Editor's evaluation: The Uso1 globular head interacts with SNAREs to maintain viability even in the absence of the coiled-coil domain
Year: 2023
Type: peer-review
Abstract: Full text Figures and data Side by side Abstract Editor's evaluation Introduction Results Discussion Materials and methods Data availability References Decision letter Author response Article and author information Metrics Abstract Uso1/p115 and RAB1 tether ER-derived vesicles to the Golgi. Uso1/p115 contains a globular-head-domain (GHD), a coiled-coil (CC) mediating dimerization/tethering, and a C-terminal region (CTR) interacting with golgins. Uso1/p115 is recruited to vesicles by RAB1. Genetic studies placed Uso1 paradoxically acting upstream of, or in conjunction with RAB1 (Sapperstein et al., 1996). We selected two missense mutations in uso1 resulting in E6K and G540S in the GHD that rescued lethality of rab1-deficient Aspergillus nidulans. The mutations are phenotypically additive, their combination suppressing the complete absence of RAB1, which emphasizes the key physiological role of the GHD. In living hyphae Uso1 recurs on puncta (60 s half-life) colocalizing partially with the Golgi markers RAB1, Sed5, and GeaA/Gea1/Gea2, and totally with the retrograde cargo receptor Rer1, consistent with Uso1 dwelling in a very early Golgi compartment from which ER residents reaching the Golgi recycle back to the ER. Localization of Uso1, but not of Uso1E6K/G540S, to puncta is abolished by compromising RAB1 function, indicating that E6K/G540S creates interactions bypassing RAB1. That Uso1 delocalization correlates with a decrease in the number of Gea1 cisternae supports that Uso1-and-Rer1-containing puncta are where the protein exerts its physiological role. In S-tag-coprecipitation experiments, Uso1 is an associate of the Sed5/Bos1/Bet1/Sec22 SNARE complex zippering vesicles with the Golgi, with Uso1E6K/G540S showing a stronger association. Using purified proteins, we show that Bos1 and Bet1 bind the Uso1 GHD directly. However, Bet1 is a strong E6K/G540S-independent binder, whereas Bos1 is weaker but becomes as strong as Bet1 when the GHD carries E6K/G540S. G540S alone markedly increases GHD binding to Bos1, whereas E6K causes a weaker effect, correlating with their phenotypic contributions. AlphaFold2 predicts that G540S increases the binding of the GHD to the Bos1 Habc domain. In contrast, E6K lies in an N-terminal, potentially alpha-helical, region that sensitive genetic tests indicate as required for full Uso1 function. Remarkably, this region is at the end of the GHD basket opposite to the end predicted to interact with Bos1. We show that, unlike dimeric full-length and CTR∆ Uso1 proteins, the GHD lacking the CC/CTR dimerization domain, whether originating from bacteria or Aspergillus extracts and irrespective of whether it carries or not E6K/G540S, would appear to be monomeric. With the finding that overexpression of E6K/G540S and wild-type GHD complement uso1∆, our data indicate that the GHD monomer is capable of providing, at least partially, the essential Uso1 functions, and that long-range tethering activity is dispensable. Rather, these findings strongly suggest that the essential role of Uso1 involves the regulation of SNAREs. Editor's evaluation This valuable manuscript explores the role of Uso1/p115, a protein that has been implicated in vesicle tethering at the ER-Golgi interface. By investigating a Uso1 mutant that allows Aspergillus cells to survive in the absence of Rab1, the authors conclude that the essential role of Uso1 is not actually tethering, but rather SNARE complex assembly mediated by the globular head domain. This convincing analysis significantly advances our understanding of Uso1 and also prompts a reevaluation of long-standing assumptions about coiled-coil proteins involved in vesicular transport. https://doi.org/10.7554/eLife.85079.sa0 Decision letter Reviews on Sciety eLife's review process Introduction Vesicular traffic at the ER/Golgi interface is the cornerstone of the secretory pathway (Barlowe and Miller, 2013; Weigel et al., 2021). In current models, in which traffic across the Golgi is driven by cisternal maturation (Day et al., 2013; Pantazopoulou and Glick, 2019), COPII vesicles generated at specialized domains of the ER fuse homotypically and heterotypically to form and feed the earliest Golgi cisternae (Rexach et al., 1994). As straightforward as this step might seem, it involves a sophisticated circuitry of regulation. Actual fusion is in part mediated by compartmental-specific sets of four-membered SNARE protein complexes (SNARE bundles) (Malsam and Söllner, 2011; Pelham, 2001; Rizo and Südhof, 2012). Most SNARES are type II single TMD proteins, whose N-terminal cytosolic domain contains nearly all the polypeptide, except a few lumenal residues. Like any other transmembrane proteins, SNAREs are synthesized in the