Endosomal sorting complexes required for transport (ESCRT) machinery of the HIV-1 budding

Human immunodeficiency virus type 1 (HIV-1) infection causes Acquired Immune Deficiency Syndrome (AIDS). According to the recent statistics published by the United Nations in 2015, there are currently 36.9 million people suffering from AIDS, and tens of millions of people have died from the disease since its discovery. Most patients are treated by a cocktail of inhibitors targeting HIV-1's enzymes protease (PR) and reverse transcriptase (RT) and integrase (IN). As of now, there exists no cure or vaccination for HIV. An interesting finding that HIV-1 hijacks the endosomal sorting complexes required for transport (ESCRT) to efficiently release new virus particles was intensively studied over the past 15 years for its potential to develop a therapeutic target. With current progress in fluorescent methodologies, it is now possible to study the dynamic of individual virions as they assemble in cells and interact with ESCRTs to facilitate their release. The goal here is to create sufficient understanding of the assembly and release mechanism for potential inhibition of HIV-1 production.

ENDOSOMAL SORTING COMPLEXES REQUIRED FOR TRANSPORT (ESCRT) MACHINERY OF THE HIV-1 BUDDING by Pei-I Ku A dissertation submitted to the faculty of The University of Utah in partial fulfillment of the requirements for the degree of Doctor of Philosophy in Physics Department of Physics and Astronomy The University of Utah December 2018 Copyright © Pei-I Ku 2018 All Rights Reserved The University of Utah Graduate School STATEMENT OF DISSERTATION APPROVAL The following faculty members served as the supervisory committee chair and members for the dissertation of Pei-I Ku . Dates at right indicate the members’ approval of the dissertation. Saveez Saffarian , Chair 10. 28. 2016 ooo Date Approved Benjamin Bromley , Member 10. 28. 2016 0000 Date Approved Markus Babst , Member 10. 28. 2016 00 Date Approved Michael Vershinin , Member Date Approved Sabina Swierczek , Member Date Approved Shanti Deemyad , Member 10. 28. 2016 000 Date Approved The dissertation has also been approved by Benjamin Bromley, Chair of the Department of Physics and Astronomy and by David B. Kieda, Dean of The Graduate School. ABSTRACT Human immunodeficiency virus type 1 (HIV-1) infection causes Acquired Immune Deficiency Syndrome (AIDS). According to the recent statistics published by the United Nations in 2015, there are currently 36.9 million people suffering from AIDS, and tens of millions of people have died from the disease since its discovery. Most patients are treated by a cocktail of inhibitors targeting HIV-1’s enzymes protease (PR) and reverse transcriptase (RT) and integrase (IN). As of now, there exists no cure or vaccination for HIV. An interesting finding that HIV-1 hijacks the endosomal sorting complexes required for transport (ESCRT) to efficiently release new virus particles was intensively studied over the past 15 years for its potential to develop a therapeutic target. With current progress in fluorescent methodologies, it is now possible to study the dynamic of individual virions as they assemble in cells and interact with ESCRTs to facilitate their release. The goal here is to create sufficient understanding of the assembly and release mechanism for potential inhibition of HIV-1 production. TABLE OF CONTENTS ABSTRACT…………………………………………………………………………….iii ACKNOWLEDGEMENTS……………………………………………………………vii Chapters 1. INTRODUCTION OF THE HIV-1 BUDDING AND ESCRT MACHINERY……..1 1.1 HIV-1 and HIV-1Budding………………………………………………….1 1.1.1 HIV-1.……………………………………………….…….............1 1.1.2 The HIV-1 budding.…………………….…....................................2 1.2 ESCRT Machinery……………………………………………….……….8 1.2.1 ESCRT-0, I, II, III, and VPS4……………………………………….9 1.2.2 ALIX and its multipathways in the HIV-1budding……...…...........18 1.2.3 UbL and Ub in the HIV-1 budding.…………………….…….......22 2. HIV-1 LIKE PARTICLES ASSEMBLY.………………………………………...…26 2.1 Abstract…………………….…………………………….………………26 2.2 Introduction.……………….………………………………….…….……27 2.3 Materials and Methods…….………………………………….…….……30 2.3.1 Fusion of VPS4 with fluorescent proteins.…………………..........30 2.3.2 Testing the functional VPS4 fusions.………………………..........30 2.3.3 VPS4A-h37-mCherry clonal cell line………………………..........31 2.3.4 Cell culture and imaging conditions....…………………….…….31 2.3.5 Microscope description………….……………………….…........32 2.3.6 Calculating the amount of Gag incorporated in single VLPs formation of the plasma membrane……………………………….33 2.3.7 Shell-filling model for VLP assembly...….………………….…….35 2.3.8 Parameter estimation.……………….…………………….…........36 2.3.9 Restrictions on parameter estimates……………………..…….......39 2.4 Results………………..…….……………………………….……………39 2.4.1 VLPs formation kinetic…………….…………………..……........41 2.4.1.1 Nucleation kinetics…………………...….………….......41 2.4.1.2 Gag polymerization………………...….………….......41 2.4.1.3 Kinetics of recruitment of VPS4….......………………..42 2.4.1.4 Lifetime of full assembly…………......………………..42 2.4.2 Pauses during HIV-1 VLPs formation……………………………..42 2.4.3 Pause are independent of Gag late domains………………………44 2.4.4 The effect of Gag concentration on the assembly kinetics………...44 2.5 Discussion……………….………...……………………….……………45 2.5.1 Model 1: Lattice stress is relieved through incorporation of defects during pauses…………………………...…………………………47 2.5.2 Model 2: An enzymatic event essential for further polymerization……………….…………………………………...47 2.6 Conclusion….………...….…………………….……………….………. 48 2.7 Acknowledgements..….………………………………………….……….49 3. VISUALIZING ALIX RECRUITMENT DURING THE HIV-1 BUDDING……….56 3.1 Abstract…………...….…………....………………...………………….56 3.2 Introduction……....….…………....………………...……………….….56 3.3 Materials and Methods ….…...………………………………………....60 3.3.1 Infectivity assay……………….………….……………………… 60 3.3.2 Cell culture and transfection…………………….……………….. 60 3.3.3 Microscope description…………….…………….…………......... 61 3.4 Results….……......…...…….…...………………………….…………... 62 3.4.1 ALIX linked to eGFP at its C terminus with a 30 amino acid helical linker supports production of HIV-1 with equal infectivity as wild-type ALIX………...…………………………...………….62 3.4.2 ALIX incorporates transiently at the end of Gag assembly during the formation of HIV-1 VLPs ……....……………………………64 3.4.3 ALIX incorporates transiently at the end of Gag assembly during the formation of EIAV VLPs …...…….………………...……….65 3.4.4 Comparison between ALIX-h30-eGFP recruitment to EIAV and HIV-1 Gag during VLP assembly …………….….…….….……. 66 3.5 Discussion….……………........................................................................66 3.6 Aknowledgements...…….………..............................................................68 4. MECHANISM OF ALIX RECRUITMENT IN HIV-1………………………………75 4.1 Abstract...………………..…………..…………………………………75 4.2 Introduction....………….…………….…...………………….…...……76 4.3 Preliminary Data and Discussion……...……...…………….……….....79 4.3.1 ALIX transient recruitment and the interaction with CHMP4…....79 4.3.2 ALIX recruitment and its activated conformation.……………......80 4.3.3 ALIX transient recruitment and the interaction with NCGag ……...81 4.4 Conclusion….………...…….………………………….…………......…82 Appendices A. TESTING THE FUNCTIONAL VPS4 FUSIONS.………………….…….………...84 v B. DUAL PENETRATION DEPTH TIRF, TIRF 360, AND THE TIRF CLAMP……..87 C. SHELL FILLING MODEL FOR VLP ASSEMBLY WITH 65% FILLING…...........90 D. RECRUITMENT OF VPS4-h37-mCHERRY ALONG WITH ALIX-h30-eGFP…..97 REFERENCES……………………………………………………………..…………99 vi ACKNOWLEDGEMENTS I would like to take the opportunity to extend my sincerest esteem to my advisor, Dr. Saveez Saffarian, for all the guidance and support he has provided to me. I have learned and grown very much as both a student and a scientist under his tutelage. I would also like to express my gratitude to my committee members, Dr. Ben Bromley, Dr. Markus Babst, Dr. Michael Vershini, Dr. Sabina Swierczek, and Dr. Shanti Deemyad, as well as Dr. Jordan Gerton, for sharing their knowledge with me. I am grateful to my coworkers, Xiaolin, Shilpa, and Akshay, for kindly sharing their expertise and friendship. Lastly, I want to express my best appreciation to my parents, my brother, and my sister for their constant love and support. Especially to my dearest darling mom, who has always stood by me like a pillar in times of need and to whom I owe my life for her endless love, encouragement, and blessings. CHAPTER 1 INTRODUCTION OF THE HIV-1 BUDDING AND ESCRT MACHINERY 1.1 HIV-1 and HIV-1 Budding 1.1.1 HIV-1 HIV-1 utilizes infected host cells to reproduce and create more virus particles through a series of steps in its life cycle. The HIV-1 life cycle begins with the virus binding to surface receptors of specific cells, such as CD4 cells, and the viral envelope fuses with the cell plasma membrane to allow the viral contents to be released into the cell 1-2. Once entering the cell, RNA (HIV-1 has two copies of plus-strand viral genomic RNA with roughly 10,000 nucleotides in length) is reverse transcriptased into double strands of DNA on the way to being transported into the host cell’s nucleus by microtubules. DNA is then integrated into a host chromosome and starts its incubation period, which could span several years 3. When the host cell is activated, the integrated DNA is transcribed and translated to form new viral RNA and viral proteins and then translocated to the cell surface to assemble at the virus budding site. The noninfectious virus particles, which are called immature virus vesicles, are first produced when the viruses bud away from the host cell membrane and are released. Then the protease enzyme cleaves the HIV-1 Gag structural polyproteins at five specific sites, which results in the structure conformational change from immature HIV-1 into mature HIV-1 particles. 2 HIV-1 includes at least nine genes to govern distinct functions during its life cycle: gag, pol, env, vif, vpr, tat, vpu, rev, and nef. The gag gene synthesizes into a precursor polyprotein Gag (also named as Pr55Gag or p55 for its 55kDa size) to direct viral construction of immature viruses. The pol gene encodes protease (PR), reverse transcriptase (RT), and integrase (IN). Co-expression of the gag and pol gene (called GagPol or p160) results in proteolytical cleavage by the viral PR before or during maturation 4 . The env gene synthesizes into two kinds of glycoproteins, surface (SU; gp120) and transmembrane (TM; gp41) glycoproteins. Each of these two proteins can combine into a 160kDa gp160 precursor protein which is in a spike shape. There are ~10 of gp160 studs on the surface of each viral envelope, and they are responsible for recognizing the binding site on the host cell membrane during entry 4-6. These three genes, gag, pol, and env, are common in all retroviruses. The other six genes are unique to HIV-1 and play a regulatory or accessory role in the HIV-1 life cycle. 1.1.2 The HIV-1 budding The budding process is the last, but essential, step of the HIV-1 life cycle, which mainly includes four independent steps: assembly, budding, release and maturation. Most retroviruses use the budding process to escape from infected cells and spread infections to neighboring cells, while some other viruses are released by the method of host-cell lysis 7. Because of the increase in understanding of Gag proteins and endogenous proteins participating in the budding events, the point of view that the budding process happened spontaneously has been challenged 8. Gag structural polyprotein, which is one of HIV-1 proteins, is a main actor in orchestrating the whole virus budding process. There are 2500 3 copies of Gag in a single ~130nm HIV-1 particle, which accounts for 50% of the virion mass 9 (the other 30% is viral membrane lipids and 20% are viral and cellular proteins, and RNA accounts for 2.5% of virions mass 4, 10 ). Gag initiates the HIV-1 budding process starting from the Gag relocation to the budding site at the plasma membrane, followed by the Gag-Gag polymerization and the Gag-cellular protein interaction, and ending with virus abscission and release as an immature virus 11-13 . Gag, alone, without any other virus proteins, RNA, or enzymes, is sufficient to form immature virus-like particles (VLPs) 14. And when Gag fuses with an extra fluoresce protein, it can still assemble as a morphologically accurate VLP 15-16. This character is especially useful for visualizing and studying the virus mechanism in vivo. Different domains of Gag polyproteins play different roles during the budding process. There are four main domains: Matric (MA), Capsid (CA), Nucleocapsid (NC), and P6, and also two small spacer peptides, SP1 and SP2 12. The MAGag domain heads the process in bringing HIV-1 Gag protein to the budding site on the plasma membrane (PM) and initiates the Gag-driven budding process by anchoring the N-terminal myristoylated group and also the basic patch into the inner bilayer leaflet of the PM. Gag proteins usually appear in a monomeric state before they migrate to the PM 4, 17 , and the conformational change through the PM binding initiates Gag-Gag and NCGag-RNA interactions18-20. Except for being the determinant of the Gagmembrane binding, MAGag also plays an important role in incorporating with Env glycoproteins into virus particles 21 . The mutation of MAGag still has efficient virus replication; however, the defects of the misallocation of most Gag proteins to the endoplasmic reticulum (ER) have been observed. Only a few of the extracellular matured VLPs have been found, and are all lacking of Env polyproteins 22-23. A couple of candidates 4 of cellular PM binding partners, which play the role of host factors to be bound by MAGag, are phosphatidylinositol-(4,5)-bisphosphate (PI(4,5)P2) 24-25 and cholesterol. The MAGag region’s positively charged character also assists the binding to the negatively charged PM 26 . It has been shown that MAGag myristoylated motif exposure and Gag polymerization trigger each other to promote the MAGag association with the PI(4,5)P2- or cholesterol-rich PM, and assemble into a hexamer of trimers which stabilizes the Gag binding into the inner leaflet of the PM 25, 27-28. The next domain, CAGag domain, can be divided into the N-terminal and the Cterminal domain (CANTD and CACTD) and are crucial for the Gag protein-protein interaction which is required for an immature virus particle assembly and mature virus particle conical viral core shell creature 4. MAGag and CAGagNTD, which are particularly important for Gag stable membrane association, have a flexible linker in between each other. CAGagNTD construct the hexametric shape connected MAGag trimers on the PM with six-number rings and large central holes 27-28 . CAGagCTD together with adjacent SP1Gag and NCGag domains are significantly important for virus particle formation 29-31 . When MAGag binds the PM and a spherical viral particle is forming, Gag lays in a radical orientation with the Cterminus P6Gag domain projecting into the center of a sphere virus particle. CAGagCTD and SP1Gag domains provide the major force for Gag multimerization through their interaction and facilitate the virus forming into immature virus particles 29, 32-33 . Although EM (electron microscopy), NMR (nuclear magnetic resonance), and X-ray crystallography have been used to study immature and mature HIV-1 particles structure for decades, cryoelectron tomographic images, which provide up to 2nm resolution, have been showing the immature and mature HIV-1 particles Gag molecules arrangement in great detail in studies 5 primarily by Briggs and Wright since 2004. They showed an immature HIV-1 particle from the different angles of CAGagNTD (green color) multimerizing into a hexametric or ringshaped lattice with a large hole in the center of the hexamer. The same pictures also display CAGagCTD (yellow color) and SP1Gag(pink color) extending these repeating hexagonal lattices of geometry away from the inner leaflet of the PM in ~8nm distance between each hexamer 34-35. A perfect repeating hexametric lattice is intuitively built into the flat-sheet geometry, so an immature HIV-1 particle needs to incorporate several irregular shapes of empty patches accordingly as defects within the hexagonal lattice in order to curve the CAGag-SP1Gag hexametric lattice into a spherical virus particle 9-10, 34. As a mature HIV-1 particle, the cleavage at the junction of MAGag-CAGagNTD by PR releases CAGag-SP1Gag hexamers from the MA domain on the inner PM. The other cleavage on the CAGagCTD and SP1Gag joints causes CAGag to rearrange into a capsid of a mature HIV-1 particle. The mature CAGag hexametric lattice introduces 12 pentameric defects, with 7 at the wide end and 5 at the narrow end, in order for CAGag to constitute a closed conical shape 36-38 . Comparing the hexametric CAGag lattice in a mature HIV-1 particle to an immature one, it was shown that the mature one has looser CAGag hexagonal structure with ~9.6nm spacing 9, 35, 38 . The third Gag main domain, NCGag domain, has been shown to play a key role in virus budding for its various important characteristics. First, NCGag is responsible for recognizing and recruiting two copies of viral genomic RNA through the binding between the NC two-zinc finger motifs and the specific packaging signal (psi) on the viral genome, and simultaneously, NCGag packaging the viral genome is important for the purpose of protection 39-40 . NCGag-RNA interaction is believed to accelerate Gag polymerization 6 during the assembly 41-43 . In contrast, the total internal reflection microscopy (TIR-FM) live image studies of Gag and RNA interaction by Jouvenet et al. have shown that the Gag assembly times do not differ either with or without the presence of RNA, but RNA dynamics did decelerate after the presence of Gag 25 . Second, NCGag zinc finger motifs associate with Bro1 Domain of ALIX, which is an adaptor protein among ESCRT (endosomal sorting complexes required for transport) machinery, to facilitate the HIV-1 budding. The two interactions of NCGag-RNA and NCGag-ALIXBro1 have been proved to function independently 44-45. In contrast, the third NCGag binding harbor to the PM showed very recently, by Sette et al., that an NCGag-PM interaction is required for an NCGagALIXBro1 interaction. Last, the C-terminus of CAGag can interact with NEDD4-2 (neural precursor cell expressed developmentally down-regulated protein 4-2), which is a member of the human NEDD4 ubiquitin E3 ligase protein family46-48. P6 domain is located at the C-terminal HIV-1 Gag. It is a critical component in the HIV-1 budding, especially for the virus and membrane separation on the PM. It was first observed by Gőttlinger et al. in 1991. They showed that the production of HIV-1 VLP after deletion of P6Gag had severely decreased, and the TEM image displayed that HIV-1 P6Gagdeleting VLPs still assembled, but failed to be released and were trapped to the cell membrane with a thin tethe 28. Later on in 1995, Eric Freed et al. followed up to discover that the highly conserved 7PTAP10 (PTA(S)P type) domain on HIV-1 P6Gag is the determinant which impairs the release of VLPs 49, and is called the “late domain” since it works at the late stage of the budding process. PTA(S)P late domain managing the VLPcell separation is found not only in some retroviruses, but also in most lentiviruses and some other enveloped viruses. Nonetheless, EIAV (equine infectious anemia virus) is an 7 exception. In 1997, the other type of late domain, YPXnL (22LYPDL26 in EIAV), was first recognized in EIAV P9Gag domain by Puffer et al. and it is still the only late domain found in EIAV so far. Puffer et al. have also proposed that these late domains, PTAP and YPXnL, promote VLPs’ release from cells by interacting with certain cellular proteins 50. In 2003, Gőttlinger et al. identified that YPXnL late domain also exists in HIV-1 P6Gag (35LYPLTSL41 in HIV-1, n=3), which is located at the C-terminal of HIV-1 P6Gag and plays a role as an auxiliary late domain in addition to the primary PTAP late domain 51. Another type of late domain with an analogous function, PPxY, was identified in many other retroviruses, but not in HIV-1 and EIAV. Shortly after these three types of late domains had been identified, a series of robust pieces of evidence indicated that these three types of late domains contributed to the final stage of the virus budding by recruiting a group of cellular protein complexes, called ESCRT machinery, and its associated proteins to form different pathways and promote the fission and release of VLPs from host cells. To date, three types of late domains P(T/S)AP, PPxY, YPXnL (where x is any residue and Xn is any sequence) are confirmed in various retroviruses and also in certain other enveloped RNA viruses33, and there might be more extra late domains to be identified 51-52 . The parallel late domains found in different RNA enveloped viruses have positional and functionally replaceable and exchangeable characteristics, which fact provides an important implication that these viruses possibly use the same mechanism with redundancy in the virus budding 52. The HIV-1 virus budding with two late domains PTAP and YPXnL in P6Gag have been well explored. PTAP domain that recruits downstream ESCRT machinery proteins by directly binding TSG101, a subunit of ESCRT-I, was identified in 2001 11, 53. In 2003, YPXnL late domain connecting with ALIX during the HIV-1 budding 8 as the second pathway has also been proved, in which they also showed YPXnL-ALIX is the main and only pathway to recruit downstream ESCRT proteins in the EIAV budding with 60X higher affinity compared to HIV-1’s 51. Sundquist Lab in the University of Utah’s showed that the effect of viral productivity of late domains mutations in EIAV and HIV154. Comparing HIV-1 and EIAV budding, it has been shown that the inhibition of the YPXnL-ALIX interaction has more significant VLPs production defect in EIAV GagP9 YPXnL mutation than in HIV-1 with the approximate proportion being 4: 1. In the HIV-1 budding, the PTAP mutant reduced the virus release number around 33 folds more than the YPXnL mutant 55. However, the YPXnL mutant impacted viral particles release relatively slightly, but the importance of YPXnL-ALIX pathway was greatly substantiated by the designed panel in which overexpression of ALIX can rescue the ALIX-dependent HIV-1 budding (HIV-1 PTAP- budding) 56. In 2003, the Sundquist and Rockefeller University’s Bieniasz labs basically examined most interactions between late domains and ESCRT proteins as well as ESCRT-ESCRT protein interactions during the virus budding 57-58. 1.2 ESCRT Machinery Although the enveloped virus Gag polyprotein alone can bind to the plasma membranes and assemble into spherical virus particles, the virus-PM abscission and virus particle release still require Gag late-domain on to hijack ESCRT machinery 11-12, 53 . ESCRT machinery is a protein complex which plays an important role in mediating not only the virus budding, but also other topological similar membrane fission events, such as formation of multivesicular bodies (MVBs) 59-60 , and the daughter cell separation in cytokinesis 59-64. ESCRT machinery caught attention first in 1992 from the discovery that 9 class E VPS (Vacuolar Protein Sorting) proteins’ mutation caused defection of vacuolar protein sorting and MVB formation 65 . Then in early 2000, a set of ~20 proteins in Saccharomyces cerevisiae yeast cell that may directly participate in the formation of MVBs was recognized as a multiprotein system, and was termed as the ESCRT complex 66. MVB is a subset of one cellular degradation process, lysosome (the other one is proteasome), where cargo proteins destined for degradation are marked by ubiquitin protein (Ub) and sorted into ILVs (intraluminal vesicles). ILVs bud into the late endosome to form MVBs, and these cargo-filled ILVs are destroyed when MVBs fuse with lysosomes 67 . ESCRT machinery dysfunction in MVB events might cause tumors or neurological diseases 68 . ESCRT machinery conducts membrane fission events mainly through four different protein complexes, ESCRT-0, I, II, and III and its associated proteins, including ALIX, VPS4, etc. They are highly conserved from yeast to eukaryotes. In human ESCRT complexes, there have been more than 30 different proteins verified, and our understanding of their complexity is still expanding. There are still some important differences in the pathways which are used in several membrane fission events, even though ESCRT proteins are extremely conserved 69-71 . To clarify some controversial and more detailed pathways, interactions, and functions in ESCRT-mediated membrane fission events, more advanced biological and physical methods are still needed. 