- Email: [email protected]
Unexpected Twist to the Z Ring
Previews 519
in Mdm2 that are specifically targeted by cyclin G-PP2A? Does cyclin G-PP2A, through dephosphorylation, activate MdmX, which is a member of the Mdm family (Michael and Oren, 2002) and a regulator of the p53-Mdm2 network? What are other cellular proteins that interact with cyclin G and are also a substrate for cyclin G-PP2A? Do cyclin G-cdk5 and cyclin G-GAK serve as a kinase to phosphorylate Mdm2 and activate Mdm2 to inhibit p53? Why do cyclin G null mice develop normally (Kimura et al., 2001)? Do cyclin G-related cyclins, that is, cyclin G2 and cyclin I, have similar activity? Is cyclin G-PP2A phosphatase activity necessary for sensitizing cancer cells to TNF␣-induced apoptosis? Finally, the new findings raise the possibility that cyclin G could be exploited for the development of cancer therapeutic agents. For example, potential strategies could include inhibition of cyclin G expression, blocking the interaction of cyclin G with the B⬘ subunit of PP2A and Mdm2, and inhibition of PP2A phosphatase activity toward Mdm2.
The p53-Mdm2 Network and Its Regulation by Cyclin G-PP2A Phosphatase
Xinbin Chen Department of Cell Biology The University of Alabama at Birmingham Birmingham, Alabama 35294 Selected Reading
of cyclin G, PP2A dephosphorylates Mdm2 less efficiently, and the resulting hyperphosphorylated Mdm2 is less efficient in destabilizing p53 (Okamoto et al., 2002). Addition of cyclin G to cyclin G⫺/⫺ MEF cells restores the activity of PP2A, and subsequently the activity of Mdm2 in degradation of p53 (Okamoto et al., 2002). These data suggest that cyclin G serves as a negative regulator of p53 by activating Mdm2 through dephosphorylation (see Figure). This also provides a mechanistic explanation for the observation that cyclin G expression is associated with growth promotion rather than arrest (Skotzko et al., 1995). These new observations provide insight into the function of cyclin G in the p53 pathway, and also raise other interesting questions that need to be addressed in the future. For example, what are the phosphorylation sites
Unexpected Twist to the Z Ring
Development in Bacillus subtilis involves a switch in the location of the cytokinetic Z ring from midcell to the pole. Time lapse photography of an FtsZ-GFP fusion reveals that this switch involves a spiral intermediate and allows identification of the specific sporulation functions involved. The Z ring is a cytoskeletal structure formed by the selfassembly of FtsZ, the ancestral homolog of eukaryotic tubulins. In most prokaryotes, cell division depends
Kanaoka, Y., Kimura, S.H., Okazaki, I., Ikeda, M., and Nojima, H. (1997). FEBS Lett. 402, 73–80. Kimura, S.H., Ikawa, M., Ito, A., Okabe, M., and Nojima, H. (2001). Oncogene 20, 3290–3300. Ko, L.J., and Prives, C. (1996). Genes Dev. 10, 1054–1072. Michael, D., and Oren, M. (2002). Curr. Opin. Genet. Dev. 12, 53–59. Morita, N., Kiryu, S., and Kiyama, H. (1996). J. Neurosci. 16, 5961– 5966. Okamoto, K., and Beach, D. (1994). EMBO J. 13, 4816–4822. Okamoto, K., and Prives, C. (1999). Oncogene 18, 4606–4615. Okamoto, K., Kamibayashi, C., Serrano, M., Prives, C., Mumby, M.C., and Beach, D. (1996). Mol. Cell. Biol. 16, 6593–6602. Okamoto, K., Li, H., Jensen, M.R., Zhang, T., Taya, Y., Thorgeirsson, S.S., and Prives, C. (2002). Mol. Cell 9, 761–771. Skotzko, M., Wu, L., Anderson, W.F., Gordon, E.M., and Hall, F.L. (1995). Cancer Res. 55, 5493–5498.
upon the Z ring, so the problem of how the cell positions the division site can be reduced to how the Z ring is positioned (Lutkenhaus and Addinall, 1997). The developmental program in Bacillus subtilis that results in sporulation offers a unique window on this problem. During vegetative growth, the Z ring is positioned at midcell and two equal-sized progeny cells are produced. However, during sporulation, an alternative, asymmetric form of cell division is induced by nutrient deprivation, yielding two cells of different sizes, the forespore and the mother cell, with different developmental fates. A few years ago it was demonstrated by immunofluorescence that Z rings form at each cell pole early in sporulation (Levin and Losick, 1996). One of these is used for the asymmetric division, while the other is eventually discarded.
