Distinct domain requirements for EAP45 in HIV budding, late endosomal recruitment, and cytokinesis

The scission of lipid membranes is a common biological process, often mediated by ESCRT complexes in concert with VPS4 assembling around the separation point. The functions of the ESCRT-I and ESCRT-III complexes are well established in certain of these cellular processes; however, the role of ESCRT-II remains contentious. Here, we devised a SNAP-tag fluorescent labelling strategy to understand the domain requirements of EAP45, the main component of ESCRT-II, in HIV egress, late endosome recruitment, and cytokinesis. We used TIRF microscopy to measure the spatial co-occurrence of the HIV structural polyprotein Gag with full length EAP45 in both fixed and live cells. Gag colocalises with the full length EAP45 comparably to ALIX, but this is lost on deletion of the EAP45 N terminus. Our findings reveal the H0 domain of the EAP45 protein is essential for linking to ESCRT-I during HIV budding and in anchoring at the late endosomal membrane, however in cytokinesis it is the Glue domain that is critical.


Introduction
ESCRTs (endosomal sorting complexes required for transport) are protein complexes involved in a variety of cellular functions, from the scission of the intercellular bridge during cytokinesis, to the formation of multi-vesicular bodies (MVB) and the extracellular budding of a number of viruses including human immunodeficiency virus (HIV). ESCRT activities involve the topological remodelling of membranous structures. The core of the ESCRT machinery comprises ESCRT-I (TSG101, VPS28, VPS37, and MVB12/UBAP1), -II (EAP20, EAP30, and EAP45) and -III (CHMP1-7) acting in a sequentially coordinated manner during membrane scission (1). The ESCRT machinery also has two other associated complexes, ESCRT-0 which precedes early cargo recruitment, and VPS4 AAA ATPase which universally acts at a late stage for membrane scission (2).
Many viruses take advantage of the properties of the ES-CRTs during their life cycle. In the case of HIV, a small number of Gag polyproteins form oligomers in the cytosol, possibly with the genomic RNA, before they are trafficked to the plasma membrane where further Gag molecules nucle-ate and membrane deformation occurs (3)(4)(5). Two late domains within the p6 domain of Gag, PTAP and YPXL, interact with the ESCRT-I protein TSG101 and ESCRT-associated protein ALIX, respectively, which then recruit ESCRT-III for the final scission to occur (3). Our understanding of the recruitment of the ESCRTs during Gag accumulation at the membrane has benefitted from biochemical analyses of budding events (6)(7)(8)(9), as well as from advanced fluorescence microscopy techniques which can examine the temporal dynamics of these events (10)(11)(12)(13). Interestingly, TSG101 and ALIX show different dynamics once engaged with Gag; TSG101 co-occurs with Gag as Gag multimerises (14), whereas ALIX is only recruited when Gag puncta reach maximum intensity (13). ESCRT-III components are recruited to the budding site and remain there for 3-5 mins (10) with recent data indicating recurrent transient recruitment of CHMP4B and VPS4 until successful cleavage occurs (5). Due to the transitory nature of these recruitments, the temporal co-occurrence between various components of ESCRT with Gag is very low, with only ∼1-4% of the total Gag at the plasma membrane detectably colocalised with an ESCRT component, as measured by direct stochastic optical reconstruction microscopy (dSTORM) (15).
