Reduced Prevalence of Epstein-Barr Virus-Related L
http://www.100md.com
病菌学杂志 2005年第15期
Department of Medicine, Brigham & Women's Hospital, Harvard Medical School, Boston, Massachusetts 02115
Department of Primate Resources, New England Primate Research Center, Harvard Medical School, Southborough, Massachusetts 01772
ABSTRACT
The recent discovery of an Epstein-Barr virus (EBV)-related lymphocryptovirus (LCV) naturally infecting common marmosets demonstrated that gamma-1 herpesviruses are not limited to human and Old World nonhuman primate hosts. We developed serologic assays to detect serum antibodies against lytic- and latent-infection marmoset LCV antigens in order to perform the first seroepidemiologic study of LCV infection in New World primates. In three different domestic colonies and in animals recently captured from the wild, we found that the seroprevalence of marmoset LCV infection was not as ubiquitous as with EBV or Old World LCV. These biologic differences in LCV infection of New World versus human and Old World primate hosts correlate with the evolution of the LCV viral gene repertoire.
TEXT
Serologic responses to lytic and latent infection antigens, such as viral capsid antigen (VCA) and EBNA-1, are widely used to document Epstein-Barr virus (EBV) infection (8, 13). Old World (4), and more recently New World (2), nonhuman primates are known to be naturally infected with related herpesviruses in the same lymphocryptovirus (LCV) genus as EBV. LCV infection in Old World primates was initially recognized by the presence of serum antibodies cross-reactive with viral antigens in EBV-infected B cells (7). As with humans, LCV seropositivity in Old World primates is highly prevalent both in nature and in domesticated colonies, with seropositivity in more than 95% of adult animals (5, 7, 9). The biology of LCV infection in Old World primates appears to be nearly identical to that of EBV infection in humans (16). This is concordant with the identical repertoire of viral genes and the high degree of sequence homology between EBV and rhesus LCV, a prototype for an Old World LCV whose genome has recently been fully sequenced (11).
It was long believed that LCV did not infect New World primates, since there was no strong evidence of EBV cross-reactive antibodies from these species. However, we recently isolated a B-cell-immortalizing herpesvirus from a spontaneous B-cell lymphoma arising in a common marmoset (Callithrix jacchus) that belonged to the LCV genus (2). PCR studies showed that LCV infection was present in three different marmoset colonies, and a similar LCV was also found in squirrel monkeys, indicating that LCV infection is present in species from both major families of New World primates (2). Studies from other laboratories have since confirmed that LCV can be found in a variety of New World species (3, 6).
The complete genome sequence of the marmoset LCV is striking for its differences from the EBV genome (10). The marmoset LCV encodes at least seven unique open reading frames (ORFs) with no sequence homology to other cell or viral genes. In most cases, there is compelling evidence that these unique genes are predecessors for certain EBV genes, e.g., they are located in the same position of the genome, they have similar transcriptional patterns, they have similar predicted secondary protein structures, or they have similar functional properties demonstrable in vitro. The marmoset LCV also lacks apparent homologs for 14 genes found in EBV and rhesus LCV. These genes are unique to herpesviruses in the LCV genus, i.e., they are not found in other alpha-, beta-, or gamma-2 herpesviruses, and appear to have been acquired in the evolution of LCV from New World primates to humans. Most of these acquired genes are lytic infection genes, and all are nonessential for EBV replication and B-cell immortalization in vitro, suggesting that they play an important role for LCV infection in vivo, rather than in vitro. Thus, the marmoset LCV appears to be a more primitive predecessor of EBV lacking a number of viral genes that have evolved in higher-order primates. Little is known about the biology of LCV infection in New World primates and whether this difference in viral gene repertoire results in biologic differences between EBV and marmoset LCV infection. Therefore, we asked whether seroprevalence of LCV infection in New World primates was as high as that seen in Old World primates and humans.
We obtained sera from common marmosets at the New England Primate Research Center (NEPRC) and tested them for reactivity to the marmoset LCV ORF59 encoding the small viral capsid antigen (sVCA), p18, or BFRF3 homologue. The carboxy terminus of the sVCA is known to be an immunodominant target in humans and rhesus macaques, and peptides derived from the carboxy terminus of either EBV or rhesus LCV BFRF3 can be used in enzyme immunoassays to detect serologic responses in virtually all LCV-infected humans or rhesus macaques (9, 12-15). Therefore, we synthesized a 24-residue peptide from the carboxy terminus of the marmoset ORF59 (Fig. 1) and tested sera from 235 common marmosets in the sVCA enzyme immunoassay (EIA). As shown in Fig. 2A, (63%) serum samples tested positive, with values at least three times above background values, and 87 (37%) serum samples tested negative. This is strikingly less than the 100% of sera from NEPRC rhesus macaques in conventional housing that test positive in a similar rhesus LCV sVCA EIA (9).