Developmental Cell 520
FtsZ Spiral Intermediate during Sporulation FtsZ spiral intermediates as visualized by time lapse photography of an FtsZ-GFP fusion and immunofluorescence microscopy (adapted from Ben-Yehuda and Losick, 2002). Schematic representation of the specific sporulation functions that induce FtsZ spirals.
The appearance of polar Z rings during sporulation is under developmental control and requires the master sporulation transcriptional factor Spo0A. Spo0H, a transcriptional regulator of late exponential growth, which is also required for sporulation, did not appear to be required. This led to a model in which gene(s) under the control of Spo0A blocked the midcell site and activated sites at the poles. Other work implicated an early sporulation gene, spoIIE, in forming the polar rings, although null mutants only displayed a delay in the switch (Khvorova et al., 1998). SpoIIE, a key regulatory protein of early cell-specific gene expression, binds FtsZ and is anchored to the membrane through N-terminal membrane-spanning domains (Lucet et al., 2000). In a study published in the April issue of Cell, BenYehuda and Losick (2002) used FtsZ-GFP in order to better monitor the switch through time lapse photography. As cells entered sporulation, Z ring formation at midcell was not inhibited as might have been expected from the earlier model. Instead, a midcell ring formed, which was then converted to a spiral that grew toward both poles (see Figure). As the spiral reached the poles it collapsed into polar rings. The same spiral pattern was observed with GFP fusions to proteins that bind to FtsZ, indicating it was not an artifact of the FtsZ-GFP fusion. In addition, this switch was entirely reversible. By putting the nutrient-starved cells into fresh medium the polar rings spiraled back to a midcell ring. Although the appearance of the FtsZ spiral was a bit of a surprise, FtsZ spirals have been observed previously. In E. coli, certain ftsZ mutants or overproduction of FtsZ results in spirals (see http://www.bio.unc.edu/ faculty/salmon/lab/mafia//mafiamovies.html for a movie showing FtsZ-GFP spiraling from an invaginating septum toward the future division sites in the progeny cells). This raises the possibility that a Z ring consists of a tight spiral that is not resolved by microscopy. Obviously, the capability of FtsZ forming spirals exists. The novelty here is that during sporulation, it is harnessed to switch the location of the Z ring within the cell. What sporulation genes are responsible for inducing the midcell ring to spiral? Reexamination of earlier results confirmed the spo0A requirement but also demonstrated that spo0H was required. What about the downstream effector genes? spoIIE proved to be the target for spo0A while ftsZ itself is the target of spo0H (see Figure). ftsZ is expressed from several promoters, one
of which utilizes spo0H and is required to boost ftsZ expression during sporulation. Removing either this ftsZ promoter or the spoIIE gene only had a minor effect on the switch; however, eliminating both dramatically dampened the switch. Thus, elevating the level of FtsZ in the presence of a sporulation-specific protein capable of anchoring FtsZ to the membrane results in the switch. No other sporulation protein appears to be required, since expressing SpoIIE in vegetative cells with elevated FtsZ led to spirals and polar Z rings. Thus, the mediators of the switch are defined even though the mechanism of spiral formation is not known. During vegetative growth, MinCD inhibits Z ring formation at the cell poles, to ensure that two equal daughter cells result from the cell division process. Thus, one might expect that in order to establish polar Z rings, a cell would have to overcome the inhibitory activity of the MinCD system (de Boer et al., 1989). One possibility is that MinCD is inactivated by a specific sporulation function; however, the