Apoptosis 2006; 11: 459–471C© 2006 Springer Science+Business Media, LLC. Manufactured in The United States.DOI: 10.1007/s10495-006-5702-1

Cell death regulation by B-cell lymphoma protein

Y. K. Verma, G. U. Gangenahalli, V. K. Singh, P. Gupta, R. Chandra, R. K. Sharma and H. G. Raj

Stem Cell Gene Therapy Research Group, Lucknow Road, Timar Pur, Delhi-110054, India (Y. K. Verma, G. U.Gangenahalli, V. K. Singh, P. Gupta); Dr. B. R. Ambedkar Centre for Biomedical Research, University of Delhi, Delhi-110007, India (R. Chandra); Institute of Nuclear Medicine and Allied Sciences, Lucknow Road, Timar Pur, Delhi-110054,India (R. K. Sharma); Department of Biochemistry, V. P. Chest Institute, Delhi-110007, India (H. G. Raj)

Published online: 9 March 2006

Bcl-2 (B Cell Lymphoma) protein is an anti-apoptoticmember of Bcl-2 family, which is comprised of pro- andanti-apoptotic members. It regulates cellular proliferationand death by inter- and intra-family interactions. It has apotential to suppress apoptotic cell death under varietyof stress conditions by modulating mitochondrial trans-membrane potential. However, prevalence of constitutivelyactivated Bcl-2 cellular activity is not always required incells; a mechanism likely exists in cells, which controlsits activity. When expression of Bcl-2 is unregulated, itgenerates lymphoma like, follicular B-cell lymphoma. Thisarticle reviews the structural and functional regulationof Bcl-2 activity at transcriptional, translational, domain,structural and post-translational level, which also accountsfor the effects of its deletion and site-directed mutants in theregulation of cellular proliferation and differentiation in vitroand in vivo. This concisely reviewed information on Bcl-2helps us to update our understanding of cell death and itsmodulation by Bcl-2 and its mutant’s interaction, which hasgained therapeutic benefits in cell growth and proliferation,particularly for sensitive human hematopoietic stem cells.

Keywords: active site; Bcl-2 family; hematopoietic stem cells;homology model; inducible expression; mutants.

Abbreviations: Gly: Glycine; Ala: Alanine; Asp: Aspartic acid;Trp: Tryptophan; Glu: Glutamic Acid; Ser: Serine; Arg: Argi-nine; Gln: glutamine; Leu: Leucine; Lys: Lysine; Val: Valine; Iso:Isoleucine; α: alpha helix; �: deletion.

Introduction

Bcl-2 is an endogenous regulator of apoptosis, also knownas programmed cell death. It was originally identifiedas the oncogene activated by the characteristic t (14:18)translocation in non-Hodgkin’s lymphoma.1 Bcl-2 is a25 kD protein localized principally on the outer membranesof mitochondria, endoplasmic reticulum and nucleus. It isthe firstly identified member of Bcl-2 family which inhibits

Y. K. Verma and G. U. Gangegenahalli are contributed equally to this work.

Correspondence to: G. U. Gangenahalli, Stem Cell Gene Ther-apy Research Group, Institute of Nuclear Medicine and AlliedSciences (INMAS), Lucknow Road, Timar Pur, Delhi-110054,India. Tel.: 011-91-23911712; Fax: 011-91-25737049; e-mail:gugdutta @rediffmail.com

cell death in a variety of stress conditions including depriva-tion of essential cytokines, heat shock, calcium ionophores,staurosporine, overdose of radiation and exposure to DNAdamaging agents.2 On the basis of cell death inhibition/promotion, Bcl-2 family proteins are divided into anti-and pro-apoptotic members respectively. Anti-apoptoticmembers (death antagonists) include Bcl-2, Bcl-XL, Bcl-W,A1, Mcl-1 and all known viral homologues (E1B19 kD-adenovirus protein, BJRF1-Epstein-Bar virus protein, LMW5.1-African swine fever virus protein and Bcl-2 homologsencoded by Herpes virus). Pro-apoptotic members (death ag-onists) include Bax, Bid, Bak, Bid, Harakiri, Bim, Bik, Badand Bcl-Xs (alternatively spliced form of Bcl-2).3–9 Fourconserved domains have so far been identified in this familytermed Bcl-2 homology (BH) domains viz. BH4, BH3,BH1 and BH2. Among these, BH1 and BH2 are highlyconserved domains located at C-terminal of these proteins,whereas BH3 and BH4 are loosely conserved.10 The presenceof one or more of these domains serves to define membershipof the protein to modulate apoptosis in functional assays.Anti-apoptotic members such as Bcl-2, Bcl-XL and Bcl-Wpossess all four BH domains. Other anti-apoptotic memberssuch as Mcl-1, BHRF-1 and KSHV-Bcl-2 possess onlyBH1, BH2 and BH3 domains. The pro-apoptotic proteinsof Bax subclass possess BH1, BH2 and BH3 domains whilethe members of the pro-apoptotic BH3 subclass such as Bid,have only BH3 region. On the basis of threading, sequenceand secondary structure comparison the Bcl-2 familymembers are also subdivided into two conformationalsubgroups, “BH3 buried” (Bcl-2 and other anti-apoptoticproteins) and “BH3 exposed” (pro-apoptotic proteins)(Table 1).5 Bcl-2 protein contains all four BH domains,BH4 (12–28 amino acids), BH3 (90–98 amino acids),BH1 (136–155 amino acids) and BH2 (188–199 aminoacids). In addition to these it also possess one X domain(192–203 amino acids) and one regulatory/flexible loopdomain (FLD) (30–89 amino acids).1 In tertiary structureBcl-2 is folded into two central hydrophobic anti-parallelalpha helices i.e., α5 and α6, whereas α6, α7, α3 and α4form flanking amphipathic helices creating a hydrophobiccleft, an interface essential for interaction of pro-apoptoticproteins such as Bax, Bik, Bid and Bim through their BH3amphipathic α-helical domain. There are multiple types of

Apoptosis · Vol 11 · No 4 · 2006 459

Y.K. Verma et al.

Table 1. Classification of Bcl-2 and its family members on the basis of cell death inhibition/promotion, presence/absence of BH domainsand BH3 domain buried/ exposed

Anti-apoptotic (cell death inhibitor)

Bcl-2 family proteins Domains (+, present; –, absent) BH3 domain

Protein (Species) BH4 BH3 BH1 BH2 buried/exposed

Bcl-2 Mammalian + + + + buried

Bcl-XL Mammalian + + + + buried

Bcl-W Mammalian + + + + buried

MCL-1 Mammalian + + + + buried

A1/BFL-1 Mammalian + + + + ?

BOO/DIVA Mammalian + − + + buried

NR-13 Mammalian + − + + ?

E1B-19K Viral − + + + ?

BHRF1 Viral − − + + ?

KS-Bcl-2 Viral − − + + ?

ORF16 Viral − − + + ?

LMW-HL Viral − − + + ?

CED-9 C. elegans + + + + buried

Pro-apoptotic (cell death promoter)

Bax Mammalian − + + + buried

Bak Mammalian − + + + buried

BOK/MTD Mammalian − + + + buried

Bcl-XS Mammalian + + − − exposed

Pro-apoptotic (BH3 only)

Bid Mammalian − + − − buried

Bad Mammalian − + − − exposed

BIK/NBK Mammalian − + − − exposed

BLK Mammalian − + − − exposed

HRK Mammalian − + − − exposed

BIM/BOD Mammalian − + − − exposed

NIP3 Mammalian − + − − exposed

NIX/BNIP3 Mammalian − + − − exposed

EGL-1 C. elegans − + − − exposed

physical intra- and inter-family interactions displayed byBcl-2. Intra-family interactions include homodimerizationand hetrodimerization. The inter-family interactions occurwith Raf-1 (Bcl-2 associating protein), BAG-1 (kinase),11

p53 binding protein (p53-BP2),12 PP2A (phosphatase)13

calcineurin,14 caspase-3 (apoptosis effecter protease), JNK(kinase),15–18 ubiquitin,19 PKC and ERK 1/2 (kinase)20 andNFAT.21

This review updates information about the regula-tion of Bcl-2 activity at transcriptional, translational, do-mains, structural and post-translational level. Additionally,the information on in vitro mutagenesis studies of Bcl-2suggests various possibilities for tapping anti-apoptoticpotential of Bcl-2 for therapeutic purposes.22 On the ba-sis of this information we have hypothesized and validatedthat the cell survival enhancing mutants of Bcl-2 are likely

to decrease the undesired death of hematopoietically origi-nated stem cells, which are the main source of bone marrowtransplantation. On the contrary, the information on Bcl-2and Bax docking may help in devising a strategy to nullifythe oncogenic effects of Bcl-2 as in B-cell lymphoma.

