Equine Infectious Anemia Virus (EIAV): what has HIV’s country cousin got to tell us? Caroline Leroux, Jean-Luc Cadoré, Ronald Montelaro

To cite this version:

Caroline Leroux, Jean-Luc Cadoré, Ronald Montelaro. Equine Infectious Anemia Virus (EIAV): what has HIV’s country cousin got to tell us?. Veterinary Research, BioMed Central, 2004, 35 (4), pp.485- 512. ￿10.1051/vetres:2004020￿. ￿hal-00902790￿

HAL Id: hal-00902790 https://hal.archives-ouvertes.fr/hal-00902790 Submitted on 1 Jan 2004

HAL is a multi-disciplinary open access L’archive ouverte pluridisciplinaire HAL, est archive for the deposit and dissemination of sci- destinée au dépôt et à la diffusion de documents entific research documents, whether they are pub- scientifiques de niveau recherche, publiés ou non, lished or not. The documents may come from émanant des établissements d’enseignement et de teaching and research institutions in France or recherche français ou étrangers, des laboratoires abroad, or from public or private research centers. publics ou privés. Copyright Vet. Res. 35 (2004) 485–512 485 © INRA, EDP Sciences, 2004 DOI: 10.1051/vetres:2004020 Review article

Equine Infectious Anemia Virus (EIAV): what has HIV’s country cousin got to tell us?

Caroline LEROUXa*, Jean-Luc CADORÉa,b, Ronald C. MONTELAROc

a UMR754 INRA-UCBL-ENVL “Rétrovirus et Pathologie Comparée”, IFR128, Université Claude Bernard Lyon I, 50 avenue Tony Garnier, 69007 Lyon, France b Département Hippique, Médecine interne École Nationale Vétérinaire de Lyon, France c Department of Molecular Genetics and Biochemistry, Department of Infectious Diseases and Microbiology, University of Pittsburgh, Pennsylvania, USA

(Received 16 December 2003; accepted 2 March 2004)

Abstract – Equine Infectious Anemia Virus (EIAV) is a lentivirus, of the Retrovirus family, with an almost worldwide distribution, infecting equids. It causes a persistent infection characterized by recurring febrile episodes associating viremia, fever, thrombocytopenia, and wasting symptoms. The disease is experimentally reproducible by inoculation of Shetland ponies or horses with EIAV pathogenic strains. Among lentiviruses, EIAV is unique in that, despite a rapid virus replication and antigenic variation, most animals progress from a chronic stage characterized by recurring peaks of viremia and fever to an asymptomatic stage of infection. The inapparent carriers remain infective for life, as demonstrated by experimental transfer of blood to naive animals. The understanding of the correlates of this immune control is of great interest in defining vaccine strategies. Research on EIAV, this “country cousin” of HIV (Human Immunodeficiency Virus), over the last five decades has produced some interesting results on natural immunological control of lentivirus replication and disease and on the nature and role of virus variation in persistence and pathogenesis. These studies are of interest in the context of HIV and efforts to develop a vaccine. This review will focus on some of the most recent results. equine infectious anemia virus / lentivirus / equids / vaccine / viral evolution

Table of contents 1. Introduction...... 486 2. EIAV genome: genetic structure and viral protein function...... 486 2.1. The structural proteins of EIAV...... 486 2.2. Function of the EIAV LTR (Long Terminal Repeat) ...... 488 2.3. Tat, Rev and S2: the accessory proteins of EIAV...... 490 3. Clinical features of the EIAV-induced disease...... 491 3.1. Control of EIAV infection ...... 491 3.2. EIA, a cyclic disease ...... 492 3.3. Physiopathology...... 493 3.4. Cell tropism and strains of virus ...... 494 4. Immune control of EIAV infection and replication...... 496 5. Viral evolution in infected animals...... 499

* Corresponding author: [email protected] 486 C. Leroux et al.

6. EIAV vaccine development...... 502 7. Conclusion...... 503

1. INTRODUCTION teins encoded by gag, pol and env, the EIAV genome contains three open reading Equine Infectious Anemia Virus (EIAV) frames encoding the Tat and Rev proteins is an RNA virus, a member of the Retroviri- generally present in lentiviruses and the dae family and of the lentivirus genus [179], S2 protein (Figs. 1A and 1B). infecting equids. Equine Infectious Anemia (EIA) was first described in France in 1843 by Ligné [92] and was associated with 2.1. The structural proteins of EIAV infection with a “filterable agent” in 1904 The gag and pol gene products are trans- [178]. This makes EIA the first animal dis- lated from the full-length viral messenger ease to be assigned a viral etiology, preced- RNA [122]. Synthesis of Gag-Pol polypro- ing by several years the major discovery of teins requires ribosomes to shift their trans- the first tumor virus by Rous [149]. EIAV lational reading frame to read through the is mechanically transmitted by insect vec- stop codon in gag. Several critical motifs tors [40, 41] or unsterile needles. The main including an AAA AAAC slippery sequence, route of transmission is by hematophagous a 5-base GC-paired segment downstream of insects of the Tabanidae family [34, 57, 64]. the slippery sequence and a pseudoknot The virus is carried on the mouthparts of the structure, affect EIAV Gag-Pol frameshift- horsefly. The receptor for EIAV on the cel- ing [19]. The assembly of Gag precursor lular membrane remains unknown. proteins on the plasma membrane is essen- For years, knowledge of the molecular tial for virus budding from the host cells. biology of EIAV has been retarded by the The EIAV Gag-precursor (Pr55gag) poly- lack of a tissue culture system. Develop- protein is cleaved by viral protease into four ment of in vitro systems [99] and produc- major internal structural proteins of the tion of viral particles led to the classification mature virion: the membrane-interacting of EIAV as a member of the Retroviridae matrix (MA) p15, the capsid (CA) p26, the family [18] and opened the possibility of RNA-binding nucleocapsid (NC) proteins biochemical and molecular studies. p11 and p9 [62, 163] (Fig. 1A). After pro- Perhaps the most intriguing feature of teolytic processing of Gag, MA remains EIAV infection is the ability of most ani- associated with the inner face of the viral mals eventually to control the viral replica- membrane, and CA condenses to form a tion. Unlike other retroviruses, long-term shell around the NC/RNA complex (for infection is associated with limited virus review [42]). The crystal structure of the replication and the absence of clinical man- matrix protein has recently been resolved ifestations. The immune mechanisms respon- [56]. Interestingly, despite the lack of sible for this control may have considerable sequence homology, there is a striking implications for vaccine design in other len- overall structure similarity with HIV-1 and tiviral infections. SIV (Simian Immunodeficiency Virus) MA, except that it does not exhibit the same 2. EIAV GENOME: GENETIC trimeric association [56]. The EIAV MA is STRUCTURE AND VIRAL compacted such that the N and C termini are PROTEIN FUNCTION in close proximity [56]. The Gag polypro- tein drives assembly and budding of retro- At 8.2 kb, the EIAV RNA genome is the viruses and small regions, namely L (Late) smallest and genetically simplest lentiviral domains, mediates its ability to bind to the genome. In addition to the structural pro- cell membrane and induces the final step of EIAV infection in equids 487 is described e; RT: reverse ed organization ed organization of the surface glycoprotein gp90 ons. LTR: Long Terminal repeat; MA: matrix; CA: capsid; NC: nucleocapsid; PR: proteas nucleocapsid; NC: CA: capsid; matrix; MA: repeat; Terminal Long ons. LTR: ase; SU: surface; TM: transmembrane. AV virion (A) and proviral genome (B). The detail proviral genome (B). The (A) and AV virion Schematic structure of the EI Figure 1. Figure (C) with the position of the variable regi (C) with the position of transcriptase; IN: integr DU: dUTPase; transcriptase; 488 C. Leroux et al. separation of the nascent viral particle processing of the polyproteins, a dUTPase (reviewed in [168]) (Fig. 2). EIAV P9 is the (p15) essential for EIAV replication in non equivalent of P6 of HIV and P2b of RSV dividing monocyte-derived macrophages, (Rous Sarcoma Virus), two proteins involved the natural host cells for EIAV expression in particle budding from the cell [127]. Ret- and replication [89, 162, 174], and an inte- roviral L domains have been mapped to grase. 3 motifs PT/SAP, PPXY or YPXL (reviewed The general organization of the env gene in [43]). EIAV is unique among the retro- is similar to that described for other retro- viruses studied to date because its L domain viruses. It codes for the surface (gp90) and is a YPDL motif located in the C-terminus transmembrane (gp45) glycoproteins, incor- region of P9 [135]. Budding of the virus porated into the virus envelope (Figs. 1A from the host cell requires membrane fis- and 2). Gp90 may interact with the cellular sion. EIAV, like HIV or MoMLV (Moloney receptor, still unknown for EIAV. EIAV Murine Leukemia Virus), requires compo- gp90 is highly glycosylated; based on the nents of the cellular vacuolar protein sorting deduced amino-acid consensus sequence of (VPS) machinery for efficient release [170, EIAVPV, 17 and 5 potential N-linked glyc- 182]. EIAV P9 interacts via its YPDL motif osylation sites (NX[S/T]) are present in with the cellular AP2 clathrin-associated gp90 and gp45 respectively [81]. During adapter protein complex [135, 136], but a the course of disease, the gp90-surface definite role for AP2 in EIAV budding glycoprotein is subjected to rapid evolu- remains to be established. AIP1/ALIX, is a tion. We showed that the mutations are binding partner of EIAV p9 [101, 166]. restricted to defined variable regions, V1 to Class E VPS proteins are involved in mul- V8, in gp90 [81] (Fig. 1C). We will discuss tivesicular body (MVB) biosynthesis [68, this point in detail later in this review. 182]. AIP1/ALIX, a mammalian orthologue of the yeast class E VPS protein Bro1, plays a key role in the MVB network by linking 2.2. Function of the EIAV LTR (Long complexes acting in the early and late stages Terminal Repeat) of the pathway. Two recent studies [166] Lentiviral Long Terminal Repeats (LTR) suggest that AIP1/ALIX may couple HIV-1 (Fig. 1A) serve as the site of transcriptional P6 and EIAV P9 to the late-acting endo- initiation; they contain three segments namely somal sorting complex ESCRT-III (Endo- U3 (unique, 3’ end), R (Repeated) and U5 somal Sorting Complexes Required for (unique, 5’ end). The U3 region contains Transport-III) [101, 166]. Interestingly, muta- several elements important for viral tran- tional analysis in the context of the infec- scription. The lentivirus promoters are reg- tious molecular clone EIAVUK revealed ulated by various cellular DNA-binding that only the N-terminal 31 amino acids of factors interacting with specific motifs on the total 51 residues are required to main- the LTR sequence. EIAV LTR contains tain replication in transfected cells and that cis-acting DNA elements corresponding to p9 is not absolutely required for efficient the methylated DNA-binding protein site budding in the context of proviral gene MDBP, two PEA2 elements, a PEA3/ets expression, suggesting that other EIAV pro- motif and an AP-1 site [16]. The EIAV pro- teins can mediate this function [20]. moter is positively regulated by PU.1 Cleavage of the EIAV Gag-Pol precur- (purine-rich element 1) recognizing the ets sor (Pr180gag/pol) yields the pol gene prod- element [17]. PU.1 expression is normally ucts (Figs. 1A and 1B), which produce var- restricted to the haematopoietic system and ious enzymatic activities including the is absolutely required for the generation of reverse transcriptase-RNaseH (p66) essen- B-lymphocytes and macrophages (reviewed tial for the conversion of viral RNA into in [93]), the natural target of EIAV infec- DNA (Fig. 2), a viral protease (p12) for the tion. EIAV infection in equids 489 Retroviral cycle.Schematic Figure 2. representation of the EIAV viral cycle the host cells. into 490 C. Leroux et al.

