The roles of programmed cell death in tumor development and cancer therapy

The Walter and Eliza Hall
Institute of Medical Research
Parkville, Victoria
Department of Medical Biology
The University of Melbourne
Parkville, Victoria

The roles of programmed cell death in tumor development and cancer therapy

by A. Strasser, Australia

Apoptosis is a process of programmed cell death responsible for the removal of no-longer-needed, damaged, or potentially dangerous cells. Apoptosis is controlled by the B-cell chronic lymphocytic leukemia/ lymphoma 2 (BCL2) family of proteins, which can be divided into 3 subgroups according to amino acid sequence, structure, and function: the BCL2-like prosurvival proteins required for cell survival, the proapoptotic BCL2-associated X (BAX)/BCL2-antagonist/killer (BAK) proteins essential for unleashing cellular destruction via the caspase cascade, and the proapoptotic BCL2 homology domain 3 (BH3)-only proteins, which are critical to initiate apoptosis signaling. It is now firmly established that mutations or other defects that cause abnormalities in the expression of proapoptotic or antiapoptotic BCL2 family members promote tumorigenesis and render cancer cells refractory to diverse chemotherapeutic drugs. This review describes current understanding of the molecular regulation of apoptosis by the BCL2 protein family, the impact of defects in this process on cancer, and finally discusses currently evolving strategies for direct therapeutic modulation of the BCL2-family–regulated apoptotic pathway for treatment of cancer.

Medicographia. 2014;36:311-318 (see French abstract on page 318)

Removal of no-longer-needed cells plays a critical role in the shaping of tissues (morphogenesis) during development of multicellular organisms.1 Moreover, throughout postnatal life, removal of aged, useless, and potentially dangerous (eg, pathogen-infected) cells is critical for tissue homeostasis and normal health. A variety of processes for removing unwanted cells have been recognized, including several mechanisms of programmed cell death, phagocytosis (cell eating), as well as shedding of cells.1 Apoptosis is a genetically controlled mechanism of programmed cell death that is evolutionarily highly conserved with closely related regulators and signaling pathways found in species as distantly related as the worm Caenorhabditis elegans and humans.2

In mammals and other vertebrates, there are 2 distinct, but ultimately converging, pathways for induction of apoptosis (Figure 1, page 312): the “death-receptor” pathway (also called the “extrinsic” pathway; mentioned here only in passing; for a review see reference 3) and the “B-cell chronic lymphocytic leukemia (CLL)/lymphoma 2 protein (BCL2)-regulated” pathway (also called “mitochondrial,” “intrinsic,” or “stress induced” pathway).2 Both pathways converge upon the activation of so-called “effector caspases” (eg, caspase 3 and caspase 7), which cleave hundreds of cellular proteins to cause the morphological and biochemical characteristics of apoptosis (eg, plasma membrane blebbing, chromatin condensation, internucleosomal DNA fragmentation) that precipitate cellular demolition.4

Figure 1
Figure 1. Two distinct, but ultimately converging,
pathways for induction of apoptosis.

Diagram showing the mediators of the BCL2-regulated and
the death receptor–initiated apoptotic pathways and their connection
through caspase 8–mediated proteolytic activation of the
proapoptotic BH3-only BCL2 family member BID.
Abbreviations: APAF-1, apoptotic protease activating factor 1;
BAK, BCL2 antagonist/killer; BAX, BCL2-associated X protein;
BCL2, B-cell chronic lymphocytic leukemia/lymphoma 2 protein;
BH3, BCL2 homology domain 3; BID, BH3-interacting-domain
death agonist; cyt C, cytochrome C; DIABLO, direct inhibitor
of apoptosis binding protein with low pI; FADD, Fas-associated
death domain; FASL, Fas ligand; IAP, inhibitor of apoptosis
binding protein; SMAC, second mitochondria-derived activator
of caspases; tBID, truncated BID; TNF, tumor necrosis factor;
TRAIL, TNF-related apoptosis-inducing ligand.

