Functional Analysis of Rev Binding Region 2, a Structural Element of the EIAV Rev Responsive Element Jerald Rudy Chavez Iowa State University

Iowa State University Capstones, Theses and Graduate Theses and Dissertations Dissertations

2015 Functional analysis of Rev binding region 2, a structural element of the EIAV Rev responsive element Jerald Rudy Chavez Iowa State University

Follow this and additional works at: https://lib.dr.iastate.edu/etd Part of the Virology Commons

Recommended Citation Chavez, Jerald Rudy, "Functional analysis of Rev binding region 2, a structural element of the EIAV Rev responsive element" (2015). Graduate Theses and Dissertations. 14323. https://lib.dr.iastate.edu/etd/14323

This Thesis is brought to you for free and open access by the Iowa State University Capstones, Theses and Dissertations at Iowa State University Digital Repository. It has been accepted for inclusion in Graduate Theses and Dissertations by an authorized administrator of Iowa State University Digital Repository. For more information, please contact [email protected].

Functional analysis of Rev binding region 2, a structural element of the EIAV Rev responsive element

Jerald Rudy Chavez

A thesis submitted to the graduate faculty

in partial fulfillment of the requirements for the degree of

MASTER OF SCIENCE

Major: Genetics

Program of Study Committee: Susan L. Carpenter, Major Professor Drena L. Dobbs Jeffrey M. Trimarchi

Iowa State University

Ames, Iowa

2015

TABLE OF CONTENTS ABSTRACT ...... iii CHAPTER 1. GENERAL INTRODUCTION ...... 1 Thesis Organization ...... 1 Introduction ...... 1 Retroviruses ...... 1 Lentiviruses and Alternative Splicing ...... 2 EIAV and Variation ...... 4 EIAV Rev...... 5 Rev Structure ...... 6 EIAV RRE ...... 7 Overall Objectives ...... 9 Figures ...... 10 References ...... 15 CHAPTER 2. FUNCTIONAL ANALYSIS OF REV BINDING REGION 2, A STRUCTURAL ELEMENT OF THE EIAV RRE ...... 20 Abstract ...... 20 Introduction ...... 21 Methods and Materials ...... 23 Results ...... 26 Discussion ...... 29 Acknowledgements ...... 30 Figures ...... 31 References ...... 36 CHAPTER 3. GENERAL DISCUSSION ...... 40 Future Studies ...... 42 References ...... 43 APPENDIX. EFFECT OF GENETIC VARIATION ON PREDICTED RNA BINDING INTERFACE OF REV PROTEIN ...... 46 References ...... 47

iii

ABSTRACT

Lentiviruses translate their structural and enzymatic proteins from unspliced and singly spliced viral mRNAs (vRNAs). Export to the cytoplasm of the incompletely spliced transcripts is mediated by the viral Rev protein. Rev binds to a highly structured region of the unspliced/singly spliced RNA, termed the Rev responsive element (RRE), multimerizes, and tethers the RNA to cellular export factors to cross the nuclear membrane. While Rev-RRE mediated nuclear export is well studied in human immunodeficiency virus type 1 (HIV-1), the interaction between the equine infectious anemia virus (EIAV) Rev and the RRE is less well studied. Chemical footprinting analysis indicates Rev interacts with two regions on the RNA, termed Rev binding region one (RBR-1) and Rev binding region 2 (RBR-2). RBR-1 contains a 56nt sequence termed the minRRE sufficient to both bind Rev and support nuclear export. The minRRE spans Rev exon 1 and an exon splicing enhancer (ESE) necessary for Rev exon 1 splicing. RBR-2 was previously functionally uncharacterized. RBR-2 contains predicted structural motifs found in several other lentiviruses, and is required for Rev high affinity binding, suggesting this region is biologically relevant and may be involved in Rev nuclear export. Here we examined the effect of deleting RBR-2 on nuclear export, virus structural gene expression, and virion production, and

Rev-SF2/ASF competition. Our findings indicate RBR-2 is neither necessary nor sufficient for nuclear export, and deletion of RBR-2 does not affect either virus structural gene expression or virion production, suggesting RBR-1 is the primary binding site and RRE for Rev. The RBR-1 sequence secondary structure and the genomic intronic location is distinct from that of HIV-1 and other lentivirus RREs. It is still unclear if RBR-2 has a functional role in virus replication. 1

CHAPTER 1. GENERAL INTRODUCTION

Thesis Organization

This thesis is a functional analysis of an RNA element important in the equine infectious anemia virus (EIAV) Rev nuclear export pathway. This thesis has three chapters and one appendix. Chapter one is a literature review that provides background information on retroviruses and EIAV, alternative splicing and the role of Rev in alternative splicing, and what is currently known about the Rev-RRE interaction in EIAV. Chapter two describes the functional analysis of RNA sequences important in Rev activity and is formatted for submission to the

Journal of General Virology. The manuscript was co-authored by Jerald Chavez, Hyelee Loyd,

Drena Dobbs and Susan Carpenter. I carried out experimental procedures, assisted in the design of experiments and interpretation of results, and am responsible for writing the manuscript.

Hyelee Loyd assisted in the construction of the pERRE-1B and pERRE-1C plasmid clones.

Drena Dobbs assisted in design of experiments and critical analysis of results. Susan Carpenter assisted in design of experiments, interpretation of results, and editing of the manuscript.

Chapter three summarizes the project and potential future studies that could be performed. The appendix includes computational analysis of the RNA binding residues for various EIAV Rev amino acid variants.

Introduction

Retroviruses

The retrovirus family consists of seven different subfamilies grouped by similarity in morphology and genetic relatedness. All retroviruses consist of two copies of the positive sense 2

RNA genome surrounded by an inner protein capsid and outer membrane envelope (Figure 1,

(1)). While retroviruses can differ by the presence or absence of accessory and regulatory genes, all retroviral genomes contain three specific genes: gag, pol, and env. The gag gene encodes interior structural proteins including the matrix, capsid and nucleocapsid proteins while pol encodes the enzymatic proteins including reverse transcriptase, protease, and integrase. Finally, env encodes the structural proteins of the envelope, including the surface unit (SU) and transmembrane protein (TM). Like all viruses, retroviruses must enter a cell, replicate their genome and protein components, assemble into new virions, and exit the infected cell. Entry is mediated by the SU proteins of the envelope upon interaction with their target cell membrane receptors. TM anchors the viral envelope proteins to the viral membrane, and also serves to initiate fusion of the viral membrane with target host cell membrane. This will release the viral core capsid and genome into the interior of the cell. The viral reverse transcriptase converts the

RNA genome into a DNA form termed the provirus, which is integrated into the host cell genome via integrase. Cellular polymerases transcribe the viral genes and upon translation, new viral RNA genomes and proteins assemble and bud off from the cell. The viral protease cleaves the Gag-Pol polyproteins into their individual proteins, making the virions mature and ready to infect new cells.

Lentiviruses and Alternative Splicing

Lentiviruses are one of seven subfamilies of the larger retrovirus family. “Lenti” in latin means slow and refers to the slow progression of the disease course that many in this virus class are known for, including human immunodeficiency virus type 1 (HIV-1). Retrovirus genomes are relatively small (8-10kb), and as such, utilize a number of strategies to maximize their coding capacity including ribosomal frameshifts, overlapping reading frames, and alternative splicing. 3

Unlike simple retroviruses that generally only express two RNAs (full length and singly spliced), lentiviruses (and spumaviruses) are so called “complex” retroviruses due to their use of multiple alternative splice sites to express high numbers of different mRNA species that code for additional regulatory and accessory proteins (1).

Some lentiviruses such as HIV-1 can generate up to 40 different mRNA species, however there are three basic classes of viral RNAs: full length RNA’s that are translated into Gag-Pol polyproteins or serve as new genomes; singly spliced RNA’s that are translated into the envelope proteins; and multi-spliced mRNAs that code for regulatory and/or accessory proteins. The cytoplasmic expression of unspliced and singly spliced mRNA species is temporally regulated by the Rev regulatory protein. The process and operations of Rev are best studied in HIV-1 (Figure

2). Rev is translated from a mutispliced RNA, imports into the nucleus via binding to Importin-

(2), and subsequently binds a structured RNA motif, known as the Rev responsive element

(RRE), which is present in the unspliced and singly spliced mRNA. After initial binding, Rev multimerizes along the RRE (3, 4) in a cooperative fashion (5). Rev then tethers the viral RNA to the cellular CRM1 protein to mediate export of the RNA cargo through the nuclear pore and to the cytoplasm (6).

The interaction between HIV-1 Rev and RRE is highly dependent on the secondary structure of the RRE (4, 7, 8). The HIV-1 RRE is a 351 nucleotide (nt) sequence that overlaps the intronic SU/TM border region (9). The HIV-1 RRE is a highly structured motif with stem regions labeled I-V (Figure 3). Early analysis indicated stem-loop II B (SLIIB) bound to Rev with similar affinity to the full RRE, at 1-3nM, suggesting SLIIB acted as the initial primary high affinity binding site for Rev (8, 10-13). The 17 amino acid (aa) arginine rich motif (ARM) of the

Rev protein inserts into the major groove site in the bulge of SLIIB and widens the groove as 4 electrostatic interactions between the ARM and the RNA form (reviewed in (14)). After initial monomer binding, Rev subsequently oligomerizes through protein-protein and low affinity protein-RNA interactions along the RNA. While early models hypothesized SLIIB acted as the primary nucleation site for Rev binding and subsequent oligomerization, more recent studies by

Daugherty et al.(15) identified Stem loop IA (SLIA) as a second high affinity Rev binding site capable of binding independent of SLIIB, however, neither site is capable of supporting nuclear export on its own.

EIAV and Variation

Equine infectious anemia virus (EIAV) is a lentivirus related to HIV-1 and the other human and non-human primate lentiviruses (16). Infection of horses by EIAV causes a rapid and dynamic disease course. Experimental infections can result in acute viremia within 5-30 days.

Thereafter, horses typically exhibit recurring cycles of fever, thrombocytopenia, viremia, and anemia anywhere from two months to one year after initial infection. While some horses exhibit prolonged initial clinical signs that can be fatal, most become clinically quiescent after the first year. This period of clinical quiescence, known as the inapparent stage of disease, is associated with low levels of viremia and the development of broadly acting immune responses believed to control, but not eradicate viral replication (reviewed in (17, 18)). This makes EIAV an excellent model to study factors involved in immunological control of lentivirus infections and mechanisms of viral persistence.