2.3. Tat, Rev and S2: the accessory mechanism [1, 9, 167]. Interestingly, recent proteins of EIAV studies on the initially named Tat protein of Small Ruminant Lentiviruses (SRLV), infect- The EIAV genome contains only three ing goats and sheep, showed that this pro- additional open reading frames with the tein has a minimal effect on the transcrip- capacity to encode additional small acces- tional level from the LTR [181]. In fact, the sory proteins (Fig. 1B). This is in contrast gene initially designed as the tat gene may with the seven additional genes contained encode a Vpr-like protein, suggesting that in the genome of the primate lentiviruses SRLV differ from the other lentiviruses in (reviewed in [173]). Functions have been their control of expression from the viral assigned to the initially designated S1, S2 LTR [180]. and S3 proteins: S1 is a transactivator pro- tein similar to HIV Tat, S2 is an accessory The S2 gene overlaps the N-terminus protein unrelated to other known lentiviral of the envelope coding region [153]. The proteins and S3 corresponds to Rev. The small S2 protein (65 amino acids) contains basal transcriptional activity from the len- 3 amino-acid motifs highly conserved dur- tiviral LTR is usually low, with the activa- ing persistent infection: a putative nucleop- tion being under the control of the Tat orin motif (GLFG), a putative SH3 domain (Transactivator of Transcription) protein. binding motif (PXXP) and a putative nuclear A Tat function has been localized in a localization domain (RRKQETKK) [86]. region between the pol and env genes of S2 is expressed during EIAV infection in EIAV (Fig. 1B) [36, 158]. The structural horses as shown by detection of antibodies analyses revealed a well defined hydropho- directed against S2 in experimentally and naturally infected animals [153]. S2 is a bic core and two flexible NH2 and COOH terminal regions [165, 185]. EIAV Tat cytoplasmic protein potentially interacting interacts with the viral LTR through the with Gag but not incorporated into the trans-activation responsive (TAR) element. EIAV particles [186]. The role of S2 remains The secondary structure of TAR is essential unclear but mutational studies of this gene for Tat action; the TAR RNA-stem loop is in the context of EIAVUK, a pathogenic closed by two U*G base pairs and the con- molecular clone [28], revealed that S2 is servation of the nucleotide sequence of the dispensable for viral replication in vitro in loop is essential [15, 59, 60]. As for HIV, it several equine cell lines and in monocyte- is clear that the transcription of EIAV DNA derived macrophages [85], but is an impor- in the cell is controlled by the interaction tant determinant of viral replication and between the viral TAR sequence and the virulence during in vivo infection [86]. Inter- viral Tat protein. Through its interaction estingly, the S2 open reading frame is absent with TAR, HIV Tat recruits transcriptional in V26, an attenuated EIAV, obtained by complexes to the viral promoter, including serial passage in primary horse macro- enzymes with histone and acetyl transferase phages [76, 191]. factor activity and the positive transcription Complex molecular mechanisms allow elongation factor b (P-TEFb) complex. The differential expression of viral protein from P-TEFb complex enhances the elongation the relatively small lentiviral genomes. One activity of the RNA polymerase II complex strategy used by lentiviruses is the alterna- and is minimally composed of the CDK9 tive RNA splicing, leading to the produc- kinase and cyclin T1 (reviewed in [11, 88]). tion of various mRNA from a single RNA A similar mechanism involving cyclin T1 precursor. This strategy requires sub-opti- has been described in the context of the mal splice sites allowing the production of EIAV Tat and TAR interaction suggesting several mRNA from a single pre-RNA. that highly divergent Tat and TAR elements Transcription from the EIAV LTR leads to control viral gene expression by the same the production of full-length genomic and EIAV infection in equids 491 singly spliced mRNA encoding the major quarantine for the rest of their life, depend- Gag, Pol and Env proteins and of small ing on local regulations. The use of this test multi-spliced transcripts encoding the Tat, allowed the efficient control of EIAV infec- S2 and Rev proteins [140, 164]. Most cel- tion in the United States. Between 1972 and lular mRNA are fully spliced before export 1984, the number of AGID-positive ani- into the cytoplasm, the need for unspliced mals decreased from 3.9% to 0.38% [130]. or partially spliced RNA has to be satisfied The major drawback of this test is that it is in lentivirus-infected cells. Rev (regulator time consuming and sometimes difficult to of expression of viral proteins) acts post- interpret. Different tests based on a compet- transcriptionally by allowing the essential itive or non-competitive enzyme-linked translocation of unspliced and partially immunosorbent assay (ELISA) have been spliced RNA from the nucleus to the cyto- developed and compared to the “Coggins plasm. In this context, functional Rev is absolutely required for expression of struc- test” [12, 37, 84, 160]. Recently, an EIAV tural proteins [100] and virus production. diagnostic assay based on fluorescence EIAV Rev mediates mRNA transport from polarization (FP) using a peptide derived the nucleus to the cytoplasm through a Rev- from the gp45 transmembrane glycoprotein responsive element (RRE) located near the has been developed [172]. This technology end of the env gene [6, 100]. A recent study is based on a change in fluorescence emis- has precisely located the EIAV-RRE to a sion distinguishing between a small mole- 55-nucleotide region proximal to the 5’ splice cule such as a fluorescently labeled peptide site of exon 3 RNA [21]. Unlike the RRE and a large complex such as the same pep- structures from other retroviruses, EIAV tide bound to an antibody. cis-acting sequences do not form a clear In France, the incidence of EIAV infec- secondary structure [100]. tion is low, with only 11 cases being detected between 1988 and 1992 but sev- eral outbreaks have developed recently 3. CLINICAL FEATURES [193]. In 1993 and 1994, a total of 59 ani- OF THE EIAV-INDUCED DISEASE mals coming from 17 herds were tested seropositive for EIAV [193]; an outbreak in 3.1. Control of EIAV infection southern France revealed the presence of 23 seropositive animals among the 60 present EIA is considered a worldwide disease in the herd [193]. The situation then stabi- but is, due to its transmission by insect vec- lized until 2000, when a new outbreak was tors, predominant in warm climates [64]. In reported with 41 infected horses coming 1970, a reliable serological test based on from four herds [192, 193] (http:// agar gel immunodiffusion (AGID) using www.aht.org.uk/icc/linksicc.html). In most p26 antigen was developed [24, 25]. The cases, the origin of the contamination “Coggins Test” rapidly gained recognition as an efficient way of detecting infected ani- remains uncertain but may be linked to the mals. It has been qualified as the standard introduction of animals into the herd. diagnostic for EIAV infection by the United EIAV-positive animals are either eutha- States Department of Agriculture (USDA) nized, kept on their farm in isolation or sent and AGID is the only test officially recog- to a research facility. In France, as in other nized by the Office International des Epiz- countries, most horses are never tested for ooties (OIE). To limit virus spread, horses EIAV antibodies. In fact, only 12 000 horses are routinely tested before being allowed are tested yearly from an estimated popula- into shows or racetracks, used for breeding tion of 350 000 equids [192]. EIAV-infected or crossing borders. EIAV-seropositive asymptomatic carriers may be the source of animals are either euthanized or kept in new infections. 492 C. Leroux et al.

Figure 3. Clinical and virological profiles of ponies experimentally infected with EIAV. Febrile episodes are defined by a rectal temperature () above 39 °C in conjunction with a reduction of the platelets (••••) below 100 000/µL of whole blood. Viral RNA load in the plasma is indicated by (♦). EIA may be divided into three phases: acute, chronic and asymptomatic. As indicated by the gray arrow on top, asymptomatic animals may go back into chronic disease following naturally- or experimentally-induced immune suppression.

3.2. EIA, a cyclic disease keys do not develop clinical EIA [30] and lower amounts of plasma associated virus EIAV is unique among lentiviruses in and/or viral nucleic acids are observed in that it causes a dynamic and defined course donkeys compared to ponies infected with of infection and recurring disease in infected the same strain of EIAV [30]. Viral RNA animals (Fig. 3). The sequences of clinical levels in ponies infected with the EIAVPV events, establishment of immune response virulent strain was 10 000 fold higher than and virus evolution have been well docu- in the donkeys infected with the same strain mented in horses and ponies (Equus cabal- during the first 20 days of infection [30]. lus) [54, 55, 83, 124, 155]. Clinical, sero- However, in vitro infection of donkey logical and virological responses after or horse monocyte-derived macrophages experimental infection of donkeys (Equus yielded the same rate of EIAV production asinus) and detection of virus and descrip- suggesting that factors other than cell per- tion of the disease in mules (Equus caballus missiveness are involved in the differential × Equus asinus) have been reported [30, clinical outcome [30]. This may be of inter- 161]. While susceptible to infection, don- est, since several lines of evidence suggest EIAV infection in equids 493 a correlation between the level of virus rep- In most but not all cases, lentiviral infec- lication and the severity of disease during tions, including HIV-1 (Human Immuno- lentivus infection in primates [115, 171]. deficiency Virus Type 1) in humans or This concept of “pathogenic threshold” SRLV (Small Ruminant LentiViruses) in postulates that the level of viral replication goats and sheep, are responsible for slowly must reach a critical level to induce disease. evolving progressive degenerative diseases EIAV is responsible for a persistent leading to the death of the host after a few months or years. In contrast, in EIAV- infection in horses that is characterized by infected equids, there is a transition from a recurring cycles of viremia and of clinical chronic to an asymptomatic state, in which episodes associating fever, anemia, oedema, the animals remain free of clinical symp- thrombocytopenia and various wasting symp- toms, but infected for the rest of their life. toms. Based on experimental infection, The animals are still infectious as demon- we classically refer to acute, chronic and strated by whole-blood transfer from inap- asymptomatic stages (Fig. 3). Initial expo- parent EIAV-infected to naive animals [65]. sure to a virulent strain usually results in an This inapparent or asymptomatic stage is acute disease characterized by hyperther- reversible. Experimental administrations of mia concomitant with a severe decrease of immunosuppressive drugs such as dexam- the platelet number. The animal may die ethasone in long-term EIAV-infected horses from the acute or chronic disease, but in lead to overall immune suppression and many cases these signs are relatively mild the recrudescence of disease [32, 79, 177]. and may be overlooked. The acute episode Many animals never show any recognizable usually subsides within a few days, then the signs of EIAV and continue to exist as animal enters the chronic stage of disease asymptomatic carriers of the virus. They characterized by the recurrence of clinical can be detected during routine serological cycles. If clinical episodes are frequent, the tests in the context of herd evaluation [63]. animal may develop classic clinical disease associating anemia, weight loss and oedema 3.3. Physiopathology [63]. In practice, the course of disease may be clearly distinct in animals infected with Thrombocytopenia is the earliest and the same strain of EIAV [29, 54, 83]. Exper- most consistent abnormality observed in febrile animals. It is commonly used to mon- imental infection of ponies with EIAVPV gives a different pattern of disease evolu- itor animals during experimental EIAV infec- tion with animals referred to as disease tions. Immune-mediated platelet destruc- progressors, experiencing multiple disease tion or direct infection of megakaryocytes have been described during thrombocyto- cycles, or nonprogressors, free from clini- penia associated with several infectious cal symptoms after the first acute viremia diseases [2, 96, 97, 195]. Thrombocytope- [54, 83]. Interestingly, the nonprogressor nia is observed during HIV-1 infection. In animals had an undetectable level of plasma contrast to other haematological disorders, virus RNA, as measured by sensitive quan- a decrease in blood platelets is an early titative RT-PCR while the progressor ponies event, generally described within the first 4 6 consistently displayed 10 to 10 copies/mL two years of HIV infection [48]. Thrombo- during the long-term asymptomatic infec- cytopenia may be due to either a decrease tion [54, 83]. These observations suggest of proliferation of megakaryocyte progen- a fragile control of infection at subclinical itors or inhibition of megakaryocyte matu- levels. The basis for the difference in steady- ration. As shown by in situ hybridization, state EIAV levels in inapparent EIAV- megakaryocytes present in the spleen do infected animals remains unclear. This may not support EIAV replication [22] despite a reflect differences in disease susceptibility. marked reduction in platelet number. In 494 C. Leroux et al.