In the death-receptor pathway, members of the tumor necrosis factor receptor (TNFR) family with an intracellular death domain (eg, TNF type 1 receptor [TNFR1]; FAS, also called apoptosis antigen 1 [APO-1] and cluster of differentiation 95 [CD95]) trigger Fas-associated death domain (FADD) adaptor- protein–mediated activation of the “initiator caspase,” caspase 8 (in humans, also caspase 10), which then proteolytically activates the effector caspases (Figure 1). In the BCL2- regulated pathway, cell death is triggered by developmental cues or cellular stressors (eg, growth factor deprivation, DNA damage; Figure 1). This causes transcriptional and/or posttranscriptional activation of BCL2 homology domain 3 (BH3)-only proteins, the proapoptotic subgroup of the BCL2 family, which initiate apoptosis by activating BCL2-associated X protein (BAX)/BCL2 antagonist/killer (BAK) (the second proapoptotic subgroup of the BCL2 family) through 2 processes (Figure 1). Activated BAX/BAK proteins oligomerize, causing mitochondrial outer membrane permeabilization (MOMP). This leads to the release of apoptogenic proteins, such as cytochrome c and second mitochondria- derived activator of caspases (Smac; also referred to as direct inhibitor of apoptosis binding protein with low pI [Diablo]), which promote apoptotic protease-activating factor 1 (Apaf-1) adaptor-protein–mediated activation of the initiator caspase, caspase 9, and amplification of effector caspase activation (Figure 1). The death-receptor and the BCL2-regulated apoptotic pathways are connected through caspase 8–mediated proteolytic activation of the BH3-only protein BH3-interacting-domain death agonist (BID) (Figure 1).


Structures and functions of the members of the 3 major subgroups of the BCL2 family of proteins

There are 3 major subgroups of the BCL2 family of proteins: (i) the prosurvival BCL2-like proteins (BCL2, BCL2-like protein 1 [BCL-XL], BCL2-like protein 2 [BCL-W], myeloid-cell leukemia sequence 1 [MCL-1], BCL2-related protein A1 [A1, also known as BFL1], BCL2-like homolog of ovary [BOO, also known as DIVA]); (ii) the BAX/BAK proteins (BAX, BAK, BCL2-related ovarian killer [BOK]); and (iii) the BH3-only proteins (BCL2-interacting mediator of cell death [BIM], BID, p53- upregulated modulator of apoptosis [PUMA], BCL2-modifying factor [BMF], BCL2-associated death promoter [BAD], BCL2-interacting killer [BIK], phorbol-12-myristate-13-acetate- induced protein 1 [NOXA], harakiri [HRK]) (Figure 2). These proteins are related to each other by the presence of at least 1 of 4 recognized BCL2 homology (BH) domains; the prosurvival members and the BAX/ BAK proteins have 4 BH domains, whereas the BH3-only proteins only have the BH3 domain. In addition, there are some proteins that contain 1 or more BH domains, but do not readily fit into any of the 3 aforementioned subgroups. For most of these “odd BH-domain–containing proteins” the functions are unknown and they will not be dealt with further in this article.

Essential functions of the different prosurvival BCL2 family members
The prosurvival BCL2 family members all promote cell survival when overexpressed (such as in transgenic mice5) and studies with genetargeted mice have revealed their essential functions. BCL2 is required for the survival of renal epithelial progenitor cells, mature lymphocytes, and melanocyte progenitors. Consequently, Bcl-2–/– mice die from polycystic kidney disease around 30 to 40 days, are lymphopenic, and turn prematurely gray.6

BCL-XL is needed for the survival of erythroid progenitors, platelets, certain neuronal populations, and developing sperm cells.7 Accordingly, Bcl-x–/– mice die around embryonic day 14 (E14) and Bcl-x+/– mice have abnormally low platelet levels and males are subfertile. cl-w–/– mice develop normally and have no overt defects in adult life with the exception of male infertility, which was ascribed to abnormal death of the supporting Sertoli cells. MCL-1 is essential for survival of undifferentiated cells in the early embryo (Mcl-1–/– embryos die prior to implantation) and for survival of cardiomyocytes, certain neuronal populations, and a broad range of hematopoietic cells, including the stem/ progenitor cells.8 The critical functions of A1/BFL1 are not fully resolved because the presence of 4 closely related, often coexpressed A1 genes makes gene targeting difficult. Studies with mice lacking 1 A1 gene, A1a, or transgenic mice expressing a short hairpin RNA (shRNA) that downregulates all A1 genes indicated that A1 plays a role in the survival of granulocytes as well as of immature B- and T-lymphoid progenitors.9 The mouse Boo/Diva gene has mutations that are predicted to render the protein nonfunctional, but it remains possible that the human protein does play a role in cell survival. There is currently very little information (eg, from analysis of double-knockout mice) on the overlapping functions of prosurvival BCL2 family members.

Figure 2
Figure 2. The members of the 3 major subgroups of the BCL2 protein family.

