Hrs and STAM Function Synergistically to Bind Ubiquitin-Modified Cargoes in Vitro

76 Biophysical Journal Volume 108 January 2015 76–84 Article

Hrs and STAM Function Synergistically to Bind Ubiquitin-Modiﬁed Cargoes In Vitro

Hirohide Takahashi,1 Jonathan R. Mayers,2 Lei Wang,2 J. Michael Edwardson,1 and Anjon Audhya2,* 1Department of Pharmacology, University of Cambridge, Tennis Court Road, Cambridge, United Kingdom; and 2Department of Biomolecular Chemistry, University of Wisconsin-Madison School of Medicine and Public Health, Madison, Wisconsin

ABSTRACT The turnover of integral membrane proteins requires a specialized transport pathway mediated by components of the endosomal sorting complex required for transport (ESCRT) machinery. In most cases, entry into this pathway requires that cargoes undergo ubiquitin-modification, thereby facilitating their sequestration on endosomal membranes by specific, ubiquitin- binding ESCRT subunits. However, requirements underlying initial cargo recognition of mono-ubiquitinated cargos remain poorly defined. In this study, we determine the capability of each ESCRT complex that harbors a ubiquitin-binding domain to bind a reconstituted integral membrane cargo (VAMP2), which has been covalently linked to mono-ubiquitin. We demonstrate that ESCRT-0, but not ESCRT-I or ESCRT-II, is able to associate stably with the mono-ubiquitinated cargo within a lipid bilayer. Moreover, we show that the ubiquitin-binding domains in both Hrs and STAM must be intact to enable cargo binding. These results indicate that the two subunits of ESCRT-0 function together to bind and sequester cargoes for downstream sorting into intralumenal vesicles.

INTRODUCTION The ESCRT machinery is composed of five multisubunit into ILVs (1–3). However, the individual roles of these complexes (ESCRT-0, ESCRT-I, ESCRT-II, ESCRT-III, ESCRT complexes in promoting cargo sequestration remain and the Vps4 complex) and several accessory proteins ill-defined. (1–3). A majority of these factors are conserved among The ubiquitin-interacting domains found within subunits eukaryotic species, suggesting that they play key roles in of the early-acting ESCRT complexes (ESCRT-0, ESCRT- regulating cellular function (4). Consistent with this idea, I, and ESCRT-II) have been demonstrated to bind ubiquitin ESCRT components have been implicated in a number of weakly in solution, with affinities ranging from ~100 to membrane scission events, including intralumenal vesicle 600 mM(2). The membrane-associated heterotetrameric (ILV) biogenesis on endosomes, cytokinetic abscission ESCRT-0 complex harbors at least eight ubiquitin-binding to separate daughter cells after division, and the budding surfaces, all of which can engage ubiquitin simultaneously of many retroviruses from the plasma membrane (5–7). In (12). In contrast, ESCRT-I contains no more than two do- each case, adaptors recruit ESCRT complexes that ulti- mains capable of interacting with ubiquitin (depending on mately bend and/or cleave the lipid bilayer. For example, the presence of mammalian UBAP1 in the complex), and in mammalian cells, Cep55 recruits the ESCRT-I complex ESCRT-II contains only a single ubiquitin-binding surface to the site of abscission, whereas viral Gag isoforms bring (16–19). Although all early-acting ESCRT complexes ESCRT-I to sites of virus budding on the plasma membrane exhibit the ability to bind to endosomally enriched phospha- (8,9). Accumulation of ESCRT-I and other downstream tidylinositol 3-phosphate (PI3P), their affinities vary signifi- complexes on endosomal membranes depends on ESCRT- cantly. At steady state, only ESCRT-0 accumulates stably on 0, which assembles as a heterotetramer of two unique sub- endosomes, whereas ESCRT-I and ESCRT-II exhibit a more units, Hrs and STAM, on lipid bilayers (10–12). In addition transient association (2). Current evidence suggests that to its role as an adaptor for ESCRT-mediated ILV formation, ESCRT-0 recruits ESCRT-I, which in turn brings ESCRT-II ESCRT-0 also plays an important role in promoting ubiqui- onto the membrane. These data suggest that a supercomplex tin-dependent cargo sorting by virtue of its ability to bind of the early-acting ESCRT machinery may form to sequester directly to ubiquitin (13–15). Both ESCRT-I and ESCRT- cargoes (13). However, assembly of such a higher-order olig- II also harbor domains that associate with ubiquitin, raising omer is unlikely to be stable, as the distributions of each com- the possibility that multiple complexes in the ESCRT ma- plex have been shown to differ in vivo (10). chinery participate in cargo recognition before deposition Recently, the ESCRT-0 complex, but not ESCRT-I, was shown to accumulate at sites of clathrin-mediated endocytosis at the plasma membrane, through its ability to interact Submitted March 25, 2014, and accepted for publication November 5, 2014. with a set of clathrin adaptors (20). In the absence of this *Correspondence: [email protected] pool of ESCRT-0, ubiquitin-modified cargoes continue to be Editor: Andreas Engel. Ó 2015 by the Biophysical Society 0006-3495/15/01/0076/9 $2.00 http://dx.doi.org/10.1016/j.bpj.2014.11.004 ESCRT-0 Sequesters Ubiquitin-Modified Cargoes 77 internalized normally, but exhibit a substantial delay in endo- Lipids (Alabaster, AL). Phosphatidylinositol 3-phosphate (PI3P) was ob- somal trafficking. These findings highlight the possibility that tained as a solution in chloroform/methanol/water (1:2:0.8) from Echelon initial recognition and/or sequestration of cargoes by ESCRT- Biosciences (Salt Lake City, UT). A lipid mixture of composition 54% PC, 30% PE, 15% PS, and 1% PI3P was produced, and the solvents were 0 in clathrin-coated pits accelerates their downstream sorting evaporated in a stream of nitrogen gas. Dried lipid (400 mg) was resus- into ILVs. Similarly, ubiquitin-modified biosynthetic cargoes pended in 200 mL of 2% CHAPS in HEPES-buffered saline (HBS; leaving the trans-Golgi network encounter a pool of ESCRT- 100 mM NaCl, 20 mM HEPES, pH 7.5), containing 1 mg of protein (VAMP2 or ubiquitin-conjugated VAMP2). The mixture was then dialyzed 0 on early endosomes, which may result in their local concentration before endosome maturation and the accompanying against 1 L of HBS for 3 days at 4 C. recruitment of downstream ESCRT complexes (1). In both of these models, cargo sequestration depends on ESCRT-0, Atomic force microscopy whereas other ESCRT components function to promote membrane bending and scission (21). Liposomes (either with or without integrated protein) were mixed with individual ESCRT complexes (final concentration of 150 nM in each case) To directly test the ability of early-acting ESCRT com- and incubated for 30 min at 37C. The suspension was then adsorbed plexes to bind cargo within a lipid bilayer, we developed a onto freshly cleaved mica for 3 min at room temperature. Imaging was per- system to measure the distribution of a mono-ubiquitin-con- formed in tapping mode under fluid (HBS) using a Bruker Multimode in- jugated substrate at nanometer-scale resolution using atomic strument (Santa Barbara, CA), controlled by a Nanoscope IIIa controller, force microscopy (AFM). Importantly, in contrast to earlier at room temperature. Cantilevers (MikroMasch HQ:NSC18/AL BS, Inno- vative Solutions Bulgaria, Sofia, Bulgaria) were tuned to between 10% to studies that used modified lipid headgroups to tether ubiqui- 20% below the peak of the resonance frequency, generally found between tin to the surface of a bilayer (22), we took advantage of a 25 and 35 kHz in fluid conditions (12,24). During the approach, the canti- reconstituted integral membrane, mono-ubiquitin-modified lever was set at 0.9 of free amplitude. During imaging, the set point value protein (VAMP2). We demonstrate that ESCRT-0, but not was kept as high as possible so that the applied force was minimized. Im- ESCRT-I or ESCRT-II, is capable of binding this model ages were captured at a scan rate of 2 Hz (unless otherwise noted), and with 512 scan lines per area. At least four cantilevers were used to generate each cargo stably, in a ubiquitin-dependent manner. Moreover, volume distribution. Data analysis was performed using commercially we show that ubiquitin-binding domains in both subunits available software (NanoScope III software; Digital Instruments, Santa of ESCRT-0 are critical for this effect, indicating that Hrs Barbara, CA). Free-standing particles were identified. Particle heights and STAM function synergistically in this process. Our and diameters were measured manually by the Nanoscope software and data define the minimal number of lipid binding modules used to calculate the molecular volume of each particle using the following equation: that must function simultaneously to restrict cargo diffusion À Á 2 2 within a physiologically relevant lipid bilayer. Vm ¼ðph=6Þ 3r þ h ; (1)