EIAV-infected horses, elevated levels of ment may participate in the impairment of platelet-bound IgG and IgM on washed CD34+ haematopoietic stem/progenitor cells platelets [22] and splenomegaly and [187]. Antibodies directed against platelets hepatomegaly at necropsy are consistent detected in the sera of HIV infected patients with immune mediated platelet destruction. may also contribute to an immune-mediated Thrombocytopenia may be explained by destruction of circulating platelets [187]. immune-complex deposition and subsequent degradation by mononuclear phagocytes. But the observation of a severe platelet 3.4. Cell tropism and strains of virus decrease in severe combined immunodefi- cient (SCID) horses, lacking functional T Experimental infections by EIAV have and B lymphocytes, suggests that non- identified active replication in various tis- immune mediated mechanisms may be sues such as the spleen, liver, lung, lymph involved during EIAV-associated throm- nodes or bone marrow [107, 143, 154], spe- bocytopenia [33]. Impairment of platelet cifically in mononuclear phagocytes. As for production may also be under control of cir- other human and animal lentiviruses, the culating factors released by infected-mac- cells of the monocyte/macrophage lineage rophages. TNFα (Tumor Necrosis Factor are the primary targets of EIAV in vivo. alpha) and TGFβ (Transforming Growth They support active viral replication during Factor beta) are produced by macrophages clinical and subclinical infection [124]. and are known to have a suppressive effect Viral RNA has also been detected in vascu- on megakaryocyte genesis. Elevation of lar endothelial cells in experimentally TNFα and TGFβ has been described in infected horses during the acute phase [125] days preceding the onset of thrombocyto- but not during subclinical infection [124]. penia [175]; Costa et al. [31] described a In vivo infections of endothelial cells may strong correlation between TNFα produc- contribute to the observed thrombocytope- tion, virulence of the infecting agent and nia in infected animals by subtle changes or clinical outcome in vaccinated and infected activation of these cells resulting in the pro- α β animals. TNF and TGF may be released motion of platelet adherence or aggrega- by macrophages upon EIAV infection and α tion, as proposed in HIV infected patients. an increase in TNF production has been Blood monocytes are permissive for EIAV shown to coincide with the onset of fever in infection [154], but monocyte differentia- African Swine Fever for example [151]. tion to mature macrophages is an essential TNFα has a direct effect on hyperthermia step for EIAV replication [102]. Whereas by directly stimulating hypothalamic pros- EIAV infects monocytes in vivo, viral taglandin biosynthesis [71]. TNFα sup- presses erythropoiesis in vitro [38, 116] and expression occurs only upon macrophage may participate in the observed haemolytic differentiation within an organ. Peripheral anemia in EIAV-infected animals in addi- blood monocytes allow virus entry and tion to the described erythrophagocytosis reverse transcription of the viral RNA but associated with the C3 fraction of the com- do not support active replication. Acute dis- plement [156]. Similar indirect mechanisms ease is associated with extensive viral rep- have been described during HIV-associated lication in tissue macrophages [154]. Inter- thrombocytopenia. Viral transcripts have estingly, the same type of restriction is been detected in vivo in megakaryocytes of observed with SRLV infection in goats and HIV-1 infected patients [96, 194] but direct sheep, for which viral expression occurs infection of haematopoietic cells does not only in macrophages [23, 46, 47, 120]. This seem to occur [97]. Bone marrow progeni- mechanism has been referred to the “Trojan tors are not infected, but increased viral rep- Horse”, where the virus silently enters the lication in the bone marrow microenviron- organ via an infected monocyte [50, 132]. EIAV infection in equids 495

A number of primary cells or cell lines by adaptation of the Wyoming strain on have been used to propagate EIAV in vitro. equine dermal fibroblasts [99]. Viruses Infection can be performed in vitro in var- derived from the CL22 clone were able to ious equine cells such as monocyte-derived establish a persistent infection in inoculated macrophages [74, 102, 137], tissue macro- horses but did not cause clinical disease phages derived from the spleen [137] or [184]. In a similar manner, viruses derived bone marrow [89], endothelial cells [104] from the pSPEIAV19 and pSPEIAV44 or fibroblasts [72, 77, 99] and in canine and clones recovered from foetal equine kidney feline transformed cell lines [8, 10, 58, 72]. (FEK) cells infected with the EIAVPV Several strains of EIAV have been stud- (pony virulent) strain, an in vitro variant of ied; they greatly vary in terms of virulence the Wyoming strain, were infectious but and ability to grow in cell culture. Most non-pathogenic for experimentally infected EIAV strains studied in laboratories world- ponies [128]. Recently, in two parallel stud- wide are derived from the highly virulent ies, replacement of the 3’ half (env and Wyoming strain [70] or from the V26 Jap- LTR) of the pSPEIAV19 with the corre- anese EIAV strain [76]. Most horse-viru- sponding regions obtained from the highly lent EIAV strains only replicate in primary pathogenic EIAVwyoming strain [129] or horse macrophage cultures and cannot rep- from the pathogenic EIAVPV variant [28] licate in tissue-derived cell lines without restored the pathogenic properties of the extensive adaptation leading to a marked viruses derived from the molecular clones. reduction of in vivo virulence [13, 76]. In These results suggest that important deter- vitro macrophage tropism tends to correlate minants of pathogenicity are present in the with higher in vivo virulence. The virulence 3’ half of pathogenic EIAV. The origin of cell adapted EIAV may be restored by of the 3’ half region determined the level serial passage through Shetland ponies [49, of pathogenicity. The inoculation of the 126]. Primary horse macrophage cultures ponies with p19/wenv17-derived viruses are intrinsically difficult to obtain and to carrying the env and LTR regions of maintain in vitro for long periods of time. EIAVwyoming induced a disease so severe New tools are now available with a reliable that the animals had to be euthanized [129]. method of isolation and maintenance of This is very similar to what is observed dur- monocyte-derived macrophages [137] and ing EIAVwyoming infection. The incidence there are recent reports of efficient replica- of disease induced in equids was found to tion of EIAV in DH82 cells, an adherent be significantly lower with EIAVUK-derived canine macrophage-like cell line [58]. Most viruses than with the parental EIAVPV infec- interestingly, these cell systems support the tion, from which we generated the clone replication of both macrophage tropic and [28]. In several other studies, EIAVUK was fibroblast-adapted EIAV strains. The atten- shown to have a disease induction rate of uation of EIAV virulence in cell culture 43% compared to 87% using the parental suggests that virulence determinants may EIAVPV [28, 86]. Genetic divergence be lost or altered when viruses are propa- between EIAVUK and EIAVPV is restricted gated in cells other than their natural target. to a replacement of arginine (R103) by tryp- To better characterize the molecular deter- tophan (W103) in the ORF S3 and to a 68 pb minants of pathogenicity and cell tropism, duplication in the EIAVUK 3’LTR [28] several laboratories have tried to develop probably originating from a minor genetic full-length EIAV molecular clones. Obtain- form in EIAVPV stock. Soon after the inoc- ing infectious and pathogenic molecular ulation of EIAVUK-derived virus in ponies, clones turned out to be challenging. A full the duplication was deleted and R103 was length clone CL22 [184] was obtained from replaced by W103 [28] suggesting that a canine foetal thymus cells (Cf2Th) infected strong in vivo selection generates virus pop- with the Malmquist strain of EIAV, derived ulations resembling the optimal EIAVPV 496 C. Leroux et al. sequence. Recently, variation in the 3’ half by intravenous infusion of EIAV-stimu- has been associated with a novel cytolytic lated donor lymphocytes in SCID foal strain of EIAV, vMA-1c, that rapidly and restores the control of viral replication in specifically kills equine dermal (ED) cells the reconstituted animal [112, 113]. Studies [106]. Several studies described the impor- examining the evolution of plasma-associ- tance of the enhancer region of the U3 ated viral RNA load together with the region of the LTR in cell tropism [14, 98, development of clinical symptoms suggest 102]. Three PU.1 binding motifs described that the threshold level for disease induc- in most strains of EIAV are required for LTR tion correlates with an RNA level in plasma expression in primary macrophages [102]. of 107 to 108 copies/mL [29, 32, 54, 83]. All Interestingly, the introduction of these these studies suggest that during the course PU.1 sites in the HIV-1 enhancer/promoter of EIAV infection, the host develops a proximal region of the LTR resulted in mac- highly effective and enduring immune rophage-specific expression [141]. Other response able to maintain viral replication enhancer motifs such as PEA-2, Oct, and below the threshold level for disease induc- CRE sites are important for expression in tion. The clearance of the primary infec- fibroblasts [105]. The repertoire of tran- tious plasma viremia correlates with the scription-factor binding motifs appears to be emergence of EIAV-specific CD8+ cyto- different in the natural target of EIAV infec- toxic T lymphocytes (CTL) and non-neu- tion and in the cells used in vitro to propa- tralizing EIAV-specific antibodies [110, gate the virus. 133, 150]. EIAV infected animals develop a strong immune response against the sur- face (gp90) and transmembrane (gp45) 4. IMMUNE CONTROL OF EIAV glycoproteins and the major core protein INFECTION AND REPLICATION p26. Although p26 is the most abundant protein in the virion, the humoral response EIAV infection results in a high-titer, against p26 is 10 to 100-fold lower than the infectious plasma viremia within three reactivity against gp90 and gp45 [51, 123]. weeks post infection. Several lines of evi- Antibodies that recognize EIAV reverse dence suggest that both humoral and cellu- transcriptase have also been described in lar EIAV-specific responses are needed to experimentally and naturally EIAV infected terminate the initial viremia. As we men- animals [35]. tioned previously, viral replication is effi- As described in HIV-1 infected patients ciently reduced to a subclinical level in ani- [80], the appearance of CD8+ CTL corre- mals evolving from the chronic stage to the lates with the clearance of the initial viremia asymptomatic stage of EIAV infection. In [110]. Later in the infection, many animals long-term inapparent EIAV-carriers, a low develop abundant CD4+ and CD8+ memory level of virus infection and replication can CTL (CTLm) recognizing EIAV antigens be detected in tissue macrophages [55, 124] [51, 54, 111, 189]. CD8+ CTL recognize and associated with plasma viremia [54, and lyse both Gag- and Env-expressing tar- 83]. The importance of the immune response gets, whereas CD4+ CTL lyse only Env- in this control was rapidly indicated by specific target cells [51]. CTL epitopes experimental immune suppression in asymp- have been mapped within Gag matrix (p15) tomatic animals leading to the recrudes- and capsid (p26) [95, 188], surface glyco- cence of disease symptoms, even after sev- protein (gp90) [94] and Rev, including the eral decades of infection [79, 177]. Arabian nuclear export and the nuclear localization foals with severe combined immunodefi- domains [114]. In a recent study, high- or ciency (SCID), lacking both B and T lym- moderate-avidity CTLm targeting Rev- phocytes, fail to control the initial viremic epitopes have been found in nonprogressor episode [33, 108, 133]. Adoptive transfer EIAV-infected animals, suggesting that CTL EIAV infection in equids 497 response against a non-variable viral pep- [32]. In a recent study, we determined the tide may be a critical element for virus sensitivity of five sequential variant enve- control [114]. Interestingly, as previously lope glycoproteins [81] to neutralization by described during HIV and SIV infections in a longitudinal panel of immune sera from primates [39, 45, 69, 119], viral evolution the source infected pony by exchanging in EIAV-gag and env genes led to CTL regions of neutralization-sensitive and neu- escape [114]. EIAV-neutralizing antibod- tralization-resistant envelopes [61]. We dem- ies that are able to block the infecting strain onstrated the influence of sequential gp90 usually emerge only after two or three variation in increasing envelope neutraliza- months post infection [5, 51, 61, 75, 123] tion resistance, identified the V3 and V4 suggesting that they are not responsible for variable regions of gp90 as the principal the termination of the acute episode. In a lon- determinants of neutralization resistance and gitudinal study, no correlation was observed suggested the importance of complex coop- between the level of neutralizing antibodies erative envelope domain interaction in defin- and the course of the infection [51]. Neu- ing this resistance [61]. tralizing antibodies do not reach a steady- state level but tend to fluctuate markedly Looking in detail at the humoral response during the course of infection [51]. The role during the progression from chronic dis- of neutralizing antibodies in controlling ease to inapparent stages, Hammond et al., EIAV replication is still unclear and com- [51, 54] described a gradual evolution of the plex. A lack of correlation between neutral- humoral response over the first 10-months izing antibody levels and protection has post infection. During this time period, been reported in several studies in infected EIAV-specific antibodies matured from [51, 117, 150] or vaccinated [66] animals. low avidity, measured by the stability of the This is in contrast with the reported effi- antibody-antigen complexes, to reach a ciency of high-titer neutralizing antibodies steady-state level of moderate to high avid- directed against the HIV-1 envelope to ity [51, 54]. A change in conformation- completely block HIV/SIV chimeric virus dependence has been reported with the ini- infection in macaque monkeys [121, 159]. tial envelope-specific antibodies mainly rec- Recrudescence of disease is, however, ognizing linear epitopes, evolving toward a associated with the emergence of neutrali- conformation-dependent response [51]. A zation-escape variants [75, 82, 83, 117], similar evolution of these immune param- suggesting that the neutralizing response is eters has been shown in experimentally in fact efficient in controlling virus replica- infected animals, regardless of their pro- tion. Three neutralization epitopes, CNT, files of disease evolution after the first acute DNT and ENT, have been mapped on the sur- viremia [54]. face glycoprotein gp90; DNT and ENT con- stitute the principal neutralizing domain Interestingly, studies in SIV or SIV/ (PND) [5]. Inoculation of ponies with virus HIV-1 infected monkeys [26, 27, 118] and particles derived from an EIAV molecular HIV-1 infected patients [27] revealed a clone lacking the PND region showed that common theme of antibody maturation dur- the animals were able to control disease in ing the first 6 to 10 months post-infection, the absence of a detectable neutralizing that is characterized by ongoing changes in response [32]. Following dexamethasone antibody titers, conformational dependence treatment, the transiently immune-suppressed and antibody avidity. Antibody-dependent ponies developed characteristic EIA, that cellular cytotoxicity (ADCC) is lacking became controlled concomitantly with the during all stages of EIAV infection [44, development of a neutralizing response, 176]. These different studies certainly indi- suggesting that a region other than the PND cate the role of an efficient and mature may be involved in the neutralizing response immune response (Fig. 4) in controlling 498 C. Leroux et al. ecific antibody ificity. Neutralizing near epitope specificity epitope near me post infection. t 10-months of infection, t 10-months and predominantly li and predominantly conformational epitope spec epitope conformational ary units according to the ti e figure represents the the e figure evolution of EIAV-sp sponse (Env-specific CTL). During the firs sponse (Env-specific avidity, non neutralizing ters are expressed in arbitr in ters are expressed utralizing and predominantly and predominantly utralizing oderate to high-avidity, ne IgG) re and of the IgG) EIAV-specific cellular e in EIAV-infected [51, 54]). Th animals (from e in EIAV-infected of infection. Immune parame infection. of adually from a population characterized by low- characterized adually from a population Maturation of the immune respons to an antibody population characterized by population characterized by m to an antibody the antibody population evolves gr months 2 to 3 after antibodies appear only response (Env-specific IgG and p26-specific Figure 4. EIAV infection in equids 499

