WASH Protein Structure
In addition to the conserved C-terminal VCA domain found in all WASP superfamily members, WASH also contains a proline-rich domain (PRD) located adjacent to the VCA, as well as a distinct N-terminal regulatory region (Linardopoulou et al. 2007; Rottner et al. 2010). As depicted in Fig. 1, this unique N-terminus of WASH is comprised of the WASH homology domain 1 (WAHD1) followed by the tubulin-binding region (TBR), and these two domains are collectively referred to as the WASH homology domain (WAHD) due to conservation among WASH orthologs (Gomez and Billadeau 2009). The WAHD1 is critical for assembly into the WASH Regulatory Complex (SHRC) and will be discussed in detail below. WASH was found to co-immunoprecipitate with tubulin following T cell receptor ligation and the TBR was identified to directly interact with microtubule dimers in vitro (Gomez and Billadeau 2009). Significantly, Drosophila WASH has been shown to bundle microtubules and mammalian WASH has been reported to interact with g-tubulin (Liu et al. 2009; Monfregola et al. 2010). Unlike other NPF family members, WASH demonstrates endosomal localization in a variety of cell lines (Derivery et al. 2009; Duleh and Welch 2010; Gomez and Billadeau 2009). The N-terminal WAHD1 of WASH mediates its targeting to endosomes via interaction with the the SHRC member, FAM21 (Family number 21 (WASH Complex Subunit FAM21C)) (Fig. 1) (Gomez and Billadeau 2009). Currently, the PRD of WASH has not been linked to any specific functions, but the PRDs of other WASP superfamily members have been reported to be responsible for binding to a number of SH3 (SRC Homology 3 Domain)-containing proteins (Stradal et al. 2004). However, the proline-rich region of WASH does not harbor any canonical type I or type II SH3-binding motifs. Importantly, these domains appear to be conserved throughout most of the currently identified WASH homologs suggesting that all of these domains may play an evolutionarily important role in WASH function (Veltman and Insall 2010).
WASH Regulatory Complex
While the WASH complex also interacts with the actin capping protein CapZ, it currently appears that CapZ is a peripheral component of the complex, as the stability of CapZ is not subject to the expression of the core SHRC members. However, the core complex members do exhibit structural interdependency for their stability (Derivery et al. 2009; Gomez and Billadeau 2009; Jia et al. 2010). The C-terminus of the SHRC member FAM21 has been shown to bind to CapZ via a capping protein interaction motif, while FAM21 itself assembles directly into the SHRC via its N-terminus (Fig. 2) (Gomez and Billadeau 2009; Jia et al. 2010). Importantly, expression of a FAM21 mutant that could not interact with CapZ was still able to organize the SHRC and localize to endosomes (Gomez and Billadeau 2009; Jia et al. 2010). This suggests that CapZ is a peripherally associated member of the SHRC that might be critical for the activities of the SHRC at endosomal structures. While the function of each individual member of the SHRC is yet to be elucidated, it is important to note that suppression of individual complex members by RNA (Ribonucleic Acid) interference in HeLa cells results in downregulation of other complex components, with suppression of FAM21, Strumpellin, and SWIP affecting all complex components, while suppression of WASH or CCDC53 results in the downregulation of each other with only limited effect on the stability of the other members of the complex (Jia et al. 2010). However, it should be noted that mouse embryonic fibroblasts and T cells deficient in WASH show marked depletion of all complex members (J.T. Piotrowski, T.S. Gomez and D.D. Billadeau, unpublished observation).
