High Levels of Sivmnd-1 Replication in Chronically Infected Mandrillus Sphinx

View metadata, citation and similar papers at core.ac.uk brought to you by CORE

provided by Elsevier - Publisher Connector Available online at www.sciencedirect.com

Virology 317 (2003) 119–127 www.elsevier.com/locate/yviro

High levels of SIVmnd-1 replication in chronically infected Mandrillus sphinx Ivona Pandrea,a,b,1 Richard Onanga,a,1 Christopher Kornfeld,c Pierre Rouquet,a Olivier Bourry,a Stephen Clifford,a Paul T. Telfer,a Kate Abernethy,a Lee TW White,a Paul Ngari,a Michaela Mu¨ller-Trutwin,c Pierre Roques,a,d Preston A. Marx,b Francois Simon,a,e and Cristian Apetreia,b,* a Laboratoire de Virologie, UGENET, SEGC, Re´serve de la Lope´, Centre de Primatologie, Centre International de Recherches Me´dicales, BP769, Franceville, Gabon b Tulane National Primate Research Center, Covington, LA 70433, USA c Unite´ de Biologie des Re´trovirus, Institut Pasteur, 29, Rue Dr. Roux, 75724 Paris Cedex 15, France d Service de Neurovirologie, Commisariod a` l’Energze Atomygue, 92265 Foulenay aux Roses, France e Hopital Charles Nicolle, 76031 Rouen, France Received 14 May 2003; returned to author for revision 17 June 2003; accepted 14 August 2003

Abstract Viral loads were investigated in SIVmnd-1 chronically infected mandrills and the results were compared with those previously observed in other nonpathogenic natural SIV infections. Four naturally and 11 experimentally SIVmnd-1-infected mandrills from a semi-free-ranging colony were studied during the chronic phase of infection. Four SIVmnd-1-infected wild mandrills were also included for comparison. Twelve uninfected mandrills were used as controls. Viral loads in all chronically infected mandrills ranged from 105 to 9 ϫ 105 copies/ml and antibody titers ranged from 200 to 14,400 and 200 to 12,800 for anti-V3 and anti-gp36, respectively. There were no differences between groups of wild and captive mandrills. Both parameters were stable during the follow-up, and no clinical signs of immune suppression were observed. Chronic SIVmnd-1-infected mandrills presented slight increases in CD20ϩ and CD28ϩ/CD8ϩ cell counts, and a slight decrease in CD4ϩ/CD3ϩ cell counts. A slight CD4ϩ/CD3ϩ cell depletion was also observed in old uninfected controls. Similar to other nonpathogenic models of lentiviral infection, these results show a persistent high level of SIVmnd-1 replication during chronic infection of mandrills, with minimal effects on T cell subpopulations. © 2003 Elsevier Inc. All rights reserved.

Keywords: SIVmnd-1; Pathogenesis; Mandrills; Viral load; Lymphocyte subsets

Introduction resemblance to the human immunodeficiency virus (HIV), SIV infections are mostly nonpathogenic in their natural More than 30 simian immunodeficiency viruses (SIVs) hosts (Beer et al., 1998; Rey-Cuille et al., 1998; Diop et al., naturally infect nonhuman African primates (Apetrei et al., 2000; Goldstein et al., 2000; Broussard et al., 2001; Silvestri 2004). Their overall seroprevalence in the wild is very high et al., 2003). A few exceptions have been documented in (estimated 18–30%) (Jolly et al., 1996; Chen et al., 1996; mandrills and African green monkeys (Pandrea et al., 2001; Souquiere et al., 2001; Peeters et al., 2002). Although they Souquiere et al., 2001; Traina-Dorge et al., 1992). Disease were named SIV because of their structural and biological is probably a rare consequence because of long-term adaptation of SIV to its natural African primate hosts, which is reflected in continuous and sustained viral replication with * Corresponding author. Tulane National Primate Research Center, minimal effects on immune function (Kaur et al., 1998; 18703 Three Rivers Road, Covington, LA 70433. Fax: ϩ1-985-871-6248. E-mail address: [email protected] (C. Apetrei). Silvestri et al., 2003). 1 These authors contributed equally to this study. There is evidence showing that SIVs have pathogenic