where h is the particle height and r the radius (25). This equation assumes MATERIALS AND METHODS that the adsorbed particles adopt the form of a spherical cap. Asymmetric particles, and particles at the bilayer edges, were excluded from the volume Recombinant protein purification analysis. More than 200 particles, from two to three independent experi- Cloning and polycistronic co-expression of wildtype and mutant C. elegans ments, were analyzed for each condition. h r ESCRT-0, ESCRT-I, and ESCRT-II was conducted as described previously To check that there were no systematic errors in measurement of and , which could skew the determination of volumes, we constructed scatter (12,23,24). Briefly, to enable purification of each ESCRT complex, a poly- h histidine tag was appended onto the amino-terminus of a single subunit to plots of against particle diameter for the various particles imaged. These enable capture using nickel affinity chromatography. All complexes were pu- are shown in Supporting Material. In all cases there was no strong correla- h rified further using size exclusion chromatography. Purification of murine tion between and diameter, excluding the possibility of systematic errors. VAMP2 or an amino-terminal translational fusion between ubiquitin and VAMP2 was conducted as described previously (25). Briefly, bacterial pel- lets were lysed by sonication and supplemented with ~1% Triton X-100 RESULTS before binding to nickel affinity resin. Proteins were washed and eluted in the presence of 1% n-octylglucoside, to maintain solubility. Purifications An AFM-based assay to study the distribution of of glutathione S-transferase (GST) and GST fusions to mono- and di-ubiqui- an integral membrane, ubiquitin-modified cargo tin were carried out using glutathione agarose beads. The immobilized GST proteins were subjected to extensive washing using 20 mM Tris (pH 7.5), The ESCRT machinery mediates the ubiquitin-dependent 500 mM NaCl, 0.1% Tween 20, 1 mM PMSF, and 1 mM DTT before a 30- turnover of integral membrane cargoes in the endolysoso- min incubation with purified ESCRT-0, ESCRT-I, or ESCRT-II. Subsequent mal system. Previous work indicates that the translation to the binding reaction, beads were washed five times using 20 mM Tris (pH fusion of monoubiquitin to a transmembrane protein results 7.5) and 150 mM NaCl and then eluted using 50 mM Tris (pH 7.5), 10 mM in its recognition by the ESCRT machinery and degradation glutathione, and 150 mM NaCl. Antibodies directed against Hrs, STAM, Mvb12, and Vps22 have been described previously (12,20,23). within the lysosome lumen (17). We sought to capitalize on this finding and develop a model cargo that could be repro- ducibly reconstituted into a lipid bilayer and visualized Reconstitution of VAMP2 into liposomes without the use of conjugated dyes or other tags, which Phosphatidylcholine (PC), phosphatidylethanolamine (PE), and phosphati- might influence associations with binding partners. AFM af- dylserine (PS) were obtained as chloroform solutions from Avanti Polar fords label-free imaging of proteins and protein complex

