Apolipoprotein L6, a Novel Proapoptotic Bcl-2 Homology 3–Only Protein, Induces Mitochondria-Mediated Apoptosis in Cancer Cells

Zhihe Liu,1 Huimei Lu,1 Zeyu Jiang,2 Andrzej Pastuszyn,1 and Chien-an A. Hu1

1Department of Biochemistry and Molecular Biology and 2Division of Biocomputing, University of New Mexico School of Medicine, Albuquerque, New Mexico

Abstract Introduction Cancer cells frequently possess defects in the genetic Apoptosis is a complex and highly regulated cell death and biochemical pathways of apoptosis. Members of the process that can be distinguished by cellular and biochemical Bcl-2 family play pivotal roles in regulating apoptosis hallmarks, including release of apoptogenic factors, activation and possess at least one of four Bcl-2 homology (BH) of caspases, chromatin condensation, and membrane blebbing. domains, designated BH1 to BH4. The BH3 domain is This cell death pathway is used by multicellular organisms to the only one conserved in proapoptotic BH3-only eliminate unwanted or injured cells and is critically important proteins and plays an important role in protein-protein for maintaining homeostasis during development and through- interactions in apoptosis by regulating homodimerization out adulthood in animals (1-4). Dysregulation of apoptosis is and heterodimerization of the Bcl-2 family members. evident in many human diseases, including cancer (5) and To date, 10 BH3-only proapoptotic proteins have been neurodegenerative disorders (6). identified and characterized in the human genome. Importantly, in mammals, there are at least three distinct but The completion of the Human Genome Project and the interactive and interconnected apoptotic pathways: mitochon- availability of various public databases and sequence dria-mediated, death receptor–initiated, and endoplasmic retic- analysis algorithms allowed us to use the bioinformatic ulum stress-mediated pathways (1, 2, 7). Activation of caspases, database-mining approach to identify one novel a group of cysteine-activated aspartate-specific proteases, is a BH3-only protein, apolipoprotein L6 (ApoL6). The common characteristic of all three apoptotic pathways (8, 9). In full-length cDNA of ApoL6 was identified, cloned, and general, activation of caspase-9 is associated with mitochondria- functionally expressed in p53-null colorectal cancer mediated apoptosis, whereas activation of caspase-8 frequently cells (DLD-1). We found that overexpression of wild-type mediates death receptor–regulated apoptosis. A connection ApoL6 induced mitochondria-mediated apoptosis in between death receptor–initiated and mitochondria-mediated DLD-1 cells characterized by release of cytochrome c pathways is made by caspase-8-cleaved and caspase-8-activated and Smac/DIABLO from mitochondria and activation of Bid, a proapoptotic member of the Bcl-2 family (1, 8, 9). The caspase-9, whereas ApoL6 BH3 domain deletion allele Bcl-2 family members play pivotal roles in regulating and did not. In addition, overexpression of ApoL6 also interconnecting all three apoptotic pathways in which they induced activation of caspase-8. Furthermore, we can either promote cell survival (antiapoptotic; e.g., Bcl-2, and showed that adenovirus harboring the full-length cDNA Bcl-xL) or induce cell death (proapoptotic; e.g., Bax, Bad, Bid, of ApoL6 induced marked apoptosis in a variety of Noxa, and PUMA). It is well recognized that cancer cells often cancer cell types, and ApoL6 recruited and interacted show altered ratios of antiapoptotic members to proapoptotic with lipid/fatty acid components during the induction members of the Bcl-2 family. Disruption of homeostasis of the of apoptosis. To our knowledge, this is the first Bcl-2 family members promotes survival and confers resistance example that intracellular overproduction of an to therapy in cancer (5, 10, 11). Each member of the Bcl-2 apolipoprotein induces marked apoptosis. (Mol Cancer protein family contains at least one to four conserved protein Res 2005;3(1):21–31) sequences, Bcl-2 homology (BH) domains designated BH1 to BH4 (10-12). Proapoptotic ‘‘BH3-only’’ proteins contain only the BH3 homology sequence attesting to the functional significance of this domain. It has been shown that the BH3-

Received 9/20/04; revised 12/6/04; accepted 12/8/04. only proteins play important roles in protein-protein interac- Grant support: Howard Hughes Medical Institute research aids to University of tions, regulation of dimerization, and protein-permeable pore New Mexico Cancer Research and Treatment Center, American Cancer Society formation of the Bcl-2 family members in apoptosis (10-12). ACS-IRG-192 grant 412488-00095, and University of New Mexico Research Allocation Committee grant C-2222-RAC (C-A.A. Hu). BH3-only proteins function either as sensors or as second The costs of publication of this article were defrayed in part by the payment of messengers of discrete apoptotic stimuli and can interact with page charges. This article must therefore be hereby marked advertisement in the multidomain Bcl-2 family proteins to either antagonize or accordance with 18 U.S.C. Section 1734 solely to indicate this fact. Requests for reprints: Chien-an A. Hu, Department of Biochemistry activate their function in apoptosis (10, 11, 13). They have been and Molecular Biology, MSC08 4670, 1 University of New Mexico, shown to induce apoptosis by binding to antiapoptotic Bcl-2 Albuquerque, NM 87131-0001. Phone: 505-272-8816; Fax: 505-272-6587. E-mail: [email protected] family members to inhibit their activity (10, 14) and other Copyright D 2005 American Association for Cancer Research. proapoptotic family members to enhance their activity (15, 16).