Mechanism of action

Bcl-2 blocks the generation of reactive oxygen species andmaintains cellular redox potentials, suggesting that it op-erates in an anti-oxidant pathway.23 It also alters intra-cellular ion fluxes that occur during apoptosis includingchanges in the partitioning of Ca2+ in the endoplasmic retic-ulum, nucleus and mitochondria.24 Other reported effects ofBcl-2 on mitochondria include inhibition of mitochondrial

460 Apoptosis · Vol 11 · No 4 · 2006

Bcl-2 in apoptosis

permeability transition, a pre-apoptotic event that leads tomitochondrial membrane swelling and loss of membranepotential.25 One theory as to how the Bcl-2 family wouldcontrol the swelling of mitochondria relates to a proposedchannel entitled the permeability transition pore (PTP).26

The PTP is a large conductance pore that evolves in mito-chondria following necrotic or apoptotic signals. The PTPis permeable to solutes with a molecular mass of ∼1500 Dawhen studied in vitro. The PTP is proposed to be composedof or influenced by clustered components of the inner andouter mitochondrial membrane including hexokinase, crea-tine kinase (CK), voltage-dependant anion channel (VDAC),adenine nucleotide translocator (ANT), the peripheral ben-zodiazepine receptor (PBR), and even the mitochondrial ma-trix cyclophilin (CycD). Opening of the PTP subsequentto death stimulus results in mitochondrial depolarization,uncoupling of oxidative phosphorylation, and swelling ofthe mitochondria. Bcl-2 has been shown to regulate andmaintain the transmembrane potential across mitochondrialmembrane through PTP. Bcl-2 and its family members suchas Bcl-XL and Bax form mitochondrial membrane pores thatact as channels for ions or possibly other molecules. By struc-tural analogy to the bacterial toxins, the core of Bcl-XL andBcl-2 consists of two central alpha helices (α5 and α6) oflargely hydrophobic residues, which insert across lipid bi-layers and contribute to the formation of a mitochondrialmembrane pore.27

Recombinant forms of both Bcl-XL and Bcl-2 form ionchannels at low pH in planar lipid bilayer or synthetic lipidvesicles, which are cation selective at physiologic pH. Theacidic pH dependence of Bcl-2/Bcl-XL channel formationand requirement of acidic lipids resemble the properties ofpore-forming domains of bacterial toxins. Bax forms poresboth at acidic and neutral pH, these pores have biophys-ical characteristics distinct from Bcl-2 channels, it dis-plays a mild Cl− selectivity in contrast to the mild K+selectivity exhibited by Bcl-2. At physiological pH thechannel-forming activity of Bax is inhibited by the pres-ence of Bcl-2.28 Membrane pores formed by Bcl-2/Bcl-XL

act to heterodimerize with Bax protein so that Bax dimers(active form) may not be able to initiate apoptotic events atthe mitochondria that would otherwise lead to the cytosolicrelease of apoptosis triggering factors such as cytochrome cand AIF through PTP.29

Since Bcl-2 is present on outer membrane of ER, manyroles have been assigned to it, which helps in regulating theprocess of apoptosis. One potential model for this hypoth-esize that presence of Bcl-2 on ER sequesters “activator”BH3 only proteins, preventing them from interactingwith Bax. “Activator” BH3 only proteins are displaced by“sensitizer” BH3 only proteins, freeing them to activate Baxwhich translocates to mitochondria and induce cytochromec release. The other model depicts that the ER lumenthat serves as a source of calcium ions, releases them oninositol triphosphate (InsP3) receptor engagement with itsligand. The resulting elevation of cytosolic calcium directly

mediates loss of mitochondrial membrane potential throughincreased uptake of calcium ions into the mitochondrialmatrix, a process that is enhanced by permeabilization ofthe outer mitochondrial membrane by tBid. Indirectly Ca2+activates calcineurin (phosphatase), which in turn activatesBad. Bcl-2 located on the ER membranes interferes withcalcium mediated apoptotic signals, either by decreasing ERluminal calcium concentration or by docking calcineurinto InsP3 receptors, thereby inhibiting InsP3-mediated cal-cium release and calcineurin mediated de-phosphorylationof phosphorylated Bad.30 Under normal condition survivalfactors such as insulin like growth factor I (IGFI) andIL-3 induces the rapid phosphorylation of Bad on two Serresidues (Ser112 and Ser136). On phosphorylation of eitherof these Ser residues, Bad no longer heterodimerizes withBcl-2/Bcl-XL and instead forms a complex in the cytosolwith 14-3-3. Bad is phosphorylated at least on Ser136 bythe Ser/Thr kinase AKT (also known as PKB). The kinasefunction of AKT is activated in response to IGF-1/IL-3 bya phosphatidylinositide-3′ kinase (PI-3 kinase) dependentmechanism. Exact mechanism of Bad phosphorylationis unknown but it is significant that Ser136 is locatedimmediately adjacent to BH3 domain of Bad that mediatesbinding to Bcl-XL suggesting a mechanism for IL-3mediated inhibition of apoptosis.31 Recent findings confirmthe association and docking of Bcl-2 with calcineurin toInsP3 receptors in brain cells.

In T-cells Bcl-2 inhibits T-cell receptor (TCR)-mediatedactivation of NFAT and induction of IL-2 expression. Thisinhibits cell cycle entry into S phase by delaying G0/G1

transition and subsequently TCR-activation-induced apop-tosis. It has been suggested that Bcl-2 inhibits NFAT ac-tivation and entry into the nucleus through calcineurin bysequestering it to intracellular membranes. Translocation ofNFAT into the nucleus increases expression of IL-2, which inturn stimulates antagonistic pathways, one leading to celldeath and the other leading to cell survival. IL-2 inducescells death through Stat2-mediated induction of the Fasligand, but promotes cell survival through Akt-mediatedinduction of Bcl-2 expression. This raises the possibilitythat increased expression of Bcl-2 may be part of a feedbackloop that attenuates InsP3-mediated calcium signals andthereby controls T-cell proliferation while maintaining cellsurvival.30

Regulation of Bcl-2 activity

Transcriptional and translational regulation. Human Bcl-2gene is unusually large and comprises of three exonsextending over ≈200 kb.32 Transcription of Bcl-2 mRNAis initiated at two promoters, P1 and P2, located approxi-mately 1400 bp or 80 bp upstream of the translational startsite respectively. P1 functions as the major transcriptionalpromoter and is characterized by multiple transcriptionalinitiation sites e.g. Sp1 transcriptional factor, GC rich

Apoptosis · Vol 11 · No 4 · 2006 461

Y.K. Verma et al.

sequences but absence of a canonical TATA element. Bycontrast P2 contains a TATA element, a CCAAT box,and an octamer motif, producing only a small fractionof Bcl-2 mRNA. Between P1 and P2 there are multiplenon-overlapping negative regulatory elements, which co-localize with regions of increased DNAse I sensitivity. AnRb protein responsive element and a Cdx protein bindingsite have been identified within this region however theircontribution to regulation of transcription is unknown.The 3′-mbr of the Bcl-2 gene locus also contains a matrixattachment site (MAR), which upon t (14; 18) translocationexhibits altered cell cycle expression of the Bcl-2 gene.33 Bcl-2 gene is transcribed into three overlapping 8.5, 5.5, and 3.5kb mRNAs. Transcripts of 5.5 kb and 3.5 kb carry two over-lapping open reading frames, one of which is 717 nucleotidelong and codes for Bcl-2 α protein while the other codes forBcl-2 β protein. Bcl-2 β (205 amino-acids, molecular mass22 kD) is identical to Bcl-2 α (239 amino acids, molecularmass 26 kD) except at the carboxyl terminus.34–36

The Bcl-2 transcription by P1 promoter is positively in-duced upon activation of mature B-cells, which is mediatedby a cyclic AMP-response element (CRE) located in the5′-flanking region.37 CRE site is also important for Bcl-2expression in neuronal cells and for oestrogen receptor acti-vation of Bcl-2 in mammary cells. The activation of Bcl-2 byNF-κB in t (14; 18) translocation in lymphoma cells has beenshown to mediated through the CRE and Sp1 binding sitedof P1 promoter. In human breast carcinoma MCF-7 cells, in-creased reporter activity was observed after co-transfectionof a vector expressing Sp1 and a reporter plasmid containingthe Sp1 binding site from the Bcl-2 promoter region.38 TheP1 promoter also contains WT-1 binding sites, which act aspositive regulators in sporadic Wilm’s tumors. Interestingly,using the bimolecular interaction (BIA) technique it hasbeen recently demonstrated that c-jun also binds at two AP-1 (activating proteins) binding motifs in Bcl-2 promoter.39

c-Jun interaction initiates Bcl-2 transcription cyclically inthe glandular cells of the human endometrium in whichlevels of Bcl-2 increases gradually during the proliferativephase and disappear soon after the beginning of secretoryphase in the menstrual cycle.40 The transcription of P2 pro-moter is positively regulated by Myb family of transcriptionfactors (in chicken T cells) and Ikaros family member Alios,which on phosphorylation by Ras bind to P2 promoter inIL-2 supplemented milieu.33 However Bcl-2 gene expres-sion is repressed by the presence of a short open readingframe located just upstream of initiating methionine whichinterferes with efficient translation.41 Its expression is alsonegatively regulated by element located upstream of the P2promoter that represses transcription initiated by P1 andheterologous promoters in transient transfection assays.42

The p53 tumor suppressor protein can also suppress tran-scription of Bcl-2 through a negative regulatory elementlocalized in the promoter region.43 Irradiation induces theactivity of p53, which in turn binds to a consensus site inBax promoter (the 5′ untranslated region) and transactivates