Biophysical Journal 108(1) 76–84 78 Takahashi et al. assembly on membranes at nanometer scale resolution. To stitution into synthetic liposomes (27). We purified a recom- generate bilayers that mimic the surface of early endosomes binant form of murine VAMP2 and used detergent (1% (26), we adsorbed liposomes composed of 54% PC, 30% n-octylglucoside) to maintain its solubility in solution. By PE, 15% PS, and 1% PI3P onto mica. Imaging of protein- dialyzing away the detergent in the presence of phospho- free bilayers by AFM revealed a relatively homogenous sur- lipids, we were able to achieve spontaneous insertion of face, with occasional gaps that exposed the underlying mica VAMP2 into membranes. To verify incorporation, we con- (Fig. 1 A), as observed previously (12,24). Importantly, the ducted a co-flotation assay, in which reconstituted vesicles supported membrane exhibited a highly uniform thickness were mixed with Accudenz density medium and overlaid of ~4 nm, consistent with the formation of a single lipid with decreasing concentrations of Accudenz. After centrifu- bilayer. gation, vesicles floated to the top of the gradient and were Previous studies indicate that transmembrane proteins collected for analysis by SDS-PAGE. We found that in the soluble NSF attachment protein receptor (SNARE) VAMP2 efficiently floated with vesicles in this assay, indi- family are amenable to recombinant expression and recon- cating that VAMP2 was membrane-embedded (Fig. 1 B). Fractions from elsewhere in the gradient were devoid of protein (Fig. 1 B). Next, VAMP2-containing liposomes were adsorbed onto freshly cleaved mica for imaging. Numerous free-standing particles were observed in the VAMP2-containing bilayers (Fig. 1 C). In addition, VAMP2 induced the formation of small pores in the bilayer (Fig. 1 C, arrows). These pores had raised edges, suggesting that they were lined with protein, unlike the larger gaps in the bilayer (Fig. 1 C, asterisks), which were also observed in the absence of VAMP2 (Fig. 1 A). A zoomed image of a section of bilayer containing both a protein-lined pore and a free-standing particle is shown in Fig. 1 D. Our subsequent analysis focused on the free-standing particles only. These particles exhibited a random distribution and had a peak molecular volume of 50 to 75 nm3 (Fig. 1 E). Given that the expected molecular volume of VAMP2, based on its amino acid composition, is 16 nm3, we estimate that a typical particle contains three or four VAMP2 molecules. We should point out caveats associated with the measurement of molecular volumes by AFM. For instance, it is well known that the geometry of the scanning AFM probe FIGURE 1 AFM imaging of VAMP2 and ubiquitin-VAMP2 integrated introduces a tendency to overestimate particle radii, a into supported lipid bilayers. (A) Representative AFM image of a supported convolution that becomes especially significant when imag- lipid bilayer assembled in the absence of protein. A shade-height scale bar is shown on the right of the image. Scale bar ¼ 200 nm. (B) Reconstituted ing under fluid, because fluid-imaging tips are blunter than liposomes containing 1 mg of either VAMP2 or Ub-VAMP2 were floated air-imaging tips. In addition, volume measurements of through an Accudenz density gradient and collected at the top by hand. Pro- particles bound to lipid bilayers do not take into account teins were then analyzed by SDS-PAGE (left). Similar amounts of protein any penetration of the bilayer by the protein, or any squash- were recovered in each case, as determined by Coomassie staining. In ing of the protein or the lipid by the tip. For this reason, contrast, we failed to recover VAMP2 or Ub-VAMP2 from other regions of the gradient, indicating that they had been efficiently incorporated into we regard the measured volumes as estimates rather than liposomes (right). The molecular mass markers are indicated by dashes precise values. Nevertheless, the reasonably close corre- on the right. From top to bottom, the markers have masses of 250, 150, spondence seen in this study between the measured and ex- 100, 75, 50, 37, 25, 20, and 10 kDa. (C) Representative AFM image of pected volumes gives us confidence that our interpretation a supported lipid bilayer assembled using liposomes containing VAMP2. of the imaging data is correct. Protein-lined pores and protein-free pores are indicated by arrows and asterisks, respectively. (D) Upper panel, zoomed image of a bilayer region We translationally fused ubiquitin to the amino-terminus containing a protein-lined pore and a free-standing particle. Lower panel, of VAMP2 and purified the recombinant protein identically a section through the bilayer at the position indicated by the line in the up- to the wildtype form. Importantly, the efficiency of reconsti- per panel is shown. Note that the edges of the pore and the free-standing tuting ubiquitin-modified VAMP2 (Ub-VAMP2) into lipo- particle protrude above the bilayer by ~0.5 nm. (E) Frequency distribution somes was similar with that of unmodified VAMP2 (Fig. 1 of molecular volumes of VAMP2 particles in the bilayer. (F) Representative AFM image of a supported lipid bilayer assembled using liposomes con- B). The protein:lipid ratio of the liposomes was ~1:400 taining ubiquitin-VAMP2. (G) Frequency distribution of molecular vol- (w/w). Following assembly of supported lipid bilayers, we umes of ubiquitin-VAMP2 particles in the bilayer. found that Ub-VAMP2 continued to distribute randomly,

