Thesis Reference

Thesis

Innate immune signaling and the contribution of different regions of capsid to HIV-1 restriction by TRIM5

LASCANO MAILLARD, Maria Josefina

Abstract

Reference

LASCANO MAILLARD, Maria Josefina. Innate immune signaling and the contribution of different regions of capsid to HIV-1 restriction by TRIM5. Thèse de doctorat : Univ. Genève, 2014, no. Sc. 4667

URN : urn:nbn:ch:unige-381368 DOI : 10.13097/archive-ouverte/unige:38136

Available at: http://archive-ouverte.unige.ch/unige:38136

Disclaimer: layout of this document may differ from the published version.

1 / 1 UNIVERSITE DE GENEVE

Département de Génétique et Evolution FACULTE DES SCIENCES

Professeur François Karch

Département de Microbiologie et Médecine moléculaire FACULTE DE MEDECINE

Professeur Jeremy Luban

Innate immune signaling and the contribution of different regions of capsid to HIV-1 restriction by TRIM5

THESE

Présentée à la Faculté des Sciences de l’Université de Genève Pour obtenir le grade de Docteur ès Sciences, mention Biologie

par

Maria Josefina Lascano Maillard

Bassecourt (JU)

Thèse n°4667

Genève 2014

Innate immune signaling and the contribution of different regions of capsid to HIV-1 restriction by TRIM5.

Par Maria Josefina Lascano Maillard

ABSTRACT

The cellular factor TRIM5α performs a dual role in the innate immunity. First, TRIM5α has an intrinsic ability to induce the AP-1 and NFκB pathways and contributes to the establishment of the LPS-mediated antiviral state. Second, it functions as a restriction factor, blocking early stages of retroviral infection in a capsid-dependent manner. The connections between these two functions of TRIM5α are debated. We investigated the conservation, in TRIM5 orthologues, of the ability to activate the innate immune pathways and analyzed the signification of this function in the context of TRIM5- mediated HIV-1 restriction. We took the advantage that there are seven TRIM5 orthologues in the mouse, with variable abilities to activate the innate immune signaling, to determine the contribution of this signal activator function to the restriction process. We found that only the TRIM5 orthologues that could activate the innate immune pathways could restrict HIV-1, when fused to the capsid (CA)-binding cyclophilin A (Cyp or CypA) domain from the owl monkey TRIM5Cyp. While human TRIM5α poses a potent blockade to N-MLV and EIAV infection, HIV-1 is much less affected in in vitro experiments using laboratory-adapted strains. Although certain gag-protease variants arising from clinical isolates show an increased sensitivity to human TRIM5α, the involvement of CA and the regions that influence the recognition of these mutants by the restriction factor are not completely understood. Here, we showed that three regions of the N- terminal domain of HIV-1 capsid are susceptible to modulate the sensitivity to human TRIM5α: the 4th and 7th helices, and the cyclophilin A (CypA)-binding loop. Taken together our data show the importance of TRIM5-mediated activation of the innate immune signaling in retroviral restriction and suggests a complex interplay between CA, CypA and TRIM5α during this process.

RESUME

Le facteur cellulaire TRIM5α possède une double fonction. Premièrement, TRIM5! montre une capacité intrinsèque d’activer les voies de signalisation cellulaires AP-1 et NFκB et contribue à l’établissement de l’état antiviral enclenché par une stimulation de lipopolisaccharide (LPS). Deuxièmement, il fonctionne en tant que facteur de restriction, en bloquant la réplication rétrovirale à un stage précoce de l’infection, de manière dépendante de la capside (CA). Les connections entre ces deux fonctions de TRIM5α font l’objet de débats. Nous avons analysé la conservation de la fonction activatrice d’immunité innée chez des orthologues de TRIM5α et nous avons tenté de déterminer la signification de cette activité dans le contexte de la restriction rétrovirale. Nous avons exploité différents orthologues murins de TRIM5α qui montrent différentes habiletés à activer les signaux cellulaires de l’immunité innée pour tenter d’établir l’importance de cette fonction dans le processus de restriction. Nous avons trouvé une corrélation entre la capacité d’un orthologue de TRIM5α à activer l’immunité innée et son habileté à bloquer le VIH-1, quand ceux-ci sont fusionnés à un domaine de TRIM5-CyclophilinA (TRIM5Cyp) de l’espèce Aotus trivirgatus, qui lie la capside. Tandis que le TRIM5α humain bloque très fortement la réplication du gammarétrovirus N-MLV et du lentivirus EIAV, le VIH-1 est beaucoup moins affecté au cours d’expériences réalisées in vitro et chez des souches adaptées au laboratoire. Bien que certains virus avec des variantes de séquences de gag-protease montrent une plus grande sensibilité au TRIM5α humain, le rôle de CA et l’influence des différentes regions de CA sur cette susceptibilité ne sont pas clairement établis. Nous avons montrés que trois régions du domaine N-terminal CA sont susceptibles de moduler la reconnaissance par TRIM5α : les hélices 4 et 7, ainsi que la boucle liant la cyclophiline A (CypA). Nos résultats montrent l’importance l’activation de l’immunité innée par TRIM5α dans la restriction rétrovirale et suggèrent une intéraction entre CA, CypA et TRIM5α durant ce processus.

3 Acknowledgments

First of all, I would like to thank my supervisor, Professor Jeremy Luban that allowed me to conduct a thesis in his lab. We spoke very interesting science together and he supported me even when I had crazy ideas that I wanted to test! Next, I want to thank Professor François Karch, to have accepted to be the co- director of my thesis. Thanks to the two members of the jury, Dr. Angela Ciuffi and Dr. Dominique Garçin, to have kindly accepted my invitation. A special thanks to the latter that, together with Professor Laurent Roux, our former dear department Director, to have welcomed me for their lab meetings and have given me advises for my research. Additionally, Dominique’s laugh is contagious! I am also very thankful to Professor William Kelley, because of his kindness and his precious help with the English language corrections on my thesis introduction. Thanks to my friend and colleague, Hanni Bartels. I will never forget the coffees we drank together, or our Pain quotidien on Fridays! Thank you to Manel and Anastasia, to the nice lunches spent together and to have adopted me after my dear friend Hanni moved away! I want to thank Madeleine for her sweetness and for the very nice dinners in her beautiful terrasse! Thanks to Stéphane that coached me in my beginnings and that was of very good support later. It was nice to speak French with him! Thanks to my other colleagues, Massimo, Alberto, Federico, Christian, Dario, and Jessica. Or should I say (the remaining one now have the capacity to understand anyways): Grazie milla! I wanted to thank my husband Julien for its support and sweetness during my entire thesis and for being part of my life. He allows me to be a better person each day. He also gave the more beautiful present I could have dream of: mi muchachito Augustin, the sunshine of our home! Thanks to my mum and my dad, Adriana and José for all the love they gave to my brothers and me, as well as for encouraging us in every moments. We are very lucky to have them! Thanks to my brothers Santiago and Clemente for all the great moments we share together and for always being present if I need support.

4 Thank you to all my family for making Augustin so happy each time that he goes to visit. Thank you to Geneviève and Jean, for their kindness and for the nice walks in the mountains as well as the table game moments spent together. Thanks to Anne-Marie, Nina and Patrick, for having me welcomed in their family, for their sweetness and for all the good moments spent together! Last but not least, I would like to dedicate this manuscript to my beloved grandmother Josefina Christe that always believed in me and supported me. She will always be present in my heart and thoughts.

5 Table of contents

CHAPTER 1

INTRODUCTION

1.1.1 Retroviruses…………………….……………………………………. 10

1.1.2 The discovery of retroviruses and the RT enzyme…..………………. 11

1.1.3 The general structure of retroviruses……………………….………..13

1.1.4 The reverse-transcription process. ….………………………………….. 15

1.1.5 The classification of retroviruses….…………………………………. 17

1.1.6 The Acquired Immunodeficiency Syndrome (AIDS) ………………… 19 1.1.7 The structures of the HIV-1 virion and genome……………………... 22

1.1.8 The HIV-1 life cycle…………….………………………………………27

1.2 TRIM5 and the innate immunity….…………………………………….33

1.2.1 The Pattern-recognition receptors….………………………………..34 1.2.2 The innate immune pathways….…………………………………….38

1.2.3 Immunity to retroviruses: restriction factors……………………...... 43

1.2.4 TRIM5-mediated retroviral restriction….……………………………50

6 1.2.5 TRIM5 is a PRR….…………………………………….……………..54 1.2.6 TRIM5 in the mouse………………………………….……………..55 1.2.7 TRIM5 take over on the acquired immunity……….………………56

1.3 Aims of the thesis………………………………….………………….57

CHAPTER 2

Introduction….………………………………….………………….……..59

2.1 The link between the two functions of TRIM5: induction of the innate immune signaling and retroviral restriction..………………….….…….………………….……………… 61

2.2 Investigation of the role of murine TRIM5 orthologues as natural restriction factors...………………….….…….………………….…………………. 101

CHAPTER 3

3 The role of the human trim5α in the restriction of HIV-1 variants that appear in vivo…..….………………………………….………………….…………….. 125

CHAPTER 4

4 Discussion..….………………………………….………………….…………155

REFERENCES………………………………….………………….…………168

ANNEX 1

A1 “TRIM5 is an innate immune sensor for the retrovirus capsid lattice”……187

7 List of Abbreviations

cDNA Complementary DNA CypA Cyclophilin A TRIM5Cyp TRIM5-cyclophilin A RF Ring finger domain BB B-box domain CC Coiled-coil domain T12A TRIM12A T12B TRIM12B T12C TRIM12C T30A TRIM30A T30B TRIM30B T30C TRIM30C T30D TRIM30D

CA Capsid MA Matrix PR Protease gRNA Genomic RNA RSV Rous sarcoma virus AIDS Aquired Immunodefficiency Syndrom HIV Human Immunodeficiency Virus HEK293T Human embryonic kidney 293T fibroblasts IRES Internal ribosome entry site PPT Polypurine tract LTR Long terminal repeats miRNA MicroRNA shRNA Short hairpin RNA MLV Murine leukemia virus EIAV Equine Infectious Anemia Virus

8 MSCV Murine stem cell virus ERV Endogenous retrovirus APC Antigen presenting cells DC Dendritic cell TLR Toll-like receptor NLR Nod-like receptor RLR RIG-I-like recptor PRR Pattern recognition receptor LPS Lipopolysaccharide mRNA Messenger RNA PIC Preintegration complex RTC Reverse transcription complex PBS Phosphate buffered saline PCR Polymerase chain reaction Luc Luciferase RT Reverse Transcriptase tRNA Transfer RNA

Chapter 1

INTRODUCTION

1.1.1 Retroviruses Viral replication requires the transcription and translation machinery, which they themselves lack, from the organisms that they infect. The common feature among all retroviruses, and also what makes them unique among other viruses, is that they reverse transcribe their RNA genome into DNA that can be inserted into the host genome 1 (figure 1). For this process, retroviruses encode an RNA-dependent DNA polymerase, an RNAseH, and a host-encoded transfer RNA (tRNA) that serves as the primer for reverse transcription.

Discovered and first studied as disease-causing agents, many decades of research has uncovered the molecular mechanisms governing spread, replication and disease progression caused by these viruses. Retroviruses can produce fast and slow-progression diseases including various types of tumors and immunodeficiency 2,3.

A principal characteristic of these viruses is that they integrate into the host genome 4. This is the reason why, they later became to be used as vectors for gene delivery into cells 5-7.

If the integration event happens in the germline, the retroviral sequences can be spread from one generation to the next, a phenomenon called vertical transmission8. This feature of retroviruses is exploited to trace the evolution of genes of the host species, which inherits retroviral sequences in a Mendelian fashion, and used in the study of speciation 9,10.

10 A subset of retroviruses lead to cell transformation and cancer, therefore they are designated as oncoviruses 11. The investigation of the corresponding insertion sites within the genome thus lead to the identification of genes involved in cell growth and tumor promotion 12. Certain oncoviruses have inherent transforming potential owing to the prior acquisition of host sequences 13This characteristic is useful for the study of genetic regulation of the cell growth 14.

Budding Maturation Binding Membrane Fusion

Expression Assembly

Reverse Transcription

Nuclear transport Integration

Figure 1: The retroviral life cycle. The main steps of the retroviral replication cycle are depicted. Blue: capsid; yellow: nucleocapsid; black bars within the nucleocapsid: RNA genome; orange bars: DNA genome. Courtesy of Prof. Jeremy Luban (adapted).

1.1.2 The discovery of retroviruses and the reverse-transcriptase enzyme

The first retroviruses were discovered at the beginning of the twentieth century as oncogenic agents affecting birds. Ellermann and Bang found that leukosis in poultry was caused by a factor present in ultra-filtered cell extracts 15, that was later called Avian Leukemia virus. A few years later Rous showed that an agent

11 present in cell-free extracts was the intrinsic cause of the sarcoma formation in the fowl 5,16. This retrovirus was named after the man who discovered it, the Rous Sarcoma Virus (RSV).

Almost thirty years later, a murine virus was found to be the agent provoking leukemia in mice and thus was called Murine Leukemia Virus (MLV) 17. MLV belongs to the gammaretrovirus genus of the retroviridae family 17.

The concept of an RNA virus converting its genetic material into a DNA form and integrating into the host genome was not yet formulated. The only information scientists had at that moment was that the viral agent did not have a DNA genome but instead was constituted of RNA 5.

In the beginning of the 1960s, the molecular biologist Howard Temin worked with the RSV and found that inhibiting the DNA synthesis blocked the viral replication 18. This led him to propose the provirus hypothesis. That is, the retroviruses have a DNA intermediate in the cells that they infect.

Later, in the year 1970 his team and another virologist involved in the MLV research published separately data showing the presence in RNA tumour virus particles - called virions - of a RNA-dependent DNA polymerase activity, by correlating the induced RNA degradation with the decrease of the DNA synthesis. This enzyme was later called the reverse-transcriptase 13,19,20.

In humans, certain types of acute leukemias were studied and the viral cause of this disease was soon investigated. A type C morphology (that will be defined in the following section) retrovirus, which close relative had been discovered to induce leukemia in the Gibbon ape was pointed out by Robert Gallo to be the cause of the human disease and it was called Human T cell Leukemia Virus (HTLV) 21.

The previous findings provided the biochemical and molecular tools that ultimately allowed the subsequent identification of the Human

12 Immunodeficiency Virus type 1 (HIV) as the agent causing the Acquired Immunodeficiency Syndrome (AIDS) 22. In the next section, I will introduce the general retroviral structure.

1.1.3 The general structure of retroviruses

It all starts with the viral RNA. The positive single-stranded RNA genome is composed of different regulatory sequences and open reading frames (ORFs) 12 (figure 2), and has a 5’ cap and a poly A tail. The regulatory elements are located at the extremities of the viral RNA and consist of repeated (R) sequences, a unique 5’ sequence (U5) containing a cis- acting attachement (att) site, a unique 3’ sequence (U3), the primer binding site (PBS), the psi (packaging signal) element (ψ) and a polypurine tract (PPT) 12 (figure 2).

The R regions are redundant in sequence and are found after the m7G5’ppp5’Gmp cap, which mimics the eukaryotic mRNA 5’cap. The U5 sequence is immediately downstream of the 5’ R sequence and contains the att sequence that is involved in proviral integration. These regions are followed by the PBS where the specific tRNA primer hybridizes and starts the transcription of the minus-strand DNA (-sDNA). The next sequence in the RNA genome is the ψ region recapitulating most of the sequences required for viral genome packaging into the viral particles. A major splice donor site, that gives rise to different subgenomic mRNAs, often closely follows this element. Subgenomic RNAs are different mRNA species created when reverse transcription jumps on the template in the 3’ to 5’ orientation. The resulting mRNAs have variable 5’ regions overlapping with the template strands at different levels but the same 3’ sequence. The generation of various mRNAs allows condensing a high amount of information 12.

The PPT, positioned at the 3’ end of the viral genome, consists of a row of purines Adenine and Guanosine, required for the initiation of the +sDNA transcription.

13 Finally, the U3 region preceding the polyA tail contains another att site and in addition a set of cis-regulatory sequences essential for viral gene expression. Given that the synthesis of the viral DNA involves a duplication of the extremities of the RNA templates with a subsequent transfer of the U5 and U3 regions, the two ends in the resulting dsDNA are identical and these are called Long Terminal Repeats (LTRs) 12.

The provirus is integrated and found in the host genome with the flanking LTRs 5. When the provirus is transcribed, the 5’ U3 region is not taken into account and the synthesis proceeds until the R to U5 boundary. In this way, the resulting viral RNA has the same genomic organization as the template from viral particles.

The viral proteins are encoded by three ORFs, namely the group antigen (gag), the polymerase (pol) and the envelope (env). These genes code for precursos that once cleaved will give rise to more than one protein. The gag ORF codes for the matrix (MA), the capsid (CA) and the nucleocapsid (NC) 12. The pol gene products are the protease (PR), the reverse-transcriptase (RT), the integrase (IN) and, in some cases, a dUTPase.

Finally, the precursor synthesized from the env gene is cleaved into the surface envelope protein (SU) and the transmembrane envelope protein (TM) 12.

Once processed from their precursors the viral proteins form the mature virion, which is able to infect susceptible cells that express the appropriate receptors. The viral core of a mature viral particle consists in the diploid RNA genome that interacts with the NC, creating a condensation, surrounded by the CA protein complex. The matrix protein that covers this core is surrounded on top by a host- derived lipid bilayer and the included SU and TM proteins 12

The viral core contains as well the pol-derived proteins that will be used for a novel round of replication, namely the PR, the RT and the IN 12.

!"# !"#

Figure 2: Schematic view of the proviral genome structure of retroviruses. The retrovirus proviral DNA is composed of untranslated regions that flank the ORFs for gag, pro, pol, env and in some cases that of accessory genes. The flanking LTRs contain U3 and U5 regions, as well as a repeat sequence (R). The 5’ region of the retroviral genome is followed by a PBS and a psi encapsidation signal. Adjacent to the last ORF, the viral RNA contain a PPT. ORF: open-reading frame; LTR: log terminal repeats; U3 and U5: unique regions 3 and 5, respectively; att: attachemetn site; PBS: primer binding site; PPT: poly-purine tract. Adapted from Fouty and Solodushko, 2011 23 .

1.1.4 The reverse-transcription process. Once the retroviral genome enters the cell, the diploid single-stranded genome that is still bound to the nucleocapsid (NC) protein, constituting the viral core, starts the process of reverse transcription 24,25.

For reverse transcription to take place, important elements contained in the viral particles are required. The central component is the reverse transcriptase enzyme, which catalyzes four different reactions: RNA-dependent and DNA- dependent DNA polymerization, DNA strand separation via its helicase function and the hydrolysis of the RNA fragments on RNA-DNA heteroduplexes 26. The viral core carries additionally a specific collection of transfer RNA (tRNA)

15 molecules, different cellular messenger RNAs (mRNAs) from previously infected cells and some ribosomal RNA (5S and 7S) 26.

Reverse transcription starts when the 3’ region of a specific tRNA is used as a primer that anneals with the PBS within the 5’ region of the viral RNA genome (figure 3). DNA synthesis continues until the 5’ extremity of the RNA strain is reached, resulting in a short DNA strand called the minus strand strong stop DNA (–ssDNA) 27.

The next step takes the advantage that the minus-strand DNA contains a repeat (R) sequence that is present at both viral genome termini and that was introduced in the newly synthesized DNA molecule by the reverse transcription of the 5’ region of the viral RNA. This confers a complementarity of the –sDNA and the 3’ end of the RNA genome that allows the transfer of the small oligonucleotide to that region, after that the RNAse H function of the RT has degraded the RNA to which the newly synthesized DNA is annealed. This marks the beginning of the elongation of the –sDNA chain, with an accompanying RNA degradation accomplished by RNAse H 27.

During the RNA dependent-DNA synthesis, the ppt permit the RNA to escape degradation and this RNA fragment is then used as a primer for the plus-strand DNA (+sDNA) polymerization that finally reaches the U5 region of the –sDNA. In the mean time, the –sDNA continues to be polymerized, with a subsequent gradual RNA degradation.

In the following step, the +sDNA synthesis proceeds until the level of the PBS complementarity is formed and the RNA and tRNA primers are degraded. When the tRNA is removed from the +sDNA a complementarity region is exposed and the second strand transfer happens where the plus and minus strands anneal. The resulting molecule is a circular DNA intermediate 27.

This point of the viral replication cycle can lead to a non-productive dead-end DNA molecule which contains a single LTR or to a productive DNA form flanked

16 HIV-1 Reverse Transcription

by two cleavage LTRs is, detected resulting while RT is from actively synthe- the strand genome; displacement the ﬁrst (or minus-strand) of the transfer plus and minus sizing DNA; instead, cleavages occur at sites can involve the R sequence at the 30 ends of where DNA synthesis pauses (Driscoll et al. either of the two RNAs (Panganiban and Fiore 27 strands and resulting in the DNA synthesis to2001; Purohit et al. 2007). Whatever the exact 1988; Hu andwards the PBS and the ppt Temin 1990b; van Wamel and . mechanism, RNase H degradation removes the Berkhout 1998; Yu et al. 1998). After this trans- 50 end of the viral RNA, exposing the newly syn- fer, minus-strand synthesis can continue along thesized minus-strand DNA (see Fig. 1). the length of the genome. As DNA synthesis The ends of the viral RNA are direct repeats, proceeds, so does RNase H degradation. How- called R. These repeats act as a bridge that allows ever, there is a purine-rich sequence in the the newly synthesized minus-strand DNA to be RNA genome, called the polypurine tract, or transferred to the 30 end of the viral RNA. Retro- ppt, that is resistant to RNase H cleavage and viruses package two copies of the viral RNA serves as the primer for the initiation of the

A R U5 pbs gag pol env ppt U3 R

B R U5 pbs gag pol env ppt U3 R

C pbs gag pol env ppt U3 R

D pbs gag pol env ppt U3 R U5

E pbs gag pol env ppt U3 R U5

F U3 R U5 pbs gag pol env ppt U3 R U5 r A www.perspectivesinmedicine.org

G U3 R U5 pbs gag pol env ppt U3 R U5 r A LTR LTR

Figure 1. Conversion of the single-stranded RNA genome of a retrovirus into double-stranded DNA. (A) The RNA genome of a retrovirus (light blue) with a tRNA primer base paired near the 50 end. (B) RT has initiated reverse transcription, generating minus-strand DNA (dark blue), and the RNase H activity of RT has degraded the RNA template (dashed line). (C) Minus-strand transfer has occurred between the R sequences at both Figure 3ends: The of the reverse genome- (seetranscription text), allowing minus-strand process. DNAThe synthesis viral to RNA continue is converted (D), accompanied to DNA by by the reverse- transcriptase enzyme. The first step is the binding of the tRNA primer to the PBS. Subsequent synthesis of RNA degradation. A purine-rich sequence (ppt), adjacent to U3, is resistant to RNase H cleavage and serves as the primer for the synthesis of plus-strand DNA (E). Plus-strand synthesis continues until the the short ﬁrst minus 18 nucleotides–sDNA of and the tRNA its arefurther copied, annealing allowing RNase with H cleavage the 3’ to remove LTR of the the tRNA viral primer. genome Most will initiate the synthesis of the retroviruses–sDNA. When the remove the entire tRNA;–sDNA reaches the PBS, the +sDNA starts to be synthesized. The resulting the RNase H of HIV-1 RT leaves the rA from the 30 end of the tRNA molecule is a dsDNA molecule with two flanking LTRs and that can be inserted into the host genome. attached to minus-strand DNA. Removal of the tRNA primer sets the stage for the second (plus-strand) transfer - (F); extension of the plus and minus strands leads to the synthesis of the complete double-stranded linear viral 28 sDNA: minusDNA- (strand DNA; +sDNA: plus strand DNAG). . From Hu and Hugues, 2012 .

Cite this article as Cold Spring Harb Perspect Med 2012;2:a006882 3 1.1.5 The classification of retroviruses

Retroviruses are members of the Retroviridae family. Depending on the morphology of the particles, their structure and their genomic sequences, the Retroviridae family can be divided into seven genera, further regrouped on the basis of their complexity.

The group-specific antigen (gag), protease (pro), pol and envelope (env) gene products are encoded by all genera of Retroviridae. Complex retroviruses carry in addition accessory genes with different regulatory functions. The simple

17 retroviruses consist of the Alpharetroviruses, Betaretroviruses and Gammaretroviruses 12.

Alpharetroviruses infects a large range of birds. They assemble at the cell membrane and possess a central spherical core (C-type morphology). The tRNA they use for the priming of reverse transcription is the one for tryptophan (tRNATrp). A typically well-studied member of this genius is the Avian Leukosis Virus (ALV) and the previously mentioned RSV 12.

Members of the Betaretroviruses infect different mammalian species including mice and primates. Morphologically, they can have either an asymmetric round core, either a cylindrical one. They contain a dUTPase gene in frame with the pro gene and they use the tRNALys. The oncovirus Mouse Mammary Tumor Virus (MMTV) is a member of this family 12.

Gammaretroviruses possess C-type virion morphology. They have two ORFs. The first one encodes the gag, pro and pol gene products; the second one encodes the envelope proteins. The tRNAs used by these retroviruses are mainly the ones for proline or glutamine. Highly documented oncogenic members of this genius include the Murine Leukemia Virus (MLV), Feline Leukemia Virus (FLV) and Gibbon Ape Leukemia Virus (GALV) 12.

The group of complex retroviruses is composed of Deltaretroviruses, Epsilonretroviruses, Lentiviruses and Spumaviruses. Deltaretroviruses and Epsilonretroviruses have a similar C-type virion morphology. The first genius is composed of members encoding two accessory proteins named rex and tax, which are involved in the synthesis and processing of viral RNA. It uses the tRNAPro. An example of this group is the oncovirus Human T-Lymphotropic Virus 1 (HTLV-1) and the closely related HTLV-2. The second genius is uses the tRNA for histidine or arginine and codes additionally for three proteins called ORFA, B and C respectively. The function of these accessory proteins are not well understood but in the case of the better-studied member Walley Dermal Sarcoma Virus (WDSV), ORFA has been shown to be an orthologue of

18 mammalian cyclin c, ORFB activates the PKC and AKT signaling and ORFC has oncolytic properties 29.

The AIDS-causing HIV-1 belongs to the genius of lentiviruses and is characterized by a conical shape of the core of the mature virion. Members of this group carry this name because of the long asymptomatic phase preceeding the first symptoms 12. HIV-1 expresses six accessory proteins that will be discussed below. These gene products control transcription, gene expression and assembly and counteract restriction factors encoded by the host 12. The primer used by lentiviruses is the tRNALys3.

In latin Spuma means foam. The members of the Spumaviruses produce vacuolization of cells, hence resulting in a foamy-like histological aspect. The human foamy virus is a well-studied member of this group. The pol gene products arise from a splice transcript. Unlike other retroviruses, this genius of viruses is characterized by virions that carry high amounts of reverse- transcribed DNA. Accessory proteins shared by the members of this group include a transcriptional transactivator. The primer used is generally tRNALys 12.

1.1.6 The Acquired Immunodeficiency Syndrome (AIDS)

The AIDS is a severe disease affecting more than 35 millions of people around the world, as published by the UNAIDS report on the global AIDS epidemics 2013 30.

In the early 1980s, young men with typical immunodeficiency symptoms were hospitalized in Los Angeles, New York and California 31,32.

As mentioned previously, biochemical and genetic tools for studying retroviruses existed in that decade and they were used by Researchers at the Institut Pasteur and in the United States to characterize the virus extracted from CD4+ T cells coming from AIDS patients. Barré-Sinoussi and colleagues isolated and described a virus that was able experimentally to infect T lymphocytes

19 extracted from cord blood 22 and called it Lymphoadenopathy Associated Virus (LAV).

The team of Robert Gallo, had suspected that the causing agent of AIDS was of retroviral origin and possessed T-cell tropism but at that time attributed it to the human tumor retrovirus HTLV-I 33. The virus was later called HTLV-III by the same team. In 1986, the virus was finally named the Human Immunodeficiency Virus (HIV), in reference to the disease it produced 34.

Transmission of HIV-1 from one person to another happens during sexual intercourse, injecting with contaminated needles, or by blood transfusion 35. Mother to child transmission during delivery or after breast-feeding is another important route of spreading 35. The first events of HIV-1 infection seem to implicate a local spreading within cells residing in the mucosa and in the epithelium, such as dendritic cells (DCs), CD4+ T cells and macrophages 36-38. Primary infected cells subsequently migrate to the lymphoid organs and seed the virus by direct cell-to-cell contact or by the release of newly produced cell-free viruses, which enter new cells 39.

When HIV-1 gp120/gp41 glycoproteins interact with the lectin receptor DC- SIGN at the surface of DCs, the virus can be either endocytosed and degraded within lysosomes or by targeting to the proteasome 40,41. Another route for entry into DCs is mediated by a host-derived glycosphingolipid present in the virion envelope that binds to an unknown receptor, with SIGLEC-1 being a potential candidate 42. This interaction allows the virus to escape degradation and join immunological synapses, from where new target CD4+ T cells can be reached 12,43,44.

During the acute phase of infection, a large fraction of CD4+ T cells are infected and high amounts of virions are synthesized and released from cells 39. As CD8+ T cells fight against the pathogen and high doses of type I interferon (IFN) and cytokines are released, infected individuals commonly experience flu-like

20 symptoms 45-47. The immune response mediated by cytotoxic T cells and B cells producing antibodies permits to moderately recover the level CD4+ T cells for a few weeks 47. At that point, HIV-1 already integrated into the host chromosomes and latent reservoirs starts to be established, and infected individuals can have a total absence of HIV-1-related symptoms for nearly ten years 47. Unfortunately, in the meantime, the virus continues to replicate and spread via the various lymphoid organs.

At the terminal stage, the disease causes a high destruction of the CD4+ T cells, which decrease below 200 cells per mm3 of blood, leading to immune suppression and the subsequent unavoidable infection by opportunistic pathogens as Candida albicans and Pneumocystis jirovesii 48.

The AIDS pandemic is likely to have originated in central Africa as a result of cross-species transmission of a chimpanzee lentivirus to humans. Studies of sequence homology between SIVcpz and HIV-1 have shown that the human lentivirus is derived from the simian one 49. The second type of HIV, named HIV- 2, is less pathogenic and transmissible and thus less frequently leads to AIDS. Although the two viruses have a similar genome organization, they are derived from different SIV strains 49. Whereas HIV-1 comes from the SIVcpz, HIV-2 arose from a zoonosis with the sootey mangabey monkey, Cercocebus atys. Instead of the Vpu accessory protein, HIV-2 possesses Vpx, which counteracts a block to reverse transcription within DCs and macrophages 50. Additionally, HIV-1 and HIV-2 highly diverge from their env sequence. In fact, it was observed that there is already 25% of divergence of the gag, pol and env sequences within the strains of each type, as reviewed by Reeves and Doms 51.

A combination of nucleoside or non-nucleoside reverse-transciptases inhibitors and protease inhibitors constitute an aggressive therapy for maintaining the virus load at a low level 52,53. The highly active antiretroviral therapy (HAART) allowed the life expectancy of individuals to reach nearly normal life spans 54. Thanks to these combined anti-retroviral therapies and efforts employed in prevention education, new infections have diminished of near 30% compared to

21 2001 (UNAIDS report, 2013). However, the pathogen is still far from being eradicated as HIV-1 has rarely been totally cleared from an individual 55,56 and a vaccine is still missing.

The reasons why the search for an effective vaccine has been unsuccessful until now could be in part the inability of the immune system to detect a dormant virus and inherent to the tropism of the virus that targets to destruction the immune cells themselves 44. Another important point that could explain the failure of the immune system to detect HIV-1 and mount a robust response is that this virus does not productively infect the DCs that are antigen-presenting cells (APCs), that prime the immune effectors to kill infected cells. Yet the antigen from these cells is being presented. It seems likely that the APC needs to be activated for the priming of effector T cells to be efficient, an unproductive infection leading to no immune activation will fail to fulfill this prerequisite 44.

The innate immunity actors and consequences of their activation will be introduced further below.

1.1.7 The structures of the HIV-1 virion and genome

The HIV-1 gag orf codes for a precursor polyprotein of 55 kDa in size, called Pr55gag, which is cleaved within the virion into the MA, the CA, the NC proteins and p6 that is involved in viral budding 57. HIV-1 membrane form a spherical particle that has a diameter of approximately 110 nanometers (figure 4). The virion contains a conical-shaped CA protein complex that is composed of 216 hexamers and 12 pentamers, linked between them by the C-terminal domains of CA 58,59(figure 4). The viral particle core is enclosed by a layer of MA proteins, in turn surrounded by a lipid bilayer coming from previous infection events.

The MA protein form hexameric higher-order complexes, which encapsulate the viral core. These complexes interact with different virion components and seem

22 to be essential for various processes. The well demonstrated bindings include the interaction with envelope bilayer through myristoylated motifs, a process essential for virion assembly at the plasma membrane 57. The MA interacts as well with phosphatidiylinositol-4,5-bisphosphate, leading to the targeting of the myristoyl tails to the plasma membrane and helping the MA to bind to the viral genome 60,61. Although subject to controversy, MA was reported to bind to the inner domain of gp41, stabilizing the interaction of the envelope into the assembling virion 57. Additionally, the Pr55gag-derived protein was shown to interact with the reverse-transcription and pre-integration complexes (RTC and PIC, respectively), suggesting a role of the MA complex in the early viral life-cycle steps.

The CA protein is divided into two structural domains. The N-terminal domain stabilizes the structure of the virion and is in the outer layer of the viral core. The C-terminal domain faces the inner space containing the genome and contributes to link the hexameric and pentameric rings together. The hexameric lattice of CA interacts with different cellular factors. Its binding to the restriction factor TRIM5 inhibits both reverse-transcription and further pre-integration steps (see below). As it will be further discussed later, the CA interacts with endogenous Cyclophilin A. When Cyclophilin A is part of a TRIM5 orthologue protein, HIV-1 is bound and strongly restricted 62. The restriction mediated by the TRIM5 protein and orthologues will be discussed in more detail in the next sections.

23 R EPORTS tecture do incorporate such machinery and thus Zvolenszky for helpful discussion, and P. Marcus and F. system for studying viral core structure and might form the basis for learning mechanisms that Scherer for their assistance in construction of the test assembly in vitro. The assembly properties of could account for our data [ J. E. Hummel and K. J. apparatus. We also thank Bell Labs for making available Holyoak, Psychol. Rev. 104, 427 (1997)]. Our goal is to the public the speech synthesizer that we used to pure recombinant HIV-1 Gag protein frag- not to deny the importance of neural networks but create our stimuli. Some subjects in experiment 1 were ments have been investigated in several lab- rather to try to characterize what properties the right tested at Amherst College; all other subjects were test- oratories (4–7). Pioneering work by Camp- sort of neural network architecture must have. ed at New York University. The parents of all partici- 25. Supported in part by an Amherst College Faculty Re- pants gave informed consent. bell and Vogt demonstrated that fragments of search Grant to P.M.V. We thank L. Bonatti, M. Brent, S. HIV-1 and Rous sarcoma virus (RSV) Gag Carey, J. Dalalakis, P. Gordon, B. Partee, V. Valian, and Z. 11 September 1998; accepted 16 November 1998 proteins that encompass the CA and NC domains can assemble into long hollow cylinders in the presence of RNA (5). Building on this, we screened for conditions that would Assembly and Analysis of support the assembly of conical (rather than cylindrical) structures. Initially, we employed Conical Models for the HIV-1 an HIV-1 RNA template that spanned sites required for genomic RNA packaging (⌿) Core and dimerization (DLS), because some models for the viral core have suggested that the Barbie K. Ganser,* Su Li,* Victor Y. Klishko, John T. Finch, genomic RNA dimer dictates the cone mor- Wesley I. Sundquist† phology (8). The protein construct included both the CA and NC domains of HIV-1 Gag, The genome of the human immunodeﬁciency virus (HIV) is packaged within an because viral core morphology can be dis- unusual conical core particle located at the center of the infectious virion. The rupted by mutations in either of these do- core is composed of a complex of the NC (nucleocapsid) protein and genomic mains, or in the short spacer peptide that RNA, surrounded by a shell of the CA (capsid) protein. A method was developed connects them (3). Finally, solution assembly for assembling cones in vitro using pure recombinant HIV-1 CA-NC fusion conditions were varied, because cylinder for- proteins and RNA templates. These synthetic cores are capped at both ends and mation is sensitive to protein and RNA con- appear similar in size and morphology to authentic viral cores. It is proposed centrations, salt, and pH (4–7). Cone forma- that both viral and synthetic cores are organized on conical hexagonal lattices, tion was assayed by transmission electron which by Euler’s theorem requires quantization of their cone angles. Electron microscopy (TEM) of negatively stained microscopic analyses revealed that the cone angles of synthetic cores were samples. indeed quantized into the ﬁve allowed angles. The viral core and most synthetic A mixture of cones (Fig. 1A, arrows) and cones exhibited cone angles of approximately 19 degrees (the narrowest of the cylinders formed spontaneously upon incuba- allowed angles). These observations suggest that the core of HIV is organized tion of a pure recombinant CA-NC fusion on the principles of a fullerene cone, in analogy to structures recently observed protein with a purified 1400-nucleotide (nt) for elemental carbon. HIV-1 RNA template in 500 mM NaCl (pH 8.0) (9). Cone:cylinder ratios as high as ϳ2:3 HIV-1 assembly is initially driven by poly- the core is composed of an RNA/NC copol- were observed under these optimized condi- merization of the Gag polyprotein, which ymer, and is surrounded by an outer shell tions. Cones also formed under physiological forms a spherical shell associated with the composed of ϳ1500 copies of CA. The con- conditions [that is, 150 mM NaCl (pH 7.2)], inner membrane of the budding particle. The ical core appears to be essential, because Gag albeit at reduced efficiencies. The synthetic three major regions of Gag all perform essen- mutations that disrupt proper core formation cones were capped at both ends, and many

tial roles in viral assembly: the NH2-terminal invariably inhibit viral infectivity (3). The appeared strikingly similar to authentic MA (matrix) region binds the membrane, the core probably organizes the viral RNA ge- HIV-1 cores (Fig. 1B). Cones formed in vitro central CA (capsid) region mediates impor- nome (and its associated enzymes) for un- varied between 100 and 300 nm in length. tant Gag-Gag interactions, and the COOH- coating and replication in the new host cell, Viral cores are typically ϳ100 nm long (8, terminal NC (nucleocapsid) region packages although these processes are not yet well 10); however, this can also vary considerably Journal of the American Chemical Society Article the viral RNA genome [reviewed in!" (1, 2)]. understood. because HIV-1 virions#" range between ϳ120 As the particle assembles, the viral protease Our initial goal was to develop a model to 260 nm in diameterHere we (11 explore). These the similarities conformational space sampled by the cleaves Gag, producing discrete MA, CA, monomeric and dimeric species of the wild-type, full-length and NC proteins, which subsequently rear- Fig. 1. CA-NC/RNA complexes HIV-1 capsid protein, CA (Figure 2), using experimental range to form the mature, infectious viral spontaneously assemble into FL NMR residual dipolar couplings (RDCs) and small- and wide- particle. During maturation, MA remains as- cones in vitro. (A) TEM image of angle solution X-ray scattering (SAXS/WAXS) data in an sociated with the inner viral membrane, while a representative field of nega- CA and NC condense about the viral RNA to tively stained particles formed ensemble simulated annealing protocol, supplemented by NMR by the CA-NC protein on a form an unusual conical structure at the cen- relaxation measurements and analytical ultracentrifugation. 1400-nt HIV-1 RNA template. Methodology was developed to treat the simultaneous ter of the virus (the “core”). The interior of Conical structures are denoted by arrows. Scale bars in Figs. 1, 2, determination of monomer and dimer ensemble structures along with optimal ensemble weights. B. K. Ganser, S. Li, V. Y. Klishko, W. I. Sundquist, and 4 are 100 nm. (B) Selected Department of Biochemistry, University of Utah, Salt thin-sectioned TEM images of an To date, attempts to study the full-length capsid protein by Lake City, UT 84132, USA. J. T. Finch, Structural authentic HIV-1 virion grown in conventional solution NMR methods have been hampered by Studies Division, UK. Medical Research Council Labo- culture (bottom) and a synthetic severe resonance line-broadening of the backbone resonances ratory of Molecular Biology, Hills Road, Cambridge, CA-NC/RNA cone assembled in of the linker residues, as well as of residues at the dimer CB2 2QH, UK. vitro (top). Electron microscope preparations of virions and synthetic cores were identical, andinterface the two as a consequence of a dynamic monomer/dimer *These authors contributed equally to this work. exchange. Although such localized line broadening is an †To whom correspondence should be addressed. E- objects are shown at the same magnification. mail: [email protected] impediment for traditional NMR structure determination, it can be circumvented, providing that a limited number of RDCs can be measured within each domain of the full-length capsid 80 1 JANUARY 1999 VOL 283 SCIENCE www.sciencemag.org protein and the structures of the individual domains in the full- length capsid and the isolated domain constructs are the same. Under these conditions, the individual domains of a multidomain protein/macromolecular assembly can be treated as Figure 1. HIV-1 capsid assembly. The capsid protein (top right) rigid bodies for ensemble simulated annealing calculations in comprises N- (green) and C-terminal (red) domains.10 During capsid 7 which RDCs arising from steric alignment provide both shape assembly, the N-terminal domains form pentameric (middle right) 18,19 11 and orientational information, while the SAXS/WAXS data and hexameric (bottom right) rings (with the N-terminal domains 20,21 shown in blue and green, respectively, and the C-terminal domains provide complementary restraints on size and shape. This shown in red in both oligomers). A model of the fully assembled hybrid approach is much less time-consuming than conven- Figure 4: Architecture of the HIV-1 core. A) The HIV-1 capsid is composed of an hexmeric lattice (green) 20 capsid (left) comprises adjacent hexamers connected to each other via tional methods of NMR structure determination and can be joined by pentameric C-terminal domain units dimers, (blue). and exactly The N 12 terminal pentamers are domain required (NTD) to and readily the transferred C terminal to other domain multidomain (CTD) proteins. are depicted in close green the cone. and 7 red, respectively. B) Pictures of the transmission electromiscroscopy (TEM) of negative stained-samples. The upper panel shows a viral particle from an ■ MATERIALSin vivo extract. The bottom pannel AND METHODS shows a in vitro incorporationassembled virion. The scale bar repr of pentamers in the mature capsid, whichesents 100 nm. A: From Deshmukh et al., 2013 in turn 59. B: Protein Expression and Purification. All full-length HIV-1 From Ganser et al., 1998 dictates the lattice58. curvature. Despite limited evidence from capsid constructs, the wild-type (CAFL, residues 1−231, plasmid previous solution NMR studies that the N- and C-terminal fi V181C pNL4-3), the disul de-linked mutant (CAFL ), and the monomeric domains do not appear to interact with one another in the full- W184/M185A mutant (CAFL ), as well as the four C-terminal domain length capsid protein,17 the overall magnitude and time scale of fi constructs, CA144−231 and CA146−231 and the corresponding disul de- V181C V181C the relative motions between the domains are unknown, and linked-linked mutants (CA144−231 and CA146−231) (see Figure 2), were The NC protein is found within the gag polyprotein or as a mature form. During the conformational space sampled by the domains has not yet subcloned in a pET-11a vector and expressed in BL21-CodonPlus been characterized. (DE3)-RIPL competent cells (Agilent Technologies). Point mutations viral assembly and budding, the NC domain is required for proper gag-to-gag contacts, and participates in the recruitement of viral RNA to the forming virion.

Within the mature virion, NC binding to the tRNALys3 facilitates the annealing of the primer to the viral genomic RNA and is essential for the reverse- transcription process to initiate 63.

The p6 gag product is a small protein containing two proline-rich motives that are responsible for recruiting the endosomal sorting complex required for transport (ESCRT) 64, promoting in this way the viral budding. Figure 2. Summary of capsid constructs used in the current work. The delineation of the N- and C-terminal domains and the location of point mutations in the various constructs are shown. The dimerization states of the constructs under the experimental conditions used for NMR and SAXS/WAXS measurements are also indicated.

16134 dx.doi.org/10.1021/ja406246z | J. Am. Chem. Soc. 2013, 135, 16133−16147

24 The gag-pol polyprotein is cleaved into the PR, RT and IN. The PR is essential to mediate the processing of the gag and gag-pol precursors. The catalytic site of PR contains an aspartic acid (Asp) that is shared by a whole family of proteases. The enzyme active form starts to appear after the cleavage at the N terminus of the PR within the gag-pol and reaches a maximum level after the cleavage at the C terminus, a process that happens under acidic conditions. Indeed, artificially extending the N terminal domain of PR renders the virions unable to infect 65.

Upon maturation, the RT is found in the form of a heterodimer, composed of the two subunits p66 and p51 of 560 and 440 amino acids, respectively. The former subunit carries the catalytic domains of the polymerase and the RNase H activities 28.

The polymerase catalyzes the initiation of reverse-transcription and the elongation of the resulting transcripts, whereas the endonuclease mediates the gradual and subsequent degradation of the viral RNA templates. Recently, the crystal structures of the RT complexed with RNA-DNA and DNA-DNA duplexes has been solved and allows the understanding of how RT-inhibitors work 66,67.

The mature form of HIV-1 IN is a dimer that is composed of two catalytic core domains (CCD), via an interaction between the N-terminal domain (NTD) and the C-terminal domain (CTD). The active site of the CCD carries Asp and glutamic acid (Glu) that are both electronegative residues. Upon interaction with viral cDNA, two IN dimers come in contact, forming a tetramer 68,69. The Pre- integration complex (PIC) that contains IN, binds to nuclear pore proteins as the Nup153, Nup160, RANBP2 and TNPO3, which allows it to be imported into the nucleus 70-74. Once there, the PIC targets the host genome via the interaction of IN with LEDGF/p75 that possess a chromatin-binding domain 75.

The lipid host-derived membrane contains bound env glycoprotein gp41 via non-covalent interactions. The TM protein gp41 interacts further with the other env-derived glycoprotein, namely gp120, which is in this way connected to the virion membrane and yet exposed to the surface (SU protein), as reviewed by

25 Wilen and colleagues 76. Heterodimers of gp41 and gp120 form further trimers that are the working module for viral fusion 77. Gp120 interacts with the specific CD4 receptor strongly expressed on the CD4+ T cells via mainly one of its five variable loops (VL1-5), VL3. This binding operates a conformational change that allows gp120 to bind one of two co-receptors depending on the cell type. CCR5 co-receptor is expressed on memory CD4+ T cells, Dendritic Cells (DCs) and macrophages. CXCR4 co-receptor is expressed on naïve and memory CD4+ T cells. The aforementioned conformational change allows additionally the gp41 monomer to expose the hydrophobic domain called the fusion peptide, representing the first step of the fusion between viral and host membranes 76.

The HIV-1 genome additionally codes for regulatory and accessory proteins. As reviewed by Karn and Stoltzfus 78, the two regulatory proteins are tat and rev that are required for proper HIV-1 provirus transcription and export of the resulting mRNAs from the nucleus to the cytoplasm, respectively.

Additional accessory proteins of HIV-1 are dispensable for the viral replication but assist in the evasion from innate and adaptive immunity. These are vif, vpr, vpu and nef. Vif and vpr both interact with cellular targets and cullin complexes to direct host defense factors to the proteasome 79. Vpu and Nef regulate the abundance of the cell surface molecules as CD4. Nef additionally targets other molecules expressed at the plasma membrane, as the CD3 receptor and Major Histocompatibility Complex (MHC) Class I molecules 79.

The structure of the HIV-1 genome is similar to that of other retroviruses in that it consists of two flanking LTRs with the previously mentioned conserved elements and a central coding region (figure 5). HIV-1 expresses one primary transcript that is either translated into a gag precursor (Pr55gag), a gag-pol precursor, or single/multiply spliced 78. Depending on the splice site used, the single splicing will result in the production of vif, vpr or vpu-env transcripts. Two different double-splicing events will result in the generation of the tat transcript and that of the bicistronic rev and nef 78.

26 !"

Figure 5: Structures of the HIV-1 genome and virion. A) The HIV-1 genome is composed of two flanking LTRs containing typcial retroviral regulatory sequences, and different ORFs coding for gag-pol, env, the regulatory proteins rev and tat, and the accessory proteins vif, vpr, vpu and nef. The rev-responsive element (RRE, light blue box) is depicted below the env ORF (green box). B) The structure of the HIV-1 mature virion is depicted with the cleaved viral products, which are indicated with arrows. A) and B) from Sakuma et al, 2012 80.

1.1.8 The HIV-1 life cycle

The first step of the HIV-1 replication cycle is the viral particle fusion with the plasma membrane of a susceptible cell that expresses specific receptors. Upon the binding of gp120 to the CD4 receptor of an immune cell such as a lymphocyte, it further interacts with a specific seven transmembrane domain G protein-coupled coreceptor depending on the virus tropism. Generally, CCR5- tropic virus dominate the first phase of the infection, and gradually the tropism of some strains can change to reach up to 50% of viral particles that use the alternative CXCR4 co-receptor 81,82. Additionally, some strains of HIV-1 are able to use other coreceptors as CCR2b, CCR3, CCR8 and the orphan receptors V28, STRL33 and GPR15 83.

27 Once the virus fuses with the membrane, the viral core is released into the cytoplasm, where reverse transcription takes place. It is still a matter of debate where the capsid uncoating may occur, before reverse-transcription, concomitantly or after completion of the process. Three major models are postulated, as reviewed in 84. One of them put forward the hypothesis that the total uncoating is necessary to activate the reverse-transcription complex (RTC). This first hypothesis was based on the observations that low-levels of capsids were recovered from HIV-1 complexes extracted from cells and on the inability to visualize the capsid complexes with Transmission Electron Microscopy (TEM) within infected cells 85-87. These findings were not able to discriminate between partial and total uncoating.

Based on the results from a study showing that CA associates with the RTC, two other schools of thoughts emerged 88. The first alternative model implies that the capsid uncoating happens gradually, upon environmental changes encountered by the viral core, like the interaction with different cellular factors at different steps of the reverse-transcription process and the production of cDNA and reverse-transcription intermediates. This hypothesis is reinforced from one side, by the observation of different sizes of HIV-1 cores isolated from early infected cells and that diverge from the mature HIV-1 complexes 88,89. From another side, the findings that altering the stability of the capsid core via mutagenesis of specific CA residues to either an increase or a decrease will affect the completion of the reverse-transcription are in agreement with this model 90. Whereas the first observation could simply represent an artifact coming from the biochemical method used for the virus isolation, the second finding uses mutations that could induce conformational changes that modify the interactions with host factors and the RTC in in vivo experiments that are not relevant in the course of a natural infection. Nevertheless, TRIM5, a factor that will be described later, blocks retroviruses at an early infection step, via accelerated CA core disassembly. This observation reinforces the model in which a tightly controlled uncoating is necessary for an efficient viral replication.

28 The second alternative model postulates that the capsid remains associated with the RTC until completion of the reverse-transcription at the nuclear pore. The finding that the capsid is required for proper nuclear import 84,91 argues in favor of this model, but does not contradict the postulate where uncoating is a gradual event. Alternatively, the uncoating could possibly not occur, as shown by the experiments done by Burdick and colleagues 92.

Nevertheless, the CA uncoating of at least a portion of particles seems to happen very early after viral entry, as shown by experiments using the CA-specific blockade of HIV-1 by TRIM5Cyp, which will be described later. TRIM5Cyp- mediated restriction is mediated by recognition of intact cores and this binding is impeded by Cyclosporine A (CsA), as refered below. Taking out the CsA treatment (CsA washout assay) after 2 hours precludes definitively restriction, indicating that the CA stability was disrupted before this time point 93. Similarly, immunofluorescence microscopy allowed detecting less particles associated with capsid after 1 hour. In the same study, the chemical inhibition of the RT, using nevirapine (NVP), in combination with the CsA washout assay showed that the reverse-transcription progression is necessary for normal uncoating, as this process did not become apparent when the inbitory drug was used 93. The restriction factor TRIM5, recognizes the capsid hexameric lattice. It is therefore possible that even if only partial uncoating happen, TRIM5Cyp wouldn’t bind to the particle anymore. Further evidence that uncoating is a rapidly started event comes from a study where another set of restriction factors, the APOBEC3 proteins, act on the nascent cDNAs, in the target cell 94. This finding suggests that the RTC is accessible very early within the viral core to proteins present in the cytoplasm, before or at the time of reverse-transcription 95.