Bax gene expression. Mice deficient in p53 exhibit reducedlevels of Bax protein and elevated levels of Bcl-2 protein incertain tissues, which validates a role for p 53 in modulatingBcl-2 expression in a physiological context.44 As discussedearlier WT-1 binding sites positively regulate transcriptionthrough P1 but it also acts as negative regulators of Bcl-2 expression in t (14; 18) translocation and HeLa cells. Inaddition to P1 and P2 promoter binding sites there are se-ries of eight DNAse I hypersensitive sites (HSS1-8), whichregulates Bcl-2 transcription. HSS1-3 are located in the 5′flanking sequence of P1, HSS5-7 are found in between P1and P2. HSS8 is mapped approximately at the junction ofexon 2 and intron 2 occurring only in REH cell line and con-tains three Myb-binding sites, activated by B-Myb. HSS1-7are found in both the Bcl-2 expressing REH and in a Bcl-2non-expressing DHL9 B-cell lines.33

The stability of Bcl-2 mRNA also plays a role in down-regulating its expression during apoptosis. It is endowedwith an adenine- and uracil-rich element (ARE) whose in-sertion downstream from stable β-globin mRNA causes anenhanced decay of β-globin transcript. The pathway leadingto the modulating activity of Bcl-2 ARE is influenced byPKC as observed on the addition of DAG and TPA, whichmarkedly attenuates the Bcl-2 ARE destabilizing potential.Conversely it is noteworthy that when C2-ceramide (apop-tosis inducer) is added to the culture medium the β-globintranscript containing the Bcl-2 ARE insertion sequence un-dergoes a dramatic increase in decay.45

Domain specific regulation. Bcl-2 is a 239 amino acid longprotein, which has been divided functionally into Bcl-2 ho-mology (BH) domains viz., BH4, BH3, BH1, BH2, an Xdomain and a regulatory loop domain46 (Figure 1).

It’s BH1 and BH2 domains along with BH3 domain forma solvent accessible hydrophobic BH3 receptor cleft essentialfor heterodimerization with pro-apoptotic proteins throughtheir amphipathic BH3 domain.47 The mutational data inBcl-XL suggests a lock and key type interaction whereinBH1 and BH2 domains within the anti-apoptotic proteincombine to act as a receptor for BH3 domain presentedby the pro-apoptotic family members. The core of Bcl-XL isformed by two α helices of largely hydrophobic amino acids,which are arranged in anti-parallel fashion and encompassedby five amphipathic helices. Sequences within the BH1,BH2 and BH3 domains including highly conserved coreresidues are juxtaposed that contribute to the formation of asolvent exposed hydrophobic groove in Bcl-XL. The solutionstructure of Bcl-XL in complex with 16 amino acids peptide(corresponding to the Bak BH3 domain) indicates that thiscleft acts as a docking site for the BH3 domain of pro-apoptotic homologs. The BH3 peptide in its bound formis α helical that lies across the elongated cleft formed byBH1, BH2 and BH3 residues in Bcl-XL. The complex isstabilized by hydrophobic interactions with apolar residues

462 Apoptosis · Vol 11 · No 4 · 2006

Bcl-2 in apoptosis

Figure 1. Bcl-2 protein line diagram showing Bcl-2 Homology domains (BH1, BH2, BH3 and BH4 domain), X domain, transmembrane domain(TM) and regulatory domain (FLD). The alpha helices predicted in Bcl-2 secondary structure are shown as α1 to α8. The interaction of eachdomain with the respective protein is shown by an arrow where BH4 domain interacts with Raf-1, Bag-1 and calcineurin; FLD domain interactswith JNK-1, PP2A and PKC; BH1, BH2 and BH3 domains form BH3 receptor cleft for interaction with pro-apoptotic proteins and TM acts asan membrane anchor to membranes of mitochondria, endoplasmic reticulum and nucleus.

that are highly conserved in BH3 domains as well as byelectrostatic interactions between charged side chains.48

By analogy to Bcl-XL structural detail and interac-tion with Bak peptide, studies have shown that BH1and BH2 domains in Bcl-2 are essential for its co-immunoprecipitation with Bax and for prolongation of cellsurvival. It has been shown that the residues viz. Gly145,Arg146 and Trp188 form an active site in Bcl-2 and itsstructural homologs.49 Substitution of Gly145 (in BH1domain) and Trp188 (in BH2 domain) with Ala disruptsthe pore forming ability of Bcl-2 protein24 and completelyabrogates its heterodimerization and consequently death re-pressor activity in IL-3 deprivation, gamma irradiation andglucocorticoid-induced apoptosis.50,51 Similarly, the substi-tution of Gly138Ala (Gly145 in Bcl-2) and Arg139Gln(Arg146 in Bcl-2) in Bcl-XL alters the accessibility andbinding properties of BH3 receptor cleft of Bcl-XL to pro-apoptotic proteins. The computer simulated Bcl-2 proteinto Bax’s BH3 domain docking has revealed that active siteresidue Arg146 of Bcl-2 forms hydrogen bonds with Asp68and Asp71 of Bax peptide. Similar interactions were ob-served in Bcl-XL and Bak peptide complex, where interac-tion between Asp83 of Bak and Arg139 of Bcl-XL stabi-lizes the complex formation. The Asp83 residue, which iscompletely conserved within Bcl-2 family, when substitutedwith Ala in Bak peptide markedly reduces the binding ofpeptide to Bcl-XL. Moreover conserved Arg139 mutationto Glu in Bcl-XL inhibits its anti-apoptotic activity andbinding to Bak protein.52 Bax mutant Asp68Ala has beenshown to retain the ability to homodimerize but failed tointeract with Bcl-2 as determined by yeast two-hybrid as-says and co-immunoprecipitation analysis using transfected293 mammalian cells. The co-expression of wild type Bcl-2with Bax mutant, Asp68Ala, rescues cells from apoptosisthus indicates the importance of Asp68 in heterodimer-ization interaction and induction of apoptosis by inhibit-

ing Bcl-2 cell survival potential.53 The other active siteresidue in Bcl-2, Trp188, docks with Lys64 of Bax peptidewhereas small Gly145 provide the space to accommodateAsp68 within the groove formed by Gly145 and Arg146.Therefore, it has been suggested that any other amino acidin place of Gly145 would sterically inhibit the entry ofAsp68 required to stably heterodimerize Bax protein withBcl-2.49

The N-terminal BH4 domain mediates the interactionof Bcl-2 with Raf-1 (protein kinase) and calcineurin (phos-phatase). Bcl-2 in association with Raf-1 induces phosphory-lation of Bad, unphosphorylated Bad abrogates the cytopro-tective functions of Bcl-2. The phophorylated Bad (whichunlike Bcl-2 has no membrane-anchoring domain) then dis-sociates from Bcl-2, moves into the cytosol in complex with14-3-3, where it cannot interfere with Bcl-2. Although Badis one potential substrate of Bcl-2/Raf-1, there exist otherunknown targets as artificially targeting active Raf-1 to mi-tochondria can also suppress apoptosis in cells that do notcontain Bad protein.54,55 The binding of Raf-1 with Bcl-2takes place after being phosphoryalted by BAG-1 protein.BAG-1 collaborates with Bcl-2 and provides a mechanismfor activating Raf-1 locally at mitochondria and other Bcl-2containing membranes in a Ras-independent manner. Theother protein that interacts with Bcl-2 at BH4 domainis Ca2+ dependent protein phosphatase, calcineurin. Cal-cineurin was recently demonstrated to dephosphorylate Bad,allowing it to translocate to mitochondria. Over-expressionof calcineurin induces apoptosis in Bcl-2 suppressible man-ner during Ca2+ mediated apoptosis. Bcl-2 redistributescalcineurin from the cytosol to intercellular membranes andprevents its interaction with phosphorylated NFAT andother substrates of calcineurin in the cytosol. The Bcl-2/calcineurin interaction is also relevant to the inhibition ofcell proliferation and translocation of NFAT into the nucleusby Bcl-2.56,57 The NFAT inducible genes are important for

Apoptosis · Vol 11 · No 4 · 2006 463

Y.K. Verma et al.

proliferation in some types of cell such as T-lymphocytes.The slow entry of these cells into the cell cycle by Bcl-2 de-creases their vulnerability to apoptotic stimuli.58 It has alsobeen demonstrated that calcium activated protease calpaincleave Bax although physiological significance of this eventis unclear.

The finding that Bcl-2 α is an active inhibitor of celldeath and is membrane bound whereas Bcl-2 β is inactiveand cytosolic indicates that the carboxy terminus contributesto the survival activity of Bcl-2. This region contains an Xdomain and a hydrophobic transmembrane domain (TM)(Figure 1). Site directed mutagenesis studies have revealedthat both domains are not only novel dimerization motifsfor the interaction of Bax with Bcl-2 but also crucial for thesurvival activity of Bcl-2. However later it was found outthat membrane attachment and therefore the hydrophobicstretch is not required for the survival activity of Bcl-2 butpart of domain X is non-redundant.1 Its interaction and roleplayed in suppressing apoptosis is yet to be elucidated.