0042-6822/$ – see front matter © 2003 Elsevier Inc. All rights reserved. doi:10.1016/j.virol.2003.08.015 120 I. Pandrea et al. / Virology 317 (2003) 119Ð127 potential. Cross-species transmission of SIVsm to macaques Results results in increased pathogenicity and induction of simian acquired immunedeficiency syndrome (sAIDS) (Letvin et Clinical and immunological follow-up of chronically al., 1985; Murphey-Corb et al., 1986). HIV-1 and HIV-2, infected SIVmnd-1 mandrills which have simian origins, are responsible for the AIDS pandemic (Peeters et al., 1989; Gao et al., 1999; Chen et al., No clinical signs of AIDS were observed during annual 1996). Experimental transmission of SIVagm from African follow-ups of chronically, naturally infected mandrills green monkeys (AGMs) and SIVIhoest from I’hoesti mon- (17B, 17D, 17G, 17D2) in captivity. Eleven experimentally keys to pig-tailed macaques has also shown pathogenic infected animals (n ϭ 11) were examined every 3 months. potential and immunosuppression (Hirsch et al., 1995, No follow-up was possible for the wild mandrills, but all 1999). In these cases, high viral loads induce massive acti- were clinically normal at the time of sampling. vation and destruction of the immune system (Alimonti et al., 2003). Therefore, plasma viral load is considered the Viral load quantification in experimentally infected best predictor of the clinical outcome in humans and ma- mandrills caques (Mellors et al., 1996; Lifson et al., 1997). In this study, we have investigated the relationship be- In HIV-infected humans and SIV-infected rhesus ma- tween SIV and its natural host in a new model, SIVmnd-1 caques the quantification of plasma viral RNA provides a chronic infection of mandrills. The probable origin of measure of the magnitude of virus replication (Mellors et SIVmnd-1 was reported to be from Cercopithecus lhoesti al., 1997; Lifson et al., 1997). For viral RNA loads in solatus to mandrills (Hirsch et al., 1999; Beer et al., 1999). plasma, a QC-RT-PCR assay specific for SIVmnd-1 was This observation raised the question as to whether or not developed (Onanga et al., 2002) using primers MP5/MP6 mandrills should be considered as long-term natural hosts of from a conserved region in integrase of SIVmnd-1 GB-1 SIVmnd-1 and whether the course of infection would be the (Nerrienet et al., 1998). There were no mismatches (Onanga et al., 2002). Our previous studies on the acute phase of same as in other nonpathogenic infections. The preponder- SIVmnd-1 infection showed that high viral loads obtained ance of evidence indicates that mandrills are a natural host by QC-RT-PCR directly correlated with both p24 Ag titers because recent studies have shown that SIVmnd-1 is highly and plasma RT activity (Onanga et al., 2002). prevalent in wild mandrills, suggesting that cross-species Experimentally infected mandrills included in this study transmission was an ancient event and that the virus infects were mandrills from a primary infection experiment (12A4, this species naturally (Souquiere et al., 2001). SIVmnd-1 2C2, 10G, 12C2) inoculated intravenously (iv) with 3000 replicates to high levels during experimental primary infec- TCID , and dams in a breastfeeding transmission experi- tion in mandrills and this replication has only slight and 50 ment (5I, 5J, 12A3, 12A7, 6B, 16C), which received 300 transitory effects on the dynamics of immune cells (Onanga TCID50 iv of the same stock. Mandrill 16E received 10 ml et al., 2002). This pattern is very similar to those observed of blood from the mandrill naturally infected with GB-1 during the primary infection of other nonhuman primate (F17). This virus was titered at 4 TCID50 on SupT1 cells, a species (Diop et al., 2000; Kaur et al., 1998). value which corresponded to 896,000 RNA copies/ml Here we address the hypothesis as to whether or not (Onanga et al., 2002). In experimentally infected mandrills, mandrills represent an intermediate level between patho- the viral set point was reached between days 30 and 60 genic and nonpathogenic SIV infections or whether or not postchallenge (Onanga et al., 2002). In this present study we the virus is adapted to the mandrill host. Only one quantified the SIVmnd-1 RNA loads in 19 samples col- SIVmnd-1 strain (GB-1) is available and it was adapted in lected between days 180 and 540 postchallenge from all 11 vitro and reported to use CXCR4 as the main coreceptor, experimentally infected mandrills (Table 1). SIVmnd-1 which is highly unusual for SIV (Schols and De Clerq, plasma viral loads are presented in Table 1. Although results 1998; Beer et al., 2001). Therefore, SIVmnd-1 (GB-1) may obtained at days 180 and 360 for mandrills 12A4, 16E, 2C2, have a higher pathogenic potential compared to SIVs which and 10G have been presented elsewhere, these points are use CCR-5 as the major coreceptor (Marx and Chen, 1998). also showed in Table 1 for comparison. Viral RNA levels However, SIVmnd-1 GB1 recovered during the primary ranged from 18,000 to 334,000 copies/ml, similar to the infection used CCR5 and Bonzo and not CXCR4 (R. range observed in SIVsm-infected sooty mangabeys (Rey- Onanga, unpublished observation). Couille et al., 1998; Ling et al., 2003; Silvestri et al., 2003) As chronic infection reflects the long-term natural his- and AGMs (Goldstein et al., 2000; Broussard et al., 2001; tory of SIV, the aim of the present study was to measure the Holzammer et al., 2001). No statistically significant differ- dynamics of RNA plasma viral load, anti-SIVmnd-1 anti- ence was observed in viral loads of mandrills in different body titers, and lymphocyte subsets in mandrills chronically cohorts. Moreover, viral loads observed during the acute infected with SIVmnd-1 to compare these data with those phase of SIVmnd-1 infection had no differences with re- previously observed in other nonpathogenic models. Data spect to the infectious dose (Onanga et al., 2002; Pandrea et from wild mandrills are also presented for comparison. al., 2002). I. Pandrea et al. / Virology 317 (2003) 119Ð127 121

Table 1 SIVmnd-1 plasma viral load quantiﬁcation (QC-RT-PCR) and antibody titers in experimentally infected, chronic naturally infected, and wild mandrills

Experimentally infected Day PIa Copies/ml anti-V3 anti-gp41 captive mandrills 12A4 D180 334,000 800 400 D360 93,750 1600 800 D540 93,750 4800 800 16E D180 150,000 400 400 D360 93,750 1600 800 D540 93,750 4800 800 2C2 D180 93,750 400 400 D360 66,964 1600 800 D540 93,750 4800 800 10G D180 66,964 1600 400 D360 66,964 4800 1600 D540 18,750 14,400 3200 12C2 D180 66,964 1600 400 5I D180 155,000 800 450 5J D180 66,964 1600 400 12A3 D180 40,000 200 200 12A7 D180 334,821 400 450 6B D150 334,821 1600 450 16C D180 334,821 800 800

Naturally infected captive Year Copies/ml anti-V3 anti-gp41 mandrills 17B 1992 93,750 4800 6400 1993 93,750 4800 6400 1995 93,750 4800 12,800 1999 ND 4800 6400 2000 93,750 4800 6400 17D 1993 ND 0 200 1999 93,750 400 800 2000 93,750 1600 1600 17D2 2000 289,235 1600 800 17G 2000 892,338 800 200

Wild mandrills Year Copies/ml anti-V3 anti-gp41 MS22-Lope 2001 334,821 1600 1600 MS23-Lope 2001 206,596 1600 400 MS25-Lope 2001 66,964 1600 800 MS21-Lope 2001 66,964 800 400

a P.I. ϭ postinoculation.

Viral loads in captive naturally infected mandrills testing. Both anti-V3 and anti-gp36 antibody titers were stable during the follow-up: 1/4800 and 1/6400, respec- Serial sampling during a 15-year seroepidemiological tively. F17D was born in 1992 and three samples were follow-up of mandrills in the semifree ranging CIRMF tested. SIVmnd-1 viral loads for mandrill F17D were sim- colony has shown a cluster of SIVmnd-1 infections in a ilar to F17B (93,750 copies/ml). However, antibody titers founder female mandrill F17 (GB1), three of her offspring for F17D were significantly lower than for 17B (Table 1). (F17B, F17D, and M17G), and two of the F17D offspring Only one plasma sample was tested from the remaining (17D1 and 17D2). Dynamics of plasma SIVmnd-1 load in animals. Mandrill 17D2 was born in 1997 and had been the founder female F17 (GB1) over a 15-year period have sampled once in 2000. Its viral load was significantly higher been published elsewhere (Pandrea et al., 2001). Here we than observed in other mandrills (289,235 copies/ml) and present the dynamics of SIVmnd-1 plasma load on serial antibody titers were significantly lower than those of 17B, samples originating from four mandrills. F17B was born in but similar to those of 17D. The remaining mandrill (M17G) 1987 and five sequential samples were tested (Table 1). was born in 1996. His plasma viral load from 2000 was the QC-RT-PCR showed a remarkably narrow range of viral highest in this study group (892,338 copies/ml). Anti- loads (93,750 copies/ml) in all but one sample. The remain- SIVmnd-1 antibody titers were low in this mandrill as well ing sample was stored at Ϫ20°C and was used for antibody (Table 1). 122 I. Pandrea et al. / Virology 317 (2003) 119Ð127

