Cath Brooksbank

Phosphorylation: The key to staying faithful

NATURE REVIEWS | MOLECULAR CELL BIOLOGY VOLUME 2 | MARCH 2001 | 167 The cascades in which mitogen-activated protein kinases (MAPKs) play a part can have radically different outcomes, making promiscuity among MAPKs dangerous to the cell.Yet each pathway comprises closely related components. How do MAPKs remain faithful to their pathways? In the 1 February issue of EMBO Journal, Takuji Tanoue and colleagues explain one mechanism. Remarkably, this relies on the identity of just two amino acids. We know of three types of MAPKs, each with their own cascade. The extracellular-signal regulated kinases (ERKs) are activated by growth factors, generally leading to proliferation, whereas p38s and Jun-Nterminal kinases (JNKs) are activated by stress signals, usually causing cellcycle arrest or apoptosis. Each MAPK has to interact with the kinases that activate it (MAPKKs), the phosphatases that inactivate it (MKPs) and its substrates, which are also kinases (MAPKAPKs). A single acidic site outside the active site — the common docking or CD site — interacts with all these molecules, but it doesn’t explain the different binding specificities of the MAPKs. Could there be another site that regulates specificity? Mutation of the CD in p38 reduced, but didn’t completely prevent, binding of the p38-specific MAPKAPK 3pk, implying that another docking site exists. By searching for charged residues and systematically mutating them, the authors identified a pair of residues on p38, Glu 160 and Asp 161, that account for this residual binding. Mutation of the CD site and this second site, which they dubbed the ED site, markedly reduced the ability of p38 to phosphorylate 3pk. The corresponding residues in ERK2 are two threonine residues. Mutation of these to Glu and Asp enabled ERK2 to bind 3pk, and mutation of ERK2’s CD site to make it identical to p38’s improved the interaction further. Extending these studies to other MAPKAPKs and MKPs revealed that, although the CD is necessary for binding, the nature of the ED regulates specificity. Together, the two sites form a groove with two pins in it. Only if it can interact with both pins can a MAPKinteracting protein do its business. Cath Brooksbank References and links ORIGINAL RESEARCH PAPER Tanoue, T. et al. Identification of a docking groove on ERK and p38 MAP kinases that regulates the specificity of docking interactions. EMBO J. 20, 466–479 (2001) FURTHER READING Tanoue, T. et al. A conserved docking motif in MAP kinases common to substrates, activators and regulators. Nature Cell Biol. 2, 110–116 (2000) WEB SITE Mammalian MAPK signalling pathways IN BRIEF

Endocytosis. Tent pegs for clathrin.

Every construction — no matter how temporary — needs sound foundations. So how are the clathrin coats that surround endocytic vesicles tethered to the plasma membrane? Two papers in the 9 February issue of Science reveal the importance of a phospholipid, phosphatidylinositol-4,5-bisphosphate (PtdIns(4,5)P2), in this process.

This way up and handle with care: Proteomics

cells that aren’t cycling. Similar results were obtained in the salivary gland, an organ that undegoes endoreduplication (cell division without cytokinesis). Datar and colleagues turned back to the wing to study whether CycD–Cdk4 exerts its effects through Rbf. As expected, overexpression of Rbf alone slowed cell division, whereas cells expressing all three proteins had near normal cell divison rates but were larger, indicating that cell growth was promoted even while the cell cycle was being slowed by Rbf. In the eye, by contrast, Rbf overexpression didn’t influence postmitotic growth, and insertion of a null Rbf allele had no effect on cell growth in either the wing or the eye. CycD–Cdk4 must, therefore, be promoting cell growth by phosphorylating targets other than Rbf. So, rather than being dedicated to getting cells through G1, CycD–Cdk4 promotes hyperplasia (increased numbers of cell divisions) in dividing cells, hypertrophy (increased cell size) in endoreduplicating cells and both in postmitotic cells — and it doesn’t need Rbf to carry out any of these functions. The solution to the next challenge — determining the growthpromoting target of CycD–Cdk4 — might well have all our eyes bulging. Cath Brooksbank References and links ORIGINAL RESEARCH PAPERS 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) | Datar, S. A. et al. The Drosophila cyclin D–Cdk4 complex promotes cellular growth. EMBO J. 19, 4543–4554 (2000) REVIEW Sherr, C. J. & Roberts, J. M. CDK inhibitors: positive and negative regulators of G1phase progression. Genes Dev. 13, 1501–1512 (1999) FURTHER READING Cockcroft, C. E. et al. Cyclin D control of growth rate in plants. Nature 405, 575–579 (2000) | Rane, S. G. et al. Loss of Cdk4 expression causes insulin-deficient diabetes and Cdk4 activation results in β-islet cell hyperplasia. Nature Genet. 22, 44–52 (1999) | Tsutsui, T. et al. Targeted disruption of CDK4 delays cell cycle entry with enhanced p27Kip1 activity. Mol. Cell Biol. 19, 7011–7019 (1999) FURTHER INFORMATION Cyclins and E2F: a Kohn interaction map | The interactive fly: cell cycle genes | Mitosis world

