CONNECTING PROLIFERATION AND APOPTOSIS IN DEVELOPMENT AND DISEASE David R. Hipfner and Stephen M. Cohen Abstract | Cells grow and divide rapidly during embryonic and postnatal development. Net tissue growth reflects the balance between the addition of new cells and the elimination of existing cells by programmed cell death. Cells compete for growth and survival factors to ensure an appropriate balance between the addition and elimination of cells. Elaborate mechanisms ensure that cells do not evade these constraints, and thereby prevent uncontrolled proliferation.
An increase in cell mass. ONCOGENE
A gene that is causally linked to the formation of tumours in vivo, usually as a result of mutant forms of the gene. BASIC HELIX–LOOP–HELIX LEUCINE-ZIPPER PROTEIN
A type of transcription factor with a basic domain, a ‘helix–loop–helix’ DNA-binding motif and a ‘leucine zipper’ dimerization domain. CYCLINS
A family of proteins, the levels of which fluctuate throughout the cell cycle. By activating cyclindependent kinases, they regulate the cell cycle. D- and E-type cyclins promote G1–S phase progression; A- and B-type cyclins regulate S–G2–M progression.
European Molecular Biology Laboratory, Meyerhofstrasse 1, 169117, Heidelberg, Germany. e-mails: [email protected]
; [email protected]
Much has been learned about the molecular mechanisms that control CELL GROWTH, cell division and cell death. Coordination of these processes is essential for normal embryonic development. In this review, we use the term ‘cell proliferation’ to denote the combination of cell growth and cell division. Although cell growth and cell division are distinct and separable processes, their rates are usually coordinated so that tissues grow by an increase in cell number but the size of constituent cells remains within a typical range. A notable exception to this occurs in endoreplicating tissues, which grow partly by replicating their DNA without cell division (this will not be discussed further; for a review, see REF. 1). Productive tissue growth requires the rate of cell proliferation to exceed that of cell death. At first glance, this might seem simple. In fact, in tissue culture, where conditions are conducive to cell survival and the control of cell proliferation rates predominates, it is simple. But, it has been known for nearly 30 years that cells in developing tissues compete with one another to survive and proliferate2. Increasing evidence, which is largely based on genetic studies in mice and fruitflies, indicates that the mechanisms regulating cell proliferation and apoptosis might also be intimately linked in growing animals. Recent evidence implicates these mechanisms in the competition for cell survival in vivo. This review will focus on the emerging connections between proliferation and apoptosis in the context of tissue growth during development.
NATURE REVIEWS | MOLECUL AR CELL BIOLOGY
The Myc connection
Myc in cell growth and division. Although the Myc gene was originally identified as an ONCOGENE, the normal function of the Myc family of transcription factors is in the control of cell proliferation in animals (reviewed in REFS 3,4). Mammals have three Myc proteins: c-Myc, NMyc and L-Myc. These proteins are BASIC HELIX–LOOP–HELIX LEUCINE-ZIPPER (bHLH-Zip) transcription factors that bind to the bHLH-Zip protein Max. Myc–Max heterodimers regulate the transcription of target genes. Genome-wide studies indicate that Myc proteins probably ‘fine-tune’ the expression of a large number of genes, rather than turning them on or off 5,6. Recent estimates from flies and mice indicate that 10–15% of all genes might be affected by Myc activity7,8. Two important groups of genes that are induced by Myc encode metabolic enzymes and proteins that are involved in the control of protein synthesis — including ribosomal proteins, translation-initiation factors and elongation factors6,9. This indicates that Myc regulates the production of the basic cellular growth machinery. Myc proteins also regulate several genes that encode cell-cycle regulators. These include regulators of G1 progression, such as D-type CYCLINS, cyclin-dependent kinase (CDK)4 (REFS 10,11) and E-type cyclins12,13, as well as E2F PROTEINS, which are essential for the initiation of DNA synthesis 7,14 (S-phase entry; FIG. 1). In flies, Myc has been shown to drive G1–S progression in cells of the WING IMAGINAL DISCS at least in part by
VOLUME 5 | OCTOBER 2004 | 8 0 5
©2004 Nature Publishing Group
Myc proteins RB Inhibitors
Promoters E2F Cyclin D
Cyclin E E2F CDK2
Figure 1 | Myc regulation of G1–S progression. Mitogenic signals that are mediated through Myc-family proteins promote G1–S progression by several mechanisms. Myc induces the expression of cyclin D and cyclin-dependent kinase (CDK)4, which together form an active kinase complex. Low-level phosphorylation of retinoblastoma (RB) family proteins by cyclin-D–CDK4 (or cyclin-D–CDK6, not shown) disrupts interactions between RB and E2F proteins, which relieves RB-mediated repression of E2F-target genes. Early E2F-dependent transcription of cyclin E (which might also be a Myc target) leads to the formation of active cyclin-E–CDK2 complexes. Cyclin-E–CDK2 functions in a positive-feedback loop to promote RB hyperphosphorylation. This releases E2F, which results in the full activation of E2F target genes that promote entry into S-phase. The activities of cyclin–CDK complexes are opposed by CDK inhibitors, which might be targets for Myc-mediated repression. p21CIP1 and p27KIP1 inactivate cyclin-E–CDK2 complexes. p21CIP1 and p27KIP1 are also sequestered by active cyclin-D–CDK complexes, which frees cyclin-E–CDK2 from their inhibitory effects. p16INK4a binds to, and disrupts, cyclin-D–CDK complexes, which blocks G1–S progression both by preventing the cyclin-D–CDK4-mediated RB protein phosphorylation and by releasing sequestered p21CIP1 and p27KIP1.
Members of a family of transcription factors that control the expression of genes that are involved in cell-cycle progression, including cyclin E, string CDC25 phosphatase and components of the DNAsynthesis machinery. WING IMAGINAL DISCS
Imaginal discs are the larval precursors of adult structures. The wing imaginal discs make the dorsal thorax and the wing appendages. HYPOMORPHIC
A mutation that reduces, but does not completely eliminate, the function of a gene. PRIMARY CELLS
Cultured cells that are derived directly from tissue (often embryonic tissue). They are distinct from transformed cell lines.
increasing the levels of Cyclin E, although this seems to be a post-transcriptional effect15,16. Cyclin E is rate-limiting for G1–S progression in flies, but is apparently not essential in mammalian cells17,18. Myc also downregulates the genes that encode the CDK inhibitors p27KIP1, p21CIP1 and p15INK4B, which inhibit G1 progression9,19–22. The reduction of p27KIP1 levels seems to be essential for c-Myc-induced cell-cycle progression23. Interestingly, the loss of p27 KIP1 alone can trigger increased organ and organism growth24–26. So, the Myc proteins seem to regulate the rates of cell growth and G1–S progression through coordinated effects on many genes and, although some details of the mechanism might differ, this role seems to be generally conserved in different systems. Myc mutants are small. A decrease in Myc activity reduces body size in mice and flies. The expression of c-Myc in mid-gestation mouse embryos correlates with patterns of proliferation — high levels of expression are observed only in rapidly proliferating tissues 27. Complete removal of c-Myc by targeted gene disruption in mice results in impaired growth of
| OCTOBER 2004 | VOLUME 5
the embryo and in embryonic death28, and a reduced level of c-Myc activity leads to decreased body size at birth29. c-Myc HYPOMORPHIC mice are smaller because they have fewer cells; cell size, however, is normal. PRIMARY CELLS that are derived from mice that lack c-Myc undergo cell-cycle arrest, but those that are derived from mice with reduced c-Myc levels still divide, albeit at a reduced rate that is dependent on their c-Myc levels29. Flies with decreased Myc activity are small because of a reduction in both cell size and cell number 15,30,31. Conversely, Myc overexpression in flies caused net tissue growth by increasing cell size 15. Originally, this was thought to occur without a change in cell number, which indicated that, in flies, the function of Myc was to control cell growth rather than proliferation per se. However, a more recent study has shown that Myc overexpression in flies drives net cell proliferation if apoptosis is prevented32. So, the main difference between the myc-mutant phenotypes in flies and mice is that a change in cell size is seen in the fly that is not seen in mice. It seems unlikely that these observed differences in cellular phenotypes reflect an intrinsic difference in the mouse and fly Myc proteins. Indeed, fly Myc can rescue the proliferation defects of c-Myc-mutant mouse fibroblasts29. But is there a fundamental difference in the way that Myc proteins affect cells in these systems? The impaired rate of tissue growth in mouse and fly Myc mutants indicates that there is probably an underlying reduction in the rate of biomass accumulation at the cellular level in both cases. Rat fibroblasts in which c-Myc has been disrupted accumulate protein and RNA 2–3-times more slowly than control cells, but they maintain a normal cell size33. This is apparently similar to the changes seen in c-Myc hypomorphic mice. Maintenance of cell size in vertebrates could be explained by a potential CELL-SIZE CHECKPOINT that delays division until cells reach their normal size, even though they grow more slowly34. It should be noted, however, that the existence of a strict cell-size-checkpoint mechanism has been questioned 35. In the fly, some experiments support the existence of a cell-size checkpoint; for example, MINUTE MUTATIONS reduce the rate of cell growth and proliferation owing to defects in ribosomal protein levels, but they do not affect cell size36. However, in most other cases, a decrease in the rate of cell growth is accompanied by a decrease in the size of dividing cells, whereas at an increased growth rate, cells divide at a larger size37. Cells that overexpress fly Myc progress five times more quickly through the G1 phase of the cell cycle compared with normal cells. Despite this, these cells divide at about the same rate as normal cells because of a compensatory increase in the length of G2 (REFS 15,38). In flies and mice, therefore, Myc probably promotes both cell growth and cell-cycle progression, at least during G1. So, the differences in the mutant phenotypes probably reflect a difference in the coupling of cell growth and cell proliferation during mammalian and fly development, rather than a fundamental difference in the Myc proteins.