Diagram showing the 3 major subgroups of the BCL2 protein family: the BCL2-like prosurvival proteins
that are essential for cell survival, the proapoptotic BAX/BAK-like proteins that are essential
for unleashing the effector phase of apoptosis and the proapoptotic BH3-only proteins that are
critical for the initiation of the BCL2-regulated apoptotic pathway. The asterisk above the TM
(transmembrane domain) of the stick model of a BH3-only protein indicated that only some (eg,
BIM), but not all (eg, BAD), BH3-only proteins have a transmembrane domain.
Abbreviations: A1, BCL2-related protein A1; BAD, BCL2-associated death promoter; BAK, BCL2
antagonist/killer; BAX, BCL2-associated X protein; BCL2, B-cell chronic lymphocytic leukemia/
lymphoma 2 protein; BCL-W, BCL2-like protein 2; BCL-XL, BCL2-like protein 1; BH3, BCL2 homology
domain; BID, BH3-interacting-domain death agonist; BIK, BCL2-interacting killer; BIM, BCL2-
interacting mediator of cell death; BMF, BCL2-modifying factor; BOK, BCL2-related ovarian killer;
HRK, harakiri; MCL-1, myeloid-cell leukemia sequence 1; NOXA, phorbol-12-myristate-13-acetateinduced
protein 1; PUMA, p53-upregulated modulator of apoptosis; TM, transmembrane domain.

Essential functions of the BAX/BAK proteins
Mice lacking either BAX or BAK alone are largely normal, with the exception of male infertility in the former and increased platelet numbers in the latter. The generation of BAX/BAK doubly deficient mice revealed remarkable functional overlap between these multi–BH-domain proapoptotic proteins; these animals have webbed paws and abnormally increased numbers of certain neuronal populations, and the few animals that survive into early adulthood develop massive lymphadenopathy and splenomegaly.10 Diverse cell types from these mice are profoundly (in certain instances perhaps completely) resistant to a broad range of apoptotic stimuli.10 BOK has substantial amino acid sequence similarity to BAX and BAK. Although no abnormalities were found in Bok–/– mice, the increased persistence of primordial follicles in ovaries of aged BOK/BAK doubly deficient mice provides the first evidence that BOX exerts a proapoptotic function that overlaps with that of BAX and BAK.11

Essential functions of the different proapoptotic BH3- only proteins
Biochemical studies have shown that BH3-only proteins differ markedly in their affinities for binding to the various prosurvival BCL2 family members (Figure 3). BIM, PUMA, and BID (after caspase-mediated processing to the so-called truncated BID [tBID] form) bind with high (low nM or even sub-nM) affinity to all prosurvival BCL2-like proteins and are therefore sometimes referred to as “promiscuous binders” (Figure 3). All other BH3-only proteins are “selective binders”; for example NOXA binds to MCL-1 and A1/BFL1, whereas BAD binds to BCL2, BCL-XL, and BCL-W12 (Figure 3). Moreover, only some BH3-only proteins (BIM, PUMA, tBID) appear able to bind the multi–BH-domain proapoptotic BAX/BAK proteins although there is disagreement in the literature on whether some others, such as BMF or NOXA, can also do this.

Figure 3
Figure 3. BH3-only proteins differ in their affinities for binding different
prosurvival BCL2 family members.

Diagram showing the differences in binding specificities of distinct BH3-only
proteins for different prosurvival BCL2 family members.
Abbreviations: BAD, BCL2-associated death promoter; BCL2, B-cell chronic
lymphocytic leukemia/lymphoma 2 protein; BCL-W, BCL2-like protein 2; BCL-XL,
BCL2-like protein 1; BH, BCL2 homology domain; BIM, BCL2-interacting mediator
of cell death; MCL-1, myeloid-cell leukemia sequence 1; NOXA, phorbol-12-
myristate-13-acetate-induced protein 1; PUMA, p53-upregulated modulator of
apoptosis; tBID, truncated BID (BH3-interacting-domain death agonist).

The “promiscuous” BH3-only proteins can all elicit apoptosis when overexpressed by themselves. Conversely, enforced expression of combinations of complimentary “select binders,” such as NOXA plus BAD, is needed to induce efficient cell killing.12 This indicates that all prosurvival BCL2-like proteins present in a cell must be neutralized by BH3-only proteins to initiate apoptosis, presumably by liberating primed BAX/BAK (this has been dubbed the “indirect activation” model). The “direct activation” model posits that some BH3-only proteins (called “direct activators,” eg, BIM, PUMA, tBID) can bind and thereby activate BAX/BAK and that these BH3-only proteins are kept in check by binding to the prosurvival BCL2-like proteins until displaced by the “indirect activators” (eg, BAD, NOXA). Experiments with gene-targeted mice carrying subtle mutations that affect the binding specificity of BIM have shown that elements of “indirect” as well as “direct” activation must contribute to its role in programmed death of hematopoietic cells.13

Experiments with gene-targeted mice have shown that loss of the “promiscuous binders,” BIM, PUMA, and BID, causes profound abnormalities, whereas loss of most other BH3-only proteins on their own has only minor impact. BIM is critical for apoptosis induced by growth factor withdrawal, deregulated calcium flux, or endoplasmic reticulum (ER) stress, and also contributes to apoptosis triggered by DNA damage. BIM-deficient mice have defects in the deletion of autoreactive T- and B-lymphoid cells and removal of activated lymphocytes during shutdown of acute as well as chronic immune responses.

