival in the infected hosts, chlamydia has also acquired the ability to evade host defense mechanisms (8Fan T. Lu H. Hu H. Shi L. McClarty G.A. Nance D.M. Greenberg A.H. Zhong G. J. Exp. Med. 1998; 187: 487-496Crossref PubMed Scopus (493) Google Scholar, 9Zhong G. Fan T. Liu L. J. Exp. Med. 1999; 189: 1931-1938Crossref PubMed Scopus (175) Google Scholar, 10Zhong G. Liu L. Fan T. Fan P. Ji H. J. Exp. Med. 2000; 191: 1525-1534Crossref PubMed Scopus (171) Google Scholar, 11Zhong G. Fan P. Ji H. Dong F. Huang Y. J. Exp. Med. 2001; 193: 935-942Crossref PubMed Scopus (321) Google Scholar, 12Greene W. Xiao Y. Huang Y. McClarty G. Zhong G. Infect. Immun. 2004; 72: 451-460Crossref PubMed Scopus (94) Google Scholar). For example, a chlamydia-secreted protein with proteolytic activity designated CPAF (chlamydial protease/proteasome-like activity factor) is responsible for degrading host transcription factors such as RFX5 required for major histocompatibility complex antigen expression (11Zhong G. Fan P. Ji H. Dong F. Huang Y. J. Exp. Med. 2001; 193: 935-942Crossref PubMed Scopus (321) Google Scholar), which may allow chlamydia to escape efficient immune detection. CPAF can also cleave cytokeratin 8 (13Dong F. Su H. Huang Y. Zhong Y. Zhong G. Infect. Immun. 2004; 72: 3863-3868Crossref PubMed Scopus (77) Google Scholar), which may facilitate chlamydial vacuole expansion and intravacuolar growth. Interestingly, CPAF is synthesized as a proenzyme by chlamydia and has to be processed into intramolecular dimers to acquire proteolytic activity (14Dong F. Sharma J. Xiao Y. Zhong Y. Zhong G. Infect. Immun. 2004; 72: 3869-3875Crossref PubMed Scopus (39) Google Scholar, 15Dong F. Pirbhai M. Zhong Y. Zhong G. Mol. Microbiol. 2004; 52: 1487-1494Crossref PubMed Scopus (44) Google Scholar). However, it is still unknown whether CPAF targets other cellular proteins and what the CPAF substrate specificity is. To evade host immune effector mechanisms, chlamydia has evolved a profound antiapoptotic activity to suppress host cell apoptosis in both productively (8Fan T. Lu H. Hu H. Shi L. McClarty G.A. Nance D.M. Greenberg A.H. Zhong G. J. Exp. Med. 1998; 187: 487-496Crossref PubMed Scopus (493) Google Scholar) and persistently (16Dean D. Powers V.C. Infect. Immun. 2001; 69: 2442-2447Crossref PubMed Scopus (121) Google Scholar) infected cells. However, the precise mechanisms of the chlamydial anti-apoptotic activity are still unknown. An obvious question is whether CPAF is also involved in the chlamydial anti-apoptotic activity. Host cell apoptosis is a highly regulated form of cell death that plays vital roles in many biological processes, including remodeling tissues during development and eliminating the infected cells during host defense responses. Two major apoptosis pathways have been identified, and they are the intrinsic (17Sprick M.R. Walczak H. Biochim. Biophys. Acta. 2004; 1644: 125-132Crossref PubMed Scopus (191) Google Scholar) and extrinsic (18Ozoren N. El-Deiry W.S. Semin. Cancer Biol. 2003; 13: 135-147Crossref PubMed Scopus (261) Google Scholar) pathways. In the intrinsic pathway, the intracellular death signals induce mitochondrial release of apoptogenic factors such as cytochrome c that participate in the formation of the apoptosome, leading to activation of downstream effector caspases, including caspase-3, -6, and -7. The activated effector caspases can cleave a wide range of cellular molecules, including the DNA-repairing enzyme poly(ADP-ribose) polymerase, and also activate endonucleases, finally leading to irreversible nuclear apoptosis. On the other hand, the extrinsic pathway is triggered by the ligation of death receptors with ligands from extracellular sources, resulting in the formation of a death-inducing signaling complex that causes activation of upstream caspases such as caspase-8. The activated caspase-8 can either directly activate the downstream effector caspases to cause terminal apoptosis in the so-called type I cells such as the B lymphoblastoid cell line SKW6.4 or simultaneously enter the intrinsic mitochondrial pathway via the generation of cleaved Bid (19Li