1.2.1 ESCRT-0, I, II, III, and VPS4 ESCRT-0 was extensively studied around 2000. It comprises two components in yeast: VPS 27 and Hse1. The yeast structure is 95% identical to ESCRT-0 in humans, where two subunits are termed as HRS (hepatocyte growth factor-regulated tyrosine kinase 10 substrate) and STAM1 and STAM2 (signal transducing adaptor molecule) in humans 72. ESCRT-0 is especially important for MVBs in the early endosome for recognizing ubiquitinated cargos (the cargos have been marked for degradation) at the endosomal membrane and recruiting the downstream ESCRT proteins. The distinct domains on these two ESCRT-0 subunits provide different performance in MVBs44: 1) N-terminal of both Hrs and STAM contain VHS domain, which was reported to have membrane binding and ubiquitinated cargo recognition roles 73-75. 2) FYVE domain on Hrs, a double zinc-finger, binds directly to the endosomal membrane where is enriched with PtdIns3P (phosphatidylinositol-3-phosphate) 76-77 . 3) UIM (Ubiquitin-interacting motif) domain is composed in both Hrs and STAM. They also bind to ubiquitinated cargos and the mutation of these UIMs defected, sorting ubiquitinated protein into vacuolar lumen 78 . 4) PSAP domain on Hrs bind to ESCRT-I. The mutation of Hrs caused less of collocalized ESCRTI on the membrane and decreased the number of MVBs 79-80 . 5) The mutual association between Hrs and STAM is through their coil-coil (CC) regions 81. 6) The C-terminal of Hrs containing clathrin box (CB) motif directly interacts with clathrin 82 . Together with all these functions—starting from ESCRT-0 being recruited to the endosomal membrane, binding and sorting the ubiquitinated cargos, and then recruiting ESCRT-I to cascade the sorting of cargos and MVBs formation—ESCRT-0 is a trigger for ESCRT pathway in the MVBs and is crucial for linking the upstream machinery of early endosomes and downstream late endosomes for MVBs biogenesis. Because of interesting comparisons in the functions of ESCRT-0 in MVBs and other membrane fission events, ESCRT-0 seems dispensable in virus budding and cytokinesis. HIV-1 Gag resembles ESCRT-0’s function by using HIV-1 P6Gag MA domain to bind the PIP2-enriched PM and using HIV-1 P6Gag 11 PTAP late domain to recruit downstream ESCRT-I 83-84 (ESCRT-I has even stronger interaction affinity to HIV-1 P6Gag PTAP compared the PSAP on Hrs to ESCRT-I). In cytokinesis, Cep55 shows a parallel scheme by utilizing the ESCRT machinery and replacing the function of ESCRT-0 38, 51 (shown in Figure 1). ESCRT-I had its first three subunits identified in the early 2000s: VPS23 (homolog of TSG101 (tumor susceptibility gene 101) in humans), VPS28 (four versions in humans: VPS28 A-D), and VPS37 66, 85 . The ESCRT-I structure did not complete until the last subunit MVB12 (multivesicular body sorting factor 12; two versions in humans, MVB12A and B) was found in 2008 86-87. In humans, ESCRT-I can express eight different stable and soluble combinations of ~350 kDA complexes with ratio of 1:1:1:1 of four subunits in heterotetrameric structure 86-87 . It is a highly asymmetric shape 176 Å long in total, and consists of a 25x55x60 Å globular headpiece tied to an extended 130 Å long and 20 Å diameter extended cylinder stalk 53 . The headpiece is built around antiparallel two helix hairpins and superimposes well with the ternary by VPS 28, and the part of TSG101, VPS 37, and MVB 12. The stalk contains four α helices, curving the stalk into a long S shape, two from TSG101 and one each from VPS27 and MVB12, but not VPS 28 87. From the structure aspect, the ESCRT-I stalk architecture is proposed to be structurally and functionally important for the cargo sorting and to serve as a spacer regulator 88 . In the virus budding, since TSG101 was recognized in 2000, it was soon demonstrated by several groups that PTAP late domain hijacks the ESCRT mechanism for the virus particle release starting from the specific interaction with TSG101. The deletion of TSG101 by RNAi causes serious defect in the HIV-1 budding 11-12, 53 . Later on, it was known that PTAP- TSG101 is one of the most essential pathways for the HIV-1 budding in which TSG101 is 12 considered as an early factor because it shows up at the beginning of Gag multimerization at budding sites. Except for this binding, the N-terminal of Ubiquitin-conjugating E2 enzymes (UEV) domain on TSG 101 also binds to ubiquitin to facilitate budding 89. The other binding partner of UEV, single PSAP motif in Hrs subunit of ESCRT-0, has also been verified 80, 84. The second domain of ESCRT-I, VPS28, binds to the C-terminus of TSG101 and plays an integral role in budding. The mutation of mutual binding sites on VPS28 or TSG101 or mutation of VPS28 impaired the HIV-1 budding 90. The C-terminal domain of VPS28 has a binding site to ESCRT-II complex 91. The third subunit of ESCRT-I, VPS37B, and C, has been shown that their overexpression can rescue HIV-1 budding lacking PTAP late domain, implying that VPS37 is also a part of pathways to enhance the virus budding 92 . The last subunit of ESCRT-I, MVB 12 has showed not affecting virus budding form its deletion 86. ESCRT-II in the yeast was reported right after ESCRT-I by Babst et al. in 2002. It contains VPS22, VPS25, and VPS36, which are termed EAP30, EAP20, and EAP45 individually in humans. In MVB, ESCRT-II plays the central bridge role in ESCRT pathway where ESCRT-II connects ESCRT-I and ESCRT-III, and also interacts with ubiquitinated cargos 93. The structure has been revealed by Hurley and Williams’s group in 2004. They showed the yeast ESCRT-II composed as a trilobal complex in ‘Y’ shape, with two copies of VPS25, one copy of VPS22, and one copy of VPS36 located at the Cterminal region 61 . The two VPS25 proteins projected away from the center of ‘Y’ and attached to VPS22 and VPS36 dimer with equivalent affinity. It has been reported that ESCRT-II in yeast is recruited by ESCRT-I directly to the membrane by the interaction 13 between N-terminal NZF domain on VPS36 (a subunit of ESCRT-II) to VPS28 (a subunit of ESCRT-I) 93-94. The downstream recruitment from ESCRT-II to ESCRT-III is by VPS25 to VPS20. The downstream recruitment from ESCRT-II to ESCRT-III is by VPS25 to VPS20 93 . However, the interaction of ESCRT-II in ESCRT machinery in humans is still not yet fully understood. We only knew that human ESCRT-II EAP45 is missing the NZF (Npl4 zinc fingers) domain, which is responsible for binding ESCRT-I 95, but ESCRT-II has still been found on the membrane. Additionally, the mutation of ESCRT-II impairs the MVBs vesicle formation in humans, but does not impair cytokinesis 61, 95-96 . The ESCRT-II involvement in the virus budding is still a controversial theory. Researchers still do not know how ESCRT-I activates ESCRT-III if ESCRT-II is not required during the HIV-1 budding process 95, 97-98 . On the other hand, it has also been shown that ESCRT-II did involve or improve the HIV-1 budding in some studies 97, 99 ESCRT-III in yeast was reported by Babst et al. in 2002, where ESCRT-III consists of four main subunits around 200~250 residue proteins in the similar sequence and structure: VPS2, VPS20, VPS24, and Snf7. Additionally, ESCRT-III also includes three extra regulatory subunits: Did2, VPS60, and Ist1 100-101 . In humans, there are about 12 proteins in ESCRT-III termed as charged multivesicular body proteins (CHMPs) and can be divided into seven families, including CHMP1A/B (Did2 or VPS46), CHMP2A/B (VPS2 or Did4), CHMP3 (VPS24), CHMP4A/B/C (Snf7 or VPS32), CHMP5 (VPS60), CHMP6 (VPS20), and CHMP7 monomeric state in cytosol 4, 102 103-104 . ESCRT-III proteins remain in an autoinhibited , and the dissociation of C-terminal acidic tail and N- terminal core structure of ESCRT-III protein stimulates ESCRT-III proteins into an active 14 state which allows ESCRT-III proteins to polymerize, to interact with the membrane or other proteins, and finally to facilitate membrane abscission 105-107. Most ESCRT-III subunits have their specific role, recruitment order, and copolymerization partners, but some subunits were found unnecessary for their possible functional redundancy according to different membrane fission events 108. The significant scale differences between different membrane fission events, from the small vesicle of MVB (< 50 nm diameter) and virus particles (~ 100 nm diameter), to the intercellular bridge during cytokinesis abscission in two equally sized domains ( > 1 um in diameter) 63, 109 , makes ESCRT-III mechanics especially difficult to come to one conclusion. Researchers generally agree that ESCRT-III is the core component for the membrane deformation and abscission in membrane fission events which starts from an ESCRT-III subunit, CHMP6 (VPS20), to associate with the membrane and bind VPS25 of ESCRT-II in MVB 69, 110 while other membrane fission events possibly utilize TSG101 or/and ALIX as upper stream binding 51, 58. Then CHMP6 plays a nucleator role to induce the second ESCRT-III subunit, CHMP4 (VPS32; Snf7), to oligomerization 69, 110-111 . CHMP4 is the most abundant subunit in 450kDa ESCRT-III complex (which is not like ESCRT-I as a 1:1:1:1 complex) 70. Hanson et al. first proposed in 2008 a geometry model from their deep-etch EM image that the overexpression of CHMP4 can polymerize into a circular array by approximately an 5nm filament in vivo of COS7 cells and cause the membrane negative-curving and outward budding 70 , and that CHMP4B can later spontaneously form a single stranded spiral filament has also been observed in vitro 73. The dysfunction of CHMP4 oligomerization caused liposome deformation not to be found anymore, even co-expressed with all three other main ESCRT-III proteins, revealing the 15 importance of CHMP4 oligomerization to membrane deformation. CHMP6 and membrane binding were also shown to be requirements of the CHMP4 oligomerization 69 . CHMP4 was also observed in vitro by Ghazi-Tabatabai et al. in 2008 and shown to form into other various structures, such as sheet, ring, and tube shapes 75 . The third ESCRT-III subunit, CHMP3 (VPS24), was studied most extensively in the crystal structure within ESCRT-III and found that CHMP3 can assemble into a flat lattice through four helical bundle cores to induce high affinity binding with membrane by exposing the highly basic surface and also reveals two potential dimerization domains which can mediate the polymerization with other ESCRT-III proteins 105. EM studies showed that the C-terminal truncated CHMP3 can co-polymerize with the C-terminal truncated fourth subunit CHMP2 (VPS2) into a helical tubular structure with an average 40 nm diameter with membrane-binding domains facing outward and with one end in domed-shape 75 . The CHMP2A-CHMP3 filament has also been observed to form a cone structure 76. A model was proposed accordingly that CHMP3 recruits CHMP2 to complete or to terminate the CHMP4 homo-oligomerization by capping their domedshape structures and subsequently associating a final ESCRT machinery member, AAAATPase VPS4, through its C-terminal MIM1 (microtubule interacting and trafficking (MIT)-interacting motifs) to close up the fission events 69, 102, 112 . The depletion of either CHMP2 or CHMP3 causes excess of ESCRT-III as a high molecular weight and fails to recruit VPS4 in vitro 102. Except the copolymerization, CHMP3 and CHMP2 alone in vitro can spontaneously form into ordered ~15nm and 100~400nm in diameter filament tubes, respectively and the overexpression of CHMP2B in vivo induces a protruding tube out of the cell surface where they are concentrated 74, 78. With the substantial range of CHMP2B 16 filament tube widths, it was proposed that CHMP2B might be a major scaffold component, instead of CHMP4, due to the various membrane fission events with different size scales 109, 113 . In addition to the core ESCRT-III proteins, CHMP1B (Did2) was also recently observed to form a spiral filament which deforms the membrane in positive manner, opposite to the CHMP4 spiral. The CHMP1B filament has also been shown to coat outside of membrane tubules no matter whether CHMP1B is alone or copolymerizing with IST1NTD. CHMP2, CHMP1, and CHMP3 all possess a C-terminal MIM1 domain, which can recruit VPS4 114. There are several distinct differences in the topologically similar membrane fission events. For example, CHMP6 is required for MVBs, but is not necessary for the virus budding and cytokinesis. In the virus budding, CHMP2 and CHMP4 play key roles, although CHMP1 and CHMP3 may also contribute modestly 4, 115-116 . In addition, CHMP4B is more important in the virus budding, but CHMP4C is more necessary during cytokinesis. With a variety of ESCRT-III demands and polymer structures being observed, several models of the ESCRT-III membrane constriction mechanism have been proposed in different membrane fission events. As for HIV-1 budding studies, the most accepted model is that the ESCRT-III helical polymer constricts the membrane neck as a spiral in the dome shape, and then has ATPase VPS4 come in the end to provide energy and complete the fission 59, 117-119 . This dome-like model is also supported by a theoretical model that ESCRT-III membrane binding affinity is larger than the deformation energy required to bend the membrane to wrap around the dome 120 . The details in this model (such as which ESCRT proteins are involved, and the recruitment order of them, etc.) or 17 other suggested models are somewhat debated in some of their aspects 108. For the highly mechanical, temporary, and also short recruitments of ESCRT-III proteins, the high tempospectral methods will be required to confirm the real and native ESCRT-III membrane fission mechanism. VPS4 is the last component, with one isoform in yeast and two in humans (VPS4A and VPS4B), which is required at the neck of membrane fission events after other ESCRT proteins are completely assembled. The N-terminal of the MIT (microtubule interacting and transport) domain of VPS4 is the binding point to be recruited at the budding sites by the C-terminal of ESCRT-III MIMs (MIT interacting motifs) 121-122 . VPS4 is a class of AAA-ATPase (ATPase associated with a variety of cellular actives) enzyme in the geometry of two stacked hexametric rings 123 where AAA-ATPase can provide mechanical energy to dissociate ESCRT-III from the membrane by converting chemical energy stored at ATP hydrolysis 122-123. Many studies show that the interaction of VPS4 and ESCRT-III and the Vps4 ATPase activity are required to complete the accurate membrane fission and catalyze multiple rounds of the particle formation. For example, when co-expressed VPS4B and the C-terminal truncated CHMP2A and CHMP3, VPS4B was seen inside the CHMP2-CHMP3 tube polymer 75 . As the co-expression of dominant negative VPS4B and CHMP4, the CHMP4 circular array filament protruded away from the cell surface with various heights and CHMP4 and VPS4B accumulated at the base of the protrusion 70. In the TIRFM (total internal reflection fluorescence microscopy) live image, it was shown that around three to five VPS4A dodecamers were recruited to each of the virus budding sites 124 and arrived at the bidding sites about 10 seconds later on average than CHMP4B and lingered for a 18 longer time as well 125 . Except for dissembling and recycling ESCRT-III back to inactive closed conformation in cytosol, VPS4 has also been suggested to contribute to final fission geometry by dissembling ESCRT-III proteins sequentially and prompting the closure of the dome, ring, or spiral constricting assembly 126 . More studies are still needed to understand these processes in mechanistic detail. 1.2.2 ALIX and its multipathways in the HIV-1 budding ALIX was first discovered targeting the apoptotic calcium-binding protein ALG-2, so it is also called AIP1 (ALG-2-interacting protein) 127-128. It belongs to a Bro1-containing mammalian protein family which includes HD-PTP, BROX, RHPN1, and RHPN2 7. Like other ESCRT machinery proteins, ALIX is involved in MVB 129 , retrovirus budding 130, and cytokinesis 61, 96 through connecting with other ESCRT components at the abscission site. Other activities requiring ALIX are apoptosis 131-132 127-128 and cytoskeleton reorganization . ALIX comprises three distinct elements, an N-terminal crescent-shaped Bro1 domain (1-358), a central V-shaped V domain (362-702), and a C-terminal proline-rich region (PRR) (702-868) 55, 133 . secondary or tertiary structure Except for the last domain, PRR, which lacks a stable 55, 134-135 , the structure of ALIXBro1-v has been well documented through small-angle X-ray scattering in protein crystals. This scattering demonstrated that monomeric ALIXBro1-v is 150~170 Å long in one dimension and less than 50 Å in the other two 55, 136. The N-terminal ALIXBro1 domain is composed of three helical hairpins to form a core tetratricopeptide repeat (TPR) into a 100 Å long banana shape 55. In virus budding, EIAV is primarily dependent on ALIX for budding, but ALIX alone is 19 not efficient for the HIV-1 budding 11-12 . Nevertheless, ALIX has a demonstrated importance in budding. Overexpression of ALIX successfully rescues HIV-1 PTAP mutants in the HIV-1 PTAP- rescue panel in which the essential budding pathway for HIV1 (PTAP-TSG101) is abolished, leaving the HIV-1 budding reliant on exogenous overexpression of components that, either directly or indirectly, are able to facilitate the HIV-1 budding 55, 130 . Furthermore, with its unique conformation and multiple binding partners, the complicity of ALIX in membrane fission events becomes very intriguing. In the conformational aspects, ALIX, like other ESCRT proteins, has an autoinhibition function which can switch the proteins between inactive (closed) and active (open) states 137 . Although the triggers to activate and recycle ALIX are unclear, many studies have shown the necessity of ALIX activation to function for membrane fission events 137-140 . When ALIX is in an inactive state, SAXS (small X-ray scattering) profiles indicate that the C-terminus unconstructed ALIXPRD stays close to ALIXBro1-V in solution and associates with arm 2 of the ALIXV or/and ALIXBro1 137, 141. ALIXV arm 1 and arm 2 have unequal lengths of 77 Å and 90 Å and stay at an angle of ~30˚ in a V shape 93 . ALIXV possesses an intriguing hinge node made by two separate helix bundles, each with polypeptide chains crossing over arm 1 and arm 2 three times to provide flexibility for ALIX conformational change 55, 136-137. When ALIX is in an activated state, ALIXPRD is released from ALIXBro1-V, and the ALIXV construct further opens its two arms through its hinge 92. Arg649 is found on ALIXV arm 2 and it joins the three linker strands. The Arg649 mutation can destabilize the ALIX closed conformation and trigger an active state. With the Arg649 mutation, the active state ALIX has been shown to have more membrane association and a higher potential to facilitate HIV-1 viral release and 20 infectivity as shown by Sundquist lab 93 . Pires et al. showed that the activated state of truncated PRD ALIXBro1-V conformer can precede dimerization through the V domain in antiparallel manner. Dimerized ALIXBro1-V has about double the length of the structure of monomeric ALIXBro1-V at the maximal protein dimension. ALIX dimerization was shown to be important to the HIV-1 budding 92. These conformational changes are accompanied or induced by ALIX involvements in the membrane fission. With the multiple ALIX binding harbors from N- to C-terminus, it performs diverse functions in ESCRT-mediated events 142. First, there are two exposed hydrophobic patches and one flexible loop located at ALIXBro1. Both of their mutations cause severe decrease of HIV-1 release. The point mutations in the middle of the first patch located at the concave surface, F199, I212D, or L216D, inhibit the interaction with CHMP4 isoforms, which possess unequal binding affinity in order of CHMP4B > A > C during the virus budding. The overexpression of ALIX with the CHMP4 binding site mutated (I212D) fails to rescue the HIV-1 PTAP- particle release 51, 57-58, 143. ALIX-CHMP4B complex in vitro was shown in EM where co-expressed, C-terminal truncated CHMP4B and ALIXBro1-V dimers formed a filament with a ladder-shape decorated in linear or circular shape 92 . In the second patch on ALIXBro1, there is a kinase docking site Y319F mutation inhibits Src binding 145 133, 144 . Although , the virus budding activities are still retained 55 . This indicates that Src is not required for virus budding activity, so the reason a second patch mutation abolishes virus budding is still not clear. The ALIXBro1 flexible loop projected away from the ALIXBro1 convex surface with Phe105 residue on the loop tip 26, whose mutation was found to impair HIV-1 release as severely as I212D mutant does, even though Phe105 mutation still retains the ALIX interaction function with both Gag and 21 CHMP4 142, 146 . We learned that Phe105 associates with the LBPA (lysobisphosphatidic acid), which is only found abundantly in endosomes 140, 147 , but it is still unclear how Phe105 engages and who the partner is to facilitate the HIV-1 budding 148. Last, an important interaction between ALIXBro1 and two CCHC-type zinc fingers motifs of NCGag was observed by Göttlinger’s lab at the University of Massachusetts. They displayed that ALIX being included in VLPs is greatly dependent on NCGag’s two conserved CCHC-type zinc finger motifs instead of YPXnL and CCHC mutant fails to rescue the HIV-1 PTAP- particle release 44 . It was also verified that this phenomenon is independent of CCHC and RNA association and that CCHC mutant does not affect GagGag interaction 44-45, 149. The EM image showed that the NCGag mutant caused the HIV-1 budding to be defective in the similar way to those of L-domain mutants 150. On the ALIXV domain, the binding partner to the Gag YPXnL late domain during virus budding is located at a conserved hydrophobic pocket with the middle residue Phe676 on arm 2 of ALIXV 51, 151 . The mutation of either YPXnL or Phe676 on one hand, or the ALIX siRNA depletion on the other, inhibits the release of EIAV markedly and causes the overexpression of ALIX fail to rescue the HIV-1 PTAP- particle release 55, 152. On the C-terminus of ALIX, PRD also serves multiple docking sites on its unstructured and flexible structure 55. First, 717PSAP720, which associates directly with the N-terminal UEV domain of TSG101, was thought of as a potential bridge to connect ESCRT-I to ESCRT-III 51, 58 . Overexpression of ALIX with 717PSAP720 mutation still rescued about the same as WT ALIX in the HIV-1 PTAP- rescue panel, indicating that ALIX-dependent viral release does not require TSG101 55, 152 . Secondly, 852PSYP855 mutation and/or Y864A and Y865A mutations on ALIXPRD keep the known ALIX 22 association partners, but losing ALIX self-multimerization ability which was shown to have much less of a contribution to HIV-1 PTAP- rescue. This self-multimerization function is required for both TSG101-ALIX interaction and the ALIX-dependent HIV-1 particle release, but not to ALIX’s function in EIAV budding and. Thirdly, phosphorylation on 718SAPS721 was shown to be necessary to cytokinesis and EIAV budding (but not MVB) and was proposed to be a trigger for ALIX activation since S718-S721 phosphorylation interrupted ALIXPRD interaction with ALIXBro1 138 . The more binding functions of ALIXPRD to other cellular factors includes Cep55 96, 152, endophilins 153, and ALG-2 154-155. 