Actin cytoskeleton: Are FH proteins local organizers?

Actin cytoskeleton: Are FH proteins local organizers?

R414 Dispatch Actin cytoskeleton: Are FH proteins local organizers? Jennifer A. Frazier and Christine M. Field The FH proteins, defined by the pres...

292KB Sizes 15 Downloads 120 Views

R414

Dispatch

Actin cytoskeleton: Are FH proteins local organizers? Jennifer A. Frazier and Christine M. Field

The FH proteins, defined by the presence of ‘formin homology’ regions, are important for a number of actindependent processes, including polarized cell growth and cytokinesis. They are large, probably multi-domain, proteins and their function may be in part mediated by an interaction with profilin. Address: Department of Biochemistry and Biophysics, UCSF Medical Center, San Francisco, California 94143–0448, USA. E-mail: [email protected] Current Biology 1997, 7:R414–R417 http://biomednet.com/elecref/09609822007R0414 © Current Biology Ltd ISSN 0960-9822

Many functions of the cell cortex, including adhesion, protrusive motility and cytokinesis, are mediated by specialized regions of the actin cytoskeleton. Each process requires the recruitment of different sets of proteins to specific cortical sites, resulting in the local modulation of actin organization and dynamics. Three questions are crucial for understanding this local functional specialization of the cortex. What are the cues that induce local specialization? How are these cues transduced to recruit specific proteins? And what is the role of these proteins in the organization and function of the actin cytoskeleton? Recently, a group of proteins that

have regions of sequence similarity to the vertebrate formins, known as formin homology (FH) proteins (S. Wasserman, personal communication), has been implicated in the organization of diverse actin assemblies, including cortical patches in budding yeast, cytokinesis rings in fission yeast, and cleavage furrows in higher eukaryotes (Table 1). The formin homology regions were first defined by Castrillon and Wasserman [1], who recognized that the Drosophila protein Diaphanous, the Bni1 protein from the budding yeast Saccharomyces cerevisiae, and the vertebrate formins contain two regions of sequence homology (Figure 1a). These domains, termed formin homology domain 1 (FH1) and formin homology domain 2 (FH2), have now been identified in a number of proteins from a variety of organisms. Most FH proteins were identified genetically, with their mutant phenotypes suggesting defects in actin mediated processes, such as polarized cell growth and cytokinesis (Table 1). However, the vertebrate formins, which have been implicated in limb development, are present primarily in the nucleus [2], and there is genetic evidence that this nuclear localization is important for their function [3]. It is not yet clear if all FH proteins are members of a protein family with related functions or rather constitute a group of distinct proteins that contain formin homology domains.

Table 1 FH-containing proteins: subcellular localization and mutant phenotype. Protein Formins

Organism

Localization

Mutant phenotype/inferred function

Vertebrates

Nucleus and cytoplasm

Limb deformities [2]

Diaphanous

D. melanogaster

Cleavage furrow†

Multinucleate cells (cytokinesis defect) [1]

Cappuccino

D. melanogaster

Unknown

Multinucleate nurse cells (cytokinesis defect), oocyte polarity [18,19]

Cyk1

C. elegans

Unknown

Required for embryonic cytokinesis‡

Bni1

S. cerevisiae

Budneck and bud tip§; mating projection tip [10]

Defects in bipolar bud site selection [20] and cytokinesis (broad neck) [21]

Unknown

Unknown (sequence from yeast data base, accession #P40450)

Putative 157 kDa protein S. cerevisiae Cdc12

S. pombe

Cytokinesis ring [4]

Defects in cytokinesis and septa formation [4]

Fus1*

S. pombe

Projection tip during conjugation [22]

Block in conjugation, cell walls separating mating partners not degraded [22]

A. nidulans

Unknown

Defects in cell polarity and cytokinesis (actin ring fails to form) [5]

FigA/SepA

† K. Oegema, unpublished data. ‡ B. Bowerman, personal communication. § J. Pringle, personal communication.* Note that Fus1p does not have consecutive proline sequences as found in other formins (Fig. 1c).