Biophysical Journal 108(1) 76–84 ESCRT-0 Sequesters Ubiquitin-Modified Cargoes 79 and we again observed pores in the bilayer (Fig. 1 F; arrows). The only difference detected between Ub-VAMP2 and wildtype VAMP2 was an increase in its molecular volume to 75 to 125 nm3 (Fig. 1 G). The addition of a ubiquitin moiety should increase the molecular volume of a VAMP2 molecule to 27 nm3. Hence, a typical particle again would contain three or four Ub-VAMP2 molecules. Together, these findings indicated that Ub-VAMP2 met all of our criteria for determining whether components of the ESCRT machinery can influence cargo distribution within a membrane. Addi- tionally, parallel analysis of unmodified VAMP2 provided an ideal control for all studies.

Ubiquitin-binding domains in both Hrs and STAM are required for cargo binding Previous work suggests that the ESCRT-0 complex acts upstream of other ESCRT components during the sorting of ubiquitin-modified cargoes (2). ESCRT-0 is composed of two subunits, Hrs and STAM, both of which contain at least two ubiquitin-binding domains with affinities ranging from ~130 to 550 mM(12). Our goal was to determine whether these low-affinity ubiquitin-binding domains could facilitate the stable recruitment of an integral membrane protein in a ubiquitin-dependent manner. We first confirmed that recombinant ESCRT-0, assembled using C. elegans Hrs and STAM, bound to supported lipid bilayers composed of 54% PC, 30% PE, 15% PS, and 1% PI3P and in the absence of VAMP2. Consistent with our previous work (12), we found that individual particles of ESCRT-0 were distributed evenly across the bilayer (Fig. 2 A). Analysis of their vol- FIGURE 2 Mutations in the ubiquitin-binding domains of ESCRT-0 do 3 umes highlighted two major peaks, at 125 to 150 nm and not affect its assembly on lipid bilayers. (A, C, E, and G) Representative 225 to 275 nm3, as indicated by the arrows in Fig. 2 B). AFM images of bilayers assembled in the presence of wildtype ESCRT- The expected volume of a heterodimer of Hrs and STAM, 0(A), ESCRT-0 harboring point mutations in the DUIM domain of Hrs based on the amino acid compositions of the two proteins, (C), ESCRT-0 harboring two-point mutations in the VHS and UIM domains 3 of STAM (E), and ESCRT-0 harboring point mutations in all ubiquitin- is 162 nm . We conclude, therefore, that the two volume binding domains (G). A shade-height scale bar is shown. Scale bar ¼ peaks correspond to heterodimers and heterotetramers of 200 nm. In panel B, a representative Coomassie stained gel of recombinant, Hrs and STAM. We next examined mutant ESCRT-0 com- wildtype ESCRT-0 is also shown. The molecular mass markers are indi- plexes, which harbor point mutations that impair the ability cated by dashes on the right. From top to bottom, the markers have masses of either Hrs or STAM (or both) to associate with ubiquitin of 250, 150, 100, 75, 50, 37, 25, 20, and 10 kDa. Hrs and STAM are highlighted, as are two contaminants (asterisks), shown previously to be heat (12). Mutation of the DUIM (double ubiquitin interacting shock proteins that do not bind to lipid bilayers (12). (B, D, F, and H) motif) within Hrs did not affect particle size or particle The frequency distributions of molecular volumes for particles observed volume distribution as compared with wildtype ESCRT- by AFM are shown. Each graph indicates the form of ESCRT-0 visualized. 0(Fig. 2, C and D). Similarly, point mutations in the Arrows indicate the volume peaks corresponding to heterodimers and het- Vps27/Hrs/STAM (VHS) (W29A) and UIM domains of erotetramers of Hrs and STAM. STAM failed to affect the assembly of ESCRT-0 on lipid bilayers, either alone or in combination with mutations in the We reasoned that if the proteins do not interact, then the Hrs DUIM domain (Fig. 2, E–H). Importantly, these data combined volume distribution will simply be an overlay highlight that both wildtype and mutant ESCRT-0 complexes of the distributions of the two proteins imaged separately. exhibit similar volume distributions, which are distinct from In contrast, if the two proteins interact on the bilayer, then those calculated for VAMP2 and Ub-VAMP2. additional, larger particles will appear in the distribution. To assess whether ESCRT-0 could interact with an inte- We first incubated either wildtype or mutant ESCRT-0 com- gral membrane cargo in bilayers, we compared the volume plexes with supported lipid bilayers containing unmodified distributions for ESCRT-0 and VAMP2 imaged together VAMP2 and determined the volumes of all particles with the distributions for the proteins imaged separately. observed in an unbiased manner. When compared with the

