Novel Multiplexed HIV/Simian Immunodeficiency Virus Antibody Detection Assay

This assay identified new simian immunodeficiency viruses in primate bushmeat.

Like most emerging infectious disease viruses, HIV is also of zoonotic origin. To assess the risk for cross-species transmission of simian immunodefi ciency viruses (SIVs) from nonhuman primates to humans in the Democratic Republic of the Congo, we collected 330 samples derived from nonhuman primate bushmeat at 3 remote forest sites. SIV prevalences were estimated by using a novel highthroughput assay that included 34 HIV and SIV antigens in a single well. Overall, 19% of nonhuman primate bushmeat was infected with SIVs, and new SIV lineages were identifi ed. Highest SIV prevalences were seen in redtailed guenons (25%) and Tshuapa red colobus monkeys (24%), representing the most common hunted primate species, thus increasing the likelihood for cross-species transmission. Additional studies are needed to determine whether other SIVs crossed the species barrier. With the newly developed assay, large-scale screening against many antigens is now easier and faster.
Humans are still hunting and butchering a wide diversity of primate species, and the possibility of additional cross-species transfers of viruses has to be considered (7,8). Recent reports showed ongoing transmission of simian retroviruses to humans in central Africa, i.e., a wide variety of simian foamy viruses and new human T-lymphotropic virus variants, closely related to viruses in co-habiting nonhuman primates, have been observed in humans who report primate hunting and butchering (9)(10)(11)(12). The description in 2009 of HIV-1 group P, closely related to SIVgor, in a patient from Cameroon living in France, shows also that our knowledge on HIV diversity and possible cross-species transmissions is still incomplete and illustrates how rapidly new viruses can spread today to other continents (6).
Given the potential pathogenicity of these lentiviruses, as illustrated by the actual HIV-1 group M pandemic that resulted from a single cross-species transmission, it is necessary to estimate to what extent humans are exposed to SIVs and whether other viruses crossed the species-barrier. SIV infection has already been identifi ed in >40 nonhuman primate (NHP) species from Africa but our knowledge on prevalence and geographic distribution remains limited; few large-scale studies on retroviral infections in wild Novel Multiplexed HIV/Simian Immunodefi ciency Virus Antibody Detection Assay primate populations have been conducted (13). SIV prevalences can vary among species and within species according to geographic areas (2,14,15), and exposure to infected primates and subsequent risk for cross-species transmission can thus differ across Africa. SIV infections were initially identifi ed on the basis of cross-reactivity with HIV antigens (8), but to increase sensitivity, SIV lineage-specifi c ELISAs have been developed. These assays must be regularly updated when new SIV lineages are discovered (14)(15)(16)(17). Therefore, they become time-consuming and bench work-consuming, use relatively large volumes of scarce biological material, and are not adapted for large-scale surveillance studies. We adapted the Multiple Analyte Profi ling technology (xMAP; Luminex Inc., Austin, TX, USA), which is a fl ow cytometry-based system (18), for simultaneous antibody detection against 34 peptides representing the actual known HIV/SIV diversity. This new assay was used to study SIV infection in primate bushmeat in the Democratic Republic of Congo (DRC), home to a wide diversity of primate species.

NHP Samples
For the validation of the HIV/SIV xMAP assay, we used 142 well-characterized samples from our NHP reference panel in which SIV infection was either confi rmed or ruled out by highly sensitive PCR approaches, and for which suffi cient plasma was available (14,15). The panel included 93 SIV-negative samples from 8 species and 49 SIV-positive samples from 9 species (Table 1). For SIV prevalence studies, 330 samples were collected during May 2009-2010 as dried blood spots (DBSs) around 3 rural cities in DRC ( Figure 1). Whole blood, collected from primate bushmeat, was spotted onto a fi lter 903 FTA card (Whatman Plc, Kent, UK). After air-drying at ambient temperature, DBSs were stored into individual envelopes at ambient temperature. Animals died 6-78 hours before sampling. All NHP samples were obtained with approval from the Ministry of Environment and Health and the National Ethics Committee. Similar to our previous studies, bushmeat samples were obtained through a strategy specifi cally designed not to increase demand (8,15).