Dispatch

Figure 1 (a)

(b)

FH1

FH2

FH1

FH2

What are the functions of the FH proteins, and how are these functions related to their conserved sequences? The function of the FH2 domain is not yet known, but recent work has shed some light on the potential function of the proline-rich FH1 domain. Proline-rich sequences have been implicated in a number of protein–protein interactions, including binding to cytoskeletal and signal transduction proteins containing Src homology 3 (SH3) domains [6,7] and WW (defined by two highly conserved tryptophan residues) domains [8], and to profilin [9], a protein involved in the regulation of the actin cytoskeleton. All known FH1 domains contain consensus binding sites for SH3 (and possibly WW) domain-containing proteins, and all except that of Fus1 contain multiple repeats of 5–12 consecutive prolines, although the number of repeats varies from six (Diaphanous) to as few as two (Cdc12) (Figure 1c). The presence of these repeats led Castrillon and Wasserman [1] to speculate that the FH1 domain of Diaphanous might function in part by interacting with profilin, an actin binding protein known to bind polyproline [9]. Two recent papers support the hypothesis that FH1 domains may mediate profilin binding and that proteins containing this domain may play a role in the local organization of the actin cytoskeleton.

(c) Formin IV 694 PPAPPTPPPLPPPLIPPPPPLPPGLGPLPPAPPIPPVCPVSPPPPPPPPPPTP VPPSDGPPPPPPPPPPLPNVLALPNSGGPPPPPPPPPPPPGLAPPPPP 744 Diaphanous 548 PSPNKLPKVNIPMPPPPPGGGGAPPPPPPPMPGRAGGGPPPPPPPPMPGRA GGPPPPPPPPGMGGPPPPPMPGMMRPGGGPPPPPMMMGPMVPVLP 596 Bni1 1291 PPPPPPPPPPVPAKLFGESLEKEKKSEDDTVKQETTGDSPAPPPPPPPPPPPP MALFGKPKGETPPPPPLPSVLSSSTDGVIPPAPPMMP 1328 Cdc12 904 PTPPPPPPLPVKTSLNTFSHPDSVNIVANDTSVAGVMPAFPPPPPPPPPLVSA AGGKFVSPAVSNNISKDDLHKTTGLTRRP 986 Fus1 806 PPPAPLPPPAPPLPTAMSSLQKFEKNDSQIFRKTIIIPENISIDDIFKFCSGSE SEVYASKIP 868

(d) VASP 118 PPPPPALPTWSVPNGPSPEEVEQQKRQQPGPSEHIERRVSNAGGPPAPPAGG PPPPPGPPPPPGPPPPPGLPPSGVPAAAHGAGGGPPPAPPLPAAQGP 216 Mena 281 PPAPPPPPPLPSGPAYASALPPPPGPPPPPPLPSTGPPPPPPPPPPLPNQA PPPPPPPPAPPLPASGIFSGSTSEDNRPLTGLAAAIAGAKLRKVSRVEDGSFP

R415

381

Homologous regions in FH proteins. (a) A representative FH protein as originally defined by Castrillon and Wasserman [1]. All FH proteins contain a proline-rich domain, FH1, and a second region of sequence homology, FH2. Most FH proteins are predicted to form coiled-coils in the regions flanking the FH1 and FH2 domains (indicated by wavy lines.) (b) Representation of an FH protein with the FH2 domain extended to include recently recognized conserved sequences. (c) Sequences of the FH1 domains of selected FH proteins. FH1 domains contain multiple repeats of 5–12 consecutive prolines (highlighted in blue) and consensus SH3 binding sites (underlined in red). SH3 binding consensus sequences are xPxxPxR, RxPxxP [6] and PxxxxPxxP [7] (single letter code, where x is any amino acid). (d) Proline-rich regions of Ena/VASP family members human VASP and Mena.