Biophysical Journal 108(1) 76–84 80 Takahashi et al. volume distribution obtained for wildtype ESCRT-0 alone, the frequency of particles exhibiting a volume larger than we found that co-incubation of ESCRT-0 and VAMP2 300 nm3 (larger than ESCRT-0 heterotetramers or Ub- reduced the frequency of particles exhibiting a volume be- VAMP2) increased dramatically as compared with wildtype tween 100 to 300 nm3 (heterodimers and heterotetramers ESCRT-0 alone. These data strongly suggest that ESCRT- of Hrs and STAM), whereas the number of particles with 0 binds directly to Ub-VAMP2 in lipid bilayers to form a volume below 100 nm3 (VAMP2 alone) increased stable complexes. (Fig. 3, A and B). We also examined mutant forms of To determine whether the association between ESCRT- ESCRT-0, which contain mutations within its ubiquitin 0 and Ub-VAMP2 was dependent on the presence of its binding domains, and obtained similar results (Fig. 3, multiple ubiquitin binding domains, we took advantage of C–H). Together, these data strongly suggest that ESCRT- ESCRT-0 complexes harboring point mutations in the Hrs 0 does not bind to VAMP2. DUIM domain, the STAM VHS domain, and the STAM In contrast to studies using unmodified VAMP2, we UIM domain. The DUIM domain of Hrs exhibits the stron- observed a striking redistribution of particle volumes gest affinity for ubiquitin (~130 mM) (12). In contrast to the when we co-incubated wildtype ESCRT-0 with Ub- wildtype complex, ESCRT-0 harboring point mutations in VAMP2 (Fig. 4, A and B). In particular, we found that the DUIM domain of Hrs failed to promote the formation of enlarged particles when mixed with Ub-VAMP2 (Fig. 4, C and D). These data indicate that the capture of monoubiquitin-modified cargoes by ESCRT-0 requires the DUIM domain of Hrs. We also tested the impact of mutations in the VHS and UIM domains of STAM, which exhibit affinities of ~280 and 550 mM for monoubiquitin, respectively (12). Interestingly, the mutant complex was not able to bind stably to Ub-VAMP2, as reflected by the absence of particles larger than 300 nm3 (Fig. 4, E and F). As expected, based on these data, mutant ESCRT-0 harboring mutations in all ubiquitin-binding sites also failed to bind to Ub-VAMP2 (Fig. 4, G and H). Taken together, these data indicate that ubiquitin binding sites in Hrs and STAM function together to facilitate stable, ubiquitin-dependent cargo binding.

ESCRT-I and ESCRT-II bind to immobilized ubiquitin but are unable to stably associate with an integral membrane, ubiquitin-modiﬁed substrate In addition to regulating ubiquitin-dependent trafﬁcking to the lysosome, ESCRT-0 also recruits the ESCRT-I complex to endosomal membranes. ESCRT-I is composed of four distinct subunits, which co-assemble with a 1:1:1:1 stoichiometry (2). The combined molecular mass of the complex is 126 kDa, giving an expected molecular volume of 153 nm3. In all organisms, including C. elegans, at least one subunit of the complex harbors a ubiquitin-binding domain (Tsg101). Although ESCRT-I binds only weakly to membranes (2), we were able to visualize its accumulation by AFM, even in the absence of ESCRT-0 (Fig. 5 A). We found FIGURE 3 ESCRT-0 does not associate stably with VAMP2. (A, C, E, G that ESCRT-I exhibited a peak molecular volume of ~150 to and ) Representative AFM images of bilayers assembled in the presence 3 of VAMP2 and either (A) wildtype ESCRT-0, (C) ESCRT-0 harboring point 200 nm when imaged on a membrane surface (Fig. 5 B). mutations in the DUIM domain of Hrs, (E) ESCRT-0 harboring two-point These data are consistent with the formation of a 1:1:1:1 mutations in the VHS and UIM domains of STAM, or (G) ESCRT- heterotetramer and indicate that ESCRT-I does not undergo 0 harboring point mutations in all ubiquitin-binding domains. A shade- further oligomerization when bound to a lipid bilayer. Upon height scale bar is shown. Scale bar ¼ 200 nm. (B, D, F, and H) The frequency distributions of molecular volumes for particles observed by addition of ESCRT-I to bilayers containing VAMP2, AFM are shown. Each graph indicates the form of ESCRT-0 visualized we observed a downward shift in the particle volume distri- together with VAMP2. bution, indicating that VAMP2 did not bind to ESCRT-I