H. Zhu H. Xu C.J. Yuan J. Cell. 1998; 94: 491-501Abstract Full Text Full Text PDF PubMed Scopus (3782) Google Scholar) to amplify the receptor-transmitted death signals to achieve terminal apoptosis in type II cells such as the T cell line Jurkat (20Scaffidi C. Fulda S. Srinivasan A. Friesen C. Li F. Tomaselli K.J. Debatin K.M. Krammer P.H. Peter M.E. EMBO J. 1998; 17: 1675-1687Crossref PubMed Scopus (2623) Google Scholar, 21Scaffidi C. Schmitz I. Zha J. Korsmeyer S.J. Krammer P.H. Peter M.E. J. Biol. Chem. 1999; 274: 22532-22538Abstract Full Text Full Text PDF PubMed Scopus (530) Google Scholar). A recent study has demonstrated that the effector caspases can also amplify cell death potency via a positive feedback pathway by promoting mitochondrial release of cytochrome c (22Lakhani S.A. Masud A. Kuida K. Porter Jr., G.A. Booth C.J. Mehal W.Z. Inayat I. Flavell R.A. Science. 2006; 311: 847-851Crossref PubMed Scopus (910) Google Scholar). It is obvious that mitochondria play a critical role in controlling eukaryotic cell apoptosis. The Bcl-2 family members regulate mitochondrial release of apoptogenic factors, including cytochrome c (23Willis S.N. Adams J.M. Curr. Opin. Cell Biol. 2005; 17: 617-625Crossref PubMed Scopus (641) Google Scholar, 24Bouchier-Hayes L. Lartigue L. Newmeyer D.D. J. Clin. Investig. 2005; 115: 2640-2647Crossref PubMed Scopus (178) Google Scholar, 25Newmeyer D.D. Ferguson-Miller S. Cell. 2003; 112: 481-490Abstract Full Text Full Text PDF PubMed Scopus (1083) Google Scholar). There are several dozens of members in the Bcl-2 family, and they share at least one common Bcl-2 homology (BH) 2The abbreviations used are: BH, Bcl-2 homology; GST, glutathione S-transferase; mAb, monoclonal antibody. 2The abbreviations used are: BH, Bcl-2 homology; GST, glutathione S-transferase; mAb, monoclonal antibody. domain. A prototype of Bcl-2 protein has four BH domains, including BH4, BH3, BH1, and BH2 from the N to C termini, and Bcl-2 family members can be categorized into three subfamilies based on their structural and functional characteristics (23Willis S.N. Adams J.M. Curr. Opin. Cell Biol. 2005; 17: 617-625Crossref PubMed Scopus (641) Google Scholar, 26Lanave C. Santamaria M. Saccone C. Gene (Amst.). 2004; 333: 71-79Crossref PubMed Scopus (69) Google Scholar, 27Tsujimoto Y. J. Cell. Physiol. 2003; 195: 158-167Crossref PubMed Scopus (449) Google Scholar). The anti-apoptotic Bcl-2 subfamily members (including Mcl-1, Bcl-2, and Bcl-xL) share three to four BH domains. The proapoptotic multidomain Bcl-2 subfamily members (including Bax and Bak) share more than one BH domain. Activation of Bax or Bak can cause mitochondrial cytochrome c release and apoptosis (28Griffiths G.J. Dubrez L. Morgan C.P. Jones N.A. Whitehouse J. Corfe B.M. Dive C. Hickman J.A. J. Cell Biol. 1999; 144: 903-914Crossref PubMed Scopus (394) Google Scholar, 29Heimlich G. McKinnon A.D. Bernardo K. Brdiczka D. Reed J.C. Kain R. Kronke M. Jurgensmeier J.M. Biochem. J. 2004; 378: 247-255Crossref PubMed Scopus (87) Google Scholar, 30Nechushtan A. Smith C.L. Lamensdorf I. Yoon S.H. Youle R.J. J. Cell Biol. 2001; 153: 1265-1276Crossref PubMed Scopus (403) Google Scholar, 31Smaili S.S. Hsu Y.T. Sanders K.M. Russell J.T. Youle R.J. Cell Death Differ. 2001; 8: 909-920Crossref PubMed Scopus (164) Google Scholar). The pro-apoptotic BH3-only Bcl-2 subfamily members (including Bid, Bad, Bik, Puma, Bim, Bmf, Noxa, and Hrk) share only the BH3 domain (32Bouillet P. Strasser A. J. Cell Sci. 2002; 115: 1567-1574Crossref PubMed Google Scholar). How these Bcl-2 family members interact with each other to regulate mitochondrial cytochrome c release is not entirely clear. It is thought that the BH3-only proteins that are normally associated with intracellular organelles can sense cell death signals from either extrinsic (e.g. Bid) or intrinsic (e.g. Bim, Bmf, Puma, and Bik) sources by undergoing transcriptional and/or post-translational changes and translocating to mitochondria (23Willis S.N. Adams J.M. Curr. Opin. Cell Biol. 2005; 17: 617-625Crossref PubMed Scopus (641) Google Scholar, 27Tsujimoto Y. J. Cell. Physiol. 