INTRODUCTION

Top Abstract Introduction Materials and Methods Results Discussion References

Nonhuman primates (NHPs) are the natural hosts of a number of exogenous retroviruses, including simian immunodeficiency virus (SIV), simian T-cell lymphotropic virus (STLV), simian foamy virus (SFV), and simian type D retrovirus (SRV) (21). The close phylogenetic relationship between humans and NHPs increases the potential for cross-species transmission of these retroviral agents (33). Retroviral zoonoses have received increased attention both because of the growing body of evidence indicating that human immunodeficiency viruses types 1 and 2 originated from cross-species transmission of closely related viruses of chimpanzees (Pan troglodytes troglodytes) and sooty mangabeys (Cercocebus atys), respectively (6), and because of recent serious consideration given to the use of NHPs as potential organ donors for human xenotransplantation (1). Although the risk factors associated with cross-species transmission remain poorly defined, human infections with both SIV and SFV have recently been identified in individuals occupationally exposed to NHPs or their tissues (10, 12). Occupational exposure to simian retroviruses is of concern not only with regard to the potential adverse health effects for individual workers who are occasionally infected but also because of the potential for introduction of animal retroviruses into the general human population through secondary transmission from infected workers to intimate contacts.

The exogenous simian type D retroviruses are a group of closely related viruses that are enzootic in many captive and wild populations of macaques (genus Macaca) (18). In macaques, SRV has been identified as the etiologic agent of an infectious immunodeficiency disease that bears some clinical resemblance to AIDS in humans (18). In infected macaques, SRV may be present in blood, saliva, urine, and other body fluids, and a significant risk of exposure can reasonably be assumed in persons employed in the care and handling of NHPs (18). This is particularly true for workers employed before the current widespread use of personal protective equipment. Furthermore, the risk of exposure to NHPs and their body fluids was recently highlighted in a study that found a high frequency of needle sticks and mucocutaneous exposures in persons who reported working primarily with macaques, the NHPs most frequently used in biomedical research (30). Although SRV is known to replicate in cells of human origin (17), the zoonotic potential of this virus has not been thoroughly investigated.

Since the early 1970s, a number of surveys have been conducted to search for evidence of human infection with SRV. Most of these, however, were carried out in groups of patients with specific diagnoses (8, 19, 25) or in populations with no specific exposure to NHPs (4, 26, 27). For the majority of these studies, the results have been inconclusive, describing only partial seroreactivity, most commonly to a single SRV gag gene product (e.g., p25 or p27) by Western immunoblotting (WB) (4, 8, 19, 25-27). Type D retroviruses have also been isolated from human cell lines in culture, but the majority of these infections have been attributed to laboratory contamination (13, 28). In one study, detection of Mason-Pfizer monkey virus-related env-pol sequences by PCR in children with Burkitt's lymphoma was reported (14), but other studies have found no evidence of type D retrovirus infection in patients with non-Hodgkin's lymphoma or other lymphoproliferative or immunosuppressive illnesses (8).

The most compelling evidence to date of human SRV infection involved a homosexual male AIDS patient with lymphoma (2). SRV was isolated from the patient's lymphoma tissue, his bone marrow was positive for integrated proviral DNA for two viral regions by PCR, and antibodies to both gag and env SRV viral gene products were detected in the patient's serum by WB and radioimmunoprecipitation. (2). Characterization of this isolate revealed a close relationship to Mason-Pfizer monkey virus, the prototype simian type D retrovirus (now called SRV serotype 3 [SRV-3]), and to SRV-1 (5). This individual had no known history of contact with NHPs or their blood or tissues, and the source of his infection remains unknown.

An ongoing survey of individuals occupationally exposed to NHPs has recently identified human infections with two other exogenous simian retroviruses, SIV and SFV (10). Here we report the findings of SRV surveillance among the same cohort.

	MATERIALS AND METHODS

Top Abstract Introduction Materials and Methods Results Discussion References

Human subjects. As part of ongoing voluntary prospective surveillance for human infections with simian retroviruses among workers occupationally exposed to NHPs or their tissues, body fluids, or viruses, serum samples from 231 workers from 13 institutions in North America were tested for antibodies against SRV. Informed consent was obtained from all participants, and each participant completed a questionnaire regarding employment and potential exposure history. Additional archived as well as follow-up blood specimens were requested and obtained for analysis from individuals found to be positive or indeterminate on initial antibody testing.