Pocket the difference.

There’s a molecular identity parade at the plasma membrane every time cells activate phosphatidylinositol 3-OH kinase (PI(3)K). This enzyme catalyses the production of the phospholipid messengers phosphatidylinositol-3,4,5-tr isphosphate (PtdIns(3,4,5)P3) and PtdIns(3,4)P2 which, in turn, recruit proteins containing pleckstrin homology (PH) domains to the membrane. These proteins choreograph a wide variety of molecular dances, from migration and membrane trafficking to adhesion. But how do they pick out tiny amounts of PtdIns(3,4,5)P3 or PtdIns(3,4)P2 from among much higher concentrations of PtdIns(4,5)P2 in the cell membrane? Some PH domains recognize only PtdIns(3,4,5)P3 but others are less discriminating and recognize both PtdIns(3,4,5)P3 and PtdIns(3,4)P2. This makes a big difference to the length of the dance because activation of PI(3)K leads to a transient blip in the amount of PtdIns(3,4,5)P3 but a much more prolonged rise in the amount of PtdIns(3,4)P2. Two papers in the August issue of Molecular Cell provide an answer, by comparing the structure of a PH domain that picks out PtdIns(3,4,5)P3 every time with one that can’t tell the difference between PtdIns(3,4,5)P3 and PtdIns(3,4)P2. Kathryn Ferguson and colleagues at the University of Pennsylvania, and Susan Lietzke and co-workers up the coast at the University of Massachusetts, both chose the PH domain from the PtdIns(3,4,5)P3-specific protein Grp1 (also known as ARNO3 or cytohesin 3) as the subject of their crystallographic analysis. Ferguson et al. also solved the structure of DAPP1 (PHISH), which binds PtdIns(3,4,5)P3 and PtdIns(3,4)P2 with similar affinity. Lietzke and colleagues compared features of the Grp1 PH domain with the promiscuous PH domain from protein kinase B (AKT). In both studies, inositol-1,3,4,5-tetrakisphosphate (Ins(1,3,4,5)P4) — the water-soluble headgroup of PtdIns(3,4,5)P3 — was used to crystallize the ligand-bound PH domains. It was no surprise that the PH domains of both Grp1 and DAPP1 are β-sandwiches, just like previously solved PH domains, but Grp1 has an extra two β-strands. These form a snug pocket accommodating the 5-phosphate (see picture), which simply isn’t there in the non-discriminating PH domains of DAPP1 and protein kinase B. This difference explains how Grp1 binds PtdIns(3,4,5)P3, but how does it stop PtdIns(3,4)P2 from binding? The answer lies in the number of hydrogen bonds necessary to stabilize the interaction. Ferguson and colleagues find that both Grp1 and DAPP1 make a total of 14 side-chain hydrogen bonds with Ins(1,3,4,5)P4, but in Grp1 these are equally distributed among the 3-, 4and 5-phosphates, whereas in DAPP1 there are no hydrogen bonds to the 5-phosphate. So, when PtdIns(3,4)P2 enters Grp1’s PH domain, there’s nothing for this extra pocket to grab hold of, and Grp1 can’t hang on tightly enough to just the 3and 4-phosphates. As well as explaining why some PH domains are fussier than others, these studies also have predictive value. With their new-found understanding of which residues are important for gripping each phosphate group, Ferguson and colleagues used sequence comparisons to predict the specificity of a previously uncharacterized PH domain, then tested their prediction (which turned out to be correct) using a simple gel filtration assay. Armed with this knowledge, it should now be possible to work out the length and strength of responses that different PH-domain proteins mediate. As Lietzke and colleagues point out, it also paves the way towards blocking PI(3)K responses therapeutically, by designing drugs that bind specific PH domains. Cath Brooksbank References and links ORIGINAL RESEARCH PAPERS Ferguson, K. M. et al. Structural basis for discrimination of 3-phosphoinositides by pleckstrin homology domains. Mol. Cell 6, 373–384 (2000) | Lietzke, S. E. et al. Structural basis of 3-phosphoinositide recognition by pleckstrin homology domains. Mol. Cell 6, 385–394 (2000) REVIEWS Lemmon, M. A. & Ferguson, K. M. Signal-dependent membrane targeting by pleckstrin homology (PH) domains. Biochem. J. 350, 1–18 (2000) | Chan, T. O. et al. AKT/PKB and other D3 phosphoinositide-regulated kinases: kinase activation by phosphoinositide-dependent phosphorylation. Annu. Rev. Biochem. 68, 965–1014 (1999) Pocket the difference S T R U C T U R A L B I O LO G Y Finally Wang and colleagues looked for other functional parallels with the Drosophila homologues. It turns out that, in both cases, the amino-terminal region is essential for function. The authors narrowed this down to just seven amino acids, which, on their own, are enough to activate procaspase 3 (albeit much less efficiently than the full-length protein). Given that the only region of sequence conservation in Grim, Reaper and Hid is the amino-terminal 14 amino acids, this work, say the authors, defines “an evolutionarily conserved structural and biochemical basis for the activation of apoptosis by Smac/DIABLO”. Alison Mitchell References and links ORIGINAL RESEARCH PAPERS Chai, J. et al. Structural and biochemical basis of apoptotic activation by Smac/DIABLO. Nature 406, 855–862 (2000) | Du, C., Fang, M., Li, Y., Li, L. & Wang, X. Smac, a mitochondrial protein that promotes cyctochrome c-dependent caspase activation by eliminating IAP inhibition. Cell 102, 33–42 (2000) | Verhagen, A. M. et al. Identification of DIABLO, a mammalian protein that promotes apoptosis by binding and antagonizing IAP proteins. Cell 102, 43–53 (2000) FURTHER READING Green, D. R. Apoptotic pathways: Paper wraps stone blunts scissors. Cell 102, 1–4 (2000) Image kindly provided by Kathryn Ferguson and Mark Lemmon, University of Pennsylvania, Pennsylvania, USA.