1.2.3 UbL and Ub in the HIV-1 budding PPxY late domain has been shown to bind E3 ubiquitin-protein ligases, NEED4 (neural precursor cell expressed developmentally down-regulated protein 4), to facilitate PPxY-dependent virus budding. However, the detailed mechanism of how NEED4-like proteins engage to the ESCRT pathway for PPxY-dependent virus budding still remains unclear 156-157 . UbL (ubiquitin-protein ligases) is an enzyme to induce protein ubiquitination, which is a common process in MVB utilizing conserved 8.5kDa Ub (ubiquitin) proteins to regulate the sorting of integral membrane proteins into the MVBs pathway. Other ESCRT-dependent events are also shown being regulated by Ub. For example, ALIX binds directly to K63-linked polyUb chains and abolishing this interaction inhibits ALIX-dependent virus budding 158 . The NEDD4 family includes 9 different proteins: NEDD4, NEDD4-2(NEDD4-L), ITCH (AIP4), SMURF1, SMURF2, WWP1, WWP2 (AIP2), NEDL1 (BUL1), and NEDL2 46. All NEDD4-like proteins own three main domains: C2 (the domain for membrane binding), WW domains (the harbor site for PPxY 23 late domain), and HECT (engage the virus budding after it is catalytically active) 159 . Surprisingly, HIV-1, which lacks PPxY late domain, was found that the overexpression of NEDD4 proteins can still rescue the HIV-1 PTAP- particle release, although there is no evidence that NEDD4 interacts with Gag directly 46-47. NEDD4-2 overexpression is capable of rescuing the HIV-1 PTAP- particle release potently despite the absence of late domains, which is different from ALIX rescue dependence on YPXnL late domain 46-47. Recently, a protein called Angiomotin (AMOT) was found directly binding to NEDD4-2 and is required for the HIV-1 budding interdependent on NEDD4-2. Depletion of AMOT inhibited the virus formation at the earlier stage when shell arc is around 165˚ vs. 320˚ of TSG101. Of the other Nedd4-like proteins, Nedd4-1 was also shown to rescue the HIV-1 PTAP- particle release to a slightly lesser degree compared to NEDD4-2. Notably, the NEDD4-1 rescue is interdependent on the ALIX-YPXnL pathway through the direct interaction with ALIX for ubiquitination (absence of ALIX or YPXnL disrupted the NEDD4-1 overexpression rescue of the HIV-1 PTAP- particle release, and deficiency of NEDD4-1 also caused the overexpression of ALIX fail to rescue the HIV-1 PTAP- particle release) 160 . For the NEDD4 family rescue, the enzymatic activity on HECT domain of both NEDD4-2 and NEDD4-1, and ubiquitination on HIV-1 Gag proteins and some ESCRT proteins are important 161-164 , which indicates that the Ub and UbL facilitate the HIV-1 budding directly or indirectly. The removal of Ub by DUb (deubiquitinating enzymes) from ubiquitinated proteins is shown in the MVBs process to be equally important to maintain a steady amount of Ub in the cell 66, 165-167. There are two kinds of endosomal DUbs, AMSH and UBPY (termed Doa4 in yeast), and both of them can be recruited by ESCRT-III (CHMP3), ALIX (Bro1 24 domain), or the ESCRT-0 subunit STAM in the MVBs pathway 165, 168 . Most of DUb’s functions are as yet poorly understood 169. Thus, Lb, UbL, and DUb’s mechanism in HIV1 is still in need of further study. 25 Figure 1. Early ESCRT pathways for different membrane fission events: MVB, HIV-1, and cytokinesis. CHAPTER 2 HIV-1 LIKE PARTICLES ASSEMBLY 2.1 Abstract HIV-1 Gag polymerizes on the plasma membrane to form VLPs with similar diameter to the wild-type virus. Using multicolor dual penetration depth TIRF microscopy, we imaged the assembly of individual VLPs from their nucleation to the recruitment of VPS4, a component of the ESCRTs that marks the final stage of VLP assembly. Dual penetration depth TIRF microscopy was used to measure the relative amount of Gag incorporation in each VLP independent of its azimuthal movement. In 60% of observed VLPs, Gag polymerization paused for 2-20 minutes during VLP formation independent of cytosolic Gag concentration. This pause distribution is unaffected by interactions with ESCRTs since VLPs assembled with late domain mutants of Gag, which do not recruit ALIX or TSG101, exhibit similar pause behavior. These pauses are indicative of a single rate limiting event required for completion of assembly independent of Gag concentration and ESCRT recruitment. We propose that the pauses are either related to incorporation of defects in the organization of Gag within the forming VLP and/or shortcomings in interactions of Gag with essential and still undefined cellular components during formation of curvature on the plasma membrane. 27 2.2 Introduction As introduction of HIV-1 Gag in Chapter 1, HIV-1 Gag alone is sufficient to create fully formed vesicles coated with Gag that bud into the extracellular space as VLPs 170 . Starting from the MA domain, which contributes to membrane binding, is essential for targeting Gag onto the inner leaflet of the plasma membrane where it contains a PIP2 binding site as well as myristoylation motif. We know that HIV-1 virions are enriched in well-ordered lipid domains and cholesterol 171-173, though the exact mechanism by which these lipids become enriched in the forming VLP is not clear 4, 174 . Therefore, aside from Gag polymerization, lipid and protein interactions on the inner leaflet of the plasma membrane play a clear role in assembly of HIV-1 virions. The CA domains in adjacent Gag proteins bind each other with strong affinity and these interactions are critical for hexagonal arrangements of Gag within immature HIV-1 virions as observed by cryo-EM 35, 175 . In the immature HIV-1 virion, Gag forms an incomplete lattice of hexagonal geometry on the inner leaflet of the plasma membrane held together mainly through CA-CA interactions 176-177. In the mature HIV-1, the CA domains are cleaved, and the HIV-1 core is assembled through CA-CA interaction. The HIV-1 core incorporated 12 pentagons to ensure forming a closed shell 176. Cryo-EM measurements of the immature HIV-1 virions, however, revealed no pentagonal facets within the lattices. Therefore, to create a closed topology, the immature virions incorporate empty patches as defects within the hexagonal lattice 34, 175, 178 . These defects could be the results of membrane fission 178. However, before the fission of the membrane, the hexagonal lattice of Gag needs to curve, which requires deviations from hexagonal lattice assembly. These observations suggest a complex relationship between Gag polymerization and membrane 28 curvature during virion formation which would likely manifest itself in the kinetics of assembly. To catalyze the fission of the host membrane and the release of the virus, ESCRTs 11-13, 51, 179 recruit to the HIV-1 assembly site by P6 two conserved late domain motif, culminating in arrival of the AAA ATPase VPS4 4, 11 . Two late domains are, 1) a PTAP motif, that functions as the primary HIV-1 binding site to ESCRT-I component Tsg101 4, 11-12, 53 ; and 2) a LYPxL motif, that binds to ALIX, an auxiliary member of the ESCRT pathway 51, 55 . Gag double mutants lacking functional PTAP and LYPxL motifs were shown to have identical kinetics to wild-type Gag during assembly of HIV-1 VLPs 15, 180. The kinetics of assembly of HIV-1 VLPs have been studied in the context of late domain mutants of Gag. Within the P6 domain of HIV-1 Gag, there are two conserved L domain motifs: Aside from TSG101 and ALIX, ubiquitin ligases have been shown to play a role in the budding of HIV-1 46, 48, 181. Assembly of individual HIV-1 VLPs has been observed using TIRF microscopy. These studies show that formation of HIV-1 VLP initiates at the plasma membrane and continues through polymerization of Gag, resulting in fully formed HIV-1 virions 15, 72. A vector system that expresses wild-type levels of Gag and Gag-Pol proteins 15 showed the same kinetics for assembly as transient transfection of Gag and Gag-mCherry 72. Therefore, the presences of Gag-Pol or fluorescent protein fusions were shown to have minimal effects on the kinetics of assembly. Recruitment of ESCRT proteins into forming HIV-1 VLPs was observed using dual-dolor TIRF microscopy 115, 124 . The two late ESCRT factors CHMP4 and VPS4 protein are transiently recruited to the VLP for 25-30 s at the last step after Gag 29 polymerization is complete 16. The arrival of ESCRT proteins is a molecular signature for the release of VLPs. Because the resolution of optical microscopy is limited, before visualization of ESCRT recruitment, the fully formed VLP was inferred either through a plateau in Gag polymerization followed by subsequent movement of the VLP 15, or through incorporation and subsequent quenching of pH-sensitive fluorescent proteins fused to Gag at the end of the imaging period 180. Assembly of HIV-1 VLPs is driven by polymerization of Gag on the inner leaflet of the plasma membrane. Gag molecules interact with each other and with the plasma membrane, and possibly with the cellular machinery during this polymerization. Exactly how the polymerization of Gag is linked to the creation of membrane curvature and what, if any, cellular components are involved during the formation of curvature is not clear. The ESCRT machinery, which is the well-characterized partner of Gag, is mostly engaged during the end stages of the VLP formation and its separation from the host cell. Because Gag polymerization is the primary driver of the VLP formation and there is an abundance of Gag within the cytosol during polymerization, the polymerization of Gag on the membrane during VLP formation should be able to proceed uninterrupted. Through monitoring the accumulation of fluorescence from fluorescently tagged Gag molecules within a forming VLP, the polymerization kinetics of Gag have been previously established 15, 180 . However, the previous measurements were performed using TIRF microscopy and therefore, it has been difficult to relate the fluorescence directly to Gag incorporation. This is because fluorescence could be affected both by azimuthal movement of the VLP and by accumulation of Gag. Here we have used a method based on dual-penetration depth TIRF microscopy, which allows decoupling of the relative amount of Gag incorporation from 30 azimuthal movements of the VLP. Our measurements of the relative Gag incorporation during the assembly are consistent with previous observations based on single-penetration depth TIRF microscopy; however, we uncovered pauses in the polymerization of Gag at various stages of the VLP formation. These pauses, characterized in this article, are best described as deviations from a simplistic Gag-driven VLP assembly model, and can be caused by possible defects in the organization of Gag within the forming lattice. They can also be caused by shortcomings in interactions of Gag with essential and still undefined cellular components during formation of curvature on the plasma membrane. 2.3 Materials and Methods 2.3.1 Fusion of VPS4 with fluorescent proteins Ten DNA constructs were created that encoded the joining of VPS4A to GFP with protein linkages of materially different character. DNA sequences were generated using four helical linkers of similar composition but varying lengths (13, 21, 29, and 37 amino acids) and one flexible linker of 32 amino acids in length. All five linkers were used to generate VPS4A fusions with GFP in both gene orders, e.g., GFP before VPS4A and VPS4A before GFP, for 10 total linkage combinations. The flexible linker avoided helixforming amino acids. 2.3.2 Testing the functional VPS4 fusions RNAi-resistant plasmids of VPS4A fusion with GFP using various linker combinations were used to rescue the knockdown of VPS4A. Production of infectious virus was used as readout for the rescue. This experiment has been described in detail in Morita 31 et al. 116 and in Appendix A. Using this test, we found that VPS4A-h37-eGFP (h37 is a 37amino-acid-long α-helix-forming linker) was the most effective plasmid to rescue the infectivity of HIV-1 produced in cells depleted of endogenous VPS4A and B. 2.3.3 VPS4A-h37-mCherry clonal cell line HeLa cells were transfected with VPS4A-h37-mCherry plasmid containing a delta CMV promoter for reduced expression 116 . Stable cells were selected through incubation with medium supplemented with 1mg/ml G418 for a week. After selection, cells were grown and maintained in medium with 0.5 mg/ml G418. Fluorescent cells were then sorted using a cell sorter and diluted into 96 well plates so each well hosts 1-2 cells. Single colonies were picked up from the 96 well plate after a week and expanded. Colonies were tested for their expression level of VPS4A-h37-mCherry and a single clone was chosen for experiments. This clone has cell doubling times identical to untrasfected control HeLa cells and no cell toxicity was observed. 2.3.4 Cell culture and imaging conditions VPS4A-h37-mCherry HeLa clonal cell line was grown in DMEM (Invitrogen) with 10% fetal calf serum, sodium pyruvate, and L-glutamine. Cells were seeded 12 hours before transfection on sterile 35mm glass coverslips at 80% confluence. Transfection was carried out using 2ul of Lipofectamine2000 (Invitrogen) and DNA plasmid mix of 450ng Gag, 450ng Gag-eGFP fused Gag, and 100ng of VPU. Three hours after transfection, the sample was supplemented in CO2 independent medium + 2.5% fetal calf serum and moved to the microscope for imaging. The cells were kept at 37 °C during the imaging. 32 2.3.5 Microscope description Live images were acquired using an iMIC Digital Microscope (TILL Photonics, Gräfelfing, Munich, Germany) controlled with LIVE ACQUISITION imaging software (TILL Photonics). Two laser wavelengths, consisting of a 488-nm diode laser (iBeam Smart 488S; Toptica Photonics, Gräfelfing, Munich, Germany) and a 561-nm diodepumped solid-state laser (Cobolt Jive, 561-nm Jive High Power; Cobolt, San Jose, CA), were used to excite GFP and mCherry, respectively. Laser beams were passed through an acousto-optical tunable filter and focused onto a fiber that delivered the light to the Yanus Digital Scan Head, utilizing the Polytrope Imaging Mode Switch (equipment by TILL Photonics; see diagram of working setup given in Appendix B). The Yanus Digital Scan Head consists of two Galvo mirrors and one spherical mirror to control the laser-beam position. The Polytrope Imaging Mode Switch rapidly switches the illumination beam path between the Epi- (wide-field), the FRAP, and the TIRF microscopy modes. It also holds the quadrant photodiode used for TIRF penetration-depth calibration. In the TIRF mode, the Yanus Digital Scan Head (TILL Photonics) is used to control the position of the focused beam in the objective’s back focal plane and can be adjusted within 0.2 ms, as shown in Appendix B. We positioned the focused beam at the edge of the back focal plane of the objective (N = 1.46, 100×, Carl Zeiss Microscopy, Thornwood, NY) to reach beyond the critical angle and achieve TIRF. The TIRF critical angle was verified by scanning the laser beam across the back aperture and measuring the reflection of the laser from the glass sample interface back into the objective and onto the quadrant photodiode. The penetration depth of the beam is calculated based on the incident angle of the beam, which is in turn measured by the position of the beam on the quadrant photodiode. Once the penetration 33 depths for the experiments are set at the beginning of acquisition, a feedback loop keeps the focus of the objective on the sample by constantly monitoring the position of the backreflected beam with respect to the original beam. We also rotated the TIRF illumination on the objective back focal plane at the rate of one turn per exposure (TIRF360) to maximize homogeneity of the TIRF images. We adjusted the laser incident angle and recorded two different penetration-depths’ excitations of eGFP (100 and 200 nm, respectively) and one penetration-depth excitation of mCherry (200 nm) every 15 s for 90 minutes on each cell. The two colors of emission were separated by a Dichrotome Image Splitter (TILL Photonics) and recorded by an electron-multiplying charge-coupled device camera (iXon 897; Andor Technology, Belfast, Ireland). Camera electron-multiplying gain is set at 20, which has been tested to have the best signal/noise. Images were analyzed by a homewritten MATLAB algorithm (The MathWorks, Natick, MA), which was previously described in Saffarian and Kirchhausen 182. 2.3.6 Calculating the amount of Gag incorporated in single VLPs formation of the plasma membrane The intensity of evanescent wave decays exponentially with the distance 𝑍away from the glass and medium interface into the cell sample 183-184. Let 𝑍 represent the center of mass position of the VLP, 𝐼𝑜 the intensity when 𝑍=0 (located at interface surface), and 𝑑 the penetration depth. Then: 𝑍 𝐼(𝑧) = 𝐼𝑜 𝑒 −𝑑 (1) The total TIRF from N Gag-eGFP molecules can be written as: 𝐹(𝑧) = 𝑁𝑞𝐼(𝑧) (2) 34 in which 𝑞 is a product of quantum yield, absorption coefficient, and the detection efficiency. To normalize out the constants, the average intensity of the first 15 data points from the single VLP formation was calculated and used to normalize the fluorescence intensity traces in each of the two penetration depth Gag-eGFP channels 185 . The results yield normalized TIRF fluorescent intensities ( 𝐹1 and 𝐹2 ). The normalized 𝐹1 and 𝐹2 collected using different penetration depth illuminations (𝑑1 and 𝑑2 ), the axial position of the center of mass of the VLP can be calculated as, 𝑑 𝑑 𝐹 𝑍 = (𝑑 1−𝑑2 ) log (𝐹2 ) 2 1 1 (3) After we calculate 𝑍 from (3), we smooth the noise by a weighted average over time as 𝑍𝑓 =0.3𝑍𝑖−1+0.4𝑍𝑖 +0.3𝑍𝑖+1. (4) Using the position 𝑍𝑓 , we estimate the average Gag-eGFP molecular number inside the VLP: 𝑁𝑖 = 𝐹𝑒 𝑍𝑓 𝑑𝑖 (5) The relative number of Gag molecules, shown in Figure 2, is calculated as the average of 𝑁1 and 𝑁2 . To provide insight into VLP assembly, we have also measured the average 𝑍𝑓 position for all VLPs in this study as presented in Appendix C. The 𝑍𝑓 increases as the VLP assembles, implying that the VLP assembly starts near the glass surface and further assembly results in an increase to the height of the center of mass of the VLP. We also measured the differences between N1 and N2 during the assembly by calculating ∆𝑁/𝑁. The standard deviation of this value over all of our profiles is 0.074. 35 2.3.7 Shell filling model for VLP assembly We constructed a deterministic model to describe how the total number of Gag proteins would change as the VLP assembles. We incorporate the spherical geometry into the derivation of our model. First, we consider how the surface area of a sphere with radius 𝑟 changes with respect to the polar angle 𝜙 as shown in Figure 3, which is given by: 2𝜋 𝜙 𝑆𝐴(𝜙) = ∫0 ∫0 𝑟 2 sin(𝜙) d𝜙d𝜃 = 2𝜋𝑟 2 (1 − cos(𝜙)) (6) Not all the VLP is covered with Gag, indeed fluorescence fluctuation analysis on purified VLPs has shown a low fraction of Gag coverage within each VLP 186 and Cryo- EM of purified immature virions has shown that only around 65% of the surface of the virion is covered with an organized Gag lattice 54. For purposes of simplicity in our model, we assumed that the vacancies within the Gag lattice are uniformly distributed and thereby the shell filling model presented here assumes that the VLP forms a Gag shell with 100 % coverage, albeit with a lower packing density. A separate model assuming 65% Gag filling of the VLP is presented in Appendix C and shows no effects on the conclusions of our study. We can approximate the surface area of a single Gag protein (𝑆𝐴𝐺) by dividing the total surface area occupied by Gag by the total number of Gag proteins in a single VLP (𝐺𝑡𝑜𝑡 ) or: 𝑆𝐴𝐺 = 4𝜋𝑟 2 (7) 𝐺𝑡𝑜𝑡 Under the assumption that Gag binds only at the cross section of the cell membrane, the total number of Gag proteins (𝑁) at some angle 𝜙 is: 𝑁(𝜙) = 𝑆𝐴(𝜙) 𝑆𝐴𝐺 = 𝐺𝑡𝑜𝑡 2 (1 − cos(𝜙)) (8) In addition, we assume that the polymerization changes the angle 𝜙 at a constant rate (𝐵). This assumption is governed by the hypothesis that Gag assembly is driven by addition of 36 new Gag monomers to the periphery of the Gag lattice, thereby being proportional to the number of gag molecules available for binding at the leading ring of the forming HIV-1 particle. Our mathematical model consists of the following system of ordinary differential equations, which is based on an application of the chain rule and the above derivation: 𝑑𝑁 d𝑡 d𝜙 d𝑡 𝑑𝑁 d𝜙 = 𝑑𝜙 d𝑡 (8a) =𝐵 (8b) We assume that the solution to (8a) is proportional to the total intensity (𝐼) of the GagGFP, which is given by equation (9) as shown in the next page. However, the resolution of the image or the filtering of the fluorescence may affect the time at which we begin to detect the fluorescence. In order to incorporate this into our solution, we start our time t at the time when the VLP becomes visible above noise, but start the shell filling model at time 𝜏1 . This condition means that for most cases, 𝜏1 < 0 as shown in Figure 4. In addition, once the VLP has completely formed, the intensity of the Gag-GFP will reach its maximum (𝐴). Therefore, we define 𝐼(𝑡)piecewise as: 𝐴 𝐼(𝑡) = 2 (1 − cos(𝐵𝑡)) (9)  0  A I (t ) =  (1 − cos(B  (t −  1 )) 2  A   if t   1 if  1  t  if t   B  B (10) + 1 + 1 2.3.8 Parameter estimation Maximum likelihood estimation (MLE) can be used to estimate the three unknown parameters 𝐴, 𝐵, and 𝜏1 of (10) 187. This method selects values of the parameters that give 37 the experimental data the greatest probability under the assumption that the error 𝜀𝑖 has a normal distribution with mean zero and variance 𝜎 2 . Under this assumption, MLE is equivalent to least squares regression, so we use a nonlinear least squares curve fitting routine in Matlab to determine our parameter values and estimate 𝜎 as the square root of the sum of the squares of the errors divided by the length of the data set. We use these parameter estimates to calculate the log-likelihood estimate of (10). The likelihood ratio test is then used to compare two nested models to determine which model is more likely to produce the experimental data 188 . The first model that we derived (10) does not include any parameters that describe the pause-like behavior in the experimental data. In order to use the likelihood ratio test, we add two unknown parameters to (10) to describe a single pause in the data. The parameter 𝜏𝑃1 defines the starting time of the pause and the parameter 𝑝1 defines the length of the pause. We assume that the parameter 𝐵 does not change due to the pause. The second model can be defined piecewise as:  0   A (1 − cos(B(t −  1 ))) 2  A I (t ) =  (1 − cos(B( 1P −  1 ))) 2 A  2 (1 − cos(B(t − p1 −  1 )))  A  if t   1 if  1  t   1P if  P1  t   1P + p1 if  P1 + p1  t  if t   B  B +  1 + p1 (11) +  1 + p1 We use the log-likelihood estimate for both (10) and (11) to calculate the test statistic. The test statistic is equal to twice the difference in the log-likelihood estimates of the two models. To adjust for multiple comparisons, we use the Bonferroni correction. We reject 38 the null model when 𝑝 < 0.05 𝑛 , where 𝑛 is the number of data sets we compare. Lastly, we consider a final case which adds two additional unknown parameters to (11) to describe two pauses in the data. The parameter 𝜏𝑃2 defines the starting time of the second pause and the parameter 𝑝2 defines the length of the second pause. Again, we assume that the parameter (𝐵) does not change due to either pause. The third model can be defined piecewise as equation (12). The curve-fitting routine in pause breaks into two consecutive pauses. In order to avoid nonidentifiability, Matlab may not be able to clearly distinguish between two distinct pauses or one long that was of 𝜏𝑃1 and 𝜏𝑃2 , so we alter the routine to estimate the seven unknown parameters of (12) using subsets of data.  0 A  (1 − cos(B(t −  1 ))) 2 A 1  (1 − cos(B( P −  1 ))) 2 A I(t ) =  (1 − cos(B(t − p1 −  1 ))) 2  A (1 − cos(B( 2 −  ))) P 1 2   A (1 − cos(B(t − p1 − p2 −  1 ))) 2  A  if t   1 if  1  t   P1 if  P1  t   P1 + p1 (12) if  P1 + p1  t   P2 if  P2  t   P2 + p2 if  P2 + p2  t  if t   B  B + p1 + p2 +  1 + p1 + p2 +  1 To do this, we first determine the most likely value of an additional pause using only the experimental data before the value 𝜏𝑃1 as we determined in (11). We then consider the experimental data after the value 𝜏𝑃1 + 𝑝1 from (11) and use these data to determine the most likely value of an additional pause. We compute the likelihood estimate associated with these different values to determine which values associated with the additional pause 39 are more likely when paired with the value of 𝜏𝑃1 and 𝑝1 from (11). Once we have assigned a value to 𝜏𝑃1 , 𝜏𝑃2 , 𝑝1 , and 𝑝2 (12), we then use the curve-fitting routine in Matlab to determine the remaining unknown values and then calculate the log-likelihood estimate of the third model. We use the likelihood ratio test to compare the third model (12) to the other two models, (10) and (11). 