Screening for antibodies to SRV. Serum specimens were obtained from coagulated blood and stored at 20 or 70°C until use. A four-tiered testing algorithm was used. Serum specimens were screened for the presence of antibodies to SRV by enzyme immunoassay (EIA) using SRV-1 and SRV-2 viral antigens as previously described (19). An optical density (OD) value that was twice the mean value of standard negative control sera run on the same plate was used as the cutoff. All specimens with OD values less than the cutoff were considered negative. All other specimens were further tested by WB. In addition, to increase the sensitivity of the EIA, sera with OD values below but within 20% of the calculated cutoff value were also further tested by WB.

Neutralization assay. Serum was tested for neutralizing activity against both SRV-1 and SRV-2 as previously described (22), with modifications. Briefly, serial twofold dilutions (1:10 to 1:640; final volume, 0.5 ml per well) of heat-inactivated (56°C for 30 min) sera with assay medium (RPMI 1640, 10% heat-inactivated fetal calf serum, L-glutamine, and Fungibact) were incubated for 30 min at 37°C with 0.5 ml of medium containing 40 50% tissue culture infective doses (TCID₅₀) of SRV-1 or 80 TCID₅₀ of SRV-2 in 24-well microtiter plates. Cultures were held at 37°C and 5% CO₂ with daily observations for cytopathic effect (CPE). Cells were harvested on day 5, and slides were prepared for IFA as previously described (17). SRV-1- and SRV-2-positive control sera were from a naturally infected rhesus macaque and cynomolgus macaque, respectively. WB-negative macaque and human sera were used as negative controls. Both CPE observations and IFA results were recorded for each well. The reported titer is the highest dilution that completely inhibited CPE and gave a negative IFA result.

Virus isolation. Peripheral blood lymphocytes (PBLs) were obtained from EDTA-treated whole blood by Ficoll-Hypaque separation. Following stimulation with phytohemagglutinin A or interleukin 2 for 48 h, 2.5 × 10⁶ to 5 × 10⁶ PBLs were cocultivated with equivalent amounts of either 2-day PHA-stimulated normal donor PBLs or the SRV permissive Raji B cell line using standard tissue culture techniques. The Raji cocultures were maintained in assay medium, and the PBL-PBL cocultures were maintained in assay medium with 10% interleukin 2. The PBL cocultures were fed with fresh 2-day PHA-stimulated PBLs every 10 days. All cocultures were maintained for 40 to 50 days and were monitored biweekly for CPE, for reverse transcriptase (RT) activity by using the Amp-RT assay (9), and for proviral SRV gag sequences by PCR as described below.

PCR analysis. PBL lysates were prepared from whole, unfractionated EDTA-treated blood specimens as previously described (31). In addition, monocyte, B-cell, CD4 T-cell, and CD8 T-cell subsets were enriched for by positive selection of the PBLs by using immunomagnetic beads directly conjugated with antibodies to CD19 (B cell), CD14 (monocytes and macrophages), CD4 (T lymphocytes and monocytes), and CD8 (T-cytotoxic and suppressor cells) (Dynal), as described elsewhere (3). All whole-cell and subset populations were lysed for PCR analysis at a concentration of 6 × 10⁶ cells per ml. In addition, DNA lysates were made from leukocyte nuclei obtained by sucrose lysis of 1 ml of whole blood, as previously described (7). Detection of human beta-globin sequences in the whole and enriched PBL fractions and the nuclear lysates was done to ensure the presence of amplifiable DNA in each sample.

