Epstein-Barr Virus Can Establish Infection in the
http://www.100md.com
病菌学杂志 2005年第17期
Basic and Clinical Virology, University of Edinburgh, Royal (Dick) School of Veterinary Studies, Summerhall, Edinburgh EH9 1QH, United Kingdom
Immunobiology Unit, Institute of Child Health, Guilford Street, London WC1N 1EH, United Kingdom
Department of Immunology, John Radcliffe Hospital, Oxford OX3 9DU, United Kingdom
Department of Immunology, Royal Free Hospital, Rowland Hill Street, London NW3 2PF, United Kingdom
Clinical Biochemistry and Immunology, Addenbrooke's Hospital, Hills Road, Cambridge CB2 2QQ, United Kingdom
Regional Immunology Department, Royal Victoria Infirmary, Queen Victoria Road, Newcastle upon Tyne NE1 4LP, United Kingdom
ABSTRACT
Epstein-Barr virus (EBV) is a ubiquitous human herpesvirus that persists in the body for life after primary infection. The primary site of EBV persistence is the memory B lymphocyte, but whether the virus initially infects nave or memory B cells is still disputed. We have analyzed EBV infection in nine cases of X-linked hyper-immunoglobulin M (hyper-IgM) syndrome who, due to a mutation in CD40 ligand gene, do not have a classical, class-switched memory B-cell population (IgD– CD27+). We found evidence of EBV infection in 67% of cases, which is similar to the infection rate found in the general United Kingdom population (60 to 70% for the relevant age range). We detected EBV DNA in peripheral blood B cells and showed in one case that the infection was restricted to the small population of nonclassical, germinal center-independent memory B cells (IgD+ CD27+). Detection of EBV small RNAs, latent membrane protein 2, and EBV nuclear antigen 3C expression in peripheral blood suggests full latent viral gene expression in this population. Analysis of EBV DNA in serial samples showed variability over time, suggesting cycles of infection and loss. Our results demonstrate that short-term EBV persistence can occur in the absence of a germinal center reaction and a classical memory B-cell population.
INTRODUCTION
Epstein-Barr virus (EBV) is a ubiquitous, tumorigenic herpesvirus that infects over 90% of the world's adult population. The virus generally infects during childhood without clinical illness, and thereafter persists in the body for life, only rarely causing disease in the immunocompetent host. This highly successful colonization of the human race is effected by close integration of the virus life cycle with host biological processes (reviewed in reference 6).
EBV enters the body via the mouth and infects B lymphocytes in oropharyngeal lymphoid tissue (tonsil, adenoids), inducing their proliferation. From this nidus of infection the virus is disseminated throughout the body in circulating infected B cells. This infection induces an EBV-specific, cytotoxic T-cell response which is essential for control of the infection and may be so florid as to cause the immunopathological symptoms of infectious mononucleosis (reviewed in reference 6). At resolution of primary infection a persistent virus carrier state is established in which around 1 to 60 x 106 circulating B lymphocytes carry viral DNA (17).
The EBV genome contains a set of unique latent genes, EBV nuclear antigens (EBNA) 1, 2, 3A, B and C, leader protein (LP), latent membrane proteins (LMP) 1, 2A and B, as well as two small EBV-encoded RNAs (EBERs) 1 and 2, which act in concert to induce proliferation and immortalization of B cells infected in vitro. EBNA2 is the main transactivator of viral and cellular genes, whereas LMP1 and 2A mimic CD40 and B-cell receptor signaling, respectively (reviewed in reference 5). B cells expressing these latent antigens can be found in tonsillar tissue, but a more restricted latent viral gene expression is seen in peripheral blood B cells (3). Although EBERs and LMP 2A mRNA are regularly detected (10, 17, 18), it now appears that in most cells the virus is transcriptionally silent (9).
Careful analyses identify the memory B-cell population (characterized by markers: CD27+, IgM–, IgD–) as the specific site of viral persistence in peripheral blood (4); however, there is still debate over whether these cells are infected directly by EBV in lymphoid tissue, or whether they mature from infected nave B cells. In one study EBV-infected nave B cells expressing EBNA2 are reported to be the predominant infected population in tonsil (11). This scenario contrasts with the finding of EBV in a predominantly memory B-cell population in infectious mononucleosis tonsils, suggesting initial infection of cells which have already traversed the germinal center (16). To attempt to resolve these issues we have examined EBV infection in patients without a classical memory B-cell population.
Hyper-immunoglobulin M (IgM) syndrome is a rare heterogeneous immunodeficiency characterized clinically by high levels of IgM and low levels of the other immunoglobulin classes in serum, and recurrent infections. Four types of X-linked hyper-IgM are now recognized, all of which cause defective immunoglobulin class switch recombination. In type 1 CD40 ligand (L) is mutated (1, 2, 7). Consequently, signaling through CD40/CD40L is defective and B cells are unable to undergo a germinal center reaction with immunoglobulin hypermutation, class switching and maturation to memory cells. Thus, the majority of circulating B cells in hyper-IgM have a nave B-cell phenotype (IgM+, IgD+, CD27–), although recently a small population of IgM+ IgD+ CD27+ B cells have been identified in both healthy individuals and hyper-IgM cases which have mutated immunoglobulin genes and are proposed to arise by a germinal center-independent mechanism (23).
In order to investigate the relationship between EBV infection and B-cell maturation we have sought evidence of EBV infection and persistence in nine cases of type 1 X-linked hyper-IgM. Our hypothesis was that, if a post-germinal center memory B-cell population is essential for EBV persistence, then hyper-IgM cases will not carry EBV long term. However, if nave B cells are the initial target of infection, then recurrent infections may occur without the establishment of persistence.
MATERIALS AND METHODS
Hyper-IgM cases and healthy controls. Nine cases of hyper-IgM were investigated. All were male, age range 2 to 44 years (Table 1). All cases were shown to have mutations/deletions in the CD40L gene and/or absence of CD40L expression on activated lymphocytes (data not shown); 2 to 60 ml of peripheral blood and throat wash or swab samples (depending on age) were obtained from each patient. Control samples were obtained from anonymous blood donors (n = 27). Control throat washes were obtained from 16 healthy laboratory workers. This study was approved by the Northern and Yorkshire Multi-Centre Ethics Committee.
Peripheral blood mononuclear cells. Peripheral blood mononuclear cells (PBMC) were separated from whole blood by routine density gradient centrifugation, washed, and counted.
Throat wash and swab samples. Throat wash samples were obtained by gargling with 10 ml normal saline. Where hyper-IgM cases were under 15 years of age, throat swabs were obtained and shaken in 2 ml normal saline. Thereafter both samples were centrifuged at 600 x g for 5 min, and the cell pellets were washed twice in saline. Both cell pellets and supernatant saline were stored at –70°C
Flow cytometry. PBMCs were analyzed on a FACScalibur (Becton Dickinson) after staining with the following antibodies: phycoerythrin-labeled CD154, Cy-chrome-labeled CD3, allophycocyanin-labeled CD19 (BD Biosciences), phycoerythrin-labeled CD27 (Dako), fluorescein isothiocyanate-labeled IgD (Southern Biotechnology Associates) according to the manufacturers' instructions. Isotype antibodies used as negative controls were: goat IgG fluorescein isothiocyanate (Southern Biotechnology Associates), mouse IgG1 phycoerythrin and fluorescein isothiocyanate (BD Biosciences). On occasions stained cell populations were sorted on a fluorescence activated cell sorter (Becton Dickinson FACS Vantage SE) and used for further analysis.