Feral mandrills

Previous reports of SIVmnd prevalence in mandrills were limited to SIVmnd-2 infections, which has a larger geographical distribution (Takehisha et al., 2001; Peeters et al., 2002). However, recent data originating from ecological and seroepidemiological studies have shown that SIVmnd-1 is readily transmitted in the wild, as 10 of 14 mandrills sampled in the Lope reserve in Gabon harbored SIVmnd-1 (Souquiere et al., 2001). In 2001, seven wild mandrills from the Lope reserve were sampled. Four were positive for SIVmnd-1 RNA. The two major aims of viral quantification in these mandrills were (1) to compare with results observed in the CIRMF colony in both naturally and experimentally infected mandrills, and (2) to determine if a high SIVmnd-1 prevalence in feral mandrills correlated with high viral loads. The results are presented in Table 1. All four wild mandrills harbored SIVmnd-1 viral loads in the same range as the captive mandrills (7 ϫ 104–3 ϫ 105 copies/ml), and also in the same range as African green monkeys and sooty mangabeys (Broussard et al., 2001; Goldstein et al., 2000; Holzammer et al., 2001; Chrakrabarti et al., 2000a; Silvestri et al., 2003). These results indicate that our primers were suitable for SIVmnd-1 quantification, as divergent viruses were amplified and viral loads were similar to experimentally inoculated animals.

Flow cytometry

Lymphocyte subsets (expressed as median and 25–75th percentile) were investigated in eight of the experimentally infected mandrills. Twelve uninfected mandrills served as ϭ Fig. 1. (a) Box-plots of the relative numbers of lymphocyte subsets in controls. These mandrills were either age-matched (n 6) mandrills. The plots correspond to young age-matched uninfected man- or old mandrills, aged over 18 years (range 18–23 years) (n drills (mean 7.9 years, range 6–11 years) (ᮀ); old uninfected mandrills ϭ 6). This last group was examined to determine if an (mean 18.6 years; range: 17–22 years) ( ); and SIVmnd-1-infected man- age-related decline of CD4 cells occurs in mandrills, similar drills (mean 7.4 years; range 3–12 years) (■). (b) Activation status of CD4ϩ and CD8ϩ cells in young age-matched SIVmnd-1 uninfected (ᮀ and to that reported in SIVsm-uninfected sooty mangabeys infected (■) mandrills, as determined by HLA-DR and/or CD28 coexpres- (Chakrabarti et al., 2000a, 2000b) and humans (Mo et al., sion. In the box plots representation, the box extends from the 25th 2003). percentile to the 75th percentile, with a horizontal line at the median, and When the haematological parameters (expressed as me- the dots extend down to the smallest value and up to the largest value. dian and 25–75th percentile) of chronically infected (n ϭ 8, mean age 8.6 years, range 3–19 years) and uninfected (n ϭ young age-matched uninfected mandrills (mean: 45.63%; 6) age-matched (mean age 8.2 years, range 4–12) monkeys range: 35.3–58.2%) (Fig. 1). The effect of age was ex- were analyzed by the Mann–Whitney test, no signiﬁcant pressed by a slight decrease in total lymphocyte in old difference was noted in absolute lymphocyte counts be- uninfected controls. Thus, mean absolute lymphocyte tween the two groups, expressed as ϫ103 cells/␮l blood (5.2 counts, expressed as ϫ103 cells/␮l blood, were 3.92 Ϯ 1.2 Ϯ 1.6 versus 4.63 Ϯ 2.32, in SIVmnd-1-infected and non- in old uninfected mandrills, which was similar to that pre- infected mandrills, respectively). However, the absolute viously reported in old sooty mangabeys (Chakrabarti et al., CD20ϩ, CD3ϩ, and CD8ϩ counts were higher in the in- 2000b). Mean percentages of CD8ϩ cells in old uninfected fected animals. mandrills was 40.1% (range: 30.1 Ϯ 51.7%). Higher absolute CD8ϩ T cell numbers were found in the CD3ϩ cells were also higher in SIVmnd-1-infected man- infected monkeys compared to age-matched uninfected drills (Fig. 1a), but this difference was not statistically mandrills (2230 Ϯ 1310 cells/␮l blood versus 1871 Ϯ 1160 signiﬁcant between the three groups. cells/␮l blood, P Ͼ 0.05). Similar data were observed in Differences were observed in the expression of activa- relative values of CD8ϩ cells, which were higher in infected tion markers when two groups of animals, SIVmnd-1-in- mandrills (mean: 52.39%, range: 39.4–62.5%) compared to fected versus uninfected, were compared (Fig. 1b). CD8ϩ I. Pandrea et al. / Virology 317 (2003) 119Ð127 123