View larger version (49K): [in this window] [in a new window]   FIG. 1.   (A) Sensitivity analysis of generic SRV gag PCR assay. The first 15 lanes contain reaction mixtures containing DNA lysates from 150,000 uninfected human PBLs spiked with DNA lysates equivalent to 15, 1.5, and 0.15 SupT1 cells infected with each respective SRV serotype; controls include DNA lysate from uninfected human PBLs (NEG) and water-only negative controls for the primary and nested PCR amplifications. (B to D) Proviral PCR analysis of SRV gag and pol sequences in DNA lysates prepared from unfractionated, fractionated, and nuclear PBLs from seropositive animal handlers. (B) Generic gag PCR; (C) generic pol PCR; (D) SRV-2-specific gag PCR. PBL-96 and PBL-97, DNA lysates prepared from unfractionated PBLs collected in 1996 and 1997, respectively, from subjects 1 and 2; NUC-96 and NUC-97, DNA lysates prepared from leukocyte nuclei collected at the same time points as the unfractionated PBLs; +CD14, +CD19, +CD8, and +CD4, DNA lysates prepared from monocyte-, B lymphocyte-, T-suppressor lymphocyte-, and T-helper lymphocyte-enriched PBL fractions, respectively; NEG, DNA lysate from uninfected human PBLs; H2O, water-only negative controls for the primary and nested PCR amplifications. The last two lanes show sensitivity controls containing DNA lysates from 150,000 uninfected human PBLs spiked with DNA lysates equivalent to 15 and 0.15 SupT1 cells infected with SRV-2. Not shown are the negative results for the unfractionated PBL samples collected from both subjects in 1998.

Animal inoculation A single juvenile rhesus macaque, negative for SIV, STLV, SFV, and SRV, was inoculated intravenously with 10 ml of heparinized whole blood obtained from an animal handler with persistent SRV seropositivity (subject 1). The experimental use and housing of this macaque was in accordance with guidelines set by the laboratory animal care and use committees at the respective institutions. Follow-up blood samples were obtained weekly for 2 weeks, then biweekly for 8 weeks, then monthly for 9 months. Also, an axillary lymph node biopsy was obtained at 5 months postinoculation. Serum samples were tested for SRV-specific antibody by WB. At each sampling time point, virus isolation was attempted by cocultivation of PBLs with both Raji and SupT1 cells, as previously described (17). Cultures were monitored for the appearance of CPE, and supernatant and cells were collected for RT assay and SRV PCR, respectively, on day 7 and weekly thereafter for 6 weeks. For PCR, 10⁷ cells were lysed in PCR cell lysis buffer and tested as described elsewhere (20). A single suspension of lymph node cells was prepared from the axillary lymph node biopsy and processed for culture and PCR as described above.

	RESULTS

Top Abstract Introduction Materials and Methods Results Discussion References

Screening for antibodies to SRV. Of 231 sera tested for SRV antibodies by EIA, 60 were found to be reactive or near the cutoff and were further tested by WB. Of these, 46 were determined to be negative and 12 were seroindeterminate, showing reactivity to a single gene product (p27). In further testing of these 12 samples by IFA, 5 were negative and 7 showed nonspecific reactivity (equal staining intensity and pattern in SRV-infected and -uninfected cells). Two of these 60 sera, however (cases 1 and 2), were found to have antibodies reactive against gag, pol, and env gene products of both SRV-1 and SRV-2 by WB on initial testing (Fig. 2). Both sera showed reactivity against p16, gp20, p24, p27, and p31 for both SRV-1 and SRV-2 (Fig. 2). Both blot-positive sera were also positive by IFA. Subsequent WB testing of two archived samples collected in 1995 from subject 1 found both to be SRV-1 positive (Fig. 3) and SRV-2 positive (data not shown), showing persistent seropositivity of at least 3 years' duration for this person. In contrast, two archived sera from subject 2 collected in 1994 were both negative for SRV antibodies (Fig. 3). In addition, the follow-up serum sample collected from subject 2 in 1997 was weakly positive against both SRV-1 (Fig. 3) and SRV-2 (data not shown), and the sample collected in 1998 was negative for SRV antibodies, indicating waning SRV antibody and eventual seroreversion after the initial positive sample, which had been collected in 1996 (Fig. 3). The apparent loss of reactivity to the putative pol gene product (p31) and the lesser gag proteins (p14 and p16), present in WB of initial samples (Fig. 2), in WB of archival and follow-up sera (Fig. 3) reflects the variable presence of these antigens, which is commonly seen in different lots of antigen.

View larger version (44K): [in this window] [in a new window]   FIG. 2.   WB reactivity against SRV-1 and SRV-2 on initial screening in sera from two persons occupationally exposed to NHPs. Lanes: +, SRV-positive control; , SRV-negative control; 1, subject 1 (sample date, 23 September 1996); 2, subject 2 (sample date, 20 November 1996). MW, molecular weight (weights are in thousands).

