dh82 cells: a macrophage cell line for the replication and study of equine infectious anemia virus

Journal of Virological Methods 95 (2001) 47–56

DH82 cells: a macrophage cell line for the replication andstudy of equine infectious anemia virus

Robert Hines a, Wendy Maury b,*a Uni�ersity of South Dakota, Lee Medical Building, 414 E Clark St., Vermillion, SD 57069, USA

b Department of Microbiology, Uni�ersity of Iowa, 3-612 Bowen Science Building, Iowa City, IA 52242, USA

Received 14 November 2000; received in revised form 19 February 2001; accepted 19 February 2001

Abstract

In vivo, tissue macrophages have been implicated as an important cell for the replication of equine infectiousanemia virus (EIAV). Laboratory investigations of EIAV/macrophage interactions, however, have been hampered bythe laborious blood monocyte isolation procedures. In addition, adherent equine macrophage cultures generally havepoor long-term viability and are resistant to transfection. This report describes an adherent canine macrophage-likecell line, DH82, that supports the replication of EIAV. This cell line was easily transfectable and supported EIAV Tattransactivation of the LTR. Electrophoretic mobility shift assays were carried out to determine which transcriptionfactor binding sites within the LTR enhancer region were bound by DH82 nuclear extracts. It was found that fivedifferent motifs were occupied. The ets motifs that are bound by PU.1 in primary macrophage nuclear extractsspecifically interacted with DH82 nuclear extracts. In addition, the PEA-2, Lvb and Oct motifs that are occupied byfibroblast nuclear extracts were also bound by DH82 nuclear extracts. Finally, the methylation-dependent bindingprotein (MDBP) site that is bound by all nuclear extracts investigated to date demonstrated specific interactions withDH82 nuclear extracts. The observation that both macrophage-specific and fibroblast-specific motifs were utilized byDH82 nuclear extracts suggested that both macrophage-adapted and fibroblast-adapted EIAV could replicate inDH82 cells. Indeed, infectivity studies demonstrated that strains of virus that exclusively replicate in macrophages canreplicate in DH82 cells and fibroblast-adapted strains of virus can also replicate in these cells. Finally, these cellscould be transfected readily with the EIAV molecular clone, pSPeiav19-2, and virus spread was detected within theculture. In conclusion, this study has identified a useful cell line that should facilitate the study of EIAV expressionand replication. © 2001 Elsevier Science B.V. All rights reserved.

Keywords: Equine infectious anemia virus; Methylation-dependent binding protein; Transfection; TAT transactivation; EIAV; DH82 cells

www.elsevier.com/locate/jviromet

1. Introduction

Equine infectious anemia virus (EIAV) is aretrovirus within the subfamily of Lentiviridae.EIAV, unlike other members of this subfamily,causes an acute disease rather than a slow pro-

* Corresponding author. Tel.: +1-319-3358021.E-mail address: [email protected] (W. Maury).

0166-0934/01/$ - see front matter © 2001 Elsevier Science B.V. All rights reserved.

PII: S 0166 -0934 (01 )00288 -9

R. Hines, W. Maury / Journal of Virological Methods 95 (2001) 47–5648

gressive one. Within 2–6 weeks of the initialinfection, viremia can be detected at high levelswithin the horse. This burst of viral production isusually short lived and is accompanied by feverand thrombocytopenia (Cheevers and McGuire,1985; Clabough et al., 1991; Sellon, 1993). Subse-quent viremic episodes can, but do not always,occur and tend to be less severe. Eventually,viremia is controlled and although culturable in-fectious virus becomes undetectable, horses re-main seropositive for life (Oaks et al., 1998) withviral and proviral sequences remaining detectableby PCR (Oaks et al., 1998; Maury et al., 1997;Langemeier et al., 1996).