Extensive work has been carried out in elucidating the function of ESCRT-I and ESCRT-III in HIV budding and in other cellular processes (16), however the role of ESCRT-II remains unclear. ESCRT-II comprises one copy of EAP30 and EAP45 (Fig. 1A) and two copies of EAP20 forming a Y shaped structure (17). Structural studies in yeast show ESCRT-I interacts with ESCRT-II via the C-terminal domain (CTD) of VPS28 and the N terminus of the VPS36 (EAP45 in metazoans) Glue and H0 linker domains (1,17,18) (Fig.  1A). A homologous region in mammalian EAP45 was also shown to interact with TSG101 by yeast two-hybrid screening (19). The remaining domains of EAP45 (WH1 and WH2) and EAP30 form the core structure of the ESCRT-II complex with EAP20 (17,20), which is responsible for recruiting ESCRT-III (17), though an alternative link is also documented (21). Although ESCRT-II's role as the canonical bridge in linking to ESCRT-I in MVB formation (22,23)  and cytokinesis is established (24), the same role has been controversial in HIV budding. An earlier siRNA knockdown study of EAP20 suggested ESCRT-II is dispensable for HIV budding in 293T cells (19), however, our recent study using CRISPR/EAP45 knockout (KO) HAP1 and T cells showed that EAP45 is required for efficient HIV budding and spreading (25,26). The specificity of this effect was demonstrated by the ability of EAP45 in trans to rescue the EAP45 KO cell-mediated virus budding and by its dependency on the H0 linker region for interacting with ESCRT-I (26). The discrepancy between these two studies may have arisen from minor amounts of residual ESCRT-II in the siRNA KD study still being sufficient for functional budding. Notably in these studies, cell proliferation was largely unperturbed by the depletion of EAP45 and yet there was up to a 10 fold decrease in virus spreading, suggesting that EAP45 is redundant for cell growth but is critical for viral multiplication and spread (26). This, together with the apparent requirement for early ESCRT proteins to ensure the specific recruitment of the Gag-pol polyprotein (26), underlines the importance of understanding the mechanistic roles of ESCRT-II in HIV budding. Further insights into these protein-protein interactions could help in developing a new class of antivirals to treat HIV. In addition to its link with ESCRT-I in viral budding, the Glue domain in EAP45 contains ubiquitin-binding sites and lipid-binding sites for membrane anchoring at the endosome (27,28). It was shown in yeast that neither phosphatidyli-nositol 3-phosphate (PI3P) nor ubiquitin-binding sites are required for endosomal anchoring of EAP45 but the link to ESCRT-I is (22). In metazoans, the Glue domain is believed to be involved in the recruitment of EAP45 at the endosomal membrane, presumably due to its multifaceted roles (28). However, direct visual evidence for this is lacking.
Building on our previous biochemical evidence supporting the role of ESCRT II in HIV budding, here we used a SNAP-tag to label EAP45 and widefield total internal reflection fluorescence (TIRF) (29) microscopy to visualise and quantify its spatial co-occurrence with Gag polyprotein at the plasma membrane in fixed cells. Furthermore, we used single particle tracking in live HeLa cells to study the dynamics of the interaction between EAP45 and Gag. Lastly, we used this labelling approach to examine the involvement of EAP45 in the endosomal membrane recruitment and during cytokinesis. Our findings show the H0 domain is crucial for anchoring at endosomes and in recruiting ESCRT-I in HIV budding, however the Glue domain is vital in the recruitment of ESCRT-II to the intercellular bridge in cytokinesis.

Results
Fusion of a SNAP tag at the N terminus of EAP45 does not affect its normal cellular localisation. The SNAP tag (∼20 kDa) is derived from the mammalian DNA repair protein O 6 -alkylguanine-DNA alkyltransferase and is widely used for imaging studies on both fixed and live cells due to the versatility of the commercially available substrates, providing a superior specificity for targeting the protein of interest (30). A slightly bigger GFP tag (∼27 kDa) was previously inserted at the N terminus of EAP45 without adversely affecting its cellular localisation (28) so we decided to SNAP label EAP45 at the same site (Fig. 1A). Overexpression of ESCRT protein can sometimes have adverse effects on cells (6,7). To preclude the possibility of the incorporation of a tag affecting the normal cellular localisation of EAP45, we transfected full length (FL) SNAP-EAP45 into HeLa cells and examined the recruitment of EAP45 during cytokinesis, one of the well-characterised activities in which ESCRT proteins are involved (31,32). A HeLa cell line constitutively expressing YFP-TSG101 was used as a control (31). We observed that TSG101 was recruited to the intercellular bridge as expected (Fig. 1B). Following substrate staining and specificity verification for the SNAP tag (Fig. S1), we also observed that EAP45 was recruited to a similar location (Fig. 1C).
The labelling method is specific, as the cells which were not successfully transfected showed no signal from the staining. These data show that the normal cellular activity of EAP45 is not affected by the insertion of an N-terminal SNAP tag.
