Xenoinfection of nonhuman primates by feline immunodeficiency virus James B. Johnston, Merle E. Olson, Erling W. Rud, and Christopher Power*

Size: px

Start display at page:

Download "Xenoinfection of nonhuman primates by feline immunodeficiency virus James B. Johnston*, Merle E. Olson*, Erling W. Rud, and Christopher Power*"

Neil Casey
6 years ago
Views:

1 Brief Communication 1109 Xenoinfection of nonhuman primates by feline immunodeficiency virus James B. Johnston*, Merle E. Olson*, Erling W. Rud, and Christopher Power* New viral infections in humans usually result from Results and discussion viruses that have been transmitted from other Although productive infection of primate cells by FIV has species as zoonoses. For example, it is accepted been demonstrated in vitro [2, 7, 8], transmission of the widely that human immunodeficiency virus (HIV) is virus to primates by natural infection has not been docu- the result of the propagation and adaptation of a mented.to assess the ability of FIV to infect primate cells simian immunodeficiency virus (SIV) from in vivo, we isolated primary peripheral blood mononuclear nonhuman primates to man [1]. Previously, we cells (PBMC) from two adult cynomolgus macaques and reported productive infection of primary human infected them with a virus derived from the Petaluma cells in vitro by feline immunodeficiency virus (FIV) strain of FIV, as described previously [2].After 3 days, [2], a lentivirus that causes an immunodeficiency FIV replication was confirmed by a reverse transcriptase syndrome in cats similar to HIV in humans [3]. The (RT) assay [2], and infected PBMC ( ) were returned present study extends these findings by to the same donor by intravenous transfusion in demonstrating that cynomolgus macaques (Macaca 3 ml phosphate-buffered saline.this protocol was chosen fasicularis) infected with FIV exhibited clinical for several reasons.first, the transmission of FIV under signs, including depletion of CD4 cells and weight natural conditions primarily involves a bite that introduces loss, that are consistent with FIV infection. The free viruses and infected cells directly into the blood- development of an antibody response to FIV gag- stream of the new host [3].Second, viral stocks produced encoded proteins and detection of virus-specific in a recipient animal s own PBMC (such stocks are termed sequences in sera, blood-derived cells, and an autologous virus) have been shown to result in higher necropsied tissue accompanied these changes. viral loads and more severe pathologies than nonautolo- Moreover, the reactivation of FIV replication from gous lentiviral preparations [9].Moreover, FIV derived latently infected cells was observed after from feline cells would contain envelope glycoproteins stimulation in vitro with phorbol esters and in vivo expressing galactosyl ( 1-3)galactosyl ( Gal).Since primates with tetanus toxoid. The proposed use of produce natural antibodies against this sugar, comwith lentiviruses in human gene therapy [4, 5] and of plement-mediated inactivation would likely reduce the nonhuman cells and organs in xenotransplantation potential for in vivo infection of primate cells by FIV [6] has raised concerns about zoonoses as potential produced by nonprimate cells [10].Finally, the use of sources of new human pathogens. Therefore, the PBMC that were productively infected with FIV allowed study of FIV infection of primate cells may provide the delivery of the virus in a manner that circumvented insight into the principles underlying retroviral the low titers of cell-free virus obtained from FIV-infected xenoinfections. primate PBMC [2]. Addresses: Departments of * Microbiology and Infectious Diseases and Clinical Neurosciences, University of Calgary, 3330 Hospital Drive Northwest, Calgary, Alberta T2N 4N1, Canada. National Laboratory for HIV Pathogenesis, Laboratory Center for Disease Control, LCDC Building, Ottawa, Ontario K1A 0L2, Canada. Correspondence: Christopher Power power@ucalgary.ca Received: 26 January 2001 Revised: 9 April 2001 Accepted: 21 May 2001 Published: 24 July 2001 Current Biology 2001, 11: Distinct clinical features, including wasting and depletion of peripheral CD4 cells, characterize the early stages of FIV infection in the natural host [3].Prior to being challenged with FIV, neither macaque showed consistent changes in body weight or blood cell population over a 6 week period.after infection, however, the initial body mass reached a reduction of 21.2% after 6 weeks for one macaque (CM1SP) and 10.7% after 3 weeks for the second subject (CM2ST) (Figure 1a).Consistent with the reported cytotoxicity after productive infection of primate cells by FIV in vitro [2, 11], a similar decreasing pattern was observed in CD4 cell levels at successive weeks postinfection (PI).FACS analysis [2] of blood samples collected over the time course revealed a sustained de /01/$ see front matter crease in CD4 cells, whose level reached a maximum 2001 Elsevier Science Ltd. All rights reserved. reduction of 72.3% at week 6 for CM1SP, while the level of CD8 cells was elevated over the same period (Figure