ER. This implies that they have to travel to compartments of the cell as distant as the plasma membrane in a conformation that precludes them from catalyzing what would be a calamitous fusion of non-cognate donor and acceptor compartments. Achieving the strictest specificity is particularly challenging in the ER-to-Golgi stage that, as the first step in the secretory pathway, represents an obligate point of transit for each and every transmembrane SNARE. Therefore, the only SNARES acting in this first step are the Qa Sed5, the Qb Bos1, the Qc Bet1, and the R-SNARE Sec22, which form the bundle mediating fusion of carriers that coalesce into cisternae (McNew et al., 2000). Given the central role played by the secretory pathway in the physiology of every eukaryotic cell, it is unsurprising that this step involves regulatory factors which are essential for cell survival. One is the SM (Sec1, Munc-18) protein Sly1, which promotes SNARE bundle formation (Bracher and Weissenhorn, 2002; Peng and Gallwitz, 2002; Thomas et al., 2019) Another is the TRAPPIII complex, which interacts with the external coat of COPII carriers and acts as a guanine nucleotide exchange factor (GEF) for RAB1 (Bracher and Weissenhorn, 2002; Cai et al., 2007; Galindo et al., 2021; Joiner et al., 2021; Lord et al., 2011; Peng and Gallwitz, 2002; Pinar and Peñalva, 2020; Riedel et al., 2018; Thomas et al., 2018; Thomas et al., 2019). This small GTPase is a key player that transiently recruits protein effectors from the cytosol to donor and acceptor membranes (Søgaard et al., 1994) and regulates SNARE assembly through an as yet undefined mechanism (Lupashin and Waters, 1997; Sapperstein et al., 1996). One RAB1 effector is a fungal protein denoted Uso1, whose highly conserved metazoan homolog is p115. These are homodimers with a globular N-terminal head and a long C-terminal coiled-coil region characteristic of tethering proteins, which bring donor and acceptor membranes into the distance at which v- and t-SNAREs can engage into the productive trans-SNARE complex that mediates membrane fusion (Cao et al., 1998; Nakajima et al., 1991; Sapperstein et al., 1996; Sapperstein et al., 1995; Seog et al., 1994; Yamakawa et al., 1996). Despite most Golgi tethers are non-essential, Uso1 is unique in that it is required for viability. uso1-1, an S. cerevisiae amber mutation truncating most, but not all the coiled-coil region is viable, yet further upstream truncation removing the complete coiled-coil is lethal, which was taken as evidence that tethering is the essential function of Uso1 (Seog et al., 1994). In addition, as the coiled-coil is predicted to mediate dimerization, it is broadly accepted that Uso1 is ‘just’ an essential homodimer that tethers vesicles to the acceptor membrane. However, genetic evidence stubbornly indicates that Uso1 plays additional functions related to SNAREs. For example, Sapperstein et al., 1996 showed that SNAREs function downstream of Uso1. Notably, the view that Uso1 is a mere RAB1 effector was challenged by the observation that overexpressing Ypt1 (yeast RAB1) rescues the lethality of uso1∆, whereas the reciprocal is not true, indicating that Uso1 acts upstream of or in conjunction with RAB1 (Sapperstein et al., 1996). Our laboratory is interested in deciphering the domains of action of RAB GTPases in the genetic and cell biological model organism Aspergillus nidulans (Pinar and Peñalva, 2021). We have previously gained mechanistic insight into the activation of RAB11 by TRAPPII by exploiting a forward genetic screen for mutations bypassing, at the restrictive temperature, the essential role of the key TRAPPII subunit Trs120 (Pinar et al., 2019; Pinar et al., 2015; Pinar and Peñalva, 2020). In this type of screen, a strain carrying a temperature-sensitive (ts) mutation in the gene-of-interest is mutagenized and strains bypassing lethality at the restrictive temperature are identified and characterized molecularly. A well-characterized, conditionally lethal rab1 mutation is available (Pinar et al., 2013), enabling us to investigate pathways collaborating with RAB1 in anterograde traffic. We isolated two uso1 missense mutations causing substitutions in the GHD. When combined together, these rescued the lethality resulting from rab1∆ and promoted the localization of the protein to early Golgi cisternae by increasing Uso1 binding to the cytosolic region of the Qa SNARE Bos1. Importantly, we show that endogenous expression of a protein consisting solely of the double mutant GHD, or overexpression of double mutant or wild-type GHD, rescues the lethality resulting from uso1∆, even though data strongly suggest that the GHD is monomeric. Our results show that one essential role of RAB1 is recruiting Uso1 to membranes, and strongly suggest that the essential role of Uso1 is not tethering membranes, but rather regulating the formation of the cognate SNARE