Equine infectious anemia virus, like all otherlentiviruses, replicates in cells of the monocyte/macrophage lineage in vivo and in vitro (Sellon etal., 1992; Bagasra and Pomerantz, 1993; Englishet al., 1993; Mori et al., 1992; Narayan and Zink,1992). Infection of macrophages is believed to beimportant in EIAV pathogenesis and high levelsof EIAV replication have been detected in tissuemacrophages during acute viremia (McGuire etal., 1971; Sellon et al., 1992, 1996). Lower levelsof replication have also been observed inmacrophages during viral persistence (Oaks et al.,1998). Virulent strains of EIAV have a strong celltropism for macrophages (McGuire et al., 1971;Sellon et al., 1992). Thus, it is important to studyEIAV/macrophage interactions. However, suchstudies have been limited due to difficult andlaborious monocyte isolation procedures and thelimited viability of these cells in vitro. Thus, es-tablishment of an infectable macrophage culturesystem has been a goal of many EIAV laborato-ries. In this study, we identified and characterizedEIAV infection of an adherent caninemacrophage cell line. This cell line, DH82, wastransfectable, supported amplification of a molec-ular infectious clone of EIAV and readily pro-duced infectious virus from both macrophage-and fibroblast-tropic strains of EIAV.

2. Material and methods

2.1. Cell lines and �irus strains used

The DH82 cell line was established in 1988

from the neoplastic progenitor cells of caninemalignant histiocytosis (ATCC No. CRL-10389)(Wellman et al., 1988). Neoplastic cells of malig-nant histiocytosis have been shown to belong tothe mononuclear-phagocyte lineage (Huhn andMeister, 1978; Vilpo et al., 1980). DH82 is anadherent cell line that is able to phagocytose latexbeads and is positive for alpha napthyl acetateesterase, acid phosphatase and Fc receptors (Well-man et al., 1988). DH82 cells were maintained inDMEM+15% FCS at 37°C and split 1:5 withtrypsin/versene two to three times per week.Cf2Th cells (ATCC No. CRL-1430) (Nelson-Reeset al., 1976) were maintained in DMEM+5%FCS at 37°C and split 1:20 with trypsin/versenetwo to three times per week. Primary equine renalendothelial cells (Maury et al., 1998) were main-tained in DMEM+40% FCS at 37°C and split1:5 with trypsin/versene two to three times perweek.

The viral strains Th.1 and MA-1 have beendescribed in detail previously (Carpenter andChesebro, 1989). Briefly, Th.1 was generated byco-culturing peripheral blood mononuclear cellsfrom an experimentally EIAV-infected horse dur-ing a febrile episode with uninfected equinemacrophages. This virus is macrophage tropic andreplicates well in equine macrophages, but notequine dermal (ED) cells (Carpenter and Chese-bro, 1989). Th.1 repeatedly passaged blindly ontoequine dermal cells followed by passage throughCf2Th cells resulted in a stock that was biologi-cally cloned to generate MA-1. MA-1 does notreplicate in horse mononuclear cells, but replicateswell in equine dermal cells with no apparent cyto-pathic effects. While the virulence of Th.1 hasnever been studied in vivo, it has been demon-strated that MA-1 does not induce clinical diseasein experimentally infected horses (Carpenter andChesebro, 1989). The macrophage-tropic molecu-lar clone pSPeiav19-2 (Payne et al., 1998) waspreviously generated by combining sequencesfrom an avirulent molecular clone and the highlyvirulent Wyoming virus field strain.