©2004 Nature Publishing Group
BCL-XL BCL2-family proteins Hid
Activated effector caspases
A mechanism that ensures that cells divide at a defined size.
Figure 2 | Simplified model of intrinsic and extrinsic apoptosis pathways. The intrinsic death pathway (left, green) is activated by the release of cytochrome c from mitochondria in response to various stresses and developmental-death cues. These deathinducing signals are sensed by BH3-ONLY-DOMAIN PROTEINS of the BCL2 family (for example, PUMA), which, in turn, bind to pro-survival family members (such as BCL2 and BCL-XL) and prevent them from interacting with pro-apoptotic family members (such as BAX). When not bound to BCL2, BAX functions, at least in part, by perturbing mitochondrial membrane permeability, thereby promoting cytochrome-c release. Cytosolic cytochrome c triggers the formation of the apoptotic-protease-activating factor1 (APAF1) and caspase-9-containing ‘apoptosome’ as well as activation of caspase-9. Once activated, caspase-9 triggers the effector-caspase cascade, which leads to cell death. The EXTRINSIC DEATH PATHWAY (right, orange) is activated by the binding of secreted ligands such as Fas ligand (FasL) to ‘death receptors’ of the tumour-necrosis-factor-receptor family, such as Fas. Receptor aggregation recruits other proteins, including the adaptor protein Fas-associated death-domain protein (FADD) and pro-caspase-8, into the ‘death-inducing signalling complex’ (DISC). Cleavage and activation of caspase-8 in the complex triggers the effectorcaspase cascade, which leads to cell death. Caspases are inactivated by inhibitor of apoptosis (IAP) proteins, which can bind to and block the active site on the caspase. IAPs are targeted for degradation by the activity of pro-apoptotic proteins like second mitochondria-derived activator of caspase (SMAC)/Diablo and its functional homologues in flies, which include Grim, Reaper and Hid.
A class of mutations that produce a dominant growth defect. Minutes are caused by the mutation of genes that encode any of several ribosomal proteins. BH3-ONLY DOMAIN PROTEINS
Members of a class of pro-apoptotic proteins in the larger BCL2 family that contain a single ‘BCL2 homology-3’ domain. EXTRINSIC DEATH PATHWAY
An apoptotic signalling pathway that is activated by binding of secreted ligands to ‘death receptors’ of the tumournecrosis-factor-receptor family. MITOGENIC
Promotes cell proliferation. INTRINSIC DEATH PATHWAY
An apoptotic signalling pathway that is activated in response to various forms of intracellular stress.
Myc-induced apoptosis. In cell culture assays, c-Myc expression is induced in response to MITOGENIC factors3. cMyc can support proliferation even if mitogenic factors are withdrawn, but further survival cues are required to suppress apoptosis39–41. Despite stimulating cell proliferation and tissue growth rates in mice, c-Myc activation can lead to a net loss of cells through apoptosis42. Similarly, overexpression of Myc in flies induces cell proliferation and compensatory apoptosis32. It seems that c-Myc sensitizes cells to the pro-apoptotic effects of diverse cellular stresses by several means (reviewed in REF. 43). These include: partial activation of the INTRINSIC DEATH PATHWAY (FIG. 2) and the release of cytochrome c from mitochondria (possibly by suppression of levels of the pro-survival proteins BCL2 and BCL-XL and/or by induction of the proapoptotic protein PUMA)44–51; synergy with extrinsic death-receptor signalling52–54; and the generation of reactive oxygen species and damage to DNA55,56. The outcome of c-Myc activation therefore depends on whether the amount of survival factors is sufficient to prevent apoptosis41,57. Cell-culture conditions allow the proliferative response to predominate, but when
NATURE REVIEWS | MOLECUL AR CELL BIOLOGY
survival factors are limiting — for example, in low serum conditions — c-Myc activation is more likely to cause apoptosis39–41. In vivo, this might provide a general safeguard against uncontrolled oncogene-induced proliferation57,58. The failure to obtain sufficient support for growth would lead to the elimination of over-proliferating cells. Indeed, c-Myc-induced tumour formation is inhibited by c-Myc-induced apoptosis42,59–62. In transgenic mice, activation of c-Myc specifically in adult pancreatic β-cells rapidly induces their proliferation, but they then undergo massive apoptosis and are eliminated. However, co-expression of the anti-apoptotic BCL-XL protein in these β-cells results in the formation of tumours42. Suppression of the intrinsic apoptosis pathway might be of direct relevance in tumour formation as increased expression of endogenous pro-survival BCL2-family members is selected for in tumours of c-Myc transgenic mice47. The p53 tumour-suppressor protein also protects against c-Myc-induced tumorigenesis. This transcription factor regulates the expression of genes that block cell-cycle progression or that promote apoptosis
VOLUME 5 | OCTOBER 2004 | 8 0 7
©2004 Nature Publishing Group
REVIEWS not expressed in embryonic tissues, and mice that lack functional ARF are viable, although they are prone to developing cancer early in adult life69,71,72. Flies do not have ARF or MDM2 genes, but have a p53 homologue that is required for apoptosis in response to radiationinduced DNA damage73. It is not known if fly p53 is responsive to strong proliferative signals, although this seems likely.
Ub Ub Ub p53
p53 target genes
Figure 3 | The ARF–MDM2–p53 response to activated oncogenes. The p53 transcription factor activates target genes that promote cell-cycle arrest or apoptosis in response to cellular stresses such as oncogene activation. p53 activity is counteracted by the E3 ubiquitin ligase MDM2. Under normal circumstances, MDM2 maintains low levels of p53 by targeting the protein for degradation by the proteasome, and by directly blocking p53 transcriptional activity. MDM2 is also a transcriptional target of p53, which functions as an auto-regulatory negative-feedback mechanism. MDM2 activity is counteracted by ARF, one of two alternative open reading frame tumour-suppressor products that are produced from the p16INK4A-ARF locus. The ARF promoter is responsive to strong proliferation signals, such as activated oncogenes, which increase ARF transcription and protein levels. Binding of ARF to MDM2 inhibits MDM2-mediated p53 transcriptional silencing and degradation. In this way, oncogenic activation leads to increased p53 activity and cell-cycle arrest or apoptosis. Ub, ubiquitin.
An anti-apoptotic protein. It is the founding member of the BCL2 family of pro- and antiapoptotic proteins. E3 UBIQUITIN PROTEIN LIGASE
An enzyme that catalyses the addition of ubiquitin to target proteins, which signals their degradation via the proteasome pathway. WNT PROTEINS
A family of highly conserved secreted signalling molecules that regulate cell–cell interactions during embryogenesis. STERILE 20 KINASE
A member of a diverse family of serine/threonine protein kinases that share homology with the yeast Ste20 kinase.