It is generally accepted that BH3-only proteins require the Moreover, we show that ApoL6 induces a BH3-dependent multidomain proapoptotic Bcl-2 family members to induce cell apoptosis and interacts with lipid/fatty acid components during death (7, 16, 17). Two lines of evidence indicate that BH3 the induction of apoptosis, suggesting its role in connecting domain–mediated protein-protein interactions are determinants lipid second messengers and cell death. of proapoptotic activity of BH3-only proteins. First, deletion alleles of the BH3 domain of some of the BH3-only proteins (e.g., PUMA and Bad) failed to induce apoptosis (18, 19). Results Second, peptides derived from BH3 domains of the BH3-only Employing a database search of open reading frame based proteins can induce apoptosis by binding to the hydrophobic on 9-residue consensus sequence constructed from 17 known pocket of the antiapoptotic proteins or by ‘‘sensitizing’’ cells to human Bcl-2 protein sequences (Fig. 1), we identified a apoptotic stimuli (14, 19). In addition, BH3-only proteins have BH3 domain embedded with the ApoL6 gene (Genbank been shown to interact with intrinsic mitochondrial membrane accession no. NM_030641). ApoL proteins belong to a newly proteins, such as the voltage-dependent anion channel or the identified, high-density lipoprotein family composed of six adenine nucleotide exchanger, to induce mitochondrial dys- members (L1-L6) that share significant sequence identity function and cytochrome c release (13, 20, 21). Importantly, within the predicted amphipathic a-helix. Evidence suggested several genes encoding BH3-only proteins (e.g., Bid, Noxa, and that these proteins play a central role in cholesterol transport PUMA) are p53 downstream targets in p53-induced apoptosis (24-29). Previous studies showed that genes encoding ApoL are and are expressed in overlapping as well as unique and specific localized to human chromosome 22q12-13 (24-26). A compar- tissues and cells (3, 10-12, 22). Gene knockout studies in mice ison of gene expression at the RNA level with ApoA-I, the have shown unique, specific roles for various Bcl-2 family major structural component of high-density lipoprotein, sug- genes in regulating cell proliferation and death in specific gests that the ApoL proteins play general and fundamental roles tissues/cells or under particular developmental, physiologic, or in lipid biochemistry (26, 27). Previous studies also showed that pathologic circumstances (16, 17, 23). Finally, the various ApoL6 is a widely expressed lipid binding protein (refs. 24-29; members of the Bcl-2 family are localized in different also in GeneCards database). We found that full-length ApoL6 subcellular compartments, thereby indicating their nonredun- cDNA could be successfully amplified from a variety of human dant roles in physiologic and pathologic states (10-12, 22, 23). cell lines, including DLD-1 (colorectal), HEK 293 (human Therefore, it is important to identify all members of the Bcl-2 embryonic kidney), and DU-145 (prostate) cells (data not family, particularly the BH3-only proteins, and to delineate the shown). Previous studies have predicted that extracellular cellular and biochemical contexts in which they contribute to ApoL6 is a protein component of high-density lipoprotein, the regulation of apoptosis. Currently, 10 BH3-only proapop- whereas intracellular ApoL6 may affect the movement of lipids totic proteins have been identified (7, 10-12, 14). Through or allow the binding of lipids to organelles (27, 30, 31). Human a database-mining approach, we identified and characterized ApoL6 encodes a 343–amino acid protein and has a calculated apolipoprotein L6 (ApoL6) as a novel BH3-only protein. molecular mass of 38.1 kDa (Fig. 1A).

FIGURE 1. A. Conceptually translated polypeptide sequence of human ApoL6. Putative domains/motifs in ApoL6 include a BH3 domain (amino acids 59- 67), a N-glycosylation site (amino acids 199-201), and a leucine zipper motif (amino acids 271- 292) and are in bold, italicized and underlined. B. Alignment and comparison of the BH3 domain in ApoL6 and other BH3-only proteins in humans. Two amino acids, L and D, shown in bold, are iden- tical and functionally conserved in all BH3 domains identified thus far. C. Schematic presentation of three different ApoL6 constructs: ApoL6, full-length cDNA of ApoL6; ApoL6.V5, COOH-terminal V5 epitope-tagged full-length cDNA of ApoL6; and ApoL6.dBH3, ApoL6 without the BH3 domain.

Because ApoL6 is a lipid binding protein possessing a BH3 Apoptosis Assays in DLD-1.ApoL6.V5 Cells by Flow domain and lipid second messengers play important roles in Cytometry initiating apoptotic pathways, we investigated intracellular To confirm and extend observations regarding cell death, ApoL6 functions in apoptosis. We first analyzed putative apoptosis was determined in induced DLD-1.ApoL6.V5 cells motifs encoded with ApoL6. Results from analysis using the using propidium iodide staining in combination with flow Protein Family database (32) and PROSITE program (33) cytometry. A significant increase (>11-fold) of the sub-G1 revealed, in addition to the putative BH3 domain (amino acids population in the induced DLD-1.ApoL6.V5 cells was 60-68), two other putative domains in ApoL6: a leucine zipper observed 6 hours after induction, indicating increased apoptosis domain (amino acids 271-292) and a N-glycosylation site (Fig. 3A). Induction of apoptosis in DLD-1.ApoL6.V5 cells (amino acids 199-201; Fig. 1A). Leucine zipper sequences play was time course dependent. As shown in Fig. 3B-D, the effect important roles in protein-protein interactions, whereas protein was apparent at 6 hours and maximal at 24 hours. The glycosylation is critical for function and stability of secreted percentage of cells that were induced to undergo apoptosis is and membrane proteins. determined by subtracting the percentage of apoptotic cells in the uninduced population; we found a >30-fold increase of Overexpression of ApoL6 Induces Cell Death in DLD-1 apoptosis 24 hours after initiation of induction (Fig. 3D). Cells To examine whether ApoL6 may play a role in apoptosis and Release of Cytochrome c and Smac/DIABLO in Apopto- if the BH3 domain of ApoL6 was important in inducing tic DLD-1.ApoL6.V5 Cells apoptosis, we tested the effect of expressing ApoL6 in ‘‘Tet- Release of Smac/DIABLO and cytochrome c from inter- Off’’-inducible DLD-1.TA14 cells, which are p53-null co- mitochondrial space to cytosol is a critical step in triggering lorectal cancer cells suitable for gene-induced apoptosis caspase activation and downstream apoptosis (7, 12). We studies (18, 34, 35). Stable transfection of inducible DLD-1 determined therefore whether we could detect Smac/DIABLO cell lines harboring full-length cDNA of wild-type ApoL6 or cytochrome c after overexpression of ApoL6 by immuno- (DLD-1.ApoL6), COOH-terminal V5 epitope-tagged ApoL6 fluorescent microscopy and immunoblotting analyses. As (DLD-1.ApoL6.V5), or ApoL6 BH3 domain deletion (DLD- shown in Fig. 4A, ApoL6 induced release of Smac/DIABLO 1.ApoL6.dBH3) were generated (Fig. 1C), cultured in the basal from mitochondria to cytosol 6 hours after the induction could D.20 medium containing 20 ng/mL doxycycline, and harvested be detected by a diffusely cytoplasmic distribution of Smac/ at hourly intervals following transfer to medium without DIABLO immunoactivity. As expected, the staining pattern of doxycycline (D.0). Semiquantitative reverse transcription-PCR cytochrome c was consistent with that of Smac/DIABLO (Fig. showed that expression of ApoL6.V5 and ApoL6.dBH3 was 4B). To further show the cytosolic localization of Smac/ f increased 4-fold 3 and 6 hours after the removal of DIABLO and cytochrome c in the apoptotic DLD-1.ApoL6.V5 doxycycline (Fig. 2A). ApoL6 transcripts were also detectable cells, immunoblotting analysis of cytosolic fractions was in the DLD-1.TA14 control cells, thereby indicating that DLD-1 conducted. As shown in Fig. 4C and D, the amounts of cells have some endogenous ApoL6 expression (Fig. 2A). To Smac/DIABLO and cytochrome c proteins were markedly confirm the induction of ApoL6 expression at the protein level, elevated (>2-fold in Smac/DIABLO and >6-fold in cytochrome immunoblotting analysis of induced DLD-1.ApoL6.V5 cells c) in the cytosolic fraction of induced DLD-1.ApoL6.V5 cells was conducted using anti-V5 antibody. As shown in Fig. 2B, the compared with that of uninduced cells. Therefore, the result of amount of ApoL6.V5 protein was greatly increased (>6-fold) 3 immunoblotting analysis was consistent with that of the hours after the induction. We investigated if overexpression of ApoL6 could promote immunofluorescent staining. cell death. As shown in Fig. 2C, induction of ApoL6 in DLD- 1.ApoL6.V5 cells had a marked effect on cell proliferation by Activation of Caspase-9 and Caspase-8 significantly reducing the growth rate by 24 hours when grown Activation of caspase-9 and/or caspase-8 is a hallmark of in the absence of doxycycline. Similar results were observed in apoptosis (1.9). We therefore investigated using immunoblot- DLD-1.ApoL6 cells (data not shown). Because ApoL6 ting whether overexpression of ApoL6 induced 37-kDa expression was Tet-Off regulated by the concentration of activated caspase-9 and/or 43/41-kDa activated caspase-8. As doxycycline in the culture medium, a cell growth study was shown in Fig. 4E, cleavage of caspase-9 and caspase-8 was conducted at various concentrations of doxycycline. In the observed 6 hours after induction. Interestingly, Bid was not presence of gradually increasing doxycycline concentrations, cleaved/activated in induced DLD-1.ApoL6.V5 cells, although such as 1 ng/mL (D.1) and 5 ng/mL (D.5), cell death was caspase-8 was activated (Fig. 4E). In contrast, overexpression decreased in DLD-1.ApoL6.V5 cells, implying that the cell of ApoL6.dBH3 construct in DLD-1 cells had no effect death rate was related to the level of ApoL6 expression. In on activation of either caspase-9 or caspase-8 or Bid (data contrast, DLD-1.ApoL6.V5 cells grown in D.20 exhibited not shown). normal growth (Fig. 2D). To investigate the importance of the BH3 domain of ApoL6 in the induction of apoptosis, we assayed AD-ApoL6 Induced Apoptosis in Four Other Different Cell the overexpression of ApoL6.dBH3 after induction in D.0. As Lines shown in Fig. 2E, expression of the ApoL6 deleted of the BH3 To confirm and further investigate functions of ApoL6 and its domain did not cause cell death. This implies that ApoL6- BH3 domain in induction of apoptosis in different cells, we induced apoptosis in DLD-1 cells is BH3 domain dependent. constructed adenovirus harboring wild-type full-length ApoL6

