AUTOINHIBITORY DOMAINS: Modular Effectors of ... - Semantic Scholar

10 Sep 2002

19:59

AR

AR170-CB18-15.tex

AR170-CB18-15.sgm

LaTeX2e(2002/01/18) P1: IBD 10.1146/annurev.cellbio.18.031502.133614

Annu. Rev. Cell Dev. Biol. 2002. 18:421–62 doi: 10.1146/annurev.cellbio.18.031502.133614 c 2002 by Annual Reviews. All rights reserved Copyright ° First published online as a Review in Advance on June 28, 2002

AUTOINHIBITORY DOMAINS: Modular Effectors of Cellular Regulation Miles A. Pufall and Barbara J. Graves Huntsman Cancer Institute, Department of Oncological Sciences, University of Utah, 2000 Circle of Hope, Salt Lake City, Utah 84112-5550; e-mail: [email protected]; [email protected]

Key Words repression, intramolecular, masking, allosteric, autoinhibition ■ Abstract Autoinhibitory domains are regions of proteins that negatively regulate the function of other domains via intramolecular interactions. Autoinhibition is a potent regulatory mechanism that provides tight “on-site” repression. The discovery of autoinhibition generates valuable clues to how a protein is regulated within a biological context. Mechanisms that counteract the autoinhibition, including proteolysis, posttranslational modifications, as well as addition of proteins or small molecules in trans, often represent central regulatory pathways. In this review, we document the diversity of instances in which autoinhibition acts in cell regulation. Seven well-characterized examples (e.g., σ 70, Ets-1, ERM, SNARE and WASP proteins, SREBP, Src) are covered in detail. Over thirty additional examples are listed. We present experimental approaches to characterize autoinhibitory domains and discuss the implications of this widespread phenomenon for biological regulation in both the normal and diseased states. CONTENTS INTRODUCTION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Definition of Autoinhibition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Autoinhibition as a Regulatory Strategy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Experimental Approaches . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . INHIBITION OF LIGAND BINDING . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Sigma 70 (σ 70 ) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Ets-1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . SNARE Proteins . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Wiscott-Aldrich Syndrome Protein (WASP) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . INHIBITION OF SUBCELLULAR LOCALIZATION . . . . . . . . . . . . . . . . . . . . . . . . ERM Proteins . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . SREBP . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . INHIBITION OF ENZYMATIC ACTIVITY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Src Kinase . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . CONCLUSIONS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Conformational Change is a Common Feature 1081-0706/02/1115-0421$14.00

422 422 424 425 426 426 432 435 438 441 442 444 447 447 450

421

10 Sep 2002

19:59

422

AR

PUFALL

AR170-CB18-15.tex

¥

AR170-CB18-15.sgm

LaTeX2e(2002/01/18)

P1: IBD

GRAVES

of Autoinhibition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Autoinhibition Adds a Layer of Regulation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A Second Role for the Autoinhibitory Domain . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Implications for Human Disease . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

451 451 452 453

INTRODUCTION Definition of Autoinhibition Autoinhibition is a widespread phenomenon that plays a key role in the regulation of proteins by facilitating the response to signaling pathways. The precise regulation of protein activities is essential for normal growth and development as well as homeostasis. A picture is emerging from a diverse set of biological phenomena that intramolecular interactions between separable elements within a single polypeptide provide a common regulatory strategy to modulate protein function. Specifically, one region of a protein interacts with another to negatively regulate its activity (Figure 1). In the clearest examples, the controlling elements make up a defined domain, the autoinhibitory domain, which mediates inhibition. The simplest route to discovery of this regulatory pathway is the characterization of the activity of fragments of a protein relative to the activity of the full-length species. The enhancement of an activity of a particular domain, such as kinase activity or DNA binding, in the absence of some other region of the protein indicates autoinhibition. With such data in hand, the deleted region can be implicated as the autoinhibitory domain. However, this observation is just the beginning. Defining the mechanism of inhibition and elucidation of how the autoinhibition is counteracted or reinforced requires extensive additional experimental investigation. Importantly, delineation of an autoinhibitory mechanism is a guide to the discovery of crucial regulatory schemes. To illustrate the widespread use of autoinhibition as a regulatory strategy, this review covers examples in which a variety of activities are inhibited. Inhibition of ligand binding is the most common class of autoinhibition. This scenario is well represented by inhibition of DNA-protein interactions, as in the case of transcription factors (e.g., σ 70 and Ets-1). Inhibition of protein-protein interactions is also frequently observed (e.g., WASP, SNARE, and ERM proteins). Cellular compartmentalization is another target of inhibition, as reported for the transcription factors SREBP and ATF6, which are sequestered in membranes, and the ERM proteins whose membrane attachment is inhibited. Src kinase illustrates autoinhibition of enzymatic activity. Within this diversity there is a common thread in the mechanism of autoinhibition: an intramolecular interaction that either directly or allosterically interferes with the function of a “targeted” domain (Figure 1A). The functional domain could be directly occluded from a necessary ligand interaction or constrained in a nonfunctional conformation by a more indirect mechanism. The modular organization of proteins facilitates regulation by autoinhibition. Individual proteins often perform a variety of functions, and separable regions of

10 Sep 2002

19:59

AR

AR170-CB18-15.tex

AR170-CB18-15.sgm

LaTeX2e(2002/01/18)

AUTOINHIBITORY DOMAINS

P1: IBD

423

Figure 1 Autoinhibition is a regulatory mechanism. (A) An autoinhibitory domain modulates the activity of a second, separable domain (center). Autoinhibition can be counteracted and reinforced by modification (left) or by association with a second molecule, partner protein (right). Autoinhibition is often identified experimentally by deletion of the autoinhibitory domain. Proteolysis is also a regulatory strategy in vivo (lower). (B) Modularity of proteins facilitates autoinhibition. A transcription factor with two functional domains (one for DNA binding and one for transactivation) illustrates the specificity of an autoinhibitory domain for a second, separable target domain. Alternatively, a specialized autoinhibitory domain mediates inhibition (top); a dual function domain mediates both inhibition and transactivation (center); or two domains are mutually inhibitory and carry out other functions (lower).

10 Sep 2002

19:59

424

AR

PUFALL

AR170-CB18-15.tex

¥

AR170-CB18-15.sgm

LaTeX2e(2002/01/18)

P1: IBD

GRAVES

proteins can carry out each function. Well-defined structural domains of a protein retain their folded state as a protein fragment and can function in isolation from other regions. Transcription factors illustrate this concept (Figure 1B). A discretely folded structural domain can often be identified that binds DNA independently of the remainder of the protein. Likewise regions within transcription factors that activate or repress transcription are functional when separated from their cognate DNA-binding domains. The modular design of proteins has several implications for the autoinhibitory phenomenon. First, autoinhibitory domains are distinct from domains that are the target of the inhibition. However, the relationship of an autoinhibitory domain to other domains varies (Figure 1B). The autoinhibitory domain can be one whose sole purpose is inhibition or it can be a domain that mediates inhibition but also performs a second activity. Alternatively, there can be mutual inhibition between two domains that possess other activities. A second feature of the modularity of proteins is that domains are often linked by flexible regions. The transition from the inhibited to the activated state usually requires this flexibility (Figure 1A). Finally, a distinguishing feature for a module that functions as an autoinhibitory domain is the set of intramolecular interactions. Although most protein modules can function in isolation, potentially acting like beads on a string even within the full-length protein, an autoinhibitory domain is structurally coupled to the targeted domain. Its inhibitory function is inextricably linked to the function of the remainder of the protein. These intramolecular interactions, which may be composed of seemingly small surface areas and weak interactions, are bolstered by the relatively high effective local concentration that is generated by tethering of the interacting parts. Although analogous inhibition might be designed to function in trans with intermolecular interactions, the affinity would need to be stronger and the specificity would need to be more finely tuned.

Autoinhibition as a Regulatory Strategy Discovery of autoinhibition provides valuable clues for the investigation of biological regulatory strategies for a particular protein. The autoinhibitory domain is an on-site repressor that restrains the targeted domain in a secure off state. In some cases this is viewed as the default state, although in others, autoinhibition is reinforced by external factors or modifications (Figure 1A) (e.g., Ets-1, Src, SNAREs). A corollary of autoinhibition is the existence of regulatory strategies that counteract the inhibition. Autoinhibition of a molecule presents a reversible barrier that prevents spurious activation of a signaling pathway. This allows a system to be primed for response only to appropriate signals. There are diverse mechanisms for counteracting inhibition (Figure 1A), the most common of which is the displacement of the inhibitory domain by a second molecule, thus replacing the intramolecular interaction with an intermolecular interaction. However, even this simple formulation is played out with surprising diversity (e.g., Ets-1, WASP, SNAREs, σ 70, and Src). Other mechanisms include proteolysis of the inhibitory domain (e.g., SREBP, ATF6), post-translational modification (ERM proteins) or

10 Sep 2002

19:59

AR

AR170-CB18-15.tex

AR170-CB18-15.sgm

LaTeX2e(2002/01/18)

AUTOINHIBITORY DOMAINS

P1: IBD

425

binding of small molecules that allosterically alter the inhibitory domain. Except in the case of proteolysis, an autoinhibitory domain also provides a mechanism to reverse activation, resetting the switch for the next cycle of activation. The widespread use of autoinhibition in different contexts emphasizes the importance of tight control of both the off and on state in biological regulatory pathways.

Experimental Approaches In spite of extremely divergent biological settings for known autoinhibitory phenomena, there are striking similarities in the experimental approaches used to decipher inhibitory mechanisms. A standard first step is to map the minimal region necessary for inhibition as well as the minimal region that shows full de-repressed activity. These protein fragments are often easily engineered with cDNA clones and produced in one of a variety of expression systems. Partial proteolysis is a useful alternative route to discovery of autoinhibition because inhibitory function often maps to a discrete region, and this strategy generates stable fragments that often make up structural domains (e.g., σ 70 and Ets-1). Quantitative assays that provide a reliable measure of the activity of the inhibited and activated fragments are necessary to establish the validity of the autoinhibitory effects. Controls for correct folding of the inhibited molecule are advisable because the low activity of the inhibited molecule could reflect a structural defect. One approach for testing the inhibitory function of the delimited minimal domain can be its transfer to a heterologous protein. However, one of the hallmarks of autoinhibitory domains is that their inhibitory effect requires specific features within the cognate, targeted domain. Thus this technique is not likely to be successful unless the recipient protein is highly related to the cognate protein. Most mechanistic models of autoinhibition predict the existence of intramolecular interactions between the inhibitory elements and the functional domain. Mutants that activate, presumably by disrupting these interactions, are informative. This intramolecular interaction model also can be tested experimentally using the separated domains. A variety of protein-protein interaction assays can demonstrate intermolecular binding of an inhibitory domain to the targeted functional domain. An interesting variant of this experiment is to test for repression in trans. The σ 70, SNARE, and WASP examples illustrate this approach. However, this experiment is not always successful. Although the tethering can often be mimicked experimentally by a high concentration of the inhibitory fragment, in some cases direct structural coupling of the inhibitory elements to the targeted domain has been found to be crucial. Some of the most illuminating and definitive findings come from structural approaches. High-resolution molecular data from NMR experiments and crystallography provide the most specific information about structure, intramolecular contacts, and conformational change (e.g., σ 70 Ets-1, Src, SNARE and ERM proteins). Conformational changes also can be detected by biophysical techniques, including circular dichroism (CD) spectroscopy and NMR approaches (e.g., SNAREs,

10 Sep 2002

19:59

426

AR

PUFALL

AR170-CB18-15.tex

¥

AR170-CB18-15.sgm

LaTeX2e(2002/01/18)

P1: IBD

GRAVES

Ets-1, and WASP), as well as lower-resolution strategies such as partial proteolysis (e.g., Ets-1) or electron microscopy (e.g., SNAREs and σ 70). These structural data can provide a framework for proposing a mechanism for autoinhibition, especially in cases in which structural data are available for both the inhibited and activated states (e.g., Ets-1, SNAREs, ERM, and WASP proteins). Structural modeling also provides a context for interpreting mutant phenotypes and directing new mutagenesis to test specific mechanistic hypotheses. One of the most interesting features of the structural data is the observation that there is structural fluidity in the inhibitory elements. Many cases show alternative conformations in the inhibited versus activated state. In contrast to experimental approaches that characterize an autoinhibition phenomenon, strategies for investigation of regulatory pathways that impinge upon autoinhibition cannot be formulated generically. The distinctive function of the protein within its biological context must be considered. Nevertheless, a detailed mechanistic and structural model of autoinhibition is an invaluable tool in exploring the critical routes to regulation.

INHIBITION OF LIGAND BINDING Ligand binding is one of the most universal properties of proteins, with a vast range of ligands from small molecules to large macromolecules, including lipids, nucleic acids, and other proteins. It is not surprising that the greatest number of examples of autoinhibition include cases in which ligand binding is inhibited. In the first two examples provided below, the ligand is DNA, and the proteins of interest play a role in regulation of gene expression. The next two examples demonstrate a role for autoinhibition in regulation of vesicle fusion and cell motility. These essential cell processes require ligand binding within context of multiprotein complexes. Table 1 cites other examples of autoinhibition that work at the level of DNA binding and assembly of multiprotein complexes within the cell.