This causes abnormal accumulation of lymphoid cells and plasma cells with a predisposition to autoimmune disease and lymphoid malignancy.14 PUMA is directly transcriptionally activated by the tumor suppressor protein p53 and essential for apoptosis triggered by DNA damage.15 PUMA also contributes substantially to apoptosis triggered by certain p53-independent stimuli, such as cytokine deprivation, ER stress, and treatment with glucocorticoids or phorbol ester.15 Activation of BID by caspase 8, leading to amplification of the caspase cascade via BAX/BAK and activation of caspase 9, is critical for FASor TNFR1-induced killing of certain cell types (called type 2; eg, hepatocytes, pancreatic βcells), but dispensable in others (called type 1; eg, thymocytes).16 BH3-only proteins have overlapping functions; this is best demonstrated by the profound resistance of cells from BIM/PUMA–doubly deficient mice and BIM/PUMA/BID–triply deficient mice to a broad range of apoptotic stimuli and the remarkable lymphadenopathy and autoimmune disease predisposition of these animals.17

Antagonistic interactions between proapoptotic and prosurvival BCL2 family members
Genetic and crystallographic studies have illuminated the functional and structural interactions between the members of the 3 subgroups of the BCL2 protein family. Remarkably, loss of BIM (even loss of 1 Bim allele) can overcome all abnormalities caused by loss of BCL2 and some of the defects (eg, failure of fetal erythropoiesis) elicited by loss of BCL-XL.18 Moreover, loss of BAK rescues the abnormal drop in platelets that is caused by BCL-XL deficiency. These functional antagonisms are mirrored by strong physical interactions of these proteins, which involve insertion of the BH3 domain of either the BH3-only or the BAX/BAK protein into the groove on the surface of the prosurvival BCL2-like protein that is formed by their BH1, BH2, and BH3 domains.19

During their activation (ie, through direct binding by BH3-only proteins or after release from the prosurvival BCL2-like proteins), BAX/BAK undergo dramatic structural changes.20 This leads to dimerization and ultimately multimerization of BAX/ BAK, which causes MOMP in a manner that is still not resolved. This constitutes the point of “no return” in the control of apoptosis signaling. Hence, getting tumor cells to this point is what needs to be achieved by anticancer drugs that directly modulate the apoptotic machinery.

The role of BCL2 family members in cancer

The study of cell death is tightly interwoven with cancer research. The first cell death regulatory gene from any species, Bcl-2, was discovered because of its recurrent chromosomal translocation (t14;18) in human follicular center B-cell lymphoma.21

Abnormalities in the expression of BCL2 family members can promote tumor development
Although chromosomal translocations involving genes for prosurvival BCL2 family members are rare in cancers other than follicular lymphoma, it has become clear that somatically acquired copy number amplifications of the genomic regions harboring the Mcl-1 and Bcl-x genes are present at relatively high frequencies in diverse human cancers.22 The amplified genomic regions in these cancers are relatively large and therefore contain additional (non–cell-death regulatory genes), but initial functional studies using RNA interference (RNAi)-mediated knockdown indicated that excess MCL-1 or BCL-XL may indeed be critical for the sustained growth of these transformed cells.22

Finally, gene expression profiling studies (using microarray or RNA-Seq technology) and protein analysis have found abnormally high levels of prosurvival BCL2-like proteins (or their messenger RNAs [mRNAs]) in a substantial number of human cancers, even though many of those are likely not to have overt chromosomal alterations of the corresponding genes. This may be due to epigenetic modifications or the fact that oncogenic pathways that are activated in these tumors cause their transcriptional induction, translational increase, or posttranslational stabilization.