HLA-DRϩ cells were increased in infected mandrills conserved region of the SIVmnd-1 genome and therefore (mean: 7%, range: 6.1–10.2) compared to age-matched un- are likely to correctly amplify divergent SIVmnd-1 strains. infected controls (mean: 5.51%, range: 3.2–7.1). CD8ϩ In humans, HIV-1 viral diversity may significantly influ- CD28ϩ cells were also higher in infected mandrills (mean: ence the performances of different commercially available 17.81%, range: 14.5–21.5%) compared to uninfected con- viral load quantification kits. Thus, bDNA designed to map trols (mean: 15.55%, range: 13.2–18.7%). No difference in the polymerase region has a better sensitivity than gag- activation markers were detected on CD4ϩ cells between based Monitor–Roche for the quantification of divergent infected and uninfected mandrills (Fig. 1b). HIV-1 group M viruses, which may vary in up to 20% of pol In summary, minor differences were detected between sequences (Damond et al., 1999). Our strategy of amplify- infected and uninfected mandrills, mostly within the CD3ϩ ing the integrase region seems more appropriate for the and CD8ϩ cells, and activation status in the latter. However, quantification of divergent viruses than a gag-based strat- these differences were not statistically significant. egy. However, in a recent study of viral loads in sooty mangabeys from the Yerkes National Primate Research Centre, gag quantification gave similar results with a UTR- Discussion based method for the quantification of SIVsm [Silvestri et al., 2003), despite the cocirculation in the Yerkes colony of In this study the viral load in SIVmnd-1 chronically four different SIVsm lineages (Ling et al., 2003; A. Kaur, C. infected mandrills was investigated. Although previous re- Apetrei, and P.A. Marx, unpublished observation). ports measured viral loads in captive African monkeys It is not clear why the prevalence of SIVmnd-1 in man- (Goldstein et al., 2000; Broussard et al., 2001; Chakrabarti drils from the Lope reserve is higher than generally reported et al., 2000a; Ling et al., 2003; Silvestri et al., 2003), to our for other natural SIV infection (Jolly et al., 1996, Peeters et knowledge, this is the first report describing SIV viral loads al., 2002). Since no differences were detected in viral loads in wild natural infected hosts. The only other study which between wild and captive mandrills, the differences in prev- investigated the viral burden in wild sooty mangabeys mea- alence between different groups of the same species or sured only cellular proviral loads and not plasma RNA different nonhuman primate species may not be related to (Chen et al., 1996). No significant differences were ob- virological or host factors but rather to behavior which served in viral loads of mandrills from experimentally and favors viral transmission (Santiago et al., 2003). SIV might naturally infected study groups. Chronic viral loads were be concentrated in specific nonhuman primate populations also similar in animals receiving different infectious doses rather than widely distributed. This is also consistent with of SIVmnd-1, showing that dose had no detectable influence the hypothesis that lentivirus transmission to humans only on the dynamics of acute (Onanga et al., 2002; Pandrea et occurred in relatively limited areas in Central and West al., 2002) or chronic SIVmnd-1 infection. A potential lim- Africa (Peeters et al., 2002). itation of our study as a model of the natural history of the In humans and macaques there is a strong correlation SIVmnd-1 infections is that all experimentally infected an- between the pattern of viral replication and the clinical imals received the same viral strain (SIVmnd-1 GB1). How- course of infection, a high viral load often predicting early ever, this is also true of SIV pathogenesis studies in rhesus disease (Mellors et al., 1997; Lifson et al., 1997). Previous macaques, which are infected with very similar SIVmac reports suggested that the lack of pathogenicity of nonhu- strains (Mansfield et al., 1995). Our results showed that man primate lentiviruses in their natural hosts might be due experimental infection reproduced similar patterns of virus- to low viral replication (Hartung et al., 1992). However, host dynamics as observed in naturally infected mandrills. several subsequent studies as well as this study have shown Viral loads described here in wild mandrills were in the that during the chronic phase of infection, primate lentivi- same range as those reported by others examining sooty ruses replicate at moderate to high levels without clinical or mangabeys and AGMs (Rey-Cuille et al., 1998; Diop et al., pathological signs (Rey-Couille et al., 1998; Diop et al., 2000; Goldstein et al., 2000; Broussard et al., 2001; Ling et 2000; Goldstein et al., 2000; Broussard et al., 2001; Hol- al., 2003; Silvestri et al., 2003). The fact that viral loads are zammer et al., 2001). Therefore the lack of pathogenicity of the same between wild and captive mandrills indicates that SIVs, including SIVmnd-1, cannot be explained by limited diet, medical care, and other factors play little role in de- replication in their natural host. termining viral load. Conversely, these latter factors plus a Anti-gp36 and anti-V3 antibody titers were performed to lack of natural predators and not a lower viral replication are evaluate antibody responses and relationship to virus load. likely to be responsible for the longer life span of captive The immunodominant region of the gp36 is one of the most monkeys compared to wild ones. conserved SIV epitopes and it is useful in diagnostic tests in Although the sequences of SIVmnd-1 strains from wild both humans and nonhuman primates. Detecting anti-gp36 mandrills have been not analyzed in this study, previous antibodies is a very effective diagnostic tool (Simon et al., analysis has shown significant viral diversity in SIVmnd-1- 2001). Although in HIV-1 infection V3 region of gp120 