Enrichment of lymphocyte subpopulations. Dynabeads (Dynal Biotech ASA, Oslo, Norway) were added to PBMCs at a bead to cell ratio of 5:1 and the CD3-enriched and CD3-depleted populations were separated by application of a magnet in accordance with the manufacturer's instructions. Both the CD3-enriched and -depleted fractions were then cultured at a concentration of 106/ml in complete RPMI for 24 h, prior to DNA extraction.
EBV detection in PBMCs and throat wash and swab samples. PCR detection of the BamHI W repeat region of the EBV genome was performed using 1 to 2 μg of DNA extracted from hyper-IgM and control PBMCs and throat wash cell pellet samples using the Easy-DNA kit (Invitrogen) following the manufacturer's instructions. On occasions where the number of cells in a specific population was below 104, 106 EBV-negative filler cells (TK-143B human osteosarcoma cells) were added prior to DNA extraction.
The primers amplify a 298-bp fragment of the W repeats; forward primer CTT TAG AGG CGA ATG GGC GC and reverse primer AGG ACC ACT TTA TAC CAG GG. PCR amplification was performed in a final volume of 100 μl with the reaction mixture containing 1x PCR buffer (50 mM KCl, 10 mM Tris-HCl [pH 9.0] and 0.1% Triton X-100), plus 1.5 mM MgCl2, 200 μM each of dATP, dTTP, dGTP, and dCTP, 0.1 μM of each primer, and 1 unit of Taq polymerase (Promega). A dilution series of an EBV-positive cell line (Namalwa, one to two EBV genomes per cell) in an EBV-negative cell line (BJAB) was used to give standard 10-fold dilutions from 1 to 106 EBV genomes per 106 cells.
PCR products were separated by electrophoresis on an agarose gel, transferred to positively charged nylon membranes, probed with an EBV-specific probe labeled with digoxigenin, and visualized using a digoxigenin luminescent kit (Roche). The density of bands detected in the Namalwa dilution series was compared with the test samples by scanning with LabWorks 4.0 (UVP BioImaging systems). This provided a semiquantitative estimate of viral load (10).
In some cases EBV DNA load was determined by quantitative competitive PCR exactly as described by Stevens et al. (20). The amplification reaction contained 50 mM KCl, 1.5 mM MgCl, 10 mM Tris, pH 8.5, 25 pmol of each primer, one of which was biotin labeled, and 1 U of Taq. Cycling conditions were 4 min at 95°C; 40 cycles at 95°C, 55°C, and 72°C for 1 min each; and finally 3 min at 72°C. Products were captured on a streptavidin-coated plate and probed with digoxigenin-labeled wild-type and internal standard probes. Optical density was measured and used to calculate copy number.
Viral isolate typing. The 39-bp repeat region of EBNA 3C was amplified in a nested PCR using outer primers 5'ACA CTT GAG TTC CAT GTC GC 3' and 5' TGTAATCACTGGCAAAGGGC 3', and inner primers 5' TAT CGC ACG AAG AAC AAC CCC 3' and 5' AGA TGT GGG AAC TGG GAG ACC 3' following the method described in Haque et al. (8). The 33-bp repeat region of LMP1 was amplified using primers 5' TTT CCA GCA AGA GTC GCT AGG 3' and 5' GGC GCA CCT GGA GGT GGT CC 3' (13); 10 μl of the product was run on a 2% agarose gel, Southern hybridized with a digoxigenin-labeled specific probe 5' CAC GGG CTC CAA TCA TCT TC 3' (EBNA3C) or 5' AGG ACC CTG ACA ACA CTG AT 3' (LMP1) and visualized using a digoxigenin luminescent kit (Roche).
PCR products were excised and purified using the QIAquick gel extraction kit (QIAGEN). Direct sequencing of the products was performed commercially by MWG Research using the forward primer.
Expression of latent viral genes. RNA was isolated from PBMCs using the RNeasy Mini Protocol (QIAGEN). RNA was DNase treated (Promega) before synthesizing cDNA from 1 μg RNA using the Thermoscript reverse transcription-PCR system (Invitrogen); 50 μg/μl random hexamer was used as primer and 2 μl of the reaction mixture was either used immediately for PCR amplification or stored at –20°C for later use. All cDNA was tested for glyceraldehyde-3-phosphate dehydrogenase and -actin expression to check the quality of the DNA, and only used for analysis of EBV gene expression when these tests were positive.
Reverse transcription-PCRs were optimized for detection of human EBV latent transcripts EBNA1, EBNA3C, and LMP2A and the late lytic transcript encoding the envelope glycoprotein (gp) 350, using the EBV-positive cell line B95-8. PCR amplification was performed in a final volume of 50 μl with the reaction mixture containing the following: 1x PCR buffer (50 mM KCl, 10 mM Tris-HCl [pH 9.0] and 0.1% Triton X-100), plus 1.5 mM MgCl2 (3 mM MgCl2 for gp350), 200 μM each of dATP, dTTP, dGTP, and dCTP, 0.1 μM of each primer and 1 unit of Taq polymerase (Promega). Amplification consisted of an initial denaturation for 2 min at 94°C in all reactions, followed by 35 cycles (except EBERs, 30 cycles) of denaturation at 94°C for 1 min, annealing at various temperatures (EBERs at 50°C, EBNA3C at 54°C, gp350, LMP1, and LMP2A at 55°C, EBNA3C nested, LMP1 nested, and LMP2A nested at 57°C) for 2 min (EBERs 1 min) and extension at 72°C for 2 min (EBERs, 1 min), with a final extension at 72°C for 10 min.
Controls included B95-8 (positive), water and untranscribed test RNA (to monitor for DNA contamination); 18 μl of PCR products was then separated by electrophoresis on an agarose gel, transferred to positively charged nylon membranes, and visualized using a digoxigenin luminescent kit (Roche). Primer and probe sequences used are as described in Hopwood et al. (10). With the exception of EBERs and gp350, each primer pair was designed to amplify across introns, giving an amplified cDNA which differs in size from possible amplified contaminating DNA (sizes of possible contaminants: LMP1, 535 bp; LMP2A, 278 bp; and EBNA3C, 386 bp).
RESULTS
Hyper-IgM cases are infected with EBV at the expected rate. Since all hyper-IgM cases were treated with regular infusions of pooled human immunoglobulin, routine serological testing could not be used to indicate past or present EBV infection. We therefore used PCR analysis to detect viral DNA as a marker of infection. We analyzed PBMCs from eight hyper-IgM cases and 20 healthy controls (18 known EBV seropositive and 2 EBV seronegative). Throat wash and swab samples were also available from eight cases.
EBV DNA from the BamHI-W repeat region of EBNA LP was amplified by PCR. This technique regularly detects EBV DNA from 1 to 10 Namalwa cells (1 to 2 EBV genome copies/cell) in a background of 106 EBV-negative cells (Fig. 1). Four of eight cases (50%) and 10 of 18 healthy controls (56%) gave a positive result in PBMCs. Five of eight throat wash and swab cell pellet samples (62%) from hyper-IgM cases and 10 of 16 (63%) from healthy controls were PCR positive. Thus, cases and controls gave similar results. Overall six out of nine hyper-IgM cases (66.7%) showed evidence of EBV infection in either PBMCs or throat wash and swab samples, with three cases being positive in both (Table 1 and Fig. 1a and b). Densitometric analysis of these results showed the viral load in hyper-IgM peripheral blood to be within the same range as that of healthy EBV carriers (<1 to 103 genomes/μg DNA) (Table 1). Neither hyper-IgM serum/plasma nor concentrated throat wash supernatant samples contained any detectable EBV DNA (data not shown).