bundle, almost certainly placing Uso1 as a component of the SNARE fusion machinery. Results Missense mutations in uso1 rescue the lethality resulting from rab1∆ rab1A136D (hereby rab1ts) mutants do not grow at 37 °C. However, when we plated UV-mutagenized conidiospores of the rab1ts mutant at this temperature, we obtained colonies showing different degrees of growth, presumably carrying mutations rescuing the lethality resulting from rab1ts. One was chosen for further characterization. By sexual crosses and parasexual genetics, this strain was shown to carry a single suppressor mutation, denoted su1rab1ts, that co-segregated with chromosome VIII. Meiotic mapping narrowed su1rab1ts to the vicinity (2 cM) of hisC. 40 kb centromere distal from hisC lies AN0706 (Figure 1A) encoding Aspergillus nidulans Uso1, a conserved effector of RAB1. Sanger sequencing revealed the presence of a G16A transition (denoted E6K) resulting in Glu6Lys substitution in the uso1 gene of su1rab1ts. Figure 1 with 1 supplement see all Download asset Open asset Characterization of mutations bypassing the essential role of RAB1. (A). Genetic map in the region surrounding uso1 with genetic markers used as landmarks for mapping. (B). Molecular identification of the nucleotide changes in suArab1ts strains. (C) and (D): growth tests showing rab1ts- and rab1∆-rescuing phenotypes, respectively, of individual mutations, and synthetic positive interaction between E6K and G540S. Strains produce either green or white conidiospores (conidiospore colors are used as genetic markers). In (C), strains were point-inoculated. In (D) conidiospores were spread on agar plates to give individual colonies. Figure 1—source data 1 https://cdn.elifesciences.org/articles/85079/elife-85079-fig1-data1-v2.docx Download elife-85079-fig1-data1-v2.docx To determine if the remaining suppressor strains were allelic to su1rab1ts, we sequenced uso1 from a further 13 isolates (Figure 1B). Of these, four were rab1ts pseudo-revertants that had acquired a functionally acceptable mutation in the altered codon, and eight carried uso1 E6K, suggesting that the screen was close to saturation. However, one mutation was found to be a different missense allele, su85rab1ts (denoted G540S) resulting in Gly540Ser substitution. Single mutant strains carrying these uso1 mutations showed no growth defect, indicating that E6K and G540S were unlikely to result in loss-of-function, and suggesting instead that mutant strains had acquired features that made them largely independent of RAB1. These findings were unexpected because in Saccharomyces cerevisiae overexpression of Uso1 does not rescue the lethality of ypt1∆ mutants (Ypt1 is the yeast RAB1 homolog) (Sapperstein et al., 1996). To demonstrate that uso1E6K and uso1G540S were causative of the suppression, we reconstructed them by homologous recombination. These reverse-genetic alleles rescued the viability of rab1ts at 37 °C to a similar extent as su1rab1ts and su85rab1ts (Figure 1C). uso1G540S was the strongest suppressor, such that rab1ts uso1G540S double mutants grew nearly as the wt at 37 °C. Nevertheless, the two alleles showed additivity, and a triple mutant carrying uso1E6K, uso1G540S, and rab1ts grew at 42 °C, unlike either single mutant (Figure 1C). These data, together with the genetic mapping above, established that uso1E6K and uso1G540S are responsible for the suppression phenotype. RAB1 recruits Uso1/p115 to COPII vesicles and early Golgi cisternae (Allan et al., 2000). Therefore, uso1E6K/G540S might, by increasing the affinity of Uso1 for RAB1, compensate for the reduction in the amount of the GTPase resulting from rab1ts. However, Figure 1D shows that both uso1E6K and uso1G540S rescue the lethality resulting from the complete ablation of rab1∆ at 30 °C, with the strongest uso1G540S suppressor rescuing viability even at 37 °C, and the double mutant rescuing rab1∆ even at 42 °C (Figure 1D). In contrast, uso1E6K/G540S did not rescue the lethality resulting from arf1∆, nor from sed5∆ or sly1∆, the syntaxin and the SM protein which are crucial for the formation of the ER/Golgi SNARE bundle (Figure 1—figure supplement 1), indicating that Uso1 plays a role acting downstream of RAB1 and upstream of or in conjunction with the SNARE machinery. This role is essential for survival (Figure 1—figure supplement 1). E6K affects a previously undetected N-terminal helix, whereas G540S is located in a loop near the end of the armadillo domain 1103-residue A. nidulans Uso1 is similar in size to 961-residue p115 (bovine) and notably shorter than S. cerevisiae Uso1p (1790 residues) (Yamakawa et al., 1996). Thus far, atomic structures of Uso1/p115 are limited to the 600–700 residue GHD, which consists of a highly conserved α-catenin-like armadillo-fold (An et al., 2009; Heo et al., 2020; Striegl et al., 2009). In silico analyses robustly predict that the approximately