is infected mandrills from the Lope reserve (Souquiere et al., highly variable, this region is significantly more conserved 2001). However, our primers were designed from a highly in SIV infection (Franchini et al., 1990). This is supported 124 I. Pandrea et al. / Virology 317 (2003) 119Ð127 by the finding that all the infected mandrills tested so far host. The mechanism(s) of this resistance to SIV of African reacted with GB-1 V3 peptide. Our SIVmnd-1 V3 peptide is monkeys should be further investigated to gain insights to able to even detect antibodies in SIVmnd-2-infected man- control HIV infection. drills or SIVdrl-infected drills (Souquiere et al., 2001; Si- mon et al., 2001). This is not surprising, as SIVmnd-2 and SIVdrl are recombinant viruses with a SIVmnd-1 envelope Materials and methods (Souquiere et al., 2001; Hu et al., 2003). The anti-V3 antibodies were tested because a decrease in titers is often Animals correlated with disease progression and a decline in serum neutralizing activity in HIV-1-infected humans (Fenouillet A total of 19 SIVmnd-1 chronically infected mandrills et al., 1995). We have also reported a loss of anti-V3 were examined, 4 of which resided in a large semi-free antibodies in an SIVmnd-1-infected mandrill with immune enclosure, and were naturally infected in captivity. Further- suppression (Pandrea et al., 2001). However, none of the more, these four (17B, 17D, 17D2, 17G) were offspring of mandrills presented here showed decreased anti-V3 titers. the founder SIVmnd-1 GB1 female with whom they were Also, there was no clear correlation between viral load and raised/housed, suggesting that vertical transmission oc- anti-V3 titers. Theoretically, a more meaningful gauge of curred. However, serial plasma samples were only exam- immune function and its relationship to viral load may have ined on two of these mandrills: F17B (1992, 1993, 1995, been neutralizing antibody responses. However, SIVmnd-1 1999, and 2000) and F17D (1993, 1999, and 2000). GB1 grows poorly in cell culture, and this has hampered a Eleven mandrills were experimentally infected in two proper evaluation of neutralizing antibody responses. Also, different experiments: mandrills 16E, 2C2, 12A4, 10G, it should be noted that in other SIV naturally infected 12C2 were intravenously infected with 3000 TCID50. Data African monkeys, neutralizing antibody responses are gen- on acute infection in these animals have been published erally low (Apetrei et al., 2004). elsewhere (Onanga et al., 2002). Mandrills 12A3, 16C, From a clinical point of view, African monkeys may be 12A7, 5I, 5J, 6B were infected intravenously with 300 similar to long-term nonprogressing HIV-infected humans. TCID50 SIVmnd-1 and viral loads were measured (day It was suggested that SIV adaptation to the nonprogressor 180). In both experiments, the inoculum was derived from status may be due to induction of a persistent infection with plasma prepared as previously described (Onanga et al., a symptom-free period exceeding the normal life span of 2002). An additional 11 uninfected mandrills were used as most monkeys (Pandrea et al., 2001). However, in African controls for flow cytometry analyses. In the study group, 5 nonhuman primates infected with SIV, the pathologic animals were male (2C2, 12A4, 17G, 12C2, 10G) and 10 mechanisms must be different from those observed in HIV- were female (17B, 17D, 17D2, 12A3, 16C, 16E, 12A7, 5I, 1-infected long-term nonprogressors, in whom the low 5J, 6B). The mean age of the animals was 8.6 (range 3–19). HIV-1 loads are due to active cellular immune responses The mean weight for the males was 30 kg and 15 kg for (Cao et al., 1995). In contrast, African monkeys have high females. Mandrill 12C2 died in 2000 of gastric dilatation, viral loads which are associated with an attenuated antiviral with no evidence of AIDS (Onanga et al., 2002). The wild cellular immune response (Chakrabarti et al., 2000a; Holz- mandrills originated from the Lope reserve in Central Ga- nagel et al., 2002; Silvestri et al., 2003). Moreover, this bon. All were adult solitary males in good health at the time equilibrium is maintained for long periods of time, without of capture. No follow-up was possible for these mandrills any clinical and immunological sign of immune suppres- because they were released. sion. Thus, our analysis of immunophenotypic markers in SIVmnd-1-infected mandrills was designed to determine if Sampling high levels of viremia have any effect on T cells. None of the parameters included in our immunophenotypic analysis Animals were anesthetized with ketamine–HCl and 7–10 showed significant differences between SIVmnd-1-infected ml of whole blood was collected in EDTA. Plasma was and uninfected mandrills. Chronically infected mandrills separated and stored at Ϫ80°C until used for semiquantita- were found to maintain relatively normal CD4ϩ counts in tive viral load analysis and serology. Whole blood was used peripheral blood, despite years of infection. The slight in- for flow cytometry as described below. crease in levels of CD8ϩ HLA-DRϩ cells observed in chronically infected mandrills was not associated with de- Quantification of viral RNA in plasma creased levels of CD8ϩ CD28ϩ cells, unlike results observed in humans (Giorgi et al., 1999; Sousa et al., 2002). RNA quantification was performed by a limiting-dilution However, this data are similar to data previously reported in RT-PCR assay specific for SIVmnd-1 (Onanaga et al., SIVsm-infected sooty mangabeys (Silvestri et al., 2003). 2002). The primers were MP5 (5Ј-CCA GAT AAG TGG Our results indicate that SIVmnd-1 is well adapted to its AAG ATA GAA AAG-3Ј) and MP6 (5Ј-GGG AGA TAT mandrill host, and that cross-species transmission of lenti- GGG AAG ATT GGG-3Ј), allowing the amplification of a viruses in the wild can be controlled by the new primate 476-bp region within the SIVmnd-1 GB1 pol gene as pre- I. Pandrea et al. / Virology 317 (2003) 119Ð127 125