FIGURE 2. ApoL6 induces apoptosis in DLD-1 cells. A. Semiquantitative reverse transcription-PCR assay of time-dependent induction of ApoL6 and ApoL6.dBH3 in stably transfected DLD-1 cells. B. Immunoblotting analysis of ApoL6 expression in DLD-1.ApoL6.V5 cells cultured in D.0 medium. Control, DLD-1.TA14 cells. Hours of induction are as indicated. Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) and h-actin were used as the loading control for reverse transcription-PCR and immunoblotting analysis, respectively. C. Time-dependent apoptosis in DLD-1.ApoL6.V5 cells cultured in D.0 medium. Hours of induction are as indicated. D. ApoL6 concentration-dependent apoptosis in DLD-1.ApoL6 cells. E. Growth curves of DLD-1.ApoL6.dBH3 cells cultured in D.20 and D.0. D, doxycycline concentration present in the medium; D.0, no doxycycline; D.1, 1 ng/mL; D.2, 2 ng/mL; D.5, 5 ng/mL; D.20, 20 ng/mL. Columns, mean (n = 3). Bars, SE. *, P < 0.05; **, P < 0.01.

(AD-ApoL6) and ApoL6 BH3 domain deletion construct (AD- induced DLD-1.ApoL6.V5 or DLD-1.ApoL6.dBH3 cells. The ApoL6.dBH3) to infect different cells. As shown in Fig. 5, AD- precipitated complexes were then deproteinized, trans-esteri- ApoL6 was able to induce apoptosis in HEK 293 cells 3 days fied, and subjected to gas chromatography-mass spectrometry after infection (Fig. 5B and E): cells infected with AD-ApoL6 analysis. We showed that there were fatty acids and fatty acid showed change of the cell morphology (Fig. 5B) and nuclear derivatives from the lipid components that could be detected in condensation/fragmentation (Fig. 5E). In contrast, control the ApoL6.V5-containing lipoprotein complex in apoptotic adenovirus or AD-ApoL6.dBH3 did not induce apoptosis in DLD-1.ApoL6.V5 cells (Fig. 6C). When compared with the HEK 293 cells (Fig. 5A, B, C, and E). In addition, we also standard profile (Fig. 6D), C16:0 and C18:0 fatty acids could be conducted the same experiment in other cell lines, DLD-1, identified in the complex. In contrast, although ApoL6.dBH3 HepG2 (liver cancer), MCF-7 (breast cancer), and DU-145, and interacted with lipids/fatty acids in induced DLD- detected the same results of ApoL6-dependent apoptosis (data 1.ApoL6.dBH3 cells, there were fewer and less fatty acids not shown). and fatty acid derivatives bound to ApoL6.dBH3 (Fig. 6F). Taken together, these results suggest that the BH3 domain in Binding of Fatty Acids/Lipids by ApoL6.V5 and ApoL6 affected binding of lipids of ApoL6 both quantitatively ApoL6.dBH3 and qualitatively. Binding and transporting of specific lipids by Recent studies showed that lipid second messengers play ApoL6 may be involved in ApoL6-induced apoptosis. important roles in initiating apoptosis (36-42). To investigate if ApoL6 interacted with lipids/fatty acids during its induction of mitochondria-mediated apoptosis and if the BH3 domain in Discussion ApoL6 affected binding of lipid/fatty acids of ApoL6, A database-mining strategy was used to identify a novel we immunoprecipitated lipoprotein complex containing human BH3-only protein, ApoL6, that when stably transfected ApoL6.V5 using agarose-conjugated anti-V5 antibody from induced classic mitochondria-mediated apoptotic cell death both induced and control DLD-1.ApoL6.V5 cells. We also characterized by release of cytochrome c and Smac/DIABLO conducted immunoprecipitation of the lipoprotein complex and activation of caspase-9, an initiator caspase in mediating containing ApoL6.V5 or ApoL6.dBH3 using polyclonal anti- apoptosis. In addition, overexpression of ApoL6 also induced ApoL6 antibody, agarose-conjugated protein A beads, and activation of caspase-8, another initiator caspase that functions