2003; 195: 158-167Crossref PubMed Scopus (449) Google Scholar, 32Bouillet P. Strasser A. J. Cell Sci. 2002; 115: 1567-1574Crossref PubMed Google Scholar). These BH3-only proteins can transmit death signals to mitochondria by inhibiting the anti-apoptotic Bcl-2 subfamily members and/or activating the pro-apoptotic multidomain Bcl-2 family members such as Bax and Bak (23Willis S.N. Adams J.M. Curr. Opin. Cell Biol. 2005; 17: 617-625Crossref PubMed Scopus (641) Google Scholar, 32Bouillet P. Strasser A. J. Cell Sci. 2002; 115: 1567-1574Crossref PubMed Google Scholar). To understand the mechanisms of the chlamydial anti-apoptotic activity, recent studies have correlated the chlamydial anti-apoptotic activity with blockade of mitochondrial cytochrome c release (8Fan T. Lu H. Hu H. Shi L. McClarty G.A. Nance D.M. Greenberg A.H. Zhong G. J. Exp. Med. 1998; 187: 487-496Crossref PubMed Scopus (493) Google Scholar, 33Fischer S.F. Harlander T. Vier J. Hacker G. Infect. Immun. 2004; 72: 1107-1115Crossref PubMed Scopus (53) Google Scholar), inhibition of pro-apoptotic Bax and Bak activation (34Xiao Y. Zhong Y. Greene W. Dong F. Zhong G. Infect. Immun. 2004; 72: 5470-5474Crossref PubMed Scopus (69) Google Scholar), and degradation of BH3-only proteins (35Dong F. Pirbhai M. Xiao Y. Zhong Y. Wu Y. Zhong G. Infect. Immun. 2005; 73: 1861-1864Crossref PubMed Scopus (88) Google Scholar, 36Ying S. Seiffert B.M. Hacker G. Fischer S.F. Infect. Immun. 2005; 73: 1399-1403Crossref PubMed Scopus (57) Google Scholar, 37Fischer S.F. Vier J. Kirschnek S. Klos A. Hess S. Ying S. Hacker G. J. Exp. Med. 2004; 200: 905-916Crossref PubMed Scopus (162) Google Scholar). However, it is still not known how the host pro-apoptotic BH3-only proteins are degraded in C. trachomatis-infected cells. In this study, we demonstrate that CPAF, a chlamydia-secreted protease factor that has been shown previously to degrade host transcription factors (9Zhong G. Fan T. Liu L. J. Exp. Med. 1999; 189: 1931-1938Crossref PubMed Scopus (175) Google Scholar, 10Zhong G. Liu L. Fan T. Fan P. Ji H. J. Exp. Med. 2000; 191: 1525-1534Crossref PubMed Scopus (171) Google Scholar, 11Zhong G. Fan P. Ji H. Dong F. Huang Y. J. Exp. Med. 2001; 193: 935-942Crossref PubMed Scopus (321) Google Scholar) and cytokeratin 8 (13Dong F. Su H. Huang Y. Zhong Y. Zhong G. Infect. Immun. 2004; 72: 3863-3868Crossref PubMed Scopus (77) Google Scholar), is also responsible for degrading the BH3-only proteins. Cell Culture and Chlamydial Infection—HeLa cells (American Type Culture Collection, Manassas, VA) were grown in a growth medium consisting of Dulbecco’s modified Eagle’s medium (Invitrogen) and 10% fetal calf serum (Novo-Tech, Grand Island, NY) in a humidified incubator supplied with 5% CO2. C. trachomatis LGV2 (L2) was grown and purified as described previously (38Xiao Y. Zhong Y. Su H. Zhou Z. Chiao P. Zhong G. J. Immunol. 2005; 174: 1701-1708Crossref PubMed Scopus (48) Google Scholar) and used to infect HeLa cells at a multiplicity of infection of 5 or as indicated in individual experiments. The infection was carried out by directly diluting the stock organisms in the growth medium. The infected cultures were harvested ∼40 h after infection or as indicated in individual experiments. For the inhibition experiments, the inhibitors lactacystin (an irreversible proteasomal inhibitor known to inhibit CPAF; Chemicon, Temecula, CA) and Ala-Ala-Phe-chloromethyl ketone (a cell-permeable serine protease inhibitor; BIOMOL, Plymouth Meeting, PA) were added to the cultures during chlamydial infection. Column Chromatography—To correlate CPAF proteolytic activity with BH3-only protein degradation, a cytosolic protein preparation from chlamydia-infected HeLa cells (L2S100) (11Zhong G. Fan P. Ji H. Dong F. Huang Y. J. Exp. Med. 2001; 193: 935-942Crossref PubMed Scopus (321) Google Scholar) was subjected to fractionation on a Mono Q ion exchange column (Amersham Biosciences). An ÅKTA Model 10 purifier (GE Healthcare) was used to run the column. A salt gradient was used to elute the Mono Q column. The eluted fractions were monitored for the presence of both CPAF and host proteasomes by Western blotting (see below), and CPAF proteolytic activity and BH3-only protein degradation were measured using a cell-free cleavage assay (see below). Cell-free Cleavage Assay—The cell-free assays were carried out as described previously (11Zhong G. Fan P. Ji H. Dong F. Huang Y. J. Exp. Med. 2001; 193: 935-942Crossref PubMed Scopus (321) Google Scholar, 13Dong F. Su H. Huang Y. Zhong Y. Zhong G. Infect. Immun. 