in response to DNA damage or cell stress, and that result in cell-cycle arrest or cell death. Mutations that impair the p53 response suppress c-Myc-induced apoptosis and enhance tumour formation in mice59,60,63–67, whereas increased p53 responsiveness suppresses tumour formation60,68. Proliferative signals and p53 activation in mammalian cells are linked by ARF — one of two alternative open-reading-frame products of the p16INK4A-ARF tumour-suppressor locus. The ARF promoter is responsive to strong proliferative signals, and ARF transcription and protein levels are increased when c-Myc is activated69,70. In turn, ARF promotes the p53 response by inhibiting MDM2, an E3 UBIQUITIN LIGASE that blocks p53induced transcription and targets p53 for degradation (FIG. 3). ARF induction therefore sets a threshold for productive proliferative signalling. Cells in which these signals exceed the intensity limit for ARF induction activate p53 and are prone to apoptosis. This indicates that the level of survival signals that a cell receives probably determines whether a proliferative stimulus is ‘normal’ or ‘abnormal’, and that the threshold might vary between different tissues, or at different times. ARF is
| OCTOBER 2004 | VOLUME 5
Overcoming Myc-induced apoptosis. Cell-intrinsic mechanisms for limiting cell proliferation — such as those described above that involve c-Myc, p53 and ARF — are clearly important in preventing tumour formation. These pathways are essential for limiting the proliferative capacity of single cells that might incur an oncogenic lesion. However, they also restrict the normal physiological proliferation of cells, which then require increased survival inputs to support this process. Intestinal epithelial stem cells generate a compartment of proliferative progenitor cells in the intestinal crypts that divide rapidly as they migrate towards the intestinal epithelial surface. After they leave the intestinal crypts, these cells stop proliferating and undergo differentiation. Maintenance of the proliferative state is dependent on a WNT signal that is transduced by the β-catenin–T-cell-factor (TCF) transcriptional-regulatory complex. Mice with a targeted deletion of both copies of the Tcf-4 gene, or with transgenic intestinal expression of the secreted Wnt inhibitor Dickkopf-1, lack the proliferative crypt-cell compartment74,75. Studies indicate that c-Myc is a transcriptional target of signalling by Wnt proteins, and that they mediate their effects on crypt-cell proliferation through c-Myc75–77. Interestingly, Wnt proteins also inhibit c-Myc-induced apoptosis in cultured intestinal epithelial cells and in tumours in vivo 78. Wnt proteins seem to regulate the expression of survivin, a member of the inhibitor of apoptosis (IAP) protein family that is involved both in blocking apoptosis and in mitotic progression79–81. Like c-Myc, survivin is highly expressed in cells that are located deep in the intestinal crypts, and expression is lost as cells move out of range of the Wnt signal and undergo cell-cycle arrest. This implies that survivin might be important for maintaining survival of the proliferating cells80,81. However, it remains to be seen whether survivin mediates the effects of Wnt proteins by blocking c-Myc-induced apoptosis. Positive and negative growth regulators
It is not known whether the mechanisms that coordinate proliferation and apoptosis also have a role during embryonic growth in mammals. However, recent work on growth regulation in the fly supports this theory. The following sections outline recent progress in the understanding of mechanisms that balance proliferation and apoptosis during fly development to allow net tissue growth. Slik, a STERILE-20 (Ste20) KINASE, behaves like a conventional oncogene, promoting proliferation and apoptosis82. Hippo, another Ste20 kinase, behaves like a tumour suppressor by limiting proliferation and by promoting apoptosis83–86. Furthermore, a microRNA, bantam, promotes cell proliferation and prevents apoptosis87,88.
©2004 Nature Publishing Group
A small RNA of ~22 nucleotides that is encoded by an endogenous gene. The microRNA regulates expression of RNAs to which it is complementary in sequence.
Promoting proliferation and apoptosis: slik. The fly slik gene encodes a Ste20 serine/threonine kinase and was identified in a screen for genes that affect the rates of tissue growth82. Like E2F and Myc17,32,89, overexpression of Slik in flies promotes cell proliferation but also induces compensatory apoptosis. The extent of Slikinduced tissue overgrowth can be strongly increased by simultaneously blocking apoptosis, which indicates that apoptosis functions as a protective mechanism against excessive proliferation during development (FIG. 4). Slikinduced proliferation and apoptosis are mediated, at least in part, by Raf. Raf is a mitogen-activated protein kinase (MAPK) kinase kinase (MAPKKK) that phosphorylates and activates the MAPKK MEK (MAPK and extracellular signal-regulated kinase (ERK) kinase; known as Sor in the fly). MEK, in turn, phosphorylates and activates the MAPK, ERK. Genetic and biochemical analyses indicate that Slik can bind to Raf and can activate it by a mechanism that does not require Slik kinase activity. The effects of Slik on cell proliferation and survival do not require ERK/MAPK activity. This indicates that Raf does not require the canonical MAPK pathway to exert its effects on proliferation and survival when activated by Slik. Like fly Myc, Raf and one of its upstream activators, the small GTPase Ras, promote G1–S progression and increase the rates of cell and tissue growth in flies90. Interestingly, Ras and Raf regulate fly Myc levels90. This is consistent with previous observations in mammalian cells91, and indicates that stabilization of Myc might contribute to the proliferationpromoting effects of Slik and Raf. One intriguing feature of Slik is that cells are triggered to undergo apoptosis when Slik levels are too high and also when they are too low. Mutants with reduced slik activity show excessive apoptosis in the imaginal discs. This indicates that the level of Slik activity might be important for cell survival. Apoptosis in slik mutants is sensitive to Raf levels.Although Raf can function through ERK/MAPK to promote cell survival by reducing and/or inactivating the pro-apoptotic protein Hid92,93, ERK/MAPK does not seem to have a large role in Slikmediated survival signalling in vivo. In both flies and mammals there is evidence for ERK/MAPK-independent survival signalling downstream of Raf 94,95. For example, expression of a form of Raf1 that cannot phosphorylate MEK1 can nevertheless restore viability and rescue the excessive apoptosis phenotype of mice that lack endogenous Raf1 (REF. 96). In the fly eye, both ERK/MAPK-dependent and -independent pathways promote cell survival during differentiation97. These pathways of survival signalling might reflect different activities of Raf proteins in suppressing extrinsic (MEK1-dependent) versus intrinsic (MEK1-independent) apoptotic pathways in response to distinct signals94. Slik protein is expressed ubiquitously at a uniform level, which indicates that regulation of Slik activity might be used to convey survival and proliferative signals. It is not known how Slik activity is regulated. Promoting proliferation and suppressing apoptosis: bantam. The bantam gene was also identified in screens for loci that affect the rate of tissue growth in flies87,98.
NATURE REVIEWS | MOLECUL AR CELL BIOLOGY
d Slik + p35
Figure 4 | Slik-driven growth is opposed by apoptosis. a | Fruitfly adult wing. The arrow indicates the region between veins three and four. b | A wing in which Slik is expressed specifically between veins three and four (arrows). Slik overexpression leads to modest overgrowth of this region. c | An overlay of a (in green) and b (in red) shows the increased distance between veins three and four, which is caused by Slik-induced overgrowth. d | A wing in which Slik is co-expressed with the apoptosis-inhibitor p35. When apoptosis is prevented using p35, Slik expression leads to a much greater overgrowth of the tissue. Reproduced with permission from REF. 82.
Increased bantam activity accelerates cell proliferation and causes net tissue overgrowth, whereas reduced bantam activity slows growth. In contrast to Myc, E2F and Slik, bantam promotes cell proliferation without inducing compensatory apoptosis. It encodes a microRNA that is expressed in actively proliferating tissues in the imaginal discs and in the central nervous system of the fly88. The spatial pattern of bantam expression is controlled by patterning signals such as those from Wingless, a fly Wnt-family protein. As bantam
VOLUME 5 | OCTOBER 2004 | 8 0 9
©2004 Nature Publishing Group
Functionally related genes with extensive sequence similarity, which indicates a common ancestor. Orthologues are often used to indicate the most closely related members of larger gene families in different species. WW DOMAIN
A protein domain that is involved in binding to prolinerich peptide motifs. SCAFFOLD
A protein that functions as a support to assemble a multiprotein complex.