Concomitantly, there are several results suggesting that CA remains associated until a late step of retroviral replication. First, experiments show that CA is the determinant for nuclear import 71,96,97. Second, CA interacts with Cyclophilin A (CypA) and this interaction is required for proper reverse-transcription but blocks nuclear-entry in some cell lines 98. Third, a study revealed that CA binds to

29 the CypA domain of the Nucleoporin protein of 358 kDa (Nup358) at the nuclear pore complex. Interaction with this protein allows nuclear entry of the PIC 99. When Nup358 or another member of the complex involved in this nuclear internalization pathway - TNPO3 or cytoplasmic CypA- are disrupted, other routes are used and result in different preferential integration sites and concomitant impaired HIV-1 replication, as reviewed by Fassati 95. Fourth, CA binds to nuclear export factors, suggesting it can localize to the nucleus 100. Fifth, CA total uncoating is not necessary for reverse-transcription to proceed, as a mutant that stabilizes the CA core still synthesizes normal levels of cDNA 93.

The observation that different natural or artificial restriction factors recognizing the CA block different pre-integration steps, including post-nuclear entry, further argues in favor of a partial and gradual uncoating that culminates into the nucleus at some step before integration 101,102 103. Despite a considerable body of work devoted to the HIV-1 uncoating process, the subcellular compartment and the viral replication step where its completion occurs remain a mystery. What is clear from previous studies is that a proper timing of CA uncoating is necessary for various steps of the viral life cycle to proceed, as shown by the CA-dependent negative effect on HIV-1 replication by restriction factors that accelerate the uncoating (see below) and by proteins that are influencing nuclear entry.

The process of the nuclear import of the PIC seems to rely on the capsid protein. In earlier studies, IN, MA and Vpr were suggested to be required for nuclear entry and the fact that these proteins carry a nuclear localization signal (NLS) supported this theory. To understand the key experiments that were performed to investigate the viral proteins involved in nuclear import of the PIC, one must consider that there is a major difference between lentiviruses and other retroviruses like the gammaretrovirus MLV, in respect to nuclear entry. As such, whereas HIV-1 can enter the nucleus of non-dividing cells, MLV is dependent on the breakdown of the nuclear membrane at the mitosis to import its PIC 104-106. This feature in fact allows the gammaretrovirus to uncoat after nuclear entry 107, perhaps conferring a protection of the RTC and the PIC from cytoplasmic sensors. Taking advantage of this difference, the IN or the CA

30 proteins of both retroviruses were exchanged, expecting that they could confer a differential ability to enter the nucleus.

When adding a NLS into the MLV MA or IN proteins, MLV could still not be imported into the nucleus of resting cells 108,109. Interestingly, however, the exchange of the MLV capsid by the one from HIV-1, transferred to the gammaretrovirus the independency from cell-cycle requirements for nuclear entry 91. Reciprocally, HIV-1 with an MLV capsid lost its ability to infect non- dividing cells.

To import the PIC into the nucleus, nuclear pore complex (NPC) proteins as Nup98, Nup153 and Nup358 and other cellular factors seems to be required, as reviewed by Fassati 95. Before nuclear entry, the PIC machinery already activates the viral DNA to be integrated 69. IN binds to both viral LTRs, recognizing internal specific sequences and catalyzes the processing at a CA dinucleotide, producing an available 3’ hydroxyl group, which constitute a cleaved donor complex (CDC) that is competent for nucleophilic attack of the target DNA once in the nucleus. This interaction leads to the DNA strand transfer in whom the 3’ ends of the viral cDNA are ligated to 5’ phosphates into the host chromatin, as reviewed by Krishnan and Engelman 69. Once the strand transfer complex is formed, repair enzymes from the host take care of joining and filling the gaps created at the 5’ ends of viral cDNA within the targeted host genome, creating the so called target site duplication (TSD) at both ends of the provirus 69.

As reviewed by Karn and Stolzfus 78, regulation of HIV-1 expression from the provirus is controlled both transcriptionally and post-transcriptionally. The two accessory proteins tat and rev are responsible for the stimulation of the transcripts elongation and the export of some mRNAs species, respectively, that would otherwise be degraded within the nucleus. Within HIV-1 LTR, the transactivation-responsive element (TAR) recruits tat and its cellular cofactor P- TEFb, resulting in transcriptional elongation, a process that is dependent on cellular elongation factors as ELL2 78.

31 The core promoter of HIV-1 contains three SP1 binding sites, a TATA box and an initiator sequence. Additionally, the HIV-1 LTR bears an NFKB binding site, which acts a viral enhancer involved in the reactivation of latency and in increased HIV-1 replication in T cells 78,110. The epigenetic regulation, dependent on acetylation and methylation of histones as well as on DNA methylation allows the virus to establish latency 111. At a later stage, HIV-1 transcripts are exported from the nucleus to the cytoplasm. Given that unspliced and incompletely spliced mRNAs are the target of nuclear enzymes that degrade them, these transcripts species have to associate with rev, that recognize a specific sequence in the env coding sequence, the rev-responsive element (RRE) and hide them from the cellular machinery. The protected transcripts are then exported through the NPC via interaction with the cellular protein Crm1 78. HIV-1 transcripts are further processed at the 3’ end and polyadenylated, in the view of being translated together with host-derived mRNAs 78.

The gag gene products that constitute the structural components of the HIV-1 virion coordinate the last phase of viral replication. Indeed, viral proteins and nucleic acid materials assembly at the viral membrane is directed by the unprocessed gag polyprotein that binds the plasma membrane, and the env protein, recruits the PR, the RT and the IN proteins and packs the viral RNA and the primer tRNALys2, 3 112, forming an immature virion. During the budding process, the plasma membrane is integrated into the viral particle, constituting a de novo lipid bilayer. Upon particle maturation, the PR protein cleaves gag, producing processed MA, CA and NC. The resulting mature virions are either released into the blood stream or directly infect new cells via cell-to-cell transmission involving the formation of virological synapses 39,113.

My thesis will focus on the activity of the restriction factor TRIM5, which will be introduced further in the next chapter, and thus I will examine the early steps of retroviral replication.

32 1.2 TRIM5 and the innate immunity

Viruses and other pathogens attack the organism by different routes. The immune response that is mounted to counteract this invasion depends on different germline-encoded and de novo synthesized factors that are produced in the context of the innate and adaptive immune response, respectively. As a first defense, all cell types that are the target of infections carry different combinations of proteins acting like sentinels that recognize specific foreign motifs or molecular signatures. These pathogen-associated molecular patterns (PAMPs) are bound by pattern-recognition receptors (PRRs), at the cell membrane, in endosomal compartments or within the cytoplasm, and this interaction results in the activation of signaling pathways that will ultimately lead to the production of inflammatory cytokines and type I Interferon (Type I IFN), as reviewed in 114The inflammatory cytokines activate immune cells and act on endothelial cells, in this way stimulating the early inflammatory response (reviewed in 115).

Secretion of Type I IFN provokes an antiviral state via the production of interferon-stimulated genes (ISGs) and the stimulation of the acquired immune system by activating immune cells, contributing to the presentation of Major Histocompatibility Complex Class I (MHC I) molecules, essential for recognition of antigens exposed by antigen-presenting cells (APCs) and promoting cytotoxic T lymphocytes (CTL) response 116.

Type I IFN stimulates the expression of a plethora of factors with different specificities to mount a broad antiviral response. The category of ISGs contains several hundreds of genes that serve as antiviral effectors or as signaling activators for processes such as apoptosis and vesicular transport 117. Many antiviral effectors are components of the intrinsic immunity, and will be discussed later. These constituvely-expressed factors are able to degrade viral components specifically in a direct way.

33 Interestingly, as it will be discussed later, some of these ISGs, including TRIM5, function as PRR them selves, showing the way by which the innate immunity can be self-amplified.

1.2.1 The Pattern-recognition receptors

Four main families of PRR are associated with detection of foreign material within a cell: one group of membrane-associated receptors and three classes of cytoplasmic PRRs.

Toll-like receptors

A first group of PRRs is composed of proteins with a transmembrane domain, an extracellular leucine rich repeat (LRR) region that recognizes specific PAMPs, and an intracellular module containing a TLR/IL-1R (TIR) domain that allows them to interact with adaptors molecules for the signal transduction 118.

With ten functional members in humans, the Toll-like receptor (TLR) family recognizes a wide variety of PAMPs. Using different adaptor molecules, TLRs bind to many types of molecules including lipids, lipoproteins, proteins and nucleic acids. The members of this family of receptors are located at different cell compartments. Whereas TLRs 1, 2, 4, 5, 6 and 10 are found at the cell surface, TLRs 3, 7, 8, and 9 are located in endosomes 114,119 (figure 6). TLR2 is found in the form of a dimer, either with TLR1, TLR6 or TLR10 mainly detecting PAMPs from Bacteria and fungi 114,120,121. When complexed with TLR1, the dimer recognizes the triacetylated lipoproteins, peptidoglycans and lipopolysaccharides 122, as reviewed by Kawai and Akira 114. The dimer composed of TLR2 and 6 is responsible for the detection of diacylated lipoproteins 123. The specific molecules recognized by TLR2-TLR10 have not been discovered yet 120,121. The other cell surface-associated PRRs, TLR4 and 5, were found to bind to lipopolysaccharides (LPS) and flagellin 124, respectively (reviewed in 125).

Within the endosomal compartment, TLRs are responsible for the sensing of foreign nucleic acids. TLR3 was discovered to be activated by a dsDNA synthetic oligonucleotide, the polyinosinic:polycytidylic acid (poly I:C) 114. Whereas TLR7 and 8 recognizes ssRNA, TLR9 binds viral DNA. TLR7 and 9 are located in the endoplasmic reticulum (ER) under resting conditions, but they translocate into the endosomal compartment upon a first TLR- mediated stimulation, for example when TLR4 binds to LPS 126.

A common feature that adaptor molecules used by all TLRs share is a TIR domain that allows TIR-TIR interactions with the corresponding receptor 114,127- 131. These signal transducers include myeloid differentiation primary response gene 88 (MyD88), TIR domain-containing adaptor protein (TIRAP), TIR domain- containing adapter-inducing IFNβ (TRIF) and TIR domain-containing adapter molecule (TRAM). Whereas TIRAP directs MyD88 towards TLR2, TRAM targets TRIF to TLR4. Both interactions of MyD88 and TRIF with a corresponding TLR dimer will result in the expression of inflammatory cytokines 114,132,133. The TRIF- TRAM-TLR complex induces additionally the production of Type I IFN 132. Most TLRs activate signaling leading to the production of inflammatory cytokines via either the MyD88- or the TRIF-mediated pathways. However, TLR4 is an exception because it requires both pathways to induce expression of the corresponding genes 134-137.

The cell possesses many means to perform signal transduction, by post- translational modifications. For example, the phosphorylation of different oligoaminoacid substrates, by protein kinases, produces conformational changes in the molecules, allowing them to interact with other proteins and activate signaling pathways 138.

Another post-translational modification involves the addition of monoubiquitin or polyubiquitin (polyUb) chains to target proteins. The best-studied types of ubiquitin chains are the ones that are linked via lysine (K) 48 and K63 of

35 ubiquitin, leading to targeting of the modified protein to the proteasome or to signal transduction complexes, respectively 139.

As review by Schulman and Harper 140, the mechanism of protein ubiquitynation starts with an ubiquitin-activating enzyme (E1), which bind to two ubiquitin (Ub) molecules via thioester bonds. In the next step, an ubiquitin-conjugating enzyme (E2) recognizes the E1-Ub complex and takes over one of the Ub molecules. Finally, an E3 Ub-ligase enzyme bound to a specific substrate interacts with the E2, facilitating the catalysis of the ubiquitination of the target protein.

The MyD88-induced signaling requires the cooperation of the IL-1 receptor- associated kinases (IRAK) 1, 2, 4 and M that interact with tumor necrosis factor (TNF) receptor associated factor 6 (TRAF6), which ubiquitylates them with K63- linked polyubiquitin chains in addition to autoubiquitinate itself 114. PolyU chains interact with, from one part TAK-1 binding protein 2 (Tab2) and 3 and from another part with the inhibitor of nuclear factor kappa-B (NFΚB) kinase gamma (IKKγ), leading eventually to the activation of mitogen-activated protein kinase (MAPK)- and NFKB-dependent pathways, which will be described later. The engagement of TLR7 and 9 can additionally stimulate TRAF3 in cooperation with TRAF6, in the MyD88 pathway and lead to the activation of IRF7 141. Finally, in some immune cells, TLRs use the MyD88-IRAK4 pathway to stimulate IRF5 142.

The TRIF-dependent pathway involves the initial activation of the Tab2-Tab3- TAK1 complex by the cooperation of TRAF6 and the kinase receptor-interacting protein 1 (RIP1) 114. TRIF dimerization can alternatively activate TRAF3 and the TBK1-IKKγ kinases leading to the IFN-regulatory factor 3 (IRF3) activation and the subsequent production of IFNβ.

36 '()(*+(&,-./(0&

!"#$%&

1#2341#25&

9:7$&

1#2341#25& 12678&

Figure 6: Simplified scheme of the role of Toll-like receptors in the innate immune signaling. TLR1, 2, 4, 5, 6 and 10 are shown at the cell surface. TLRs 3 and 7-9 are depicted on endosomal vesicles. The PAMPs that stimulates each PRR is indicated adjacent to the rectangles that symbolize the extracellular leucine rich repeat (LRR) domains of the TLRs. Sensing of the different PAMPs activate the innate immune cascades AP- 1, NF-κB, IRF3, IRF7 and/or IRF5, leading to the production of type I IFN and inflammatory cytokines. Adapted from Van Duin et al., 2006 143.

The cytosolic PRRs

Upon entry of a pathogen within the cell, different types of cytosolic PRRs sense PAMPs. The retinoic acid inducible gene 1 (RIG-1)- like family of receptors (RLR) recognizes viral RNA molecules in different conformations 144. Another well-studied family of receptors is the NOD-like group (NLR), recognizing bacterial products such as peptidoglycans and flagellin 145.

These two first families of receptors activate the innate immune system by the stimulation of the MAPK- and NFκB-dependent pathways 145-149. Additionally, RLRs are able to activate IRF 3 and/or 7, similar to TLRs 148,150. Some cytosolic DNA-sensors were identified recently, such as stimulator of IFN genes protein (STING), IFNγ-induced protein 16 (IFI16), the cyclic GMP-AMP synthase (cGAS) and the DNA helicases DDX41 and DHX9/DHX36 151-154. These proteins mostly

37 signal by activating differentially the IRF3-, IRF7- and/or NFKB-dependent pathways 151,153,154.

Highlighting the role of the sensing of viral DNA during immunity to retroviruses, HIV-1 capsid binds the cellular cyclophilin A and CPSF6 as cofactors to escape to recognition of the reverse-transcription products by as yet unidentified cytosolic sensors that would otherwise restrict replication in a type I IFN-dependent manner 155.

1.2.2 The innate immune pathways

Two important routes that PRRs use to activate the innate immune response are via the AP-1- and the NFκB- mediated signaling (figure 5). The activation of the MAPK-dependent pathway is initiated with the stimulation of different MAPK kinase kinases (MAP3Ks) and continues with a cascade of subsequent phosphorylations of a target MAPK kinase (MAP2K) that will in turn act on a specific MAPK 115. Three routes of the MAPK signaling pathway have been extensively studied and are involved in one or both of the pro-inflammatory and anti-inflammatory processes during the immune response: the extracellular signal-regulated kinase (ERK), Jun N-terminal Kinase (JNK) and protein of 38 kDa (p38) pathways (figure 7). The corresponding studies are mainly based on the effect of the stimulation of TLRs.

The ERK1 and ERK2 branch involves the activation, upon engagement of a TLR, of the tumor progression locus 2 (TPL2) 156. This MAP3K in turns activate MAPK kinase1 (MKK1) and MKK2. This pathway leads to the production of from one side the pro-inflammatory cytokines tumor TNF-α and interleukin 1 beta (IL-1β), from the other side the anti-inflammatory IL-10 115,157,158. ERK1 and ERK2 additionally have a repressive action on the expression of IL-12 and the two antiviral proteins IFNβ and inducible nitric oxide synthase (iNOS) 158,159.

The stimulation of the complex formed by the MAP3K TAK1 and Tab2/Tab3, results in the phosphorylation of the components of three main pathways. The

38 first route involves the activation if the IKK complex, leading to the enhancement of the ERK1/2- and NFκB- mediated pathways. The targeting of, from one side MKK4 and 7, from another side MKK3 and 6, will result in the up-regulation of JNK- and p38- stimulated genes, respectively 115.

JNK-mediated signaling was shown to contribute in myeloid cells to the establishment of the acute inflammatory M1 macrophage phenotype 160.

The p38 MAPK isoform alpha (p38α), for his part, has dual roles. Depending on the cell line and the type of induced-damage examined, p38α enhances or decreases the inflammatory response 115,161,162.

The ERK1/2 MAPKs phosphorylate the nuclear transcription factors c-AMP response-element binding protein (CREB), cMyc and cFos. The latter protein associates with one of the targets of JNK, cJun, to form one type of activator protein- 1 (AP-1) transcription complex. Alternatively, cJun can homodimerize or associates with the ATF2 transcription factor that is phosphorylated by p38 115.

The different AP-1 transcription factor complexes are involved in different contrary processes such as cell proliferation and apoptosis, inflammatory potentiation and modulation 163. These opposite effects are mediated in part by differential combinations of transcription factors that will form heterodimers with different target genes. Consistently, the binding of cJun with cFos or ATF2 will result in cell growth or neuronal apoptosis, respectively 164. An alternative way to direct AP-1 in one or another pathway is by the expression of JNK targets that antagonize the action of cJun-containing complexes, as it seems to be the case for JunB, in the context of cJun-promoted cell growth 165,166. The AP-1 transcritption factors additionally activate a plethora of target genes that are involved in the innate immune response and the inflammatory process. Dimers of Jun and Fos can cooperate with the nuclear factor of activated T cells (NFAT) to activate the expression of cytokines as IL-2, IL-3, IL-4, IL-5 and IL-13, that of molecules playing a role in humoral immunity such as CD25, as well as that of

39 the inducer of inflammatory prostaglandins, cyclooxygenase 2 (COX2) 167. Additionally, AP-1 stimulates the expression of the type I IFN, alone or in combination with other transcription factors such as signal transducers and activators of transcription 4 (STAT4) 168-170. Inflammatory cytokines in turn up- regulate the MAPK-dependent pathway 163.

As Type I IFN production induces, among other pathways, AP-1-mediated signaling and the expression of ISGs, it constitutes the second wave of innate immune activation and the MAPK-mediated pathway is thus an early response component. Similarly, NFκB activity is stimulated by inflammatory cytokines, including IFNβ 171-173, to induce successively the expression of genes involved in the innate immune response.

The NFκB family of transcription factors is composed of six members, namely p50, p52, RelA, RelB, c-Rel and v-Rel 172. Similar to the AP-1 family, these proteins associate in different combinations of homo- or hetero-dimers. All the NFκB proteins are composed of an N-terminal Rel-homology domain (RHD) that allows them to bind to DNA as well as to dimerize, and a nuclear-localization signal (NLS) that directs them to the nucleus when free from the inhibitor. Except for p50 and p52, the other NFκB members carry a C-terminal transcription activation domain (TAD) that allows them to stimulate the synthesis of gene transcripts. For this reason, p50 and p52 need to heterodimerize with other proteins as RelA or RelB to act as transactivators 174- 176. When found in a homodimeric form, p50 and p52 act as transcriptional repressors 173.

In most unstimulated cells, the NFκB complexes are bound to inhibitor proteins called inhibitors of κB (IκB) and reside in the cytoplasm. Upon activation of the upstream kinases IκB kinases (IKKs), IκB is phosphorylated then degraded and the NFκB transcription complex can enter the nucleus to bind particular sites in the promoter of specific genes 177.

40 The IKK proteins are activated via two alternative pathways, namely the canonical and non-canonical signaling routes 173. The conventional pathway is initiated with most of the stimuli leading to NFκB activation, such as ligand binding to a TLR and the TNFα interaction with its receptor. Upon recognition of TNFα by the TNF 1 receptor (TNF1R), the TNF1R-associated protein with DEATH domain (TRADD) acts as a scaffold for the recruitment of Fas-associated protein with DEATH domain (FADD) and TRAF2 or TRAF5 173. The E3 ubiquitin-ligase activity of TRAF2 or TRAF5, similar to that of TRAF6, mediates the activation of RIP1 by K63-linked polyubiquitination 178. This protein modification allows RIP1 to interact with the TAK1 complex via Tab2 and Tab3, which subsequently leads to phosphorylation of the IKKβ component within the IKK complex also formed by IKKα and IKKγ (NEMO) 179,180. In this way activated, the IKK complex phosphorylates the IκB, targeting it to the proteasome and releasing the NFκB complexes to allow them to enter the nucleus and activate target genes (figure 6).

In the case of the TLR4 signaling via MyD88, the scaffold protein is TRAM and the activation of the TAK1 complex involves TRAF6 173, as seen above for the AP- 1 pathway. The TAK1 kinase is thus a central component of the innate immune response and mediates the convergence of different stimulated receptors as PRRs and cytokine receptors to the activation of the MAPK and NFκB pathways. This constitutes a means of producing cytokines, antiviral proteins, and negative feed-back loops, all of which are important in the context of infection and in the setting and the modulation of the inflammation.

Other cytokines from the TNF family, such as CD40L and lymphotoxin-β (LT-β), activate the non-canonical NFκB pathway, via the phosphorylation of an IKKα dimer by the NFκB-inducing kinase (NIK) 181. The phosphorylated IKKα interacts with the p100 NFκB precursor and process it into p52. The dimer composed of p52 and RelB is then competent to migrate into the nucleus and regulate gene expression 181.

41 As reviewed by Bonizzi and Karin 172, the NFκB transcription factors up-regulate the expression of cytokines, chemokines, other proinflammatory and/or antiviral molecules as iNOS, COX2, as well as that of adhesion molecules, essential for cell- to-cell contacts during the immune response. Thus, similar to AP-1, depending on different combinations of transcription factors, NFκB-mediated pathway can lead to apoptosis or survival and this property allows these proteins to play a role at different stages of development and survival of immune cells such as neutrophils, DCs, natural killer (NK) cells, T lymphocytes and B cells 173.

*.% :;% :;% :;%

!,-$% !,-.% !"#$% !,'(%

&''()$% &''/)0% &''.)1% <''4% <''5%

9*&=%

*+'()$% 2.34)5)6)7% 89'()$).%

,">(%?@ABCD@E2FGB%HAD?G@C% 9IJ-%?@ABCD@E2FGB%HAD?G@C%

Figure 7: Simplified representation of the MAPK and NFKB pathways. When an E3 ligase (green hexagone) such as TRAF6 or TRIM5 synthesizes poly-ubquitin chains (blue circles, Ub), the TAB2/TAB3/TAK1 complex gets activated and stimulates kinases of the p38 and JNK family, resulting in the activation of different complexes of AP-1 transcription factors (red and green imbricated shapes). The IKK complex is also activated by TAK1 and results in the stimulation of different NFκB transcription factors (yellow and blue imbricated shapes). The activation of ERK1 and ERK2 is mediated via the stimulation of the TPL2-dependent pathway.

1.2.3 Immunity to retroviruses: restriction factors

The replication ability of retroviruses in different cells depends on many cellular factors. The first considered factor is the entry of the retrovirus into the cell cytoplasm, via recognition of the corresponding receptor. For example, as discussed previously, HIV-1 entry requires the recognition of the CD4 receptor and a coreceptor, principally CXCR4 or CCR5. The subsequent steps of the viral life cycle exploit host proteins in a species-dependent way to proceed, as highlighted by the inability of HIV-1 to productively infect murine cell lines that have been engineered to express human CD4 182 and taking into account that the murine CXCR4 can be used as a coreceptor by HIV-1 183. Importantly, murine cells have a cyclin T1 protein, that HIV-1 Tat does not bind because of a species- specific polymorphism, thus precluding the employment of this cofactor required for the transactivation of LTR-directed expression 184. When circumventing this post-entry blocks by expression of human Cyclin T1, some murine cell lines proceed into viral gene transcription, but further steps are blocked, as mRNA export and processing, as well as virion assembly 185,186. These blockades are rescued upon fusion of murine and human cells, showing that there are factors exerting a positive effect on viral replication late steps that are not present in the mouse 187.

Interestingly, in contrast to fibroblasts, murine T cells do not support HIV-1 reverse transcription 183. The blockade of a pre-integration step of the viral replication strongly recalls other phenotypes observed in mice and primates. The cell tropism is not only dictated by the presence or the absence of positive cofactors in a cell.

The first indirect report of a negative factor influencing retroviral replication was in 1957 by C. Friend who discovered that a genetic transmissible trait

43 dictated the susceptibility of different strains of mice to MLV 188. The factor mediating this blockade was later genetically mapped on chromosome 4 and called Fv-1 189,190. The two alleles of the gene were Fv1B and Fv1N that conferred the resistance to N tropic MLV (N-MLV) and B-MLV, respectively. Nineteen years after Friend’s discovery, the blockade by Fv-1 was determined to act after reverse-transcription, but before integration 191 (figure 8).

In parallel with the previous findings, a potent blockade of HIV-1 infection was observed in monkey cells 192,193 and was termed lentiviral susceptibility factor 1 (Lv1). Few years later, an Fv-1B-like restriction was observed in cells from primates including human, and from dog, pig and cow that potently blocked N- MLV early post-entry 194. The host protein responsible for this blockade was called restriction factor 1 (Ref1).

In common to all restriction factors, the barrier to retroviruses was capsid- specific, dominant and saturable by a high amount of viral particles 195-199. The factors at the origin of these blockades were all cloned and it appeared that Lv1 and Ref1 were products of the same gene, TRIM5 62,200-203.

I will first briefly describe the best-studied restriction factors affecting retroviral replication and will then focus on TRIM5.

44 Reverse Transcription

TRIM5

Nuclear Import Fv1

Figure 8: The blockade of early retroviral replication steps by TRIM5 and Fv1. The restriction factors TRIM5 and Fv1 inhibit retroviral replication at an early post-entry step. Whereas TRIM5 can act before reverse-transcription and nuclear import (solid black bars), Fv1 only targets the latter step. The virion core is represented by the blue conical shape. RNA and DNA species within the core are depicted as two black and blue bars, respectively. Courtesy of Prof. Jeremy Luban (adapted).

APOBEC3 proteins

Among a family of cytidine deaminases, the apolipoprotein B mRNA-editing enzyme catalytic polypeptide-like 3G (APOBEC3G) inhibits the replication of HIV-1 by associating with assembling virions, via its N terminal zinc-binding deaminase domain that interacts with the viral RNA and gag polyprotein 204,205. Once in a target cell, APOBEC3G recognizes cytosine residues within C-C dinucleotides on newly synthesized minus-strand viral cDNA and induces their deamination, transforming it into a uracil 206. The resulting viral genome contains guanine to adenine mutations, leading to replication catastrophe 207. In addition, the catalytic activity of APOBEC3G is required for the blockade of HIV-1 integration. The mechanism relies on the interference with the tRNALys3 primer dissociation, leading to the formation of abnormal 3’LTRs and thus a subsequent defect in its targeting to the host genome 208. Finally, APOBEC3G targets HIV-1 at

45 the reverse-transcription step by impeding the tRNALys3 to prime the viral RNA, although it is not clear whether this is in a deaminase-dependent way 209-212. However, despite these potent restrictions, HIV-1 evolved a mean to counteract APOBEC3G by orchestrating its degradation by the viral accessory protein Vif in a proteasome-dependent pathway 206.

Other members of the APOBEC family have similar deamination activity and restrict HIV-1 infection 213. While APOBEC3A has been linked to the inhibition of HIV-1 in monocytes 214, APOBEC3B is not expressed in primary lymphoid cells but still renders HIV-1 particles less infectious, when expressed transiently in the virus-producing cells 213,215. The other APOBEC proteins exerting anti-HIV-1 activity include APOBEC3C that inhibits the infectivity of some strains of HIV-1 216, APOBEC3D/E that is counteracted by vif 217 and APOBEC3F 213,218,219.

Tetherin

The tetherin restriction factor, named in that way because it “tethers” HIV-1 virions to the cell surface, impeding their release 220. By homodimerizing via the extracellular coiled-coil domain, tetherin engage a second monomer bound to the viral membrane 221.

Tetherin is induced by type I IFN and its action is counteracted by HIV-1 Vpu 220. In turn, the activation of the NFκB pathway by tetherin upon HIV-1 infection 222 results in the production of type I IFN.

SAMHD1

At first associated with the Aicardi-Goutières autoimmunity syndrome 223 the sterile alpha motif (SAM) and histidine-aspartic (HD) domains-containing protein 1 (SAMHD1) was subsequently investigated for its role in mediating the innate immunity to retroviruses. This restriction factor was found to decrease dNTP levels and to block HIV-1 reverse-transcription 224. As for APOBEC3G and tetherin, some retroviral accessory proteins neutralize SAMHD1. Indeed, Vpx

46 from HIV-2 and SIV degrades SAMHD1 by targeting this factor to the proteasome 225.

MX2

The IFN-induced myxovirus resistance 2 (MX2) protein restrict HIV-1 infection in a capsid-dependent way 103. Although the mechanism of retroviral inhibition remains unknown, the transient expression of the restriction factor decreased 2- LTR circles formation and integration 103, suggesting that this protein inhibits HIV-1 nuclear entry.

ZAP

The first retroviral target of the zinc-finger antiviral protein (ZAP) to be discovered was MLV 226. In this study, the abundance of MLV transcripts was decreased in rodent cells expressing the endogenous protein. It was later found that HIV-1 was similarly restricted by the human ZAP orthologue, which induced specific mRNA uncapping and degradation of the retroviral transcripts 227.

MOV10

Discovered in murine strains as the site of MLV provirus integration 2. the MOV10 gene encodes a protein with seven helicase motives 228. The human orthologue of MOV10 inhibits HIV-1 at various replication steps. Although the mechanism by which MOV10 reduces HIV-1 virion production is still unclear, it could involve the inhibition of gag expression 229 and this could have a link with the observed association of MOV10 orthologues from mammals with the RNA interference (RNAi) system 230 that may silence viral gene expression. At a second level, virion-associated MOV10 from the producer cells restricts HIV-1 reverse-transcription in the target cells 229,231.

47 ADAR-1

The adenosine deaminase acting on RNA protein 1 (ADAR1) induces the deamination of adenosine into inosine on a double-stranded RNA substrate 232. It was recently found that this enzyme has a restriction activity on HIV-1, inhibiting the expression of viral proteins via the post-transcriptional mRNA editing inducing a defect on nuclear export of the respective messengers of gag, pol and env 233.

Fv1

The sequence of the murine restriction factor Fv1 is derived from a gag gene from the endogenous retrovirus family ERV-L present across mammalian genomes, as revealed by the approximated 60% of homology with the sequence of human ERV-L (HERV-L) 234. The resulting capsid-like protein recognizes specific capsids of MLV strains. The different alleles of Fv1, N and B, differ only in three residues within a small motif associated with their restriction capacity 235 and recognize differentially a residue at position 110 of the amino acid sequence of the CA protein of B- and N-MLV, respectively 236. For the binding between the MLV capsid and Fv1 to happen, the gag polyprotein must be mature and cleaved from p12 and NC 237. The direct binding was shown using a biochemical method where capsid-coated lipid nanotubes were subjected to immunoprecipitation with Fv1 proteins. As observed by negative staining and electron microscopy, the capsid units assembled in an ordered manner in vitro, dependent on the typical retroviral β-sheet formation on the N terminus of the CA protein 238. Importantly, the binding results were in agreement with the specific restriction pattern of the distinct Fv1 alleles 238.

TRIM5

The large family of TRIpartite Motif (TRIM) proteins comprises approximately 100 members in human 239. Besides a few exceptions, they all share the conservation of three main modules in a precise order (reviewed in 240). The Ring-finger (RF) domain is found at the N-terminal part of a TRIM protein. This

48 domain is formed of zinc-coordinating motifs, that allows the formation of a “cross-brace” structure, involving cysteins and a histidine that contact two Zn atoms 241-243. Present in several families of proteins, RF domains often confer binding to an E2- ubiquitin conjugating enzyme and function as E3-ubiquitin ligases to themselves and to other substrates, as reviewed by Deshaies and Joazeiro 243. Indeed, many TRIM family members have been described to display E3-ubiquitin ligase activity, including TRIM5, TRIM21 and TRIM25 244-246, as reviewed in 240. The nuclear magnetic resonance (NMR) structure of the human TRIM5α RING finger has been solved 247 and showed that the core of the domain is composed of the majority of the hydrophobic residues and is located between two β-sheets and an α-helix.

As a second motif, TRIM proteins carry one or two B-box domains, which share a similar ternary conformation. Indeed, B-boxes also coordinate Zn atoms in a crossed configuration and form two β-sheets, followed by a α helix, as shown by the study of MID1 (TRIM18) and human TRIM5α 248-250. The TRIM B-box1 always precedes B-box2 and is never found alone 240. In contrast, many TRIMs possess only the B-box2, as exemplified by TRIM5 240. The function of B-box domains is not completely understood. However, the study of the B-box2 of MuRF1 (TRIM62) revealed a surface hydrophobic patch with polar residues on a dimer interface 251. This finding allowed the manipulation of the equivalent residues of TRIM5 B-box2 and showed that this domain is important for protein higher- order multimerization 250,252.

As a third motif, the Coiled-coil (CC) domain is the last module of the tripartite RING finger-B-box-CC (RBCC) motif. It contains appropriately spaced hydrophobic residues on amphipathic α-helices, forming two putative leucine- zipper motives, which are responsible for TRIM protein-to-protein interactions 253,254. This domain was shown to participate in the higher-order multimerization, homo- and hetero-dimer formation of TRIM proteins as well as to be required for their concentration into discrete cellular compartments such as the nuclear and cytoplasmic bodies (NB and CB, respectively) 255-259.

The TRIM proteins vary in their C-terminal domain composition. As reviewed by Ozato and colleagues 240, there are ten distinct C-terminal TRIM domains, found alone or in different combinations, and participating to functions as variable as localization to microtubules, binding to a retroviral capsid or interaction with histones and transcriptional repression. The discovery that the TRIM5 gene was the determinant for HIV-1 restriction in monkey cells 62,200,201,203, motivated the study of the involvement of TRIM proteins in the innate immunity. Importantly, TRIM25 was found to induce the K63 polyubiquitination RIG-I, essential for the RNA viruses-triggered signal transduction 245. Although TRIM1, TRIM19 and TRIM22 were already shown to have antiviral properties (as reviewed by Nisole and colleagues 260), a large screen, looking at 55 TRIM proteins revealed that this feature was shared by many other TRIM family members 261. Notably, TRIM11 and TRIM15 were found to restrict the release of HIV-1 and MLV, respectively, with TRIM15 recognizing the gag protein in the producer cell in a B-box- dependent way 261.

The role of TRIM proteins in the innate immune response is further emphasized by the fact that their expression is up-regulated upon type I IFN treatment or induction of TLR by agonists and that they differentially activate the AP-1, NFκB and IFNβ promoters 262-266.

1.2.4 TRIM5-mediated retroviral restriction

Six isoforms have been described for the human TRIM5 gene, namely α, γ, δ, ε, ι and κ 267. Only TRIM5α have been shown to inhibit retroviral replication. Moreover, TRIM5ι is the second more abundant transcript in some human cell lines and, similar to the isoforms γ, δ and κ, down-regulates TRIM5α levels and correspondingly modulates the anti N-MLV restriction activity 267. An important particularity of TRIM5α is that it is the only isoform that possess a C-terminal PRY-SPRY domain (figure 9).

50 B># !"# $$# %%# &!'(&!'#

5!6789:#5;,<=>$:#5;,<=>%:# 5;,

B># !"# $$# %%# %)*+#

5!678%)*#

!"# $$# %%#

5;,<=>+:#5;,<=>$:#230#5!678#

Figure 9: Schematic representation of the TRIM5 orthologues and structure of the proteins. TRIM5 proteins are composed of an RBCC motif, comprising the RING finger (RF), the B-Box (BB) and the Coiled- Coil (CC). Some TRIM5 orthologues carry a C-terminal capsid-binding domain. Whereas most of TRIM5 proteins bear a C-terminal PRYSPRY domain, some primate species carry a Cyclophilin A (CypA) module. The linker 2 (L2) separates the CC from the C terminal domain of TRIM5 proteins.

The PRYSPRY is found on other TRIM5 paralogues and structural studies have revealed that it forms a dimer interface via a donor sequence and an acceptor strand from different protein targets 268. This interface is composed of six variable loops (VLs) that mediate the binding to different specific substrates 269. Importantly, the PRYSPRY domain of TRIM21 recognizes immunoglobulin G (IgG) with contacting residues found in the VL4, where the E405 and E406 of TRIM5α conferring species-specific N-MLV activity are positioned 269,270.

For its part, the specificity of the restriction of HIV-1 by rhesus macaque TRIM5α involves residues in the VL1 271. Indeed, the PRYSPRY domain is the determinant of TRIM5α retroviral restriction specificity 271-273 and directly binds to the retroviral capsid 274,275.

51 The binding between TRIM5α and particulate capsid protein was never observed. Notably, whereas mature virions could saturate the restriction phenotype, monomeric capsid did not have any effect on retroviral blockade 199,237,276,277. It became later evident that the restriction factor recognized the CA in complexes forming an hexameric lattice 271,274. Soon after viral entry into the host cell cytoplasm, TRIM5α orthologues bind to the retroviral capsid 62,200 and can subsequently mediate a blockade at two steps of the viral life cycle.

The use of single-cycle infection assays allows a safe and precise way to determine what steps of the viral life cycle are affected by a cellular factor in non-permissive (restrictive) cells. This method uses combinations of three-part or two-part vectors composed of the viral genome, the packaging genes and an envelope. Once expressed in producer cells, the viral proteins form particles that are used to infect target cells. The challenging of target cells with these virions will result in a single-round of infection, given that no complete retroviral genome is provided and thus the virus is replication-incompetent. Cells can also be transduced (figure 10), referring to the transfer of DNA by retroviral vectors. When using bi-cistronic vectors coding for a gene of interest and for an antibiotic resistance gene, cells can be selected with the specific antibiotic and give rise to stable cell lines expressing constitutively the gene which function wants to be studied.

First, TRIM5α impedes reverse-transcription to proceed, as revealed by the comparison of viral cDNA accumulation between permissive and non-permissive cell lines 194,195,202,277. Indeed, the quantification of the reverse-transcripts in HeLa cells transduced with a bi-cistronic vector revealed that the cells that stably expressed rhesus TRIM5α contained at least ten fold less early and late HIV-1 cDNA products than the cells that had been transduced with the vector that only carries the antibiotic resistance gene 62,200.

The TRIM5 gene has been subjected to strong positive selection on residues of the PRYSPRY domain 278,279. In at least two independent events, the cyclophilin A (CypA) cDNA has inserted between exons 7 and 8 or in the 3’ Untranslated

52 region (3’UTR) of the TRIM5 gene via a LINE-1-mediated retrotransposition, replacing the PRY-SPRY domain and forming a fusion protein (TRIM5Cyp) 62,280- 283 (figure 8). Whereas in the case of the New world monkey Aotus trivirgatus (owl monkey), TRIM5Cyp potently blocks HIV-1, FIV and SIV from the African green monkey (SIVagm), the version from Macaca mulata restricts HIV-2 and FIV 62,282,284.

Similar to TRIM5α, owl monkey TRIM5Cyp binds to the retroviral capsid soon after entry and blocks the reverse-transcription 62,285,286. The cylophilin A (CypA) domain is responsible for the binding to HIV-1 capsid, as evidenced by the examination of the effect of deletion mutants or the addition of CsA 285,287,288. Although it was shown that the monomeric CA protein p24 could bind to cyclophilins A and B 289, the saturation of the TRIM5Cyp-mediated blockade requires completely processed virion cores 68 indicating that the assembled hexameric capsid is the target of the restriction factor.

TRIM5 proteins can additionally target a second and later step of the retroviral life cycle, as evidenced by the treatment of non-permissive cells with proteasome inhibitors such as MG132 290-293. Indeed, after addition of MG132, an increase in reverse-transcripts was observed, but the restriction was still not affected. More profound examination of the abundance of different viral products showed that 2-LTR circles, a marker for nuclear import, were affected 291,292, suggesting that a step before integration is also affected by TRIM5 orthologues.

The previous findings allow establishing a two-step model of TRIM5-mediated retroviral blockade (figure 7). First, inhibition of the reverse-transcription is concomitant with the observed proteasome-dependent TRIM5-mediated disassembly of the capsid 290,293. In the second step, a proteasome-independent mechanism is responsible for a block before, or at, retroviral cDNA nuclear import.

53 1.2.5 TRIM5 is a PRR

The expression of many TRIM proteins has been induced upon type I IFN or TLR agonist treatment 262-264,294,295. Importantly, TRIM5α and TRIM5Cyp transcripts are up-regulated following the addition of Type I IFN or LPS, in cells expressing the corresponding receptors 262,263,296. Moreover, we found that TRIM5 is required for the establishment of the TLR4-mediated antiviral state 265.

In the same study, we showed that TRIM5 function as a PRR for the retroviral core. Notably, TRIM5 synthesizes unanchored K63-polyUb chains that activate the TAK1-TAB2-TAB3 complex leading to the stimulation of MAPK- and NFκB- mediated signaling. Likewise, TRIM5Cyp was additionally able to stimulate AP-1 and NFκB promoters. This activity of TRIM5 alone is enhanced when it binds a restriction-sensitive capsid. Furthermore, the specific E2 ligase Ubc13 and TAK1 were found to be necessary for TRIM5-mediated restriction 265. In agreement with these findings, another study revealed the importance of the E3 ubiquitin ligase function to the TRIM5-mediated retroviral blockade, although in contrast with the previous study, autoubiquitination was the proposed limiting process 247. Conversely, a study showed that the E3 ligase function of some TRIM5Cyp is not required for retroviral blockade, as evidenced by the conserved restriction ability of TRIM5Cyp mutants were the RING and B-box domains were deleted 288. Furthermore, the binding of the capsid to TRIM5Cyp precludes the rescue by CsA treatment, as soon as 15 minutes after virus entry 286, showing that for this TRIM5 orthologue, the restriction happens very quickly and that it is less likely that the products of the TRIMCyp-induced innate immune pathways could contribute to this blockade.

Although major progress has been made on the comprehension of the TRIM5- mediated restriction mechanism, it remains unclear how similarly the different TRIM5 orthologues rely on the activation of the innate immune signaling for performing retroviral blockade. Notably, the TAK1-TAB2-TAB3 complex could be used as a platform that helps to recruit other cellular factors important for direct TRIM5-mediated restriction. Alternatively, some cellular protein produced by

54 the activation of the MAPK- and/or NFκB- dependent pathways could induce the degradation of sequestered capsid within a TRIM5 complex that would be potentially resistant to CsA, as proposed by Lukic and Campbell 297. The fact that the retroviral restriction mediated by TRIM5 is a two-step mechanism further complicates the interpretation of the data obtained to date.

1.2.6 TRIM5 in the mouse

In the mouse, the TRIM5 locus has expanded, leading to the generation of eight TRIM5 orthologues in the C57BL/6 laboratory strain 298. Based on sequence conservation, two groups of TRIM5 orthologues can be distinguished in the mouse. The murine TRIM5-like proteins are paralogues in respect of each other. In the C57BL/6 strain successive duplications of Trim12 gave rise to Trim12A, B and C. Three other genes emerged from Trim30. This second group is composed of Trim30A, B, C and D. The fourth Trim30-like gene, Trim30E, is now annotated as a pseudogene in the GenBank database (see Gene ID: 625321).

The different murine TRIM5 orthologues are RBCC proteins, though the CC domain of Trim30-1 is not complete (figure 7). Additionally, Trim12B and C as well as Trim30A, C and D carry a PRYSPRY domain 298. Although attempts were made to find retroviruses that could be restricted by the different murine TRIM5 orthologues were unsuccessful 298, the presence of a C-terminal PRYSPRY suggests that they could possibly recognize a retroviral capsid and block infectivity, as other TRIM5 orthologues do. Additionally, the study previously mentioned on murine T cell lines that potently block HIV-1 raise the question on the involvement of the different TRIM5 proteins in this restriction phenotype that is independent of the Fv1 alleles 183. Intriguingly, the blockade seems to happen before reverse-transcription. Furthermore, it would be interesting to investigate the ability of these TRIM5 proteins as potential innate immunity inducers, to evaluate the conservation of this feature of TRIM5α among mammals.

55 Retroviral vectors

AntibioR'' IRES GOI

!"#$ *&+,%-+(' %&%'()$ .)$)(+-&''

:;<' *+,$ !"##"$%"&'()$$' -$.**/$

/)$$'$%&)'.0",$1' )234)..%&5'06)' 789'

Figure 10: Schematic view of the cell transduction method. Bi-cistronic retroviral vector containing an antibiotic resistance (AntibioR), followed by an internal ribosome entry site (IRES) and the ORF for the gene of interest (GOI), can be used together with vectors containing the packaging genes(gagpol) and the enveloppe (env) to produce viral-like particles (VLPs). Mammalian cells are infected with the GOI- containing VLPs and 48 hours later are subjected to antibiotic selection. After one week in antibiotic- containing medium, a cell line stably expressing the GOI can be used for subsequent studies. Yellow arrows: Promoter.

1.2.7 TRIM5 take over on the acquired immunity Innate immunity to retroviruses is demonstrated in the case of MLV and mouse mammary tumor virus (MMTV) in some murine strains, showing partial dependence on TLR7, suggesting that viral RNA detection contributes to the antiviral response 44,299,300. In a class of DCs, the plasmacytoid DCs (pDCs), HIV-1 is also sensed by TLR7, leading to type I IFN secretion 301-303. However, the conventional myeloid DCs (cDCs) that are important for priming naïve HIV-1- specific CD4+ T cells 44,46 fail to be productively infected by HIV-1 and thus in inducing innate immune signaling, necessary for the maturation of the APC 304, (reviewed in 44). Failure of DCs maturation was suggested to be the cause of the induction of immune inhibitor molecules by primed CD4+ T cells in the context of HIV-1 infection 44. Although CD8+ cytotoxic T cells (CTLs) exert some effect on HIV-1, as demonstrated by the high-rate of mutations in the CTL-directed

56 antigens of the HIV-1 sequence 305,306, these cells fail to kill the majority of infected cells even upon forced expression of the provirus by histone deacetylases 44.

Contributing to the difficulty to the innate immune sensing of HIV-1 in humans, TRIM5α blocks laboratory strains very poorly, as assessed by single-cycle infectivity assays 307. As TRIM5α functions as a PRR, the consequence of binding less efficiently the HIV-1 capsid is that the innate immune signaling is not activated or very poorly stimulated. In agreement, challenging human DCs with HIV-1 failed to show production of inflammatory cytokines, in contrast of when using restriction-sensitive viruses 44,265. However, HIV-1 strains derived from clinical isolates show variable susceptibility to human TRIM5α, as evidenced by comparing the infectivities of different gag-proteases sequences cloned in a HIV- 1 vector background on control or TRIM5-disrupted cell lines 307,308. The restriction by TRIM5α could reach 15 fold with some gag sequences. Interestingly, the mutants with increased sensitivity to TRIM5α bear mutations located in epitopes targeted by CTLs, suggesting that they were induced to escape the cellular immunity response 308.

Although the clinical isolates tested in the previous studies seem to rely on capsid mutations for TRIM5α-acquired sensitivity, they also carry different other mutations in the matrix, nucleocapsid and protease sequences. It would be of interest to investigate whether the strains that become sensitive to TRIM5α do so mainly because of the capsid mutations that would reveal an altered recognition by TRIM5α and would potentially result in a stronger induction of the innate immune signaling.

1.3 Aims of the thesis

TRIM5 is a cellular protein that has dual roles. First, it acts as a restriction factor, blocking the reverse-transcription and the nuclear entry of retroviruses 62,200,290-

57 292,309. Second, it functions as a signaling molecule that stimulates the MAPK- and NFκB- dependent innate immune pathways and is essential to the LPS-mediated antiviral state 265. The second role of TRIM5 is accentuated when it recognizes the retroviral capsid of a mature virion, acting as a PRR 265. The link between the two roles has been suggested by our previous study 265. Conversely, another study suggested that the E3 ligase function of TRIM5Cyp is not required for restriction 288. An important difference between the two studies resides in the fact that the second team used a feline kidney epithelial cell line (CRFK), which expresses a TRIM5 orthologue. As it was shown that TRIM5 is able to homo- and hetero-dimerize 310, likely via the coiled-coil domain, it would not be surprising that the TRIM5Cyp with the RING and B-box deletions could associate with the feline TRIM5. This binding could allow TRIM5Cyp to use the N-terminal domains of the feline orthologue to induce the appropriate signaling cascade.

The first aim of my thesis was to investigate further the requirement of the induction of the innate immune signaling by TRIM5 for its retroviral restriction function. I analyzed the conservation of the innate immune inducer feature of TRIM5 in primate and murine orthologues, as well as designed particular deletion and point mutants to identify the domains required for this feature. I next used strong and weak inducers of the innate immune promoters to fuse them with an HIV-1 CA-binding domain and evaluated their ability to restrict HIV-1. The ability of murine TRIM5 orthologues to restrict retroviruses was also investigated.

Human TRIM5α can restrict some HIV-1 strains from clinical isolates. However, it is not clear if the mutations in the capsid sequence exclusively dictate the restriction phenotype or if substitutions in other sites of the gag-protease sequences can contribute.

The second aim of my thesis was to determine if the mutations in the capsid of HIV-1 strains that were restricted in human cell lines could recapitulate the sensitivity to TRIM5α.

Chapter 2

THE ROLE OF THE MURINE TRIM5 ORTHOLOGUES IN INNATE IMMUNITY AND IN RETROVIRAL RESTRICTION

Introduction

Component of the innate immune response, TRIM5α is activated upon TLR4 engagement and is required for the establishment of the LPS-mediated antiviral state 265.

Another stimuli leading to the activation of the TRIM5-dependent innate immune signaling is the hexameric capsid lattice of a restriction-sensitive retrovirus that is directly recognized by a specific TRIM5 orthologue 265, which thus functions as a PRR.

The retroviral restriction mediated by TRIM5 is still an incompletely characterized mechanism. Nevertheless, it is known that TRIM5-sensitive retroviruses are blocked prior to reverse-transciption 194,195,200,202,277. However, the use of a proteasome inhibitor from one part, and artificial constructs consisting of fusions of different TRIM proteins to the HIV-1 binding-Cyclophilin A domain from another part, lead to the discovery that capsid-dependent restriction by a given TRIM5-like protein additionally happens in a second step of the viral life cycle, after the completion of reverse-transcription and before integration 102,290-293.

The blockade to the reverse-transcription involves the accelerated uncoating of the retroviral capsid, as shown with assays that examine the fate of pelletable capsid complexes in TRIM5-expressing cells 275,311. A link between the loss of particulate capsid and the ability to restrict the reverse-transcription step was shown recently. Indeed, some rhesus TRIM5α variants carrying mutations in the RING finger domain lost the capacity to induce the degradation of capsid upon

59 entry into the cell 312. This inability to accelerate the uncoating of the retroviral capsid correlated with the loss of a reverse-transcription blockade, but not of the second step before nuclear entry. These findings suggest that the promotion of premature uncoating of the capsid is important for the first step of TRIM5- mediated restriction.

The domains of TRIM5 that are required for restriction are still a matter of debate. Notably, comparison of the restriction by rhesus TRIM5α and owl monkey TRIM5Cyp showed that they require different domains depending on the orthologue examined 288. In that study, TRIM5α required the RING finger and B-box domains for retroviral restriction, whereas TRIM5Cyp with a deletion of the two N-terminal domains was still competent for the blockade. However, these data did not discriminate between the two steps of TRIM5-mediated blockade. It is therefore possible that TRIM5Cyp versions without the RF and Bbox domains could only block one of the steps of the viral life cycle but not the other one. The requirements of the Bbox for higher-order multimerization 250,252,313 and that of the CC for the dimerization 253,257,314 of TRIM5α suggests a model in which restriction needs a higher-order assembly on top of the lower-order multimerization. In the case of TRIM5Cyp, the Linker 2 (L2) region could account for the assembly of mutlimers in higher-order complexes 315, explaining why the RF and the Bbox domains are, in this case, dispensable for restriction. The dimerization to which the CC contributes was shown to be essential for the retroviral restriction via TRIM5 316. In agreement with this finding, the deletion of the CC domain in rhesus or human TRIM5α precluded its ability to block HIV-1 and N-MLV infections, respectively 254,272. We and others showed that some of the RF domain functions from TRIM5 and TRIM5Cyp are required for the inhibition of a restriction-sensitive retrovirus at least at one of the steps of the viral life cycle 200,265,272,288,312.

The B-box, the CC and the L2 region all contribute to the assembly of TRIM5 complexes into cytosolic concentrations termed cytoplasmic bodies (CBs) 315,317.