viously described (Nerrienet et al., 1998). Since the virus Table 2 used in this study was derived from the SIVmnd-1 GB1, the Antibody panel used for evaluating host immune dynamics in mandrills possibility of mismatches was minimal. chronically infected with SIVmnd-1 RNA was extracted from plasma using the QIAamp mAB-PEa mAB-FITCb Cell type defined Viral RNA Mini kit (Qiagen, Courtaboeuf, France). The (clone, source) (clone, source) RNA extracted from 540 ␮l of plasma was subjected to IgG1 (PN0639 ITc) Control seven 5- to 10-fold serial dilutions which were used for the IgG1 (BDd) Control RT-PCR (one step RT-PCR kit, Qiagen). Reverse transcrip- IgG1 (BD) IgG1 (IT) Control tion was performed at 50°C for 30 min in the presence of IgG1 (BD) IgG2a (BD) Control CD20 (Leu-16 BD) B cells MP5 and MP6, followed by 47 cycles of 95°C for 15 s, CD4 (Leu-3a BD) CD3 (FN-18 BSe) T cells, T helper cells 55°C for 45 s, and 72°C for 1 min, with a final extension of CD8 (Leu-2a BD) CD3 (FN-18 BS) T cells, T cytotoxic cells 7 min at 72°C. The absolute target copy numbers were CD28 (Leu-28BD) CD4 (Leu-3 BD) Costimulatory T helper cells determined by amplification in parallel of known amounts CD28 (Leu-28BD) CD8 (Leu 2a BD) Costimulatory T cytotoxic cells CD4 (Leu-3a BD) CD8 (Leu 2a BD) T helper, T cytotoxic cells of a SIVmnd-1 RNA external standard. The standard was ϩ CD4 (Leu-3a BD) HLA-DR (BD) Activated MHC II CD4 cells obtained by amplifying the 476-bp region of SIVmnd-1 CD8 (Leu-2a BD) HLA-DR (BD) Activated MHC II CD8ϩ cells GB1 pol from F17 genomic DNA. The PCR product was a Phycoerythrin (PE). cloned into the pCR 2.1 vector using the topoTA Cloning b Ј Ј Fluorescein isothiocyanate (FITC). Kit (Invitrogen, Groningen, The Netherlands). The 5 -3 c ImmunoTech (IT). sense orientation of the insert was verified by sequence d BioSource (BS). analyses. A clone containing the insert in 5Ј-3Ј orientation e Becton Dickinson (BD). was chosen and subjected to PCR using the primer pair pCRT7S 5Ј (GCG TAA TAC GAC TCA CTA TAG GGA tested against gp36 and V3 peptide, respectively. The cutoff GAG GA GCTC GAG CGG CCG CCA GTG TG) and was established at 0.1 o.d. pCRT7AS (5Ј-ATT ACG CCA AGC TTG GTA CCG AG- 3Ј), which hybridize to regions just upstream and down- Flow cytometry analysis stream of the pCR2.1 plasmid polylinker sequence. Cycling conditions for this PCR consisted of an initial DNA dena- To quantify lymphocyte subsets and examine lympho- turation step, followed by 40 cycles that included a dena- cyte activation in infected versus uninfected animals, a turation step at 92°C for 30 s, primer annealing at 62°C for variety of antibodies against cell-surface markers were used, 45 s, and DNA synthesis at 72°C for 45 s, with a final as summarized in Table 2. Samples from noninfected extension of 7 min at 72°C. The resulting PCR product (M5H, M2I, F12A3, F5J, F16C, F12A7, and F5I), and contained the cloned insert together with a T7 promoter infected (F17, F17D, M2C2, M12A4, and M10G) mandrills sequence at its 5Ј end. After clean up of the PCR product were analyzed by flow cytometry. (PCR purification kit, Qiagen, Courtaboeuf, France) this One hundred microliters of EDTA-treated whole blood ␮ DNA was used as a template for an in vitro transcription was incubated in the dark with 5 l antibody for 30 min at (MEGAscript kit, Ambion, Austin, TX). The resulting RNA 4°C. Red blood cells were lysed using a standard ammo- was DNase digested and nonincorporated nucleotides were nium chloride lysing solution. Cells were then washed twice eliminated using a Chroma Spin-30 DEPC H20 Column in phosphate-buffered saline (PBS) and resuspended in PBS (Clontech, Palo Alto, CA). Size and purity of the synthe- with 1% paraformaldehyde. Data were acquired using a sized RNA were verified on a denaturating-PAGE gel. The FACsCalibur flow cytometer (Immunocytometry System, concentration of the in vitro transcript was determined by a Becton Dickinson) and analyzed with Cell Quest software highly sensitive and specific Gene Quant II spectrophoto- (Immunocytometry System, Becton Dickinson). metric operation system as already described (Onanga et al., Statistical analysis and data presentation 2002). Aliquots were stored at Ϫ80°C until use. Immedi- ately before usage, one aliquot of RNA was thawed and Flow cytometriy data were analyzed for significant dif- diluted in RNase free H2O and five dilutions of the RNA ferences (P Ͻ 0.05) using the Mann–Whitney test. Corre- were used as external standard in each experiment. The lations were calculated and expressed as the Spearman co- ϫ 2 sensitivity of the assay was 2 10 RNA copies/ml of efficient of correlation (rs). Data were displayed using the plasma. box-and-whisker plot method (Fig. 1). Primate immunodeficiency viruses ELISA (PIV-ELISA) Acknowledgments Antibody titers to both gp36 and V3 peptide were measured using a PIV-ELISA, as previous described (Simon et This work was supported by grants from Agence Natio- al., 2001). Twofold serial dilutions from 1/100 to 1/12,800 nale pour la Recherche sur le SIDA (ANRS) and NIH and from 1/100 to 1/43,200 were done for each sample (RO1-AI44596). 126 I. Pandrea et al. / Virology 317 (2003) 119Ð127