Hyper-IgM cases show variable detection of EBV DNA over time. Since our findings indicated that hyper-IgM cases are infected with EBV, we analyzed serial blood and throat wash samples from four individuals to determine whether this infection persisted over time. We received samples from two time points from cases 1, 2, and 3 and at four time points from case 8. Each case showed a pattern which varied over time, with one set of samples being negative at at least one time point (Table 2). Only case 8 showed positive blood and throat wash samples on more than one occasion; the first two sets of samples taken 6 months apart were positive, but the samples taken 25 months later were negative. The final blood sample taken 9 months later was again positive (Table 2). These results contrast with our findings in PBMC from 5 healthy EBV seropositive individuals. Over a similar time period controls showed a relatively stable EBV load with levels varying from detectable to undetectable in only one individual (Table 3). Similarly EBV DNA in throat wash from healthy donors varied over time from detectable to nondetectable in one of four donors (Table 3).
Further analysis to detect EBV DNA was only possible on cases 3 and 8, where repeat blood samples with detectable EBV DNA were available.
EBV DNA is only present in the CD27+, IgD+ B-cell fraction in hyper-IgM. PBMCs from case 8 were used to identify the cell type carrying EBV DNA. PBMCs were separated into CD3-enriched and -depleted fractions using magnetic beads. Fractions were cultured for 24 h prior to DNA preparation and PCR. EBV DNA was detected in the CD3-depleted but not CD3-enriched fraction indicating that the virus is present in peripheral B cells (Fig. 2).
Circulating B cells from hyper-IgM cases mainly consist of short lived IgD+ CD27– nave cells, however, a small population of IgD+ CD27+ B cells has been reported which, although not classical memory B cells, do have mutated immunoglobulin genes (22). Flow cytometric analysis on PBMCs from hyper-IgM PBMCs stained for CD19, CD27, and IgD showed that the majority of CD19+ cells were IgD+ (86.7 to 96.4%). A small population (3.1 to 5.4%) of CD27+, IgD+ cells could be detected. The mean percentage of CD27+ IgD+ cells in six healthy controls was 35% (Fig. 3a).
In order to identify the B-cell subset carrying EBV in hyper-IgM, PBMCs from case 3 were further separated into IgD+ CD27+ and IgD– CD27+ subsets using fluorescence-activated cell sorting. PCR analysis detected EBV DNA exclusively in the IgD+ CD27+ subset. The PCR was repeated on a second occasion and gave the same result. Quantitative PCR indicated a ratio of 1 EBV genome per 30 B cells in this population (Fig. 3b), however, since the cell populations were incubated for 24 h prior to DNA extraction, it is possible that the lytic cycle had been initiated, accounting for this high viral load.
PBMCs from four healthy controls in which EBV DNA was detected by PCR were separated by fluorescence-activated cell sorting on the basis of IgD and CD27 and processed exactly as above. EBV DNA was not detected in the IgD+ CD27+ population in any of these samples.
Hyper-IgM PBMCs express latent viral genes. Having detected EBV in hyper-IgM B cells, we wished to determine which latent viral genes were expressed in these infected cells. RNA from three hyper-IgM PBMCs samples (nos. 3 and 8) was subjected to reverse transcription-PCR to amplify EBER1, LMP1, LMP2A, EBNA3C, and gp350 sequences. EBERs were detected in both cases, with additional detection of LMP2A in one case (no. 3) and EBNA 3C in another (no. 8) (Fig. 4). LMP1 and gp350 expression was not detected in any sample tested (data not shown). Similar RNA analysis on PBMCs from 18 healthy EBV-seropositive controls detected EBERs in 78% and LMP2A in 67%, but no LMP1, EBNA3C, or gp350 expression.
Hyper-IgM case 8 has EBV isolates with identical numbers of EBNA 3C repeats over time. Only case 8 showed evidence of EBV infection sustained over time (Table 2), and we therefore used the EBV-positive blood and throat wash samples taken 6 months apart to distinguish between persistent infection and reinfection in this case. PCR amplification of a repeat sequence in the EBNA 3C gene known to vary in size between viral isolates (8) gave an amplified product from the two throat wash samples, but not from the PBMCs. The same size bands were amplified from both samples which were estimated by size comparison with control samples with known numbers of repeats to contain 6 repeat sequences (Fig. 5). Excision of the bands and sequencing of the products confirmed that the two products were identical. However, results from a series of control samples showed that 26% of individual isolates have six repeats in this region. We therefore attempted to amplify a repeat sequence in the LMP1 gene to corroborate this finding, but this was unsuccessful (data not shown).
DISCUSSION
X-linked hyper-IgM is a rare congenital immunodeficiency affecting around one in a million of the general population. We obtained samples from nine hyper-IgM cases, and using a sensitive, semiquantitative PCR technique demonstrated EBV DNA in 50% of PBMC samples and 62% of throat wash and swab samples. This gives an overall level of EBV positivity in hyper-IgM cases of 67%. Taking account of the age spread of the cases (2 to 44 years), the expected level of seropositivity in healthy individuals in the United Kingdom would be 60 to 70%, and thus we have detected the expected level of EBV infection in the hyper-IgM group. Levels of EBV DNA in hyper-IgM PBMCs and throat wash samples also fell within the normal range for healthy controls (<1 to 103 and <1 to 106genomes/μg DNA, respectively) (Table 1). Since it is highly unlikely that all six EBV-positive hyper-IgM cases were experiencing primary EBV infection at the time that the samples were taken (all were negative for IgM antibodies to EBV capsid antigen), these results indicate that EBV can infect and persist in the absence of a classical memory B-cell population. Additionally the high levels of EBV DNA detected in hyper-IgM throat wash samples (<105 genomes/μg DNA) suggest that the virus can replicate in vivo.
Analysis of lymphocyte subsets in one hyper-IgM case revealed EBV DNA exclusively in the B-cell population, and in another case, where more detailed analysis was possible, EBV was restricted to IgD+ CD27+ B cells. This recently described B-cell subset represents around 5 to 25% of the circulating B-cell population and has many characteristics of classical memory B cells, including CD27 expression and mutated heavy chain genes (22), but its origins and longevity are presently unknown. Since this subset is found in hyper-IgM cases as well as healthy controls, a germinal center-independent pathway is proposed for its ontogeny (22). Cells of this phenotype are also found in the splenic marginal zone (but not in lymph nodes) and it is postulated that they differentiate from nave B cells in this microenvironment (23). Recent evidence suggests that they have a role in T-independent responses because their absence following splenectomy confers susceptibility to severe infections with encapsulated bacteria (15).
Joseph et al. (12) characterized the phenotype of circulating EBV-carrying cells in healthy individuals as IgD– CD27+ CD5– and provided quantitative evidence for its tight restriction within this population. They specifically analyzed the small IgD+ CD27+ subset and showed that EBV infection of this population is very rare (12).
Having detected EBV DNA in hyper-IgM B cells, we next investigated the latent EBV gene expression in the infected cells. We detected EBERs expression in two cases, with LMP2A additionally expressed in one case and EBNA 3C in the other. Expression of EBERs and LMP2A is well documented in PBMCs in EBV persistence in healthy individuals (10, 17, 18; this study), but detection of EBNA 3C, denoting full latent viral gene expression, has only previously been reported in primary infection (infectious mononucleosis) (10, 21) and in the immunocompromised host (10, 19).