EIAV replication. But, correlates of immune virus present in the plasma [81, 90] or that protection remain to be clearly defined. in various reservoir tissues [142]. These apparently contradictory reported results may be due to the origin of the viruses stud- 5. VIRAL EVOLUTION ied and the duration of infection. A hyper- IN INFECTED ANIMALS variable region of the U3 LTR has been described in the context of cell-adapted Lentiviruses are among the most rapidly EIAV. For example, the MA-1 variant of evolving genomes. Three polymerization EIAVwyoming was selected by in vitro adap- events occur during the retroviral cycle: the tation in equine dermal cells [14]. LTR var- generation of a double-stranded DNA from iants have been shown to be rapidly elimi- the viral RNA by the reverse transcriptase, nated in vivo [29, 91]. LTR variations are the replication of the integrated viral DNA associated with in vitro adaptation to non- by cellular DNA polymerase and the RNA natural target cells such as non-equine cells synthesis by cellular RNA polymerase II. or fibroblasts; in vivo replication probably Low fidelity of the lentiviral reverse tran- requires a rapid adaptation of the LTR. scriptase, which lacks proofreading activity Mutations in rev have been reported in sev- [134, 147] and recombination of genomes eral studies [3, 7, 81]. Interestingly, a rev within coinfected cells [148] contribute to variant present during chronic EIA had sig- an elevated mutation rate. EIAV RT is as nificantly higher Rev-mediated nuclear export error prone as HIV-1 RT and shows effi- activity than the variant sequenced during cient mismatch extension during the copy- afebrile periods, suggesting that the genetic ing of both RNA and DNA templates [4]. variation of rev may contribute to EIAV The fidelity of EIAV RT, as with other RT disease progression [3, 7]. enzymes, is sequence related; purine-pyri- It is well established that cycles of dis- midine mispairs are more efficiently extended ease correlate with the emergence of anti- than the purine-purine mispairs [4]. A high genic variants of the surface glycoprotein level of variations can be observed between [78, 117]. Virus neutralization assays estab- and within the lentivirus-infected host. lished that plasma neutralizes virus isolates Lentiviruses exist in vivo as complex pop- recovered during earlier febrile episodes, ulations of related, but not identical, viral but fails to neutralize virus isolates recov- genotypes or quasispecies. LTR regions, ered from subsequent febrile episodes [117, rev and env have been shown to accumulate 152]. Comprehensive studies on the evolu- nucleotide mutations. Studies on the evolu- tion of the surface gp90 glycoprotein revealed tion of the LTR during EIAV infections yield some controversial results. A hyper- that mutations occur rapidly after experi- variable region has been described in the U3 mental infection of ponies with the Ameri- region of the LTR, by comparing the highly can EIAVPV [81] or horses with the Japa- nese virulent EIAV strain V70 [190]. These pathogenic EIAVwyoming strain and the bio- logical MA-1 variant selected by in vitro two parallel studies with viruses from dif- replication on equine dermal (ED) cells ferent geographical origins showed that [14]. The same region has been defined as these mutations are not randomly distrib- a hot spot of mutation by comparing U3 uted along gp90 but delineate variable LTR of isolates obtained from naturally regions [81, 190]. We defined eight varia- infected animals [103]. This interesting ble regions, rapidly evolving after EIAV study highlighted the natural variation of inoculation (Fig. 1C). field isolates of EIAV. In contrast, we New predominant quasispecies are asso- described a very low rate of mutation of the ciated with each clinical episode and the cir- LTR during experimental infection of culating viral populations are completely ponies with the EIAVPV strain either in the replaced between febrile episodes (Fig. 5) 500 C. Leroux et al.

Figure 5. Evolution of the surface glycoprotein gp90 in an experimentally infected pony (from [81, 83]). Phylogenetic study of gp90 deduced amino-acid sequences of longitudinal viral isolates obtained from plasma during EIAV-induced fevers I (Q), II (‹), III (F), IV (▲), V (❍) shows that recrudescence of fever is associated with the emergence of new viral populations during the chronic phase of EIA. Evolution of gp90 continues during the asymptomatic stage of infection as shown by viral populations recovered from monocyte-derived macrophages during the asymptomatic stage (▼), suggesting a continuous viral replication. Branch lengths are proportional to the distance between the sequences. EIAV infection in equids 501

Figure 6. Evolution of the third variable region (V3) of gp90 during the course of infection in a persistently EIAVPV-infected pony (adapted from [83]). Deduced amino-acid sequences of the V3 region recovered during febrile episodes of chronic EIA or from monocyte-derived macrophages obtained during the asymptomatic stage are compared to the inoculated strain EIAVPV. Only amino-acid residues different from the EIAVPV sequence are indicated. Dots indicate residues identical to the EIAVPV sequence; dashes indicate amino-acid deletions; triangles (T) indicate potential N-glycosylation sites (NX[S/T]). The gray box indicates the Principal Neutralizing Domain (PND). [81, 190]. Point mutations, insertions and [81, 83]. The propensity for variations in deletions are observed during the dynamic gp90 glycosylation sites suggests that this evolution of the gp90 quasispecies, mainly post-translational modification is an impor- in the V3 region corresponding to the PND tant determinant of gp90 immunogenicity. (Fig. 6). Up to 80% of the amino acids of A widely accepted paradigm in lentivi- V3 have been replaced in vivo in EIAVPV rus evolution is that the rate of variation is infected ponies [83]. Interestingly, we showed directly correlated to the level of virus rep- that a large deletion of up to 15 amino-acids, lication. To test this hypothesis, we fol- removing virtually all the PND region, did lowed the EIAV gp90 evolution over a not alter the competence for in vitro [81] or 2.5 year period in ponies that differed in vivo replication [32]. Multiple mutation, markedly in clinical progression. Surpris- repositioning or creation of potential N- ingly, the level of gp90 variation was inde- linked glycosylation sites (NX[S:T]) are pendent of the number of disease cycles observed during long term infection (Fig. 6) (only one in the nonprogressor ponies and 502 C. Leroux et al.

Table I. Accumulation of mutation over a 2.5 year period in EIAV gp90 during the course of infection in disease progressor or nonprogressor EIAVPV-infected ponies (adapted from [83]).

% of envelope divergence from the inoculated EIAVPV measured at the deduced amino-acid level (± mean standard deviation) during the different stages of the disease

Non progressor ponies: Fever # 1 Asymptomatic stage 561 0.11 No chronic disease NA ± 0.14 562 0.28 No chronic disease 8.14 ± 0.21 ± 0.37 Progressor ponies: Fever # 1 Fever # 2 Fever # 3 Fever # 4 Fever # 5 Fever # 6 Asymptomatic stage 564 0.41 1.53 1.54 2.24 7.86 5.24 5.08 ± 0.34 ± 0.61 ± 0.55 ± 0.34 ± 0.96 ± 0.15 ± 0.17 567 0.41 1.01 5.12 3.22 7.46 7.67 7.44 ± 0.13 ± 0.38 ± 0.42 ± 0.35 ± 0.69 ± 2.29 ± 0.48

Asympt: asymptomatic stage of infection, 2.5 years post infection. NA: not applicable. up to six in the progressor ponies) and of the high priority in human and veterinary med- apparent steady-state level of virus replica- icine. Despite international efforts and inter- tion during long-term infection [83]. Regard- esting results, no vaccine is actually avail- less of the low levels of plasma viremia in able against HIV infection. Most naturally long term infected ponies, the virus quasi- or experimentally EIAV-infected animals species evolve constantly [83], suggesting successfully control viral replication and an active replication in reservoir tissues disease within a few months. This innate such as the liver, bone marrow or spleen ability to control lentivirus replication sug- [55] (Tab. I). Altogether, these studies show gests that a vaccine against EIAV may be that mutations are clearly associated with effective in controlling disease develop- the recurrence of EIA but, alone, are not ment. This made EIAV a unique model to sufficient to trigger viremia and disease. test various vaccine strategies. Unfortu- Disease progression does not correlate with nately, EIAV-vaccine development has only the amount of accumulated mutations, as been moderately successful and classical shown by the extent of mutation observed vaccines based on inactivated or attenuated in nonprogressor animals (Tab. I) [83]. As whole virus and on viral recombinant pro- we mentioned before, cellular and humoral tein failed to elicit a broadly protective responses are not clearly different in pro- immune response. Twenty years ago, Chi- gressor and nonprogressor ponies, suggest- nese scientists reported a successful vac- ing that other parameters control virus rep- cine trial using a live attenuated EIAV lication [54] and that emergence of a new strain produced by serial passages on don- viral population is not itself sufficient to key leukocyte cells [157]. But no independ- induce clinical symptoms of EIA. ent investigations have been possible on this very promising vaccine trial and the 6. EIAV VACCINE DEVELOPMENT results are not easily accessible to the inter- national scientific community. Inactivated The development of an efficacious vac- vaccine using the prototype cell-adapted cine against lentiviral infections remains a Wyoming strain [99] efficiently protected EIAV infection in equids 503