Viruses employ a number of strategies to evade or even suppress the host immune response. In lentiviruses, antigenic variation of the HIV-1 envelope is a well-studied mechanism of immune evasion and represents a significant challenge to the development of an effective vaccine (reviewed in (19)). Antigenic variation in lentiviruses can result from a number of viral 5 and cellular mechanisms including: high error rate of reverse transcriptase (20, 21), recombination between viral genomes (22), and hyper-editing of viral genomes due to enzymes such as APOBEC3G (23). The EIAV surface (SU) envelope gene has been shown to be a highly variable region (24), and unique viral envelope variants are isolated from recurring febrile cycles

(25-29) suggesting antigenic variation mediates immune escape. Later stage isolates also appear to be increasingly resistant to neutralizing antibodies (25, 29, 30).

EIAV Rev

In addition to variation in the SU region, significant genetic variation has been found in the Rev/TM overlapping reading frame (31-33). Unique dominant rev genotypes are found during different stages of disease (33). Phylogenetic analysis of Rev variants from experimentally infected horses indicated that variants clustered into two major clades; clade A comprised of chronic, inapparent, and late stage recrudescent Revs variants, and clade B comprised of mostly inoculum strain variants. Partition analysis of these groups indicated the two major subpopulations of Rev variants cycled in dominance over the course of infection, suggesting that each group varied in selective advantage (34). The presence of unique Rev genotypes and the expansion and contraction of subpopulations of Rev variants suggests Rev could facilitate immune evasion by regulating viral gene expression. In support of this, analysis of Rev activity during experimental EIAV infection indicated that changes in Rev export phenotype was associated with changes in stage of disease (33, 34).

EIAV Rev is functionally homologous to HIV-1 Rev and regulates structural gene expression (31, 35). EIAV Rev is a 165 aa protein possessing a unique domain organization compared to the prototype HIV-1 (Figure 4). EIAV Rev includes a nuclear export signal (NES, aa 31-55)(36), a bi-partite RNA binding domain comprised of two small arginine rich motifs 6

(ARMs), with the second small C’-terminal ARM also functioning as the nuclear localization signal (NLS)(37). Mutations in HIV-1 Rev’s NES have been reported in asymptomatic HIV-1 infected patients, leading to attenuated Rev export activity that is hypothesized to be involved in decreased viral replication (38). However, a majority of the genetic variation in EIAV Rev appears to localize outside the functional domain (32). The highest frequency of mutations in

Rev occurs in a hypervariable region (aa range 105-142) just proximal to the second N-terminal

ARM, and single point mutations in this region can alter Rev nuclear export phenotype (32). It is unclear how these mutations alter Rev nuclear export levels, however, multiple changes in this region alter the electrostatic charge of the local hypervariable region, which could feasibly alter the strength of the Rev-RRE interactions (32).

Rev Structure

Early modeling of the tertiary structure of EIAV Rev suggested that the two short ARMs of the bipartite RNA binding domain were positioned next to each other in the Rev monomer, forming a single binding RNA binding interface (39). However, more recent studies suggests that Rev could adopt an elongated structure placing the two short ARMs on opposing sides of the protein, preventing formation of a single RNA binding interface (40). Structure prediction and mutational analysis suggests that Rev dimerization is required for RNA binding. Docking prediction suggests that formation of the dimer juxtaposes RNA binding domain 1 from one Rev protein to RNA binding domain 2 of the second protein to form a single RNA binding interface on each side of the dimer (40). The details of how the protein interacts with the EIAV RRE are unclear. In particular, it is not known if dimerization of Rev results in two RNA interfaces, and if so, do both interfaces interact with distinct regions of the RRE? 7

EIAV RRE

Initial analysis indicated a 1700 nt sequence containing the majority of the envelope coding region, termed ERRE-All, possessed the highest Rev responsiveness in functional assays of export capacity (41). Subsequent mapping indicated that the first 555 nt of the ERRE-All, called ERRE-1, supported 52% the export activity of the full ERRE-All, while the remaining downstream sequence (termed ERRE-2) supported only 17% the activity of ERRE-All, suggesting ERRE-1 was the primary interaction site for Rev (41). ERRE-1 is located in the 5’

SU region of the envelope gene and spans Rev exon 1, unlike most characterized lentivirus responsive elements which do not overlap Rev coding regions and are located either at, or downstream of the SU/TM border,.

ERRE-1 spans the first coding exon of Rev and contains a 56nt minRRE sequence which is the smallest sequence known to both bind Rev and support nuclear export (42). The minRRE contains guanine/adenine/purine (GAR) sequence repeats that serve as a binding site for Rev and as an exon splicing enhancer (ESE) necessary for the inclusion of exon three of the bi-cistronic

RNA that codes for Tat and Rev (42, 43). EIAV Rev is translated from exon three and four of the bi-cistronic RNA. The presence of EIAV Rev is known to cause skipping of exon 3, resulting in

Tat only mRNAs (43). In addition, mutation in the ESE affects both Rev binding and splicing of exon three (42). This suggests that Rev regulates exon 3 splicing by binding to the ESE and blocking splicing factors access to the ESE. Support for this competition model came from subsequent analysis that showed that increasing amounts of the SR protein SF2/ASF (involved in both constitutive and alternative eukaryotic splicing (44)) specifically inhibited Rev export activity and EIAV replication in a dose dependent fashion (42, 45). 8

Computational predictions combined with chemical probing analysis indicated the

ERRE-1 sequence possessed a rather unstructured RNA topology (Figure 5)(46) compared to the highly ordered HIV-1 (10), FIV (47), and SIV (48) RRE secondary structures. Chemical footprint analysis of the RRE indicated that Rev interacted with two regions on the RRE, Rev binding region 1 (RBR-1) and Rev binding region 2 (RBR-2) (Figure 5)(46). RBR-1 spans Rev exon 1 and contains the minRRE/ESE sequence previously established to bind Rev and support nuclear export. The function of RBR-2 is not known. Like the HIV-1 RRE, RBR-2 is located in an envelope intronic region between rev exon 1 and 2. The EIAV Rev protein has a 19 nM binding affinity for the full ERRE-1 sequence. However deletion of RBR-2 results in ~250X decrease in Rev binding affinity for RRE (5.2 uM). In addition, unlike RBR-1 and the minRRE,which are highly unstructured, RBR-2 contains more stem structured regions reminiscent of the HIV-1 RRE. RBR-2 also contains predicted structural motifs similar to SLIIB which is also found in several other lentiviral genomes, suggesting a biological relevance for these structures (46). The requirement for high affinity binding and structural conservation data together suggests that RBR-2 could play a role in Rev mediated nuclear export and gene expression. Subsequent footprint analysis of SF2/ASF on the RRE indicated SF2/ASF bound two regions like Rev, with the first binding region partially overlapping RBR-1, and the second

SF2/ASF binding site just prior to RBR-2 (45) (Figure 5). It is unclear how or if RBR-2 is involved in SF2/ASF-Rev functional competition, however, the fine mapping data of Rev and

SF2/ASF on the RRE and the fact that RBR-2 facilitates Rev high affinity binding to the RRE could suggest RBR-2 provides Rev a competitive advantage for the RRE in competition with

SF2/ASF.

Overall Objectives

The overall objective of this study is to examine the role of RBR-2 in EIAV replication.

We hypothesize that RBR-2 is involved in EIAV Rev mediated export and virus production, and that RBR-2 provides Rev a competitive advantage in SF2/ASF competition for the ESE/RRE. To test these hypotheses, we conducted the following

1) Determine the effect of deleting RBR-2 on Rev mediated nuclear export and EIAV

gene expression.

2) Determine the effect of deleting RBR-2 on EIAV virion production

3) Determine the effect of deleting RBR-2 on Rev-SF2/ASF competition.

Figures

Figure 1. Cartoon depiction of retrovirus virion structure (adapted from (1)). Outer membrane consists of a lipid membrane embedded with SU/TM envelope trimers necessary for virus entry into a cell. Inner capsid proteins surround the two RNA genome copies and the rest of the viral structural and enzymatic proteins necessary to replicate upon entry into a target cell. CA = capsid, MA = matrix, NC = nucleocapsid, RT = reverse transcriptase,

PR = protease, IN = integrase, SU = surface unit, TM = transmembrane protein.

Figure 2. Cartoon depiction of the cellular role of Rev in nuclear export. Depiction is based on HIV-1 Rev prototype

(49). After transcription of the provirus, the viral pre-mRNA is spliced by the host spliceosome and exported to the cytoplasm. These mRNAs are translated into regulatory proteins such as Tat and Rev. Rev imports back into the nucleus, where it binds to the structured RNA element called the Rev responsive element (RRE) in partially spliced or unspliced viral RNA. Rev tethers these RNAs to nuclear export factors that then mediate transport of the RNAs to the cytoplasm, where they can be translated into the structural proteins of the virus or serve as new viral genomes.

Picture is courtesy of Mike Belshan.

Figure 3. Predicted HIV-1 RRE secondary structure (15). Stem nomenclature is from (50) and numbering corresponds to (51).

Figure 4. Domain organization of EIAV Rev and HIV-1 Rev proteins. EIAV domain numbers according to (32, 40),

HIV-1 domain numbers according to (49). Lengths are to relative scale. Abbreviations : NES = nuclear export signal, RBD = RNA binding domain, NER = non-essential region, NLS = nuclear localization signal.

Figure 5. Two distinct regions of the ERRE undergo structural transitions in the presence of bound EIAV Rev and

SF2/ASF (reproduced from (45)). Consensus chemical modification patterns, based on at least 3 experiments in which several different primers were used to probe the complete ERRE RNA, mapped onto the RNA secondary structure. Ribonucleotides that consistently displayed enhanced modification with either kethoxal or DMS upon protein binding are circled: bold circle (strong) and thin circle (mild). Regions protected from hydroxyl radical cleavage in the presence of protein are denoted by a thick line. Purine-rich motifs are highlighted in green. Blue circles represent general areas modified in the presence of Rev, pink circles represent general areas modified in the presence of SF2/ASF.

References

1. Coffin JM, Hughes SH, Varmus HE. 1997. Retroviruses. Cold Spring Harbor Laboratory Press, New York.

2. Truant R, Cullen BR. 1999. The arginine-rich domains present in human immunodeficiency virus type 1 Tat and Rev function as direct importin beta-dependent nuclear localization signals. Molecular and Cellular Biology 19:1210-1217.