Screening for Cross-Reactive HIV/SIV Antibodies
All DBS samples were screened with the new HIV/ SIV multiplex microbead immunoassay technology, i.e., xMAP. Similarly as for the SIV ELISAs, we used peptides covering the immunodominant region of the gp41 transmembrane and V3-loop region from all major SIV/ HIV lineages known at the time we conducted this study (Table 2). To avoid interpeptide and intrapeptide cross-linking, the 2 cysteins of the gp41 peptides were cyclicized during synthesis. For SIVcol, gp41 peptides could not be synthesized because of their low solubility; the V3-loop peptide was used to identify corresponding antibodies (14,15). Peptides were covalently coupled on carboxylfunctionalized fl uorescent polystyrene beads (Luminex Inc.) by using the Bio-Plex Amine Coupling Kit (Bio-Rad Laboratories, Marnes-la-Coquette, France) according to the manufacturer's instructions. Unreacted sites were blocked with blocking buffer from the Amine Coupling Kit (Bio-Rad Laboratories). Peptide-coupled microsphere preparations were counted by using a hemocytometer and stored in the dark at 4°C. Before use, peptide-coupled beads were vortexed (30 s), sonicated (30 s), and diluted to 4,000 beads/μL. Dilution and washing buffer consisted of phosphate-buffered saline (PBS) containing 0.75 mol/L NaCl, 1% (wt/vol) bovine serum albumin (Sigma Aldrich, St. Quentin Fallavier, France), 5% (vol/vol) fetal bovine serum (Gibco-Invitrogen, Cergy Pontoise, France), and 0.05% (vol/vol) Tween-20 (Sigma-Aldrich).
Assays were performed in 96 well fl at-bottomed fi lter plates (Millipore, Tullagreen, Ireland). Plates were prewet with 100 μL assay buffer; 50 μL of bead mixture was subsequently added to each well. Liquid was aspirated with a vacuum manifold and wells were washed with 100 μL of assay buffer. Wells were then incubated with 100 μL plasma (diluted 1:200) for 60 min at room temperature in the dark on a plate shaker (300 rpm/min). For DBSs, 100 μL of eluate, obtained after overnight incubation of two 6-mm disks in 1 mL of hypertonic PBS, was added to Data were analyzed by using BioPlex Software Manager version 5.0 (Bio-Rad Laboratories). For each bead set, >100 events were read and results were expressed as median fl uorescence intensity (MFI) per 100 beads. The cutoff value was calculated for each peptide as the mean MFI for all antibody-negative reference serum samples plus 5 SD as an adaptation of the strategy defi ned for ELISA and was set at 200 MFI corresponding to a consensus value for all peptides (19). Sensitivity and specifi city were calculated for homologous (same species) and heterologous (different species) antibody detection.
DBS samples were also tested for HIV cross-reactive antibodies by using the INNO-LIA HIV Confi rmation Test (Innogenetics, Gent, Belgium) as described (10). This test confi guration includes HIV-1 and HIV-2 recombinant proteins and synthetic peptides that are coated as discrete lines on a nylon strip.

DNA Extraction and NHP Species Confi rmation
Total DNA was extracted from all DBSs by using the Nuclisens MiniMAG Extraction Kit (Biomerieux, Craponne, France) according to the manufacturer's instructions. Minor changes consisted of increasing the incubation time (2 h) of the viral lysis step to increase DNA release (20). Species identifi cation recorded in the fi eld was confi rmed on all samples by amplifying a 386-bp mitochondrial DNA fragment of the 12S rRNA gene with  primers 12S-L1091 and 12S-H1478 (21). PCR products were purifi ed by electrophoresis on a 1% agarose gel and directly sequenced (ABI PRISM Big Dye Terminator Cycle sequencing Ready Reaction Kit with amplitaq FS DNA polymerase) on an automated sequencer (ABI 3130XL, Applied Biosystems, Courtaboeuf, France). Sequences were then assembled by using the software package Lasergene (DNASTAR, Inc, Madison, WI, USA).

Molecular Characterization and Phylogenetic Analyses of SIVs
PCR analyses were performed on SIV antibodypositive samples by using described conditions with universal HIV/SIV and SIV lineage-specifi c primers in pol or env shown in Table 3 (8,15,(22)(23)(24). PCR products were purifi ed by electrophoresis on a 1% agarose gel and directly sequenced as described above. Newly derived SIV nucleotide sequences were aligned with reference sequences of the different HIV/SIV lineages with MEGA4 and ClustalX version 2 (25) and minor manual adjustments when necessary. Nucleotide sites that could not be unambiguously aligned were excluded from the analyses. Appropriate models of evolution were selected for each data set by using Topali software (26) and maximum-likelihood phylogenies were reconstructed by using PhyML (27). The analyses were performed by using discrete gamma distribution and generalized time-reversible model. The starting tree was obtained by using PhyML. One hundred bootstrap replications were performed to assess confi dence in topology. New sequences have been deposited in GenBank under accession nos. JN020273-JN020279 and GU989632.   Table 4 summarizes the sensitivity and specifi city of homologous and heterologous antibody detection of the xMAP assay on the same reference panel that was used for the SIV lineage-specifi c ELISAs (17,18