Sigma 70 (σ 70) DNA binding by the prokaryotic transcription factor σ 70 is repressed by autoinhibition. σ 70 only binds DNA in complex with RNA polymerase. This first example is described in detail to illustrate the methodology for discovery and characterization of an autoinhibitory phenomenon. SIGMA SUBUNIT DIRECTS PROMOTER SELECTIVITY OF RNA POLYMERASE HOLOENZYME Transcriptional initiation in prokaryotes requires the direct binding of the

multisubunit RNA polymerase holoenzyme to transcriptional control elements. Biochemical and genetic data indicate that the sigma subunit within the holoenzyme directly binds the two conserved promoter elements, known as the –35 and –10 regions, which reflects their distance from the start site of transcription. The

10 Sep 2002

19:59

AR

AR170-CB18-15.tex

AR170-CB18-15.sgm

LaTeX2e(2002/01/18)

P1: IBD

AUTOINHIBITORY DOMAINS

427

TABLE 1 Further examples of autoinhibition Proteina LIGAND BINDING NF-κB, transcription factor

Leu3p, transcription factor

Inhibited activity Transcription activation by CBP/p300 binding Transcription activation

Inhibitory domain/ activation (repression)b C-terminal region masks N-terminal interaction domain/PKA phosphorylation Internal region masks activation/activate by alpha-isopropylmalate DNA binding domain/ response element binding Masking TPR arrays/ Brf1 association C-terminal region of SNAPc p190 subunit/ association with Oct-1 N-terminal region/SNAPc binding to N-terminal region, TFIIB association

Nuclear receptors

Transcription activation

TFIIIc131, pol III transcription factor SNAPc, small nuclear RNA-activating complex TBP, TATA-binding protein

TFIIIB70 binding

Hoxb-1, Drosophila Lab (labial)— homeodomain transcription factor IRF4 (Pip), transcription factor

DNA binding

Hexapeptide motif/ EXD protein binding

DNA binding and transcription activation DNA binding

C-terminal region/PU.1 binding to C-terminal region C terminus/ phosphorylation C-terminal region/ phosphorylation during viral infection PF/PN motif/SH3 domain C-terminal region of cyclinT1/ binding of TATA-SF1 to cyclinT1 N-terminal pseudoligand/ Bcl-XL binding N-terminal region/ acidic phospholipids

p53, transcription factor IRF 3, interferon regulatory factor, transcription factor Esx1, homeodomain transcription factor PTEFb, transcription elongation factor Bid, BH3 interacting domain protein Vinculin, actinmembrane linker

DNA binding

DNA binding, DNA bending

DNA binding, activation, nuclear localization DNA binding, nuclear entry RNA (TAR element) binding Apoptosis via protein:protein F-actin binding

Referencec (Zhong et al. 1998)

(Wang et al. 1999) (Lefstin & Yamamoto 1998) (Moir et al. 2002) (Mittal et al. 1999) (Kuddus & Schmidt 1993, Mittal & Hernandez 1997, Zhao & Herr 2002) (Chan et al. 1996)

(Brass et al. 1996) (Ko & Prives 1996) (Lin et al. 1999) (Yan et al. 2000) (Fong & Zhou 2000) (Tan et al. 1999) (Bakolitsa et al. 1999, Johnson & Craig 2000) (Continued)

10 Sep 2002

19:59

428

AR

PUFALL

AR170-CB18-15.tex

¥

AR170-CB18-15.sgm

LaTeX2e(2002/01/18)

P1: IBD

GRAVES

TABLE 1 (Continued) Proteina

Inhibitory domain/ Inhibited activity activation (repression)b

Alpha-actinin

Titin binding

ENZYMATIC ACTIVITY PAS kinase, period, Kinase aryl hydrocarbon, single-minded homology domain Vav, GDP-GTP GEF activity exchange factor (GEF) CKI epsilon, casein Kinase kinase I epsilon GSK3beta, glycogen Kinase synthase kinase 3beta MRCK, myotonic dystrophy kinaserelated Cdc42-binding kinase Pak1, (S. pombe)—p21GTPase activated protein kinase, also MIHCK PKA/PKG, cAMP- and cGMP-dependent kinases SNF1, protein kinase

Ephb2 receptor, ephrin B1 binding receptor tyrosine kinase Cystathionine betasynthase

Calcineurin, CN, protein phosphatase 2A

Kinase

Referencec

Z-repeat motif pseudoligand/PIP2 binds actin-binding domain

(Young & Gautel 2000)

PAS domain/metabolites (proposed)

(Rutter and McKnight 2001)

N-terminal extension binds GTPase binding site/ tyrosine phosphorylation C-terminal extension/ phosphatase N-terminal phosphopseudosubstrate/(inhibited by phosphorylation) Distal coiled-coil domain binds kinase domain/ phorbol ester

(Aghazadeh et al. 2000) (Cegielska et al. 1998) (Dajani et al. 2001) (Tan et al. 2001)

Kinase

Regulatory domain/ Cdc42 binding

(Brzeska et al. 2001, Tu & Wigler 1999)

Kinase

Pseudosubstrate/cAMP or cGMP

(Francis et al. 2002)

Kinase

Regulatory domain binds catalytic domain/SNF4 binds in low glucose Juxtamembrane domain binds catalytic domain/phosphorylation C-terminal domain/ S-adenosyl-L-methionine binding

(Jiang & Carlson 1997)

Kinase

Condense homocysteine and serine in cysteine biosynthesis Phosphatase

(WybengaGroot et al. 2001) (Janosik et al. 2001)

N-terminal pseudosubstrate/ (Tokoyoda calmodulin binding et al. 2000) (Continued )

10 Sep 2002

19:59

AR

AR170-CB18-15.tex

AR170-CB18-15.sgm

LaTeX2e(2002/01/18)

AUTOINHIBITORY DOMAINS

P1: IBD

429

TABLE 1 (Continued) Proteina

Inhibited activity

Inhibitory domain/ activation (repression)b

NADP malate dehydrogenase

Reduction of oxaloacetate to L-malate

C-terminal disulfide bridge occludes active site/ reduction of disulfide bond

(Krimm et al. 1999)

13 phosphates constitute NES/dephosphorylation by calcineurin reveals NLS Intramembrane domain sequestration/proteolysis IBB pseudoligand/NLS and/or importin beta

(Okamura et al. 2000)

SUBCELLULAR LOCALIZATION NFAT1, nuclear factor Nuclear of activated T-cells localization transcription factor Tubby, transcription Nuclear factor localization Importin alpha Nuclear localization signal binding Notch, transcription Nuclear factor localization

Relish NF-κB, transcription factor

Nuclear localization

Transmembrane domain anchor/binding of intracellular domain to DSL ligand stimulating intramembrane proteolysis C terminus/proteolytic cleavage

Referencec

(Santagata et al. 2001) (Catimel et al. 2001, Kobe 1999) (Mumm & Kopan 2000)

(Stoven et al. 2000)

a

Examples are included based on personal knowledge of the authors and PubMed searches using the terms autoinhibition, autoinhibitory, self regulating, and intrasteric inhibition. This resulting list is extensive although not comprehensive.

b

The majority of examples are stimuli that counteract autoinhibition. Those that reinforce autoinhibition are in parentheses.

c

References cited are representative of most recent work, reviews, or key publications and are not intended to be comprehensive.

–35 region is recognized by sigma factors as double-stranded DNA. The –10 region is the site of duplex melting, and the ability of sigma to bind single-stranded DNA is critical for transcription initiation. Sigma subunits can be divided into two general classes. The primary sigmas, including σ 70 of Escherichia coli, direct transcription of most genes during normal growth, whereas more specialized sigma subunits, such as the heat shock responsive σ 32 of E. coli, direct RNA polymerase to a specific subset of genes (Gross et al. 1998). The discovery that σ 70 binds directly to DNA provided an important breakthrough in the understanding of promoter recognition. Initial work failed to detect DNA binding activity by σ 70 subunit when it was isolated from the holoenzyme. In 1992, proteolytic fragments of σ 70 were found to bind DNA in isolation and display the same promoter selectivity as holoenzyme. Consistent with previous genetic analysis, quantitative binding studies identified two DNA-binding domains within σ 70 that target the –10 and –35 promoter elements. Most significantly for our discussion, these early studies AUTOINHIBITION REGULATES SIGMA 70 DNA BINDING

10 Sep 2002

19:59

430

AR

PUFALL

AR170-CB18-15.tex

¥

AR170-CB18-15.sgm

LaTeX2e(2002/01/18)

P1: IBD

GRAVES

Figure 2 Autoinhibition of DNA binding by σ 70 is counteracted by RNA polymerase. (A) Domain structure of σ 70. Conventional nomenclature for E. coli RNA polymerase subunit σ 70 (regions 1.1 to 4.2) designates regions of conservation among related sigma factors. RNA polymerase holoenzyme binds to –10 and −35 elements in prokaryotic promoters via regions 2.1–2.4 and 4.2, respectively. Region 1.1 functions as an autoinhibitory domain (Dombroski et al. 1992). (B) Model of σ 70 autoinhibition and RNA polymerase binding to promoter DNA. In the inhibited form of σ 70, intramolecular interactions between regions 1.1 and the C-terminal region mask region 4.2, the −35 DNA-binding domain (left). Forming holoenzyme, σ 70 binds directly to RNA polymerase, thus making a specific set of contacts (middle). Both σ 70 and RNA polymerase undergo structural rearrangements in the transition from free holoenzyme to a closed complex formation (not shown) and finally open complex, (right). These changes include establishment of a new set of intramolecular and intermolecular interactions (Burgess & Anthony 2001, Gruber et al. 2001, Mekler et al. 2002, Naryshkin et al. 2000). In addition, σ 70 binds the –35 and –10 promoter elements and aids in the melting of DNA for open complex formation (Wilson & Dombroski 1997).

suggested an autoinhibitory phenomenon and mapped the inhibitory function to the N-terminal region of σ 70 (Dombroski et al. 1992) (Figure 2A). These initial investigations led to a model of autoinhibition in which repressive interactions between autoinhibitory domain and the two DNA-binding domains prevent σ 70 DNA binding. Furthermore, the relief of autoinhibition was modeled as a switch from these repressive intramolecular interactions to a set of intermolecular interactions between σ 70 and RNA polymerase. A decade of additional investigation have validated this model and led to new mechanistic insights (Figure 2B). The autoinhibitory domain was mapped by experiments in which N-terminal fragments were added in trans to the C-terminal fragments that are active in DNA binding. The region necessary for trans inhibition spans the first 100 amino acids. As expected, inhibition in trans requires a large molar excess of the inhibitory fragment. The trans experiments also delineated targets of inhibition. The

10 Sep 2002

19:59

AR

AR170-CB18-15.tex

AR170-CB18-15.sgm

LaTeX2e(2002/01/18)

AUTOINHIBITORY DOMAINS

P1: IBD

431

autoinhibitory domain inhibits binding to the –35 promoter element, whereas a fragment bearing only the DNA-binding domain for the –10 promoter region is not inhibited in trans. It is proposed that inhibition of the –10 binding region may require structural coupling to the autoinhibitory domain (Dombroski et al. 1993). RNA POLYMERASE COUNTERACTS AUTOINHIBITION Interaction with RNA polymerase relieves the autoinhibition of σ 70, thus enabling DNA binding. A variety of genetic and biochemical studies indicate that a large region of σ 70, encompassing both DNA-binding domains, contacts RNA polymerase (Lesley & Burgess 1989, Sharp et al. 1999). Most significantly, the intermolecular contacts of the autoinhibitory domain map to a specific structural feature on RNA polymerase. Furthermore, a subset of these polymerase-sigma interactions is also negatively regulated by the intramolecular contacts of the autoinhibitory domain (Gruber et al. 2001). This observation led to a two-step model of binding in which an initial contact with RNA polymerase induces a conformational change in σ 70 that facilitates access to additional contact surfaces (Figure 2B). The conformational changes in σ 70 are a critical component of the mechanism that activates DNA binding. Changes were monitored initially by chemical reactivity of cysteines engineered individually in both the inhibitory domain and the DNA-binding domain (Callaci et al. 1998). A second biophysical approach utilized luminescence resonance energy transfer to measure changes in intramolecular distances (Callaci et al. 1999). These studies identified conformational changes in both the inhibitory domain and the two DNA-binding domains and suggest that polymerase induces a change in the distance between the two DNA-binding regions. In summary, both genetic and biochemical observations indicate that, in addition to a simple unmasking step, a more complex set of conformation changes in σ 70 plays a role in polymerase-induced DNA binding. In addition to its role in autoinhibition, the N-terminal region of σ 70 plays a role in transcription initiation. Promoter recognition can be detected with RNA polymerase containing a sigma that lacks the autoinhibitory domain; however, addition of the autoinhibitory domain in trans stimulates initiation (Vuthoori et al. 2001, Wilson & Dombroski 1997). Such a role is consistent with the extensive set of sigma-polymerase interactions that are revealed during the transition from closed to open complex formation (Gruber et al. 2001). Additional high-resolution modeling has lent insight into the role of σ 70 within the RNA polymerase holoenzyme. Experiments performed with fluorescence resonance energy transfer technology have deduced the position of σ 70 in holoenzyme. Surprisingly, the autoinhibitory domain of σ 70 is positioned near the polymerase active site. This segment of σ 70 is proposed to be a molecular mimic of double-stranded DNA because it is displaced from this position by DNA in the open complex. Initial contacts made between σ 70 and RNA polymerase are rearranged to form a stable complex similar to the open complex (Mekler et al. 2002).

10 Sep 2002

19:59

432

AR

PUFALL

AR170-CB18-15.tex

¥

AR170-CB18-15.sgm

LaTeX2e(2002/01/18)

P1: IBD

GRAVES

AUTOINHIBITION PROVIDES ADDED REGULATION OF PROMOTER SELECTIVITY Autoinhibition appears to be a common control mechanism among various bacterial sigma factors. In addition to σ 70, the heat shock factor σ 32 of E. coli also has an N-terminal inhibitory domain (Dombroski et al. 1993). Also, the developmental sigma factor of Bacillus subtilis, σ K, is autoinhibited (Kroos et al. 1989). In each case, autoinhibition tightly links DNA binding of the sigma factor to its association with RNA polymerase; isolated sigma factors cannot occupy promoters alone. Importantly, autoinhibition provides a route to activation of specialized sigmas. For example, σ K of B. subtilis is activated by proteolysis that is tightly coupled to signaling (Cutting et al. 1991, Rudner & Losick 2002). Exchangeable sigma factors provide specificity for gene expression in bacteria. The use of autoinhibition in this setting helps guarantee tight control.

Ets-1 DNA binding by the eukaryotic transcription factor Ets-1 is repressed by autoinhibitory elements. A detailed structural model for the autoinhibition mechanism illustrates the central role of conformational change. Autoinhibition of Ets-1 is counteracted by a protein partnership and reinforced by phosphorylation. This regulatory strategy provides Ets-1 a degree of specificity within the ets family of transcription factors. ETS-1 REGULATES TRANSCRIPTION BY SEQUENCE-SPECIFIC DNA BINDING In eukaryotes, transcription factors that display sequence-specific DNA binding are often used to selectively direct the transcriptional machinery to promoters. These proteins orchestrate differential gene expression in response to physiological cues and developmental events. Although each factor is required to direct transcriptional control of a subset of specific genes, there is a remarkable level of conservation among subclasses of these proteins. Specifically, the DNA-binding domains of many regulatory transcription factors are conserved, which has resulted in similar DNA-binding properties. Specificity of action among these factors requires strategies that distinguish individual family members. Autoinhibition represents a strategy to regulate DNA binding and provide such specificity. The ets gene family encodes regulatory transcription factors that illustrate this specificity conundrum. The human genome encodes 25 ets genes. Each ETS protein bears a highly conserved DNA-binding domain (ETS domain) that displays a winged helix-turn-helix motif and recognizes a core sequence motif 50 GGAA/T 30 . The proteins show significant diversity outside of the ETS domain. Furthermore, genetic studies show that individual ets genes can display unique biological properties. The founding member of the ets gene family, Ets-1, illustrates the many layers of regulation that generate specificity for its DNA-binding activity (Graves & Petersen 1998, Sharrocks 2001). AUTOINHIBITION OF ETS-1 DNA BINDING Ets-1 is regulated by an intramolecular conformational switch that forms the basis of an autoinhibitory mechanism.