View larger version (54K): [in this window] [in a new window]   FIG. 3.   WB results for SRV-1 on archival and follow-up serum samples from two persons identified as SRV seropositive on initial screening. Lanes: +, positive monkey control; , negative monkey control (left) and negative human control (right); 1 to 7, subject 1 samples from 1 May 1995, 11 June 1995, 23 September 1996, 24 September 1996, 13 January 1997, 27 January 1997, and 31 July 1998, respectively; 8 to 12, subject 2 samples from 26 April 1994, 24 October 1994, 20 November 1996, 30 January 1997, and 18 December 1998, respectively. MW, molecular weight (weights are in thousands).

Patient histories. In the questionnaire and a personal interview, subject 1 described handling a variety of NHPs, including African green monkeys (AGM), pig-tailed macaques, and squirrel monkeys, for 23 years. During that time, subject 1 reported multiple exposures to NHP blood, body fluids, and fresh tissue, including two bite wounds by an AGM prior to 1975 that required suturing and a cut during necropsy of a squirrel monkey. Subject 1 had been shown previously to be infected with SFV originating from an AGM (10). Subject 2 reported working with NHPs such as rhesus and cynomolgus macaques, baboons, and owl monkeys for about 5 years and described receiving scratch wounds from a rhesus and a cynomolgus macaque in 1994 and 1996, respectively. In addition, subject 2 reported receiving scratch wounds from cages and pans used to house these macaques. Both subjects reported performing invasive procedures with NHPs, including venipuncture, tooth extraction, and surgery. Both subject 1 and subject 2 have remained in apparent good health. Spouses of the subjects were not available for testing to determine the transmissibility of the observed seroreactivity.

Neutralizing activity. Neutralization assays were performed with sera collected from both persons, using samples with evidence of stronger SRV seropositivity in each case, to determine if the observed seroreactivity was due to a specific SRV serotype. Serum collected from subject 1 was found to have a moderately high titer of neutralizing antibody (1:160) against SRV-2 (Table 2). Similarly, subject 2 had detectable neutralizing antibody (1:40) against SRV-2 but at lower levels than that seen in subject 1 (Table 2). These data suggest that both subjects may have been exposed to SRV-2.

Virus isolation. Attempts to isolate SRV from PBL of both seropositive persons, including at two time points for subject 1, by coculture with either normal donor PBLs or Raji cells were negative. During the 40 to 50 days in culture, there was no evidence of any markers of retroviral infection, such as CPE or RT activity. In addition, PCR analysis of tissue culture cells collected biweekly was negative for SRV gag sequences, suggesting the absence of latent, nonproductive infection of either the donor PBLs or Raji cells.

PCR analysis. Abundant -globin sequences were readily amplified from all unfractionated and fractionated PBL and nuclear PCR lysates, indicating the absence of PCR inhibitors and adequate DNA integrity in the test material (data not shown). All attempts to detect SRV proviral DNA from both PBL lysates from both subjects using generic SRV primers for both gag and pol sequences were negative. These negative PCR results were seen in four PBL samples from subject 1 and in two PBL samples from subject 2, both collected longitudinally over a period of 2 to 3 years (Fig. 1B and C). However, the negative results seen for subject 2 were with blood specimens collected from two time points in 1997 and 1998 that tested weakly positive and negative for SRV antibodies, respectively.

Animal inoculation. The single juvenile rhesus macaque inoculated intravenously with 10 ml of heparinized blood from subject 1 remains healthy and seronegative 21 months postinoculation. All attempts at virus isolation by cocultivation of monkey PBLs and peripheral lymph node cells with permissive cell lines and PCR analysis of monkey PBL and peripheral lymph node cells for SRV proviral DNA were negative.

	DISCUSSION

Top Abstract Introduction Materials and Methods Results Discussion References

The results of this study provide evidence that occupational exposure to NHPs may result in an increased risk of infection with SRV. To our knowledge, this is the first report of human sera demonstrating seroreactivity against SRV gag, pol, and env gene products. We demonstrate persistent seropositivity in one animal handler spanning 3 years and evidence of seroreversion in a second person occurring over a 2-year period. The persistent seropositivity observed in case 1 suggests continuous exposure to SRV antigens. However, the inability to isolate SRV from PBL cultures or detect evidence of SRV-infected cells in the circulation of either person despite the use of highly sensitive PCR methods suggests very low-level viremias. Collectively, these data suggest the presence of a persistent infection in subject 1 and a possible transient infection in subject 2.