3. Malim MH, Cullen BR. 1991. HIV-1 structural gene expression requires the binding of multiple Rev monomers to the viral RRE: implcations for HIV-1 latency. Cell 65:241- 248.

4. Olsen H, Cochrane A, Dillon P, Nalin C, Rosen C. 1990. Interaction of the human immunodeficiency virus type 1 Rev protein with a structured region in env mRNA is dependent on multimer formation mediated through a basic stretch of amino acids. Genes & Development. 4:1357-1364.

5. Jain C, Belasco JG. 2001. Structural model for the cooperative assembly of HIV-1 Rev multimers on the RRE as deduced from analysis of assembly-defective mutants. Molecular Cell 7:603-614.

6. Neville M, Stutz F, Lee L, Davis LI, Rosbash M. 1997. The importin-beta family member Crm1p bridges the interaction between Rev and the nuclear pore complex during nuclear export. Current Biology 7:767-775.

7. Heaphy S, Dingwall C, Ernberg I, Gait MJ, Green SM, Karn J, Lowe AD, Singh M, Skinner MA. 1990. HIV-1 regulator of virion expression (Rev) protein binds to an RNA stem-loop structure located within the Rev response element region. Cell 60:685-693.

8. Malim MH, Tiley LS, McCarn DF, Rusche JR, Hauber J, Cullen BR. 1990. HIV-1 structural gene expression requires binding of the Rev trans-activator to its RNA target sequence. Cell 60:675-683.

9. Solomin L, Felber BK, Pavlakis GN. 1990. Different sites of interaction for Rev, Tev, and Rex proteins within the Rev-responsive element of human-immunodeficiency-virus type-1. Journal of Virology 64:6010-6017.

10. Kjems J, Brown M, Chang DD, Sharp PA. 1991. Structural-analysis of the interaction between the human-immunodeficiency-virus Rev protein and the Rev response element. Proceedings of the National Academy of Sciences of the United States of America 88:683-687.

11. Cook K, Sue , Fisk GJ, Hauber J, Usman N, Daly TJ, Rusche JR. 1991. Characterization of HIV-1 Rev protein: binding stoichiometry and minimal RNA substrate. Nucleic Acids Research 19:1577-1583.

12. Huang X, Hope TJ, Bond BL, McDonald D, Grahl K, Parslow TG. 1991. Minimal Rev- response element for type 1 human immunodeficiency virus. Journal of Virology 65:2131-2134.

13. Heaphy S, Finch JT, Gait MJ, Karn J, Singh M. 1991. Human-immunodeficiency-virus type-1 regulator of virion expression, Rev, forms nucleoprotein filaments after binding to a purine-rich bubble located within the Rev-responsive region of viral messenger-RNAs. Proceedings of the National Academy of Sciences of the United States of America 88:7366-7370.

14. Weiss MA, Narayana N. 1999. RNA recognition by arginine-rich peptide motifs. Biopoly 48:167-180.

15. Daugherty MD, D'Orso I, Frankel AD. 2008. A solution to limited genomic capacity: using adaptable binding surfaces to assemble the functional HIV Rev oligomer on RNA. Molecular Cell 31:824-834.

16. Katzourakis A, Tristem M, Pybus OG, Gifford RJ. 2007. Discovery and analysis of the first endogenous lentivirus. Proceedings of the National Academy of Sciences of the United States of America 104:6261-6265.

17. Leroux C, Cadoré JL, Montelaro RC. 2004. Equine infectious anemia virus (EIAV): what has HIV's country cousin got to tell us? Vet Research 35:485-512.

18. Cheevers WP, McGuire TC. 1985. Equine infectious-anemia virus - immunopathogenesis and persistence. Reviews of Infectious Diseases 7:83-88.

19. Kwong PD, Mascola JR, Nabel GJ. 2012. The changing face of HIV vaccine research. Journal of the International Aids Society 15.

20. Preston BD, Poiesz BJ, Loeb LA. 1988. Fidelity of HIV-1 reverse transcriptase. Science 242:1168-1171.

21. Roberts JD, Bebenek K, Kunkel TA. 1988. The accuracy of reverse transcriptase from HIV-1. Science 242:1171-1173.

22. Robertson D, Sharp P, McCutchan F, Hahn B. 1995. Recombination in HIV-1. Nature 374:124-126.

23. Mangeat B, Turelli P, Caron G, Friedli M, Perrin L, Trono D. 2003. Broad antiretroviral defence by human APOBEC3G through lethal editing of nascent reverse transcripts. Nature 424:99-103. 17

24. Lichtenstein DL, Issel CJ, Montelaro RC. 1996. Genomic quasispecies associated with the initiation of infection and disease in ponies experimentally infected with equine infectious anemia virus. Journal of Virology 70:3346-3354.

25. Howe L, Leroux C, Issel CJ, Montelaro RC. 2002. Equine infectious anemia virus envelope evolution in vivo during persistent infection progressively increases resistance to in vitro serum antibody neutralization as a dominant phenotype. Journal of Virology 76:10588-10597.

26. Leroux C, Issel CJ, Montelaro RC. 1997. Novel and dynamic evolution of equine infectious anemia virus genomic quasispecies associated with sequential disease cycles in an experimentally infected pony. Journal of Virology 71:9627-9639.

27. Pang H, Kong X-G, Sentsui H, Kono Y, Sugiura T, Hasegawa A, Akaski H. 1997. Genetic variation of envelope gp90 gene of equine infectious anemia virus isolated from an experimentally infected horse. Journal of Veterinary Medical Science 59:1089-1095.

28. Leroux C, Craigo JK, Issel CJ, Montelaro RC. 2001. Equine infectious anemia virus genomic evolution in progressor and nonprogressor ponies. Journal of Virology 75:4570- 4583.

29. Sponseller BA, Sparks WO, Wannemuehler Y, Li Y, Antons AK, Oaks JL, Carpenter S. 2007. Immune selection of equine infectious anemia virus env variants during the long- term inapparent stage of disease. Virology 363:156-165.

30. Zheng Y-H, Sentsui H, Nakaya T, Kono Y, Ikuta K. 1997. In vivo dynamics of equine infectious anemia viruses emerging during febrile episodes: insertions/duplications at the principle neutralizing domain. Journal of Virology 71:5031-5039.

31. Alexandersen S, Carpenter S. 1991. Characterization of variable regions in the envelope and S3 open reading frame of equine infectious anemia virus. Journal of Virol 65:4255- 4262.

32. Sparks WO, Dorman KS, Liu SJ, Carpenter S. 2008. Naturally arising point mutations in non-essential domains of equine infectious anemia virus Rev alter Rev-dependent nuclear-export activity. Journal of General Virology 89:1043-1048.

33. Belshan M, Baccam P, Oaks JL, Sponseller BA, Murphy SC, Cornette J, Carpenter S. 2001. Genetic and biological variation in equine infectious anemia virus Rev correlates with variable stages of clinical disease in an experimentally infected pony. Virology 279:185-200.

34. Baccam P, Thompson RJ, Li Y, Sparks WO, Belshan M, Dorman KS, Wannemuehler Y, Oaks JL, Cornette JL, Carpenter S. 2003. Subpopulations of equine infectious anemia virus Rev coexist in vivo and differ in phenotype. Journal of Virology 77:12122-12131. 18

35. Rosin-Arbesfeld R, Rivlin M, Noiman S, Mashiah P, Yaniv A, Miki T, Tronick SR, Gazit A. 1993. Structural and functional characterization of Rev-like transcripts of equine infectious anemia virus. Journal of Virology. 67:5640-5646.

36. Fridell RA, Partin KM, Carpenter S, Cullen BR. 1993. Identification of the activation domain of equine infectious anemia virus Rev. Journal of Virology 67:7317-7323.

37. Lee JH, Murphy SC, Belshan M, Sparks WO, Wannemuehler Y, Liu SJ, Hope TJ, Dobbs D, Carpenter S. 2006. Characterization of functional domains of equine infectious anemia virus Rev suggests a bipartite RNA-binding domain. Journal of Virology 80:3844-3852.

38. Iversen AKN, Shpaer EG, Rodrigo AG, Hirsch MS, Walker BD, Sheppard HW, Merigan TC, Mullins JI. 1995. Persistence of attenuated rev genes in a human immunodeficiency virus type 1-infected asymptomatic individual. Journal of Virology 69:5743-5753.

39. Ihm Y, Sparks WO, Lee JH, Cao HB, Carpenter S, Wang CZ, Ho KM, Dobbs D. 2009. Structural model of the Rev regulatory protein from equine infectious anemia virus. Plos One 4:e4178.

40. Umunnakwe CN, Loyd H, Cornik K, Chavez JR, Dobbs D, Carpenter S. In Press. Computational modeling suggests dimerization of equine infectious anemia virus Rev is required for RNA binding. Retrovirology.

41. Belshan M, Harris ME, Shoemaker AE, Hope TJ, Carpenter S. 1998. Biological characterization of Rev variation in equine infectious anemia virus. Journal of Virology 72:4421-4426.

42. Belshan M, Park GS, Bilodeau P, Stoltzfus CM, Carpenter S. 2000. Binding of equine infectious anemia virus Rev to an exon splicing enhancer mediates alternative splicing and nuclear export of viral mRNAs. Molecular and Cellular Biology 20:3550-3557.

43. Gontarek RR, Derse D. 1996. Interactions among SR proteins, an exonic splicing enhancer, and a lentivirus Rev protein regulate alternative splicing. Molecular and Cellular Biology 16:2325-2331.

44. Long JC, Caceres JF. 2009. The SR protein family of splicing factors: master regulators of gene expression. Biochemical Journal 417:15-27.

45. Park GS, Lee J, Dobbs D, Carpenter S. SF2/ASF can inhibit equine infectious anemia virus Rev-mediated nuclear export activity and productive virus replication.In preparation

46. Lee JH, Culver G, Carpenter S, Dobbs D. 2008. Analysis of the EIAV Rev-responsive element (RRE) reveals a conserved RNA motif required for high affinity Rev binding in both HIV-1 and EIAV. Plos One 3:e2272.

47. Na H, Huisman W, Ellestad KK, Phillips TR, Power C. 2010. Domain- and nucleotide- specific Rev response element regulation of feline immunodeficiency virus production. Virology 404:246-260.