2 1110 Current Biology Vol 11 No 14 Figure 1 the primary targets of FIV in its feline host [3], also represented the principal primate cell type infected in vivo. Moreover, the severity of the CD4 cell depletion observed in FIV-infected macaques indicated that FIV infection was highly toxic to select primate cells in vivo. For the confirmation of the in vivo infection of primate cells by FIV, the presence of the FIV pol gene in blood and tissue was assessed by nested PCR and RT-PCR amplification of genomic DNA and total cellular RNA, respectively, with the primers and conditions described previously [2].Although FIV was not detectable in either macaque prior to infection (week 0), FIV-specific sequences were detected in PBMC from CM1SP until 9 weeks and in CM2ST until 4 weeks (Figure 2a).Similarly, RT-PCR revealed the presence of FIV RNA in sera collected from each animal over the same period, although viral sequences were only detectable for 4 weeks in CM1SP and 2 weeks in CM2ST (Figure 2a).An analysis of necropsied tissue from each macaque demonstrated that FIV sequences were also present in RNA and DNA isolated from spleen from CM1SP and in RNA from both lymph node and spleen tissue from CM2ST, but not in any other tissues investigated (Figure 2b).It is unlikely that these results represented residual cells from the initial transfusion of infected PBMC.Animals were transfused with only cells, of which approximately 10% were infected [2], and the cytotoxicity associated with FIV infection would have reduced this number further. In addition, the abundance of viral DNA detected in PBMC varied over the time course, and this variation indicates changes in the number of infected cells. FIV infection is associated with weight loss and depletion of CD4 cells. Cynomolgus macaques were infected with FIV, body weight was The above findings extended earlier in vitro results by measured, and blood, from which mononuclear cells were isolated, was harvested at successive weeks PI. The same parameters were demonstrating that FIV could infect nonhuman primate assessed for 6 weeks before infection. CM2ST received intramuscular cells in vivo and induce symptoms characteristic of FIV injections of tetanus toxoid (0.25 ml) at weeks 9 and 10 PI (arrows), infection.however, differences were observed between while CM1SP remained untreated after the initial infection. (a) Weight hosts.although CM1SP exhibited sustained weight loss loss, expressed as a percentage of the initial weight of each macaque (approximately 28 pounds), was observed with both animals and and CD4 cell depletion over the 12 week period, these reached maximal values of 11% 20%. (b) CD4 and CD8 cell parameters were decreased only transiently for CM2ST levels in mononuclear cells isolated at each time point were determined and began to return to preinfection values after week 3 by flow cytometry. Results are expressed as percent changes relative (Figure 1).Similarly, FIV remained detectable in the seto the abundance of CD4 (approximately 20%) and CD8 (approximately 30%) cells detected in each animal prior to infection. rum and PBMC of CM1SP for a longer period of time Cells incubated either in the absence of antibodies or with 1 g/ml than CM2ST (Figure 2), and this result suggests that of isotype-matched murine IgG 1 served as controls for each animal. CM2ST was able either to clear the virus or to control A marked decline in CD4 cells exceeded 70% and was detected the infection more efficiently.since latent infection of in both animals. primate cells has been observed in vitro [12], the question of viral persistence was addressed by the stimulation of PBMC isolated from each macaque with phorbol-12-1b).no change in the level of CD8 cells was observed myristate-13-acetate (PMA; Sigma, Oakville, Ontario, for CM2ST, but CD4 levels decreased by 76.5% within Canada). Like primate lentivirus genomes, the Petaluma the first 3 weeks PI (Figure 1b).Compared to preinfection genome contains promoter elements that bind phorbol values, monocyte and neutrophil levels were decreased ester-activated transcription factors, such as members of by 10% and 5%, respectively, after infection in both macaques, but the abundance of these cells did not vary over was not detectable by RT-PCR analysis of PBMC har- the activated protein (AP)-1 family [13].Although FIV the time course.these results suggested that CD4 cells, vested from CM2ST at week 6 PI, viral RNA (Figure 2c)