In order to test whether the seroprevalence for marmoset LCV infection determined by the sVCA EIA was accurate, we looked for serum antibodies against the marmoset LCV EBNA-1 homologue, ORF39. A Flag-tagged marmoset LCV ORF39 was expressed from a recombinant vaccinia virus and used to infect BSC-40 cells. ORF39 was affinity purified from cell lysates with a Flag monoclonal antibody and used as the antigen for an EIA. The same NEPRC serum samples were tested for ORF39 reactivity, with 152 (65%) samples testing positive and 83 (35%) of the samples testing negative (Fig. 2B). Thus, by serologic testing against a latent infection antigen, there were still a significant number of animals with no serologic evidence of marmoset LCV infection.
When the sVCA and ORF39 reactivities were compared, 128 (54%) of the animals were seropositive against both lytic and latent antigens, and 64 (28%) of the animals were negative for reactivity to both antigens (Table 1). The remaining 42 animals (18%) were positive in only one assay, with approximately an equal number positive in either the ORF39 or sVCA assay. The observation that more than one-fourth of the animals showed no serological reactivity to immunodominant lytic and latent antigens suggests that LCV infection is not as prevalent in marmosets as in humans or other Old World primates. In order to test whether other viral antigens may be more immunodominant than ORF39 and sVCA in marmosets, a subset of EIA positive and negative sera were tested by immunoblotting against cell lysates of marmoset LCV-infected cells induced by treatment with phorbol ester and butyrate or not induced for lytic replication. The EIA negative sera failed to detect any significant proteins in the cell lysates, consistent with the interpretation that these animals were naive for marmoset LCV infection (data not shown). The high percentage of seronegative animals did not appear to be due to an unusually high number of younger animals in this group, since the average age of the animals in the double-negative group (4.7 years old) was actually higher than that of the double-positive group (3.9 years old). The average age of animals with discrepant ORF39 and sVCA results was 2.2 years old, suggesting the possibility that these animals may be closer to the time of acquiring LCV infection.
In order to validate the serologic data, we used DNA PCR of peripheral blood mononuclear cells to look for evidence of persistent marmoset LCV infection. In our previous studies, approximately 60% from a small group of domestic animals were PCR positive for marmoset LCV DNA in peripheral blood mononuclear cells (2). However, since these studies were not linked to serologic tests, it was not known whether PCR-negative animals were not infected versus infected with undetectable levels of DNA. In order to increase the sensitivity of the PCR amplification, we designed new primers from the major internal repeat (5'-GGACCCGAGGAACTGTCTAAAAC-3' and 5'-TGTGTAGGCACCAAGTCTGCC-3') instead of using the previously described primers from a single-copy gene (2). We obtained additional blood samples from 24 animals for serologic and DNA PCR testing in parallel. Eighteen animals were seropositive (10 positive for both sVCA and ORF39, 3 positive for sVCA only, and 5 positive for ORF39 only), and 6 animals were seronegative by both serologic tests. Peripheral blood mononuclear cells from all 18 seropositive animals were also positive by marmoset LCV DNA PCR, and DNA PCR was negative for 5 of the 6 seronegative animals. These results show that DNA PCR or the combination of sVCA and ORF39 serologic tests are sensitive assays for LCV infection that detect almost identical populations. In addition, a significant number of animals are negative for both DNA PCR and sVCA/ORF39 serologic responses, suggesting that they have not been infected. Whether a minor subpopulation of animals can be LCV infected with little or no serologic response requires further studies.
To determine whether the lower seroprevalence of marmoset LCV infection was common to other marmoset colonies, sera from two geographically separated marmoset colonies were obtained. In these two colonies, 37% and 47% of the animals tested positive by the sVCA EIA (n = 165 and 126, respectively). The results support the findings that LCV infection may not be as ubiquitous among marmosets as it is among humans and Old World primates.