C-terminal half of Uso1/p115 consists of a coiled-coil that mediates tethering and dimerization, but this region has not been characterized beyond low-resolution EM studies (Yamakawa et al., 1996). Neither crystal structures nor predictions provided information about the N-terminal extension in which Glu6Lys lies. Thus, we used AlphaFold2, imposing the condition that the protein is a dimer (see below). Figure 2A shows a model with the highest confidence scores (see Figure 2—figure supplement 1). Like their relatives, Uso1 from A. nidulans contains an N-terminal GHD including a previously unnoticed short α-helix in which Glu6 is affected by E6K lies. This N-terminal extension is followed by ~34 α-helices arranged into 12 tandem repetitions of armadillo repeats (ARM1-ARM12; residues 17 through 564), each containing three right-handed α-helices except for the first two repeats. Altogether, the armadillo repeats resemble the shape of a jai alai basket. Downstream of the GHD, AlphaFold2 predicts a long extended CC between residues 674 through 1082, which would mediate dimerization (see below) (Figure 2A). The CC ends at a conserved CTR (Figure 2A and D), which includes an also C-terminal segment rich in acidic residues. In Uso1, twelve out of the last seventeen amino acids are Asp/Glu (Figure 2A and D). The CC and CTR regions will be collectively denoted the coiled-coil domain (CCD). Readers should note that the model of the CCD is highly speculative (see Figure 2—figure supplement 1), and that it is displayed with the sole purpose of visually depicting the relative size of this region compared to the globular domain, which may help to consider the potential effects of this extended region in the sedimentation data discussed below. Figure 2 with 1 supplement see all Download asset Open asset Localization of the amino acid substitutions within the Uso1 AlphaFold2 structure. (A). AlphaFold2 cartoon representations of A. nidulans Uso1 dimer. Note that the depiction of the coiled-coil domain (CCD) is highly speculative. It is included with the sole purpose of visually appreciating the relative sizes of the globular-head-domain (GHD) and the CCD domains. Confidence estimations for Uso1 models are detailed in Figure 2—figure supplement 1. Colors are as in the scheme below: Cyan, N-terminal tail; pink, globular head domain; gray, coiled-coil; yellow, C-terminal region (CTR). (B). Position of the Gly6Lys and Gly540Ser substitutions. Only the GHD of dimeric full-length Uso1 is shown. Distances between mutated residues are displayed in angstroms. (C). The N-terminal amphipathic α-helix affected by the Glu6Lys substitution. (D). Amino acid alignment of fungal sequences with mammalian p115 showing strong conservation within the CTR: ANIDU, Aspergillus nidulans; PRUBE, Penicillium rubens; TREES, Thrichoderma ressei; SSCLE, Sclerotinia scleriotorum; MORYZ, Magnaporthe oryzae; CIMM, Coccidioides immitis; BOVIN, Bos taurus. Even though the GHD contains Glu6 and Gly540 in the N-terminal helix and at the beginning of armadillo α-helix 29, respectively, intramolecular or intermolecular (in the context of a homodimer, see below) distances between these residues are long, arguing against the possibility that they bind a common target as components of the same interaction surface (Figure 2B). Indeed, the synthetic positive effect of the mutations would be consistent with their rescuing viability through different mechanisms. The previously unnoticed short α-helix predicted by AlphaFold between Phe2 and Lys12 is amphipathic (Figure 2C). Glu6 lies on the polar side of this helix, such that Glu6Lys increases its overall positive charge (three of the four polar residues are Lys or Arg). Coiled-coil mediated dimerization of Uso1: the globular head as isolated from bacteria is monomeric It has been proposed that p115 alternates between closed and open conformations to hide or expose a RAB1 binding site present in the coiled-coil region (Beard et al., 2005). This would be mediated by intramolecular interactions between the globular domain and the C-terminal acidic region, which would be disrupted by the competitive binding of golgins GM130 and giantin to the latter. We addressed whether E6K/G540S promotes a conformational change in Uso1, or, alternatively, a change in the oligomerization status of the protein, by analytical ultracentrifugation. We designed carrying a C-terminal (Figure to the full-length protein with or E6K and G540S substitutions. The included wild-type and of Uso1 lacking the CTR and The to wild-type and of the globular domain, denoted Uso1 GHD and Uso1 The to the Uso1 proteins were in purified by and and by sedimentation ultracentrifugation. These revealed that all protein were and they were used to determine the In addition, by we the of the With these we the of the different proteins Figure with 1 supplement see all Download asset Open