mainly in death receptor–mediated apoptosis. Interestingly, Bid understand how multifunctional proteins may bring together was not activated by caspase-8 in ApoL6-induced apoptosis, different signaling mechanisms to mediate induction of suggesting that caspase-8 activates other downstream caspases apoptosis. Furthermore, analysis of ApoL6 will prove useful independent of mitochondria (Fig. 7). Our investigation to investigate the dynamic interactions of the Bcl-2 family furthermore showed that ApoL6 is a bona fide BH3-only members in cancer genetics and ultimately in designing thera- proapoptotic molecule, and the deletion of the BH3 domain peutic strategies to combat cancer and possibly other diseases. effectively abolished the apoptotic activity of ApoL6. Previ- ously, BH3 domains have been shown to be essential for other proapoptotic BH3-only proteins, such as PUMA and Bad Materials and Methods (18, 19). Finally, we showed that AD-ApoL6 is a potent inducer Cell Lines, Chemicals, and Culture Media of apoptosis in a variety of cells, such as DLD-1 and HEK 293 DLD-1.TA14, a p53-null and Tet-Off-inducible colorectal (Figs. 2 and 5). This suggests that ApoL6 may be a general cancer cell line (34), was a gift from Dr. Bert Vogelstein apoptosis effector. Thus, it is reasonable to speculate that (Johns Hopkins Medical Institutes, Baltimore, MD). HEK dysregulation of ApoL6 in cells might derail apoptosis and 293 cell line was purchased from American Type Culture cause tumorigenesis. Collection (Manassas, VA) and cultured in DMEM. Except The observation that ApoL6 may be a proapoptotic for hygromycin B (Calbiochem, La Jolla, CA) and doxycycline component could provide a link between lipid signaling and (Sigma Chemical Co., St. Louis, MO), all components of the apoptosis. Based on the predicted amphipathic a-helix structure cell culture were obtained from Cellgro (Mediatech, Inc., and the pattern of expression in various tissues, plasma ApoL is Herndon, VA). suggested to mediate cholesterol exchange between tissues and blood (24-29). It has been suggested, moreover, that apolipo- Functional Genomics and Biocomputing Approaches to proteins are also involved in the intracellular lipid transport for Identify Novel BH3-Only Proteins the supply of lipid components to the membranes of organelles, To search human databases for novel BH3-only proteins, a critical step in the maintenance of membrane fluidity and the 9–amino acid sequences of all BH3 domains of the 17 integrity (30, 31). Importantly, recent studies have shown that known human BH3-containing proteins were retrieved from lipid messengers, such as ceramide, sphingolipids, ganglioside the (a) published literature (up to August 2003; refs. 7, 10, 11, GD3, and oxidized cholesterol, play important roles in inducing 14), (b) PROSITE motif database (http://us.expasy.org/prosite/; matrix metalloproteinase and initiating apoptosis (36-43). ref. 32), and (c) National Center for Biotechnology Information However, how those lipid messengers are transported to particular intracellular sites (e.g., mitochondria, endoplasmic reticulum, and lysosome) and what mechanisms they employ to induce apoptosis remain largely unknown. Recently, another BH3-only proapoptotic molecule, Bid, was reported to bind monolysocardiolipin, suggesting a connection among lipid metabolism, relocation of Bid to mitochondria, and induction of apoptosis (43-45). We propose that ApoL6 is multifunctional and plays some additional role in transporting intracellular lipid messengers during apoptotic processes. Consistent with this idea, we showed that intracellular ApoL6 bound specific fatty acids (C16:0 and C18:0) and/or lipids in the ApoL6-lipoprotein complex of induced DLD-1.ApoL6.V5 cells (Fig. 6C). The BH3 domain in ApoL6 affected binding of lipids of ApoL6 both quantitatively and qualitatively. It will be important to determine the identities and functions of the complex lipids/ fatty acids in the ApoL6.V5-lipoprotein complex. The gene encoding ApoL6 is localized to chromosome 22q12.3, a high-susceptibility locus for schizophrenia (25, 46, 47). Interestingly, expression of ApoL1, ApoL2, ApoL4, and ApoL6 is significantly up-regulated (>1.41-fold) in the brain of schizophrenia (47). Expression of ApoL2 was also found to be up-regulated >1.60- and >1.53-fold in bipolar and major depressive disorders, respectively (47). It will be interesting to see if overexpression of ApoL6 induces apoptosis in brain cells and schizophrenic phenotypes in animal models. In summary, members of the Bcl-2 family of proteins play pivotal roles in regulating apoptosis and are excellent FIGURE 3. Flow cytometry analysis of apoptosis in DLD-1.ApoL6 cells candidates in the restoring of functional apoptosis back to by propidium iodide staining. DLD-1.ApoL6.V5 cells were grown to 50% confluence in D.20 medium and then switched to D.0 medium. Cells were tumor cells. Based on our results, ApoL6, a BH3-only and a harvested hours after induction: (A) 0 hour, (B) 6 hours, (C) 12 hours, and lipid-binding protein, sets the stage for investigators to (D) 24 hours.

FIGURE 4. ApoL6-induced release of cytochrome c (Cyt. C) and Smac/DIABLO and activation of caspase-9. Immunoflu- orescent microscope analysis of subcellular localization of (A) Smac/DIABLO and (B)cytochrome c in DLD-1.ApoL6.V5 cells. White triangles, cytosolic localization of antigens in apoptotic cells. C. Immunoblotting analysis of Smac/DIABLO and cytochrome c in the cytosolic fractions of uninduced and induced DLD-1.ApoL6.V5 cells. D. Relative amounts of Smac/DIA- BLO and cytochrome c in the cytosolic fractions of induced versus uninduced DLD- 1.ApoL6.V5 cells. Columns, mean of at least three independent experiments; bars, SE. *, P < 0.05; **, P < 0.01. E. Immuno- blotting analysis of activation of caspase-9 in induced DLD- 1.ApoL6.V5 cells grown in D.0 medium. Hours after induction are as indicated. h-actin was used as a loading control.