60 The overexpression of TRIM5 proteins was found to induce their localization and concentration into CBs 200,257. The ability of a TRIM5 to form CBs upon transient expression in the absence of restriction-sensitive viruses was shown not to be required for retroviral blockade 286,318. However, another team showed that rhesus TRIM5α associated with HIV-1 virions in structures similar to CBs 290. In fact, Sastri and Campbell proposed that it is the ability to induce CBs around the retroviral particle that dictates the capacity to restrict a specific retrovirus, and that the preexisting CBs reflects this tendency of TRIM5 to form protein aggregates around the viral core 319.

In our previous report, we found that TRIM5 activates AP-1 and NFκB pathways 265, in a RING-dependent manner. This chapter aims to investigate the importance of the signaling-inducing function of TRIM5 for its restriction activity.

2.1 The link between the two functions of TRIM5: induction of the innate immune signaling and retroviral restriction.

In the present study, we found that the innate immune signaling function of TRIM5 is conserved among mammals, as revealed by the examination of simian, feline and murine orthologues. The requirements of the different domains for the signaling feature of TRIM5 were found to diverge from one orthologue to the other. In order to confirm the involvement of the innate immune signaling in TRIM5- mediated retroviral restriction, we fused strong and weak AP-1 and NFκB inducers to an HIV-1 binding-CypA domain and examined the ability of these artificial proteins to restrict HIV-1. We found that only the strong inducers of the AP-1 pathway could elicit retroviral restriction.

The following data are unpublished results. I performed all the experiments except the immunofluorescence imaging and the cloning of some primate TRIM5 orthologues into the pcDNA3.1(-) expression vector.

Retroviral restriction by non-human orthologues of TRIM5 correlates with the ability to activate AP-1 and NF-κB

Josefina Lascano1, Pradeep Uchil2, Walther Mothes2 and Jeremy Luban3

Unpublished

1Department of Microbiology and Molecular Medicine, University of Geneva, 1 Rue Michel Servet, CH-1211 Geneva 4, Switzerland 2Section of Microbial Pathogenesis, Yale University School of Medicine, New Haven, CT 06536, USA 3Program in Molecular Medicine, University of Massachusetts Medical School, 373 Plantation Street, Biotech II, Suite 319, Worcester, Massachusetts 01605, USA

*Correspondence to: Jeremy Luban Program in Molecular Medicine University of Massachusetts Medical School 373 Plantation Street Biotech II, Suite 319 Worcester, MA 01605 USA Phone: +1-508-856-6899 Fax: +1-508-856-8289 Email: [email protected]

62 ABSTRACT

The restriction factor TRIM5 blocks retroviruses at an early step of infection. Restriction depends on the recognition of a specific retroviral capsid and results in the stimulation of innate immune genes. The requirement of individual domains of TRIM5α in retroviral inhibition has been investigated previously and coincides with those important for inducing the AP-1 and NFκB promoters. Importantly, TRIM5α recognizes a particular retrovirus by the mean of its PRYSPRY domain that binds to the corresponding retroviral capsid. However, the link between the intrinsic ability of TRIM5 to stimulate the MAPK- and NFΚB- dependent innate immune pathways and the capacity to restrict a bound retrovirus is debated. Here we confirm, using seven murine TRIM5 orthologues that stimulate differentially the innate immune promoters and were fused to an HIV-1 capsid-binding domain, that restriction by TRIM5, in addition to the binding to a specific retroviral caspid, requires the ability to stimulate the innate immune signaling.

63 INTRODUCTION

TRIM5α is a restriction factor that blocks retroviruses’ replication at the reverse transcription stage and before nuclear import of the pre-integration complex 1-6.

The restriction is mediated in a species’ specific manner and involves the binding of the factor to a particular retroviral capsid 1,5,7.

The human TRIM5α protein is organized into four domains, namely the RING finger (RF), the B-box (BB), the Coiled-Coil (CC) and the PRYSPRY 5,8. A linker 2

(L2) region separates the two C-terminal domains.

The RF domain is a zinc-coordinating motif that promotes binding between proteins 9-12 and exhibits intrinsic E3 ubiquitin ligase activity 13,14. Indeed, TRIM5 catalyzes the synthesis of free Lysine 63-linked ubiquitin chains 13,15. These newly synthesized molecules are involved in cell signaling 15-17.

The BB domain confers to TRIM5 the ability to form higher-order assemblies 18-

20.

TRIM proteins can multimerize via the interactions between the CC domains 21-

23. TRIM5 dimers form the blocks for higher-order complexes, to which the L2 region contribute 1,24. The L2 region was found to promote the formation of cytoplasmic bodies and to be essential for TRIM5α-mediated restriction 24.

At its C-terminal extremity, TRIM5 bears a PRYSPRY domain, involved in the binding to the retroviral capsid 1,3,8. In the New World owl monkey Aotus trivirgatus, the cyclophilin A cDNA was inserted by LINE-1-mediated retrotransposition between the exons 7 and 8 of the Trim5 gene, replacing the

PRYSPRY domain 5,7,10,12. The Cyclophilin A domain is responsible for the binding to the HIV-1 capsid, allowing its subsequent restriction 5,14. Substitution of the

64 Histidine residue by a Glutamine at the position 126 of the Cyclophilin A protein or domain, abolishes its binding to the HIV-1 capsid 9,11,13.

We previously showed that the restriction of retroviruses by TRIM5 requires additionally the RF-dependent activation of the MAPK and NFκB pathways

13,16,17, essential components of the innate immune response.

Furthermore, Uchil and colleagues found recently that 14 out 42 human TRIM proteins were able to induce the AP-1 and NFκB promoters 15,18-20. Importantly, the anti-NMLV function of TRIM1 and TRIM62 was shown to be dependent on the activation of the innate immune pathways 15,21-23.

Conversely, another study argues against the requirement of the RF domain by the owl monkey TRIM5-Cyp to restrict HIV-1 25. Notably, however, the ability to induce the MAPK and NFκB pathways by the different deletion mutants analyzed by that team was not assessed.

Here, we aimed to confirm that the capacity to induce innate immune promoters is a feature of TRIM5 proteins that is conserved among mammals and we aimed to correlate this function with an ability to restrict specific retroviruses.

To further dissect the potential differences between TRIM5α and TRIM5-Cyp in the usage of their different domains to efficiently restrict the retroviral life cycle, we examined the ability of different deletion mutants from the owl monkey orthologue to induce the AP-1 and NFκB promoters.

In the laboratory mouse strain C5BL6J, the TRIM5 locus has expanded, giving rise to seven TRIM5 orthologues 26. We aimed to test the capacity of the mouse

65 TRIM5 orthologues to activate the AP-1 and NFκB pathways, and correlate this feature to their ability to restrict a specific retrovirus. For this, we engineered a fusion of the seven TRIM5 orthologues that induce the innate immune signaling with variable strengths, to the Cyclophillin A domain from Aotus trivirgatus’

Trim5-Cyp.

Here we show that murine TRIM5 proteins fused to the Cyclophilin A domain restrict HIV-1 in a manner that is dependent on the binding to the capsid and on the ability to activate the MAPK and NFκB-dependent pathways. Additionally, we found that the ability of murine TRIM5 orthologues to form CBs does not correlate with their capacity to induce the innate immune promoters.

MATERIALS AND METHODS

Drugs, reagents and antibodies.

The TAK-1 inhibitor, 5-Z-7-oxoeaenol and the puromycin drug for the selection of the FUPI-positive CRFK cell lines were purchased from Sigma-Aldrich.

The TAK-1 inhibitor was diluted into dimethylsulfoxyde (DMSO) and used at a concentration of 300 nM in this study. DMSO was then used as a vehicle added to the well of the control condition.

The polyvinylidene difluoride (PVDF) membrane and the β-mercaptoethanol were purchased from Bio-Rad. The ECL Western Blotting Detection Reagents were from GE-Healthcare.

The primary anti-c-Myc, anti-β-actin and anti-GAPDH antibodies from mouse were purchased form Sigma. The secondary anti-mouse antibody was from

Santa-Cruz Biotechnologies.

66 The Protease and Phosphatase Inhibitor Cocktail was from Roche.

Plasmids, vectors and viruses.

The FUPI plasmid is derived from pFUW 27 and carry the Ubiquitin promoter driving the expression of a puromycin resistance cassette followed by ECMV

IRES, as described previously 11.

FUPI three parts virus was obtained by transfecting 293FT cells, plated in 10cm plates, with pFUPI plasmid containing either no insert or different TRIM-Cyp constructs, psPAX2 (gagpol) and pMD2G (envelope) plasmids 11.

The pcDNA3.1(-) plasmid was purchased from Invitrogen and used to clone the different mouse Trim5 orthologues cDNAs.

The different mouse Trim5 orthologues ORFs were obtained by PCR using specific primers and a cDNA template reverse-transcribed from C57BL6J murine embryonic fibroblasts (MEFs)-derived RNA (see primers used in the Supp. Table

1). The plasmid pcDNA3.1(-) containing the different primate Trim5α ORFs were cloned previously in our laboratory 13,28. The feline Trim5 orthologue was cloned from a cDNA template prepared from CRFK-derived total RNA, using specific primers (Supp. Table 1).

HIV-1-GFP three parts virus was prepared by transfecting the 293FT cells in 10 cm plates with pWPTS-GFP plasmid 29, pPAX2 (packaging genes) and pMD2.G

(envelope) plasmid 30.

MLV-GFP three part virus was prepared by transfecting the 293FT cells in 10 cm plates with pLNC-GFP 31, pCG-gagpol and pMD2.G.

The vectors for the retroviral gene expression, the packaging genes and the envelope were transfected at a ratio of 3:2:1.

67 Lipofectamine 2000 (Invitrogen) or polyethilenimine (PEI) (Sigma Inc) were used as transfection agents.

Briefly, 30 μg of DNA were mixed with 60 μl of lipofectmine 2000 or PEI

(1mg/ml) in 1 ml of Opti-MEM (Invitrogen), incubated for 30 minutes and added to the cells.

At 48 hours post-transfection, virus supernatants were harvested.

Cloning into pcDNA3.1(-).

The human TRIM5α , the rhesus monkey TRIM5α and the owl monkey

TRIM5Cyp were previously cloned into the pMIG or pMIP plasmids in our laboratory from human TE671, fetal rhesus monkey kidney (FRhK4) and owl monkey kidney (OMK) cell lines, respectively 5,28 and transferred into pcDNA3.1(-) 13.

The various owl monkey TRIM5Cyp mutants were synthesized by site-directed mutagenesis from the pcDNA3.1(-) TRIM5Cyp construct, using the XbaI 5’

(except for the ΔRF) and NotI 3’ primers from Supp. Table 1 with different combinations of the following internal primers: ΔRF-Xba5’: caactctagagccaccATGCGGATCAGTTACTCGTCT; ΔBB: Forward:

GGGCAGAAGGTTGATCACCACCAGACATTCCTTGTG, Reverse:

CACAAGGAATGTCTGGTGGTGATCAACCTTCTGCCC ; RBCC-Not3’: accagcggccgcCTAGAGCACTCTCACGGACTG; 1-264-Not3’: accagcggccgcTTACTGCAAAGTCACTTTCTCAAT; 1-277-Not3’: accagcggccgcTTAAAATATTCTCCTTTTTTCATTAA; 1-299-Not3’: accagcggccgcCTACCAGTAGCGTTGGACTTC; ΔCypA-Not3’: accagcggccgcCTAGGCTGATGCTACAAGGTCC.

68 Macaca nemestrina TRIM5Cyp was amplified by PCR from the pLPCX TRIM5Cyp plasmid (gift from Theodora Hatziioannou, Aaron Diamond AIDS Research center, New York), using the specific primers from the Suppl. Table 1.

1).

The various murine TRIM5-Cyp fusion constructs were produced by overlapping

PCR from the corresponding pcDNA3.1(-) murine TRIM5 orthologue together with a pcDNA3.1(-) owl monkey TRIM5Cyp template. The primers used where the NheI 5’ for each murine TRIM5 orthologue together with specific internal primers (see below), and the NotI 3’ for the owl monkey TRIM5Cyp (Suppl. Table

1) in combination with the following internal oligoaminoacids: Linker:

TCTGGTGGCGGTGGCTCGGGCGGAGGTGGGTCGGGTGGCGGCGGATCAG; Forward linker-Cyp fusion: GCGGCGGATCA ATGGTCAATCCT; Reverse-linker-Cyp fusion:

AGGATTGACCATTGATCCGCCGC; TRIM12A-linker: Forward: GCTCATCGCTAC

TCTGGTGGCGGT, Reverse: ACCGCCACCAGAGTAGCGATGAGC; TRIM12B-/C- linker: Forward: CGCTACTCTGGTGGCGGTGGCTCG, Reverse:

GCCACCAGAGTAGCGTTGAGCC; TRIM30A-linker: Forward:

GGGAAGCATTACTCTGGTGGCGGT, Reverse: ACCGCCACCAGAGTAATGCTTCCC;

TRIM30B-linker: Forward: GGGATTTGGTCTGGTGGCGGTGGCTCGG, Reverse:

ACCGCCACCAGACCAAATCCCAGGAA; TRIM30C-linker: Forward:

ACGATATTCTGGTGGCGGTGGCTCGGG, Reverse:

ACCGCCACCAGAATATCGTCGGACATA; TRIM30D-linker: Forward:

AGCAATACTCTGGTGGCGGTGGC, Reverse: CACCAGAGTATTGCTGAACATCCA.

69 The feline TRIM5 orthologue was cloned into pcDNA31(-) from a cDNA template prepared from CRFK-derived total RNA, using specific primers (Supp. Table 1).

The different C-terminal Myc-tagged murine TRIM5 orthologues where amplified by PCR of the corresponding pcDNA3.1(-) constructs using the following 3’ read- through primers (to delete the stop codon before the C-terminal tag): TRIM12A: accatgcggccgcGTAGCGATGAGCCTCTGTGAC; TRIM12B: accatgcggccgcAGAGTCTGGC CAGCAAATTGTCATCG; TRIM12C: accatgcggccgcAGAGTCTGGC CAGCAAATTGTCATGG; TRIM30A: accatgcggccgcGGAGGGTGGCCCGCATATAG; TRIM30B: accatgcggccgcCCAAATCCCAGGAAGTAAA; TRIM30C: accatgcggccgcTTCTTTTGACTGTGTTTCCACAG; TRIM30D: accatgcggccgcGGATGGTGGTCCGCATA. The resulting ORFs were cloned into a pcDNA3.1(-)- C terminal Myc tag plasmid 13.

Cloning into pFUPI.

The various murine TRIM5 orthologues and the corresponding Cyp-fusions were subcloned from pcDNA3.1(-) with a NheI/NotI double-digestion into pFUPI digested with XbaI/NotI.

Cell lines.

The HEK-293 cell line was kindly provided by Prof. Walther Mothes (Yale School of Medicine). The C57BL6J MEF cell line was a gift from Dr. Massimo Pizzato

(University of Trento).

70 The human 293 FT and the feline renal fibroblast (CRFK) cell lines were obtained from Invitrogen and ATCC, respectively.

All three lines were maintained in high glucose Dulbecco’s modified Eagle medium (D-MEM) (Invitrogen) supplemented with L-glutamine (2 mM), penicillin (100 units/ml), streptomycin (100 μg/ml) (GIBCO) and FBS (10%)

(PAA).

Cells were passed every 3 days, when reaching 80% of confluency and were diluted out 20 times.

For stable cell lines production, 800 ul of fresh viral supernatant was added to each well of a 6 wells plate, containing 50000 cells per well. The cells were expanded for 4 days without any selection drug. The FUPI-postive CRFK cells were then selected for one week in complete medium containing 2μg/ml of puromycin.

Total RNA extraction and complementary DNA (cDNA) synthesis.

The total RNAs from the MEF and CRFK cell lines were prepared by the lysis of one to five millions of cells and separation from the DNA and proteins using the

RNeasy mini kit from Qiagen. The samples were further treated with RNase inhibitor Ribolock (Fermentas) and subjected to a reverse-transcription PCR, using the M-MLV reverse-transcriptase and RT buffer from Promega. The RNA template (up to 5 μg) was incubated with random hexamers for 5 minutes at

70°C. The second step was the incubation of the RNA-primers solution at 25°C for 10 minutes. Then, the first-strand cDNA synthesis was performed for 50 minutes at 42°C and terminated by incubating the reaction at 70°C for 3 minutes.

Infections with GFP-viruses.

CRFK cells were seeded at 1500 cells per well in a 96 well plate. The next day, 50

μl of HIV-1- or MLV-GFP virus were added at serials three fold dilutions to each well.

After 48 hours, the virus-containing medium was aspirated, the cells washed with Phosphate Buffered Saline (PBS) (Invitrogen) and 50 μl of trypsin was added to each well. The cells were subsequently collected into 1.2 ml tubes

(Scientific Specialties Inc.) and subjected to Fluorescence Activated Cell Sorting

(FACS), to determine the percentage of infected (GFP-positive) cells.

Luciferase assays.

One day prior to transfection, HEK-293 cells were plated on white 96-wells plates (Perkin-Elmer) at a density of 2.5 x 10^4 cells per well.

After 24 hours, cells were transfected with 25 ng of Firefly-luciferase reporter plasmid, 5 ng of the internal control reporter plasmid pRL-TK (Promega) and 25 to 50 ng of the pcDNA3-1(-) plasmid containing a cDNA of interest or the empty plasmid. Lipofectamine 2000 was used as the transfection agent.

Briefly, 30 ng of DNA were mixed with 0.5 μl of lipofectamine 2000 in 100 μl of

Opti-MEM. After 30 minutes of incubation, the reactions were added to the cells.

48 hours post-transfection, the cells were lysed and assayed with the Dual-Glo luciferase system (Promega), according to the manufacturer’s instructions. The luciferase activity of each well was measured using the Veritas Microplate

Luminometer (Turner Biosystem). For each well, the Firefly luciferase activity

72 was normalized to the Renilla luciferase reading and plotted as a fold-induction, compared to the empty pcDNA3.1 (-).

Each condition was done in quintuplets.

The Prl promoter AP-1-luc construct was a gift from Dr. Ruslan Medzhitov (Yale

School of Medicine), and NFκB-luc was kindly provided by Dr. Jurgen Brojatsch

(Albert Einstein College of Medicine).

Treatment of cells with the TAK1 inhibitor.

After 4 to 6 hours of incubation with the transfection complexes, the medium of

HEK293 cells was replaced with medium containing the 5-Z-7-oxozeaenol drug

(300nM final) or dimethylsulfoxide (DMSO, Sigma). As the inhibitory drug is very unstable in a solution, the medium was replaced with fresh drug-containing medium every 12 hours for a total of 36 hours.

SDS-PAGE and Immunobotting.

HEK293 cells were transfected at a confluence of 10^6 cells per well into 6 wells plates with 4 ug of the different pcDNA3.1(-) Myc-tagged constructs. The medium was replaced 6 to 8 hours post-transfection. After 48 hours, the medium was aspirated, the cells washed with ice-cold PBS and lysed with 200 ul of 1 × Laemmli sample buffer (62.5 mM Tris, pH 6.8, 2 % SDS, 10 % glycerol, 357.5 mM 2- mercaptoethanol (2-ME), 0.0025 % bromophenol blue (Bio-Rad)), containing

Protease and Phosphatase Inhibitor Cocktail (EDTA-free) and 5 mM EDTA. The cellular extracts were incubated at 98 °C for 5 min, centrifuged at 14, 000 × g for 2 min, and 20 µl were loaded onto a 10 % polyacrylamide gel. SDS-PAGE was run for

1 hour at 100 V. Proteins were transferred onto an Immuno-Blot polyvinylidene

73 fluoride (PVDF) membrane (Bio-Rad) overnight at 30 V (constant voltage), at 4 °C.

The membrane was subsequently washed 3 times 5 minutes with deionized water and blocked 1 hour at 4 °C in TBST 5% - milk prepared with non-fat dry milk. TBST buffer was prepared with 50 mM Tris, pH 7.4, 150 mM NaCl and 0.1 % Tween 20.

The membrane was then washed 3 times 5 minutes with TBST buffer. The primary antibody was diluted to 1 ug/ml in 5 % TBST-milk and poured above The membrane was incubated with 1µg/ml of the primary antibody diluted in 5% TBST-milk 4 hours at 4°. The membrane was washed 3 times 5 minutes with TBST. The secondary antibody was diluted 1:10,000 in 5% TBST-milk and used to incubate the membrane at room temperature for 1 hour on an agitator. The membrane was washed 3 times 10 minutes and developed using either the ECL Western Blotting Detection Reagents.

The Fujifilm machine was used to expose the membrane for 30 seconds to 30 minutes, depending on the strength of the signal.

Immunofluorescence and imaging of the cytoplasmic bodies.

HEK293 cells were transfected using FUGENE6 (Promega) with 10 ng of myc- tagged murine TRIM5 orthologues expression constructs in a poly-lysine coated

8-well Nunc® Lab-Tek® II 1.5 borosilicate coverglass bottom chambers and fixed after 24-36 h with 4% paraformaldehyde. Murine TRIM5 orthologues-myc were visualized using rabbit monoclonal antibodies to the myc tag (1:2,000 dilution; Sigma-Aldrich), followed by secondary antibodies to rabbit immunoglobulin conjugated to Alexa Fluor 488 (1:5,000 dilution; Invitrogen).

The nuclei were stained with propidium iodide. The images were acquired by using a 60 X objective lens (NA 1.4) on a Volocity spinning disc confocal microscope with z-spacing of 0.15 μm and processed using the Volocity software

74 package 6.3.1 (Perkin-Elmer). Z-Sections were combined to flatten and obtain extended focus image.

RESULTS

Activation of innate immune signaling is a conserved feature of TRIM5 orthologues.

In a previous study, overexpression of human TRIM5α or owl monkey TRIM5Cyp was found to activate luciferase reporters for AP-1 and NF-κB 13. To determine if this activity of TRIM5 is conserved among mammalian orthologues, we first cloned the open reading frames (ORFs) of TRIM5alpha orthologues from three non-human primates, Macaca mulata, Cercopithecus aetiops and Cercopithecus tantalus, as well as the TRIM5Cyp from Macaca nemestrina, into a pcDNA3.1(-) plasmid, which bears a CMV promoter. In addition, the TRIM5 orthologue from cat was cloned.

We then co-transfected human embryonic kidney fibroblast (HEK293) cells with the empty pcDNA3.1(-) or the plasmids containing the different TRIM5 coding sequences with plasmids bearing AP-1 or NFκB promoters directing the expression of a reporter luciferase gene from the firefly beetle Photinus spiralis.

As an internal control for transfection efficiency, we included a third plasmid bearing the thymidine kinase (TK) promoter that is constitutively expressed in eukariotic cells, driving the expression of a reporter luciferase gene coming from the Renilla reniformis cnidarian species. This luciferase enzyme uses another luciferin substrate from the one from firefly to induce bioluminescence, allowing to distinguish between the two reporter genes and successively measure the

Renilla’s luciferase after the Firefly’s.

75 In this assay, all the TRIM5 orthologues that we tested could strongly induce the

AP-1 and NFKB-promoters, when compared to the empty plasmid (figure 1).

This result suggests that the ability of TRIM5 to induce innate immune pathways is a conserved mechanism in primate and in cats.

Differential requirements for the RF and BB domains for human TRIM5α- and owl monkey TRIM5-Cyp-mediated activation of the innate immune signaling.

We previously showed that human TRIM5α required the RF domain to activate the AP-1 promoter, in the luciferase reporter assay, as revealed by the inability of a RF deletion and a point mutant in this domain 13.

When the BB domain is deleted, human TRIM5α is at least 5 fold less able to induce the AP-1 -luciferase reporter (data not shown).

Here, we aimed to examine the effect of deleting particular domains of the owl monkey TRIM5Cyp (Figure 2A).

In this view, we cloned the different deletion mutants in the pcDNA3.1(-) plasmid and transfected it together with the reporter plasmids, as described above.

The results showed that TRIM5Cyp is dispensable from the RF and BB domains to activate the AP-1 promoter (figure 2C).

Conversely, the RBCC protein completely lost the ability to activate the innate immune promoters. Keeping a fragment comprising the L2 region of TRIM5Cyp

(Figure 2B) restored the activity more than 10 fold (Figure 2C).

76 The L2 region was analyzed previously for its contribution to the formation of

CBs and restriction 24. Authors found that the KKPV and RRV motives of rhesus

TRIM5α L2 were critical for both CBs formation and antiviral activity.

Here we show that, at least a part of the first motif is sufficient to weakly rescue the TRIM5Cyp-mediated activation of the AP-1 promoter, but that both motives contributed to a 10 fold-rescue.

Altogether, these data indicate that the requirements for the different domains of a TRIM5 protein vary between the PRY-SPRY- or the CypA-containing orthologues.

TRIM12A, TRIM12B and TRIM12C activate the AP-1- and NFκB-luciferase reporters.

We next wanted to confirm the conservation of the innate immune-activating function of TRIM5 in Mus musculus, because this species contain various TRIM5 orthologues.

Indeed, the syntenic TRIM5 cluster in mouse has expanded by successive duplications of the Trim12 and Trim30 genes, which encode proteins highly similar to human TRIM5 26. In the C57BL6J laboratory strain, the expansion of the Trim5 locus gave rise to seven Trim5-like genes, namely Trim12A, Trim12B,

Trim12C, Trim30A, Trim30B, Trim30C and Trim30D 26.

We cloned the seven mouse TRIM5 orthologues in the pcDNA3.1(-) vector, and we tested them for the induction of AP-1- and NFκB-luciferase reporters by overexpression in HEK293 cells.

77 Trim12A, Trim12B and Trim12C activated AP-1- and NFκB-luciferase promoters as well or better than human TRIM5α when compared to empty pcDNA3.1(-), while Trim30C only activated NFκB (figure 3). Conversely, Trim30A, Trim30B and Trim30D were poorly efficient to stimulate either of the reporters.

The levels of the expression of the seven mouse TRIM5 orthologues do not seem to account for a reduced ability to stimulate the AP-1- and NFκB promoters, as assessed by over-expression of the corresponding Myc-tagged constructs in

293FT cells, with the potential exception of Trim30C (figure 4).

The activation of the MAPK pathway by murine TRIM5 orthologues requires TAK1.

The activation of the AP-1-luciferase by human TRIM5α is dependent on TAK1 activity 13. TAK1 forms a complex with the Tab2 and Tab3 proteins that stimulate the MAPK-dependent pathway.

We wanted to investigate whether the activation of AP-1-luciferase by TRIM5 orthologues from the mouse was also relying on the stimulation of TAK1. Hence, we overexpressed the different murine TRIM5 orthologues in HEK293 cells and then treated them with a TAK1 inhibitor drug, 5-Z-7-oxoeaenol. The AP-1- luciferase activation was assessed as mentioned above.

We found that the activation of AP-1-luciferase by the different Trim5 orthologues was reduced wherever the cells were treated with the TAK1 inhibitor drug but not when they were treated with the vehicle solution (DMSO)

(figure 5).

78 We conclude that the murine TRIM5 orthologues stimulate the AP-1-luciferase activity through TAK1 activity.

TRIM12A-, TRIM12B- TRIM12C- and TRIM30D-Cyp restrict HIV-1.

The mouse Trim5 orthologues tested to date do not confer resistance to HIV-1 when over-expressed in the feline cell line CRFK 26.

In order to test the correlation between the capacity of activating signaling cascades by a given Trim5 orthologue and its ability to restrict a specific retrovirus, we engineered fusion constructs of the RF, BB and CC motives from

TRIM12A and TRIM30A with the Cyclophilin domain of TRIM5-Cyp from Aotus trivirgatus, known to bind the HIV-1 capsid lattice, or with the same Cyclophilin domain bearing the H126Q substitution that impedes this binding.

We cloned the corresponding cDNAs into the FUPI plasmid, which carries a puromycin resistance gene, and used them to generate CRFK stable cell lines expressing the various consructs or the empty FUPI vector.

TRIM12A and TRIM30A by them selves did not restrict HIV-1 in challenged CRFK cells, as assessed by the percentage of GFP-positive cells by the Fluorescent

Activated Cell Sorting (FACS) method.

Interestingly, TRIM12A-Cyclophilin (TRIM12A-Cyp) but not TRIM30A-

Cyclophilin (TRIM30A-Cyp) restricted HIV-1, reaching 2 logs of inhibition. As expected, the cyclophilin domain H126Q mutation abolished the restriction ability of TRIM12A-Cyp (figure 6A).

79 The lack of restriction by TRIM12A-Cyp H126Q wasn’t explained by any drastic reduction of the activation of neither AP-1- nor NFKB-luciferase promoters, as shown in figure 7. The inhibition of HIV-1 infection by TRIM12A-Cyp was not due to the activation of a constitutive antiviral state, as assessed by the examination of NDV and MLV infectivity in stable CRFK cell lines expressing either TRIM12A-

Cyp, TRIM30A-Cyp or puromycin resistance-only (Supp. Figure 1).

These findings indicate that the capacity of TRIM12A-Cyp in blocking HIV-1 requires the binding to its capsid and suggests that the activation of the innate immune signaling is also required for this function.

To confirm the role of the innate immune signaling in the HIV-1 restriction, we tested the effect of the fusions of the other five murine TRIM5 orthologues.

Again, TRIM12B, TRIM12C, TRIM30B, TRIM30C and TRIM30D were not able to block HIV-1 infection with, whenever present, their corresponding C terminal domains. However, the Cyp fusion versions of TRIM12B, TRIM12C and TRIM30D conferred a strong inhibition of HIV-1 to the CRFK cells they were expressed in

(figures 6B and 6C).

Given that TRIM30D does not strongly activate the AP-1 or NFκB pathways per se, we tested the ability of the Cyp-fused proteins to induce the innate immune promoters. Interestingly, among all the other TRIM30-Cyp constructs, TRIM30D-

Cyp was the more efficient to activate the AP-1 promoter (figure 7) and had a modest effect on the NFκB promoter. This finding would explain the fact that this artificial protein is competent for retroviral restriction.

Comparison of the protein levels between Myc-tagged murine TRIM5-Cyp constructs showed that there was no correlation between the protein levels and the ability to restrict (Supp. Figure 2).

80 We conclude that the restriction of a specific retrovirus by a given TRIM5 orthologue does not only involve the binding to a specific capsid but also requires that the factor is efficient enough in activating the MAPK- and NFκB- dependent innate immune signaling pathways.

The ability of a TRIM5 orthologues to form cytoplasmic bodies does not dictate their capacity to activate the MAPK/NFKB pathways.

Upon ectopic expression, TRIM5α concentrates to cytoplasmic aggregates called cytoplasmic bodies (CBs) 4,32.

The functional significance of this particular protein compartmentalization has been debated. Notably, studies analyzing the ability of different TRIM5α orthologues to restrict HIV-1 suggested that the presence of such bodies, in conditions where the virus was absent, did not represent a pre requisite for the capability to block the retrovirus when it entered the cell 33,34.

However, another team showed that, upon infection, rhesus TRIM5α interacted with HIV-1 virions and formed de-novo cytoplasmic bodies around the viral particles 35, arguing for a functional role of the formation of TRIM5α concentrations in the subsequent retroviral restriction.

In order to evaluate the role of CBs formation in TRIM5-mediated activation of

AP-1-luciferase reporter, we performed Immunofluorescence imaging on 293T cells expressing Myc-tagged versions of each murine TRIM5 orthologues and correlated these observations to their ability to induce the innate immune promoter.

As indicated in Table 1 and shown in Supplementary figure 3, all the murine

TRIM5 orthologues formed CBs, except TRIM30B and TRIM30C. The fact that

81 TRIM30A does not induce strongly AP-1-luciferase and still forms CBs indicated that the ability of a given TRIM5 to concentrate into cytoplasmic aggregates is not sufficient to activate innate immune signaling. Reciprocally, TRIM30C does not form CBs but induces the AP-1 promoter more than 10 fold, suggesting that the formation of these protein aggregates are not required to activate the innate immune signaling.

These data showed that there was no correlation between the formation of CBs by TRIM5 proteins and their ability to induce innate immune promoters.

DISCUSSION

The restriction factors responsible for the blockade of the N-tropic MLV in human and in the mouse are TRIM5α and Fv1, respectively. These factors have very different structures, the former being a TRIpartite Motif family member, the latter being derived from a gag gene 36.

In 2009, Tareen et al. depicted, along with the previously described Trim30, six other

Trim5 orthologues, resulting from duplications of the Trim12 and Trim30 genes 26.

Here we demonstrate that three out of the seven murine TRIM5 orthologues have an intrinsic ability to stimulate the innate immune signaling and that this feature correlates to the capacity of restricting a retrovirus like HIV-1.

Which are the evolutionary pressures that induced the laboratory mouse to undergo such an expansion of the Trim5 locus is still unknown. Restriction of Moloney, N- tropic or B-tropic MLV variants by the mouse Trim5 orthologues could not be demonstrated, as well as the retriction of HIV-1, HIV-2, SIV, FIV and EIAV, which are targets of different primate TRIM5 orthologues (this study and 26). Nonetheless,

82 with the exception of Trim12A and Trim30B, all mouse Trim5 genes possess a

PRYSPRY domain that could potentially recognize a retroviral capsid.

We previously reported an E3-ligase dependency for TRIM5-mediated restriction 13.

Indeed, the central components for the innate immune activation, TAK1 and the E2 and E1 ubiquitination enzymes, were pinpointed to be essential for TRIM5Cyp- mediated restriction of HIV-1 13.

Conversely, other studies show data arguing for a dispensable role of the E3-ligase function from TRIM5 proteins in the retroviral restriction process, by using deletion mutants or by depleting endogenous E1ubiquitin –activating enzyme 25,37. However, in the study by Diaz-Griffero and colleagues, TRIM5Cyp did not require the RF or the BB domains, but rhesus TRIM5α did 25. Their data did not include any correlation with the activation of the innate immune signaling by the different deletion mutants of

TRIM5Cyp 25. In this report, we show that human TRIM5α and TRIM5Cyp from the owl monkey also have different requirements of their domains for their ability to signal. The findings that the RF and BB are dispensable for the activation of the innate immune promoters by TRIM5Cyp tested in the present study suggest that other domains from this protein orthologue could account for this feature, as for example the presence of the L2 region or another sequence between the RBCC and the CypA domain. This would explain why Diaz-Griffero and colleagues 25 observed an independency of TRIM5Cyp from the two N-terminal domains for HIV-1 restriction.

The discrepancy between our previous findings and the study from Perez-

Caballero and colleagues 37 could be accounted by the usage of cell lines from primate and rodents, respectively, that could express different factors regulating

83 TRIM5-mediated restriction and that would be differentially sensitive to ubiquitination.

Nonetheless, altogether, our results indicate that the ability of a given TRIM5 to activate the innate immune signaling correlates with its capacity to restrict a TRIM5- sensitive retrovirus.

Additionally, we found that the ability of the various murine TRIM5 orthologues to form cytoplasmic bodies did not correlated with their capacity to induce the innate immune signaling.

Interestingly, most of the murine TRIM5 orthologues that activated the innate immune promoters and formed CBs could restrict HIV-1 when fused to CypA. The only exception was TRIM30D that did not activated AP-1-luciferase but still restricted HIV-1 in the CypA-fusion version. However, TRIM30D-CypA induced more strongly the innate immune promoter than all the other TRIM30-Cyp constructs.

These observations suggest that, with the obvious condition of carrying a capsid- recognition domain, the double capacity to robustly induce the innate immune signaling and CBs formation by a given TRIM5 protein could dictate its restriction ability. It is interesting to note that the L2 region that was attributed to CBs formation and retroviral restriction 24 is not strictly required for AP-1-luciferase activation by the owl monkey TRIM5Cyp (Figure 2), indicating that the induction of the innate immune signaling does not require that TRIM5α concentrate into CBs. This emphasizes that the two features are separable.

Furthermore, the fact that TRIM30C activates the NFκB-promoter, comparable to human TRIM5α and murine TRIM12B, and that AP-1 activation by TRIM30D-Cyp is

84 stronger than for the other innate immune promoter, points toward a more important role for the MAPK pathway in the TRIM5-mediated retroviral restriction. Further studies will be necessary to determine which components of the MAPK pathway are playing an important role in retroviral restriction by TRIM5 and how these actors orchestrate the blockades of one or both of the steps of the viral replication targeted by the restriction factor.

FIGURES AND TABLES Name of the AP-1-luc CBs formation HIV-1 construct induction restriction by the corresponding TRIMCyp construct TRIM12A ++++ YES YES TRIM12B ++ YES YES TRIM12C +++ YES YES TRIM30A / YES NO TRIM30B / NO NO TRIM30C + NO NO TRIM30D / (TRIM30D- YES YES Cyp:++)

TABLE 1:The comparison of the abilities of the different murine TRIM5 orthologues at activating the innate immune signaling, forming cytoplasmic bodies and restricting HIV-1. The following codes were used for the different AP-1-luciferase (AP1-luc) fold-activation: ++++: more than 100 fold. +++: more than 45 fold. ++: more than 35 fold. +: more than 10 fold. /: less than 10 fold. As a comparison, human TRIM5α activated more than 45 fold.

86 !"# $"# AP-1-luciferase NFκB-luciferase

1000 100 e e s s a a e e r

r 100 c c n i n i

- 10 - d l d l

10 o o F F

1 1

Ct At Ct At Fc Hs Ca Mn Fc Hs Mm Ca Mn Mm

MEKK-1 MEKK-1

pcDNA 3.1 (-) pcDNA 3.1 (-)

FIG. 1. Induction of AP-1- and NFκB-luciferase reporters by simian and feline

TRIM5 orthologues. The empty plasmid pcDNA3.1(-) is used as a negative control and the MAPK pathway-activator MEKK-1 as a positive control. A) The AP-1- luciferase activity was measured. B) The NFκB-luciferase stimulation was assessed.

Hs: Homo sapiens. Mm: Macaca mulata. Ct: Cercopithecus tantalus. Ca:

Cercopithecus aetiops. At: Aotus trivirgatus. Mn: Macaca nemestrina. Fc: Felis catus.

87 !"#

,-* .*/%0* !%($12*!%($* &'()*+#,# !"#$%&'($()*+*

$"#

%"# AP-1-luciferase e

s 100 a e r c n i -

d 10 l o F

RF BB WT ! ! CypA RBCC ! aa1-265aa1-277aa1-299 pcDNA 3.1 (-)

FIG 2. The domains required for the stimulation of the AP-1 promoter by

TRIM5Cyp. A) A schematic representation of the domain composition of TRIM5-

Cyp. B) The alignment of the Linker 2 regions of Macaca mulata (Mm, rhesus monkey), human (Hs) and Aotus trivirgatus TRIM5 orthologues. The blue squares indicate the residues that were attributed previously to to the abilities to form cytoplasmic bodies (CBs) and restrict (Sastri, 2010, Virology). C) The AP-1- luciferase activity was measured, as described above. WT: the wild-type owl monkey

TRIM5Cyp. ΔRF: RF deletion mutant. ΔBB: BB deletion mutant. RBCC: deletion of the C terminal part of TRIM5Cyp, not including the CC. 1-264, 1-277 and 1-299: mutants with increasing portions of the segment connecting the RBCC with the CypA domain of TRIM5Cyp, including the L2 region. ΔCypA: Cyclophilin A deletion

88 mutant. The orange, blue and green stars indicate the last amino acid included in the aa1-265, aa1-277 and aa1-299 constructs, respectively.

!"# $"# AP-1-luciferase NFKB-luciferase 150 40 e e s s 30 a 100 a e e r r c c n n 20 i i - - d d l l 50 o o 10 F F

0 0

hT5 hT5 T12A T12B T12C T30A T30B T30C T30D T12A T12B T12C T30A T30B T30C T30D

pcDNA3.1(-) pcDNA3.1(-) TRIM5 orthologues TRIM5 orthologues

FIG. 3. The ability of murine TRIM5 orthologues to activate the AP-1- and

NFκB-luciferase reporters. The empty plasmid pcDNA3.1(-) is used as a negative control and human TRIM5α as a positive control. A) The AP-1-luciferase activity was measured. B) The NFκB-luciferase stimulation was assessed. hT5: human TRIM5α. T12A-T30D: TRIM12A-TRIM30D.

89 *(+,# *(+-# *(+.# *'/,# *'/-# *'/.# *'/%#

!"#$%&#

'!#$%&#

()#$%&#

0&1234#

FIG 4. The different murine TRIM5 orthologues have similar protein levels.

Comparison of the expression of the different mouse TRIM5 proteins, by over- expressing C terminal- Myc constructs in 293FT cells. The cell lysates were subjected to an SDS-PAGE and immunoblotted with anti-c-Myc antibody. Expected molecular weights (approximately): T12A: 37kD, T12B, T12C, T30A, T30C and T30D: 57kD and T30B: 17kD.

FIG. 5. The ability of the TRIM12 proteins to activate the AP-1-luciferase reporter is dependent on TAK-1. The empty plasmid pcDNA3.1(-) is used as a negative control. The TAK1 inhibitor 5-Z-7-oxoeaenol was used at 300 nM, as described in the Materials and method section. The AP-1-luciferase activity was measured.

91 !"# $"# HIV-1 in CRFK cells HIV-1 in CRFK cells

102 102 s l

1 l s

l 10 e l c e 1 10 c e

v e i v t

i FUPI(-) i t i FUPI (-) s

s 0 10 o TRIM12B o TRIM12A p - p -

P TRIM12B-Cyp P

TRIM12A-Cyp F 0

F 10

G TRIM12C

TRIM12A-Cyp H136Q

-1 % % TRIM12C-Cyp 10 TRIM30A TRIM30A-Cyp

TRIM30A-Cyp H136Q 10-1 10-2 0.001 0.010 0.100 1.000 0.01 0.10 1 Dilution of viral supernatant Dilution of viral supernatant %"# HIV-1 in CRFK cells 102

101

s TRIM12A-Cyp l l

e TRIM30A-Cyp c

e TRIM30B v i

t TRIM30B-Cyp i 0 s 10 TRIM30C o p

- TRIM30C-Cyp P

F TRIM30D

G TRIM30D-Cyp %

10-1

10-2 0.001 0.010 0.100 1 Dilution of viral supernatant

FIG. 6. Artificial TRIM5Cyp constructs differentially restrict HIV-1. A)

Restriction of HIV-1 by the TRIM12A-Cyp fusion is dependent on the Cyclophilin A domain binding to the HIV-1 capsid. B) TRIM12B- and TRIM12C-Cyp fusions are able to restrict HIV-1. D) TRIM30D-Cyp blocks HIV-1. CRFK cells transduced with the different constructs were challenged with an HIV-GFP virus bearing the VSVg envelope. The percentage of GFP-positive cells was determined by FACS.

92 !"# $"# AP-1-luciferase NFκB-luciferase

100 e e

s 10 s a a e e r

r 10 c c n n i i - -

d 1 d l l 1 o o F F

0.1 0.1

TRIM12A TRIM30A TRIM12A TRIM30A pcDNA3.1(-) pcDNA3.1(-) TRIM12A-Cyp TRIM30A-Cyp TRIM12A-Cyp TRIM30A-Cyp

TRIM12A-Cyp H136QTRIM30A-Cyp H136Q TRIM12A-Cyp H136QTRIM30A-Cyp H136Q %"# AP-1-luciferase &"# NFκB-luciferase 100 e e s s 100 a a 10 e e r r c c n n i i - - d d 10 l l 1 o o F F

1 0.1

TRIM12BTRIM12CTRIM30BTRIM30CTRIM30D TRIM12BTRIM12CTRIM30BTRIM30CTRIM30D

pcDNA 3.1TRIM12B-Cyp (-) TRIM12C-CypTRIM30B-CypTRIM30C-CypTRIM30D-Cyp pcDNA 3.1TRIM12B-Cyp (-) TRIM12C-CypTRIM30B-CypTRIM30C-CypTRIM30D-Cyp

FIG 7. The murine TRIM5-Cyp constructs have differential abilities to induce

the AP-1 and NFκB promoters. A) and B) TRIM12A-Cyp, TRIM30A-Cyp and the

corresponding H136Q mutants were tested for AP-1- and NFκB-luciferase

activation.C) and D) TRIM12B-, TRIM12C- and the various TRIM30-Cyp constructs

were assayed for induction of the AP-1 and NFκB promoters.

REFERENCES

1. Stremlau, M., Perron, M., Welikala, S. & Sodroski, J. Species-specific variation in the B30.2(SPRY) domain of TRIM5alpha determines the potency of human immunodeficiency virus restriction. J. Virol. 79, 3139– 3145 (2005).

93 2. Anderson, J. L. et al. Proteasome Inhibition Reveals that a Functional Preintegration Complex Intermediate Can Be Generated during Restriction by Diverse TRIM5 Proteins. J. Virol. 80, 9754–9760 (2006). 3. Sebastian, S. & Luban, J. TRIM5alpha selectively binds a restriction- sensitive retroviral capsid. Retrovirology 2, 40 (2005). 4. Stremlau, M. et al. The cytoplasmic body component TRIM5alpha restricts HIV-1 infection in Old World monkeys. Nature 427, 848–853 (2004). 5. Sayah, D. M., Sokolskaja, E., Berthoux, L. & Luban, J. Cyclophilin A retrotransposition into TRIM5 explains owl monkey resistance to HIV-1. Nature 430, 569–573 (2004). 6. Diaz-Griffero, F. et al. Modulation of retroviral restriction and proteasome inhibitor-resistant turnover by changes in the TRIM5alpha B-box 2 domain. J. Virol. 81, 10362–10378 (2007). 7. Sawyer, S. L., Wu, L. I., Emerman, M. & Malik, H. S. Positive selection of primate TRIM5alpha identifies a critical species-specific retroviral restriction domain. Proc Natl Acad Sci USA 102, 2832–2837 (2005). 8. Nisole, S., Stoye, J. P. & Saïb, A. TRIM family proteins: retroviral restriction and antiviral defence. Nat Rev Microbiol 3, 799–808 (2005). 9. Braaten, D., Ansari, H. & Luban, J. The hydrophobic pocket of cyclophilin is the binding site for the human immunodeficiency virus type 1 Gag polyprotein. J. Virol. 71, 2107–2113 (1997). 10. Saurin, A. J., Borden, K. L., Boddy, M. N. & Freemont, P. S. Does this have a familiar RING? Trends Biochem Sci 21, 208–214 (1996). 11. Neagu, M. R. et al. Potent inhibition of HIV-1 by TRIM5-cyclophilin fusion proteins engineered from human components. J Clin Invest 119, 3035– 3047 (2009). 12. Borden, K. L., Campbelldwyer, E. J., Carlile, G. W., Djavani, M. & Salvato, M. S. Two RING finger proteins, the oncoprotein PML and the arenavirus Z protein, colocalize with the nuclear fraction of the ribosomal P proteins. J. Virol. 72, 3819–3826 (1998). 13. Pertel, T. et al. TRIM5 is an innate immune sensor for the retrovirus capsid lattice. Nature 472, 361–365 (2011). 14. Lorick, K. L. et al. RING fingers mediate ubiquitin-conjugating enzyme (E2)-dependent ubiquitination. Proc Natl Acad Sci USA 96, 11364–11369 (1999). 15. Uchil, P. D. et al. TRIM protein-mediated regulation of inflammatory and innate immune signaling and its association with antiretroviral activity. J. Virol. 87, 257–272 (2013). 16. Hershko, A. & Ciechanover, A. The ubiquitin system. Annu Rev Biochem 67, 425–479 (1998). 17. Mukhopadhyay, D. & Riezman, H. Proteasome-independent functions of ubiquitin in endocytosis and signaling. Science 315, 201–205 (2007). 18. Li, X. & Sodroski, J. The TRIM5 B-Box 2 Domain Promotes Cooperative Binding to the Retroviral Capsid by Mediating Higher-Order Self- Association. J. Virol. 82, 11495–11502 (2008). 19. Li, X., Song, B., Xiang, S.-H. & Sodroski, J. Functional interplay between the B-box 2 and the B30.2(SPRY) domains of TRIM5alpha. Virology 366, 234– 244 (2007). 20. Diaz-Griffero, F. et al. A B-box 2 surface patch important for TRIM5alpha

94 self-association, capsid binding avidity, and retrovirus restriction. J. Virol. 83, 10737–10751 (2009). 21. Reymond, A. et al. The tripartite motif family identifies cell compartments. EMBO J 20, 2140–2151 (2001). 22. Mische, C. C. et al. Retroviral restriction factor TRIM5alpha is a trimer. J. Virol. 79, 14446–14450 (2005). 23. Javanbakht, H. et al. The ability of multimerized cyclophilin A to restrict retrovirus infection. Virology 367, 19–29 (2007). 24. Sastri, J. et al. Identification of residues within the L2 region of rhesus TRIM5alpha that are required for retroviral restriction and cytoplasmic body localization. Virology 405, 259–266 (2010). 25. Diaz-Griffero, F. et al. Comparative requirements for the restriction of retrovirus infection by TRIM5 alpha and TRIMCyp. Virology 369, 400–410 (2007). 26. Tareen, S. U., Sawyer, S. L., Malik, H. S. & Emerman, M. An expanded clade of rodent Trim5 genes. Virology 385, 473–483 (2009). 27. Lois, C., Hong, E. J., Pease, S., Brown, E. J. & Baltimore, D. Germline transmission and tissue-specific expression of transgenes delivered by lentiviral vectors. Science 295, 868–872 (2002). 28. Sebastian, S., Sokolskaja, E. & Luban, J. Arsenic counteracts human immunodeficiency virus type 1 restriction by various TRIM5 orthologues in a cell type-dependent manner. J. Virol. 80, 2051–2054 (2006). 29. Zufferey, R. et al. Self-inactivating lentivirus vector for safe and efficient in vivo gene delivery. J. Virol. 72, 9873–9880 (1998). 30. Berthoux, L. et al. As2O3 Enhances Retroviral Reverse Transcription and Counteracts Ref1 Antiviral Activity. Journal of … (2003). 31. Cowan, S. et al. Cellular inhibitors with Fv1-like activity restrict human and simian immunodeficiency virus tropism. Proc Natl Acad Sci USA 99, 11914–11919 (2002). 32. Campbell, E. M. et al. TRIM5 alpha cytoplasmic bodies are highly dynamic structures. Mol. Biol. Cell 18, 2102–2111 (2007). 33. Perez-Caballero, D., Hatziioannou, T., Yang, A., Cowan, S. & Bieniasz, P. D. Human tripartite motif 5alpha domains responsible for retrovirus restriction activity and specificity. J. Virol. 79, 8969–8978 (2005). 34. Song, B. et al. TRIM5alpha association with cytoplasmic bodies is not required for antiretroviral activity. Virology 343, 201–211 (2005). 35. Campbell, E. M., Perez, O., Anderson, J. L. & Hope, T. J. Visualization of a proteasome-independent intermediate during restriction of HIV-1 by rhesus TRIM5alpha. 180, 549–561 (2008). 36. Best, S., Le Tissier, P., Towers, G. & Stoye, J. P. Positional cloning of the mouse retrovirus restriction gene Fv1. Nature 382, 826–829 (1996). 37. Perez-Caballero, D., Hatziioannou, T., Zhang, F., Cowan, S. & Bieniasz, P. D. Restriction of human immunodeficiency virus type 1 by TRIM-CypA occurs with rapid kinetics and independently of cytoplasmic bodies, ubiquitin, and proteasome activity. J. Virol. 79, 15567–15572 (2005).

95 SUPPLEMENTARY FIGURES AND TABLES

Gene Forward Primer (NheI 5’/XbaI 5’) Reverse Primer (NotI 3’) name TRIM5Cyp caactctagagccaccATGGCTTCCA accagcggccgcTTAAAGTTGTCC (Aotus GAATCCTGGT ACAGTCAGC trivirgatus)

Gene ID: 50301247

AY646198

TRIM5Cyp caactctagagccaccATGGCTTCTG accagcggccgcTTATTCGAGTTG (Macaca GAATCCTGGTTA TCCACAGTCAGC nemestrin a)

TRIM5 caactctagagccaccATGGCTTCTG accagcggccgcTCAAGAGCTTG (Cercopith GAATCCTGCTT GTGAGCACA ecus tantalus)

AY593973. 2 TRIM5 caactctagagccaccATGGCTTCTG accagcggccgcTCAAGAGCTTG (Cercopith GAATCCTGGTT GTGAGCACA ecus aetiops)

AY669399. 1 Trim12A caacgctagcgccaccATGGCTTCA accagcggccgcCTAGTAGCGAT CAATTCATGAAGAA GAGCCTCTGTGAC Gene ID: 76681

NM_02383 5.2 Trim12B caacgctagcgccaccATGGCTTCA accagcggccgcTTAAGAGTCTGG « Trim5 » CAATTCATGAAGA C CAGCAAATTGTCATC

Gene ID: G 667823

XM_99271 4.5 Trim12C caacgctagcgccaccATGGCTTCA accagcggccgcTTAAGAGTCTGG CAATTCATGAAGA C CAGCAAATTGTCATG Gene ID: 319236 G

NM_00114 6007.1 Trim30A caacgctagcgccaccATGGCCTCA accagcggccgcTTAGGAGGGTG TCAGTCCTGGAGATG GCCCGCATATAG Gene ID: 20128

NM_00909 9.2 Trim30B caacgctagcgccaccATGGCCTCA accagcggccgcTCACCAAATCCC TCAATCCTCG AGGAAGTAAA Gene ID: 244183

NM_17564 8.2 Trim30C caacgctagcgccacc accagcggccgcTCATTCTTTTGA ATGATGGCCTCCT CAGCTCT CTGTGTTTCCACAG Gene ID: 434219

XM_48598 0.5 Trim30D caacgctagcgccaccATGGCCTCA accagcggccgcTTAGGATGGTG TCAGTCCTGGAGAT GTCCGCATA Gene ID: 209387

NM_00116 7828.1 TRIM5 caactctagagccaccATGGCTTCTG accagcggccgcTCAGTGTGGAAT (Felis AACTCCTGAAA CACGTGAGC catus)

Gene ID: 100302544

NM_00116 3659.1

Supp. Table 1. The primers used in this study. The sequences for the forward (5’) and reverse (3’) primers are indicated in the corresponding line for each murine and feline gene. Names and alternative names, as well as, whenever available, the GeneID and the mRNA accession numbers are indicated.