10 Sep 2002

19:59

AR

AR170-CB18-15.tex

AR170-CB18-15.sgm

LaTeX2e(2002/01/18)

AUTOINHIBITORY DOMAINS

P1: IBD

433

Autoinhibitory elements were discovered in Ets-1 by the observation that proteolysis of Ets-1 or deletion of regions flanking the ETS domain enhanced DNAbinding activity (Graves et al. 1998, Hagman & Grosschedl 1992, Lim et al. 1992). Quantitative analyses showed that deletion of either the N-terminal or Cterminal flanking regions fully activated DNA binding, suggesting that two regions function together to mediate autoinhibition (Jonsen et al. 1996) (Figure 3A). Structural studies using NMR spectroscopy identified three inhibitory helices within these regions (Skalicky et al. 1996). These NMR studies also detected structural coupling between inhibitory helices, as well as a connection between the inhibitory elements and helix H1 of the ETS domain (Figure 3B). These initial studies established the existence of an autoinhibitory module with interacting elements but did not provide a clear mechanism of inhibition. Most strikingly, the data did not support a simple model in which the inhibitory elements would mask the major DNA-binding surface of the ETS domain, the helix-turn-helix motif. Additional structural studies provided further insight into the mechanism of autoinhibition. Protease sensitivity measurements and CD spectroscopy analysis of alpha helicity showed that the inhibitory elements undergo a change upon DNA binding that involves the dramatic unfolding of one of the inhibitory helices (Petersen et al. 1995) (Figure 3B). The energy required for this conformational change is likely the basis of the inhibitory mechanism. High-resolution molecular models of the ETS domain in contact with DNA provided further insight (Batchelor et al. 1998; Garvie et al. 2001; Kodandapani et al. 1996; Mo et al. 1998, 2000). Helix H1 of the ETS domain directly binds DNA via a set of phosphate contacts. Based on this structural information and mutational analysis of the DNA-protein interaction, a key role has been proposed for helix H1. The model suggests that helix H1 can exist in one of two conformations. In the absence of DNA, this helix packs with the inhibitory elements to form a metastable inhibitory module. In the presence of DNA, the N terminus of helix H1 contacts the phosphate backbone via a precisely positioned hydrogen bond. This contact is dependent on the macrodipole of the helix and relies on the strict positioning of the entire structural element. Interconversions between the proposed alternative states would require conformational change (Graves et al. 1998, Wang et al. 2001). REGULATORY PATHWAYS ACT THROUGH AUTOINHIBITION The autoinhibitory mechanism functions as a rheostat-type switch. DNA binding is either enhanced by regulatory pathways that counteract inhibition or further inhibited by regulatory pathways that reinforce the inhibition. Ets-1 binds with a variety of partner proteins to form ternary complexes on DNA (Dickinson et al. 1999, Garvie et al. 2001, Sheridan et al. 1995, Sieweke et al. 1998). These partnerships provide additional specificity because the appropriate promoter provides an Ets-1 binding site as well as a properly positioned binding site for the partner protein. In the case of RUNX1 (previously CBFα2 and AML1), cooperative DNA binding has been characterized (Goetz et al. 2000, Gu et al. 2000, Kim et al. 1999). RUNX1 enhances

10 Sep 2002

19:59

434

AR

PUFALL

AR170-CB18-15.tex

¥

AR170-CB18-15.sgm

LaTeX2e(2002/01/18)

P1: IBD

GRAVES

Figure 3 Autoinhibition of Ets-1 DNA binding is reinforced by phosphorylation and counteracted by a protein partnership. (A) Domain structure of eukaryotic transcription factor Ets-1. Pointed (PNT) domain functions in protein interactions. ETS domain bears a winged helix-turn-helix motif and displays sequence-specific DNA binding. Autoinhibitory elements flank the ETS domain. The serine-rich region is the site of inhibitory phosphorylation. Secondary structure of ETS domain and inhibitory elements are provided by NMR (Donaldson et al. 1996, Skalicky et al. 1996) and crystallographic analyses (Garvie et al. 2001). (B) Model of Ets-1 autoinhibition and regulation of DNA binding. The autoinhibitory module structure, including the position of helix HI-1, is modeled from genetic and biochemical analyses (crystallographic data for HI-2, H4, and H5; C.W. Garvie, personal communication) (center). Two inhibitory alpha helices, HI-1 and HI-2, cooperate to inhibit DNA binding by making intramolecular contacts with the ETS domain, specifically helix H1. DNA binding disrupts intramolecular contacts between the ETS domain and the autoinhibitory module, characterized by the unfolding of HI-1 (left). Phosphorylation of the serine-rich region in response to calcium release in activated T cells further inhibits DNA binding by stabilizing the autoinhibitory module (upper right) (Cowley & Graves 2000). A protein partnership with RUNX1 counteracts the inhibitory mechanism by interacting with the autoinhibitory module (Goetz et al. 2000) (lower right). The DNA-binding domain of RUNX1 is adapted from Bravo et al. (2001). Hatched oval represents the N-terminal region necessary for cooperative DNA binding for which there is no structural information.

10 Sep 2002

19:59

AR

AR170-CB18-15.tex

AR170-CB18-15.sgm

LaTeX2e(2002/01/18)

AUTOINHIBITORY DOMAINS

P1: IBD

435

Ets-1 binding affinity for DNA with the same magnitude as autoinhibition (approximately tenfold). Furthermore, this cooperativity requires an intact inhibitory module. These results suggest that DNA binding cooperativity functions by counteracting the autoinhibitory mechanism (Figure 3B). On the other hand, a Ca2+dependent phosphorylation of Ets-1 dramatically reduces DNA binding (Cowley & Graves 2000, Rabault & Ghysdael 1994). These sites of phosphorylation lie near the inhibitory elements, and an intact autoinhibitory domain is required for the added repression. Protease sensitivity measurements and CD spectroscopy indicate that the added phosphates stabilize the inhibitory module (Cowley & Graves 2000) (M. Pufall, unpublished observations). This reduction in binding affinity is thought either to abrogate DNA binding altogether or to alter the specificity of DNA binding by rendering Ets-1 more selective for promoters that provide a partner binding site. Regulation of gene expression requires that highly related DNA binding proteins execute target site specificity. Within the ets family, autoinhibition appears to be one strategy for distinguishing different family members. Autoinhibition has been detected in several other ETS proteins, including Elk-1 (Price et al. 1995) and PEA3 (Bojovic & Hassell 2001, Greenall et al. 2001); however, the autoinhibitory elements are different in each of these proteins. Interestingly, helix H1 of the ETS domain is highly conserved in all of these family members. It is possible that the ETS domain is a conserved target of inhibition via the helix H1-DNA contact, whereas the mechanism that affects this DNA contact may have diverged among the various family members during evolution.

SNARE Proteins The SNARE complex is a heterotrimeric assemblage that mediates membrane fusion. Assembly of this complex is modulated by an autoinhibitory domain present in the syntaxin1a component. This example is distinguished by a structural model that shows two conformations of the syntaxin component, one in the inhibited and one in the activated state. SNARE PROTEINS MEDIATE MEMBRANE FUSION Membrane fusion is required for a variety of cellular functions including transport of molecules between membranebound organelles, organelle growth, and inheritance during mitosis. Membrane fusion also is a central step in synaptic transmission in the nervous system. Preceding fusion, vesicles must be docked to the recipient membrane, a process mediated by a set of membrane-associated proteins, termed SNAREs (soluble NSF attachment protein receptor where NSF is N-ethyl-maleimide-sensitive fusion protein) (Chen & Scheller 2001, Misura et al. 2000a). There are many different SNARE proteins distributed in specific membrane locations within the cell, and both the SNAREs and other membrane-associated proteins appear to contribute to specificity of membrane fusion (Pelham 2001, Waters & Pfeffer 1999). Representatives from the three major families of SNARE proteins form the SNARE complex (Figure 4). The incoming vesicle brings a v-SNARE of the

10 Sep 2002

19:59

436

AR

PUFALL

AR170-CB18-15.tex

¥

AR170-CB18-15.sgm

LaTeX2e(2002/01/18)

P1: IBD

GRAVES

Figure 4 Autoinhibition of syntaxin and binding of n-Sec1 regulates SNARE core assembly. (A) Domain structure of syntaxin1a. The N-terminal inhibitory module of syntaxin1a autonomously folds into three alpha helices. The SNARE motif forms three alpha helices in the inhibited form but only one extended alpha helix in the activated state within the SNARE core complex. SNARE, soluble N-ethylmaleimide-sensitive fusion attachment protein receptor; TM, transmembrane region. (B) Model of syntaxin1a autoinhibition and SNARE core complex assembly (adapted from Chen & Scheller 2001). The N-terminal inhibitory module of syntaxin1a interacts intramolecularly with the three alpha helices in the C-terminal region, inhibiting assembly of the SNARE complex. The n-Sec1 protein reinforces this autoinhibition by binding both the N and C domains (left) (Misura et al. 2000b). A structural rearrangement leads to association of one helix from syntaxin1a, two helices of SNAP-25, and one helix from VAMP. SNAP-25 and VAMP acquire structure in the complex and syntaxin1a changes structure as intramolecular contacts are displaced by intermolecular contacts (right) (Sutton et al. 1998). Syntaxin family and SNAP-25 family are t-SNARES. Yeast homologs include Sso1p and Sec9p, respectively. The v-SNAREs includes VAMPs, neuronal synaptobrevin, Sec22b and yeast Ykt6p. n-Sec1/munc18 homologs include yeast Sec1, Drosophila ROP, C. elegans UNC-13. SNAP-25, 25 kDa synatopsomeassociated protein; VAMP, vesicle associated membrane protein.

VAMP family, whereas, the target membrane has two t-SNARE proteins, one of the syntaxin family and one of the SNAP-25 family. Each SNARE protein has a membrane attachment domain, and each contributes one or two (in case of SNAP-25) alpha helices to the SNARE core complex. Crystal structure of a SNARE core with synaptobrevin-II, syntaxin1a, and SNAP-25B shows a four-helix bundle that can be described generally as a coiled-coil (Sutton et al. 1998). The formation

10 Sep 2002

19:59

AR

AR170-CB18-15.tex

AR170-CB18-15.sgm

LaTeX2e(2002/01/18)

AUTOINHIBITORY DOMAINS

P1: IBD

437

of the heterotrimer brings the two membranes into close proximity to facilitate membrane fusion. AUTOINHIBITION REGULATES SNARE COMPLEX ASSEMBLY Autoinhibition of the syntaxin class of t-SNAREs regulates the assembly of the SNARE complex. Proteins of the syntaxin family are composed of two structural domains (Figure 4A). The C-terminal domain binds the SNARE core by contributing a single alpha helix to the four-helix bundle. The N-terminal domain is not necessary for SNARE core formation. Instead, this domain impairs formation of a binary complex between syntaxin and SNAP25, as well as binding of VAMP. Early studies reported that adding the N terminus of syntaxin1a in trans inhibited the participation of the C terminus in SNARE core formation (Calakos et al. 1994). More recently, kinetic studies on the yeast syntaxin homolog, Sso1p, demonstrated that assembly is regulated. The binary complex of Sso1p and the yeast SNAP25 homolog, Sec9p, shows equal stability with either the full length or the C-terminal fragment of Sso1p; however, the association rate constant of binary complex formation is ∼1000-fold higher with the isolated C-terminal fragment (Nicholson et al. 1998). These studies established the N-terminal region as an autoinhibitory domain that regulates SNARE complex assembly. The working model of autoinhibition suggests that intramolecular interactions between the N- and C-terminal domains of syntaxin compete for intermolecular interactions with the SNARE core (Figure 4B). Structural data and a variety of biochemical approaches show that intramolecular interactions link the N- and C-terminal domains. In addition to straightforward binding assays, NMR experiments suggest that interactions between N- and C-terminal domains of syntaxin1a induce folding of the C-terminal fragment, which appears largely unstructured in isolation. Furthermore, chemical shift changes in the N terminus are observed in the presence of the C terminus (Dulubova et al. 1999). The full picture of intramolecular interactions comes into focus from crystallographic studies of the yeast syntaxin Sso1p. A three-helix bundle within the terminal domain interacts with a set of short helical segments within the C-terminal domain (Munson et al. 2000) (Figure 4B, left). The conformation of the C terminus differs from the single helix that participates in SNARE core (Figure 4B, right). Thus SNARE complex formation, which overcomes autoinhibition, requires both a conformational change in the syntaxin and a switch from intramolecular to intermolecular interactions. AUTOINHIBITION AS A REGULATORY STRATEGY A non-SNARE protein, n-Sec1 (also called munc18) provides regulation of complex assembly. A crystallographic analysis of a complex of syntaxin1a and n-Sec1 shows both the N- and C-terminal domains of syntaxin1a interact with n-Sec1 (Figure 4B). The structure is similar to that observed in the absence of n-Sec1, with the N- and C-terminal domains retaining the inhibitory conformation that occludes the binding of other SNAREs to syntaxin (Misura et al. 2000b). These structural data suggest a role for n-Sec1 in repression, possibly stabilizing the inhibitory conformation. In contrast, genetic analyses in Caenorhabditis elegans and Drosophila suggest roles

10 Sep 2002

19:59

438

AR

PUFALL

AR170-CB18-15.tex

¥

AR170-CB18-15.sgm

LaTeX2e(2002/01/18)

P1: IBD

GRAVES

for n-Sec1 homologs in both negative and positive regulation of membrane fusion (Richmond et al. 2001, Wu et al. 1998). To accommodate these apparently confusing observations, n-Sec1 is proposed to prime the conformational change in the syntaxin C terminus or to provide a structural platform for SNARE assembly (Misura et al. 2000b, Munson et al. 2000). The N-terminal regions of v-SNAREs may also provide a regulatory device. Structural studies of the N-terminal domain of two VAMPs show a profilin fold that is entirely different from that observed in syntaxin (Gonzalez et al. 2001, Tochio et al. 2001). Nevertheless, preliminary analysis of the yeast VAMP Ykt6p suggests that this domain negatively regulates SNARE complex formation (Tochio et al. 2001). Thus multiple SNARE proteins may contribute to regulation of SNARE complex assembly by an autoinhibitory mechanism. Autoinhibition provides an important level of control for SNARE assembly. There is considerable controversy concerning the promiscuity of SNARE assembly; however, it is likely that some degree of specificity is required for orderly membrane fusion events. On-site repression of assembly by autoinhibition could favor specific interactions. Future work is needed to ascertain whether regulatory routes that counteract the inhibitory mechanism connect membrane fusion and signaling.

Wiscott-Aldrich Syndrome Protein (WASP) WASP proteins regulate actin assembly via activation of the Arp2/3 complex. The activation function is masked by autoinhibition. This example illustrates how autoinhibition can be synergistically counteracted by several signaling pathways. WASP FUNCTIONS IN ACTIN DYNAMICS The dynamic assembly and disassembly of actin filaments controls the shape of much of the membrane within a cell. In addition to providing a structural scaffold, actin filaments are involved in changing membrane shape in cell division, vesicular transport, and motility. Each of these processes requires characteristic networks of actin filaments (Chen et al. 2000). The WASP family of proteins serves as a paradigm for the translation of cellular signals into directed actin polymerization (Wear et al. 2000). The WASP family of proteins catalyzes actin nucleation and polymerization at the membrane by activating the Arp2/3 complex, which is composed of seven polypeptides that activate actin filament growth by increasing the rate of actin nucleation, branching, and polymerization. The complex alone, however, nucleates actin poorly and requires the presence of an activator, including members of the WASP family of proteins (Higgs & Pollard 2000; Zalevsky et al. 2001a,b). The conserved C terminus of all WASP family members, termed the VCA domain (verprolin homology, cofilin homology, acidic region), is necessary for this activation (Figure 5A). The VCA domain has been shown to bind directly to the Arp2/3 complex, inducing a conformational change that converts it into an active form (Volkmann et al. 2001, Zalevsky et al. 2001b). Although fully active

10 Sep 2002

19:59

AR

AR170-CB18-15.tex

AR170-CB18-15.sgm

LaTeX2e(2002/01/18)

AUTOINHIBITORY DOMAINS

P1: IBD

439

Figure 5 Autoinhibition of WASP regulates its activation of the Arp2/3 complex and actin polymerization. (A) Domain structure of WASP/N-WASP [adapted from (Higgs & Pollard 2001)]. The enabled/VASP homology (EVH1), basic region (BR), Rho GTPase-binding domain, (GBD), and proline-rich region each bind ligands that activate WASP/N-WASP. The BR and GBD comprise the minimal autoinhibitory domain regulating the VCA domain. The VCA domain stimulates actin polymerization and branching by activating the Arp2/3 complex. The hematopoietic-restricted WASP and more ubiquitously expressed N-WASP are ∼50% identical and share a similar gene structure. The differences in N-WASP, including extended proline-rich and acidic regions, as well as an extra V domain, influence the activity of the protein in vitro; however, the in vivo significance is not known. V, verprolin homology domain; C, cofilin homology domain, also known as the central or connector region; A, acidic domain. Note: Sometimes V and C together are referred to as WASP homology domain (W or WH1) (Higgs & Pollard 2001). (B) Model of autoinhibition and activation by chemotactic stimuli [adapted from (Prehoda et al. 2000, Rohatgi et al. 2000)]. Intramolecular interactions between inhibitory elements, BR-GBD, and the VCA domain mask the activating function of N-WASP (left). Chemotactic agents stimulate exchange of GDP to GTP in the Rho GTPase Cdc42 and up-regulate phosphatidylinositol-4,5bisphosphate (PIP2) levels in the membrane. Activated GTP-Cdc42 and PIP2 displace the inhibitory elements GBD and BR, allowing the VCA domain to activate the Arp2/3 complex, which induces both de novo actin polymerization and actin branching, both necessary for the formation of filopodia (Miki et al. 1998). Trimerization of Arp2/3 (black circles) and V domain–bound actin (gray circles) is proposed to stimulate actin polymerization (right) (Robinson et al. 2001).