asset and oligomerization status of the different Uso1 by sedimentation The different the sedimentation of the protein with of the scheme of the in Figure and of showing the of the protein The the of the used to relative is the of the from predicted of the Figure data 1 Download and E6K/G540S full-length showed the same sedimentation that the mutations do not a conformational change that would have been in changes in to in Molecular from the that these full-length proteins are homodimers (Figure in with of the conserved CTR did not result in any change in the sedimentation (Figure and 1 and irrespective of the presence or absence of the a model in which the CTR would interact with the GHD to a closed conformation (Beard et al., 2005). In addition, the of the proteins to a that the acidic region is not involved in dimerization the GHD, whether wild-type or as a monomer (Figure and which has below. In contrast, the with a predicted of as a dimer of (Figure The sedimentation of the CCD is markedly than that of the monomeric GHD, suggesting an These together with the dimeric of the lacking the showed that dimerization is mediated by the The absence of residues to the CCD CTR domains in the GHD and the monomeric of the compared to the full-length Uso1 dimer did not result in a decrease in sedimentation which from to in the wild-type that the change in from in full-length Uso1 to only of the (Figure These data are in with AlphaFold2 predictions depicting Uso1 as a dimer with a globular head and an extended coiled-coil that would sedimentation of the protein As with full-length Uso1, the double substitution did not the sedimentation of the GHD (Figure and To the that GHD is a monomer irrespective of the presence or absence of the mutations, we sedimentation different protein from to (Figure supplement In all the GHD as a of Uso1 GHD lacking the showed a similar that the monomeric of the mutant is not to the at the C-terminal dimerization (Figure supplement 1C). Therefore, sedimentation did not any change in or structures between wild-type and mutant GHD, which is for the of genetic data that will be discussed below. In Uso1 is a The C-terminal acidic region is for dimerization and does not mediate an between closed and open The globular domain of Uso1 is a The coiled-coil domain of Uso1 is a the double E6K G540S substitution does not any conformational in Uso1, nor does it result in a change in the oligomerization of the to the early Golgi in a The compartments of the Golgi are not in the of which appear as structures in different of by et al., et al., Pantazopoulou and Peñalva, 2011; Pinar et al., 2013; and Pelham, Uso1 is predicted to to the Golgi, studies of its localization in are limited et al., et al., Therefore, we the A. nidulans uso1 gene with Figure and 1 depicting a of a as as of in Figure show that to puncta the (see Figure and Figure supplement 1). These puncta are and than for other markers of the Golgi, which that they might domains rather than complete Notably, revealed that Uso1 puncta are and with (Figure That this did not that the puncta of was established with which revealed a similar of Uso1 irrespective of whether or was used Therefore, we with only of we a 2 with for consisting of These to Uso1 puncta across the complete of the hyphae (Figure However, as the of Uso1 puncta made analysis of Uso1 maturation we them with the of representations generated with combined with observation of in (Figure and Figure supplement 1). The in Figure (see a The Figure shows to this for We which an of the of Uso1 in puncta of (Figure Figure with 1 supplement see all Download asset Open asset localization of Uso1. (A). to by (B). of a and its in for (C). showing the of Uso1 to (D). of of Uso1 in these of one such with a and with the 1 Download asset This be played in because does may the for Download as Download as Download as of a 2 Download asset This be played in because does may the for Download as Download as Download as showing the of in which were acquired at a of 1 every Download asset This be played in because does may the for Download as Download as Download as of at 2 showing the of 2 Download asset This be played in because does may the for Download as Download as Download as Single of The contains acquired at and have proposed that the of and proteins between ER and the early Golgi through a mechanism et al., structures in size and ER with COPII we cells with et al., 2019; et al., 2019). The shown in Figure and show that the two markers are but only in a few they showed These did not as they were found to in the or (Figure and These have not been further with but at the very least we can conclude that the are
Source: eLife
Author: Benjamin S Glick
Institution: University of Chicago
Language: English
Subfield: Cell Biology
Domain: Life Sciences
Open Access status: gold
Estimated APC paid: $2,500
OpenAlex ID: W4380990568