The flexible loop domain (FLD) lying in between BH4and BH3 domains lacks a defined structure and is not con-served among Bcl-2 family members. FLD deleted mutantof Bcl-XL and Bcl-2 displayed enhanced ability to inhibitapoptosis without impairing their heterodimerization withpro-apoptotic proteins.59 FLD deleted mutant also displayeda quantitative difference in its ability to inhibit apopto-sis. Full length Bcl-2 was unable to prevent anti-IgM in-duced cell death of the immature B cell line WEHI-231.In contrast, the mutant protected WEHI-231 cells fromdeath. Substantial differences were observed in the abilityof WEHI-231 cells to phosphorylate the mutant comparedwith full length Bcl-2.60

Structural regulation. The structure of two isoforms of Bcl-2(α and β) that differ by two amino acids was determinedby NMR spectroscopy. Because wild-type Bcl-2 behavedpoorly in solution, the structures were determined by usingBcl-2/Bcl-XL chimeras in which part of the putative un-structured loop of Bcl-2 was replaced with a shortened loopfrom Bcl-XL. These chimeric proteins have a low pI and aresoluble compared with the wild-type protein. The structureof the two Bcl-2 isoforms consists of 6 α helices with a hy-drophobic groove on the surface similar to that observed forthe homologous protein Bcl-XL. Comparison of the Bcl-2structures to that of Bcl-XL reveals that the overall fold ofthese proteins is same however there are differences in thestructural topology and electrostatic potential of the bind-ing groove. Although the structures of the two isoforms ofBcl-2 are virtually identical but differences were observedin the ability of the proteins to bind to a 25-residue pep-tide from the pro-apoptotic Bad protein and a 16-residuepeptide from the pro-apoptotic Bak protein. These resultssuggest that there are subtle differences in the hydrophobicbinding groove in Bcl-2 that govern differences in anti-apoptotic activity for the two isoforms.61 Similar types ofinteractions were observed in other experimentally solved

complexes such as 1BXL, 1PQ1 and 1G5J. Although thereis a difference in binding affinity of Bcl-2 family proteinstowards pro-apoptotic Bax, Bad, Bim and Bak peptides,due to difference in residues lining the hydrophobic recep-tor cleft,62,63 however, the position of predicted active siteresidues remain conserved sequentially as well as structurallyin each of Bcl-2 structural homologs. The residues formingthe active site provides the basic structural skeleton ontowhich pro-apoptotic proteins sit whereas the interactionsin between other residues decide the specificity and efficacyof Bcl-2 family members for heterodimerization.49 For ex-ample, the three-dimensional structure of 1Q59 does notcontain prominent hydrophobic groove that mediates bind-ing to pro-apoptotic family members but it does binds toBax, Bak and to Bad with low affinity. This binding maybe attributed to the predicted active site residues, which arestructurally conserved in 1Q59 protein also.64

Out of 6 α-helices the two core helices viz., α5 and α6in Bcl-2α form two central hydrophobic anti-parallel chan-nels in the mitochondria, endoplasmic reticulum and nu-clear membrane similar to the bacterial toxin regulatingCa2+ flux and protein translocation (p53-BP2 translocationacross nucleus). These helices presumably insert perpendic-ularly across the lipid bilayer with the protein undergoingprofound conformational changes, perhaps opening like anumbrella, with the surrounding amphipathic helices fold-ing upwards and resting on top of the membrane. Consistentwith this idea removal of α5 and α6 from Bcl-2 preventschannel formation in vitro and abolishes its anti-apoptotic ef-fect in cell (Figure 2). The helices α3, α4, α6 and α7 encom-passing BH4 and BH3 domains, form flanking amphipathichelices creating a hydrophobic cleft, an interface essentialfor interaction of pro-apoptotic proteins through their BH3

Figure 2. Bcl-2 α three-dimensional model obtained by homologymodeling at SWISS-MODEL server,65 depicting α helices, one un-structured loop, Bcl-2 homology domains and an X domain. Thehomology model was obtained to identify the active site on Bcl-2since only chimeric NMR model of Bcl-2 is solved.66

464 Apoptosis · Vol 11 · No 4 · 2006

Bcl-2 in apoptosis

amphipathic α-helical domain. For example, the BH3 do-main only molecule Bim which is localized on tubule-associated dynein motor complex (through association withLC8 dynein light chain) translocates to mitochondria andinteract at BH3 receptor cleft of Bcl-2 following apoptoticstimulus.67

The N-terminal region of Bcl-2 folds in an α helical con-formation in the presence of SDS micelle. The structure waspredicted by the computerized secondary structure predic-tion and circular dichroism spectroscopy of synthetic Bcl-2peptide. Systematic mutagenesis of this region shows thatonly five hydrophobic and aromatic residues (Ile 14, Val 15,Tyr 18, Ile 19, Leu 23) within it are specifically required forits ability to inhibit apoptosis. These five residues are clus-tered on one face of this helix mediating the protein-proteininteraction of the apoptotic pathway. Similar helical motifis present in other anti-apoptotic proteins only and not inthe pro-apoptotic proteins.19

By using two-dimensional peptide mapping and sequenc-ing three residues (Ser70, Ser87 and Thr69) within FLD ofBcl-2 were found to be phosphorylated in response to micro-tubule damaging agent (paclitaxel), which also arrest T-cellsat G2/M. Changing these sites to Ala conferred more anti-apoptotic activity on Bcl-2 following physiological deathsignals as well as paclitaxel treatment indicatingthat hyper-phosphorylation is inactivating. An examination of cyclingcells enriched by elutriation for distinct phases of the cellcycle revealed that Bcl-2 gets phosphorylated at the G2/Mphase of the cell cycle. It was noted that ASK1 and JNK1were normally activated at G2/M phase and JNK is capableof phosphorylating Bcl-2. Expression of a series of wild typeand dominant-negative kinase indicated that ASK1/Jun N-terminal protein kinase 1 (JNK1) phosphorylates Bcl-2 invivo. Moreover the combination of dominant negative ASK1,(dnASK1), dnMKK7 and dnJNK1 inhibited paclitaxel-induced Bcl-2 phosphorylation. Similar effect was observedfor FLD deleted mutant of Bcl-2, which completely in-hibits paclitaxel-induced apoptosis.68 This study suggeststhat FLD is the potential site for stress response kinases,which hyper-phosphorylate Bcl-2 during cell cycle progres-sion as a normal physiologic process to inactivate Bcl-2 atG2/M.69

On the contrary, a single phosphorylation of Bcl-2 atS70 is essential for its full and potent cell survival activity.Bcl-2 protein gets phosphorylated by IL-3, PKC and ery-thropoietin on receiving signal in the normal conditions tomaintain cellular homeostasis.70 The Ser70Ala Bcl-2 mu-tant fails to be phosphorylated after IL-3 or Bryo stimula-tion and is unable to support prolonged cell survival eitherupon IL-3 deprivation or etoposide treatment. In contrasta Ser70Glu substitution suppresses the etoposide-inducedapoptosis more potently than wild type Bcl-2 in the absenceof IL-3. JNK-1 (SAK) was found to be latently activated onIL-3 withdrawal, which phosphorylates Bcl-2 at Ser70 andco-localizes with it.71 This data suggests that JNK-1 is thepotent Bcl-2 kinase which upregulates/downregulates Bcl-2

activity by phosphorylation/hyper-phosphorylation respec-tively depending upon the cell survival/ death signal.

Based upon the above studies Bcl-2 and other anti-apoptotic proteins are assigned as BH3 buried, this indi-cates that the hydrophobic face of Bcl-2 (containing BH3domain) is buried in the protein but on receiving deathstimulus it gets converted to an active conformation by ex-posure of the BH3 domain and potentially hydrophobic sur-faces. This hypothesis may be consolidated by the fact thatcleavage of amino-terminus of Bcl-2 (predicted to exposeBH3 domain surface) can convert them into pro-apoptoticmolecules whose BH3 domain is always exposed (BH3exposed).72

Post-translational regulation. The phosphorylation of Bcl-2has been a dynamic process, which involves both kinase andphosphatases, and a mechanism likely exists to rapidly andreversibly regulate its activity and functional consequences.The co-localization or translocation of the Bcl-2 with ki-nases (PKC and ERK1/2) and phosphatase (PP2A) fromcytosol to mitochondrial membrane explains the rapid andreversible regulation of Bcl-2 phosphorylation.73–78 Stud-ies have revealed that PP2A is a physiologic Bcl-2 phos-phatase since okadic acid and sphingolipid-ceramide sensi-tize PP2A and co-localize with Bcl-2 at the mitochondrialmembrane.79,80 The molecular interaction of Bcl-2 and Bcl-XL with caspase 3 (Cysteine rich Aspartyl Proteases) affectstheir anti-apoptotic induction by cleaving Bcl-2 at Asp34and Bcl-XL at Asp64. Caspases recognize tetra-peptide mo-tifs and cleave their substrates on the carboxyl side of anAsp residue (P1 position). Individual caspases have distinctsubstrate specificity because they recognize different aminoacids at three residues directly N-terminal of the Asp (P2,P3 and P4 positions).81 The caspase-3 mediated cleavageconverts anti-apoptotic Bcl-2 into a Bax like pro-apoptoticprotein that heterodimerizes with wild type Bcl-2 and in-duces apoptosis.82 These mechanism, dephosphorylation anddegradation, exist to downregulate the Bcl-2 function whenit is not required to maintain the cellular homeostasis.