2004; 72: 3863-3868Crossref PubMed Scopus (77) Google Scholar). The enzyme preparations or purified enzymes were mixed with the desired substrate in the presence or absence of the inhibitor lactacystin or the solvent Me2SO alone (Sigma). All reactions were carried out in phosphate-buffered saline at 37 °C for 1 h or as indicated in individual experiments. The enzyme sources included the cytosolic S100 samples made from HeLa cells with (L2S100) or without (HeLaS100) chlamydial infection, the column-fractionated samples, the remaining supernatants after antibody depletion, the antibody-precipitated pellets, and the various versions of recombinant glutathione S-transferase (GST)-CPAF created as described previously (14Dong F. Sharma J. Xiao Y. Zhong Y. Zhong G. Infect. Immun. 2004; 72: 3869-3875Crossref PubMed Scopus (39) Google Scholar, 15Dong F. Pirbhai M. Zhong Y. Zhong G. Mol. Microbiol. 2004; 52: 1487-1494Crossref PubMed Scopus (44) Google Scholar). The substrate preparations included crude nuclear extracts containing RFX5 and cytosolic extracts containing keratin 8 and various BH3-only proteins as well as a purified recombinant GST-human Pumaα fusion protein (see below). The nuclear extract was made as described previously (10Zhong G. Liu L. Fan T. Fan P. Ji H. J. Exp. Med. 2000; 191: 1525-1534Crossref PubMed Scopus (171) Google Scholar). Briefly, normal HeLa cells were homogenized in a Dounce homogenizer to break cytoplasmic membranes, and pellets were repeatedly washed with a buffer consisting of 1% Nonidet P-40 and 150 mm NaCl in 50 mm Tris (pH 8.0) plus a protease inhibitor mixture (phenylmethylsulfonyl fluoride at a final concentration of 1 mm, leupeptin at 20 μm, pepstatin A at 1.6 μm, and aprotinin at 1.7 μg/ml; all from Sigma) to remove cytosolic/membrane proteins as much as possible. The final washed nuclear pellets were extracted with a buffer consisting of 0.5 m NaCl and 1% Triton X-100 in 20 mm Tris (pH 8.0). The cytosolic extract was made as described previously (13Dong F. Su H. Huang Y. Zhong Y. Zhong G. Infect. Immun. 2004; 72: 3863-3868Crossref PubMed Scopus (77) Google Scholar) by extracting 1–2 × 107 HeLa cells with 1 ml of a buffer consisting of 1% Nonidet P-40, 0.5% Triton X-100, and 150 mm NaCl in 50 mm Tris (pH 8.0) plus the protease inhibitor mixture. The GST-human Pumaα fusion was generated by cloning the human Pumaα cDNA (accession number AF354654; ncbi.nlm.nih.gov/entrez) into a pGEX-4T1 vector (kindly provided by Dr. Chun Wu, Abgent, San Diego, CA), and the GST fusion protein expressed in bacteria was purified with glutathione-agarose beads (Amersham Biosciences) following the manufacturer’s instructions. Western Blot Assay—Western blot assays were carried out as described previously (39Zhong G. Reis e Sousa C. Germain R.N. Proc. Natl. Acad. Sci. U. S. A. 1997; 94: 13856-13861Crossref PubMed Scopus (132) Google Scholar). All samples for Western blot assays were separated by SDS-PAGE, and the separated proteins were transferred onto a nitrocellulose membrane for immunostaining. The primary antibodies used for Western blotting included mouse monoclonal antibody (mAb) M20 (which recognizes keratin 8; IgG1; Sigma); mAb MCP21 (specific to host 20 S proteasome α-subunit HC3; IgG1; AFFINITI Research Products Ltd., Mamhead Castle, UK); mAb 54b (which recognizes the CPAF N terminus; IgG1) (11Zhong G. Fan P. Ji H. Dong F. Huang Y. J. Exp. Med. 2001; 193: 935-942Crossref PubMed Scopus (321) Google Scholar); and rabbit polyclonal antibodies against RFX5 (Rockland Immunochemicals, Inc., Gilbertsville, PA) (10Zhong G. Liu L. Fan T. Fan P. Ji H. J. Exp. Med. 2000; 191: 1525-1534Crossref PubMed Scopus (171) Google