∆ ponies against a challenge with the homol- on EIAVUK S2, the molecular clone ogous EIAV strain, but failed to prevent EIAVUK carrying two inactivating stop infection with an heterologous strain [66]. codons in the S2 open reading frame [87]. ∆ However, the level of replication of the Immunization with EIAVUK S2 provided challenge strain was significantly reduced. protection from disease and detectable infec- A subunit vaccine enriched in EIAV enve- tion after challenge with a homologous lope glycoproteins not only failed to protect EIAV strain, using low- or high-dose expo- the vaccinated ponies against homologous sure [87]. The ability to confer sterile pro- or heterologous strains but had a high tection remains to be demonstrated with potential to enhance viral replication and to a heterologous strain. A vigorous CTL exacerbate clinical disease in challenged response is supposed to be a key parameter animals [66]. Disease enhancement has in the host immune response for the control been described in FIV [67, 131, 144] or of lentiviral infection. Targeting antigens CAEV [73, 109] vaccine trials. In a similar into the phagocytic pathway is a strategy to manner, a recombinant subunit vaccine, induce a cellular response. In this context, containing a baculovirus-expressed EIAV- a particulate vaccine, using gradient-puri- envelope surface glycoprotein, enhanced fied EIAV complexed to iron oxide beads, the disease in challenged ponies [138, 183]. was evaluated. Ponies vaccinated and chal- The severity of the disease after challenge lenged with an infectious EIAV strain have ranges from a lack of clinical signs to severe a delayed progression to disease and a lower exacerbation, with a correlation between virus load, but in most cases an absence of the level of virus replication and the sever- a CTL activity [52]. The protection after ity of the disease [138]. Antibody-depend- virus challenge has been correlated with a ent enhancement (ADE) and complement- strong anamnestic response associating an mediated antibody-dependent enhancement increase in antibody titer, a conformation- (C’-ADE) have been proposed as mecha- dependent envelope-specific response and nisms of disease enhancement. ADE is a emergence of a strong neutralizing activity phenomenon in which virus-specific anti- [52]. Recently, lipopeptide immunization bodies enhance the entry of virus into with MHC class I-restricted CTL epitopes monocytes/macrophages and granulocytic from the surface glycoprotein proved to cells through the interaction with Fc and/or have a protective effect against EIAV- complement receptors (reviewed in [169]). induced disease after challenge with a path- But in vitro ADE assays failed to correlate ogenic strain [145, 146]. Taken together, with the severity of the clinical symptoms these EIAV vaccines trials are encouraging observed in immunized ponies [139]. but correlates of protection remain to be In a study comparing the humoral clearly defined. Protection against disease responses at the day of challenge elicited by seems to be a reachable goal. Are the vac- attenuated, inactivated whole virus or enve- cinated ponies, like those in the asympto- lope subunit vaccines, Hammond et al. [53] matic phase, capable of carrying infectious tried to determine the immune correlates EIAV or can vaccine immunity prevent of protection. No single parameter of the virus transmission? humoral response (titer, avidity index and conformational ratio) provided a statisti- 7. CONCLUSION cally reliable correlate of protection. But interestingly, the level of protection after In conclusion, the rapid progression and challenge does correlate with the develop- the clear demarcation of the disease in ment of a mature antibody response [53] EIAV-infection in equids offers a unique as described during EIAV infection [51]. model with which to characterize natural Encouraging results have been described immunologic control of lentivirus replica- with a live attenuated EIAV vaccine based tion and disease and to elucidate the nature 504 C. Leroux et al. and role of viral variation in persistence and Rev coexist in vivo and differ in phenotype, pathogenesis. One EIAV-intriguing feature J. Virol. 77 (2003) 12122–12131. is the ability of most animals to eventually [4] Bakhanashvili M., Hizi A., Fidelity of DNA synthesis exhibited in vitro by the reverse control viral replication. Long-term infec- transcriptase of the lentivirus equine infec- tion is characterized by a fragile control of tious anemia virus, Biochemistry 32 (1993) infection to subclinical levels. This control 7559–7567. may be reversed by transient immunosup- [5] Ball J.M., Rushlow K.E., Issel C.J., Montelaro pression, suggesting a tight control of virus R.C., Detailed mapping of the antigenicity of replication by the immune system. the surface unit glycoprotein of equine infec- tious anemia virus by using synthetic peptide Recurring episodes of EIA are associ- strategies, J. Virol. 66 (1992) 732–742. ated with the dynamic evolution and repli- [6] Belshan M., Harris M.E., Shoemaker A.E., cation of distinct EIAV quasispecies, that Hope T.J., Carpenter S., Biological character- temporarily escape the established immune ization of Rev variation in equine infectious response. The evolution of the disease can anemia virus, J. Virol. 72 (1998) 4421–4426. [7] Belshan M., Baccam P., Oaks J.L., Sponseller be viewed as a fragile balance between the B.A., Murphy S.C., Cornette J., Carpenter S., rapid viral evolution and an efficient immune Genetic and biological variation in equine response. infectious anemia virus Rev correlates with variable stages of clinical disease in an exper- Clearance of the primary infectious plasma imentally infected pony, Virology 279 (2001) viremia correlates with the emergence 185–200. + of EIAV-specific CD8 cytotoxic T lym- [8] Benton C.V., Brown B.L., Harshman J.S., phocytes and non-neutralizing antibodies. Gilden R.V., In vitro host range of equine Then, a gradual evolution of the humoral infectious anemia virus, Intervirology 16 response clearly develops over the first 10- (1981) 225–232. months post infection. Interestingly, stud- [9] Bieniasz P.D., Grdina T.A., Bogerd H.P., Cullen B.R., Highly divergent lentiviral Tat ies in EIAV-infected equids, and SIV or proteins activate viral gene expression by a SIV/HIV-1 infected monkeys and HIV-1 common mechanism, Mol. Cell. Biol. 19 infected patients revealed a common theme (1999) 4592–4599. of antibody maturation, that is character- [10] Bouillant A.M., Nielsen K., Ruckerbauer ized by ongoing changes in antibody titers, G.M., Samagh B.S., Hare W.C., The persist- ent infection of a canine thymus cell line by conformational dependence and antibody equine infectious anaemia virus and prelimi- avidity. nary data on the production of viral antigens, Development of an efficacious vaccine J. Virol. Methods 13 (1986) 309–321. against lentiviruses, such as HIV-1 and [11] Brigati C., Giacca M., Noonan D.M., Albini A., HIV Tat, its targets and the control of viral HIV-2, is a major challenge of this century. gene expression, FEMS Microbiol. Lett. 220 The innate ability of EIAV-infected equids (2003) 57–65. to efficiently control viral replication within [12] Burki F., Rossmanith W., Rossmanith E., a few months could help us reach this goal. Equine lentivirus, comparative studies on four serological tests for the diagnosis of equine infectious anaemia, Vet. Microbiol. 33 (1992) REFERENCES 353–360. [13] Carpenter S., Chesebro B., Change in host cell [1] Albrecht T.R., Lund L.H., Garcia-Blanco tropism associated with in vitro replication of M.A., Canine cyclin T1 rescues equine infec- equine infectious anemia virus, J. Virol. 63 tious anemia virus tat trans-activation in (1989) 2492–2496. human cells, Virology 268 (2000) 7–11. [14] Carpenter S., Alexandersen S., Long M.J., [2] Axthelm M.K., Krakowka S., Canine distem- Perryman S., Chesebro B., Identification of a per virus-induced thrombocytopenia, Am. J. hypervariable region in the long terminal Vet. Res. 48 (1987) 1269–1275. repeat of equine infectious anemia virus, J. [3] Baccam P., Thompson R.J., Li Y., Sparks W.O., Virol. 65 (1991) 1605–1610. Belshan M., Dorman K.S., Wannemuehler Y., [15] Carvalho M., Derse D., Mutational analysis of Oaks J.L., Cornette J.L., Carpenter S., Sub- the equine infectious anemia virus Tat- respon- populations of equine infectious anemia virus sive element, J. Virol. 65 (1991) 3468–3474. EIAV infection in equids 505

[16] Carvalho M., Derse D., Physical and func- [28] Cook R.F., Leroux C., Cook S.J., Berger S.L., tional characterization of transcriptional con- Lichtenstein D.L., Ghabrial N.N., Montelaro trol elements in the equine infectious anemia R.C., Issel C.J., Development and characteri- virus promoter, J. Virol. 67 (1993) 2064– zation of an in vivo pathogenic molecular 2074. clone of equine infectious anemia virus, J. [17] Carvalho M., Derse D., The PU.1/Spi-1 proto- Virol. 72 (1998) 1383–1393. oncogene is a transcriptional regulator of a [29] Cook R.F., Cook S.J., Berger S.L., Leroux C., lentivirus promoter, J. Virol. 67 (1993) 3885– Ghabrial N.N., Gantz M., Bolin P.S., Mousel 3890. M.R., Montelaro R.C., Issel C.J., Enhance- ment of equine infectious anemia virus viru- [18] Charman H.P., Bladen S., Gilden R.V., Coggins lence by identification and removal of subop- L., Equine infectious anemia virus: evidence timal nucleotides, Virology 313 (2003) 588– favoring classification as a retrovirus, J. Virol. 603. 19 (1976) 1073–1079. [30] Cook S.J., Cook R.F., Montelaro R.C., Issel [19] Chen C., Montelaro R.C., Characterization of C.J., Differential responses of Equus caballus RNA elements that regulate gag-pol ribos- and Equus asinus to infection with two path- omal frameshifting in equine infectious ane- ogenic strains of equine infectious anemia mia virus, J. Virol. 77 (2003) 10280–10287. virus, Vet. Microbiol. 79 (2001) 93–109. [20] Chen C., Li F., Montelaro R.C., Functional [31] Costa L.R., Santos I.K., Issel C.J., Montelaro roles of equine infectious anemia virus Gag p9 R.C., Tumor necrosis factor-alpha production in viral budding and infection, J. Virol. 75 and disease severity after immunization with (2001) 9762–9770. enriched major core protein (p26) and/or [21] Chung H., Derse D., Binding sites for Rev and infection with equine infectious anemia virus, ASF/SF2 map to a 55-nucleotide purine-rich Vet. Immunol. Immunopathol. 57 (1997) exonic element in equine infectious anemia 33–47. virus RNA, J. Biol. Chem. 276 (2001) 18960– [32] Craigo J.K., Leroux C., Howe L., Steckbeck 18967. J.D., Cook S.J., Issel C.J., Montelaro R.C., [22] Clabough D.L., Gebhard D., Flaherty M.T., Transient immune suppression of inapparent Whetter L.E., Perry S.T., Coggins L., Fuller carriers infected with a principal neutralizing F.J., Immune-mediated thrombocytopenia in domain-deficient equine infectious anaemia virus induces neutralizing antibodies and low- horses infected with equine infectious anemia ers steady-state virus replication, J. Gen. virus, J. Virol. 65 (1991) 6242–6251. Virol. 83 (2002) 1353–1359. [23] Clements J.E., Gabuzda D.H., Gdovin S.L., [33] Crawford T.B., Wardrop K.J., Tornquist S.J., Cell type specific and viral regulation of visna Reilich E., Meyers K.M., McGuire T.C., A virus gene expression, Virus Res. 16 (1990) primary production deficit in the thrombocy- 175–183. topenia of equine infectious anemia, J. Virol. [24] Coggins L., Norcross N.L., Immunodiffusion 70 (1996) 7842–7850. reaction in equine infectious anemia, Cornell [34] Cupp E.W., Kemen M.J., The role of stable Vet. 60 (1970) 330. flies and mosquitoes in the transmission of [25] Coggins L., Norcross N.L., Nusbaum S.R., equine infectious anemia virus, Proc. Annu. Diagnosis of equine infectious anemia by Meet. US Anim. Health Assoc. 84 (1980) immunodiffusion test, Am. J. Vet. Res. 33 362–367. (1972) 11–18. [35] DeVico A.L., Issel C.J., Le Grice S.F., Payne [26] Cole K.S., Rowles J.L., Jagerski B.A., S.L., Montelaro R.C., Sarngadharan M.G., Murphey-Corb M., Unangst T., Clements High prevalence of serum antibodies to equine J.E., Robinson J., Wyand M.S., Desrosiers infectious anemia virus reverse transcriptase, R.C., Montelaro R.C., Evolution of envelope- AIDS Res. Hum. Retroviruses 9 (1993) 7–11. specific antibody responses in monkeys [36] Dorn P.L., Derse D., Cis- and trans-acting reg- experimentally infected or immunized with ulation of gene expression of equine infec- simian immunodeficiency virus and its asso- tious anemia virus, J. Virol. 62 (1988) 3522– ciation with the development of protective 3526. immunity, J. Virol. 71 (1997) 5069–5079. [37] dos Reis J.K., Melo L.M., Rezende M.R., [27] Cole K.S., Murphey-Corb M., Narayan O., Leite R.C., Use of an ELISA test in the erad- Joag S.V., Shaw G.M., Montelaro R.C., Com- ication of an equine infectious anaemia focus, mon themes of antibody maturation to simian Trop. Anim. Health Prod. 26 (1994) 65–68. immunodeficiency virus, simian-human immu- [38] Dufour C., Corcione A., Svahn J., Haupt R., nodeficiency virus, and human immunodefi- Poggi V., Beka'ssy A.N., Scime R., Pistorio ciency virus type 1 infections, J. Virol. 72 A., Pistoia V., TNF-alpha and IFN-gamma are (1998) 7852–7859. overexpressed in the bone marrow of Fanconi 506 C. Leroux et al.