Common marmosets are typically housed in smaller units than Old World primates, so a lower prevalence of marmoset LCV infection could be due to segregation of seropositive and seronegative animals in domesticated colonies. Therefore, we examined the housing patterns of animals in relation to seropositivity. Out of 91 animals in 37 cages at the NEPRC, 5 cages contained all sVCA-seropositive animals, 7 cages contained all seronegative animals, and 25 cages contained both seropositive and seronegative animals. Thirty of forty-three seropositive animals were housed in cages with both seropositive and seronegative animals. Similarly, 29 out of 48 seronegative animals were housed in cages with both seropositive and seronegative animals. Thus, a significant portion of seronegative animals (19 of 48; 40%) were segregated with other naive animals, suggesting that housing practices may contribute to a lower seroprevalence of marmoset LCV infection. However, the large percentage of mixed cages and large number of seronegative animals in mixed cages (60%) also suggest that LCV infection may not be readily transmitted among marmosets. In contrast, virtually all newborn Old World primates, such as rhesus macaques and baboons, turn seropositive within 1 year when housed with other seropositive animals (5, 9). In order to eliminate potential bias from domestic housing, sera collected from common marmosets shortly after capture from the wild were also tested. Twelve out of 24 animals (50%) tested positive by the sVCA EIA, indicating reduced seroprevalence among New World primates in the wild, similar to animals in domestic colonies.
These are the first serologic studies of LCV infection in New World primates. Historically, the failure to reliably detect EBV cross-reactive antibodies in New World primates was probably due to the degree of sequence divergence between EBV and marmoset LCV genes, exacerbated by additional divergence between human and marmoset immunoglobulins. Thus, important technical aspects in these studies were the use of antigens derived from marmoset LCV sequences and anti-human immunoglobulin secondary reagents that had not been absorbed for reactivity to immunoglobulins from other mammalian species. The combined use of lytic and latent antigens that are immunodominant in EBV and rhesus LCV infection identified largely identical positive and negative populations among NEPRC animals. ORF39- and ORF59-negative sera did not react with any other specific bands on immunoblots with LCV-infected cell lysates induced for viral replication, consistent with LCV-naive hosts. Reduced seroprevalence of marmoset LCV infection was consistently found in two other domestic colonies and in animals recently captured from the wild.
These results suggest that LCV infection may not be as prevalent in marmosets as in humans and in Old World primates, such as rhesus macaques. Results of the seroprevalence studies with these larger populations are consistent with our previous data obtained using nested PCR amplification of peripheral blood lymphocytes from a smaller number of animals at the Wisconsin and New England Primate Research Centers, 60% and 44% positivity, respectively (2). Analysis of the current data suggests that age and housing may have had some impact on the prevalence of seronegative animals, but these factors do not completely account for the significant differences between the ubiquitous LCV infection of adult humans and rhesus macaques versus a significant seronegative subpopulation in different marmoset colonies and in the wild.
Additional studies will be required to confirm whether the LCV seroprevalence is also reduced in other New World primate species. A lower seroprevalence of New World LCV infection versus EBV and Old World LCV infection could be due to less efficient transmission, i.e., a decreased rate of new infection, or LCV infection that is less persistent, i.e., transient infection or loss of infection. Limited studies of serial serum samples from NEPRC marmosets did not reveal any evidence for loss of seropositivity that may result from infection that fails to persist (data not shown). Thus, we believe it is more likely that marmoset LCV infection may be transmitted less efficiently than EBV or rhesus LCV infection, e.g., less viral shedding, less-efficient infection, or better host resistance. The evolution of genes in EBV and rhesus LCV that are not present in marmoset LCV may provide a genetic basis for this biologic phenotype. Genes encoding a viral interleukin 10 and CSF-1R homologue, BCRF1 and BARF1, are also absent in the marmoset LCV. These are soluble cytokines produced during lytic LCV infection that could inhibit local immune responses to immunogenic viral antigens during viral replication. One could speculate that the lack of these immunomodulators could affect transmission by decreasing the amount of viral replication and shedding from the oral cavity or reducing the ability for new viral infection to evade oral mucosal immune responses. The marmoset LCV does not have apparent homologues for EBERs and BARF0. These latent infection genes are not essential for B-cell immortalization in vitro but may be important for successful or efficient B-cell infection in vivo. The marmoset LCV also lacks a BDLF3 homologue. This encodes a 150-kDa glycoprotein, and paradoxically, deletion of this glycoprotein enhances epithelial cell infection in vitro (1). Four of the EBV genes missing in marmoset LCV have unknown function (BLLF2, LF3, ECRF4, and BNLF2), so it remains to be determined whether any of these genes may contribute to efficiency of virus transmission. Development of a genetic system and an experimental infection model for marmoset LCV may make it possible to introduce these genes into marmoset LCV in order to directly test whether these viral genes might improve the efficiency of LCV infection in marmosets. Alternatively, deletion of these viral genes from rhesus LCV may result in a reduced efficiency of LCV infection in Old World primates.