3 Brief Communication 1111 Figure 2 and elevated RT activity (not shown) were detected in cultures of these cells after stimulation with PMA.Similar results were observed with PBMC obtained from CM1SP at week 12 PI (Figure 2c) and suggest that latently infected cells persisted in both animals.to investigate this hypothesis in vivo, CM2ST was immunized at weeks 9 and 10 PI with tetanus toxoid (TTX; Connaught Labs, Seattle, Washington), a compound shown previously to activate lentiviral replication in latently infected cells [14]. Following treatment with TTX, weight loss was observed again in CM2ST and had decreased approximately 10% by week 12 PI (Figure 1a).Although a similar decline in CD4 cells was not detected in these cells, the increase that had been observed from week 3 to week 9 was arrested (Figure 1b).Moreover, FIV genome, which was not detectable in either PBMC or serum beyond 4 weeks PI before CM2ST was immunized, was again detected at week 12 (Figure 3a).These results indicate that FIV persisted in immunocompetent primate hosts at levels below the limit of detection of the PCR protocol (approximately 10 copies [2]) for a minimum of 3 months after initial exposure to the virus.moreover, an activating stimulus could induce further replication and spread of the virus. Detection of FIV-specific sequences in infected macaques. Necropsied tissue, PBMC, and sera were harvested from FIV- infected macaques, and genomic DNA and total RNA, from which cdna was prepared, were isolated. The presence of viral sequences was determined by nested PCR amplification and Southern blot detection of the FIV pol gene. Amplification of a water blank (W) and the GAPDH gene served as controls for contamination and equal template loading, respectively. Representative blots from three reactions are shown. (a) PCR analysis detected FIV sequences for up to 9 weeks in PBMC and 4 weeks in serum isolated from CM1SP. FIV sequences were detected for up to 4 weeks in PBMC and 2 weeks in serum isolated from CM2ST. After treatment with tetanus toxoid at weeks 9 and 10 PI, viral sequences were again found in PBMC and in serum from CM2ST. (b) Bone marrow (BM), bowel (B), brain (BR), kidney (K), lung (L), lymph node (LN), and spleen (S) tissue was harvested at week 12 PI. Lymph nodes were obtained from the bowel and were remote from the site of injection. FIV sequences were detected in genomic DNA and RNA isolated from spleen tissue from CM1SP and in RNA isolated from spleen and lymph node from CM2ST. (c) PBMC were isolated from CM1SP and CM2ST at weeks 12 and 6, respectively, and cultured for 48 hr in the presence ( ) or absence (-) of PMA (50 ng/ml). RNA isolated from FIV-infected feline PBMC (C) and from each PBMC preparation prior to the creation of cdna ( ) served as controls. PMA stimulation induced FIV gene expression in PBMC harvested from either animal. Because numerous host factors, such as virus-specific immune responses, influence the persistence of lentiviruses in vivo, the production of anti-fiv antibodies by infected macaques was investigated by Western blot analysis [15]. Using sera collected from CM2ST at weeks 6 and 12 PI as sources of antibodies, we detected prominent bands at 25 kda and 55 kda in protein extracts derived from the blood of a Petaluma-infected adult cat, with additional bands observed at higher molecular weights (Figure 3b and 3c).Similar bands were detectable in protein extracts from DH5 bacteria expressing Petaluma gag-encoded proteins but not in protein isolated from untransformed DH5 bacteria.a monoclonal antibody raised against the Petaluma p25 capsid protein (clone 43-1E2; kindly pro- vided by Dr.N.Pedersen, University of California, Davis, California) detected the same bands at 25 kda and 55 kda (Figure 3d).These findings suggest that antibodies against the FIV p25 capsid protein and the p55 Gag polyprotein were present in serum from CM2ST.No immu- noreactivity was detected with serum collected from CM2ST before infection (Figure 3a) or from CM1SP at any of the time points (not shown).these findings indi- cate that antibody responses to FIV varied between host animals and suggest that, compared to the sustained changes exhibited by CM1SP, the transient CD4 cell depletion and weight loss observed with CM2ST may have resulted from differences in the ability to mount an adaptive immune response and control the spread of the virus.this concept is supported by previous reports that a rapid and sustained decrease, analogous to that observed with CM1SP, in circulating CD4 cells abrogates the