Scholar), BimL (catalog number sc-11425, Santa Cruz Biotechnology), Bax (catalog number sc-493, Santa Cruz Biotechnology, Inc.), Bik (catalog number 4592, Cell Signaling Technology, Inc., Beverly, MA), Pumaα (catalog number P4618, Sigma; catalog number 3041, ProSci Inc.), and polyclonal gαpumaα (catalog number sc-1987, Santa Cruz Biotechnology). The binding of these primary antibodies was probed with the corresponding secondary antibodies conjugated to horseradish peroxidase and visualized using standard enhanced chemiluminescence (ECL, Amersham Biosciences). Immunoprecipitation Assay—The immunoprecipitation assay was carried out as described previously (40Zhong G. Castellino F. Romagnoli P. Germain R.N. J. Exp. Med. 1996; 184: 2061-2066Crossref PubMed Scopus (71) Google Scholar). For this study, the assay was used to deplete antigens from extracts with specific antibodies. L2S100 samples were mixed with protein A/G-agarose-immobilized antibody complexes. After a 1-h incubation at room temperature, the agarose pellets were spun down, and the remaining supernatants were re-precipitated with a fresh set of the corresponding antibody complexes to completely remove the corresponding antigens in the supernatants. The final remaining supernatants and the precipitates were used as the source of enzymes in the cell-free degradation assay. mAb 54b, mAb MC22 (IgG3) (11Zhong G. Fan P. Ji H. Dong F. Huang Y. J. Exp. Med. 2001; 193: 935-942Crossref PubMed Scopus (321) Google Scholar), and mAb MCP21 were used to precipitate CPAF, the chlamydial major outer membrane protein, and host proteasome α-subunit HC3, respectively. Correlation of BH3-only Protein Degradation with CPAF Activity—To test whether a previously identified chlamydial protease factor, CPAF, is involved in the degradation of the pro-apoptotic BH3-only proteins in chlamydia-infected cells, we used a column chromatography approach to analyze the relationship between these two (Fig. 1). The C. trachomatis-infected cell cytosolic extracts (L2S100) were subjected to fractionation via Mono Q column chromatography because it is known that CPAF can both bind to and be eluted from a Mono Q column (10Zhong G. Liu L. Fan T. Fan P. Ji H. J. Exp. Med. 2000; 191: 1525-1534Crossref PubMed Scopus (171) Google Scholar, 11Zhong G. Fan P. Ji H. Dong F. Huang Y. J. Exp. Med. 2001; 193: 935-942Crossref PubMed Scopus (321) Google Scholar). The elution peaks of both CPAF protein and CPAF activity (measured as RFX5 degradation) overlapped with those of BH3-only protein degradation, including the degradation of Pumaα, Pumaβ, Bik, and BimL, whereas the elution peak of host proteasomes was delayed, demonstrating a correlation between BH3-only protein degradation and CPAF, but not host proteasomes. However, this experiment could not determine whether CPAF indeed participates in the degradation of the host BH3-only proteins. CPAF Is Required for BH3-only Protein Degradation in C. trachomatis-infected Cells—To test whether CPAF is necessary for degrading the BH3-only proteins, we first tested whether lactacystin, an irreversible proteasomal inhibitor that is known to inhibit CPAF activity, is able to block BH3-only protein degradation activity in the C. trachomatis-infected cell cytosol (Fig. 2A). When a small amount of the C. trachomatis-infected cell cytosol sample (L2S100) was used as the source of enzyme, a partial degradation of Pumaα and Pumaβ (lane 3) was observed, whereas a large amount caused complete disappearance of Pumaα and Pumaβ (lane 4). However, addition of lactacystin (but not the solvent Me2SO alone) to the reaction mixture even containing the large amount of L2S100 restored most of Pumaα and Pumaβ (lanes 5 and 6), demonstrating that lactacystin-sensitive proteolytic activity is mainly responsible for the degradation of the BH3-only proteins. We next used a depletion experimental approach to further evaluate the necessity of CPAF to degrade BH3-only proteins (Fig. 2B). Intact L2S100 degraded both Pumaα and Pumaβ (lanes 2 and 3). However, the residual supernatant after precipitation with mAb 54b-conjugated beads lost the degradation