(ftp://ftp.ncbi.nih.gov/genomes/H_sapiens/; ref. 48) protein cards; ref. 49), the Protein Family database at the Sanger sequence databases. The BH3 domain consensus sequence Institute (http://www.sanger.ac.uk/Software/Pfam/; ref. 33), the was designed as NH2-terminal-(LM)(ERAKQL)(RAQICEY) PROSITE database, and trEST, trGEN, or Hits programs (50), (MILAFSE)(ASG)D(DEQKGR)(LFMIVD)(NDVHSEQA)- were used to conduct sequence analyses. COOH-terminal. The annotated human open reading frames were downloaded from the National Center for Biotechnology Generation of Stably Transfected and Inducible ApoL6 Information database. The search program for the BH3 domain Genes in DLD-1.TA14 Cells was written in Perl (http://www.perldoc.com/perl5.6/). To A previously described Tet-Off-inducible system was analyze if newly identified protein possessed other putative used to generate ApoL6-inducible, ApoL6.V5-inducible, or domains/motifs, its protein sequence was retrieved and public ApoL6.dBH3-inducible cell lines in DLD-1.TA14 (34, 35). sequence analysis programs, such as the LocusLink database Briefly, PCR amplified 1,032-bp human ApoL6 cDNA con- of National Center for Biotechnology Information (http:// taining the full-length open reading frame (ApoL6) or ApoL6 NCBI.NIH.GOV/LocusLink) and GeneCards database of Weiz- with a COOH-terminal V5 tag (ApoL6.V5) was cloned into mann Institute of Science (http://bioinformatics.weizmann.ac.il/ pBI-EGFP vector (Clontech, Palo Alto, CA; Fig. 1C). The

sequences were confirmed by sequencing. The primers used for PCR was conducted using gene-specific primers and a amplifying ApoL6 were reverse 5V-TCCTGAGGAACATTC- standard protocol as described (36, 37). The primers were ATGTAAAC-3V (bp +1,339 to +1,317) and forward 5V-GCT- reverse 5V-TCCTGAGGAACATTCATGTAAAC-3V(bp +1,339 GCCACAGAGGCTGATGGACAAC-3V (bp À16 to +9). The to +1,317) and forward 5V-GTCTTCTGACCACTGGCCAAG- primers used for amplifying ApoL6.V5 were reverse 5V- 3V (bp +902 to +922). GATAGGAGCTCGCTAGCTCACGTAGAATCGAGACC- GAGGAGAGGGTTAGGGATAGGCTTACCTGTAAACTG- Site-Specific Deletion Mutagenesis TACATACACACAGAC-3V (bp +1,323 to +1,300) and forward PCR-based site-directed mutagenesis kit (ExSite kit, 5V-GATACGACGCGTTGCCACAGAGGCTGATGGACAAC- Stratagene, La Jolla, CA) was used to generate the BH3 3V (bp À14 to +9). The resulting plasmids were pBI-ApoL6 and domain deletion allele (deletion of the 9–amino acid BH3 pBI-ApoL6.V5. To make stably transfected DLD-1 cells, pBI- domain; amino acids 60-68; bp +178 to +204; Fig. 1C) of ApoL6 or pBI-ApoL6.V5 was cotransfected with pIND-hygro ApoL6 as described in manufacturer’s manual (Stratagene). In (Invitrogen, Carlsbad, CA) into DLD-1.TA14 cells using the brief, we incorporated the following steps to ensure success in DMSO-enhanced protocol as described (51, 52). Cells were amplifying the designed construct: (a) increasing plasmid selected in D.20 medium composed of McCoy’s 5A supple- template (pBI-ApoL6) concentration f1,000-fold over con- mented with 10% fetal bovine serum, 1Â antibacterial ventional PCR conditions, (b) reducing the number of PCR antimycotic solution, 400 Ag/mL G418, 250 Ag/mL hygrom- cycles from 25-30 to 5-10, (c) adding the restriction ycin B, and 20 ng/mL doxycycline (D.20). Approximately 40 endonuclease DpnI to select against parental DNA, (d) using individual clones of each transfection were screened for Taq Extender in the PCR mix for increased reliability for inducibility of green fluorescent protein. Those that had a low PCR, (e) using Pfu DNA polymerase to polish the ends of the background in D.20 and uniform green fluorescent protein PCR product, and (f) increasing efficient intramolecular expression in induction medium D.0 (same medium as D.20 ligation in the presence of T4 DNA ligase. The primers that but without doxycycline) were selected. These candidates were we used to make ApoL6 deletion construct were forward 5V- then assessed for inducible expression of ApoL6 transcripts AAAACCCACAGAAATTCACC and reverse 5V- by semiquantitative reverse transcription-PCR and immuno- (p)CTTGTCAATGTTCCCTTTAGAACACCCAAAA. We blotting analyses. Total RNA was isolated from cells using a then confirmed the deletion construct pBI-ApoL6.dBH3 by Purescript kit (Gentra Systems, Minneapolis, MN). Reverse sequencing and used it to generate inducible cell lines in transcription was conducted using random hexamers and DLD-1 as described above.

FIGURE 5. Induction of cell death by ApoL6 in HEK 293 cells treated with adenovirus-ApoL6 (AD-ApoL6) for 3 days. Adenovirus only and AD- ApoL6.dBH3 served as control. Same titer of viruses of adenovirus (A and D), AD-ApoL6 (B and E), and AD-ApoL6.dBH3 (C and F) were used. Green fluorescent protein indicated cells that were infected with adenoviruses (A-C). Cells were treated with Hoechst nuclear staining 3 days after infection (D-F). Apoptotic cells were characterized by cell morphology (roundup and detached cells as indicated by white triangles in B) and nuclear staining (fragmented and condensed nuclei as indicated by red triangles in E).

FIGURE 6. Analysis of fatty acid components extracted from ApoL6- containing lipoprotein complex by gas chromatography-mass spectrometry. Extracted, trans-esterified fatty acids were subjected to gas chromatography-mass spectrosco- py analysis. A. Agarose-conjugated anti-V5 antibody alone, (B) complex isolated from uninduced DLD- 1.ApoL6.V5 cells, (C) complex isolated from induced DLD- 1.ApoL6.V5 cells, (D) standard profile of known fatty acids, (E) complex isolated from uninduced DLD- 1.ApoL6.dBH3 cells, (F) complex isolated from induced DLD- 1.ApoL6.dBH3 cells and (G) standard (STD): A, 16:0; B, 16:1 cis; C, 16:1 trans; D, 18:0; E, 18:1 cis; F, 18:1 trans; G, 18:2 cis, cis; H, 18:2 trans, trans; I, 19:0 (internal standard). Peaks representing 16:0 and 18:0 were detected in samplesofinduced DLD-1.ApoL6.V5 and DLD-1.ApoL6.dBH3 cells. Other peaks in induced DLD-1.ApoL6.V5 cells and DLD-1.ApoL6.dBH3, however, because of the lack of internal standards of fatty acid derivatives, were unknown.