Prevalence and Genetic Diversity of SIVs in DRC
Because 4 of the 7 NHP species or subspecies are only present in DRC, they are most likely infected with SIVs that have not been documented, and antibody detection will thus depend on extent of cross-reactivity with antigens of known HIV/SIV lineages. Therefore, we screened all samples with the new SIV/HIV xMAP assay and with the HIV line immune assay (INNO-LIA) confi rmation assay, which we previously used to detect a large diversity of SIV infections (8). Table 5 shows the number of SIV-positive and indeterminate samples per species. SIV infection was documented in 6 species and the overall prevalence was 19% (64/330) ranging from 0% to 25% per species. Highest SIV prevalences were seen in red-tailed guenons (25%) and Tshuapa red colobus (24%), which represent 70% (226/330) of the primate bushmeat  collected in this study. We also observed HIV/SIV crossreactive antibodies in De Brazza monkeys (20%), Wolf's monkeys (12 %), black mangabeys (3%), and Angola pied colobus (4%). In addition, 3% (10/330) of the samples were considered as indeterminate for SIV. Notably, all samples from Tsuapa red colobus were only reactive in the xMAP assay and showed strong cross-reactivity with SIVwrc antigens from western red colobus (Piliocolobus badius), illustrating clearly the need for including a wide variety of SIV antigens to uncover new SIV lineages.

Genetic Diversity of SIVS in DRC
To confi rm SIV infection and document SIV diversity, all SIV-positive and indeterminate samples were subjected to PCR amplifi cation. Although DNA integrity was suffi cient to confi rm the primate species in all DBSs, proviral SIV DNA could only be amplifi ed in pol (400 bp) for 8 samples, most likely because of DNA degradation related to long and suboptimal storage at ambient temperature in the fi eld and the fact that animals died several days before sampling. SIV infection was confi rmed in 4 red-tailed guenons, 1 Wolf's monkey, 1 De Brazza monkey, and 2 Tshuapa red colobus. Phylogenetic tree analysis shows the presence of new SIV lineages in Wolf's monkeys and Tshuapa red colobus (Figure 2). SIVwol is close to SIVden obtained from Dent's monkeys (C. mona denti), which are found in eastern DRC but without overlapping habitats with Wolf's monkeys in central DRC (28). SIVtrc from Tshuapa red colobus forms a separate lineage although related to SIVkrc from Kibale red colobus in eastern Africa (29). SIVasc from red-tailed guenons forms a speciesspecifi c lineage with SIVasc described in a capitive animal, but a high genetic diversity is seen (30). The reported pol sequence from a captive black mangabey housed in the zoo in Kinshasa, also falls within the SIVasc radiation (31). Finally, SIVdeb clustered within the species-specifi c SIVdeb lineage observed for De Brazza monkeys across central Africa (32).