The FH1 domain is typically a 100 amino acid region characterized by a high proline content. The FH2 domain is a conserved region of 130 amino acids. Although the FH1 and FH2 domains were originally described as distinct domains separated by a 160 residue linker region, recent sequence comparisons [4,5] show that the intervening regions are in fact also conserved, exhibiting 44–67% sequence similarity. In view of this, it now seems reasonable to extend the FH2 domain to include the sequences between the two domains (Figure 1b). Most FH proteins also contain regions predicted to fold as coiled-coils in two stretches that flank the formin homology region. This coiled-coil could mediate homodimerization or even oligomerization.

In a screen for genes required for efficient mating in S. cerevisiae, Evangelista et al. [10] isolated two independent mutant alleles of the gene encoding the FH protein Bni1. They found that, under conditions in which most wild type cells had projections with cortical actin patches concentrated at their tips, cells containing a BNI1 deletion remained depolarized with randomly distributed actin patches. Cells overexpressing an amino-terminally truncated Bni1 also had a depolarized actin cytoskeleton, with an unusual abundance of actin cables and cortical actin patches. Overexpression of this truncated Bni1 was lethal, but could be suppressed in a dosage-dependent manner by both profilin and tropomyosin. Bni1 and profilin were shown to be capable of interacting in yeast cells, by the ‘two-hybrid’ test, and bacterially-expressed profilin was found to bind a tagged fragment of Bni1 in yeast extracts. Taken together, these data suggest that there is an interaction between an FH protein and profilin, and that this interaction may be important for the polarization of the actin cytoskeleton in S. cerevisiae. An interaction between an FH protein and profilin has also been implicated in the formation of the actin-based cell division ring in the fission yeast Schizosaccharomyces pombe. Chang et al. [4] found that cdc12, a gene required for actin-ring assembly and septum formation in S. pombe, encodes an FH protein. Cells with a deletion of cdc12 had normal actin patches during interphase, but delocalized actin patches instead of an actin ring during mitosis. Cdc12 was found to localize to the actin ring but not other actin structures in wild type cells, consistent with a role in

R416

Current Biology, Vol 7 No 7

Figure 2 (a) External signal: polarized growth in response to pheromone

(b) Internal signal: actin ring assembly in S.pombe

α Factor Receptor

SH3 protein Profilin

Plasma membrane

Bni1 Gβγ

Cdc42Gβγ GTP

PPPPP

Local actin effects Bud 6 ?

SH3 Profilin protein Cdc12

Local actin effects

PPPPP

Signal from nucleus? Nucleus

© 1997 Current Biology

Two models for possible roles of FH proteins — Bni1 in (a) and Cdc12 in (b) — in transducing signals to the actin cytoskeleton from either extrinsic or intrinsic cues. Note that binding sites on the FH proteins for an SH3 protein and profilin may overlap.

ring assembly. The S. pombe homologue of profilin, Cdc3, also localizes to the actin ring and is required for ring assembly and septum formation [11]. Chang et al. demonstrated that these two proteins interact, by showing that cdc3 and cdc12 mutants show a synthetic lethal genetic interaction, that profilin is required for the localization of Cdc12, and that an FH-containing fragment of Cdc12 binds to profilin in a polyproline-dependent manner. These two papers [4,10] provide the first direct evidence that FH proteins are involved in organizing the actin cytoskeleton, by showing that actin patches fail to localize in S. cerevisiae bni1 mutants and that the actin ring fails to assemble in S. pombe cdc12 mutants. They also demonstrate that proteins containing formin homology domains can interact with profilin, although the significance of this interaction in the function of these proteins requires further characterization of the role of profilin in vivo. The effects of profilin on actin polymerization in vitro are complex. Profilin can inhibit actin polymerization by sequestering actin monomers, but there is also evidence that profilin can promote actin polymerization by stimulating exchange of the nucleotide bound to actin or competing with other actin-sequestering proteins such as thymosin b4 [12]. Proteins containing formin homology domains are not the only ones likely to function in part by recruiting profilin. Profilin is also thought to bind the proline-rich repeats present in members of the ‘Enabled/vasodilatorstimulated phosphoprotein’ (Ena/VASP) family (Figure 1d). Like FH proteins, the proteins in this family colocalize with actin-rich structures and are thought to affect the actin cytoskeleton. A number of these proteins have multiple binding partners — for example, the protein Mena, in addition to binding to profilin, has been shown to bind ActA, zyxin, and the SH3 domains of the protein kinases Abl and Src [13]. The in vivo relevance of these interactions is, however, unclear.