2.2. Transfections and reporter gene assays

DH82 cells (2.8×105) were plated in a six-well

R. Hines, W. Maury / Journal of Virological Methods 95 (2001) 47–56 49

tray with DMEM+15% FCS and allowed togrow overnight. Cells were transfected the fol-lowing day with either a �-galactosidase expres-sion vector (pCMV/� gal), or with 3 �g of anLTR/CAT construct in combination with 200 ngof pCMV/� gal using the GenePORTER™transfection reagent (GTS San Diego, CA, USA)or the calcium phosphate procedure (Graham etal., 1973). All transfections containing the LTR/CAT reporter plasmid were carried out both inthe presence and absence of 1 �g of an EIAVtat expression plasmid, pRSV-Etat, as noted(Dorn et al., 1990). In wells where pRSV-Etatwas not added, 1 �g of salmon sperm DNA wasadded to maintain equivalent concentrations ofDNA in all transfections. Transfections usingGenePORTER™ were carried out in a total vol-ume of 1 ml serum-free DMEM per well follow-ing manufacturer’s instructions. DH82 cellstransfected by the calcium phosphate procedure(Graham et al., 1973) were transfected with 5 �gof CMV-� gal onto six-well trays containing2.8×105 DH82 cells/well. At 4–6 h post trans-fection, cells were glycerol shocked by treatingthem for 2 min with 15% glycerol in phosphatebuffered saline (PBS). At 48 h post transfection,cells were harvested and assayed either for �-galactosidase activity or � galactosidase andchloramphenicol acetyltransferase activity.Transfections were all done in duplicate and re-peated three times.

�-Galactosidase assays were carried out usingthe �-galactosidase enzyme assay system (cat.no. E2000; Promega Madison, WI, USA) permanufacturer’s instructions. Optical densitieswere read at 414 nm using a 96-well platereader. For chloramphenicol acetyltransferase as-says, cell lysates were normalized for �-galactosi-dase activity and incubated with [14C]-chloram-phenicol and acetyl coenzyme A as described byGorman et al. (1982). All chloramphenicolacetyltransferase assays were performed in thelinear range of the assay. Acetylated and un-acetylated [14C]-chloramphenicol was separatedby thin-layer chromatography using Kodak thin-layer sheets. The acetylation pattern was iden-tified and quantified using the Packard Instrum-ents Instant Imager and the amount of chloram-

phenicol acetyltransferase activity was expressedas the percent acetylation/hour. The fold activa-tion in the presence of Tat was determined bydividing the rate of acetylation in the presenceof Tat by the rate of acetylation in the absenceof Tat.

2.3. Infection studies and immunostaining of in-fected cells

DH82 cells (4.5×104) per well were plated in24 well plates with DMEM+15% FCS and al-lowed to grow overnight. The following day, 1000infectious virions from titered stocks of MA-1 andTh.1 were added to their respective wells andgrown in DMEM+15% FCS at 37°C. At timepoints 24, 48, 72 and 96 h post-infection, cellswere fixed by adding 1 ml of 75:25% — acetone/water for 10 min and then washed 3× with PBS.To stain cells, horse anti-EIAV antisera diluted1:700 in DMEM+5% FCS was used as the pri-mary antibody solution. The secondary antibodysolution consisted of HRP-conjugated goat anti-horse antibody (ICN) diluted 1:500 in DMEM+5% FCS. To visualize antigen-expressing cells,aminoethylcarbazole was used as a substrate forthe peroxidase reaction and positivity was scoredmicroscopically in each well. Percent of cells in-fected was determined by the equation X= (P/T)100 where X=percent positive, P= thenumber of positive cells in a given field andT= the total number of cells in the same field. Totiter the various EIAV strains, serial dilutions ofviral stocks were added to cells and incubated for48 h at 37°C. Cells were subsequently fixed andstained as described above. Titers were deter-mined by the equation T= (1000 �l×p)/s whereT= titer, p=number of positive cells in the well,and s= the volume of supernatant added to eachwell.

2.4. Nuclear extract preparations and elec-trophoretic mobility shift assays

Nuclear extracts were made from DH82 cellsusing a procedure outlined by Dignam et al.(1983) and modified by Quinn et al. (1987). Dou-


ble-stranded oligonucleotides were made by an-nealing complimentary synthetic oligonucleotideswith 5� overhanging ends. Oligonucleotides wereblunt ended using three unlabeled deoxynucle-otide triphosphates, one 32P-labeled deoxynucle-otide triphosphate and the Klenow fragment ofDNA polymerase I. Radiolabeled oligonucle-otides (20 000 cpm) were incubated with approxi-mately 10 �g of DH82 nuclear extract in a totalvolume of 20 �l containing 4 �g poly(dI-dC), 100mM KCl and 1mM MgCl2 for 20–25 min. Boundvs. unbound oligonucleotides were separated on a1× TBE 5% polyacrylamide gel (80:1 — acryl-amide/bis) and run at 150 V. Binding was deter-mined to be specific based on the fact that a200-fold molar excess of unlabeled self was able tocompete off the bound complex whereas the samemolar excess of a non-specific oligonucleotide wasunable to do so.