The findings seen in subject 1 are consistent with data obtained from macaques that are naturally infected with SRV (18, 20, 29). SRV infection in macaques is thought to result in persistent infection, although SRV may become latent in the face of a vigorous immune response (18). In macaque populations where SRV is enzootic, it has been observed that the PBLs of some animals with strong antibody responses by immunoblotting show no evidence of viral presence by either cocultivation or PCR (20). It has been suggested that PBLs may not be the optimal tissue to analyze for detection of latent SRV infections. In one study, SRV proviral DNA was detected in bone marrow and other tissues from infected, seropositive macaques, although their PBLs repeatedly tested negative (24). Unfortunately, such specimens were not available from the seropositive persons in the present study. The mechanisms determining activation and latency of SRV infection are not well understood. It is interesting that the only human case of active SRV infection thus far reported occurred in a patient with severe immunosuppression from human immunodeficiency virus infection/AIDS and lymphoma (2).

The observed seroreversion in subject 2 is similar to that previously observed in a person who sustained a needle stick exposure to SIV (12). These data suggest that cross-species transmission of simian retroviruses may not always result in the establishment of a lifelong persistent infection.

Although there is significant antigenic cross-reactivity among the five recognized SRV serotypes, based on neutralization data, both seropositive humans in our study showed evidence of exposure to SRV-2. The presence of neutralizing antibody against SRV-2, but not SRV-1, provides evidence of a virus-specific immune response rather than nonspecific neutralization, as both the SRV-1 and SRV-2 viruses used in the neutralization assay were produced in the same cell line (SupT1). These data are also consistent with the high prevalence of SRV-2 among captive macaque species, particularly pig-tailed (Macaca nemestrina) and cynomolgus (Macaca fascicularis) macaques. Contact with at least one of these two macaque species was reported by both subject 1 and subject 2.

The risk for developing disease in SRV-infected humans or for secondary transmission to other humans is unknown. Spouses of both persons were unavailable for testing to determine whether this SRV seroreactivity is transmissible to humans, and while the absence of disease in both subjects is reassuring, our information is limited by the small sample size, the relatively short length of follow-up, and the absence of clinical observations coincident with infection and seroconversion. However, the anti-SRV neutralizing-antibody titers seen in both seropositive persons may offer them some degree of immunologic protection from viremia and subsequent disease manifestation, as has been recently observed in SRV-2-infected macaques with similar antibody titers and low or undetectable viral loads (29).

Subject 1 has shown persistent antibody to SRV over the 3-year period for which archived or follow-up serum samples were available, and it is possible that the duration of seropositivity may actually be much longer. Subject 1 has previously been found to be persistently infected with SFV (10). There is no antigenic cross-reactivity between SRV and SFV, and three other humans described as being persistently infected with SFV all tested SRV negative (10). To our knowledge, the finding of evidence of infection with two different simian retroviruses in a single individual is unprecedented. In contrast to STLV and SIV, both SRV and SFV are readily isolated from saliva and oropharyngeal secretions of infected NHPs (11, 16, 31), and penetrating bite wounds would represent a potentially efficient route of transmission for both agents. Transmission of SRV and SFV from a single exposure is unlikely, as the SFV infection in this person was derived from an AGM (Chlorocebus aethiops), a species not known to be a host for SRV. However, dual infection with two different simian retroviruses does serve to underscore the potential for cumulative exposure to multiple agents over a long career involving the handling of NHPs.

The findings of this study add to the body of existing data regarding the potential risk of cross-species transmission of simian retroviruses to humans in persons occupationally exposed to NHPs and their tissues or body fluids. These findings reemphasize the importance of occupational safety practices, including the use of personal protective equipment to reduce the risk of exposure to these viruses, and of efforts to establish and maintain retrovirus-specific pathogen-free colonies of macaques. The precautions taken against herpesvirus B (cercopithecine herpesvirus 1) transmission from macaques to humans may also limit the transmission of SRV and other retroviruses (30). However, our findings also serve to reinforce the need for continued and expanded surveillance for cross-species transmission among occupationally at-risk workers.