97 !"# NDV in CRFK cells

101 s l l e c

e v i

t FUPI(-) i

s 100 o TRIM12A-Cyp p -

P TRIM30A-Cyp F G

10-1 0.001 0.010 Dilution of viral supernatant

$"# MLV in CRFK cells

102 s l l e c

e v i

t FUPI(-) i

s 101 o TRIM12A-Cyp p -

P TRIM30A-Cyp F G

100 0.001 0.010 0.100 Dilution of viral supernatant

Supp. Figure 1. The effect of TRIM12A-Cyp on HIV-1 infection is not due to the induction of a general antiviral state. CRFK cells were challenged with A)

NDV-GFP or B) MLV-GFP virus. The percentage of GFP-positive cells was analyzed by FACS.

98 -./01234#52678&99:;#6<=48.>684#/=#(?)-#6:@@4#

'# (# )# *# +# !# ,#

!"#$%&#

ABC%D#

Supp. Figure 2. The absence of a correlation between a difference in the protein level and a decreased ability to restrict by murine TRIM5-Cyp proteins. HEK293 cells were transfected with Myc-tagged murine TRIM5-Cyp constructs. The cell lysates were subjected to an SDS-PAGE and immunoblotted with anti-c-Myc antibody. 1: TRIM12A-Cyp. 2: TRIM12B-Cyp. 3: TRIM12C-Cyp. 4:

TRIM30A-Cyp. 5: TRIM30B-Cyp. 6: TRIM30C-Cyp. 7: TRIM30D-Cyp. The different expected bands are indicated with red arrows in each case. The Glyceraldehyde 3- phosphate dehydrogenase (GAPDH) enzyme is ubiquitously expressed in eukaryotic cells and was used as a loading control.

99 !"#$% !"#&% !"#'%

!()$% !()&% !()'% !()*%

Supp. Figure 3. The visualization of cytoplasmic bodies formation by murine

TRIM5 orthologues. Examples of CBs and NBs are depicted with light-blue and white arrows, respectively. The nuclei were stained with propidium iodide. The various TRIM5-Myc constructs were detected with anti-Myc antibody, following by rabbit immunoglobulin conjugated to Alexa Fluor 488. The white bar corresponds to 17 μm.

100

2.2 Investigation of the role of murine TRIM5 orthologues as natural restriction factors.

In the previously mentioned study from Baumann and colleagues, the murine thymoma-derived T cell line, TA3, was found to block HIV-1 infection at the reverse-transcription step and before nuclear entry 183. Given that the mouse genome carries various TRIM5 orthologues, it is conceivable that the anti-HIV-1 phenotype in TA3 cells represents a novel Lv1- like activity. Recently, it was found that murine TRIM5 orthologues were unable to restrict HIV-1, among other retroviruses tested 298. This study did not examine, however, the cases of TRIM12A, TRIM12B and TRIM30C. Furthermore, the murine TRIM5 proteins where stably expressed in CRFK cells and thus it was possible that essential murine cofactors were missing or that the endogenous feline TRIM5 could exert a dominant-negative effect, similar to other TRIM5 proteins 267,310,320. In order to investigate the involvement of murine TRIM5 proteins as natural restriction factors, we used two different strategies. We first wished to repeat the results from the study by Tareen and colleagues and include the analysis of the non-tested TRIM12A, TRIM12B and TRIM30C for their ability to restrict HIV-1, as well as examining a cell line that does not express a functional TRIM5. Thus, in addition to use CRFK cells, we also transduced the canine cell line CF2Th, because the genome from the Canis lupus familiaris carries a disrupted TRIM5 ORF. None of the murine TRIM5 orthologues that we tested could confer anti-HIV-1 activity to CRFK or CF2Th cell lines. The murine genome carries multiple copies of functional endogenous MLVs (eMLVs). We wanted to investigate a potential role of murine TRIM5 orthologues in the defense against these retroviruses. As TRIM5-mediated restriction is capsid-dependent, we wanted to design an MLV-based plasmid with restriction sites that would allow the convenient sub-cloning of the equivalent of a 244 amino acids-long portion of different capsids of eMLVs into the gag gene of Moloney MLV (MoMLV). To this end, we performed site-directed mutagenesis on

101 a MoMLV-based vector called pCG gagpol, in order to create unique 5’ PstI and 3’ SalI sites, without affecting the resulting amino acid sequence.

The resulting modified pCG gagpol was called pCG PstI/SalI and was equally efficient to produce infectious virus compared to the unmodified version when used in combination with the vectors for GFP-genome (pLNCG) and envelope (pMD2.G). The newly designed plasmid was used to clone a corresponding portion of the capsid gene of a TA3 cells-derived MLV transcript, as well as that of N-MLV and B-MLV. The specific pattern of N- versus B-MLV restriction by human TRIM5α was confirmed. No murine or human cell line tested was found to restrict the TA3 cells-derived capsid.

As previously discussed in this this chapter, the murine TRIM12 family members all activate the MAPK pathway in contrast to the TRIM30 proteins. In addition, TRIM12B and TRIM12C carry a C-terminal PRYSPRY domain. We therefore hypothesized that if the murine TA3 cell line indeed possesses an Lv1-like activity, the best candidates from murine TRIM5 orthologues to fulfill this function are TRIM12B and TRIM12C. In a second approach, we aimed to down-regulate the expression of the corresponding murine genes in TA3 cells. As the murine T cell line restricts HIV-1 and expresses the puromycin resistance and Cyclin T1 genes simultaneously, we therefore planned to use an MLV-based retroviral vector driving the expression of the blasticidin resistance (BasltiR) and a micro RNA (miRNA)-based short hairpin RNA (shRNA) composed of two almost complementary 22 nucleotides-long sequences with one unique mismatch, spaced by a linker sequence and targeting specifically the mRNA of a gene of interest (figure 6). We took the advantage that the recombinant Murine Stem Cell Virus (MSCV) vector infect murine T cells efficiently 321 and combined a 3’UTR deleted plasmid with a murine Cyclophilin (Cyp) Promoter-driving-GFP expression vector, in order to have the strong murine promoter in the same time than reducing the well established retroviral 3’UTR-mediated transcriptional interference 321-323. The resulting plasmid was called pMIC ΔU3-GFP. The Cyp promoter-driven GFP expression cassette from the second vector was replaced by the bi-cistronic BlastiR/ miRNA30-based shRNA (mir30) by the usage of

102 specific restriction sites for cloning (figure 6), resulting in a pMIC ΔU3-BlastiR- miR30 vector. In parallel, we designed specific TRIM12A/B/C- targeting sequences on the miR30 background. The pMIC ΔU3 vector expressing the GFP ORF was tested and found to infect five to seven fold more efficiently the TA3 cell line than the HIV-1 based vector pAGM. We conclude that the pMIC ΔU3-GFP vector can be used to efficiently transduce TA3 cells. In contrast, we cannot conclude on the efficiency of the expression of the pMIC ΔU3-BlastiR-miR30, as a gene marker version of this construct was not obtained. Further experiments would be needed to test the efficiency of the gene silencing plasmid that we designed. In parallel, we used an HIV-1-based vector, pAPM, to clone the different target sites we designed. We tested two different target sites in the case of Trim12A and Trim12C, and one target site for Trim12B. Surprisingly, Trim12C ts2 rescued HIV-1 transduction 3-7 fold compared to the control (Scrambled, Scr) KD. The quantification of the relative mRNA levels by RT-PCR, confirmed that whereas the Trim12C target site 2 cell line contained nearly 10 fold less Trim12C transcripts than the Scr KD, the target site 1 cell line was modestly affected. These results suggest that the reduction of TRIM12C mRNA levels coincides with an impaired blockade to HIV-1 in TA3 cells. As the rescue of HIV-1 levels as compared to that in MEF cells was not complete, we propose two hypotheses. In the first one, the KD efficiency could be low, due to the usage of an HIV-1 based vector and given that the puromycin resistance is conferred by a previous selection (puroR-CycT1 vector). In the second hypothesis, TRIM12C could contribute to HIV-1 restriction but other factors could be sufficient for retroviral blockade. In the future, the silencing of the three different TRIM12s could be examined, using the pMIC ΔU3-BlastiR-miR30 vector, in order to gain insights into their roles as natural murine restriction factors. In addition, the cloning of different MLV-derived capsid sequences into the pCG PstI/SalI plasmid that we designed could be used to test a potential capsid-specific restriction phenotype.

103 Materials and methods

Chemicals, reagents and drugs.

The agarose, the Tris-HCl, the acetic acid and the EDTA were from Sigma-Aldrich and used to prepare the 10X TAE buffer (400 mM Tris-HCl, 200 mM acetic acid and 10 mM EDTA) and agarose gels. The TAE buffer was diluted to 1X to dissolve the agarose in order to prepare agarose gels, and to run nucleic acids by electrophoresis. All the restriction enzymes were purchased from New England Biolabs (NEB). The puromycin antibiotic was from Sigma-Aldrich.

Cell lines.

The human fetal kidney epithelial 293FT cells (ATCC) and HeLa cell lines, the murine MEF cells , the feline renal epithelial CRFK cells and the canine new born thymus-derived CF2Th cell line were maintained in Dulbecco’s modified Eagle medium (DMEM) from Invitrogen, containing 10% fetal calf serum (FBS), 20mM L-glutamine, 1000 units/ml penicillin and 1g/ml of streptomycin. When reaching 80% confluency, they were diluted 20 times into new 10 cm plates. The murine TA3 cell line derived from a thymoma of p53-defficient C5BL/6J x 129/SV6 cross mice was a gift from Dr. KewalRamani (National Cancer Institute, Frederick, Maryland, USA). TA3 cells were maintained in Roswell Park Memorial Institute (RPMI) medium-1640 from Invitrogen, supplemented as DMEM, and grown into 12.5 cm2 flasks (BD Falcon).

Viruses.

Two-parts and three-parts-derived HIV-1-GFP viruses were produced by co- transfection of NL4.3-GFP env- and MD2.G plasmids, and pWPTS-GFP, psPAX2 and MD2.G plasmids, respectively. The MLV-GFP three parts viruses were produced by co-transfection of pLNCG, pCG gagpol vectors and MD2.G.

104 Briefly, 3x106 293FT cells were plated in 10 cm plates (BD Biosciences) one day prior to transfection. The different plasmids for two- or three-parts viruses were transfected with lipofectamine (Life Technologies) or Polyethylenimine (PEI). For two-parts 2 viruses, NL4.3-GFP env- and MD2.G were transfected at a ratio of 2:1. The mixes containing the DNA and the transfection agent were incubated in Opti-MEM for 30 minutes prior to be added to the cells. For the three-parts viruses, the GFP-genome was transfected together with the packaging genes-carrying plasmid and the Vesicular stomatitis virus G protein envelope-coding genes (MD2.G), at a ratio of 3:2:1, respectively. The viruses for the establishment of stable cell lines were produced by the co- transfection of an HIV-1-based vector (pFUPI or pAPM) with psPAX2 and MD2.G, at a ratio of 3:2:1, respectively. Six to eight hours later, the medium was replaced with fresh medium. Forty-eight hours post-transfection, the viral supernatants were collected and centrifuged at 500xg for 5 minutes, to purify the supernatant from cell debris. The viral supernatants were transferred to 1.5 ml or 2 ml microcentrifuge tubes and stored at -70°C.

Stable cell line generation.

The FUPI and APM plasmids were described previously (265,324 and in section 2.1). The pFUPI vector was used to produce virus bearing different expression constructs for the establishment of cell lines stably expressing the proteins of interest. The different murine Trim5 orthologues were cloned from pcDNA3.1(-) (see above in the section 2.1) into pFUPI by using the NheI/BamHI restriction sites and ligated into XbaI/BamHI. Two different target sites for Trim12A and C and one target site for Trim12B were cloned into pAPM via the MluI/NotI restriction sites. As the two vectors encode the puromycin resistance gene (PuroR), cells were selected for one week with puromycin-containing medium.

105 One day prior to transduction CRFK and CF2Th cell lines were plated into 6 wells plates at a concentration of 25x103 cells/ml. They were subsequently transduced with 1 ml of fresh FUPI-containing viral supernatant. 48 hours post-infection, the cells were diluted 10 times and plated in medium containing 2 and 3 μg/ml puromycin for the CRFK and CF2Th, respectively. The murine TA3 cells that grow in suspension were plated at 25x103 cells/ml into 6 ml plates on the day of the transfection and were challenged with 2 ml of fresh pAPM-containing viral supernatants. Two days later, they were resuspended in 10ug/ml puromcyin-containing medium.

Infection with GFP viruses and FACS.

CRFK and CF2Th cell lines were plated at 25x103 cells/ml into 96 well plates the day prior to the virus challenging. Serial three- or five-fold dilutions of viruses were added to the wells. For the TA3 and MEF cell lines, 24 well plates were used to plate cells at a low confluence. The same day, cells were challenged with serial three- or five-fold dilutions of the different viral supernatant. Forty-eight to ninety-six hours post-infection, the feline and canine cells were collected into 1.2 ml FACS tubes from the SSI company, and murine cell lines were transferred to 5 ml FACS tubes from BD Falcon . TA3 cells were centrifuged, washed with PBS, re-centrifuged and re-suspended in 500 μl of PBS. The cells were then analysed with the FACS flow cytometer (FACSCalibur, BD Biosciences).

Design of the pCG PstI/SalI and pMICΔU3 vectors and cloning.

The SnabI/XhoI fragment from the pCG gagpol (gift from Dr. Massimo Pizzato, Univerity of Trento, Italy) was subcloned into the pcDNA3.1(-) plasmid. A SnaBI/BrsGI fragment with a mutagenized sequence that destroyed the PstI site, was amplified by PCR. A second mutagenesis generated a new PstI site within a BsrGI/XhoI fragment. Both fragments were cloned into pcDNA3.1(-). When the

106 sequence of the SnaBI/XhoI fragment was confirmed to contain the new PstI site, the DNA piece was cloned back into pCG gagpol (Figure 2). Mutagenesis was used to destroy the SalI site present within a XhoI/MfeI fragment that was amplified by PCR and cloned back into the pCG gagpol. Using the MfeI and BlpI unique sites, a new SalI site was created by PCR and ligated back into the pCG gagpol plasmid. The pCG gagpol palsmids obtained were sequenced and the positive clone was expanded and called pCG PstI/SalI. For the generation of the pMICΔU3-PuroMiR and pMICΔU3-BlastiMiR, we subcloned the BamHI/NcoI fragment from the pMSCV-mCyp-GFP into the BglII/NcoI sites of pMIG-ΔU3, destroying the BglII site. This plasmid was called pMICΔU3- GFP (Figure 6). The NcoI/SalI sites were used to clone a PuroR-MiR-30 shRNA cassette into the corresponding sites of pMICΔU3-GFP, resulting in the replacement of GFP. The BlastiR ORF was amplified by PCR, using primers containing the BspHI and NotI sites, respectively, and cloned into the NcoI/NotI restriction sites of pMICΔU3- PuroMiR, replacing the PuroR.

The different target sites from Trim12A, B and C were cloned into pAPM, via the XhoI/EcoRI sites. The following primers were used for amplification of the MiR30 shRNA sequences: miR-30 XhoI 5′: 5′-AAGGCTCGAGAAGGTATATTGCTGTTGACAGTGAG- 3′ and miR-30 EcoRI 3′: 5′-AGCCCCTTGAATTCCGAGGCAGTAGGCA-3′. The PCR reaction was performed in a 50 μl total volume with the AccuPrime Pfx SuperMix (Invitrogen), 1M Betaine (Sigma-Aldrich), using the primers and a MiR30 shRNA template that was synthesized by Microsynth. The PCR program included an initial step at 95 °C for 5 min, followed by 30 cycles of: 1) 95 °C for 15 s, 2) 55 °C for 30 s and 3) 75 °C for 30 s. The reaction was terminated by 5 min at 75 °C. All the other PCR were performed as above, except that 30-35 cycles were used and contained the following steps: 1) 95°C for 5 minutes, 2) 54°C for 30 s and 3) 68°C, for 1 min per kilobase pairs.

Primers and MiR30 shRNA sequences (target sequences are in pink):

Name Sequence

107 BalstiB 5’-caactcatgagccaccATGGCCAAGCCTTTGTCTC-3’ spHI5’

Blasti 5’- accagcggccgcTTAGCCCTCCCACACATAACCAGA-3’ NotI3’ pCGSn 5’-TGGCAGTACATCTACGTATTAG-3’ ab5’ pCGBs 5’- GGTCAAGCCCTTTGTACACCCTA-3’ rGI5’ pCGBs 5’- TAGGGTGTACAAAGGGCTTGACC-3’ rGI3’ pCGXh 5’- TGCCGCTTTTCCCCTCGAGCGCC-3’ oI5’ pCGXh 5’- GGCGCTCGAGGGGAAAAGCGGCA-3’ oI3’ pCGMf 5’- CAGGACAATTGACCTGGACC-3’ eI5’ pCGMf 5’-CCAGGTCAATTGTCCTGAGA-3’ eI3’ pCGBl 5’- CAGCCCGCTGAGCGGATGTCC-3’ pI3’ PstIkill 5’- TCaGCtGAgTGGCCAACCTTTAACGT-3’ For

PstIkill 5’-AGGTTGGCCAcTCaGCtGAGCAGAAGGTAACCC-3’ Rev PstIcre 5’- GACAGCTgCAgTACTGGCCGTTCTCC-3’ ateFor PstIcre 5’-AGTACTGCAGCTGTCCGTTTCCTCCTGCG-3’ ateR SalIkill 5’- ACTCTTTGTTGATGAGAAGCAGGGCTACGCCAA-3’ For SalIkill 5’-CCCTGCTTCTCaTCaACAAAGAGTTCAAA-3’ Rev SalIcre 5’-TcGtcGACATAGAGAGATGA-3’ ateFor SalIcre 5’-CTCTATGTCGAcGacgatctCTTTCTT-3’ ateRev MLVca 5’-acagctGcaG TATTGGCCGTTTTCCTCCTC-3’ psids- PstI5’ N/B- 5’-ctatgtcGAcgaCGGTCCCTTTCTTTCTCT-3’ SalI3’ XM8- 5’-ctatgtcGAcTGCGGTCCCTTTCTTTCTCTCT-3’ SalI3’ Trim1 5’- 2A ts1 TGCTGTTGACAGTGAGCGCTGAGTCCTAAGTACTGGCAAATAGTGAAGCC

108 ACAGATGTATTTGCCAGTACTTAGGACTCAATGCCTACTGCCTCGGA-3’

Trim1 5’- 2A ts2 TGCTGTTGACAGTGAGCGCGAAGGCAGCCACAAACTCAAATAGTGAAGCC ACAGATGTATTTGAGTTTGTGGCTGCCTTCTTGCCTACTGCCTCGGA-3’

Trim1 5’- 2B ts2 TGCTGTTGACAGTGAGCGCGCACCTGACTTCCACACATATTAGTGAAGCCA CAGATGTAATATGTGTGGAAGTCAGGTGCATGCCTACTGCCTCGGA-3’

TRIM1 5- 2C ts1 TGCTGTTGACAGTGAGCGCCACCCAGTTATCCAATCAGGATAGTGAAGCCA CAGATGTATCCTGATTGGATAACTGGGTGTTGCCTACTGCCTCGGA-3’

Trim1 5’- 2C ts2 TGCTGTTGACAGTGAGCGCAGAAGATACTGTGTTCGGAAGTAGTGAAGCC ACAGATGTACTTCCGAACACAGTATCTTCTTTGCCTACTGCCTCGGA-3’

Scram 5’- bled TGCTGTTGACAGTGAGCGATCTCGCTTGGGCGAGAGTAAGTAGTGAAGCC ACAGATGTACTTACTCTCGCCCAAGCGAGAGTGCCTACTGCCTCGGA-3’ RPL4 5’-GCAACATCCCTGGTATTACTCTGCT-3’ RT for RPL4 5’-GTGCATGGGCAGGTTATAGTTACTC-3’ RT rev CXCL1 5’-CCAAGTGCTGCCGTCATTTTC-3’ 0 RT for CXCL1 5’-GGCTCGCAGGGATGATTTCAA-3’ 0 RT rev Trim1 5’-CCAACCAGCAGCCCCTCC-3’ 2A RT for Trim1 5’-CGCTGAAAACTTCTCCTCTAGCCT-3’ 2A RT rev Trim1 5’-CTTGCTCAGGCATCATATGTGC-3’ 2B RT for Trim1 5’-CTTTTTCTTTCCCCCTGCCTC-3’ 2B RT rev Trim1 5’-TGCTGTGTTATAGTGTCCC-3’

109 2C RT for Trim1 5’-ATTTGGGGTCTTTATTGTGACT-3’ 2C RT rev

Genomic DNA and mRNA extractions, cDNA synthesis and RT-PCR.

Genomic DNA was isolated from 5x106 TA3 cells, using the Blood & Cell Culture DNA Mini Kit (Qiagen), according to the manufacturer’s instructions. The RNA fraction containing mRNAs was extracted from 107 TA3 cells from each pAPM-derived line, with the RNAzol-RT reagent (Molecular Research Center, Inc.), according to the manufacturer’s protocol. The mRNA samples were subsequently subjected to a treatment with the DNAseI enzyme (Life Tecnologies) and purified on columns (Qiagen). For complementary DNA (cDNA) first-strand synthesis, the purified mRNA samples were reverse-transcribed with random hexamers, using the MoMLV- reverse-transcription (MMLV-RT) enzyme (Promega). The starting volume of water and RNA was calculated in order to carry the same amount of RNA in all the samples, as determined from the individuals optical densities (ODs). A reaction containing the water, RNA and the random hexamers was incubated at 70°C for 5 min and subsequently cooled on ice for another 5 min. The deoxyribonucleotides (dNTPs), the 5x reaction buffer and the MMLV-RT were then added to the reactions. All the samples were incubated at 25°C for 10 min, 42°C for 50 min and finally at 70°C for 3 min, in order to inactivate the enzyme. The semi-quantitative Real-Time PCR was performed as follows: The different reactions containing 10 μl of cDNA (up to 5 μg) were incubated each with 10 μl of PCR mix containing 5 μM of each specific primer, 2X PCR buffer, 2X BSA, 10 mM MgCl2, 0.4 mM of each dNTP, 1/10000 SyBrGreen and 1 unit of HotStart Taq Polymerase (Promega), in 96 wells plates (Axygen). The PCR program consisted in a first incubation at 50°C for 2 min and at 95°C for 10 min, followed by 39 cycles of 15 s at 95°C, 30 s at 60°C and 30 s at 72°C. The

110 reactions were terminated at 72°C for 10 s. The SybrGreen signal was recorded at the end of each annealing step. The reactions and the subsequent quantifications were performed with the Bio- Rad CFX96 real-time PCR detection system.

Results

A) The assessment of the ability of the murine TRIM5 orthologues to restrict HIV-1 when transiently expressed in CRFK or CF2Th cell lines.

Previously, three out of the seven Trim5 orthologues were tested for anti-HIV-1 activity 298. That study showed that TRIM12C, TRIM30A and TRIM30D were unable to restrict HIV-1 when expressed in CRFK cells. We wanted to repeat the previous results and include TRIM12A, TRIM12B, TRIM30A and TRIM30C in the screening for HIV-1-targeting restriction factors. We cloned the different murine Trim5 orthologues into the HIV-1-based FUPI plasmid (pFUPI), that was described earlier 324 and transduced CRFK and CF2Th cell lines with the empty pFUPI (pFUPI(-)) or with the vector containing different Trim5 orthologues. As a positive control, the HIV-1-restricting TRIM5α from rhesus monkey or TRIM30D-Cyp was used. We then performed single-cycle infection assays, using HIV-1-GFP and measured the percentage of GFP-positive cells, 2-3 days post-infection. As seen in Figure 1, none of the murine TRIM5 orthologues blocked HIV-1 in either CRFK or CF2Th cell lines. In contrast, rhesus monkey TRIM5α conferred to CRFK nearly 10 fold more protection from HIV-1 than the pFUPI(-) cell line. Similarly, the rhesus Trim5α and TRIM30D-Cyp constructs restricted HIV-1 more than 10 fold in CF2Th cells. Introducing the different pFUPI-murineTrim5 constructs in HeLa cells yielded similar results (data not shown). These data suggest that the murine TRIM5 orthologues are unable to restrict HIV-1 at least when expressed in heterologous cells.

111 !"# $"#

HIV-1 in CRFK cells HIV-1 in CRFK cells 100 100 s l l s l e l 10 e c

10 e FUPI (-) e

FUPI (-) v i v t i

i TRIM12B t

i TRIM12A s

s 1

o TRIM12C

o TRIM12B p p - - TRIM12C rhesus monkey TRIM5α P P 1 F F TRIM30A G G

rhesus monkey 0.10 % TRIM5α %

0.1 0.01 0.0 0.1 0.2 0.3 0.0 0.2 0.4 0.6 Dilution of viral supernatant Dilution of viral supernatant %"# &"#

HIV-1 in CF2Th cells HIV-1 in CF2Th cells 100 100 s l s l l l e e c

10 10 e FUPI(-) e

FUPI (-) v i v i t

i TRIM30A t

i TRIM12A s s

o TRIM30C

o TRIM12B p p - - TRIM12C TRIM30D P P 1

1 F F TRIM30A TRIM30D-Cyp G G

rhesus monkey TRIM5α % %

0.1 0.1 0.00 0.05 0.10 0.15 0.00 0.05 0.10 0.15 0.20 0.25 Dilution of viral supernatant Dilution of viral supernatant Figure 2: The efect of the ectopic expression of the different Trim5 orthologues in heterologous cells on HIV-1 infection. Cells were challenged with HIV-1-GFP and 48 hours post-infection, the percentages of GFP-positive cells were assessed by FACS. A) CRFK cells were transduced with FUPI vectors to stably express the different Trim12 genes, Trim30A and rhesus monkey Trim5α. B) Six serial dilutions of HIV-1 were examined for Trim12B, Trim12C and rhesus monkey Trim5α cell lines. C) Stable CF2Th cells stably expressing the different Trim12 genes, Trim30A and rhesus monkey Trim5α were generated D) The effect of the Trim30 genes on HIV-1 infectivity was separately assayed and compared to the restrictive Trim30D- Cyp. FUPI(-) was used each time as a negative control.

B) The generation of an MLV-based vector with unique PstI and SalI restriction sites in the capsid gene.

The murine genome contains several copies of endogenous MLV sequences that are still coding for the gagpol-derived proteins. Although they often lack a functional envelope, these endogenous retroviruses could still induce detection by the innate immune system and subsequent inflammation. Notably, the link between the expression of certain retroviruses and autoinflammatory diseases development was suggested by a considerable number of reports.

112 Interestingly, the human endogenous retrovirus Fc1 (HERV-Fc1) and human TRIM5α polymorphisms were associated with the etiology of multiple sclerosis 325. In these instances, TRIM5 could recognize the capsid of endogenous retroviruses and initiate a cascade of MAPK and NFκB pathways. We therefore reasoned that murine TRIM5-like proteins could potentially restrict endogenous MLVs. We therefore wanted to produce an MLV-based vector in which the capsid sequence could be easily replaced with that of different MLV strains. For this, we designed unique restriction sites that would not disturb the resulting protein sequence. We used the Vector NTI (Life technologies) software to find unique sites in the portion of the sequence of the pCG gagpol vector that was known. As a fraction of the MLV-derived plasmid was not sequenced, we individually tested the different restriction sites for unique cutting in single-enzyme reactions. Among the unique-cutter enzymes, PstI and SalI were found to fit the sequence requirements for allowing conservative mutagenesis. Therefore, the two preexisting sites outside the capsid gene were killed by mutagenesis and two new sites were created flanking the region of capsid that wanted to be replaced (Figure 2). The infectivity of the resulting pCG PstI/SalI vector, when co-produced with the GFP-genome and the envelope plasmids, was compared to that of the original pCG gagpol backbone. As shown in Figure 3, the infectivities derived from each vector were similar, as assessed in MEF and TA3 cells. We then sub-cloned the N- and B-MLV capsids in the newly designed plasmid, in order to determine if the Fv1- or TRIM5- sensitivity was conserved in this context. The specific pattern of N- and B-MLV infectivity in different cell lines was consistent with the Fv1 and Trim5α alleles that were present. Indeed, murine MEF and TA3 cell lines are derived from the C57BL/6J mice and C57BL/6J X 129/SV6 crosses, respectively and thus carry the Fv1b allele 183,326(http://www.informatics.jax.org/marker/MGI:95595). Importantly, in the Fv1b or TRIM5α- expressing cell lines that we tested, N-MLV was restricted but B-MLV was as infectious as MoMLV (Figure 5).

113 Thus, it is conceivable that other potential TRIM5-restricted capsids from different MLV strains could retain their sensitivity in the pCG PstI/SalI backbone. A recent screening of human cell lines identified a contamination with replication-competent xenotropic MLV, likely to have come from recombination of non-infectious endogenous MLV from murine cell lines 327. One of the viruses that were sequenced, named EKVX, was quite poorly infectious in the cell lines tested, compared to the other detected strains. As human TRIM5α restricts N-MLV, we wondered if the apparent blockade in human cells to the EKVX virus was dependent on the restriction factor. We used specific primers that flanked the sequence of the capsid fragment we wanted to clone in the pCG PstI/SalI. We then tried to amplify capsid sequences from TA3 cells-derived genomic DNA by PCR, using the PstI forward and SalI reverse primers (see materials and methods section). The PCR reaction was run on an agarose gel and yielded a band of the expected size (Figure 4). We therefore extracted the band from the agarose gel, digested the product with the PstI and SalI enzymes and cloned the fragment into pCG PstI/SalI. The amino acid sequence of one of the clones was found to be 100% identical to three NCBI entries, notably the EKVX protein. The corresponding capsid was called XM8 and subsequently tested for restriction on MEF, TA3 and HeLa cells. We found that XM8 could infect both murine and human cell lines, as effectively as MoMLV (Figure 5). However, we did not test the ability of murine TRIM5 orthologues to restrict this construct when expressed transiently in heterologous cell lines like CF2Th. In the future, we could assess the sensitivity of this and other capsids to the different murine TRIM5 orthologues. We conclude that the pCG PstI/SalI vector that we designed is suitable for the cloning of capsids from MLV strains and a reproducible tool to test the MLV capsid-specific restriction by TRIM5 proteins.

114 %>?@$+%(6&9**+ Original pCG gagpol vector 8.=$+%/0/)*+

7'+ ,3-8$+%((9*+ :;<$+%'(&&*+ 1234$$$+%5/6(*+

!"#$ %&#$'($)*+',$ -./012345678$780958$ :132;54$ -./012345678$<54=>$

!"#$%&'()*+ 8A;B$+ ,-.$+%/&0/*+ %'/))*+

pCG PstI/SalI vector

7'+ :;<$+%'065*+ ,-.$+%&C95*+

!"#$ %&#$'($)*+',$ -./012345678$780958$ :132;54$ -./012345678$<54=>$

Figure 3: Maps of the pCG gagpol and pCG PstI/SalI vectors. The positions of the unique restriction sites used for mutagenesis or sub-cloning into the pcDNA3.1(-) vector are indicated. The CMV promoter drives the expression of gag-pol, which is flanked of regulatory sequences (yellow rectangles) consiting in the 5’UTR of the thymidine KINASE (TK) gene of Herpes Simplex Virus (HSV) and the human β-Globin intron and PolyA tail. HSV-1 TK 5’UTR: possesses a cis-acting element responsible for increased transcription, independent of viral promoter or trans-acting viral factors, so can be used in cmv promoter-based vectors 328. hBeta globin intron: improves the expression of cmv promoter-driven transcription.

115 !"#

MLV capsids in MEF cells 100 s l l e c

e pCG gagpol v i t

i pCG PstI/SalI s o p - P F G 10 %

0.01 0.10 1 Dilution of viral supernatant $"# MLV capsids in TA3 cells

100 s l l

e pCG gagpol c

e pCG PstI/SalI v i t i s o p - P

F 10 G

0.0 0.5 1.0 1.5 2.0 Dilution of viral supernatant

Figure 3: The comparison of the ability of the original pCG gagpol and the newly designed pCG PstI/SalI to infect murine cells. The two vectors were used to produce virus in 293FT cells and assayed on MEF cells (A) and in TA3 cells (B).

116 *###&'(& )"#&'(& "##&'(& !"#$!"%&'(&

Figure 4: The amplification of a MLV capsid fragment from TA3 gDNA. The product was obtained by PCR and yeilded a band of the expected size (709 bp). The different sizes from the DNA ladder are indicated in base pairs (bp).

117

..:.ù-\ :, l'( G ,.,9^-\J \.l\ iJJ , \Cé -l* r.Í t^-,,\L: :1lz J \

/r,\ LlY Ar.* > (t

fi)ìJ, \ -,( * I\ t) ['xit 1i' Z\ I 'Y\.'7" ,17 ,',l €.1,L \. ì\rl\. ? \) r' r\ 4

n. a' , ,_1 _r ti 2

ì- fi /l c, * 'fv\<) 'i I - rtl !"# $"# MLV capsids in MEF cells MLV capsids in TA3 cells

s 10 l l s e l l c

e 10

e pCG PstI/SalI pCG PstI/SalI c

v i e

t N-MLV v i N-MLV

1 i s t

i B-MLV

o B-MLV s p o - XM8 1 XM8 p P - F P

G 0.10 F

G %

% 0.1 0.01 0.001 0.010 0.100 1 0.001 0.010 0.100 1 Dilution of viral supernatant Dilution of viral supernatant %"# MLV capsids in HeLa cells 100 s l l e

c 10

e pCG PstI/SalI v i t

i N-MLV s

o B-MLV p

- 1 XM8 P F G

% 0.1

0.001 0.010 0.100 1 10 Dilution of viral supernatant

Figure 4: The cell tropism of different MLV capsids. The two Fv1b-expressing MEF and TA3 cell lines A) and B), respectively, were tested for the restriction of MoMLV (pCG PstI/SalI), N- and B-MLV and XM8 capsids-associated viruses. C) The TRIM5α-containing HeLa cell line was assayed for the restriction of the different MLV capsids-bearing viruses.

C) The combination of two MSCV-based vectors in one: attempts to create a plasmid for an efficient delivery of shRNA into the murine TA3 cell line.

Originally called MESV, the pMSCV plasmid was derived from the myeloproliferative sarcoma virus (MPSV) that is closely related to MoMLV 329,330. The MLV-derived vector had specific deletions and mutations in the LTR sequence that permitted to increase the range of cellular tropism for the efficient transduction of embryonic stem cells (ESCs) 329,331. A previous study from our laboratory reported the usage of the pMSCV vector that contained a GFP expression cassette (pMIG) to efficiently transduce murine primary T cells 321. The authors cloned the Cyclophilin A (CypA) promoter

118 downstream of the packaging signal sequence of the pMSCV, given that CypA is strongly and constitutively expressed in murine cells. As the retroviral LTR-mediated transcription from the integrated provirus can interfere with the mammalian promoter-driven expression, the U3 enhancer region was deleted in pMIG by digesting the vector with NheI/XbaI 321, resulting in pMIG-ΔU3. We wanted to combine the two previously described plasmids in order to obtain a vector able to transduce the TA3 cell line, with strong constitutive expression of the bi-cistronic Blasti-MiR sequence in the same time than reducing the interference by the viral LTR. To this end, we sub-cloned the BamHI/NcoI fragment encompassing the entire CypA promoter into the BglII/NcoI sites of pMIG-ΔU3 backbone, resulting in the pMICΔU3-GFP vector (Figure 6). We then aimed to test the transduction efficiency of the pMICΔU3-GFP vector on the T cell line TA3 and compare it to the HIV-1 derived pAGM. As shown in figure 7, pMICΔU3-GFP was 5 to 7 fold more efficient to infect TA3 cells than was pAGM. Whereas the MLV-derived virus could infect nearly 10% of the cells at a viral dilution of 1:10, the HIV-1-GFP particles could only reach 5 to 7% from 1:2 to 2:1 concentrations. Using higher concentrations of virus could be an issue, since they are more likely to induce non-specific cytotoxicity 332,333. In addition, the detected mean fluorescence for the pMICΔU3-GFP virus was from 2 to 3.5 fold higher than pAGM on the different viral dilutions that we assayed (figure 7B). We subsequently cloned a PuroR-MiR sequence from the HIV-1-derived pAPM into the newly designed plasmid, into the unique NcoI and SalI sites (figure 6). We called this vector pMICΔU3-PuroMiR. Then, the PuroR sequence was replaced by the one for BlastiR by cloning a BspHI/NotI fragment into the NcoI/NotI sites. Finally, we aimed to clone different target sequences to silence each of the three Trim12 genes from the mouse. Two different target sites were designed for each TRIM5 orthologue (see the Materials and methods section). The ability of the pMICΔU3-BlastiMiR vector to establish resistance to Blasticidin in TA3 cells was not assessed here.

119 What we can conclude from our data is that the pMICΔU3-GFP vector can infect TA3 cells more efficiently than pAPM and thus represents a useful backbone for the transduction of this murine T lymphocytes-derived cell line. In the future, we could test the pMICΔU3-BlastiMiR for the capacity to confer Blasticidin resistance to TA3 cells and other T cell lines. Subsequently, the efficiency of gene silencing using this vector could be assessed.

*$ *$ %&'$ !"#$%&'(!()*'% !"#$ %&'$ %&'$ !"#$ %&'$ ()$

35-6$ <=>6$ *$ %&'$ !"#$%&'(!()*'% !"#$ %&'$

35-6$7889:;$ 3-A6$ .>+6$ <=>6$78?@9;$ *$ %&'$ !"#$%&'(!()*'% #+,-'$ ./')0$$12'34$ %&'$

B1DE6$ 3-A6$

B>=1C'$ 3-A6$ .>+6$ <=>6$ *$ %&'$ !"#$%&'(!()*'% B>=1C'$ ./')0$$12'34$ %&'$

Figure 5: The maps of the pMICΔU3-PuroMiR and pMICΔU3-BlastiMiR vectors derived from pMIC and pMIGΔU3. A bi-cistronic element coding for the puromycin resistance gene(PuroR) and the RNA polymerase II (RNA pol II)-driven MiRNA-based shRNA (MiR30 shRNA) was cloned into the pMICΔU3 vector via the NcoI and SalI unique restriction sites. The RNA pol II promoter is shown in red. The PuroR was replaced by the blasticidin resistance gene (BlastiR) using the BspHI/NotI enzymes, killing the NcoI site.

Ψ: Psi packaging signal. The U3 region within the viral LTR is shown as a dark purple square.

120 !"#

TA3 cells 100 s

l pAGM l

e pMIC-!U3-GFP c

v 10 i t i s o p - 1 P F G

% 0.1 0.01 0.10 1 10 Dilution of viral supernatant $"#

TA3 cells e c

n 50

e pAGM c

s 40 pMIC-!U3-GFP e r o

u 30 ﬂ

a 20 e m

e 10 v i t a l 0 e R 0.02 0.10 0.50 2.50 Dilution of viral supernatant

Figure 6: The comparison of the transduction efficiencies in TA3 cells between the HIV-1 based and the MSCV-based vectors. VSVg-pseudotyped viral particles were produced from pAGM and pMICΔU3-GFP and used to transduce TA3 cells. A) The percentage of GFP-positive cells was counted by FACS. B) The mean fluorescences that were determined by FACS were compared.

D) The dowregulation of the Trim12 genes in TA3 cells, using an HIV-1 based vector.

We first aimed to repeat the results obtained by the previous study by Baumann and colleagues, on the blockade of HIV-1 in TA3 cells 183. Therefore, we compared the infectivites of HIV-1 and MoMLV on TA3 and MEF cells. Similar to what the authors found, we obtained a profound block of HIV-1 infection in TA3 cells, for similar viral titers, compared to MEFs. The ratio of HIV- 1/MoMLV infectivities was up to 47 fold more elevated in MEFs than in TA3 cells (Figure 8A). Previously, we showed that the HIV-1 based pAPM vector could be used to efficiently silence Trim5α in human cell lines 265. In parallel to the creation of the pMICΔU3-BlastiMiR vector, we aimed to test the effect of the miRNA target sequences (ts) for the three Trim12 genes on HIV-1 infection, using pAPM.

121 We cloned the different miRNA sequences that were described above, into the pAPM backbone, via the MluI/NotI unique restriction sites. The transduction of TA3 cells with the pAPM-Scrambled (pAPM-Scr) control or with the constructs containing Trim12A ts1/ts2, Trim12B ts2 and Trim12C ts1 resulted in a similar low HIV-1/MoMLV infections ratio. Surprisingly, the second target site specific for Trim12C rescued HIV-1 levels (Figures 8B and 8C), rising the HIV-1/MoMLV ratio up to 8 fold compared to the control pAPM-luc (Figure 8D). Curiously, the stable cell lines expressing Trim12A ts2 and Trim12C ts1 were less infected by HIV-1 or MoMLV, as assessed by experiments where a equal number of cells were used to challenge with either of the two viruses on the same day (Figures 8B and 8C). We suggest that this could be due to a potential higher number of cells that were plated for these two cell lines prior to retroviral infection. The percentage of GFP-positive cells obtained with MoMLV challenging was thus essential to the normalization of HIV-1-GFP between the various cell lines. In order to check for the specificity of Trim12C knock-down (KD), we determined the mRNA levels of the different Trim12 genes from the pAPM-Scr, Trim12C ts1 and Trim12C ts2 cell lines, using semi-quantitative Real-Time PCR (RT-PCR). The ribosomal protein L4 (RPL4) is a house-keeping gene and therefore was used as an internal control for normalizing the gene expression in each cell line. As shown in Figure 8E, Trim12C was modestly down-regulated in the Trim12C ts1 cell line compared to the control. In contrast, Trim12C ts2 showed nearly 10 fold reduction of the Trim12C transcript and only a modest decrease of the mRNA levels of the two other Trim12 genes (Figure 8E). This result suggests that the silencing by Trim12C ts2 is specific for the Trim12C gene. In conclusion, we found that TRIM12C silencing correlates with a rescue of HIV-1 infection in TA3 cells. The fact that the blockade to HIV-1 transduction is not totally overcome when Trim12C is silenced suggests that other factors may participate to the retroviral restriction or that the requirements for cellular co-

122 factors potentially less abundant in TA3 cells have an additive negative effect on HIV-1 pre-integration steps. Alternatively, the usage of an HIV-1 based vector for establishing a stable cell line from murine T cells may result in a lower transgene expression (see Figure 7) and thus in a less efficient gene silencing. This technical issue could be overcome by using the MLV-derived pMICΔU3-BlastiMiR vector that we designed. In the future, we could silence the different Trim12C genes, in addition of other murine Trim5 orthologues, in TA3 cells with the pMICΔU3-BlastiMiR vector and assess the resulting effects on HIV-1 infectivity. Additionally, as Trim12 ts2 recognizes a sequence in the 3’UTR of the corresponding mRNA, we could assess if Trim12C KD has a direct role on HIV-1 restriction in TA3 cells, by over- expressing THE Trim12C coding sequence in the KD cell line in order to detect a potential re-establishment of the retroviral blockade.

!"# $"# HIV-1 in TA3 cells HIV-1/MoMLV infectivity ratio 10 s

l Control 1000 l MEF e T12A ts1 c

TA3 e T12A ts2

100 v i t

i T12B ts2 e s

g 1

o T12C ts1

n 10 a p -

h T12C ts2 P c -

1 F d l G o

F 0.1 % 0.1 1 2 3 0.01 0.10 1 Serial 10 folds dilutions of viral supernatant Dilution of viral supernatant %"# &"# MLV in TA3 cells HIV-1/MoMLV ratio 100 10 s

l Control l

e 8 T12A ts1 e c

g e T12A ts2 n a v 10 6 i h t

i T12B ts2 c - s 4 d

o T12C ts1 l p o -

1 T12C ts2 F

P 2 F G 0 % 0.1 0.01 0.10 1 T12A ts1 T12A ts2 T12C ts1 T12C ts2 T12B ts2 Dilution of viral supernatant Scrambled

'"# Normalization to Scrambled KD 1.5 Scrambled s l

e T12C ts1 v e l 1.0 T12C ts2 A N R m

v 0.5 i t a l e R 0.0

12A 12B 12C RPL4 CXCL10

123 Figure 7: The silencing of Trim12C in TA3 cells rescues HIV-1 levels in TA3 cells. HIV-1- and MoMLV- GFP viruses were used to infect MEF and/or TA3 cells. A) The ratio of HIV-1- to MoMLV-derived infectivities was compared between MEF and TA3 cells. B) TA3 cells stably transduced with different pAPM contructs confering the resistance to puromycin as well as targeting different Trim12 genes were challenged with HIV-1-GFP (B) or MoMLV-GFP (C) and the percentage of GFP-positive cells was determined by FACS. D) The infectivity ratios from HIV-1 to MoMLV was compared in the different TA3 cell lines. E) The mRNA levels of the different Trim12 genes from the Scrambled (control), Trim12C ts1 and Trim12C ts2 knock-down (KD) cell lines were assessed by semi-quantitative RT-PCR. The RPL4 house-keeping gene was used as an internal control. The levels of the CXCL10 cytokine were also compared. A 22 base pairs sequence with no predicted target gene in mouse was used as a control pAPM-transduced TA3 cell line (control or Scrambled) in all the TA3 silencing experiments.

124 Chapter 3

Mutations in the HIV-1 capsid modulate the human TRIM5α-mediated recognition.

Although HIV-1 is generally considered as poorly restricted by the human version of TRIM5α, it was previously shown that gag-protease sequences derived from clinical isolates were more susceptible to the restriction factor 308. Mutations in the capsid but also in the matrix and protease sequences could potentially all contribute to the observed phenotype. We aimed to investigate the level of contribution of the capsid sequence to the increased sensitivity to TRIM5α and determine which were the regions that were targeted by the restriction factor. We therefore cloned the capsid sequences from the clinical isolates-derived variants into the p8.9 WT plasmid and compared the infectivities of the resulting viruses on cells transduced with a luciferase- or Trim5α-targeting miRNA-based shRNA and determined their susceptibility to the presence of human TRIM5α. The sequence of the capsids of the more restricted viruses were aligned and the common mutations were used to design a mutant with a minimal number of substitutions that would retain the sensitivity to human TRIM5α. We found that mutations in the helices 4 and 7, as well as in the CypA-binding loop could recapitulate at least the half of the higher acquired TRIM5α-mediated restriction. Finally, some residues were found to abolish the susceptibility to human TRIM5α. These data point toward an important role of three distinct regions of the N- terminal domain of the HIV-1 capsid that modulates the recognition by human TRIM5α. Recognition of a specific capsid can stimulate TRIM5α-dependent innate immune pathways 265. In the future, we could determine the effect of the HIV-1 restricted variants on the amplitude of TRIM5α-mediated innate immune signaling activation. Additionally, these data could be interpreted on future structural analyses that could shed a light on the link between CypA and TRIM5α during retroviral restriction.

125 The following are unpublished results. I personally performed all the experiments, except the cloning of the luciferase- and TRIM5α-targeting sequences into the pAPM vector.

126 Residues in HIV-1 capsid that modulates the sensitivity to human TRIM5α

Josefina Lascano2, Thomas Pertel3 and Jeremy Luban1

Unpublished

1Program in Molecular Medicine, University of Massachusetts Medical School, 373 Plantation Street, Biotech II, Suite 319, Worcester, Massachusetts 01605, USA

2Department of Microbiology and Molecular Medicine, University of Geneva, 1 Rue Michel Servet, CH-1211 Geneva 4, Switzerland

3Department of Neurology, Harvard Institutes of Medicine, Center for Neurologic Diseases, HIM 780, 77 Avenue Louis Pasteur, Boston MA 02115

127 ABSTRACT

Capsid-containing sequences from clinical HIV-1 isolates have been shown to exhibit differential increased sensitivities to the human TRIM5α restriction factor, when compared to that of the typical laboratory-adapted strains. Mutation of different residues of the capsid sequence had divergent impacts on the sensitivity to TRIM5α, depending on the mutations already present in the virus that were suggested to have arisen from Cytotoxic T lymphocytes (CTL) responses in previous hosts. In the present study, we investigated the unique contribution of the capsid sequence to the previously observed phenotype, by cloning them into a laboratory-adapted strain HIV-1 background (WT). Furthermore, we aimed to characterize the involvement of different regions of the capsid (CA) and define smaller motives that would recapitulate the most the gained sensitivity to human TRIM5α. We show that the capsids derived from the different clinical isolates conferred the TRIM5α-restricted phenotype to the WT background virus, at similar levels that the gag-protease-containing viruses analyzed previously. In addition, the D71-A86-I96-K132-V135-M136-V141-R158 mutations were present in the more restricted strains and were sufficient to recapitulate the half of the sensitivity to TRIM5α observed previously. The helical regions 4 and 7 from the N-terminal domain of the HIV-1 capsid, as well as the Cyclophilin A-binding loop were found to carry most of the mutations. We conclude that these regions of the HIV-1 CA have an impact on the virus susceptibility to the TRIM5α expressed in its natural host.

128 INTRODUCTION

In the early course of the infection, HIV-1 peptides coming from infected cells or from phagocytized infected cells are presented as viral antigens to CD8+- and CD4+-T cells on the Major histocompatibility complex (MHC) I and II, respectively. The priming of a CD8+-T cell leads to the activation of its subsequent cytotoxic T lymphocyte (CTL) function. The HIV-1-specific CD8+ T cells participates in the control of the acute viremia and leads to the production of IFN-γ and IL-2 1-3. As a result of the selective pressure induced by the CTL responses, mutations are introduced in the HIV-1 sequence within or near the targeted epitopes 4,5 6-8. In the acute phase of infection, HIV-1-specific CD8+ T lymphocytes target viral epitopes that are likely to escape by permitted mutations 9. As HIV-1 escape variants appear, the CTL responses starts to decrease 1,9. Potentially contributing to the weakness of viral control, the conventional dendritic cells (cDCs) do not sense HIV-1 and are not activated when they present a given antigen to CD4+- and CD8+-T cells, limiting their activation and expansion, as seen on ex-vivo experiments 10-13. The innate immune response to HIV-1 includes the Toll-like receptor 7 (TLR7) that senses the genomic DNA, APOBEC3 proteins that lead to mutation catastrophe on the viral genome and the viral DNA sensor IFI16 14-17. Functioning as a pattern recognition receptor for the retroviral capsid, the rhesus macaque and owl monkey versions of TRIM5α robustly block HIV-1 18-22. In human, it is not clear what is the impact of TRIM5α during HIV-1 infection. Notably, the laboratory-adapted strains show a modest 1.5 to 3 fold-restriction by human TRIM5α 23-25. Nevertheless, strains derived from different subtypes of HIV-1 are restricted from 1.2 to 19.5 fold 26. Recently, it was shown that CTL-induced escape mutations alters the sensitivity of HIV-1 to human TRIM5α and this effect seemed to be dependent on the capsid (CA) sequence 27, although mutations outside the capsid, including in the matrix and the protease, could also influence the phenotype observed (Figure 1).