Abnormalities in genes encoding proapoptotic BCL2 family members have also been found in human cancers. Loss of both alleles of Bim was reported in nearly 20% of human mantle cell B lymphoma.23 Moreover, abnormally low levels of BIM and PUMA, in part ascribed to hypermethylation of the genes for these BH3-only proteins, has been observed in several human cancers, including renal cell carcinoma or Burkitt Lymphoma.24 Loss of Bax or Bak genes appear to be rare in human cancers because mutations in 4 alleles, which would be needed to obliterate the BAX/BAK checkpoint,10 is an unlikely event. Curiously, however, loss of the region harboring the Bok gene was found in several human cancers.22

Studies with transgenic and gene-targeted (knockout) mice have confirmed and extended the findings from the studies of human cancers. Overexpression of BCL2 or its prosurvival relatives in lymphoid cells promotes lymphomagenesis, particularly in combination with oncogenic lesions that deregulate the control of cell proliferation, such as by c-MYC overexpression.25 Similarly, loss of the BH3-only proteins BIM, PUMA, or multi– BH-domain proapoptotic BAX also promote tumorigenesis.26

The role of prosurvival BCL2 family members expressed under endogenous control in the development and sustained growth of tumors
Although it is well established that overexpression of prosurvival BCL2 family members can promote tumorigenesis, there are still only few reports from studies on the importance of these proteins, expressed under endogenous control, for the development and sustained growth of cancers. Interestingly, BCL-XL and MCL-1, but not BCL2 were found to be critical for c-MYC–driven pre-B/B lymphoma and acute myelocytic leukemia (AML).27 A likely explanation for this may be that BCLXL and MCL-1, but not BCL2, are expressed in the cell populations from which these hematological malignancies emerge (leukemia/lymphoma-initiating stem cells). MCL-1, under endogenous control, appears to play a particularly prominent role in tumor development, since it was also found to be essential for the induction and sustained growth of AML driven by diverse oncogenes, such as MLL-ENL.28

The role of the proapoptotic BCL2 family members in anticancer drug–induced killing of tumor cells
It has long been known that chemotherapeutic drugs and γ-radiation can induce apoptosis in cancer cells. Studies with transgenic mice or tumor-derived cell lines transduced with expression vectors revealed that abnormally increased levels of prosurvival BCL2 family members can render nontransformed as well as cancerous cells resistant to a broad range of anticancer therapeutics.5 Studies using gene-targeted mice or shRNA expression vectors have revealed the roles of the BH3-only proteins and BAX/BAK in the killing of cancer cells by chemotherapeutic drugs. Combined loss of BAX and BAK rendered experimentally transformed cells resistant to a broad range of chemotherapeutic drugs and γ-radiation (Figure 4, page 316). The BH3-only proteins PUMA and, to a lesser extent, NOXA, which are directly transcriptionally regulated by the tumor suppressor p53, were shown to be critical for the killing of lymphoma and certain other neoplastic cells by DNA damage–inducing anticancer therapies29 (Figure 4). Unexpectedly, BIM, which does not appear to be directly regulated by p53, also contributes to this response29 (Figure 4). PUMA and BIM were shown to mediate glucocorticoid-induced killing of lymphomas and leukemias, and BIM as well as BMF were critical for the response of tumor cells to paclitaxel or histone deacetylase (HDAC) inhibitors (Figure 4). How BIM, PUMA, and BMF are induced in response to these nongenotoxic drugs is not clear. BIM and, to a lesser extent, BAD are essential for the killing of diverse cancer cells by inhibitors of oncogenic kinases, such as imatinib for inhibition of the breakpoint cluster region (BCR)-Abelson tyrosine kinase (ABL) fusion protein (BCR-ABL) in chronic myelocytic leukemia (CML), erlotinib/ gefitinib for the inhibition of mutant epidermal growth factor receptor (EGFR) in lung cancer, and serine/threonine– protein kinase B-Raf (B-RAF) or mitogen-activated extracellular signal–regulated kinase (MEK) inhibitors for the treatment of melanoma or colon carcinomas bearing B-Raf mutations30 (Figure 4). Interestingly, a polymorphism in Bim that attenuates its expression is enriched among East Asian patients with CML or lung cancer, who show poor de novo responses to inhibitors of oncogenic kinases (imatinib or erlotinib/gefitinib, respectively).31

Figure 4
Figure 4. Different anticancer agents activate distinct
BH3-only proteins to initiate apoptosis in tumor cells.

Diagram showing which currently used anticancer agents activate
which BH3-only proteins to initiate apoptosis in tumor cells.
Abbreviations: BAD, BCL2-associated death promoter; BH, BCL2
homology domain; BIK, BCL2-interacting killer; BIM, BCL2-interacting
mediator of cell death; BMF, BCL2-modifying factor; HDAC, histone
deacetylase; HRK, harakiri; NOXA, phorbol-12-myristate-13-acetateinduced
protein 1; p53, protein 53; PUMA, p53-upregulated modulator
of apoptosis; tBID, truncated BID (BH3-interacting-domain
death agonist).