3. Results

3.1. DH82 cells are highly transfectable and areresponsi�e to EIAV Tat

For a macrophage cell line to be useful for thestudy of EIAV, not only does it need to support aproductive infection of the virus, it is importantthat it be transfectable as well. Two methods wereexamined to evaluate the transfection efficiency ofDH82 cells. Both liposome-based and calciumphosphate-based transfection procedures yieldedreadily detectable levels of �-gal activity whentransfected with either 200 ng or 5 �g, respec-tively, of a CMV �-gal expression vector (data notshown). The liposome-based transfection gaveconsistently higher levels of �-gal activity, so thisprotocol was used throughout the remainder ofthe study. DH82 cells were transfected with in-creasing concentrations of the pCMV/�-gal. Thetitration curve generated demonstrated a dose-de-pendent level of �-galactosidase expression (Fig.1a).

Because of the species specificity of EIAV Tattransactivation (Maury et al., 1994), it was impor-tant to determine if these canine cells could sup-port transactivation of the LTR by EIAV Tat.

Transient transfections using an LTR/CAT re-porter gene in the presence and absence of EIAVTat were therefore performed. As shown in Fig.1b, the increase of chloramphenicol acetyltrans-ferase activity in the presence of Tat was approxi-mately 170 times basal levels demonstrating thatDH82 cells strongly support EIAV Tat-transactivation.

3.2. DH82 nuclear extracts interact withnumerous motifs within the EIAV LTR enhancer

The enhancer region within the EIAV LTR hasbeen shown to be hypervariable both in vivo andin vitro (Payne et al., 1999; Carpenter et al., 1991;Maury et al., 1997). As a consequence of thishypervariation, the LTR appears be a major de-terminant of EIAV cell tropism because of thecell-specific usage of transcription factors by theLTR (Maury, 1994; Maury et al., 2000). In pri-mary macrophages, the macrophage-specific, etstranscription factor family member, PU.1, ex-hibits binding to multiple sites within the EIAVenhancer and this interaction is important forLTR expression in these cells (Maury, 1994).However, in fibroblasts other partially overlap-ping transcription factor binding sites have beenshown to be important for viral expression(Maury et al., 2000). In those cells, the PEA-2,Lvb, octamer, and cAMP response element(CRE) sites are bound by nuclear proteins todrive viral replication. Electrophoretic mobilityshift assays were performed to identify transcrip-tion factor motifs within the EIAV enhancer thatwere bound by DH82 nuclear extracts. A series ofdouble-stranded oligonucleotides were used todefine the DH82 nuclear protein/DNA interac-tions (Fig. 2a). The transcription factor bindingsites present within each oligonucleotide havebeen previously characterized (Maury, 1994;Maury et al., 2000). DH82 nuclear extract boundspecifically to oligonucleotides 91, 111, 116, 129and 3�PU.1 (Fig. 2b). The 3� PU.1 oligonucleotidecontains an ets site that is adjacent to the invari-ant TATA box within the LTR. This motif isbound by macrophage nuclear extracts, but is notbound by nuclear extract from other cell typesthat can be infected productively with EIAV. In


addition to the ets site, DH82 nuclear extractsinteracted with oligonucleotides that contain themethylation-dependent binding site, PEA-2, Lvband octamer sites. Fibroblast nuclear extract hadbeen shown previously to interact with thesesites (Carvalho and Derse, 1993; Maury et al.,2000). Interestingly, the CRE site that is occu-

pied by fibroblast nuclear extracts did not inter-act with DH82 nuclear extracts. In total,electrophoretic mobility shift assay data indi-cated that DH82 cells contain transcription fac-tors that interact with motifs that had previouslybeen characterized as either macrophage- orfibroblast-specific.