Protein-protein interactions: Phosphothreonine lego.

Does parkin have other substrates, and is one of CDCrel1’s normal functions to block dopamine release? If it is, blockage of CDCrel-1’s breakdown in ARJP would provide a satisfactory explanation and a promising therapeutic target for this form of Parkinson’s. Cath Brooksbank References and links ORIGINAL RESEARCH PAPER Zhang, Y. et al. Parkin functions as an E2-dependent ubiquitin protein ligase and promotes the degradation of the synaptic vesicle-associated protein, CDCrel-1. Proc. Natl Acad. Sci. USA 97, 13354–13359 (2000) FURTHER READING Shimura, H. et al. Familial Parkinson disease gene product, parkin, is a ubiquitin-protein ligase. Nature Genet. 25, 302–305 (2000) H I G H L I G H T S

Proteomics: This way up and handle with care

cells that aren’t cycling. Similar results were obtained in the salivary gland, an organ that undegoes endoreduplication (cell division without cytokinesis). Datar and colleagues turned back to the wing to study whether CycD–Cdk4 exerts its effects through Rbf. As expected, overexpression of Rbf alone slowed cell division, whereas cells expressing all three proteins had near normal cell divison rates but were larger, indicating that cell growth was promoted even while the cell cycle was being slowed by Rbf. In the eye, by contrast, Rbf overexpression didn’t influence postmitotic growth, and insertion of a null Rbf allele had no effect on cell growth in either the wing or the eye. CycD–Cdk4 must, therefore, be promoting cell growth by phosphorylating targets other than Rbf. So, rather than being dedicated to getting cells through G1, CycD–Cdk4 promotes hyperplasia (increased numbers of cell divisions) in dividing cells, hypertrophy (increased cell size) in endoreduplicating cells and both in postmitotic cells — and it doesn’t need Rbf to carry out any of these functions. The solution to the next challenge — determining the growthpromoting target of CycD–Cdk4 — might well have all our eyes bulging. Cath Brooksbank References and links ORIGINAL RESEARCH PAPERS 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) | Datar, S. A. et al. The Drosophila cyclin D–Cdk4 complex promotes cellular growth. EMBO J. 19, 4543–4554 (2000) REVIEW Sherr, C. J. & Roberts, J. M. CDK inhibitors: positive and negative regulators of G1phase progression. Genes Dev. 13, 1501–1512 (1999) FURTHER READING Cockcroft, C. E. et al. Cyclin D control of growth rate in plants. Nature 405, 575–579 (2000) | Rane, S. G. et al. Loss of Cdk4 expression causes insulin-deficient diabetes and Cdk4 activation results in β-islet cell hyperplasia. Nature Genet. 22, 44–52 (1999) | Tsutsui, T. et al. Targeted disruption of CDK4 delays cell cycle entry with enhanced p27Kip1 activity. Mol. Cell Biol. 19, 7011–7019 (1999) FURTHER INFORMATION Cyclins and E2F: a Kohn interaction map | The interactive fly: cell cycle genes | Mitosis world