Flipping the BCL2-regulated apoptotic switch for cancer therapy

Since efficient induction of apoptosis appears critical for the responses of many cancers to a broad range of therapeutics and since many cancers bear mutations that impair efficient induction of apoptosis (eg, mutations in p53, disabling efficient induction of PUMA, or overexpression of prosurvival BCL2 family members), substantial efforts are being undertaken to develop drugs that can directly flip the BCL2–regulated apoptotic switch (Figure 5).

Direct inhibition of prosurvival BCL2 family members by BH3 mimetics
One approach involves the generation of small molecular weight compounds that mimic the action of the BH3-only proteins, binding in the groove on the surface of prosurvival BCL2-like proteins, thereby inhibiting their antiapoptotic activity (Figure 5). ABT-737 and the structurally closely related, orally available ABT-263 bind to BCL2, BCLXL, and BCL-W; ABT-199 binds only to BCL2;32 and WEHI-539 selectively inhibits BCL-XL.33 These BH3 mimetics can kill certain tumor cells (eg, CLL) as single agents, but in many other cancers they need to be combined with standard anticancer therapeutics or inhibitors of oncogenic kinases for efficient killing. Functional and biochemical studies have revealed that ABT- 737 initiates apoptosis mostly by displacing BIM (and possibly other BH3-only proteins) from BCL2, thereby allowing it to bind and neutralize the other prosurvival BCL2-like proteins that are present in those cancer cells.34 Excitingly, both ABT-263 and ABT-199 are currently showing early promise in clinical trials for CLL and certain other cancers.32

Of course, not only tumor cells, but also nontransformed cells in healthy tissues are affected by BH3 mimetics and their response will be dose limiting. ABT-263 and ABT-737 cause a drop in platelets, consistent with their ability to bind BCL-XL, which is critical for platelet survival. It will of course be important to determine to what extent BH3 mimetics will augment the toxicity to healthy tissues that is caused by the anticancer therapeutics that they are likely to be combined with in the clinic. I predict that combinations of BH3 mimetics with inhibitors of oncogenic kinases could be particularly promising, because the latter should only affect the cancerous, but not the healthy, cells.30 Since MCL-1 and A1/BFL1 are abnormally highly expressed in certain cancers, it may also be of interest to develop BH3 mimetics that target these prosurvival BCL2 family members. Since MCL-1 is critical for the survival of several essential and hard-to-replace cell types, such as cardiomyocytes, it has been argued that targeting MCL-1 will not be a viable option for cancer therapy. It is, however, important to remember that irreversible loss of a protein does not equate to transient inhibition of a protein. Hence, targeting MCL-1 with BH3 mimetics may still be a viable strategy for cancer therapy, particularly in tumor cells exquisitely dependent on this prosurvival protein.

Figure 5
Figure 5. Apoptosis regulators targeted for
cancer therapy.

Diagram showing which components of the BCL2-regulated
and the death-receptor–initiated apoptotic pathways are being
targeted by anticancer therapeutics that are currently being
developed (indicated with a red circle).
Abbreviations: APAF-1, apoptotic protease activating factor 1;
BAK, BCL2 antagonist/killer; BAX, BCL2-associated X protein;
BCL2, B-cell chronic lymphocytic leukemia/lymphoma 2 protein;
BH, BCL2 homology domain; cyt C, cytochrome C; DIABLO,
direct inhibitor of apoptosis binding protein with low pI; FADD,
Fas-associated death domain; FASL, FAS ligand; FLIP, FADDlike
interleukin 1β–converting enzyme (FLICE)-inhibitory protein;
IAP, inhibitor of apoptosis; tBID, truncated BID (BH3-interactingdomain
death agonist); TNF-α, tumor necrosis factor α; TNFR1,
tumor necrosis factor type 1 receptor; TRAIL, TNF-related
apoptosis-inducing ligand; TRAILR, TRAIL receptor.

Finally, certain cancers, such as B-cell lymphomas in immune-suppressed transplant recipients and Kaposi sarcoma, are associated with infection by viruses that carry BCL2-like prosurvival proteins (eg, the Epstein-Barr virus BCL2 homolog, BHRF1, and the Kaposi sarcoma–associated virus BCL2 homolog, Ks-BCL2). If these proteins are critical for the sustained survival of these tumor cells, targeting them with BH3 mimetics may be an attractive strategy. Since it should be possible to engineer such compounds so that they do not bind to the human BCL2-like prosurvival proteins, they would be expected to cause minimal collateral damage. Thus it might even be possible to combine them with chemotherapeutic drugs that act nonspecifically, for example, by causing DNA damage (eg, cyclophosphamide, etoposide).