10 Sep 2002

19:59

440

AR

PUFALL

AR170-CB18-15.tex

¥

AR170-CB18-15.sgm

LaTeX2e(2002/01/18)

P1: IBD

GRAVES

in isolation, the activity of the VCA domain is inhibited in the context of the full-length protein by the N terminus, suggesting an autoinhibitory mechanism (Higgs & Pollard 2001) (Figure 5B). The N terminus, in turn, makes contacts with Cdc42 and phosphatidyl inositol-4,5-biphosphate (PIP2), which are anchored in the membrane. Much work has been done on many WASP family members, but the member that most clearly illustrates autoinhibition is N-WASP, which is discussed below. MULTIPLE SIGNALING PATHWAYS COUNTERACT AUTOINHIBITION Inhibition of the activity of N-WASP by the N-terminal region can be counteracted by a wide variety of cellular inputs. The N terminus is composed of four domains: the Enabled/VASP homology 1 domain (EVH1), a basic region (BR), the GTPase-binding domain (GBD), and the proline-rich domain (sometimes referred to as the PRD) (Higgs & Pollard 2001) (Figure 5A). Each of these domains has been implicated in serving as a binding surface for activating molecules: EVH1 is related to pleckstrin homology domains and binds proline-rich regions, the BR binds PIP2, the GBD binds Rho GTPases such as Cdc42, and the proline-rich region binds SH3 domains (Abdul-Manan et al. 1999; Fedorov et al. 1999; Fukuoka et al. 2001; Prehoda et al. 2000; Rohatgi et al. 2000, 2001; Rudolph et al. 1998). Membrane-bound PIP2 activates WASP alone or synergistically with either an SH3 domain–containing protein or Cdc42 (Fukuoka et al. 2001; Prehoda et al. 2000; Rohatgi et al. 2000, 2001) (Figure 5B). Thus N-WASP is a control point for a variety of cell signals directing targeted actin polymerization at the membrane. MECHANISTIC MODELING Despite the variety of possible activating inputs, deletion analysis of N-WASP revealed a minimally sized autoinhibitory region (Prehoda et al. 2000, Rohatgi et al. 2000). Fragments of the N terminus were added in trans to gauge their inhibitory effect on the VCA domain. Although the GDB partially inhibits VCA activity, fragments containing both the GDB and the BR are necessary to fully inhibit the VCA domain in actin polymerization assays in vitro. Surprisingly, although both regions are necessary for full inhibition, only the GBD binds directly to the VCA domain (Figure 5B). This binding is not enhanced by the addition of other domains, nor do other domains interact with the VCA domain independently. These results suggested an inhibitory mechanism. GDB is hypothesized to mediate direct contact with the VCA, an interaction that provides only partial inhibition. The BR is then postulated to enhance the inhibitory effect of the GBD, without directly binding the VCA or affecting the affinity of the GBD for the VCA. The inhibitory module may function by occluding Arp2/3 binding; however, support of this mechanism is controversial. One group showed that N-terminal fragments containing the GBD-BR prevent binding of Arp2/3 to the VCA domain (Rohatgi et al. 2000), whereas another group showed the opposite (Prehoda et al. 2000). Strikingly, the latter group also demonstrated an interaction between the BR and Arp2/3 bound to the VCA region. This interaction suggests a different

10 Sep 2002

19:59

AR

AR170-CB18-15.tex

AR170-CB18-15.sgm

LaTeX2e(2002/01/18)

AUTOINHIBITORY DOMAINS

P1: IBD

441

model in which the Arp2/3 module could in fact bind to the VCA region but could not be activated owing to the direct interaction with the BR (Prehoda et al. 2000). In this alternative model, the BR either masks the activity of the Arp2/3 complex, or prevents it from undergoing the structural transition necessary for activation. The exact role of the GBD in this model is not yet clear. Structural studies provide more detail of this intramolecular interaction and further highlight the metastable nature of the inhibitory elements. NMR spectroscopy revealed that stabilization of the GDB is critical to both autoinhibition and activation by Cdc42. The GBD alone was shown by both CD and NMR spectroscopy to be unstructured in the absence of binding partners. Addition of alcoholic solvents, which stabilize protein folding, induced the formation of two anti-parallel beta sheets and an alpha helix. Strikingly, highly similar secondary structural elements were induced in the presence of either the activating ligand, Cdc42, or the inhibited target, the VCA domain. Biochemical studies further indicated that the binding of the GBD to Cdc42 and the VCA are mutually exclusive. Together, these studies indicate that the GBD is stabilized either intramolecularly by the VCA domain to create an inhibited WASP or intermolecularly by Cdc42 to create an activated WASP. Importantly, GBD and the VCA contacts only involve the coffilin (C) homology region (Kim et al. 2000). This leaves open the possibility that either model for Arp2/3 binding could be correct. Arp2/3 could bind to WASP through part of the C domain and A domain even in its inhibited form, or it could clash sterically in the C domain (Abdul-Manan et al. 1999, Kim et al. 2000). Amide exchange experiments using a GBD-VCA fragment indicate that the GBD is only partially displaced by Cdc42, which again suggests that multiple inputs are necessary to relieve autoinhibition (Buck et al. 2001). The mechanistic and structural work provide a valuable model of counteracting autoinhibition, although the exact mechanism of inhibition of Arp2/3 activation has not yet been resolved. ROUTE TO SPECIFICITY The variable N termini of WASP family members might provide a route to specificity in directing actin polymerization. WASP, N-WASP, and Scar proteins are all homologous in their VCA and proline-rich regions, but they diverge significantly in their inhibitory N termini (Higgs & Pollard 2001, Machesky et al. 1999). Interestingly, however, Scar proteins, similar to WASP and N-WASP, also localize to the leading edge of cells (Hahne et al. 2001). An attractive hypothesis that highlights the potential of autoinhibitory modules is that Scar, WASP, and N-WASP have similar functions but are regulated differently by cellular signals due to differences in their N termini. This possibility awaits further study.

INHIBITION OF SUBCELLULAR LOCALIZATION The compartmentalization of proteins within a cell is essential for the regulation and execution of a variety of activities. Our two examples illustrate opposite scenarios. In the first example, membrane attachment is inhibited, thus preventing

10 Sep 2002

19:59

442

AR

PUFALL

AR170-CB18-15.tex

¥

AR170-CB18-15.sgm

LaTeX2e(2002/01/18)

P1: IBD

GRAVES

localization to the plasma membrane. In the second example, a transcription factor is sequestered in the ER membrane, thereby preventing its nuclear localization. Table 1 lists additional examples of regulation of subcellular compartmentalization. In particular, regulation of nuclear trafficking is a common target for autoinhibition.

ERM Proteins ERM proteins link the plasma membrane to the actin cytoskeleton. The domains for membrane interaction and actin binding are autoinhibited by reciprocal intramolecular interactions. This example has a high-resolution structural model of a fully inhibited molecule dramatically illustrating intramolecular masking as an autoinhibition mechanism. ERM PROTEINS LINK PLASMA MEMBRANE TO ACTIN CYTOSKELETON The function of the cytoskeleton requires attachment to a variety of cellular anchors. A key linkage is the physical connection between cytoskeletal actin and the plasma membrane. Actin-membrane linkages are critical for cell adhesion, organization of cell surface receptors, cell cortical resilience, and the formation of specialized membrane structures including microvilli and nerve growth cones. The ERM proteins, ezrin, radixin, and moesin crosslink the membrane to the actin cytoskeleton (Bretscher 1999, Bretscher et al. 2000, Mangeat et al. 1999). These proteins display an autoinhibitory mechanism that regulates both actin binding and membrane attachment. The ERM proteins are characterized by two conserved domains that provide the link between the plasma membrane and the actin cytoskeleton. The N-terminal domain, termed N-ERM associate domain (N-ERMAD), displays a FERM domain that functions in membrane binding (Figure 6A). The N-ERMAD binds directly either to integral membrane proteins such as CD44 and I-CAM-1,-2, or to an adaptor protein such as EBP50, which binds membrane proteins (Bretscher 1999). The FERM domain is well structured with three connected subdomains, as determined by crystallographic studies of moesin and radixin (Edwards & Keep 2001, Hamada et al. 2000, Pearson et al. 2000). The C-terminal domain, termed C-ERMAD, binds F-actin at its extreme terminus (Pestonjamasp et al. 1995, Turunen et al. 1994) (Figure 6A). AUTOINHIBITION REGULATES BOTH MEMBRANE AND ACTIN BINDING ERM proteins display intramolecular interactions that mask both the FERM domain interaction with membrane and the interaction of the C terminus with F-actin. Initially discovered by binding assays with protein fragments (Gary & Bretscher 1995, Henry et al. 1995, Martin et al. 1995), autoinhibition is now best understood from a high-resolution structural study of a complex between N-ERMAD and C-ERMAD fragments of moesin (Pearson et al. 2000). The C-ERMAD forms an extended structure that contacts a large surface area of the FERM domain in

10 Sep 2002

19:59

AR

AR170-CB18-15.tex

AR170-CB18-15.sgm

LaTeX2e(2002/01/18)

AUTOINHIBITORY DOMAINS

P1: IBD

443

Figure 6 Autoinhibition of ERM proteins regulates membrane attachment to actin cytoskeleton. (A) Domain structure of ERM proteins (moesin shown). The N-ERMAD is composed of a FERM domain (F1, F2, and F3 subdomains), which binds membraneassociated proteins. The C-ERMAD includes an actin-binding site (hatched ). These two halves are linked by a putative coiled-coil motif denoted the helical domain. (B) Model of ERM protein autoinhibition and activation (adapted from Pearson et al. 2000). Structural information for the inhibited form is derived from crystal structure of moesin N-ERMAD and C-ERMAD fragments in complex without helical region linkage (Pearson et al. 2000). With the modeled intact protein (shown), the C-ERMAD interacts intramolecularly with the N-ERMAD, masking both membrane attachment and actin binding activities (left). The binding of PIP2 to the FERM domain and phosphorylation of the C-ERMAD activates ERM proteins to bind membrane-associated protein and F-actin, thus crosslinking the plasma membrane and the actin cytoskeleton (right). ERM, ezrin, radixin, moesin: three proteins encoded by paralogous genes; FERM, 4.1 (four point one), ezrin, radixin, moesin homology domain; ERMAD, ERM-association domain.

a meandering manner (Figure 6B). The interface is composed of residues that are highly conserved among the ERM proteins. Most importantly, the C-terminal part of the interface overlaps with the F-actin-binding site, thereby, masking actin association. The transition from the inhibited to activated state involves the switch from intramolecular to intermolecular interactions with F-actin (Figure 6B). The conformation of the C-terminal region may change in the presence of actin (Pearson et al. 2000). The intramolecular interactions that inhibit actin binding also inhibit EBP50 binding (Reczek & Bretscher 1998). In this case, the FERM domain also switches from intramolecular to intermolecular interactions to attach to the membrane. Thus the ERM proteins have two masked activities. Interestingly,

10 Sep 2002

19:59

444

AR

PUFALL

AR170-CB18-15.tex

¥

AR170-CB18-15.sgm

LaTeX2e(2002/01/18)

P1: IBD

GRAVES

intermolecular interactions between N-ERMADs and C-ERMADs from different ERM proteins are also detected in cells and in vitro, and the resulting head-to-tail oligomers may be functionally important (Gary & Bretscher 1995, Gautreau et al. 2000, Nguyen et al. 2001). ACTIVATION OF ERM PROTEINS Both post-translational modifications and ligand binding counteract autoinhibition of ERM proteins. Phosphorylation of a specific threonine within the C-ERMAD weakens the intramolecular interaction (Matsui et al. 1998, Simons et al. 1998). Consistent with this role, this threonine, as mapped by crystallography, is located within the intramolecular interface, but with some solvent exposure (Pearson et al. 2000). In addition to affecting intramolecular interactions, the phosphorylation of this threonine may play a role in the transition from oligomeric to active monomeric forms of ezrin (Gautreau et al. 2000). Binding of phospholipids is another route to activation. Phospholipids bind the FERM domain; PIP2 specifically enhances binding of ERM proteins to CD44 (Hirao et al. 1996, Niggli et al. 1995) and synergizes with phosphorylation to stimulate actin binding (Nakamura et al. 1999). A crystal structure of the radixin FERM domain in complex with IP3 (the head group of PIP2) suggests phospholipid binding might allosterically disrupt intramolecular interactions (Hamada et al. 2000). The diversity of subdomains in the FERM domain and the modular nature of the C-ERMAD suggest that other mechanisms of activation of ERM proteins remain to be elucidated (Gautreau et al. 2002). Microfilament attachment to the plasma membrane plays an important role in dynamic membrane events, such as adhesion, motility, and establishment of polarity. The ERM proteins have emerged as central players in many of these processes, having both a structural and regulatory role. It is interesting that both actin binding and membrane binding are targets of autoinhibition. Thus the subcellular localization of the ERM proteins is regulated by autoinhibition. Finally, the dormant monomeric form of the ERM proteins facilitates the cellular response to changing conditions and supports the dynamic nature of the cortical cytoskeleton and the cellular membrane.