ability to induce polar rings in vegetative cells suggests that this may not be necessary. In E. coli, it is known that overexpressing FtsZ or expressing an FtsZ mutant that has reduced GTPase activity, and therefore produces more stable polymers, can overcome MinCD and lead to polar Z rings (Pichoff and Lutkenhaus, 2001). A recent study in B. subtilis observed much the same (Levin et al., 2001). Manipulation of various factors that are expected to increase FtsZ polymer stability led to polar Z rings. Whether spirals are involved is not known; they may have been missed. The presence of the spiral intermediate emphasizes that we know little about the structure of the Z ring and how is it maintained. FtsZ readily assembles into protofilaments, but the importance of lateral interactions that may lead to a higher order structure or proteins that may bundle filaments are still largely an enigma. In E. coli, the presence of either of two completely unrelated proteins, FtsA or ZipA, that have in common the ability to bind to the C terminus of FtsZ, is sufficient to support formation of the Z ring (Pichoff and Lutkenhaus, 2002). The mechanism underlying this is unknown. A recent analysis of the Z ring in E. coli using FRAP and FtsZ-GFP showed that the Z ring is as dynamic as its big brother the microtubule, turning over every 30 s (Stricker et al., 2002). Are FtsZ spirals this dynamic? How is such a dynamic structure localized long enough to carry out its function? Perhaps new insights will
Previews 521
emerge as the regulation of this cytoskeletal structure is unraveled.
Khvorova, A., Zhang, L., Higgins, M.L., and Piggot, P.J. (1998). J. Bacteriol. 180, 1256–1260. Levin, P.A., and Losick, R. (1996). Genes Dev. 10, 478–488.
Joe Lutkenhaus Department of Microbiology Molecular Genetics and Immunology University of Kansas Medical Center Kansas City, Kansas 66160
Levin, P.A., Schwartz, R.L., and Grossman, A.D. (2001). J. Bacteriol. 183, 5449–5452.
Selected Reading
Pichoff, S., and Lutkenhaus, J. (2001). J. Bacteriol. 183, 6630–6635.
Ben-Yehuda, S., and Losick, R. (2002). Cell 109, 257–266.
Pichoff, S., and Lutkenhaus, J. (2002). EMBO J. 21, 685–693.
de Boer, P.A., Crossley, R., and Rothfield, L.I. (1989). Cell 56, 641–649.
Stricker, J., Maddox, P., Salmon, E.D., and Erickson, H.P. (2002). Proc. Natl. Acad. Sci. USA 99, 3171–3175.
RhoGAP: The Next Big Thing for Small Mice?
and Lawrence, 2000). Moreover, there are so many examples in which growth and proliferation are physiologically uncoupled that one is forced to at least consider the possibility that these two processes might generally be regulated independently. For example, reductive cleavages in most early embryos represent cell division without growth, whereas the opposite occurs in a hypertrophic muscle cell or a megakaryocyte, the giant bone marrow cell that sheds platelets. Interestingly, the latter cell represents an example of the rule that DNA content directly correlates with cell size, as it is the repeated rounds of DNA synthesis unaccompanied by cell division that accounts for the bulk of the megakaryocyte. Another principle guiding the determination of organ size is that alterations in proliferation are compensated by inverse changes in cell size, such that the volume of the compartment remains unchanged. Thus, we are presented with an intriguing paradox: while the major factor responsible for differences in size among organisms is the number rather the size of the constituent cells, the primary determinant of organ size within an organism is the rate of growth and not proliferation. Over the past several years, a number of lines of investigation have converged on the evolutionarily conserved insulin/IGF-1 signaling pathway as a critical regulator of cell growth (Day and Lawrence, 2000). Some of the most informative data derive from genetic experiments in the fruit fly, Drosophila melanogaster, though