does not affect spatial patterning per se, its function might be to mediate the effects of Wingless and other patterning cues by influencing cell-proliferation rates in distinct spatial patterns. MicroRNAs control post-transcriptional gene expression by binding to complementary sequences in target mRNAs. Computational prediction of possible bantam targets has identified the pro-apoptotic gene hid 88. bantam negatively regulates the production of Hid protein through complementary sequences in the hid mRNA 3′ untranslated region. As strong proliferative signals induce apoptosis, one likely function of bantammediated Hid repression could be to suppress apoptosis that might be induced by the proliferation-inducing activity of bantam itself. bantam can also suppress proliferation-induced apoptosis that is caused by the expression of other genes, including E2f and its cofactor Dp88. Suppression of proliferation-induced apoptosis might be essential for allowing increased tissue growth to result from cell proliferation. On the other hand, suppression of apoptosis alone is not sufficient to generate tissue overgrowth, so there are probably more bantam targets that regulate cell proliferation. Unlike the situation for Myc or the target of rapamycin (Tor) pathway in flies99,100, cell sizes remain normal during bantam-induced tissue growth. (Tor is a protein kinase that controls the capacity of cells to produce protein in response to nutritional cues and insulin signalling.) This is also the case for growth that is induced by the fly Cyclin-D–Cdk4 complex101,102. Cyclin-D–Cdk4/6 can promote cell-cycle progression by phosphorylating and inhibiting the G1–S-phase inhibitor retinoblastoma (Rb) protein. The effects of Cyclin-D–Cdk4 on cellular growth have recently been attributed to the regulation of Hypoxia-inducible factor-1 (Hif-1) prolyl hydroxylase, which controls cellular growth rates by an unknown means103. It is not known whether Cyclin-D–Cdk4 also has anti-apoptotic activity, although, by analogy to bantam, this seems likely. Similar to Cyclin-D–Cdk4, bantam must also coordinately promote cell-cycle progression and stimulate cellular growth rates. bantam probably blocks the expression of an inhibitor of cell proliferation (or it might separately block inhibitors of cell-cycle progression and cell growth). However, at present, none of the known tumour-suppressor genes or negative-regulators of growth are predicted to be direct targets for regulation by bantam. Identification of such a bantam target might therefore uncover a new growth repressor. Although there is no clear-cut bantam ORTHOLOGUE in mammals, related DNA sequences have been identified, but their functions have not yet been tested. Promoting apoptosis and suppressing proliferation: hippo. The fly hippo, salvador/shar-pei and warts/lats genes were identified in genetic screens as negative regulators of tissue growth83–86,104–107. The three proteins function together in a complex. Hippo is a Ste20 kinase and Warts belongs to the nuclear Dbf2-related (NDR) family of serine/threonine protein kinases that have been implicated in the control of cell division and
| OCTOBER 2004 | VOLUME 5
morphogenesis108,109. Salvador is a WW-DOMAIN-containing protein that functions as a SCAFFOLD, facilitating the phosphorylation of Warts by Hippo83–86,106. The activity of this complex might be further modulated through tyrosine phosphorylation of Warts by Csk, the fly homologue of the carboxy-terminal Src kinase110. Interestingly, a mammalian homologue of Hippo (MST1) has been shown to phosphorylate histone H2B, and this modification might be involved in regulating chromatin condensation that occurs during apoptosis111. This indicates that Hippo might also function independently of Salvador and Warts. Mutants of hippo, salvador and warts have almost identical effects on the growth of the fly 83–86,104–107. The loss of any of these genes causes dramatic overgrowth of developing tissues and enlarged adult structures (FIG. 5). Mutant cells grow and proliferate faster than usual to produce enlarged tissues that have more cells of normal size. Increased proliferation of any of these mutant cells is associated with elevated levels of Cyclin-E mRNA and protein, which indicates that, in flies, these genes normally limit proliferation partly by reducing Cyclin-E levels to restrict G1–S progression83,84,86,105,106. However, the Hippo–Salvador–Warts complex must also target other growth regulators, as manipulation of the cell cycle alone is not sufficient to increase the growth rates of imaginal discs17. Proliferation that is induced in the absence of hippo, salvador or warts leads to massive tissue overgrowth because it is not accompanied by a compensatory increase in apoptosis. This is due to the increased expression of the anti-apoptotic protein Diap1, a member of the IAP family of proteins. Under normal conditions, Hippo–Salvador–Warts activity keeps the levels of Diap1 low by the suppression of diap1 transcription and by phosphorylation and destabilization of the Diap1 protein83,86 (FIG. 6). The increased Diap1 levels that result from removing Hippo–Salvador–Warts activity is sufficient to suppress apoptosis that is induced by the pro-apoptotic proteins Hid, Reaper or Grim84,86,106. Conversely, elevated levels of Hippo promote apoptosis, in part by induction of the Hid protein84. Therefore, the Hippo–Salvador–Warts complex functions during normal development to limit proliferation rates by controlling the expression of cell-cycle regulators and by promoting apoptosis. Control at the tissue level: cell competition
In addition to cell-intrinsic mechanisms that limit the growth capacity of an individual cell, growth during development relies on ‘social controls’ that are exerted at the level of tissues and organs (reviewed in REF. 112). Cells compete for survival factors in culture as well as in vivo. For instance, neuronal and glial cells are often produced in excess in the nervous system. Cells that can obtain sufficient amounts of survival signals from their respective target cells survive, whereas the remainder are eliminated by apoptosis112. Genetic studies in the fly have provided evidence that the control of proliferation and survival at the tissue level depends on CELL COMPETITION (BOX 1). Using GENETIC MOSAICS, it was shown that fasterproliferating cells out-competed slower-proliferating,
©2004 Nature Publishing Group
Figure 5 | The Hippo–Salvador–Warts complex suppresses tissue growth. a | Wild-type fly. b–d | Flies in which most of the head is composed of cells that lack Hippo (Hpo) (b), Warts (Wts) (c), or Salvador (Sav) (d). Loss of any of these genes leads to dramatic overgrowth of the eye and head. e–h | Cross-sections through developing eyes of flies of the genotypes that are depicted in a–d, respectively. The cells are stained with an antibody to demarcate cell outlines. e | Non-committed precursor cells normally undergo cell-cycle arrest at an earlier stage of eye development, whereas other cell types differentiate. Later, these cells differentiate into pigment cells, which form a single layered honeycomb-like meshwork between the clusters of photoreceptor cells. Excess cells are eliminated by apoptosis. f–h | In hippo (f), warts (g) or salvador (h) mutant eyes, the non-committed cells fail to arrest, and so overproliferate, and extra cells are not eliminated by apoptosis. Therefore, the eyes are overgrown because they contain too many pigment cells. Reproduced with permission from REF. 84 © (2003) Macmillan Magazines Ltd.
A phenomenon that is observed in imaginal discs — slowly dividing cells are eliminated by apoptosis owing to competition with faster-dividing cells. GENETIC MOSAICS
Tissues that are composed of cells of different genotypes. They are usually generated by controlled recombination during mitosis.
but otherwise healthy, cells during the growth of imaginal discs2. It was subsequently shown that competition might be due to a difference in the rates of cell growth. Removal of one copy of any of several genes that encode ribosomal proteins caused the resultant heterozygous mutant cells (Minutes; see BOX 1) to be out-competed by wild-type cells that had two normal copies of the gene36,113. As expected, the reduced biosynthetic capacity of myc hypomorphic mutant cells renders them susceptible to competition when they are mixed with normal cells15. Conversely, increasing Myc expression levels in the fly is sufficient to induce cell competition that leads to the elimination of neighbouring wild-type cells32,114. Other mutations that have cell-intrinsic effects on proliferation can also lead to cell competition. For example, ras- or slik-mutant cells are out-competed in mosaic tissues unless they are given a compensatory growth advantage by other means16,82. Competition for survival factors that are present at limiting levels has been proposed as a mechanism that can explain cell competition112. The secreted signalling proteins Dpp (Decapentaplegic, a member of the bone-morphogenetic-protein family) and Wingless have been implicated in the control of cell survival in imaginal discs115–118. These secreted molecules are present at limiting levels in the imaginal discs, as increased expression of either molecule leads to disc overgrowth118–120. Cells that show impaired Wingless or Dpp signalling activate the Jun amino-terminal kinase (JNK) pathway, which promotes apoptosis118,121. The idea that cells compete for ligands has recently gained support from studies of Dpp signalling in cells of wing
NATURE REVIEWS | MOLECUL AR CELL BIOLOGY
imaginal discs that were subjected to cell competition owing to altered expression levels of a Minute gene or of myc 114,122. In these studies, cells that proliferated more slowly showed signs of reduced Dpp-signalling activity, including a partial failure to repress brinker, a target of Dpp signalling. Increased brinker expression induced JNK activity and caused apoptosis. Survival of these cells could be improved by activating Dpp signalling or by expressing Rab5, a small GTPase that promotes endocytosis. These observations imply that slower-proliferating cells might be less efficient at acquiring or internalizing the Dpp ligand, which therefore leads to reduced signalling activity114,122. However, a similar study of cell competition that was induced by differences in myc levels came to a different conclusion32. The authors of this study found that Dpp signalling was not affected in the out-competed cells. The apoptosis that arose from competition seemed to result primarily from increased expression of hid in the slower-proliferating cells, rather than from JNK activation. The competitive effects that were induced by Myc expression were strongest when wild-type cells were in close proximity to the faster-growing Mycexpressing cells. But Myc-overexpressing cells also out-competed slower-proliferating cells located at a distance, although the competitive effects were decreased32. The induction of hid expression in these distant slower-proliferating cells could still reflect an increased ‘consumption’ of limited survival ligands by the faster-proliferating cells. However, these results might indicate that Dpp is not necessarily the relevant survival-promoting ligand.
VOLUME 5 | OCTOBER 2004 | 8 1 1
©2004 Nature Publishing Group
Apoptosis P Diap1
Figure 6 | Involvement of the Hippo–Salvador–Warts complex in apoptosis and proliferation. Hippo (Hpo), Salvador (Sav) and Warts (Wts) function in a complex to suppress tissue growth. Both Salvador and Warts are substrates for serine phosphorylation by Hippo. Salvador functions as a scaffold, binding to both Hippo and Warts and facilitating Warts phosphorylation. Warts is also phosphorylated on a conserved carboxy-terminal tyrosine residue by carboxy-terminal Src kinase (Csk), and this phosphorylation is essential for Warts function in vivo. The Hippo–Salvador–Warts complex promotes apoptosis by blocking transcription of the gene that encodes the Diap1 apoptosis inhibitor, as well as by phosphorylating and destabilizing Diap1 protein. Cells that lack Hippo–Salvador–Warts are resistant to apoptosis owing to increased levels of Diap1 protein. Hippo–Salvador–Warts regulate the cell cycle by suppressing cyclin-E transcription by an unknown mechanism, and thereby blocking G1–S-phase progression. It is not yet clear how the loss of these activities leads to increased rates of tissue growth in Hippo–Salvador–Warts mutants. P, phosphate.