ACKNOWLEDGMENTS
This work was funded by a grant from the U.S. Public Health Service (CA89172). Services from the New England Primate Research Center were supported by a base grant to the institution (USPHS P51RR00168).
All animal experiments were performed with approval from the Committee on Animals for Harvard Medical School, and animals were maintained in compliance with federal and institutional guidelines for animal care. We thank Jose Augusto Muniz and the Centro Nacional de Primatas for providing sera from marmosets shortly after capture from the wild.
REFERENCES
Borza, C. M., and L. M. Hutt-Fletcher. 1998. Epstein-Barr virus recombinant lacking expression of glycoprotein gp150 infects B cells normally but is enhanced for infection of epithelial cells. J. Virol. 72:7577-7582.
Cho, Y., J. Ramer, P. Rivailler, C. Quink, R. L. Garber, D. R. Beier, and F. Wang. 2001. An Epstein-Barr-related herpesvirus from marmoset lymphomas. Proc. Natl. Acad. Sci. USA 98:1224-1229.
Ehlers, B., A. Ochs, F. Leendertz, M. Goltz, C. Boesch, and K. Matz-Rensing. 2003. Novel simian homologues of Epstein-Barr virus. J. Virol. 77:10695-10699.
Frank, A., W. A. Andiman, and G. Miller. 1976. Epstein-Barr virus and nonhuman primates: natural and experimental infection. Adv. Cancer Res. 23:171-201.
Jenson, H. B., Y. Ench, S. J. Gao, K. Rice, D. Carey, R. C. Kennedy, J. R. Arrand, and M. Mackett. 2000. Epidemiology of herpesvirus papio infection in a large captive baboon colony: similarities to Epstein-Barr virus infection in humans. J. Infect. Dis. 181:1462-1466.
Jenson, H. B., Y. Ench, Y. Zhang, S. J. Gao, J. R. Arrand, and M. Mackett. 2002. Characterization of an Epstein-Barr virus-related gammaherpesvirus from common marmoset (Callithrix jacchus). J. Gen. Virol. 83:1621-1633.
Landon, J. C., and L. B. Malan. 1971. Seroepidemiologic studies of Epstein-Barr virus antibody in monkeys. J. Natl. Cancer Inst. 46:881-884.
Milman, G., A. L. Scott, M. S. Cho, S. C. Hartman, D. K. Ades, G. S. Hayward, P. F. Ki, J. T. August, and S. D. Hayward. 1985. Carboxyl-terminal domain of the Epstein-Barr virus nuclear antigen is highly immunogenic in man. Proc. Natl. Acad. Sci. USA 82:6300-6304.
Rao, P., H. Jiang, and F. Wang. 2000. Cloning of the rhesus lymphocryptovirus viral capsid antigen and Epstein-Barr virus-encoded small RNA homologues and use in diagnosis of acute and persistent infections. J. Clin. Microbiol. 38:3219-3225.
Rivailler, P., Y. G. Cho, and F. Wang. 2002. Complete genomic sequence of an Epstein-Barr virus-related herpesvirus naturally infecting a new world primate: a defining point in the evolution of oncogenic lymphocryptoviruses. J. Virol. 76:12055-12068.
Rivailler, P., H. Jiang, Y. G. Cho, C. Quink, and F. Wang. 2002. Complete nucleotide sequence of the rhesus lymphocryptovirus: genetic validation for an Epstein-Barr virus animal model. J. Virol. 76:421-426.
Shedd, D., A. Angeloni, J. Niederman, and G. Miller. 1995. Detection of human serum antibodies to the BFRF3 Epstein-Barr virus capsid component by means of a DNA-binding assay. J. Infect. Dis. 172:1367-1370.
van Grunsven, W. M., A. Nabbe, and J. M. Middeldorp. 1993. Identification and molecular characterization of two diagnostically relevant marker proteins of the Epstein-Barr virus capsid antigen complex. J. Med. Virol. 40:161-169.
van Grunsven, W. M., W. J. Spaan, and J. M. Middeldorp. 1994. Localization and diagnostic application of immunodominant domains of the BFRF3-encoded Epstein-Barr virus capsid protein. J. Infect. Dis. 170:13-19.
van Grunsven, W. M., E. C. van Heerde, H. J. de Haard, W. J. Spaan, and J. M. Middeldorp. 1993. Gene mapping and expression of two immunodominant Epstein-Barr virus capsid proteins. J. Virol. 67:3908-3916.