SREBP The eukaryotic transcription factor SREBP (sterol response element-binding protein) is sequestered in the ER membrane. Proteolyis within the membrane releases SREBP and allows nuclear localization. Coincidentally, the membrane-localized protease that releases SREBP shows similarity to the protease that cleaves the B. subtilis sigma factor σ K to counteract its autoinhibition (Brown et al. 2000). SREBP REGULATES CHOLESTEROL HOMEOSTASIS Homeostatic levels of substances in the cell must be strictly maintained. Many of the substances, such as oxygen, metals, and cholesterol, are necessary at certain levels but are detrimental at higher levels. The regulation of cholesterol in particular has been carefully studied in this

10 Sep 2002

19:59

AR

AR170-CB18-15.tex

AR170-CB18-15.sgm

LaTeX2e(2002/01/18)

AUTOINHIBITORY DOMAINS

P1: IBD

445

regard. Cholesterol is necessary for membrane fluidity, assembly of signaling rafts, and as a biosynthetic precursor for hormones. Excess levels of cholesterol form crystals that are extremely toxic to cells (Brown & Goldstein 1997). Cholesterol levels are regulated by SREBP, whose activity is in turn repressed by an autoinhibitory mechanism. SREBP, a transcription factor that binds DNA via a basic helix-loop-helix leucine zipper (bHLH-ZIP) motif, is maintained in an inactive form as a two-pass transmembrane protein (Brown & Goldstein 1997, 1999) (Figure 7). As opposed to other autoinhibitory mechanisms described in this review, this transmembrane domain does not function by an intramolecular masking mechanism but by sequestering the transcription factor in the endoplasmic reticulum (ER), away from its site of action. Proteolysis triggered by low cholesterol liberates SREBP from its membrane tether, allowing it to translocate to the nucleus, where it regulates genes encoding enzymes essential for the cholesterol biosynthetic pathway (Brown & Goldstein 1997). REGULATED INTRAMEMBRANE PROTEOLYSIS COUNTERACTS THE AUTOINHIBITION OF SREBP Liberation of SREBP and other membrane-tethered proteins is exe-

cuted by regulated intramembrane proteolysis (RIP), which involves the sequential cleavage of a transmembrane protein. The first cleavage shortens the portion of the protein projecting into the extracytosolic space. The second cleavage then cuts within the transmembrane domain, which disrupts anchoring (Brown et al. 2000). In the case of SREBP, conditions of low cholesterol are sensed by SCAP (SREBP cleavage-activating protein), which binds the C-terminal domain of SREBP and transports it from the ER to the Golgi (Figure 7B, left). In the lumen of the cis-Golgi, site 1 protease (S1P) cleaves SREBP between the two transmembrane domains (DeBose-Boyd et al. 1999, Nohturfft et al. 1999). The N terminus of the protein, which contains the transcription factor and a single-pass type 2 transmembrane domain, is subsequently cleaved by site 2 protease (S2P) within the transmembrane region (Rawson et al. 1997). Next, the SREBP C-terminal fragment diffuses into the cytoplasm and translocates to the nucleus to regulate transcription (Brown & Goldstein 1997). RIP REGULATES OTHER MEMBRANE-BOUND PROTEINS Other targets of RIP include apolipoprotein (APP), NOTCH, Ire1, and ATF6 (Brown et al. 2000). Of these, the RIP of the transcription factor ATF6 has striking similarity to SREBP. ATF6 is sequestered in the ER by a type 2, single-pass transmembrane domain and released after sequential cleavage by S1P and S2P (Ye et al. 2000) (Figure 7B, right). The transcriptionally competent N-terminal fragment translocates to the nucleus and activates genes regulated by ER stress response elements (Haze et al. 1999, Yoshida et al. 1998). The sensor for detecting unfolded proteins has not been identified. Although different from the autoinhibitory mechanisms described elsewhere that require intramolecular interactions, this autoinhibitory mechanism is otherwise very similar. First, if their transmembrane autoinhibitory domains are deleted, these transcription factors are constitutively active. Second, autoinhibition via the

10 Sep 2002

19:59

446

AR

PUFALL

AR170-CB18-15.tex

¥

AR170-CB18-15.sgm

LaTeX2e(2002/01/18)

P1: IBD

GRAVES

Figure 7 Autoinhibition of nuclear localization of transcription factors SREBP and ATF6 is counteracted by regulated intramembrane proteolysis, RIP. (A) Domain structure of ATF6/SREBP [adapted from (Brown & Goldstein 1997, Brown et al. 2000, Li et al. 2000)]. SREBP is a two-pass transmembrane protein. The portion of the protein N terminal of the transmembrane domain is a transcription factor (TF ) that binds DNA via a bHLH-ZIP domain at sterol responsive elements. The C terminus is necessary for sterol-responsive cleavage (Brown & Goldstein 1997). Activating transcription factor (ATF6) is a type 2 transmembrane domain protein. Like SREBP, the part of the protein N terminal to the transmembrane domain is a transcription factor, in this case binding endoplasmic reticulum stress response elements via a bZIP domain (Yoshida et al. 2001). TM, transmembrane domain; bHLH-ZIP, basic helix-loop-helix zipper motif; b-ZIP, basic leucine zipper motif. (B) Model of autoinhibition of SREBP and ATF6 and activation by RIP (adapted from Brown et al. 2000). SREBP is sequestered in the ER until the sterol cleavage-activating protein (SCAP) senses low cholesterol levels. SCAP then chaperones SREBP to the cis-Golgi for cleavage by site 1 protease (S1P). After S1P cleavage, the N terminus is cleaved by site 2 protease (S2P) within the transmembrane domain, liberating the transcription factor from the membrane. ATF6 is cleaved by S1P in response to the ER stress response induced by increased levels of unfolded protein. ATF6 also is subsequently cleaved by S2P in the transmembrane domain, releasing the transcription factor from autoinhibition.

10 Sep 2002

19:59

AR

AR170-CB18-15.tex

AR170-CB18-15.sgm

LaTeX2e(2002/01/18)

AUTOINHIBITORY DOMAINS

P1: IBD

447

transmembrane domain is overcome in response to cell signals: sterol starvation or excess unfolded protein levels. In addition, these examples show a particularly exquisite level of specificity. SREBP and ATF6 are cleaved by the same two proteases, yet the cellular signals are different, and there appears to be no cross talk (Haze et al. 1999, Ye et al. 2000).

INHIBITION OF ENZYMATIC ACTIVITY The vast literature on enzymes presents a plethora of strategies for regulation. Many are regulated by autoinhibition as defined in this review. The classic zymogen is an excellent example (Khan & James 1998). In this case, an autoinhibitory domain inhibits protease activity, and activation is mediated by proteolysis of the inhibitory elements. In several examples presented in Table 1, small molecules or post-translational modifications counteract autoinhibition of enzymes. Examples also illustrate the autoinhibitory domain as a pseudosubstrate that binds the catalytic site and is displaced during activation. We have chosen to highlight the classic Src kinase story in which the autoinhibitory regions act allosterically to block kinase activity and for which there is a high-resolution structure of the inhibited state.

Src Kinase Kinase activity of Src is repressed by two inhibitory regions. The replacement of these inhibitory intramolecular interactions with analogous intermolecular interactions leads to de-repression. This example illustrates how relatively weak interactions function in cis-acting autoinhibition, whereas stronger intermolecular interactions are necessary for activation in trans. THE ROLE OF SRC KINASE IN SIGNALING Responses to stimuli propogated at the cell surface in metazoans often involves the sequential phosphorylation of proteins within a signaling pathway. Each protein in the pathway is thus potentially able to be regulated on two levels: kinase activity and association with its downstream substrate. One well-studied example that involves autoinhibition of both of these types of regulation is Src kinase (Gonfloni et al. 2000). Src, and eight other related family members, are non-receptor tyrosine kinases that can be activated by cell surface receptor tyrosine kinases (RTKs), as well as by downstream targets in growth and proliferation pathways (Wybenga-Groot et al. 2001). Each family member can be functionally divided roughly into two components. The N-terminal component has a membrane-anchoring myristylation site and Src homology domains 2 and 3 (SH2 and SH3) domains (Figure 8A). The C-terminal component contains the kinase domain, as well as a C-terminal extension, which interacts intramolecularly with the SH2 and SH3 domains to mediate autoinhibition of the kinase domain (Figure 8B). Activation of Src kinase is mediated by the binding of the SH2 and SH3 domains to a ligand, such as

10 Sep 2002

19:59

448

AR

PUFALL

AR170-CB18-15.tex

¥

AR170-CB18-15.sgm

LaTeX2e(2002/01/18)

P1: IBD

GRAVES

Figure 8 Autoinhibition of Src kinase activity is regulated by ligands for SH2 and SH3 domains. (A) Domain structure of Src kinase (adapted from Pawson 1997). The kinase activity is mediated by a two-part kinase domain that shows homology to other tyrosine kinases. The Src homology domains (SH2 and SH3) and the C-terminal extension cooperate to form an autoinhibitory module that inhibits the activity of the kinase domain. Phosphorylation of tyrosine 527 in the C-terminal extension is required for autoinhibition. The myristylation site at the N terminus is required for membrane localization and proper signaling. (B) Model of autoinhibition and activation (adapted from Xu et al. 1999). The SH2 and SH3 domains constrain the kinase domain in an inactive conformation through intramolecular interactions. The SH3 domain interacts with a pseudoligand in the linker connecting the N-terminal kinase lobe to the SH2 domain. The SH2 domain binds a pseudoligand containing the phosphotyrosine 527 in the C-terminal extension (left). These intramolecular interactions are displaced by high-affinity intermolecular interactions between the SH2 and SH3 domains and activating ligands. Consensus motifs for ligands are noted (right). Disruption of these intramolecular interactions allows the kinase domain to adopt an active conformation, resulting in an activating phosphorylation at tyrosine 416 (Sicheri et al. 1997; Xu et al. 1997, 1999).

10 Sep 2002

19:59

AR

AR170-CB18-15.tex

AR170-CB18-15.sgm

LaTeX2e(2002/01/18)

AUTOINHIBITORY DOMAINS

P1: IBD

449

the phospho-tyrosines in the cytoplasmic domain of RTKs (Kuriyan & Cowburn 1997, Moran et al. 1990, Wybenga-Groot et al. 2001) (Figure 8B). The activated Src kinase then phosphorylates substrates that trigger the Ras and PI3 kinase pathways, which induce growth and cell proliferation (Aftab et al. 1997). IDENTIFICATION OF INHIBITORY ELEMENTS Autoinhibition of Src kinase was discovered in the context of an oncogenic version of Src. v-Src kinase, encoded by Rous sarcoma virus, is responsible for cell transformation by this retrovirus (Stehelin et al. 1976). The cellular homolog, c-Src, is unable to transform cells (Parker et al. 1984). The only difference between the two proteins is a missing C-terminal extension that is phosphorylated at tyrosine 527 in c-Src (Cooper et al. 1986) (Figure 8). Mutation of Y527 abolished phosphorylation and activated the kinase activity as well as its transformation capacity (Superti-Furga et al. 1993). These early studies identified this phosphorylated C-terminal extension as an autoinhibitory domain. Additional experimental analyses elucidated the basic components of Src autoinhibition and activation. Inhibitory intramolecular interactions involving the C-terminal extension and the SH2 and SH3 domains are replaced by intermolecular interactions with SH2 and SH3 ligands of other proteins (Figure 8B). The involvement of these domains in autoinhibition was suggested by the observation that Src kinase could be activated by deletion or mutation of SH2 and SH3 domains (Hirai & Varmus 1990a,b,c). In addition, these domains likely worked with the C-terminal extension in autoinhibition. However, a key to understanding this mechanism is the ligand specificity of the SH2 and SH3 domains. These domains were initially implicated in protein-protein interactions by binding studies with specific peptide ligands. The SH2 domains bind phospho-tyrosine-containing peptides with a consensus Tyr-Glu-Glu-Leu, whereas SH3 domains bind peptides with Pro-X-X-Pro-X-Arg consensus in an unusual PPII helix conformation. Binding of SH2 and SH3 peptide ligands was shown to activate Src kinase. Furthermore, polypeptides that contained both SH2 and SH3 ligands were demonstrated to be synergistic activators of Src kinase (Alexandropoulos & Baltimore 1996). This synergistic activation suggested that the two domains also cooperate in autoinhibition. A working model emerged upon discovery that the phosphorylated C-terminal extension specifies a cryptic SH2 ligand. Thus these early studies defined the inhibitory elements, implicated intramolecular interactions, and identified a route to counteract the autoinhibition. STRUCTURAL STUDIES PROVIDE A CLEAR PICTURE Biochemical and structural studies led to a more detailed mechanistic model of autoinhibition (Figure 8B) (Hubbard 1999). The phosphorylated C-terminal extension, with its SH2 cryptic ligand sequence, was shown by binding assays to interact intramolecularly with the SH2 domain, and disruption of this interaction by mutagenesis activates the kinase (Liu et al. 1993). Crystal structures of c-Src and Src family member Hck finally showed the full set of intramolecular interactions. The kinase domain of

10 Sep 2002

19:59

450

AR

PUFALL

AR170-CB18-15.tex

¥

AR170-CB18-15.sgm

LaTeX2e(2002/01/18)

P1: IBD

GRAVES

c-Src forms a bilobed structure with the SH2 and SH3 domains extended from the N-terminal kinase lobe and the phospho-tyrosine-containing C-terminal extended from the C-terminal lobe. Satisfyingly, the structures confirmed the intramolecular interaction of the SH2 domain with the C-terminal extension. More surprisingly, the SH3 domain was shown to interact with a cryptic binding motif in the linker between the N-terminal kinase lobe and the SH2 domain. Although this motif contained only one of the two prolines in the consensus Pro-X-X-Pro motif, it adopted the characteristic left handed PPII conformation. This imperfect helix formed a weak pseudoligand for the SH3 domain (Pawson 1997; Sicheri et al. 1997; Xu et al. 1997, 1999). Constraint of the SH2 domain by the phosphorylated C-terminal extension was then shown by mutational analysis to precisely position this imperfect helix for interaction with the SH3 domain (Young et al. 2001). The combination of these two weak pseudoligand interactions thus locks the N- and C-terminal lobes of the kinase domain in a closed position, thereby inhibiting kinase activity (Gonfloni et al. 2000; Hubbard 1999; Superti-Furga & Gonfloni 1997; Xu et al. 1997, 1999; Young et al. 2001). INHIBITION AFFECTS TARGETING The detailed structural model of the inhibited state elucidates the mechanism of autoinhibition of the kinase activity, as well as inhibition of kinase targeting (Figure 8B). The interaction surfaces of the SH2 and SH3 domains are occupied in the inhibited form of the kinase. Mutations engineered in the protein that disrupt these intramolecular interactions, although activating to the kinase activity, also disrupt intermolecular interaction with activating molecules or downstream substrates (Hirai & Varmus 1990a,b,c; Pawson 1997). Disruption of intermolecular interactions with target sequences breaks the signal transduction chain, preventing Src from transforming cells despite the increased kinase activity (Brown & Cooper 1996). The presence of the intramolecular pseudoligands provides exquisite control of targeting and activation. However, the intramolecular interactions of the SH2 and SH3 domains with the non-consensus pseudoligands are weak in the wild-type kinase, and can be disrupted in the presence of a high-affinity ligand (Liu et al. 1993). Similar intramolecular interactions in other Src family members are hypothesized to mediate specificity. Variations in SH2 and SH3 domains of family members bind with high affinity to different sequences than Src, which directs targeting of family members to a different set of substrates (Brown & Cooper 1996).

CONCLUSIONS The seven examples articulated in this review illustrate the central role of autoinhibitory domains in the control of proteins within regulatory pathways. This on-site negative control assures tight regulation; the latent activity can be fully utilized only when activated within a regulatory pathway. In many cases, the autoinhibition phenomenon has led investigators to discover a mode of regulation within the protein. The additional cases cited in Table 1 illustrate the widespread deployment of autoinhibition. The number of entries, although not comprehensive,

10 Sep 2002

19:59

AR

AR170-CB18-15.tex

AR170-CB18-15.sgm

LaTeX2e(2002/01/18)

AUTOINHIBITORY DOMAINS

P1: IBD

451

emphasizes the versatility of this strategy and suggests that autoinhibition plays a role in the regulation of a large number of proteins.