48. Berchtold S, Hornung U, Aepinus C. 1995. The activation domain of simian immunodeficiency virus SIVmac239 Rev protein is structurally and functionally analogous to the HIV-1 Rev activation domain. Virology 211:285-289.

49. Pollard VW. 1998. The HIV-1 Rev protein. Annual Review of Microbiol 52:491-532.

50. Charpentier B, Stutz F, Rosbash M. 1997. A dynamic in vivo view of the HIV-1 Rev- RRE interaction. Journal of Molecular Biology. 266:950-962.

51. Malim MH, Hauber J, Le S-Y, Maizel JV, Cullen BR. 1989. The HIV-1 Rev trans- activator acts through a structured target sequence to activate nuclear export of unspliced viral mRNA. Nature 338:254-256.

CHAPTER 2. FUNCTIONAL ANALYSIS OF REV BINDING REGION 2, A STRUCTURAL ELEMENT OF THE EIAV RRE

Jerald Chavez, Hyelee Loyd, Drena Dobbs, Susan Carpenter

Abstract

Retrovirus replication depends on RNA-protein interactions that mediate key steps in the virus life cycle. In lentiviruses, nuclear export of incompletely spliced viral RNA requires interactions between the viral Rev protein and a viral sequence known as the Rev Responsive

Element (RRE). In equine infectious anemia virus (EIAV), Rev binds two regions in the RRE, designated Rev-binding region-1 (RBR-1) and Rev-binding region-2 (RBR-2). RBR-1 is sufficient to mediate RNA export and overlaps an exonic splicing enhancer that binds the alternative splicing factor SF2/ASF. RBR-2 is structurally similar to other lentiviral RRE structures required for export, and is required for EIAV Rev high affinity binding to the RNA; however, the functional significance of RBR-2 is unknown. Here, we used Rev-dependent nuclear export assays and a Rev dependent EIAV virus gene expression assay to assess the role of RBR-2 in Rev dependent nuclear export and virus gene expression. There was no significant difference in export activity with or without RBR-2, whereas deletion of RBR-1 resulted in complete loss of nuclear export. Deletion of RBR-2 also had no effect on SF2/ASF inhibition of

Rev activity in transient assays, suggesting RBR-2 does not provide a competitive advantage to

Rev in the presence of SF2/ASF. Further, deletion of RBR-2 from the Rev dependent Gag/Pol expression construct did not result in any significant changes in to EIAV virus gene expression or virion production. Together, these results indicate RBR-2 does not play a role in Rev mediated nuclear export or virion replication. 21

Introduction

Equine infectious anemia virus (EIAV) is a lentivirus that primarily infects tissue macrophages (1). EIAV induces a rapid and dynamic clinical disease course. Initial acute viremia occurs 5-30 days post infection, followed by a chronic period characterized by repeated cycles of fever, thrombocytopenia, anemia, and viremia over a period of two months to one year.

After, equids enter into the inapparent stage of disease, characterized by clinical quiescence and the development of a broadly acting immune response to control virus replication (reviewed in

(2, 3)). However, like all lentiviruses, EIAV causes a lifelong persistent infection, and some inapparent carriers can recrudesce years to decades later. Therefore, EIAV is an excellent model to study both the mechanisms of lentiviral persistence and factors involved in immunological control of the virus.

Due to their limited genome capacity, retroviruses utilize many strategies to maximize their coding capacity, including overlapping reading frames, ribosomal frameshifts, and alternative splicing. Lentiviruses translate the structural and enzymatic Gag, Pol, and Env proteins from unspliced and singly spliced mRNAs. Eukaryotes prevent release of incompletely spliced RNA through multiple forms of RNA surveillance including formation of retention and splicing complexes (RES) (4) and nonsense mediated decay (5, 6). Lentiviruses, encode the Rev protein for the export of unspliced/incompletely viral RNAs (vRNAs). Rev binds a cis-acting element in the viral RNAs known as the Rev responsive element (RRE)(7-9), multimerizes (10-

12), and tethers the RNA to the CRM1 nuclear export factor to mediate export of its cargo (13).

Secondary structure is a primary determinant of interaction between Rev and the RRE (14-16) as mutations that disrupt key structural features can significantly reduce Rev binding (17-19). Early 22 analysis of the HIV-1 RRE secondary structure indicated Stem Loop IIB (SLIIB) was the primary binding site for Rev, though more recent analysis suggests that a secondary site, Stem loop IA (SLIA) can bind with similar affinity to that of SLIIB (20). Interaction between Rev and the RRE can induce structural changes in both the RNA and protein (21-23).

EIAV Rev is functionally homologous to the HIV-1 Rev protein. EIAV Rev is a 165 amino acid (aa) protein, and possesses a unique structural domain organization, including an N- terminal nuclear export signal (NES) and a bipartite RNA binding domain (RBD) (24). The 555 nt EIAV RRE is located in the 5’region of envelope and overlaps with exon 1 of Rev. Two regions of the RRE have been shown to be structurally altered in the presence of Rev, Rev binding region 1 (RBR-1) and Rev binding region two (RBR-2) (Figure 1) (25). RBR-1 is located in Rev exon 1, lacks extensive secondary RNA structure, and contains a purine rich exon splicing enhancer (ESE) (26, 27) that overlaps a 57 nt sequence shown to be sufficient for export.

The ESE region of RBR-1 contains binding sites for both Rev and alternative splicing factors, and mutations within this region have been shown to decrease both splicing between the flanking exons and binding of Rev to the RRE (26-28). Unlike RBR-1, RBR-2 is located in the envelope intronic region between Rev exon 1 and 2, and contains predicted structural motifs found in several other lentiviruses (25). In addition, deletion of RBR-2 resulted in ~250 fold decrease in

Rev binding affinity for the RRE (25). The functional significance of RBR-2 is not known; however the structural conservation and requirement for high affinity binding suggests that RBR-

2 is involved in Rev mediated nuclear export. In this study, we explored the functional role of

RBR-2 in Rev-dependent nuclear export and virion production.

Methods and Materials

Cells

HEK293T cells were plated in 12 well plates or in 6 well plates in Dulbecco’s Modified

Eagles Medium (DMEM) supplemented with 10% Fetal Bovine Serum (FBS), Penicillin,

Streptomycin, and L-glutamine and were kept at 37C with 5% CO2. Cells were transfected 24 hours after seeding with plasmid DNA using TransIT-LT1 (Mirus). Media was changed 24 hours post transfection, and cells and supernatants were harvested 48 hours post transfection.

Plasmid construction

The plasmids pCH21SL, pERRE-1, pDM138 have been previously described (29, 30). pERRE-1B and pERRE-1C were constructed by PCR amplification of EIAV nt 5281 to 5624 and nt 5606 to 5791 with primers containing Cla I restriction sites. The 5’ pERRE-1B primer was GGATCGATGGTTTGATATATGGGAGTA and the 3’ pERRE-1B primer was

GGATCGATCCCTATATAATGTTGCTG (Cla I site underlined). The 5’ pERRE-1C primer was GGATCGATCAGCAACATTATATAGGG and the 3’ pERRE-1C primer was

GGATCGATCTCTCTATGATAAGCTTC. Amplified fragments were purified, digested, and ligated to Cla I digested pDM138 vector.

The EIAV provirus core plasmid, pEV53B, was previously described (31). To construct pEV53BRBR2, EIAV nt 2584-5586 (nt numbering based on CL22 genome, Genbank accession M87581.1) was PCR amplified with a forward primer upstream of a natural EcoRV site and a reverse primer containing a Hind III restriction site. The 5’ pEV53BRBR2 primer was ACAGCTTTCACTATTCCCTCC, and the 3’ pEV53BRBR2 primer was

ACCCAAGCTTCCCTATATAATGTTGCTG (Hind III site underlined). Amplified fragments were purified, digested, and ligated to EcoRV-HindIII digested pEV53B. 24

pSF2/ASF encodes the human SF2/ASF alternative splicing factor, and was previously described (27).

Transient CAT Assays

Chloramphenicol acetyltransferase (CAT) assays were carried out as previously described (29). Briefly, HEK293T cells were seeded in 12 well plates at 3 x 105 cells per well.

Cells were co-transfected 24 hr post seeding with 50ng pCH21SL (Rev) or pCDNA3.1(–), 100ng of pCH110 (-Gal), 230ng pUC18, and 100ng of pDM138 CAT based reporter plasmid (32). At

48 hr post transfection, cells were lysed by rapid freeze/thaw cycles in 0.2 ml 0.25M Tris buffer

(pH 7.5). -Gal assays were performed with 50ul lystate to normalize for transfection efficiency.

Normalized lysate volumes were assayed for CAT protein levels via a CAT ELISA (Roach # 11

363 727 001). Results were normalized to pERRE-1, and results represent at least two independent transfections done in triplicate.

SF2/ASF competition assays

HEK293T cells were seeded in 12 well plates at 3 x 105 cells per well. Cells were co- transfected 24hr post seeding with 50ng pCH21SL (Rev) or pCDNA3.1 – and 100ng of pCH100

(-Gal), varying amounts of pSF2/ASF, 100ng of pDM138 CAT based reporter plasmid (32), and enough pUC18 to normalize to 650 ug total DNA transfected per well. CAT assays were carried out as above. Competition was measured by calculating the inhibitory dose of pSF2/ASF necessary to reduce Rev activity to 50% (ID50) of that observed in the absence of pSF2/ASF.

Values were calculated using Graphpad Prism 6 software using a non-linear regression model.

Significance was calculated by an unpaired T Test using the PRISM software program (Version

6.01, Graphpad).

EIAV Gag Expression and Virion Production

HEK293T cells were seeded at 7 x 105 cells per well in 6 well plates. At 24 hr post seeding, cells were transfected with varying amounts of pEV53B, pEV53BRBR2, or pEV53BRev and pUC18 was added to total 1.6 ug of DNA per well. Cells and supernatants were collected 48 hr post transfection. Cells were lysed with RIPA buffer (150 mM NaCl, 10 mM NaPO4, 0.5% Triton X-100, 0.1% SDS, 0.5% SDS), and supernatants were clarified by centrifugation at 1000xg for 10min.

For pseudovirus production, cells were co-transfected 24 hr post seeding with 1.6 ug pEV53B, pEV53BRBR2, or pEV53BRev and 1.6 ug pSIN6.1Luc (reporter genome) and pcDNAENV (EIAV19 envelope) as previously described (33). Supernatants were collected 48 hr post transfection and were clarified by centrifugation at 1000xg for 10min.