Tent pegs for clathrin

Every construction — no matter how temporary — needs sound foundations. So how are the clathrin coats that surround endocytic vesicles tethered to the plasma membrane? Two papers in the 9 February issue of Science reveal the importance of a phospholipid, phosphatidylinositol-4,5-bisphosphate (PtdIns(4,5)P2), in this process.

Networking with proteins

cells that aren’t cycling. Similar results were obtained in the salivary gland, an organ that undegoes endoreduplication (cell division without cytokinesis). Datar and colleagues turned back to the wing to study whether CycD–Cdk4 exerts its effects through Rbf. As expected, overexpression of Rbf alone slowed cell division, whereas cells expressing all three proteins had near normal cell divison rates but were larger, indicating that cell growth was promoted even while the cell cycle was being slowed by Rbf. In the eye, by contrast, Rbf overexpression didn’t influence postmitotic growth, and insertion of a null Rbf allele had no effect on cell growth in either the wing or the eye. CycD–Cdk4 must, therefore, be promoting cell growth by phosphorylating targets other than Rbf. So, rather than being dedicated to getting cells through G1, CycD–Cdk4 promotes hyperplasia (increased numbers of cell divisions) in dividing cells, hypertrophy (increased cell size) in endoreduplicating cells and both in postmitotic cells — and it doesn’t need Rbf to carry out any of these functions. The solution to the next challenge — determining the growthpromoting target of CycD–Cdk4 — might well have all our eyes bulging. Cath Brooksbank References and links ORIGINAL RESEARCH PAPERS 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) | Datar, S. A. et al. The Drosophila cyclin D–Cdk4 complex promotes cellular growth. EMBO J. 19, 4543–4554 (2000) REVIEW Sherr, C. J. & Roberts, J. M. CDK inhibitors: positive and negative regulators of G1phase progression. Genes Dev. 13, 1501–1512 (1999) FURTHER READING Cockcroft, C. E. et al. Cyclin D control of growth rate in plants. Nature 405, 575–579 (2000) | Rane, S. G. et al. Loss of Cdk4 expression causes insulin-deficient diabetes and Cdk4 activation results in β-islet cell hyperplasia. Nature Genet. 22, 44–52 (1999) | Tsutsui, T. et al. Targeted disruption of CDK4 delays cell cycle entry with enhanced p27Kip1 activity. Mol. Cell Biol. 19, 7011–7019 (1999) FURTHER INFORMATION Cyclins and E2F: a Kohn interaction map | The interactive fly: cell cycle genes | Mitosis world