References Giorgi, J.V., Hultin, L.E., McKeating, J.A., Johnson, T.D., Owens, B., Jacobson, L.P., Shih, R., Lewis, J., Wiley, D.J., Phair, J.P., Wolinsky, ϩ S.M., Detels, R., 1999. Shorter survival in advanced human immuno- Alimonti, J.B., Ball, T.B., Fowke, K.R., 2003. Mechanisms of CD4 T cell deficiency virus type 1 infection is more closely associated with T death in human immunodeficiency virus infection and AIDS. J. Gen. lymphocyte activation than with plasma virus burden or virus chemo- Virol. 84, 1649–1661. kine coreceptor usage. J. Infect. Dis. 179, 859–870. Apetrei, C., Robertson, D.L., Marx, P.A., 2004. The history of SIVs and Goldstein, S., Ourmanov, I., Brown, C.R., Beer, B.E., Elkins, W.R., AIDS: epidemiology, phylogeny and biology of isolates from naturally Plishka, R., Buckler-White, A., Hirsch, V.M., 2000. Wide range of SIV infected non-human primates (NHP) in Africa. Front. Biosci. 9, viral load in healthy African green monkeys naturally infected with 225–254. simian immunodeficiency virus. J. Virol. 74, 11744–11753. Beer, B., Denner, J., Brown, C.R., Norkley, S., zur Megede, J., Coulibaly, Hartung, S., Boller, K., Cichutek, K., Norley, S.G., Kurth, R., 1992. C., Plesker, R., Holzammer, S., Baier, M., Hirsch, V.M., Kurth, R., Quantitation of a lentivirus in its natural host: simian immunodefi- 1998. Simian immunodeficiency virus of African green monkeys is ciency virus in African green monkeys. J. Virol. 66, 2143–2149. apathogenic in the newborn natural host. J. Acquir. Immune. Defic. Hirsch, V.M., Campbell, B.J., Bailes, E., Goeken, R., Brown, C., Elkins, Syndr. Hum. Retrovirol. 18, 210–220. W.R., Axthelm, M., Murphey-Corb, M., Sharp, P.M., 1999. Charac- Beer, B.E., Bailes, E., Goeken, R., Dapolito, G., Coulibaly, C., Norley, terization of a novel simian immunodeficiency virus (SIV) from S.G., Kurth, R., Gautier, J.P., Gautier-Hion, A., Vallet, D., Sharp, P.M., L’Hoest monkeys (Cercopithecus l’hoesti): implications for the origin Hirsch, V.M., 1999. Simian immunodeficiency virus (SIV) from sun- of SIVmnd and other primate lentiviruses. J. Virol. 73, 1036–1045. tailed monkeys (Cercopithecus solatus): evidence for host-dependent Hirsch, V.M., Dapolito, G., Johnson, P.R., Elkins, W.R., London, W.T., evolution of SIV within the C. lhoesti superspecies. J. Virol. 73, Montali, R., Goldstein, S., Brown, C., 1995. Induction of AIDS by 7734–7744. simian immunodeficiency virus from an African green monkey: spe- Beer, B.E., Foley, B.T., Kuiken, C.L., Tooze, Z., Goeken, R.M., Brown, cies-specific variation in pathogenicity correlates with the extent of in C.R., Hu, J., St Claire, M., Korber, B.T., Hirsch, V.M., 2001. Charac- vivo replication. J. Virol. 69, 955–967. terization of novel simian immunodeficiency viruses from red-capped Holzammer, S., Holznagel, E., Kaul, A., Kurth, R., Norley, S., 2001. High mangabeys from Nigeria (SIVrcmNG409 and -NG411). J. Virol. 75, virus loads in naturally and experimentally SIVagm-infected African 12014–12027. green monkeys. Virology 283, 324–331. Broussard, S.R., Staprans, S.I., White, R., Whitehead, E.M., Feinberg, Holznagel, E., Norley, S., Holzammer, S., Coulibaly, C., Kurth, R., 2002. M.B., Allan, J.S., 2001. Simian immunodeficiency virus replicates to Immunological changes in simian immunodeficiency virus (SIVagm)- high levels in naturally infected African green monkeys without induc- infected African green monkey (AGM): expanded cytotoxic T lympho- ing immunologic or neurologic disease. J. Virol. 75, 2262–2275. cyte, natural killer and B cell subsets in the natural host of SIVagm. Cao, Y., Qin, L., Zhang, L., Safrit, J., Ho, D.D., 1995. Virologic and J. Gen. Virol. 83, 631–640. immunologic characterization of long-term survivors of human immu- Hu, J., Switzer, W.M., Foley, B.T., Robertson, D.L., Goeken, R.M., nodeficiency virus type 1 infection. N. Engl. J. Med. 332, 201–208. Korber, B.T., Hirsch, V.M., Beer, B.E., 2003. Characterization and Chakrabarti, L.A., Lewin, S.R., Zhang, L., Gettie, A., Luckay, A., Martin, comparison of recombinant simian immunodeficiency virus from drill L.N., Skulsky, E., Ho, D.D., Cheng-Mayer, C., Marx, P.A., 2000a. (Mandrillus leucophaeus) and mandrill (Mandrillus sphinx) isolates. Normal T-cell turnover in sooty mangabeys harboring active simian J. Virol. 77, 4867–4880. immunodeficiency virus infection. J. Virol. 74, 1209–1223. Jolly, C., Phillips-Conroy, J.E., Turner, T.R., Broussard, S., Allan, J.S., Chakrabarti, L.A., Lewin, S.R., Zhang, L., Gettie, A., Luckay, A., Martin, 1996. SIVagm incidence over two decades in a natural population in a L.N., Skulsky, E., Ho, D.D., Cheng-Mayer, C., Marx, P.A., 2000b. natural population of Ethiopian grivet monkeys (Cercopithecus ae- Age-dependent changes in T cell homeostasis and SIV load in sooty thiops aethiops). J. Med. Primatol. 25, 78–83. mangabeys. J. Med. Primatol. 29, 158–165. Kaur, A., Grant, R.M., Means, R.E., McClure, H., Feinberg, M., Johnson, Chen, Z., Telfer, P., Gettie, A., Reed, P., Zhang, L., Ho, D.D., Marx, P.A., P., 1998. Diverse host responses and outcomes following simian im- 1996. Genetic characterization of new West African simian immuno- munodeficiency virus SIVmac239 infection in sooty mangabeys and deficiency virus SIVsm: geographic clustering of household-derived rhesus macaques. J. Virol. 72, 9597–9611. SIV strains with human immunodeficiency virus type 2 subtypes and Letvin, N.L., Daniel, M.D., Sehgal, P.K., Desrosiers, R.C., Hunt, R.D., genetically diverse viruses from a single feral sooty mangabey troop. Waldron, L.M., MacKey, J.J., Schmidt, D.K., Chalifoux, L.V., King, J. Virol. 71, 3617–3627. N.W., 1985. Induction of AIDS-like disease in macaque monkeys with Damond, F., Apetrei, C., Descamps, D., Brun-Vezinet, F., Simon, F., 1999. T-cell tropic retrovirus STLV-III. Science 230, 71–73. HIV-1 subtypes and plasma RNA quantification. AIDS 13, 286–288. Lifson, J.D., Nowak, M.A., Goldstein, S., Rossio, J.L., Kinter, A., Diop, O.M., Gueye, A., Dias-Tavares, M., Kornfeld, C., Faye, A., Ave, P., Vasquez, I., Wiltrout, T.A., Brown, C., Schneider, D., Wahl, L., Lloyd, Huerre, M., Corbet, S., Barre´-Sinoussi, F., Mu¨ller-Trutwin, M.C., A.L., Williams, J., Elkins, W.R., Fauci, A.S., Hirsch, V.M., 1997. The 2000. High levels of viral replication during primary simian immuno- extent of early viral replication is a critical determinant of the natural deficiency virus SIVagm infection are rapidly and strongly controlled history of simian immunodeficiency virus infection. J. Virol. 71, 9508– in African green monkeys. J. Virol. 74, 7538–7547. 9514. Fenouillet, E., Blanes, N., Benjouad, A., Gluckman, J.C., 1995. Anti-V3 Ling, B., Santiago, M.L., Meleth, S., Gormus, B., McClure, H.M., Apetrei, antibody reactivity correlates with clinical stage of HIV-1 infection and C., Hahn, B.H., Marx, P.A., 2003. Noninvasive detection of new simian with serum neutralizing activity. Clin. Exp. Immunol. 99, 419–24. immunodeficiency virus lineages in captive sooty mangabeys: ability to Franchini, G., Markham, P., Gard, E., Fargnoli, K., Keubaruwa, S., Ja- amplify virion RNA from fecal samples correlates with viral load in godzinski, L., Robert-Guroff, M., Lusso, P., Ford, G., Wong-Staal, F., plasma. J. Virol. 77, 2214–2226. Gallo, R.C., 1990. Persistent infection of rhesus macaques with a Mansfield, K.G., Lerche, N.W., Gardner, M.B., Lackner, A.A., 1995. molecular clone of human immunodeficiency virus type 2: evidence of Origins of simian immunodeficiency virus infection in macaques at The minimal genetic drift and low pathogenic effects. J. Virol. 64, 4462– New England Regional Primate research Center. J. Med. Primatol. 24, 4467. 116–122. Gao, F., Bailes, E., Robertson, D.L., Chen, Y., Rodenburg, C.M., Michael, Marx, P.A., Chen, Z., 1998. The function of simian chemokine receptors in S.F., Cummins, L.B., Arthur, L.O., Peeters, M., Shaw, G.M., Sharp, the replication of SIV. Semin. Immunol. 10, 215–223. P.M., Hahn, B.H., 1999. Origin of HIV-1 in the chimpanzee Pan Mellors, J.W., Munoz, A., Giorgi, J.V., Margolick, J.B., Tassoni, C.J., troglodytes troglodytes. Nature 397, 436–441. Gupta, P., Kingsley, L.A., Todd, J.A., Saah, A.J., Detels, R., Phair, J.P., I. Pandrea et al. / Virology 317 (2003) 119Ð127 127