Fig. 1. (a) Effect of DNA concentration on �-galactosidase reporter gene activity in DH82 cells. 100, 200, 400, 600, 800, or 1000ng of the plasmid CMV �-gal were transfected into DH82 cells and �-galactosidase activity was assayed at 48 h, post transfection.(b) EIAV Tat responsiveness of DH82 cells. Responsiveness of EIAV Tat transactivation was measured by transfecting 3 �g of anEIAV LTR/chloramphenicol acetyltransferase reporter plasmid in the presence or absence of 1 �g of a RSV/EIAV Tat expressionvector. At 48 h post transfection, cells were harvested and chloramphenicol acetyltransferase activity was measured. Chlorampheni-col acetyltransferase activity was normalized to �-galactosidase activity. These data were from a single representative experimentdone in triplicate.


Fig. 2. The interaction of DH82 nuclear extracts with EIAV transcription factor binding motifs. (a) The oligonucleotides and sensestrand sequence of probes and competitors used in the electrophoretic mobility shift assays. The names of previously identifiedtranscription factor binding motifs are noted and the sequences are bolded within each oligonucleotide. (b) Binding of DH82 nuclearextract to EIAV enhancer region oligonucleotides in an electrophoretic mobility shift assay. The oligonucleotides used as probes aredisplayed. Competitor oligonucleotides demonstrate specificity of binding. Lanes either contained no competitor (None), a 200-foldmolar excess of unlabeled but otherwise identical oligonucleotide (Self), or a 200-fold molar excess of an unlabeled non-specificoligonucleotide either 580 from the Gag region or 91 from the beginning of the enhancer region. Specific binding of DH82 nuclearextract was observed for oligonucleotides 91, 111, 116, 129 and 3�PU.1.


Fig. 3. The ability of DH82 cells to support productive EIAV infection. (a) A time course of infection with the macrophage tropicstrain Th.1 and the fibroblast tropic strain MA-1 demonstrating that these viral strains were able to infect and spread within aculture of DH82 cells. One thousand infectious virions initiated infection and cells were fixed at the stated time points. Cells wereimmunostained for viral antigens and the percentages of antigen-positive cells were scored for each time point. By 96 h post infectionboth strains of the virus had infected a similar number of cells. (b) DH82-grown MA-1 was titered onto naı̈ve DH82 cells andprimary equine renal endothelial cells (EREC). A stock of MA-1 was generated in DH82 cells. Supernatants were collected,centrifuged twice to remove cellular debris and titered onto primary equine renal endothelial cells and DH82 cells. Viral titers weredetermined by EIAV antigen immunostaining. Standard deviations are shown in parentheses.

3.3. DH82 cells are infectable with macrophage-and fibroblast-tropic strains of EIAV

Because DH82 cells contained a transcriptionfactor repertoire that was a composite of bothmacrophages and fibroblasts, it was necessary todetermine if both fibroblast-and macrophage-tropic strains of the virus could replicate in thesecells. Therefore, cells were infected with twostrains of EIAV that had previously been charac-terized for cell tropism. Th.1 is a macrophage-tropic strain of EIAV and MA-1 is an avirulent,fibroblast-tropic strain derived from Th.1 (Car-penter and Chesebro, 1989). DH82 cells wereinfected with 1000 infectious virions per well of

titered Th.1 and MA-1. At 24, 48, 72 and 96 hpost-infection, the cells were stained and scoredfor the percentage of antigen positive cells (Fig.3a). By 96 h post-infection, both Th.1 and MA-1infected DH82 cells were approximately 10% pos-itive. Over time the infection continued to spreadthrough the culture, with 20–60% of the cellseventually becoming positive for EIAV antigens;however, cultures never became 100% infected.Nonetheless, this experiment demonstrated thatDH82 cells were permissive to infection by notonly macrophage-tropic strains, but also tofibroblast-tropic strains of the virus, highlightingthe versatility of these cells. Repeated attempts todetect infection of a stock of the highly virulent


Wyoming strain of EIAV, 2078, in DH82 cellswas unsuccessful.