Western Blots

Lysates and supernatants from transfected cells were harvested 48 hrs post transfection and 50ug of total protein from each lysate was resolved by SDS-PAGE (12% polyacrylamide) and transferred to polyvinylidene difluoride (PVDF) membranes. Membranes were probed with either mouse anti-Gag 6A-1(Washington State University) hybridoma supernatant or mouse anti-

-Actin (SIGMA A1978) primary antibodies, and a peroxidase conjugated sheep anti mouse secondary antibody (Jackson Immuno Research #515-0350062). Secondary antibodies were visualized following incubation ECL substrate (Pierce # 32209).

Virus-like particles (VLPs) were pelleted from clarified supernatants at 100,000 xg for

1hr, resuspended in 40ul 1xPBS, and analyzed by Western Blot as described above.

RNA isolation & qRT-PCR

For RNA isolation, 140ul of producer cell supernatant was DNase I treated using 1U of

DNase I in a reaction volume of 160 ul with 1X DNase I reaction buffer (Invitrogen Cat# 18068-

015). After incubation for 15 min at room temperature, the reaction was stopped by adding

EDTA to 2mM concentration in a final volume of 180 ul, and samples were heat treated at 65°C for 10min. A total of 180ul volumes of DNase treated supernatant was then used to isolate RNA using QIAamp Viral RNA mini kit (QIAGEN Cat# 52904) according to manufacturer’s protocols. RNA was eluted in 60 ul of elution buffer and 10 ul purified RNA was subjected to second DNAse I treatment and diluted to 100 ul with H2O.

Virion production was quantified by qRT-PCR for the firefly luciferase reporter RNA using RNA-to-Ct one step kit (Applied Biosystems # 4392938) according to manufacturer’s protocols. PCR reaction contained 8 ul of purified diluted RNA in 20ul volumes with the following primers and probes: 5’-TATGAAGAGATACGCCCTGGTT-3’, Reverse Primer : 5’-

GCCCATATCGTTTCATAGCTTC-3’, and FAM Probe : 5’-

FAM/TATGAAGAGATACGCCCTGGTT-3’. The initial reverse transcriptase step included one round of 48C for 15 min, 95C for 10min, followed by 40 cycles of 95C for 10sec, 57C for 50sec. the standard curve was calculated from serial ten-fold dilutions of firefly luciferase plasmid

DNA (108 to 102 copy/ul).

Results

RBR-2 is not essential for Rev-mediated nuclear export

RBR-2 was previously shown to be required for high affinity binding to the EIAV RRE and contains a predicted structural motif found in several other lentiviral genomes, suggesting 27 biological relevance (25). To determine if RBR-2 plays a role in Rev mediated nuclear export, we utilized a CAT based reporter assay (29). EIAV RRE sequences containing RBR-1 or RBR-2 were cloned into the pDM138 reporter plasmid (pERRE-1B or pERRE-1C respectively, Figure

2). Nuclear export activity was normalized to wild type pERRE-1, which contains both RBR-1 and RBR-2. Results indicated that nuclear export activity of pERRE-1B was equivalent to the activity of pERRE-1 (Figure 3A), indicating that deletion of RBR-2 does not diminish Rev export activity. In contrast pERRE-1C showed no export activity above the background of pDM138, demonstrating that RBR-2 alone is not able to mediate Rev-dependent nuclear export.

Together, these results indicate that RBR-2 is neither sufficient nor necessary for Rev export activity in transient assays.

To confirm these results in a more virus like system, we utilized the pEV53B lentivirus vector system (Olsen, 1998) to examine the effect of RBR-2 deletion on EIAV Gag expression using varying amounts of Rev and pEV53B vector. Gag expression gradually increased as the amount of transfected pEV53B increased from 400 to 1600 ng (Figure 3B). In comparison, p53BRBR2 Gag expression over the same titration range appeared to be identical to pEV53B, indicating deletion of RBR-2 did not alter Gag expression (Figure 3B). To confirm that Gag expression was Rev-dependent, we constructed pEV53BRev, in which the second exon of Rev is deleted. As expected, significant amounts of Gag were not detected in cells transfected with pEV53BRev, indicating that Gag expression from pEV53B-based plasmids is Rev-dependent.

Together, these results indicate deletion of RBR-2 from an EIAV-based vector had no effect on

Gag expression.

RBR-2 is not required for Virion Production

In addition to nuclear export, Rev has been suggested to have other roles in regulating virus gene expression, including translation and packaging (34). Therefore, we wished to determine if deletion of RBR-2 had any effect on virion production. Equal amounts of pEV53B or pEV53BRBR2 were transfected into 293T cells and both intracellular Gag expression and production of virus like particle (VLP) were analyzed using Western Blot (Figure 4a). Similar to the previous assays, deletion of RBR-2 had no effect on either intracellular Gag expression or

VLP production.

To further investigate if deletion of RBR-2 had any effect on packaging of viral genomes, we generated an EIAV pseudotyped reporter virus using the pEV53B and pSIN6.1Luc reporter plasmids (33). RNA was isolated from producer cell supernatants and luciferase RNA was quantified by qRT-PCR. Results indicated similar levels of packaged RNA in supernatant from cells transfected with pEV53B and pEV53BRBR2 (Figure 4B). pEV53BRev produced 1000 fold less RNA in comparison to pEV53B and pEV53BRBR2, and levels equivalent to the negative control in which pUC18 is substituted for pEV53B. Therefore, deletion of RBR-2 had no effect on packaging. Together, these studies indicate that RBR-2 does not play a role in late stages of EIAV replication.

Deletion of RBR-2 does not affect SF2/ASF mediated inhibition of Rev export activity

The EIAV RRE overlaps an ESE required for exon 3 inclusion in EIAV mRNA (27, 28).

Previous studies have shown that SF2/ASF can inhibit both Rev activity (27) and EIAV replication (35), and conversely Rev can inhibit SF2/ASF dependent splicing (26, 27). SF2/ASF may compete with Rev for binding to the RRE, as deletion of SF2/ASF RNA recognition motif 29

(RRM) abrogates inhibition of Rev activity and EIAV replication. Because RBR-2 is required for high affinity binding of Rev to the RRE, we hypothesized that RBR-2 may provide Rev a competitive advantage in the presence of SF2/ASF. To test this, we determined the effect of deleting RBR-2 on Rev export activity in the presence of increasing concentrations of SF2/ASF

(Figure 5). SF2/ASF inhibited 50% the Rev export activity using ERRE-1 (ID50) with 95ng plasmid, while deletion of RBR-2 resulted in non-significant 1.3 fold decrease to 77ng

(significance calculated by unpaired T Test), indicating RBR-2 does not provide Rev a competitive advantage for the RRE in the presence of SF2/ASF.

Discussion

Like other lentivirus RREs, RBR-2 is located an intronic region of the EIAV envelope gene and is required for Rev high affinity binding to the RRE. In addition, RBR-2 contains predicted structural motifs conserved across several other lentivirus (25). We hypothesized that

RBR-2 was a component of the EIAV RRE and played a role in Rev mediated nuclear export and subsequent viral structural gene expression. However, the results showed that RBR-2 is neither necessary nor sufficient for nuclear export in transient assays, and deletion of RBR-2 does not subsequently affect Gag gene expression or virus production. Therefore, the functional significance, if any, of RBR-2 is unknown.

Early models of the HIV-1 Rev-RRE interaction suggested that Stem loop IIB acted as the primary high binding site for Rev (18, 36, 37), and upon binding, initiated a nucleation event wherein Rev multimerized across the RRE. However, more recent analysis using natively purified Rev has identified Stem loop IA (SLIA) as a second high affinity site that can bind Rev on its own and support multimerization (38). SLIIB or SLIA alone do not support nuclear export

(38-40), suggesting that nuclear export requires additional sequences and/or structures from the 30 larger RRE acting in concert with these high affinity sites. Like the HIV-1 RRE, RBR-2 is located in the intronic region between Rev exon one and two. In addition, similar to SLIIB or

SLIA, RBR-2 is required for high affinity binding, and does not support Rev-mediated nuclear export on its own. However, the result that deletion of RBR-2 does not affect Rev export activity at all is surprising considering that deletion of this region results in a 250 fold loss in Rev binding affinity to the RRE. We had previously reported significant genetic variation in the Rev/TM overlapping reading frame (29, 30, 41), and demonstrated that variation could alter Rev nuclear export activity (29). Further, changes in Rev export phenotype correlated with changes in clinical stages of disease, suggesting that Rev export phenotype may contribute to variant selection in vivo (41, 42). Most of the variation in Rev occurred outside essential functional domains of Rev, and it is not clear how genetic changes alter Rev phenotype. Interestingly, variation in charged amino acids were commonly found in the non-essential region, located upstream of the second RNA binding domain. Many of these charged changes modulated nuclear export activity (30). The present studies were done using a single Rev genotype, and it is possible that charge changes in the Rev protein at or near the RNA binding domains of

Rev could alter nuclear export capacity by changing the strength of ionic interactions at the protein-RNA interface. If so, RBR-2 could be more important for Revs with lower RNA binding affinities. Further analysis of export activity with and without RBR-2 using variant Revs with charge changes is necessary to explore this hypothesis.

Acknowledgements

We thank Dr. Bob Mealey at Washington State University for his kind gift of the Anti-EIAV

Gag antibody hybridomas. This work was supported by the National Institute of Health grant

CA128568 and the Alliance for Graduate Education and the Professoriate (AGEP).

Figures

Figure 1. Two distinct regions of the ERRE undergo structural transitions in the presence of bound EIAV Rev protein. Consensus chemical modification patterns, based on at least 3 experiments in which several different primers were used to probe the complete ERRE RNA, mapped onto the RNA secondary structure. Ribonucleotides that consistently displayed enhanced modification with either kethoxal or DMS upon Rev binding are circled: bold circle (strong) and thin circle (mild). Regions protected from hydroxyl radical cleavage in the presence of Rev are denoted by a thick line. Purine-rich motifs are highlighted in green. Re-print of figure from (25). Shaded blue regions indicate general regions of structural alterations in the presence of Rev. The 56 nt MinRRE is outlined in yellow, with overlapping exon splicing enhancer position defined by dashed blue line.

Figure 2. A) Schematic of the pDM138 CAT reporter plasmid. X indicates location of multiple cloning site for insertion of RRE sequences. B) Schematic of RRE sequences inserted into the pDM138 reporter construct. pERRE-

1 contains the 555nt ERRE-1 sequence previously described (27). pERRE-1B has a deletion of RBR-2, while pERRE-1 has a deletion of RBR-1. Nucleotide numbers are based on the pFL85 proviral clone (43).