SNAP-EAP45 rescues HIV budding in HAP1-EAP45 KO cells. We then verified the rescue functions of the expressor in HIV budding using a similar procedure to that which we reported previously (26). Transient expression of SNAP-FL EAP45 with WT provirus into the HAP1-EAP45 KO cells shows rescue of the virus budding defect ( Fig. 2A). This was accompanied by restoration of a normal rate of cleavage at the last step of Gag polyprotein processing (p24/p2), consistent with the effect observed with the non-tagged EAP45 ( Fig. 2B) (26). This suggests that fusion of a SNAP tag does not affect the function of EAP45 in rescuing HIV export from HAP1-EAP45 KO cells. To fluorescently label Gag for imaging, we used a piGFP construct in which a GFP tag was inserted between MA and CA of the Gag polyprotein, and has previously been shown to produce GFP-Gag intracellularly (35). To verify this construct in HAP1 EAP45 KO cells, we co-transfected KO cells with piGFP plasmid, VSV-G envelope, and FL EAP45 expressors. Two days post transfection the supernatants were harvested and used to transduce HeLa cells, and then GFP positive cells were counted (Fig. 2C)   Raw dual-colour images captured with the TIRF microscope were registered using calibration images from multicolour fluorescent beads and then segmented using the Trainable Weka Segmentation (33) plugin in ImageJ/Fiji (34). A nearest-neighbour distance analysis was implemented in Matlab to quantify the spatial proximity of Gag and EAP45 or other protein of interest. Particles found within one pixel (∼117 nm) from a particle in the opposite channel were labelled as "associated". (C) Co-occurrence analysis applied to cells expressing ALIX and Gag, as well as cells with SNAP-FL EAP45 and Gag (the same FL EAP45 values in Fig. 4E) for direct comparison). Representative images of the mCherry-ALIX cells with GFP-Gag are shown, along with insets showing instances of co-occurrence between the two markers. Yellow arrowheads in the insets, shown throughout this study, highlight puncta in both channels which show visible co-occurrence. The plot shows the percentage of associated Gag spots. Each data point represents a cell (n = 8 cells for ALIX, and n = 12 cells for FL EAP45) processed with the pipeline illustrated in (B). Welch's two sample t-test was used to compare the co-occurrence results from ALIX and Gag, to those of FL EAP45 and Gag.
that of the empty vector (EV). These data confirm that the GFP labelled virus works comparably with the wild-type untagged viruses in HAP1 EAP45 KO cells.
Co-occurrence analysis using widefield TIRF microscopy. After verifying the specificity of the SNAP label for EAP45 and the GFP label for Gag, we devised an imaging and analysis procedure to observe and measure the cooccurrence of EAP45 and Gag at the plasma membrane using widefield TIRF microscopy, as illustrated in Fig. 3A. To validate this procedure, we used a nearest-neighbour distance analysis (36,37) to search for the maximal co-occurrence from two different fluorescent labels on the same species. HeLa cells were transfected with piGFP, followed by staining with an anti-GFP nanobody conjugated with AlexaFluor647 dye. The two different fluorescent markers were expected to perfectly overlap in space, allowing us to test the maximal experimentally attainable colocalisation (Fig. S2A). Images were segmented to detect particles in both colour channels, and the center-to-center distances between Gag and nanobody particles were calculated (Fig. 3B). Any Gag particles found to have a neighbouring nanobody within a given search radius were labelled as "associated". To account for different densities of Gag particles per cell, the number of associated particles was expressed as a percentage of the total number of detected particles. A search radius of up to 3 pixels (350 nm) was used, and the percentage of associated Gag particles was plotted as a function of this radius (
B). The percentage of associated particles at a 1 pixel was ∼75%, increasing to ∼90% within 2 and 3 pixels. These percentages of association provide an estimated upper bound for the maximum proximity expected for a sample in which a marker of interest directly binds to Gag particles. Following validation using this ideal sample, we examined the cooccurrence of Gag and ALIX (Fig. 3C). ALIX is well known to be transiently recruited to the budding site by Gag during viral egress for subsequent recruitment of other ESCRT components (13). We transfected piGFP into an mCherry-ALIX expressing HeLa cell line (31). One day post-transfection, the cells were fixed and examined using TIRF microscopy. The nearest-neighbour analysis procedure described above (Fig.  3B) was applied to search for ALIX particles in the vicinity of Gag particles. Based on the average size of clusters of ES-CRT proteins (∼150 nm) (14), the diameter of HIV budding virions (∼120 nm) (38), and the diffraction-limited resolution of our microscope (2 pixels=234 nm), we selected a 1-pixel perimeter as a constraint for defining interactions between the two clusters. Our data show that around 2.5% of ALIX puncta associated with at least one Gag punctum within a 1pixel distance (Fig. 3C). This colocalisation rate is similar to the ∼1-4% reported previously for various ESCRT components (15).