These data imply that, during the growth of a tissue, there might be a complex interplay between the proliferative capacity of a cell and its ability to obtain the secreted signalling molecules that are needed for survival. When cell-proliferation rates are manipulated, cells that over-proliferate might be unable to obtain sufficient quantities of the limiting survival signals, and therefore die. Cells that proliferate too slowly might also get less of these ligands in competition with their neighbours, and so they also die. Furthermore, many secreted signalling molecules, including Dpp, Wingless, Epidermal growth factor (EGF) ligands and Hedgehog also regulate cell proliferation rates in imaginal discs. Activation of any of these pathways stimulates cell proliferation and tissue overgrowth, whereas impaired signalling reduces growth rates16,90,116–118,123. These effects on proliferation might occur partly through regulation of Myc, as Dpp90, Wingless15 and EGF-receptor signalling90 can control Myc levels either transcriptionally or post-translationally. Cells, therefore, are in competition with their neighbours for the combined proliferative and survival signals that are necessary for their continued existence. Cell competition and regulation of final organ size? Cell competition for growth factors has typically been interpreted as a quality-control mechanism.
| OCTOBER 2004 | VOLUME 5
Elimination of slowly-dividing cells ensures that only the most ‘fit’ cells contribute to the final organism. However, more recent observations indicate that cell competition might be an important part of the mechanism that regulates the final size to which a tissue or organ can grow32. In the fly, the growth regulators Cyclin-D–Cdk4 and Phosphatidylinositol-3-kinase (Pi3k) can drive overgrowth of a tissue when they are expressed in restricted subsets of cells within the tissue, an effect not seen after expression of Myc15,101,114,124. The ability of these proteins to promote tissue overgrowth is inversely correlated with their ability to induce cell competition. For example, expression of Myc in a subset of cells induces competition in neighbouring cells, whereas Cyclin-D–Cdk4 or Pi3k expression does not32. Elimination of cells by competitioninduced apoptosis might therefore be essential for limiting tissue size. Three further observations support this conclusion. First, if the apoptosis of normal cells that is induced by competition with neighbouring Myc-overexpressing cells is blocked, then Myc overexpression leads to tissue overgrowth. Second, if cell competition is eliminated by ubiquitous expression of a myc transgene, which artificially enforces a uniform rate of proliferation in cells within a tissue, then the increased Myc activity also leads to tissue overgrowth. Third, if cell competition is reintroduced in this background by inactivation of the myc transgene in subsets of cells, the tissue overgrowth that is induced by ubiquitous expression of Myc is reduced. Competitioninduced apoptosis might therefore provide a means of regulating tissue size, and signals that induce tissue growth without competition could override the normal mechanisms that restrict tissue growth32. Apoptosis — enforcing an optimum speed?
Recent studies investigating genes that regulate growth in the fly have identified a remarkable flexibility in the coupling of proliferation and apoptosis during development. Rapid tissue growth requires both proliferation and suppression of apoptosis. Some proteins that drive proliferation, such as Myc and Slik, are dependent on survival signals to maintain the viability of proliferating cells. Other molecules, such as bantam, are capable of providing both signals. The limiting levels of survival ligands ensure that excessive activity of proliferative cues, such as Myc and Slik, leads to apoptosis rather than tissue overgrowth, and that cells are eliminated through competition. Tumour-suppressor proteins like the Hippo–Salvador–Warts complex function as a brake; they constrain growth rates by slowing proliferation and promoting apoptosis (FIG. 6). Together, these mechanisms ensure that the relative rates of cell addition and cell loss are balanced, which leads to the formation of organs that are of the appropriate size and shape. During wing growth in the fly, only 1–2% of cells undergo apoptosis at any given time125, and preventing this cell death does not increase the average final wing size. As a result of such observations, the contribution of apoptosis to tissue growth has been overlooked until
©2004 Nature Publishing Group
Box 1 | Visualizing cell competition in vivo
The term ‘cell competition’ was first used to describe the behaviour of fruitfly cells that were heterozygous for mutations in individual ribosomal proteins (termed Minutes) in an otherwise wild-type fly 2. Loss of one copy of a Minute gene impairs the rate at which cells proliferate. Consequently, Minute+/– animals grow slowly. However, they typically grow to the normal final body size with only minor defects, which indicates that the Minute mutations do not have other adverse effects36. Clones of Minute+/– cells are eliminated when they are surrounded by wild-type cells as a result of their slower proliferation rate 2,113. This might reflect the inability of the slowly dividing cells to obtain sufficient levels of survival signals when in competition with their faster-dividing neighbours114,122. The competitive effects are strongest at the interface between slowerand faster-dividing cells, indicating that local cell–cell interactions might be involved113. Cells that are located at a distance from the faster-dividing cells are, however, also affected32. Other mutations that impair proliferation rates, such as loss of Slik, can also cause cells to be out-competed. The left panel shows a wing disc in which wild-type clones were generated by mitotic recombination. Wild-type cells that are induced to lose expression of their only copy of a green fluorescent protein (GFP) transgene (and therefore appear black; indicated by arrows) grow to form cell clones of the same size as the wild-type twin clones with two copies of the GFP transgene that are induced at the same time (bright green; indicated by arrowheads). All other cells contain one copy of the GFP transgene and appear light green. The middle panel shows a wing disc in which slik-mutant clones were generated by mitotic recombination. These slik-mutant cells (which lack GFP) proliferate slowly. As a result, slik-mutant clones are eliminated by cell competition and are not observed in the discs. Only the wild-type clones remain (indicated by arrowheads). The right panel shows a wing disc in which slik-mutant clones were generated by mitotic recombination and were given a proliferative advantage. When slikmutant cells (which lack GFP; as indicated by the arrow) are given a proliferative advantage relative to neighbouring cells, they are no longer subject to competition and form large clones. In this case, the surrounding cells proliferate at a slower rate than the slik-mutant cells as they are missing one copy of a Minute gene (Minute+/–), whereas the slik-mutant cells are Minute+/+. The Minute –/– twin clones are absent because the Minute mutation is lethal.
recently. However, blocking apoptosis in the developing wing does lead to an increased variability in tissue size, although the average tissue size remains unchanged. This implies that apoptosis contributes to the precision of size control32. Proliferation-induced apoptosis sets a maximum rate at which cells can proliferate, whereas elimination of slow-growing cells by competition-induced apoptosis sets a minimum proliferation rate. Together, these mechanisms might enforce an optimal rate of tissue growth to ensure the fidelity of final tissue size regulation. These mechanisms are important in the control of growth rates, but the question of how final tissue and organ size is defined remains unanswered.
NATURE REVIEWS | MOLECUL AR CELL BIOLOGY
Much of the work on the links between proliferation and apoptosis in vertebrates has focused on the role of these pathways in cell transformation and tumour formation. A failure to regulate proliferation together with suppression of apoptosis are the minimal requirements for a cell to become cancerous 57. In the context of aberrant growth control, p53- and mitochondria-dependent apoptotic pathways have an important protective role against undesired cell proliferation in mammals. A similar proliferation-induced apoptosis mechanism seems to exist in flies, which protects against overgrowth of tissues during development. As there is no ARF orthologue in flies, the regulation is probably different. It will be interesting to see how proliferative signals in flies sensitize cells to apoptosis, and whether this is important for limiting tissue size during development. The recent characterization of growth regulators in flies such as Slik, bantam, and Hippo–Salvador–Warts further underscores the importance of coordinating proliferation and apoptosis for productive growth. Although some of the effectors of cell-cycle regulation and apoptosis downstream of these molecules have been identified, it is largely unknown how they regulate growth rates. Vertebrate orthologues of Slik (lymphocyte-oriented kinase (LOK) and Ste20-like kinase (SLK)), Hippo (mammalian Ste20-like-1 and -2 (MST1 and MST2)), Salvador (WW45) and Warts (large tumour suppressor (LATS)1 and LATS2) have been identified. In some cases, the conservation of function across species has been confirmed83,126. Mice that lack Lats1 are prone to developing tumours127, and both LATS1 and WW45 are mutated in some human cancers and cell lines. These observations indicate that this protein complex might also be important in diseases that are associated with dysregulated proliferation106,128. The existence of functional orthologues of these and other growth-regulatory gene products in both flies and mammals further supports the idea that the regulation of tissue growth is a conserved process, although the mechanistic details might vary across species. The data from flies indicate that growth regulation probably depends on the limitation of the levels of secreted ligands that promote proliferation and cell survival. There is clear evidence that secreted signalling proteins are involved in the control of proliferation and cell survival in mammalian cancers. For example, inappropriate regulation of Wnt, Hedgehog or EGF signalling can contribute to tumours in mice and has been implicated in human cancers129–134. Similarly, secreted signalling molecules are required to support cell survival and proliferation during mammalian embryonic growth135,136. It remains to be seen whether competition for limiting levels of signalling proteins is an important mechanism for balancing apoptosis and proliferation during growth of vertebrate embryos in a manner that is comparable to cell competition in the fly.
VOLUME 5 | OCTOBER 2004 | 8 1 3
©2004 Nature Publishing Group
REVIEWS 1. 2.
5. 6. 7. 8.