4 1112 Current Biology Vol 11 No 14 Figure 3 Detection of anti-fiv antibodies in sera from infected macaques. Protein extracts were prepared from cultures of DH5 bacteria transformed with plasmids encoding FIV gagencoded polyproteins (capsid), separated by SDS-PAGE, and transferred to nitrocellulose membranes. Protein isolated from nontransformed DH5 cells (-C) and blood from a FIV-infected adult cat ( C) served as negative and positive controls, respectively. Representative immunoblots from three experiments for each time point are shown. (a,b,c) Membranes were incubated with serum (1:10 dilution) collected from each FIV-infected macaque at weeks (a) 0, (b) 6, and (c) 12 PI. Immunoreactivity was observed with the use of serum collected from CM2ST at weeks 6 and 12 PI as a source of antibodies (shown), but no such immunoreactivity was observed with serum collected from CM1SP at any of the timepoints (data not shown). (d) Membranes were incubated with a monoclonal antibody raised against the FIV p25 capsid protein. ability of macaques infected with simian-human immuno- to suggest that FIV represents a health hazard.for example, deficiency viruses (SHIV) to develop a detectable antibody the design of the current study circumvents many of response against the virus, whereas a marked but the natural barriers to cross-species infection by FIV, transient decrease in CD4 cells still allows for rapid while the small number of animals used makes definitive seroconversion [9, 16].However, only antibodies to FIV conclusions about the properties of FIV infection of primate gag-encoded proteins were investigated in the present cells difficult.however, in the context of xenotransgag-encoded study, and it is conceivable that antibodies to other FIV plantation and gene therapy based on viral vectors, the antigens were present in sera from CM1SP.In addition, transmission barriers between animal viruses and humans seroconversion may have been observed if the infection are also greatly reduced due to immunosuppressive therapies had been followed beyond 12 weeks or the animal had designed to prevent rejection of the foreign tissue been treated with tetanus toxoid, which would stimulate or adverse immune responses to the vector [6].Thus, FIV FIV replication. may provide a model for assessing the mechanisms by which lentivirus xenoinfections emerge. The abundance of studies demonstrating the capacity of viruses to cross species raises questions about ongoing Acknowledgements transmissions and renders the study of the adaptations The authors thank Heather Finch and Colleen Geary for animal care assistance. required for viruses to be transmitted from one host spe- These studies were supported by the Canadian Institutes for Health Research, the Natural Sciences and Engineering Research Council of Cancies to another increasingly relevant.a recent report has ada, and the Alberta Heritage Foundation for Medical Research. demonstrated the absence of FIV in veterinary workers at risk for exposure to the virus [17].However, FIV infection was assessed solely by serological tests, confirmation References 1. Gao F, Bailes E, Robertson DL, Chen Y, Rodenburg CM, Michael of direct exposure to the virus was limited, and prolonged SF, et al.: Origin of HIV-1 in the chimpanzee Pan troglodytes periods between potential exposure and assessment of troglodytes. Nature 1999, 397: Johnston J, Power C: Productive infection of human peripheral infection existed.as demonstrated above, the most obviblood mononuclear cells by feline immunodeficiency virus: ous effects of FIV infection in primates were observed implications for vector development. J Virol 1999, 73:2491- early after exposure.moreover, FIV-specific antibodies Bendinelli M, Pistello M, Lombardi S, Poli A, Garzelli C, Matteucci were not detected in the macaque, in which evidence D, et al.: Feline immunodeficiency virus: an interesting model of FIV infection was most prominent.the lack of such for AIDS studies and an important cat pathogen. Clin Microbiol detection suggests that seroconversion is not indicative Rev 1995, 8: Naldini L, Blomer U, Gallay P, Ory D, Mulligan R, Gage FH, et al.: of prior exposure to the virus. Although the results presented here support the ability of FIV to infect primate cells in vivo, it is not our intent In vivo gene delivery and stable transduction of nondividing cells by a lentiviral vector. Science 1996, 272: Poeschla EM, Wong Staal F, Looney DJ: Efficient transduction of nondividing human cells by feline immunodeficiency virus lentiviral vectors. Nat Med 1998, 4:

5 6. Patience C, Takeuchi Y, Weiss RA: Zoonosis in xenotransplantation. Curr Opin Immunol 1998, 10: Miyazawa T, Kawaguchi Y, Kohmoto M, Sakuragi J, Adachi A, Fukasawa M, et al.: Production of feline immunodeficiency virus in feline and non-feline non-lymphoid cell lines by transfection of an infectious molecular clone. J Gen Virol 1992, 73: Dow SW, Dreitz MJ, Hoover EA: Feline immunodeficiency virus neurotropism: evidence that astrocytes and microglia are the primary target cells. Vet Immunol Immunopathol 1992, 35: Shinohara K, Sakai K, Ando S, Ami Y, Yoshino N, Takahashi E, et al.: A highly pathogenic simian/human immunodeficiency virus with genetic changes in cynomolgus monkey. J Gen Virol 1999, 80: Takeuchi Y, Liong S, Bieniasz PD, Jager U, Porter CD, Friedmann T, et al.: Sensitization of rhabdo-, lenti-, and spumaviruses to human serum by galactosyl(alpha1-3)galactosylation. J Virol 1997, 71: Poeschla EM, Looney DJ: CXCR4 is required by a nonprimate lentivirus: heterologous expression of feline immunodeficiency virus in human, rodent, and feline cells. J Virol 1998, 72: Ikeda Y, Tomonaga K, Kawaguchi Y, Kohmoto M, Inoshima Y, Tohya Y, et al.: Feline immunodeficiency virus can infect a human cell line (MOLT-4) but establishes a state of latency in the cells. J Gen Virol 1996, 77: Inoshima Y, Kohmoto M, Ikeda Y, Yamada H, Kawaguchi Y, Tomonaga K, et al.: Roles of the auxiliary genes and AP-1 binding site in the long terminal repeat of feline immunodeficiency virus in the early stage of infection in cats. J Virol 1996, 70: Ostrowski MA, Krakauer DC, Li Y, Justement SJ, Learn G, Ehler LA, et al.: Effect of immune activation on the dynamics of human immunodeficiency virus replication and on the distribution of viral quasispecies. J Virol 1998, 72: Johnston JB, Jiang Y, van Marle G, Mayne MB, Ni W, Holden J, et al.: Lentivirus infection in the brain induces matrix metalloproteinase expression: role of envelope diversity. J Virol 2000, 74: Harouse JM, Gettie A, Tan RC, Blanchard J, Cheng-Mayer C: Distinct pathogenic sequela in rhesus macaques infected with CCR5 or CXCR4 utilizing SHIVs. Science 1999, 284: Butera ST, Brown J, Callahan ME, Owen SM, Matthews AL, Weigner DD, et al.: Survey of veterinary conference attendees for evidence of zoonotic infection by feline retroviruses. JAmVet Med Assoc 2000, 217: Brief Communication 1113

Human Immunodeficiency Virus

Human Immunodeficiency Virus Virion Genome Genes and proteins Viruses and hosts Diseases Distinctive characteristics Viruses and hosts Lentivirus from Latin lentis (slow), for slow progression of disease