2.3.9 Restrictions on parameter estimates In order to prevent the curve-fitting routine in Matlab from identifying a pause in the data once an equilibrium has been reached, we restrict the data to only include values where 𝐼(𝑡) < 0.9𝐴. We also prevent the routine from identifying a pause before the start time of the experimental data. 2.4 Results As Gag concentration within the cytoplasm rises, Gag proteins assemble on the plasma membrane to nucleate the formation of a VLP. These assembly sites become visible between 3-7 hours posttransfection of Gag and Gag-eGFP in Hela cells. The nucleation starts with a single site on the membrane and culminates in as many as 100-500 sites by 10 hours posttransfection. Nucleation of each VLP is followed by a period of Gag polymerization that results in increased fluorescence from the VLP on the plasma membrane. The polymerization of Gag is temporarily interrupted by a pause in Gag polymerization in 60% of VLPs, the identification and characterization of these pauses is the focus of this paper. Once the polymerization resumes and the VLP is complete, the fluorescence intensity reaches a plateau implying full assembly of the Gag within the VLP. 40 This plateau is followed by recruitment of ESCRT components to facilitate the release of the VLP. The Gag filling fraction is defined as the ratio of Gag at the moment of interest to the total Gag incorporated at the end of VLP formation. We measure the Gag filling fraction during each stage of the assembly to understand the dynamics of VLP formation. The intensity of the VLP in TIRF is a function of its number of Gag proteins as well as its azimuthal location within the penetration depth. To determine the amount of Gag incorporated independent of the TIRF field, we measured the TIRF intensity of the same VLP under two different penetration depths and used the ratio of these fluorescence signals to de-convolve the relative Gag incorporation in each VLP as explained in Appendix C and shown in Figure 2. To ensure only fully formed VLPs were analyzed, only VLPs which were imaged from their nucleation, stayed separated from other VLPs by at least 1 um, and recruited VPS4 at the end of assembly within the observation window of our microscopy were included in our analysis (58 spots from 7 cells). The experimentally derived relative Gag incorporations were fit by the shell filling model as explained in the methods. The theoretical fit allows the measurement of the Gag filling fraction corrected for the amount of Gag present during nucleation, although this initial Gag level is below the signal to noise level of detection. Our model assumes new Gag proteins are added at the periphery of the formed Gag lattice during the polymerization. We also assume that Gag assembly and curvature creation follow each other, therefore the periphery of the Gag lattice is assumed to be identical to the periphery of the forming VLP as shown in Figure 3 and explained in Appendix C. This model has the advantage of having a single rate constant for Gag addition (B=3.1×10-3 ± 2×10-3radians/second as defined by equation (8a) and measured for 58 41 VLP assembly events). The model was modified to allow incorporation of pauses as explained in more detail in methods. 2.4.1 VLP formation kinetics 2.4.1.1 Nucleation kinetics We imaged the assembly of Gag into VLPs on the plasma membrane of HeLa cells. During nucleation, the VLP is composed of a few Gag proteins assembling on the plasma membrane. Our measurements do not have the required signal to noise to pick the exact nucleation event, therefore we use the extrapolations of the shell filling model to estimate nucleation kinetics. Based on extrapolations of the shell filling model, we estimate the Gag filling fraction at the moment in which VLP emerges out of the background to be 12±10% for the 58 spots analyzed. The nucleation time 𝜏1 =-5.5 ± 4.1 minutes was measured for the 58 VLPs and is defined as the estimated time required for the VLP to nucleate and appear above signal to noise (Figure 4). The Gag filling fraction is defined using the theoretical fit as the ratio of Gag at the moment of interest to the total Gag assembled within the VLP as fit by the theoretical model. This definition allows an estimate of the Gag filling fraction at the first moment in which the VLP gets above signal to noise as shown in Figure 4. 2.4.1.2 Gag polymerization Once the forming VLP has enough signal to rise above background, Gag continues to polymerize until a plateau is reached and VPS4 is recruited to facilitate fission. The position of this plateau is determined by fitting the shell filling model. Gag polymerization time is therefore determined as the time between the first appearances of the VLP above 42 noise to the moment that maximum Gag filling fraction is achieved as determined by the fit to the shell filling model. This time is 𝜏2 =23.7 ± 11 minutes and is measured for all the 58 VLPs (Figure 4). The average Gag polymerization time for profiles with pauses is 26 ± 11 minutes measured for 37 VLPs while the average for profiles with no pause is 20 ± 9 minutes measured for 21 VLPs. 2.4.1.3 Kinetics of recruitment of VPS4 After full assembly of Gag into the VLP, VPS4 is recruited to assist with membrane fission. We have timed the recruitment of VPS4 with respect to the time of maximum Gag filling fraction. The VPS4 recruitment time for profiles with pauses is 𝜏3 =1.2 ± 10 minutes measured for 37 VLPs while the average recruitment time for profiles with no pause is 𝜏3 =4.8 ± 11 minutes measured for 21 VLPs. Figure 4e shows the distribution of 𝜏3 for all the profiles. 2.4.1.4 Lifetime of full assembly Lifetime of full assembly is defined as the time measured from nucleation to the moment of VPS4 recruitment at the end of Gag assembly. This lifetime for profiles with pauses is 33 ± 14 minutes measured for 37 VLPs while the average recruitment time for profiles with no pause is 30 ± 14 minutes measured for 21 VLPs. 2.4.2 Pauses during HIV-1 VLPs formation In 60% of the observed VLPs, the polymerization of Gag paused for a period longer than 2 minutes. After this pause, the polymerization resumed with kinetics consistent with 43 the shell filling model. Pauses are distributed between 2-20 minutes with an average of 4 minutes with a single exponential distribution and a time constant of 4.7 minutes (Figure 5). Our signal to noise at the initial stages of assembly is limited and therefore, we cannot detect any pauses below 20% of Gag filling fraction. We examined the associated Gag filling fractions of 40 pauses from 37 profiles. If pauses occur at a particular curvature or specific number of Gag molecules, then Gag filling fraction during the pause will identify the critical curvature or amount of Gag that generate pauses. Figure 5 shows the distribution of the observed pauses with respect to the Gag filling fraction and although pauses between 40-60% Gag filling fractions are slightly more prevalent, pauses can happen anywhere along the filling fraction. If the pauses during formation of VLPs occur independently, double and triple pauses would be common. We have identified 5% (3 out of 58) VLPs with a double pause as shown in Figure 5. Because the probability of having a single pause longer than 2 minutes is 60%, an independent model predicts the probability of two pauses longer than 2 minutes as 36%. The observed 5% double pause probability is significantly below this ratio, indicating that the incorporation of a single pause may reduce the probability of incorporating a second pause within the VLP. We would like to add that our current method is only sensitive to pauses that are longer than 2 minutes and thereby, we cannot exclude the presence of short pauses (< 2 minutes) in addition to the observed long pauses. In addition, we cannot detect pauses that occur near each other or near the beginning or end of VLP formation, making it possible that the deficit of double pauses is due to limited detection power. 44 2.4.3 Pauses are independent of Gag late domains To test whether the pauses are caused by interactions of Gag with members of the ESCRT pathway, we repeated our experiments using Gag mutants with double mutations deleting the PTAP and YPXL motifs which mediate the interactions with two major ESCRT proteins ALIX and TSG101. The results from the study of these mutants are shown in supplementary information and in Appendix C. For the mutant data, average values are τ1 (nucleation time) = -11.36±9.13 minutes, τ2 (polymerization time) = 25.75±17.29 minutes. The fraction of profiles with a pause is 71%. The average pause length is 6.04±3.47 minutes. These data suggest that there is no significant change in either the percentile of profiles with pauses and or the average length of the pauses, and that the pause mechanism is independent of interactions with the ESCRT pathway. 2.4.4 The effect of Gag concentration on the assembly kinetics The cytosolic concentration of Gag within the cell increases during the 1.5 hr imaging window. We have analyzed the rate of assembly (B), VLP lifetime (τ), and Pause length as shown in Figure 6 from single cells. We analyzed the distribution of these parameters for VLPs that formed within the first 45 minutes of the acquisition, therefore ensuring all of the events recruited VPS4 by the end of the 1.5 hours imaging period. We found that the rate of assembly (B) increased from 1.4×10-3 radians/second during the first 10 minutes to 0.0034 radians/second for VLPs starting between 30-40 minutes into the acquisition. Similarly, the lifetime of the VLPs decreased from 42 to 27 minutes between the same periods. This result is in agreement with similar observations from Jouvenet et al. 45 The pause length remains surprisingly constant during the 1.5 hours of observation, therefore suggesting that the pauses are not dependent on Gag concentration. 2.5 Discussion Here we report our measurements of polymerization of Gag into fully formed VLPs. Our measurements on kinetics of VLP formation are consistent with previous results. In experiments reported by Baumgartel et al. where the VLP formation was observed in cells for a duration of 90 minutes, the VPS4 spikes were observed mainly between 20-45 minutes after the initiation of the VLP on the membrane 190 which is consistent with our data. Jouvenet et al. reported polymerization time of HIV-1 VLPs as 420 minutes with an average of 12 minutes. In these experiments, however, the Gag proteins were labeled with the less bright mCherry fluorescent protein and the total acquisition on the cell membrane was 45 minutes, which can inherently discriminate against longer assembly times 115 . The initial observations of Gag assembly during VLP formation assumed completion of the assembly upon reaching a plateau, which can be misleading when traces with pauses within the assembly are analyzed. Our data show 60% of VLPs paused during Gag polymerization for 2-20 minutes and that pause lengths are distributed exponentially. The frequency of double pause events was 5%, much lower than expected if the pauses were independent from one another, although our methods might have lower power to detect double pauses. Pauses within the assembly of Gag can also be seen in data from other laboratories, for example Figure 4c, 5b, and 5d from Jouvenet et al. 115; however, since Jouvenet et al. data were acquired using single penetration depth TIRF, the presence of the pauses in fluorescence intensity could 46 not be directly linked to pauses in the incorporation of molecules as the fluorescence signal is not only a function of Gag accumulation but also the azimuthal VLP motion. Our method of imaging based on two penetration depth TIRF microscopy allows the direct measurement of Gag incorporation and identification of pauses within the HIV-1 Gag VLP assembly. The pauses in the Gag assembly show that polymerization of Gag does not always proceed perfectly and there is a kinetic defect during the VLP assembly. Given that the VLP is assembled by polymerization of thousands of molecules, it is peculiar that polymerization would come to an abrupt stop and then resume a few minutes later. The exponential distribution of pause durations indicates that the VLP assembly is halted waiting for a single event. Assembly of the VLP is a complex event orchestrated by polymerization of Gag, deformations in the plasma membrane, and possibly interactions with cellular factors during the assembly. In a simple way, once can either expect the pause to be caused by defects directly related to Gag-Gag interactions during polymerization or the defect can be caused by a failure of interaction between Gag and the cellular machinery involved during VLP assembly. Although there are many possible events that can cause this particular behavior, we propose the following two models to suggest possible culprits in pause formation. 47 2.5.1 Model 1: Lattice stress is relieved through incorporation of defects during pauses The Gag lattice in immature HIV-1 virions is formed through hexagonal Gag assemblies covering the internal membrane of the VLP. Because hexagons cannot cover a curved surface, defects incorporate within this lattice allowing for the closure of the VLP. Incorporation of a defect relieves the stress within the lattice and allows the closure of the VLP. Assuming that Gag polymerization is a Brownian ratchet 191, addition of new Gag molecules to the lattice is limited by the Brownian fluctuations of the membrane. When a section of the lattice polymerizes as a long range hexagonal lattice, this Gag lattice will create stress within the VLP and dampen the Brownian movement of the membrane required for assembly of new Gag molecules at the periphery. This results in a pause, waiting for large enough Brownian movement that will allow incorporation of a big defect. This defect will relax the membrane and allow continuation of the assembly. In this model, once the lattice has been relaxed through incorporation of a pause, the secondary pause becomes less likely as shown in Figure 7. 2.5.2 Model 2: An enzymatic event essential for further polymerization We do not know of a particular host protein enzyme that is required for membrane remodeling. However, ubiquitin ligases influence the budding of HIV-1 181, 192-193 . Recently finding that Ubiquitin ligase NEDD4L and AMOT facilitated virus budding interpedently and AMOT depletion inhibited the virus budding from early assembly stage 194 , but ubiquitin dependent budding of Gag is not yet well understood; it is not clear at what stage of VLP formation the ubiquitin effects play a role and how they contribute to 48 membrane remodeling during VLP formation. Here we propose that the pauses may be caused by temporary shortage of ubiquitin-related enzymatic activity. Once the ubiquitin pathway is better understood, one way to test this hypothesis will be to study the effect of crucial enzymes within the pathway on VLP assembly and pause formation. 2.6 Conclusion Our method of imaging based on two penetration depth TIRF microscopy allows the identification of pauses within the HIV-1 Gag VLP assembly. Although VLP formation was imaged previously, none of the previous studies had the capability to calculate the relative number of added molecules with high sensitivity to demonstrate that the observed pauses in fluorescence intensity are actually due to a pause in assembly versus assembly combined by movement of the VLP out of the TIRF field. We have observed that the pauses within the assembly of HIV-1 have an exponential distribution with a decay time of 4.7 minutes. Our signal to noise resolution did not allow quantification of pauses below 2 minutes, therefore we cannot rule out the presence of a faster rate below 2 minutes. The kinetics above 2 minutes are governed through a single rate, which means there is a single stochastic process which has to be overcome for continuation of the VLP assembly. Expression of Gag within the cytoplasm of host cell is sufficient for assembly of HIV-1 VLPs; however, attempts to reconstruct VLP formation in vitro have not been successful. A simple Gag centric model of polymerization would predict that assembly of HIV-1 VLPs would proceed uninterrupted given that the concentration of Gag within the cytosol during assembly is sufficient. The characterized pauses within the assembly of 49 VLP, which occur in the presence of sufficient Gag within the cytosol and in the absence of any interactions with ESCRT components, argue that the assembly is not only limited by Gag availability but also by other cellular or biophysical factors. Proper identification of these elements may be required for reconstructing VLP formation in vitro as well as a complete understanding of the VLP assembly process. 2.7 Acknowledgements We thank Dr. Wesley Sundquist for helping us with testing of the VPS4 plasmids and helpful discussions. We also thank Dr. Michael B. Landesman for constructive comments and reading of the manuscript. This work was supported by NSF grants 1121972 (SS) and a Complex Systems grant from the James S. McDonnell Foundation (FRA) 50 Figure 2. Multicolor dual-penetration-depth TIRF microscopy. HeLa cells stably expressing VPS4A-h37-mCherry were transfected with Gag and Gag-eGFP plasmids and imaged 3–7 h posttransfection. (A) Assembly of an HIV VLP captured by imaging with 100- and 200-nm TIRF excitation in 488 nm for GFP followed by 200- nm TIRF excitation in 561 nm for mCherry every 15 s. Assembly initiates de novo and concludes with VPS4A recruitment at the end. (B) A schematic model of VLP formation and VPS4 recruitment. (C) Fluorescence intensity signal from the same VLP as shown in panel A. (D) The Gag filling fraction calculated for the same traces shown in panel B as explained in Materials and Methods. 51 Figure 3. Shell-filling model of VLP assembly. (A) Gag polymerizes on the plasma membrane to form a VLP. (B) We have approximated this process by assuming that new Gag molecules can only join at the periphery of the forming VLP. We also assume that the curvature of the VLP is always identical to the fully formed VLP, therefore parts of the lattice that are built already have the curvature that they would have in the final VLP. 52 Figure 4. Kinetics of VLP formation. (A) Definition of various timings on the graph of Gag polymerization (blue dot) and VPS4 (red) simulated data and Gag polymerization shell-filling model fitting (blue solid). We define τ as the VLP formation lifetime, τ1 as the nucleation time defined by the period of time that Gag VLP had started to assemble but intensity was still below the detection, τ2 as the Gag polymerization time, and τ3 as the recruitment time of VPS4 after Gag polymerization was complete. (B–E) Histograms of different time periods, τ, τ1, τ2, and τ3 from all 58 VLP formation profiles. Average values are τ = 31.9 ± 14 min, τ1 = −5.5 ± 4.1 min, τ2 = 23.7 ± 11 min, and τ3 = 2.54 ± 11 min. 53 Figure 5. Kinetics of pauses. (A–C) Three different Gag polymerization behaviors, including no-pause, single pause, and double pause of Gag polymerization. (Blue dot) Gag polymerization experimental data; (blue line) shell-filling model fitting; and (dark pink line) pauses. (D) Frequency of three different behaviors. Probability of having a single pause is 60%, double pause is 5%. (E) Histogram of pause length in min. (Dark blue line) Singleexponential distribution with a time constant of 4.7 min. (F) Histogram of Gag filling fraction in pause (%). Note that we cannot efficiently detect pauses below 20% filling fraction. 54 Figure 6. HIV-1 polymerase rate, completion time, and pause length A) The plot shows polymerization rate from panel B versus time of the VLP formation during the imaging period. The rate of polymerization increases with the concentration of Gag as expected. (B) The VLP formation time drops with the increasing Gag concentration. (C) The probability of a pause or the pause length is not changed during the observation period in the same cells. 55 Figure 7. Two models for pause formation during Gag polymerization. The mechanism governing the pause is a single stochastic event. We hypothesize that this single event can either be incorporation of a defect that will relax the stress within the curving hexagonal Gag lattice and allow the polymerization to continue (A), or evidence of a molecular event such as ubiquitination of Gag that is essential for continuation of polymerization (B). Both of these models satisfy the data by providing a single stochastic event that is unique during polymerization. CHAPTER 3 VISUALIZING ALIX RECRUITMENT DURING THE HIV-1 BUDDING 3.1 Abstract Polymerization of Gag on the inner leaflet of the plasma membrane drives the assembly of Human Immunodeficiency Virus 1 (HIV-1). Gag recruits components of the endosomal sorting complexes required for transport (ESCRT) to facilitate membrane fission and virion release. ESCRT assembly is initiated by recruitment of ALIX and TSG101/ESCRT-I, which bind directly to the viral Gag protein and then recruit the downstream ESCRT-III and VPS4 factors to complete the budding process. In contrast to previous models, we show that ALIX is recruited transiently at the end of Gag assembly, and that most ALIX molecules are recycled into the cytosol as the virus buds, although a fraction remain within the virion. Our experiments imply that ALIX is recruited to the neck of the assembling virion and is mostly recycled after virion release. 