Wang, F., P. Rivailler, P. Rao, and Y. Cho. 2001. Simian homologues of Epstein-Barr virus. Philos. Trans. R. Soc. Lond. B Biol. Sci. 356:489-497.(Mark H. Fogg, Angela Carv)
Department of Primate Resources, New England Primate Research Center, Harvard Medical School, Southborough, Massachusetts 01772
ABSTRACT
The recent discovery of an Epstein-Barr virus (EBV)-related lymphocryptovirus (LCV) naturally infecting common marmosets demonstrated that gamma-1 herpesviruses are not limited to human and Old World nonhuman primate hosts. We developed serologic assays to detect serum antibodies against lytic- and latent-infection marmoset LCV antigens in order to perform the first seroepidemiologic study of LCV infection in New World primates. In three different domestic colonies and in animals recently captured from the wild, we found that the seroprevalence of marmoset LCV infection was not as ubiquitous as with EBV or Old World LCV. These biologic differences in LCV infection of New World versus human and Old World primate hosts correlate with the evolution of the LCV viral gene repertoire.
TEXT
Serologic responses to lytic and latent infection antigens, such as viral capsid antigen (VCA) and EBNA-1, are widely used to document Epstein-Barr virus (EBV) infection (8, 13). Old World (4), and more recently New World (2), nonhuman primates are known to be naturally infected with related herpesviruses in the same lymphocryptovirus (LCV) genus as EBV. LCV infection in Old World primates was initially recognized by the presence of serum antibodies cross-reactive with viral antigens in EBV-infected B cells (7). As with humans, LCV seropositivity in Old World primates is highly prevalent both in nature and in domesticated colonies, with seropositivity in more than 95% of adult animals (5, 7, 9). The biology of LCV infection in Old World primates appears to be nearly identical to that of EBV infection in humans (16). This is concordant with the identical repertoire of viral genes and the high degree of sequence homology between EBV and rhesus LCV, a prototype for an Old World LCV whose genome has recently been fully sequenced (11).
It was long believed that LCV did not infect New World primates, since there was no strong evidence of EBV cross-reactive antibodies from these species. However, we recently isolated a B-cell-immortalizing herpesvirus from a spontaneous B-cell lymphoma arising in a common marmoset (Callithrix jacchus) that belonged to the LCV genus (2). PCR studies showed that LCV infection was present in three different marmoset colonies, and a similar LCV was also found in squirrel monkeys, indicating that LCV infection is present in species from both major families of New World primates (2). Studies from other laboratories have since confirmed that LCV can be found in a variety of New World species (3, 6).
The complete genome sequence of the marmoset LCV is striking for its differences from the EBV genome (10). The marmoset LCV encodes at least seven unique open reading frames (ORFs) with no sequence homology to other cell or viral genes. In most cases, there is compelling evidence that these unique genes are predecessors for certain EBV genes, e.g., they are located in the same position of the genome, they have similar transcriptional patterns, they have similar predicted secondary protein structures, or they have similar functional properties demonstrable in vitro. The marmoset LCV also lacks apparent homologs for 14 genes found in EBV and rhesus LCV. These genes are unique to herpesviruses in the LCV genus, i.e., they are not found in other alpha-, beta-, or gamma-2 herpesviruses, and appear to have been acquired in the evolution of LCV from New World primates to humans. Most of these acquired genes are lytic infection genes, and all are nonessential for EBV replication and B-cell immortalization in vitro, suggesting that they play an important role for LCV infection in vivo, rather than in vitro. Thus, the marmoset LCV appears to be a more primitive predecessor of EBV lacking a number of viral genes that have evolved in higher-order primates. Little is known about the biology of LCV infection in New World primates and whether this difference in viral gene repertoire results in biologic differences between EBV and marmoset LCV infection. Therefore, we asked whether seroprevalence of LCV infection in New World primates was as high as that seen in Old World primates and humans.
We obtained sera from common marmosets at the New England Primate Research Center (NEPRC) and tested them for reactivity to the marmoset LCV ORF59 encoding the small viral capsid antigen (sVCA), p18, or BFRF3 homologue. The carboxy terminus of the sVCA is known to be an immunodominant target in humans and rhesus macaques, and peptides derived from the carboxy terminus of either EBV or rhesus LCV BFRF3 can be used in enzyme immunoassays to detect serologic responses in virtually all LCV-infected humans or rhesus macaques (9, 12-15). Therefore, we synthesized a 24-residue peptide from the carboxy terminus of the marmoset ORF59 (Fig. 1) and tested sera from 235 common marmosets in the sVCA enzyme immunoassay (EIA). As shown in Fig. 2A, (63%) serum samples tested positive, with values at least three times above background values, and 87 (37%) serum samples tested negative. This is strikingly less than the 100% of sera from NEPRC rhesus macaques in conventional housing that test positive in a similar rhesus LCV sVCA EIA (9).