129 The N-terminal domain of CA, which is recognized by TRIM5α 28-31, is composed of an initial β-hairpin, seven α-helices and one cyclophilin A (CypA)-binding loop located between the 4th and 5th helices 32. Modulating the CA-CypA interactions in CTL escape variants lead to differential effects on the sensitivity to TRIM5α 33. Importantly, the virus derived from the clinical isolate NRC10-5, became less susceptible to TRIM5α when the binding to CypA was prevented 33, indicating that CypA facilitated somehow the recognition of the NRC10-5-derived virus by the restriction factor. Interestingly, the same variant was found to carry mutations within the epitope KK10 on the helix 7 and in the CypA-binding loop, that were contributing to the increase in TRIM5α-mediated restriction 27. It is unclear, however, which regions of the capsid could mostly account for the observed sensitivity to human TRIM5α. In the present study, we aimed to investigate the level of contribution of CA to the TRIM5α-mediated restriction as well as investigate the influence of different regions of p24 to the phenotype. We found that CA is an important contributor of the gained sensitivity to TRIM5α. Additionally, we designed a variant with a smaller number of mutations that show at least half of the susceptibility to TRIM5α. Most of the substitutions lied within the 4th and the 7th helices, as well as in the CypA-binding loop. In addition, in the minimal variant background, the mutation in the CypA-binding region or one of the mutations in the 7th helix completely abolished the TRIM5α- mediated restriction. These data show that on HIV-1 CA, residues present on the α-helices 4 and 7, as well as on the CypA-binding loop, can influence the recognition by human TRIM5α.

MATERIALS AND METHODS

Plasmids, vectors and viruses.

130 The HIV-1-derived pWPTS vector contains a enhanced GFP ORF under the control of the EF1α promoter 34. The p8.9NdSB packaging vector was described previously 35 and contains the gag-pol sequences. The third vector necessary for virus production is the Vesicular Stomatitis Virus (VSV) G protein envelope

(MD2.G).

HIV-1-GFP three parts viruses were prepared by transfecting the 293FT cells with pWPTS-GFP plasmid {Berthoux, 2004 #954}, the different p8.9NdSB containing capsid sequences from different variants and pMD2.G {Berthoux,

2003 #955}.

N-/B- MLV-GFP three part viruses were prepared by transfecting the 293FT cells with pLNC-GFP 29, pCG-N/-B and pMD2.G.

The pAPM luciferase and pAPM Trim5 ts2 KD, were cloned previously 22.

The vectors for the retroviral gene expression, the packaging genes and the envelope were transfected at a ratio of 3:2:1, respectively, using

Polyethylenimine (PEI) (Sigma Inc.).

Briefly, 2 μl of PEI for 1 μg of DNA were mixed in Opti-MEM (Invitrogen) and incubated for 30 minutes prior to be added to the cells.

At 48 hours post-transfection, virus supernatants were harvested.

Cloning.

The different capsid sequences were designed on the basis of ClustalW alignments and synthesized by GeneScript USA Inc., and were flanked by 5’NotI and 3’ApaI restriction sites (see Supplementary Figure 1).

131 The primers used for the amplification of the sequences by PCR were the following: NotI forward 5’-CCTTGGCTTCTTATGCGACGG-3’ and ApaI reverse: 5’-

GTGGGAAGGCCAGATCTTCC-3’.

The PCR program included an initial step at 95 °C for 5 min, followed by 30 cycles of: 1) 95°C for 5 minutes, 2) 54°C for 30 s and 3) 68°C, for 1 min per kilobase pairs. The reaction was terminated by 5 min at 75 °C.

Cell lines.

The human fibrosarcoma epithelial cell line HT1080 was obtained from ATCC.

They were maintained in high glucose Dulbecco’s modified Eagle medium (D-

MEM) with 1x MEM NEAA and 1x GlutaMAX-I (Invitrogen) supplemented with

10% FBS.

Cells were passed every 3 days, when reaching 80% of confluency.

For stable cell lines production, 800 ul of fresh viral supernatant was added to each well of a 6 wells plate, containing 50000 cells per well. The cells were expanded for 4 days without any selection drug. The APM-postive HT1080 cells were then selected for one week in complete medium containing 2ug/ml of puromycin.

Infections with GFP-viruses.

HT1080 cell lines were seeded at 1500 cells per well in a 96 well plate. The next day, 50 ul of HIV-1- or MLV-GFP viruses were added at serials three fold dilutions to each well.

132 After 48 hours, the virus-containing medium was aspirated, the cells washed with PBS and tripsine was added to each well. The cells were subsequently collected and subjected to Fluorescence Activated Cell Sorting (FACS) with the

FACSCalibur flow cytometer (BD Biosciences), to determine the percentage of infected (GFP-positive) cells.

Solutions for SG-PERT.

The 2X lysis buffer was composed of 0.25% Triton X-100, 50mM KCl, 100 mM Tris-HCl pH 7.4, 40% glycerol and 2 μl of RNAse inhibitor (Ribolock Fermentas) per 100 μl of 2X lysis buffer.

The 2X reaction buffer was composed of 10mM (NH4)2SO4, 40mM KCl, 40 mM

Tris-HCL pH 8.3, 10 mM MgCl2, 0.2 mg/ml BSA, 1/10000 SyBrGreen I, 400 μM dNTPs, 1 μM of forward and reverse primers and 7 ρmoles/ml of MS2 RNA.

The 10X dilution buffer was composed of 50 mM (NH4)2SO4, 200 mM KCl and 200 mM Tris-HCl pH 8.3. The HotStart Taq polymerase (Promega) was added immediately before performing the SG-PERT assay at a concentration of 0.02 units/μl.

SG-PERT ASSAY.

Two days post-transfection viruses were collected, centrifuged 5 minutes at

1500xg and 5 μl of each supernatant were incubated with an equal volume of 2X lysis buffer. The reactions were incubated for 10 min at Room Temperature (RT) and 90 μl of dilution buffer were added to each of them. After that, 10 μl of each diluted sample was loaded on a well of a 96-wells plate and mixed with an equal amount of 2X PCR reaction buffer.

133 The PCR programe that was used is the following: 1) The RT reaction was performed at 42°C for 20 min, 2) 95°C for 2 min (in order to activate the hotStart

Taq), 3) 95°C for 5 s 3) 40 cycles of amplification comprising a denaturation step at 95°C for 5 s, an annealing step at 58°C for 5 s, followed by extension at 72°C for 20 s and finished by an recording of the SyBrGreen signal at 80°C for 7 s, and

4) 65°C for 5 s and a 95°C for 5 s incrementing step (where another acquisition of the SyBrGreen signal was performed), in order to establish a melting curve. All the reactions and the quantifications were performed using the Bio-Rad CFX96 real-time PCR detection system.

RESULTS

The capsid sequences derived from clinical isolates NRC10-5, NRC10-5 (T41S) and NRC10-6 are sufficient to confer viral susceptibility to human TRIM5α.

In the previous study by Battivelli and colleagues, gag-protease sequences derived from clinical isolates of HIV-1 were found to show an increased susceptibility to human TRIM5α 27. Indeed, NRC2, NRC10-6 and NRC10-5, as well as a variant of NRC10-5 with the Threonine mutation at position 41 of CA reverted to Serine, were more restricted by human TRIM5α than was the laboratory-adapted strain (NL4.3). Although this phenotype was suggested to be

CA-dependent, many mutations outside the capsid, notably in the matrix (MA) and in the protease (PR) sequences could potentially contribute to the acquired characteristic, as shown in Figure 1 and resumed in Table 1.

134 In order to investigate the contribution of the variations within the CA protein in the increased susceptibility to human TRIM5α, we cloned the different capsid sequences in the p8.9NdSB vector, using the NotI/ApaI unique restriction sites.

As shown in Figure 3A, the silencing of TRIM5α by miRNA-based shRNA rescued the infectivity of the viruses containing the capsid sequences from NRC10-6,

NRC10-5 and NRC10-5 (T41S). Compared to WT HIV-1, the restriction by

TRIM5α was higher, ranging from 2-3 fold for NRC10-6 to 8-10 fold for NRC10-5

(T41S). NRC10-5-derived virus showed an intermediary phenotype with 3-7 fold more susceptibility to TRIM5α than the WT.

The data suggest that the CA of HIV-1 variants can be better recognized by human TRIM5α, in naturally occurring mutations, as shown by the cases of the

NRC10-6 and NRC10-5 viruses.

HIV-1 variant name 260/280 ratio 260/230 ratio

WT 1.9 2.48 D71E 1.91 2.3

A86V 2.13 3.13 I96M 1.9 2.37 K132R 1.9 2.4

V135I 1.95 2.56

M136L 1.99 2.75 V141I 1.89 2.34 R158K 1.92 2.35

KVM 1.9 2.4

Table 1: The ratios of DNA/protein and DNA/RNA from each mutant or WT DNA sample, as determined by the optical densities (ODs). All the DNA samples showed similar purities.

135

!"# $"#

Figure 8: The alignments of the matrix (A) and protease (B) amino acid sequences from the different HIV-1 variants from Battivelli and colleagues (ref). From the top to the bottom, NL4.3 (WT), NRC10-5, NRC10-6 and NTRC2. The sequence of the WT capsid sequence is shown on the top of the alignment of the matrix sequences, to indicate the start of p24.

The contribution of different regions of HIV-1 CA to the increased recognition by human TRIM5α.

We further aimed to investigate the regions of HIV-1 CA that participate to the gained susceptibility to human TRIM5α by the different NRC viruses.

The restriction factor recognizes the hexameric lattice of the retroviral capsid, composed by hexamers and pentamers of the N-terminal domain of the viral protein 31,36,37.

136 We thus reasoned that changes residing within this region of CA would be more likely to have an impact on the sensibility to TRIM5α-mediated restriction.

The alignment of the different sequences of N-terminal domain of the capsid showed revealed different mutations that were shared among the more restricted variants (Figure 2). As NRC2 was previously shown to have a modest increased recognition by human TRIM5α, we used the sequence of this virus, together with that of the WT to help us to discriminate between mutations having a higher contribution to the gained TRIM5α restriction. Following this rationale, the combination of the six mutations E71D-V86A-M96I-R132K-I135V-

L136M (hereafter called DAIKVM) was a good candidate for recapitulating the phenotype observed with the NRC10-6 variant. In addition, the I141V-K158R double substitution was shared among the more restricted variants (NRC10-5 and NRC10-5 (T41S), and we therefore aimed to include them in the DAIKVM background, resulting in the DAIKVMVR mutant. All the mutations, except the

M96I, were distributed between the helices 4 and 7 and the cyclophilin A- binding loop (Figure 2).

We therefore cloned these two new mutant capsid sequences in the p8.9NdSB vector as described above and tested the virus for rescue by TRIM5α knockdown (KD). We found that DAIKVM mutant could not account for the NRC10-6 susceptibility to human TRIM5α (Figure 3B). Conversely, DAIKVMVR was able to recapitulate at least the half of the effect of the restriction factor on NRC10-5 and

NRC10-5 (T41S). The efficiency of Trim5α KD was confirmed by the potent rescue of N-MLV, as opposed to that of B-MLV (Figure 3B).

The levels of infectivity of NRC10-5 (T41S) and DAIKVMVR were low compared to the WT and NRC10-6 viruses and the silencing of Trim5α did not completely

137 rescue this defect (Table 2). The N-MLV control retained a much lower defect of infectivity in the Trim5α KD-background (data not shown), showing that the efficiency of the silencing could not account for the poor remaining infectivity of the two HIV-1 variants. Interestingly, DAIKVMVR showed a 2 fold lower defect than NRC10-5 (T41S), that is the same magnitude of difference observed for the susceptibility to human TRIM5α between the two variants. This observation suggests that the mechanism governing a better recognition of the capsids of the two HIV-1 mutants is also influenced by other factors that have an impact on

HIV-1 infection.

In order to gain more insights into the contribution of the different regions of the

N-terminal domain of CA, we aimed to revert back individually each of the mutations within the DAIKVMVR background. We additionally included in the screening a triple mutant containing the KVM motif, which is concentrated in the seventh α-helix, so as to evaluate a potential major contribution by this region.

As indicated in Table 3, the viruses derived from most of the point mutants presented a defect before or at budding that prevented the presence of reverse- transcriptase (RT) in the supernatants of virus-producer cells, as assessed by

SyBr-green-based- qPCR-based product enhanced RT (SG-PERT) assay. This defect was not due to a lower purity of the DNA samples used for producing the virus, as the 260/280 and 260/230 ratios were similar for all the variants (data not shown). Only the A86V and K132R revertants, as well as the variant containing the KVM motif produced viral-like particles (VLPs) at a similar extent than that of WT. We therefore could only examine these three variants for their impact on TRIM5α restriction.

138 We found that the reversion of the A86 and K132 mutations, which are located in the CypA-binding loop and the α-helix 7 respectively, to the WT residues, almost completely abolished the sensitivity of DAIKVMVR to human TRIM5α (Figure 4).

The KVM motif was not sufficient to confer a better recognition by the restriction factor.

These data suggest that the mutations affecting the 4th and 7th helical regions as well as the CypA-binding loop modulate the recognition of HIV-1 capsid by human TRIM5α.

Figure 9: The alignment of the sequences of the capsid N-terminal domain from the different HIV-1 variants from the study by Battivelli and colleagues (ref). Common mutations between the three more restricted variants are indicated with yellow stars above the corresponding amino acid. The position of the mutation shared only by the two more restricted variant is depicted with a pink star. The other mutation that is only common to the more restricted viruses is in the C-terminal domain of the capsid and therefore it is not indicated in this figure. The orange and blue squares define the helices 4 and 7, respectively. The green bar is depicted above the Cyclophilin A-binding loop.

139

D HT1080 cells

!"# K

5 WT

M 10

I NRC10-6 R

T NRC10-5

y NRC10-5 T41S b

e s a e r c n i -

d 1 l

o 1 0.3 F Dilution of viral supernatant

$"# HT1080 cells

D 384.5 K 100 !

5 WT M

I NRC10-5 T41S R

T DAIKVM

y DAIKVMVR b N-MLV e

s B-MLV a 1 e r c n i -

d 0.1 l

o 1 0.3 F MOI

Figure 10: The mutations in the capsid sequence of HIV-1 variants are sufficient to confer TRIM5α sensitivity to a p8.9-based virus. HT1080 cells stably expressing a miRNA-based shRNA targeting luciferase (control) or Trim5α were challenged with increasing doses of viruses. A) The capsid sequences derived from the gag-proteases sequences from Battivelli and colleaguesshow variable susceptibilities to human TRIM5α, as determined by the effect of TRIM5α silencing in HT1080 cells. B) The mutant with a smaller number of substitutions (DAIKVMVR) confers to NL4.3 at least the half of TRIM5α sensitivity compared to the most susceptible variant (NRC10-5 T41S), in HT1080 cells. All the data are represented as fold-increase by TRIM5α knock-down (KD).

140 HT1080 cells

260.2 72.4 D 307.5 K

5 WT

M 10

I K132R R

T KVM only

y N b B e

s A86V a e r c n i -

d 1 l o 1/5 F 1/45 1/15 Dilution of viral supernatant

Figure 11: The reversion of mutations in the CyclophilinA-binding loop or in the helix 7 abolishes the sensitivity to TRIM5α. HT1080 cells stably expressing a miRNA targeting luciferase (control) or Trim5α were challenged with increasing doses of viruses. The TRIM5α-restricted N-MLV was used as a positive control. The unrestricted B-MLV was used as a negative control. The data are represented as fold-increase by TRIM5α knock-down (KD).

DISCUSSION

CypA can have different impacts on HIV-1 infectivity. Notably, it was shown that in cells from non-human primates, impeding the CypA-CA binding could rescue HIV-1 levels 24,38. The observation that CypA can block HIV-1 nuclear import in some human cell lines and promote the reverse-trasncription in others 39, argues in favor of the existence of a restriction factor, that is CypA-dependent and that is only expressed in particular cells. Originally, the analysis of a human cell line that lost the N-MLV restriction phenotype and where HIV-1 was CypA-independent, showed that TRIM5α sequence and expression were not altered 40. These results showed that TRIM5α was not the CypA-dependent restriction factor, yet it is conceivable, however, that TRIM5α could bind this CypA co-factor.

141 In this model, the interaction between the HIV-1 capsid and CypA would, however, impede the early recognition of CA by TRIM5α and lead to the rescue of the reverse-transcription blockade. Interestingly, the TRIM5α version of the Aotus trivirgatus monkey is a fusion of the RBCC with the CypA domain, which binds to the HIV-1 capsid lattice, leading to the restriction of the reverse-transcription and nuclear entry 21,41. The fact that rhesus macaque TRIM5α is also able to restrict HIV-1 at the two different steps 41,42 further suggests that the restriction factor and CypA may bind to each other or share a common cofactor, necessary for the later step blockade. Curiously, it was previously shown that, while different NRC variants were more sensitive to human TRIM5α when the binding to CypA was impeded, NRC10-5 was less 33,43. These observations suggest that CypA promoted the recognition of NRC10-5 by TRIM5α, indicating that in some instances, the CypA-capsid binding is not competitive towards this restriction factor. The fact that only viral DNA synthesis was affected suggests that the CypA-TRIM5α collaboration is, in this case, at the first blockade step. Here, we show that mutations in the CypA-binding loop, and in the α-helical regions 4 and 7, influences the sensitivity of HIV-1 to the TRIM5α version that is expressed in its natural host. Importantly, we found that the V86 residue located in the CypA-binding region have a strong effect on TRIM5α-mediated recognition. This data is in agreement with previous studies 33,43, where the recognition of NRC10-5 by TRIM5α was influenced by CypA. Interestingly, previous studies correlated the V86M and V86E mutations to a decrease in CypA-dependent human TRIM5α-mediated inhibition of HIV-1 nuclear import 44 and in rhesus TRIM5α-dependent restriction 45. Further studies will be needed to determine if the V86 residue dictates vulnerability to TRIM5α in a CA context-dependence and if different restriction steps are affected. The finding that three mutations are concentrated in the seventh α-helix and that reverting one of them (K132) abolishes the sensitivity to human TRIM5α argues in favor of an important role of this region for the recognition by the human restriction factor.

142 We previously showed that TRIM5α-mediated recognition of the retroviral capsid induces the amplification of TRIM5α-dependent innate immune pathways 22. It would be interesting to investigate in the future the impact of the NRC10-5 variants, as well as that of DAIKVMVR, in the stimulation of innate immune signaling by TRIM5α. The magnitude of the response may be proportional to the differential recognition of HIV-1 by the restriction factor. Given that the TRIM5α orthologues from other species robustly restricts HIV-1, it may well be that a strong innate immune stimulation is only carried when the CA is strongly recognized by TRIM5α. Nevertheless, the study of the relations between human TRIM5α and CypA in the context of HIV-1 recognition could help us to better understand the mechanism of restriction. Furthermore, our findings could serve as a basis for future structural studies of the different regions of the N-terminal domain of HIV-1 CA and their significance on TRIM5α-mediated recognition.

143 REFERENCES

1. Goonetilleke, N. et al. The first T cell response to transmitted/founder virus contributes to the control of acute viremia in HIV-1 infection. Journal of Experimental Medicine 206, 1253–1272 (2009). 2. Zimmerli, S. C. et al. HIV-1-specific IFN-gamma/IL-2-secreting CD8 T cells support CD4-independent proliferation of HIV-1-specific CD8 T cells. Proc Natl Acad Sci USA 102, 7239–7244 (2005). 3. Kim, G. J., Lee, H. S., Hong, K.-J. & Kim, S. S. Dynamic correlation between CTL response and viral load in primary human immunodeficiency virus-1 infected Koreans. Virol. J. 7, 239 (2010). 4. Kearney, M. et al. Human Immunodeficiency Virus Type 1 Population Genetics and Adaptation in Newly Infected Individuals. J. Virol. 83, 2715– 2727 (2009). 5. Henn, M. R. et al. Whole Genome Deep Sequencing of HIV-1 Reveals the Impact of Early Minor Variants Upon Immune Recognition During Acute Infection. PLoS Pathog 8, –e1002529 (2012). 6. Borrow, P. et al. Antiviral pressure exerted by HIV-l-specific cytotoxic T lymphocytes (CTLs) during primary infection demonstrated by rapid selection of CTL escape virus. Nat Med 3, 205–211 (1997). 7. Cao, J., McNevin, J., Malhotra, U. & McElrath, M. J. Evolution of CD8+ T cell immunity and viral escape following acute HIV-1 infection. J Immunol 171, 3837–3846 (2003). 8. Jones, N. A. et al. Determinants of human immunodeficiency virus type 1 escape from the primary CD8+ cytotoxic T lymphocyte response. J Exp Med 200, 1243–1256 (2004). 9. Liu, Y., McNevin, J. P., Holte, S., McElrath, M. J. & Mullins, J. I. Dynamics of Viral Evolution and CTL Responses in HIV-1 Infection. PLoS ONE 6, e15639 (2011). 10. Hou, B., Reizis, B. & DeFranco, A. L. Toll-like receptors activate innate and adaptive immunity by using dendritic cell-intrinsic and -extrinsic mechanisms. Immunity 29, 272–282 (2008). 11. Spörri, R. & Sousa, C. R. E. Inflammatory mediators are insufficient for full dendritic cell activation and promote expansion of CD4+ T cell populations lacking helper function. Nat Immunol 6, 163–170 (2005). 12. Kratky, W., Sousa, C. R. E., Oxenius, A. & Spörri, R. Direct activation of antigen-presenting cells is required for CD8+ T-cell priming and tumor vaccination. (2011). 13. Luban, J. Innate immune sensing of HIV-1 by dendritic cells. Cell Host Microbe 12, 408–418 (2012). 14. Fonteneau, J. F. et al. Human immunodeficiency virus type 1 activates plasmacytoid dendritic cells and concomitantly induces the bystander maturation of myeloid dendritic cells. J. Virol. 78, 5223–5232 (2004). 15. Beignon, A. S. et al. Endocytosis of HIV-1 activates plasmacytoid dendritic cells via toll-like receptor-viral RNA interactions. J Clin Invest 115, 3265– 3275 (2005).

144 16. Mangeat, B. et al. Broad antiretroviral defence by human APOBEC3G through lethal editing of nascent reverse transcripts. Nature 424, 99–103 (2003). 17. Monroe, K. M., Yang, Z., Johnson, J. R., Geng, X. & Doitsh, G. IFI16 DNA Sensor Is Required for Death of Lymphoid CD4 T Cells Abortively Infected w ith HIV. (2013). 18. Cowan, S. et al. An intracellular block to primate lentivirus replication. 99, 11549–11551 (2002). 19. Hatziioannou, T., Perez-Caballero, D., Yang, A., Cowan, S. & Bieniasz, P. D. Retrovirus resistance factors Ref1 and Lv1 are species-specific variants of TRIM5alpha. Proc Natl Acad Sci USA 101, 10774–10779 (2004). 20. Stremlau, M. et al. The cytoplasmic body component TRIM5alpha restricts HIV-1 infection in Old World monkeys. Nature 427, 848–853 (2004). 21. Sayah, D. M., Sokolskaja, E., Berthoux, L. & Luban, J. Cyclophilin A retrotransposition into TRIM5 explains owl monkey resistance to HIV-1. Nature 430, 569–573 (2004). 22. Pertel, T. et al. TRIM5 is an innate immune sensor for the retrovirus capsid lattice. Nature 472, 361–365 (2011). 23. Hatziioannou, T., Perez-Caballero, D., Cowan, S. & Bieniasz, P. D. Cyclophilin interactions with incoming human immunodeficiency virus type 1 capsids with opposing effects on infectivity in human cells. J. Virol. 79, 176–183 (2005). 24. Keckesova, Z., Ylinen, L. M. J. & Towers, G. J. Cyclophilin A renders human immunodeficiency virus type 1 sensitive to Old World monkey but not human TRIM5 alpha antiviral activity. J. Virol. 80, 4683–4690 (2006). 25. Sokolskaja, E., Berthoux, L. & Luban, J. Cyclophilin A and TRIM5alpha independently regulate human immunodeficiency virus type 1 infectivity in human cells. J. Virol. 80, 2855–2862 (2006). 26. Kaumanns, P., Hagmann, I. & Dittmar, M. T. Retrovirology 3, 79 (2006). 27. Battivelli, E. et al. Gag cytotoxic T lymphocyte escape mutations can increase sensitivity of HIV-1 to human TRIM5alpha, linking intrinsic and acquired immunity. J. Virol. 85, 11846–11854 (2011). 28. Owens, C. M., Yang, P. C., Göttlinger, H. & Sodroski, J. Human and simian immunodeficiency virus capsid proteins are major viral determinants of early, postentry replication blocks in simian cells. J. Virol. 77, 726–731 (2002). 29. Cowan, S. et al. Cellular inhibitors with Fv1-like activity restrict human and simian immunodeficiency virus tropism. Proc Natl Acad Sci USA 99, 11914–11919 (2002). 30. Forshey, B. M., Shi, J. & Aiken, C. Structural requirements for recognition of the human immunodeficiency virus type 1 core during host restriction in owl monkey cells. J. Virol. 79, 869–875 (2004). 31. Stremlau, M. et al. Specific recognition and accelerated uncoating of retroviral capsids by the TRIM5alpha restriction factor. Proc Natl Acad Sci USA 103, 5514–5519 (2006). 32. Gamble, T. R. et al. Crystal structure of human cyclophilin A bound to the amino-terminal domain of HIV-1 capsid. Cell 87, 1285–1294 (1996). 33. Matsuoka, S., Dam, E., Lecossier, D., Clavel, F. & Hance, A. J. Modulation of HIV-1 infectivity and cyclophilin A-dependence by Gag sequence and

145 target cell type. Retrovirology 6, 21 (2009). 34. Zufferey, R. et al. Self-inactivating lentivirus vector for safe and efficient in vivo gene delivery. J. Virol. 72, 9873–9880 (1998). 35. Berthoux, L., Sebastian, S., Sokolskaja, E. & Luban, J. Lv1 inhibition of human immunodeficiency virus type 1 is counteracted by factors that stimulate synthesis or nuclear translocation of viral cDNA. J. Virol. 78, 11739–11750 (2004). 36. Shi, J. & Aiken, C. Saturation of TRIM5 alpha-mediated restriction of HIV-1 infection depends on the stability of the incoming viral capsid. Virology 350, 493–500 (2006). 37. Deshmukh, L. et al. Structure and Dynamics of Full-Length HIV-1 Capsid Protein in Solution. J. Am. Chem. Soc. (2013). doi:10.1021/ja406246z 38. Berthoux, L., Sebastian, S., Sokolskaja, E. & Luban, J. Cyclophilin A is required for TRIM5{alpha}-mediated resistance to HIV-1 in Old World monkey cells. Proc Natl Acad Sci USA 102, 14849–14853 (2005). 39. De Iaco, A. Microsoft Word - First_page.docx. 1–159 (2012). 40. Sayah, D. M. & Luban, J. Selection for loss of Ref1 activity in human cells releases human immunodeficiency virus type 1 from cyclophilin A dependence during infection. J. Virol. 78, 12066–12070 (2004). 41. Anderson, J. L. et al. Proteasome Inhibition Reveals that a Functional Preintegration Complex Intermediate Can Be Generated during Restriction by Diverse TRIM5 Proteins. J. Virol. 80, 9754–9760 (2006). 42. Wu, X., Anderson, J. L., Campbell, E. M., Joseph, A. M. & Hope, T. J. Proteasome inhibitors uncouple rhesus TRIM5alpha restriction of HIV-1 reverse transcription and infection. Proc Natl Acad Sci USA 103, 7465– 7470 (2006). 43. Battivelli, E. et al. Strain-specific differences in the impact of human TRIM5alpha, different TRIM5alpha alleles, and the inhibition of capsid- cyclophilin A interactions on the infectivity of HIV-1. J. Virol. 84, 11010– 11019 (2010). 44. Veillette, M. et al. The V86M mutation in HIV-1 capsid confers resistance to TRIM5a by abrogation of cyclophilin A-dependent restriction and enhancement of viral nuclear import. Retrovirology 10, 1–1 (2013). 45. Soll, S. J., Wilson, S. J., Kutluay, S. B., Hatziioannou, T. & Bieniasz, P. D. Assisted Evolution Enables HIV-1 to Overcome a High TRIM5α-Imposed Genetic Barrier to Rhesus Macaque Tropism. PLoS Pathog 9, e1003667 (2013).

146 SUPPLEMENTARY INFORMATION

NRC10-6: GCGGCCGCTGGTGAGAGATGGGTGCGAGAGCGTCGGTATTAAGCGGGGGAGAATTAGATAAATGGG AAAAAATTCGGTTAAGGCCAGGGGGAAAGAAACAATATAAACTAAAACATATAGTATGGGCAAGCA GGGAGCTAGAACGATTCGCAGTTAATCCTGGCCTTTTAGAGACATCAGAAGGCTGTAGACAAATACT GGGACAGCTACAACCATCCCTTCAGACAGGATCAGAAGAACTTAGATCATTATATAATACAATAGCA GTCCTCTATTGTGTGCATCAAAGGATAGATGTAAAAGACACCAAGGAAGCCTTAGATAAGATAGAG GAAGAGCAAAACAAAAGTAAGAAAAAGGCACAGCAAGCAGCAGCTGACACAGGAAACAACAGCCAG GTCAGCCAAAATTACCCTATAGTGCAGAACCTCCAGGGGCAAATGGTACATCAGGCCATATCACCTA GAACTTTAAATGCATGGGTAAAAGTAATAGAAGAGAAGGCTTTCAACCCAGAAGTAATACCCATGT TTACAGCATTATCAGAAGGAGCCACCCCACAAGATTTAAATACCATGCTAAACACAGTGGGGGGACA TCAAGCAGCCATGCAAATGTTAAAAGACACCATCAATGAGGAAGCTGCAGAATGGGATAGATTGCA TCCAGCGCATGCAGGGCCTATTGCACCAGGCCGAATAAGAGAACCAAGGGGAAGTGACATAGCAGGA ACTACTAGTACCCTTCAGGAACAAATAGCATGGATGACAAATAATCCACCTATCCCAGTAGGAGAAA TCTATAAAAAGTGGATAGTTATGGGATTAAATAAAATAGTAAGAATGTATAGCCCTACCAGCATTC TGGACATAAAACAAGGACCAAAGGAACCCTTTAGAGACTATGTAGACCGATTCTATAAAACTCTAA GAGCCGAGCAAGCTACACAAGAGGTAAAAAATTGGATGACAGAAACCTTGTTGGTCCAAAATGCGA ACCCAGATTGTAAGACTATTTTAAAAGCATTGGGACCAGCAGCGACACTAGAAGAAATGATGACAG CATGTCAGGGAGTGGGGGGACCCGGCCATAAAGCAAGAGTTTTGGCTGAAGCAATGAGCCAAGTAAC AAATCCAGCTACCATAATGATACAGAAAGGCAATTTTAGGAACCAAAGAAAGACTGTTAAGTGTTT CAATTGTGGCAAAGAAGGGCACATAGCCAAAAATTGCAGGGCCC

NRC10-5: GCGGCCGCTGGTGAGAGATGGGTGCGAGAGCGTCGGTATTAAGCGGGGGAGAATTAGATAAATGG GAAAAAATTCGGTTAAGGCCAGGGGGAAAGAAACAATATAAACTAAAACATATAGTATGGGCAA GCAGGGAGCTAGAACGATTCGCAGTTAATCCTGGCCTTTTAGAGACATCAGAAGGCTGTAGACAA ATACTGGGACAGCTACAACCATCCCTTCAGACAGGATCAGAAGAACTTAGATCATTATATAATAC AATAGCAGTCCTCTATTGTGTGCATCAAAGGATAGATGTAAAAGACACCAAGGAAGCCTTAGATA AGATAGAGGAAGAGCAAAACAAAAGTAAGAAAAAGGCACAGCAAGCAGCAGCTGACACAGGAAAC AACAGCCAGGTCAGCCAAAATTACCCTATAGTGCAGAACCTCCAGGGGCAAATGGTACATCAGGCC ATATCACCTAGAACTTTAAATGCATGGGTAAAAGTAATAGAAGAGAAGGCTTTCAACCCAGAAGT AATACCCATGTTTACAGCATTATCAGAAGGAGCCACCCCACAAGATTTAAATACCATGCTAAACAC AGTGGGGGGACATCAAGCAGCTATGCAAATGTTAAAAGACACCATCAATGAGGAAGCTGCAGAAT GGGATAGATTGCATCCAGCGCATGCAGGGCCTATTGCACCAGGCCAGATAAGAGAACCAAGGGGA AGTGACATAGCAGGAACTACTAGTACCCTTCAGGAACAAATAGCATGGATGACAAATAATCCACC TATCCCAGTAGGAGAAATCTATAAAAAATGGATAGTTATGGGATTAAATAAAGTAGTAAGAATG TATAGCCCTACCAGCATTCTGGACATAAAACAAGGACCAAGAGAACCCTTTAGAGACTATGTAGA CCGATTCTATAAAACTCTAAGAGCCGAGCAAGCTACACAAGAGGTAAAAAATTGGATGACAGAAA

147 CCTTGTTGGTCCAAAATGCGAACCCAGATTGTAAGACTATTTTAAAAGCATTGGGACCAGCAGCG ACACTAGAAGAAATGATGACAGCATGTCAGGGAGTGGGGGGACCCGGCCATAAAGCAAGAGTTTT GGCTGAAGCAATGAGCCAAGTAACAAATCCAGCTACCATAATGATACAGAAAGGCAATTTTAGGA ACCAAAGAAAGACTGTTAAGTGTTTCAATTGTGGCAAAGAAGGGCACATAGCCAAAAATTGCAGG GCCC

NRC10-5 (T41S): GCGGCCGCTGGTGAGAGATGGGTGCGAGAGCGTCGGTATTAAGCGGGGGAGAATTAGATAAATGG GAAAAAATTCGGTTAAGGCCAGGGGGAAAGAAACAATATAAACTAAAACATATAGTATGGGCAA GCAGGGAGCTAGAACGATTCGCAGTTAATCCTGGCCTTTTAGAGACATCAGAAGGCTGTAGACAA ATACTGGGACAGCTACAACCATCCCTTCAGACAGGATCAGAAGAACTTAGATCATTATATAATAC AATAGCAGTCCTCTATTGTGTGCATCAAAGGATAGATGTAAAAGACACCAAGGAAGCCTTAGATA AGATAGAGGAAGAGCAAAACAAAAGTAAGAAAAAGGCACAGCAAGCAGCAGCTGACACAGGAAAC AACAGCCAGGTCAGCCAAAATTACCCTATAGTGCAGAACCTCCAGGGGCAAATGGTACATCAGGCC ATATCACCTAGAACTTTAAATGCATGGGTAAAAGTAATAGAAGAGAAGGCTTTCAACCCAGAAGT AATACCCATGTTTTCAGCATTATCAGAAGGAGCCACCCCACAAGATTTAAATACCATGCTAAACAC AGTGGGGGGACATCAAGCAGCTATGCAAATGTTAAAAGACACCATCAATGAGGAAGCTGCAGAAT GGGATAGATTGCATCCAGCGCATGCAGGGCCTATTGCACCAGGCCAGATAAGAGAACCAAGGGGA AGTGACATAGCAGGAACTACTAGTACCCTTCAGGAACAAATAGCATGGATGACAAATAATCCACC TATCCCAGTAGGAGAAATCTATAAAAAATGGATAGTTATGGGATTAAATAAAGTAGTAAGAATG TATAGCCCTACCAGCATTCTGGACATAAAACAAGGACCAAGAGAACCCTTTAGAGACTATGTAGA CCGATTCTATAAAACTCTAAGAGCCGAGCAAGCTACACAAGAGGTAAAAAATTGGATGACAGAAA CCTTGTTGGTCCAAAATGCGAACCCAGATTGTAAGACTATTTTAAAAGCATTGGGACCAGCAGCG ACACTAGAAGAAATGATGACAGCATGTCAGGGAGTGGGGGGACCCGGCCATAAAGCAAGAGTTTT GGCTGAAGCAATGAGCCAAGTAACAAATCCAGCTACCATAATGATACAGAAAGGCAATTTTAGGA ACCAAAGAAAGACTGTTAAGTGTTTCAATTGTGGCAAAGAAGGGCACATAGCCAAAAATTGCAGG GCCC

DAIKVM: GCGGCCGCTGGTGAGAGATGGGTGCGAGAGCGTCGGTATTAAGCGGGGGAGAATTAGATAAATGGG AAAAAATTCGGTTAAGGCCAGGGGGAAAGAAACAATATAAACTAAAACATATAGTATGGGCAAGCA GGGAGCTAGAACGATTCGCAGTTAATCCTGGCCTTTTAGAGACATCAGAAGGCTGTAGACAAATACT GGGACAGCTACAACCATCCCTTCAGACAGGATCAGAAGAACTTAGATCATTATATAATACAATAGCA GTCCTCTATTGTGTGCATCAAAGGATAGATGTAAAAGACACCAAGGAAGCCTTAGATAAGATAGAG GAAGAGCAAAACAAAAGTAAGAAAAAGGCACAGCAAGCAGCAGCTGACACAGGAAACAACAGCCAG GTCAGCCAAAATTACCCTATAGTGCAGAACCTCCAGGGGCAAATGGTACATCAGGCCATATCACCTA GAACTTTAAATGCATGGGTAAAAGTAGTAGAAGAGAAGGCTTTCAGCCCAGAAGTAATACCCATGT TTTCAGCATTATCAGAAGGAGCCACCCCACAAGATTTAAATACCATGCTAAACACAGTGGGGGGACA TCAAGCAGCCATGCAAATGTTAAAAGACACCATCAATGAGGAAGCTGCAGAATGGGATAGATTGCA

148 TCCAGCGCATGCAGGGCCTATTGCACCAGGCCAGATAAGAGAACCAAGGGGAAGTGACATAGCAGGA ACTACTAGTACCCTTCAGGAACAAATAGGATGGATGACACATAATCCACCTATCCCAGTAGGAGAAA TCTATAAAAAATGGATAGTTATGGGATTAAATAAAATAGTAAGAATGTATAGCCCTACCAGCATTC TGGACATAAGACAAGGACCAAAGGAACCCTTTAGAGACTATGTAGACCGATTCTATAAAACTCTAA GAGCCGAGCAAGCTTCACAAGAGGTAAAAAATTGGATGACAGAAACCTTGTTGGTCCAAAATGCGA ACCCAGATTGTAAGACTATTTTAAAAGCATTGGGACCAGGAGCGACACTAGAAGAAATGATGACAG CATGTCAGGGAGTGGGGGGACCCGGCCATAAAGCAAGAGTTTTGGCTGAAGCAATGAGCCAAGTAAC AAATCCAGCTACCATAATGATACAGAAAGGCAATTTTAGGAACCAAAGAAAGACTGTTAAGTGTTT CAATTGTGGCAAAGAAGGGCACATAGCCAAAAATTGCAGGGCCC

DAIKVMVR: GCGGCCGCTGGTGAGAGATGGGTGCGAGAGCGTCGGTATTAAGCGGGGGAGAATTAGATAAATGG GAAAAAATTCGGTTAAGGCCAGGGGGAAAGAAACAATATAAACTAAAACATATAGTATGGGCAA GCAGGGAGCTAGAACGATTCGCAGTTAATCCTGGCCTTTTAGAGACATCAGAAGGCTGTAGACAA ATACTGGGACAGCTACAACCATCCCTTCAGACAGGATCAGAAGAACTTAGATCATTATATAATAC AATAGCAGTCCTCTATTGTGTGCATCAAAGGATAGATGTAAAAGACACCAAGGAAGCCTTAGATA AGATAGAGGAAGAGCAAAACAAAAGTAAGAAAAAGGCACAGCAAGCAGCAGCTGACACAGGAAAC AACAGCCAGGTCAGCCAAAATTACCCTATAGTGCAGAACCTCCAGGGGCAAATGGTACATCAGGCC ATATCACCTAGAACTTTAAATGCATGGGTAAAAGTAGTAGAAGAGAAGGCTTTCAGCCCAGAAGT AATACCCATGTTTTCAGCATTATCAGAAGGAGCCACCCCACAAGATTTAAATACCATGCTAAACAC AGTGGGGGGACATCAAGCAGCCATGCAAATGTTAAAAGACACCATCAATGAGGAAGCTGCAGAAT GGGATAGATTGCATCCAGCGCATGCAGGGCCTATTGCACCAGGCCAGATAAGAGAACCAAGGGGA AGTGACATAGCAGGAACTACTAGTACCCTTCAGGAACAAATAGGATGGATGACACATAATCCACC TATCCCAGTAGGAGAAATCTATAAAAAATGGATAGTTATGGGATTAAATAAAGTAGTAAGAATG TATAGCCCTACCAGCATTCTGGACATAAGACAAGGACCAAGAGAACCCTTTAGAGACTATGTAGA CCGATTCTATAAAACTCTAAGAGCCGAGCAAGCTTCACAAGAGGTAAAAAATTGGATGACAGAAA CCTTGTTGGTCCAAAATGCGAACCCAGATTGTAAGACTATTTTAAAAGCATTGGGACCAGGAGCG ACACTAGAAGAAATGATGACAGCATGTCAGGGAGTGGGGGGACCCGGCCATAAAGCAAGAGTTTT GGCTGAAGCAATGAGCCAAGTAACAAATCCAGCTACCATAATGATACAGAAAGGCAATTTTAGGA ACCAAAGAAAGACTGTTAAGTGTTTCAATTGTGGCAAAGAAGGGCACATAGCCAAAAATTGCAGG GCCC

AIKVMVR: GCGGCCGCTGGTGAGAGATGGGTGCGAGAGCGTCGGTATTAAGCGGGGGAGAATTAGATAAATGG GAAAAAATTCGGTTAAGGCCAGGGGGAAAGAAACAATATAAACTAAAACATATAGTATGGGCAA GCAGGGAGCTAGAACGATTCGCAGTTAATCCTGGCCTTTTAGAGACATCAGAAGGCTGTAGACAA ATACTGGGACAGCTACAACCATCCCTTCAGACAGGATCAGAAGAACTTAGATCATTATATAATAC AATAGCAGTCCTCTATTGTGTGCATCAAAGGATAGATGTAAAAGACACCAAGGAAGCCTTAGATA AGATAGAGGAAGAGCAAAACAAAAGTAAGAAAAAGGCACAGCAAGCAGCAGCTGACACAGGAAAC AACAGCCAGGTCAGCCAAAATTACCCTATAGTGCAGAACCTCCAGGGGCAAATGGTACATCAGGCC

149 ATATCACCTAGAACTTTAAATGCATGGGTAAAAGTAGTAGAAGAGAAGGCTTTCAGCCCAGAAGT AATACCCATGTTTTCAGCATTATCAGAAGGAGCCACCCCACAAGATTTAAATACCATGCTAAACAC AGTGGGGGGACATCAAGCAGCCATGCAAATGTTAAAAGAGACCATCAATGAGGAAGCTGCAGAAT GGGATAGATTGCATCCAGCGCATGCAGGGCCTATTGCACCAGGCCAGATAAGAGAACCAAGGGGA AGTGACATAGCAGGAACTACTAGTACCCTTCAGGAACAAATAGGATGGATGACACATAATCCACC TATCCCAGTAGGAGAAATCTATAAAAAATGGATAGTTATGGGATTAAATAAAGTAGTAAGAATG TATAGCCCTACCAGCATTCTGGACATAAGACAAGGACCAAGAGAACCCTTTAGAGACTATGTAGA CCGATTCTATAAAACTCTAAGAGCCGAGCAAGCTTCACAAGAGGTAAAAAATTGGATGACAGAAA CCTTGTTGGTCCAAAATGCGAACCCAGATTGTAAGACTATTTTAAAAGCATTGGGACCAGGAGCG ACACTAGAAGAAATGATGACAGCATGTCAGGGAGTGGGGGGACCCGGCCATAAAGCAAGAGTTTT GGCTGAAGCAATGAGCCAAGTAACAAATCCAGCTACCATAATGATACAGAAAGGCAATTTTAGGA ACCAAAGAAAGACTGTTAAGTGTTTCAATTGTGGCAAAGAAGGGCACATAGCCAAAAATTGCAGG GCCC DIKVMVR: GCGGCCGCTGGTGAGAGATGGGTGCGAGAGCGTCGGTATTAAGCGGGGGAGAATTAGATAAATGG GAAAAAATTCGGTTAAGGCCAGGGGGAAAGAAACAATATAAACTAAAACATATAGTATGGGCAA GCAGGGAGCTAGAACGATTCGCAGTTAATCCTGGCCTTTTAGAGACATCAGAAGGCTGTAGACAA ATACTGGGACAGCTACAACCATCCCTTCAGACAGGATCAGAAGAACTTAGATCATTATATAATAC AATAGCAGTCCTCTATTGTGTGCATCAAAGGATAGATGTAAAAGACACCAAGGAAGCCTTAGATA AGATAGAGGAAGAGCAAAACAAAAGTAAGAAAAAGGCACAGCAAGCAGCAGCTGACACAGGAAAC AACAGCCAGGTCAGCCAAAATTACCCTATAGTGCAGAACCTCCAGGGGCAAATGGTACATCAGGCC ATATCACCTAGAACTTTAAATGCATGGGTAAAAGTAGTAGAAGAGAAGGCTTTCAGCCCAGAAGT AATACCCATGTTTTCAGCATTATCAGAAGGAGCCACCCCACAAGATTTAAATACCATGCTAAACAC AGTGGGGGGACATCAAGCAGCCATGCAAATGTTAAAAGACACCATCAATGAGGAAGCTGCAGAAT GGGATAGATTGCATCCAGTGCATGCAGGGCCTATTGCACCAGGCCAGATAAGAGAACCAAGGGGA AGTGACATAGCAGGAACTACTAGTACCCTTCAGGAACAAATAGGATGGATGACACATAATCCACC TATCCCAGTAGGAGAAATCTATAAAAAATGGATAGTTATGGGATTAAATAAAGTAGTAAGAATG TATAGCCCTACCAGCATTCTGGACATAAGACAAGGACCAAGAGAACCCTTTAGAGACTATGTAGA CCGATTCTATAAAACTCTAAGAGCCGAGCAAGCTTCACAAGAGGTAAAAAATTGGATGACAGAAA CCTTGTTGGTCCAAAATGCGAACCCAGATTGTAAGACTATTTTAAAAGCATTGGGACCAGGAGCG ACACTAGAAGAAATGATGACAGCATGTCAGGGAGTGGGGGGACCCGGCCATAAAGCAAGAGTTTT GGCTGAAGCAATGAGCCAAGTAACAAATCCAGCTACCATAATGATACAGAAAGGCAATTTTAGGA ACCAAAGAAAGACTGTTAAGTGTTTCAATTGTGGCAAAGAAGGGCACATAGCCAAAAATTGCAGG GCCC

DAKVMVR: GCGGCCGCTGGTGAGAGATGGGTGCGAGAGCGTCGGTATTAAGCGGGGGAGAATTAGATAAATGG GAAAAAATTCGGTTAAGGCCAGGGGGAAAGAAACAATATAAACTAAAACATATAGTATGGGCAA GCAGGGAGCTAGAACGATTCGCAGTTAATCCTGGCCTTTTAGAGACATCAGAAGGCTGTAGACAA ATACTGGGACAGCTACAACCATCCCTTCAGACAGGATCAGAAGAACTTAGATCATTATATAATAC

150 AATAGCAGTCCTCTATTGTGTGCATCAAAGGATAGATGTAAAAGACACCAAGGAAGCCTTAGATA AGATAGAGGAAGAGCAAAACAAAAGTAAGAAAAAGGCACAGCAAGCAGCAGCTGACACAGGAAAC AACAGCCAGGTCAGCCAAAATTACCCTATAGTGCAGAACCTCCAGGGGCAAATGGTACATCAGGCC ATATCACCTAGAACTTTAAATGCATGGGTAAAAGTAGTAGAAGAGAAGGCTTTCAGCCCAGAAGT AATACCCATGTTTTCAGCATTATCAGAAGGAGCCACCCCACAAGATTTAAATACCATGCTAAACAC AGTGGGGGGACATCAAGCAGCCATGCAAATGTTAAAAGACACCATCAATGAGGAAGCTGCAGAAT GGGATAGATTGCATCCAGCGCATGCAGGGCCTATTGCACCAGGCCAGATGAGAGAACCAAGGGGA AGTGACATAGCAGGAACTACTAGTACCCTTCAGGAACAAATAGGATGGATGACACATAATCCACC TATCCCAGTAGGAGAAATCTATAAAAAATGGATAGTTATGGGATTAAATAAAGTAGTAAGAATG TATAGCCCTACCAGCATTCTGGACATAAGACAAGGACCAAGAGAACCCTTTAGAGACTATGTAGA CCGATTCTATAAAACTCTAAGAGCCGAGCAAGCTTCACAAGAGGTAAAAAATTGGATGACAGAAA CCTTGTTGGTCCAAAATGCGAACCCAGATTGTAAGACTATTTTAAAAGCATTGGGACCAGGAGCG ACACTAGAAGAAATGATGACAGCATGTCAGGGAGTGGGGGGACCCGGCCATAAAGCAAGAGTTTT GGCTGAAGCAATGAGCCAAGTAACAAATCCAGCTACCATAATGATACAGAAAGGCAATTTTAGGA ACCAAAGAAAGACTGTTAAGTGTTTCAATTGTGGCAAAGAAGGGCACATAGCCAAAAATTGCAGG GCCC

DAIVMVR: GCGGCCGCTGGTGAGAGATGGGTGCGAGAGCGTCGGTATTAAGCGGGGGAGAATTAGATAAATGG GAAAAAATTCGGTTAAGGCCAGGGGGAAAGAAACAATATAAACTAAAACATATAGTATGGGCAA GCAGGGAGCTAGAACGATTCGCAGTTAATCCTGGCCTTTTAGAGACATCAGAAGGCTGTAGACAA ATACTGGGACAGCTACAACCATCCCTTCAGACAGGATCAGAAGAACTTAGATCATTATATAATAC AATAGCAGTCCTCTATTGTGTGCATCAAAGGATAGATGTAAAAGACACCAAGGAAGCCTTAGATA AGATAGAGGAAGAGCAAAACAAAAGTAAGAAAAAGGCACAGCAAGCAGCAGCTGACACAGGAAAC AACAGCCAGGTCAGCCAAAATTACCCTATAGTGCAGAACCTCCAGGGGCAAATGGTACATCAGGCC ATATCACCTAGAACTTTAAATGCATGGGTAAAAGTAGTAGAAGAGAAGGCTTTCAGCCCAGAAGT AATACCCATGTTTTCAGCATTATCAGAAGGAGCCACCCCACAAGATTTAAATACCATGCTAAACAC AGTGGGGGGACATCAAGCAGCCATGCAAATGTTAAAAGACACCATCAATGAGGAAGCTGCAGAAT GGGATAGATTGCATCCAGCGCATGCAGGGCCTATTGCACCAGGCCAGATAAGAGAACCAAGGGGA AGTGACATAGCAGGAACTACTAGTACCCTTCAGGAACAAATAGGATGGATGACACATAATCCACC TATCCCAGTAGGAGAAATCTATAAAAGATGGATAGTTATGGGATTAAATAAAGTAGTAAGAATG TATAGCCCTACCAGCATTCTGGACATAAGACAAGGACCAAGAGAACCCTTTAGAGACTATGTAGA CCGATTCTATAAAACTCTAAGAGCCGAGCAAGCTTCACAAGAGGTAAAAAATTGGATGACAGAAA CCTTGTTGGTCCAAAATGCGAACCCAGATTGTAAGACTATTTTAAAAGCATTGGGACCAGGAGCG ACACTAGAAGAAATGATGACAGCATGTCAGGGAGTGGGGGGACCCGGCCATAAAGCAAGAGTTTT GGCTGAAGCAATGAGCCAAGTAACAAATCCAGCTACCATAATGATACAGAAAGGCAATTTTAGGA ACCAAAGAAAGACTGTTAAGTGTTTCAATTGTGGCAAAGAAGGGCACATAGCCAAAAATTGCAGG GCCC

DAIKMVR:

151

GCGGCCGCTGGTGAGAGATGGGTGCGAGAGCGTCGGTATTAAGCGGGGGAGAATTAGATAAATGG GAAAAAATTCGGTTAAGGCCAGGGGGAAAGAAACAATATAAACTAAAACATATAGTATGGGCAA GCAGGGAGCTAGAACGATTCGCAGTTAATCCTGGCCTTTTAGAGACATCAGAAGGCTGTAGACAA ATACTGGGACAGCTACAACCATCCCTTCAGACAGGATCAGAAGAACTTAGATCATTATATAATAC AATAGCAGTCCTCTATTGTGTGCATCAAAGGATAGATGTAAAAGACACCAAGGAAGCCTTAGATA AGATAGAGGAAGAGCAAAACAAAAGTAAGAAAAAGGCACAGCAAGCAGCAGCTGACACAGGAAAC AACAGCCAGGTCAGCCAAAATTACCCTATAGTGCAGAACCTCCAGGGGCAAATGGTACATCAGGCC ATATCACCTAGAACTTTAAATGCATGGGTAAAAGTAGTAGAAGAGAAGGCTTTCAGCCCAGAAGT AATACCCATGTTTTCAGCATTATCAGAAGGAGCCACCCCACAAGATTTAAATACCATGCTAAACAC AGTGGGGGGACATCAAGCAGCCATGCAAATGTTAAAAGACACCATCAATGAGGAAGCTGCAGAAT GGGATAGATTGCATCCAGCGCATGCAGGGCCTATTGCACCAGGCCAGATAAGAGAACCAAGGGGA AGTGACATAGCAGGAACTACTAGTACCCTTCAGGAACAAATAGGATGGATGACACATAATCCACC TATCCCAGTAGGAGAAATCTATAAAAAATGGATAATTATGGGATTAAATAAAGTAGTAAGAATG TATAGCCCTACCAGCATTCTGGACATAAGACAAGGACCAAGAGAACCCTTTAGAGACTATGTAGA CCGATTCTATAAAACTCTAAGAGCCGAGCAAGCTTCACAAGAGGTAAAAAATTGGATGACAGAAA CCTTGTTGGTCCAAAATGCGAACCCAGATTGTAAGACTATTTTAAAAGCATTGGGACCAGGAGCG ACACTAGAAGAAATGATGACAGCATGTCAGGGAGTGGGGGGACCCGGCCATAAAGCAAGAGTTTT GGCTGAAGCAATGAGCCAAGTAACAAATCCAGCTACCATAATGATACAGAAAGGCAATTTTAGGA ACCAAAGAAAGACTGTTAAGTGTTTCAATTGTGGCAAAGAAGGGCACATAGCCAAAAATTGCAGG GCCC

DAIKVVR: GCGGCCGCTGGTGAGAGATGGGTGCGAGAGCGTCGGTATTAAGCGGGGGAGAATTAGATAAATGGG AAAAAATTCGGTTAAGGCCAGGGGGAAAGAAACAATATAAACTAAAACATATAGTATGGGCAAGCA GGGAGCTAGAACGATTCGCAGTTAATCCTGGCCTTTTAGAGACATCAGAAGGCTGTAGACAAATACT GGGACAGCTACAACCATCCCTTCAGACAGGATCAGAAGAACTTAGATCATTATATAATACAATAGCA GTCCTCTATTGTGTGCATCAAAGGATAGATGTAAAAGACACCAAGGAAGCCTTAGATAAGATAGAG GAAGAGCAAAACAAAAGTAAGAAAAAGGCACAGCAAGCAGCAGCTGACACAGGAAACAACAGCCAG GTCAGCCAAAATTACCCTATAGTGCAGAACCTCCAGGGGCAAATGGTACATCAGGCCATATCACCTA GAACTTTAAATGCATGGGTAAAAGTAGTAGAAGAGAAGGCTTTCAGCCCAGAAGTAATACCCATGT TTTCAGCATTATCAGAAGGAGCCACCCCACAAGATTTAAATACCATGCTAAACACAGTGGGGGGACA TCAAGCAGCCATGCAAATGTTAAAAGACACCATCAATGAGGAAGCTGCAGAATGGGATAGATTGCA TCCAGCGCATGCAGGGCCTATTGCACCAGGCCAGATAAGAGAACCAAGGGGAAGTGACATAGCAGGA ACTACTAGTACCCTTCAGGAACAAATAGGATGGATGACACATAATCCACCTATCCCAGTAGGAGAAA TCTATAAAAAATGGATAGTCCTGGGATTAAATAAAGTAGTAAGAATGTATAGCCCTACCAGCATTC TGGACATAAGACAAGGACCAAGAGAACCCTTTAGAGACTATGTAGACCGATTCTATAAAACTCTAA GAGCCGAGCAAGCTTCACAAGAGGTAAAAAATTGGATGACAGAAACCTTGTTGGTCCAAAATGCGA ACCCAGATTGTAAGACTATTTTAAAAGCATTGGGACCAGGAGCGACACTAGAAGAAATGATGACAG CATGTCAGGGAGTGGGGGGACCCGGCCATAAAGCAAGAGTTTTGGCTGAAGCAATGAGCCAAGTAAC

152 AAATCCAGCTACCATAATGATACAGAAAGGCAATTTTAGGAACCAAAGAAAGACTGTTAAGTGTTT CAATTGTGGCAAAGAAGGGCACATAGCCAAAAATTGCAGGGCCC

DAIKVMR: GCGGCCGCTGGTGAGAGATGGGTGCGAGAGCGTCGGTATTAAGCGGGGGAGAATTAGATAAATGGG AAAAAATTCGGTTAAGGCCAGGGGGAAAGAAACAATATAAACTAAAACATATAGTATGGGCAAGCA GGGAGCTAGAACGATTCGCAGTTAATCCTGGCCTTTTAGAGACATCAGAAGGCTGTAGACAAATACT GGGACAGCTACAACCATCCCTTCAGACAGGATCAGAAGAACTTAGATCATTATATAATACAATAGCA GTCCTCTATTGTGTGCATCAAAGGATAGATGTAAAAGACACCAAGGAAGCCTTAGATAAGATAGAG GAAGAGCAAAACAAAAGTAAGAAAAAGGCACAGCAAGCAGCAGCTGACACAGGAAACAACAGCCAG GTCAGCCAAAATTACCCTATAGTGCAGAACCTCCAGGGGCAAATGGTACATCAGGCCATATCACCTA GAACTTTAAATGCATGGGTAAAAGTAGTAGAAGAGAAGGCTTTCAGCCCAGAAGTAATACCCATGT TTTCAGCATTATCAGAAGGAGCCACCCCACAAGATTTAAATACCATGCTAAACACAGTGGGGGGACA TCAAGCAGCCATGCAAATGTTAAAAGACACCATCAATGAGGAAGCTGCAGAATGGGATAGATTGCA TCCAGCGCATGCAGGGCCTATTGCACCAGGCCAGATAAGAGAACCAAGGGGAAGTGACATAGCAGGA ACTACTAGTACCCTTCAGGAACAAATAGGATGGATGACACATAATCCACCTATCCCAGTAGGAGAAA TCTATAAAAAATGGATAGTTATGGGATTAAATAAAATAGTAAGAATGTATAGCCCTACCAGCATTC TGGACATAAGACAAGGACCAAGAGAACCCTTTAGAGACTATGTAGACCGATTCTATAAAACTCTAA GAGCCGAGCAAGCTTCACAAGAGGTAAAAAATTGGATGACAGAAACCTTGTTGGTCCAAAATGCGA ACCCAGATTGTAAGACTATTTTAAAAGCATTGGGACCAGGAGCGACACTAGAAGAAATGATGACAG CATGTCAGGGAGTGGGGGGACCCGGCCATAAAGCAAGAGTTTTGGCTGAAGCAATGAGCCAAGTAAC AAATCCAGCTACCATAATGATACAGAAAGGCAATTTTAGGAACCAAAGAAAGACTGTTAAGTGTTT CAATTGTGGCAAAGAAGGGCACATAGCCAAAAATTGCAGGGCCC

DAIKVMV: GCGGCCGCTGGTGAGAGATGGGTGCGAGAGCGTCGGTATTAAGCGGGGGAGAATTAGATAAATGGG AAAAAATTCGGTTAAGGCCAGGGGGAAAGAAACAATATAAACTAAAACATATAGTATGGGCAAGCA GGGAGCTAGAACGATTCGCAGTTAATCCTGGCCTTTTAGAGACATCAGAAGGCTGTAGACAAATACT GGGACAGCTACAACCATCCCTTCAGACAGGATCAGAAGAACTTAGATCATTATATAATACAATAGCA GTCCTCTATTGTGTGCATCAAAGGATAGATGTAAAAGACACCAAGGAAGCCTTAGATAAGATAGAG GAAGAGCAAAACAAAAGTAAGAAAAAGGCACAGCAAGCAGCAGCTGACACAGGAAACAACAGCCAG GTCAGCCAAAATTACCCTATAGTGCAGAACCTCCAGGGGCAAATGGTACATCAGGCCATATCACCTA GAACTTTAAATGCATGGGTAAAAGTAGTAGAAGAGAAGGCTTTCAGCCCAGAAGTAATACCCATGT TTTCAGCATTATCAGAAGGAGCCACCCCACAAGATTTAAATACCATGCTAAACACAGTGGGGGGACA TCAAGCAGCCATGCAAATGTTAAAAGACACCATCAATGAGGAAGCTGCAGAATGGGATAGATTGCA TCCAGCGCATGCAGGGCCTATTGCACCAGGCCAGATAAGAGAACCAAGGGGAAGTGACATAGCAGGA ACTACTAGTACCCTTCAGGAACAAATAGGATGGATGACACATAATCCACCTATCCCAGTAGGAGAAA TCTATAAAAAATGGATAGTTATGGGATTAAATAAAGTAGTAAGAATGTATAGCCCTACCAGCATTC TGGACATAAGACAAGGACCAAAGGAACCCTTTAGAGACTATGTAGACCGATTCTATAAAACTCTAA GAGCCGAGCAAGCTTCACAAGAGGTAAAAAATTGGATGACAGAAACCTTGTTGGTCCAAAATGCGA

153 ACCCAGATTGTAAGACTATTTTAAAAGCATTGGGACCAGGAGCGACACTAGAAGAAATGATGACAG CATGTCAGGGAGTGGGGGGACCCGGCCATAAAGCAAGAGTTTTGGCTGAAGCAATGAGCCAAGTAAC AAATCCAGCTACCATAATGATACAGAAAGGCAATTTTAGGAACCAAAGAAAGACTGTTAAGTGTTT CAATTGTGGCAAAGAAGGGCACATAGCCAAAAATTGCAGGGCCC

KVM: GCGGCCGCTGGTGAGAGATGGGTGCGAGAGCGTCGGTATTAAGCGGGGGAGAATTAGATAAATGGG AAAAAATTCGGTTAAGGCCAGGGGGAAAGAAACAATATAAACTAAAACATATAGTATGGGCAAGCA GGGAGCTAGAACGATTCGCAGTTAATCCTGGCCTTTTAGAGACATCAGAAGGCTGTAGACAAATACT GGGACAGCTACAACCATCCCTTCAGACAGGATCAGAAGAACTTAGATCATTATATAATACAATAGCA GTCCTCTATTGTGTGCATCAAAGGATAGATGTAAAAGACACCAAGGAAGCCTTAGATAAGATAGAG GAAGAGCAAAACAAAAGTAAGAAAAAGGCACAGCAAGCAGCAGCTGACACAGGAAACAACAGCCAG GTCAGCCAAAATTACCCTATAGTGCAGAACCTCCAGGGGCAAATGGTACATCAGGCCATATCACCTA GAACTTTAAATGCATGGGTAAAAGTAGTAGAAGAGAAGGCTTTCAGCCCAGAAGTAATACCCATGT TTTCAGCATTATCAGAAGGAGCCACCCCACAAGATTTAAATACCATGCTAAACACAGTGGGGGGACA TCAAGCAGCCATGCAAATGTTAAAAGAGACCATCAATGAGGAAGCTGCAGAATGGGATAGATTGCA TCCAGTGCATGCAGGGCCTATTGCACCAGGCCAGATGAGAGAACCAAGGGGAAGTGACATAGCAGGA ACTACTAGTACCCTTCAGGAACAAATAGGATGGATGACACATAATCCACCTATCCCAGTAGGAGAAA TCTATAAAAAATGGATAGTTATGGGATTAAATAAAATAGTAAGAATGTATAGCCCTACCAGCATTC TGGACATAAGACAAGGACCAAAGGAACCCTTTAGAGACTATGTAGACCGATTCTATAAAACTCTAA GAGCCGAGCAAGCTTCACAAGAGGTAAAAAATTGGATGACAGAAACCTTGTTGGTCCAAAATGCGA ACCCAGATTGTAAGACTATTTTAAAAGCATTGGGACCAGGAGCGACACTAGAAGAAATGATGACAG CATGTCAGGGAGTGGGGGGACCCGGCCATAAAGCAAGAGTTTTGGCTGAAGCAATGAGCCAAGTAAC AAATCCAGCTACCATAATGATACAGAAAGGCAATTTTAGGAACCAAAGAAAGACTGTTAAGTGTTT CAATTGTGGCAAAGAAGGGCACATAGCCAAAAATTGCAGGGCCC

154

Chapter 4

GENERAL DISCUSSION

4.1 Introduction

The innate immune response is composed of complementary and interplaying routes. One route consists in the establishment of a general antiviral state that results from the detection of many different bacterial, fungal and viral molecules by PRRs. The second route involves intrinsic restriction factors and their direct targeting of viral components. Members of the TRIM family of proteins are involved in the regulation or the establishment of the innate immune signaling and some contribute to the production of an antiviral state 240,260,265,266.

In the recent years, some TRIM proteins were discovered to act themselves as PRRs for different viral molecules or viral-induced structures 265,334,335, activating the innate immune signaling upon their recognition. TRIM21, for example, recognizes antibody-bound virus by the binding of the PRYSPRY domain to the IgG molecule and mediate the so-called antibody- dependent intracellular neutralization (AIDN) 335. In common with this protein, TRIM5α contains a PRYSPRY that is essential for its PRR function and stimulates the innate immune signaling upon engagement. However, whereas TRIM21-mediated viral restriction completely rely on the proteasomal degradation, retroviral restriction by TRIM5α is a two steps mechanism and only the first one requires the proteasome 275,286,292,334.

155

Indeed, TRIM5α recognizes the hexameric retroviral capsid lattice and mediates a block at the reverse-transcription and pre-nuclear entry steps. Whereas the inhibition of the proteasome function precluded the restriction of the viral DNA synthesis, the blockade to the nuclear import of the pre-integration complex remained unaffected 292.

The particularity of TRIM5α resides in the fact that it is required for both the LPS-mediated establishment of a general antiviral state and the direct restriction of a specific retrovirus 265. At the basis of its dual functionality, TRIM5α binds to the TAK1 complex and activates the MAPK and NFκB pathways, in a similar manner than the TLR signaling component TRAF6 336. TRIM5α carry an intrinsic ability to stimulate the innate immune signaling, as shown by the activation of the AP-1- and NFκB- luciferase reporters when ectopically expressed in HEK 293T cells 265. Similarly to TRAF6, TRIM5α has a RING finger domain with an E3-ubiquitin ligase function and synthesizes free K63-linked poly-Ubiquitin chains that stimulate TAK1, resulting in the activation of the innate immune pathways 180,265.

The finding that silencing TAK1 or either of the two components of the E2 complex, Ubc13 and Uev1A, relieved TRIM5-mediated restriction 265, suggests that the intrinsic ability of TRIM5 to activate the innate immune response plays an important role in the blockade of specific retroviruses. Conversely, other studies found that the deletion of the RING domain in the owl monkey TRIM5Cyp or depleting the E1 enzyme in cells did not prevent retroviral restriction 272,288.

When ectopically expressed, TRIM5α localize to cytoplasmic aggregates that can be visualized by Immunofluorescence 200,257,337. The role of cytoplasmic bodies formation in TRIM5α-mediated restriction is still a matter of debate.

Even though the TRIM5 orthologues from some Old world and New world monkeys potently restrict HIV-1 in cell culture systems, the significance of the

156 retroviral blockade was not confirmed in vivo. However, given the strong positive-selection on which the PRYSPRY of the TRIM5 gene is subject 273, it is likely that the restriction factor has been and is presently confronted to a challenging number of retroviruses that shape its sequence. Originally considered to block very poorly HIV-1 infection, as determined with laboratory-adapted HIV-1 strains, human TRIM5α was shown to exhibit different degrees of restriction to some HIV-1 variants coming from clinical isolates 307,338. Given that, in the second study, the variants arose upon CTL pressure, it is therefore possible that TRIM5α-mediated HIV-1 restriction has a role in the control of HIV-1 in vivo.

The fact that HIV-1 can successfully replicate in human argues in favor of a poor effect of human TRIM5α, even in the instances where it exerts a stronger restriction after CTL-induced mutation of the CA. The lack of a restrictive TRIM5α when the first infection events arise precludes the clearance of the virus. It therefore appears important do further dissect the mechanisms of TRIM5α- mediated recognition and how this binding is influenced by the interaction with other cellular factors.

4.2 The ability of TRIM5 orthologues to induce the innate immune signaling correlates with the retroviral restriction capacity.

The owl monkey TRIM5Cyp strongly recognizes the HIV-1 capsid by the mean of the Cyclophilin A (CypA) domain and impose a potent block to viral replication 62. We showed that over-expression of human TRIM5α and owl monkey TRIM5Cyp lead to the activation of the AP-1- and NFκB promoters (Annex II) 265. In chapter 2.1, I showed that this feature of TRIM5 is conserved among primate, murine and feline orthologues suggesting that this function is widespread in mammals. In the mouse, tandem duplications of the TRIM12 and TRIM30 in the cluster corresponding to that of TRIM5 lead to the generation of seven orthologues 298.

157

We found that the different murine TRIM5 orthologues accounted for differential abilities to induce the innate immune promoters. Notably, the three TRIM12 proteins (TRIM12A, B and C) potently induced the promoters, in a TAK1-dependent manner, while the four TRIM30s were weak activators. We aimed to take advantage of these differential signaling activities of the different murine proteins in order to test the importance of TRIM5-mediated signaling in the retroviral restriction. As shown in chapter 2.1, when fused to the CypA domain, only the strong AP-1 and NFκB inducers could restrict HIV-1. These data confirm our previous study that showed the importance of TAK1 and the Ubc13/Uev1A E2 enzymes in the TRIM5α-mediated retroviral restriction (Annex I) 265.

Furthermore, a study showed that the murine restriction factor Fv1 can also be fused to the CypA domain and confer the ability to restrict HIV-1 101. Interestingly, we found that both Fv1n and Fv1b alleles could induce the AP-1 and NFκB promoters (data not shown). These findings further argue in favor of an important role of the innate immune signaling in TRIM5α-mediated retroviral restriction.

As mentioned previously, our findings are in contrast with other studies that showed the independence of from one part, the RING domain for TRIM5Cyp- mediated restriction and, for another part, the E1 enzymes 286,288. Given that, in the first study, the RING-deleted mutant was not assessed for its signaling ability, the possibility remained that this domain was dispensable for the activation of the innate immune promoters by TRIM5Cyp. In chapter 2.1, I showed that TRIM5Cyp did not require the RING or the B-box to induce the AP-1 pathway, opposite to what we observed for human TRIM5α (Annex I) 265. These findings illustrate the differences between TRIM5 and TRIM5Cyp that could explain the discrepancies with the RING-deletion study. Indeed, if the RING-deleted mutant could still bind to TAK1 and the E2 enzymes, the outcome of the restriction would be the same.

158 The contradictions observed between our data and that of the second study could rely, for its part, on the usage of different cell lines for each analysis. Whereas we evaluated the role of the innate immune signaling on human and feline cells (chapter 2) 265, the other team based their conclusions in the data obtained in murine cells 286. One could imagine that a factor that would be degraded by the activation of some innate immune pathway by TRIM5α, playing a negative impact on the retroviral restriction, may not be present in murine cells. Alternatively, the stimulation of the innate immune promoters being the consequence of an interaction with TAK1 and the E2 enzymes, it is possible that the sole binding to this protein platform is essential for the potential interaction with one or more other unknown factors that would have an effector function on retroviral restriction by TRIM5α.

Another question remaining is whether the stimulation of the signaling would target the first or the second step of the viral life cycle blocked by TRIM5Cyp. Given that the reverse-transcription starts very early upon entry into the cell but take 8-12 hours to complete 84,93, it is therefore conceivable that the products induced by the activation of the innate immune signaling could target any of the two steps. These questions could be investigated in future experiments.

The reasons why TRIM5α and TRIM5Cyp rely on different domains for the activation of the innate immune promoters remain unknown. The RING domain was shown to mediate the E3-ligase activity of TRIM5α 246,316,339,340. How the Δ-RF TRM5Cyp can substitute for the RING for the binding to TAK1 and the E2 enzymes is not clear. It is possible that the ectopically expressed TRIM5Cyp could interact with the endogenous TRIM5α or with an unknown E3 ligase, as for example a TRAF protein or another TRIM.

The B-box domains additionally contribute to the TRIM5α-mediated activation of the innate immune promoters (data not shown), and correlates with the observed decrease of restriction by rhesus TRIM5α 288. The function of this

159 domain is not well characterized. However, it was shown that it promoted the formation of TRIM5α higher-order assemblies 250,252,341. It is conceivable that if human TRIM5α show a partial dependence on this domain for inducing the AP-1 promoter, the reason would be that assemblies of various TRIM5α dimers could amplify the activation of the innate immune signaling.

We found that the L2 region played an important role in TRIM5Cyp-mediated activation of AP-1. If the owl monkey TRIM5Cyp L2 could mediate higher-order assemblies, this would explain why the B-Box is not required for its ability to activate the innate immune promoters. Interestingly, a role for the L2 in the formation of higher- order multimer and in retroviral restriction was previously suggested by two studies 315,317. While we did not examine directly, the contribution of the CC domain in the owl monkey TRIM5Cyp-mediated AP1-1 activation, we analyzed the phenotype of the RBCC protein. Interestingly, this construct was not able to induce the promoters.

The CC domain is responsible for mediating the dimerization between TRIM5α monomers 253,254. Thus, the result with the TRIM5Cyp RBCC suggests that the L2 and the other C-terminal sequences play an important role in signaling activation. The possibility remains, however, that the RBCC construct is not expressed. This was not checked in our work. In the future, the individual effect of the CC could be evaluated. It is very likely that this domain will play a very important role for two reasons: 1) the fact that the L2 contributed to the signaling suggested that the formation of higher order assemblies would be important, and this complexes requires the formation of dimers and 2) it was shown that CC-CypA fusions could restrict, whereas L2-CypA or CypA alone could not 288.

The importance of the formation of CBs by TRIM5α for retroviral restriction is not clear. One study showed that TRIM5Cyp constructs that did not form bodies when ectopically expressed still retained the ability to restrict 286. In a second

160 study, the use of the Hsp90 inhibitor geldanamycin that impeded the formation of CBs, did not precluded restriction by ectopically expressed TRIM5α variants. It is possible that these aggregates are an artificial effect of the ectopic expression of TRIM5, with no function. However, another study showed that TRIM5α trafficked between the CBs and the diffuse cytoplasm staining in an active way 337. The same team reported the visualization by Immunofluorescence of TRIM5α CBs that formed de novo around restricted viral particles 290. Furthermore, the finding that L2 mutants that could not form CBs lost their ability to restrict reinforces the idea that these structures contributes to restriction 315. The contribution of CBs formation to the ability to induce the innate immune signaling is unkown.

We found three different scenarios of murine TRIM5 orthologues in respect to the CBs formation: 1) those that activated the AP-1 promoter and concentrated to the CBs, 2) those that did not activate the MAPK pathway and still formed CBs and 3) those that did not signal and did not localized to the CBs (Chapter 2). These data suggested that the ability to activate the innate immune pathway did not correlate with the capacity to form CBs. Interestingly, the murine TRIM5 orthologues that both activated the innate immune promoter and formed CBs also were able to restrict HIV-1 when fused to the CypA domain, with the exception of TRIM30D. However, TRIM30D-Cyp activated the AP-1-luciferase more strongly than all the other TRIM30-Cyp fusions, raising the possibility that the two conditions of the formation of CBs and the activation of the AP-1 pathway could be fulfilled in this last case. More experiments will be needed to determine if the different TRIM5-Cyp variants exhibit the same localization pattern than their RBCC/ RBCC-PRYSPRY counterparts.

Additionally, we found that two motives in the L2, previously shown to be required for the localization of rhesus monkey TRIM5α to cytoplasmic bodies (CBs) and restriction 315, where strongly contributing to the innate immune

161 activation but were not totally essential. Indeed, deleting the half of the first motif and the whole second motif still induced AP-1 near 4 fold over an empty plasmid control. The fact that the different L2-deleted TRIM5Cyp retained some signaling activity in our hands suggested that the lack of CBs did not preclude the ability to activate the innate immune promoters. However, as we worked with TRIM5Cyp and not TRIM5α, the outcome on CBs formation could be different. The localization of the different L2-deletion mutants TRIM5Cyp in CBs could be evaluated in future studies and correlated to their ability to induce the AP-1 pathway.

4.3 The potential role of TRIM12C in the restriction of HIV-1 in murine cells.

In chapter 2.2, I showed that the silencing of TRIM12C in murine TA3 cells correlated with a rescue of HIV-1 infection, compared to the cells expressing the scrambled target sequence control. This finding provides the appealing possibility that TRIM12C could participate to the robust HIV-1 blockade observed in TA3 cells and other murine T cell lines 183. However, the ectopic expression of this murine TRIM5 orthologue in cell lines from different species failed to induce HIV-1 restriction (chapter 2). These findings suggest that in the hypothetical case that TRIM12C contributes to HIV-1 restriction, another murine factor would be necessary.

Interestingly, a previous study found that murine TRIM8 could block HIV-1 at an early replication step and at a magnitude similar to what we observed for the rescue of HIV-1 by the silencing of TRIM12C 261. These data raise the possibility that TRIM12C collaborates with TRIM8 to restrict HIV-1 in TA3 cells. Alternatively, TRIM12C could potentially up-regulate the levels of expression of TRIM8 and the silencing of the murine TRIM5 orthologue would thus result in decreased levels of the HIV-1 restriction factor. Curiously, TRIM8 does not carry a PRYSPRY domain that would recognize the HIV-1 CA. Other domains of TRIM8 could be involved in HIV-1-specific restriction or, alternatively, TRIM8 could bind to TRIM12C that could potentially recognize the PRYSPRY. In the study

162 where TRIM8 was expressed ectopically in human cells, the possibility remains that this protein could heterodimerize with the endogenous TRIM5α, which restricts 4 fold HIV-1 in the same experiment 261, and potentiate the HIV-1 blockade.

The fact that the restriction phenotype observed in TA3 cells is not saturable by incoming viral particles further argues for the models in which various TRIMs, including a TRIM that does not carry a PRYSPRY would be involved in HIV-1 restriction. Alternatively, the HIV-1 blockade in TA3 cells could be due to the absence of a cofactor necessary for HIV-1 replication and would not be the product of a restriction factor. Indeed, the effects observed with the TRIM12C KD could be non-specific and represent an experimental artifact.

In the future, the introduction of a non-targetable TRIM12C into TRIM12C- silenced TA3 cells in order to detect a potential restriction rescue and the analysis of the potential contribution of TRIM8 would contribute to answer to the question whereas the phenotype observed in murine T cells is due to a murine TRIM5 orthologue.

4.3 The pCG PstI/SalI vector can be used to evaluate the capsid-dependent restriction of Mo- N- and B-MLV.

In chapter 2.2, I additionally showed that the Mo-MLV-derived pCG gagpol plasmid that was engineered to carry unique PstI and SalI restriction sites flanking a region of MLV capsid could be used to reproduce the restriction specificity of Fv1 alleles and human TRIM5α. These data suggest that we could use this plasmid for future experiments in order to evaluate the restriction of different MLV capsids by Fv1 alleles and different TRIM5 orthologues. The possibility remains that some other capsid sequences would not be expressed or would present defects in the protein folding or assembly in this particular Mo- MLV context.

163 Nevertheless, the newly engineered plasmid is a useful tool to reproducibly evaluate the blockade of Mo-, N- and B-MLV by different restriction factors and to assess the efficiency of human TRIM5 KD cell lines. The fact that the capsid sequence is the only region that varies from the three strains of MLV in the pCG PstI/SalI context, further could allow the evaluation of the effects of specific capsid-recognition, as for example the induction of the innate immune pathways during TRIM5α-mediated restriction.

4.4 The pMICΔU3-GFP plasmid mediates an improved transduction of TA3 cells.

As murine TA3 cells block HIV-1, we designed a plasmid that was derived from the MSCV vector, in order to establish KD cell lines. We found, as shown in the chapter 2.2, that the combination of the pMIG-Cyp and pMIG-ΔU3 plasmids from the study by Asmal and colleagues 321 could be used to produce virus that could transduce 5 to 7 fold more efficiently TA3 cells than the HIV-1-derived pAGM- cotaining virus. We cloned the PuroR-MiR and BlastiR-MiR sequences downstream of the CypA promoter. In the future, these new plasmids should be evaluated for their capacity to confer the corresponding antibiotic resistance. We can conclude that the pMICΔU3 combination vector can be used to effectively transduce TA3 cells and potentially perform stable gene silencing in these murine cells.

4.5 The residues in the helices 4 and 7, as well as in the CypA-binding loop of the N-terminal domain of the HIV-1 CA could influence the recognition by human TRIM5α.

In chapter 3, I showed that the human TRIM5α-mediated restriction of the HIV-1 variants derived from the NRC10-5 clinical isolate was strongly influenced by residues located in three regions of the N-terminal domain of CA. The reversions

164 of the mutation in the CypA-binding loop and of one of the mutations in the 7th helix to the WT residues were found to rescue HIV-1 infectivity. All together, these data indicated that the residues mutated in the three regions of the N- terminal region of CA were important for the recognition by human TRIM5α, in the context of the minimal D71-A86-I96-K132-V135-M136-V141-R158 mutant. The significance of the effect of the different mutations from each domain in the gained sentivity of HIV-1 to human TRIM5α is not known. The fact that previous studies found a role for CypA-binding in the recognition of the initial variant derived from the clinical isolates NRC10-5 by human TRIM5α is in agreement with our data showing the abolishment of the sensitivity to TRIM5α, in the context of the NRC10-5-derived D71-A86-I96-K132-V135-M136- V141-R158 mutant, by the reversion of a the A86 mutation in the CypA-binding loop.

Interestingly, the V86 residue was shown to be involved in the Van der Waals interactions of HIV-1 CA with CypA, whereas other residues were found to mediate the hydrogen bonds-interactions - which are stronger forces - with the host protein 342. It is therefore conceivable that the V86A mutation may influence the recognition of CA by endogenous CypA, potentially resulting in a conformational change of the CypA-CA complex and alteration of the kinetics and stability of the retroviral core 343.

We can only speculate why CypA could influence the recognition of the NRC10- 5-derived variants by human TRIM5α in the same cells where other TRIM5α- sensitive variants did not show any positive effect on restriction. Potentially, CypA could be competing less for the binding of HIV-1 CA before reverse- transcription while still interacting with it at a later pre-nuclear entry level, where the second step of TRIM5α-mediated restriction could be reinforced by the mean of CypA-binding. Indeed, it was shown in unpublished data that CypA promotes reverse-transcription and in some contexts, can inhibit nuclear entry 98. Further studies would be needed to determine the relationship between TRIM5α and CypA and their potential contribution in retroviral restriction.

165

In regard of the contribution of the helix 7 to the TRIM5α-sensitive phenotype, we can hypothesize that the R132 mutation could induce an altered stability of the capsid hexameric lattice that would render HIV-1 more susceptible to human TRIM5α. Indeed, the X-ray crystallography of the HIV-1 CA hexamers revealed that the assembly of the hexameric lattice is promoted, in part, by the interaction between the helix 7 of the N-terminal domain and the helix 11 of the C-terminal domain 344.

The influence of the different HIV-1 CA regions in the recognition by TRIM5α could be further evaluated in future experiments. Our data could be used in combination with structural analyses in order to gain more insights into the interactions between the retroviral capsid, the restriction factor and CypA that govern the TRIM5α-mediated blockade.

4.6 Final word.

Among TRIM proteins, many of them can restrict early, late or both stages of retroviral replication 261. Notably, TRIM11 and TRIM15 could inhibit HIV-1 assembly and release from more than 20 fold to near 80 fold. Considering the outcome of the HIV-1 infection in human, these data show that the critical events are early in the viral replication cycle and that even though human carries two potent anti-HIV-1 restriction factors, the targeting of this viral step by these proteins is not enough to control the infection, highlighting the importance of TRIM5α function. It therefore appears fundamental to further elucidate the mechanism of TRIM5α- mediated restriction.

In this thesis, I showed that the ability to induce the innate immune signaling by a given TRIM5 orthologue plays an important role in retroviral restriction. Additionally, our work highlighted the ability of different regions of the HIV-1 capsid to modulate the recognition by human TRIM5α.

166 Altogether, the data presented in this thesis contributes to a better understanding of the requirements of the different features of TRIM5α for its ability to restrict a bound retrovirus and the different regions of the capsid that could facilitate its recognition.

Future experiments could potentially detect components of the innate immune response that are playing an effector function during TRIM5α-mediated restriction, leading to the hypothetical pharmacological exploitation of these components to increase the restriction of HIV-1 in human.

167

REFERENCES

1. Coffin, J. M., Hughes, S. H., Varmus, H. E., Boeke, J. D. & Stoye, J. P. Retrotransposons, Endogenous Retroviruses, and the Evolution of Retroelements. (Cold Spring Harbor Laboratory Press, 1997). 2. Mooslehner, K., Müller, U., Karls, U., Hamann, L. & Harbers, K. Structure and expression of a gene encoding a putative GTP-binding protein identified by provirus integration in a transgenic mouse strain. Mol Cell Biol 11, 886–893 (1991). 3. Coffin, J. M. in Genetic Diversity and Evolution of Retroviruses 176, 143– 164 (Springer Berlin Heidelberg, 1992). 4. Coffin, J. M. Structure, Replication, and Recombination of Retrovirus Genomes: Some Unifying Hypotheses. J Gen Virol 42, 1–26 (1979). 5. Coffin, J. M., Hughes, S. H. & Varmus, H. E. Retroviruses. (Cold Spring Harbor Laboratory Press, 1997). 6. Robertson, E. E., Bradley, A. A., Kuehn, M. M. & Evans, M. M. Germ-line transmission of genes introduced into cultured pluripotential cells by retroviral vector. Nature 323, 445–448 (1986). 7. Morgenstern, J. P. & Land, H. Advanced mammalian gene transfer: high titre retroviral vectors with multiple drug selection markers and a complementary helper-free packaging cell line. Nucleic Acids Res 18, 3587–3596 (1990). 8. Fine, P. E. P. Vectors and vertical transmission: an epidemiologic perspective. Annals of the New York Academy of Sciences 266, 173–194 (1974). 9. Johnson, W. E. & Coffin, J. M. Constructing primate phylogenies from ancient retrovirus sequences. in (1999). 10. Jern, P., Sperber, G. O. & Blomberg, J. Divergent patterns of recent retroviral integrations in the human and chimpanzee genomes: probable transmissions between other primates and chimpanzees. J. Virol. 80, 1367–1375 (2006). 11. Weiss, R. A. R. Introduction to oncogenesis by RNA tumor viruses. Haematol Blood Transfus 23, 423–427 (1978). 12. Goff, S. Goff: Retroviridae: The Retroviruses and Their Replicatio... - Google Scholar. (1999). 13. Coffin, J. M. J. et al. Proposal for naming host cell-derived inserts in retrovirus genomes. J. Virol. 40, 953–957 (1981). 14. Maeda, N. N., Fan, H. H. & Yoshikai, Y. Y. Oncogenesis by retroviruses: old and new paradigms. Rev. Med. Virol. 18, 387–405 (2007). 15. Ellermann, V. & Bang, O. Experimentelle Leukämie bei Hühnern. (1908). 16. Rous, P. A SARCOMA OF THE FOWL TRANSMISSIBLE BY AN AGENT SEPARABLE FROM THE TUMOR CELLS. J Exp Med 13, 397–411 (1911). 17. Weiss, R. A. The discovery of endogenous retroviruses. Retrovirology 3, 67 (2006). 18. Temin, H. M. H. THE EFFECTS OF ACTINOMYCIN D ON GROWTH OF ROUS SARCOMA VIRUS IN VITRO. Virology 20, 577–582 (1963). 19. Temin, H. M. H. & Mizutani, S. S. RNA-dependent DNA polymerase in

168 virions of Rous sarcoma virus. Nature 226, 1211–1213 (1970). 20. Baltimore, D. D. RNA-dependent DNA polymerase in virions of RNA tumour viruses. Nature 226, 1209–1211 (1970). 21. Gallo, R. C. R. & Meyskens, F. L. F. Evidence for the presence of RNA tumor virus components in human leukemia. Prog Exp Tumor Res 21, 216–249 (1977). 22. Barré-Sinoussi, F. et al. Isolation of a T-lymphotropic retrovirus from a patient at risk for acquired immune deficiency syndrome (AIDS). Science 220, 868–871 (1983). 23. Fouty, B. & Solodushko, V. in The Glucocorticoid Receptor in Retroviral Infection (InTech, 2011). doi:10.5772/21722 24. Rein, A. Nucleic acid chaperone activity of retroviral Gag proteins. rnabiology 7, 700–705 (2010). 25. Sleiman, D. et al. Initiation of HIV-1 reverse transcription and functional role of nucleocapsid-mediated tRNA/viral genome interactions. Virus Res 169, 324–339 (2012). 26. Katz, R. A. & Skalka, A. M. The Retroviral Enzymes. Annu Rev Biochem 63, 133–173 (1994). 27. Flint, S. J., Enquist, L. W., Racaniello, V. R. & Skalka, A. M. Flint: Reverse transcription and integration - Google Scholar. Principles of Virology ASM Press (2004). 28. Hu, W. S. & Hughes, S. H. HIV-1 Reverse Transcription. Cold Spring Harb Perspect Med 2, a006882–a006882 (2012). 29. Rovnak, J. J. & Quackenbush, S. L. S. Walleye dermal sarcoma virus: molecular biology and oncogenesis. Viruses 2, 1984–1999 (2010). 30. http://www.unaids.org/en/media/unaids/contentassets/documents /epidemiology/2013/gr2013/UNAIDS_Global_Report_2013_en.pdf. 31. Centers for Disease Control and Prevention. Centers for Disease Control and Prevention: Pneumocystis... - Google Scholar. Morbidity and Mortality Weekly Report (1981). 32. Valdiserri, R. O. Commentary: thirty years of AIDS in America: a story of infinite hope. AIDS Educ Prev 23, 479–494 (2011). 33. Gallo, R. C. et al. Isolation of human T-cell leukemia virus in acquired immune deficiency syndrome (AIDS). Science 220, 865–867 (1983). 34. Coffin, J. J. et al. What to call the AIDS virus? Nature 321, 10–10 (1986). 35. Hansasuta, P. HIV-1 transmission and acute HIV-1 infection. British Medical Bulletin 58, 109–127 (2001). 36. Haase, A. T. Targeting early infection to prevent HIV-1 mucosal transmission. Nature 464, 217–223 (2010). 37. Perreau, M. et al. Follicular helper T cells serve as the major CD4 T cell compartment for HIV-1 infection, replication, and production. Journal of Experimental Medicine 210, 143–156 (2013). 38. Saba, E. et al. HIV-1 sexual transmission: early events of HIV-1 infection of human cervico-vaginal tissue in an optimized ex vivo model. Mucosal Immunol 3, 280–290 (2010). 39. Pantaleo, G. & Fauci, A. S. Immunopathogenesis of HIV infection. Annu Rev Microbiol 50, 825–854 (1995). 40. Moris, A. et al. DC-SIGN promotes exogenous MHC-I-restricted HIV-1

169 antigen presentation. Blood 103, 2648–2654 (2004). 41. Piguet, V. & Steinman, R. M. The interaction of HIV with dendritic cells: outcomes and pathways. Trends Immunol 28, 503–510 (2007). 42. Izquierdo-Useros, N., Lorizate, M. & Puertas, M. C. Siglec-1 is a novel dendritic cell receptor that mediates HIV-1 trans-infection through recognition of viral membrane gangliosides. PLoS Biol (2012). 43. Izquierdo-Useros, N. et al. Sialyllactose in viral membrane gangliosides is a novel molecular recognition pattern for mature dendritic cell capture of HIV-1. PLoS Biol 10, e1001315 (2012). 44. Luban, J. Innate immune sensing of HIV-1 by dendritic cells. Cell Host Microbe 12, 408–418 (2012). 45. Mackewicz, C. E., Ortega, H. W. & Levy, J. A. CD8+ cell anti-HIV activity correlates with the clinical state of the infected individual. J Clin Invest 87, 1462–1466 (1991). 46. Lubong Sabado, R. et al. In vitro priming recapitulates in vivo HIV-1 specific T cell responses, revealing rapid loss of virus reactive CD4 T cells in acute HIV-1 infection. PLoS ONE 4, e4256 (2009). 47. McMichael, A. & Dorrell, L. The immune response to HIV. Medicine 37, 321–325 (2009). 48. Lackner, A. A., Lederman, M. M. & Rodriguez, B. HIV pathogenesis: the host. (2012). 49. Sharp, P. M. & Hahn, B. H. Origins of HIV and the AIDS Pandemic. Cold Spring Harb Perspect Med 1, a006841–a006841 (2011). 50. Sharkey, M. in Intrinsic Immunity 371, 105–122 (Springer Berlin Heidelberg, 2013). 51. Reeves, J. D. J. & Doms, R. W. R. Human immunodeficiency virus type 2. J Gen Virol 83, 1253–1265 (2002). 52. Montaner, J. S. et al. A randomized, double-blind trial comparing combinations of nevirapine, didanosine, and zidovudine for HIV-infected patients: the INCAS Trial. Italy, The Netherlands, Canada and Australia Study. JAMA 279, 930–937 (1998). 53. Richman, D. D. et al. The challenge of finding a cure for HIV infection. Science 323, 1304–1307 (2009). 54. Johnson, L. F. et al. Life Expectancies of South African Adults Starting Antiretroviral Treatment: Collaborative Analysis of Cohort Studies. PLoS Med. 10, e1001418 (2013). 55. Hütter, G. et al. Long-term control of HIV by CCR5 Delta32/Delta32 stem- cell transplantation. N Engl J Med 360, 692–698 (2009). 56. Persaud, D., Gay, H., Ziemniak, C. & Chen, Y. H. Absence of detectable HIV- 1 viremia after treatment cessation in an infant. … England Journal of … (2013). 57. Bell, N. M. & Lever, A. M. L. HIV Gag polyprotein: processing and early viral particle assembly. Trends in Microbiology 21, 136–144 (2013). 58. Ganser, B. K. B., Li, S. S., Klishko, V. Y. V., Finch, J. T. J. & Sundquist, W. I. W. Assembly and analysis of conical models for the HIV-1 core. Science 283, 80–83 (1998). 59. Deshmukh, L. et al. Structure and Dynamics of Full-Length HIV-1 Capsid Protein in Solution. J. Am. Chem. Soc. (2013). doi:10.1021/ja406246z 60. Saad, J. S. et al. Structural basis for targeting HIV-1 Gag proteins to the

170 plasma membrane for virus assembly. Proc Natl Acad Sci USA 103, 11364–11369 (2006). 61. Alfadhli, A., Still, A. & Barklis, E. Analysis of human immunodeficiency virus type 1 matrix binding to membranes and nucleic acids. J. Virol. 83, 12196–12203 (2009). 62. Sayah, D. M., Sokolskaja, E., Berthoux, L. & Luban, J. Cyclophilin A retrotransposition into TRIM5 explains owl monkey resistance to HIV-1. Nature 430, 569–573 (2004). 63. Saadatmand, J. J. & Kleiman, L. L. Aspects of HIV-1 assembly that promote primer tRNA(Lys3) annealing to viral RNA. Virus Res 169, 340–348 (2012). 64. Garrus, J. E. J. et al. Tsg101 and the Vacuolar Protein Sorting Pathway Are Essential for HIV-1 Budding. Cell 107, 11–11 (2001). 65. Louis, J. M., Ishima, R., Torchia, D. A. & Weber, I. T. in HIV-1 Protease: Structure, Dynamics, and Inhibition 55, 261–298 (Elsevier, 2007). 66. Das, K., Martinez, S. E., Bauman, J. D. & Arnold, E. HIV-1 reverse transcriptase complex with DNA and nevirapine reveals non-nucleoside inhibition mechanism. Nature Structural & Molecular Biology 19, 253–259 (2012). 67. Lapkouski, M., Tian, L., Miller, J. T., Le Grice, S. F. J. & Yang, W. Complexes of HIV-1 RT, NNRTI and RNA/DNA hybrid reveal a structure compatible with RNA degradation. Nature Structural & Molecular Biology 20, 230–236 (2013). 68. Forshey, B. M., Shi, J. & Aiken, C. Structural requirements for recognition of the human immunodeficiency virus type 1 core during host restriction in owl monkey cells. J. Virol. 79, 869–875 (2004). 69. Krishnan, L. L. & Engelman, A. A. Retroviral Integrase Proteins and HIV-1 DNA Integration. J Biol Chem 287, 40858–40866 (2012). 70. Craigie, R. & Bushman, F. D. HIV DNA integration. Cold Spring Harb Perspect Med 2, a006890–a006890 (2012). 71. De Iaco, A. A. & Luban, J. J. Inhibition of HIV-1 infection by TNPO3 depletion is determined by capsid and detectable after viral cDNA enters the nucleus. Retrovirology 8, 98–98 (2011). 72. Matreyek, K. A. & Engelman, A. The requirement for nucleoporin NUP153 during human immunodeficiency virus type 1 infection is determined by the viral capsid. J. Virol. 85, 7818–7827 (2011). 73. Matreyek, K. A. & Engelman, A. Viral and cellular requirements for the nuclear entry of retroviral preintegration nucleoprotein complexes. Viruses 5, 2483–2511 (2012). 74. Christ, F. F. et al. Transportin-SR2 imports HIV into the nucleus. Curr Biol 18, 1192–1202 (2008). 75. Cherepanov, P. et al. HIV-1 integrase forms stable tetramers and associates with LEDGF/p75 protein in human cells. J Biol Chem 278, 372–381 (2003). 76. Wilen, C. B., Tilton, J. C. & Doms, R. W. HIV: Cell Binding and Entry. Cold Spring Harb Perspect Med 2, a006866–a006866 (2012). 77. Eckert, D. M. & Kim, P. S. Mechanisms of viral membrane fusion and its inhibition. Annu Rev Biochem 70, 777–810 (2001). 78. Karn, J. & Stoltzfus, C. M. Transcriptional and Posttranscriptional

171 Regulation of HIV-1 Gene Expression. Cold Spring Harb Perspect Med 2, a006916–a006916 (2012). 79. Malim, M. H. & Emerman, M. HIV-1 accessory proteins--ensuring viral survival in a hostile environment. Cell Host Microbe 3, 388–398 (2008). 80. Sakuma, T., Barry, M. A. & Ikeda, Y. Lentiviral vectors: basic to translational. Biochem J 443, 603–618 (2012). 81. Schuitemaker, H. et al. Biological phenotype of human immunodeficiency virus type 1 clones at different stages of infection: progression of disease is associated with a shift from monocytotropic to T-cell-tropic virus population. J. Virol. 66, 1354–1360 (1992). 82. Berger, E. A. HIV entry and tropism: the chemokine receptor connection. AIDS 11 Suppl A, S3–16 (1997). 83. Berson, J. F. & Doms, R. W. Structure-function studies of the HIV-1 coreceptors. Semin Immunol 10, 12–12 (1998). 84. Arhel, N. Revisiting HIV-1 uncoating. Retrovirology 7, 96 (2010). 85. Grewe, C., Beck, A. & Gelderblom, H. R. HIV: early virus-cell interactions. J Acquir Immune Defic Syndr 3, 965–974 (1990). 86. Karageorgos, L., Li, P. & Burrell, C. Characterization of HIV replication complexes early after cell-to-cell infection. AIDS Res. Hum. Retroviruses 9, 817–823 (1993). 87. Iordanskiy, S., Berro, R., Altieri, M., Kashanchi, F. & Bukrinsky, M. Intracytoplasmic maturation of the human immunodeficiency virus type 1 reverse transcription complexes determines their capacity to integrate into chromatin. Retrovirology 3, 4 (2006). 88. McDonald, D. Visualization of the intracellular behavior of HIV in living cells. J Cell Biol 159, 441–452 (2002). 89. Nermut, M. V. & Fassati, A. Structural analyses of purified human immunodeficiency virus type 1 intracellular reverse transcription complexes. J. Virol. 77, 8196–8206 (2003). 90. Forshey, B. M. B., Schwedler, von, U. U., Sundquist, W. I. W. & Aiken, C. C. Formation of a human immunodeficiency virus type 1 core of optimal stability is crucial for viral replication. J. Virol. 76, 5667–5677 (2002). 91. Yamashita, M. & Emerman, M. Capsid is a dominant determinant of retrovirus infectivity in nondividing cells. J. Virol. 78, 5670–5678 (2004). 92. Burdick, R. C., Hu, W. S. & Pathak, V. K. Nuclear import of APOBEC3F- labeled HIV-1 preintegration complexes. in (2013). 93. Hulme, A. E. A., Perez, O. O. & Hope, T. J. T. Complementary assays reveal a relationship between HIV-1 uncoating and reverse transcription. Proc Natl Acad Sci USA 108, 9975–9980 (2011). 94. Koning, F. A., Goujon, C., Bauby, H. & Malim, M. H. Target cell-mediated editing of HIV-1 cDNA by APOBEC3 proteins in human macrophages. J. Virol. 85, 13448–13452 (2011). 95. Fassati, A. Multiple roles of the capsid protein in the early steps of HIV-1 infection. Virus Res 170, 15–24 (2012). 96. Krishnan, L., Matreyek, K. A., Oztop, I. & Lee, K. The requirement for cellular transportin 3 (TNPO3 or TRN-SR2) during infection maps to human immunodeficiency virus type 1 capsid and not integrase. Journal of … (2010). 97. Valle-Casuso, J. C. et al. TNPO3 is required for HIV-1 replication after

172 nuclear import but prior to integration and binds the HIV-1 core. J. Virol. 86, 5931–5936 (2012). 98. De Iaco, A. Microsoft Word - First_page.docx. 1–159 (2012). 99. Schaller, T. et al. HIV-1 capsid-cyclophilin interactions determine nuclear import pathway, integration targeting and replication efficiency. PLoS Pathog 7, e1002439 (2011). 100. Zhou, L. et al. Transportin 3 promotes a nuclear maturation step required for efficient HIV-1 integration. PLoS Pathog 7, e1002194 (2011). 101. Schaller, T., Ylinen, L. M. J., Webb, B. L. J., Singh, S. & Towers, G. J. Fusion of cyclophilin A to Fv1 enables cyclosporine-sensitive restriction of human and feline immunodeficiency viruses. J. Virol. 81, 10055–10063 (2007). 102. Yap, M. W., Dodding, M. P. & Stoye, J. P. Trim-cyclophilin A fusion proteins can restrict human immunodeficiency virus type 1 infection at two distinct phases in the viral life cycle. J. Virol. 80, 4061–4067 (2006). 103. Goujon, C. et al. Human MX2 is an interferon-induced post-entry inhibitor of HIV-1 infection. Nature 502, 559–562 (2013). 104. Lewis, P., Hensel, M. & Emerman, M. Human immunodeficiency virus infection of cells arrested in the cell cycle. EMBO J 11, 3053–3058 (1992). 105. Roe, T., Reynolds, T. C., Yu, G. & Brown, P. O. Integration of murine leukemia virus DNA depends on mitosis. EMBO J 12, 2099–2108 (1993). 106. Lewis, P. F. & Emerman, M. Passage through mitosis is required for oncoretroviruses but not for the human immunodeficiency virus. J. Virol. 68, 510–516 (1994). 107. Fassati, A. & GOFF, S. P. Characterization of intracellular reverse transcription complexes of Moloney murine leukemia virus. J. Virol. 73, 8919–8925 (1999). 108. Deminie, C. A. & Emerman, M. Functional exchange of an oncoretrovirus and a lentivirus matrix protein. J. Virol. 68, 4442–4449 (1994). 109. Seamon, J. A., Jones, K. S., Miller, C. & Roth, M. J. Inserting a nuclear targeting signal into a replication-competent Moloney murine leukemia virus affects viral export and is not sufficient for cell cycle-independent infection. J. Virol. 76, 8475–8484 (2002). 110. Alcamí, J. J. et al. Absolute dependence on kappa B responsive elements for initiation and Tat-mediated amplification of HIV transcription in blood CD4 T lymphocytes. EMBO J 14, 1552–1560 (1995). 111. Hakre, S., Chavez, L., Shirakawa, K. & Verdin, E. Epigenetic regulation of HIV latency. Curr Opin HIV AIDS 6, 19–24 (2011). 112. Sundquist, W. I. & Kräusslich, H.-G. HIV-1 assembly, budding, and maturation. Cold Spring Harb Perspect Med 2, a006924–a006924 (2012). 113. Chen, B. K. T cell virological synapses and HIV-1 pathogenesis. Immunologic research (2012). 114. Kawai, T. & Akira, S. The role of pattern-recognition receptors in innate immunity: update on Toll-like receptors. Nat Immunol 11, 373–384 (2010). 115. Arthur, J. S. C. & Ley, S. C. Mitogen-activated protein kinases in innate immunity. Nat Rev Immunol 13, 679–692 (2013). 116. Honda, K., Yanai, H. & Takaoka, A. Regulation of the type I IFN induction: a current view. International … (2005). 117. de Veer, M. J. et al. Functional classification of interferon-stimulated