Biophysical Journal 108(1) 76–84 ESCRT-0 Sequesters Ubiquitin-Modiﬁed Cargoes 81

FIGURE 5 ESCRT-I does not associate stably with VAMP2 or ubiquitin- conjugated VAMP2. (A, C, and E) Representative AFM images of bilayers assembled in the presence of either (A) ESCRT-I alone, (C) ESCRT-I together with VAMP2, or (E) ESCRT-I together with Ub-VAMP2. A shade-height scale bar is shown. Scale bar ¼ 200 nm. (B, D, and F) The frequency distributions of molecular volumes for particles observed by AFM FIGURE 4 ESCRT-0 associates stably with ubiquitin-conjugated are shown. Each graph indicates the proteins present in the experiment. In VAMP2. (A, C, E, and G) Representative AFM images of bilayers assem- panel B, a representative Coomassie stained gel of recombinant ESCRT-I is bled in the presence of Ub-VAMP2 and either (A) wildtype ESCRT-0, also shown. The molecular mass markers are indicated by dashes on the (C) ESCRT-0 harboring point mutations in the DUIM domain of Hrs, (E) right. From top to bottom, the markers have masses of 50, 37, 25, and ESCRT-0 harboring two-point mutations in the VHS and UIM domains 20, and 10 kDa. The ESCRT-I components, Tsg101, Mvb12, Vps37, and of STAM, or (G) ESCRT-0 harboring point mutations in all ubiquitin-bind- Vps28, are highlighted. ing domains. A shade-height scale bar is shown. Scale bar ¼ 200 nm. (B, D, F, and H) the frequency distributions of molecular volumes for particles observed by AFM are shown. Each graph indicates the form of ESCRT- which assemble with a 1:2:1 stoichiometry in solution. 0 visualized together with Ub-VAMP2. The combined molecular mass of this complex is 116 kDa, giving an expected molecular volume of (Fig. 5, C and D). A similar downward redistribution of 140 nm3. The Vps36 subunit harbors a single ubiquitin in- particles was observed when ESCRT-I was incubated with teracting motif within its amino terminus. Using a GST Ub-VAMP2 (Fig. 5, E and F). To confirm that the intact pull-down assay, we confirmed that intact C. elegans ESCRT-I complex can bind to ubiquitin, we immobilized ESCRT-II binds directly to ubiquitin (Fig. 6 B). Further- GST alone or a mixture of mono- and di-ubiquitin fused more, we found that ESCRT-II assembles on lipid bilayers to GST on glutathione agarose beads, and incubated them using AFM, exhibiting a peak molecular volume of with purified ESCRT complexes (Fig. 6 A). We found that ~200 nm3 in the presence of 54% PC, 30% PE, 15% PS, ESCRT-I exhibited preferential binding to beads harboring and 1% PI3P, consistent with our previous work (24; see GST-tagged ubiquitin as compared with GST alone, as also Fig. 7, A and B). The measured volume is somewhat observed for ESCRT-0 components (Fig. 6 B). Together, higher than the expected volume; nevertheless, because of these data strongly suggest that ESCRT-I is incapable of the caveats associated with volume measurement by binding stably to an integral membrane cargo modified AFM, we suggest that the volume peak represents a single only by mono-ubiquitin. ESCRT-II complex. In contrast to our results with The ESCRT-II complex acts downstream of ESCRT-I and ESCRT-0, we found that neither the addition of VAMP2 harbors three distinct subunits, Vps22, Vps25, and Vps36, or Ub-VAMP2 resulted in an upward shift in particle

Biophysical Journal 108(1) 76–84 82 Takahashi et al.

FIGURE 6 The intact, early acting ESCRT complexes bind to ubiquitin. (A) GST alone or a mixture of GST fused to mono-ubiquitin or di-ubiquitin was immobilized on glutathione agarose beads. Fractions of the beads were resuspended in sample buffer following purification and separated by SDS- PAGE. Silver staining was conducted to determine the proteins associated with the resin. A molecular mass marker is shown in the left-most lane. From top to bottom, the markers have masses of 75, 50, 37, 25, and FIGURE 7 ESCRT-II does not associate stably with VAMP2 or ubiqui- 20 kDa. (B) Purified ESCRT-0, ESCRT-I, and ESCRT-II were incubated tin-conjugated VAMP2. (A, C, and E) Representative AFM images of bila- with the beads described in panel A. Following extensive washing, beads yers assembled in the presence of either (A) ESCRT-II alone, (C) ESCRT-II were eluted using reduced glutathione, and eluates were separated by together with VAMP2, or (E) ESCRT-II together with Ub-VAMP2. A SDS-PAGE. The presence of each ESCRT complex was determined by shade-height scale bar is shown. Scale bar ¼ 200 nm. (B, D, and F) The fre- immunoblotting, using antibodies against ESCRT-0 (Hrs and STAM), quency distributions of molecular volumes for particles observed by AFM ESCRT-I (Mvb12), and ESCRT-II (Vps22). To determine the relative are shown. Each graph indicates the proteins present in the experiment. amount of each ESCRT complex that bound specifically to GST-ubiquitin, In panel B, a representative Coomassie stained gel of recombinant dilutions of the fractions that bound to GST-ubiquitin were evaluated and ESCRT-II is also shown. The molecular mass markers are indicated by compared with the total amount bound to GST alone (100%). The dilution dashes on the right. From top to bottom, the markers have masses of 50, series enabled us to generate a standard curve, against which the amount of 37, 25, and 20 kDa. The ESCRT-II subunits, Vps36, Vps22, and Vps25, protein that bound to GSTalone could be measured. In all cases, the ESCRT are highlighted. complexes exhibited a preference for binding to GST-ubiquitin as compared with GST alone. including activated epidermal growth factor receptor, and volume (Fig. 7, C–F). These data indicate that ESCRT-II is mutations in c-Cbl that impair its ability to ubiquitin-modify not able to bind stably to integral membrane proteins, irre- substrates have been linked to oncogenic transformation spective of the presence of ubiquitin. Together, our findings (29). Similarly, numerous mutations that perturb transport highlight the specific ability of ESCRT-0 to bind and through the endocytic system are also associated with can- sequester integral membrane proteins in a manner depen- cer, potentially by affecting receptor degradation (30). dent on the presence of ubiquitin. Together, these data underscore a need to better understand the mechanisms that regulate ubiquitin-dependent receptor sorting. Mechanisms by which the ESCRT machinery handles DISCUSSION ubiquitin-modified cargoes have long been debated. In Modification of many transmembrane proteins by ubiquitin one widely cited model, ESCRT-0 delivers substrates conjugation serves as a sorting signal for transport into the directly to ESCRT-I and/or ESCRT-II (31). However, struc- lumen of multivesicular endosomes. Unlike proteasomal tural and thermodynamic studies fail to support this hand- targeting, which typically utilizes polyubiquitin chains, off scenario (32). Additionally, our data do not support the conjugation of one or a few ubiquitin moieties is sufficient possibility that ESCRT-0 transfers cargoes to other com- for entry into ILVs (28). In mammalian cells, the E3 ubiqui- plexes in the pathway, as neither ESCRT-I nor ESCRT-II tin ligase c-Cbl modifies several cell surface receptors, was capable of stably binding a mono-ubiquitin-modified