Wild-type ApoL6 Induces Dose-Dependent and Time Flow Cytometry Analysis of Apoptotic DLD-1.ApoL6 Cells Course–Dependent Apoptosis by Propidium Iodide Taking advantage of the inducibility of the Tet-Off system, Fragmentation of genomic DNA to sub-G1 DNAwas assessed dose-dependent and time-dependent apoptosis studies in DLD- with the aid of fluorescence-activated cell sorting (Becton 1.ApoL6 or DLD-1.ApoL6.V5 cells were conducted. Briefly, Dickinson, Franklin Lakes, NJ) analysis as described (34). stably transfected cells were grown in D.20 medium. For DLD-1.ApoL6 or DLD-1.ApoL6.V5 cells were cultured in D.20 ApoL6 dose-dependent effect, adherent cells were rinsed in medium until 80% confluence and then switched to D.0 medium. PBS and re-fed with the induction medium containing Total cells were collected at indicated time points. Cells were gradually decreased amount of doxycycline (e.g., 5 ng/mL, resuspended in 100 AL PBS, fixed with 1 mL acetone-methanol D.5; 1 ng/mL, D.1; and 0 ng/mL, D.0), and adherent cells were (1:1; À20jC), and washed with PBS. Each pellet was then harvested after 24 hours. For time course–dependent study, resuspended in 400 AL PBS containing 1 mg/mL RNase and adherent cells were rinsed in PBS and re-fed with D.0 induction incubated on ice for 1 hour. After the incubation with 20 AL medium, and adherent cells were harvested at the indicated time propidium iodide solution (2 mg/mL in PBS; Sigma) for 0.5 points as described previously (34, 35). Cells cultured in D.20 hour on ice, cells were subjected to fluorescence-activated cell were used as control. sorting assay.

Assay of Cytochrome c and Smac/DIABLO Release by Bio-Rad, Hercules, CA). For ApoL6 fusion protein, mono- Immunofluorescent Microscopy clonal anti-V5 purified mouse immunoglobulin (1:2,000, ApoL6-induced release of cytochrome c and Smac/ Sigma Chemical Co.) and goat antirabbit HP-labeled DIABLO from the mitochondrial intermembrane space into secondary antibodies (1:5,000, Bio-Rad) were used. the cytosol was analyzed using monoclonal anti–cytochrome c antibody (PharMingen, San Diego, CA) or monoclonal AD-ApoL6 Infection anti-Smac/DIABLO antibody (Zymed, South San Francisco, Construction, amplification, and assaying for titers of CA) to localize cytochrome c or Smac/DIABLO in DLD- adenoviruses harboring ApoL6 alleles and viral infection of 1.ApoL6.V5 cells. Briefly, cells were grown on glass mammalian cells were done as described previously (34). A coverslips coated with collagen (Roche, Indianapolis, IN). variety of cells were infected with control adenovirus, AD- After induction of ApoL6 for 6 hours, cells were rinsed and ApoL6, and AD-ApoL6.dBH3 with a range of concentrations then fixed with 4% (w/v) paraformadehyde solution at room from 0.01 to 5 multiplicity of infection (or plaque-forming temperature. Coverslips were rinsed twice with HBS and unit per cell, which was estimated to be 0.2-100 virus blocked with 3% (w/v) bovine serum albumin in TBS plus particles per cell) for 3 days. Cell numbers and apoptosis Tween 20 for 0.5 hour at room temperature. After incubation were measured over time by cell counting, analysis of cell overnight at 4jC in 3% (w/v) bovine serum albumin morphology, and Hoechst nuclear staining (34). containing the anti–cytochrome c antibody (1:150) or anti- Generation of Antipeptide Antibodies against Human Smac/DIABLO antibody (1:100), coverslips were rinsed with ApoL6 HBS four times and incubated in 3% (w/v) bovine serum To raise antipeptide antisera specifically against human albumin in TBS plus Tween 20 containing Texas red–labeled ApoL6, we used a peptide corresponding to the codons 310 to secondary antibody (1:200, Jackson ImmunoResearch, Soham 324 of ApoL6 (TCETEAYWKELREHV) to immunize rabbits Cambridgeshire, United Kingdom) for 1 hour at room (Anaspec, San Jose, CA). Antibody was affinity purified by temperature with protection against light. Finally, coverslips chromatography on peptide (antigen)-linked Sepharose and were rinsed with HBS, mounted onto slides, and subjected to designated as anti-ApoL6. fluorescent microscope analysis. Coimmunoprecipitation of ApoL6-Lipid Complex and Assay of Expression of ApoL6.V5 Fusion Protein, Fatty Acid Analysis by Gas Chromatography-Mass Caspase-9 and Caspase-8 Activation, and Release of Spectrometry Cytochrome c and Smac/DIABLO by Immunoblotting To identify ApoL6 lipid/fatty acid binding components, To prepare total cellular extracts, cells were rinsed twice lipoprotein complexes harboring ApoL6.V5 or ApoL6.dBH3 with PBS and gently scraped off from culture dish in 1Â in total protein extracts of DLD-1.ApoL6.V5 or DLD- PBS with a rubber policeman. Total cellular extracts were 1.ApoL6.dBH3 cells after 24-hour induction were immuno- obtained by dissolving and sonicating cells in radioimmuno- precipitated by agarose-immobilized anti-V5 antibody beads precipitation assay buffer solution [1% sodium deoxycholate, (Sigma Chemical Co.) or anti-ApoL6 antibody and agarose- 0.1% SDS, 1% Triton X-100, 0.01 mol/L Tris-HCl (pH 8.0), protein A beads (BD Biosciences, San Diego, CA). Lipopro- 0.14 mol/L NaCl]. To prepare cytosolic fractionation, cells teins were then deproteinized, subjected to trans-esterification were gently scraped off from culture dish, centrifuged down reaction by 0.5 mL of 14% boron trifluoride in methanol, at 200 Â g for 5 minutes, washed twice with 1.0 mL cold heated for 15 minutes at 80jC, cooled down on ice, and then PBS, and resuspended in 200 to 300 AL of buffer A added 1 mL water and 1.5 mL hexane (53). The hexane layer [250 mmol/L sucrose, 20 mmol/L HEPES-KOH (pH 7.4), containing fatty acid methyl esters was transferred to another 10 mmol/L KCl, 1.5 mmol/L Na-EGTA, 1.5 mmol/L Na- vial and dried with anhydrous sodium sulfate. Aliquots of