anemia patients and TNF-alpha suppresses [48] Goldsweig H.G., Grossman R., William D., erythropoiesis in vitro, Blood 102 (2003) Thrombocytopenia in homosexual men, Am. 2053–2059. J. Hematol. 21 (1986) 243–247. [39] Evans D.T., O'Connor D.H., Jing P., Dzuris [49] Gutekunst D.E., Becvar C.S., Responses in J.L., Sidney J., da Silva J., Allen T.M., Horton horses infected with equine infectious anemia H., Venham J.E., Rudersdorf R.A., Vogel T., virus adapted to tissue culture, Am. J. Vet. Pauza C.D., Bontrop R.E., DeMars R., Sette Res. 40 (1979) 974–977. A., Hughes A.L., Watkins D.I., Virus-specific [50] Haase A.T., Pathogenesis of lentivirus infec- cytotoxic T-lymphocyte responses select for tions, Nature 322 (1986) 130–136. amino-acid variation in simian immunodefi- [51] Hammond S.A., Cook S.J., Lichtenstein D.L., ciency virus Env and Nef, Nat. Med. 5 (1999) Issel C.J., Montelaro R.C., Maturation of the 1270–1276. cellular and humoral immune responses to [40] Foil L.D., Meek C.L., Adams W.V., Issel C.J., persistent infection in horses by equine infec- Mechanical transmission of equine infectious tious anemia virus is a complex and lengthy anemia virus by deer flies (Chrysops flavidus) process, J. Virol. 71 (1997) 3840–3852. and stable flies (Stomoxys calcitrans), Am. J. [52] Hammond S.A., Cook S.J., Falo L.D.J., Issel Vet. Res. 44 (1983) 155–156. C.J., Montelaro R.C., A particulate viral pro- [41] Foil L.D., Adams W.V., McManus J.M., Issel tein vaccine reduces viral load and delays pro- C.J., Bloodmeal residues on mouthparts of gression to disease in immunized ponies chal- Tabanus fuscicostatus (Diptera: Tabanidae) lenged with equine infectious anemia virus, and the potential for mechanical transmission Virology 254 (1999) 37–49. of pathogens, J. Med. Entomol. 24 (1987) [53] Hammond S.A., Raabe M.L., Issel C.J., 613–616. Montelaro R.C., Evaluation of antibody [42] Freed E.O., HIV-1 gag proteins: diverse func- parameters as potential correlates of protec- tions in the virus life cycle, Virology 251 tion or enhancement by experimental vac- (1998) 1–15. cines to equine infectious anemia virus, Virol- [43] Freed E.O., Viral late domains, J. Virol. 76 ogy 262 (1999) 416–430. (2002) 4679–4687. [54] Hammond S.A., Li F., McKeon B.M.S., [44] Fujimiya Y., Perryman L.E., Crawford T.B., Cook S.J., Issel C.J., Montelaro R.C., Immune Leukocyte cytotoxicity in a persistent virus responses and viral replication in long-term infection: presence of direct cytotoxicity but inapparent carrier ponies inoculated with absence of antibody-dependent cellular cyto- equine infectious anemia virus, J. Virol. 74 toxicity in horses infected with equine infec- (2000) 5968–5981. tious anemia virus, Infect. Immun. 24 (1979) [55] Harrold S.M., Cook S.J., Cook R.F., Rushlow 628–636. K.E., Issel C.J., Montelaro R.C., Tissue sites [45] Geels M.J., Cornelissen M., Schuitemaker H., of persistent infection and active replication of Anderson K., Kwa D., Maas J., Dekker J.T., equine infectious anemia virus during acute Baan E., Zorgdrager F., van den Burg R., van disease and asymptomatic infection in exper- Beelen M., Lukashov V.V., Fu T.M., Paxton imentally infected equids, J. Virol. 74 (2000) W.A., van der Hoek L., Dubey S.A., Shiver 3112–3121. J.W., Goudsmit J., Identification of sequential [56] Hatanaka H., Lourin O., Rao Z., Fry E., viral escape mutants associated with altered Kingsman A., Stuart D.I., Structure of equine T-cell responses in a human immunodefi- infectious anemia virus matrix protein, J. ciency virus type 1-infected individual, J. Virol. 76 (2002) 1876–1883. Virol. 77 (2003) 12430–12440. [57] Hawkins J.A., Adams W.V., Cook L., Wilson [46] Gendelman H.E., Narayan O., Molineaux S., B.H., Roth E.E., Role of horse fly (Tabanus Clements J.E., Ghotbi Z., Slow, persistent rep- fuscicostatus Hine) and stable fly (Stomoxys lication of lentiviruses: role of tissue macro- calcitrans L.) in transmission of equine infec- phages and macrophage precursors in bone tious anemia to ponies in Louisiana, Am. J. marrow, Proc. Natl. Acad. Sci. USA 82 (1985) Vet. Res. 34 (1973) 1583–1586. 7086–7090. [58] Hines R., Maury W., DH82 cells: a macro- [47] Gendelman H.E., Narayan O., Kennedy- phage cell line for the replication and study of Stoskopf S., Kennedy P.G., Ghotbi Z., Clements equine infectious anemia virus, J. Virol. J.E., Stanley J., Pezeshkpour G., Tropism of Methods 95 (2001) 47–56. sheep lentiviruses for monocytes: susceptibil- [59] Hoffman D.W., White S.W., NMR analysis of ity to infection and virus gene expression the trans-activation response (TAR) RNA ele- increase during maturation of monocytes to ment of equine infectious anemia virus, macrophages, J. Virol. 58 (1986) 67–74. Nucleic Acids Res. 23 (1995) 4058–4065. EIAV infection in equids 507

[60] Hoffman D.W., Colvin R.A., Garcia-Blanco vating protein breakdown in muscle or lipol- M.A., White S.W., Structural features of the ysis in adipose tissue, J. Clin. Invest. 81 (1988) trans-activation response RNA element of 1384–1389. equine infectious anemia virus, Biochemistry [72] Klevjer-Anderson P., Cheevers W.P., Crawford 32 (1993) 1096–1104. T.B., Characterization of the infection of [61] Howe L., Leroux C., Issel C.J., Montelaro equine fibroblasts by equine infectious ane- R.C., Equine infectious anemia virus enve- mia virus, Arch. Virol. 60 (1979) 279–289. lope evolution in vivo during persistent infec- [73] Knowles D. Jr., Cheevers W., McGuire T., tion progressively increases resistance to in Stem T., Gorham J., Severity of arthritis is pre- vitro serum antibody neutralization as a dom- dicted by antibody response to gp135 in inant phenotype, J. Virol. 76 (2002) 10588– chronic infection with caprine arthritis- 10597. encephalitis virus, J. Virol. 64 (1990) 2396– [62] Hussain K.A., Issel C.J., Rwambo P.M., Arnizaut 2398. A.B., Ball J.M., Schnorr K.L., Montelaro R.C., [74] Kobayashi K., Kono Y., Propagation and titra- Identification of gag precursor of equine tion of equine infectious anemia virus in horse infectious anaemia virus with monoclonal leukocyte culture, Natl. Inst. Anim. Health Q. antibodies to the major viral core protein, p26, (Tokyo) 7 (1967) 8–20. J. Gen. Virol. 69 (1988) 1719–1724. [63] Issel C.J., Coggins L., Equine infectious ane- [75] Kono Y., Viremia and immunological responses mia: current knowledge, J. Am. Vet. Med. in horses infected with equine infectious ane- Assoc. 174 (1979) 727–733. mia virus, Natl. Inst. Anim. Health Q. (Tokyo) 9 (1969) 1–9. [64] Issel C.J., Foil L.D., Studies on equine infec- tious anemia virus transmission by insects, J. [76] Kono Y., Kobayashi K., Changes in patho- Am. Vet. Med. Assoc. 184 (1984) 293–297. genicity of equine infectious anemia virus during passages in horse leukocyte cultures, [65] Issel C.J., Adams W.V.J., Meek L., Ochoa R., Natl. Inst. Anim. Health Q. (Tokyo) 10 (1970) Transmission of equine infectious anemia 106–112. virus from horses without clinical signs of dis- ease, J. Am. Vet. Med. Assoc. 180 (1982) [77] Kono Y., Yoshino T., Propagation of equine 272–275. infectious anemia virus in horse kidney cell cultures, Natl. Inst. Anim. Health Q. (Tokyo) [66] Issel C.J., Horohov D.W., Lea D.F., Adams 14 (1974) 155–162. W.V.J., Hagius S.D., McManus J.M., Allison A.C., Montelaro R.C., Efficacy of inactivated [78] Kono Y., Kobayashi K., Fukunaga Y., Anti- whole-virus and subunit vaccines in prevent- genic drift of equine infectious anemia virus ing infection and disease caused by equine in chronically infected horses, Arch. Gesamte infectious anemia virus, J. Virol. 66 (1992) Virusforsch. 41 (1973) 1–10. 3398–3408. [79] Kono Y., Hirasawa K., Fukunaga Y., Taniguchi [67] Karlas J.A., Siebelink K.H., Peer M.A., Huisman T., Recrudescence of equine infectious ane- W., Cuisinier A.M., Rimmelzwaan G.F., mia by treatment with immunosuppressive Osterhaus A.D., Vaccination with experimen- drugs, Natl. Inst. Anim. Health Q. (Tokyo) 16 tal feline immunodeficiency virus vaccines, (1976) 8–15. based on autologous infected cells, elicits [80] Koup R.A., Safrit J.T., Cao Y., Andrews C.A., enhancement of homologous challenge infec- McLeod G., Borkowsky W., Farthing C., Ho tion, J. Gen. Virol. 80 (1999) 761–765. D.D., Temporal association of cellular immune [68] Katzmann D.J., Odorizzi G., Emr S.D., responses with the initial control of viremia in Receptor downregulation and multivesicular- primary human immunodeficiency virus type 1 body sorting, Nat. Rev. Mol. Cell Biol. 3 syndrome, J. Virol. 68 (1994) 4650–4655. (2002) 893–905. [81] Leroux C., Issel C.J., Montelaro R.C., Novel [69] Kaur A., Alexander L., Staprans S.I., Denekamp and dynamic evolution of equine infectious L., Hale C.L., McClure H.M., Feinberg M.B., anemia virus genomic quasispecies associated Desrosiers R.C., Johnson R.P., Emergence of with sequential disease cycles in an experi- cytotoxic T lymphocyte escape mutations mentally infected pony, J. Virol. 71 (1997) in nonpathogenic simian immunodeficiency 9627–9639. virus infection, Eur. J. Immunol. 31 (2001) [82] Leroux C., Chastang J., Greenland T., Mornex 3207–3217. J.F., Genomic heterogeneity of small rumi- [70] Kemeny L.J., Mott L.O., Pearson J.E., Titra- nant lentiviruses: existence of heterogeneous tion of equine infectious anemia virus. Effect populations in sheep and of the same lentiviral of dosage on incubation time and clinical genotypes in sheep and goats, Arch. Virol. 142 signs, Cornell Vet. 61 (1971) 687–695. (1997) 1125–1137. [71] Kettelhut I.C., Goldberg A.L., Tumor necro- [83] Leroux C., Craigo J.K., Issel C.J., Montelaro sis factor can induce fever in rats without acti- R.C., Equine infectious anemia virus genomic 508 C. Leroux et al.