Cell division: Foreman in the histone factory

NATURE REVIEWS | MOLECULAR CELL BIOLOGY VOLUME 1 | NOVEMBER 2000 | 83 The cell, like any factory, must often step up supply to meet demand. For example, during S phase of the cell cycle, the supply of histones has to be increased to decorate the newly synthesized DNA. Two papers in Genes and Development explain how the kinase CDK2 and its regulatory partner cyclin E coordinate the synthesis of histones and DNA through a CDK2 substrate called NPAT. Humans have two clusters of histone genes, on chomosomes 1 and 6, but the transcription factors driving histone expression vary. Zhao and colleagues reasoned that there must be a ‘master regulator’ of histone expression and set out to find it. Having previously identified NPAT in a screen for cyclin E–CDK2 substrates, they used immunofluorescence to study its cellular localization. This revealed two tiny dots of NPAT in non-S-phase cells, but four in S phase. This localization overlapped with that of coilin, a component of a nuclear organelle called the Cajal body or coiled body (see picture) — a finding corroborated by Ma and colleagues. Cajal bodies often associate with histone gene clusters, so this finding provided an intriguing link between cyclin E–CDK2 and histone genes. Furthermore, fluorescence in situ hybridization showed that NPAT’s association with the histone gene cluster on chromosome 1 was cell cycle dependent, explaining why the number of NPAT dots increases during S phase. Next, Zhao et al. found a large increase in gene expression driven by the histone H4 promoter when the NPAT gene was cotransfected into the cells. NPAT also enhanced expression from the H2B and H3 promoters. For the H4 promoter, the authors narrowed down the NPAT-responsive region to a sequence that binds a putative transcription factor called H4TF-2. Mutations in this sequence that abolish H4TF-2 binding blocked the effect of NPAT, whereas cotransfection of cyclin Eand CDK2expressing plasmids enhanced NPATmediated transcriptional activation of histone genes. Ma and colleagues determined the CDK2 phosphorylation sites on NPAT, then used phospho-NPATspecific antibodies to show that phospho-NPAT colocalizes with both cyclin E and coilin in Cajal bodies, and that the combination of phospho-NPAT and cyclin E–CDK2 is present in Cajal bodies only during S phase. Furthermore, mutation of NPAT’s CDK2 phosphorylation sites to alanine reduced NPAT’s ability to activate transcription from the histone 2B promoter. So cyclin E–CDK2 gives the orders, and NPAT ensures that they’re carried out. Appreciating NPAT’s skills will be our next lesson in this tour of the histone factory: how does NPAT manage its staff — presumably the histone-gene-specific transcription factors — and does it have other teams with responsibilities beyond histone production? Cath Brooksbank Foreman in the histone factory C E L L D I V I S I O N HIGHLIGHTS ADVISORS

Foreman in the histone factory: Cell division

NATURE REVIEWS | MOLECULAR CELL BIOLOGY VOLUME 1 | NOVEMBER 2000 | 83 The cell, like any factory, must often step up supply to meet demand. For example, during S phase of the cell cycle, the supply of histones has to be increased to decorate the newly synthesized DNA. Two papers in Genes and Development explain how the kinase CDK2 and its regulatory partner cyclin E coordinate the synthesis of histones and DNA through a CDK2 substrate called NPAT. Humans have two clusters of histone genes, on chomosomes 1 and 6, but the transcription factors driving histone expression vary. Zhao and colleagues reasoned that there must be a ‘master regulator’ of histone expression and set out to find it. Having previously identified NPAT in a screen for cyclin E–CDK2 substrates, they used immunofluorescence to study its cellular localization. This revealed two tiny dots of NPAT in non-S-phase cells, but four in S phase. This localization overlapped with that of coilin, a component of a nuclear organelle called the Cajal body or coiled body (see picture) — a finding corroborated by Ma and colleagues. Cajal bodies often associate with histone gene clusters, so this finding provided an intriguing link between cyclin E–CDK2 and histone genes. Furthermore, fluorescence in situ hybridization showed that NPAT’s association with the histone gene cluster on chromosome 1 was cell cycle dependent, explaining why the number of NPAT dots increases during S phase. Next, Zhao et al. found a large increase in gene expression driven by the histone H4 promoter when the NPAT gene was cotransfected into the cells. NPAT also enhanced expression from the H2B and H3 promoters. For the H4 promoter, the authors narrowed down the NPAT-responsive region to a sequence that binds a putative transcription factor called H4TF-2. Mutations in this sequence that abolish H4TF-2 binding blocked the effect of NPAT, whereas cotransfection of cyclin Eand CDK2expressing plasmids enhanced NPATmediated transcriptional activation of histone genes. Ma and colleagues determined the CDK2 phosphorylation sites on NPAT, then used phospho-NPATspecific antibodies to show that phospho-NPAT colocalizes with both cyclin E and coilin in Cajal bodies, and that the combination of phospho-NPAT and cyclin E–CDK2 is present in Cajal bodies only during S phase. Furthermore, mutation of NPAT’s CDK2 phosphorylation sites to alanine reduced NPAT’s ability to activate transcription from the histone 2B promoter. So cyclin E–CDK2 gives the orders, and NPAT ensures that they’re carried out. Appreciating NPAT’s skills will be our next lesson in this tour of the histone factory: how does NPAT manage its staff — presumably the histone-gene-specific transcription factors — and does it have other teams with responsibilities beyond histone production? Cath Brooksbank Foreman in the histone factory C E L L D I V I S I O N HIGHLIGHTS ADVISORS