Edgar, B. A. & Orr-Weaver, T. L. Endoreplication cell cycles: more for less. Cell 105, 297–306 (2001). Morata, G. & Ripoll, P. Minutes: mutants of Drosophila autonomously affecting cell division rate. Dev. Biol. 42, 211–221 (1975). Grandori, C., Cowley, S. M., James, L. P. & Eisenman, R. N. The Myc/Max/Mad network and the transcriptional control of cell behavior. Annu. Rev. Cell Dev. Biol. 16, 653–699 (2000). Pelengaris, S., Khan, M. & Evan, G. c-MYC: more than just a matter of life and death. Nature Rev. Cancer 2, 764–776 (2002). Eisenman, R. N. Deconstructing myc. Genes Dev. 15, 2023–2030 (2001). Levens, D. Disentangling the MYC web. Proc. Natl Acad. Sci. USA 9, 5757–5759 (2002). Fernandez, P. C. et al. Genomic targets of the human c-Myc protein. Genes Dev. 17, 1115–1129 (2003). Orian, A. et al. Genomic binding by the Drosophila Myc, Max, Mad/Mnt transcription factor network. Genes Dev. 17, 1101–1114 (2003). Zeller, K. I., Jegga, A. G., Aronow, B. J., O’Donnell, K. A. & Dang, C. V. An integrated database of genes responsive to the Myc oncogenic transcription factor: identification of direct genomic targets. Genome Biol. 4, R69 (2003). Bouchard, C. et al. Regulation of cyclin D2 gene expression by the Myc/Max/Mad network: Myc-dependent TRRAP recruitment and histone acetylation at the cyclin D2 promoter. Genes Dev. 15, 2042–2047 (2001). Hermeking, H. et al. Identification of CDK4 as a target of c-MYC. Proc. Natl Acad. Sci. USA 97, 2229–2234 (2000). Perez-Roger, I., Solomon, D. L., Sewing, A. & Land, H. Myc activation of cyclin E/Cdk2 kinase involves induction of cyclin E gene transcription and inhibition of p27Kip1 binding to newly formed complexes. Oncogene 14, 2373–2381 (1997). O’Connell, B.C. et al. A large scale genetic analysis of c-Myc-regulated gene expression patterns. J. Biol. Chem. 278, 12563–12573 (2003). Sears, R., Ohtani, K. & Nevins, J. R. Identification of positively and negatively acting elements regulating expression of the E2F2 gene in response to cell growth signals. Mol. Cell. Biol. 17, 5227–5235 (1997). Johnston, L. A., Prober, D. A., Edgar, B. A., Eisenman, R. N. & Gallant, P. Drosophila myc regulates cellular growth during development. Cell 98, 779–790 (1999). First papers (see also reference 30) implicating the Myc oncogene in the control of cellular growth and size in animal development. Prober, D. A. & Edgar, B. A. Ras1 promotes cellular growth in the Drosophila wing. Cell 100, 435–446 (2000). Neufeld, T. P., de la Cruz, A. F., Johnston, L. A. & Edgar, B. A. Coordination of growth and cell division in the Drosophila wing. Cell 93, 1183–1193 (1998). Shows that, as in yeast, cell division and cellular growth are separable processes in higher eukaryotes, and that tissue size is not measured in terms of the number of cell divisions, but in terms of mass. Geng, Y. et al. Cyclin E ablation in the mouse. Cell 114, 431–443 (2003). Claassen, G. F. & Hann, S. R. A role for transcriptional repression of p21CIP1 by c-Myc in overcoming transforming growth factor β-induced cell-cycle arrest. Proc. Natl Acad. Sci. USA 97, 9498–9503 (2000). Coller, H. A. et al. Expression analysis with oligonucleotide microarrays reveals that MYC regulates genes involved in growth, cell cycle, signaling, and adhesion. Proc. Natl Acad. Sci. USA 97, 3260–3265 (2000). Yang, W. et al. Repression of transcription of the p27Kip1 cyclin-dependent kinase inhibitor gene by c-Myc. Oncogene 20, 1688–1702 (2001). Staller, P. et al. Repression of p15INK4b expression by Myc through association with Miz-1. Nature Cell Biol. 3, 392–399 (2001). Martins, C. P. & Berns, A. Loss of p27Kip1 but not p21Cip1 decreases survival and synergizes with MYC in murine lymphomagenesis. EMBO J. 21, 3739–3748 (2002). Fero, M. L. et al. A syndrome of multiorgan hyperplasia with features of gigantism, tumorigenesis, and female sterility in p27Kip1-deficient mice. Cell 85, 733–744 (1996). Kiyokawa, H. et al. Enhanced growth of mice lacking the cyclin-dependent kinase inhibitor function of p27Kip1. Cell 85, 721–732 (1996). Nakayama, K. et al. Mice lacking p27Kip1 display increased body size, multiple organ hyperplasia, retinal dysplasia, and pituitary tumors. Cell 85, 707–720 (1996). Schmid, P., Schulz, W. A. & Hameister, H. Dynamic expression pattern of the myc protooncogene in midgestation mouse embryos. Science 243, 226–229 (1989).
28. Davis, A. C., Wims, M., Spotts, G. D., Hann, S. R. & Bradley, A. A null c-myc mutation causes lethality before 10.5 days of gestation in homozygotes and reduced fertility in heterozygous female mice. Genes Dev. 7, 671–682 (1993). 29. Trumpp, A. et al. c-Myc regulates mammalian body size by controlling cell number but not cell size. Nature 414, 768–773 (2001). Shows that mammalian c-MYC controls organ and body size by regulating cell numbers. 30. Gallant, P., Shiio, Y., Cheng, P. F., Parkhurst, S. M. & Eisenman, R. N. Myc and Max homologs in Drosophila. Science 274, 1523–1527 (1996). 31. Bourbon, H. M. et al. A P-insertion screen identifying novel X-linked essential genes in Drosophila. Mech. Dev. 110, 71–83 (2002). 32. de la Cova, C., Abrili, M., Bellosta, P., Gallant, P. & Johnston, L. A. Drosophila Myc regulates organ size by inducing cell competition. Cell 117, 107–116 (2004). Together with references 114 and 122, provides strong evidence that cell competition contributes significantly to the control of organ size. 33. Mateyak, M. K., Obaya, A. J., Adachi, S. & Sedivy, J. M. Phenotypes of c-Myc-deficient rat fibroblasts isolated by targeted homologous recombination. Cell Growth Differ. 8, 1039–1048 (1997). 34. Thomas, G. An encore for ribosome biogenesis in the control of cell proliferation. Nature Cell Biol. 2, E71–E72 (2000). 35. Conlon, I. & Raff, M. Differences in the way a mammalian cell and yeast cells coordinate cell growth and cell-cycle progression. J. Biol. 2, 7 (2003). 36. Lambertsson, A. The minute genes in Drosophila and their molecular functions. Adv. Genet. 38, 69–134 (1998). 37. Saucedo, L. J. & Edgar, B. A. Why size matters: altering cell size. Curr. Opin. Genet. Dev. 12, 565–571 (2002). 38. Reis, T. & Edgar, B. A. Negative regulation of dE2F1 by cyclin-dependent kinases controls cell cycle timing. Cell 117, 253–264 (2004). 39. Askew, D. S., Ashmun, R. A., Simmons, B. C. & Cleveland, J. L. Constitutive c-myc expression in an IL-3-dependent myeloid cell line suppresses cell cycle arrest and accelerates apoptosis. Oncogene 6, 1915–1922 (1991). 40. Evan, G. I. et al. Induction of apoptosis in fibroblasts by c-myc protein. Cell 69, 119–128 (1992). References 39 and 40 report the striking finding that an oncogene induces apoptosis. 41. Harrington, E. A., Bennett, M. R., Fanidi, A. & Evan, G. I. c-Myc-induced apoptosis in fibroblasts is inhibited by specific cytokines. EMBO J. 13, 3286–3295 (1994). 42. Pelengaris, S., Khan, M. & Evan, G. I. Suppression of Mycinduced apoptosis in β cells exposes multiple oncogenic properties of Myc and triggers carcinogenic progression. Cell 109, 321–334 (2002). Shows that complex angiogenic and invasive tumours can be generated by driving proliferation and blocking apoptosis, without the need for additional mutations. 43. Nilsson, J. A. & Cleveland, J. L. Myc pathways provoking cell suicide and cancer. Oncogene 22, 9007–9021 (2003). 44. Juin, P., Hueber, A. O., Littlewood, T. & Evan, G. c-Mycinduced sensitization to apoptosis is mediated through cytochrome c release. Genes Dev. 13, 1367–1381 (1999). 45. Maclean, K. H., Keller, U. B., Rodriguez-Galindo, C., Nilsson, J. A. & Cleveland, J. L. c-Myc augments γ irradiation-induced apoptosis by suppressing Bcl-XL. Mol. Cell. Biol. 23, 7256–7270 (2003). 46. Iaccarino, I., Hancock, D., Evan, G. & Downward, J. c-Myc induces cytochrome c release in Rat1 fibroblasts by increasing outer mitochondrial membrane permeability in a Bid-dependent manner. Cell Death Differ. 10, 599–608 (2003). 47. Eischen, C. M., Woo, D., Roussel, M. F. & Cleveland, J. L. Apoptosis triggered by Myc-induced suppression of Bcl-XL or Bcl-2 is bypassed during lymphomagenesis. Mol. Cell. Biol. 21, 5063–5070 (2001). 48. Eischen, C. M. et al. Bcl-2 is an apoptotic target suppressed by both c-Myc and E2F-1. Oncogene 20, 6983–6993 (2001). 49. Mitchell, K. O. et al. Bax is a transcriptional target and mediator of c-myc-induced apoptosis. Cancer Res. 60, 6318–6325 (2000). 50. Juin, P. et al. c-Myc functionally cooperates with Bax to induce apoptosis. Mol. Cell. Biol. 22, 6158–6169 (2002). 51. Morrish, F., Giedt, C. & Hockenbery, D. c-MYC apoptotic function is mediated by NRF-1 target genes. Genes Dev. 17, 240–255 (2003). 52. Hueber, A. O. et al. Requirement for the CD95 receptorligand pathway in c-Myc-induced apoptosis. Science 278, 1305–1309 (1997). 53. Amanullah, A., Liebermann, D. A. & Hoffman, B. Deregulated c-Myc prematurely recruits both Type I and II