Analysis of serial samples from four hyper-IgM cases shows variable EBV detection over time (Table 2), whereas we and others have shown a fairly consistent level of EBV DNA in PBMCs from healthy donors over years (Table 3) (10, 17). One obvious explanation for this is that repeated cycles of infection and loss occur in hyper-IgM as EBV infected cells are eliminated by cytotoxic T lymphocytes by virtue of their expression of the latent viral genes, particularly the immunodominant EBNA 3 complex. However, our findings in hyper-IgM cases are reminiscent of those of Kim et al. (14), who used the MHV-68-infected mouse model to show that latent virus was progressively lost in CD40 null B cells, which could not undergo a germinal center reaction and mature into memory cells, and preferentially maintained in long-lived, isotype-switched CD40+ B cells (14). Thus, the loss of EBV from the IgD+ CD27+ B-cell subset could be due its inability to retain and/or replicate the viral genome, or to the inherently short life span of these B cells. These possibilities cannot be differentiated until more is known about this B-cell subset.
Identification of identical EBNA 3 repeat sequences in two hyper-IgM throat wash isolates taken 6 months apart suggests viral persistence, however, only detailed longitudinal studies of EBV isolates from PBMCs and throat wash samples from hyper-IgM cases and their partner(s) and contacts could resolve this issue.
ACKNOWLEDGMENTS
We thank Shonna MacCall for valuable assistance with flow cytometry and David Gray and Harry White for helpful discussions. We thank J. M. Middeldorp (Department of Pathology, Free University Hospital, Amsterdam, The Netherlands) for the bacterial DNA-based vector with modified DNA sequences used in the quantitative PCR for EBV detection.
This work was supported by Wellcome Trust project grant number 057133.
REFERENCES
Allen, R. C., R. J. Armitage, M. E. Conley, H. M. E. Rosenblatt, N. A. Jenkins, N. G. Copeland, M. A. Bedell, S. Edelhoff, C. M. Disteche, and D. K. Simoneaux. 1993. CD40 ligand gene defects responsible for X-linked hyper-IgM syndrome. Science 259:990-993.
Aruffo, A., M. Farrington, D. Hollenbaugh, X. Li, A. Milatovich, S. Nonoyama, J. Bajorath, L. S. Grosmaire, R. Stenkamp, and M. Neubauer. 1993. The CD40 ligand gp39 is defective in activated T cells from patients with X-linked hyper-IgM syndrome. Cell 72:291-300.
Babcock, G. J., D. Hochberg, and D. A. Thorley-Lawson. 2000. The expression pattern of Epstein-Barr virus latent genes in vivo is dependent upon the differentiation stage of the infected B-cell. Immunity 13:497-506.
Babcock, G. J., L. L. Decker, M. Volk, and D. A. Thorley-Lawson. 1998. EBV persistence in memory B cells in vivo. Immunity. 9:395-404.
Bornkamm, G. W., and W. Hammerschmidt. 2001. Molecular virology of Epstein-Barr virus. Phil. Trans. R. Soc. Lond. B 356:437-459.
Crawford, D. H. 2001. Biology and disease associations of Epstein-Barr virus. Phil. Trans. R. Soc. Lond. B 356:461-473.
Di Santo, J. P., J. Y. Bonnefoy, J. F. Gauchat, A. Fischer, and G. de Saint Basile. 1993. CD40 ligand mutations in X-linked immunodeficiency with hyper-IgM. Nature 361:541-543.
Haque, T., J. A. Thomas, K. I. Falk, R. Parratt, B. J. Hunt, M. Yacoub, and D. H. Crawford. 1996. Transmission of donor Epstein-Barr virus (EBV) in transplanted organ causes lymphoproliferative disease in EBV seronegative recipients. J. Gen. Virol. 77:1169-1172.
Hochberg, D., J. M. Middeldorp, M. Catalina, J. L. Sullivan, K. Luzuriaga, and D. A. Thorley-Lawson. 2004. Demonstration of the Burkitt's lymphoma Epstein-Barr virus phenotype in dividing latently infected memory cells in vivo. Proc. Natl. Acad. Sci. USA 101:239-244.
Hopwood, P. A., L. Brookes, R. Parratt, B. J. Hunt, M. Bokhari, J. A. Thomas, M. Yacoub, and D. H. Crawford. 2002. Persistent Epstein-Barr virus infection: unrestricted latent and lytic viral gene expression in healthy immunosuppressed transplant recipients. Transplantation 74:194-202.
Joseph, A. M., G. J. Babcock, and D. A. Thorley-Lawson. 2000. Cells expressing the Epstein-Barr virus growth program are present in and restricted to the nave B-cell subset of healthy tonsils. J. Virol. 74:9964-9971.
Joseph, A. M., G. J. Babcock, and D. A. Thorley-Lawson. 2000. EBV persistence involves strict selection of latently infected B cells. J. Immunol. 165:2975-2981.
Khanim, F., Q.-Y. Yao, G. Niedobitek, S. G. Sihota, A. B. Rickinson, and L. S. Young. 1996. Analysis of Epstein-Barr virus gene polymorphisms in normal donors and in virus-associated tumors from different geographical locations. Blood. 88:3491-3501.
Kim, I.-J., E. Flao, D. L. Woodland, F. E. Lund, T. D. Randall, and M. A. Blackman. 2003. Maintenance of long term -herpesvirus B-cell latency is dependent on CD40-mediated development of memory B cells. J. Immunol. 171:886-892.
Kreutzmann, S., M. M. Rosado, H. Weber, U. Germing, O. Tournilhac, H.-H. Peter, R. Berner, A. Peters, T. Boehm, A. Plebani, I. Quinti, and R. Carsetti. 2003. Human immunoglobulin M memory B cells controlling Streptococcus pneumoniae infections are generated in the spleen. J. Exp. Med. 197:939-945.
Kurth, J., M.-L. Hansmann, K. Rajewsky, and R. Kuppers. 2003. Epstein-Barr virus-infected B cells expanding in germinal centers of infectious mononucleosis patients do not participate in the germinal center reaction. Proc. Natl. Acad. Sci. USA 100:4730-4735.
Miyashita, E., B. Yang, K. M. C. Lam, D. H. Crawford, and D. A. Thorley-Lawson. 1995. A novel form of Epstein-Barr virus latency in normal B cells in vivo. Cell 80:593-601.
Qu, L., and D. T. Rowe. 1992. Epstein-Barr virus latent gene expression in uncultured peripheral blood lymphocytes. J. Virol. 66:3715-3724.
Qu, L., M. Green, S. Webber, J. Reyes, D. Ellis, and D. Rowe. 2000. Epstein-Barr virus gene expression in the peripheral blood of transplant recipients with persistent circulating viral loads. J. Infect. Dis. 182:1013-1021.
Stevens, S. J., M. B. Vervoort, A. J. Van Den Brule, P. L. Meenhorst, C. J. Meijer, and J. M. Middeldorp. 1999. Monitoring of Epstein-Barr virus DNA load in peripheral blood by quantitative competitive PCR. J. Clin. Microbiol. 37:2852-2857.
Tierney, R. J., N. Steven, L. S. Young, and A. B. Rickinson, A.B. 1994. Epstein-Barr virus latency in blood mononuclear cells: analysis of viral gene transcription during primary infection and in the carrier state. J. Virol. 68:7377-7385.
Weller, S., A. Faili, C. Garcia, M. C. Braun, F. Le Deist, G. de Saint Basile, O. Hermine, A. Fischer, C.-A. Reynaud, and J.-C. Weill. 2001. CD40-CD40L independent immunoglobulin gene hypermutation suggests a second B-cell diversification pathway in humans. Proc. Natl. Acad. Sci. USA 98:1166-1170.