Other approaches to activate the BCL2-regulated pathway for cancer therapy
In addition to targeting prosurvival BCL2 family members directly, it may also be possible to act upon them indirectly by targeting one of their regulators. In the case of MCL-1, which has a short half-life due in part to ubiquitin-dependent proteasomal degradation, indirect targeting might be achieved by inhibiting a deubiquitinase that delays this degradation. Of course, since such regulators act not only upon MCL-1, but many other proteins as well, unwanted side effects may arise from effects on other client proteins.


In conclusion, the first cell death regulator, BCL2,21 and its function35 were discovered nearly 30 and 25 years ago, respectively. Since then, the field has made enormous progress, defining the components of the cell death pathways, their functions in health and disease (particularly cancer), and their structures. This has culminated in the development of the first 2 drugs that directly flip the apoptotic switch and they are showing early promise in clinical trials for treatment of CLL and certain other cancers. Having been involved in cell death research for nearly 25 years, I hope that these drugs and future compounds will contribute to the armamentarium for treating cancer and possibly other diseases.

Acknowledgments: I thank all present and past members of my laboratory and colleagues from The Walter and Eliza Hall Institute for their work and intellectual input. My current work is supported by the National Health and Medical Research Council, Australia (Program Grant 1016701 and Fellowship 1020363) and the Leukemia & Lymphoma Society (Specialized Center of Research 7001-13) and was made possible through Victorian State Government Operational Infrastructure Support and Australian Government NHMRC IRIISS.