Conformational Change is a Common Feature of Autoinhibition The identification of autoinhibitory domains derives primarily from functional studies; however, structural data are essential for mechanistic modeling. There are three levels of structural data. The first step is defining the structure of the domain that executes the targeted activity as well as the structure of the inhibitory elements. Even secondary structure data of these elements can be useful. Second, the mapping of possible intramolecular interactions between inhibitory elements and the targeted domain provides the proof of the basic phenomenon. Finally, a highresolution three-dimensional description of the inhibited and activated structure provides a complete framework for a mechanistic model of autoinhibition. Several of the examples discussed have substantial structural data. The SNARE protein syntaxin1a has a complete structural model of inhibited and activated states. The ERM proteins and Src kinase have a complete model of the inhibited state. Ets-1, WASP, and σ 70 have structural information for both the inhibited and activated species, although not a complete three-dimensional model. Structural data for many proteins highlight the essential role of conformational change in the autoinhibitory mechanism. In the simplest model, the inhibitory domain sterically masks the active site on the targeted domain, and a conformational change unmasks the active site during activation. Most cases approximate this simple model, showing two juxtaposed domains moving with respect to each other. However, the conformational changes are usually more complex. In several cases we noted conformational flexibility of elements within a single domain. For example, one helix within the inhibitory module of Ets-1 unfolds upon DNA binding. In the case of syntaxin1a, there are two alternate conformations of the targeted domain, one for the inhibited state and one for the SNARE complex. In Src, the inhibitory elements repress kinase activity by allosterically constraining the entire catalytic domain. In WASP, the structure of the inhibitory GDB domain is induced by both the intramolecular and intermolecular interactions. σ 70 shows structural changes within the DNA binding domain and the autoinhibitory domain during activation. These well-characterized examples emphasize the dynamic nature of proteins. These rich conformational repertoires enable a variety of regulatory strategies, autoinhibition being an excellent example.

Autoinhibition Adds a Layer of Regulation The autoinhibitory mechanism is a constitutive damper that serves to repress an activity, safeguarding against inappropriate function, which means that the activity will be executed only in the presence of an activating signal that is sufficiently strong to overcome the inhibition. This review discussed proteins that carry out a wide variety of functions in a cell. In every case, activity requires exquisite control. DNA-binding proteins must localize and bind to appropriate

10 Sep 2002

19:59

452

AR

PUFALL

AR170-CB18-15.tex

¥

AR170-CB18-15.sgm

LaTeX2e(2002/01/18)

P1: IBD

GRAVES

promoters; membrane fusions events must accurately connect cellular organelles to surface membranes; and cytoskeletal elements must expand to meet the needs of the cell while connecting to the membrane. In most cases a route to activation has been identified. Proteolysis is the simple solution as exhibited by SREBP. Ligand binding is the more common route, as seen in many other examples. Nevertheless, each case provides a unique picture. WASP and ERM similarly illustrate how multiple pathways could function synergistically in activation. In ERM proteins, two different activities are inhibited; each inhibited domain binds a ligand, either actin or a membrane-associated protein, upon activation. The SNARE protein syntaxin1a has a clearly defined inhibitory domain, and the only activating ligand is the SNARE complex itself. In several cases, the inhibition is reinforced. Phosphorylation of Ets-1 strengthens the negative regulation of DNA binding by stabilizing the inhibitory domain. An auxiliary protein, n-Sec1, stabilizes the inhibitory intramolecular interactions of the SNARE protein syntaxin. The phosphorylation of the inhibitory domain of Src is essential for autoinhibition. This reinforcement can drive the autoinhibition to an even more secure level of control, and possibly make autoinhibition itself subject to regulation. Autoinhibition not only creates an on-off state but also the potential for selectivity. The need for this activity is most dramatic in cases in which families of related proteins must be assigned to specific tasks. There are multiple types of sigma factors in each bacterium; Ets-1 belongs to a family of related transcription factors; there are related SNARE proteins in all membrane compartments; and ERM, WASP, and Src all bear highly conserved domains that are found in a variety of related proteins. It is proposed that autoinhibition can help distinguish among these related proteins. In the most general model, all factors would be repressed such that a specific signaling mechanism would provide activation. Elucidation of valid, undoubtedly more complex, scenarios await more comparative analyses of related proteins.

A Second Role for the Autoinhibitory Domain The autoinhibitory domain often acquires a new function in the activated molecule. In the case of Ets-1, the activating protein RUNX1 interacts with the inhibitory elements and is a transcriptional regulator. In this way, the intermolecular interactions formed by the autoinhibitory domain stabilize components that play a role in transcriptional regulation. Likewise, in the case of σ 70, the autoinhibitory domain contacts RNA polymerase and participates in the transcription initiation mechanism. The inhibitory elements of WASP bind activating ligands that provide important membrane linkages. The mutually inhibitory domains of the ERM proteins provide binding sites for both ends of the membrane-cytoskeletal crosslink. Finally, the SH2 and SH3 domains of Src can function in the activated state by facilitating docking of Src to substrates with cognate ligands. In each case, one could argue that the second function of the autoinhibitory domain, which is displayed in the activated state, is also masked within the inhibited state by intramolecular

10 Sep 2002

19:59

AR

AR170-CB18-15.tex

AR170-CB18-15.sgm

LaTeX2e(2002/01/18)

AUTOINHIBITORY DOMAINS

P1: IBD

453

interactions. Therefore, the identification of the inhibitory domain and the target of repression is biased by the assay used in the initial characterization. With this perspective one can predict that there are undoubtedly many more examples of autoinhibition to be discovered.

Implications for Human Disease Autoinhibition can be disrupted in disease states. In a genetic disease, a mutated gene could be altered within the region that encodes the inhibitory element or the surface that interacts with the autoinhibitory domain. The mutated gene could direct synthesis of a protein that is constitutively active, having lost the negative control afforded by the intramolecular network. Oncogenic versions of both src and ets-1, which were originally discovered in acutely transforming retroviruses, have such inactivated autoinhibitory domains. In both cases, a C-terminal deletion removes inhibitory elements leading to enhanced activity and providing the oncogenic activity of the avian retroviruses RSV and E-26 virus, respectively (Lim et al. 1992, Sefton & Hunter 1986). A more subtle disruption is implicated in the case of a human disease and WASP. An inherited single-point mutation in the GBD correlates with myeloid disease in the form of severe congenital neutropenia. The position of the mutation suggests that it could result in release of VCA inhibition, leading to constitutive activation of Arp2/3 and mis-regulated actin polymerization (Devriendt et al. 2001). Chromosome translocations are another route to release of autoinhibition. In reciprocal translocations there are cases in which the two breakpoints disrupt two genes, and the translocation creates a new genetic locus that encodes a fusion protein. Many such translocations have been associated with human leukemia (Look 1997). Retention of functional domains from two disparate proteins and their juxtaposition in a new context often leads to a gain-of-function phenotype that is oncogenic. Most relevant to our discussion is the potential loss of the original structural context that might have provided autoinhibition. In this scenario, the retained functional domain would have enhanced activity. The oncogenic fusion protein Pax3-FKHR fits this prediction (Bennicelli et al. 1999). Discovery of other such examples should be possible with more quantitative functional assays. Irrespective of the type of mutation, the general picture within these disease contexts is similar; the mutation alters a regulatory function that is specific to the function of a single protein, resulting in a highly specific disease phenotype that involves gain of function rather than loss of function. Because an autoinhibitory mechanism is specific to the regulation of a particular activity within a unique protein, it is suitable for targeted therapeutics. An autoinhibitory domain could be targeted for either activation or further repression by a small molecule. The drug would be directed to a single protein to intervene with its unique regulatory mechanism. An understanding of the full repertoire of regulatory strategies for a particular protein, including autoinhibition, should facilitate such pharmaceutical research efforts.

10 Sep 2002

19:59

454

AR

PUFALL

AR170-CB18-15.tex

¥

AR170-CB18-15.sgm

LaTeX2e(2002/01/18)

P1: IBD

GRAVES

ACKNOWLEDGMENTS This project is supported in part by the National Institutes of Health with research funding to B.J.G. (R01 GM38663) and fellowship support to M.A.P. (T32 CA93247, GM08537). Support from the Huntsman Cancer Foundation is also gratefully acknowledged. We thank current and past members of the Graves laboratory who have contributed to the Ets-1 autoinhibition story including Jeannine Petersen, Matt Jonsen, Marc Gillespie, Jean Hsu, Mary Nelson and Hong Wang. We also thank our long-standing collaborators on the Ets-1 studies, Nancy Speck (Dartmouth College), Lawrence McIntosh (University of British Columbia), and Tom Alber (UC, Berkeley). We thank University of Utah colleages Mary Beckerle and Janet Shaw for advice on cell biology sections of the review and Valerie Orlemann for assistance with manuscript preparation. The Annual Review of Cell and Developmental Biology is online at http://cellbio.annualreviews.org

LITERATURE CITED Abdul-Manan N, Aghazadeh B, Liu GA, Majumdar A, Ouerfelli O, et al. 1999. Structure of Cdc42 in complex with the GTPasebinding domain of the ‘Wiskott-Aldrich syndrome’ protein. Nature 399:379–83 Aftab DT, Kwan J, Martin GS. 1997. Rasindependent transformation by v-Src. Proc. Natl. Acad. Sci. USA 94:3028–33 Aghazadeh B, Lowry WE, Huang XY, Rosen MK. 2000. Structural basis for relief of autoinhibition of the Dbl homology domain of proto-oncogene Vav by tyrosine phosphorylation. Cell 102:625–33 Alexandropoulos K, Baltimore D. 1996. Coordinate activation of c-Src by SH3- and SH2binding sites on a novel p130Cas-related protein, Sin. Genes Dev. 10:1341–55 Bakolitsa C, de Pereda JM, Bagshaw CR, Critchley DR, Liddington RC. 1999. Crystal structure of the vinculin tail suggests a pathway for activation. Cell 99:603–13 Batchelor A, Piper D, Charles de la Brousse F, McKnight S, Wolberger C. 1998. The structure of GABPα/β: an ETS domain-ankryin repeat heterodimer bound to DNA. Science 279:1037–41 Bennicelli JL, Advani S, Schafer BW, Barr FG. 1999. PAX3 and PAX7 exhibit conserved cis-

acting transcription repression domains and utilize a common gain of function mechanism in alveolar rhabdomyosarcoma. Oncogene 18:4348–56 Bojovic BB, Hassell JA. 2001. The PEA3 Ets transcription factor comprises multiple domains that regulate transactivation and DNA binding. J. Biol. Chem. 276:4509–21 Brass AL, Kehrli E, Eisenbeis CF, Storb U, Singh H. 1996. Pip, a lymphoid-restricted IRF, contains a regulatory domain that is important for autoinhibition and ternary complex formation with the Ets factor PU.1. Genes Dev. 10:2335–47 Bravo J, Li Z, Speck NA, Warren AJ. 2001. The leukemia-associated AML1 (Runx1)CBF beta complex functions as a DNAinduced molecular clamp. Nat. Struct. Biol. 8:371–78 Bretscher A. 1999. Regulation of cortical structure by the ezrin-radixin-moesin protein family. Curr. Opin. Cell Biol. 11:109–16 Bretscher A, Chambers D, Nguyen R, Reczek D. 2000. ERM-Merlin and EBP50 protein families in plasma membrane organization and function. Annu. Rev. Cell Dev. Biol. 16: 113–43 Brown MS, Goldstein JL. 1997. The SREBP

10 Sep 2002

19:59

AR

AR170-CB18-15.tex

AR170-CB18-15.sgm

LaTeX2e(2002/01/18)

AUTOINHIBITORY DOMAINS pathway: regulation of cholesterol metabolism by proteolysis of a membrane-bound transcription factor. Cell 89:331–40 Brown MS, Goldstein JL. 1999. A proteolytic pathway that controls the cholesterol content of membranes, cells, and blood. Proc. Natl. Acad. Sci. USA 96:11041–48 Brown MS, Ye J, Rawson RB, Goldstein JL. 2000. Regulated intramembrane proteolysis: a control mechanism conserved from bacteria to humans. Cell 100:391–98 Brown MT, Cooper JA. 1996. Regulation, substrates and functions of src. Biochim. Biophys. Acta 1287:121–49 Brzeska H, Young R, Tan C, Szczepanowska J, Korn ED. 2001. Calmodulin-binding and autoinhibitory domains of Acanthamoeba myosin I heavy chain kinase, a p21-activated kinase (PAK). J. Biol. Chem. 276:47468–73 Buck M, Xu W, Rosen MK. 2001. Global disruption of the WASP autoinhibited structure on Cdc42 binding. Ligand displacement as a novel method for monitoring amide hydrogen exchange. Biochemistry 40:14115–22 Burgess RR, Anthony L. 2001. How sigma docks to RNA polymerase and what sigma does. Curr. Opin. Microbiol. 4:126–31 Calakos N, Bennett MK, Peterson KE, Scheller RH. 1994. Protein-protein interactions contributing to the specificity of intracellular vesicular trafficking. Science 263:1146–49 Callaci S, Heyduk E, Heyduk T. 1998. Conformational changes of Escherichia coli RNA polymerase sigma70 factor induced by binding to the core enzyme. J. Biol. Chem. 273: 32995–3001 Callaci S, Heyduk E, Heyduk T. 1999. Core RNA polymerase from E. coli induces a major change in the domain arrangement of the sigma 70 subunit. Mol. Cell 3:229–38 Catimel B, Teh T, Fontes MR, Jennings IG, Jans DA, et al. 2001. Biophysical characterization of interactions involving importinalpha during nuclear import. J. Biol. Chem. 276:34189–98 Cegielska A, Gietzen KF, Rivers A, Virshup DM. 1998. Autoinhibition of casein kinase I epsilon (CKI epsilon) is relieved by protein

P1: IBD

455

phosphatases and limited proteolysis. J. Biol. Chem. 273:1357–64 Chan SK, Popperl H, Krumlauf R, Mann RS. 1996. An extradenticle-induced conformational change in a HOX protein overcomes an inhibitory function of the conserved hexapeptide motif. EMBO J. 15:2476–87 Chen H, Bernstein BW, Bamburg JR. 2000. Regulating actin-filament dynamics in vivo. Trends Biochem. Sci. 25:19–23 Chen YA, Scheller RH. 2001. SNAREmediated membrane fusion. Nat. Rev. Mol. Cell Biol. 2:98–106 Cooper JA, Gould KL, Cartwright CA, Hunter T. 1986. Tyr527 is phosphorylated in pp60csrc: implications for regulation. Science 231:1431–34 Cowley DO, Graves BJ. 2000. Phosphorylation represses Ets-1 DNA binding by reinforcing autoinhibition. Genes Dev. 14:366–76 Cutting S, Driks A, Schmidt R, Kunkel B, Losick R. 1991. Forespore-specific transcription of a gene in the signal transduction pathway that governs Pro-sigma K processing in Bacillus subtilis. Genes Dev. 5:456–66 Dajani R, Fraser E, Roe SM, Young N, Good V, et al. 2001. Crystal structure of glycogen synthase kinase 3 beta: structural basis for phosphate-primed substrate specificity and autoinhibition. Cell 105:721–32 DeBose-Boyd RA, Brown MS, Li WP, Nohturfft A, Goldstein JL, Espenshade PJ. 1999. Transport-dependent proteolysis of SREBP: relocation of site-1 protease from Golgi to ER obviates the need for SREBP transport to Golgi. Cell 99:703–12 Devriendt K, Kim AS, Mathijs G, Frints SG, Schwartz M, et al. 2001. Constitutively activating mutation in WASP causes X-linked severe congenital neutropenia. Nat. Genet. 27:313–17 Dickinson LA, Trauger JW, Baird EE, Dervan PB, Graves BJ, Gottesfeld JM. 1999. Inhibition of Ets-1 DNA binding and ternary complex formation between Ets-1, NF-κB and DNA by a designed DNA-binding ligand. J. Biol. Chem. 274:12765–73 Dombroski AJ, Walter WA, Gross CA. 1993.