3.2 Introduction As the HIV-1 Gag protein concentration increases in the cytosol, the protein polymerizes on the plasma membrane to create nascent virion. Late assembly domains on C-terminus of Gag are required for efficient budding of infectious virions 195-196 . These motifs function by interacting directly with early acting components of the endosomal 57 sorting complexes required for transport (ESCRT) pathway. Specifically, the Gagp6 PTAP motif interacts with the TSG101 subunit of the ESCRT-I complex 197-200 and the YPXnL motif interacts with ALIX 201-204. ALIX also interacts with upstream NC Gag elements 205206 . In most cell types, the PTAP element appears to be the dominant late assembly domain 195, 207 . However, mutations that remove the PTAP element can be substantially rescued by ALIX overexpression, implying that ALIX and ESCRT-I can function redundantly, at least in some contexts 208-209. Furthermore, some related retroviruses like the Equine Infectious Anemia Virus (EIAV) lack ESCRT-I binding sites and appear to bud exclusively through ALIX, implying that ALIX alone can initiate assembly of the ESCRT machinery 201-203, 210. The HIV-1 Gagp6 YPXnL late assembly domain binds in a pocket on the inner face of the second arm of the ALIXV domain 208, 211 and series of positively charged residues on the ALIXBro1 domain interact with GagNC 212 . In addition to these Gag interactions, the PSAP motif within the ALIXPRR can bind TSG101/ESCRT-I 201-203, and the second arm of the ALIXV can bind Lys-63 polyubiquitin chains 213-214. The PSAP motif within ALIX has not been shown to contribute to ALIX recruitment to sites of virus budding, but ubiquitin binding contributes to ALIX recruitment and/or bud functions 208, 213, 215 . Current models hold that ALIX is activated through release of PRR autoinhibition and ALIXV domain opening. These conformational changes may increase membrane binding affinity and/or induce protein dimerization 218 216-217 which encourage ALIX binding to ESCRT-III subunits of the CHMP4 family through interactions at its ALIXBro1 208, 218 , 219-221. Functional and imaging analyses have demonstrated central roles in the HIV-1 budding for two of the ESCRT-III families; CHMP4 (three members, CHMP4A-C) and CHMP2 (two members, CHMP2A, B). Two additional ESCRT-III families appear to 58 perform accessory roles; CHMP1 (two members, CHMP1A, B) and CHMP3 114, 222 . Although mechanistic details are lacking, the ESCRT-III subunits are thought to polymerize into filaments that constrict the bud neck and promote membrane fission 222-227 114, . The ESCRT-III filaments also recruit essential AAA ATPase, VPS4, which is thought to remodel and/or disassemble the filaments 126, 228-229 . VPS4 is essential for The HIV-1 budding and is recruited at the end stages of virion assembly 115, 190, 227 , 230. In recent years, imaging techniques such as total internal reflection fluorescence microscopy (TIR-FM) have been used to visualize viral assembly and budding from living cells. TIR-FM provides excellent spatial and temporal resolution and is ideally suited for visualizing HIV-1 assembly at the plasma membrane because the excitation field at the coverslip-cell interface decays exponentially, and therefore only excites fluorophores that are within about 100 nm of the plasma membrane 185, 231-232 . This property eliminates confounding background fluorescence from molecules that are deeper within the cell. TIR-FM has been used to define many dynamic features of viral assembly including virion assembly 180, 189, 230, 233, viral genome capture 234, and ESCRT protein recruitment 115, 190 . Elegant TIR-FM studies of HIV-1 and EIAV assembly have shown that the ESCRT- III and VPS4 proteins arrive late, after Gag assembly is nearly complete and immediately preceding the budding event. As the VLP buds, the ESCRT-III and VPS4 proteins appear to be released back into the cytoplasm rather than remaining within the virion 115, 190, 230. A recent super resolution imaging study has challenged this model arguing that, instead, the bulk of the recruited CHMP4B remains within the released virion 235. This discrepancy has important implications for the mechanism of the HIV-1 budding because retention of CHMP4B implies that the ESCRT-III assembly constricts the bud neck from the virion- 59 proximal side, whereas release of ESCRT-III proteins back into the cytoplasm implies that the ESCRT-III filaments constrict the bud neck from the cell-proximal side. In contrast to the late-acting ESCRT factors, the dynamic recruitment of the earlyacting ESCRT proteins remains less well characterized because previous experiments lacked the sensitivity required to visualize ALIX or ESCRT-I recruitment to sites of the HIV-1 budding. It has, however, been possible to visualize ALIX recruitment to sites of EIAV p9Gag assembly 115, presumably because ALIX binds more efficiently to the YPXnL motif of EIAV p9Gag than to the one of HIV-1 p6Gag 236. In the EIAV Gag studies, ALIX recruitment was seen to initiate early and accumulate steadily in parallel with Gag accumulation 115 . Unlike ESCRT-III and VPS4, most of the ALIX molecules remained associated with the budded EIAV virions, suggesting different dynamic recruitment profiles for the early-acting ALIX factor versus the late-acting ESCRT-III and VPS4 factors. The importance of HIV-1 as both a medical problem and as the leading model system for studying virus budding prompted us to attempt to visualize ALIX recruitment to sites of the HIV-1 budding. Using a fluorescently-labeled ALIX protein that is fully functional in the HIV-1 budding, we were able to visualize ALIX recruitment to sites of HIV-1 Gag VLP assembly. The recruitment profiles revealed that ALIX is recruited late in the assembly process and that most, but not all, of the ALIX molecules are released back into the cytoplasm upon VLP release. 60 3.3 Materials and Methods 3.3.1 Infectivity assay Virions were produced in 293T cells (4×105 cells/well in 6-well plates) following calcium-phosphate (Clontech) co-transfection of the following plasmids: 1.35 µg of an HIV-1 RΔ8.2 construct (either wild-type or incorporating a 7LIRL10 instead of 7PTAP10 in the p6 domain of Gag (∆PTAP)195), 1.35 µg of pLOX-GFP237, 0.4 µg of pCMV-VSV-G, and 1 µg of the negative control empty vector (pCI-FLAG-EV) or ALIX expression vector. The medium (2 ml/plate) was replaced 8 hours after transfection, and the supernatant was harvested 16 hours later and syringe-filtered through 0.45 µm membranes. Viral titers were measured using flow cytometry/fluorescence-activated cell sorting (FACS) to detect green fluorescent protein (eGFP) expression from the packaged pLOX-GFP vector in transduced HeLa cells. For western blotting, cells were harvested with cold PBS and lysed in 200 µL of RIPA (50 mMTris, 150 mMNaCl, 0.1% SDS, 0.5 % sodium deoxycholate, 1% NP40, and complete protease inhibitor cocktail (Sigma)). Virions were collected by pelleting 1 mL of medium through a 250 µL 20% sucrose cushion. The supernatant was removed, and the virion pellet was dissolved in 60 µL of SDS-PAGE loading buffer. Western blots were performed using anti-CA, 1:3000 (NIH AIDS Research & Reference Reagent Program, Catalog #3537) and anti-ALIX 1:5000 208 3.3.2 Cell culture and transfection 293T and HeLa cells were maintained in Dulbecco's modified Eagle's medium supplemented with fetal calf serum (10%), sodium pyruvate (1 mM), and L-glutamine (2 mM). HeLa cell lines stably expressing eGFP-tagged ALIX were generated by transfection 61 with Lipofectamine 2000 (Invitrogen), followed by Geneticin (G418, Invitrogen) selection (0.5 mg/mL) and flow cytometry sorting. Single-cell clones expressing the proper levels of the fusion protein were selected for experiments. For TIR-FM experiments, HeLa cells (80% confluence in 35mm plates) were transfected with untagged Gag and Gag-mCherry in a 1:1 ratio using Lipofectamine 2000. 3.3.3 Microscope description Live images were acquired using an iMIC Digital Microscope made by TILL photonics controlled by TILL’s Live Acquisition imaging software as previously described 230 . Two wavelengths of laser, 488 nm diode laser (Toptic photonics, iBeam smart 488S) and 561 n diode-pumped solid state (DPSS) laser (Cobolt Jive, 561nm Jive High Power), were used to excite eGFP and mCherry, respectively. Laser beams passed through an AOTF (acousto-optical tunable filter) and focused into a fiber which delivers the light to a TILL Yanus digital scan head and then a Polytrope II optical mode switch (Diagram shown in supplementary materials). Yanus consists of two Galvo-mirrors and one spherical mirror to control the laser beam position. The Polytrope rapidly switches illumination beam path between Epi (widefield), FRAP, and TIRF microscopy modes. It also holds the quadrant photodiode used for TIRF penetration depth calibration, which was set to 150 nm for the experiments in this manuscript. In the TIRF mode, Yanus is used to control the position of the focused beam in the objective’s back focal plane and can be adjusted within 0.2ms. We positioned the focused beam at the edge of the back focal plane of the objective (N=1.46, 100X, Zeiss) to reach beyond the critical angle and achieve TIRF. TIRF critical angle was verified by scanning the laser beam across the back aperture and measuring the reflection 62 of the laser from the Glass sample interface back into the objective and onto the quadrant photodiode 230. The penetration depth of the beam is calculated based on the incident angle of the beam which is in turn measured by the position of the beam on the quadrant photodiode. Once the penetration depths for the experiments are set at the beginning of acquisition, a feedback loop keeps the focus of the objective on the sample by constantly monitoring the position of the back reflected beam with respect to the original beam. We also rotated the TIRF illumination on the objective back focal plane 1 turn / exposure (TIRF360) to maximize homogeneity of the TIRF images. 3.4 Results 3.4.1 ALIX linked to eGFP at its C terminus with a 30 amino acid helical linker supports production of HIV-1 with equal infectivity as wild-type ALIX Six DNA constructs were created that encoded the joining of ALIX to eGFP with protein linkages of materially different character. The rationale behind constructing linkers was to create some distance between eGFP and ALIX. However, since it is not clear how ALIX fits into the functioning ESCRT machinery, a library of linkers was created so that various lengths and orientations may be tested. DNA sequences were generated using two helical linkers of similar composition, but of varying length (20 and 30 amino acids), and one flexible linker of 28 amino acids in length 238. As shown in Figure 8, all three linkers were used to generate ALIX fusions with eGFP in both gene orders, i.e. eGFP before ALIX and ALIX before eGFP, for six total linkage combinations. 63 Infectivity of HIV-1 virions is significantly affected by a ∆PTAP mutation (motif 7LIRL10 instead of 7PTAP10 195). This inefficiency, which is due to inability to recruit TSG101, can be rescued by overexpressing ALIX 208-209 . We used this system to test the proper assembly of HIV-1 PTAP- in the presence of the six ALIX fusion constructs. Figure 9a shows the infectivity of HIV-1 PTAP- virions harvested from 293 cells under overexpression of various ALIX fusion proteins; the constructs are overexpressed to similar levels as shown in Figure 8. The ALIX-h30-eGFP (CT.3 #7) as shown in the figure supports assembly of HIV-1 PTAP- virions with equivalent infectivity to wild-type ALIX. Western blots on harvested virions as shown in Figure 9b show incorporation of ALIXh30-eGFP into harvested virions. We further tested the specificity of incorporation of ALIX-h30-eGFP into VLPs by transfecting ALIX-h30-eGFP and HIV-1∆YP into 293T cells. As shown in Figure 10 , the incorporation of ALIX into VLPs was dependent on its expression level and is substantially diminished in VLPs produced by HIV-1∆YP mutant (YP to SR mutation 201). The ALIX-h30-eGFP was transfected into Hela cells and used to create a cell line with stable expression of ALIX-h30-eGFP. This cell line was then cloned and cells from a single clone were grown for experiments. In this clone, ALIX-h30-eGFP is expressed stably at a similar level of expression when compared to wild-type ALIX as shown in Figure 11. A minimal amount of ALIX as well as ALIX-h30-eGFP is released from these cells in the absence of Gag due most likely to cellular basal microvesicle formation. This clone, named HeLa-C1, was further used in the imaging experiments. 64 3.4.2 ALIX incorporates transiently at the end of Gag assembly during the formation of HIV-1 VLPs The HeLa-C1 Cells were transfected with a mixture of Gag and Gag-mCherry plasmids as explained in methods. HIV-1 Gag VLPs were observed forming between 4-5 hours posttransfection. Cells were imaged for 90 minutes with a frame rate of 20 seconds /frame. TIRF images in both 488 and 561 channels corresponding to ALIX-eGFP and GagmCherry were acquired independently with 100mSec in between the two acquisitions. As shown in Figure 12 (panels A and B), HIV-1 VLPs nucleated on the plasma membrane and the fluorescence intensity grew over time, indicating Gag polymerization and incorporation within the VLP. When Gag polymerization was complete, the fluorescence intensity detected from the VLP reached a plateau, indicating completion of assembly 230 . We analyzed the incorporation of Gag and ALIX in 44 VLPs from 5 cells. In all events, ALIXh30-eGFP was recruited into the VLP near the final plateau of Gag assembly. To better determine the relationship between the incorporation of ALIX and Gag polymerization kinetics, we used the shell filling model developed previously to fit the Gag polymerization profiles to determine the fluorescence intensity at the completion of assembly 230. As shown in Figure 13e, based on this analysis, in 40 out of 44 observed VLPs, ALIX-h30-eGFP was recruited when 100% of the Gag was assembled, in 4 out of 44 VLPs, ALIX-h30-eGFP was observed when 90% of the Gag was assembled and lower percentiles were not observed. As shown in Figure 11a and 11b, two distinct behaviors of ALIX were observed. In 34 of the 44 observed VLPs, less than 20% of recruited ALIX was retained within the VLP after fission. This leftover ALIX was not reliably detectable given the background fluorescence. 10 of the 44 observed VLPs retained a detectable fraction of the ALIX-h30- 65 eGFP (Figure 13c). To further investigate the timing of recruitment of ALIX, we transfected Hela-C1 cells with VPS4-h37-mCherry 230 and HIV-1 Gag expressing plasmids. In these cells, the recruitment of ALIX is observed overlapping with VPS4 as shown in Appendix D. The temporal resolution of 15 seconds used in these experiments, however, is not sufficient to resolve the relative timing of recruitment of ALIX with respect to VPS4. 3.4.3 ALIX incorporates transiently at the end of Gag assembly during the formation of EIAV VLPs EIAV assembly is driven by EIAV Gag which has a very similar domain distribution to HIV-1. EIAV, however, has a p9 domain that lacks the PTAP-TSG101 binding motif, and instead binds strongly to ALIX through its YPXL motif. The affinity of EIAV p9Gag to ALIX is higher than that of HIV-1 p6Gag 236. To measure the recruitment of ALIX into EIAV Gag VLPs, we imaged the assembly of EIAV VLPs in the Hela-C1 cell line expressing ALIX-h30-eGFP. As shown in Figure 11, panels C and D, polymerization of EIAV Gag reaches a plateau similar to HIV-1 Gag polymerization, at which time ALIX-h30-eGFP is observed recruiting into the budding VLPs. We analyzed the incorporation of EIAV Gag and ALIX in 33 EIAV VLPs in 5 cells, and in all events, ALIX-h30-eGFP was recruited into the VLPs near the final plateau of Gag assembly (Figure 12d). In 6 of the 33 observed VLPs, less than 20% of recruited ALIX was retained within the VLP; this leftover ALIX was not reliably detectable above the background. 27 of the 33 observed VLPs, however, retained a detectable amount of ALIX-h30-eGFP, with 18 retaining more than 80% of recruited ALIX, as shown in Figure 12b. The enhanced 66 retention of ALIX within EIAV Gag VLPs is consistent with the higher affinity of EIAV p9Gag compared to HIV-1 p6Gag. 3.4.4 Comparison between ALIX-h30-eGFP recruitment to EIAV and HIV-1 Gag during VLP assembly As shown in Figure 12, ALIX recruitment to both HIV-1 and EIAV Gag is through transient incorporation of ALIX at the end of Gag polymerization. The EIAV and HIV-1 profiles are only different in terms of the retention level of ALIX within the VLPs. We defined the retention factor as the ratio of maximum intensity reached by ALIX at the peak of recruitment to the intensity of ALIX remaining in the VLPs after the initial transient recruitment. Figure 12 demonstrates that while the retention factor for HIV-1 is 10%, the EIAV Gag has a much higher average retention factor of 40%, which is consistent with the stronger affinity of the EIAV p9Gag region for ALIX compared to that of HIV-1. While it is challenging to observe ALIX recruitment before the peak of recruitment in individual profiles, we aligned all the ALIX fluorescence intensity profiles based on the position of the recruitment peak of ALIX to show the averaged intensity from all profiles (Figure 12a). Based on this averaging, we can observe a slight linear recruitment of ALIX as early as 5 minutes prior recruitment peak. 3.5 Discussion Our data show that ALIX is recruited transiently into forming HIV-1 as well as EIAV VLPs after polymerization of Gag is complete. ALIX interacts directly with Gag through the GagP6 208, 211, GagNC 205-206, 212, as well as indirectly through ubiquitin binding 67 213-215 and TSG101 201-203. Since Gag is abundant within the VLPs, the sudden recruitment of ALIX into formed VLPs cannot be fully explained only by its direct interactions with Gag. The observed ALIX recruitment at the end of Gag assembly supports a model of recruitment where ALIX is recruited to the neck of the formed VLP. It was shown by imaging ESCRT recruitment onto giant unilamellar vesicles (GUV) that during invagination of membrane, elements of ESCRT-I and -II proteins are localized to the periphery of the invagination and remain on the cytosolic side of the membrane even after secession of the formed vesicle 239. These experiments are conducted in the context of the MVB pathway and the function of ESCRT machinery in MVBs is dominated by ubiquitin binding. Early studies identified a link between ubiquitin and retrovirus release 161, 240-241 and recent reports are unraveling more about the role of ubiquitin in The HIV-1 budding. Indeed, ubiquitin ligases were clearly shown to influence the budding of HIV-1 181, 192-193 and ubiquitin conjugation to Gag appears to be essential for ESCRT mediated The HIV-1 budding 164, 215 213-214, 242 . It is therefore possible that ubiquitin linked interactions might assist/control the . Interestingly, ALIX binds specifically to ubiquitin through its V domain specific recruitment of ALIX onto the neck of the fully assembled Gag lattice during virus budding. An ALIX fusion with GFP at the Bro1 domain was shown to recruit from the beginning of the EIAV Gag assembly exhibits defects in infectivity assays 202 115 . The Bro1 domain fusion to GFP, however, . When average recruitment profiles are analyzed from the experiments presented here as shown in Figure 12, our data also support some earlier recruitment of ALIX during the VLPs formation; however, this amount is minimal 68 compared to the major recruitment at the end of assembly. ALIX can be activated through opening of its ALIXV domain and potential dimerization which results in higher affinity for the membrane and CHMP4 216-218; therefore, it is possible that labeling with fluorescent proteins at the Bro1 domain can activate ALIX. During HIV-1 and EIAV budding, our results show that a portion of ALIX is retained within released virions, with a higher retention rate in EIAV compared to HIV-1, consistently with the higher affinity of EIAV p9 Gag to ALIX when compared to HIV-1 p6Gag 236. The removal of a significant portion of ALIX after initial incorporation into the formed VLP is still puzzling. Initially, we speculated that the loss of ALIX signal is due to self-quenching of eGFP based on the close proximity of packaged eGFPs trapped within the VLP after fission of the membrane 243-244. Although we still cannot completely rule out some self-quenching effects, given that EIAV has very similar VLP size to HIV-1 and the ALIX signal at the end of EIAV assembly is mostly retained, the self-quenching of eGFP cannot convincingly explain the loss of eGFP signal, especially in the HIV-1 case. Therefore, it is reasonable to assume that the loss of eGFP signal has to do with dissociation of ALIX from the VLP during and/or after fission of the membrane. ESCRT III proteins have been shown to polymerize into spiral structures on the plasma membrane 222-224 ; it is therefore possible that ALIX would dissociate from the Gag lattice due to the forces applied during either polymerization of CHMP4, CHMP2, or recruitment of VPS4. 3.6 Acknowledgements We thank Dr. Wesley Sundquist for helpful discussions and help with editing the manuscript. 69 Figure 8. ALIX eGFP fusion protein library. ALIX was fused to eGFP using three designed linkers (f, h20 and h30). (A) Shows the designation of the fusion proteins with the orientation of the linker fusions. (B) The amino acid sequence of the three linkers. 70 Figure 9. Characterization of the ALIX fusion proteins with eGFP. A) Cells expressing ALIX fusions (lanes 5–10), Wild-Type ALIX (lanes 2 and 4), or control (lanes 1 and 3) where infected with Wild-Type (Lanes 1 and 2) as well as DPTAP HIV-1 RD8.2 (Lanes 3–10). Assembled virions were harvested 16 hours post infection and viral titers were measured using FACS to detect eGFP expression from the packaged pLOX-GFP veactor in transduced HeLa cells. B) Western blot analysis of cells and collected virions from the same experiment presented in A. 71 Figure 10. Incorporation of ALIX-h30-eGFP is dependent on the Gag late domains. 293T cells were transfected with either RD8.2 variants (WT or DYP) alone or with increasing amounts of the ALIX-h30-GFP expressing vector. 24 hours posttransfection, cell lysates and VLPs were analyzed by western blot using the indicated antibodies. 72 Figure 11. VLP formation in HeLa-C1 cells stably expressing ALIXh30-eGFP. HeLa and HeLa-C1 cells were transfected with DR8.2 variants (WT or DPTAP). 24 hours posttransfection, cell lysates and VLPs were analyzed by western blot using the indicated antibodies. 73 Figure 12. Incorporation of ALIX-h30-eGFP into forming HIV-1 and EIAV VLPs. HeLa-C1 cells stably expressing ALIX-h30-eGFP were transfected with either HIV-1 Gag and Gag-mCherry plasmids (A and B) or EIAV Gag and Gag-mCherry plasmids (C and D). In all, ALIX arrives at or near the completion of Gag assembly. 74 Figure 13. Characterization of ALIX recruitment. A) Temporal averaging of ALIX fluorescence intensity, with all profiles lined up so the peak recruitment is positioned at time zero. This averaging shows early ALIX recruitment up to 5 minutes before appearing as a spike. The retained ALIX is also visible in HIV1. B) Distribution of ALIX retention in EIAV VLPs. C) Distribution of ALIX retention in HIV-1 VLPs. (D and E) show percentile of total accumulated Gag within the VLP at the time of peak ALIX recruitment for EIAV and HIV-1, respectively. CHAPTER 4 MECHANISM OF ALIX RECRUITMENT IN HIV-1 4.1 Abstract Endosomal sorting complexes required for transport (ESCRTs) are protein complexes that facilitate the release of most cellular processes requiring fission of budding membranes including multivesicular of budding vesicles into the extracellular environment. ESCRTs are essential in body formation, envelope virus release, and exosome release. During The HIV-1 budding, HIV-1 Gag assembles on the plasma membrane and drives the outward deformation of the membrane toward forming vesicles. HIV-1 like particles (VLPs) can be generated by expressing only the HIV-1 Gag polyprotein in the cytosol. These VLPs consist of ~2000 copies of Gag and form vesicles with an average diameter of 120 nanometers. Their assembly on the plasma membrane takes ~20-30 minutes. HIV-1 utilizes ESCRTs extensively for the release of progeny virions into extracellular space, and VLPs serve as a good model system for studying ESCRT recruitment on the plasma membrane. Each Gag polyprotein is equipped with two late domain motifs which biochemically interact with TSG101 and ALIX, two of the early ESCRT components. Given the stoichiometry of Gag and its steady accumulation in the forming VLPs, it has been expected that ALIX would be recruited into the forming VLPs along with Gag during the assembly. Using live imaging to observe real-time assembly of VLPs in the presence of fluorescently tagged functional ALIX proteins, we have observed 76 that ALIX is recruited transiently at the end of Gag assembly and is mostly recycled after VLP release. This observation implies high spatial and temporal regulation for recruitment of ALIX during budding. In this work-in-process project, we have further dissected the essential interactions that govern ALIX recruitment and show that interaction of ALIXBro1 domain with CHMP4 is critical for recruitment of ALIX into budding sites. We will further discuss the emerging regulatory mechanism essential for recruitment of ALIX to the plasma membrane during vesicle budding. 