Weller, S., A. Faili, S. Aoufouchi, Q. Guéranger, M. Braun, C.-A. Reynaud, and J.-C. Weill. 2003. Hypermutation in human B cells in vivo and in vitro. Ann. N. Y. Acad. Sci. 987:158-165.(Margaret Conacher, Robin )
Immunobiology Unit, Institute of Child Health, Guilford Street, London WC1N 1EH, United Kingdom
Department of Immunology, John Radcliffe Hospital, Oxford OX3 9DU, United Kingdom
Department of Immunology, Royal Free Hospital, Rowland Hill Street, London NW3 2PF, United Kingdom
Clinical Biochemistry and Immunology, Addenbrooke's Hospital, Hills Road, Cambridge CB2 2QQ, United Kingdom
Regional Immunology Department, Royal Victoria Infirmary, Queen Victoria Road, Newcastle upon Tyne NE1 4LP, United Kingdom
ABSTRACT
Epstein-Barr virus (EBV) is a ubiquitous human herpesvirus that persists in the body for life after primary infection. The primary site of EBV persistence is the memory B lymphocyte, but whether the virus initially infects nave or memory B cells is still disputed. We have analyzed EBV infection in nine cases of X-linked hyper-immunoglobulin M (hyper-IgM) syndrome who, due to a mutation in CD40 ligand gene, do not have a classical, class-switched memory B-cell population (IgD– CD27+). We found evidence of EBV infection in 67% of cases, which is similar to the infection rate found in the general United Kingdom population (60 to 70% for the relevant age range). We detected EBV DNA in peripheral blood B cells and showed in one case that the infection was restricted to the small population of nonclassical, germinal center-independent memory B cells (IgD+ CD27+). Detection of EBV small RNAs, latent membrane protein 2, and EBV nuclear antigen 3C expression in peripheral blood suggests full latent viral gene expression in this population. Analysis of EBV DNA in serial samples showed variability over time, suggesting cycles of infection and loss. Our results demonstrate that short-term EBV persistence can occur in the absence of a germinal center reaction and a classical memory B-cell population.
INTRODUCTION
Epstein-Barr virus (EBV) is a ubiquitous, tumorigenic herpesvirus that infects over 90% of the world's adult population. The virus generally infects during childhood without clinical illness, and thereafter persists in the body for life, only rarely causing disease in the immunocompetent host. This highly successful colonization of the human race is effected by close integration of the virus life cycle with host biological processes (reviewed in reference 6).
EBV enters the body via the mouth and infects B lymphocytes in oropharyngeal lymphoid tissue (tonsil, adenoids), inducing their proliferation. From this nidus of infection the virus is disseminated throughout the body in circulating infected B cells. This infection induces an EBV-specific, cytotoxic T-cell response which is essential for control of the infection and may be so florid as to cause the immunopathological symptoms of infectious mononucleosis (reviewed in reference 6). At resolution of primary infection a persistent virus carrier state is established in which around 1 to 60 x 106 circulating B lymphocytes carry viral DNA (17).
The EBV genome contains a set of unique latent genes, EBV nuclear antigens (EBNA) 1, 2, 3A, B and C, leader protein (LP), latent membrane proteins (LMP) 1, 2A and B, as well as two small EBV-encoded RNAs (EBERs) 1 and 2, which act in concert to induce proliferation and immortalization of B cells infected in vitro. EBNA2 is the main transactivator of viral and cellular genes, whereas LMP1 and 2A mimic CD40 and B-cell receptor signaling, respectively (reviewed in reference 5). B cells expressing these latent antigens can be found in tonsillar tissue, but a more restricted latent viral gene expression is seen in peripheral blood B cells (3). Although EBERs and LMP 2A mRNA are regularly detected (10, 17, 18), it now appears that in most cells the virus is transcriptionally silent (9).
Careful analyses identify the memory B-cell population (characterized by markers: CD27+, IgM–, IgD–) as the specific site of viral persistence in peripheral blood (4); however, there is still debate over whether these cells are infected directly by EBV in lymphoid tissue, or whether they mature from infected nave B cells. In one study EBV-infected nave B cells expressing EBNA2 are reported to be the predominant infected population in tonsil (11). This scenario contrasts with the finding of EBV in a predominantly memory B-cell population in infectious mononucleosis tonsils, suggesting initial infection of cells which have already traversed the germinal center (16). To attempt to resolve these issues we have examined EBV infection in patients without a classical memory B-cell population.
Hyper-immunoglobulin M (IgM) syndrome is a rare heterogeneous immunodeficiency characterized clinically by high levels of IgM and low levels of the other immunoglobulin classes in serum, and recurrent infections. Four types of X-linked hyper-IgM are now recognized, all of which cause defective immunoglobulin class switch recombination. In type 1 CD40 ligand (L) is mutated (1, 2, 7). Consequently, signaling through CD40/CD40L is defective and B cells are unable to undergo a germinal center reaction with immunoglobulin hypermutation, class switching and maturation to memory cells. Thus, the majority of circulating B cells in hyper-IgM have a nave B-cell phenotype (IgM+, IgD+, CD27–), although recently a small population of IgM+ IgD+ CD27+ B cells have been identified in both healthy individuals and hyper-IgM cases which have mutated immunoglobulin genes and are proposed to arise by a germinal center-independent mechanism (23).
In order to investigate the relationship between EBV infection and B-cell maturation we have sought evidence of EBV infection and persistence in nine cases of type 1 X-linked hyper-IgM. Our hypothesis was that, if a post-germinal center memory B-cell population is essential for EBV persistence, then hyper-IgM cases will not carry EBV long term. However, if nave B cells are the initial target of infection, then recurrent infections may occur without the establishment of persistence.
MATERIALS AND METHODS
Hyper-IgM cases and healthy controls. Nine cases of hyper-IgM were investigated. All were male, age range 2 to 44 years (Table 1). All cases were shown to have mutations/deletions in the CD40L gene and/or absence of CD40L expression on activated lymphocytes (data not shown); 2 to 60 ml of peripheral blood and throat wash or swab samples (depending on age) were obtained from each patient. Control samples were obtained from anonymous blood donors (n = 27). Control throat washes were obtained from 16 healthy laboratory workers. This study was approved by the Northern and Yorkshire Multi-Centre Ethics Committee.
Peripheral blood mononuclear cells. Peripheral blood mononuclear cells (PBMC) were separated from whole blood by routine density gradient centrifugation, washed, and counted.
Throat wash and swab samples. Throat wash samples were obtained by gargling with 10 ml normal saline. Where hyper-IgM cases were under 15 years of age, throat swabs were obtained and shaken in 2 ml normal saline. Thereafter both samples were centrifuged at 600 x g for 5 min, and the cell pellets were washed twice in saline. Both cell pellets and supernatant saline were stored at –70°C
Flow cytometry. PBMCs were analyzed on a FACScalibur (Becton Dickinson) after staining with the following antibodies: phycoerythrin-labeled CD154, Cy-chrome-labeled CD3, allophycocyanin-labeled CD19 (BD Biosciences), phycoerythrin-labeled CD27 (Dako), fluorescein isothiocyanate-labeled IgD (Southern Biotechnology Associates) according to the manufacturers' instructions. Isotype antibodies used as negative controls were: goat IgG fluorescein isothiocyanate (Southern Biotechnology Associates), mouse IgG1 phycoerythrin and fluorescein isothiocyanate (BD Biosciences). On occasions stained cell populations were sorted on a fluorescence activated cell sorter (Becton Dickinson FACS Vantage SE) and used for further analysis.
Enrichment of lymphocyte subpopulations. Dynabeads (Dynal Biotech ASA, Oslo, Norway) were added to PBMCs at a bead to cell ratio of 5:1 and the CD3-enriched and CD3-depleted populations were separated by application of a magnet in accordance with the manufacturer's instructions. Both the CD3-enriched and -depleted fractions were then cultured at a concentration of 106/ml in complete RPMI for 24 h, prior to DNA extraction.