1. Hotchkiss RS, Strasser A, McDunn JE, Swanson PE. Cell death. New Engl J Med. 2009;361(16):1570-1583.
2. Strasser A, O’Connor L, Dixit VM. Apoptosis signaling. Ann Rev Biochem. 2000;69:217-245.
3. Strasser A, Jost PJ, Nagata S. The many roles of FAS receptor signaling in the immune system. Immunity. 2009;30(2):180-192.
4. Timmer JC, Salvesen GS. Caspase substrates. Cell Death Differ. 2007;14(1): 66-72.
5. Strasser A, Harris AW, Cory S. Bcl-2 transgene inhibits T cell death and perturbs thymic self-censorship. Cell. 1991;67:889-899.
6. Veis DJ, Sorenson CM, Shutter JR, Korsmeyer SJ. Bcl-2-deficient mice demonstrate fulminant lymphoid apoptosis, polycystic kidneys, and hypopigmented hair. Cell.1993;75:229-240.
7. Motoyama N, Wang FP, Roth KA, et al. Massive cell death of immature hematopoietic cells and neurons in Bcl-x deficient mice. Science.1995;267:1506-1510.
8. Opferman J, Iwasaki H, Ong CC, et al. Obligate role of anti-apoptotic MCL-1 in the survival of hematopoietic stem cells. Science. 2005;307(5712):1101-1104.
9. Ottina E, Grespi F, Tischner D, et al. Targeting antiapoptotic A1/Bfl-1 by in vivo RNAi reveals multiple roles in leukocyte development in mice. Blood. 2012; 119(25):6032-6042.
10. Lindsten T, Ross AJ, King A, et al.The combined functions of proapoptotic Bcl-2 family members Bak and Bax are essential for normal development of multiple tissues. Mol Cell. 2000;6(6):1389-1399.
11. Ke F, Bouillet P, Kaufmann T, Strasser A, Kerr J, Voss AK. Consequences of the combined loss of BOK and BAK or BOK and BAX. Cell Death Dis. 2013;4:e650.
12. Chen L, Willis SN, Wei A, et al. Differential targeting of pro-survival Bcl-2 proteins by their BH3-only ligands allows complementary apoptotic function. Mol Cell. 2005;17(3):393-403.
13. Merino D, Giam M, Hughes PD, et al. The role of BH3-only protein Bim extends beyond inhibiting Bcl-2-like prosurvival proteins. J Cell Biol. 2009;186(3): 355-362.
14. Bouillet P, Metcalf D, Huang DC, et al. Proapoptotic Bcl-2 relative Bim required for certain apoptotic responses, leukocyte homeostasis, and to preclude autoimmunity. Science. 1999;286(5445):1735-1738.
15. Villunger A, Michalak EM, Coultas L, et al. p53- and drug-induced apoptotic responses mediated by BH3-only proteins puma and noxa. Science. 2003;302 (5647):1036-1038.
16. Jost PJ, Grabow S, Gray D, et al. XIAP discriminates between type I and type II FAS-induced apoptosis. Nature. 2009;460(7258):1035-1039.
17. Erlacher M, Laabi V, Manzl C, et al. Puma cooperates with Bim, the rate-limiting BH3-only protein in cell death during lymphocyte development, in apoptosis induction. J Exp Med. 2006;203(13):2939-2951.
18. Bouillet P, Cory S, Zhang LC, Strasser A, Adams JM. Degenerative disorders caused by Bcl-2 deficiency are prevented by loss of its BH3-only antagonist Bim. Dev Cell. 2001;1(5):645-653.
19. Liu X, Dai S, Zhu Y, Marrack P, Kappler JW. The structure of a Bcl-xL/Bim fragment complex: Implications for Bim function. Immunity. 2003;19(3):341-352.
20. Czabotar PE, Westphal D, Dewson G, et al. Bax crystal structures reveal how BH3 domains activate Bax and nucleate its oligomerization to induce apoptosis. Cell. 2013;152(3):519-531.
21. Tsujimoto Y, Finger LR, Yunis J, Nowell PC, Croce CM. Cloning of the chromosome breakpoint of neoplastic B cells with the t(14;18) chromosome translocation. Science. 1984;226(4678):1097-1099.
22. Beroukhim R, Mermel C, Porter D, et al. The landscape of somatic copy-number alteration across human cancers. Nature. 2010;463(7283):899-905.
23. Tagawa H, Karnan S, Suzuki R, et al. Genome-wide array-based CGH for mantle cell lymphoma: identification of homozygous deletions of the proapoptotic gene BIM. Oncogene. 2005;24(8):1348-1358.
24. Garrison SP, Jeffers JR, Yang C, et al. Selection against PUMA gene expression in Myc-driven B-cell lymphomagenesis. Mol Cell Biol. 2008;28(17):5391-5402.
25. Strasser A, Harris AW, Bath ML, Cory S. Novel primitive lymphoid tumours induced in transgenic mice by cooperation between myc and bcl-2. Nature. 1990;348(6299):331-333.
26. Egle A, Harris AW, Bouillet P, Cory S. Bim is a suppressor of Myc-induced mouse B cell leukemia. Proc Natl Acad Sci U S A. 2004;101(16):6164-6169.
27. Kelly PN, Grabow S, Delbridge AR, Strasser A, Adams JM. Endogenous Bcl-xL is essential for Myc-driven lymphomagenesis in mice. Blood. 2011;118(24): 6380-6286.
28. Glaser S, Lee EF, Trounson E, et al. Anti-apoptotic Mcl-1 is essential for the development and sustained growth of acute myeloid leukemia. Genes Dev. 2012; 26:120-125.
29. Happo L, Cragg MS, Phipson B, et al. Maximal killing of lymphoma cells by DNAdamage inducing therapy requires not only the p53 targets Puma and Noxa but also Bim. Blood. 2010;116(24):5256-5267.
30. Cragg MS, Harris C, Strasser A, Scott CL. Unleashing the power of inhibitors of oncogenic kinases through BH3 mimetics. Nat Rev Cancer. 2009;9(5):321-326.
31. Ng KP, Hillmer AM, Chuah CT, et al. A common BIM deletion polymorphism mediates intrinsic resistance and inferior responses to tyrosine kinase inhibitors in cancer. Nat Med. 2012;18(4):521-528.
32. Souers AJ, Leverson JD, Boghaert ER, et al. ABT-199, a potent and selective BCL-2 inhibitor, achieves antitumor activity while sparing platelets. Nat Med. 2013;19(2):202-208.
33. Lessene G, Czabotar PE, Sleebs BE, et al. Discovery, structure-guided design and validation of a novel, potent and selective inhibitor of the pro-survival BCL-2 family member BCL-XL. Nat Chem Biol. 2013;9(6):390-397.
34. Merino D, Khaw SL, Glaser SP, et al. Bcl-2, Bcl-xL, and Bcl-w are not equivalent targets of ABT-737 and navitoclax (ABT-263) in lymphoid and leukemic cells. Blood. 2012;119(24):5807-5816.
35. Vaux DL, Cory S, Adams JM. Bcl-2 gene promotes haemopoietic cell survival and cooperates with c-myc to immortalize pre-B cells. Nature. 1988;335:440-442.

Keywords: apoptosis; BCL2 family; cancer; caspase; intrinsic pathway; p53