confirmatory information has begun to emerge in mammals. The critical signaling components include, in addition to the insulin receptor and its substrate chico (the fruit fly homolog of mammalian insulin receptor substrate 1), phosphoinositide 3⬘-kinase, phosphoinositide-dependent protein kinase 1, Akt/protein kinase B, and S6 protein kinase and PHAS1/4EBP1, though the latter two translational regulators might well be more responsive to nutritional cues than insulin. Suggesting a surprisingly linear cascade, overexpression of each member of this pathway yields a virtually equivalent phenotype, an increase in compartment size mediated by a disproportionate augmentation in cell growth. The lipid phosphatase PTEN functions as a negative regulator of the pathway. In the present report, Sordella et al. have generated
The size of an organism is determined by the number and size of its constituent cells. The insulin/IGF-1 signaling systems have been long recognized to play a critical role in the determination of body size. Now the generation of mice deficient for a RhoGAP suggests that this small G protein might also regulate the growth of animals. In a time marked by impressive advances in the elucidation of both principles and details governing development, one of the most poorly understood processes is the determination of the mature size of organs and organisms. While we all accept the consistency by which human arms are bilaterally roughly equal in size and smaller than legs and that humans are invariably larger than mice, the molecular mechanisms that ensure the fidelity of these phenomena are only beginning to be understood. Thus, the report by Sordella et al. in this issue of Developmental Cell, pointing to a surprising pathway that regulates cell size, represents a potentially important contribution to the solution of this problem (Sordella et al., 2002). Three fundamental processes principally regulate biological mass: cell proliferation, death, and growth (used here to indicate the determination of cell size irrespective of rates of division). Whereas the first two have received substantial attention in recent decades, only in the last several years have scientists begun to address the problem of regulation of cell growth. Perhaps the relatively constant size of most cells has caused many of us to forget that proliferation must be accompanied by a commensurate, perfectly matched increase in growth, or over time cell size will drift. The simplistic but rather tempting view that growth always follows passively as a direct result of proliferation has never fit the data particularly well, but only recently has it been refuted using modern cellular and genetic strategies (Day
Lucet, I., Feucht, A., Yudkin, M.D., and Errington, J. (2000). EMBO J. 19, 1467–1475. Lutkenhaus, J., and Addinall, S.G. (1997). Annu. Rev. Biochem. 66, 93–116.
in Mdm2 that are specifically targeted by cyclin G-PP2A? Does cyclin G-PP2A, through dephosphorylation, activate MdmX, which is a member of the Mdm family (Michael and Oren, 2002) and a regulator of the p53-Mdm2 network? What are other cellular proteins that interact with cyclin G and are also a substrate for cyclin G-PP2A? Do cyclin G-cdk5 and cyclin G-GAK serve as a kinase to phosphorylate Mdm2 and activate Mdm2 to inhibit p53? Why do cyclin G null mice develop normally (Kimura et al., 2001)? Do cyclin G-related cyclins, that is, cyclin G2 and cyclin I, have similar activity? Is cyclin G-PP2A phosphatase activity necessary for sensitizing cancer cells to TNF␣-induced apoptosis? Finally, the new findings raise the possibility that cyclin G could be exploited for the development of cancer therapeutic agents. For example, potential strategies could include inhibition of cyclin G expression, blocking the interaction of cyclin G with the B⬘ subunit of PP2A and Mdm2, and inhibition of PP2A phosphatase activity toward Mdm2.