Discussion
In this study, we used a novel high throughput immune assay that included 34 HIV and SIV antigens in a single well to evaluate prevalence and genetic diversity of SIVs from NHPs at the primate/human interface in DRC. Overall, we showed that ≈20% of NHP bushmeat is infected with SIVs and identifi ed new SIV lineages. Highest SIV prevalences were seen among the most commonly hunted primate species.
Although SIV lineage-specifi c ELISAs were highly sensitive and specifi c (14)(15)(16)(17), with the increasing number of new SIV lineages and the high genetic diversity within SIV lineages, a large number of antigens must be included. Using a large set of SIV antigens is necessary, especially when new species are tested for which no SIVs have been reported and when antibody detection is based on crossreactivity with antigens from heterologous SIV/HIV lineages. To reduce increasing workload and volumes of scarce biological material, we adapted xMAP technology to enable a single sample to be tested simultaneously for multiple peptides (18). We used the same gp41 and V3loop peptides as in the SIV lineage-specifi c ELISAs and the same reference panel of NHP samples to validate the assay (14,15). We updated the assay with antigens of SIVgor from gorillas (3) and HIV-1 groups M, N, and O. The homologous reactivity for some gp41 peptides, especially SIVmnd, was lower in the xMAP assay (95.9%) compared with that of ELISAs (97.5%) (15). However, these samples were also only weakly reactive in the gp41 ELISA, suggesting that all antigens in a single well could slightly reduce sensitivity when mismatches are present in corresponding gp41 sequences. The combination of 34 peptides in a single well detected SIV infection in the reference panel with 100% Figure 2. Phylogenetic relationships of the newly derived simian immunodefi ciency virus (SIV) sequences in pol to representatives of the other SIV lineages. Newly identifi ed strains in this study are in red and reference strains are in black. Unrooted trees were inferred from 350-bp nucleotides. Analyses were performed by using discrete gamma distribution and a generalized time reversible model. The starting tree was obtained by using phyML (27). One hundred bootstrap replications were performed to assess confi dence in topology. Numbers at nodes are from 100 bootstrap replicates; only those >90% are shown with an asterisk. Scale bar represents nucleotide replacements per site. sensitivity and 97.9% specifi city and reduced workload and volumes of biological material. The need for including a wide diversity of SIV antigens was clearly illustrated by identifi cation of SIVtrc in Tshuapa red colobus samples, which showed negative results with the INNO-LIA HIV confi rmatory assay.
No extensive studies have been conducted on SIV infection in monkeys from DRC, which harbors many species because of the geographic barriers constituted by the Congo, Ubangui, and Kasai Rivers (28). Overall, ≈20% of primate bushmeat was SIV infected, and as observed in previous studies, prevalences varied per species (17,18). We confi rmed SIV infection in De Brazza monkeys and red-tailed guenons (14,15,30) and identifi ed new SIV lineages in Wolf's monkeys (SIVwol) and Tshuapa red colobus (SIVtrc). De Brazza monkeys seem widely infected with SIVdeb across central Africa: 20% in DRC and 40% in Cameroon (15). A high genetic diversity is seen in the SIVasc lineage, and this lineage also includes the previously reported SIVbkm sequence from a captive black mangabey from the zoo in Kinshasa, DRC (31). Attempts to amplify an SIV in a wild black mangabey in our study were unsuccesful, and more studies are needed to clarify whether the initial SIVbkm infection is caused by contamination from a red-tailed monkey in captivity or in the wild because black mangabeys share habitats with Cercopithecus species. Finally, only full-length genome sequences will enable understanding of the evolutionary history of the new SIVwol and SIVtrc viruses.
In addition to many other factors, risk for cross-species transmissions most likely depends on frequency of human contacts with infected primates and on prevalences in frequently hunted species (33). For example, SIVcpzPtt and SIVsmm prevalences are highest (30% and 50%, respectively) in areas in west-central and western Africa where precursors of HIV-1 M (M and N) and HIV-2 (A and B) have been identifi ed in chimpanzees and mangabeys, respectively (2,34). In contrast to our study on SIV prevalences in primate bushmeat in Cameroon, in which we showed <3% seroprevalence (15), we observed in DRC high rates of SIV infection (19%) and highest prevalences in the 2 most commonly hunted species. This observation shows clearly that exposure to SIV differs across Africa and that the likelihood for SIVs to cross the species barrier could be higher in DRC than in Cameroon. In addition, contemporary simian foamy virus infections from Angolan pied colobus and red-tailed guenons have been identifi ed in persons living around an area where we collected samples from primate bushmeat, thus confi rming ongoing crossspecies transmission with simian retroviruses (35).
Given the enormous size of the country and absence of road infrastructure, persons in DRC rely on bushmeat for subsistence in many areas. Among ≈70 million inhabitants, 60% live in rural areas (36). DRC is recognized as the area where the HIV-1 group M epidemic originated (37), but exact conditions associated with epidemic spread of HIV in this part of the world are still incomplete. The internal and regional armed confl icts in DRC that started in 1997 have led to profound socioeconomic changes and internal displacement of human populations (38). The long period of civil unrest that followed the outbreak of those confl icts damaged the health care system, and today 70% of the population has little or no access to health care, including HIV/AIDS services (36). New epidemic outbreaks, especially with diseases having a long incubation period, can go unrecognized for a long period. In addition, these SIV strains are not recognized by commercial HIV-1/ HIV-2 screening assays. A multiplying factor is the presence of logging and mining industries in remote areas that provide favorable conditions for increased human contact with primates and exchanges between urban and rural settlements. High HIV prevalence was reported around logging industries and in displaced population groups (39,40), which could lead to recombinants between HIVs and SIVs and enable more effi cient adaptation and replication in the new host in addition to rapid and further spread to other geographic areas.
Prevalences and high exposure are among the factors that most likely play a role in the transmission of certain SIVs and other simian retroviruses to humans, but viral and host factors also play a role in establishing effi cient infection and disease. Given the ongoing contacts between infected NHP and African populations through hunting and butchering, it is likely that SIV cross-transmissions are still occurring. Viruses can remain unrecognized because of a weak health infrastructure and because they are not detected by commercial HIV screening assays. With increasing travel, new viruses can reach other areas rapidly, which have favorable conditions for epidemic spread. It will be necessary to determine whether other SIVs crossed the species barrier, especially in human populations exposed to highly infected primates and in populations with risk behavior that is associated with high epidemic spread. With the new assay that we developed in this study, large-scale screening against a wide variety of antigens is now easier and faster.