Activation of Arp2/3 by WASP and WAVE is constitutively inhibited by either auto-inhibition (WASP) or by assembly of WAVE into the pentameric WRC (WAVE Regulatory Complex) (Ramesh and Geha 2009; Stradal and Scita 2006). As would be expected, recombinant full length WASH, which is highly unstable and aggregates in solution, or the C-terminal VCA domain was found to strongly induce Arp2/3-mediated actin polymerization (Derivery et al. 2009; Duleh and Welch 2010; Linardopoulou et al. 2007; Liu et al. 2009). However, it has been subsequently shown that assembly of WASH into the recombinant SHRC abrogates WASH-mediated Arp2/3 actin polymerization suggesting that like other WASP superfamily members, WASH is constitutively inhibited (Fig. 2) (Jia et al. 2010). While it is currently unclear how the actin polymerization activity of WASH is activated in vivo, it appears that the complex may exist in both active and inactive conformations as the denatured SHRC allowed WASH-mediated Arp2/3 actin polymerization to proceed in vitro (Jia et al. 2010).
Significantly, the SHRC is composed of proteins that have been subject to limited investigation. However, it has been recently demonstrated that the SHRC components contain significant structural homology to members of the WRC with similarity found between SHRC/WRC members: SWIP/SRA1 (Specifically Rac-associated protein 1), Strumpellin/NAP1, and helical regions within the N-termini of WASH and WAVE, as well as CCDC53 and HSPC300 (Heat-Shock Protein C300) (Jia et al. 2010). No significant structural conservation was found between the remaining SHRC member FAM21 and the WRC member Abi1/2 (Jia et al. 2010). However, it should be noted that both members contain highly helical N-terminal regions, which are necessary for complex assembly (Jia et al. 2010). Importantly, it was found that the conserved region in WASH was required for SHRC assembly and the structurally conserved region in CCDC53 was both necessary and sufficient for complex stability. In the same report, it was also found via electron microscopy that the WASH complex and the WAVE complex take on similar structures suggesting the possibility that they assemble in a similar fashion (Jia et al. 2010). Furthermore, the previous structural predictions of WRC assembly were recently confirmed by analysis of the actual crystal structure of the WRC, which demonstrated that the two large highly helical proteins, Sra1 and Nap1, interact and form an elongated pseudo-symmetric dimer forming a platform upon which a trimer containing WAVE1, Abi2, and HSPC300 assembles (Chen et al. 2010). Importantly, the two structurally conserved helical regions of WAVE1 and HSPC300, found in WASH and CCDC53 respectively, interact with two helices of Abi2, as well as making extensive interactions with Sra1. Mutation of key residues aimed at disrupting this four helical bundle destabilizes the entire WRC (Chen et al. 2010). Lastly, the structure of the WRC indicates that the VCA of WAVE1 makes significant contacts with Sra1 as well as intramolecular associations resulting in inhibited activity toward Arp2/3 (Chen et al. 2010). Although a high-resolution structure of the SHRC remains to be made, these many structural similarities, coupled with the overall similar shape and size of the SHRC and the WRC, suggest that the SHRC is analogously organized as a large SWIP:Strumpellin platform upon which a helical bundle consisting of the N-termini of WASH:CCDC53:FAM21 assemble, with the WASH VCA sequestered by a similar mechanism (Fig. 2).
WASH Localization and Function
It was initially demonstrated that Drosophila wash was required for viability and that flies lacking wash were unable to reach the prepupal stage (Linardopoulou et al. 2007). Drosophila WASH has been further shown to bundle and crosslink both actin filaments and microtubules (MT) at the fly oocyte cortex as well as playing a critical role in maintaining the actin cytoskeleton in the nurse cells surrounding the oocyte (Liu et al. 2009). Work in this same model has reported that WASH, via the N-terminal WAHD1 domain, associates directly with the GTPase Rho1 and actin nucleation factor Spire while associating indirectly with the protein Cappuccino (Liu et al. 2009). While Drosophila WASH was reported to function as a Rho effector, mammalian WASH does not appear to function as a Rho effector (Jia et al. 2010; Liu et al. 2009). However, the GTPase Rac1 has been shown to have a weak interaction with the mammalian SHRC, yet it appears that Rac1 is unable to activate the SHRC (Jia et al. 2010). Thus it is currently unclear how the SHRC or WASH itself is activated in vivo. Future studies aimed at identifying the signaling pathways regulating WASH activity will greatly enhance our understanding of the function of WASH in cellular processes.