The p53-Mdm2 Network and Its Regulation by Cyclin G-PP2A Phosphatase
Xinbin Chen Department of Cell Biology The University of Alabama at Birmingham Birmingham, Alabama 35294 Selected Reading
of cyclin G, PP2A dephosphorylates Mdm2 less efficiently, and the resulting hyperphosphorylated Mdm2 is less efficient in destabilizing p53 (Okamoto et al., 2002). Addition of cyclin G to cyclin G⫺/⫺ MEF cells restores the activity of PP2A, and subsequently the activity of Mdm2 in degradation of p53 (Okamoto et al., 2002). These data suggest that cyclin G serves as a negative regulator of p53 by activating Mdm2 through dephosphorylation (see Figure). This also provides a mechanistic explanation for the observation that cyclin G expression is associated with growth promotion rather than arrest (Skotzko et al., 1995). These new observations provide insight into the function of cyclin G in the p53 pathway, and also raise other interesting questions that need to be addressed in the future. For example, what are the phosphorylation sites
Unexpected Twist to the Z Ring
Development in Bacillus subtilis involves a switch in the location of the cytokinetic Z ring from midcell to the pole. Time lapse photography of an FtsZ-GFP fusion reveals that this switch involves a spiral intermediate and allows identification of the specific sporulation functions involved. The Z ring is a cytoskeletal structure formed by the selfassembly of FtsZ, the ancestral homolog of eukaryotic tubulins. In most prokaryotes, cell division depends
Kanaoka, Y., Kimura, S.H., Okazaki, I., Ikeda, M., and Nojima, H. (1997). FEBS Lett. 402, 73–80. Kimura, S.H., Ikawa, M., Ito, A., Okabe, M., and Nojima, H. (2001). Oncogene 20, 3290–3300. Ko, L.J., and Prives, C. (1996). Genes Dev. 10, 1054–1072. Michael, D., and Oren, M. (2002). Curr. Opin. Genet. Dev. 12, 53–59. Morita, N., Kiryu, S., and Kiyama, H. (1996). J. Neurosci. 16, 5961– 5966. Okamoto, K., and Beach, D. (1994). EMBO J. 13, 4816–4822. Okamoto, K., and Prives, C. (1999). Oncogene 18, 4606–4615. Okamoto, K., Kamibayashi, C., Serrano, M., Prives, C., Mumby, M.C., and Beach, D. (1996). Mol. Cell. Biol. 16, 6593–6602. Okamoto, K., Li, H., Jensen, M.R., Zhang, T., Taya, Y., Thorgeirsson, S.S., and Prives, C. (2002). Mol. Cell 9, 761–771. Skotzko, M., Wu, L., Anderson, W.F., Gordon, E.M., and Hall, F.L. (1995). Cancer Res. 55, 5493–5498.
upon the Z ring, so the problem of how the cell positions the division site can be reduced to how the Z ring is positioned (Lutkenhaus and Addinall, 1997). The developmental program in Bacillus subtilis that results in sporulation offers a unique window on this problem. During vegetative growth, the Z ring is positioned at midcell and two equal-sized progeny cells are produced. However, during sporulation, an alternative, asymmetric form of cell division is induced by nutrient deprivation, yielding two cells of different sizes, the forespore and the mother cell, with different developmental fates. A few years ago it was demonstrated by immunofluorescence that Z rings form at each cell pole early in sporulation (Levin and Losick, 1996). One of these is used for the asymmetric division, while the other is eventually discarded.
Developmental Cell 520
FtsZ Spiral Intermediate during Sporulation FtsZ spiral intermediates as visualized by time lapse photography of an FtsZ-GFP fusion and immunofluorescence microscopy (adapted from Ben-Yehuda and Losick, 2002). Schematic representation of the specific sporulation functions that induce FtsZ spirals.