In order to test whether the seroprevalence for marmoset LCV infection determined by the sVCA EIA was accurate, we looked for serum antibodies against the marmoset LCV EBNA-1 homologue, ORF39. A Flag-tagged marmoset LCV ORF39 was expressed from a recombinant vaccinia virus and used to infect BSC-40 cells. ORF39 was affinity purified from cell lysates with a Flag monoclonal antibody and used as the antigen for an EIA. The same NEPRC serum samples were tested for ORF39 reactivity, with 152 (65%) samples testing positive and 83 (35%) of the samples testing negative (Fig. 2B). Thus, by serologic testing against a latent infection antigen, there were still a significant number of animals with no serologic evidence of marmoset LCV infection.
When the sVCA and ORF39 reactivities were compared, 128 (54%) of the animals were seropositive against both lytic and latent antigens, and 64 (28%) of the animals were negative for reactivity to both antigens (Table 1). The remaining 42 animals (18%) were positive in only one assay, with approximately an equal number positive in either the ORF39 or sVCA assay. The observation that more than one-fourth of the animals showed no serological reactivity to immunodominant lytic and latent antigens suggests that LCV infection is not as prevalent in marmosets as in humans or other Old World primates. In order to test whether other viral antigens may be more immunodominant than ORF39 and sVCA in marmosets, a subset of EIA positive and negative sera were tested by immunoblotting against cell lysates of marmoset LCV-infected cells induced by treatment with phorbol ester and butyrate or not induced for lytic replication. The EIA negative sera failed to detect any significant proteins in the cell lysates, consistent with the interpretation that these animals were naive for marmoset LCV infection (data not shown). The high percentage of seronegative animals did not appear to be due to an unusually high number of younger animals in this group, since the average age of the animals in the double-negative group (4.7 years old) was actually higher than that of the double-positive group (3.9 years old). The average age of animals with discrepant ORF39 and sVCA results was 2.2 years old, suggesting the possibility that these animals may be closer to the time of acquiring LCV infection.
In order to validate the serologic data, we used DNA PCR of peripheral blood mononuclear cells to look for evidence of persistent marmoset LCV infection. In our previous studies, approximately 60% from a small group of domestic animals were PCR positive for marmoset LCV DNA in peripheral blood mononuclear cells (2). However, since these studies were not linked to serologic tests, it was not known whether PCR-negative animals were not infected versus infected with undetectable levels of DNA. In order to increase the sensitivity of the PCR amplification, we designed new primers from the major internal repeat (5'-GGACCCGAGGAACTGTCTAAAAC-3' and 5'-TGTGTAGGCACCAAGTCTGCC-3') instead of using the previously described primers from a single-copy gene (2). We obtained additional blood samples from 24 animals for serologic and DNA PCR testing in parallel. Eighteen animals were seropositive (10 positive for both sVCA and ORF39, 3 positive for sVCA only, and 5 positive for ORF39 only), and 6 animals were seronegative by both serologic tests. Peripheral blood mononuclear cells from all 18 seropositive animals were also positive by marmoset LCV DNA PCR, and DNA PCR was negative for 5 of the 6 seronegative animals. These results show that DNA PCR or the combination of sVCA and ORF39 serologic tests are sensitive assays for LCV infection that detect almost identical populations. In addition, a significant number of animals are negative for both DNA PCR and sVCA/ORF39 serologic responses, suggesting that they have not been infected. Whether a minor subpopulation of animals can be LCV infected with little or no serologic response requires further studies.
To determine whether the lower seroprevalence of marmoset LCV infection was common to other marmoset colonies, sera from two geographically separated marmoset colonies were obtained. In these two colonies, 37% and 47% of the animals tested positive by the sVCA EIA (n = 165 and 126, respectively). The results support the findings that LCV infection may not be as ubiquitous among marmosets as it is among humans and Old World primates.