TIRF microscopy shows high spatial proximity between EAP45 and Gag compared to functionally compromised mutants. Having validated our imaging and cooccurrence analysis procedure, we imaged HeLa cells transfected with piGFP and SNAP-EAP45 followed by staining with an AlexaFluor647-conjugated SNAP substrate (Fig. 4A-E). The cells were fixed at one day post transfection before imaging. We observed both EAP45 and Gag puncta at the plasma membrane, showing that EAP45 is recruited to and/or is active in the proximity of the plasma membrane. This is consistent with the observation made using quantitative EM analysis showing that components of ESCRTs are located at the plasma membrane (39). The nearest-neighbour distance analysis was applied between FL EAP45 and Gag (n=12 cells, Fig. 4A,C,& E) and compared to that of ALIX and Gag (Fig. 3C). We observed around 3% of EAP45 molecules associated with Gag within our 1-pixel threshold similar to the co-occurrence rate observed between ALIX and Gag (Fig.  3C), suggesting that in Hela cells both EAP45 and ALIX are recruited to the budding sites close to the plasma membrane in a similar abundance. The Glue/H0 deletion mutant (∆G∆H) is deficient at rescuing the HIV budding defect in EAP45 KO cells, as the H0 linker region responsible for this rescue by binding ESCRT-I is absent (26). As expected, removal of the Glue and H0 domains results in lower spatial proximity between EAP45 and Gag particles within a diffraction-limited region than that measured for the FL EAP45 (n = 15 cells, Fig. 4B,D,E, p = 0.026 between FL EAP45 and ∆G∆H), reinforcing the notion that this region plays a pivotal role in linking with Gag. The PTAP domain of Gag p6 is involved in recruiting TSG101 and thus cascading the recruitment of ESCRT-II and ESCRT-III in HIV budding. We hypothesised that the removal of the PTAP late domain (∆PTAP) would impair the co-occurrence between EAP45 and Gag. Our data however show that even though there is a decrease in the averaged co-occurrence between FL EAP45 with PTAP mutants (n = 11 cells), this is not significantly different from that of WT Gag and FL EAP45 (Fig. 4E, p = 0.36), indicating that EAP45 also co-occurs with Gag independently of PTAP. Consistently, this co-occurrence with the PTAP mutant relies on an intact Glue and H0 linker region, as the removal of this region completely abrogates this phenomenon (n = 14 cells, Fig. 4E, p = 0.018 between FL EAP45 and ∆PTAP-∆G∆H).
To further confirm the specificity of the co-occurrence observed between EAP45 and Gag, we also examined them in HAP1-EAP45 KO cells, with subsequent supplementation in trans with FL EAP45, followed by image acquisition and analysis ( Fig. 4F-J). We tested the proximity of Gag and FL EAP45 (n = 8 cells) versus that of Gag and ∆G∆H EAP45 (n=9 cells). Our data show a significantly higher (p = 0.018) number of EAP45 particles within the association threshold of Gag for FL (mean ∼6%) than for the ∆G∆H mutant (mean ∼ 1%), as illustrated in Fig. 4J, reinforcing the notion that the N terminus of EAP45 is required for the co-occurrence of EAP45 and Gag by providing a bridge linking EAP45 to ESCRT-I (26). The findings from the nearest-neighbour analysis in both HAP1-EAP45 KO and HeLa cells are consistent with the evidence obtained biochemically showing the N terminus encompassing the H0 linker region in EAP45 plays an important role in HIV budding.
Single particle tracking in live HeLa cells shows transient co-occurrence of Gag and EAP45. The assembly of Gag particles at the cell membrane during HIV budding has been reported to occur within a time frame of ∼10 mins (10,40), and the dynamics of ESCRT protein recruitment by Gag particles have been reported for TSG101 (10), VPS4A (5,11,14), some CHMP proteins of ESCRT-III (5, 10) and ALIX (13), with an association time at the membrane for different components varying from a few seconds up to 20 mins.
To further validate the observed association between Gag and EAP45, and to analyse its dynamics, we used TIRF microscopy in live HeLa cells to visualise the temporal interaction between GFP-Gag and SNAP-FL EAP45 at the plasma membrane. The HeLa cells were transfected with GFP-Gag and Gag expressors at a 1 : 5 ratio, similarly to previous reports (13,14), with SNAP-FL EAP45. The substrate for SNAP-EAP45 was added at >10 hours post-transfection when the Gag puncta are predominantly detectable at the plasma membrane (35). We analysed 20-min long live cell movies using single particle tracking. A sub-pixel fitting routine to calculate the centre position of the puncta was applied in which the distance between the centres of an EAP45 particle and a Gag particle were plotted over time. In the live cell movies, both Gag and EAP45 clusters move simultaneously, making it difficult to see how their relative positions are correlated over time. To visualise the relative motion of the particles more clearly, we developed an algorithm that tracked the position of a Gag particle and repositioned the field of view such that the detected particle remained in the centre of the image from frame to frame. We used the Gag particle position as a reference point against which the motion of the EAP45 cluster could be measured. We call this algorithm a "co-moving frame analysis", and its steps are summarised in Fig. S3. For each region of interest (ROI) extracted, the distance between the Gag and EAP45 tracks was plotted as a function of time. The particles were considered to be "associated" if the distance between them was within the diffractionlimited resolution of our microscope, in this case ∼100 nm (2 pixels). Our analysis shows a transient association, or spatial co-occurrence, between Gag and EAP45 particles over a time scale of 1 to 5 mins, with a mean consecutive association time of ∼1.7 mins (Fig. 5C). This is consistent with the time frame of co-occurrence reported for other ESCRT proteins. Within 14 analysed ROIs (Fig. S4), there appear to be three classes of movements between EAP45 and Gag: i) EAP45 associates with Gag multiple times (Fig. 5A-B top row; 7 out of 14 events); ii) EAP45 approaches Gag and remains associated with Gag for a span of time before leaving again (

EAP45, similar to that of Gag and other ESCRT components during viral egress.