4.2 Introduction ESCRT complex is used in membrane fission events that have similar topology, but there are still crucial differences. Even in the similar mechanism of retrovirus budding of EIAV and HIV-1, the distinct late domains make the ESCRT machinery pathway greatly varied. EIAV ESCRT-mediated budding is the simple model to study retrovirus budding where EIAV P9Gag has only one late domain, YPXnL, recruit CHMP4-CHMP2-VPS4 via ALIX to carry out the EIAV budding membrane fission. In EIAV ALIX budding pathway, ALIX plays an earlier-acting element with higher ALIX retention rate in complete VLPs, and the rest of ESCRT proteins except CHMP4, CHMP 2, and VPS4 might be dispensable 50-51, 57-58, 137, 204, 208, 245-247 . However, The HIV-1 budding mechanism, P6Gag PTAP late domain is responsible for recruiting TSG101 as the main pathway and YPXnL-ALIX is an auxiliary pathway. In our previous study, ALIX was recruited only transiently at the end of HIV-1 Gag assembly 248 which is discrepant with the PTAP-TSG101 mechanism, where TSG101 shows up at the beginning as an early factor during the HIV-1 budding 125 and is also consistent with the ALIX recruitment manner in EIAV 248 . Moreover, TSG101 is 77 recruited to a position closer to the center of Gag assemblage than ALIX and the other ESCRT-III proteins 125. Multiple pathways complicated the ESCRT mechanism in HIV-1 in some contexts for the pathway’s independence, redundancy, and the difficulty of dissection [63, 64]. Plus, a number of ALIX protein’s multiple characters entangled the HIV-1 budding to the next level, as well as to the other membrane fission events. First, ALIX binds to both YPXnL late domain and NC domain directly to Gag proteins 57. ALIX siRNA depletion has also been shown to markedly inhibit the release of EIAV but not HIV-1. Although the ALIXYPXnL pathway in the HIV-1 budding is less essential than the PTAP-TSG101 main pathway, it is clear that ALIX plays potentially important roles in efficient the HIV-1 budding, since the YPXnL mutants caused morphogenesis aberrant in cell-type-dependent defect (delay virus replication in T-cell lines and in primary T cells and macrophages) in virus replication 56 and overexpression of full-length ALIX 55 or ALIXBro1 45 potently rescued the defect of the PTAP-TSG101 pathway. NCGag two CCHC-type zinc fingers motifs were shown to interact with ALIXBro1 and, interestingly, ALIX is included into VLPs greatly dependent on NCGag rather than YPXnL 44. Furthermore, NCGag facilitating the HIV-1 budding is greatly dependent on the association between NCGag and the PM 249. Second, there is important interaction between ALIXBro1 and CHMP4. It has been shown that overexpression of ALIXI212D (CHMP4B binding site) failed to rescue the HIV-1 PTAP- particle release 55 and ALIX point mutations of CHMP4 binding sites inhibited The HIV-1 budding and abscission 96, 173, 208. Additionally, EM showed that ALIX bound to Cterminal truncated CHMP4 polymer and formed a ladder-decorated filament in vitro for possibly increasing stability 136, 250 . Third, the inactive (closed) and active (open) 78 conformation of ALIX protein and monomer and dimer conformation are decisive for ALIX’s activation. Although it is still unknown what intrigues ALIX active switch or induce dimerization, more evidence showed that ALIX activation or dimerization conformation is needed for ALIX to function properly. The conformational change has ALIX PRD released from ALIXBro1-V and ALIXV arms open which might permit or promote some specific binding sites or increase the affinity between ALIX and Gag/ESCRT proteins and the membrane to facilitate The HIV-1 budding 136-137 . Association of CHMP4 to ALIXBro1 does not cause the conformational change of ALIXBro1, but binding to YPXnL and membrane or phosphorylation have been proposed to be potential triggers for ALIX activation 61, 137 . Additionally, the trigger to inactivate or recycle the activated ALIX back to cytosol is also undefined. Lastly, a number of interactions of ALIX with other ESCRT-related proteins remain enigmatic for the detail mechanism, such as the direct interactions of the mysterious Ub and DUb mechanism to ALIXV and ALIXBro1, respectively, to complete HIV-1 budding, the possibly indirect relation from ALIX to UbL through NCGag, the bindings to TSG101 and the selfmultimerization of ALIX 50-51, 57 . Mutation of self-multimerization fails to rescue the overexpression of ALIX in HIV-1 PTAP- particle release panel, but does not affect EIAV and cytokinesis, corresponding to the overexpression of ALIX with high enough levels to overcome the low affinity between ALIX-YPXnL in HIV-1, which might emphasize the affinity’s significance for ALIX function 55, 152 . ALIX have more widespread influence than we currently recognize. Much work remains to be done before we gain a mechanistic understanding of ALIX. 79 Combining with early ALIX studies in HIV-1 and MVB, we proposed the model where the observed ALIX recruitment at the end of Gag assembly only comes to the neck of the formed VLP. The removal of a significant portion of ALIX after initial incorporation into the formed VLP is still puzzling, but we suspect that ALIX dissociates from the Gag lattice due to the forces applied during either polymerization of CHMP4 or CHMP2, or else during recruitment of VPS4. The main challenging questions here in ALIX-involved retrovirus budding are 1) How is ALIX activated and recycled back to the cytosol; 2) What is the main function of ALIX coming in the end of virus budding; and 3) What is the details of late-ALIX mechanism with other HIV-1 budding related proteins, including NCGag, CHMP4, and Ub, Dub and the PM at the budding site neck? 4.3 Preliminary Data and Discussion 4.3.1 ALIX transient recruitment and the interaction with CHMP4 In retrovirus budding and cytokinesis, CHMP4 binding sites’ mutations on ALIX inhibited neck membrane abscission in both events 55, 96, 173 . ALIX is thought of as an adaptor to recruit CHMP4 to function at the abscission neck. In our proposed model, the observed ALIX recruitment at the end of Gag assembly only comes to the neck of the formed VLP, and is transiently recruited with the following removal by the downstream component of CHMP4-CHMP2-VPS4. Therefore, we tested whether CHMP4 binding site mutations produce the VLPs that contain more ALIX retention. We mutated F199D and I212D on ALIXBro1 which can totally block the interaction between ALIX and CHMP4 isoforms. We coexpressed GagWT and exogenous HA-ALIX with F199D and I212D mutants (HA-ALIXΔCHMP4) in 293 cells. Surprisingly, we harvested 80 the same amount of VLPs, but having only ~½ fold amount of HA-ALIXΔCHMP4 retained inside the VLPs compared to the amount of HA-ALIXWT (see in Figure 14 lane 1 and lane 2) contrasts with our assumption that CHMP4 is recycled back to the cytosol and might carry away ALIX away from the neck of the budding site together. This result can be reasoned as either less HA-ALIXΔCHMP4 recruiting to the neck at the beginning or ALIXΔCHMP4 not playing an upstream recruiting role to CHMP4. In a previous in vitro EM study, active dimeric ALIX bound to C-truncated CHMP4 to form stabilized polymers. In the MVB study, CHMP4 proteins can apparently recruit ALIX to the endosome where it begins function. Our near future project would be a live image experiment to characterize the recruitment order of ALIX and CHMP4, as well as CHMP2 and VPS4. It might also help us to understand the recycling mechanism of ALIX during the HIV-1 budding. 4.3.2 ALIX transient recruitment and its activated conformation The activation conformation is required for ALIX to function in HIV-1 budding 136. In order to exclude the possibility that ALIXΔCHMP4 low retention in the VLP was caused by the inactivation of ALIX, we coexpressed GagWT and exogenous HA-ALIXR649E (the point mutation to induce open V ALIX structure), which resulted in about 2 times of the amount of HA-ALIXR649E retained in the VLPs compared to the amount of HA-ALIXWT (see lane 1 and lane 3 in Figure 14), which is inconsistent with the result that the overexpression of ALIXR649E rescued the HIV-1 PTAP- particle release more effectively 137 . However, we had the same amount of ALIX retention between ALIXR649E and ALIXΔCHMP4+R649E (compare lane 3 and 4 in Figure 14), which might suggest that ALIX 81 activation plays a heavier role than the CHMP4 binding site in ALIX recruitment or in the ALIX residue left inside the VLPs. With activation or dimerization of ALIX and overexpression of ALIX contributing to HIV-1 budding 55, 136-137, they might all depend on increasing the low affinity of YPXnLALIX pathway to enhance the productivity of VLPs. Moreover, the ALIX self-dimerization is required for ALIX-dependent HIV-1 budding, but is dispensable in EIAV budding and cytokinesis where ALIX has higher affinity 152. An interesting future project will be to test the overexpression of ALIX or ALIXR649E in HIV-1 Gag PTAP- rescue panel. We might be able to have the same ALIX recruitment behavior in HIV-1 and in EIAV, where ALIX has earlier recruitment and higher retention in a live TIRF image, and ALIX might also be located closer to the Gag assemblage center in a super-resolution experiment. 4.3.3 ALIX transient recruitment and the interaction with NCGag The important interaction between ALIXBro1 and NCGag two CCHC-type zinc fingers motifs displayed that ALIX being included into HIV-1VLPs greatly depends on NCGag instead of YPXnL 44. This is consistent with our previous study, where ALIX residue percentage inside VLPs showed that ALIX might have two recruitment paths, recruitment, more residue, and more centeral position, the other with NCGag which has ALIX being recruited late and located at the neck. With different affinity or activation needed, ALIX shows the different recruitment timing, residue amount, and ALIX position during EIAV or The HIV-1 budding. Two recruitment manners can be reduced to one by increasing the affinity in HIV-1 or decreasing the affnity in EIAV and abolishing the interaction of NC to ALIX. 82 4.4 Conclusion ALIX is an early ESCRT-associated protein. ALIX binds both HIV-1 Gag and CHMP4, which forms constrictive polymers on the neck of the budding virions and along with VPS4 is responsible for the final steps of virion release, but mutations of ALIX binding sites on Gag are shown to have only a minor effect on infectivity of released HIV1 virions from cells. Based on the interactions of ALIX with the p6 domain of Gag, it was expected that ALIX would be recruited into the forming virion from the beginning of Gag lattice assembly. We have shown that contrary to this expectation, ALIX arrives late during the formation of HIV-1 virions and therefore, we hypothesized that the interactions responsible for recruiting ALIX into the forming virions are regulated during the assembly of the virions. Here we further show that mutations within the V domain of ALIX, which are suggested to stabilize the open active form of ALIX, result in an increase of ALIX recruitment into the forming virions while, surprisingly, mutations within ALIX that abrogate its binding to CHMP4 result in a major decrease in recruitment of ALIX into the forming VLPs. Based on these observations, we propose that activation of ALIX and its dimerization stabilize the nascent binding of CHMP4 molecules to the neck of the budding virions and bridge the gap between CHMPs and Gag during the assembly of CHMP4 on the neck of the virion. This new model is compatible with the reported role of ALIX in stabilizing ESCRT-III elements during MVB formation. More studies are still in progress to resolve the late-recruitment ALIX mechanism. 83 Figure 14. The western blot of the ALIX recruitment into VLPs. At lane 2, we show the GagWT expression in 293 Cell with exogenous expression of WT HA-ALIX, ALIX F199D+I212D, ALIX R649, and ALIX R199D+I212D+R649. APPENDIX A TESTING THE FUNCTIONAL VPS4 FUSIONS A.1 Cell culture 293T and HeLa-TZM cells were maintained in DMEM (Invitrogen) with 10% FCS. They were obtained from Drs. J. C. Kappes and X. Wu, AIDS Research and Reference Reagent Program, Division of AIDS, NIAID, NIH. A.2 siRNAs, expression vectors and antibodies The following siRNAs were used to silence VSP4A and B in the virus-producer cells: VPS4A-200: CCGAGAAGCUGAAGGAUUAtt and VPS4B-425: CCAAAGAAGCACUGAAAGAtt. The corresponding proteins were detected using our antibodies rabbit anti-VPS4A UT289 (1/500) and antiVPS4B UT292 (1/500). The HIV-1 Gag proteins were detected by western blotting using our own rabbit anti-HIV-1 CA and MA antisera (both mixed at 1:1,000). The antibodies and their working conditions were described in Morita et al. (2010). The secondary antibody was an anti-rabbit IgG (donkey) polyclonal conjugated to IRdye800 (1:10,000, Rockland) and an Odyssey scanner (LiCorBiosciences) was used to visualize the western blot signals. The R9 plasmid WISP9696 (Swingler et al., 1997) was used to express HIV-1. Ten DNA constructs were obtained from Protein Fusion Technologies, LLC., that encoded the joining VPS4A to GFP with 85 protein linkages of the following character: four helical linkers of similar composition but varying lengths (13, 21, 29, and 37 amino acids) and one flexible linker of 32 amino acids in length. All five linkers were used to generate VPS4A fusions with GFP in both gene orders, e.g., GFP before VPS4A and VPS4A before GFP, for ten total linkage combinations. The flexible linker avoided helix-forming amino acids. A3. Rescue of expression of VPS4A GFP fusions and virus production This experiment has been described in detail in Morita et al. (2011). Briefly, the experiments were performed following this time course: t=0, 293T cells were seeded at 2 x 105 cells/well in 6-well plates; t=24 hours, transfection of 293T cells with 10 nMsiRNA and 7.5 l lipofectamine RNAi max (Invitrogen); t=48 hours, change media and cotransfection of 293T cells with 10 nMsiRNA, siRNA-resistant construct expressing VPS4A or mutants (0.7 g), and HIV-1 viral expression vector (0.5 g of HIV-1 R9 vector) using 10 l lipofectamine 2000 (Invitrogen); and t=96 hours, harvest cells and culture supernatant for analysis. Infectious titers were determined using HeLa-TZM indicator cells and counting the β- galactosidase-positive foci formed. We observed lower expressions as the linker became longer on the C terminus of VPS4. The plasmid VPS4A-h37eGFP (V37G) was chosen for further experiments based on its performance on the rescue as shown in Figure 15. 86 Figure 15 Viral infectivity (panel 1) and western blots showing virus release (panel 2, “Virus”, anti-CA, and anti-MA), cellular Gag protein levels (panel 3, “Cell”, anti-CA, and anti-MA), endogenous VPS4A and B (panel 3, anti-VPS4A, and anti- VPS4B) and cellular GAPDH levels (panel 3, anti-GAPDH) are presented. Cells expressing HIV-1 and treated with control siRNAs (panel 1, lane 1) and with siRNAs that target VPS4A and B (lanes 2 to 13) were transfected to express an empty vector (lanes 1 and 2) or a siRNA-resistant VPS4A or VPS4A fusion proteins (lanes 4 to 13). APPENDIX B DUAL PENETRATION DEPTH TIRF, TIRF 360, AND THE TIRF CLAMP The beam path of the laser is shown in Figure16. There are two main components to control the light path, TILL Yanus (digit scan head) and Polytrope (Image mode switch). Yanus contains two galvanometric and one spherical mirror to control the laser beam position. The focused beam can be moved within the back focal plane of the objective by Yanus by turning its Galvo mirror. This feature is used to adjust the penetration depth by placing the focused beam closer or further from the central optical axis and also move the focus point circularly on the focal plane in order to collect more homogeneous image (A turn/exposure; called TIRF360). The Polytrope is composed of one Galvo Mirror and one motorized lens along with multiple fixed mirrors. The light path can be chosen by adjusting the Galvo mirror, in the TIRF path shown in Figure 16, the laser beam in focused on the back focal plane of the objective. Perfect focus is achieved by moving the motorized lens. The penetration depth of the TIRF beam is measured directly by measuring the angle of reflection of the beam. The reflected beam is collected from the back aperture of the objective onto a quadrant photo diode. The position of the beam on the diode determines the angle of reflection and the penetration depth. This position is also sensitive to the drift between the coverglass and the objective lens. A fast feedback loop is used to keep the 88 reflected beam on the same position during long experiments resulting in keeping the glass interface in focus always. The Yanus and Polytrope can position the laser beam anywhere in the objective’s back foal plane within 0.2ms. 89 Figure 16. TILL photonics Yanus and Polytrope Beam Path. The blue line indicates the path of the laser beam. The quadrant photo detector is shown besides the Polytrope; however, in the instrument it resides within the Polytrope. This simplification was made to make the path easier for visualization. APPENDIX C SHELL FILLING MODEL FOR VLP ASSEMBLY WITH 65% FILLING It has previously been shown that Gag assembles into an incomplete shell such that only a fraction of the VLP surface is covered by Gag (1). This fraction is estimated to be two-thirds of a sphere (2). One possible assumption is that the initial 65% of the VLP is fully coated with Gag. This assumption will present a different setting for the shell filling model in which Gag only assembles up to that. In this model identical to the 100% filling model described in the main text, we can approximate the surface area of a single Gag protein (SAG) by dividing the partial surface area of a sphere by the total number of Gag proteins in a single VLP (Gtot ) or: Under the assumption that Gag binds only at the cross section of the cell membrane, the total number of Gag proteins (N) at some angle is shown as equation (1) (compare to equation (7) in the main text): For this model, I (t) from equation (10) in the main text is modified to equation (2). Parameter estimation of the 65% model was conducted as with the 100% filling model described in the text. The following equation was used instead of equation (11) in the main text for the fits. Lastly, we consider a case which adds two additional unknown parameters to (4) to describe two pauses in the data. The third model can be defined piecewise as the equations set in the next page (in comparison to equation (12) in the main text). The results of the analysis with the 65% shell filling model are presented in Figures 91 17 and 18. In general, this analysis did not significantly change the results presented in Figures 4 and 5 of the Chapter 2. 92 93 Figure 17. Kinetics of VLP formation analyzed by the 65% shell filling model. (A) Definition of various timings on the graph of Gag polymerization (blue dot) and VPS4 (red) simulated data and Gag polymerization shell filling model fitting (blue solid). We define τ as the VLP formation lifetime, τ1 as the nucleation time defined by the period of time that Gag VLP had started to assemble but intensity still below the detection, τ2 as the Gag polymerization time, and τ2 as the recruitment time of VPS4 after Gag polymerization was complete. (B-E) Histograms of different time periods, and from all 58 VLP formation profiles. Average values are τ=33.08±15.88 minutes, τ1=-7.04±6.23 minutes, τ2=19.88±10.40 minutes and τ3= 6.33±10.50 minutes. 94 Figure18. Kinetics of pauses with the 65% shell filling model (A-C). Three different Gag polymerization behaviors, including no-pause, single pause, and double pause of Gag polymerization. The blue dot shows the Gag polymerization experimental data, the solid blue line is shell filling model fitting, and the dark pink line highlights the pauses. (D) Frequency of three different behaviors. Probability of having a single pause is 65%, double pause is 5%. (E) Histogram of pause length in minutes. The average pause length is 4.09±3.69 minutes. (F) Histogram of Gag filling fraction in pause (%). Note that we cannot efficiently detect pauses below 20% filling fraction. 95 Figure 19. Kinetics of VLP formation and pauses during polymerization of Gag late domain double mutant. For the mutant data, average values are τ1=-11.36±9.13 minutes, τ2=25.75±17.29 minutes. The fraction of profiles with a pause is 71% as shown in D. The average pause length is 6.04±3.47 minutes as shown in E. 96 Figure 20. Average center of mass position of the VLP Zf in nanometers versus time of assmely. The average Zf was calculated on all VLP profiles in this study. Since the lifetime of VLP formations in individual profiles varies, we first calculated the average Zf profile of the each VLP along 20 equally spaced time intervals covering the whole lifetime of the VLP. The 20 resulting Zf values from each VLP were then used to calculate an average of Zf over the entire population of VLP formation profiles and is shown in this figure. APPENDIX D RECRUITMENT OF VPS4-h37-mCHERRY ALONG WITH ALIX-h30-eGFP See Figure 21. Recruitment of VPS4-h37-mCherry along with ALIX-h30-eGFP in Hela-C1 cells. Cells were transfected with, Gag, and VPS4-h37-mCherry plasmids and imaged 5 hours posttransfection. The time resolution of these experiments is 15 seconds and therefore insufficient for separating the recruitment of ALIX from VPS4. 98 Figure 21. Alix and VPS4 recruitment at the HIV-1 budding sites. (A) Shows the recruitment dynamics of one spot. (B-C) shows the normalized and average recruitment profiles for 22 spots; all of them had both ALIX and VPS4 corecruitment. REFERENCES 1. Wyatt, R.; Sodroski, J., The HIV-1 envelope glycoproteins: fusogens, antigens, and immunogens. Science 1998, 280 (5371), 1884-8. 2. Chan, D. C.; Kim, P. S., HIV entry and its inhibition. Cell 1998, 93 (5), 681-4. 3. Levy, J. A., HIV pathogenesis and long-term survival. AIDS 1993, 7 (11), 1401-10. 4. Sundquist, W. I.; Krausslich, H. G., HIV-1 assembly, budding, and maturation. Cold Spring Harb Perspect Med 2012, 2 (7), a006924. 5. Chertova, E.; Bess, J. W., Jr.; Crise, B. J.; Sowder, I. R.; Schaden, T. M.; Hilburn, J. M.; Hoxie, J. A.; Benveniste, R. E.; Lifson, J. D.; Henderson, L. E.; Arthur, L. O., Envelope glycoprotein incorporation, not shedding of surface envelope glycoprotein (gp120/SU), Is the primary determinant of SU content of purified human immunodeficiency virus type 1 and simian immunodeficiency virus. J Virol 2002, 76 (11), 5315-25. 6. Zhu, P.; Chertova, E.; Bess, J., Jr.; Lifson, J. D.; Arthur, L. O.; Liu, J.; Taylor, K. A.; Roux, K. H., Electron tomography analysis of envelope glycoprotein trimers on HIV and simian immunodeficiency virus virions. Proc Natl Acad Sci U S A 2003, 100 (26), 15812-7. 7. Votteler, J.; Sundquist, W. I., Virus budding and the ESCRT pathway. Cell Host Microbe 2013, 14 (3), 232-41. 8. 256. Gottlinger, H. G., How HIV-1 hijacks ALIX. Nat Struct Mol Biol 2007, 14 (4), 254- 9. Briggs, J. A.; Krausslich, H. G., The molecular architecture of HIV. J Mol Biol 2011, 410 (4), 491-500. 10. Carlson, L. A.; Briggs, J. A.; Glass, B.; Riches, J. D.; Simon, M. N.; Johnson, M. C.; Muller, B.; Grunewald, K.; Krausslich, H. G., Three-dimensional analysis of budding sites and released virus suggests a revised model for HIV-1 morphogenesis. Cell Host Microbe 2008, 4 (6), 592-9. 11. Garrus, J. E.; von Schwedler, U. K.; Pornillos, O. W.; Morham, S. G.; Zavitz, K. H.; Wang, H. E.; Wettstein, D. A.; Stray, K. M.; Cote, M.; Rich, R. L.; Myszka, D. G.; 100 Sundquist, W. I., Tsg101 and the vacuolar protein sorting pathway are essential for HIV-1 budding. Cell 2001, 107 (1), 55-65. 12. Martin-Serrano, J.; Zang, T.; Bieniasz, P. D., HIV-1 and Ebola virus encode small peptide motifs that recruit Tsg101 to sites of particle assembly to facilitate egress. Nat Med 2001, 7 (12), 1313-9. 13. Bieniasz, P. D., The cell biology of HIV-1 virion genesis. Cell Host Microbe 2009, 5 (6), 550-8. 14. Gheysen, D.; Jacobs, E.; de Foresta, F.; Thiriart, C.; Francotte, M.; Thines, D.; De Wilde, M., Assembly and release of HIV-1 precursor Pr55gag virus-like particles from recombinant baculovirus-infected insect cells. Cell 1989, 59 (1), 103-12. 15. Ivanchenko, S.; Godinez, W. J.; Lampe, M.; Krausslich, H. G.; Eils, R.; Rohr, K.; Brauchle, C.; Muller, B.; Lamb, D. C., Dynamics of HIV-1 assembly and release. PLoS Pathog 2009, 5 (11), e1000652. 16. Jouvenet, N.; Simon, S. M.; Bieniasz, P. D., Visualizing HIV-1 assembly. J Mol Biol 2011, 410 (4), 501-11. 17. Kutluay, S. B.; Bieniasz, P. D., Analysis of the initiating events in HIV-1 particle assembly and genome packaging. PLoS Pathog 2010, 6 (11), e1001200. 18. Chukkapalli, V.; Oh, S. J.; Ono, A., Opposing mechanisms involving RNA and lipids regulate HIV-1 Gag membrane binding through the highly basic region of the matrix domain. Proc Natl Acad Sci U S A 2010, 107 (4), 1600-5. 19. Datta, S. A.; Heinrich, F.; Raghunandan, S.; Krueger, S.; Curtis, J. E.; Rein, A.; Nanda, H., HIV-1 Gag extension: conformational changes require simultaneous interaction with membrane and nucleic acid. J Mol Biol 2011, 406 (2), 205-14. 