activity (lane 4), whereas the antibody pellet retained the degradation activity (lane 5). Because of the co-migration of the immunoglobulin light chain and Pumaα, only the Pumaβ degradation was clearly measurable by Western blotting. Precipitation with antibody recognizing either host proteasomes (mAb MCP21) or the chlamydial major outer membrane protein (mAb MC22) failed to remove the degradation activity from the L2S100 supernatants. The low level of degradation activity associated with the control antibody pellets may be due to the nonspecific binding of CPAF to the beads. CPAF Is Sufficient for BH3-only Protein Degradation in C. trachomatis-infected Cells—The next question we asked is whether CPAF alone is able to degrade the BH3-only proteins. We first used the CPAF fusion protein purified from bacteria as the source of enzyme to carry out the cell-free degradation assay (Fig. 3A). The GST-CPAF fusion protein completely degraded both Pumaα and Pumaβ, but left a degradation fragment (lane 4). The residual degradation fragment was not uniquely generated by the GST-CPAF fusion protein, but represented an intermediate fragment of Puma digested by both GST-CPAF and endogenous CPAF. This is because the degradation fragment left in the GST-CPAF digestion mixture was equivalent to the longer degradation fragment generated by endogenous CPAF when used at a low concentration (Fig. 2A, lane 3; and Fig. 3B, lane 4). More important, the GST-CPAF-mediated degradation of Puma was completely inhibited by lactacystin, but not by Me2SO (Fig. 3A, lanes 5 and 6), suggesting that a CPAF-like activity in the GST-CPAF fusion protein was responsible for degrading Puma. To exclude the possible involvement of any contaminating proteolytic activity from the bacteria in the degradation of the host BH3-only proteins, we compared the degradation activity of wild-type GST-CPAF with that of two mutant GST-CPAF fusion proteins (GST-CPAF(YVAD) and GST-CPAF(L281G)). Both GST-CPAF(YVAD) and GST-CPAF(L281G)) have been shown previously to lack CPAF activity (14Dong F. Sharma J. Xiao Y. Zhong Y. Zhong G. Infect. Immun. 2004; 72: 3869-3875Crossref PubMed Scopus (39) Google Scholar, 15Dong F. Pirbhai M. Zhong Y. Zhong G. Mol. Microbiol. 2004; 52: 1487-1494Crossref PubMed Scopus (44) Google Scholar). The three GST fusion proteins were immobilized on beads and monitored for total protein amount on a Coomassie Blue-stained gel. All three fusion proteins were loaded onto the gel in three different bead volumes, and protein staining revealed that the three GST fusion proteins were present in equivalent amounts (Fig. 4A). A duplicate set of bead aliquots was used as the source of enzyme to degrade the BH3-only proteins in a cell-free degradation assay (Fig. 4B). Because cytokeratin 8 (keratin 8) is a known substrate of CPAF (13Dong F. Su H. Huang Y. Zhong Y. Zhong G. Infect. Immun. 2004; 72: 3863-3868Crossref PubMed Scopus (77) Google Scholar), it was used as a positive control. Wild-type GST-CPAF displayed some degradation activity for both Puma and Bik when 8 μl of beads was used and strong degradation activity when 32 μl of beads was used. However, neither of the mutant GST-CPAF proteins showed any degradation activity regardless of the bead volumes used. Because all three GST-CPAF fusion proteins were similarly purified from bacterial cultures, this experiment demonstrated that it is unlikely that bacterial contaminants contributed to BH3-only protein degradation in any significant way. In all of the above experiments, a HeLa cell cytosolic extract was used as the source of substrate containing not only BH3-only proteins, but also many other cytosolic proteins from HeLa cells. It is unlikely that the cytosolic extract proteins contributed to BH3-only protein degradation because, in all of the above experiments, incubation of the cytosolic extract alone (Fig. 3A, lane 1) or with a HeLaS100 preparation (lane 2) failed to degrade the BH3-only proteins in the cytosolic extract. However, to further exclude the potential contribution of the cytosolic extract proteins to the CPAF-mediated BH3-only protein