EBV detection in PBMCs and throat wash and swab samples. PCR detection of the BamHI W repeat region of the EBV genome was performed using 1 to 2 μg of DNA extracted from hyper-IgM and control PBMCs and throat wash cell pellet samples using the Easy-DNA kit (Invitrogen) following the manufacturer's instructions. On occasions where the number of cells in a specific population was below 104, 106 EBV-negative filler cells (TK-143B human osteosarcoma cells) were added prior to DNA extraction.
The primers amplify a 298-bp fragment of the W repeats; forward primer CTT TAG AGG CGA ATG GGC GC and reverse primer AGG ACC ACT TTA TAC CAG GG. PCR amplification was performed in a final volume of 100 μl with the reaction mixture containing 1x PCR buffer (50 mM KCl, 10 mM Tris-HCl [pH 9.0] and 0.1% Triton X-100), plus 1.5 mM MgCl2, 200 μM each of dATP, dTTP, dGTP, and dCTP, 0.1 μM of each primer, and 1 unit of Taq polymerase (Promega). A dilution series of an EBV-positive cell line (Namalwa, one to two EBV genomes per cell) in an EBV-negative cell line (BJAB) was used to give standard 10-fold dilutions from 1 to 106 EBV genomes per 106 cells.
PCR products were separated by electrophoresis on an agarose gel, transferred to positively charged nylon membranes, probed with an EBV-specific probe labeled with digoxigenin, and visualized using a digoxigenin luminescent kit (Roche). The density of bands detected in the Namalwa dilution series was compared with the test samples by scanning with LabWorks 4.0 (UVP BioImaging systems). This provided a semiquantitative estimate of viral load (10).
In some cases EBV DNA load was determined by quantitative competitive PCR exactly as described by Stevens et al. (20). The amplification reaction contained 50 mM KCl, 1.5 mM MgCl, 10 mM Tris, pH 8.5, 25 pmol of each primer, one of which was biotin labeled, and 1 U of Taq. Cycling conditions were 4 min at 95°C; 40 cycles at 95°C, 55°C, and 72°C for 1 min each; and finally 3 min at 72°C. Products were captured on a streptavidin-coated plate and probed with digoxigenin-labeled wild-type and internal standard probes. Optical density was measured and used to calculate copy number.
Viral isolate typing. The 39-bp repeat region of EBNA 3C was amplified in a nested PCR using outer primers 5'ACA CTT GAG TTC CAT GTC GC 3' and 5' TGTAATCACTGGCAAAGGGC 3', and inner primers 5' TAT CGC ACG AAG AAC AAC CCC 3' and 5' AGA TGT GGG AAC TGG GAG ACC 3' following the method described in Haque et al. (8). The 33-bp repeat region of LMP1 was amplified using primers 5' TTT CCA GCA AGA GTC GCT AGG 3' and 5' GGC GCA CCT GGA GGT GGT CC 3' (13); 10 μl of the product was run on a 2% agarose gel, Southern hybridized with a digoxigenin-labeled specific probe 5' CAC GGG CTC CAA TCA TCT TC 3' (EBNA3C) or 5' AGG ACC CTG ACA ACA CTG AT 3' (LMP1) and visualized using a digoxigenin luminescent kit (Roche).
PCR products were excised and purified using the QIAquick gel extraction kit (QIAGEN). Direct sequencing of the products was performed commercially by MWG Research using the forward primer.
Expression of latent viral genes. RNA was isolated from PBMCs using the RNeasy Mini Protocol (QIAGEN). RNA was DNase treated (Promega) before synthesizing cDNA from 1 μg RNA using the Thermoscript reverse transcription-PCR system (Invitrogen); 50 μg/μl random hexamer was used as primer and 2 μl of the reaction mixture was either used immediately for PCR amplification or stored at –20°C for later use. All cDNA was tested for glyceraldehyde-3-phosphate dehydrogenase and -actin expression to check the quality of the DNA, and only used for analysis of EBV gene expression when these tests were positive.
Reverse transcription-PCRs were optimized for detection of human EBV latent transcripts EBNA1, EBNA3C, and LMP2A and the late lytic transcript encoding the envelope glycoprotein (gp) 350, using the EBV-positive cell line B95-8. PCR amplification was performed in a final volume of 50 μl with the reaction mixture containing the following: 1x PCR buffer (50 mM KCl, 10 mM Tris-HCl [pH 9.0] and 0.1% Triton X-100), plus 1.5 mM MgCl2 (3 mM MgCl2 for gp350), 200 μM each of dATP, dTTP, dGTP, and dCTP, 0.1 μM of each primer and 1 unit of Taq polymerase (Promega). Amplification consisted of an initial denaturation for 2 min at 94°C in all reactions, followed by 35 cycles (except EBERs, 30 cycles) of denaturation at 94°C for 1 min, annealing at various temperatures (EBERs at 50°C, EBNA3C at 54°C, gp350, LMP1, and LMP2A at 55°C, EBNA3C nested, LMP1 nested, and LMP2A nested at 57°C) for 2 min (EBERs 1 min) and extension at 72°C for 2 min (EBERs, 1 min), with a final extension at 72°C for 10 min.
Controls included B95-8 (positive), water and untranscribed test RNA (to monitor for DNA contamination); 18 μl of PCR products was then separated by electrophoresis on an agarose gel, transferred to positively charged nylon membranes, and visualized using a digoxigenin luminescent kit (Roche). Primer and probe sequences used are as described in Hopwood et al. (10). With the exception of EBERs and gp350, each primer pair was designed to amplify across introns, giving an amplified cDNA which differs in size from possible amplified contaminating DNA (sizes of possible contaminants: LMP1, 535 bp; LMP2A, 278 bp; and EBNA3C, 386 bp).
RESULTS
Hyper-IgM cases are infected with EBV at the expected rate. Since all hyper-IgM cases were treated with regular infusions of pooled human immunoglobulin, routine serological testing could not be used to indicate past or present EBV infection. We therefore used PCR analysis to detect viral DNA as a marker of infection. We analyzed PBMCs from eight hyper-IgM cases and 20 healthy controls (18 known EBV seropositive and 2 EBV seronegative). Throat wash and swab samples were also available from eight cases.
EBV DNA from the BamHI-W repeat region of EBNA LP was amplified by PCR. This technique regularly detects EBV DNA from 1 to 10 Namalwa cells (1 to 2 EBV genome copies/cell) in a background of 106 EBV-negative cells (Fig. 1). Four of eight cases (50%) and 10 of 18 healthy controls (56%) gave a positive result in PBMCs. Five of eight throat wash and swab cell pellet samples (62%) from hyper-IgM cases and 10 of 16 (63%) from healthy controls were PCR positive. Thus, cases and controls gave similar results. Overall six out of nine hyper-IgM cases (66.7%) showed evidence of EBV infection in either PBMCs or throat wash and swab samples, with three cases being positive in both (Table 1 and Fig. 1a and b). Densitometric analysis of these results showed the viral load in hyper-IgM peripheral blood to be within the same range as that of healthy EBV carriers (<1 to 103 genomes/μg DNA) (Table 1). Neither hyper-IgM serum/plasma nor concentrated throat wash supernatant samples contained any detectable EBV DNA (data not shown).