173 genes identified using microarrays. J Leukoc Biol 69, 912–920 (2001). 118. Akira, S. S. & Takeda, K. K. Toll-like receptor signalling. Nat Rev Immunol 4, 499–511 (2004). 119. Mansell, A. & Jenkins, B. J. Dangerous liaisons between interleukin-6 cytokine and toll-like receptor families: A potent combination in inflammation and cancer. Cytokine and Growth Factor Reviews 24, 249– 256 (2013). 120. Hasan, U. U. et al. Human TLR10 is a functional receptor, expressed by B cells and plasmacytoid dendritic cells, which activates gene transcription through MyD88. J Immunol 174, 2942–2950 (2005). 121. Guan, Y. et al. Human TLRs 10 and 1 Share Common Mechanisms of Innate Immune Sensing but Not Signaling. J Immunol 184, 5094–5103 (2010). 122. Jin, M. S. et al. Crystal structure of the TLR1-TLR2 heterodimer induced by binding of a tri-acylated lipopeptide. Cell 130, 1071–1082 (2007). 123. Kang, J. Y. et al. Recognition of Lipopeptide Patterns by Toll-like Receptor 2-Toll-like Receptor 6 Heterodimer. Immunity 31, 873–884 (2009). 124. Cullender, T. C. et al. Innate and adaptive immunity interact to quench microbiome flagellar motility in the gut. Cell Host Microbe 14, 571–581 (2013). 125. Akira, S., Uematsu, S. & Takeuchi, O. Pathogen recognition and innate immunity. Cell 124, 783–801 (2006). 126. Kim, Y.-M., Brinkmann, M. M., Paquet, M.-E. & Ploegh, H. L. UNC93B1 delivers nucleotide-sensing toll-like receptors to endolysosomes. Nature 452, 234–U80 (2008). 127. Bovijn, C. et al. Identification of binding sites for myeloid differentiation primary response gene 88 (MyD88) and Toll-like receptor 4 in MyD88 adapter-like (Mal). Journal of Biological Chemistry 288, 12054–12066 (2013). 128. Xu, Y. et al. Structural basis for signal transduction by the Toll/interleukin-1 receptor domains. Nature 408, 111–115 (2000). 129. Dunne, A., Ejdeback, M., Ludidi, P. L., O apos Neill, L. A. J. & Gay, N. J. Structural complementarity of Toll/interleukin-1 receptor domains in Toll-like receptors and the adaptors Mal and MyD88. J Biol Chem 278, 41443–41451 (2003). 130. Gautam, J. K., Ashish, Comeau, L. D., Krueger, J. K. & Smith, M. F. Structural and functional evidence for the role of the TLR2 DD loop in TLR1/TLR2 heterodimerization and signaling. J Biol Chem 281, 30132–30142 (2006). 131. Jiang, Z., Georgel, P., Li, C. & Choe, J. Details of Toll-like receptor: adapter interaction revealed by germ-line mutagenesis. in (2006). 132. Yamamoto, M., Sato, S., Hemmi, H., Hoshino, K. & Kaisho, T. Role of adaptor TRIF in the MyD88-independent toll-like receptor signaling pathway. Science (2003). 133. Akira, S., Takeda, K. & Kaisho, T. Toll-like receptors: critical proteins linking innate and acquired immunity. Nat Immunol 2, 675–680 (2001). 134. Kagan, J. C. & Medzhitov, R. Phosphoinositide-mediated adaptor recruitment controls Toll-like receptor signaling. Cell 125, 943–955 (2006). 135. Kagan, J. C. et al. TRAM couples endocytosis of Toll-like receptor 4 to the

174 induction of interferon-beta. Nat Immunol 9, 361–368 (2008). 136. Tanimura, N., Saitoh, S., Matsumoto, F., Akashi-Takamura, S. & Miyake, K. Roles for LPS-dependent interaction and relocation of TLR4 and TRAM in TRIF-signaling. Biochem Biophys Res Commun 368, 94–99 (2008). 137. Rowe, D. C. et al. The myristoylation of TRIF-related adaptor molecule is essential for Toll-like receptor 4 signal transduction. Proc Natl Acad Sci USA 103, 6299–6304 (2006). 138. Nag, K. & Chaudhary, A. Mediators of Tyrosine Phosphorylation in Innate Immunity: From Host Defense to Inflammation onto Oncogenesis. Curr Signal Transduct Ther 4, 76–81 (2009). 139. Bhoj, V. G. & Chen, Z. J. Ubiquitylation in innate and adaptive immunity. Nature 458, 430–437 (2009). 140. Schulman, B. A. & Harper, J. W. Ubiquitin-like protein activation by E1 enzymes: the apex for downstream signalling pathways. Nat. Rev. Mol. Cell Biol. 10, 319–331 (2009). 141. Kawai, T. & Akira, S. Toll-like Receptor and RIG-1-like Receptor Signaling. Annals of the New York Academy of Sciences 1143, 1–20 (2008). 142. Takaoka, A. et al. Integral role of IRF-5 in the gene induction programme activated by Toll-like receptors. Nature 434, 243–249 (2005). 143. Van Duin, D., Medzhitov, R. & Shaw, A. C. Triggering TLR signaling in vaccination. Trends Immunol (2006). 144. Hornung, V. et al. 5'-Triphosphate RNA is the ligand for RIG-I. Science 314, 994–997 (2006). 145. Shaw, P. J., Lamkanfi, M. & Kanneganti, T.-D. NOD-like receptor (NLR) signaling beyond the inflammasome. Eur. J. Immunol. 40, 624–627 (2010). 146. Kato, H. et al. Length-dependent recognition of double-stranded ribonucleic acids by retinoic acid-inducible gene-I and melanoma differentiation-associated gene 5. Journal of Experimental Medicine 205, 1601–1610 (2008). 147. Kato, H. et al. Differential roles of MDA5 and RIG-I helicases in the recognition of RNA viruses. Nature 441, 101–105 (2006). 148. Sun, Q. et al. The Specific and Essential Role of MAVS in Antiviral Innate Immune Responses. Immunity 24, 633–642 (2006). 149. Kumar, H. et al. Essential role of IPS-1 in innate immune responses against RNA viruses. J Exp Med 203, 1795–1803 (2006). 150. Kawai, T. & Akira, S. The roles of TLRs, RLRs and NLRs in pathogen recognition. International Immunology 21, 317–337 (2009). 151. Zhang, Z. et al. The helicase DDX41 senses intracellular DNA mediated by the adaptor STING in dendritic cells. Nat Immunol 12, 959–965 (2011). 152. Abe, T. et al. STING recognition of cytoplasmic DNA instigates cellular defense. Mol. Cell 50, 5–15 (2013). 153. Unterholzner, L. et al. IFI16 is an innate immune sensor for intracellular DNA. Nat Immunol (2010). doi:10.1038/ni.1932 154. Kim, T. et al. Aspartate-glutamate-alanine-histidine box motif (DEAH)/RNA helicase A helicases sense microbial DNA in human plasmacytoid dendritic cells. Proc Natl Acad Sci USA 107, 15181–15186 (2010). 155. Rasaiyaah, J. et al. HIV-1 evades innate immune recognition through

175 specific cofactor recruitment. Nature 503, 402–405 (2013). 156. Gantke, T., Sriskantharajah, S. & Sadowski, M. IκB kinase regulation of the TPL-2/ERK MAPK pathway. Immunological … (2012). 157. Kaiser, F. et al. TPL-2 negatively regulates interferon-beta production in macrophages and myeloid dendritic cells. Journal of Experimental Medicine 206, 1863–1871 (2009). 158. Mielke, L. A., Elkins, K. L., Wei, L. & Starr, R. Tumor progression locus 2 (Map3k8) is critical for host defense against Listeria monocytogenes and IL-1β production. The Journal of … (2009). 159. Dumitru, C. D. et al. TNF-alpha induction by LPS is regulated posttranscriptionally via a Tpl2/ERK-dependent pathway. Cell 103, 1071–1083 (2000). 160. Martinez, F. O., Helming, L. & Gordon, S. Alternative activation of macrophages: an immunologic functional perspective. Annual review of … (2009). 161. O apos Keefe, S. J. et al. Chemical genetics define the roles of p38alpha and p38beta in acute and chronic inflammation. J Biol Chem 282, 34663– 34671 (2007). 162. Kim, C., Sano, Y., Todorova, K., Carlson, B. A. & Arpa, L. The kinase p38α serves cell type–specific inflammatory functions in skin injury and coordinates pro-and anti-inflammatory gene expression. Nature (2008). 163. Wisdom, R. AP-1: One Switch for Many Signals. Exp Cell Res 253, 6–6 (1999). 164. Yuan, Z. et al. Opposing Roles for ATF2 and c-Fos in c-Jun-Mediated Neuronal Apoptosis. Mol Cell Biol 29, 2431–2442 (2009). 165. Chiu, R., Angel, P. & Karin, M. Jun-B differs in its biological properties from, and is a negative regulator of, c-Jun. Cell 59, 979–986 (1989). 166. Shaulian, E. E. & Karin, M. M. AP-1 in cell proliferation and survival. Oncogene 20, 2390–2400 (2001). 167. Macián, F. F., López-Rodríguez, C. C. & Rao, A. A. Partners in transcription: NFAT and AP-1. Oncogene 20, 2476–2489 (2001). 168. Barbulescu, K. et al. Cutting edge: IL-12 and IL-18 differentially regulate the transcriptional activity of the human IFN-γ promoter in primary CD4+ T lymphocytes. J Immunol 160, 3642–3647 (1998). 169. Park, W.-R. W. et al. A mechanism underlying STAT4-mediated up- regulation of IFN-gamma induction inTCR-triggered T cells. International Immunology 16, 295–302 (2004). 170. Baccala, R., Hoebe, K., Kono, D. H., Beutler, B. & Theofilopoulos, A. N. TLR- dependent and TLR-independent pathways of type I interferon induction in systemic autoimmunity. Nat Med 13, 543–551 (2007). 171. Beg, A. A., Finco, T. S., Nantermet, P. V. & Baldwin, A. S. Tumor necrosis factor and interleukin-1 lead to phosphorylation and loss of I kappa B alpha: a mechanism for NF-kappa B activation. Mol Cell Biol 13, 3301– 3310 (1993). 172. Bonizzi, G. & Karin, M. The two NF-κB activation pathways and their role in innate and adaptive immunity. Trends Immunol 25, 280–288 (2004). 173. Hayden, M. S., West, A. P. & Ghosh, S. NF-κB and the immune response. Oncogene 25, 6758–6780 (2006). 174. Suhasini, M. M. & Pilz, R. B. R. Transcriptional elongation of c-myb is

176 regulated by NF-kappaB (p50/RelB). Oncogene 18, 7360–7369 (1999). 175. Sarnico, I. et al. NF-kappaB dimers in the regulation of neuronal survival. Int. Rev. Neurobiol. 85, 351–362 (2009). 176. Bonizzi, G. et al. Activation of IKKalpha target genes depends on recognition of specific kappaB binding sites by RelB:p52 dimers. EMBO J 23, 4202–4210 (2004). 177. Kopp, E. B. & Ghosh, S. in NF-κB and Rel Proteins in Innate Immunity 58, 1–27 (Elsevier, 1995). 178. Alvarez, S. E. et al. Sphingosine-1-phosphate is a missing cofactor for the E3 ubiquitin ligase TRAF2. Nature 465, 1084–1088 (2010). 179. Adhikari, A., Xu, M. & Chen, Z. J. Ubiquitin-mediated activation of TAK1 and IKK. Oncogene 26, 3214–3226 (2007). 180. Xia, Z.-P. et al. Direct activation of protein kinases by unanchored polyubiquitin chains. Nature 461, 114–119 (2009). 181. Oeckinghaus, A., Hayden, M. S. & Ghosh, S. Crosstalk in NF-κB signaling pathways. Nat Immunol 12, 695–708 (2011). 182. Tachibana, K. et al. CXCR4/fusin is not a species-specific barrier in murine cells for HIV-1 entry. J Exp Med 185, 1865–1870 (1997). 183. Baumann, J. G. et al. Murine T cells potently restrict human immunodeficiency virus infection. J. Virol. 78, 12537–12547 (2004). 184. Garber, M. E. et al. The interaction between HIV-1 Tat and human cyclin T1 requires zinc and a critical cysteine residue that is not conserved in the murine CycT1 protein. Genes Dev 12, 3512–3527 (1998). 185. Bieniasz, P. D. & Cullen, B. R. Multiple blocks to human immunodeficiency virus type 1 replication in rodent cells. J. Virol. 74, 9868–9877 (2000). 186. Zheng, Y. H. Human p32 protein relieves a posttranscriptional block to HIV replication in murine cells (vol 5, pg 611, 2003). Nat Cell Biol 5, 839– 839 (2003). 187. Mariani, R. et al. Mouse-human heterokaryons support efficient human immunodeficiency virus type 1 assembly. J. Virol. 75, 3141–3151 (2001). 188. FRIEND, C. Cell-free transmission in adult Swiss mice of a disease having the character of a leukemia. J Exp Med 105, 307–318 (1957). 189. Pincus, T. T., Hartley, J. W. J. & Rowe, W. P. W. A major genetic locus affecting resistance to infection with murine leukemia viruses. I. Tissue culture studies of naturally occurring viruses. J Exp Med 133, 1219–1233 (1971). 190. Pincus, T. T., Rowe, W. P. W. & Lilly, F. F. A major genetic locus affecting resistance to infection with murine leukemia viruses. II. Apparent identity to a major locus described for resistance to friend murine leukemia virus. J Exp Med 133, 1234–1241 (1971). 191. Jolicoeur, P. & Baltimore, D. Effect of Fv-1 gene product on proviral DNA formation and integration in cells infected with murine leukemia viruses. CORD Conference Proceedings 73, 2236–2240 (1976). 192. Shibata, R., Sakai, H., Kawamura, M., Tokunaga, K. & Adachi, A. Early replication block of human immunodeficiency virus type 1 in monkey cells. J Gen Virol 76, 2723–2730 (1995). 193. Himathongkham, S. S. & Luciw, P. A. P. Restriction of HIV-1 (Subtype B) Replication at the Entry Step in Rhesus Macaque Cells. Virology 219, 4–4 (1996).

177 194. Towers, G. et al. A conserved mechanism of retrovirus restriction in mammals. Proc Natl Acad Sci USA 97, 12295–12299 (2000). 195. Owens, C. M., Yang, P. C., Göttlinger, H. & Sodroski, J. Human and simian immunodeficiency virus capsid proteins are major viral determinants of early, postentry replication blocks in simian cells. J. Virol. 77, 726–731 (2002). 196. Towers, G., Collins, M. & Takeuchi, Y. Abrogation of Ref1 retrovirus restriction in human cells. J. Virol. 76, 2548–2550 (2002). 197. Cowan, S. et al. An intracellular block to primate lentivirus replication. 99, 11549–11551 (2002). 198. Cowan, S. et al. Cellular inhibitors with Fv1-like activity restrict human and simian immunodeficiency virus tropism. Proc Natl Acad Sci USA 99, 11914–11919 (2002). 199. Hatziioannou, T., Cowan, S., Goff, S. P., Bieniasz, P. D. & Towers, G. J. Restriction of multiple divergent retroviruses by Lv1 and Ref1. EMBO J 22, 385–394 (2003). 200. Stremlau, M. et al. The cytoplasmic body component TRIM5alpha restricts HIV-1 infection in Old World monkeys. Nature 427, 848–853 (2004). 201. Keckesova, Z., Ylinen, L. M. J. & Towers, G. J. The human and African green monkey TRIM5alpha genes encode Ref1 and Lv1 retroviral restriction factor activities. Proc Natl Acad Sci USA 101, 10780–10785 (2004). 202. Perron, M. J. et al. TRIM5alpha mediates the postentry block to N-tropic murine leukemia viruses in human cells. Proc Natl Acad Sci USA 101, 11827–11832 (2004). 203. Yap, M. W., Nisole, S., Lynch, C. & Stoye, J. P. Trim5alpha protein restricts both HIV-1 and murine leukemia virus. Proc Natl Acad Sci USA 101, 10786–10791 (2004). 204. Cen, S. et al. The interaction between HIV-1 Gag and APOBEC3G. J Biol Chem 279, 33177–33184 (2004). 205. Khan, M. A. et al. Analysis of the contribution of cellular and viral RNA to the packaging of APOBEC3G into HIV-1 virions. Retrovirology 4, 48 (2007). 206. Mehle, A. et al. Vif overcomes the innate antiviral activity of APOBEC3G by promoting its degradation in the ubiquitin-proteasome pathway. J Biol Chem 279, 7792–7798 (2004). 207. Mangeat, B. et al. Broad antiretroviral defence by human APOBEC3G through lethal editing of nascent reverse transcripts. Nature 424, 99– 103 (2003). 208. Mbisa, J. L. J. et al. Human immunodeficiency virus type 1 cDNAs produced in the presence of APOBEC3G exhibit defects in plus-strand DNA transfer and integration. J. Virol. 81, 7099–7110 (2007). 209. Zheng, Y.-H., Jeang, K.-T. & Tokunaga, K. Host restriction factors in retroviral infection: promises in virus-host interaction. Retrovirology 9, 112 (2012). 210. Bishop, K. N., Verma, M., Kim, E.-Y., Wolinsky, S. M. & Malim, M. H. APOBEC3G inhibits elongation of HIV-1 reverse transcripts. PLoS Pathog 4, e1000231 (2008). 211. Guo, F. et al. The interaction of APOBEC3G with human

178 immunodeficiency virus type 1 nucleocapsid inhibits tRNA3Lys annealing to viral RNA. J. Virol. 81, 11322–11331 (2007). 212. Iwatani, Y. et al. Deaminase-independent inhibition of HIV-1 reverse transcription by APOBEC3G. Nucleic Acids Res 35, 7096–7108 (2006). 213. Bishop, K. N., Holmes, R. K., Sheehy, A. M. & Davidson, N. O. Current Biology - Cytidine Deamination of Retroviral DNA by Diverse APOBEC Proteins. Current biology (2004). 214. Peng, G. et al. Myeloid differentiation and susceptibility to HIV-1 are linked to APOBEC3 expression. Blood 110, 393–400 (2007). 215. Doehle, B. P., Schäfer, A. & Cullen, B. R. Human APOBEC3B is a potent inhibitor of HIV-1 infectivity and is resistant to HIV-1 Vif. Virology (2005). 216. Bourara, K., Liegler, T. J. & Grant, R. M. Target Cell APOBEC3C Can Induce Limited G-to-A Mutation in HIV-1. PLoS Pathog 3, e153 (2007). 217. Dang, Y., Wang, X., Esselman, W. J. & Zheng, Y. H. Identification of APOBEC3DE as Another Antiretroviral Factor from the Human APOBEC Family. J. Virol. 80, 10522–10533 (2006). 218. Liddament, M. T., Brown, W. L., Schumacher, A. J. & Harris, R. S. APOBEC3F properties and hypermutation preferences indicate activity against HIV-1 in vivo. Curr Biol 14, 1385–1391 (2004). 219. Holmes, R. K., Koning, F. A. & Bishop, K. N. APOBEC3F Can Inhibit the Accumulation of HIV-1 Reverse Transcription Products in the Absence of Hypermutation COMPARISONS WITH APOBEC3G. Journal of Biological … (2007). 220. Neil, S. J. D., Sandrin, V., Sundquist, W. I. & Bieniasz, P. D. An interferon- alpha-induced tethering mechanism inhibits HIV-1 and Ebola virus particle release but is counteracted by the HIV-1 Vpu protein. Cell Host Microbe 2, 193–203 (2007). 221. Fitzpatrick, K. et al. Direct restriction of virus release and incorporation of the interferon-induced protein BST-2 into HIV-1 particles. PLoS Pathog 6, e1000701–e1000701 (2010). 222. Galão, R. P., Le Tortorec, A., Pickering, S., Kueck, T. & Neil, S. J. D. Innate sensing of HIV-1 assembly by Tetherin induces NFκB-dependent proinflammatory responses. Cell Host Microbe 12, 633–644 (2012). 223. Rice, G. I. et al. Mutations involved in Aicardi-Goutières syndrome implicate SAMHD1 as regulator of the innate immune response. Nat Genet 41, 829–832 (2009). 224. Baldauf, H.-M. et al. SAMHD1 restricts HIV-1 infection in resting CD4(+) T cells. Nat Med 18, 1682–1687 (2012). 225. Ahn, J. et al. HIV/simian immunodeficiency virus (SIV) accessory virulence factor Vpx loads the host cell restriction factor SAMHD1 onto the E3 ubiquitin ligase complex CRL4DCAF1. J Biol Chem 287, 12550– 12558 (2012). 226. Gao, G. Inhibition of Retroviral RNA Production by ZAP, a CCCH-Type Zinc Finger Protein. Science 297, 1703–1706 (2002). 227. Zhu, Y. et al. Zinc-finger antiviral protein inhibits HIV-1 infection by selectively targeting multiply spliced viral mRNAs for degradation. (2011). 228. Koonin, E. V. E. A new group of putative RNA helicases. Trends Biochem

179 Sci 17, 495–497 (1992). 229. Burdick, R. et al. P body-associated protein Mov10 inhibits HIV-1 replication at multiple stages. J. Virol. 84, 10241–10253 (2010). 230. Meister, G. et al. Identification of novel argonaute-associated proteins. Curr Biol 15, 2149–2155 (2005). 231. Furtak, V. et al. Perturbation of the P-body component Mov10 inhibits HIV-1 infectivity. PLoS ONE 5, e9081–e9081 (2009). 232. O apos Connell, M. A. & Keller, W. Purification and properties of double- stranded RNA-specific adenosine deaminase from calf thymus. Proc Natl Acad Sci USA 91, 10596–10600 (1994). 233. Biswas, N. et al. ADAR1 is a novel multi targeted anti-HIV-1 cellular protein. Virology 422, 265–277 (2012). 234. Best, S., Le Tissier, P., Towers, G. & Stoye, J. P. Positional cloning of the mouse retrovirus restriction gene Fv1. Nature 382, 826–829 (1996). 235. Bishop, K. N., Bock, M., Towers, G. & Stoye, J. P. Identification of the regions of Fv1 necessary for murine leukemia virus restriction. J. Virol. 75, 5182–5188 (2001). 236. Kozak, C. A. C. & Chakraborti, A. A. Single Amino Acid Changes in the Murine Leukemia Virus Capsid Protein Gene Define the Target ofFv1Resistance. Virology 225, 6–6 (1996). 237. Dodding, M. P., Bock, M., Yap, M. W. & Stoye, J. P. Capsid processing requirements for abrogation of Fv1 and Ref1 restriction. J. Virol. 79, 10571–10577 (2005). 238. Hilditch, L. et al. Ordered assembly of murine leukemia virus capsid protein on lipid nanotubes directs specific binding by the restriction factor, Fv1. Proceedings of the National Academy of Sciences 108, 5771– 5776 (2011). 239. Han, K., Lou, D. I. & Sawyer, S. L. Identification of a Genomic Reservoir for New TRIM Genes in Primate Genomes. PLoS Genet 7, e1002388 (2011). 240. Ozato, K., Shin, D.-M., Chang, T.-H. & Morse, H. C. TRIM family proteins and their emerging roles in innate immunity. Nat Rev Immunol 8, 849– 860 (2008). 241. Freemont, P. S. P. The RING finger. A novel protein sequence motif related to the zinc finger. Annals of the New York Academy of Sciences 684, 174–192 (1993). 242. Borden, K. L. et al. The solution structure of the RING finger domain from the acute promyelocytic leukaemia proto-oncoprotein PML. EMBO J 14, 1532–1541 (1995). 243. Deshaies, R. J. & Joazeiro, C. A. P. RING Domain E3 Ubiquitin Ligases. Annu Rev Biochem 78, 399–434 (2009). 244. Sabile, A. et al. Regulation of p27 degradation and S-phase progression by Ro52 RING finger protein. Mol Cell Biol 26, 5994–6004 (2006). 245. Gack, M. U. et al. TRIM25 RING-finger E3 ubiquitin ligase is essential for RIG-I-mediated antiviral activity. Nature 446, 916–920 (2007). 246. Diaz-Griffero, F. et al. Rapid turnover and polyubiquitylation of the retroviral restriction factor TRIM5. Virology 349, 300–315 (2006). 247. Lienlaf, M. et al. Contribution of E3-ubiquitin ligase activity to HIV-1 restriction by TRIM5alpha(rh): structure of the RING domain of TRIM5alpha. J. Virol. 85, 8725–8737 (2011).

180 248. Massiah, M. A. et al. Solution Structure of the MID1 B-box2 CHC(D/C)C2H2 Zinc-binding Domain: Insights into an Evolutionarily Conserved RING Fold. J Mol Biol 369, 1–10 (2007). 249. Massiah, M. A., Simmons, B. N., Short, K. M. & Cox, T. C. Solution structure of the RBCC/TRIM B-box1 domain of human MID1: B-box with a RING. J Mol Biol 358, 532–545 (2006). 250. Diaz-Griffero, F. et al. A B-box 2 surface patch important for TRIM5alpha self-association, capsid binding avidity, and retrovirus restriction. J. Virol. 83, 10737–10751 (2009). 251. Mrosek, M. et al. Structural Analysis of B-Box 2 from MuRF1: Identification of a Novel Self-Association Pattern in a RING-like Fold †‡. Biochemistry 47, 10722–10730 (2008). 252. Li, X. & Sodroski, J. The TRIM5 B-Box 2 Domain Promotes Cooperative Binding to the Retroviral Capsid by Mediating Higher-Order Self- Association. J. Virol. 82, 11495–11502 (2008). 253. Mische, C. C. et al. Retroviral restriction factor TRIM5alpha is a trimer. J. Virol. 79, 14446–14450 (2005). 254. Javanbakht, H. et al. Characterization of TRIM5alpha trimerization and its contribution to human immunodeficiency virus capsid binding. Virology 353, 234–246 (2006). 255. Peng, H. Z. et al. Reconstitution of the KRAB-KAP-1 repressor complex: A model system for defining the molecular anatomy of RING-B box-coiled- coil domain-mediated protein-protein interactions. J Mol Biol 295, 1139– 1162 (2000). 256. Grignani, F. et al. Formation of PML/RAR alpha high molecular weight nuclear complexes through the PML coiled-coil region is essential for the PML/RAR alpha-mediated retinoic acid response. Oncogene 18, 6313– 6321 (1999). 257. Reymond, A. et al. The tripartite motif family identifies cell compartments. EMBO J 20, 2140–2151 (2001). 258. Xu, L. BTBD1 and BTBD2 colocalize to cytoplasmic bodies with the RBCC/tripartite motif protein, TRIM5δ. Exp Cell Res 288, 84–93 (2003). 259. Short, K. M., Hopwood, B., Yi, Z. & Cox, T. C. MID1 and MID2 homo- and hetero-dimerise to tether the rapamycin-sensitive PP2A regulatory subunit, Alpha 4, to microtubules: implications for the clinical variability of X-linked Opitz GBBB syndrome and other developmental disorders. Bmc Cell Biology 3, (2002). 260. Nisole, S., Stoye, J. P. & Saïb, A. TRIM family proteins: retroviral restriction and antiviral defence. Nat Rev Microbiol 3, 799–808 (2005). 261. Uchil, P. D., Quinlan, B. D., Chan, W.-T., Luna, J. M. & Mothes, W. TRIM E3 ligases interfere with early and late stages of the retroviral life cycle. PLoS Pathog 4, e16 (2008). 262. Martinez, F. O., Gordon, S., Locati, M. & Mantovani, A. Transcriptional profiling of the human monocyte-to-macrophage differentiation and polarization: New molecules and patterns of gene expression. J Immunol 177, 7303–7311 (2006). 263. Rajsbaum, R., Stoye, J. P. & O apos Garra, A. Type I interferon-dependent and -independent expression of tripartite motif proteins in immune cells. Eur. J. Immunol. 38, 619–630 (2008).

181 264. Carthagena, L. et al. Human TRIM gene expression in response to interferons. PLoS ONE 4, e4894 (2009). 265. Pertel, T. et al. TRIM5 is an innate immune sensor for the retrovirus capsid lattice. Nature 472, 361–365 (2011). 266. Uchil, P. D. et al. TRIM protein-mediated regulation of inflammatory and innate immune signaling and its association with antiretroviral activity. J. Virol. 87, 257–272 (2013). 267. Battivelli, E. et al. Modulation of TRIM5 Activity in Human Cells by Alternatively Spliced TRIM5 Isoforms. J. Virol. 85, 7828–7835 (2011). 268. Grütter, C. et al. Structure of the PRYSPRY-domain: implications for autoinflammatory diseases. FEBS Lett. 580, 99–106 (2006). 269. James, L. C., Keeble, A. H., Khan, Z., Rhodes, D. A. & Trowsdale, J. Structural basis for PRYSPRY-mediated tripartite motif (TRIM) protein function. Proc Natl Acad Sci USA 104, 6200–6205 (2007). 270. Perron, M. J., Stremlau, M. & Sodroski, J. Two surface-exposed elements of the B30.2/SPRY domain as potency determinants of N-tropic murine leukemia virus restriction by human TRIM5alpha. J. Virol. 80, 5631–5636 (2006). 271. Stremlau, M., Perron, M., Welikala, S. & Sodroski, J. Species-specific variation in the B30.2(SPRY) domain of TRIM5alpha determines the potency of human immunodeficiency virus restriction. J. Virol. 79, 3139– 3145 (2005). 272. Perez-Caballero, D., Hatziioannou, T., Yang, A., Cowan, S. & Bieniasz, P. D. Human tripartite motif 5alpha domains responsible for retrovirus restriction activity and specificity. J. Virol. 79, 8969–8978 (2005). 273. Sawyer, S. L., Wu, L. I., Emerman, M. & Malik, H. S. Positive selection of primate TRIM5alpha identifies a critical species-specific retroviral restriction domain. Proc Natl Acad Sci USA 102, 2832–2837 (2005). 274. Sebastian, S. & Luban, J. TRIM5alpha selectively binds a restriction- sensitive retroviral capsid. Retrovirology 2, 40 (2005). 275. Stremlau, M. et al. Specific recognition and accelerated uncoating of retroviral capsids by the TRIM5alpha restriction factor. Proc Natl Acad Sci USA 103, 5514–5519 (2006). 276. Munk, C., Brandt, S. M., Lucero, G. & Landau, N. R. A dominant block to HIV-1 replication at reverse transcription in simian cells. Proc Natl Acad Sci USA 99, 13843–13848 (2002). 277. Besnier, C., Takeuchi, Y. & Towers, G. Restriction of lentivirus in monkeys. Proc Natl Acad Sci USA 99, 11920–11925 (2002). 278. Sawyer, S. L., Emerman, M. & Malik, H. S. Discordant evolution of the adjacent antiretroviral genes TRIM22 and TRIM5 in mammals. PLoS Pathog 3, e197 (2007). 279. Liu, H. F. et al. Adaptive evolution of primate TRIM5 alpha, a gene restricting HIV-1 infection. Gene 362, 109–116 (2005). 280. Kuang, Y.-Q. et al. Genotyping of TRIM5 locus in northern pig-tailed macaques (Macaca leonina), a primate species susceptible to Human Immunodeficiency Virus type 1 infection. Retrovirology 6, 58–58 (2008). 281. Virgen, C. A., Kratovac, Z., Bieniasz, P. D. & Hatziioannou, T. Independent genesis of chimeric TRIMS-cyclophilin proteins in two primate species. Proc Natl Acad Sci USA 105, 3563–3568 (2008).

182 282. Wilson, S. J. et al. Independent evolution of an antiviral TRIMCyp in rhesus macaques. Proc Natl Acad Sci USA 105, 3557–3562 (2008). 283. Brennan, G., Kozyrev, Y. & Hu, S.-L. TRIMCyp expression in Old World primates Macaca nemestrina and Macaca fascicularis. Proc Natl Acad Sci USA 105, 3569–3574 (2008). 284. Zhang, F., Hatziioannou, T., Perez-Caballero, D., Derse, D. & Bieniasz, P. D. Antiretroviral potential of human tripartite motif-5 and related proteins. Virology 353, 396–409 (2006). 285. Diaz-Griffero, F. et al. Requirements for capsid-binding and an effector function in TRIMCyp-mediated restriction of HIV-1. Virology 351, 404– 419 (2006). 286. Perez-Caballero, D., Hatziioannou, T., Zhang, F., Cowan, S. & Bieniasz, P. D. Restriction of human immunodeficiency virus type 1 by TRIM-CypA occurs with rapid kinetics and independently of cytoplasmic bodies, ubiquitin, and proteasome activity. J. Virol. 79, 15567–15572 (2005). 287. Towers, G. J. et al. Cyclophilin A modulates the sensitivity of HIV-1 to host restriction factors. Nat Med 9, 1138–1143 (2003). 288. Diaz-Griffero, F. et al. Comparative requirements for the restriction of retrovirus infection by TRIM5 alpha and TRIMCyp. Virology 369, 400– 410 (2007). 289. Luban, J., BOSSOLT, K. L., FRANKE, E. K., KALPANA, G. V. & GOFF, S. P. Human-Immunodeficiency-Virus Type-1 Gag Protein Binds to Cyclophilin-a and Cyclophilin-B. Cell 73, 1067–1078 (1993). 290. Campbell, E. M., Perez, O., Anderson, J. L. & Hope, T. J. Visualization of a proteasome-independent intermediate during restriction of HIV-1 by rhesus TRIM5alpha. 180, 549–561 (2008). 291. Wu, X., Anderson, J. L., Campbell, E. M., Joseph, A. M. & Hope, T. J. Proteasome inhibitors uncouple rhesus TRIM5alpha restriction of HIV-1 reverse transcription and infection. Proc Natl Acad Sci USA 103, 7465– 7470 (2006). 292. Anderson, J. L. et al. Proteasome Inhibition Reveals that a Functional Preintegration Complex Intermediate Can Be Generated during Restriction by Diverse TRIM5 Proteins. J. Virol. 80, 9754–9760 (2006). 293. Diaz-Griffero, F. et al. Modulation of retroviral restriction and proteasome inhibitor-resistant turnover by changes in the TRIM5alpha B-box 2 domain. J. Virol. 81, 10362–10378 (2007). 294. TISSOT, C. & MECHTI, N. Molecular-Cloning of a New Interferon-Induced Factor That Represses Human-Immunodeficiency-Virus Type-1 Long Terminal Repeat Expression. J Biol Chem 270, 14891–14898 (1995). 295. Der, S. D., Zhou, A. M., Williams, B. & Silverman, R. H. Identification of genes differentially regulated by interferon alpha, beta, or gamma using oligonucleotide arrays. Proc Natl Acad Sci USA 95, 15623–15628 (1998). 296. Carthagena, L. et al. Implication of TRIM alpha and TRIMCyp in interferon-induced anti-retroviral restriction activities. Retrovirology 5, 59 (2008). 297. Lukic, Z. & Campbell, E. M. The cell biology of TRIM5α. Curr HIV/AIDS Rep 9, 73–80 (2012). 298. Tareen, S. U., Sawyer, S. L., Malik, H. S. & Emerman, M. An expanded clade of rodent Trim5 genes. Virology 385, 473–483 (2009).

183 299. Browne, E. P. & Littman, D. R. Myd88 is required for an antibody response to retroviral infection. PLoS Pathog 5, e1000298–e1000298 (2009). 300. Kane, M. et al. Innate immune sensing of retroviral infection via Toll-like receptor 7 occurs upon viral entry. Immunity 35, 135–145 (2011). 301. Beignon, A. S. et al. Endocytosis of HIV-1 activates plasmacytoid dendritic cells via toll-like receptor-viral RNA interactions. J Clin Invest 115, 3265– 3275 (2005). 302. Fonteneau, J. F. et al. Human immunodeficiency virus type 1 activates plasmacytoid dendritic cells and concomitantly induces the bystander maturation of myeloid dendritic cells. J. Virol. 78, 5223–5232 (2004). 303. Smed-Sorensen, A. et al. Differential susceptibility to human immunodeficiency virus type 1 infection of myeloid and plasmacytoid dendritic cells. J. Virol. 79, 8861–8869 (2005). 304. Spörri, R. & Sousa, C. R. E. Inflammatory mediators are insufficient for full dendritic cell activation and promote expansion of CD4+ T cell populations lacking helper function. Nat Immunol 6, 163–170 (2005). 305. Henn, M. R. et al. Whole Genome Deep Sequencing of HIV-1 Reveals the Impact of Early Minor Variants Upon Immune Recognition During Acute Infection. PLoS Pathog 8, –e1002529 (2012). 306. Kearney, M. et al. Human Immunodeficiency Virus Type 1 Population Genetics and Adaptation in Newly Infected Individuals. J. Virol. 83, 2715– 2727 (2009). 307. Battivelli, E. et al. Strain-specific differences in the impact of human TRIM5alpha, different TRIM5alpha alleles, and the inhibition of capsid- cyclophilin A interactions on the infectivity of HIV-1. J. Virol. 84, 11010– 11019 (2010). 308. Battivelli, E. et al. Gag cytotoxic T lymphocyte escape mutations can increase sensitivity of HIV-1 to human TRIM5alpha, linking intrinsic and acquired immunity. J. Virol. 85, 11846–11854 (2011). 309. Nisole, S., Lynch, C., Stoye, J. P. & Yap, M. W. A Trim5-cyclophilin A fusion protein found in owl monkey kidney cells can restrict HIV-1. Proc Natl Acad Sci USA 101, 13324–13328 (2004). 310. Berthoux, L., Sebastian, S., Sayah, D. M. & Luban, J. Disruption of human TRIM5alpha antiviral activity by nonhuman primate orthologues. J. Virol. 79, 7883–7888 (2005). 311. Perron, M. J. et al. The human TRIM5alpha restriction factor mediates accelerated uncoating of the N-tropic murine leukemia virus capsid. J. Virol. 81, 2138–2148 (2007). 312. Roa, A. et al. RING domain mutations uncouple TRIM5α restriction of HIV-1 from inhibition of reverse transcription and acceleration of uncoating. J. Virol. 86, 1717–1727 (2012). 313. Li, X., Song, B., Xiang, S.-H. & Sodroski, J. Functional interplay between the B-box 2 and the B30.2(SPRY) domains of TRIM5alpha. Virology 366, 234–244 (2007). 314. Javanbakht, H. et al. The ability of multimerized cyclophilin A to restrict retrovirus infection. Virology 367, 19–29 (2007). 315. Sastri, J. et al. Identification of residues within the L2 region of rhesus TRIM5alpha that are required for retroviral restriction and cytoplasmic

184 body localization. Virology 405, 259–266 (2010). 316. Langelier, C. R. et al. Biochemical characterization of a recombinant TRIM5alpha protein that restricts human immunodeficiency virus type 1 replication. J. Virol. 82, 11682–11694 (2008). 317. Ganser-Pornillos, B. K. et al. Hexagonal assembly of a restricting TRIM5alpha protein. Proceedings of the National Academy of Sciences 108, 534–539 (2011). 318. Song, B. et al. TRIM5alpha association with cytoplasmic bodies is not required for antiretroviral activity. Virology 343, 201–211 (2005). 319. Sastri, J. & Campbell, E. M. Recent insights into the mechanism and consequences of TRIM5α retroviral restriction. AIDS Res. Hum. Retroviruses 27, 231–238 (2011). 320. Schaller, T., Hué, S. & Towers, G. J. An active TRIM5 protein in rabbits indicates a common antiviral ancestor for mammalian TRIM5 proteins. J. Virol. 81, 11713–11721 (2007). 321. Asmal, M. et al. Production of ribosome components in effector CD4+ T cells is accelerated by TCR stimulation and coordinated by ERK-MAPK. Immunity 19, 535–548 (2003). 322. Cullen, B. R., Lomedico, P. T. & Ju, G. Transcriptional interference in avian retroviruses--implications for the promoter insertion model of leukaemogenesis. Nature 307, 241–245 (1984). 323. Boerkoel, C. F. & Kung, H.-J. Transcriptional interaction between retroviral long terminal repeats (LTRs): mechanism of 5'LTR suppression and 3'LTR promoter activation of c-myc in avian B- cell lymphomas. J. Virol. 66, 4814–4823 (1992). 324. Neagu, M. R. et al. Potent inhibition of HIV-1 by TRIM5-cyclophilin fusion proteins engineered from human components. J Clin Invest 119, 3035– 3047 (2009). 325. Nexø, B. A. et al. The etiology of multiple sclerosis: genetic evidence for the involvement of the human endogenous retrovirus HERV-Fc1. PLoS ONE 6, e16652 (2011). 326. Jacks, T. et al. Tumor spectrum analysis in p53-mutant mice. Curr Biol 4, 1–7 (1994). 327. Sfanos, K. S. et al. Identification of Replication Competent Murine Gammaretroviruses in Commonly Used Prostate Cancer Cell Lines. PLoS ONE 6, e20874 (2011). 328. The DNA sequence coding for the 5' untranslated region of herpes simplex virus type 1 ICP22 mRNA mediates a high level of gene expression. 75 ( Pt 7), 1693–1702 (1994). 329. Grez, M., Akgün, E., Hilberg, F. & Ostertag, W. Embryonic stem cell virus, a recombinant murine retrovirus with expression in embryonic stem cells. Proc Natl Acad Sci USA 87, 9202–9206 (1990). 330. Stacey, A., Arbuthnott, C., Kollek, R., Coggins, L. & Ostertag, W. Comparison of myeloproliferative sarcoma virus with Moloney murine sarcoma virus variants by nucleotide sequencing and heteroduplex analysis. J. Virol. 50, 725–732 (1984). 331. Laker, C. et al. Host cis-mediated extinction of a retrovirus permissive for expression in embryonal stem cells during differentiation. J. Virol. 72, 339–348 (1998).

185 332. Burns, J. C., Friedmann, T., Driever, W., Burrascano, M. & Yee, J. K. Vesicular stomatitis virus G glycoprotein pseudotyped retroviral vectors: concentration to very high titer and efficient gene transfer into mammalian and nonmammalian cells. Proc Natl Acad Sci USA 90, 8033– 8037 (1993). 333. Castellani, S., Di Gioia, S., Trotta, T., Maffione, A. B. & Conese, M. Impact of lentiviral vector-mediated transduction on the tightness of a polarized model of airway epithelium and effect of cationic polymer polyethylenimine. J. Biomed. Biotechnol. 2010, 103976 (2010). 334. Mallery, D. L. et al. Antibodies mediate intracellular immunity through tripartite motif-containing 21 (TRIM21). Proceedings of the National Academy of Sciences 107, 19985–19990 (2010). 335. Watkinson, R. E., Tam, J. C. H., Vaysburd, M. J. & James, L. C. Simultaneous Neutralization and Innate Immune Detection of a Replicating Virus by TRIM21. J. Virol. 87, 7309–7313 (2013). 336. Baud, V. et al. Signaling by proinflammatory cytokines: oligomerization of TRAF2 and TRAF6 is sufficient for JNK and IKK activation and target gene induction via an amino-terminal effector domain. Genes Dev 13, 1297–1308 (1999). 337. Campbell, E. M. et al. TRIM5 alpha cytoplasmic bodies are highly dynamic structures. Mol. Biol. Cell 18, 2102–2111 (2007). 338. Kaumanns, P., Hagmann, I. & Dittmar, M. T. Retrovirology 3, 79 (2006). 339. Lorick, K. L. et al. RING fingers mediate ubiquitin-conjugating enzyme (E2)-dependent ubiquitination. Proc Natl Acad Sci USA 96, 11364–11369 (1999). 340. Maegawa, H., Miyamoto, T., Sakuragi, J.-I., Shioda, T. & Nakayama, E. E. Contribution of RING domain to retrovirus restriction by TRIM5alpha depends on combination of host and virus. Virology 399, 212–220 (2010). 341. Li, X. et al. Unique features of TRIM5alpha among closely related human TRIM family members. Virology 360, 419–433 (2007). 342. Gamble, T. R. et al. Crystal structure of human cyclophilin A bound to the amino-terminal domain of HIV-1 capsid. Cell 87, 1285–1294 (1996). 343. Bon Homme, M., Carter, C. & Scarlata, S. The Cysteine Residues of HIV-1 Capsid Regulate Oligomerization and Cyclophilin A-Induced Changes. Biophysical Journal 88, 2078–2088 (2005). 344. Pornillos, O. et al. X-ray structures of the hexameric building block of the HIV capsid. Cell 137, 1282–1292 (2009).

186 ANNEX I

My personal contribution to this paper was for the figure 2b and supplementary figures 1d and 1e.

187 LETTER doi:10.1038/nature09 976 TRIM5 is an innate immune sensor for the retrovirus capsid lattice

Thomas Pertel1, Ste´phane Hausmann1, Damien Morger2, Sara Zu¨ger2, Jessica Guerra1, Josefina Lascano1, Christian Reinhard1, Federico A. Santoni1, Pradeep D. Uchil3, Laurence Chatel4, Aure´lie Bisiaux5, Matthew L. Albert5, Caterina Strambio-De-Castillia1, Walther Mothes3, 1 2 1 Massimo Pizzato , Markus G. Gru¨tter & Jeremy Luban

TRIM5 is a RING domain-E3 ubiquitin ligase that restricts infec- Lipopolysaccharide (LPS), a pathogen-associated molecular pattern tion by human immunodeficiency virus (HIV)-1 and other retro- (PAMP) recognized by the pattern recognition receptor (PRR) TLR4- viruses immediately following virus invasion of the target cell MD-2, activates AP-1 and NF-kB-signalling and this culminates in the cytoplasm1,2. Antiviral potency correlates with TRIM5 avidity for expression of inflammatory genes like those perturbed by TRIM5 the retrovirion capsid lattice3,4 and several reports indicate that knockdown10,11. Monocyte-derived dendritic cells (MDDC), macro- TRIM5 has a role in signal transduction5–7, but the precise mech- phages (MDM) and THP-1 cells were challenged with LPS and induc- anism of restriction is unknown8. Here we demonstrate that TRIM5 tion of the AP-1- and NF-kB-dependent genes CXCL9, CXCL10, promotes innate immune signalling and that this activity is amp- CCL8, IL6, IL8 and PTGS2 (also known as COX2), was found to be lified by retroviral infection and interaction with the capsid lattice. attenuated by TRIM5 knockdown (Fig. 1e and f and Supplementary Acting with the heterodimeric, ubiquitin-conjugating enzyme Fig. 2d and e). These results demonstrate that TRIM5 activates MAPK- UBC13–UEV1A (also known as UBE2N–UBE2V1), TRIM5 cata- and NF-kB-dependent genes and makes a major contribution to LPS lyses the synthesis of unattached K63-linked ubiquitin chains that signalling and gene induction (Supplementary Fig. 1f). activate the TAK1 (also known as MAP3K7) kinase complex and Given the contribution of TRIM5 to the production of inflammatory stimulate AP-1 and NFkB signalling. Interaction with the HIV-1 mediators by LPS, the effect of TRIM5 on the previously reported capsid lattice greatly enhances the UBC13–UEV1A-dependent E3 anti-HIV-1 activity of LPS12 was examined. Transduction of MDDC, activity of TRIM5 and challenge with retroviruses induces the tran- MDM or THP-1 macrophages by vesicular stomatitis virus (VSV) scription of AP-1 and NF-kB-dependent factors with a magnitude G-pseudotyped HIV-1 was blocked by LPS, by other PAMPs, and by that tracks with TRIM5 avidity for the invading capsid. Finally, type 1 IFN (Supplementary Fig. 3a–c). TRIM5 mRNA increased TAK1 and UBC13–UEV1A contribute to capsid-specific restriction tenfold in response to these factors (Supplementary Fig. 3d, e), but this by TRIM5. Thus, the retroviral restriction factor TRIM5 has two increase was not sufficient for the anti-HIV-1 state (Supplemen- additional activities that are linked to restriction: it constitutively tary Fig. 3f, g). Nonetheless, TRIM5 knockdown rescued HIV-1 from promotes innate immune signalling and it acts as a pattern recog- LPS, although not from type 1 IFN, and the magnitude rescue corre- nition receptor specific for the retrovirus capsid lattice. lated with the efficiency of TRIM5 knockdown (Fig. 1g and Sup- To determine if TRIM5 contributes to signal transduction, the effect plementary Fig. 4a and b). These phenotypes were indistinguishable of ectopic human TRIM5a expression on transcriptional reporters in from those observed with knockdown of IRF3, a critical transcrip- HEK-293 cells was examined. TRIM5 stimulated either of two luciferase tion factor that acts proximal to IFNB1 (ref. 10; Fig. 1h and Sup- reporters for AP-1 with a magnitude comparable to that of MAVS or plementary Fig. 4c). In contrast, knockdown of STAT2, a factor the AP-1 transcription factor c-Jun (Fig. 1a and Supplementary Fig. 1a). that acts downstream of the type I IFN receptor, blocked the anti- TRIM5 also stimulated NF-kB (Fig. 1b) but minimally activated HIV-1 activity of either LPS or type 1 IFN (Fig. 1i and Supplemen- IFNB1-, or IRF3-dependent, luciferase reporters (Fig. 1c and Sup- tary Fig. 4d). plementary Fig. 1b and c). The TRIM5–cyclophilin A fusion protein Rescue from LPS seems to be independent of capsid-recognition by from owl monkey1 activated AP-1 and NF-kB to similar levels as TRIM5 in that TRIM5 knockdown rescued a molecular clone of simian human TRIM5a (Supplementary Fig. 1d, e). Although TRIM5 was immunodeficiency virus (SIVMAC), a retrovirus that differs greatly not sufficient to activate IFNB1, induction of IFNB1 by IRF3 was greatly from HIV-1 in terms of its sensitivity to TRIM5-mediated restric- enhanced by TRIM5 (Fig. 1c), consistent with the fact that IFNB1 tion1,2, as well as two non-retroviruses, the rhabdovirus vesicular transcription requires NF-kB and AP-1, as well as IRF3 (Sup- stomatitis virus and the paramyxovirus Newcastle disease virus plementary Fig. 1f)9. (Fig. 1j and Supplementary Fig. 4e–i). Although TRIM5 is not suf- To determine if endogenous TRIM5 regulates AP-1 and NF-kB sig- ficient to activate IFNB1 (Fig. 1c), it promotes the first wave of innate nalling pathways, the effect of TRIM5 knockdown was assessed in mye- immune signalling upstream of IFNB1 and thereby contributes to the loid cells. THP-1 cells were transduced with lentiviral vectors engineered antiviral state established by LPS (Supplementary Fig. 1f). to confer puromycin-resistance and to express RNA polymerase II (Pol To understand how TRIM5 activates AP-1 and NF-kB, 20 candidate II)-driven, microRNA-based short hairpin RNAs (shRNAs) targeting proteins, selected on the basis of signalling activity above the MAPK/ either TRIM5 or control RNAs (Supplementary Fig. 2a–c). Pools of NF-kB bifurcation in the LPS signalling pathway, were tested for the puromycin-resistant cells were generated with each knockdown vector ability to immunoprecipitate with TRIM5. Strong signal was observed and global expression profiles were assessed. The effect of TRIM5 with TAK1, TAB2, and TAB3 (Fig. 2a and Supplementary Fig. 5a), all knockdown was extraordinarily specific in that, of 25,000 genes probed, components of the TAK1 kinase complex that phosphorylates proximal only 33 were significantly decreased (Fig. 1d). The majority of these were MAPK and NF-kB kinases in response to LPS11. Like TRIM5, TAK1 NF-kB- and AP-1-responsive inflammatory mediators, 70% being potently activated AP-1 and modestly activated NF-kB (Fig. 2b). 5Z-7- inflammatory chemokines and cytokines (Supplementary Table 1). oxozeaenol, a TAK1-inhibitor, blocked AP-1 induction by TRIM5 or

1Department of Microbiology and Molecular Medicine, University of Geneva, Geneva CH-1211, Switzerland. 2Department of Biochemistry, University of Zurich, Zurich CH-8057, Switzerland. 3Section of Microbial Pathogenesis, Yale University School of Medicine, New Haven, Connecticut 06536, USA. 4Novimmune SA, Geneva CH-1228, Switzerland. 5Institut Pasteur, Inserm U818, Paris 75724, France. 2 1 A P R I L 2 0 1 1 | V O L 4 7 2 | N A T U R E | 3 6 1 ©2011 Macmillan Publishers Limited. All rights reserved

188 RESEARCH LETTER

a102 AP-1 luc b NF-⎢B luc c 1,500 d THP-1 IFNB1 luc 2 10 1,000

101 1 KD 10 500 Fold change Fold change Fold change Fold

TRIM5 100 100 0

Control KD

Control KD TRIM5 KD Control KD TRIM5 KD ) 1 e 20 8,000 150 4,000 1.5 f – 12,000 3,000 400 6,000

15 3, 00 6,000 (pg 9,000 0 2,0 00 300 4,000 units)