The appearance of polar Z rings during sporulation is under developmental control and requires the master sporulation transcriptional factor Spo0A. Spo0H, a transcriptional regulator of late exponential growth, which is also required for sporulation, did not appear to be required. This led to a model in which gene(s) under the control of Spo0A blocked the midcell site and activated sites at the poles. Other work implicated an early sporulation gene, spoIIE, in forming the polar rings, although null mutants only displayed a delay in the switch (Khvorova et al., 1998). SpoIIE, a key regulatory protein of early cell-specific gene expression, binds FtsZ and is anchored to the membrane through N-terminal membrane-spanning domains (Lucet et al., 2000). In a study published in the April issue of Cell, BenYehuda and Losick (2002) used FtsZ-GFP in order to better monitor the switch through time lapse photography. As cells entered sporulation, Z ring formation at midcell was not inhibited as might have been expected from the earlier model. Instead, a midcell ring formed, which was then converted to a spiral that grew toward both poles (see Figure). As the spiral reached the poles it collapsed into polar rings. The same spiral pattern was observed with GFP fusions to proteins that bind to FtsZ, indicating it was not an artifact of the FtsZ-GFP fusion. In addition, this switch was entirely reversible. By putting the nutrient-starved cells into fresh medium the polar rings spiraled back to a midcell ring. Although the appearance of the FtsZ spiral was a bit of a surprise, FtsZ spirals have been observed previously. In E. coli, certain ftsZ mutants or overproduction of FtsZ results in spirals (see http://www.bio.unc.edu/ faculty/salmon/lab/mafia//mafiamovies.html for a movie showing FtsZ-GFP spiraling from an invaginating septum toward the future division sites in the progeny cells). This raises the possibility that a Z ring consists of a tight spiral that is not resolved by microscopy. Obviously, the capability of FtsZ forming spirals exists. The novelty here is that during sporulation, it is harnessed to switch the location of the Z ring within the cell. What sporulation genes are responsible for inducing the midcell ring to spiral? Reexamination of earlier results confirmed the spo0A requirement but also demonstrated that spo0H was required. What about the downstream effector genes? spoIIE proved to be the target for spo0A while ftsZ itself is the target of spo0H (see Figure). ftsZ is expressed from several promoters, one
of which utilizes spo0H and is required to boost ftsZ expression during sporulation. Removing either this ftsZ promoter or the spoIIE gene only had a minor effect on the switch; however, eliminating both dramatically dampened the switch. Thus, elevating the level of FtsZ in the presence of a sporulation-specific protein capable of anchoring FtsZ to the membrane results in the switch. No other sporulation protein appears to be required, since expressing SpoIIE in vegetative cells with elevated FtsZ led to spirals and polar Z rings. Thus, the mediators of the switch are defined even though the mechanism of spiral formation is not known. During vegetative growth, MinCD inhibits Z ring formation at the cell poles, to ensure that two equal daughter cells result from the cell division process. Thus, one might expect that in order to establish polar Z rings, a cell would have to overcome the inhibitory activity of the MinCD system (de Boer et al., 1989). One possibility is that MinCD is inactivated by a specific sporulation function; however, the ability to induce polar rings in vegetative cells suggests that this may not be necessary. In E. coli, it is known that overexpressing FtsZ or expressing an FtsZ mutant that has reduced GTPase activity, and therefore produces more stable polymers, can overcome MinCD and lead to polar Z rings (Pichoff and Lutkenhaus, 2001). A recent study in B. subtilis observed much the same (Levin et al., 2001). Manipulation of various factors that are expected to increase FtsZ polymer stability led to polar Z rings. Whether spirals are involved is not known; they may have been missed. The presence of the spiral intermediate emphasizes that we know little about the structure of the Z ring and how is it maintained. FtsZ readily assembles into protofilaments, but the importance of lateral interactions that may lead to a higher order structure or proteins that may bundle filaments are still largely an enigma. In E. coli, the presence of either of two completely unrelated proteins, FtsA or ZipA, that have in common the ability to bind to the C terminus of FtsZ, is sufficient to support formation of the Z ring (Pichoff and Lutkenhaus, 2002). The mechanism underlying this is unknown. A recent analysis of the Z ring in E. coli using FRAP and FtsZ-GFP showed that the Z ring is as dynamic as its big brother the microtubule, turning over every 30 s (Stricker et al., 2002). Are FtsZ spirals this dynamic? How is such a dynamic structure localized long enough to carry out its function? Perhaps new insights will
Previews 521
emerge as the regulation of this cytoskeletal structure is unraveled.
Khvorova, A., Zhang, L., Higgins, M.L., and Piggot, P.J. (1998). J. Bacteriol. 180, 1256–1260. Levin, P.A., and Losick, R. (1996). Genes Dev. 10, 478–488.
Joe Lutkenhaus Department of Microbiology Molecular Genetics and Immunology University of Kansas Medical Center Kansas City, Kansas 66160
Levin, P.A., Schwartz, R.L., and Grossman, A.D. (2001). J. Bacteriol. 183, 5449–5452.
Selected Reading
Pichoff, S., and Lutkenhaus, J. (2001). J. Bacteriol. 183, 6630–6635.