20. Jones, C. P.; Datta, S. A.; Rein, A.; Rouzina, I.; Musier-Forsyth, K., Matrix domain modulates HIV-1 Gag's nucleic acid chaperone activity via inositol phosphate binding. J Virol 2011, 85 (4), 1594-603. 21. 9. Ono, A., HIV-1 assembly at the plasma membrane. Vaccine 2010, 28 Suppl 2, B55- 22. Facke, M.; Janetzko, A.; Shoeman, R. L.; Krausslich, H. G., A large deletion in the matrix domain of the human immunodeficiency virus gag gene redirects virus particle assembly from the plasma membrane to the endoplasmic reticulum. J Virol 1993, 67 (8), 4972-80. 101 23. Reil, H.; Bukovsky, A. A.; Gelderblom, H. R.; Gottlinger, H. G., Efficient HIV-1 replication can occur in the absence of the viral matrix protein. The EMBO journal 1998, 17 (9), 2699-708. 24. Ono, A.; Ablan, S. D.; Lockett, S. J.; Nagashima, K.; Freed, E. O., Phosphatidylinositol (4,5) bisphosphate regulates HIV-1 Gag targeting to the plasma membrane. Proc Natl Acad Sci U S A 2004, 101 (41), 14889-94. 25. Saad, J. S.; Miller, J.; Tai, J.; Kim, A.; Ghanam, R. H.; Summers, M. F., Structural basis for targeting HIV-1 Gag proteins to the plasma membrane for virus assembly. Proc Natl Acad Sci U S A 2006, 103 (30), 11364-9. 26. Ehrlich, L. S.; Fong, S.; Scarlata, S.; Zybarth, G.; Carter, C., Partitioning of HIV-1 Gag and Gag-related proteins to membranes. Biochemistry 1996, 35 (13), 3933-43. 27. Alfadhli, A.; Barklis, R. L.; Barklis, E., HIV-1 matrix organizes as a hexamer of trimers on membranes containing phosphatidylinositol-(4,5)-bisphosphate. Virology 2009, 387 (2), 466-72. 28. Chukkapalli, V.; Ono, A., Molecular determinants that regulate plasma membrane association of HIV-1 Gag. J Mol Biol 2011, 410 (4), 512-24. 29. Krausslich, H. G.; Facke, M.; Heuser, A. M.; Konvalinka, J.; Zentgraf, H., The spacer peptide between human immunodeficiency virus capsid and nucleocapsid proteins is essential for ordered assembly and viral infectivity. J Virol 1995, 69 (6), 3407-19. 30. Borsetti, A.; Ohagen, A.; Gottlinger, H. G., The C-terminal half of the human immunodeficiency virus type 1 Gag precursor is sufficient for efficient particle assembly. J Virol 1998, 72 (11), 9313-7. 31. Accola, M. A.; Strack, B.; Gottlinger, H. G., Efficient particle production by minimal Gag constructs which retain the carboxy-terminal domain of human immunodeficiency virus type 1 capsid-p2 and a late assembly domain. J Virol 2000, 74 (12), 5395-402. 32. Dorfman, T.; Bukovsky, A.; Ohagen, A.; Hoglund, S.; Gottlinger, H. G., Functional domains of the capsid protein of human immunodeficiency virus type 1. J Virol 1994, 68 (12), 8180-7. 33. Liang, C.; Hu, J.; Russell, R. S.; Roldan, A.; Kleiman, L.; Wainberg, M. A., Characterization of a putative alpha-helix across the capsid-SP1 boundary that is critical for the multimerization of human immunodeficiency virus type 1 gag. J Virol 2002, 76 (22), 11729-37. 102 34. Briggs, J. A.; Riches, J. D.; Glass, B.; Bartonova, V.; Zanetti, G.; Krausslich, H. G., Structure and assembly of immature HIV. Proc Natl Acad Sci U S A 2009, 106 (27), 11090-5. 35. Wright, E. R.; Schooler, J. B.; Ding, H. J.; Kieffer, C.; Fillmore, C.; Sundquist, W. I.; Jensen, G. J., Electron cryotomography of immature HIV-1 virions reveals the structure of the CA and SP1 Gag shells. The EMBO journal 2007, 26 (8), 2218-26. 36. Li, S.; Hill, C. P.; Sundquist, W. I.; Finch, J. T., Image reconstructions of helical assemblies of the HIV-1 CA protein. Nature 2000, 407 (6802), 409-13. 37. Briggs, J. A.; Wilk, T.; Welker, R.; Krausslich, H. G.; Fuller, S. D., Structural organization of authentic, mature HIV-1 virions and cores. The EMBO journal 2003, 22 (7), 1707-15. 38. Ganser-Pornillos, B. K.; von Schwedler, U. K.; Stray, K. M.; Aiken, C.; Sundquist, W. I., Assembly properties of the human immunodeficiency virus type 1 CA protein. J Virol 2004, 78 (5), 2545-52. 39. Lu, K.; Heng, X.; Summers, M. F., Structural determinants and mechanism of HIV1 genome packaging. J Mol Biol 2011, 410 (4), 609-33. 40. Kuzembayeva, M.; Dilley, K.; Sardo, L.; Hu, W. S., Life of psi: how full-length HIV-1 RNAs become packaged genomes in the viral particles. Virology 2014, 454-455, 362-70. 41. Burniston, M. T.; Cimarelli, A.; Colgan, J.; Curtis, S. P.; Luban, J., Human immunodeficiency virus type 1 Gag polyprotein multimerization requires the nucleocapsid domain and RNA and is promoted by the capsid-dimer interface and the basic region of matrix protein. J Virol 1999, 73 (10), 8527-40. 42. Muriaux, D.; Mirro, J.; Harvin, D.; Rein, A., RNA is a structural element in retrovirus particles. Proc Natl Acad Sci U S A 2001, 98 (9), 5246-51. 43. Cimarelli, A.; Sandin, S.; Hoglund, S.; Luban, J., Basic residues in human immunodeficiency virus type 1 nucleocapsid promote virion assembly via interaction with RNA. J Virol 2000, 74 (7), 3046-57. 44. Popov, S.; Popova, E.; Inoue, M.; Gottlinger, H. G., Human immunodeficiency virus type 1 Gag engages the Bro1 domain of ALIX/AIP1 through the nucleocapsid. J Virol 2008, 82 (3), 1389-98. 45. Dussupt, V.; Javid, M. P.; Abou-Jaoude, G.; Jadwin, J. A.; de La Cruz, J.; Nagashima, K.; Bouamr, F., The nucleocapsid region of HIV-1 Gag cooperates with the PTAP and LYPXnL late domains to recruit the cellular machinery necessary for viral budding. PLoS Pathog 2009, 5 (3), e1000339. 103 46. Chung, H. Y.; Morita, E.; von Schwedler, U.; Muller, B.; Krausslich, H. G.; Sundquist, W. I., NEDD4L overexpression rescues the release and infectivity of human immunodeficiency virus type 1 constructs lacking PTAP and YPXL late domains. J Virol 2008, 82 (10), 4884-97. 47. Usami, Y.; Popov, S.; Popova, E.; Gottlinger, H. G., Efficient and specific rescue of human immunodeficiency virus type 1 budding defects by a Nedd4-like ubiquitin ligase. J Virol 2008, 82 (10), 4898-907. 48. Weiss, E. R.; Popova, E.; Yamanaka, H.; Kim, H. C.; Huibregtse, J. M.; Gottlinger, H., Rescue of HIV-1 release by targeting widely divergent NEDD4-type ubiquitin ligases and isolated catalytic HECT domains to Gag. PLoS Pathog 2010, 6 (9), e1001107. 49. Huang, M.; Orenstein, J. M.; Martin, M. A.; Freed, E. O., p6Gag is required for particle production from full-length human immunodeficiency virus type 1 molecular clones expressing protease. J Virol 1995, 69 (11), 6810-8. 50. Puffer, B. A.; Parent, L. J.; Wills, J. W.; Montelaro, R. C., Equine infectious anemia virus utilizes a YXXL motif within the late assembly domain of the Gag p9 protein. J Virol 1997, 71 (9), 6541-6. 51. Strack, B.; Calistri, A.; Craig, S.; Popova, E.; Gottlinger, H. G., AIP1/ALIX is a binding partner for HIV-1 p6 and EIAV p9 functioning in virus budding. Cell 2003, 114 (6), 689-99. 52. Morita, E.; Sundquist, W. I., Retrovirus budding. Annu Rev Cell Dev Biol 2004, 20, 395-425. 53. VerPlank, L.; Bouamr, F.; LaGrassa, T. J.; Agresta, B.; Kikonyogo, A.; Leis, J.; Carter, C. A., Tsg101, a homologue of ubiquitin-conjugating (E2) enzymes, binds the L domain in HIV type 1 Pr55(Gag). Proc Natl Acad Sci U S A 2001, 98 (14), 7724-9. 54. Briggs, J. A. G.; Riches, J. D.; Glass, B.; Bartonova, V.; Zanetti, G.; Kräusslich, H.-G., Structure and assembly of immature HIV. Proceedings of the National Academy of Sciences 2009, 106 (27), 11090-11095. 55. Fisher, R. D.; Chung, H. Y.; Zhai, Q.; Robinson, H.; Sundquist, W. I.; Hill, C. P., Structural and biochemical studies of ALIX/AIP1 and its role in retrovirus budding. Cell 2007, 128 (5), 841-52. 56. Fujii, K.; Munshi, U. M.; Ablan, S. D.; Demirov, D. G.; Soheilian, F.; Nagashima, K.; Stephen, A. G.; Fisher, R. J.; Freed, E. O., Functional role of Alix in HIV-1 replication. Virology 2009, 391 (2), 284-92. 104 57. Martin-Serrano, J.; Yarovoy, A.; Perez-Caballero, D.; Bieniasz, P. D., Divergent retroviral late-budding domains recruit vacuolar protein sorting factors by using alternative adaptor proteins. Proc Natl Acad Sci U S A 2003, 100 (21), 12414-9. 58. von Schwedler, U. K.; Stuchell, M.; Muller, B.; Ward, D. M.; Chung, H. Y.; Morita, E.; Wang, H. E.; Davis, T.; He, G. P.; Cimbora, D. M.; Scott, A.; Krausslich, H. G.; Kaplan, J.; Morham, S. G.; Sundquist, W. I., The protein network of HIV budding. Cell 2003, 114 (6), 701-13. 59. Hurley, J. H.; Hanson, P. I., Membrane budding and scission by the ESCRT machinery: it's all in the neck. Nat Rev Mol Cell Biol 2010, 11 (8), 556-66. 60. Peel, S.; Macheboeuf, P.; Martinelli, N.; Weissenhorn, W., Divergent pathways lead to ESCRT-III-catalyzed membrane fission. Trends Biochem Sci 2011, 36 (4), 199210. 61. Carlton, J. G.; Martin-Serrano, J., Parallels between cytokinesis and retroviral budding: a role for the ESCRT machinery. Science 2007, 316 (5833), 1908-12. 62. Carlton, J. G.; Martin-Serrano, J., The ESCRT machinery: new functions in viral and cellular biology. Biochem Soc Trans 2009, 37 (Pt 1), 195-9. 63. Elia, N.; Sougrat, R.; Spurlin, T. A.; Hurley, J. H.; Lippincott-Schwartz, J., Dynamics of endosomal sorting complex required for transport (ESCRT) machinery during cytokinesis and its role in abscission. Proc Natl Acad Sci U S A 2011, 108 (12), 4846-51. 64. Guizetti, J.; Schermelleh, L.; Mantler, J.; Maar, S.; Poser, I.; Leonhardt, H.; MullerReichert, T.; Gerlich, D. W., Cortical constriction during abscission involves helices of ESCRT-III-dependent filaments. Science 2011, 331 (6024), 1616-20. 65. Raymond, C. K.; Howald-Stevenson, I.; Vater, C. A.; Stevens, T. H., Morphological classification of the yeast vacuolar protein sorting mutants: evidence for a prevacuolar compartment in class E vps mutants. Molecular biology of the cell 1992, 3 (12), 1389-402. 66. Katzmann, D. J.; Babst, M.; Emr, S. D., Ubiquitin-dependent sorting into the multivesicular body pathway requires the function of a conserved endosomal protein sorting complex, ESCRT-I. Cell 2001, 106 (2), 145-55. 67. Piper, R. C.; Katzmann, D. J., Biogenesis and function of multivesicular bodies. Annu Rev Cell Dev Biol 2007, 23, 519-47. 68. Hurley, J. H., The ESCRT complexes. Crit Rev Biochem Mol Biol 2010, 45 (6), 463-87. 105 69. Saksena, S.; Wahlman, J.; Teis, D.; Johnson, A. E.; Emr, S. D., Functional reconstitution of ESCRT-III assembly and disassembly. Cell 2009, 136 (1), 97-109. 70. Teis, D.; Saksena, S.; Emr, S. D., Ordered assembly of the ESCRT-III complex on endosomes is required to sequester cargo during MVB formation. Dev Cell 2008, 15 (4), 578-89. 71. Wollert, T.; Hurley, J. H., Molecular mechanism of multivesicular body biogenesis by ESCRT complexes. Nature 2010, 464 (7290), 864-9. 72. Jouvenet, N.; Simon, S. M.; Bieniasz, P. D., Imaging the interaction of HIV-1 genomes and Gag during assembly of individual viral particles. Proc Natl Acad Sci U S A 2009, 106 (45), 19114-9. 73. Mao, Y.; Nickitenko, A.; Duan, X.; Lloyd, T. E.; Wu, M. N.; Bellen, H.; Quiocho, F. A., Crystal structure of the VHS and FYVE tandem domains of Hrs, a protein involved in membrane trafficking and signal transduction. Cell 2000, 100 (4), 447-56. 74. Mizuno, E.; Kawahata, K.; Kato, M.; Kitamura, N.; Komada, M., STAM proteins bind ubiquitinated proteins on the early endosome via the VHS domain and ubiquitininteracting motif. Molecular biology of the cell 2003, 14 (9), 3675-89. 75. Hirst, J.; Lui, W. W.; Bright, N. A.; Totty, N.; Seaman, M. N.; Robinson, M. S., A family of proteins with gamma-adaptin and VHS domains that facilitate trafficking between the trans-Golgi network and the vacuole/lysosome. The Journal of cell biology 2000, 149 (1), 67-80. 76. Misra, S.; Hurley, J. H., Crystal structure of a phosphatidylinositol 3-phosphatespecific membrane-targeting motif, the FYVE domain of Vps27p. Cell 1999, 97 (5), 65766. 77. Raiborg, C.; Bremnes, B.; Mehlum, A.; Gillooly, D. J.; D'Arrigo, A.; Stang, E.; Stenmark, H., FYVE and coiled-coil domains determine the specific localisation of Hrs to early endosomes. Journal of cell science 2001, 114 (Pt 12), 2255-63. 78. Bilodeau, P. S.; Urbanowski, J. L.; Winistorfer, S. C.; Piper, R. C., The Vps27p Hse1p complex binds ubiquitin and mediates endosomal protein sorting. Nature cell biology 2002, 4 (7), 534-9. 79. Bache, K. G.; Brech, A.; Mehlum, A.; Stenmark, H., Hrs regulates multivesicular body formation via ESCRT recruitment to endosomes. The Journal of cell biology 2003, 162 (3), 435-42. 80. Lu, Q.; Hope, L. W.; Brasch, M.; Reinhard, C.; Cohen, S. N., TSG101 interaction with HRS mediates endosomal trafficking and receptor down-regulation. Proc Natl Acad Sci U S A 2003, 100 (13), 7626-31. 106 81. Asao, H.; Sasaki, Y.; Arita, T.; Tanaka, N.; Endo, K.; Kasai, H.; Takeshita, T.; Endo, Y.; Fujita, T.; Sugamura, K., Hrs is associated with STAM, a signal-transducing adaptor molecule. Its suppressive effect on cytokine-induced cell growth. J Biol Chem 1997, 272 (52), 32785-91. 82. Raiborg, C.; Bache, K. G.; Mehlum, A.; Stang, E.; Stenmark, H., Hrs recruits clathrin to early endosomes. The EMBO journal 2001, 20 (17), 5008-21. 83. Im, Y. J.; Kuo, L.; Ren, X.; Burgos, P. V.; Zhao, X. Z.; Liu, F.; Burke, T. R., Jr.; Bonifacino, J. S.; Freed, E. O.; Hurley, J. H., Crystallographic and functional analysis of the ESCRT-I /HIV-1 Gag PTAP interaction. Structure 2010, 18 (11), 1536-47. 84. Pornillos, O.; Higginson, D. S.; Stray, K. M.; Fisher, R. D.; Garrus, J. E.; Payne, M.; He, G. P.; Wang, H. E.; Morham, S. G.; Sundquist, W. I., HIV Gag mimics the Tsg101recruiting activity of the human Hrs protein. The Journal of cell biology 2003, 162 (3), 425-34. 85. Babst, M.; Odorizzi, G.; Estepa, E. J.; Emr, S. D., Mammalian tumor susceptibility gene 101 (TSG101) and the yeast homologue, Vps23p, both function in late endosomal trafficking. Traffic 2000, 1 (3), 248-58. 86. Morita, E.; Sandrin, V.; Alam, S. L.; Eckert, D. M.; Gygi, S. P.; Sundquist, W. I., Identification of human MVB12 proteins as ESCRT-I subunits that function in HIV budding. Cell Host Microbe 2007, 2 (1), 41-53. 87. Kostelansky, M. S.; Schluter, C.; Tam, Y. Y.; Lee, S.; Ghirlando, R.; Beach, B.; Conibear, E.; Hurley, J. H., Molecular architecture and functional model of the complete yeast ESCRT-I heterotetramer. Cell 2007, 129 (3), 485-98. 88. Weiss, E. R.; Gottlinger, H., The role of cellular factors in promoting HIV budding. J Mol Biol 2011, 410 (4), 525-33. 89. Sundquist, W. I.; Schubert, H. L.; Kelly, B. N.; Hill, G. C.; Holton, J. M.; Hill, C. P., Ubiquitin recognition by the human TSG101 protein. Molecular cell 2004, 13 (6), 7839. 90. Martin-Serrano, J.; Zang, T.; Bieniasz, P. D., Role of ESCRT-I in retroviral budding. J Virol 2003, 77 (8), 4794-804. 91. Kostelansky, M. S.; Sun, J.; Lee, S.; Kim, J.; Ghirlando, R.; Hierro, A.; Emr, S. D.; Hurley, J. H., Structural and functional organization of the ESCRT-I trafficking complex. Cell 2006, 125 (1), 113-26. 92. Stuchell, M. D.; Garrus, J. E.; Muller, B.; Stray, K. M.; Ghaffarian, S.; McKinnon, R.; Krausslich, H. G.; Morham, S. G.; Sundquist, W. I., The human endosomal sorting 107 complex required for transport (ESCRT-I) and its role in HIV-1 budding. J Biol Chem 2004, 279 (34), 36059-71. 93. Babst, M.; Katzmann, D. J.; Snyder, W. B.; Wendland, B.; Emr, S. D., Endosomeassociated complex, ESCRT-II, recruits transport machinery for protein sorting at the multivesicular body. Dev Cell 2002, 3 (2), 283-9. 94. Gill, D. J.; Teo, H.; Sun, J.; Perisic, O.; Veprintsev, D. B.; Emr, S. D.; Williams, R. L., Structural insight into the ESCRT-I/-II link and its role in MVB trafficking. The EMBO journal 2007, 26 (2), 600-12. 95. Langelier, C.; von Schwedler, U. K.; Fisher, R. D.; De Domenico, I.; White, P. L.; Hill, C. P.; Kaplan, J.; Ward, D.; Sundquist, W. I., Human ESCRT-II complex and its role in human immunodeficiency virus type 1 release. J Virol 2006, 80 (19), 9465-80. 96. Morita, E.; Sandrin, V.; Chung, H. Y.; Morham, S. G.; Gygi, S. P.; Rodesch, C. K.; Sundquist, W. I., Human ESCRT and ALIX proteins interact with proteins of the midbody and function in cytokinesis. The EMBO journal 2007, 26 (19), 4215-27. 97. Carlson, L. A.; Hurley, J. H., In vitro reconstitution of the ordered assembly of the endosomal sorting complex required for transport at membrane-bound HIV-1 Gag clusters. Proc Natl Acad Sci U S A 2012, 109 (42), 16928-33. 98. Zhadina, M.; Bieniasz, P. D., Functional interchangeability of late domains, late domain cofactors and ubiquitin in viral budding. PLoS Pathog 2010, 6 (10), e1001153. 99. Meng, B.; Ip, N. C.; Prestwood, L. J.; Abbink, T. E.; Lever, A. M., Evidence that the endosomal sorting complex required for transport-II (ESCRT-II) is required for efficient human immunodeficiency virus-1 (HIV-1) production. Retrovirology 2015, 12, 72. 100. Babst, M.; Katzmann, D. J.; Estepa-Sabal, E. J.; Meerloo, T.; Emr, S. D., Escrt-III: an endosome-associated heterooligomeric protein complex required for mvb sorting. Dev Cell 2002, 3 (2), 271-82. 101. Bajorek, M.; Schubert, H. L.; McCullough, J.; Langelier, C.; Eckert, D. M.; Stubblefield, W. M.; Uter, N. T.; Myszka, D. G.; Hill, C. P.; Sundquist, W. I., Structural basis for ESCRT-III protein autoinhibition. Nat Struct Mol Biol 2009, 16 (7), 754-62. 102. Lata, S.; Schoehn, G.; Solomons, J.; Pires, R.; Gottlinger, H. G.; Weissenhorn, W., Structure and function of ESCRT-III. Biochem Soc Trans 2009, 37 (Pt 1), 156-60. 103. Zamborlini, A.; Usami, Y.; Radoshitzky, S. R.; Popova, E.; Palu, G.; Gottlinger, H., Release of autoinhibition converts ESCRT-III components into potent inhibitors of HIV-1 budding. Proc Natl Acad Sci U S A 2006, 103 (50), 19140-5. 108 104. Lata, S.; Roessle, M.; Solomons, J.; Jamin, M.; Gottlinger, H. G.; Svergun, D. I.; Weissenhorn, W., Structural basis for autoinhibition of ESCRT-III CHMP3. J Mol Biol 2008, 378 (4), 818-27. 105. Muziol, T.; Pineda-Molina, E.; Ravelli, R. B.; Zamborlini, A.; Usami, Y.; Gottlinger, H.; Weissenhorn, W., Structural basis for budding by the ESCRT-III factor CHMP3. Dev Cell 2006, 10 (6), 821-30. 106. Lin, Y.; Kimpler, L. A.; Naismith, T. V.; Lauer, J. M.; Hanson, P. I., Interaction of the mammalian endosomal sorting complex required for transport (ESCRT) III protein hSnf7-1 with itself, membranes, and the AAA+ ATPase SKD1. J Biol Chem 2005, 280 (13), 12799-809. 107. Shim, S.; Merrill, S. A.; Hanson, P. I., Novel interactions of ESCRT-III with LIP5 and VPS4 and their implications for ESCRT-III disassembly. Molecular biology of the cell 2008, 19 (6), 2661-72. 108. Hanson, P. I.; Roth, R.; Lin, Y.; Heuser, J. E., Plasma membrane deformation by circular arrays of ESCRT-III protein filaments. The Journal of cell biology 2008, 180 (2), 389-402. 109. Guizetti, J.; Gerlich, D. W., ESCRT-III polymers in membrane neck constriction. Trends in cell biology 2012, 22 (3), 133-40. 110. Teo, H.; Perisic, O.; Gonzalez, B.; Williams, R. L., ESCRT-II, an endosomeassociated complex required for protein sorting: crystal structure and interactions with ESCRT-III and membranes. Dev Cell 2004, 7 (4), 559-69. 111. Fyfe, I.; Schuh, A. L.; Edwardson, J. M.; Audhya, A., Association of the endosomal sorting complex ESCRT-II with the Vps20 subunit of ESCRT-III generates a curvaturesensitive complex capable of nucleating ESCRT-III filaments. J Biol Chem 2011, 286 (39), 34262-70. 112. Teis, D.; Saksena, S.; Judson, B. L.; Emr, S. D., ESCRT-II coordinates the assembly of ESCRT-III filaments for cargo sorting and multivesicular body vesicle formation. The EMBO journal 2010, 29 (5), 871-83. 113. Bodon, G.; Chassefeyre, R.; Pernet-Gallay, K.; Martinelli, N.; Effantin, G.; Hulsik, D. L.; Belly, A.; Goldberg, Y.; Chatellard-Causse, C.; Blot, B.; Schoehn, G.; Weissenhorn, W.; Sadoul, R., Charged multivesicular body protein 2B (CHMP2B) of the endosomal sorting complex required for transport-III (ESCRT-III) polymerizes into helical structures deforming the plasma membrane. J Biol Chem 2011, 286 (46), 40276-86. 114. Morita, E.; Sandrin, V.; McCullough, J.; Katsuyama, A.; Baci Hamilton, I.; Sundquist, Wesley I., ESCRT-III Protein Requirements for HIV-1 Budding. Cell Host & Microbe 2011, 9 (3), 235-242. 109 115. Jouvenet, N.; Zhadina, M.; Bieniasz, P. D.; Simon, S. M., Dynamics of ESCRT protein recruitment during retroviral assembly. Nature cell biology 2011, 13 (4), 394-401. 116. Morita, E.; Sandrin, V.; McCullough, J.; Katsuyama, A.; Baci Hamilton, I.; Sundquist, W. I., ESCRT-III protein requirements for HIV-1 budding. Cell Host Microbe 2011, 9 (3), 235-42. 117. Effantin, G.; Dordor, A.; Sandrin, V.; Martinelli, N.; Sundquist, W. I.; Schoehn, G.; Weissenhorn, W., ESCRT-III CHMP2A and CHMP3 form variable helical polymers in vitro and act synergistically during HIV-1 budding. Cellular microbiology 2013, 15 (2), 213-26. 118. Cashikar, A. G.; Shim, S.; Roth, R.; Maldazys, M. R.; Heuser, J. E.; Hanson, P. I., Structure of cellular ESCRT-III spirals and their relationship to HIV budding. Elife 2014, 3. 119. Baumgartel, V.; Muller, B.; Lamb, D. C., Quantitative live-cell imaging of human immunodeficiency virus (HIV-1) assembly. Viruses 2012, 4 (5), 777-99. 120. Fabrikant, G.; Lata, S.; Riches, J. D.; Briggs, J. A.; Weissenhorn, W.; Kozlov, M. M., Computational model of membrane fission catalyzed by ESCRT-III. PLoS Comput Biol 2009, 5 (11), e1000575. 121. Stuchell-Brereton, M. D.; Skalicky, J. J.; Kieffer, C.; Karren, M. A.; Ghaffarian, S.; Sundquist, W. I., ESCRT-III recognition by VPS4 ATPases. Nature 2007, 449 (7163), 7404. 122. Scott, A.; Chung, H. Y.; Gonciarz-Swiatek, M.; Hill, G. C.; Whitby, F. G.; Gaspar, J.; Holton, J. M.; Viswanathan, R.; Ghaffarian, S.; Hill, C. P.; Sundquist, W. I., Structural and mechanistic studies of VPS4 proteins. The EMBO journal 2005, 24 (20), 3658-69. 123. Babst, M.; Wendland, B.; Estepa, E. J.; Emr, S. D., The Vps4p AAA ATPase regulates membrane association of a Vps protein complex required for normal endosome function. The EMBO journal 1998, 17 (11), 2982-93. 124. Baumgartel, V.; Ivanchenko, S.; Dupont, A.; Sergeev, M.; Wiseman, P. W.; Krausslich, H. G.; Brauchle, C.; Muller, B.; Lamb, D. C., Live-cell visualization of dynamics of HIV budding site interactions with an ESCRT component. Nature cell biology 2011, 13 (4), 469-74. 125. Bleck, M.; Itano, M. S.; Johnson, D. S.; Thomas, V. K.; North, A. J.; Bieniasz, P. D.; Simon, S. M., Temporal and spatial organization of ESCRT protein recruitment during HIV-1 budding. Proc Natl Acad Sci U S A 2014, 111 (33), 12211-6. 126. Saksena, S.; Wahlman, J.; Teis, D.; Johnson, A. E.; Emr, S. D., Functional Reconstitution of ESCRT-III Assembly and Disassembly. Cell 2008, 136 (1), 97-109. 110 127. Vito, P.; Pellegrini, L.; Guiet, C.; D'Adamio, L., Cloning of AIP1, a novel protein that associates with the apoptosis-linked gene ALG-2 in a Ca2+-dependent reaction. J Biol Chem 1999, 274 (3), 1533-40. 128. Missotten, M.; Nichols, A.; Rieger, K.; Sadoul, R., Alix, a novel mouse protein undergoing calcium-dependent interaction with the apoptosis-linked-gene 2 (ALG-2) protein. Cell Death Differ 1999, 6 (2), 124-9. 129. Saksena, S.; Sun, J.; Chu, T.; Emr, S. D., ESCRTing proteins in the endocytic pathway. Trends Biochem Sci 2007, 32 (12), 561-73. 130. Usami, Y.; Popov, S.; Popova, E.; Inoue, M.; Weissenhorn, W.; H, G. G., The ESCRT pathway and HIV-1 budding. Biochem Soc Trans 2009, 37 (Pt 1), 181-4. 131. Cabezas, A.; Bache, K. G.; Brech, A.; Stenmark, H., Alix regulates cortical actin and the spatial distribution of endosomes. Journal of cell science 2005, 118 (Pt 12), 262535. 132. Pan, S.; Wang, R.; Zhou, X.; He, G.; Koomen, J.; Kobayashi, R.; Sun, L.; Corvera, J.; Gallick, G. E.; Kuang, J., Involvement of the conserved adaptor protein Alix in actin cytoskeleton assembly. J Biol Chem 2006, 281 (45), 34640-50. 133. Kim, J.; Sitaraman, S.; Hierro, A.; Beach, B. M.; Odorizzi, G.; Hurley, J. H., Structural basis for endosomal targeting by the Bro1 domain. Dev Cell 2005, 8 (6), 93747. 134. Fujii, K.; Hurley, J. H.; Freed, E. O., Beyond Tsg101: the role of Alix in 'ESCRTing' HIV-1. Nat Rev Microbiol 2007, 5 (12), 912-6. 135. Odorizzi, G., The multiple personalities of Alix. Journal of cell science 2006, 119 (Pt 15), 3025-32. 136. Pires, R.; Hartlieb, B.; Signor, L.; Schoehn, G.; Lata, S.; Roessle, M.; Moriscot, C.; Popov, S.; Hinz, A.; Jamin, M.; Boyer, V.; Sadoul, R.; Forest, E.; Svergun, D. I.; Gottlinger, H. G.; Weissenhorn, W., A crescent-shaped ALIX dimer targets ESCRT-III CHMP4 filaments. Structure 2009, 17 (6), 843-56. 