Biophysical Journal 108(1) 76–84 ESCRT-0 Sequesters Ubiquitin-Modified Cargoes 83 substrate. Alternatively, ESCRT-0 may act as a central Mol. Biol. 49:242–261. http://dx.doi.org/10.3109/10409238.2014. nucleating factor, sequestering ubiquitin-modified cargoes 881777. within subdomains on endosomal membranes, and subse- 3. Raiborg, C., and H. Stenmark. 2009. The ESCRT machinery in endosomal sorting of ubiquitylated membrane proteins. Nature. 458: quently aiding the recruitment of downstream ESCRT 445–452. complexes (12,32). In this model, eventual release 4. Leung, K. F., J. B. Dacks, and M. C. Field. 2008. Evolution of the mul- of ESCRT-0 from the site of vesicle formation would tivesicular body ESCRT machinery; retention across the eukaryotic require coordination with the internalization of cargoes lineage. Traffic. 9:1698–1716. into ILVs, which may involve ubiquitin-binding domains 5. Hanson, P. I., and A. Cashikar. 2012. Multivesicular body morphogen- esis. Annu. Rev. Cell Dev. Biol. 28:337–362. in ESCRT-I and/or ESCRT-II to restrict cargo diffu- 6. Agromayor, M., and J. Martin-Serrano. 2013. Knowing when to cut and sion. Further studies are necessary to validate such a run: mechanisms that control cytokinetic abscission. Trends Cell Biol. model. 23:433–441. Although our studies indicate that ESCRT-II forms 7. Votteler, J., and W. I. Sundquist. 2013. Virus budding and the ESCRT individual heterotetrameric complexes on membranes, pathway. Cell Host Microbe. 14:232–241. previous work suggests that the lipid environment can in- 8. Carlton, J. G., and J. Martin-Serrano. 2007. Parallels between cytoki- nesis and retroviral budding: a role for the ESCRT machinery. Science. fluence ESCRT complex assembly. Specifically, the pres- 316:1908–1912. ence of 25 mol % cholesterol was shown to promote 9. VerPlank, L., F. Bouamr, ., C. A. Carter. 2001. Tsg101, a homologue ESCRT-II clustering in vitro, even in the absence of its of ubiquitin-conjugating (E2) enzymes, binds the L domain in HIV endogenous recruitment factor ESCRT-I (33). On average, type 1 Pr55(Gag). Proc. Natl. Acad. Sci. USA. 98:7724–7729. the presence of cholesterol caused ~50 ESCRT-II com- 10. Bache, K. G., A. Brech, ., H. Stenmark. 2003. Hrs regulates multivesicular body formation via ESCRT recruitment to endosomes. J. Cell plexes to coalesce on a model membrane, and these clus- Biol. 162:435–442. tered complexes were capable of binding to ubiquitin. This 11. Katzmann, D. J., C. J. Stefan, ., S. D. Emr. 2003. Vps27 recruits finding is consistent with our data that numerous low-af- ESCRT machinery to endosomes during MVB sorting. J. Cell Biol. finity ubiquitin-binding domains must be present to stably 162:413–423. recruit and sequester ubiquitin-modified cargoes. How- 12. Mayers, J. R., I. Fyfe, ., A. Audhya. 2011. ESCRT-0 assembles as a heterotetrameric complex on membranes and binds multiple ubiquiti- ever, it remains unclear whether such a large number of nylated cargoes simultaneously. J. Biol. Chem. 286:9636–9645. ESCRT-II complexes assemble simultaneously on the en- 13. Shields, S. B., and R. C. Piper. 2011. How ubiquitin functions with dosome in vivo. Based on biochemical data and localiza- ESCRTs. Traffic. 12:1306–1317. tion studies, the majority of ESCRT-II is found in the 14. Stuffers, S., A. Brech, and H. Stenmark. 2009. ESCRT proteins in phys- cytoplasm at steady state, making it an unlikely candidate iology and disease. Exp. Cell Res. 315:1619–1626. to cluster cargoes before their deposition into ILVs (34). 15. Mayers, J. R., and A. Audhya. 2012. Vesicle formation within endo- By contrast, ESCRT-0 binds to membranes stably, and somes: an ESCRT marks the spot. Commun. Integr. Biol. 5:50–56. 16. Agromayor, M., N. Soler, ., R. L. Williams. 2012. The UBAP1 our data indicate that its multiple ubiquitin-binding do- subunit of ESCRT-I interacts with ubiquitin via a SOUBA domain. mains can bind cargo efficiently, unlike other ESCRT Structure. 20:414–428. complexes. Together, our findings strongly suggest that 17. Stringer, D. K., and R. C. Piper. 2011. A single ubiquitin is sufficient ESCRT-0 acts upstream of ESCRT-I and ESCRT-II to for cargo protein entry into MVBs in the absence of ESCRT ubiquiti- bind ubiquitin-modified cargoes during multivesicular en- nation. J. Cell Biol. 192:229–242. . dosome biogenesis. 18. Slagsvold, T., R. Aasland, , H. Stenmark. 2005. Eap45 in mammalian ESCRT-II binds ubiquitin via a phosphoinositide-interacting GLUE domain. J. Biol. Chem. 280:19600–19606. 19. Hierro, A., J. Sun, ., J. H. Hurley. 2004. Structure of the ESCRT-II SUPPORTING MATERIAL endosomal trafficking complex. Nature. 431:221–225. 20. Mayers, J. R., L. Wang, ., A. Audhya. 2013. Regulation of ubiquitin- Seven figures are available at http://www.biophysj.org/biophysj/ dependent cargo sorting by multiple endocytic adaptors at the plasma supplemental/S0006-3495(14)01185-0. membrane. Proc. Natl. Acad. Sci. USA. 110:11857–11862. 21. Hurley, J. H., and P. I. Hanson. 2010. Membrane budding and scission by the ESCRT machinery: it’s all in the neck. Nat. Rev. Mol. Cell Biol. ACKNOWLEDGMENTS 11:556–566. 22. Wollert, T., and J. H. Hurley. 2010. Molecular mechanism of multive- This work was supported by a grant from the NIH (GM088151 to AA) and a sicular body biogenesis by ESCRT complexes. Nature. 464:864–869. Newton International Fellowship (to H. T.). 23. Audhya, A., I. X. McLeod, ., K. Oegema. 2007. MVB-12, a fourth subunit of metazoan ESCRT-I, functions in receptor downregulation. PLoS ONE. 2:e956. REFERENCES 24. Fyfe, I., A. L. Schuh, ., A. Audhya. 2011. Association of the endosomal sorting complex ESCRT-II with the Vps20 subunit of ESCRT-III generates a curvature-sensitive complex capable of nucleating 1. Henne, W. M., N. J. Buchkovich, and S. D. Emr. 2011. The ESCRT ESCRT-III filaments. J. Biol. Chem. 286:34262–34270. pathway. Dev. Cell. 21:77–91. 25. Schneider, S. W., J. La¨rmer, ., H. Oberleithner. 1998. Molecular 2. Schuh, A. L., and A. Audhya. 2014. The ESCRT machinery: from the weights of individual proteins correlate with molecular volumes plasma membrane to endosomes and back again. Crit. Rev. Biochem. measured by atomic force microscopy. Pflugers Arch. 435:362–367.