To determine if infected DH82 cells producedinfectious virus, cells were infected with MA-1and maintained in culture for 3–4 weeks. Super-natants were collected from these cultures andtitered on to naive DH82 cells as well as primaryequine renal endothelial cells. As shown in Fig.3b, the titers of virus were similar in both primaryequine renal endothelial cells and DH82 cells dif-fering only by a factor of 2. This demonstrates theability of EIAV-infected DH82 cells to produceinfectious virus.

3.4. DH82 cells support replication of infectiousmolecular clones of EIAV

Several molecular clones of EIAV have recentlybeen developed (Payne et al., 1998; Cook et al.,1998; Payne et al., 1994; Whetter et al., 1990).These clones have allowed laboratories to identifyand characterize regions of EIAV important for invivo virulence. However, due to the limited capac-ity of these clones to replicate in fibroblastic celllines and the poor transfectability of primarymacrophages and endothelial cells, these molecu-lar clones have not proven to be useful tools fordissecting the genome in vitro. To determine if theEIAV molecular clone pSPeiav19-2 (Payne et al.,

1998) could be productively introduced intoDH82 cells, this construct was transfected intoDH82 cells and the spread of virus was monitoredby periodic EIAV antigen immunostaining. Thismolecular clone expressed virus antigens andspread within the culture. By day 26, post-trans-fection, the number of infected cells within theculture plateaued at approximately 6% (Fig. 4).

Because of the possibility that the limited infec-tivity of EIAV in these cells was due to a permis-sive subpopulation within the culture, the DH82cells were cloned biologically by introducing asingle cell into each well of a series of 96 welltrays. These subclones were tested for enhancedinfectivity and transfection efficiency. Interest-ingly, transfection and virus spread of pSPeiav19-2 DNA through these subcloned cultures resultedin infection of approximately 5% of the cells (datanot shown). Thus, biological cloning of the DH82population did not enhance the infectivity of thecell population.

4. Discussion

The absence of a transfectable and EIAV-in-fectable macrophage cell line has hampered stud-ies on EIAV pathogenesis. This study identified amacrophage cell line, DH82, that is easily trans-

Fig. 4. The spread of the molecular clone, pSPeiav19-2, in DH82 cells. pSPeiav19-2 was transfected into DH82 cells on day 0. Thepercentage of EIAV antigen positive cells was monitored by periodic immunostaining. Over the course of 26 days, virus was foundto spread in the culture. By day 26, 5.8% of the cells were positive for virus antigen.


fectable and infectable with EIAV. Furthermore,these canine cells were found to be EIAV Tatresponsive. This is important to note becauseEIAV Tat does not work in either human ormurine cells due to the incompatibility of thecellular cyclin T1 (Bieniasz et al., 1999; Taube etal., 2000). This incompatibility rules out effec-tively most of the macrophage cell lines available.

This study also characterized the interactionsbetween the EIAV enhancer region and DH82nuclear proteins. The enhancer motifs that werebound by DH82 nuclear extracts have beenshown previously to be utilized in a cell-specificfashion. The utilization of both macrophage-spe-cific and fibroblast-specific motifs by DH82 cellssuggested that both fibroblast-tropic andmacrophage-tropic strains of the virus wouldreplicate in these cells. Indeed, it was found thatboth Th.1 and MA-1 strains of EIAV did repli-cate similarly in DH82 cells.

Finally, these cells could be transfected with theinfectious molecular clone, pSPeiav19-2, and aspreading infection of this virus could be estab-lished. The ability of this cell to serve as both aninitial cell for transfection and subsequently allowinfectious clones to spread within the culture willbe useful for molecular studies of EIAV. Thus,DH82 cells may serve as an excellent cell fordetermining viral titers of EIAV. Furthermore,DH82 cells provide the opportunity to study thevirus in the context of a macrophage-like cell.