137. Zhai, Q.; Landesman, M. B.; Chung, H. Y.; Dierkers, A.; Jeffries, C. M.; Trewhella, J.; Hill, C. P.; Sundquist, W. I., Activation of the retroviral budding factor ALIX. J Virol 2011, 85 (17), 9222-6. 138. Sun, S.; Sun, L.; Zhou, X.; Wu, C.; Wang, R.; Lin, S. H.; Kuang, J., Phosphorylation-Dependent Activation of the ESCRT Function of ALIX in Cytokinetic Abscission and Retroviral Budding. Dev Cell 2016, 36 (3), 331-43. 111 139. Eikenes, A. H.; Malerod, L.; Christensen, A. L.; Steen, C. B.; Mathieu, J.; Nezis, I. P.; Liestol, K.; Huynh, J. R.; Stenmark, H.; Haglund, K., ALIX and ESCRT-III coordinately control cytokinetic abscission during germline stem cell division in vivo. PLoS Genet 2015, 11 (1), e1004904. 140. Bissig, C.; Gruenberg, J., ALIX and the multivesicular endosome: ALIX in Wonderland. Trends in cell biology 2014, 24 (1), 19-25. 141. Zhou, X.; Pan, S.; Sun, L.; Corvera, J.; Lin, S. H.; Kuang, J., The HIV-1 p6/EIAV p9 docking site in Alix is autoinhibited as revealed by a conformation-sensitive anti-Alix monoclonal antibody. Biochem J 2008, 414 (2), 215-20. 142. Zhai, Q.; Landesman, M. B.; Robinson, H.; Sundquist, W. I.; Hill, C. P., Structure of the Bro1 domain protein BROX and functional analyses of the ALIX Bro1 domain in HIV-1 budding. PLoS One 2011, 6 (12), e27466. 143. Katoh, K.; Shibata, H.; Hatta, K.; Maki, M., CHMP4b is a major binding partner of the ALG-2-interacting protein Alix among the three CHMP4 isoforms. Arch Biochem Biophys 2004, 421 (1), 159-65. 144. McCullough, J.; Fisher, R. D.; Whitby, F. G.; Sundquist, W. I.; Hill, C. P., ALIXCHMP4 interactions in the human ESCRT pathway. Proc Natl Acad Sci U S A 2008, 105 (22), 7687-91. 145. Schmidt, M. H.; Dikic, I.; Bogler, O., Src phosphorylation of Alix/AIP1 modulates its interaction with binding partners and antagonizes its activities. J Biol Chem 2005, 280 (5), 3414-25. 146. Sette, P.; Mu, R.; Dussupt, V.; Jiang, J.; Snyder, G.; Smith, P.; Xiao, T. S.; Bouamr, F., The Phe105 loop of Alix Bro1 domain plays a key role in HIV-1 release. Structure 2011, 19 (10), 1485-95. 147. Matsuo, H.; Chevallier, J.; Mayran, N.; Le Blanc, I.; Ferguson, C.; Faure, J.; Blanc, N. S.; Matile, S.; Dubochet, J.; Sadoul, R.; Parton, R. G.; Vilbois, F.; Gruenberg, J., Role of LBPA and Alix in multivesicular liposome formation and endosome organization. Science 2004, 303 (5657), 531-4. 148. Bissig, C.; Lenoir, M.; Velluz, M. C.; Kufareva, I.; Abagyan, R.; Overduin, M.; Gruenberg, J., Viral infection controlled by a calcium-dependent lipid-binding module in ALIX. Dev Cell 2013, 25 (4), 364-73. 149. Popov, S.; Popova, E.; Inoue, M.; Gottlinger, H. G., Divergent Bro1 domains share the capacity to bind human immunodeficiency virus type 1 nucleocapsid and to enhance virus-like particle production. J Virol 2009, 83 (14), 7185-93. 112 150. Dussupt, V.; Sette, P.; Bello, N. F.; Javid, M. P.; Nagashima, K.; Bouamr, F., Basic residues in the nucleocapsid domain of Gag are critical for late events of HIV-1 budding. J Virol 2011, 85 (5), 2304-15. 151. Mahul-Mellier, A. L.; Hemming, F. J.; Blot, B.; Fraboulet, S.; Sadoul, R., Alix, making a link between apoptosis-linked gene-2, the endosomal sorting complexes required for transport, and neuronal death in vivo. J Neurosci 2006, 26 (2), 542-9. 152. Carlton, J. G.; Agromayor, M.; Martin-Serrano, J., Differential requirements for Alix and ESCRT-III in cytokinesis and HIV-1 release. Proc Natl Acad Sci U S A 2008, 105 (30), 10541-6. 153. Chatellard-Causse, C.; Blot, B.; Cristina, N.; Torch, S.; Missotten, M.; Sadoul, R., Alix (ALG-2-interacting protein X), a protein involved in apoptosis, binds to endophilins and induces cytoplasmic vacuolization. J Biol Chem 2002, 277 (32), 29108-15. 154. Trioulier, Y.; Torch, S.; Blot, B.; Cristina, N.; Chatellard-Causse, C.; Verna, J. M.; Sadoul, R., Alix, a protein regulating endosomal trafficking, is involved in neuronal death. J Biol Chem 2004, 279 (3), 2046-52. 155. Shibata, H.; Yamada, K.; Mizuno, T.; Yorikawa, C.; Takahashi, H.; Satoh, H.; Kitaura, Y.; Maki, M., The penta-EF-hand protein ALG-2 interacts with a region containing PxY repeats in Alix/AIP1, which is required for the subcellular punctate distribution of the amino-terminal truncation form of Alix/AIP1. J Biochem 2004, 135 (1), 117-28. 156. Bouamr, F.; Melillo, J. A.; Wang, M. Q.; Nagashima, K.; de Los Santos, M.; Rein, A.; Goff, S. P., PPPYVEPTAP motif is the late domain of human T-cell leukemia virus type 1 Gag and mediates its functional interaction with cellular proteins Nedd4 and Tsg101 [corrected]. J Virol 2003, 77 (22), 11882-95. 157. Harty, R. N.; Brown, M. E.; Wang, G.; Huibregtse, J.; Hayes, F. P., A PPxY motif within the VP40 protein of Ebola virus interacts physically and functionally with a ubiquitin ligase: implications for filovirus budding. Proc Natl Acad Sci U S A 2000, 97 (25), 13871-6. 158. Dowlatshahi, D. P.; Sandrin, V.; Vivona, S.; Shaler, T. A.; Kaiser, S. E.; Melandri, F.; Sundquist, W. I.; Kopito, R. R., ALIX is a Lys63-specific polyubiquitin binding protein that functions in retrovirus budding. Dev Cell 2012, 23 (6), 1247-54. 159. Martin-Serrano, J.; Eastman, S. W.; Chung, W.; Bieniasz, P. D., HECT ubiquitin ligases link viral and cellular PPXY motifs to the vacuolar protein-sorting pathway. The Journal of cell biology 2005, 168 (1), 89-101. 113 160. Sette, P.; Jadwin, J. A.; Dussupt, V.; Bello, N. F.; Bouamr, F., The ESCRTassociated protein Alix recruits the ubiquitin ligase Nedd4-1 to facilitate HIV-1 release through the LYPXnL L domain motif. J Virol 2010, 84 (16), 8181-92. 161. Patnaik, A.; Chau, V.; Wills, J. W., Ubiquitin is part of the retrovirus budding machinery. Proceedings of the National Academy of Sciences 2000, 97 (24), 13069-13074. 162. Schubert, U.; Ott, D. E.; Chertova, E. N.; Welker, R.; Tessmer, U.; Princiotta, M. F.; Bennink, J. R.; Krausslich, H. G.; Yewdell, J. W., Proteasome inhibition interferes with gag polyprotein processing, release, and maturation of HIV-1 and HIV-2. Proc Natl Acad Sci U S A 2000, 97 (24), 13057-62. 163. Strack, B.; Calistri, A.; Accola, M. A.; Palu, G.; Gottlinger, H. G., A role for ubiquitin ligase recruitment in retrovirus release. Proc Natl Acad Sci U S A 2000, 97 (24), 13063-8. 164. Sette, P.; Nagashima, K.; Piper, R.; Bouamr, F., Ubiquitin conjugation to Gag is essential for ESCRT-mediated HIV-1 budding. Retrovirology 2013, 10 (1), 79. 165. Wright, M. H.; Berlin, I.; Nash, P. D., Regulation of endocytic sorting by ESCRTDUB-mediated deubiquitination. Cell biochemistry and biophysics 2011, 60 (1-2), 39-46. 166. Wing, S. S., Deubiquitinating enzymes--the importance of driving in reverse along the ubiquitin-proteasome pathway. Int J Biochem Cell Biol 2003, 35 (5), 590-605. 167. Losko, S.; Kopp, F.; Kranz, A.; Kolling, R., Uptake of the ATP-binding cassette (ABC) transporter Ste6 into the yeast vacuole is blocked in the doa4 Mutant. Molecular biology of the cell 2001, 12 (4), 1047-59. 168. Luhtala, N.; Odorizzi, G., Bro1 coordinates deubiquitination in the multivesicular body pathway by recruiting Doa4 to endosomes. The Journal of cell biology 2004, 166 (5), 717-29. 169. Isono, E.; Nagel, M. K., Deubiquitylating enzymes and their emerging role in plant biology. Front Plant Sci 2014, 5, 56. 170. Candela Blanes, A.; Hernandez Blasco, L.; Martin Serrano, C.; Romero Candeira, S., [Pneumothorax as a complication of acupuncture]. An Med Interna 1995, 12 (8), 4123. 171. Feng, Z.; Wu, C. F.; Zhou, X.; Kuang, J., Alternative polyadenylation produces two major transcripts of Alix. Arch Biochem Biophys 2007, 465 (2), 328-35. 172. Waheed, A. A.; Freed, E. O., Lipids and membrane microdomains in HIV-1 replication. Virus Res 2009, 143 (2), 162-76. 114 173. Usami, Y.; Popov, S.; Gottlinger, H. G., Potent rescue of human immunodeficiency virus type 1 late domain mutants by ALIX/AIP1 depends on its CHMP4 binding site. J Virol 2007, 81 (12), 6614-22. 174. Kaimbo, W. K.; Bifuko, Z. A.; Longo, M. B.; Dralands, L.; Missotten, L., Color vision in 42 Congolese patients with tuberculosis receiving ethambutol treatment. Bull Soc Belge Ophtalmol 2002, (284), 57-61. 175. Kaimbo Wa Kaimbo, D.; Missotten, L., Glaucoma in Congo. Bull Soc Belge Ophtalmol 1997, 267, 21-6. 176. Ganser-Pornillos, B. K.; Yeager, M.; Sundquist, W. I., The structural biology of HIV assembly. Current opinion in structural biology 2008, 18 (2), 203-17. 177. Ganser, B. K.; Li, S.; Klishko, V. Y.; Finch, J. T.; Sundquist, W. I., Assembly and analysis of conical models for the HIV-1 core. Science 1999, 283 (5398), 80-3. 178. Munshi, U. M.; Kim, J.; Nagashima, K.; Hurley, J. H.; Freed, E. O., An Alix fragment potently inhibits HIV-1 budding: characterization of binding to retroviral YPXL late domains. J Biol Chem 2007, 282 (6), 3847-55. 179. Lv, M.; Chen, L.; Qin, T.; Zhang, X.; Liu, P.; Yang, J., Angiomotin promotes breast cancer cell proliferation and invasion. Oncol Rep 2015, 33 (4), 1938-46. 180. Jouvenet, N.; Bieniasz, P. D.; Simon, S. M., Imaging the biogenesis of individual HIV-1 virions in live cells. Nature 2008, 454 (7201), 236-40. 181. Sette, P.; Jadwin, J. A.; Dussupt, V.; Bello, N.; Bouamr, F., The ESCRT-associated protein Alix recruits the ubiquitin ligase Nedd4-1 to facilitate HIV-1 release through the LYPXnL L domain motif. J. Virol. 2010, JVI.00634-10. 182. Saffarian, S.; Kirchhausen, T., Differential evanescence nanometry: live-cell fluorescence measurements with 10-nm axial resolution on the plasma membrane. Biophysical journal 2008, 94 (6), 2333-42. 183. Allersma, M. W.; Bittner, M. A.; Axelrod, D.; Holz, R. W., Motion Matters: Secretory Granule Motion Adjacent to the Plasma Membrane and Exocytosis. Mol. Biol. Cell 2006, 17 (5), 2424-2438. 184. Axelrod, D., Total internal reflection fluorescence microscopy in cell biology. In Biophotonics, Pt B, 2003; Vol. 361, pp 1-33. 185. Saffarian, S.; Kirchhausen, T., Differential evanescence nanometry: Live-cell fluorescence measurements with 10-nm axial resolution on the plasma membrane. Biophysical Journal 2008, 94 (6), 2333-2342. 115 186. Chen, Y.; Wu, B.; Musier-Forsyth, K.; Mansky, L. M.; Mueller, J. D., Fluorescence Fluctuation Spectroscopy on Viral-Like Particles Reveals Variable Gag Stoichiometry. Biophysical journal 2009, 96 (5), 1961-1969. 187. Adler, F. R., Modeling the Dynamics of Life: Calculus and Probability for Life Scientists. Thomson Brooks/Cole: Belmont, CA, 2005. 188. Zar, J. H., Biostatistical Analysis. Prentice-Hall/Pearson: Upper Saddle River, NJ, 2010. 189. Jouvenet, N.; Bieniasz, P. D.; Simon, S. M., Imaging the biogenesis of individual HIV-1 virions in live cells. 2008, 454 (7201), 236-240. 190. Baumgartel, V.; Ivanchenko, S.; Dupont, A.; Sergeev, M.; Wiseman, P. W.; Krausslich, H.-G.; Brauchle, C.; Muller, B.; Lamb, D. C., Live-cell visualization of dynamics of HIV budding site interactions with an ESCRT component. Nature cell biology 2011, 13 (4), 469-474. 191. Peskin, C.; Odell, G.; Oster, G., Cellular motions and thermal fluctuations: the Brownian ratchet. Biophys. J. 1993, 65, 316-324. 192. Weiss, E. R.; Popova, E.; Yamanaka, H.; Kim, H. C.; Huibregtse, J. M.; Göttlinger, H., Rescue of HIV-1 Release by Targeting Widely Divergent NEDD4-Type Ubiquitin Ligases and Isolated Catalytic HECT Domains to Gag. PLoS Pathog 2010, 6 (9), e1001107. 193. Chung, H.-Y.; Morita, E.; von Schwedler, U.; Müller, B.; Kräusslich, H.-G.; Sundquist, W. I., NEDD4L Overexpression Rescues the Release and Infectivity of Human Immunodeficiency Virus Type 1 Constructs Lacking PTAP and YPXL Late Domains. Journal of Virology 2008, 82 (10), 4884-4897. 194. Mercenne, G.; Alam, S. L.; Arii, J.; Lalonde, M. S.; Sundquist, W. I., Angiomotin functions in HIV-1 assembly and budding. Elife 2015, 4. 195. Huang, M.; Orenstein, J. M.; Martin, M. A.; Freed, E. O., p6Gag is required for particle production from full-length human immunodeficiency virus type 1 molecular clones expressing protease. Journal of Virology 1995, 69 (11), 6810-8. 196. Göttlinger, H. G.; Dorfman, T.; Sodroski, J. G.; Haseltine, W. A., Effect of mutations affecting the p6 gag protein on human immunodeficiency virus particle release. Proceedings of the National Academy of Sciences 1991, 88 (8), 3195-3199. 197. Garrus, J. E.; von Schwedler, U. K.; Pornillos, O. W.; Morham, S. G.; Zavitz, K. H.; Wang, H. E.; Wettstein, D. A.; Stray, K. M.; Côté, M.; Rich, R. L.; Myszka, D. G.; Sundquist, W. I., Tsg101 and the Vacuolar Protein Sorting Pathway Are Essential for HIV1 Budding. Cell 2001, 107 (1), 55-65. 116 198. Martin-Serrano, J.; Zang, T.; Bieniasz, P. D., HIV-1 and Ebola virus encode small peptide motifs that recruit Tsg101 to sites of particle assembly to facilitate egress. Nat Med 2001, 7 (12), 1313-1319. 199. VerPlank, L.; Bouamr, F.; LaGrassa, T. J.; Agresta, B.; Kikonyogo, A.; Leis, J.; Carter, C. A., Tsg101, a homologue of ubiquitin-conjugating (E2) enzymes, binds the L domain in HIV type 1 Pr55Gag. Proceedings of the National Academy of Sciences 2001, 98 (14), 7724-7729. 200. Demirov, D. G.; Ono, A.; Orenstein, J. M.; Freed, E. O., Overexpression of the Nterminal domain of TSG101 inhibits HIV-1 budding by blocking late domain function. Proceedings of the National Academy of Sciences 2002, 99 (2), 955-960. 201. Strack, B.; Calistri, A.; Craig, S.; Popova, E.; Göttlinger, H. G., AIP1/ALIX Is a Binding Partner for HIV-1 p6 and EIAV p9 Functioning in Virus Budding. Cell 2003, 114 (6), 689-699. 202. Martin-Serrano, J.; Yaravoy, A.; Perez-Caballero, D.; Bieniasz, P. D., Divergent retroviral late-budding domains recruit vacuolar protein sorting factors by using alternative adaptor proteins. Proceedings of the National Academy of Sciences 2003, 100 (21), 1241412419. 203. von Schwedler, U. K.; Stuchell, M.; Müller, B.; Ward, D. M.; Chung, H.-Y.; Morita, E.; Wang, H. E.; Davis, T.; He, G.-P.; Cimbora, D. M.; Scott, A.; Kräusslich, H.G.; Kaplan, J.; Morham, S. G.; Sundquist, W. I., The Protein Network of HIV Budding. Cell 2003, 114 (6), 701-713. 204. Vincent, O.; Rainbow, L.; Tilburn, J.; Arst, J., Herbert N.; Peñalva, M. A., YPXL/I Is a Protein Interaction Motif Recognized by Aspergillus PalA and Its Human Homologue, AIP1/Alix. Molecular and cellular biology 2003, 23 (5), 1647-1655. 205. Dussupt, V.; Javid, M. P.; Abou-Jaoudé, G.; Jadwin, J. A.; de La Cruz, J.; Nagashima, K.; Bouamr, F., The Nucleocapsid Region of HIV-1 Gag Cooperates with the PTAP and LYPX_nL Late Domains to Recruit the Cellular Machinery Necessary for Viral Budding. PLoS Pathog 2009, 5 (3), e1000339. 206. Popov, S.; Popova, E.; Inoue, M.; Göttlinger, H. G., Human Immunodeficiency Virus Type 1 Gag Engages the Bro1 Domain of ALIX/AIP1 through the Nucleocapsid. Journal of Virology 2008, 82 (3), 1389-1398. 207. Demirov, D. G.; Orenstein, J. M.; Freed, E. O., The Late Domain of Human Immunodeficiency Virus Type 1 p6 Promotes Virus Release in a Cell Type-Dependent Manner. Journal of Virology 2002, 76 (1), 105-117. 117 208. Fisher, R. D.; Chung, H.-Y.; Zhai, Q.; Robinson, H.; Sundquist, W. I.; Hill, C. P., Structural and Biochemical Studies of ALIX/AIP1 and Its Role in Retrovirus Budding. Cell 2007, 128 (5), 841-852. 209. Usami, Y.; Popov, S.; Gottlinger, H. G., Potent Rescue of Human Immunodeficiency Virus Type 1 Late Domain Mutants by ALIX/AIP1 Depends on Its CHMP4 Binding Site. J. Virol. 2007, 81 (12), 6614-6622. 210. Puffer, B. A.; Parent, L. J.; Wills, J. W.; Montelaro, R. C., Equine infectious anemia virus utilizes a YXXL motif within the late assembly domain of the Gag p9 protein. Journal of Virology 1997, 71 (9), 6541-6. 211. Lee, S.; Joshi, A.; Nagashima, K.; Freed, E. O.; Hurley, J. H., Structural basis for viral late-domain binding to Alix. Nat Struct Mol Biol 2007, 14 (3), 194-199. 212. Sette, P.; Dussupt, V.; Bouamr, F., Identification of the HIV-1 NC Binding Interface in Alix Bro1 Reveals a Role for RNA. Journal of Virology 2012, 86 (21), 1160811615. 213. Dowlatshahi, Dara P.; Sandrin, V.; Vivona, S.; Shaler, Thomas A.; Kaiser, Stephen E.; Melandri, F.; Sundquist, Wesley I.; Kopito, Ron R., ALIX Is a Lys63-Specific Polyubiquitin Binding Protein that Functions in Retrovirus Budding. Developmental Cell 2012, 23 (6), 1247-1254. 214. Keren-Kaplan, T.; Attali, I.; Estrin, M.; Kuo, L. S.; Farkash, E.; Jerabek-Willemsen, M.; Blutraich, N.; Artzi, S.; Peri, A.; Freed, E. O.; Wolfson, H. J.; Prag, G., Structure-based in silico identification of ubiquitin-binding domains provides insights into the ALIXV:ubiquitin complex and retrovirus budding. EMBO J 2013, 32 (4), 538-551. 215. Joshi, A.; Munshi, U.; Ablan, S. D.; Nagashima, K.; Freed, E. O., Functional Replacement of a Retroviral Late Domain by Ubiquitin Fusion. Traffic 2008, 9 (11), 19721983. 216. Zhai, Q.; Landesman, M. B.; Chung, H.-Y.; Dierkers, A.; Jeffries, C. M.; Trewhella, J.; Hill, C. P.; Sundquist, W. I., Activation of the Retroviral Budding Factor ALIX. J. Virol. 2011, JVI.02653-10. 217. Zhou, X.; Pan, S.; Sun, L.; Corvera, J.; Lee, Y.-C.; Lin, S.-H.; Kuang, J., The CHMP4b- and Src-docking sites in the Bro1 domain are autoinhibited in the native state of Alix. Biochem J 2009, 418 (2), 277-284. 218. Pires, R.; Hartlieb, B.; Signor, L.; Schoehn, G.; Lata, S.; Roessle, M.; Moriscot, C.; Popov, S.; Hinz, A.; Jamin, M.; Boyer, V.; Sadoul, R.; Forest, E.; Svergun, D. I.; Göttlinger, H. G.; Weissenhorn, W., A Crescent-Shaped ALIX Dimer Targets ESCRT-III CHMP4 Filaments. Structure 2009, 17 (6), 843-856. 118 219. Katoh, K.; Shibata, H.; Hatta, K.; Maki, M., CHMP4b is a major binding partner of the ALG-2-interacting protein Alix among the three CHMP4 isoforms. Archives of Biochemistry and Biophysics 2004, 421 (1), 159-165. 220. Kim, J.; Sitaraman, S.; Hierro, A.; Beach, B. M.; Odorizzi, G.; Hurley, J. H., Structural Basis for Endosomal Targeting by the Bro1 Domain. Developmental Cell 2005, 8 (6), 937-947. 221. McCullough, J.; Fisher, R. D.; Whitby, F. G.; Sundquist, W. I.; Hill, C. P., ALIXCHMP4 interactions in the human ESCRT pathway. Proceedings of the National Academy of Sciences 2008, 105 (22), 7687-7691. 222. Effantin, G.; Dordor, A.; Sandrin, V.; Martinelli, N.; Sundquist, W. I.; Schoehn, G.; Weissenhorn, W., ESCRT-III CHMP2A and CHMP3 form variable helical polymers in vitro and act synergistically during HIV-1 budding. Cellular Microbiology 2013, 15 (2), 213-226. 223. Hanson, P. I.; Roth, R.; Lin, Y.; Heuser, J. E., Plasma membrane deformation by circular arrays of ESCRT-III protein filaments. J. Cell Biol. 2008, 180 (2), 389-402. 224. Bodon, G.; Chassefeyre, R.; Pernet-Gallay, K.; Martinelli, N.; Effantin, G.; Hulsik, D. L.; Belly, A.; Goldberg, Y.; Chatellard-Causse, C.; Blot, B.; Schoehn, G.; Weissenhorn, W.; Sadoul, R., Charged Multivesicular Body Protein 2B (CHMP2B) of the Endosomal Sorting Complex Required for Transport-III (ESCRT-III) Polymerizes into Helical Structures Deforming the Plasma Membrane. Journal of Biological Chemistry 2011, 286 (46), 40276-40286. 225. Hurley, J. H.; Hanson, P. I., Membrane budding and scission by the ESCRT machinery: it's all in the neck. Nat Rev Mol Cell Biol 2010, 11 (8), 556-566. 226. Martin-Serrano, J.; Neil, S. J. D., Host factors involved in retroviral budding and release. Nat Rev Micro 2011, 9 (7), 519-531. 227. Votteler, J.; Sundquist, Wesley I., Virus Budding and the ESCRT Pathway. Cell Host & Microbe 2013, 14 (3), 232-241. 228. Babst, M.; Sato, T. K.; Banta, L. M.; Emr, S. D., Endosomal transport function in yeast requires a novel AAA-type ATPase, Vps4p. EMBO J 1997, 16 (8), 1820-1831. 229. Lata, S.; Schoehn, G.; Jain, A.; Pires, R.; Piehler, J.; Gottlinger, H. G.; Weissenhorn, W., Helical Structures of ESCRT-III Are Disassembled by VPS4. Science 2008, 321 (5894), 1354-1357. 230. Ku, P.-I.; Miller, Anna K.; Ballew, J.; Sandrin, V.; Adler, Frederick R.; Saffarian, S., Identification of Pauses during Formation of HIV-1 Virus Like Particles. Biophysical journal 2013, 105 (10), 2262-2272. 119 231. Sund, S. E.; Swanson, J. A.; Axelrod, D., Cell Membrane Orientation Visualized by Polarized Total Internal Reflection Fluorescence. Biophys. J. 1999, 77 (4), 2266-2283. 232. Axelrod, D., Chapter 7 Total Internal Reflection Fluorescence Microscopy. In Methods in Cell Biology, Dr. John, J. C.; Dr. H. William Detrich, III, Eds. Academic Press: 2008; Vol. Volume 89, pp 169-221. 233. Ivanchenko, S.; Godinez, W. J.; Lampe, M.; Kräusslich, H.-G.; Eils, R.; Rohr, K.; Bräuchle, C.; Müller, B.; Lamb, D. C., Dynamics of HIV-1 Assembly and Release. PLoS Pathog 2009, 5 (11), e1000652. 234. Jouvenet, N.; Simon, S. M.; Bieniasz, P. D., Imaging the interaction of HIV-1 genomes and Gag during assembly of individual viral particles. Proceedings of the National Academy of Sciences 2009, 106 (45), 19114-19119. 235. Van Engelenburg, S. B.; Shtengel, G.; Sengupta, P.; Waki, K.; Jarnik, M.; Ablan, S. D.; Freed, E. O.; Hess, H. F.; Lippincott-Schwartz, J., Distribution of ESCRT Machinery at HIV Assembly Sites Reveals Virus Scaffolding of ESCRT Subunits. Science 2014. 236. Zhai, Q.; Fisher, R. D.; Chung, H.-Y.; Myszka, D. G.; Sundquist, W. I.; Hill, C. P., Structural and functional studies of ALIX interactions with YPXnL late domains of HIV1 and EIAV. Nat Struct Mol Biol 2008, 15 (1), 43-49. 237. Salmon, P.; Oberholzer, J.; Occhiodoro, T.; Morel, P.; Lou, J.; Trono, D., Reversible Immortalization of Human Primary Cells by Lentivector-Mediated Transfer of Specific Genes. Mol Ther 2000, 2 (4), 404-414. 238. Wriggers, W.; Chakravarty, S.; Jennings, P. A., Control of protein functional dynamics by peptide linkers. Peptide Science 2005, 80 (6), 736-746. 239. Wollert, T.; Hurley, J. H., Molecular mechanism of multivesicular body biogenesis by ESCRT complexes. Nature 2010, advance online publication. 240. Strack, B.; Calistri, A.; Accola, M. A.; Palù, G.; Göttlinger, H. G., A role for ubiquitin ligase recruitment in retrovirus release. Proceedings of the National Academy of Sciences 2000, 97 (24), 13063-13068. 241. Schubert, U.; Ott, D. E.; Chertova, E. N.; Welker, R.; Tessmer, U.; Princiotta, M. F.; Bennink, J. R.; Kräusslich, H.-G.; Yewdell, J. W., Proteasome inhibition interferes with Gag polyprotein processing, release, and maturation of HIV-1 and HIV-2. Proceedings of the National Academy of Sciences 2000, 97 (24), 13057-13062. 242. Pashkova, N.; Gakhar, L.; Winistorfer, Stanley C.; Sunshine, Anna B.; Rich, M.; Dunham, Maitreya J.; Yu, L.; Piper, Robert C., The Yeast Alix Homolog Bro1 Functions as a Ubiquitin Receptor for Protein Sorting into Multivesicular Endosomes. Developmental Cell 2013, 25 (5), 520-533. 120 243. Weissleder, R.; Tung, C.-H.; Mahmood, U.; Bogdanov, A., In vivo imaging of tumors with protease-activated near-infrared fluorescent probes. Nat Biotech 1999, 17 (4), 375-378. 244. Yoo, L.; Park, J.-S.; Kwon, K. C.; Kim, S.-E.; Jin, X.; Kim, H.; Lee, J., Fluorescent viral nanoparticles with stable in vitro and in vivo activity. Biomaterials 2012, 33 (26), 6194-6200. 245. Zwan, M. D.; Villemagne, V. L.; Dore, V.; Buckley, R.; Bourgeat, P.; Veljanoski, R.; Salvado, O.; Williams, R.; Margison, L.; Rembach, A.; Macaulay, S. L.; Martins, R.; Ames, D.; van der Flier, W. M.; Ellis, K. A.; Scheltens, P.; Masters, C. L.; Rowe, C. C., Subjective Memory Complaints in APOEvarepsilon4 Carriers are Associated with High Amyloid-beta Burden. J Alzheimers Dis 2015, 49 (4), 1115-22. 246. Chen, C.; Vincent, O.; Jin, J.; Weisz, O. A.; Montelaro, R. C., Functions of early (AP-2) and late (AIP1/ALIX) endocytic proteins in equine infectious anemia virus budding. J Biol Chem 2005, 280 (49), 40474-80. 247. Tanzi, G. O.; Piefer, A. J.; Bates, P., Equine infectious anemia virus utilizes host vesicular protein sorting machinery during particle release. J Virol 2003, 77 (15), 8440-7. 248. Ku, P. I.; Bendjennat, M.; Ballew, J.; Landesman, M. B.; Saffarian, S., ALIX is recruited temporarily into HIV-1 budding sites at the end of gag assembly. PLoS One 2014, 9 (5), e96950. 249. Sette, P.; O'Connor, S. K.; Yerramilli, V. S.; Dussupt, V.; Nagashima, K.; Chutiraka, K.; Lingappa, J.; Scarlata, S.; Bouamr, F., HIV-1 Nucleocapsid Mimics the Membrane Adaptor Syntenin PDZ to Gain Access to ESCRTs and Promote Virus Budding. Cell Host Microbe 2016, 19 (3), 336-48. 250. Sandrin, V.; Sundquist, W. I., ESCRT requirements for EIAV budding. Retrovirology 2013, 10, 104.