10 Sep 2002

19:59

456

AR

PUFALL

AR170-CB18-15.tex

¥

AR170-CB18-15.sgm

LaTeX2e(2002/01/18)

P1: IBD

GRAVES

Amino-terminal amino acids modulate sigma-factor DNA-binding activity. Genes Dev. 7:2446–55 Dombroski AJ, Walter WA, Record MT Jr, Siegele DA, Gross CA. 1992. Polypeptides containing highly conserved regions of transcription initiation factor sigma 70 exhibit specificity of binding to promoter DNA. Cell 70:501–12 Donaldson LW, Petersen JM, Graves BJ, McIntosh LP. 1996. Solution structure of the ETS domain from murine Ets-1: a winged helix-turn-helix DNA binding motif. EMBO J. 15:125–34 Dulubova I, Sugita S, Hill S, Hosaka M, Fernandez I, et al. 1999. A conformational switch in syntaxin during exocytosis: role of munc18. EMBO J. 18:4372–82 ˚ crystal Edwards SD, Keep NH. 2001. The 2.7 A structure of the activated FERM domain of moesin: an analysis of structural changes on activation. Biochemistry 40:7061–68 Fedorov AA, Fedorov E, Gertler F, Almo SC. 1999. Structure of EVH1, a novel prolinerich ligand-binding module involved in cytoskeletal dynamics and neural function. Nat. Struct. Biol. 6:661–65 Fong YW, Zhou Q. 2000. Relief of two builtin autoinhibitory mechanisms in P-TEFb is required for assembly of a multicomponent transcription elongation complex at the human immunodeficiency virus type 1 promoter. Mol. Cell. Biol. 20:5897–907 Francis SH, Poteet-Smith C, Busch JL, RichieJannetta R, Corbin JD. 2002. Mechanisms of autoinhibition in cyclic nucleotidedependent protein kinases. Front. Biosci. 7:D580–92 Fukuoka M, Suetsugu S, Miki H, Fukami K, Endo T, Takenawa T. 2001. A novel neural Wiskott-Aldrich syndrome protein (N-WASP) binding protein, WISH, induces Arp2/3 complex activation independent of Cdc42. J. Cell Biol. 152:471–82 Garvie CW, Hagman J, Wolberger C. 2001. Structural studies of Ets-1/Pax5 complex formation on DNA. Mol. Cell 8:1267–76 Gary R, Bretscher A. 1995. Ezrin self-

association involves binding of an N-terminal domain to a normally masked C-terminal domain that includes the F-actin binding site. Mol. Biol. Cell 6:1061–75 Gautreau A, Louvard D, Arpin M. 2000. Morphogenic effects of ezrin require a phosphorylation-induced transition from oligomers to monomers at the plasma membrane. J. Cell Biol. 150:193–203 Gautreau A, Louvard D, Arpin M. 2002. ERM proteins and NF2 tumor suppressor: the Yin and Yang of cortical actin organization and cell growth signaling. Curr. Opin. Cell Biol. 14:104–9 Goetz TL, Gu TL, Speck NA, Graves BJ. 2000. Auto-inhibition of Ets-1 is counteracted by DNA binding cooperativity with core-binding factor alpha2. Mol. Cell. Biol. 20:81–90 Gonfloni S, Weijland A, Kretzschmar J, Superti-Furga G. 2000. Crosstalk between the catalytic and regulatory domains allows bidirectional regulation of Src. Nat. Struct. Biol. 7:281–86 Gonzalez LC Jr, Weis WI, Scheller RH. 2001. A novel snare N-terminal domain revealed by the crystal structure of Sec22b. J. Biol. Chem. 276:24203–11 Graves BJ, Cowley DO, Goetz TL, Petersen JM, Jonsen MD, Gillespie ME. 1998. Autoinhibition as a transcriptional regulatory mechanism. In Cold Spring Harbor Symposia on Quantitative Biology LXIII, ed. B Stillman, pp. 621–29. Cold Spring Harbor, NY: Cold Spring Harbor Press Graves BJ, Petersen JM. 1998. Specificity within the ets family of transcription factors. In Advances in Cancer Research, ed. GV Woude, G Klein, pp. 1–55. San Diego: Academic Greenall A, Willingham N, Cheung E, Boam DS, Sharrocks AD. 2001. DNA binding by the ETS-domain transcription factor PEA3 is regulated by intramolecular and intermolecular protein-protein interactions. J. Biol. Chem. 17:17 Gross CA, Chan C, Dombroski A, Gruber T, Sharp M, et al. 1998. The functional and

10 Sep 2002

19:59

AR

AR170-CB18-15.tex

AR170-CB18-15.sgm

LaTeX2e(2002/01/18)

AUTOINHIBITORY DOMAINS regulatory roles of sigma factors in transcription. Cold Spring Harbor Symp. Quant. Biol. 63:141–55 Gruber TM, Markov D, Sharp MM, Young BA, Lu CZ, et al. 2001. Binding of the initiation factor sigma(70) to core RNA polymerase is a multistep process. Mol. Cell 8:21–31 Gu TL, Goetz TL, Graves BJ, Speck NA. 2000. Auto-inhibition and partner proteins, core-binding factor beta (CBFbeta) and Ets-1, modulate DNA binding by CBFalpha2 (AML1). Mol. Cell. Biol. 20:91–103 Hagman J, Grosschedl R. 1992. An inhibitory carboxyl terminal domain in Ets-1 and Ets-2 mediates differential binding of ETS family factors to promoter sequences of the mb-1 gene. Proc. Natl. Acad. Sci. USA 89:8889– 93 Hahne P, Sechi A, Benesch S, Small JV. 2001. Scar/WAVE is localised at the tips of protruding lamellipodia in living cells. FEBS Lett. 492:215–20 Hamada K, Shimizu T, Matsui T, Tsukita S, Hakoshima T. 2000. Structural basis of the membrane-targeting and unmasking mechanisms of the radixin FERM domain. EMBO J. 19:4449–62 Haze K, Yoshida H, Yanagi H, Yura T, Mori K. 1999. Mammalian transcription factor ATF6 is synthesized as a transmembrane protein and activated by proteolysis in response to endoplasmic reticulum stress. Mol. Biol. Cell 10:3787–99 Henry MD, Gonzalez Agosti C, Solomon F. 1995. Molecular dissection of radixin: distinct and interdependent functions of the amino- and carboxy-terminal domains. J. Cell Biol. 129:1007–22 Higgs HN, Pollard TD. 2000. Activation by Cdc42 and PIP(2) of Wiskott-Aldrich syndrome protein (WASp) stimulates actin nucleation by Arp2/3 complex. J. Cell Biol. 150:1311–20 Higgs HN, Pollard TD. 2001. Regulation of actin filament network formation through ARP2/3 complex: activation by a diverse array of proteins. Annu. Rev. Biochem. 70:649– 76

P1: IBD

457

Hirai H, Varmus HE. 1990a. Mutations in src homology regions 2 and 3 of activated chicken c-src that result in preferential transformation of mouse or chicken cells. Proc. Natl. Acad. Sci. USA 87:8592–96 Hirai H, Varmus HE. 1990b. SH2 mutants of c-src that are host dependent for transformation are trans-dominant inhibitors of mouse cell transformation by activated c-src. Genes Dev. 4:2342–52 Hirai H, Varmus HE. 1990c. Site-directed mutagenesis of the SH2- and SH3-coding domains of c-src produces varied phenotypes, including oncogenic activation of p60c-src. Mol. Cell. Biol. 10:1307–18 Hirao M, Sato N, Kondo T, Yonemura S, Monden M, et al. 1996. Regulation mechanism of ERM (ezrin/radixin/moesin) protein/plasma membrane association: possible involvement of phosphatidylinositol turnover and Rhodependent signaling pathway. J. Cell Biol. 135:37–51 Hubbard SR. 1999. Src autoinhibition: let us count the ways. Nat. Struct. Biol. 6:711– 14 Janosik M, Kery V, Gaustadnes M, Maclean KN, Kraus JP. 2001. Regulation of human cystathionine beta-synthase by S-adenosylL-methionine: evidence for two catalytically active conformations involving an autoinhibitory domain in the C-terminal region. Biochemistry 40:10625–33 Jiang R, Carlson M. 1997. The Snf1 protein kinase and its activating subunit, Snf4, interact with distinct domains of the Sip1/Sip2/Gal83 component in the kinase complex. Mol. Cell. Biol. 17:2099–106 Johnson RP, Craig SW. 2000. Actin activates a cryptic dimerization potential of the vinculin tail domain. J. Biol. Chem. 275:95–105 Jonsen MD, Petersen JM, Xu QP, Graves BJ. 1996. Characterization of the cooperative function of inhibitory sequences in Ets-1. Mol. Cell. Biol. 16:2065–73 Khan AR, James MN. 1998. Molecular mechanisms for the conversion of zymogens to active proteolytic enzymes. Protein Sci. 7:815–36

10 Sep 2002

19:59

458

AR

PUFALL

AR170-CB18-15.tex

¥

AR170-CB18-15.sgm

LaTeX2e(2002/01/18)

P1: IBD

GRAVES

Kim AS, Kakalis LT, Abdul-Manan N, Liu GA, Rosen MK. 2000. Autoinhibition and activation mechanisms of the Wiskott-Aldrich syndrome protein. Nature 404:151–58 Kim W-Y, Sieweke M, Ogawa D, Wee H-J, Englemeier U, et al. 1999. Mutual activation of Ets-1 and AML1 DNA binding by direct interaction of their autoinhibitory domains. EMBO J. 18:1609–20 Ko LJ, Prives C. 1996. p53: Puzzle and paradigm. Genes Dev. 10:1054–72 Kobe B. 1999. Autoinhibition by an internal nuclear localization signal revealed by the crystal structure of mammalian importin alpha. Nat. Struct. Biol. 6:388–97 Kodandapani R, Pio F, Ni C-Z, Piccialli G, Klemsz M, et al. 1996. A new pattern for helix-turn-helix recognition revealed by the PU.1 ETS-domain-DNA complex. Nature 380:456–60 Krimm I, Goyer A, Issakidis-Bourguet E, Miginiac-Maslow M, Lancelin JM. 1999. Direct NMR observation of the thioredoxinmediated reduction of the chloroplast NADPmalate dehydrogenase provides a structural basis for the relief of autoinhibition. J. Biol. Chem. 274:34539–42 Kroos L, Kunkel B, Losick R. 1989. Switch protein alters specificity of RNA polymerase containing a compartment-specific sigma factor. Science 243:526–29 Kuddus R, Schmidt MC. 1993. Effect of the non-conserved N-terminus on the DNA binding activity of the yeast TATA binding protein. Nucleic Acids Res. 21:1789– 96 Kuriyan J, Cowburn D. 1997. Modular peptide recognition domains in eukaryotic signaling. Annu. Rev. Biophys. Biomol. Struct. 26:259– 88 Lefstin JA, Yamamoto KR. 1998. Allosteric effects of DNA on transcriptional regulators. Nature 392:885–88 Lesley SA, Burgess RR. 1989. Characterization of the Escherichia coli transcription factor sigma 70: localization of a region involved in the interaction with core RNA polymerase. Biochemistry 28:7728–34

Li M, Baumeister P, Roy B, Phan T, Foti D, et al. 2000. ATF6 as a transcription activator of the endoplasmic reticulum stress element: thapsigargin stress-induced changes and synergistic interactions with NF-Y and YY1. Mol. Cell. Biol. 20:5096–106 Lim F, Kraut N, Frampton J, Graf T. 1992. DNA binding by c-Ets-1, but not v-Ets, is repressed by an intramolecular mechanism. EMBO J. 11:643–52 Lin R, Mamane Y, Hiscott J. 1999. Structural and functional analysis of interferon regulatory factor 3: localization of the transactivation and autoinhibitory domains. Mol. Cell. Biol. 19:2465–74 Liu X, Brodeur SR, Gish G, Songyang Z, Cantley LC, et al. 1993. Regulation of c-Src tyrosine kinase activity by the Src SH2 domain. Oncogene 8:1119–26 Look AT. 1997. Oncogenic transcription factors in the human acute leukemias. Science 278:1059–64 Machesky LM, Mullins RD, Higgs HN, Kaiser DA, Blanchoin L, et al. 1999. Scar, a WASprelated protein, activates nucleation of actin filaments by the Arp2/3 complex. Proc. Natl. Acad. Sci. USA 96:3739–44 Mangeat P, Roy C, Martin M. 1999. ERM proteins in cell adhesion and membrane dynamics. Trends Cell Biol. 9:187–92 Martin M, Andreoli C, Sahuquet A, Montcourrier P, Algrain M, Mangeat P. 1995. Ezrin NH2-terminal domain inhibits the cell extension activity of the COOH-terminal domain. J. Cell Biol. 128:1081–93 Matsui T, Maeda M, Doi Y, Yonemura S, Amano M, et al. 1998. Rho-kinase phosphorylates COOH-terminal threonines of ezrin/radixin/moesin (ERM) proteins and regulates their head-to-tail association. J. Cell Biol. 140:647–57 Mekler V, Kortkhonjia E, Mukhopadhyay J, Knight J, Revyakin A, et al. 2002. Structural organization of bacterial RNA polymerase holoenzyme and the RNA polymerasepromoter open complex. Cell 108:599–614 Miki H, Sasaki T, Takai Y, Takenawa T. 1998. Induction of filopodium formation by a

10 Sep 2002

19:59

AR

AR170-CB18-15.tex

AR170-CB18-15.sgm

LaTeX2e(2002/01/18)

AUTOINHIBITORY DOMAINS WASP-related actin-depolymerizing protein N-WASP. Nature 391:93–96 Misura KM, May AP, Weis WI. 2000a. Protein-protein interactions in intracellular membrane fusion. Curr. Opin. Struct. Biol. 10:662–71 Misura KM, Scheller RH, Weis WI. 2000b. Three-dimensional structure of the neuronalSec1-syntaxin 1a complex. Nature 404:355– 62 Mittal V, Hernandez N. 1997. Role for the amino-terminal region of human TBP in U6 snRNA transcription. Science 275:1136–40 Mittal V, Ma B, Hernandez N. 1999. SNAP(c): a core promoter factor with a built-in DNAbinding damper that is deactivated by the Oct-1 POU domain. Genes Dev. 13:1807– 21 Mo Y, Vaessen B, Johnston K, Marmorstein R. 1998. Structures of SAP-1 bound to DNA targets from the E74 and c-fos promoters: insights into DNA sequence discrimination by Ets proteins. Mol. Cell 2:201–12 Mo Y, Vaessen B, Johnston K, Marmorstein R. 2000. Structure of the Elk-1-DNA complex reveals how DNA-distal residues affect ETS domain recognition of DNA. Nat. Struct. Biol. 7:292–97 Moir RD, Puglia KV, Willis IM. 2002. Autoinhibition of TFIIIB70 binding by the tetratricopeptide repeat-containing subunit of TFIIIC. J. Biol. Chem. 277:694–701 Moran MF, Koch CA, Anderson D, Ellis C, England L, et al. 1990. Src homology region 2 domains direct protein-protein interactions in signal transduction. Proc. Natl. Acad. Sci. USA 87:8622–26 Mumm JS, Kopan R. 2000. Notch signaling: from the outside in. Dev. Biol. 228:151–65 Munson M, Chen X, Cocina AE, Schultz SM, Hughson FM. 2000. Interactions within the yeast t-SNARE Sso1p that control SNARE complex assembly. Nat. Struct. Biol. 7:894– 902 Nakamura F, Huang L, Pestonjamasp K, Luna EJ, Furthmayr H. 1999. Regulation of F-actin binding to platelet moesin in vitro by both phosphorylation of threonine 558 and