Ben-Yehuda, S., and Losick, R. (2002). Cell 109, 257–266.
Pichoff, S., and Lutkenhaus, J. (2002). EMBO J. 21, 685–693.
de Boer, P.A., Crossley, R., and Rothfield, L.I. (1989). Cell 56, 641–649.
Stricker, J., Maddox, P., Salmon, E.D., and Erickson, H.P. (2002). Proc. Natl. Acad. Sci. USA 99, 3171–3175.
RhoGAP: The Next Big Thing for Small Mice?
and Lawrence, 2000). Moreover, there are so many examples in which growth and proliferation are physiologically uncoupled that one is forced to at least consider the possibility that these two processes might generally be regulated independently. For example, reductive cleavages in most early embryos represent cell division without growth, whereas the opposite occurs in a hypertrophic muscle cell or a megakaryocyte, the giant bone marrow cell that sheds platelets. Interestingly, the latter cell represents an example of the rule that DNA content directly correlates with cell size, as it is the repeated rounds of DNA synthesis unaccompanied by cell division that accounts for the bulk of the megakaryocyte. Another principle guiding the determination of organ size is that alterations in proliferation are compensated by inverse changes in cell size, such that the volume of the compartment remains unchanged. Thus, we are presented with an intriguing paradox: while the major factor responsible for differences in size among organisms is the number rather the size of the constituent cells, the primary determinant of organ size within an organism is the rate of growth and not proliferation. Over the past several years, a number of lines of investigation have converged on the evolutionarily conserved insulin/IGF-1 signaling pathway as a critical regulator of cell growth (Day and Lawrence, 2000). Some of the most informative data derive from genetic experiments in the fruit fly, Drosophila melanogaster, though confirmatory information has begun to emerge in mammals. The critical signaling components include, in addition to the insulin receptor and its substrate chico (the fruit fly homolog of mammalian insulin receptor substrate 1), phosphoinositide 3⬘-kinase, phosphoinositide-dependent protein kinase 1, Akt/protein kinase B, and S6 protein kinase and PHAS1/4EBP1, though the latter two translational regulators might well be more responsive to nutritional cues than insulin. Suggesting a surprisingly linear cascade, overexpression of each member of this pathway yields a virtually equivalent phenotype, an increase in compartment size mediated by a disproportionate augmentation in cell growth. The lipid phosphatase PTEN functions as a negative regulator of the pathway. In the present report, Sordella et al. have generated
The size of an organism is determined by the number and size of its constituent cells. The insulin/IGF-1 signaling systems have been long recognized to play a critical role in the determination of body size. Now the generation of mice deficient for a RhoGAP suggests that this small G protein might also regulate the growth of animals. In a time marked by impressive advances in the elucidation of both principles and details governing development, one of the most poorly understood processes is the determination of the mature size of organs and organisms. While we all accept the consistency by which human arms are bilaterally roughly equal in size and smaller than legs and that humans are invariably larger than mice, the molecular mechanisms that ensure the fidelity of these phenomena are only beginning to be understood. Thus, the report by Sordella et al. in this issue of Developmental Cell, pointing to a surprising pathway that regulates cell size, represents a potentially important contribution to the solution of this problem (Sordella et al., 2002). Three fundamental processes principally regulate biological mass: cell proliferation, death, and growth (used here to indicate the determination of cell size irrespective of rates of division). Whereas the first two have received substantial attention in recent decades, only in the last several years have scientists begun to address the problem of regulation of cell growth. Perhaps the relatively constant size of most cells has caused many of us to forget that proliferation must be accompanied by a commensurate, perfectly matched increase in growth, or over time cell size will drift. The simplistic but rather tempting view that growth always follows passively as a direct result of proliferation has never fit the data particularly well, but only recently has it been refuted using modern cellular and genetic strategies (Day
Lucet, I., Feucht, A., Yudkin, M.D., and Errington, J. (2000). EMBO J. 19, 1467–1475. Lutkenhaus, J., and Addinall, S.G. (1997). Annu. Rev. Biochem. 66, 93–116.