100 10 1.0 6,000

expression expression 00 200 4,000 2, 1,0 00 2,000 5 0 3,000 (relative 00 100 mRNA 50 0 2,000 0 0 0 TRIM5 CXCL10 CCL81, IL6 OAZ1 Concentration CXCL10 CXCL9 IL6 IL8 0 g h 12 i 25 THP-1 j

THP-1 THP-1 12 STAT2 KD 10 TRIM5 KD 10 IRF3 KD 20 Vehicle 10 KD: 8 8 LPS rescue rescue rescue 8 15 6 6 rescue fold 6 fold fold

4 10 TRIM5 1: 1: 1: 1 4 - - - 4 - 2 5 fold

HIV 2 HIV HIV 2 0 THP HIV-1 SIV VSV NDV

Figure 1 | TRIM5 promotes innate immune signalling. a–c, HEK-293 the culture supernatant, 24 h after LPS treatment (mean 6 s.d., n 5 3). cells transfected with the indicated pcDNA-based expression plasmids RNA and protein data are representative of at least three separate donors. and g–j, THP-1 macrophages transduced with miR30-based lentivirus KD luciferase reporters for AP-1 (a), NF-kB (b) or IFNB1 (c). Bars show mean vectors targeting either TRIM5 (g and j), IRF3 (h), or STAT2 (i), were luciferase activity 6 s.d. (n 5 6). d, Global expression profile comparing treated for 24 h with the indicated compounds and challenged with VSV- TRIM5 knockdown (KD) to control KD THP-1 macrophages. Triangles G pseudotyped HIV-1 luciferase reporter virus (g–i) or with the indicate inflammatory genes significantly downregulated in TRIM5 KD. e, indicated green fluorescent protein (GFP) reporter viruses (j). Data are qRT–PCR for the indicated mRNAs collected from MDDCs 2 to 8 h after expressed as fold-change compared to control KD cells, with s.e.m (n 5 4). LPS treatment, depending on the peak values for that gene. Shown are All data are representative of at least three independent experiments. the means 6 s.e.m TAK(n 5 3)-1 re witholativeu tot e unffecttre aonted AP ce-l1ls. indu f, Conctioncentr byation the ofdownstream the indicated effec tor and expressing hygromycin-resistance. The pools of puromycin/ cp-rJotune ins(Fi ing. 2b). TAK1 knockdown blocked AP-1 activation by TRIM5 hygromycin double-resistant THP-1 cells were then challenged with (Fig. 2c), but not by c-Jun (Fig. 2c and Supplementary Fig. 5c). TRIM5 HIV-1. TAK1 knockdown rescued HIV-1 transduction and nascent knockdown blocked LPS-induced TAK1 autophosphorylation on HIV-1 cDNA synthesis (Fig. 2g, h). This effect was specific to the cells threonine 187 (Fig. 2d), a post-translational modification required for with TRIM5Cyp-mediated restriction activity because TAK1 knock- TAK1 activation11. Like TRIM5 knockdown, TAK1 knockdown rescued down had no effect on HIV-1 transduction in the non-restrictive, HIV-1 from the LPS-induced antiviral state (Fig. 2e and Supplementary H436Q control cells (Fig. 2g). Fig. 5d), and either TAB2 or TAB3 acted synergistically with TRIM5 to The contribution of TAK1 to restriction of N-tropic murine leuke- activate AP-1 (Fig. 2f). These results indicate that TRIM5 and the TAK1 mia virus (MLV) by human TRIM5a was examined using miR30- kinase complex cooperate to promote signal transduction, and given based knockdown vectors in THP-1, HeLa and HT1080 cells. that TAK1 phosphorylates both IkB kinases (IKKs) and mitogen- Inhibition of both N-tropic and B-tropic MLV infection by the activated protein kinase kinases (MKKs)11, explains how TRIM5 TAK1 knockdown was observed, perhaps because, unlike HIV-1, activates both MAPK and NF-kB signalling pathways. infection with MLV is cell-cycle dependent14, and these viruses were The well-characterized restriction of HIV-1 by owl monkey sensitive to growth inhibitory effects of the knockdown. This precluded TRIMCyp1,13 (a TRIM5–CypA fusion protein) was exploited to deter- assessment of capsid-specific effects on reporter gene transduction, mine if TAK1 contributes to TRIM5-mediated, capsid-specific restric- although nascent viral cDNA synthesized after infection of THP-1 cells tion. Pools of THP-1 cells were selected for puromycin-resistance after was rescued by the TAK1 knockdown in an N-tropic MLV-specific transduction with a bicistronic lentiviral vector encoding owl monkey manner (Supplementary Fig. 5g). Similar non-specific effects on MLV TRIM5Cyp13. As shown previously, these cells were resistant to infec- were observed after transfection of double stranded RNA (dsRNA) tion with wild-type HIV-1, but not to the HIV-1 G89V capsid mutant, oligonucleotides targeting TAK1. Like HIV-1, equine infectious and the infectivity of wild-type HIV-1 was rescued by cyclosporine13 anaemia virus (EIAV) is a lentivirus that infects non-dividing cells, (Supplementary Fig. 5e). Control cells transduced with a vector bear- but it differs from HIV-1 in that it is relatively sensitive to human ing TRIM5Cyp(H436Q), a mutant that does not bind HIV-1 capsid TRIM5a-mediated restriction15. Transfection of dsRNAs targeting and does not restrict HIV-1 (ref. 13), were infected with efficiency TAK1 rescued EIAV transduction almost to the same level as the equal to that of cells transduced with the empty vector. THP-1 cells TRIM5 knockdown (Fig. 2i). These results indicate that TAK1 con- transduced with either wild-type or H436Q mutant TRIM5Cyp were tributes to capsid-specific restriction mediated by TRIM5. then subjected to a second round of selection after transduction with AP-1 induction by TRIM5 was impaired by mutants of the RING E3 miR30-based knockdown vectors targeting TAK1 or luciferase control ubiquitin (Ub)-ligase domain (Fig. 3a). This raised the1 que89s tion which

a b AP-1 luc c a 250 AP-1 luc b150 c d 〈 AP -1 luc 6

IK 2 Vehicle AP-1 luc 25 AP-1 luc THP1

IKK 〈 10 20

Ctrl KD

MEK Ctrl KD 200 UBC13 KD 

TAK1 inhibitor TAK1 KD IKK K  K1 TAK1 TAB1 TAB2 TAB3 TAB3  20 15 100 UBC13 KD

150 rescue IP:Flag 15 4 IB:Myc IgH 101 10

100 fold 10

Fold change Fold change Fold 50 1: 5 change Fold IB:Myc change Fold - 50 change Fold 5 2 100 HIV pcDNA TRIM5 0 0 d Control KD TRIM5 KD pcDNA TRIM5 Vehicle LPS min 0 15 30 45 0 15 30 45 e 8 f 150 pTAK1 THP-1 AP-1 luc

TAK1 KD TAK1 6 g IB: Ub TRIM5Cyp

rescue 100

4 e 60 f 250 fold

2 AP-1 luc g 10 TRIM5Cyp AP-1 luc 1: 50

- 200 Fold change Fold WT 2 170

1 HIV 40 10 150 72 cells

Vehicle LPS

+ 100 55 100

20 change Fold

GFP 50 Fold change Fold

34 of

–1 0 10 h TAK1 KD i 2 50 Ctrl KD 0 10 TRIM5Cyp UBC13 KD TAK1 KD H436Q 5 Ctrl KD 40 5 10 11 TRIM5 KD

1 10

cells Percentage

10 30 UEV1A + 1: 1: P Percentage Percentage - per

1 cells 1

Ctrl KD - 20 h TRIM5Cyp i GF

0 j

10 〈

THP

TAK1 KD of – + – +

THP 10

Ubc13 KD Copies RING –1 HT1080: 10 0 ZsGreen TRIM5 C15/18A ⊗ TRAF6 TRAF2 4 E3: – – TRIM5Cyp TRAF6 100 101 102 10 0.25 0.50 0.75 1.00 HIV-1: RT units TRIM5Cyp: WT H436Q EIAV dilution 170 E2: – + + + 170 pTAK1 72 Figure 2 | The TAK1 kinase complex interacts biochemically and 72 TAK1 functionally with TRIM5. a, HEK-293T cells were co-transfected with 55 55 Myc- tagged human TRIM5a and the indicated Flag-tagged constructs. 34 Shown are immunoblots (IB) with anti-Myc antibody after 34 immunoprecipitation with anti-Flag (upper panel), or of total cell lysate 11 Ub 11 (bottom panel). b, c, and f, HEK- 293 cells were transfected with the IB: K63 IB: TRIM5 IB: Ub indicated pcDNA-based expression plasmids and an AP-1 luciferase reporter and show the effect of TAK1 inhibitor 5Z-7-oxozeaenol (b) or Figure 3 | TRIM5 acts with UBC13–UEV1A to synth esize free K63- TAK1 KD (c). TAK1 KD and control KD THP-1 macrophages were treated linked Ub chains that activate TAK1. a–c, e and f, HEK-293 cells were with LPS for the indicated times and immunoblotted with anti-TAK1 transfected with an AP-1 luciferase reporter and the indicated pcDNA- antibody (lower panel) or anti-phospho-TAK1 antibody (upper panel) based expression plasmids. Bars show mean 6 s.d. (n 5 6). In c, HEK-293 (d), or, cells were treated 24 h with LPS or vehicle and challenged with an cells had stable UBC13 KD or control KD. d, UBC13 KD or control KD THP- HIV-1 luciferase reporter virus (e). The results in (e) are reported as 1 macrophages were treated for 24 h with LPS or vehicle and challenged fold rescue due to TAK1 KD, with respect to control KD. g and h, THP-1 with an HIV-1 luciferase reporter virus. Shown is the fold rescue due to cells were transduced with lentiviral vectors encoding owl monkey UBC13 KD, with respect to the control KD. g-j, Products of in vitro TRIM5Cyp, either wild-type (WT) or the H436Q mutant. Pools of each reactions with ATP, Ub, UBE1, UBC13– UEV1A, and the indicated E3 Ub were then transduced with lentiviral KD vectors targeting either TAK1, ligases were revealed by immunoblot for total Ub (g, i, and j), K63-linked UBC13 or control, and challenged with an HIV-1-GFP reporter vector. Ub chains (left panel of h), or TRIM5 (right panel of h). E3 ubiquitin ligases Infectivity was monitored included purified owl monkey TRIM5Cyp (g, h, and j), or the indicated by FACS (g) or by PCR for synthesis of full-length viral cDNA (h). i, Flag-tagged proteins immunoprecipitated from HEK-293T cells (i). j, In HT1080 cells were transfected with dsRNA oligonucleotides targeting vitro Ub reactions like those in (i) were incubated with purified TAK1 ofTR I Mthe5, TmanyAK1 or E2 UE V1A Ub -andcon jcuhgallateinge denz wiythmes EIA Vmi-GgFPht re bepo rreterle vveantcto r .for kinase complex. Products were probed in immunoblot with the indicated TRIM5-mediated effects on signal transduction. Among candidate antibodies. E2s, UBC13 synergized with TRIM5 to activate AP-1 (Fig. 3b). Fig. 7). No procedure has been reported to date for the production Interestingly, the TAK1 kinase complex is activated by the heterodi- of purified, full-length, recombinant TRIM5 protein17. Purified meric E2 UBC13–UEV1A11. Knockdown of UBC13 or UEV1A TRIM5Cyp was incubated with purified Ub, E1 and the E2 Ub- severely blocked AP-1 activation by TRIM5 (Fig. 3c and Supplemen- conjugases UBC13 and UEV1A, and reaction products were separated tary Fig. 6a–c), rescued HIV-1 from the LPS-induced antiviral state in by SDS–polyacrylamide gel electrophoresis (SDS–PAGE). With THP-1 macrophages (Fig. 3d and Supplementary Fig. 6d), and rescued increasing TRIM5Cyp concentration, monomeric Ub was progres- HIV-1 and EIAV from TRIM5-mediated restriction (Fig. 2g–i). sively depleted and the yield of Ub chains increased (Fig. 3g and The UBC13–UEV1A E2 heterodimer is notable in that it generates Supplementary Fig. 8a). Synthesis of Ub chains was ATP-dependent K63-linked Ub chains that are unlinked to substrates; these free Ub and required both UBC13 and UEV1A. chains multimerize and activate the TAK1 kinase complex via the Ub The Ub chains generated by TRIM5Cyp were detected with anti- binding components, TAB2 and TAB3 (ref. 11). Ub in which all lysines body specific for K63-linked Ub and immunoblot showed TRIM5Cyp except K63 are mutated to arginine (Ub K63) activated AP-1 and NF- to be a monomer with no detectable incorporation into the Ub chains kB (Fig. 3e and Supplementary Fig. 6e, f), and enhanced the ability of (Fig. 3h). To obtain an independent assessment of their identity, reac- TRIM5 to activate AP-1 (Fig. 3f). K48-only Ub did not have these tion products were isolated by PAGE and analysed by matrix-assisted activities (Fig. 3e, f), nor did wild-type Ub, perhaps because of the laser desorption/ionization and tandem mass spectrometry (Sup- dominance of competing Ub metabolic pathways and the tight regu- plementary Figs 8a and 9a–c). These methods identified peptides cor- lation of K63 chains within cells16. These experiments indicate that the responding to K63-linked Ub and failed to detect conjugates with heterodimeric E2 UBC13–UEV1A and the K63-linked Ub chains that other Ub lysines or peptides corresponding to TRIM5Cyp, confirming it produces have a role in TRIM5-mediated signalling. that reaction products were free, unattached K63 Ub chains. Because TRIM5 interacted biochemically and functionally with Additionally, synthesis of Ub chains was undetectable with a Ub TAK1, TAB2, TAB3, UBC13, UEV1A and K63-Ub, the ability of mutant in which K63 was mutated to arginine (Supplementary Fig. TRIM5 to synthesize K63-linked Ub chains was assessed. A purifica- 8b). Conversely, Ub was efficiently incorporated into chains when all tion protocol was established that yielded 0.5 mg of soluble, full-length, lysines except K63 were mutated to arginine (Supplementary Fig. 8b), owl monkey TRIM5Cyp from 1 l of Sf9 cell culture (Supplementary indicating that K63 was necessary and sufficient to fo190rm the Ub chains.

THP-1 Human TRIM5a, produced by transfection of 293T cells and a THP-1 b 160 50 140 4 CXCL10 enriched by immunoprecipitation, catalysed the synthesis of free 140 VSV G only 120 40 3

K63 Ub chains like those of TRIM5Cyp, in a RING domain-dependent change) change) 120 manner (Fig. 3i and Supplementary Fig. 8c, d). It had at least as much Unrestricted capsid 100

(fold (fold 100 30 2 activity as TRAF6 (Fig. 3i), an E3 Ub ligase previously reported to Restricted capsid 80 80

mRNA 20 mRNA synthesize unattached K63 chains that activate TAK1 (ref. 11). 60 1 PTGS2 IL6 CCL8 KD: Ctrl TRIM5 TRAF2, a close paralogue of TRAF6 that does not interact with MDDC MDDC

100 9 6 ) 4 50 c d 1 300 400 UBC13 (ref. 18), lacked activity (Fig. 3i). – ml 5 40 Free K63-linked Ub chains generated by TRAF6 result in TAK1 7 350

10 (pg 3 4 change) 30 200 autophosphorylation on threonine 187 (ref. 11), a modification 10 5 300 3 (fold required for TAK1 activation. To test the effect of K63-linked Ub chains 2 20 3 2 250 100 1 10

generated by TRIM5 on TAK1 activation, the essential components of a mRNA

1 1 1 Concentration 200 1 TAK1 kinase complex, TAK1, TAB1 and TAB2 (ref. 11), were purified CXCL9 CXCL10 IFIT1 IFIT2 IL8 CCL5 CXCL9 CXCL10 OMK OMK and combined (Supplementary Fig. 10). This complex was then incu- 25 100 70 120 100 8 e 3.5 f h bated with Ub, UBC13–UEV1A, and either TRAF6 or purified owl 3.0 20 60 100 levels 6 50 80 monkey TRIM5Cyp. TAK1 phosphorylation was observed in response change) 2.5 15 10 40 60 10 4

10 protein (fold 2.0 to the K63-linked Ub chains synthesized by either TRAF6 or by 30 40 2 TRIM5Cyp (Fig. 3j). Kinase activity required the TAK1-associated 1.5 5 20 20

mRNA TRIM5 dimers 1.0 0 1 Relative 10 0 1 0 TAB1, the Ub receptor TAB2, and UBC13–UEV1A (Fig. 3j). These PTGS2 IFIT1 IFIT2 IL8 CCL2 CCL4 CXCL10 g experiments show that, like TRAF6 (ref. 11), TRIM5 synthesizes free inhib: – – – – – + + + + + – – – – – + + + + + K63-linked Ub chains that activate TAK1 autophosphorylation. min: 0 3 6 9 12 0 3 6 9 12 0 3 6 9 12 0 3 6 9 12 If TRIM5 were a PRR specific for the retroviral capsid lattice, infec- UBC13–UEV1A tion with retroviruses would activate signalling, the magnitude of which would correlate with TRIM5 avidity for the capsid of the chal- Ub K63-Ub lenge virus. To determine if this is the case, myeloid cells were challenged with pairs of retroviruses that differ with respect to TRIM5 TAB2 TAB2 3,4 TAK1 TAK1 avidity for the capsid and the subsequent induction of NF-kB- and T5 PO PO MAPK-dependent genes was assessed. VSV G-pseudotyped N-tropic 4 4 and B-tropic MLV vectors, normalized for exogenous reverse tran- CA scriptase activity and for titre on non-restrictive MDTF cells19, were No capsid Plus capsid used to challenge THP-1 macrophages. The multiplicity of infection of Figure 4 | Retrovirus capsid sensing by TRIM5. THP-1 cells (a and the non-restricted B-tropic MLV on cycling THP-1 cells was 0.1. b), MDDCs (c and d), or owl monkey kidney cells (OMK; e and f), mRNA was harvested from the THP-1 cells and processed by reverse were challenged with matched pairs of VSV G-pseudotyped particles transcription and quantitative PCR (qRT–PCR). Greater induction of bearing PTGS2, CXCL10, CCL8 and IL6 mRNA was observed after challenge retrovirion capsids that are restricted by the TRIM5 orthologue with N-MLV than with B-MLV (Fig. 4a, b), in correlation with the endogenous to that cell type (black bars), or unrestricted (white bars), or VSV G-derived particles that are devoid of capsid (grey bars). Restricted higher avidity of human TRIM5a for the capsid of N-tropic MLV than 3 capsids were from N-tropic MLV (a–d) or HIV-1 (e and f). Unrestricted for the capsid of B-tropic MLV . TRIM5 knockdown suppressed the capsids were B-tropic MLV (a and b), N/B-tropic MLV (c and d) or higher inflammatory gene induction by N-MLV, indicating its SIVMAC239 (e and f). Particles bore viral genomes in a, b, e and f, but not in c dependence upon endogenous TRIM5 (Fig. 4b). Similar differential and d. mRNA was harvested for qRT–PCR (a–c and e) and reported as induction of PTGS2, CXCL10, CCL8 and IL6 mRNAs by N-tropic and fold change versus media control. Protein in the supernatant was B-tropic MLV was observed after challenge of MDDCs or MDMs quantified (d and f). Bars show means 6 s.d. (n 5 3), and are (Supplementary Fig. 11). representative of at least three independent experiments. Retroviral cDNA activates innate immune signalling under some g, Immunoblots with the indicated antibodies of products from in vitro conditions20. Restriction by human TRIM5a results in N-MLV cDNA time- course with ATP, Ub, UBE1 (E1), UBC13–UEV1A (E2) and purified levels that are an order of magnitude lower than for B-MLV cDNA21 so owl monkey TRIM5Cyp, with or without assembled HIV-1 capsid- the experiments described above might underestimate the effect of A14C/E45C, and with or without competitive inhibitor MeIle4CsA. (h) Schematic showing entry of an HIV-1 virion core27 (courtesy of Pornillos N-MLV capsid on TRIM5-mediated signalling. Therefore, MDDCs and Yeager) into the target cell cytoplasm where it induces dimeric were challenged with matched pairs of virus-like particles (VLPs) TRIM5 to form a hexameric lattice25 with increased E3 Ub ligase activity. devoid of the viral genome that serves as the reverse transcription With UBC13–UEV1A, TRIM5 synthesizes free K63 Ub chains that are template. VLPs bearing N-MLV capsid activated CXCL9, CXCL10, resecogretnionize dof by IL TA8, BC2,C whL2,ich C CmuL4lti amnerid CzesXC andL10 a cptivroatetesins the, w aTAs hKi1g hkiern aseafte r IFIT1 and IFIT2 mRNAs from 5- to 55-fold over the levels in untreated cochmalpllenex.ge with the restricted virus (Fig. 4e, f). MDDCs (Fig. 4c). Inflammatory gene induction was not detected with TRIM5 senses retrovirus capsids in the target cell cytoplasm VLPs bearing the unrestricted NB-MLV capsid22 (Fig. 4c). As with the (Fig. 4a–f) and activates MAPK- and NF-kB-dependent transcription mRNA, soluble IL8, CCL5, CXCL9 and CXCL10 protein was differ- via the synthesis of TAK1-activating, K63-linked Ub chains (Fig. 3g–j). If entially induced by N-MLV (Fig. 4d). these observations were linked functionally, interaction with capsid To determine if differential gene induction after retrovirus challenge would be expected to stimulate the synthesis of Ub chains by was peculiar to N-tropic MLV, a similar experiment was performed TRIM5Cyp. Soluble, recombinant HIV-1 capsid or capsid hexamers with OMK, a kidney cell line from the owl monkey, Aotus trivirgatus. generated by the oxidation of recombinant capsid bearing strategically- The TRIM5 orthologue in this species restricts HIV-1 but not SIV1. placed cysteine substitutions (A14C/E45C/W184A/M185A)23 had no VSV G-pseudotyped HIV-1 and SIV vectors, normalized for exogenous effect on the synthesis of K63-linked Ub chains (data not shown). reverse transcriptase activity and for titre on HeLa cells, were used to Current models of the HIV-1 capsid lattice are based on cylinders challenge OMK cells. The multiplicity of infection of the unrestricted generated under high salt with either capsid or capsid-nucleocapsid SIV on OMK cells was 0.3. Among the MAPK- and NF-kB-dependent fusion protein17,24; both preparations were generated but the high salt gene products that were detectable in this species using human probes, necessary to maintain capsid cylinders blocked E3 Ub ligase activity. transcriptional activation of PTGS2, IFIT1 and IFIT2 mRNAs, and Capsid cylinders were then assembled with A14C/E19145C- substituted

3 6 4 | N A T U R E | V O L 4 7 2 | 2 1 A P R I L 2 0 1 1 ©2011 Macmillan Publishers Limited. All rights reserved LETTER RESEARCH capsid protein in 1 M NaCl and the cysteines were oxidized. These 8. Luban, J. & Cyclophilin, A. TRIM5, and resistance to human immunodeficiency virus type 1 infection. J. Virol. 81, 1054–1061 (2007). oxidized cylinders were stable in the absence of salt (Supplementary 9. Panne, D., Maniatis, T. & Harrison, S. C. An atomic model of the interferon- Fig. 12a) and greatly stimulated the production of K63-linked Ub b enhanceosome. Cell 129, 1111–1123 (2007). chains by TR IM5Cyp (Fig. 4g). No Ub-linked products were detected 10. Ishii, K. J., Koyama, S., Nakagawa, A., Coban, C. & Akira, S. Host innate immune with anti-capsid (p24) or anti-TRIM5 antibodies, indicating that the receptors and beyond: making sense of microbial infections. Cell Host Microbe 3, 352–363 (2008). reaction products were unattached Ub chains (Fig. 4g). 11. Xia, Z. P. et al. Direct activation of protein kinases by unanchored Finally, two factors that disrupt the HIV-1 capsid-TRIM5Cyp inter- polyubiquitin chains. Nature 461, 114–119 (2009). action and block restriction activity—a non-immunosuppressive cyclos- 12. Kornbluth, R. S., Oh, P. S., Munis, J. R., Cleveland, P. H. & Richman, D. D. 1 13 Interferons and bacterial lipopolysaccharide protect macrophages from porine analogue or the TRIM5Cyp(H436Q) mutant protein —each productive infection by human immunodeficiency virus in vitro. J. Exp. Med. eliminated the enhancement of E3 Ub ligase activity by the A14C/E45C 169, 1137–1151 (1989). capsid cylinders, without effect on the baseline activity in the absence of 13. Neagu, M. R. et al. Potent inhibition of HIV-1 by TRIM5-cyclophilin fusion capsid (Fig. 4g and Supplementary Figs 7d and 12b–d). proteins engineered from human components. J. Clin. Invest. 119, 3035– 14. 3047 (2009). Roe, T., Reynolds, T. C., Yu, G. & Brown, P. O. Integration of murine The experiments presented here demonstrate that TRIM5 is a mul- leukemia virus DNA depends on mitosis. EMBO J. 12, 2099–2108 (1993). tifunctional component of the innate immune system. In addition to 15. Berthoux, L., Sebastian, S., Sokolskaja, E. & Luban, J. Cyclophilin A is required for functioning as a retroviral capsid-specific restriction factor, TRIM5 TRIM5a-mediated resistance to HIV-1 in Old World monkey cells. Proc. Natl Acad. Sci. USA 102, 14849–14853 (2005). synthesizes K63 Ub chains that activate TAK1 and inflammatory tran- 16. Zeng, W. et al. Reconstitution of the RIG-I pathway reveals a signaling role of scription, most probably via multimerization of the TAK1-associated unanchored polyubiquitin chains in innate immunity. Cell 141, 315–330 Ub-binding protein TAB2 (ref. 11; Fig. 4h). This activity was greatly 17. (2010). Langelier, C. R. et al. Biochemical characterization of a recombinant increased by the hexameric capsid lattice, a molecular signature of TRIM5a protein that restricts human immunodeficiency virus type 1 replication. J. Virol. 82, 11682–11694 (2008). HIV-1 and other retroviruses. TRIM5, then, satisfies criteria for a bona 18. Yin, Q., Lamothe, B., Darnay, B. G. & Wu, H. Structural basis for the lack of E2 fide PRR10. Interestingly, TRIM5 spontaneously forms an hexagonal interaction in the RING domain of TRAF2. Biochemistry 48, 10558–10567 lattice that is complementary to the capsid lattice25, but the efficiency of 19. (2009). Sokolskaja, E., Berthoux, L. & Luban, J. Cyclophilin A and TRIM5a independently regulate human immunodeficiency virus type 1 infectivity in TRIM5 lattice formation is greatly stimulated by the capsid hexameric human cells. J. Virol. 80, 2855–2862 (2006). 25 lattice (Fig. 4h). Little is known about how the innate immune system 20. Yan, N., Regalado-Magdos, A. D., Stiggelbout, B., Lee-Kirsch, M. A. & Lieberman, J. detects retroviruses26 and the discovery that TRIM5 acts as a PRR is an The cytosolic exonuclease TREX1 inhibits the innate immune response to important step towards filling this critical gap. The cellular factors human immunodeficiency virus type 1. Nature Immunol. 11, 1005–1013 21. (2010). Perron, M. J. et al. TRIM5a mediates the postentry block to N-tropic required for TRIM5 E3 activity and inflammatory gene induction, murine leukemia viruses in human cells. Proc. Natl Acad. Sci. USA 101, UBC13, UEV1A and TAK1, also promoted capsid-specific restriction 11827–11832 (2004). 22. activity, indicating that the multiple functions of TRIM5 are mechan- Ulm, J. W., Perron, M., Sodroski, J. & Mulligan, R. C. Complex determinants within the Moloney murine leukemia virus capsid modulate susceptibility of the istically linked. Identification of relevant TAK1 substrates will inform virus to Fv1 and Ref1-mediated restriction. Virology 363, 245–255 (2007). future attempts to pinpoint the mechanism of restriction. 23. Pornillos, O. et al. X-ray structures of the hexameric building block of the HIV capsid. Cell 137, 1282–1292 (2009). 24. METHODS SUMMARY Ganser, B. K., Li, S., Klishko, V. Y., Finch, J. T. & Sundquist, W. I. Assembly and 1,13,19 analysis of conical models for the HIV-1 core. Science 283, 80–83 (1999). Plasmids, cells and viruses. These methods were described previously 25. Ganser-Pornillos, B. K. et al. Hexagonal assembly of a restricting TRIM5a or are protein. Proc. Natl Acad. Sci. USA 108, 534–539 (2011). 26. detailed in the Supplementary Medzhitov, R. & Littman, D. HIV immunology needs a new direction. Nature Information. 455, 27. Recombinant protein. Production of full-length, soluble, TRIM5Cyp is 591 (2008). described in the supplement. CA A14C E45C was produced and assembled Pornillos, O., Ganser-Pornillos, B. K. & Yeager, M. Atomic-level modelling of the HIV capsid. Nature 469, 424–427 (2011). into tubes as described23,24. Supplementary Information is linked to the online version of the paper at www.nature.com/nature. Microarray. Illumina HumanHT-12 V3.0 expression bead chips were probed with RNA from TRIM5 knockdown THP-1 cells. Data set and Acknowledgements We thank D. Baltimore, M. J. Birrer, J. Brojatsch, A. Cimarelli, methods are available at the Gene Expression Omnibus A. DeIaco, S. Elledge, M. Emerman, W. Ferlin, D. Garcin, S. Ghosh, O. Haller, T. Hatziioannou, J. Hiscott, A. Iwasaki, D. Kolakofsky, M. Kosco-Vilbois, H. Malik, (Recewwwived.ncbi 31.n Julm.lyn 20ih.g10;ov/ accgeo)epted un d3er Mar acches - sion number GSE25041. 2011. R. Medzhitov, M. R. Neagu, G. Napolitani, P. Palese, D. Pinschewer, O. Pornillos, L. Roux, 1. Sayah, D. M., Sokolskaja, E., Berthoux, L. & Luban, J. Cyclophilin A O. Schwartz, M. Strubin, V. Studer, W. Sundquist, G. Towers, D. Trono, J. Tschopp, retrotransposition into TRIM5 explains owl monkey resistance to HIV-1. M. Yeager, M. Zufferey, and the Functional Genomics Center (Zu¨ rich), for ideas, Nature 430, 569–573 (2004). technical assistance, and reagents. This work was supported by NIH grant 2. Stremlau, M. et al. The cytoplasmic body component TRIM5a restricts RO1AI59159 to J.L., NIH grant R21AI087467 to W.M., Swiss National Science HIV-1 infection in Old World monkeys. Nature 427, 848–853 (2004). Foundation grant 3100A0-128655 to J.L. and 3100A0-122342 to M.G. and UZH 3. Sebastian, S. & Luban, J. TRIM5a selectively binds a restriction-sensitive Forschungskredit 54041402 to S.Z.

retroviral capsid. Retrovirology 2, 40 (2005). Author Contributions T.P., S.H., J.G., C.R., C.S., M.P., W.M., M.G.G. and J.L. designed the 4. Stremlau, M. et al. Specific recognition and accelerated uncoating of retroviral experiments; T.P., S.H., D.M., S.Z., J.G., J.La., C.R., F.A.S., M.P., A.B., P.D.U. and L.C. capsids by the TRIM5a restriction factor. Proc. Natl Acad. Sci. USA 103, performed the experiments. All authors contributed to the assembly and writing 5514–5519 (2006). of the manuscript. 5. Berthoux, L. et al. As2O3 enhances retroviral reverse transcription and counteracts Ref1 antiviral activity. J. Virol. 77, 3167–3180 (2003). Author Information Reprints and permissions information is available at 6. Shi, M. et al. TRIM30a negatively regulates TLR-mediated NF-kB activation by www.nature.com/reprints. The authors declare no competing financial targeting TAB2 and TAB3 for degradation. Nature Immunol. 9, 369–377 interests. Readers are welcome to comment on the online version of this 7. (2008). Tareen, S. U. & Emerman, M. Human Trim5a has additional activities article at www.nature.com/nature. Correspondence and requests for that are uncoupled from retroviral capsid recognition. Virology 409, 113– materials should be addressed to J.L. ([email protected]). 120 (2011).

192 SUPPLEMENTARY INFORMATION doi:10.1038/nature09976

a 3 b 3 c 4 10 10 10 Prl promoter ISRE luc IFNB1 luc

6 × AP-1 site e e e 3

g 10 2 g 2 g n 10 n 10 n a a a h h h 2 c c c 10

d d d

l 1 l 1 l o 10 o 10 o

F 1 F F 10

0 0 0 10 10 10 pcDNA TRIM5 c-Jun

3 d 10 e AP-1 luc NF-⎢B luc

1 e e 10 2 ng 10 ng a a h h c c

d d l 1 l 10 Fo Fo

0 0 10 10

LPS TLR4/MD2 IFN IFNR

 TRIM5 STAT1/2 IRF9 NF-⎢B AP1 IRF3

inﬂamm antiviral IFN chemo/cyto ISGs  Supplementary Figure 1. a-e Luciferase assays of HEK-293 cells transfected

with the indicated luciferase reporter plasmids and pcDNA3.1 expression plasmids

for the indicated genes. f, The role of TRIM5 in signaling. Horizontal double line,

cytoplasmic membrane; vertical dashed line separates the two waves of innate

signaling.

W W W . N A T U R E . C O M / N A T U R E | 1

193 RESEARCH SUPPLEMENTARY INFORMATION

a pAPM ⎠

CMVp R U5 RRE cPPT SFFVp PuroR miR-30 shRNA WPRE ⊗U3 R U5

3 2 b 10 3 c 10 TRIM5 10 TRIM5

A 1

A ll 10 ce

RN 2

10 RN 2

10 +

m 0 m

P 10 control KD, B-MLV F ve ve G i i TRIM5 KD, B-MLV t t a % 1 a 1

l -1 10 l 10 10 control KD, N-MLV e e R R TRIM5 KD, N-MLV 10-2 0 1 2 100 100 10-1 10 10 10 Relative RT units

d 15 25 8 30 1.5

20 60 THP-1 RNA 10 6 20 1.0

m 15 l e r control KD 10 40 1: 1: 0.5 P 5 4 10 H

TRIM5 KD T 5

0 0 2 20 0 TRIM5 CXCL10 CCL8 IL6 PTGS2 OAZ1

120 800 150 control KD MDM A 90 600 TRIM5 KD RN 100

ve 400 i t a

l 50 e 30 200 R 0 0 0 CXCL10 IL6 CCL8

Supplementary Figure 2. a, Schematic of the lentiviral vector pAPM where

both puromycin N-acetyltransferase and microRNA-based shRNA expression are

driven by the spleen focus forming virus (SFFV) pol II promoter. b, TE671 cells,

transduced with pAPM stably expressing the indicated shRNAs, were assayed for

TRIM5 knockdown by qRT-PCR. c, cycling THP-1 monocytes, stably expressing

pAPM shRNA targeting TRIM5 or control shRNA, were challenged with VSV-G-

pseudotyped N- or B-MLV GFP reporter vectors and assayed by FACS 72 h post-

transduction. d, qRT-PCR for the indicated mRNAs harvested 2 to 8 hrs after LPS-

treatment of THP-1 macrophages, depending on the peak values for that gene. e, qRT-

PCR for the indicated mRNAs harvested 2 to 8 hrs after LPS-treatment of MDM,

depending on the peak values for that gene.

194

2 | W W W . N A T U R E . C O M / N A T U R E SUPPLEMENTARY INFORMATION RESEARCH

a b 1 c 7

10 10 1 MDDC MDM THP-1 lls 10 vehicle

ce 6 10

lls + 0 LPS P 0 10 ce

10 U + G L 5 P 10 % R F 1: 1:

- -1 10 -1 4 V 10 I % 10 H -2 10 3 10 -2 0 1 2 10 10 10 10 Relative RT units

THP-1: TRIM5 d 2 e 2 10 MDM: TRIM5 10

A A

RN m

m 1 10 ve

1 i 10 t ve a i l

t e a R l

e R 100 0 10

8 2 VSV f 10 g 10 HIV-1

7 vector 10 1 vector

lls 10 TRIM5 TRIM5 6 ce

U 10 vector + LPS + L 0 P 10 vector + LPS

R 5 F 10 TRIM5 + LPS G TRIM5 + LPS

-1 4 % 10 10

103 10-2 0 1 2 -3 -2 -1 0 1 2 10-1 10 10 10 10 10 10 10 10 10 Relative RT units Relative MOI virus dilution

Supplementary Figure 3. MDDC (a) or MDM (b) were treated with the

indicated compounds for 24 h, challenged with a VSV-G-pseudotyped HIV-1 GFP reporter minimal vector, and assayed by FACS 72 h post-transduction. TPA-

differentiated THP-1 cells (c) were treated with ultrapure E. coli K12 LPS for 24 h and challenged with a VSV-G-pseudotyped HIV-1 luciferase reporter virus, and

assayed by luciferase assay 72 h post-transduction. MDM (d) or TPA-differentiated

THP-1 cells (e) were treated with the indicated compounds for 16 h, and qRT-PCR

was performed for TRIM5. TPA-differentiated THP-1 cells stably expressing human TRIM5α, or empty vector control cells, were challenged with an HIV-1 luciferase

reporter virus (f) or Vesicular Stomatitis Virus bearing a GFP reporter (g). HIV-1 was

assayed by luciferase assay 72 h

after challenge and VSV was assayed by FACS 20 h after challenge. 195

W W W . N A T U R E . C O M / N A T U R E | 3 RESEARCH SUPPLEMENTARY INFORMATION

7 a 10 ctrl KD, vehicle b S

c P KD

T5 KD, vehicle

L s

t 6 i 10 ctrl KD, LPS 10 m KD

T5 KD, LPS o un r 2 f c R = 0.98 e u 5 l 10

u 1

control IRF3 - V I esc

4 r H 10

l 5 1: IRF3 o P f

H 103 T

1: - β-actin

V I 2 H 10 1 100 101 102 Relative RT units 1 THP1: fold T5 KD 10 THP-1

KD KD

KD KD 2 d e 10 SIV

T2 T2 lls A A T T ce

+ control S control S 10

P F G STAT2 %

1: 100 ctrl KD, vehicle P - H T5 KD, vehicle actin T ctrl KD, LPS T5 KD, LPS 10-1 IFN-β 0 1 2 − − + + 10-1 10 10 10 Relative RT units THP-1

MDM: LPS 4 f 10

s control KD TRIM5 KD

ll 3 10 100 ce

+ 2 P 10 F

1 H 10 % -

: V

I 1.98 0 0.047 10 S -1 FL2 10 0 1 2 3 4 0 1 2 3 4 10 10 10 10 10 10 10 10 10 10 1 GFP KD: Ct rl T5

h i g MDDC 20 MDDC

10 e T5 KD e IRF3 KD u

u 8 15 esc control KD control IRF3 KD r esc

d l d

6 l 10 o f o

f : IRF3 : V I

4 V I

S 5 S 2 β-actin 0

MDDC

Supplementary Figure 4. a, CD4/CCR5-THP-1 macrophages transduced with

lentiviral vectors expressing pol II-driven microRNA-based shRNAs targeting TRIM5

or control, were treated with LPS or vehicle, and challenged with CCR5- tropic HIV-1

luciferase reporter virus. b, Correlation between TRIM5 KD efficiency, as measured by

qRT-PCR, and magnitude rescue of HIV-1 infectivity

196

4 | W W W . N A T U R E . C O M / N A T U R E SUPPLEMENTARY INFORMATION RESEARCH

from LPS. Immunoblot of whole-cell lysates from TPA-differentiated THP-1 cells transduced with pAPM expressing shRNA targeting IRF3 (c), STAT2 (d) or control shRNA. STAT2 KD and control KD cells were either treated with recombinant IFN-β for 16 h, or left untreated, prior to lysis. e, THP-1 macrophages were treated with LPS or vehicle and challenged with a VSV G- pseudotyped SIVMAC239GFP reporter virus.

MDM (f) or MDDCs (g-i), transduced with lentiviral KD vectors targeting TRIM5 (f and h) or IRF3 (g and i), were treated for 20 hrs with LPS (f) or as indicated (h and i), and then challenged with an SIVMAC239GFP reporter virus. Results are reported as percent infected (i) or as fold-change compared to control (h and i). Lysates were probed with the indicated antibodies in (g).

197

W W W . N A T U R E . C O M / N A T U R E | 5 RESEARCH SUPPLEMENTARY INFORMATION

b KD

a IK IKK KD 〈

MEK

 〈 AK1 ZsGreen IKK K  K1 T AK1 control T

AB1 AB2 AB3 AB3

175− T T T T 

80− TAK1 58− IP: FLAG 46− IB: FLAG - actin 30− −IgH HEK-293

2 10 e KD

d 4

10 KD c

control KD 1 10

e 3 TAK1 KD cells

10 AK1 + control T ng a GFP h 2 AP-1 luc c

10 % 0 TAK1 10

d vector

l AoT5Cyp-WT

Fo 1 10 AoT5Cyp-H436Q - 10 -1 100 0 1 2 actin 10-1 10 10 10 pcDNA c-Jun Relative RT units THP-1

f 8 Ctrl KD A

DN TRIM5 KD

V TAK1 KD

L 6 M - N

f o

e u 4 esc r

d l o f

P 2 H T

Supplementary Figure 5. a, Immunoblot of the indicated FLAG-tagged

proteins immunopreciptated from transfected 293T cells. b, Immunoblot of TAK1

knockdown in HEK-293 cells. c, Luciferase assay of TAK1 or control KD HEK-293

cells transfected with an AP-1 luciferase reporter plasmid, and either empty pcDNA or a c-Jun expression plasmid. d, Immunoblot of TAK1 KD TPA- differentiated THP-1

cells. e, THP-1 cells expressing either wild type AtTRIM5Cyp or the H436Q mutant

(or an empty vector control line) were challenged with a VSV-G-pseudotyped HIV-1 GFP reporter virus and analyzed by FACS 72 after transduction. f, PCR for full-length

N-tropic MLV viral cDNA in TRIM5, TAK1, or control KD THP-1 cells.

198

6 | W W W . N A T U R E . C O M / N A T U R E SUPPLEMENTARY INFORMATION RESEARCH

3 a 10 c 80 b AP-1 luc control KD

KD e Ubc13 KD 2 60

ng 10 ease a r h c

c AP-1 luc

control Ubc13 40 i d

l 1 d

10 l Fo

Ubc13 Fo 20

0 10 β-actin pcDNA c-Jun 0

HEK-293 d 5 f

only only

NF-�B luc

KD e

4 wt K48 K63

a h Ub Ub Ub - - - c 3

control Ubc13 d l pcDNA HA HA HA

Fo Ubc13 2

1 β-actin pcDNA Ub WT Ub K48 Ub K63 260 − 160 − THP-1 110 − 80 − 60 − 50 − 40 −

30 − 20 −

15 − 10 −

IB: HA

Supplementary Figure 6. a, Immunoblot of UBC13 KD in HEK-293 cells. b,

Luciferase assay of UBC13 or control KD HEK-293 cells transfected with an AP-1 luciferase reporter plasmid, and either empty pcDNA or a c-Jun expression plasmid. c, Luciferase assay of control KD, UBC13 KD, or UEV1A KD HEK-293 cells

transfected with an AP-1 luciferase reporter plasmid, and a human TRIM5〈 expression plasmid. d, Immunoblot of UBC13 KD in TPA- differentiated THP-1 cells. e, HEK-

293 cells were transfected with the indicated plasmids, along with an NF-⎢B luciferase reporter. f, Immunoblot of whole cell lysates from HEK-293 cells transfected with the indicated plasmids.

199

W W W . N A T U R E . C O M / N A T U R E | 7 RESEARCH SUPPLEMENTARY INFORMATION

a AtTRIM5Cyp HsTRIM5〈 AtTRIM5Cyp HsTRIM5〈 1 2 3 4 5 1 2 3 4 5 1 2 3 4 5 1 2 3 4 5

−MBP-TRIM5Cyp 72−

55− 1) GB1 2) NusA 43− 3) ZZ 4) Trx 34− 5) MBP

Pellets Elution

b c TRIM5Cyp WT

ml ml ml ml ml ml ml ml ml

1.5 1 12.0 12.5 13.0 13.5 14.0 14.5 15.0 15.5 16.0

(kDa)

mass

50−

(mAu) TRIM5Cyp molecular

d TRIM5Cyp H436Q unit)

(arb.

Absorption index

kDa 44 440 158

Refractive − − − Elution volume (mL) Elution volume (mL)

Supplementary Figure 7. a, Coomassie-stained SDS-PAGE gels showing

insoluble (Pellets) or soluble (Elution) protein fractions. N-terminal tag fusions to Owl

monkey TRIM5Cyp and human TRIM5〈 were screened for soluble expression in transfected Sf9 cells. The fusion constructs all carried an N- terminal His-tag. To detect

soluble expressed constructs, fusion constructs were batch purified on Ni-NTA beads

and analyzed by SDS-PAGE after elution with 250 mmol⁄L imidazole (Elution). Insoluble fractions were obtained by centrifugation of lysed cells (Pellets). b, Gel

filtration chromatogram showing the elution profile of recombinant TRIM5Cyp protein on a Superdex 200 column, and a coomassie-stained SDS-PAGE gel of the indicated

fractions (insert). The apparent molecular mass was estimated by comparison to

standard molecular weight markers depicted in the chromatogram. c, d, SEC-MALS results. The

elution profiles are depicted in refractive index detection and the calculated 200

8 | W W W . N A T U R E . C O M / N A T U R E SUPPLEMENTARY INFORMATION RESEARCH

molecular masses obtained from the light scattering data are shown as distributions across the peaks. Comparable molecular masses were obtained for

TRIM5Cyp WT (c) and TRIM5Cyp H436Q (d) showing that both proteins behave the same in solution.

201

W W W . N A T U R E . C O M / N A T U R E | 9 RESEARCH SUPPLEMENTARY INFORMATION

E2 A

TRIM5Cyp b Ub a No No WT WT K48R K63R K48 K63 No + + ATP − + + + + + +

130− −UBE1 170− 72− 72− 55− −T5Cyp 55− 34− 26− −UEV1A 34− −UBC13 17− −Ub 2 11− −Ub 11− Coomassie IB: Ub

c 〈〈 RING RING

ZsGreen TRIM5 C15/18A ⊗ TRAF6 TRAF2 d ZsGreen TRIM5 C15/18A ⊗ TRAF6 TRAF2

170− 170− 72− 72− 55− 55−

34− 34−

IB: FLAG IB: K63

Supplementary Figure 8. a, Products of in vitro reactions with ATP, purified

Ubiquitin, UBE1, the UBC13/UEV1A E2 complex, and increasing amounts of the E3

Ubiqutin Ligase AtTRIM5Cyp, revealed by Coomassie. Gel pieces in the higher

molecular weight (> 170 kDa) and lower molecular weight (40 – 70 kDa) regions were

cut out separately (black boxes) and analyzed by MALDI-MS/MS after in-gel tryptic

digestion. b, Immunoblot of products from in vitro reactions with ATP, purified WT

Ubiquitin, or the indicated Ubiquitin mutants, UBE1, the UBC13/UEV1A E2 complex,

and AtTRIM5Cyp. c, Immunoblot of the indicated FLAG-tagged proteins

immunoprecipitated from lysates of transfected HEK- 293 cells. d, Immunoblot of

products of in vitro reactions with ATP, purified Ubiquitin, UBE1, the UBC13/UEV1A

E2 complex, and the indicated transfected FLAG-tagged proteins immunoprecipitated

from HEK-293 lysates.

202

1 0 | W W W . N A T U R E . C O M / N A T U R E SUPPLEMENTARY INFORMATION RESEARCH

Supplementary Figure 9. Identification of K63-linked polyubiquitin chains formed by the E3-ligase activity of AtTRIM5Cyp: Gel pieces containing products from an in vitro ubiquitinylation assay are analyzed by MS after in-gel tryptic digestion. (a) MALDI spectrum of tryptically digested fragments: Mass over charge ratios (m⁄z) of signals matching theoretical peptide fragments of Ubiquitin are labeled in the spectrum. Numbers below the indicated m⁄z ratios correspond to residues in the Ubiquitin sequence (shown on top of the spectrum). Detected signals cover 94 % of the Ubiquitin sequence. Fragments with detected masses matching Ubiquitin peptides carrying a modified lysine are marked as red numbers in the spectrum and were further analyzed by MS/MS. MS/MS spectra of the peptides 2244.2 m⁄z (b) and 2513.4 m⁄Z (c)

203

W W W . N A T U R E . C O M / N A T U R E | 1 1 RESEARCH SUPPLEMENTARY INFORMATION

show signals for fragment ions (a-, b-, and y-ions labelled in different colours) with mass differences corresponding to single amino acids in the peptide sequence (shown on top of the spectrum). The mass differences K* match a lysine modified with amino acids GG or LRGG. (b) MS/MS spectrum identifying fragment 2244.2 m⁄z as Ubiquitin peptide 55-72 with additional amino acids GG linked to K63 (see inset). (c) MS/MS spectrum identifying fragment 2513.4 m⁄z as Ubiquitin peptide 55-72 with additional amino acids LRGG linked to K63 (see inset).

204

1 2 | W W W . N A T U R E . C O M / N A T U R E SUPPLEMENTARY INFORMATION RESEARCH

a Coomassie Elution M L FT W1 W5 1 2 3 4 5 R M

-250

170- -150

-100

72- -75

55- -50

-37 34-

-25

FLAG TAB2 TAB1 TAK1 b c d e 170- 170- 170- 170-

72- 72- 72- 72-

55- 55- 55- 55-

34- 34- 34- 34-

Supplementary Figure 10. Figure S10: Purification of recombinant TAK1 complex components. In panel a, FS-TAK1, FS-TAB1, and FS-TAB2 were produced in 293T and purified by Streptactin Sepharose chromatography as described under Experimental Procedures. Aliquots (1 µL) of the soluble lysate (lane L), the Streptactin Sepharose flow-through (lane FT), and wash 1 (lane W1) fractions, and aliquots (10 µl) of the wash 5 (lane W5), and the 2.5 mM Desthiobiotin eluate fractions were analyzed by SDS−PAGE. The polypeptides were visualized by staining the gel with Coomassie Blue dye. The positions and sizes (kDa) of marker polypeptides are indicated. In panel b, c, d, and e, aliquots (10 µl) were analyzed by immunoblotting with the indicated antibodies.

205

W W W . N A T U R E . C O M / N A T U R E | 1 3 RESEARCH SUPPLEMENTARY INFORMATION

a MDM

10 4

A 8 3

ve 2 i t 4

a l e 1 R 2

0 0 CXCL10 CCL8 N-MLV

B-MLV b MDDC

5 10

A 4 8 RN m

3 6

i t 2 4 a l

e R 1 2

0 0 PTGS2 IL6

Supplementary Figure 11. qRT-PCR for the indicated mRNAs from MDM (a) or MDDC (b). mRNA was harvested 6 h after challenge with VSV-G- pseudotyped N-tropic or B-tropic MLV. Data are expressed as fold change versus media control.

206

1 4 | W W W . N A T U R E . C O M / N A T U R E SUPPLEMENTARY INFORMATION RESEARCH

b control capsid control capsid min: 0 3 6 9 12 0 3 6 9 12 0 3 6 9 12 0 3 6 9 12

130− −UBE1 72− −TRIM5Cyp 55−

34− 26− −UEV1A / capsid −UBC13

17− −Ub2

11− − Ub TRIM5Cyp TRIM5Cyp H436Q Coomassie

control capsid c TRIM5Cyp TRIM5Cyp H436 Q d min: 0 3 6 9 12 0 3 6 9 12 0 3 6 9 12 0 3 6 9 12

170− 170−

72− 72− 55− 55− IB: Ub

34− 34−

11− 11−

IB: Ub IB: Ub IB: CA

IB: TRIM5 TRIM5Cyp H436Q

Supplementary Figure 12. a, Electron micrograph of HIV-1 capsid tubes made

with oxidized A14C/E45C mutant in the absence of salt (magnification, 33,000x). b,

Coomassie-stained SDS-PAGE gels of samples from the in vitro ubiquitylation time-

course reactions using recombinant TRIM5Cyp WT or the H436Q mutant carried out in the presence of UBE1 and UBC13/UEV1A, with or without the addition of

assembled HIV-1 capsid (CA A14C E45C). c,

Immunoblot of samples from an in vitro ubiquitylation time-course reaction using recombinant TRIM5Cyp WT or the H436Q mutant, carried out in the presence

of UBE1 and UBC13/UEV1A. d, Immunoblot of samples from an in vitro

ubiquitylation time-course reaction using the recombinant TRIM5Cyp H436Q mutant, carried out in the presence of UBE1 and UBC13/UEV1A, with or without

the addition of assembled HIV-1 capsid (CA A14C E45C).

W W W . N A T U R E . C O M / N A T U R E | 1 5

207