Acknowledgements

We thank Dr C. Martin Stoltzfus for his helpfulcomments on this manuscript. This work wassupported by grants CA 72063 and AI44638 fromthe National Institutes of Health to WJM.

References

Bagasra, O., Pomerantz, R.J., 1993. Human immunodeficiencyvirus type 1 provirus is demonstrated in peripheral bloodmonocytes in vivo: a study utilizing an in situ polymerasechain reaction. AIDS Res. Hum. Retroviruses 9, 69–76.

Bieniasz, P.D., Grdina, T.A., Bogerd, H.P., Cullen, B.R.,1999. Highly divergent lentiviral Tat proteins activate viral

gene expression by a common mechanism. Mol. Cell. Biol.19, 4592–4599.

Carpenter, S., Alexandersen, S., Long, M.J., Perryman, S.,Chesebro, B., 1991. Identification of a hypervariable regionin the long terminal repeat of equine infectious anemiavirus. J. Virol. 65, 1605–1910.

Carpenter, S., Chesebro, B., 1989. Change in host cell tropismassociated with in vitro replication of equine infectiousanemia virus. J. Virol. 63, 2492–2496.

Carvalho, M., Derse, D., 1993. Physical and functional char-acterization of transcriptional control elements in theequine infectious anemia virus promoter. J. Virol. 67,2064–2074.

Cheevers, W.P., McGuire, T.C., 1985. Equine infectious ane-mia: immunopathogenesis and persistence. Rev. Infect.Dis. 7, 83–88.

Clabough, D.L., Gebhard, D., Flaherty, M.T., et al., 1991.Immune-mediated thromocytopenia in horses infected withequine infectious anemia virus. J. Virol. 65, 6242–6251.

Cook, R.F., Leroux, C., Cook, S., et al., 1998. Developmentand characterization of an in vivo pathogenic molecularclone of equine infectious anemia virus. J. Virol. 72, 1383–1393.

Dignam, J.D., Martin, P.L., Shastry, B.S., Roeder, R.F., 1983.Eukaryotic gene transcription with purified components.Methods Enzymol. 101, 582–599.

Dorn, P., LaSilva, L., Martarano, L., Derse, D., 1990. Equineinfectious anemia virus Tat: insights into the structure,function and evolution of lentivirus trans-activatorproteins. J. Virol. 64, 1616–1624.

English, R.V., Johnson, C.M., Gebhard, D.H., Tompkins,M.B., 1993. In vivo lymphocyte tropism of feline im-munodeficiency virus. J. Virol. 67, 5175–5186.

Gorman, C., Moffat, L.F., Howard, B., 1982. Recombinantgenomes which express chloramphenicol acetyltransferasein mammalian cells. Mol. Cell. Biol. 2, 1044–1051.

Graham, F.L., van der Eb, A.J., 1973. A new technique forthe assay of infectivity of human adenovirus 5 DNA.Virology 52, 456–467.

Huhn, D., Meister, P., 1978. Malignant histiocytosis. Morpho-logic and cytochemical findings. Cancer 42, 1341–1349.

Langemeier, J.L., Cook, S.J., Cook, R.F., Rushlow, K.E.,Montelaro, R.C., Issel, C.J., 1996. Detection of equineinfectious anemia viral RNA in plasma samples fromrecently infected and long term inapparent carrier animalsby PCR. J. Clin. Microbiol. 34, 1481–1487.

Maury, W., 1994. Monocyte maturation controls expression ofequine infectious anemia virus. J. Virol. 68, 6270–6279.

Maury, W., Bradley, S., Wright, B., Hines, R., 2000. Cellspecificity of the transcription-factor repertoire used by alentivirus: motifs important for expression of equine infec-tious anemia virus in nonmonocytic cells. Virology 267,267–278.

Maury, W., Carpenter, S., Graves, K., Chesebro, B., 1994.Cellular and viral specificity of equine infectious anemiavirus tat transactivation. Virology 200, 632–642.