Figure 3. RBR-2 is not necessary for Rev nuclear export activity or Gag expression. A) Rev nuclear export activity using pDM138 CAT reporter constructs containing the indicated RRE sequences was assessed in transient transfection assays in HEK293T cells as previously described (29). Experiments were performed in triplicate, and results represent at least 2 independent transfections. Results for each RRE are normalized to pERRE-1. B) Gag expression in HEK293T cells transfected with pEV53B, pEV53BRBR2, or pEV53BRev. Lysates were collected

48 hr post transfection. Gag gene expression was measured via Western Blot using anti-Gag antibody or -Actin to confirm equal loading among samples. The site of EIAV gag gene products are indicated on the right. 34

Figure 4. RBR-2 is not required for virus like particle or virion production. A) Production of VLPs. 2ug of pEV53B, pEV53BRBR2, and pEV53BRev were transfected into HEK293T cells and harvested 48hrs post transfection.

Supernatants were clarified, and VLPs were sedimented by ultracentrifugation, and analyzed by Western Blot. B)

Production of pseudovirus. 1.6ug of pEV53B, pEV53BRBR2, and pEV53BRev were cotransfected with 0.8ug

EIAV envelope and 1.6ug of pLUC luciferace reporter genome. Virion particles were collected from supernatants, lysed and quantified by qRT-PCR. 35

Figure 5. RBR-2 does not provide Rev a competitive advantage for RRE in the presence of SF2/ASF. The activity of pERRE-1 and pERRE-1B were measured using the CAT transient transfection assay in the presence of increasing amounts of pSF2/ASF. Assays were performed in triplicate, and results represent at least 2 independent transfections.

References

1. Sellon DC, Perry ST, Coggins L, Fuller FJ. 1992. Wild-type equine infectious anemia virus replicates in vivo predominantly in tissue macrophages, not in peripheral blood monocytes. Journal of Virology 66:5906-5913.

2. Cheevers WP, McGuire TC. 1985. Equine infectious anemia virus: immunopathogenesis and persistence. Reviews of Infectious Disease 7:83-88.

3. Leroux C, Cadoré JL, Montelaro RC. 2004. Equine infectious anemia virus (EIAV): what has HIV's country cousin got to tell us? Veterinary Research 35:485-512.

4. Galy V, Gadal O, Fromont-Racine M, Romano A, Jacquier A, Nehrbass U. 2004. Nuclear retention of unspliced mRNAs in yeast is mediated by perinuclear Mlp1. Cell 116:63-73.

5. Chang Y-F, Imam JS, Wilkinson ME. 2007. The nonsense-mediated decay RNA surveillance pathway. Annual Review of Biochemistry 76:51-74.

6. Wen J, Brogna S. 2008. Nonsense-mediated mRNA decay. Biochemical Society Transactions 36:514-516.

7. Hadzopouloucladaras M, Felber BK, Cladaras C, Athanassopoulos A, Tse A, Pavlakis GN. 1989. The Rev (TRS ART) protein of human immunodeficiency virus type-1 affects viral messenger-RNA and protein expression via a cis-acting sequence in the env region. Journal of Virology 63:1265-1274.

8. Daly TJ, Cook KS, Gray GS, Maione TE, Rusche JR. 1989. Specific binding of HIV-1 recombinant rev protein to the Rev-responsive element invitro. Nature 342:816-819.

9. Pfeifer K, Weiler BE, Ugarkovic D, Bachmann M, Schroder HC, Muller WEG. 1991. Evidence for a direct interaction of Rev protein with nuclear-envelope messenger RNA- translocation system. European Journal of Biochemistry 199:53-64.

10. Malim MH, Cullen BR. 1991. HIV-1 structural gene-expression requires the binding of multiple Rev monomers to the viral RRE - implications for HIV-1 latency. Cell 65:241- 248.

11. Pond SJK, Ridgeway WK, Robertson R, Wang J, Millar DP. 2009. HIV-1 Rev protein assembles on viral RNA one molecule at a time. Proceedings of the National Academy of Sciences of the United States of America 106:1404-1408.

12. Daelemans D, Costes SV, Cho EH, Erwin-Cohen RA, Lockett S, Pavlakis GN. 2004. In vivo HIV-1 Rev multimerization in the nucleolus and cytoplasm identified by fluorescence resonance energy transfer. Journal of Biological Chemistry 279:50167- 50175. 37

13. Neville M, Stutz F, Lee L, Davis LI, Rosbash M. 1997. The importin-beta family member Crm1p bridges the interaction between Rev and the nuclear pore complex during nuclear export. Current Biology 7:767-775.

14. Olsen H, Cochrane A, Dillon P, Nalin C, Rosen C. 1990. Interaction of the human immunodeficiency virus type 1 rev protein with a structured region in env mRNA is dependent on multimer formation mediated through a basic stretch of amino acids. Genes & Development 4:1357-1364.

15. Heaphy S, Dingwall C, Ernberg I, Gait MJ, Green SM, Karn J, Lowe AD, Singh M, Skinner MA. 1990. HIV-1 regulator of virion expression (Rev) protein binds to an RNA stem-loop structure located within the Rev response element region. Cell 60:685-693.

16. Malim MH, Tiley LS, McCarn DF, Rusche JR, Hauber J, Cullen BR. 1990. HIV-1 structural gene expression requires binding of the Rev trans-activator to its RNA target sequence. Cell 60:675-683.

17. Olsen HS, Nelbock P, Cochrane AW, Rosen CA. 1990. Secondary structure is the major determinant for interaction of HIV Rev protein with RNA. Science 247:845-848.

18. Heaphy S, Finch JT, Gait MJ, Karn J, Singh M. 1991. Human-immunodeficiency-virus type-1 regulator of virion expression, Rev, forms nucleoprotein filaments after binding to a purine-rich bubble located within the Rev-responsive region of viral messenger-RNAs. Proceedings of the National Academy of Sciences of the United States of America 88:7366-7370.

19. Olsen HS, Beidas S, Dillon P, Rosen CA, Cochrane AW. 1991. Mutational analysis of the HIV-1 Rev protein and its target sequence, the Rev responsive element. Journal of Acquired Immune Deficiency Syndromes and Human Retrovirology 4:558-567.

20. Daugherty MD, Booth DS, Jayaraman B, Cheng Y, Frankel AD. 2010. HIV Rev response element (RRE) directs assembly of the Rev homooligomer into discrete asymmetric complexes. Proceedings of the National Academy of Sciences of the United States of America 107:12481-12486.

21. Kjems J, Calnan BJ, Frankel AD, Sharp PA. 1992. Specific binding of a basic peptide from HIV-1 Rev. Embo Journal 11:1119-1129.

22. Battiste JL, Tan R, Frankel AD, Williamson JR. 1994. Binding of an HIV Rev peptide to Rev responsive element RNA induces formation of purine-purine base pairs. Biochemistry 33:2741-2747.

23. Peterson RD, Feigon J. 1996. Structural change in Rev responsive element RNA of HIV- 1 on binding Rev peptide. Journal of Molecular Biology 264:863-877.

24. Lee JH, Murphy SC, Belshan M, Sparks WO, Wannemuehler Y, Liu SJ, Hope TJ, Dobbs D, Carpenter S. 2006. Characterization of functional domains of equine infectious anemia virus Rev suggests a bipartite RNA-binding domain. Journal of Virology 80:3844-3852.

25. Lee JH, Culver G, Carpenter S, Dobbs D. 2008. Analysis of the EIAV Rev-responsive element (RRE) reveals a conserved RNA motif required for high affinity Rev binding in both HIV-1 and EIAV. Plos One 3:e2272.

26. Gontarek RR, Derse D. 1996. Interactions among SR proteins, an exonic splicing enhancer, and a lentivirus Rev protein regulate alternative splicing. Molecular and Cellular Biology 16:2325-2331.

27. Belshan M, Park GS, Bilodeau P, Stoltzfus CM, Carpenter S. 2000. Binding of equine infectious anemia virus Rev to an exon splicing enhancer mediates alternative splicing and nuclear export of viral mRNAs. Molecular and Cellular Biology 20:3550-3557.

28. Chung HK, Derse D. 2001. Binding sites for Rev and ASF/SF2 map to a 55-nucleotide purine-rich exonic element in equine infectious anemia virus RNA. Journal of Biological Chemistry 276:18960-18967.

29. Belshan M, Harris ME, Shoemaker AE, Hope TJ, Carpenter S. 1998. Biological characterization of Rev variation in equine infectious anemia virus. Journal of Virology 72:4421-4426.

30. Sparks WO, Dorman KS, Liu SJ, Carpenter S. 2008. Naturally arising point mutations in non-essential domains of equine infectious anemia virus Rev alter Rev-dependent nuclear-export activity. Journal of General Virology 89:1043-1048.

31. Olsen JC. 1998. Gene transfer vectors derived from equine infectious anemia virus. Gene Therapy 5:1481-1487.

32. Hope TJ, Bond BL, McDonald D, Klein NP, Parslow TG. 1991. Effector domains of human immunodeficiency virus type 1 Rev and human T-cell leukemia virus type 1 Rex are functionally interchangeable and share an essential peptide motif. Journal of Virology 65:6001-6007.

33. Tallmadge RL, Brindley MA, Salmans J, Mealey RH, Maury W, Carpenter S. 2008. Development and characterization of an equine infectious anemia virus Env-pseudotyped reporter virus. Clinical and Vaccine Immunology 15:1138-1140.

34. Groom HCT, Anderson EC, Lever AM. 2009. Rev: beyond nuclear export. Journal of General Virology 90:1303-1318.

35. Park GS, Lee J, Dobbs D, Carpenter S. SF2/ASF can inhibit equine infectious anemia virus Rev-mediated nuclear export activity and productive virus replication.In preparation

36. Cook K, Sue , Fisk GJ, Hauber J, Usman N, Daly TJ, Rusche JR. 1991. Characterization of HIV-1 Rev protein: binding stoichiometry and minimal RNA substrate. Nucleic Acids Res. 19:1577-1583.

37. Huang X, Hope TJ, Bond BL, McDonald D, Grahl K, Parslow TG. 1991. Minimal Rev- response element for type 1 human immunodeficiency virus. Journal of Virology. 65:2131-2134.