degradation, we next used Pumaα purified from bacteria in the form of a GST fusion protein as the substrate in a cell degradation assay (Fig. 5). The purified full-length GST-Pumaα fusion protein was detected both on a Coomassie Blue-stained gel (Fig. 5A, left panel) and on a Western blot with a Puma-specific antibody (right panel). When the purified GST-Pumaα fusion protein was used in the degradation assay (Fig. 5B), the fusion protein was degraded by L2S100 (lane 2), but not by HeLaS100 (lane 3), suggesting that the fusion protein maintained the recognition site for CPAF. When GST-CPAF was used as the source of enzyme, GST-Pumaα was degraded in a dose-dependent manner. Notably, the GST-CPAF-mediated degradation of GST-Pumaα was sensitive to lactacystin inhibition, further confirming that CPAF activity was responsible for degrading GST-Pumaα. Finally, an inhibition experiment was carried out to demonstrate that the CPAF-mediated degradation of Puma occurs during chlamydial infection (Fig. 6). Because lactacystin is the only known inhibitor that can inhibit CPAF activity and is also cell-permeable, we added lactacystin to cell cultures during chlamydial infection and monitored the level of Puma protein. Although Puma was degraded in chlamydia-infected cells (lane 4), lactacystin treatment completely blocked the Puma degradation induced by chlamydial infection (lane 5), and a control protease inhibitor (Ala-Ala-Phe-chloromethyl ketone) failed to do so (lane 6). It has been shown previously that degradation of the pro-apoptotic BH3-only proteins contributes significantly to the chlamydial anti-apoptotic activity (35Dong F. Pirbhai M. Xiao Y. Zhong Y. Wu Y. Zhong G. Infect. Immun. 2005; 73: 1861-1864Crossref PubMed Scopus (88) Google Scholar, 36Ying S. Seiffert B.M. Hacker G. Fischer S.F. Infect. Immun. 2005; 73: 1399-1403Crossref PubMed Scopus (57) Google Scholar, 37Fischer S.F. Vier J. Kirschnek S. Klos A. Hess S. Ying S. Hacker G. J. Exp. Med. 2004; 200: 905-916Crossref PubMed Scopus (162) Google Scholar). Although there is still some dispute over whether all BH3-only proteins are degraded in chlamydia-infected cells, the degradation of Puma seems to be most obvious, and overexpression of Puma can overcome the chlamydial anti-apoptotic activity (37Fischer S.F. Vier J. Kirschnek S. Klos A. Hess S. Ying S. Hacker G. J. Exp. Med. 2004; 200: 905-916Crossref PubMed Scopus (162) Google Scholar). We have therefore mainly used Puma as a representative of the BH3-only Bcl-2 proteins in this study. Because we have shown previously (9Zhong G. Fan T. Liu L. J. Exp. Med. 1999; 189: 1931-1938Crossref PubMed Scopus (175) Google Scholar, 10Zhong G. Liu L. Fan T. Fan P. Ji H. J. Exp. Med. 2000; 191: 1525-1534Crossref PubMed Scopus (171) Google Scholar, 11Zhong G. Fan P. Ji H. Dong F. Huang Y. J. Exp. Med. 2001; 193: 935-942Crossref PubMed Scopus (321) Google Scholar) that CPAF, a chlamydia-secreted protease factor, can degrade various host molecules, we have further investigated the potential role of CPAF in the degradation of host BH3-only proteins. The following observations have led us to conclude that CPAF is responsible for the degradation of pro-apoptotic BH3-only proteins. First, CPAF is secreted into the host cell cytosol (11Zhong G. Fan P. Ji H. Dong F. Huang Y. J. Exp. Med. 2001; 193: 935-942Crossref PubMed Scopus (321) Google Scholar), where the pro-apoptotic BH3-only proteins are located. The time course of CPAF-mediated RFX5 (10Zhong G. Liu L. Fan T. Fan P. Ji H. J. Exp. Med. 2000; 191: 1525-1534Crossref PubMed Scopus (171) Google Scholar) and keratin 8 (13Dong F. Su H. Huang Y. Zhong Y. Zhong G. Infect. Immun. 2004; 72: 3863-3868Crossref PubMed Scopus (77) Google Scholar) degradation also parallels that of BH3-only protein degradation (35Dong F. Pirbhai M. Xiao Y. Zhong Y. Wu Y. Zhong G. Infect. Immun. 2005; 73: 1861-1864Crossref PubMed Scopus (88) Google Scholar, 36Ying S. Seiffert B.M. Hacker G. Fischer S.F. Infect. Immun. 