Rinaldo Jr, C.R., 1997. Plasma viral load and CD4ϩ lymphocytes as munodeficiency virus replicates to high levels in sooty mangabeys prognostic markers of HIV-1 infection. Ann. Intern. Med. 126, 946– without inducing disease. J. Virol. 72, 3872–3886. 954. Santiago, M.L., Lukasik, M., Kamenya, S., Li, Y., Bibollet-Ruche, F., Mo, R., Chen, J., Han, Y., Bueno-Cannizares, C., Misek, D.E., Lescure, Bailes, E., Muller, M.N., Emery, M., Goldenberg, D.A., Lwanga, J.S., P.A., Hanash, S., Yung, R.L., 2003. T cell chemokine receptor expres- Ayouba, A., Nerrienet, E., McClure, H.M., Heeney, J.L., Watts, D.P., sion in aging. J. Immunol. 170, 895–904. Pusey, A.E., Collins, D.A., Wrangham, R.W., Goodall, J., Brookfield, Murphey-Corb, M., Martin, L.N., Rangan, S.R., Baskin, G.B., Gormus, J.F., Sharp, P.M., Shaw, G.M., Hahn, B.H., 2003. Foci of endemic B.J., Wolf, R.H., Andes, W.A., West, M., Montelaro, R.C., 1986. simian immunodeficiency virus infection in wild-living eastern chim- Isolation of an HTLV-III-related retrovirus from macaques with simian panzees (Pan troglodytes schweinfurthii). J. Virol. 77, 7545–7562. AIDS and its possible origin in asymptomatic mangabeys. Nature 321, Schols, D., De Clerq, E., 1998. The simian immunodeficiency virus mnd 435–437. (GB-1) strain uses CXCR4, not CCR5, as coreceptor for entry in Nerrienet, E., Amouretti, X., Mu¨ller-Trutwin, M.C., Poaty-Mavoungou, V., human cells. J. Gen. Virol. 79, 2203–2205. Bedjabaga, I., Nguyen, H.T., Dubreuil, G., Corbet, S., Wickings, E.J., Silvestri, G., Sodora, D.L., Koup, R.A., Paiardini, M., O’Neil, S.P., Mc- Barre-Sinoussi, F., Georges, A.J., Georges-Courbot, M.C., 1998. Phy- logenetic analysis of SIV and STLV type I in mandrills (Mandrillus Clure, H.M., Staprans, S.I., Feinberg, M.B., 2003. Nonpathogenic SIV sphinx): indications that intracolony transmissions are predominantly infection of sooty mangabeys is characterized by limited bystander the result of male-to-male aggressive contacts. AIDS Res. Hum. Ret- immunopathology despite chronic high-level viremia. Immunity 18, roviruses 14, 785–796. 441–452. Onanga, R., Kornfeld, C., Pandrea, I., Estaquier, J., Souquie`re, S., Rouquet, Simon, F., Souquiere, S., Damond, F., Kfutwah, A., Makuwa, M., Leroy, P., Poaty-Mavoungou, V., Bourry, O., M’Boup, S., Barre´-Sinoussi, F., E., Rouquet, P., Berthier, J.L., Rigoulet, J., Lecu, A., Telfer, P.T., Simon, F., Apetrei, C., Roques, P., Mu¨ller-Trutwin, M.C., 2002. High Pandrea, I., Plantier, J.C., Barre-Sinoussi, F., Roques, P., Muller- levels of viral replication contrasts with only transient changes in Trutwin, M.C., Apetrei, C., 2001. Synthetic peptide strategy for the CD4ϩ and CD8ϩ cell numbers during the early phase of experimental detection of and discrimination among highly divergent primate lenti- infection with simian immunodeficiency virus SIVmnd-1 in Mandrillus viruses. AIDS. Res. Hum. Retroviruses 17, 937–952. sphinx. J. Virol. 76, 10256–10263. Souquie`re, S., Bibollet-Ruche, F., Robertson, D.L., Makuwa, M., Apetrei, Pandrea, I., Onanga, R., Rouquet, P., Bourry, O., Ngari, P., Wickings, E.J., C., Onanga, R., Kornfeld, C., Plantier, J.C., Gao, F., Abernethy, K., Roques, P., Apetrei, C., 2001. Chronic SIV infection ultimately causes White, L.J.T., Karesh, W., Telfer, P.T., Wickings, E.J., Maucle`re, P., immunodeficiency in African non-human primates. AIDS 15, 2461– Marx, P.A., Barre´-Sinoussi, F., Hahn, B.H., Mu¨ller-Trutwin, M.C., 2462. Simon, F., 2001. Wild Mandrillus sphinx are carriers of two types of Pandrea, I., Onanga, R., Makuwa, M., Rouquet, P., Bourry, O., Ngari, P., lentiviruses. J. Virol. 75, 7086–7096. Bedjabaga, I., Apetrei, C., Roques, P., 2002. Lack of breast-feeding Sousa, A.E., Carneiro, B., Meier-Schellersheim, M., Grossman, Z., Vic- transmission during SIVmnd-1 primary infection in Mandrillus sphinx. torino, R.M., 2002. CD4 T cell depletion is linked directly to immune International Conference on AIDS, Barcelona, Spain, Abstr. We- activation in the pathogenesis of HIV-1 and HIV-2 but only indirectly PeA5743. to viral load. J. Immunol. 169, 3400–3406. Peeters, M., Courgnaud, V., Abela, B., Auzel, P., Pourrut, X., Bibollet- Takehisa, J., Harada, Y., Ndembi, N., Mboudjeka, I., Taniguchi, Y., Nga- Ruche, F., Loul, S., Liegeois, F., Butel, C., Koulagna, D., Mpoudi- Ngole, E., Shaw, G.M., Hahn, B.H., Delaporte, E., 2002. Risk to human nsop, C., Kuate, S., Zekeng, L., Ibuki, K., Shimada, T., Bikandou, B., health from a plethora of simian immunodeficiency viruses in primate Yamaguki-Kabata, Y., Miura, T., Ikeda, M., Ichimura, H., Kaptue, L., bushmeat. Emerg. Infect. Dis. 8, 451–457. Hayami, M., 2001. Natural infection of wild-born mandrills (Mandril- Peeters, M., Honore, C., Huet, T., Bedjabaga, L., Ossari, S., Bussi, P., lus sphinx) with two different types of simian immunodeficiency virus. Cooper, R.W., Delaporte, E., 1989. Isolation and partial characteriza- AIDS Res. Hum. Retroviruses 17, 1143–1154. tion of an HIV-related virus occurring naturally in chimpanzees in Traina-Dorge, V., Blanchard, J., Martin, L., Murphey-Corb, M., 1992. Gabon. AIDS 3, 625–630. Immunodeficiency and lymphoproliferative disease in an African green Rey-Cuille´, M.A., Berthier, J.L., Bomsel-Demontoy, M.C., Chaduc, Y., monkey dually infected with SIV and STLV-I. AIDS Res. Hum. Ret- Montagner, L., Hovanessian, A.G., Chakrabarti, L., 1998. Simian im- roviruses 8, 97–100.