The Glue domain is not required for the localisation of EAP45 at the late endosome but is required for cytokinesis. ESCRT-II components are also involved in MVB formation and cytokinesis. It has been hypothesised the Glue domain plays an important role in those activities. However, the exact domain requirements of EAP45 remain ill-defined.
We used the specific labelling from SNAP tag to investigate this in more detail. EAP45 largely shows a cytosolic distribution and colocalises with RAB7, a late endosome marker, consistent with previous observations (Fig. 6A) (28). Unexpectedly, EAP45 still localises to late endosomes despite removal of the Glue domain (∆G) (Fig. 6A middle row), sug-gesting that the Glue domain, which has binding sites for both ubiquitin and lipids, is not required for the recruitment of EAP45 to late endosomes. Further deletion of the H0 linker region (∆G∆H) greatly reduces the observed colocalisation ( Fig. 6A bottom), suggesting the H0 linker region rather than the Glue domain on its own is crucial in anchoring EAP45 at late endosomes. The H0 linker region is required for binding to subunits of ESCRT-I in vitro (19,20), therefore it is possible that ESCRT-I colocalises with and recruits EAP45 at late endosomes. We used a HeLa YFP-TSG101 cell line to examine this possibility. Upon transfection of SNAP-FL EAP45 expressor, we observed a significant cytosolic colocalisation of EAP45 with TSG101 at late endosomes (Fig.  6B). This colocalisation is also observed in the ∆G mutant (Fig. 6B middle) but not in the ∆G∆H mutant (  bottom), suggesting that ESCRT-I plays an important role in the recruitment of ESCRT-II to late endosomes and that this recruitment depends on an intact H0 linker region. In addition to its role in endosomal membrane anchoring, EAP45 is recruited to the intercellular bridge during cytokinesis (Fig.  1C) (41). To better understand which domains in EAP45 are required for this, we transfected expressors of EAP45 and the derived mutants into HeLa cells and imaged the cellular distribution of EAP45 at the intercellular bridge during cytokinesis ( Fig. 6C-E). Full length EAP45 is recruited to the intercellular bridge as expected, but this colocalisation is significantly decreased when the Glue domain (∆G) is removed, suggesting it is important for efficient recruitment in cytokinesis in contrast to the findings in late endosomes. Further removal of the H0 linker region (∆G∆H) does not significantly decrease recruitment, further stressing the importance of the Glue domain in cytokinesis.