Pathways to apoptosis insensitive to Bcl-2. Bcl-2 and its anti-apoptotic family members such as Bcl-XL are unable toblock all physiologically induced pathways to apoptosis inmammalian cells. For example, Bcl-2 does not prolong thesurvival of mutant SCID and rag−/− thymocytes, which donot express T cell antigen receptor (TCR) chains and neitherBcl-2 nor Bcl-XL can block deletion of self-antigen-specificimmature B cells and T cells. Members of the tumor necrosisfactor receptor (TNF-R) family play critical roles in apop-tosis of lymphocytes that have been chronically activatedvia their antigen receptors. Ligation of the TCR/CD3 com-plex on proliferating normal T cells or T lymphoma linesinduces expression of CD95 ligand (FasL/APO-1L) and itsreceptor CD95 (Fas/APO-1). This triggers apoptosis by au-tocrine or paracrine CD95 stimulation83 and this death can-not be blocked by Bcl-2. Experiments in some transformed

Apoptosis · Vol 11 · No 4 · 2006 465

Y.K. Verma et al.

Table 2. ∗FLD: flexible loop domain; BH: Bcl-2 Homology domain; BH3: Bcl-2 Homology domain 3; BH4: Bcl-2 Homology domain 4.

S. noResidue(s) mutated(Domain)

Type ofmutation(point/deleted) Normal function

Effect of mutation with respect toanti-apoptotic activity

1 Asp34Ala (∗FLD) Point Cleavage by caspase-3 Loss of interaction with caspase-3, activityenhanced

2 Ser70Ala (∗FLD) Point Required for normal survival function Abrogation of survival function

3 Ser70Glu (∗FLD) Point Required for normal survival function Suppress apoptosis in the serum deprivedconditions

4 Gly145Ala (∗BH3receptor cleft)

Point Homodimerization, andheterodimerization withpro-apoptotic proteins

Loss of heterodimerization,homodimerization intact, complete loss ofactivity in IL3 deprivation, gammairradiation, glucocorticoid inducedapoptosis

5 Trp188Ala (∗BH3receptor cleft)

Point Homodimerization, andheterodimerization withpro-apoptotic proteins

Loss of heterodimerization,homodimerization intact, complete loss ofactivity in IL3 deprivation, gammairradiation, glucocorticoid inducedapoptosis

6 Val93Iso (BH3) Point Unknown Non-Hodgkin’s lymphoma

7 Iso14Ala (∗BH4) Point Required for multimerization of Bcl-2and unknown components ofapoptotic pathway

Decreased anti-apoptotic activity

8 Val15Ala (∗BH4) Point Required for multimerization of Bcl-2and unknown components ofapoptotic pathway

Decreased anti-apoptotic activity

9 Yal18Ala (∗BH4) Point Required for multimerization of Bcl-2and unknown components ofapoptotic pathway

Decreased anti-apoptotic activity

10 Iso19Ala (∗BH4) Point Required for multimerization of Bcl-2and unknown components ofapoptotic pathway

Decreased anti-apoptotic activity

11 Leu23Ala (∗BH4) Point Required for multimerization of Bcl-2and unknown components ofapoptotic pathway

Decreased anti-apoptotic activity

12 ∗BH4 Deletion Required for interaction with Raf-1(phosphatase), Bag-1 andcalcineurin (phosphatase)

Loss of activity

13 ∗Flexible loop Deletion Contains many phosphorylation sitesregulating Bcl-2 anti-apoptoticactivity

Enhanced cell survival potential

14 Outside ∗BH domains Deletion - - - - - - - - - - - - - No effect

15 Transmembrane region Deletion For insertion into membranes Not as potent anti-apoptotic as wild type

16 X domain Deletion Required for survival function Loss of activity

17 α5 and α6 Deletion Required for the ionic channelformation in mitochondrial, ER,and nuclear membrane

Abrogation of anti-apoptotic activity

cell lines (e.g. L929, WEHI 164, NIH 3T3 or 293 T)demonstrated that none of the anti-apoptotic proteins Bcl-2,Bcl-XL, Bcl-W and adenovirus protein E1B19 kDa affordedany protection against apoptosis triggered by ligation of p55TNF-RI, DR3 and possibly other members of the TNF-Rfamily. Since Bcl-2 and its family proteins are not localizedto the plasma membrane, the site where caspase activationoccurs after ligation of CD95 (and possibly also after crosslinking of related receptors),84 it therefore makes sense that

Bcl-2 and its homologs cannot block this pathway to apop-tosis.

It is also suggested that the cleavage of Bcl-2 by capase-3/6/7/9 on TNFα induced cell death is responsible for Bcl-2inability to inhibit cells death in murine pro-B-cell lineFL5.12.85 But no difference was observed in between wildtype Bcl-2 and its cleavage resistant mutant (Asp34Ala)in combination with caspase inhibitor (zVAD-fmk) to re-sist TNF-α-induced apoptosis. Therefore it was suggested

466 Apoptosis · Vol 11 · No 4 · 2006

Bcl-2 in apoptosis

that Bcl-2 cleavage is not important for the inhibition ofTNF-α-induced cell death but do not preclude an involve-ment in a post commitment phase of apoptosis. Simulta-neously it was discovered that there was cleavage of Bcl-2on TNF-α-induced cell death in endothelial cells due toubiquitin-dependent proteosome complex. This cleavage oc-curs at putative mitogen-activted protein (MAP) kinase siteson Bcl-2 protein (Thr56, Thr74, Ser87), mimicking phos-phorylation of this site abolishes degradation, suggesting alink between the MAP kinase pathway to the proteosomepathway.19 These data suggest that the pathways which areinsensitive to Bcl-2 regulation, firstly inactivate/cleave (bycaspase or ubiquitin) this protein so that it doesn’t interferewith the normal process of cell death.

In vitro mutagenesis studies of Bcl-2

As discussed earlier the point mutations Gly145Ala andTrp188Ala, in BH1 and BH2 domains respectively, severelyimpair Bcl-2 cell survival function. In addition anti-apoptotic activity is lost as a result of deletion of the BH4region. Alterations in the corresponding regions withinBcl-XL

86 had similar consequences. Surprisingly point mu-tations of highly conserved residues in BH4 had no effecton cell survival. Mutations in Bcl-2 outside the BH do-mains generally had no effect on anti-apoptotic activity.The exceptions known to date are the mutation of non-conserved Ser residues (Ser70), which abolishes survivalfunction and that deletion of FLD, which increases the cellsurvival activity of Bcl-2.87 Various Bcl-2 mutants havingpositive/negative effect over its cell death suppressing activ-ity are listed in the Table 2. However there are mutations inBcl-2, which have insignificant effect on its function, are notlisted.

Discussion and future direction

The anti-apoptotic nature of prototypic Bcl-2 is responsi-ble for various undesired and desired cellular consequences.There is a delicate balance exists in between pro- and anti-apoptotic proteins which determines cellular fate. This anti-apoptotic nature of Bcl-2 protects cells from plethora of cel-lular insults, which are threatening to cell. However underspecial circumstances Bcl-2 is unable to block the apop-tosis, in addition, unregulated Bcl-2 expression leads tolymphoma. This multifaceted character of Bcl-2 to inter-act with variety of cellular targets such as calcineurin, p53,NFAT etc. makes it an important target to tap its potentialtherapeutically. On the basis of above discussion the celldeath suppressing potential of Bcl-2 and its mutants maybe explored to either suppress or induce apoptosis depend-ing upon the circumstantial requirement. We will discusshere the implications of both of these cell fate-determiningproperties of Bcl-2.

Figure 3. Optimum activation of caspase-9 after 24 h of irradiation(1.5 Gy). The cells were gamma irradiated at dose rate of 0.318Gy/min in “Gamma Cell 220” machine. The X axis represents the time(in hours) and Y axis the O.D. at 405 nm. The O.D. was measuredfor the cleavage of caspase-9 substrate in caspase-9 detection kit(Calbiochem). The sample was prepared according to the instructionmanual. The activated caspase-9 cleaves its substrate (LEHD-pNA),which generates color readable by light spectrophotometer at 405nm wavelength (unpublished observations).

Implications in cell death suppression

Experiments in which Bcl-2 was over-expressed in growthfactor-dependent cell lines gave the first clue to its functionthat it protects IL-3-dependent murine hematopoetic celllines from cytokine withdrawal. The surviving cells ceaseproliferation and accumulate in the quiescent G0 state ofthe cell cycle.88 Over-expression of Bcl-2 prolongs survivalof myeloid lineage of transgenic mice (lacking M-CSF/CSF-1 factor), facilitates89 T-cell development in mice lackingIL-7 receptors and promotes accumulation of B-cells.90 Inaddition to over-expression mediated cell death suppres-sion, it inhibits cell death induced by many other apoptoticsignals that are relevant to the evolution and/or chemore-sistance of tumor cells. Its elevated expression provides pro-tection against ionizing radiation and many commonly usedchemotherapeutic drugs, conferring a multidrug resistantphenotype to cells.91 Bcl-2 expressing cells exhibit enhancedprotection against apoptosis induced by hypoxia and loss ofsubstrate attachment,92 particularly important to the sur-vival of metastatic tumor cells.