| OCTOBER 2004 | VOLUME 5
CD95/Fas apoptotic pathways associated with terminal myeloid differentiation. Oncogene 21, 1600–1610 (2002). 54. Klefstrom, J., Verschuren, E. W. & Evan, G. c-Myc augments the apoptotic activity of cytosolic death receptor signaling proteins by engaging the mitochondrial apoptotic pathway. J. Biol. Chem. 277, 43224–43232 (2002). 55. Vafa, O. et al. c-Myc can induce DNA damage, increase reactive oxygen species, and mitigate p53 function: a mechanism for oncogene-induced genetic instability. Mol. Cell 9, 1031–1044 (2002). 56. Tanaka, H. et al. E2F1 and c-Myc potentiate apoptosis through inhibition of NF-κB activity that facilitates MnSOD-mediated ROS elimination. Mol. Cell 9, 1017–1029 (2002). 57. Green, D. R. & Evan, G. I. A matter of life and death. Cancer Cell 1, 19–30 (2002). 58. Evan, G. & Littlewood, T. A matter of life and cell death. Science 281, 1317–1322 (1998). 59. Schmitt, C. A., McCurrach, M. E., de Stanchina, E., Wallace-Brodeur, R. R. & Lowe, S. W. INK4a/ARF mutations accelerate lymphomagenesis and promote chemoresistance by disabling p53. Genes Dev. 13, 2670–2677 (1999). 60. Jacobs, J. J. et al. Bmi-1 collaborates with c-Myc in tumorigenesis by inhibiting c-Myc-induced apoptosis via INK4a/ARF. Genes Dev. 13, 2678–2690 (1999). 61. Strasser, A., Harris, A. W., Bath, M. L. & Cory, S. Novel primitive lymphoid tumours induced in transgenic mice by cooperation between myc and bcl-2. Nature 348, 331–333 (1990). 62. Eischen, C. M., Roussel, M. F., Korsmeyer, S. J. & Cleveland, J. L. Bax loss impairs Myc-induced apoptosis and circumvents the selection of p53 mutations during Mycmediated lymphomagenesis. Mol. Cell. Biol. 21, 7653–7662 (2001). 63. Blyth, K. et al. Synergy between a human c-myc transgene and p53 null genotype in murine thymic lymphomas: contrasting effects of homozygous and heterozygous p53 loss. Oncogene 10, 1717–1723 (1995). 64. Elson, A., Deng, C., Campos-Torres, J., Donehower, L. A. & Leder, P. The MMTV/c-myc transgene and p53 null alleles collaborate to induce T-cell lymphomas, but not mammary carcinomas in transgenic mice. Oncogene 11, 181–190 (1995). 65. Russell, J. L. et al. ARF differentially modulates apoptosis induced by E2F1 and Myc. Mol. Cell. Biol. 22, 1360–1368 (2002). 66. Eischen, C. M., Weber, J. D., Roussel, M. F., Sherr, C. J. & Cleveland, J. L. Disruption of the ARF–Mdm2–p53 tumor suppressor pathway in Myc-induced lymphomagenesis. Genes Dev. 13, 2658–2669 (1999). Shows that induction of tumours by Myc in vivo selects for disruption of ARF–MDM2–p53 regulation. 67. Fridman, J. S. et al. Tumor promotion by Mdm2 splice variants unable to bind p53. Cancer Res. 63, 5703–5706 (2003). 68. Alt, J. R., Greiner, T. C., Cleveland, J. L. & Eischen, C. M. Mdm2 haplo-insufficiency profoundly inhibits Myc-induced lymphomagenesis. EMBO J. 22, 1442–1450 (2003). 69. Zindy, F. et al. Myc signaling via the ARF tumor suppressor regulates p53-dependent apoptosis and immortalization. Genes Dev. 12, 2424–2433 (1998). 70. Zindy, F. et al. Arf tumor suppressor promoter monitors latent oncogenic signals in vivo. Proc. Natl Acad. Sci. USA 100, 15930–15935 (2003). 71. Serrano, M. et al. Role of the INK4a locus in tumor suppression and cell mortality. Cell 85, 27–37 (1996). 72. Kamijo, T. et al. Tumor suppression at the mouse INK4a locus mediated by the alternative reading frame product p19ARF. Cell 91, 649–659 (1997). 73. Sogame, N., Kim, M. & Abrams, J. M. Drosophila p53 preserves genomic stability by regulating cell death. Proc. Natl Acad. Sci. USA 100, 4696–4701 (2003). 74. Korinek, V. et al. Depletion of epithelial stem-cell compartments in the small intestine of mice lacking Tcf-4. Nature Genet. 19, 379–383 (1998). 75. Pinto, D., Gregorieff, A., Begthel, H. & Clevers, H. Canonical Wnt signals are essential for homeostasis of the intestinal epithelium. Genes Dev. 17, 1709–1713 (2003). 76. van de Wetering, M. et al. The α-catenin/TCF-4 complex imposes a crypt progenitor phenotype on colorectal cancer cells. Cell 111, 241–250 (2002). 77. He, T.C. et al. Identification of c-MYC as a target of the APC pathway. Science 281, 1509–1512 (1998). 78. You, Z. et al. Wnt signaling promotes oncogenic transformation by inhibiting c-Myc-induced apoptosis. J. Cell Biol. 157, 429–440 (2002). 79. Altieri, D. C. Validating survivin as a cancer therapeutic target. Nature Rev. Cancer 3, 46–54 (2003).
©2004 Nature Publishing Group
REVIEWS 80. Zhang, T. et al. Evidence that APC regulates survivin expression: a possible mechanism contributing to the stem cell origin of colon cancer. Cancer Res. 61, 8664–8667 (2001). 81. Kim, P. J., Plescia, J., Clevers, H., Fearon, E. R. & Altieri, D. C. Survivin and molecular pathogenesis of colorectal cancer. Lancet 362, 205–209 (2003). 82. Hipfner, D. R. & Cohen, S. M. The Drosophila Sterile-20 kinase Slik controls cell proliferation and apoptosis during imaginal disc development. PLoS Biol. 1, 244–256 (2003). 83. Wu, S., Huang, J., Dong, J. & Pan, D. hippo encodes a Ste-20 family protein kinase that restricts cell proliferation and promotes apoptosis in conjunction with salvador and warts. Cell 114, 445–456 (2003). 84. Udan, R. S., Kango-Singh, M., Nolo, R., Tao, C. & Halder, G. Hippo promotes proliferation arrest and apoptosis in the Salvador/Warts pathway. Nature Cell Biol. 5, 914–920 (2003). 85. Pantalacci, S., Tapon, N. & Leopold, P. The Salvador partner Hippo promotes apoptosis and cell-cycle exit in Drosophila. Nature Cell Biol. 5, 921–927 (2003). 86. Harvey, K. F., Pfleger, C. M. & Hariharan, I. K. The Drosophila Mst ortholog, hippo, restricts growth and cell proliferation and promotes apoptosis. Cell 114, 457–467 (2003). References 83–86 clearly show the close coordination of apoptosis and proliferation control in vivo through the characterization of a tumour-suppressor protein complex in flies that might be relevant to cancer. 87. Hipfner, D. R., Weigmann, K. & Cohen, S. M. The bantam gene regulates Drosophila growth. Genetics 161, 1527–1537 (2002). 88. Brennecke, J., Hipfner, D. R., Stark, A., Russell, R. B. & Cohen, S. M. bantam encodes a developmentally regulated microRNA that controls cell proliferation and regulates the proapoptotic gene hid in Drosophila. Cell 113, 25–36 (2003). Characterization of the first microRNA that controls growth and apoptosis. 89. Du, W., Xie, J. E. & Dyson, N. Ectopic expression of dE2F and dDP induces cell proliferation and death in the Drosophila eye. EMBO J. 15, 3684–3692 (1996). 90. Prober, D. A. & Edgar, B. A. Interactions between Ras1, dMyc, and dPI3K signaling in the developing Drosophila wing. Genes Dev. 16, 2286–2299 (2002). 91. Sears, R., Leone, G., DeGregori, J. & Nevins, J. R. Ras enhances Myc protein stability. Mol. Cell 3, 169–179 (1999). 92. Bergmann, A., Agapite, J., McCall, K. & Steller, H. The Drosophila gene hid is a direct molecular target of Rasdependent survival signaling. Cell 95, 331–341 (1998). 93. Kurada, P. & White, K. Ras promotes cell survival in Drosophila by downregulating hid expression. Cell 95, 319–329 (1998). 94. Alavi, A., Hood, J. D., Frausto, R., Stupack, D. G. & Cheresh, D. A. Role of Raf in vascular protection from distinct apoptotic stimuli. Science 301, 94–96 (2003). 95. Hindley, A. & Kolch, W. Extracellular signal regulated kinase (ERK)/mitogen activated protein kinase (MAPK)independent functions of Raf kinases. J. Cell Sci. 115, 1575–1581 (2002). 96. Huser, M. et al. MEK kinase activity is not necessary for Raf-1 function. EMBO J. 20, 1940–1951 (2001). 97. Yang, L. & Baker, N. E. Cell cycle withdrawal, progression, and cell survival regulation by EGFR and its effectors in the differentiating Drosophila eye. Dev. Cell 4, 359–369 (2003). 98. Raisin, S., Pantalacci, S., Breittmayer, J. P. & Leopold, P. A new genetic locus controlling growth and proliferation in Drosophila melanogaster. Genetics 164, 1015–1025 (2003). 99. Johnston, L. A. & Gallant, P. Control of growth and organ size in Drosophila. Bioessays 24, 54–64 (2002). 100. Oldham, S. & Hafen, E. Insulin/IGF and target of rapamycin signaling: a TOR de force in growth control. Trends Cell Biol. 13, 79–85 (2003).