Maury, W., Oaks, L.J., Bradley, S., 1998. Equine endothelialcells support productive infection of equine infectious ane-mia virus. J. Virology 72, 9291–9297.

Maury, W., Perryman, S., Oaks, L.J., et al., 1997. Localizedsequence heterogeneity in the long terminal repeats of invivo isolates of equine infectious anemia virus. J. Virol. 71,4929–4937.

McGuire, T.C., Crawford, T.B., Henson, J.B., 1971. Immu-nofluorescent localization of equine infectious anemia virusin tissue. Am. J. Pathol. 62, 283–294.

Mori, K., Ringler, D.J., Kodama, T., Derosiers, R.C., 1992.Complex determinants of macrophage tropism in env ofsimian immunodeficiency virus. J. Virol. 66, 2067–2075.

Narayan, O., Zink, M.C., 1992. Infection and dysfunction ofmonocytes induced by experimental inoculation of calveswith bovine immunodeficiency-like virus. J. Acquired Im-mune Defic. Syndr. 5, 1009–1015.

Nelson-Rees, W.A, Owens, R.B., Arnstein, P., Kniazeff, A.J.,1976. Source, alterations, characteristics and use of a newdog cell line (Cf2Th). In Vitro 12, 665–669.

Oaks, L.J, McGuire, T.C., Ulibarri, C., Crawford, T.B., 1998.Equine infectious anemia virus is found in tissuemacrophages during subclinical infection. J. Virol. 72,7263–7269.

Payne, S.L., La Celle, K., Pei, X.F., et al., 1999. Long terminalrepeat sequences of equine infectious anaemia virus are amajor determinant of cell tropism. J. Gen. Virol. 80,755–759.

Payne, S.L., Qi, X.M., Shao, H., Dwyer, A., Fuller, F.J., 1998.Disease induction by virus derived from molecular clonesof equine infectious anemia virus. J. Virol. 72, 483–487.

Payne, S.L., Rausch, J., Rushlow, K., et al., 1994. Characteri-zation of infectious molecular clones of equine infectiousanemia virus. J. Gen. Virol. 75, 425–429.

Quinn, J., Holbrook, N., Levens, D., 1987. Binding of acellular protein to the gibbon ape leukemia virus enhancer.Mol. Cell. Biol. 7, 2735–2744.

Sellon, D.C., 1993. Equine infectious anemia. Vet. Clin. NorthAm. Equine Pract. 9, 321–336.

Sellon, D.C., Perry, S.T., Coggins, L., Fuller, F.J., 1992.Wild-type equine infectious anemia virus replicates in vivopredominantly in tissue macrophages, not in peripheralblood monocytes. J. Virol. 66, 5906–5913.

Sellon, D.C., Walker, K.M., Russell, K.E., Perry, S.T., Cov-ington, P., Fuller, F.J., 1996. Equine infectious anemiavirus replication is upregulated during differentiation ofblood monocytes from acutely infected horses. J. Virol. 70,590–594.

Taube, R., Fujinaga, K., Irwin, D., Wimmer, J., Geyer, M.,Peterlin, B.M., 2000. Interactions between equine cyclinT1,Tat and TAR are disrupted by a leucine-to-valine substitu-tion found in human cyclin T1. J. Virol. 74, 892–898.

Vilpo, J.A., Klemi, P., Lassila, O., et al., 1980. Cytological andfunctional characterization of three cases of malignanthistiocytosis. Cancer 46, 1795–1801.

Wellman, M., Krakowka, S., Jacobs, R., Kociba, G., 1988. Amacrophage-monocyte cell line from a dog with malignanthistiocytosis. In Vitro Cell. Dev. Biol. 24, 223–229.

Whetter, L., Archambouit, D., Perry, S., et al., 1990. Equineinfectious anemia virus derived from a molecular clonepersistently infects horses. J. Virol. 64, 5750–5756.

.

dh82 cells: a macrophage cell line for the replication and study of equine infectious anemia virus

Documents