Bcl-2 cell death suppressing activity has been shown toenhance life span of various cellular parts of hematopoi-etic compartment. Its systemic over-expression in thehematopoietic system protects transgenic mice from theconsequences of ionizing irradiation.93 It plays a criticalrole in the development of lymphoid progenitor cells fromthe Hematopoietic Stem Cells (HSC).94 It also supports theoptimal proliferation of progenitor HSC. The role of StemCell Factor (SCF/c-kit) in apoptosis could be replaced byits over-expression. Cytokine withdrawal induced apoptosiswas dramatically blocked by Bcl-2 in TF-1 cells according toDNA fragmentation, morphological observation and viable

Apoptosis · Vol 11 · No 4 · 2006 467

Y.K. Verma et al.

cell counting. These cells show a significant proliferation inresponse to IL-5 that was comparable to the proliferationactivity obtained by SCF co-stimulation or IL5Rα over-expression. These cells were not factor-independent, how-ever, they survived for a longer period of time and re-enteredproliferation rapidly in the presence of cytokines. Strongsuppression of apoptosis by exogenous over-expression ofBcl-2 in TF-1 cells suggests an important role in the reg-ulation of cytokine withdrawal induced apoptosis.95 Thesefeatures make Bcl-2 an interesting molecule to study andapply its cell death suppressing potential in unwanted deathof HSC on growth factor withdrawal and therapeutic irradi-ation.

On the basis of these studies there is a possibility to useBcl-2 as a therapeutic gene to suppress apoptosis in HSC.However its unregulated expression is a cause of concern,it causes lymphoma generation in vitro as well as in vivo.Therefore in view of the transformation potential of Bcl-2any protocol to introduce such gene into HSC must includefail-safe methods to either remove/completely block expres-

sion in genetically engineered cells. Among the possiblestrategies that can be used is the Cre/loxP system. Cre, un-der control of an inducible promoter could be obtained onthe same constructs as the candidate gene, flanked by loxP.Such a construct could also contain a negative selectablemarker (such as thymidine kinase) to ablate the cells thathave not lost inserted construct.22 On the same ground theoncogenic mutants of Bcl-2 such as Asp34Ala, Ser70Gluand �FLD may be therapeutically inducted for inhibitingthe unwanted HSC death under stress conditions such asserum deprivation and irradiation.

To validate presumed strategical consequences we gener-ated site-directed mutants of Bcl-2 at Asp34 and Ser70 toAla and Glu respectively. To regulate their expression thesemutants were subcloned into inducible mammalian expres-sion system also known as Tet-On system.96 The controlledexpression of Bcl-2 mutants enabled us to regulate geneexpression in a precise, reversible and quantitative man-ner to the varying concentration of the inducer (doxycy-cline). The effect of these mutants was evaluated in human

Figure 4. Suppression of caspase-9 induction after 1.5 Gy of irradiation by overexpressing Bcl-2 mutants {Asp34Ala (a) and Ser70Glu (b)}at 1 ug/ml concentration of inducer. The X-axis represents radiation dose and the Y-axis corresponds to the caspase-9 activation measuredon visible light spectrophotometer at wavelength of 405 nm. (a) The mutant Asp34Ala shows 50% decrease in the caspase-9 activation withrespect to wild type Bcl-2 over-expression whereas (b) Ser70Glu mutant shows 68% decrease in caspase-9 activation with respect to wildtype Bcl-2 over-expression at the same inducer concentration (unpublished observations).

468 Apoptosis · Vol 11 · No 4 · 2006

Bcl-2 in apoptosis

hematopoietic stem cell line KG-1a.97 We standardized theoptimum radiation dose for apoptosis induction in KG-1acells. The optimum dose was found out to be 1.5 Gy, whichwas sufficient to induce apoptosis after 24 h (unpublishedobservations) (Figure 3). The induction of caspase-9 waschosen as a parameter for detecting apoptosis. Caspase-9 be-longs to the family of caspase-3, it gets activated throughthe binding of Apaf-1 (apoptotic protease activating factor1), cytochrome c and ATP (together known as apoptosomecomplex). Apaf-1 and cytochrome c is released from mito-chondria when its transmembrane potential gets perturbeddue to pores formed by the over-expression of Bax protein(pro-apoptotic) induced upon irradiation. Caspase-9 in turnconverts zymogen/inactive form pro-caspase-3 to active formcaspase-3, which subsequently activates Caspase ActivatedDNase (CAD) by degrading its inhibitory subunit (ICAD).The CAD protein is actually responsible for cleaving thecellular DNA.98

The regulated over-expression of mutant Asp34Ala shows50% decrease in caspase-9 activation whereas Ser70Glu mu-tant regulated over-expression is responsible for 68% de-crease in caspase-9 activity (Figure 4). The higher cell deathsuppressing activity of Ser70Glu mutant is attributed to thepresumed enhanced heterodimerization between this mu-tant and pro-apoptotic protein Bax.99 The maximum sup-pression in caspase-9 induction was observed at the 1 ug/mldose of inducer (doxycycline) (unpublished data). These re-sults suggest that these mutants are likely candidates for thesuppression of undesired apoptosis (by tapping their onco-genic potential) in human HSC, which have implications inbone marrow transplantation.22

Implications in enhancing cell death

The wide spread involvement of Bcl-2 in cancer has mo-tivated efforts to develop drugs that inhibit the functionof Bcl-2 or counteract its deregulation thus restoring sus-ceptibility to apoptosis in tumor cells. It may be possibleto develop agents that inhibit the hyper-phosphorylationof Bad that may occur in tumor cells, or molecules thatmimic the antagonistic effect of heterodimerization withBH-3 containing pro-apoptotic Bcl-2 homologs. Such BH3mimics could bind the BH-3 receptor cleft in Bcl-2 and an-tagonize its function by preventing interaction with othercell death regulatory proteins. A more direct approach ofinhibiting the expression of Bcl-2 with anti-sense oligonu-cleotide is already being tested in clinical trials for treatmentof lymphoma, suggestive of an anti-tumor response in somepatients.100 The anti-sense oligonucleotide therapy may bemade more specific by using the recently derived informa-tion in which Gly145, Arg146 and Trp188 residues werefound to form an active site in Bcl-2 and its family members.The antisense may be directed against these residues alongwith Bcl-2 specific residues in BH3 receptor cleft so that theantisense has more affinity and specificity towards tumor in-

ducing Bcl-2. The same antisense may also be labeled withradioisotopes for use in various types of tumors imaging inwhich Bcl-2 is over-expressed.

Acknowledgment

We are thankful to Director, Institute of Nuclear Medicineand Allied Sciences, Lucknow Road, Timar Pur, Delhi-110054, India and Dr. Vani Brahamchari of Dr. B. R.Ambedkar Centre for Biomedical Research (ACBR), Uni-versity of Delhi, Delhi-110007, India, for their support.Authors also thank DRDO for project financial support. Mr.Yogesh Kr. Verma in particular thanks Council for Scientificand Industrial Research (CSIR), India, for awarding researchfellowship for PhD from ACBR.

References

1. Cleary ML, Smith SD, Sklar J. Cloning and structural analysisof cDNAs for Bcl-2 and a hybrid Bcl-2/immunoglobulin tran-script resulting from the t (14;18) translocation. Cell 1986; 47:19–28.

2. Reed JC. Bcl-2 and the regulation of programmed cell death. J.Cell Biol 1994; 124: 1–6.

3. Adams JM, Cory S. The Bcl-2 protein family: Arbiters of cellsurvival. Science 1998; 281: 1322–1326.

4. Green DR, Reed JC. Mitochondria and apoptosis. Science 1998;281(5381): 1309–1312.

5. Gross A, McDonnell JM, Korsmeyer SJ. Bcl-2 family membersand the mitochondria in apoptosis. Genes Dev 1999; 13: 1899–1911.

6. Budihardjo I, Oliver H, Lutter M, Luo X, Wang X. Biochemicalpathways of caspase activation during apoptosis. Annu Rev CellDev Biol 1999; 15: 269–290.

7. Heiden VMG, Thompson CB. Bcl-2 proteins: Regulators ofapoptosis or of mitochondrial homeostasis? Nat Cell Biol 1999;1(8): E209–E216.

8. Hengartner MO. The biochemistry of apoptosis. Nature 2000;407: 770–776.

9. Strasser A, O’Connor L, Dixit VM. Apoptosis signaling. AnnuRev Biochem 2000; 69: 217–245.

10. Borner C, Martinou I, Mattmann C. The protein Bcl-2α doesnot require membrane attachment, but two conserved do-mains to suppress apoptosis. J Cell Biol 1994; 126: 1059–1068.

11. Yang X, Tang SC, Pater A. Cloning and characterization ofhuman BAG-1 gene promoter: Upregulation by tumor-derivedp 53 mutants. Oncogene 1999; 18: 4546–4553.

12. Naumouski L, Cleary ML. The p53 binding protein 53BP2 alsointeracts with Bcl-2 and impedes cell cycle progression at G2/M.Mol Cell Biol 1996; 16(7): 3884–3892.

13. Deng X, Ito T, Carr B, Mumby M, May Jr. WS Reversiblephosphorylation of Bcl-2 following interleukin 3 or bryostatin1 is mediated by direct interaction with protein phosphatase 2A.J Biol Chem 1998; 273: 34157–34163.