There is some indication that FH proteins may also have multiple binding partners. Bni1 has been shown to interact with activated Rho1 [14], activated Cdc42 and possibly Bud6 [4]. There is also evidence that one of the mouse formins can interact with several SH3 and WW domain-containing proteins, including cortactin, an actin binding protein enriched in cortical structures. [15]. The similarities between FH proteins and Ena/VASP family members suggest that these proteins could function in an analogous manner, in that they bring together profilin and other cytoskeletal proteins at a specific location in the cell cortex, resulting in modification of the actin cytoskeleton. How far can we currently go toward ascribing specific molecular functions to the FH proteins? Figure 2 illustrates two possible roles of FH proteins in transducing signals to the actin cytoskeleton from either extrinsic or intrinsic cues. In these models, the FH protein acts as a kind of scaffold that recruits other components, although a more active role for these proteins is not ruled out. FH proteins may, for example, function in a manner similar to the S. cerevisiae Ste5 protein, which acts as a scaffold that binds to and organizes the interactions between three different kinases in the mitogen-activated protein (MAP) kinase cascade. Recent data, however, show that Ste5 does not simply act as a passive scaffold — dimerization of Ste5 may in fact be the signal that activates the pathway [16]. It seems plausible that FH proteins similarly act as an ‘active scaffold’. One can speculate that the activation of Cdc42 causes local aggregation of an FH protein, resulting in recruitment of profilin and other actin-modifying factors. There are many questions yet to be answered about FH proteins. The subcellular localization of most proteins is not yet known. This is crucial information for deciphering the role of these proteins in local organization of the actin cytoskeleton. It is also important for distinguishing nuclear versus cytoskeletal functions. Although the vertebrate formins localize to the nucleus, the observation that

Dispatch

a formin interacts with cortactin and other SH3-containing proteins suggests that they may also have a more direct cytoskeletal role. Certainly, nuclear localization does not preclude cytoskeletal function; for example, the actin binding protein anillin, implicated in cleavage furrow assembly, cycles between the nucleus during interphase and the cleavage furrow during mitosis [17]. A number of questions about FH proteins may be resolved with their biochemical characterization. The existence of a potential coiled-coil region suggests they may be able to form oligomeric structures. A combination of hydrodynamic studies and immunoprecipitation, or conventional purification, should be able to determine if FH proteins exist as monomers, dimers or oligomers, and whether they exist alone or in stable complexes with other proteins. Finally, it would be interesting to test the ‘active scaffold’ idea by asking if the aggregation state of FH proteins, and their constellation of bound proteins, is affected by signaling inputs, such as the binding of activated Cdc42. Acknowledgements We thank Arshad Desai, Tim Mitchison, Karen Oegema and Matt Welch for stimulating discussions and for their comments on the manuscript.