Hyper-IgM cases show variable detection of EBV DNA over time. Since our findings indicated that hyper-IgM cases are infected with EBV, we analyzed serial blood and throat wash samples from four individuals to determine whether this infection persisted over time. We received samples from two time points from cases 1, 2, and 3 and at four time points from case 8. Each case showed a pattern which varied over time, with one set of samples being negative at at least one time point (Table 2). Only case 8 showed positive blood and throat wash samples on more than one occasion; the first two sets of samples taken 6 months apart were positive, but the samples taken 25 months later were negative. The final blood sample taken 9 months later was again positive (Table 2). These results contrast with our findings in PBMC from 5 healthy EBV seropositive individuals. Over a similar time period controls showed a relatively stable EBV load with levels varying from detectable to undetectable in only one individual (Table 3). Similarly EBV DNA in throat wash from healthy donors varied over time from detectable to nondetectable in one of four donors (Table 3).
Further analysis to detect EBV DNA was only possible on cases 3 and 8, where repeat blood samples with detectable EBV DNA were available.
EBV DNA is only present in the CD27+, IgD+ B-cell fraction in hyper-IgM. PBMCs from case 8 were used to identify the cell type carrying EBV DNA. PBMCs were separated into CD3-enriched and -depleted fractions using magnetic beads. Fractions were cultured for 24 h prior to DNA preparation and PCR. EBV DNA was detected in the CD3-depleted but not CD3-enriched fraction indicating that the virus is present in peripheral B cells (Fig. 2).
Circulating B cells from hyper-IgM cases mainly consist of short lived IgD+ CD27– nave cells, however, a small population of IgD+ CD27+ B cells has been reported which, although not classical memory B cells, do have mutated immunoglobulin genes (22). Flow cytometric analysis on PBMCs from hyper-IgM PBMCs stained for CD19, CD27, and IgD showed that the majority of CD19+ cells were IgD+ (86.7 to 96.4%). A small population (3.1 to 5.4%) of CD27+, IgD+ cells could be detected. The mean percentage of CD27+ IgD+ cells in six healthy controls was 35% (Fig. 3a).
In order to identify the B-cell subset carrying EBV in hyper-IgM, PBMCs from case 3 were further separated into IgD+ CD27+ and IgD– CD27+ subsets using fluorescence-activated cell sorting. PCR analysis detected EBV DNA exclusively in the IgD+ CD27+ subset. The PCR was repeated on a second occasion and gave the same result. Quantitative PCR indicated a ratio of 1 EBV genome per 30 B cells in this population (Fig. 3b), however, since the cell populations were incubated for 24 h prior to DNA extraction, it is possible that the lytic cycle had been initiated, accounting for this high viral load.
PBMCs from four healthy controls in which EBV DNA was detected by PCR were separated by fluorescence-activated cell sorting on the basis of IgD and CD27 and processed exactly as above. EBV DNA was not detected in the IgD+ CD27+ population in any of these samples.
Hyper-IgM PBMCs express latent viral genes. Having detected EBV in hyper-IgM B cells, we wished to determine which latent viral genes were expressed in these infected cells. RNA from three hyper-IgM PBMCs samples (nos. 3 and 8) was subjected to reverse transcription-PCR to amplify EBER1, LMP1, LMP2A, EBNA3C, and gp350 sequences. EBERs were detected in both cases, with additional detection of LMP2A in one case (no. 3) and EBNA 3C in another (no. 8) (Fig. 4). LMP1 and gp350 expression was not detected in any sample tested (data not shown). Similar RNA analysis on PBMCs from 18 healthy EBV-seropositive controls detected EBERs in 78% and LMP2A in 67%, but no LMP1, EBNA3C, or gp350 expression.
Hyper-IgM case 8 has EBV isolates with identical numbers of EBNA 3C repeats over time. Only case 8 showed evidence of EBV infection sustained over time (Table 2), and we therefore used the EBV-positive blood and throat wash samples taken 6 months apart to distinguish between persistent infection and reinfection in this case. PCR amplification of a repeat sequence in the EBNA 3C gene known to vary in size between viral isolates (8) gave an amplified product from the two throat wash samples, but not from the PBMCs. The same size bands were amplified from both samples which were estimated by size comparison with control samples with known numbers of repeats to contain 6 repeat sequences (Fig. 5). Excision of the bands and sequencing of the products confirmed that the two products were identical. However, results from a series of control samples showed that 26% of individual isolates have six repeats in this region. We therefore attempted to amplify a repeat sequence in the LMP1 gene to corroborate this finding, but this was unsuccessful (data not shown).
DISCUSSION
X-linked hyper-IgM is a rare congenital immunodeficiency affecting around one in a million of the general population. We obtained samples from nine hyper-IgM cases, and using a sensitive, semiquantitative PCR technique demonstrated EBV DNA in 50% of PBMC samples and 62% of throat wash and swab samples. This gives an overall level of EBV positivity in hyper-IgM cases of 67%. Taking account of the age spread of the cases (2 to 44 years), the expected level of seropositivity in healthy individuals in the United Kingdom would be 60 to 70%, and thus we have detected the expected level of EBV infection in the hyper-IgM group. Levels of EBV DNA in hyper-IgM PBMCs and throat wash samples also fell within the normal range for healthy controls (<1 to 103 and <1 to 106genomes/μg DNA, respectively) (Table 1). Since it is highly unlikely that all six EBV-positive hyper-IgM cases were experiencing primary EBV infection at the time that the samples were taken (all were negative for IgM antibodies to EBV capsid antigen), these results indicate that EBV can infect and persist in the absence of a classical memory B-cell population. Additionally the high levels of EBV DNA detected in hyper-IgM throat wash samples (<105 genomes/μg DNA) suggest that the virus can replicate in vivo.
Analysis of lymphocyte subsets in one hyper-IgM case revealed EBV DNA exclusively in the B-cell population, and in another case, where more detailed analysis was possible, EBV was restricted to IgD+ CD27+ B cells. This recently described B-cell subset represents around 5 to 25% of the circulating B-cell population and has many characteristics of classical memory B cells, including CD27 expression and mutated heavy chain genes (22), but its origins and longevity are presently unknown. Since this subset is found in hyper-IgM cases as well as healthy controls, a germinal center-independent pathway is proposed for its ontogeny (22). Cells of this phenotype are also found in the splenic marginal zone (but not in lymph nodes) and it is postulated that they differentiate from nave B cells in this microenvironment (23). Recent evidence suggests that they have a role in T-independent responses because their absence following splenectomy confers susceptibility to severe infections with encapsulated bacteria (15).
Joseph et al. (12) characterized the phenotype of circulating EBV-carrying cells in healthy individuals as IgD– CD27+ CD5– and provided quantitative evidence for its tight restriction within this population. They specifically analyzed the small IgD+ CD27+ subset and showed that EBV infection of this population is very rare (12).
Having detected EBV DNA in hyper-IgM B cells, we next investigated the latent EBV gene expression in the infected cells. We detected EBERs expression in two cases, with LMP2A additionally expressed in one case and EBNA 3C in the other. Expression of EBERs and LMP2A is well documented in PBMCs in EBV persistence in healthy individuals (10, 17, 18; this study), but detection of EBNA 3C, denoting full latent viral gene expression, has only previously been reported in primary infection (infectious mononucleosis) (10, 21) and in the immunocompromised host (10, 19).