Common marmosets are typically housed in smaller units than Old World primates, so a lower prevalence of marmoset LCV infection could be due to segregation of seropositive and seronegative animals in domesticated colonies. Therefore, we examined the housing patterns of animals in relation to seropositivity. Out of 91 animals in 37 cages at the NEPRC, 5 cages contained all sVCA-seropositive animals, 7 cages contained all seronegative animals, and 25 cages contained both seropositive and seronegative animals. Thirty of forty-three seropositive animals were housed in cages with both seropositive and seronegative animals. Similarly, 29 out of 48 seronegative animals were housed in cages with both seropositive and seronegative animals. Thus, a significant portion of seronegative animals (19 of 48; 40%) were segregated with other naive animals, suggesting that housing practices may contribute to a lower seroprevalence of marmoset LCV infection. However, the large percentage of mixed cages and large number of seronegative animals in mixed cages (60%) also suggest that LCV infection may not be readily transmitted among marmosets. In contrast, virtually all newborn Old World primates, such as rhesus macaques and baboons, turn seropositive within 1 year when housed with other seropositive animals (5, 9). In order to eliminate potential bias from domestic housing, sera collected from common marmosets shortly after capture from the wild were also tested. Twelve out of 24 animals (50%) tested positive by the sVCA EIA, indicating reduced seroprevalence among New World primates in the wild, similar to animals in domestic colonies.
These are the first serologic studies of LCV infection in New World primates. Historically, the failure to reliably detect EBV cross-reactive antibodies in New World primates was probably due to the degree of sequence divergence between EBV and marmoset LCV genes, exacerbated by additional divergence between human and marmoset immunoglobulins. Thus, important technical aspects in these studies were the use of antigens derived from marmoset LCV sequences and anti-human immunoglobulin secondary reagents that had not been absorbed for reactivity to immunoglobulins from other mammalian species. The combined use of lytic and latent antigens that are immunodominant in EBV and rhesus LCV infection identified largely identical positive and negative populations among NEPRC animals. ORF39- and ORF59-negative sera did not react with any other specific bands on immunoblots with LCV-infected cell lysates induced for viral replication, consistent with LCV-naive hosts. Reduced seroprevalence of marmoset LCV infection was consistently found in two other domestic colonies and in animals recently captured from the wild.
These results suggest that LCV infection may not be as prevalent in marmosets as in humans and in Old World primates, such as rhesus macaques. Results of the seroprevalence studies with these larger populations are consistent with our previous data obtained using nested PCR amplification of peripheral blood lymphocytes from a smaller number of animals at the Wisconsin and New England Primate Research Centers, 60% and 44% positivity, respectively (2). Analysis of the current data suggests that age and housing may have had some impact on the prevalence of seronegative animals, but these factors do not completely account for the significant differences between the ubiquitous LCV infection of adult humans and rhesus macaques versus a significant seronegative subpopulation in different marmoset colonies and in the wild.
Additional studies will be required to confirm whether the LCV seroprevalence is also reduced in other New World primate species. A lower seroprevalence of New World LCV infection versus EBV and Old World LCV infection could be due to less efficient transmission, i.e., a decreased rate of new infection, or LCV infection that is less persistent, i.e., transient infection or loss of infection. Limited studies of serial serum samples from NEPRC marmosets did not reveal any evidence for loss of seropositivity that may result from infection that fails to persist (data not shown). Thus, we believe it is more likely that marmoset LCV infection may be transmitted less efficiently than EBV or rhesus LCV infection, e.g., less viral shedding, less-efficient infection, or better host resistance. The evolution of genes in EBV and rhesus LCV that are not present in marmoset LCV may provide a genetic basis for this biologic phenotype. Genes encoding a viral interleukin 10 and CSF-1R homologue, BCRF1 and BARF1, are also absent in the marmoset LCV. These are soluble cytokines produced during lytic LCV infection that could inhibit local immune responses to immunogenic viral antigens during viral replication. One could speculate that the lack of these immunomodulators could affect transmission by decreasing the amount of viral replication and shedding from the oral cavity or reducing the ability for new viral infection to evade oral mucosal immune responses. The marmoset LCV does not have apparent homologues for EBERs and BARF0. These latent infection genes are not essential for B-cell immortalization in vitro but may be important for successful or efficient B-cell infection in vivo. The marmoset LCV also lacks a BDLF3 homologue. This encodes a 150-kDa glycoprotein, and paradoxically, deletion of this glycoprotein enhances epithelial cell infection in vitro (1). Four of the EBV genes missing in marmoset LCV have unknown function (BLLF2, LF3, ECRF4, and BNLF2), so it remains to be determined whether any of these genes may contribute to efficiency of virus transmission. Development of a genetic system and an experimental infection model for marmoset LCV may make it possible to introduce these genes into marmoset LCV in order to directly test whether these viral genes might improve the efficiency of LCV infection in marmosets. Alternatively, deletion of these viral genes from rhesus LCV may result in a reduced efficiency of LCV infection in Old World primates.
ACKNOWLEDGMENTS
This work was funded by a grant from the U.S. Public Health Service (CA89172). Services from the New England Primate Research Center were supported by a base grant to the institution (USPHS P51RR00168).