WASH may function in a variety of endosomal sorting and maturation pathways but there is much to clarify in the field. WASH has been reported to colocalize with epidermal growth factor (EGF) following internalization, and suppression of WASH may result in defective EGF transport to late endosomes, but there have been conflicting reports over whether WASH actually plays a role in the degradation of EGFR (Epidermal Growth Factor Receptor) (Duleh and Welch 2010; Gomez and Billadeau 2009). WASH has also been implicated in the regulation of transferrin recycling in mouse fibroblasts suppressed for WASH, but no such role was found using transformed human cells (Derivery et al. 2009; Duleh and Welch 2010). Additionally, WASH has been reported to regulate the trafficking of the cation-independent mannose-6-phosphate receptor (CI-MPR) (Gomez and Billadeau 2009). Under steady-state conditions, the CI-MPR is primarily localized to the trans-Golgi network (TGN) due to the action of the mammalian retromer complex, consisting of Vps35 (Vacuolar protein sorting-associated protein 35), Vps26 (Vacuolar protein sorting-associated protein 26), Vps29 (Vacuolar protein sorting-associated protein 29), and various sorting nexin (SNX) proteins, which is responsible for “retrograde” endosome to Golgi transport (Anitei et al. 2010; Bonifacino and Hurley 2008). Importantly, the SHRC localizes with VPS35, SNX1, and SNX2 (Gomez and Billadeau 2009), and depletion of Vps35, which effectively removes Vps26 and Vps29 from endosomes, and results in SHRC dispersal (Harbour et al. 2010) (T.S. Gomez and D.D. Billadeau, unpublished observation). This suggests that retromer accumulation on endosomal subdomains leads to SHRC recruitment and function at these specific endosomal structures. Importantly, suppression of WASH results in CI-MPR dispersal from its normal localization near the TGN to endosomes (Gomez and Billadeau 2009). Consistent with missorting of CI-MPR by the retromer in WASH-depleted cells, the total levels of CI-MPR were also reduced indicating that CI-MPR was likely being shunted into the lysosomal pathway, a feature consistent with retromer deregulation (Gomez and Billadeau 2009). Thus it appears that WASH participates in retromer-mediated endosomal transport of cargo to the Golgi (Fig. 3b, c). It is important to note that so far WASH does not appear to regulate endocytosis, as it has been shown that WASH fails to localize with clathrin at the site of the plasma membrane or affect TCR (T cell Receptor) internalization rate (Gomez and Billadeau 2009). While there is no clear consensus on the exact function of WASH in each of these cellular trafficking pathways, the current reports are suggestive of WASH plays a key role in the regulation of actin assembly on early endosomes and the subsequent membrane remodeling that occurs during endosomal sorting and trafficking (Fig. 3b, c).
Interestingly, all three current reports on the function of mammalian WASH noted that upon suppression of WASH, the normally spherical early endosomes became enlarged and elongated resulting in the endosomal network taking on a tubulation phenotype (Derivery et al. 2009; Duleh and Welch 2010; Gomez and Billadeau 2009). It was reported that depolymerization of microtubules in WASH-depleted cells prevented this endosomal tubulation from occurring (Derivery et al. 2009). Additionally, blocking the VCA domain of WASH in vivo induced endosome tubulation (Derivery et al. 2009). Coupled with the fact that WASH is known to interact with microtubules via its TBR and the C-terminal VCA domain has been shown to induce Arp2/3-mediated actin nucleation, it seems likely that the ability of WASH to interact with these two key cytoskeleton components is a critical part of its ability to influence endosomal membrane shape and dynamics (Derivery et al. 2009; Gomez and Billadeau 2009). Finally, it has been reported that the GTPase dynamin, which possesses a well-characterized role in membrane fission, immunoprecipitates with endogenous WASH, which suggests the possibility that dynamin may also play a role in WASH-mediated endosomal fission (Derivery et al. 2009). It has been widely proposed that WASH is required to mediate actin polymerization at the site of endosomal sorting and that this polymerization provides the physical force required to drive the constriction and eventual scission of endosomal membranes allowing for the proper cellular sorting and trafficking of endosomal cargo (Fig. 3b, c) (Derivery et al. 2009; Duleh and Welch 2010; Gomez and Billadeau 2009). This model, which requires further elucidation, if proved correct, could closely mimic the role of N-WASP-mediated Arp2/3 activity in driving membrane scission during endocytosis at the plasma membrane (Takenawa and Suetsugu 2007).