Discussion
The roles of ESCRT-II in MVB formation, cytokinesis, and HIV budding have all been areas of controversy, with previous reports suggesting it is redundant (19,31,32  ESCRT-II in a bridging role between ESCRT-I and ESCRT-III as originally reported in cargo sorting (17,27). Our previous data showed that expressing EAP45 in trans rescues an HIV budding defects in HAP1 EAP45 KO cells and biochemical analysis attributed this to the H0 linker domain in EAP45 (26). Using TIRF microscopy we now expand these observations and provide additional clues as to the functions of the different domains of EAP45 in various roles. We observed EAP45 expressed in close proximity to, or at, the plasma membrane in concert with the other ESCRT members documented previously (39). The relatively low co-occurrence rate (around 1 to 6%) between ALIX and Gag (Fig. 3C) and EAP45 and Gag (Fig. 4E & J) is consistent with the rate reported by others using dSTORM to measure both endogenously and exogenously expressed ESCRT components with Gag (1.5 to 3.4%) (15). Considering the maximum experimentally attainable colocalisation measured using the Alex-aFluor647 nanobody targeting GFP (Fig. S3), this number is an underestimate of the total events by a factor of 25%. Such a low rate of co-occurrence however is not uncommon (15) and is most likely attributable to the documented transient nature of recruitment of ESCRT proteins. Our data in HAP1 EAP45 KO cells show a higher co-occurrence rate between FL EAP45 and Gag than in HeLa cells (Fig. 4). This highlights the necessity of EAP45's involvement in HIV budding in HAP1 EAP45 KO cells, since in HeLa cells endogenously expressed but undetected EAP45 could also function de novo. Interestingly, the co-occurrence between EAP45 and Gag does not strongly depend on its canonical recruiting motif PTAP, as ∆PTAP Gag reduces the co-occurrence but not to a statistically significant level (Fig. 4E). Given that the level of recruitment and dynamics at the plasma membrane between WT and ∆PTAP Gag are similar (10,11,42), this observation could be explained by the recruitment of EAP45 by ALIX via its interaction with TSG101 of ESCRT-I in the absence of the PTAP motif (43). Our co-occurrence analysis between EAP45 and Gag confirms that the N terminal region of EAP45 is crucial in interacting with Gag at the plasma membrane. This is consistent with the biochemical evidence whereby deletion of the Glue and H0 regions completely abolishes the rescue effect mediated by EAP45 in the HAP1 EAP45 KO cells (26), substantiating the importance of this region in linking to ESCRT-I in HIV budding. Interestingly, this interaction is also important in aiding the recruit-ment of ESCRT-II at the endosomal membrane (Fig. 6B). However, it is less important in the context of ubiquitin and lipid binding, which mirrors previous reports on the recruitment of ESCRT-II at the endosomal membrane in yeast in which H0 mediated anchoring appears more critical than the Glue domain integrity (22). The Glue domain however plays a crucial role in cytokinesis (Fig. 6C-E) possibly by interaction with the unique local lipid components (44). The Glue domain thus unexpectedly plays distinctive roles in different cellular processes. Our live imaging experiments revealed the dynamic recruitment of EAP45 to the Gag assembling sites (Fig. 5 &  7). On average, EAP45 is within the association distance of Gag particles for a period of 1.7 mins. Remarkably, for 30% of the total recording time (∼5 mins) EAP45 was localised within ∼100 nm of Gag. We observed three distinct classes of movements of EAP45. Some show a similar behaviour to CHMP4B/VPS4 and are recruited multiple times until scission happens (5); others behave similarly to TSG101 in that, once recruited, they serve as an adaptor for the recruitment of downstream components and therefore spend a relatively longer time with Gag (14). The last class in which EAP45 stays with Gag once recruited may be attributable to the molecules that are packaged within the virions, similarly to that observed with ALIX and TSG101 (13,45). A future study is needed to pinpoint which class of dynamic interaction(s) is necessary and responsible for forming of infectious virus particles as well as the recruitment kinetics of ESCRT-II in the context of other ESCRT proteins.
In summary, we used a SNAP-tag label to image the main component of ESCRT-II, EAP45, and studied its spatial and temporal co-occurrence with Gag in HIV budding, as well as the domain requirements for its anchoring at the endosomal membrane and in cytokinesis. Our data strengthen the importance of ESCRT-II in linking to ESCRT-I and -III in HIV budding and illustrate similarities to and contrasts with the processes of MVB formation and cytokinesis.

Materials and Methods
Plasmids and cells. All cells used in this study were grown in DMEM with 10% FCS at 37°C in 5% CO 2 incubator. HeLa YFP-TSG101 and mCherry-ALIX cells were gifts from Juan Martin-Serrano at KCL. pBH10∆BglII-WT and pCMV-VSV-G, pEF-EAP45-HA were reported previously (25). piGFP was a gift from Benjamin Chen at Mount Sinai and was described previously (35). A sole Gag expressor (HVPgagpro-) was also described previously (46). SNAP-FL-EAP45, SNAP-∆G-EAP45 and SNAP-∆G∆H-EAP45 were constructed by VectorBuilder. The pBH10Gag-GFP-PTAP mutant was cloned by ligating the BssHII and SpeI digested fragments from the piGFP vector to the same sites in pBH10Gag-PTAP, which had been rendered Gag/pol deficient by creating a deletion between BclI and BalI sites.