14. Reed JC. Double identity for proteins of the Bcl-2 family. Nature1997; 387: 773–776.

15. Cheng EH, Kirsch DG, Clem RJ. Conversion of Bcl-2 to aBax-like death effector by caspases. Science 1997; 278: 1966–1968.

16. Bryan WJ, Lawrence HB. Bcl-2 and caspase inhibition cooper-ate to inhibit tumor necrosis factor-α-induced cell death in a

Apoptosis · Vol 11 · No 4 · 2006 469

Y.K. Verma et al.

Bcl-2 cleavage-independent fashion. J Biol Chem 1999; 274(26):18552–18558.

17. Clem RJ, Cheng EHY, Karp CL. Modulation of cell death byBcl-XL through caspase interaction. PNAS USA 1998; 95: 554–559.

18. Nicholson DW, Thornberry NA. Caspases: Killer proteases.Trends Biochem Sci 1997; 22: 299–306.

19. Dimmler S, Breitschopf K, Haendeler J, Zeiher AM. Dephos-phorylation targets Bcl-2 for ubiquitin-dependent degradation:A link between the apoptosome and the proteosome pathway. JExp Med 1999; 189(11): 1815–1822.

20. Ruvolo PP, Deng X, Ito T, Carr BK, May WS. Ceramide inducesBcl-2 dephosphorylation via a mechanism involving mitochon-drial PP2A. J Biol Chem 1999; 274(29): 20296–20300.

21. Shibasaki F, Kondo E, Akagi T, Mckeon F. Suppression of signal-ing through transcription factor NFAT by interaction betweencalcineurin and Bcl-2. Nature 1997; 386: 728–731.

22. Domen J, Weissman I. Self-renewal, differentiation or death:Regulation and manipulation of hematopoietic stem cell fate.Molecular Medicine Today 1999; 5(5): 201–208.

23. Hockenbery DM, Oltvai ZN, Yin XM, Milliman CL, KorsmeyerSJ. Bcl-2 functions in an antioxidant pathway to prevent apop-tosis. Cell 1993; 75: 241–251.

24. Baffy G, Miyashita T, Williamson Jr. Apoptosis induced bywithdrawal of interleukin-3 (IL-3) from an IL-3-dependenthematopoietic cell line is associated with repartitioning of intra-cellular calcium and is blocked by enforced Bcl-2 oncoproteinproduction. J Biol Chem 1993; 268(9): 6511–6519.

25. Marchetti P. Mitochondrial permeability transition is a centralcoordinating event of apoptosis. J Exp Med 1996; 184:1155–1160.

26. Shimizu S, Konishi A, Kodama T, Tsujimoto Y. BH4 domainof antiapoptotic Bcl-2 family members closes voltage-dependentanion channel and inhibits apoptotic mitochondrial changes andcell death. PNAS USA 2000; 97: 3100–3105.

27. Muchmore SW. X-ray and NMR structure of human Bcl-XL,an inhibitor of programmed cell death. Nature 1996; 381: 335–341.

28. Antonsson B, Conti F, Ciavatta AM, et al. Inhibition of Baxchannel-forming activity by Bcl-2. Science 1997; 277(5324):370–372.

29. Susin SA, Lorenzo HK, Zamzami N, et al. Molecular characteri-zation of mitochondrial apoptosis-inducing factor. Nature 1999;397: 441–446.

30. Thomenius MJ, Distelhorst CW. Bcl-2 on the endoplasmic retic-ulum: Protecting the mitochondria from a distance. J Cell Sci2003; 116: 4493–4499.

31. Maraskovsky E, O’Reilly L, Teepe M, Corocran L, Peschon J,Strasser A. Bcl-2 can rescue T lymphocyte development inInterleukin-7 receptor-deficient mice but can not in mutantrag−/− mice. Cell 1997; 89: 1011–1019.

32. Lang G, Gombert WM, Gould HJ. A transcriptional element inthe coding sequence of the human Bcl-2 gene. Immunology 2005;114: 25–36.

33. Seto M, Jaeger U, Hockett RD, et al. Alternative promoterand exons, somatic mutations and deregulation of the Bcl-2-Igfusion gene in lymphoma. EMBO J 1988; 7: 123–131.

34. Nunez G, London L, Hockenbery D, Alexander M, McKearnJP, Korsmeyer SJ. Deregulated Bcl-2 gene expression selectivelyprolongs survival of growth factor-deprived hematopoietic celllines. J Immunol 1990; 144: 3602–3610.

35. Tsujimoto Y, Croce CM. Analysis of the structure, transcripts,and protein products of Bcl-2, the gene involved in humanfollicular lymphoma. PNAS USA 1986; 83: 5214–5218.

36. Batistatou A, Merry DE, Korsmeyer SJ, Greene LA. Bcl-2 affectssurvival but not neuronal differentiation of PC12 cells. J Neurosci1993; 13: 4422–4428.

37. Wilson BE, Mochon E, Boxer LM. Induction of Bcl-2 expressionby phosphorylated CREB proteins during B-cell activation andrescue from apoptosis. Mol Cell Biol 1996; 16: 5546–5556.

38. Dong L, Wang W, Wang F, et al. Mechanisms of transcriptionalactivation of Bcl-2 gene expression by 17beta-estradiol in breastcancer cells. J Biol Chem 1999; 274(45): 32099–32107.

39. Li ZL, Abe H, Ueki K, Kumagai K, Araki R, Otsuki Y. Identi-fication of c-Jun as Bcl-2 transcription factor in human uterineendometrium. J Histochem Cytochem 2003; 51(12): 1601–1609.

40. Otsuki Y, Misaki O, Sugimoto O, Ito Y, Tsujimoto Y, Akao Y.Cyclic Bcl-2 gene expression in human uterine endometriumduring menstrual cycle. Lancet 1994; 344: 28–29.

41. Savill J, Mooney A, Hughes J. What role does apoptosis play inprogression of renal disease? Curr Opin Nephrol Hypertens 1996;5(4): 369–374.

42. Young RL, Korsmeyer SJ. A negative regulatory element inthe Bcl-2 5′-untranslated region inhibits expression from anupstream promoter. Mol Cell Biol 1993; 13(6): 3686–3697.

43. Miyashita T, Harigai M, Hanada M, Reed JC. Identification ofa p53-dependent negative response element in the Bcl-2 gene.Cancer Res 1994a; 54: 3131–3135.

44. Miyashita T, Krajewski S, Krajewska M, et al. Tumor suppressorp53 is a regulator of Bcl-2 and Bax gene expression in vitro andin vivo. Oncogene 1994; 9: 1799–1805.

45. Schiavone I, Rosini P, Quattrone A, Donnini M, Lapucci A, CittiL. A conserved AU-rich element in the 3′ untranslated region ofBcl-2 mRNA is endowed with a destabilizing function that isinvolved in Bcl-2 down-regulation during apoptosis. The FASEBJournal 2000; 14: 174–184.

46. Chen-Levy Z, Nourse J, Cleary ML. The Bcl -2 candidate proto-oncogene product is a 24-kilodalton integral-membrane proteinhighly expressed in lymphoid cell lines and lymphomas car-rying the t(14;18) translocation. Mol Cell Biol 1989; 9: 701–710.

47. Petros AM, Olejniczak ET, Fesik SW. Structural biology of theBcl-2 family of proteins. Biochimica et Biophysica Acta. MoleculCell Research 2004; 1644: 83–94.

48. Sattler M, Liang H, Nettesheim D, Meadows RP, Harlan JE.Structure of Bcl-XL/Bak peptide complex: Recognition betweenregulators of apoptosis. Science 1997; 275: 983–986.

49. Gurudutta GU, Verma YK, Singh VK, et al. Structural conser-vation of residues in BH1 and BH2 domains of Bcl-2 familyproteins. FEBS Lett 2005 Jul 4; 579(17): 3503–3507.

50. Borner C. Dissection of functional domains in Bcl-2α by site-directed mutagenesis. Biochem Cell Biol 1994; 72: 463–469.

51. Diaz JL. A common binding site mediates heterodimerizationand homodimerization of Bcl-2 family members. J Biol Chem1996; 272(17): 11350–11355.

52. Fesik SW. Structure of Bcl-XL-Bak peptide complex: Recogni-tion between regulators of apoptosis. Science 1997; 275: 983–986.

53. Zha H. Heterodimerization independent functions of cell deathregulatory proteins Bax and Bcl-2 in yeast and mammalian cells.J Biol Chem 1997; 272: 31482–31488.

54. Wang HG, Rapp UR, Reed JC. Bcl-2 targets the protein kinaseRaf-1 to mitochondria. Cell 1996; 87: 629–638.

55. Wang HG, Takayama S, Rapp UR, Reed JC. Bcl-2 interactingprotein, BAG-1, binds to and activates the kinase Raf-1. PNASUSA 1996; 93: 7063–7068.

56. Pietenpol JA, Papadopoulos N, Markowitz S, Willson JK,Kinzler KW, Vogelstein B. Paradoxical inhibition of solid tumorcell growth by Bcl-2. Cancer Res 1994; 54: 3714–3717.