References 1. Castrillon D, Wasserman S: diaphanous is required for cytokinesis in Drosophila and shares domains of similarity with the products of the limb deformity gene. Development 1994, 120:3367-3377. 2. Trumpp A, Blundell P, de la Pompa J, Zeller R: The chicken limb deformity gene encodes nuclear proteins expressed in specific cell types during morphogenesis. Genes Dev 1992, 6:14-28. 3. Chan D, Leder P: Genetic evidence that formins function within the nucleus. J Biol Chem 1996, 271:23472-23477. 4. Chang F, Drubin D, Nurse P: cdc12p, a protein required for cytokinesis in fission yeast, is a component of the cell division ring and interacts with profilin. J Cell Biol 1997, 137:169-182. 5. Harris S, Hamer L, Sharpless K, Hamer J. The Aspergillus nidulans sepA gene encodes an FH1/2 protein involved in cytokinesis and the maintenance of cellular polarity. EMBO J 1997, in press. 6. Lim W, Richards F, Fox R. Structural determinants of peptidebinding orientation and of sequence specificity in SH3 domains. Nature 1994, 372:375-379. 7. Ren R, Mayer B, Cicchetti P, Baltimore D: Identification of a 10amino acid SH3 binding site in two Abl SH3 binding proteins and its presence in formins and a muscarinic acetylcholine receptor. Science 1993, 259:1157-1161. 8. Sudol M: The WW module competes with the SH3 domain? Trends Bioch Sci 1996, 21:161-163. 9. Machesky L, Pollard T: Profilin as a potential mediator of membranecytoskeleton communication. Trends Cell Biol 1993, 3:381-385. 10. Evangelista M, Blundell, K, Longtine M, Chow C, Adames N, Pringle J, Peter M, Boone C: Bni1p, a yeast formin linking cdc42p and the actin cytoskeleton during polarized morphogenesis. Science 1997, 276:118-122. 11. Balasubramian M, Hirani B, Burke J, Gould K: The Schizosaccharomyces pombe cdc3+ gene encodes a profilin essential for cytokinesis. J Cell Biol 1994, 125:1289-1301. 12. Theriot J, Mitchison T: The three faces of profilin. Cell 1993, 75:835-838. 13. Gertler F, Niebuhr K, Reinhard M, Wheland J, Soriano P: Mena, a relative of VASP and Drosophila enabled, is implicated in the control of microfilament dynamics. Cell 1996, 87:227-239. 14. Kohno H, Tanaka K, Mino A, Umikawa M, Imamura H, Fujiwara T, Fujita Y, Htotta K, Qadota H, Watanabe T, Ohya Y, Takai Y: Bni1p implicated in cytoskeletal control is a putative target of Rho1p small GTP binding proteins in Saccharomyces cerevisiae. EMBO J 15:6060-6968.

R417

15. Chan D, Bedford M. Leder P: Formin binding proteins bear WWP/WW domains that bind proline-rich peptides and functionally resemble SH3 domains. EMBO J 1996, 15:1045-1054. 16. Yablonski D, Marbach I, Levitzki A: Dimerization of Ste5, a mitogenactivated proteins kinase cascade scaffold protein, is required for signal transduction. Proc Natl Acad Sci USA 1996, 93:13864-13869. 17. Field C, Alberts B: Anillin, a contractile ring proteins that cycles from the nucleus to the cell cortex. J Cell Biol 1995, 131:165-178. 18. Emmons S, Phan H, Calley J, Chen W, James B, Manseau L: cappuccino, a Drosophila maternal effect gene required for polarity of the egg and embryo, is related to the vertebrate limb deformity locus. Genes Dev 1995, 9:2482-2494. 19. Manseau L, Calley J, Phan H: Profilin is required for posterior patterning of the Drosophila oocyte. Development 1996, 122:2109-2116. 20. Zahner J, Harkins H, Pringle J: Genetic analysis of the bipolar pattern of bud site selection in the yeast Saccharomyces cerevisiae. Mol Cell Biol 1996, 16:1857-1870. 21. Jansen R, Dowzer C, Michaelis C, Galova M, Nasymth K: Mother cellspecific HO expression in budding yeast depends on the unconventional myosin myo4p and other cytoplasmic proteins. Cell 1996, 84:687-697. 22. Petersen J, Weilguny D, Egel R, Nielsen O: Characterization of fus1 of Schizosaccharomyces pombe: a developmentally controlled function needed for conjugation. Mol Cell Biol 1995, 15:3697-3707.