All animal experiments were performed with approval from the Committee on Animals for Harvard Medical School, and animals were maintained in compliance with federal and institutional guidelines for animal care. We thank Jose Augusto Muniz and the Centro Nacional de Primatas for providing sera from marmosets shortly after capture from the wild.
REFERENCES
Borza, C. M., and L. M. Hutt-Fletcher. 1998. Epstein-Barr virus recombinant lacking expression of glycoprotein gp150 infects B cells normally but is enhanced for infection of epithelial cells. J. Virol. 72:7577-7582.
Cho, Y., J. Ramer, P. Rivailler, C. Quink, R. L. Garber, D. R. Beier, and F. Wang. 2001. An Epstein-Barr-related herpesvirus from marmoset lymphomas. Proc. Natl. Acad. Sci. USA 98:1224-1229.
Ehlers, B., A. Ochs, F. Leendertz, M. Goltz, C. Boesch, and K. Matz-Rensing. 2003. Novel simian homologues of Epstein-Barr virus. J. Virol. 77:10695-10699.
Frank, A., W. A. Andiman, and G. Miller. 1976. Epstein-Barr virus and nonhuman primates: natural and experimental infection. Adv. Cancer Res. 23:171-201.
Jenson, H. B., Y. Ench, S. J. Gao, K. Rice, D. Carey, R. C. Kennedy, J. R. Arrand, and M. Mackett. 2000. Epidemiology of herpesvirus papio infection in a large captive baboon colony: similarities to Epstein-Barr virus infection in humans. J. Infect. Dis. 181:1462-1466.
Jenson, H. B., Y. Ench, Y. Zhang, S. J. Gao, J. R. Arrand, and M. Mackett. 2002. Characterization of an Epstein-Barr virus-related gammaherpesvirus from common marmoset (Callithrix jacchus). J. Gen. Virol. 83:1621-1633.
Landon, J. C., and L. B. Malan. 1971. Seroepidemiologic studies of Epstein-Barr virus antibody in monkeys. J. Natl. Cancer Inst. 46:881-884.
Milman, G., A. L. Scott, M. S. Cho, S. C. Hartman, D. K. Ades, G. S. Hayward, P. F. Ki, J. T. August, and S. D. Hayward. 1985. Carboxyl-terminal domain of the Epstein-Barr virus nuclear antigen is highly immunogenic in man. Proc. Natl. Acad. Sci. USA 82:6300-6304.
Rao, P., H. Jiang, and F. Wang. 2000. Cloning of the rhesus lymphocryptovirus viral capsid antigen and Epstein-Barr virus-encoded small RNA homologues and use in diagnosis of acute and persistent infections. J. Clin. Microbiol. 38:3219-3225.
Rivailler, P., Y. G. Cho, and F. Wang. 2002. Complete genomic sequence of an Epstein-Barr virus-related herpesvirus naturally infecting a new world primate: a defining point in the evolution of oncogenic lymphocryptoviruses. J. Virol. 76:12055-12068.
Rivailler, P., H. Jiang, Y. G. Cho, C. Quink, and F. Wang. 2002. Complete nucleotide sequence of the rhesus lymphocryptovirus: genetic validation for an Epstein-Barr virus animal model. J. Virol. 76:421-426.
Shedd, D., A. Angeloni, J. Niederman, and G. Miller. 1995. Detection of human serum antibodies to the BFRF3 Epstein-Barr virus capsid component by means of a DNA-binding assay. J. Infect. Dis. 172:1367-1370.
van Grunsven, W. M., A. Nabbe, and J. M. Middeldorp. 1993. Identification and molecular characterization of two diagnostically relevant marker proteins of the Epstein-Barr virus capsid antigen complex. J. Med. Virol. 40:161-169.
van Grunsven, W. M., W. J. Spaan, and J. M. Middeldorp. 1994. Localization and diagnostic application of immunodominant domains of the BFRF3-encoded Epstein-Barr virus capsid protein. J. Infect. Dis. 170:13-19.
van Grunsven, W. M., E. C. van Heerde, H. J. de Haard, W. J. Spaan, and J. M. Middeldorp. 1993. Gene mapping and expression of two immunodominant Epstein-Barr virus capsid proteins. J. Virol. 67:3908-3916.
Wang, F., P. Rivailler, P. Rao, and Y. Cho. 2001. Simian homologues of Epstein-Barr virus. Philos. Trans. R. Soc. Lond. B Biol. Sci. 356:489-497.(Mark H. Fogg, Angela Carv)