evolution in progressor and nonprogressor recombinant retroviral vectors, J. Virol. 73 ponies, J. Virol. 75 (2001) 4570–4583. (1999) 2762–2769. [84] Lew A.M., Thomas L.M., Huntington P.J., A [95] Lonning S.M., Zhang W., McGuire T.C., comparison of ELISA, FAST-ELISA and Gag protein epitopes recognized by CD4(+) gel diffusion tests for detecting antibody to T-helper lymphocytes from equine infectious equine infectious anaemia virus, Vet. Micro- anemia virus-infected carrier horses, J. Virol. biol. 34 (1993) 1–5. 73 (1999) 4257–4265. [85] Li F., Puffer B.A., Montelaro R.C., The S2 [96] Louache F., Bettaieb A., Henri A., Oksenhendler gene of equine infectious anemia virus is dis- E., Farcet J.P., Bierling P., Seligmann M., pensable for viral replication in vitro, J. Virol. Vainchenker W., Infection of megakaryo- 72 (1998) 8344–8348. cytes by human immunodeficiency virus in [86] Li F., Leroux C., Craigo J.K., Cook S.J., Issel seropositive patients with immune thrombo- C.J., Montelaro R.C., The S2 gene of equine cytopenic purpura, Blood 78 (1991) 1697– infectious anemia virus is a highly conserved 1705. determinant of viral replication and virulence [97] Louache F., Henri A., Bettaieb A., Oksenhendler properties in experimentally infected ponies, E., Raguin G., Tulliez M., Vainchenker W., J. Virol. 74 (2000) 573–579. Role of human immunodeficiency virus rep- [87] Li F., Craigo J.K., Howe L., Steckbeck J.D., lication in defective in vitro growth of hemat- Cook S., Issel C., Montelaro R.C., A live opoietic progenitors, Blood 80 (1992) 2991– attenuated equine infectious anemia virus 2999. proviral vaccine with a modified S2 gene pro- [98] Madden C.R., Shih D.S., Analysis of the long vides protection from detectable infection terminal repeat from a cytopathic strain of by intravenous virulent virus challenge of equine infectious anemia virus, Virology 225 experimentally inoculated horses, J. Virol. 77 (1996) 395–399. (2003) 7244–7253. [99] Malmquist W.A., Barnett D., Becvar C.S., [88] Liang C., Wainberg M.A., The role of Tat in Production of equine infectious anemia anti- HIV-1 replication: an activator and/or a sup- gen in a persistently infected cell line, Arch. pressor? AIDS Rev. 4 (2002) 41–49. Gesamte Virusforsch. 42 (1973) 361–370. [89] Lichtenstein D.L., Rushlow K.E., Cook R.F., [100] Martarano L., Stephens R., Rice N., Derse D., Raabe M.L., Swardson C.J., Kociba G.J., Equine infectious anemia virus trans-regula- Issel C.J., Montelaro R.C., Replication in vitro and in vivo of an equine infectious ane- tory protein Rev controls viral mRNA stabil- mia virus mutant deficient in dUTPase activ- ity, accumulation, and alternative splicing, J. ity, J. Virol. 69 (1995) 2881–2888. Virol. 68 (1994) 3102–3111. [90] Lichtenstein D.L., Issel C.J., Montelaro R.C., [101] Martin-Serrano J., Yaravoy A., Perez-Caballero Genomic quasispecies associated with the D., Bieniasz P.D., Divergent retroviral late- initiation of infection and disease in ponies budding domains recruit vacuolar protein experimentally infected with equine infec- sorting factors by using alternative adaptor tious anemia virus, J. Virol. 70 (1996) 3346– proteins, Proc. Natl. Acad. Sci. USA 100 3354. (2003) 12414–12419. [91] Lichtenstein D.L., Craigo J.K., Leroux C., [102] Maury W., Monocyte maturation controls Rushlow K.E., Cook R.F., Cook S.J., Issel expression of equine infectious anemia virus, C.J., Montelaro R.C., Effects of long terminal J. Virol. 68 (1994) 6270–6279. repeat sequence variation on equine infec- [103] Maury W., Perryman S., Oaks J.L., Seid tious anemia virus replication in vitro and in B.K., Crawford T., McGuire T., Carpenter S., vivo, Virology 263 (1999) 408–417. Localized sequence heterogeneity in the long [92] Ligné M., Mémoire et observations sur une terminal repeats of in vivo isolates of equine maladie de sang, connue sous le nom infectious anemia virus, J. Virol. 71 (1997) d'anhémie hydrohémie, cachexie acquise du 4929–4937. cheval, Rec. Med. Vet. Ec. Alfort 1843 [104] Maury W., Oaks J.L., Bradley S., Equine (1843) 30–44. endothelial cells support productive infection [93] Lloberas J., Soler C., Celada A., The key role of equine infectious anemia virus, J. Virol. 72 of PU.1/SPI-1 in B cells, myeloid cells and (1998) 9291–9297. macrophages, Immunol. Today 20 (1999) [105] Maury W., Bradley S., Wright B., Hines R., 184–189. Cell specificity of the transcription-factor [94] Lonning S.M., Zhang W., Leib S.R., repertoire used by a lentivirus: motifs impor- McGuire T.C., Detection and induction of tant for expression of equine infectious ane- equine infectious anemia virus-specific cyto- mia virus in nonmonocytic cells, Virology toxic T-lymphocyte responses by use of 267 (2000) 267–278. EIAV infection in equids 509

[106] Maury W., Wright P.J., Bradley S., Charac- blood cell kinetics and induces anemia in terization of a cytolytic strain of equine infec- vivo, Faseb J. 3 (1989) 1637–1643. tious anemia virus, J. Virol. 77 (2003) 2385– [117] Montelaro R.C., Parekh B., Orrego A., Issel 2399. C.J., Antigenic variation during persistent [107] McGuire T.C., Crawford T.B., Henson J.B., infection by equine infectious anemia virus, Immunofluorescent localization of equine a retrovirus, J. Biol. Chem. 259 (1984) infectious anemia virus in tissue, Am. J. 10539–10544. Pathol. 62 (1971) 283–294. [118] Montelaro R.C., Cole K.S., Hammond S.A., [108] McGuire T.C., Banks K.L., Poppie M.J., Maturation of immune responses to lentivi- Combined immunodeficiency in horses: char- rus infection: implications for AIDS vaccine acterization of the lymphocyte defect, Clin. development, AIDS Res. Hum. Retroviruses Immunol. Immunopathol. 3 (1975) 555–566. 14 Suppl. 3 (1998) S255–259. [109] McGuire T.C., Adams D.S., Johnson G.C., [119] Moore C.B., John M., James I.R., Christiansen Klevjer-Anderson P., Barbee D.D., Gorham F.T., Witt C.S., Mallal S.A., Evidence of J.R., Acute arthritis in caprine arthritis- HIV-1 adaptation to HLA-restricted immune encephalitis virus challenge exposure of vac- responses at a population level, Science 296 cinated or persistently infected goats, Am. J. (2002) 1439–1443. Vet. Res. 47 (1986) 537–540. [120] Narayan O., Kennedy-Stoskopf S., Sheffer [110] McGuire T.C., Tumas D.B., Byrne K.M., D., Griffin D.E., Clements J.E., Activation of Hines M.T., Leib S.R., Brassfield A.L., caprine arthritis-encephalitis virus expres- O'Rourke K.I., Perryman L.E., Major histo- sion during maturation of monocytes to mac- compatibility complex-restricted CD8+ cyto- rophages, Infect. Immun. 41 (1983) 67–73. toxic T lymphocytes from horses with equine [121] Nishimura Y., Igarashi T., Haigwood N., infectious anemia virus recognize Env and Sadjadpour R., Plishka R.J., Buckler-White Gag/PR proteins, J. Virol. 68 (1994) 1459– A., Shibata R., Martin M.A., Determination 1467. of a statistically valid neutralization titer in [111] McGuire T.C., Zhang W., Hines M.T., Henney plasma that confers protection against sim- P.J., Byrne K.M., Frequency of memory ian-human immunodeficiency virus chal- cytotoxic T lymphocytes to equine infectious lenge following passive transfer of high- anemia virus proteins in blood from carrier titered neutralizing antibodies, J. Virol. 76 horses, Virology 238 (1997) 85–93. (2002) 2123–2130. [112] McGuire T.C., Fraser D.G., Mealey R.H., [122] Noiman S., Yaniv A., Sherman L., Tronick Cytotoxic T lymphocytes and neutralizing S.R., Gazit A., Pattern of transcription of the antibody in the control of equine infectious genome of equine infectious anemia virus, J. anemia virus, Viral Immunol. 15 (2002) 521– Virol. 64 (1990) 1839–1843. 531. [123] O'Rourke K., Perryman L.E., McGuire T.C., [113] Mealey R.H., Fraser D.G., Oaks J.L., Cantor Antiviral, anti-glycoprotein and neutralizing G.H., McGuire T.C., Immune reconstitution antibodies in foals with equine infectious prevents continuous equine infectious ane- anaemia virus, J. Gen. Virol. 69 (1988) 667– mia virus replication in an arabian foal with 674. severe combined immunodeficiency: lessons [124] Oaks J.L., McGuire T.C., Ulibarri C., Crawford for control of lentiviruses, Clin. Immunol. T.B., Equine infectious anemia virus is found 101 (2001) 237–247. in tissue macrophages during subclinical [114] Mealey R.H., Zhang B., Leib S.R., Littke infection, J. Virol. 72 (1998) 7263–7269. M.H., McGuire T.C., Epitope specificity is [125] Oaks J.L., Ulibarri C., Crawford T.B., Endothe- critical for high and moderate avidity cyto- lial cell infection in vivo by equine infectious toxic T lymphocytes associated with control anaemia virus, J. Gen. Virol. 80 (1999) 2393– of viral load and clinical disease in horses 2397. with equine infectious anemia virus, Virol- [126] Orrego A., Issel C.J., Montelaro R.C., Adams ogy 313 (2003) 537–552. W.V.J., Virulence and in vitro growth of a [115] Mellors J.W., Rinaldo C.R. Jr., Gupta P., cell-adapted strain of equine infectious ane- White R.M., Todd J.A., Kingsley L.A., Prog- mia virus after serial passage in ponies, Am. nosis in HIV-1 infection predicted by the J. Vet. Res. 43 (1982) 1556–1560. quantity of virus in plasma, Science 272 [127] Parent L.J., Bennett R.P., Craven R.C., Nelle (1996) 1167–1170. T.D., Krishna N.K., Bowzard J.B., Wilson [116] Moldawer L.L., Marano M.A., Wei H., Fong C.B., Puffer B.A., Montelaro R.C., Wills Y., Silen M.L., Kuo G., Manogue K.R., Vlas- J.W., Positionally independent and exchange- sara H., Cohen H., Cerami A., et al., Cachec- able late budding functions of the Rous sar- tin/tumor necrosis factor-alpha alters red coma virus and human immunodeficiency 510 C. Leroux et al.

virus Gag proteins, J. Virol. 69 (1995) 5455– severe enhancement, Virology 245 (1998) 5460. 151–162. [128] Payne S.L., Rausch J., Rushlow K., Monte- [139] Raabe M.L., Issel C.J., Montelaro R.C., In laro R.C., Issel C., Flaherty M., Perry S., vitro antibody-dependent enhancement assays Sellon D., Fuller F., Characterization of are insensitive indicators of in vivo vaccine infectious molecular clones of equine infec- enhancement of equine infectious anemia tious anaemia virus, J. Gen. Virol. 75 (1994) virus, Virology 259 (1999) 416–427. 425–429. [140] Rasty S., Dhruva B.R., Schiltz R.L., Shih [129] Payne S.L., Qi X.M., Shao H., Dwyer A., D.S., Issel C.J., Montelaro R.C., Proviral Fuller F.J., Disease induction by virus DNA integration and transcriptional patterns derived from molecular clones of equine of equine infectious anemia virus during per- infectious anemia virus, J. Virol. 72 (1998) sistent and cytopathic infections, J. Virol. 64 483–487. (1990) 86–95. [130] Pearson J.E., Gipson C.A., Standardization [141] Reed-Inderbitzin E., Maury W., Cellular spe- of equine infectious anemia immunodiffu- cificity of HIV-1 replication can be control- sion and cELISA tests and their application led by LTR sequences, Virology 314 (2003) to control of the disease in the United States, 680–695. Equine Vet. Science 8 (1988) 60–61. [142] Reis J.K., Craigo J.K., Cook S.J., Issel C.J., [131] Pedersen N.C., Johnson L., Birch D., Theilen Montelaro R.C., Characterization of EIAV G.H., Possible immunoenhancement of per- LTR variability and compartmentalization in sistent viremia by feline leukemia virus enve- various reservoir tissues of long-term inap- lope glycoprotein vaccines in challenge- parent carrier ponies, Virology 311 (2003) exposure situations where whole inactivated 169–180. virus vaccines were protective, Vet. Immu- [143] Rice N.R., Lequarre A.S., Casey J.W., Lahn nol. Immunopathol. 11 (1986) 123–148. S., Stephens R.M., Edwards J., Viral DNA in [132] Peluso R., Haase A., Stowring L., Edwards horses infected with equine infectious ane- M., Ventura P., A Trojan Horse mechanism mia virus, J. Virol. 63 (1989) 5194–5200. for the spread of visna virus in monocytes, [144] Richardson J., Broche S., Baud S., Leste- Virology 147 (1985) 231–236. Lasserre T., Femenia F., Levy D., Moraillon [133] Perryman L.E., O'Rourke K.I., McGuire A., Pancino G., Sonigo P., Lymphoid activa- T.C., Immune responses are required to ter- tion: a confounding factor in AIDS vaccine minate viremia in equine infectious anemia development? J. Gen. Virol. 83 (2002) 2515– lentivirus infection, J. Virol. 62 (1988) 3073– 2521. 3076. [145] Ridgely S.L., McGuire T.C., Lipopeptide [134] Preston B.D., Poiesz B.J., Loeb L.A., Fidelity stimulation of MHC class I-restricted mem- of HIV-1 reverse transcriptase, Science 242 ory cytotoxic T lymphocytes from equine (1988) 1168–1171. infectious anemia virus-infected horses, Vaccine 20 (2002) 1809–1819. [135] Puffer B.A., Parent L.J., Wills J.W., Montelaro R.C., Equine infectious anemia virus utilizes [146] Ridgely S.L., Zhang B., McGuire T.C., a YXXL motif within the late assembly Response of ELA-A1 horses immunized domain of the Gag p9 protein, J. Virol. 71 with lipopeptide containing an equine infec- (1997) 6541–6546. tious anemia virus ELA-A1-restricted CTL epitope to virus challenge, Vaccine 21 (2003) [136] Puffer B.A., Watkins S.C., Montelaro R.C., 491–506. Equine infectious anemia virus Gag polypro- tein late domain specifically recruits cellular [147] Roberts J.D., Bebenek K., Kunkel T.A., The AP-2 adapter protein complexes during vir- accuracy of reverse transcriptase from HIV-1, ion assembly, J. Virol. 72 (1998) 10218– Science 242 (1988) 1171–1173. 10221. [148] Robertson D.L., Hahn B.H., Sharp P.M., [137] Raabe M.R., Issel C.J., Montelaro R.C., Recombination in AIDS viruses, J. Mol. Equine monocyte-derived macrophage cul- Evol. 40 (1995) 249–259. tures and their applications for infectivity and [149] Rous P., A sarcoma of the fowl transmissible neutralization studies of equine infectious by an agent separable from the tumor cells, anemia virus, J. Virol. Methods 71 (1998) J. Exp. Med. 13 (1911) 397–411. 87–104. [150] Rwambo P.M., Issel C.J., Adams W.V. Jr., [138] Raabe M.L., Issel C.J., Cook S.J., Cook R.F., Hussain K.A., Miller M., Montelaro R.C., Woodson B., Montelaro R.C., Immunization Equine infectious anemia virus (EIAV) with a recombinant envelope protein (rgp90) humoral responses of recipient ponies and of EIAV produces a spectrum of vaccine effi- antigenic variation during persistent infec- cacy ranging from lack of clinical disease to tion, Arch. Virol. 111 (1990) 199–212. EIAV infection in equids 511