P1: IBD

459

polyphosphatidylinositides. Mol. Biol. Cell 10:2669–85 Naryshkin N, Revyakin A, Kim Y, Mekler V, Ebright RH. 2000. Structural organization of the RNA polymerase-promoter open complex. Cell 101:601–11 Nguyen R, Reczek D, Bretscher A. 2001. Hierarchy of merlin and ezrin N- and C-terminal domain interactions in homo- and heterotypic associations and their relationship to binding of scaffolding proteins EBP50 and E3KARP. J. Biol. Chem. 276:7621–29 Nicholson KL, Munson M, Miller RB, Filip TJ, Fairman R, Hughson FM. 1998. Regulation of SNARE complex assembly by an N-terminal domain of the t-SNARE Sso1p. Nat. Struct. Biol. 5:793–802 Niggli V, Andreoli C, Roy C, Mangeat P. 1995. Identification of a phosphatidylinositol4,5–bisphosphate-binding domain in the N-terminal region of ezrin. FEBS Lett. 376:172–76 Nohturfft A, DeBose-Boyd RA, Scheek S, Goldstein JL, Brown MS. 1999. Sterols regulate cycling of SREBP cleavage-activating protein (SCAP) between endoplasmic reticulum and Golgi. Proc. Natl. Acad. Sci. USA 96:11235–40 Okamura H, Aramburu J, Garcia-Rodriguez C, Viola JP, Raghavan A, et al. 2000. Concerted dephosphorylation of the transcription factor NFAT1 induces a conformational switch that regulates transcriptional activity. Mol. Cell 6:539–50 Parker RC, Varmus HE, Bishop JM. 1984. Expression of v-src and chicken c-src in rat cells demonstrates qualitative differences between pp60v-src and pp60c-src. Cell 37:131–39 Pawson T. 1997. New impressions of Src and Hck. Nature 385:582–83, 25 Pearson MA, Reczek D, Bretscher A, Karplus PA. 2000. Structure of the ERM protein moesin reveals the FERM domain fold masked by an extended actin binding tail domain. Cell 101:259–70 Pelham HR. 2001. SNAREs and the specificity of membrane fusion. Trends Cell Biol. 11:99–101

10 Sep 2002

19:59

460

AR

PUFALL

AR170-CB18-15.tex

¥

AR170-CB18-15.sgm

LaTeX2e(2002/01/18)

P1: IBD

GRAVES

Pestonjamasp K, Amieva MR, Strassel CP, Nauseef WM, Furthmayr H, Luna EJ. 1995. Moesin, ezrin, and p205 are actinbinding proteins associated with neutrophil plasma membranes. Mol. Biol. Cell 6:247– 59 Petersen JM, Skalicky JJ, Donaldson LW, McIntosh LP, Alber T, Graves BJ. 1995. Modulation of transcription factor Ets-1 DNA binding: DNA-induced unfolding of an alpha helix. Science 269:1866–69 Prehoda KE, Scott JA, Mullins RD, Lim WA. 2000. Integration of multiple signals through cooperative regulation of the N-WASP-Arp2/3 complex. Science 290: 801–6 Price MA, Rogers AE, Treisman R. 1995. Comparative analysis of the ternary complex factors Elk-1, SAP-1a and SAP-2 (ERP/NET). EMBO J. 14:2589–601 Rabault B, Ghysdael J. 1994. Calcium-induced phosphorylation of ETS1 inhibits its specific DNA binding activity. J. Biol. Chem. 269:28143–51 Rawson RB, Zelenski NG, Nijhawan D, Ye J, Sakai J, et al. 1997. Complementation cloning of S2P, a gene encoding a putative metalloprotease required for intramembrane cleavage of SREBPs. Mol. Cell 1:47– 57 Reczek D, Bretscher A. 1998. The carboxylterminal region of EBP50 binds to a site in the amino-terminal domain of ezrin that is masked in the dormant molecule. J. Biol. Chem. 273:18452–58 Richmond JE, Weimer RM, Jorgensen EM. 2001. An open form of syntaxin bypasses the requirement for UNC-13 in vesicle priming. Nature 412:338–41 Robinson RC, Turbedsky K, Kaiser DA, Marchand JB, Higgs HN, et al. 2001. Crystal structure of Arp2/3 complex. Science 294:1679– 84 Rohatgi R, Ho HY, Kirschner MW. 2000. Mechanism of N-WASP activation by CDC42 and phosphatidylinositol 4,5-bisphosphate. J. Cell Biol. 150:1299–310 Rohatgi R, Ma L, Miki H, Lopez M, Kirch-

hausen T, et al. 1999. The interaction between N-WASP and the Arp2/3 complex links Cdc42-dependent signals to actin assembly. Cell 97:221–31 Rohatgi R, Nollau P, Ho HY, Kirschner MW, Mayer BJ. 2001. Nck and phosphatidylinositol 4,5-bisphosphate synergistically activate actin polymerization through the N-WASPArp2/3 pathway. J. Biol. Chem. 276:26448– 52 Rudner DZ, Losick R. 2002. A sporulation membrane protein tethers the pro-sigmaK processing enzyme to its inhibitor and dictates its subcellular localization. Genes Dev. 16:1007–18 Rudolph MG, Bayer P, Abo A, Kuhlmann J, Vetter IR, Wittinghofer A. 1998. The Cdc42/Rac interactive binding region motif of the Wiskott Aldrich syndrome protein (WASP) is necessary but not sufficient for tight binding to Cdc42 and structure formation. J. Biol. Chem. 273:18067–76 Rutter J, Michnoff CH, Harper SM, Gardner KH, McKnight SL. 2001. PAS kinase: an evolutionarily conserved PAS domainregulated serine/threonine kinase. Proc. Natl. Acad. Sci. USA 98:8991–96 Santagata S, Boggon TJ, Baird CL, Gomez CA, Zhao J, et al. 2001. G-protein signaling through tubby proteins. Science 292:2041– 50 Sefton BM, Hunter T. 1986. From c-src to vsrc, or the case of the missing C terminus. Cancer Surv. 5:159–72 Sharp MM, Chan CL, Lu CZ, Marr MT, Nechaev S, et al. 1999. The interface of sigma with core RNA polymerase is extensive, conserved, and functionally specialized. Genes Dev. 13:3015–26 Sharrocks AD. 2001. The ETS-domain transcription factor family. Nat. Rev. Mol. Cell Biol. 2:827–37 Sheridan PL, Sheline CT, Cannon K, Voz ML, Pazin MJ, et al. 1995. Activation of the HIV1 enhancer by the LEF-1 HMG protein on nucleosome assembled DNA in vitro. Genes Dev. 9:2090–104 Sicheri F, Moarefi I, Kuriyan J. 1997. Crystal

10 Sep 2002

19:59

AR

AR170-CB18-15.tex

AR170-CB18-15.sgm

LaTeX2e(2002/01/18)

AUTOINHIBITORY DOMAINS structure of the Src family tyrosine kinase Hck. Nature 385:602–9 Sieweke MH, Tekotte H, Jarosch U, Graf T. 1998. Cooperative interaction of Ets-1 with USF-1 required for HIV-1 enhancer activity in T cells. EMBO J. 17:1728–39 Simons PC, Pietromonaco SF, Reczek D, Bretscher A, Elias L. 1998. C-terminal threonine phosphorylation activates ERM proteins to link the cell’s cortical lipid bilayer to the cytoskeleton. Biochem. Biophys. Res. Commun. 253:561–65 Skalicky JJ, Donaldson LW, Petersen JM, Graves BJ, McIntosh LP. 1996. Structural coupling of the inhibitory regions flanking the ETS domain of murine Ets-1. Protein Sci. 5:296–309 Stehelin D, Varmus HE, Bishop JM, Vogt PK. 1976. DNA related to the transforming gene(s) of avian sarcoma viruses is present in normal avian DNA. Nature 260:170–73 Stoven S, Ando I, Kadalayil L, Engstrom Y, Hultmark D. 2000. Activation of the Drosophila NF-kappaB factor Relish by rapid endoproteolytic cleavage. EMBO Rep. 1:347– 52. Superti-Furga G, Fumagalli S, Koegl M, Courtneidge SA, Draetta G. 1993. Csk inhibition of c-Src activity requires both the SH2 and SH3 domains of Src. EMBO J. 12:2625–34 Superti-Furga G, Gonfloni S. 1997. A crystal milestone: the structure of regulated Src. BioEssays 19:447–50 Sutton RB, Fasshauer D, Jahn R, Brunger AT. 1998. Crystal structure of a SNARE complex ˚ resinvolved in synaptic exocytosis at 2.4 A olution. Nature 395:347–53 Tan I, Seow KT, Lim L, Leung T. 2001. Intermolecular and intramolecular interactions regulate catalytic activity of myotonic dystrophy kinase-related Cdc42-binding kinase alpha. Mol. Cell. Biol. 21:2767–78 Tan KO, Tan KM, Yu VC. 1999. A novel BH3-like domain in BID is required for intramolecular interaction and autoinhibition of pro-apoptotic activity. J. Biol. Chem. 274:23687–90 Tochio H, Tsui MM, Banfield DK, Zhang M.

P1: IBD

461

2001. An autoinhibitory mechanism for nonsyntaxin SNARE proteins revealed by the structure of Ykt6p. Science 293:698–702 Tokoyoda K, Takemoto Y, Nakayama T, Arai T, Kubo M. 2000. Synergism between the calmodulin-binding and autoinhibitory domains on calcineurin is essential for the induction of their phosphatase activity. J. Biol. Chem. 275:11728–34 Tu H, Wigler M. 1999. Genetic evidence for Pak1 autoinhibition and its release by Cdc42. Mol. Cell. Biol. 19:602–11 Turunen O, Wahlstrom T, Vaheri A. 1994. Ezrin has a COOH-terminal actin-binding site that is conserved in the ezrin protein family. J. Cell Biol. 126:1445–53 Volkmann N, Amann KJ, Stoilova-McPhie S, Egile C, Winter DC, et al. 2001. Structure of Arp2/3 complex in its activated state and in actin filament branch junctions. Science 293:2456–59 Vuthoori S, Bowers CW, McCracken A, Dombroski AJ, Hinton DM. 2001. Domain 1.1 of the sigma(70) subunit of Escherichia coli RNA polymerase modulates the formation of stable polymerase/promoter complexes. J. Mol. Biol. 309:561–72 Wang D, Zheng F, Holmberg S, Kohlhaw GB. 1999. Yeast transcriptional regulator Leu3p. Self-masking, specificity of masking, and evidence for regulation by the intracellular level of Leu3p. J. Biol. Chem. 274:19017–24 Wang H, McIntosh LP, Graves BJ. 2001. Inhibitory module of Ets-1 allosterically regulates DNA binding through a dipolefacilitated phosphate contact. J. Biol. Chem. 277:2225–33 Waters MG, Pfeffer SR. 1999. Membrane tethering in intracellular transport. Curr. Opin. Cell Biol. 11:453–59 Wear MA, Schafer DA, Cooper JA. 2000. Actin dynamics: assembly and disassembly of actin networks. Curr. Biol. 10:R891–95 Wilson C, Dombroski AJ. 1997. Region 1 of sigma70 is required for efficient isomerization and initiation of transcription by Escherichia coli RNA polymerase. J. Mol. Biol. 267:60–74

10 Sep 2002

19:59

462

AR

PUFALL

AR170-CB18-15.tex

¥

AR170-CB18-15.sgm

LaTeX2e(2002/01/18)

P1: IBD

GRAVES

Wu MN, Littleton JT, Bhat MA, Prokop A, Bellen HJ. 1998. ROP, the Drosophila Sec1 homolog, interacts with syntaxin and regulates neurotransmitter release in a dosagedependent manner. EMBO J. 17:127–39 Wybenga-Groot LE, Baskin B, Ong SH, Tong J, Pawson T, Sicheri F. 2001. Structural basis for autoinhibition of the Ephb2 receptor tyrosine kinase by the unphosphorylated juxtamembrane region. Cell 106:745–57 Xu W, Doshi A, Lei M, Eck MJ, Harrison SC. 1999. Crystal structures of c-Src reveal features of its autoinhibitory mechanism. Mol. Cell 3:629–38 Xu W, Harrison SC, Eck MJ. 1997. Threedimensional structure of the tyrosine kinase c-Src. Nature 385:595–602 Yan YT, Stein SM, Ding J, Shen MM, AbateShen C. 2000. A novel PF/PN motif inhibits nuclear localization and DNA binding activity of the ESX1 homeoprotein. Mol. Cell. Biol. 20:661–71 Ye J, Rawson RB, Komuro R, Chen X, Dave UP, et al. 2000. ER stress induces cleavage of membrane-bound ATF6 by the same proteases that process SREBPs. Mol. Cell 6:1355–64 Yoshida H, Haze K, Yanagi H, Yura T, Mori K. 1998. Identification of the cis-acting endoplasmic reticulum stress response element responsible for transcriptional induction of mammalian glucose-regulated proteins. Involvement of basic leucine zipper transcription factors. J. Biol. Chem. 273:33741–49

Yoshida H, Matsui T, Yamamoto A, Okada T, Mori K. 2001. XBP1 mRNA is induced by ATF6 and spliced by IRE1 in response to ER stress to produce a highly active transcription factor. Cell 107:881–91 Young MA, Gonfloni S, Superti-Furga G, Roux B, Kuriyan J. 2001. Dynamic coupling between the SH2 and SH3 domains of cSrc and Hck underlies their inactivation by C-terminal tyrosine phosphorylation. Cell 105:115–26 Young P, Gautel M. 2000. The interaction of titin and alpha-actinin is controlled by a phospholipid-regulated intramolecular pseudoligand mechanism. EMBO J. 19:6331–40 Zalevsky J, Grigorova I, Mullins RD. 2001a. Activation of the Arp2/3 complex by the Listeria acta protein. Acta binds two actin monomers and three subunits of the Arp2/3 complex. J. Biol. Chem. 276:3468–75 Zalevsky J, Lempert L, Kranitz H, Mullins RD. 2001b. Different WASP family proteins stimulate different Arp2/3 complexdependent actin-nucleating activities. Curr. Biol. 11:1903–13 Zhao X, Herr W. 2002. A regulated twostep mechanism of TBP binding to DNA: A solvent-exposed surface of TBP inhibits TATA box recognition. Cell 108:615– 27 Zhong H, Voll RE, Ghosh S. 1998. Phosphorylation of NF-kappa B p65 by PKA stimulates transcriptional activity by promoting a novel bivalent interaction with the coactivator CBP/p300. Mol. Cell 1:661–71