101. Datar, S. A., Jacobs, H. W., de la Cruz, A. F., Lehner, C. F. & Edgar, B. A. The Drosophila cyclin D–Cdk4 complex promotes cellular growth. EMBO J. 19, 4543–4554 (2000). 102. Meyer, C. A. et al. Drosophila Cdk4 is required for normal growth and is dispensable for cell cycle progression. EMBO J. 19, 4533–4542 (2000). 103. Frei, C. & Edgar, B. A. Drosophila cyclin D/Cdk4 requires hif-1 prolyl hydroxylase to drive cell growth. Dev. Cell 6, 241–251 (2004). 104. Justice, R. W., Zilian, O., Woods, D. F., Noll, M. & Bryant, P. J. The Drosophila tumor suppressor gene warts encodes a homolog of human myotonic dystrophy kinase and is required for the control of cell shape and proliferation. Genes Dev. 9, 534–546 (1995). 105. Kango-Singh, M. et al. Shar-pei mediates cell proliferation arrest during imaginal disc growth in Drosophila. Development 129, 5719–5730 (2002). 106. Tapon, N. et al. salvador promotes both cell cycle exit and apoptosis in Drosophila and is mutated in human cancer cell lines. Cell 110, 467–478 (2002). 107. Xu, T., Wang, W., Zhang, S., Stewart, R. A. & Yu, W. Identifying tumor suppressors in genetic mosaics: the Drosophila lats gene encodes a putative protein kinase. Development 121, 1053–1063 (1995). 108. Tamaskovic, R., Bichsel, S. J. & Hemmings, B. A. NDR family of AGC kinases — essential regulators of the cell cycle and morphogenesis. FEBS Lett. 546, 73–80 (2003). 109. Dan, I., Watanabe, N. M. & Kusumi, A. The Ste20 group kinases as regulators of MAP kinase cascades. Trends Cell Biol. 11, 220–230 (2001). 110. Stewart, R. A., Li, D. M., Huang, H. & Xu, T. A genetic screen for modifiers of the lats tumor suppressor gene identifies C-terminal Src kinase as a regulator of cell proliferation in Drosophila. Oncogene 22, 6436–6444 (2003). 111. Cheung, W. L. et al. Apoptotic phosphorylation of histone H2B is mediated by mammalian sterile twenty kinase. Cell 113, 507–517 (2003). 112. Conlon, I. & Raff, M. Size control in animal development. Cell 96, 235–244 (1999). 113. Simpson, P. & Morata, G. Differential mitotic rates and patterns of growth in compartments in the Drosophila wing. Dev. Biol. 85, 299–308 (1981). Provides evidence that faster-dividing cells out compete those that divide more slowly in developing tissues. Sets the stage for much of the recent progress in understanding size control. 114. Moreno, E. & Basler, K. dMyc transforms cells into supercompetitors. Cell 117, 117–129 (2004). 115. Burke, R. & Basler, K. Dpp receptors are autonomously required for cell proliferation in the entire developing Drosophila wing. Development 122, 2261–2269 (1996). 116. Martin-Castellanos, C. & Edgar, B. A. A characterization of the effects of Dpp signaling on cell growth and proliferation in the Drosophila wing. Development 129, 1003–1013 (2002). 117. Johnston, L. A. & Sanders, A. L. Wingless promotes cell survival but constrains growth during Drosophila wing development. Nature Cell Biol. 5, 827–833 (2003). 118. Giraldez, A. J. & Cohen, S. M. Wingless and Notch signaling provide cell survival cues and control cell proliferation during wing development. Development 130, 6533–6543 (2003). 119. Nellen, D., Burke, R., Struhl, G. & Basler, K. Direct and longrange action of a DPP morphogen gradient. Cell 85, 357–368 (1996). 120. Lecuit, T. et al. Two distinct mechanisms for long-range patterning by Decapentaplegic in the Drosophila wing. Nature 381, 387–393 (1996). 121. Adachi-Yamada, T., Fujimura-Kamada, K., Nishida, Y. & Matsumoto, K. Distortion of proximodistal information causes JNK-dependent apoptosis in Drosophila wing. Nature 400, 166–169 (1999).
NATURE REVIEWS | MOLECUL AR CELL BIOLOGY
122. Moreno, E., Basler, K. & Morata, G. Cells compete for decapentaplegic survival factor to prevent apoptosis in Drosophila wing development. Nature 416, 755–759 (2002). References 114 and 122 provide evidence that cells might compete for limiting levels of a survival ligand in vivo. 123. Duman-Scheel, M., Weng, L., Xin, S. & Du, W. Hedgehog regulates cell growth and proliferation by inducing Cyclin D and Cyclin E. Nature 417, 299–304 (2002). 124. Leevers, S. J., Weinkove, D., MacDougall, L. K., Hafen, E. & Waterfield, M. D. The Drosophila phosphoinositide 3-kinase Dp110 promotes cell growth. EMBO J. 15, 6584–6594 (1996). 125. Milan, M., Campuzano, S. & Garcia-Bellido, A. Developmental parameters of cell death in the wing disc of Drosophila. Proc. Natl Acad. Sci. USA 94, 5691–5696 (1997). 126. Tao, W. et al. Human homologue of the Drosophila melanogaster lats tumour suppressor modulates CDC2 activity. Nature Genet. 21, 177–181 (1999). 127. St John, M. A. et al. Mice deficient of Lats1 develop softtissue sarcomas, ovarian tumours and pituitary dysfunction. Nature Genet. 21, 182–186 (1999). 128. Hisaoka, M., Tanaka, A. & Hashimoto, H. Molecular alterations of h-warts/LATS1 tumor suppressor in human soft tissue sarcoma. Lab Invest. 82, 1427–1435 (2002). 129. Sibilia, M. et al. The EGF receptor provides an essential survival signal for SOS-dependent skin tumor development. Cell 102, 211–220 (2000). 130. Li, Y., Hively, W. P. & Varmus, H. E. Use of MMTV-Wnt-1 transgenic mice for studying the genetic basis of breast cancer. Oncogene 19, 1002–1009 (2000). 131. Oro, A. E. et al. Basal cell carcinomas in mice overexpressing sonic hedgehog. Science 276, 817–821 (1997). 132. Taipale, J. & Beachy, P. A. The Hedgehog and Wnt signalling pathways in cancer. Nature 411, 349–354 (2001). 133. Thayer, S. P. et al. Hedgehog is an early and late mediator of pancreatic cancer tumorigenesis. Nature 425, 851–856 (2003). 134. Berman, D. M. et al. Widespread requirement for Hedgehog ligand stimulation in growth of digestive tract tumours. Nature 425, 846–851 (2003). 135. Sun, X., Mariani, F. V. & Martin, G. R. Functions of FGF signalling from the apical ectodermal ridge in limb development. Nature 418, 501–508 (2002). 136. Megason, S. G. & McMahon, A. P. A mitogen gradient of dorsal midline Wnts organizes growth in the CNS. Development 129, 2087–2098 (2002).
Acknowledgments The authors would like to thank L. Johnston for kindly sharing results before publication, and J. Brennecke, S. Pizette, S. Szuplewski, D. Neubueser and M. Treier for discussions, comments and suggestions.
Competing interests statement The authors declare no competing financial interests.
Online links DATABASES The following terms in this article are linked online to: Entrez: http://www.ncbi.nlm.nih.gov/Entrez/ ARF | MDM2 | Myc | slik Flybase: http://flybase.bio.indiana.edu/ bantam | brinker | Hid | Hippo |salvador | Slik | warts | Wingless SwissProt: http://www.ca.expasy.ch c-Myc | L-Myc | N-Myc | p53 | PUMA Access to this links box is available online.
VOLUME 5 | OCTOBER 2004 | 8 1 5
©2004 Nature Publishing Group