[151] Salguero F.J., Ruiz-Villamor E., Bautista [162] Steagall W.K., Robek M.D., Perry S.T., M.J., Sanchez-Cordon P.J., Carrasco L., Fuller F.J., Payne S.L., Incorporation of Gomez-Villamandos J.C., Changes in mac- uracil into viral DNA correlates with reduced rophages in spleen and lymph nodes during replication of EIAV in macrophages, Virol- acute African swine fever: expression of ogy 210 (1995) 302–313. cytokines, Vet. Immunol. Immunopathol. 90 [163] Stephens R.M., Casey J.W., Rice N.R., (2002) 11–22. Equine infectious anemia virus gag and pol [152] Salinovich O., Payne S.L., Montelaro R.C., genes: relatedness to visna and AIDS virus, Hussain K.A., Issel C.J., Schnorr K.L., Rapid Science 231 (1986) 589–594. emergence of novel antigenic and genetic [164] Stephens R.M., Derse D., Rice N.R., Cloning variants of equine infectious anemia virus and characterization of cDNAs encoding during persistent infection, J. Virol. 57 equine infectious anemia virus tat and puta- (1986) 71–80. tive Rev proteins, J. Virol. 64 (1990) 3716– [153] Schiltz R.L., Shih D.S., Rasty S., Montelaro 3725. R.C., Rushlow K.E., Equine infectious ane- [165] Sticht H., Willbold D., Bayer P., Ejchart A., mia virus gene expression: characterization Herrmann F., Rosin-Arbesfeld R., Gazit A., of the RNA splicing pattern and the protein Yaniv A., Frank R., Rosch P., Equine infec- products encoded by open reading frames S1 tious anemia virus Tat is a predominantly hel- and S2, J. Virol. 66 (1992) 3455–3465. ical protein, Eur. J. Biochem. 218 (1993) [154] Sellon D.C., Perry S.T., Coggins L., Fuller 973–976. F.J., Wild-type equine infectious anemia [166] Strack B., Calistri A., Craig S., Popova E., virus replicates in vivo predominantly in tis- Gottlinger H.G., AIP1/ALIX is a binding sue macrophages, not in peripheral blood partner for HIV-1 p6 and EIAV p9 function- monocytes, J. Virol. 66 (1992) 5906–5913. ing in virus budding, Cell 114 (2003) 689– [155] Sellon D.C., Fuller F.J., McGuire T.C., The 699. immunopathogenesis of equine infectious [167] Sune C., Goldstrohm A.C., Peng J., Price anemia virus, Virus Res. 32 (1994) 111–138. D.H., Garcia-Blanco M.A., An in vitro tran- [156] Sentsui H., Kono Y., Complement-mediated scription system that recapitulates equine hemolysis of horse erythrocytes treated with infectious anemia virus tat-mediated inhibi- equine infectious anemia virus, Arch. Virol. tion of human immunodeficiency virus type 1 95 (1987) 53–66. Tat activity demonstrates a role for positive [157] Shen R., Wang Z.M., Development and use transcription elongation factor b and associ- of an equine infectious aneamia donkey leu- ated proteins in the mechanism of Tat activa- kocyte attenuated vaccine, in: Tashjian R. tion, Virology 274 (2000) 356–366. (Ed.), Equine Infectious Anemia: a national [168] Swanstrom R., Wills J.W., Synthesis, assem- review of policies, programs, and future bly, and processing of viral proteins, in: objectives, Amarillo, Texas, 1985, pp. 135– Coffin J.M., Hughes S.H., Varmus H.E. 148. (Eds.), Retroviruses, Cold Spring Harbor [158] Sherman L., Gazit A., Yaniv A., Kawakami Laboratory Press, Cold Spring Harbor, New T., Dahlberg J.E., Tronick S.R., Localization York, 1997, pp. 263–334. of sequences responsible for trans-activation [169] Takada A., Kawaoka Y., Antibody-depend- of the equine infectious anemia virus long ent enhancement of viral infection: molecu- terminal repeat, J. Virol. 62 (1988) 120–126. lar mechanisms and in vivo implications, [159] Shibata R., Igarashi T., Haigwood N., Buckler- Rev. Med. Virol. 13 (2003) 387–398. White A., Ogert R., Ross W., Willey R., Cho [170] Tanzi G.O., Piefer A.J., Bates P., Equine M.W., Martin M.A., Neutralizing antibody infectious anemia virus utilizes host vesicu- directed against the HIV-1 envelope glyco- lar protein sorting machinery during particle protein can completely block HIV-1/SIV chi- release, J. Virol. 77 (2003) 8440–8447. meric virus infections of macaque monkeys, [171] Ten Haaft P., Verstrepen B., Uberla K., Nat. Med. 5 (1999) 204–210. Rosenwirth B., Heeney J., A pathogenic [160] Soutullo A., Verwimp V., Riveros M., Pauli threshold of virus load defined in simian R., Tonarelli G., Design and validation of an immunodeficiency virus- or simian-human ELISA for equine infectious anemia (EIA) immunodeficiency virus-infected macaques, diagnosis using synthetic peptides, Vet. J. Virol. 72 (1998) 10281–10285. Microbiol. 79 (2001) 111–121. [172] Tencza S.B., Islam K.R., Kalia V., Nasir [161] Spyrou V., Papanastassopoulou M., Psychas M.S., Jolley M.E., Montelaro R.C., Develop- V., Billinis C., Koumbati M., Vlemmas J., ment of a fluorescence polarization-based Koptopoulos G., Equine infectious anemia in diagnostic assay for equine infectious anemia mules: virus isolation and pathogenicity virus, J. Clin. Microbiol. 38 (2000) 1854– studies, Vet. Microbiol. 95 (2003) 49–59. 1859. 512 C. Leroux et al.

[173] Thomas D.A., Furman P.A., Purification and infectious anemia virus derived from a molec- kinetic characterization of equine infectious ular clone persistently infects horses, J. Virol. anemia virus reverse transcriptase, Biochem. 64 (1990) 5750–5756. Biophys. Res. Commun. 180 (1991) 1365– [185] Willbold D., Rosin-Arbesfeld R., Sticht H., 1371. Frank R., Rosch P., Structure of the equine [174] Threadgill D.S., Steagall W.K., Flaherty infectious anemia virus Tat protein, Science M.T., Fuller F.J., Perry S.T., Rushlow K.E., 264 (1994) 1584–1587. Le Grice S.F., Payne S.L., Characterization of equine infectious anemia virus dUTPase: [186] Yoon S., Kingsman S.M., Kingsman A.J., growth properties of a dUTPase-deficient Wilson S.A., Mitrophanous K.A., Character- mutant, J. Virol. 67 (1993) 2592–2600. ization of the equine infectious anaemia virus S2 protein, J. Gen. Virol. 81 (2000) 2189– [175] Tornquist S.J., Oaks J.L., Crawford T.B., 2194. Elevation of cytokines associated with the thrombocytopenia of equine infectious anae- [187] Zauli G., Re M.C., Davis B., Sen L., Visani mia, J. Gen. Virol. 78 (1997) 2541–2548. G., Gugliotta L., Furlini G., La Placa M., Impaired in vitro growth of purified (CD34+) [176] Tschetter J.R., Byrne K.M., Perryman L.E., McGuire T.C., Control of equine infectious hematopoietic progenitors in human immuno- anemia virus is not dependent on ADCC deficiency virus-1 seropositive thrombocy- mediating antibodies, Virology 230 (1997) topenic individuals, Blood 79 (1992) 2680– 275–280. 2687. [177] Tumas D.B., Hines M.T., Perryman L.E., [188] Zhang W., Lonning S.M., McGuire T.C., Davis W.C., McGuire T.C., Corticosteroid Gag protein epitopes recognized by ELA-A- immunosuppression and monoclonal anti- restricted cytotoxic T lymphocytes from body-mediated CD5+ T lymphocyte deple- horses with long-term equine infectious ane- tion in normal and equine infectious anaemia mia virus infection, J. Virol. 72 (1998) 9612– virus-carrier horses, J. Gen. Virol. 75 (1994) 9620. 959–968. [189] Zhang W., Auyong D.B., Oaks J.L., McGuire [178] Vallée H., Carré H., Sur la nature infectieuse T.C., Natural variation of equine infectious de l'anémie du cheval, C. R. Acad. Sci. 139 anemia virus Gag protein cytotoxic T lym- (1904) 331–333. phocyte epitopes, Virology 261 (1999) 242– [179] Van Regenmortel M.H., Mayo M.A., Fauquet 252. C.M., Maniloff J., Virus nomenclature: con- [190] Zheng Y.H., Nakaya T., Sentsui H., Kameoka sensus versus chaos, Arch. Virol. 145 (2000) M., Kishi M., Hagiwara K., Takahashi H., 2227–2232. Kono Y., Ikuta K., Insertions, duplications [180] Villet S., Bouzar B.A., Morin T., Verdier G., and substitutions in restricted gp90 regions of Legras C., Chebloune Y., Maedi-visna virus equine infectious anaemia virus during and caprine arthritis encephalitis virus genomes febrile episodes in an experimentally infected encode a Vpr-like but no Tat protein, J. Virol. horse, J. Gen. Virol. 78 (1997) 807–820. 77 (2003) 9632–9638. [191] Zheng Y.H., Sentsui H., Kono Y., Ikuta K., [181] Villet S., Faure C., Bouzar B.A., Morin T., Mutations occurring during serial passage of Verdier G., Chebloune Y., Legras C., Lack of Japanese equine infectious anemia virus in trans-activation function for Maedi-visna primary horse macrophages, Virus Res. 68 virus and Caprine arthritis encephalitis virus (2000) 93–98. Tat proteins, Virology 307 (2003) 317–327. [192] Zientara S., L'anémie infectieuse des [182] Von Schwedler U.K., Stuchell M., Muller B., équidés: déjà 3 foyers en 1999, Bull. G.T.V. Ward D.M., Chung H.Y., Morita E., Wang 1 (1999) 39–40. H.E., Davis T., He G.P., Cimbora D.M., Scott A., Krausslich H.G., Kaplan J., Morham [193] Zientara S., L'anémie infectieuse des S.G., Sundquist W.I., The protein network of équidés: la situation épidémiologique en HIV budding, Cell 114 (2003) 701–713. France depuis 1993, Bull. G.T.V. 1 (1999) 41–43. [183] Wang S.Z., Rushlow K.E., Issel C.J., Cook R.F., Cook S.J., Raabe M.L., Chong Y.H., [194] Zucker-Franklin D., Cao Y.Z., Megakaryo- Costa L., Montelaro R.C., Enhancement of cytes of human immunodeficiency virus- EIAV replication and disease by immuniza- infected individuals express viral RNA, Proc. tion with a baculovirus-expressed recom- Natl. Acad. Sci. USA 86 (1989) 5595–5599. binant envelope surface glycoprotein, Virol- [195] Zucker-Franklin D., Seremetis S., Zheng ogy 199 (1994) 247–251. Z.Y., Internalization of human immunodefi- [184] Whetter L., Archambault D., Perry S., Gazit ciency virus type I and other retroviruses by A., Coggins L., Yaniv A., Clabough D., megakaryocytes and platelets, Blood 75 Dahlberg J., Fuller F., Tronick S., Equine (1990) 1920–1923.