38. Daugherty MD, D'Orso I, Frankel AD. 2008. A solution to limited genomic capacity: using adaptable binding surfaces to assemble the functional HIV Rev oligomer on RNA. Molecular Cell 31:824-834.

39. Kjems J, Sharp PA. 1993. The basic domain of Rev from human immunodeficiency virus type 1 specifically blocks the entry of U4/U6.U5 small nuclear ribonucleoprotein in spliceosome assembly. Journal of Virology 67:4769-4776.

40. Mann DA, Mikaelian I, Zemmel RW, Green SM, Lowe AD, Kimura T, Singh M, Butler PJG, Gait MJ, Karn J. 1994. A molecular rheostat: co-operative Rev binding to Stem I of the Rev-responsive element modulates human immunodeficiency virus type-1 late gene expression. Journal of Molecular Biology 241:193-207.

41. Belshan M, Baccam P, Oaks JL, Sponseller BA, Murphy SC, Cornette J, Carpenter S. 2001. Genetic and biological variation in equine infectious anemia virus Rev correlates with variable stages of clinical disease in an experimentally infected pony. Virology 279:185-200.

42. Carpenter S, Dobbs D. 2010. Molecular and biological characterization of equine infectious anemia virus Rev. Current HIV Research 8:87-93.

43. Martarano L, Stephens R, Rice N, Derse D. 1994. Equine infectious anemia virus trans- regulatory protein Rev controls viral mRNA stability, accumulation, and alternative splicing. Journal of Virology 68:3102-3111.

CHAPTER 3. GENERAL DISCUSSION

While the HIV-1 Rev protein and RRE have been extensively examined, the EIAV Rev-

RRE interaction has not been well studied. Recent chemical probing and footprinting analysis of the EIAV RRE indicated EIAV Rev bound two separate regions of the RRE, termed RBR-1 and

RBR-2 (1). RBR-1 contains sequences previously shown to be involved in Rev binding, nuclear export, and exon recognition (ESE) (2-4). The interaction of Rev with RBR-2 was previously functionally uncharacterized. Deletion of RBR-2 resulted in ~250 fold loss in Rev binding affinity for the RRE, and predicted structural motifs in RBR-2 are found in several other lentivirus genomes, suggesting biological relevance. We hypothesized that RBR-2 was involved in Rev mediated export and virion production, and would provide Rev a competitive advantage in competition with SF2/ASF. In contrast with this hypothesis, we found that deletion of RBR-2 had no effect on Rev mediated nuclear export or virion production. Rather, RBR-1 retained the almost all of the export activity of ERRE-1 (both RBR-1 and RBR-2). In addition, deletion of

RBR-2 had relatively little effect on Rev-SF2/ASF competition. The results are surprising considering that deletion of RBR-2 results in a 250 fold loss in Rev binding affinity to the RRE.

Early reports from studies of the HIV-1 RRE identified stemloop IIB as the Rev high affinity binding site (5-8), however, more recent studies identified stemloop IA (SLIA) as a second Rev binding site with similar binding affinity (9). While both sites are necessary to facilitate maximum nuclear export activity, neither isolated stem can facilitate nuclear export on their own (9), suggesting that the binding of both stems and other structures in the RRE are necessary to facilitate nuclear export. In addition to two different RRE binding sites, Rev utilizes different sets of amino acids in the arginine rich motif (ARM) to interact with SLIIB and SLIA

(9). RBR-2 of the EIAV RRE is required for Rev high affinity binding, contains structural stem 41 elements found in several other lentiviruses, and is located in the intron between the two coding exons of Rev, suggesting that this site, at the very least would be involved in, if not be required for nuclear export. Similar to SLIIB and SLIA, RBR-2 is unable to facilitate export independent of other cis-acting sequences in the RRE. In contrast to the comparison however, is that deletion of the entire RBR-2 sequence has no discernable effect on nuclear export or virion production, while mutation of SLIA or SLIID structure alone can significantly reduce nuclear export (9). The minRRE/ESE sequence in RBR-1 had previously been established to bind Rev and support export. However, compared to SLIB and SLIA, RBR-1 in the context of ERRE-1 maintains an unstructured RNA topology (1), more reminiscent of binding of a RNA recognition motif binding site (10). Because the minRRE also serves as an ESE (2-4), these structural constraints on the minRRE may be necessary to maintain the sequence/structure for binding to alternative splicing factors like SF2/ASF required inclusion of exon three of the bi-cistronic RNA necessary for translation of Rev. While it is still unclear what the molecular details of the EIAV Rev-RRE interactions are, the current data suggests that if RBR-2 does have a role in virus replication, it is independent of RBR-1. In addition, previous data suggests EIAV Rev dimerization is necessary for RNA binding, and that dimerization could result in two independent RNA interfaces on opposing sides of the dimer (11), which could interact with different sites on the RNA. However, even if this is the case, the interaction with either of the two possible interfaces and RBR-2 is not apparently required for nuclear export or virion production.

We previously reported that unique Rev genotypes can be isolated from different stages of EIAV infection (12), and that partition analysis indicated that subgroups or quasispecies of

Rev variants expanded and contracted during different stages of EAIV disease (13), suggesting that these Revs possessed a selective advantage that could facilitate immune evasion via 42 modulation of virus gene expression. In support of this, changes in Rev export phenotype correlated with changes in stage of disease (12). Natural variation in the hypervariable region of

Rev proximal to the second ARM does frequently change electrostatic charge in the area and can alter export phenotype (14). It is possible that this change in export activity could reflect a change in the charge attraction of the protein interface to the RNA. Our analysis of RBR-2 and nuclear export utilized a single Rev amino acid variant, H21. This variant exhibits relatively high levels of nuclear export activity (15), which could reflect higher binding affinity for the RRE.

Future Studies

The studies in this thesis attempt to functionally characterize the role of RBR-2 in EIAV

Rev nuclear export and Rev-SF2/ASF competition. The lack of effect on Rev export activity and virus gene expression by deletion of RBR-2 suggests RBR-2 does not serve a role in virus replication and that RBR-1 is the primary binding site necessary to support nuclear export.

However, these studies utilized only a single Rev variant, H21, associated with relatively high activity (15). The fact that multiple charge changes in Rev near the N-terminal ARM are associated with changes in Rev activity suggests that these charge changes could alter nuclear export activity by changing binding affinity of Rev to the RRE (Appendix 1). To test this hypothesis, it would need to be established if changes in charge in the Rev HVR correlate with changes in binding affinity via a filter binding or gel shift assays. In addition, if changes in binding affinity affect export capacity, then it would suggest that Rev variants with different affinities for the RRE would behave differently with and without RBR-2. Specifically, Rev variants with higher binding affinities may rely less on binding to RBR-2 for nuclear export and more on the primary binding site in RBR-1, whereas Rev variants with lower affinities would require RBR-2 interaction to support a threshold of activity. This could be tested by comparing 43 and correlating the various mutations that alter charge in the Rev protein with changes in binding affinity and nuclear export activity in the presence and absence of RBR-2. These studies would establish if there is a functional consequence of Rev RRE binding affinity and provide functional relevance and determine if RBR-2 is involved in nuclear export in the context of Rev variation.

The possibility that Rev charge changes might alter binding affinity of Rev may also have a drastic effect on Rev-SF2/ASF competition. Our current model suggests that SF2/ASF is functionally competitive with Rev for binding to the RRE/ESE region (16). If charge changes affect export activity through changes in binding affinity, Revs with lower RRE binding affinities may require RBR-2 in order to efficiently compete with SF2/ASF. Further studies need to compare Revs with different binding affinities in SF2/ASF competition assays to determine if these changes in affinity have a functional significance. Subsequent analysis would need to address if Rev variants with lower RRE affinities require RBR-2 for maximum competition efficiency with SF2/ASF.

References

1. Lee JH, Culver G, Carpenter S, Dobbs D. 2008. Analysis of the EIAV Rev-responsive element (RRE) reveals a conserved RNA motif required for high affinity rev binding in both HIV-1 and EIAV. Plos One 3:e2272.

2. Belshan M, Park GS, Bilodeau P, Stoltzfus CM, Carpenter S. 2000. Binding of equine infectious anemia virus Rev to an exon splicing enhancer mediates alternative splicing and nuclear export of viral mRNAs. Molecular and Cellular Biology 20:3550-3557.

3. Chung HK, Derse D. 2001. Binding sites for Rev and ASF/SF2 map to a 55-nucleotide purine-rich exonic element in equine infectious anemia virus RNA. Journal of Biological Chemistry 276:18960-18967. 4. Gontarek RR, Derse D. 1996. Interactions among SR proteins, an exonic splicing enhancer, and a lentivirus Rev protein regulate alternative splicing. Molecular and Cellular Biology 16:2325-2331. 44

5. Cook K, Sue , Fisk GJ, Hauber J, Usman N, Daly TJ, Rusche JR. 1991. Characterization of HIV-1 Rev protein: binding stoichiometry and minimal RNA substrate. Nucleic Acids Res. 19:1577-1583.

6. Malim MH, Tiley LS, McCarn DF, Rusche JR, Hauber J, Cullen BR. 1990. HIV-1 structural gene expression requires binding of the Rev trans-activator to its RNA target sequence. Cell 60:675-683.

7. Heaphy S, Finch JT, Gait MJ, Karn J, Singh M. 1991. Human-immunodeficiency-virus type-1 regulator of virion expression, Rev, forms nucleoprotein filaments after binding to a purine-rich bubble located within the Rev-responsive region of viral messenger-RNAs. Proceedings of the National Academy of Sciences of the United States of America 88:7366-7370.

8. Huang X, Hope TJ, Bond BL, McDonald D, Grahl K, Parslow TG. 1991. Minimal Rev- response element for type 1 human immunodeficiency virus. Journal of Virology 65:2131-2134.

9. Daugherty MD, D'Orso I, Frankel AD. 2008. A solution to limited genomic capacity: Using adaptable binding surfaces to assemble the functional HIV Rev oligomer on RNA. Molecular Cell 31:824-834.

10. Messias AC, Sattler M. 2004. Structural basis of single-stranded RNA recognition. Accounts of Chemical Research 37:279-287.

11. Umunnakwe CN, Loyd H, Cornik K, Chavez JR, Dobbs D, Carpenter S. In Press. Computational Modeling Suggests Dimerization of Equine Infectious Anemia Virus Rev is Required for RNA Binding Retrovirology.