2005; 73: 1399-1403Crossref PubMed Scopus (57) Google Scholar, 37Fischer S.F. Vier J. Kirschnek S. Klos A. Hess S. Ying S. Hacker G. J. Exp. Med. 2004; 200: 905-916Crossref PubMed Scopus (162) Google Scholar) in chlamydia-infected cells. Furthermore, CPAF activity in the chlamydia-infected cell cytosol co-eluted from the Mono Q column with the BH3-only protein degradation activity (Fig. 1). Second, in an antibody-mediated depletion experiment, only the CPAF-specific antibody was able to pull the BH3-only degradation activity from the supernatant to the antibody pellet, whereas other control antibodies failed to do so (Fig. 2), demonstrating the necessity of CPAF for the degradation of the BH3-only proteins. Third, a CPAF fusion protein expressed in bacteria was able to degrade the BH3-only proteins, whereas the lack-of-function mutant CPAF fusion proteins (15Dong F. Pirbhai M. Zhong Y. Zhong G. Mol. Microbiol. 2004; 52: 1487-1494Crossref PubMed Scopus (44) Google Scholar) similarly prepared from bacteria failed to do so (Fig. 4). Fourth, bacterium-expressed CPAF also degraded the human BH3-only protein Pumaα purified from a bacterial expression system (Fig. 5). Finally, lactacystin, a known inhibitor of CPAF, prevented Puma degradation during chlamydial infection. These results demonstrate that CPAF is both necessary and sufficient for degrading host BH3 proteins. Chlamydia may be selected to target the BH3-only proteins for inactivating host apoptosis. The pro-apoptotic BH3-only proteins are normally sequestrated from their pro-apoptotic activity by associating with various cytoplasmic organelles and functioning as the silent sensors for intracellular stress signals. The chlamydial inclusion has to expand during chlamydial intracellular growth, and the inclusion expansion inevitably alters the intracellular organization and disrupts the normal distribution of intracellular organelles, which may set off the BH3-only proteins to deliver death/stress signals to mitochondria. It is known that host cell apoptosis is an effector mechanism that controls intracellular infection, and apoptosis of the infected cells during the reticulate body stage may lead to the termination of chlamydial infection. Such a strong selection pressure may allow only the chlamydial organisms able to prevent host cell apoptosis to complete intracellular growth. By the time the inclusion expansion starts to disturb the intracellular environment, CPAF is already secreted into and accumulated in the host cell cytosol. Indeed, at 16–24 h after infection, obvious BH3-only protein degradation is observed (35Dong F. Pirbhai M. Xiao Y. Zhong Y. Wu Y. Zhong G. Infect. Immun. 2005; 73: 1861-1864Crossref PubMed Scopus (88) Google Scholar, 36Ying S. Seiffert B.M. Hacker G. Fischer S.F. Infect. Immun. 2005; 73: 1399-1403Crossref PubMed Scopus (57) Google Scholar, 37Fischer S.F. Vier J. Kirschnek S. Klos A. Hess S. Ying S. Hacker G. J. Exp. Med. 2004; 200: 905-916Crossref PubMed Scopus (162) Google Scholar). Efforts are under way to further evaluate whether CPAF alone in the absence of chlamydial infection can degrade the BH3-only proteins inside cells and prevent host cells from undergoing apoptosis. The challenge for this experiment is the difficulty in expressing active CPAF in mammalian cells. Because CPAF is synthesized as a proenzyme and has to be processed into an intramolecular dimer to acquire activity (14Dong F. Sharma J. Xiao Y. Zhong Y. Zhong G. Infect. Immun. 2004; 72: 3869-3875Crossref PubMed Scopus (39) Google Scholar, 15Dong F. Pirbhai M. Zhong Y. Zhong G. Mol. Microbiol. 2004; 52: 1487-1494Crossref PubMed Scopus (44) Google Scholar), it has been difficult to generate constructs coding for active CPAF in mammalian cells. CPAF is known to degrade the transcription factors RFX5 and USF-1 and cytokeratin 8. This study has added the BH3-only Bcl-2 family members to the list of CPAF substrates. An obvious question is what determines the substrate specificity of CPAF. When these known substrates are aligned, no common primary amino acid sequence motifs can be identified. We are in the process of determining CPAF substrate specificities using various biochemical approaches, which may potentially allow us to predict new cellular targets of CPAF.