Transfection. For transfection, the HeLa cells were transfected with Fugene HD (Promega) and HAP-EAP45 KO cells were transfected with Turbofectin (OriGene). Both the SNAP-expressor rescue experiment and the transfection of piGFP in HAP1-EAP45 KO cells were done similarly as described before (26). Briefly, 150 ng of pBH10∆BglII-WT or 250 ng of piGFP was co-transfected with 45 ng of SNAP-FL-EAP45 or pEF-EAP45-HA, together with pCMV-VSV-G, respectively, in a well of a 24 well plate. The supernatant was harvested for virion purification and cells were lysed for immunoblotting at two days post transfection. Immunoblotting. Immunoblotting was done essentially as reported previously (26). Briefly, at 2 days post transfection, supernatants were harvested for virion purification and monolayers were lysed in cell culture lysis buffer (CCLR, Promega) for immunoblotting. The primary antibodies used in this study were anti-p24 (NIBSC), anti-GAPDH (Abcam), anti-HA (Thermofisher), and secondary antibodies used were goat anti-mouse IgG-HRP (Cell Signalling Technology) and goat anti-rabbit IgG-HRP (Thermofisher). Densitometric analysis of the target bands was done in ImageJ (34) with traces of p24 and p24-p2 highlighted to show the differences in the final cleavage product.

Confocal microscopy.
HeLa or HeLa-YFP TSG101 cells were seeded into 8-well slides (EZ slides, Merck). At 24hour post seeding, the cells were transfected with 100 ng of either SNAP-FL, SNAP-∆G, or SNAP-∆G∆H-EAP45 expressor. This was followed by fixation with PFA and staining with SNAP-Surface AlexaFluor647 (NEB) at 4 µM concentration for SNAP-EAP45 and DAPI for nuclei on the following day. A primary antibody to RAB7 (Abcam) or anti-Tubulin (Sigma) was also added on some occasions followed by staining with the secondary antibody conjugated with either AlexaFluor488 or AlexaFluor647 (Thermofisher) before imaging using Leica SP5 confocal fluorescence microscope.
Widefield TIRF fixed cell imaging. Fixed cell images using widefield TIRF microscopy were acquired with a custom-built microscope based on an Olympus (Center Valley, PA) IX-73 frame with a 488-nm laser (Coherent Sapphire 488-300 CW CDRH), a 561-nm laser (Cobolt Jive 500 561 nm), and a 647-nm laser (MPB Communications Inc. VFL-P-300-647-OEM1-B1). Laser light entering the microscope frame was reflected from a dichroic mirror (Chroma ZT488/561/647rpc) into a 100x 1.49 NA oil objective lens (Olympus UAPON100XOTIRF), and then onto the sample. Light emitted by the sample passed through the dichroic and a set of 25 mm band-pass filters (Semrock FF01-525/45-25, Semrock FF01-600/37-25, and Semrock FF01-680/42-25 respectively for the 488, 561, and 647-nm laser lines) before reaching the microscope side port. Images were then relayed onto two cameras (Andor iXon Ultra 897) by a 1.3x magnification Cairn Twincam image splitter with a dichroic beam splitter (T565spxr-UF3) which allows imaging of emitted wavelengths between 390-555 nm in the transmission port and 575-1000 nm in the reflection port. The pixel size in the sample plane was measured to be 117 nm.
Nearest-neighbour distance analysis of fixed cell images. We used an object-based nearest-neighbour analysis method in which we segment the registered images for each colour channel, locate the centroid of the fluorescent spots, and search for particles in the opposite colour channel within a neighbourhood of 3 pixels (∼350 nm). A distance threshold within the diffraction-limited resolution of our microscope (1 pixel∼117 nm) was selected to designate instances in which these molecules could be interacting, and any particles found within this threshold from the centroid of the opposite channel were labelled as "associated". Raw dual-colour images were registered to correct for chromatic offset using the DoM_Utrecht plugin version 1.1.6 (https://github.com/ekatrukha/DoM_Utrecht) in Fiji (34), using 7 pixels as the maximum shift between images, a value of 15 for the SNR filter, and a PSF standard deviation of 1 pixel. The registered images were then segmented using the Trainable Weka Segmentation plugin from Fiji/ImageJ (33) with the default Fast Random Forest classifier. Three classes were defined (small blobs, large blobs, and background) and images were annotated to identify single fluorescent particles in focus (small blobs), large aggregates of particles (large blobs), and background. The segmented images were then imported into a custom Matlab (The MathWorks, Natick, USA) script which implemented a nearest neighbour algorithm to check, for each identified segmented spot, the distance from its nearest neighbours in the opposite colour channel. The function natsortfiles was used to sort the data (47). All distances for each image were recorded and exported in a .csv file. The .csv files for all experiments were imported into an R (R Core Team, 2013) script that was used to plot the files and run the statistical analysis on the different experimental conditions.