57. Linette GP, Li Y, Roth K, Korsmeyer SJ. Cross talk between celldeath and cell cycle progression: Bcl-2 regulates NFAT-mediatedactivation. PNAS USA 1996; 93: 9545–9552.

58. Muller C, Kowenz-Leutz E, Grieser-Ade S, Graf T, Leutz A.NF-M (chicken C/EBPb) induces eosinophilic differentiation

470 Apoptosis · Vol 11 · No 4 · 2006

Bcl-2 in apoptosis

and apoptosis in a hematopoietic progenitor cell line. EMBO J1995; 14: 6127–6135.

59. Zhang J, Alter N, Reed JC, Borner C, Obeid LM, Hannun YA.Bcl-2 interrupts the ceramide-mediated pathway of cell death.Biochemistry 1996; 93(11): 5325–5328.

60. Chang BS, Minn AJ, Muchmore SW, Fesik SW, Thompson CB.Identification of a novel regulatory domain in Bcl-XL and Bcl-2.EMBO J 1997; 16: 968–977.

61. Petros AM, Medek A, Nettesheim DG, Kim DH, Yoon HS,Swift K, et al. Solution structure of the antiapoptotic proteinBcl-2. PNAS USA 2001; 98(6): 3012–3017.

62. Liu X. The structure of a Bcl-XL/Bim fragment complex: Im-plications for Bim function. Immunity 2003; 19: 341–352.

63. Fesik SW. Insight into programmed cell death through struc-tural biology. Cell 2000; 103: 273–282.

64. Huang Q. Solution structure of the BHRF1 protein fromEpstein-Barr virus, a homolog of human Bcl-2. J Mol Biol 2003;332: 1123–1130.

65. Peitsch MC. ProMod and Swiss-Model: Internet-based tools forautomated comparative p rotein modelling. Biochem Soc Trans1996; 24: 274–279.

66. Verma YK, Gurudutta GU, Singh VK, et al. Homology modelof anti-apoptotic protein Bcl-2. Bioinformatics India 2006; 3(3):in press.

67. Day CL, Puthalakath H, Skea G, et al. Localization of dyneinlight chains 1 and 2 and their pro-apoptotic ligands. Biochem J2004; 377(3): 597–605.

68. Antonsson B, Martinou JC. The Bcl-2 protein family. Exp CellRes 2000; 256: 50–57.

69. Yamamoto K, Ichijo H, Korsmeyer SJ. Bcl-2 is phosphorylatedand inactivated by an ASK1/Jun N-terminal protein kinase path-way normally activated at G2/M. Molecul Cellular Biol 1999;19(12): 8469–8478.

70. Ruvolo PP, Deng X, May WS Jr. Phosphorylation of Bcl-2 andregulation of apoptosis. Leukemia 2001; 15: 515–522.

71. Meier P, Finch A, Evan G. Apoptosis in development. Nature2000; 407: 796–801.

72. McDonnell JM, Fushman D, Milliman CL, Korsmeyer SJ,Cowburn D. Solution structure of the proapoptotic moleculeBID: a structural basis for apoptotic agonists and antagonists.Cell 1999; 96: 625–634.

73. Haldar S, Jena N, Croce CM. Inactivation of Bcl-2 by phospho-rylation. PNAS USA 1995; 92: 4507–4511.

74. Haldar S, Chintapalli J, Croce CM. Taxol induces Bcl-2 phos-phorylation and death of prostate cancer cells. Cancer Res 1996;56: 1253–1255.

75. Haldar S, Basu A, Croce CM. Serine-70 is one of the critical sitesfor drug-induced Bcl-2 phosphorylation in cancer cells. CancerRes 1998; 58: 1609–1615.

76. Blagosklonny MV, Schulte T, Nguyen P, Trepel J, Neckers LM.Taxol-induced apoptosis and phosphorylation of Bcl-2 proteininvolves c-Raf-1 and represents a novel c-Raf-1 signal transduc-tion pathway. Cancer Res 1996; 56: 1851–1854.

77. Blagosklonny MV, Giannakakou P, El-Deiry WS, et al. Raf-1/Bcl-2 phosphorylation: A step from microtubule damage tocell death. Cancer Res 57: 130–135.

78. Blagosklonny MV, Chuman Y, Bergan RC, Fojo T. Mitogenactivated protein kinase pathway is dispensable for microtubule-active drug-induced Raf-1/Bcl-2 phosphorylation and apoptosisin leukemia cells. Leukemia 1999; 13: 1028–1036.

79. Hannun YA. The sphingomyelin cycle and the second messengerfunction of ceramide. J Biol Chem 1994; 269: 3125–3128.

80. Jarvis WD, Grant S, Kolesnick RN. Ceramide and the inductionof apoptosis. Clin Cancer Res 1996; 2: 1–6.

81. Nicholson DW, Ali A, Thornberry NA, et al. Identification andinhibition of the ICE/CED-3 protease necessary for mammalianapoptosis. Nature 1995; 376: 37–43.

82. Johnson BW, Boise LH. Bcl-2 and caspase inhibition cooperateto inhibit tumor necrosis factor-induced cell death in a Bcl-2 cleavage-independent fashion. J Biol Chem 1999; 274(26):18552–18558.

83. Russel JH, Rush B, Weaver C, Wang R. Mature T cells ofautoimmune lpr/lpr mice have a defect in antigen-stimulatedsuicide. PNAS USA 1993; 90: 4409–4413.

84. Kischkel FC, Hellbardt S, Behrmann I, et al. Cytotoxicity de-pendent APO-1 (Fas/CD95)-associated proteins form a death-inducing signaling complex (DISC) with the receptor. EMBO J1995; 14: 5579–5588.

85. Johnson BW, Boise LH. Bcl-2 and caspase inhibitor (zVAD-fmk) cooperate to inhibit tumor necrosis factor-α-induced celldeath in a Bcl-2 cleavage independent fashion. J Biol Chem 1999;274(26): 18552–18558.

86. Cheng EH, Levine B, Boise LH. Thompson CB, HardwickJM. Bax-independent inhibition of apoptosis by Bcl-XL.Nature 1996; 379: 554–556.

87. Cheng EHY, Nicholas J, Bellows DS, Hayward GS, Guo HG,Reitz MS, et al. A Bcl-2 homolog encoded by Kaposi sarcoma-associated virus, human herpesvirus 8, inhibits apoptosis butdoes not heterodimerize with Bax or Bak. PNAS USA 1997; 94:690–694.

88. Vaux DL, Cory S, Adams JM. Bcl-2 gene promotes hamatopoi-etic cell survival and cooperates with c-myc to immortalize pre-Bcells. Nature 1988; 335: 440–442.

89. Lagasse E, Weissman IL. Enforced expression of Bcl-2 in mono-cytes rescues macrophages and partially reverses osteopetrosis inop/op mice. Cell 1997; 89(7): 1021–1031.

90. Maraskovsky E, O’Reilly L, Teepe M, Corocran L, Peschon J,Strasser A. Bcl-2 can rescue T lymphocyte development inInterleukin-7 receptor-deficient mice but can not in mutantrag−/− mice. Cell 1997; 89: 1011–1019.

91. Reed JC. Mechanisms of Bcl-2 family protein function and dys-function in health and disease. Behring Inst Mit 1996; 97: 72–100.

92. Frisch SM, Francis H. Disruption of epithelial cell-matrix inter-actions induces apoptosis. J Cell Biol 1994; 124: 619–626.

93. Domen J, Gandy KL. Systemic overexpression of Bcl-2 inthe hematopoietic system protects transgenic mice from theconsequences of lethal irradiation. Blood 1998; 91: 2272–2282.

94. Tsukasaki K, Tsushima H, Yamamura M, et al. Integration pat-terns of HTLV-I provirus in relation to the clinical course ofATL: Frequent clonal change at crisis from indolent disease.Blood 1997; 89(3): 948–956.

95. Huang HM, Li JC, Hsieh YC, Yen HFY, Yen JJY. Optimalproliferation of a hematopoietic progenitor cell line requireseither costimulation with stem cell factor or increase of receptorexpression that can be replaced by overexpression of Bcl-2. Blood1999; 93(8): 2569–2577.

96. Gossen M, Bujard H. Tight control of gene expression in mam-malian cells by tetracycline-responsive promoters. PNAS USA1992; 89(12): 5547–5551.

97. Francis K, Ramakrishna R, Holloway W, Palsson BO. Two newpseudopod morphologies displayed by the human hematopoietickG-1a progenitor cell line and by primary human CD34+ cells.Blood 1998; 92(10): 3616–3623.

98. Wyllie AH. The genetic regulation of apoptosis. Curr Opin GenDev 1995; 5: 97–104.

99. Deng X, Ruvolo P, Carr B, May Jr WS. Survival function ofERk1/2 as IL-3 activated, staurosporine-resistant Bcl-2 kinases.PNAS USA 2000; 97(4): 1578–1583.

100. Rao S, Watkins S, Cunningham D, et al. Phase II study ofISIS 3521, an antisense oligodeoxynucleotide to protein kinaseC alpha, in patients with previously treated low-grade non-Hodgkin’s lymphoma. Ann Oncol 1997; 15: 1413–1418.

Apoptosis · Vol 11 · No 4 · 2006 471