EDTA, 1 mmol/L MgCl2, 1 mmol/L DTT, and protease the hexane phase were analyzed by gas chromatography-mass inhibitors]. Cells were incubated on ice for 30 minutes and then disrupted by Dounce homogenizer. Homogenates were centrifuged at 800 Â g for 10 minutes at 4jC. Supernatants were further centrifuged at 22,000 Â g for 15 minutes at 4jC. The resulting supernatants were used for immunoblotting analysis. Protein concentrations were measured using the BCA protein assay reagent Kit (Pierce Biotechnology, Rockford, IL). Proteins were separated using 10% SDS- PAGE (20 Ag protein per lane) and transferred to polyvinylidene difluoride membranes (Millipore, Bedford, MA), which were then incubated with polyclonal anti- activated caspase-9 antibody (1:1,000, Cell Signaling Tech- nology, Beverly, MA), monoclonal anti–cytochrome c (1:1,00), and monoclonal anti-Smac/DIABLO (1:500). Sub- FIGURE 7. Hypothetical model of ApoL6-induced apoptosis in DLD-1 sequently, blots were incubated with goat antirabbit or anti- cells. Intracellular accumulation of ApoL6 induces activation of both mouse HP-labeled secondary antibodies (1:5,000 for both, caspase-8 and mitochondria-mediated apoptosis.

spectrometry (Hewlett-Packard 5890 II, DB-225 column and 20. Zamzami N, El Hamel C, Maisse C, et al. Bid acts on the permeability the mass selective detector 5972A, Hewlett-Packard, Palo transition pore complex to induce apoptosis. Oncogene 2000;19:6342 – 50. 21. Sugiyama T, Shimizu S, Matsuoka Y, et al. Activation of mitochondrial Alto, CA) in scan mode. The injector temperature was set at voltage-dependent anion channel by a pro-apoptotic BH3-only protein Bim. 250jC, detector at 280jC, and oven at 70jC for 1 minute and Oncogene 2002;21:4944 – 56. the temperature was increased from 70j to180jCat20jC/min, 22. Fei P, Bernhard EJ, El-Deiry WS. Tissue-specific induction of p53 targets 180 to 220jCat3jC/min, and 220jC for 3 minutes. Methyl in vivo. Cancer Res 2002;62:7316 – 27. nonadecanoate (C19:0) was used as an internal standard for 23. Villunger A, Michalak EM, Coultas L, et al. p53- and drug-induced apoptotic responses mediated by BH3-only proteins puma and noxa. Science 2003; each experiment. Several known fatty acids were used to gen- 302:1036 – 8. erate a standard profile. 24. Duchateau PN, Pullinger CR, Cho MH, et al. Apolipoprotein L gene family: tissue-specific expression, splicing, promoter regions; discovery of a new gene. J Lipid Res 2001;42:620 – 30. Statistical Analysis 25. Page NM, Butlin DJ, Lomthaisong K, Lowry PJ. The human apolipoprotein F L gene cluster: identification, classification, and sites of distribution. Genomics Values are means SE. Data were analyzed by Student’s t 2001;74:71 – 8. test. Differences with P < 0.05 were considered significant. 26. Monajemi H, Fontijn RD, Pannekoek H, Horrevoets AJG. The apolipoprotein L gene cluster has emerged recently in evolution and is expressed in human vascular tissue. Genomics 2002;79:539 – 46. Acknowledgments 27. Duchateau PN, Movsesyan I, Yamashita S, et al. Plasma apolipoprotein L We thank Drs. Michael Wilson and Jeffrey Griffith for commenting on the concentrations correlate with plasma triglycerides and cholesterol levels in article and the colleagues of the DNA Service Core and Flow Cytometry normolipidemic, hyperlipidemic, and diabetic subjects. J Lipid Res 2000;41: Facility at University of New Mexico Health Sciences Center for their kind 1232 – 6. assistance. 28. Duchateau PN, Pullinger CR, Orellana RE, et al. Apolipoprotein L, a new human high density lipoprotein apolipoprotein expressed by the pancreas. Identification, cloning, characterization, and plasma distribution of apolipoprotein L. J Biol Chem 1997;272:25576 – 82. References 29. Horrevoets AJ, Fontijn RD, Zonneveld AJV, et al. Vascular endothelial genes 1. Hengartner MO. The biochemistry of apoptosis. Nature 2000;407:770 – 6. that are responsive to tumor necrosis factor-a in vitro are expressed in 2. Ferri KF, Kroemer G. Organelle-specific initiation of cell death pathways. atherosclerotic lesions, including inhibitor of apoptosis protein-1, stannin, and Nat Cell Biol 2001;3:E255 – 63. two novel genes. Blood 1999;93:3418 – 31. 3. Vogelstein B, Lane D, Levine AJ. Surfing the p53 network. Nature 2000; 30. Schroeder F, Gallegos AM, Atshaves BP, et al. Recent advances in membrane 408:307 – 10. microdomains: rafts, caveolae, and intracellular cholesterol trafficking. Exp Biol 4. Jacobson MD, Wei M, Raff MC. Programmed cell death in animal Med (Maywood) 2001;226:873 – 90. development. Cell 1997;88:347 – 54. 31. Huijbregts RP, Topalof L, Bankaitis VA. Lipid metabolism and regulation of 5. Johnston RW, Ruefli AA, Lowe SW. Apoptosis: a link between cancer membrane trafficking. Traffic 2000;1:195 – 202. genetics and chemotherapy. Cell 2002;108:153 – 64. 32. Falquet L, Pagni M, Bucher P, et al. The PROSITE database, its status in 6. Vila M, Przedborski S. Targeting programmed cell death in neurodegenerative 2002. Nucleic Acids Res 2000;30:235 – 8. diseases. Nat Rev Neurosci 2003;4:365 – 75. 33. Bateman A, Birney E, Cerruti L, et al. The Pfam protein families database. 7. Scorrano L, Korsmeyer SJ. Mechanisms of cytochrome c release by Nucleic Acids Res 2002;30:276 – 80. proapoptotic BCL-2 family members. Biochem Biophys Res Commun 2003; 34. Yu J, Zhang L, Hwang PM, et al. Identification and classification of 304:437 – 44. p53-regulated genes. Proc Natl Acad Sci U S A 1999;96:14517 – 22. 8. Boatright KM, Renatus M, Scott FL, et al. A unified model for apical 35. Donald SP, Sun XY, Hu CA, et al. Proline oxidase, encoded by p53-induced caspase activation. Mol Cell 2003;11:529 – 41. gene-6, catalyzes the generation of proline-dependent reactive oxygen species. 9. Earnshaw WC, Martins LM, Kaufmann SH. Mammalian caspases: structure, Cancer Res 2001;61:1810 – 5. activation, substrates, and functions during apoptosis. Annu Rev Biochem 1999; 36. Merrill AH Jr. De novo sphingolipid biosynthesis: a necessary, but 68:383 – 424. dangerous, pathway. J Biol Chem 2002;277:25843 – 6. 10. Cory S, Adams JM. The Bcl2 family: regulators of the cellular life-or-death 37. Hannun YA, Obeid LM. The ceramide-centric universe of lipid-mediated cell switch. Nat Rev Cancer 2002;2:647 – 56. regulation: stress encounters of the lipid kind. J Biol Chem 2002;277:25847 – 50. 11. Huang DC, Strasser A. BH3-only proteins—essential initiators of apoptotic 38. Hatch GM. Cardiolipin: biosynthesis, remodeling and trafficking in the heart cell death. Cell 2000;103:839 – 42. and mammalian cells [review]. Int J Mol Med 1998;1:33 – 41. 12. Wang X. The expanding role of mitochondria in apoptosis. Genes Dev 39. Kolesnick RN, Kronke M. Regulation of ceramide production and apoptosis. 2001;15:2922 – 33. Annu Rev Physiol 1998;60:643 – 65. 13. Esposti MD, Dive C. Mitochondrial membrane permeabilisation by Bax/Bak. 40. Esposti MD. Lipids, cardiolipin and apoptosis: a greasy license to kill. Cell Biochem Biophys Res Commun 2003;304:455 – 61. Death Differ 2002;9:234 – 6. 14. Letai A, Bassik MC, Walensky LD, et al. Distinct BH3 domains either 41. Garcia-Ruiz C, Colell A, Morales A, et al. Trafficking of ganglioside GD3 sensitize or activate mitochondrial apoptosis, serving as prototype cancer to mitochondria by tumor necrosis factor-a. J Biol Chem 2002;277:36443 – 8. therapeutics. Cancer Cell 2002;2:183 – 92. 42. Mertens A, Holvoet P. Oxidized LDL and HDL: antagonists in athero- 15. Kuwana T, Mackey MR, Perkins G, et al. Bid, Bax, and lipids cooperate to thrombosis. FASEB J 2001;15:2073 – 84. form supramolecular openings in the outer mitochondrial membrane. Cell 2002; 43. Lutter M, Fang M, Luo X, et al. Cardiolipin provides specificity for targeting 111:331 – 42. of tBid to mitochondria. Nat Cell Biol 2000;2:754 – 61. 16. Zong WX, Lindsten T, Ross AJ, et al. BH3-only proteins that bind pro- 44. Esposti MD, Erler JT, Hickman JA, Dive C. Bid, a widely expressed survival Bcl-2 family members fail to induce apoptosis in the absence of Bax and proapoptotic protein of the Bcl-2 family, displays lipid transfer activity. Mol Cell Bak. Genes Dev 2001;15:1481 – 6. Biol 2001;21:7268 – 76. 17. Cheng EH, Wei MC, Weiler S, et al. BCL-2, BCL-X(L) sequester BH3 45. Esposti MD, Cristea IM, Gaskell SJ, et al. Proapoptotic Bid binds to domain-only molecules preventing BAX- and BAK-mediated mitochondrial monolysocardiolipin, a new molecular connection between mitochondrial apoptosis. Mol Cell 2001;8:705 – 11. membranes and cell death. Cell Death Differ 2003;10:1300 – 9. 18. Yu J, Zhang L, Hwang PM, Kinzler KW, Vogelstein B. PUMA induces 46. Mimmack ML, Ryan M, Baba H, et al. Gene expression analysis in the rapid apoptosis of colorectal cancer cells. Mol Cell 2001;7:673 – 82. schizophrenia: reproducible up-regulation of several members of the apolipo- 19. Tsujimoto Y. Cell death regulation by the Bcl-2 protein family in the protein L family located in a high-susceptibility locus for schizophrenia on mitochondria. J Cell Physiol 2003;195:158 – 67. chromosome 22. Proc Natl Acad Sci U S A 2002;99:4680 – 5.