Biophysical Journal 108(1) 76–84 84 Takahashi et al.

26. van Meer, G., D. R. Voelker, and G. W. Feigenson. 2008. Membrane 31. Hurley, J. H., and S. D. Emr. 2006. The ESCRT complexes: structure lipids: where they are and how they behave. Nat. Rev. Mol. Cell Biol. and mechanism of a membrane-trafficking network. Annu. Rev. Bio- 9:112–124. phys. Biomol. Struct. 35:277–298. 27. Weber, T., B. V. Zemelman, ., J. E. Rothman. 1998. SNAREpins: 32. Nickerson, D. P., M. R. Russell, and G. Odorizzi. 2007. A concentric minimal machinery for membrane fusion. Cell. 92:759–772. circle model of multivesicular body cargo sorting. EMBO Rep. 8:644–650. 28. Clague, M. J., H. Liu, and S. Urbe´. 2012. Governance of endocytic trafficking and signaling by reversible ubiquitylation. Dev. Cell. 23:457–467. 33. Boura, E., V. Ivanov, ., J. H. Hurley. 2012. Endosomal sorting complex required for transport (ESCRT) complexes induce phase-separated 29. Ogawa, S., L. Y. Shih, ., M. Sanada. 2010. Deregulated intracellular microdomains in supported lipid bilayers. J. Biol. Chem. 287:28144– signaling by mutated c-CBL in myeloid neoplasms. Clin. Cancer Res. 28151. 16:3825–3831. 34. Babst, M., D. J. Katzmann, ., S. D. Emr. 2002. Endosome-associated 30. Wegner, C. S., L. M. Rodahl, and H. Stenmark. 2011. ESCRT proteins complex, ESCRT-II, recruits transport machinery for protein sorting at and cell signalling. Traffic. 12:1291–1297. the multivesicular body. Dev. Cell. 3:283–289.

Biophysical Journal 108(1) 76–84