Analysis of serial samples from four hyper-IgM cases shows variable EBV detection over time (Table 2), whereas we and others have shown a fairly consistent level of EBV DNA in PBMCs from healthy donors over years (Table 3) (10, 17). One obvious explanation for this is that repeated cycles of infection and loss occur in hyper-IgM as EBV infected cells are eliminated by cytotoxic T lymphocytes by virtue of their expression of the latent viral genes, particularly the immunodominant EBNA 3 complex. However, our findings in hyper-IgM cases are reminiscent of those of Kim et al. (14), who used the MHV-68-infected mouse model to show that latent virus was progressively lost in CD40 null B cells, which could not undergo a germinal center reaction and mature into memory cells, and preferentially maintained in long-lived, isotype-switched CD40+ B cells (14). Thus, the loss of EBV from the IgD+ CD27+ B-cell subset could be due its inability to retain and/or replicate the viral genome, or to the inherently short life span of these B cells. These possibilities cannot be differentiated until more is known about this B-cell subset.
Identification of identical EBNA 3 repeat sequences in two hyper-IgM throat wash isolates taken 6 months apart suggests viral persistence, however, only detailed longitudinal studies of EBV isolates from PBMCs and throat wash samples from hyper-IgM cases and their partner(s) and contacts could resolve this issue.
ACKNOWLEDGMENTS
We thank Shonna MacCall for valuable assistance with flow cytometry and David Gray and Harry White for helpful discussions. We thank J. M. Middeldorp (Department of Pathology, Free University Hospital, Amsterdam, The Netherlands) for the bacterial DNA-based vector with modified DNA sequences used in the quantitative PCR for EBV detection.
This work was supported by Wellcome Trust project grant number 057133.
REFERENCES
Allen, R. C., R. J. Armitage, M. E. Conley, H. M. E. Rosenblatt, N. A. Jenkins, N. G. Copeland, M. A. Bedell, S. Edelhoff, C. M. Disteche, and D. K. Simoneaux. 1993. CD40 ligand gene defects responsible for X-linked hyper-IgM syndrome. Science 259:990-993.
Aruffo, A., M. Farrington, D. Hollenbaugh, X. Li, A. Milatovich, S. Nonoyama, J. Bajorath, L. S. Grosmaire, R. Stenkamp, and M. Neubauer. 1993. The CD40 ligand gp39 is defective in activated T cells from patients with X-linked hyper-IgM syndrome. Cell 72:291-300.
Babcock, G. J., D. Hochberg, and D. A. Thorley-Lawson. 2000. The expression pattern of Epstein-Barr virus latent genes in vivo is dependent upon the differentiation stage of the infected B-cell. Immunity 13:497-506.
Babcock, G. J., L. L. Decker, M. Volk, and D. A. Thorley-Lawson. 1998. EBV persistence in memory B cells in vivo. Immunity. 9:395-404.
Bornkamm, G. W., and W. Hammerschmidt. 2001. Molecular virology of Epstein-Barr virus. Phil. Trans. R. Soc. Lond. B 356:437-459.
Crawford, D. H. 2001. Biology and disease associations of Epstein-Barr virus. Phil. Trans. R. Soc. Lond. B 356:461-473.
Di Santo, J. P., J. Y. Bonnefoy, J. F. Gauchat, A. Fischer, and G. de Saint Basile. 1993. CD40 ligand mutations in X-linked immunodeficiency with hyper-IgM. Nature 361:541-543.
Haque, T., J. A. Thomas, K. I. Falk, R. Parratt, B. J. Hunt, M. Yacoub, and D. H. Crawford. 1996. Transmission of donor Epstein-Barr virus (EBV) in transplanted organ causes lymphoproliferative disease in EBV seronegative recipients. J. Gen. Virol. 77:1169-1172.
Hochberg, D., J. M. Middeldorp, M. Catalina, J. L. Sullivan, K. Luzuriaga, and D. A. Thorley-Lawson. 2004. Demonstration of the Burkitt's lymphoma Epstein-Barr virus phenotype in dividing latently infected memory cells in vivo. Proc. Natl. Acad. Sci. USA 101:239-244.
Hopwood, P. A., L. Brookes, R. Parratt, B. J. Hunt, M. Bokhari, J. A. Thomas, M. Yacoub, and D. H. Crawford. 2002. Persistent Epstein-Barr virus infection: unrestricted latent and lytic viral gene expression in healthy immunosuppressed transplant recipients. Transplantation 74:194-202.
Joseph, A. M., G. J. Babcock, and D. A. Thorley-Lawson. 2000. Cells expressing the Epstein-Barr virus growth program are present in and restricted to the nave B-cell subset of healthy tonsils. J. Virol. 74:9964-9971.
Joseph, A. M., G. J. Babcock, and D. A. Thorley-Lawson. 2000. EBV persistence involves strict selection of latently infected B cells. J. Immunol. 165:2975-2981.
Khanim, F., Q.-Y. Yao, G. Niedobitek, S. G. Sihota, A. B. Rickinson, and L. S. Young. 1996. Analysis of Epstein-Barr virus gene polymorphisms in normal donors and in virus-associated tumors from different geographical locations. Blood. 88:3491-3501.
Kim, I.-J., E. Flao, D. L. Woodland, F. E. Lund, T. D. Randall, and M. A. Blackman. 2003. Maintenance of long term -herpesvirus B-cell latency is dependent on CD40-mediated development of memory B cells. J. Immunol. 171:886-892.
Kreutzmann, S., M. M. Rosado, H. Weber, U. Germing, O. Tournilhac, H.-H. Peter, R. Berner, A. Peters, T. Boehm, A. Plebani, I. Quinti, and R. Carsetti. 2003. Human immunoglobulin M memory B cells controlling Streptococcus pneumoniae infections are generated in the spleen. J. Exp. Med. 197:939-945.
Kurth, J., M.-L. Hansmann, K. Rajewsky, and R. Kuppers. 2003. Epstein-Barr virus-infected B cells expanding in germinal centers of infectious mononucleosis patients do not participate in the germinal center reaction. Proc. Natl. Acad. Sci. USA 100:4730-4735.
Miyashita, E., B. Yang, K. M. C. Lam, D. H. Crawford, and D. A. Thorley-Lawson. 1995. A novel form of Epstein-Barr virus latency in normal B cells in vivo. Cell 80:593-601.
Qu, L., and D. T. Rowe. 1992. Epstein-Barr virus latent gene expression in uncultured peripheral blood lymphocytes. J. Virol. 66:3715-3724.
Qu, L., M. Green, S. Webber, J. Reyes, D. Ellis, and D. Rowe. 2000. Epstein-Barr virus gene expression in the peripheral blood of transplant recipients with persistent circulating viral loads. J. Infect. Dis. 182:1013-1021.
Stevens, S. J., M. B. Vervoort, A. J. Van Den Brule, P. L. Meenhorst, C. J. Meijer, and J. M. Middeldorp. 1999. Monitoring of Epstein-Barr virus DNA load in peripheral blood by quantitative competitive PCR. J. Clin. Microbiol. 37:2852-2857.
Tierney, R. J., N. Steven, L. S. Young, and A. B. Rickinson, A.B. 1994. Epstein-Barr virus latency in blood mononuclear cells: analysis of viral gene transcription during primary infection and in the carrier state. J. Virol. 68:7377-7385.
Weller, S., A. Faili, C. Garcia, M. C. Braun, F. Le Deist, G. de Saint Basile, O. Hermine, A. Fischer, C.-A. Reynaud, and J.-C. Weill. 2001. CD40-CD40L independent immunoglobulin gene hypermutation suggests a second B-cell diversification pathway in humans. Proc. Natl. Acad. Sci. USA 98:1166-1170.
Weller, S., A. Faili, S. Aoufouchi, Q. Guéranger, M. Braun, C.-A. Reynaud, and J.-C. Weill. 2003. Hypermutation in human B cells in vivo and in vitro. Ann. N. Y. Acad. Sci. 987:158-165.(Margaret Conacher, Robin )