47. Sutcliffe JG, Thomas EA. The neurobiology of apolipoproteins in psychiatric 51. Geraghty MT, Vaughn D, Nicholson AJ, et al. Mutations in the Delta1- disorders. Mol Neurobiol 2002;26:369 – 88. pyrroline 5-carboxylate dehydrogenase gene cause type II hyperprolinemia. Hum 48. Pruitt KD, Maglott DR. RefSeq and LocusLink: NCBI gene-centered Mol Genet 1998;7:1411 – 5. resources. Nucleic Acids Res 2001;29:137 – 40. 52. Hu CA, Lin WW, Obie C, Valle D. Molecular enzymology of mammalian 49. Rebhan M, Chalifa-Caspi V, Prilusky J, Lancet D. GeneCards: encyclopedia Delta1-pyrroline-5-carboxylate synthase. Alternative splice donor utilization for genes, proteins and diseases. Rehovot (Israel): Weizmann Institute of Science, generates isoforms with different sensitivity to ornithine inhibition. J Biol Chem Bioinformatics Unit and Genome Center; 1997. 1999;274:6754 – 62. 50. Pagni M, Iseli C, Junier T, Falquet L, Jongeneel V, Bucher P. TrEST, trGEN 53. Morrison WR, Smith LM. Preparation of fatty acid methyl esters and and Hits: access to databases of predicted protein sequences. Nucleic Acids Res dimethyl acetals from lipids with boron fluoride methanol. J Lipid Res 1964; 2001;29:148 – 51. 5:84 – 91.

Mol Cancer Res 2005;3(1). January 2005 Downloaded from mcr.aacrjournals.org on September 27, 2021. © 2005 American Association for Cancer Research. Apolipoprotein L6, a Novel Proapoptotic Bcl-2 Homology 3− Only Protein, Induces Mitochondria-Mediated Apoptosis in Cancer Cells 1 1 Howard Hughes Medical Institute research aids to University of New Mexico Cancer Research and Treatment Center, American Cancer Society ACS-IRG-192 grant 412488-00095, and University of New Mexico Research Allocation Committee grant C-2222-RAC (C-A.A. Hu).

Zhihe Liu, Huimei Lu, Zeyu Jiang, et al.

Mol Cancer Res 2005;3:21-31.

Updated version Access the most recent version of this article at: http://mcr.aacrjournals.org/content/3/1/21

Cited articles This article cites 51 articles, 15 of which you can access for free at: http://mcr.aacrjournals.org/content/3/1/21.full#ref-list-1

Citing articles This article has been cited by 7 HighWire-hosted articles. Access the articles at: http://mcr.aacrjournals.org/content/3/1/21.full#related-urls

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://mcr.aacrjournals.org/content/3/1/21. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.