It should be briefly noted that WASH has been found to localize to the site of Salmonella cellular entry and that depletion of WASH in fibroblasts reduced the level of Salmonella invasion (Hänisch et al. 2010). Furthermore, the SHRC member, FAM21, has been reported to facilitate vaccina virus entry into HeLa cells via dynamin-mediated endocytosis and that knockdown of FAM21 resulted in reduced cellular entry of vaccina virus (Huang et al. 2008). Consistent with other reports of WASH localization, it appears that FAM21-mediated vaccina virus penetration occurs through clathrin-independent endocytosis (Gomez and Billadeau 2009; Huang et al. 2008). It is possible that actin polymerization mediated by the WASH complex may play an important role in pathogen-directed membrane remolding and subsequent intracellular invasion. It is unclear if this function of WASH is related to its regulation of endosomal sorting.
WASH is a recently identified member of the WASP superfamily of Arp2/3 NPFs. A highly evolutionarily conserved protein, the gene for WASH in humans and higher order primates has variable copy numbers with gene copies located within subtelomeric regions. The protein itself has been shown to posses several conversed domains: an N-terminal WAHD1, a C-terminal VCA, and a central PRD preceded by a TBR (Fig. 1). Like other WASP superfamily members, WASH is integrated into a multiprotein complex and the activity and stabilization of WASH is regulated and dependent upon the pentameric SHRC (Fig. 2). The SHRC has been found to contain a high degree of homology to the pentameric WRC, which regulates its subcellular localization and activity toward Arp2/3. Unlike the WRC however, it does not appear that the SHRC functions downstream of Rac1. While no activator of WASH has been established, the intrinsic activity of WASH toward Arp2/3 is inhibited upon assembly into the SHRC, suggesting that WASH activation is likely to be regulated in vivo by currently undefined molecular mechanisms. Unique among the WASP superfamily, WASH localizes to early endosomal subdomains enriched in retromer subunits and is responsible for endosomal localization and activity of Arp2/3 (Fig. 3a). The N-terminal region of the SHRC member FAM21 mediates this endosomal localization. Depletion of WASH has been shown to result in a tubulation phenotype in which endosomes lose their spherical shape and become enlarged and elongated. It appears that WASH plays a key role in the sorting and trafficking of early endosomes, and that as a result WASH may specifically help regulate EGFR and transferrin trafficking and degradation pathways. It is also clear that WASH is involved in the proper function of retromer-mediated transport from endosomes to the Golgi. As a recently discovered NPF, there is much to be learned in the future about the role of WASH in cellular signaling and function. As the extent of endosomal-based signaling becomes better understood, the mechanisms by which WASH regulates cytoskeleton dynamics and endosomal trafficking will be even more important to understand.
- Jia D, Gomez TS, Metlagel Z, Umetani J, Otwinowski Z, Rosen MK, et al. WASH and WAVE actin regulators of the Wiskott-Aldrich syndrome protein (WASP) family are controlled by analogous structurally related complexes. Proc Natl Acad Sci U S A. 2010;107:10442–7.PubMedPubMedCentralCrossRefGoogle Scholar
- Monfregola J, Napolitano G, D’urso M, Lappalainen P, Ursini MV. Functional characterization of Wiskott-Aldrich Syndrome Protein and Scar Homolog (WASH), a bi-modular nucleation-promoting factor able to interact with biogenesis of lysosome-related organelle subunit 2 (BLOS2) and -Tubulin. J Biol Chem. 2010;285:16951–7.PubMedPubMedCentralCrossRefGoogle Scholar