12. Belshan M, Baccam P, Oaks JL, Sponseller BA, Murphy SC, Cornette J, Carpenter S. 2001. Genetic and biological variation in equine infectious anemia virus rev correlates with variable stages of clinical disease in an experimentally infected pony. Virology 279:185-200.

13. Baccam P, Thompson RJ, Li Y, Sparks WO, Belshan M, Dorman KS, Wannemuehler Y, Oaks JL, Cornette JL, Carpenter S. 2003. Subpopulations of equine infectious anemia virus Rev coexist in vivo and differ in phenotype. Journal of Virology 77:12122-12131.

14. Sparks WO, Dorman KS, Liu SJ, Carpenter S. 2008. Naturally arising point mutations in non-essential domains of equine infectious anemia virus Rev alter Rev-dependent nuclear-export activity. Journal of General Virology 89:1043-1048. 15. Belshan M, Harris ME, Shoemaker AE, Hope TJ, Carpenter S. 1998. Biological characterization of Rev variation in equine infectious anemia virus. Journal of Virology 72:4421-4426.

16. Park GS, Lee J, Dobbs D, Carpenter S. SF2/ASF can inhibit equine infectious anemia virus Rev-mediated nuclear export activity and productive virus replication.In preparation.

APPENDIX. EFFECT OF GENETIC VARIATION ON PREDICTED RNA BINDING INTERFACE OF REV PROTEIN

Activity of R1 Phylogenetic Group Rev Variant NES 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 100% Clade B R1 D P Q G P L E S D Q W C R V L R Q S L P E E K I S S Q T C I A R R H L G P G P T Q H T P S 95% G134D ------+ - - + ------+ - + - + + + - - + + - - + + - + + + + + + + 100% R143H ------+ - - + ------+ - + - + + + - - + + - - + + - + + + + + + + N/A AADAA ------+ - - + ------+ - + - + + + - - + + - - + + - + + + + + - + 95% Clade B R2 ------+ - - + ------L ------+ + - - + + - + + + + + + + 200% G110D ------+ - - + ------+ - + - + + + - - + + - - + + - + + + + + + + 55% Clade B R26 ------+ - - + ------+ - + - + + + - - + + - - + + - + + + + + + + 50% Clade B R103 ------+ - - + ------L ------+ + - - - + D + + + + + + + 160% Clade A R32 ------+ - - + ------+ - + - + + + - - + + - - + + - + + + + + + + 125% Clade A R17 ------+ - - + ------+ - + - + + + - - + + - - + + - + + + + + + + 220% Clade A R4 ------+ - - + ------L ------+ + - - + + - + + + + + + + 190% Clade A R45 ------+ - - + ------+ - + - + + + - - + + - - + + - + + + + + + + 165% Clade A R71 ------+ - - + ------+ - + - + + + - - + + - - + + - + + + + + + + 200% H21 ------+ - - + ------+ - P + - + + - - + + - - + + - + + + + + + + 160% Clade A R93 ------+ - - + ------+ - P + - + + - - + + - - + + - + + + + + + + 150% D135G ------+ - - + ------+ - + - + + + - - + + - - + + - + + + + + + + 175% Clade B R51 ------+ - - + ------+ - + - + + + - - + + - - + + - + + + + + + + 160% Clade A R72 ------+ - - + ------+ - P + - + + - - + + - - + + - + + + + + + + 180% Q138R ------+ - - + ------+ - + - + + + - - + + - - + + - + + + + + + + 150% Clade A R42 ------+ - - + ------+ - + - + + + - - + + - - + + - + + + + + + + 160% Clade A R12 ------+ - - + ------+ - + - + + + - - + + - - + + - + + + + + + + 165% Clade A R53 ------+ - - + ------+ - + - + + + - - + + - - + + - + + + + + + +

Activity of R1 Phylogenetic Group Rev Variant RBD1 Predicted Coil-Coil 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 100% Clade B R1 R R D R W I R E Q I L Q A E V L Q E R L E W R I R G V Q Q V A K E L G E V N R G I W R E L 95% G134D + + + + + - + ------+ - - + ------+ - - - + - - 100% R143H + + + + + - + ------+ - - + ------+ - - - + - - N/A AADAA A A + A A - + ------+ - - + ------+ - - - + - - 95% Clade B R2 + + + + + - + ------+ - - + ------+ - - - + - - 200% G110D + + + + + - + ------+ - - + ------D - - - + - - - + - - 55% Clade B R26 + + - + + - G ------+ - - + ------+ - - - + - - 50% Clade B R103 + + + + + - + ------+ - - + ------+ - - - + - - 160% Clade A R32 + + + + + - + ------+ - - + - - - + - - - - A - + - - - + - - 125% Clade A R17 + + + + + - + ------+ - - + ------+ - - - + - - 220% Clade A R4 + + + + + - + ------+ - - + ------+ - - - + - - 190% Clade A R45 + + + + + - + ------+ - - + ------G - - + - - - + - - 165% Clade A R71 + + + + + - + ------+ - - + - A - - - - D - - - + - - - + - - 200% H21 + + + + + - + G ------+ - - + ------+ - - - + - - 160% Clade A R93 + + + + + - + ------+ - - + - A - - - - D - - - + - - - + - - 150% D135G + + + + + - + ------+ - - + ------+ - - - + - - 175% Clade B R51 + + + + + - + ------+ - - + ------G - - + - - - + - - 160% Clade A R72 + + + + + - + ------+ - - + ------D - - - + - - - + - - 180% Q138R + + + + + - + ------+ - - + ------+ - - - + - - 150% Clade A R42 + + + + + - + ------+ - - + - - - + - - - - A - + - - - + - - 160% Clade A R12 + + + + + - + ------+ - - + ------+ - - - + - - 165% Clade A R53 + + + + + - + ------+ - - + ------+ - - - + - -

Activity of R1 Phylogenetic Group Rev Variant Non-Essential RBD2/NLS 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 145 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 100% Clade B R1 H F R E D Q R G D F S A W G D Y Q Q A Q E R R W G E Q S S P R V L R P G D S K R R R K H L 95% G134D - - + - - - + ------D - - - - - + - + + + + + + + + + + - - + + + + + + + + + + + - 100% R143H - - + - - - + ------+ - H + - - + + + + + + - - + + + + + + + + + + + - N/A AADAA - - + - - - + ------+ - + - + + + + + + + + + + - - + + + + + + + + + + + - 95% Clade B R2 - - + - - - + ------+ - + - + + + + + + + + + + - - + + + + + + + + + + + - 200% G110D - - + - - - + ------+ - + - + + + + + + + + + + - - + + + + + + + + + + + - 55% Clade B R26 - - + - - - + ------+ - + - + + + + + + + + + + - - + + + + + + + + + + + - 50% Clade B R103 - - + - - - + ------+ - + - + + + + + + + + + + - - + + + + + + + + + + + - 160% Clade A R32 - - + - - - + ------D G + - R - + + + + + + + + + + + + - - + + + + + + + + + + + - 125% Clade A R17 - - + - - - + ------D G + - R - + + + + + + + + + + + + - - + + + + + + + + + + + - 220% Clade A R4 - - + - - - + ------D G + - R - + + + + + + + + + + + + - - + + + + + + + + + + + - 190% Clade A R45 - - + - - - + ------D G + - R - + + + + + + + + + + + + - - + + + + + + + + + + + - 165% Clade A R71 - - + - - - K ------D G + - R - + + + + + + + + + + + + - - + + + + + + + + + + + - 200% H21 Y - + - - - + - - - - - + - G + + R - + - + L - - - + + + + + - - + + + + + + + + + + + - 160% Clade A R93 - - + - - - K - - - - - + - G + + R - + - + H - - + + + + + + - - + + + + + + + + + + + - 150% D135G - - + - - - + - - - - - + - G + + + - + - + + + + + + + + + + - - + + + + + + + + + + + - 175% Clade B R51 - - + - - - + - - - - - + - G + + + - + - + + + + + + + + + + - - + + + + + + + + + + + - 160% Clade A R72 - - + - - - K - - - - - + - - + + R - + + + + + + + + + + + + - - + + + + + + + + + + + - 180% Q138R - - + - - - + - - - - - + - - + + R - + + + + + + + + + + + + - - + + + + + + + + + + + - 150% Clade A R42 - - + - - - + - - - - - + - G + + R + + + + + + + + + + + + + - - + + + + + + + + + + + - 160% Clade A R12 - - + - - - + - - - - - + - G + + R + + + + + + + + + + + + + - - + + + + + + + + + + - 165% Clade A R53 - - + - - - K - - - - - + - G + + R + + + + + + + + + + + + + - - + + + + + + + + + + + -

Figure A1. Effect of genetic variation on predicted RNA binding interface of Rev protein. Rev Amino acid sequences (exon 2 only) were uploaded to Bind N Server (1) for RNA interface predictions. Dots represent identical residues to R1 Rev variant. Green colored residues are predicted not to bind to RNA, red colored residues are predicted to bind RNA. Rev export activity percentages are relative R1 variant and are based on previous data

(2,3,4). Phylogenetic groups are based on (5). Domain abbreviation: NES = nuclear export signal, RBD = RNA binding domain, NLS = nuclear localization signal. 47

References

1. Wang L and Brown SJ. 2006.BindN: a web-based tool for efficient prediction of DNA and RNA binding sites in amino acid sequences. Nucleic Acids Research 44:W243- W248.

2. Belshan M, Harris ME, Shoemaker AE, Hope TJ, Carpenter S. 1998. Biological characterization of Rev variation in equine infectious anemia virus. Journal of Virology 72:4421-4426.

3. Belshan M, Baccam P, Oaks JL, Sponseller BA, Murphy SC, Cornette J, Carpenter S. 2001. Genetic and biological variation in equine infectious anemia virus Rev correlates with variable stages of clinical disease in an experimentally infected pony. Virology 279:185-200.

4. Sparks WO, Dorman KS, Liu SJ, Carpenter S. 2008. Naturally arising point mutations in non-essential domains of equine infectious anemia virus Rev alter Rev-dependent nuclear-export activity. Journal of General Virology 89:1043-1048.

5. Baccam P, Thompson RJ, Li Y, Sparks WO, Belshan M, Dorman KS, Wannemuehler Y, Oaks JL, Cornette JL, Carpenter S. 2003. Subpopulations of equine infectious anemia virus Rev coexist in vivo and differ in phenotype. Journal of Virology 77:12122-12131.