Widefield TIRF live cell imaging. Live cell images using widefield TIRF microscopy were acquired with a custombuilt microscope based on an Olympus (Center Valley, PA) IX-71 frame with a 488-nm laser (Toptica iBeam SMART), a 561-nm laser (Coherent OBIS LS), and a 640nm laser (Cobolt MLD). Laser light entering the microscope frame was reflected from a dichroic mirror (Chroma ZT405/488/561/640rpc) into a 100x 1.49 NA oil objective lens (Olympus UAPON100XOTIRF), and then onto the sample. Light emitted by the sample passed through the dichroic and a set of 25 mm band-pass filters (Semrock FF01-525/45-25, Semrock FF01-600/37-25, and Semrock FF01-680/42-25 respectively for the 488, 561, and 647-nm laser lines) before reaching the microscope side port. Images were then relayed onto an sCMOS camera (Hamamatsu Orca Flash v4.0). Live cell TIRF images were acquired in two colour channels every 5 or 10 seconds with a 100 ms exposure time over 15-20 minutes. The pixel size at the image plane was is ∼50 nm (6.45 µm pixels at the camera de magnified by the 100x objective and a 1.3x relay lens).
Single particle tracking in live cell images. The dualcolour live cell movies were registered to correct for chromatic offset using the DoM_Utrecht plugin version 1.1.6 (https://github.com/ekatrukha/DoM_Utrecht) in Fiji (34), and regions of interest of ∼80x80 pixels (∼1.3x1.3 µm 2 ) were manually extracted in EAP45 particles were found in the vicinity of a Gag particle. The Gag particles and the EAP45 particles were tracked independently using Track-Mate v4.0.0 (48) using a radius of 4 pixels (∼200 nm), a threshold value of 2 for particle detection with sub-pixel localisation, and the LAP tracker for tracking. The resulting tracks were imported into Matlab for nearest neighbour analysis. A custom Matlab script was written to import Track-Mate tracks. The distances between x,y positions of the tracks in each channel are calculated and output as a .csv file. For ease of visualisation, a new frame is generated centred around the x,y position of the Gag track such that the Gag particle is always at the center of the frame, and the position of the EAP45 particle is shown relative to the new frame of reference. This co-moving frame analysis makes it easy to visualise the motion of two particles over time in a 2D image. We show the average image of the immobilised Gag particle, and the relative motion of EAP45 on top of this image, colour coded as a function of time.
Statistical Analysis. For the case in which only two experimental data sets were compared, such as in Fig. 3C for ALIX and FL EAP45, and Fig. 4J for FL EAP45 and ∆G∆H, Welch's two sample t-test was used with ns: non-significance, * for p < 0.05, ** for p <0.01, and *** for p < 0.001. Boxplots were overlaid on the raw data points, with the bold horizontal line showing the mean, and whiskers represent the largest and smallest value within 1.5 times the interquartile range above 75th and 25th percentile, respectively. For the case in which more than two conditions were compared (Fig. 4E), a One-Way ANOVA with Dunnett Comparison was used to compare the FL EAP45 condition to each of the functionally compromised mutants (∆G∆H, ∆PTAP, ∆PTAP-∆G∆H) with ns: non-significance, * for p < 0.05, ** for p <0.01, and *** for p < 0.001. All statistical tests were performed using R.
Data availability statement. The Matlab (Natick, MA) code used for the nearest-neighbour analysis for the fixed cell co-occurrence experiments, and the co-moving frame analysis for the live cell tracking experiments are available at our online repository. The R scripts used to generate the plots in Fig. 3, 4, 5, S1, S2, and S4 can also be found in this repository.     S3. Co-moving frame analysis for single particle tracking data. (A) Registered dual colour still frame from live cell movie showing a single EAP45 particle in the vicinity of a Gag particle. (B) The centre position of the tracks for a Gag particle (cyan cross) and EAP45 particle (magenta circle) overlaid on a grey scale still frame of the Gag channel. A dotted rectangle indicates a region around the centre position of the Gag channel which is extracted in each frame to immobilise the Gag particle, and the position of the EAP45 molecule is then plotted relative to the new frame of reference in (C). (D) shows the track of the EAP45 particle motion colour coded from start to finish of the track duration, overlaid on an average image of the Gag channel.  . S4. Co-moving frame analysis reveals three classes of EAP45 movement relative to Gag. The motion of the EAP45 relative to the Gag cluster can be classified into three categories: ((A), cyan box) EAP45 particles which hover in the vicinity of the Gag cluster for short spans of time, and either oscillate in and out of proximity, or move away from the Gag cluster, ((B), green box) EAP45 particles which approach the virus from a distance and remain in the 2-pixel vicinity of the virus particle for several minutes, and ((C), red box) EAP45 particles which remain within 2 pixels of the virus particle throughout the observation time interval.