Abstract
In this paper we demonstrate, for the first time, that Epstein-Barr virus (EBV)-infected cells expressing the lymphoblastoid growth program are present in healthy carriers of the virus. Previously we observed that latently infected naive B cells are present in tonsils only when viral replication is detected, suggesting that these may represent newly infected B cells. We have tested this idea by performing a reverse transcription-PCR analysis for the expression of latent genes (EBNA2 and the EBNA3s) that are characteristically expressed only by newly infected cells expressing the growth latency program. EBNA2 expression is regularly detected in purified naive (IgD+) tonsillar B cells (13 of 16 tonsils tested) but was never found in the IgD− population (0 of 16). More detailed analysis revealed that the mRNAs for the latent genes EBNA1 (3 of 3 tonsils tested), EBNA3a (3 of 5), EBNA3b (3 of 5), EBNA3c (3 of 5), LMP1 (6 of 6), and LMP2 (5 of 6) were also present in the IgD+ population, but the EBNA1Q-K transcript, characteristic of nonlymphoblastoid forms of latency, was never detected (0 of 6). Finally, we demonstrate that the latently infected naive (IgD+) cells express CD80 (B7.1), a marker characteristically expressed on activated naive lymphoblasts but absent from resting naive B cells. The infected naive (IgD+) population in the tonsil therefore has the viral and cellular phenotype of a B-cell directly infected with EBV—an activated lymphoblast expressing the growth program.
Epstein-Barr virus (EBV) is a human, B-lymphotropic herpesvirus that is best known for its capacity to immortalize normal B cells in vitro and for its association with a number of human neoplasias, including both lymphomas and carcinomas (for reviews see references 14 and 27). EBV immortalizes B cells in vitro by infecting them and driving them to become proliferating lymphoblasts (34) through the expression of nine latent proteins under the control of the transcription factor EBNA2 (38, 39, 42). This state of infection is referred to as the growth program (35) or latency 3 (27), and EBNA2 expression is a specific characteristic of EBV-infected B cells using this program. Like other members of the herpesvirus family, EBV also has the capacity to establish a life-long, persistent infection. Despite the pathogenic potential of the virus, life-long infection is benign in the overwhelming majority of the infected population.
Recent studies have begun to unravel the mechanism of persistent infection in healthy individuals, which is at variance with the pathogenic behavior classically associated with EBV. In the peripheral blood of healthy carriers, the virus is tightly restrained, being found only in resting memory B cells (4, 19). The only latent gene to be consistently expressed in the peripheral blood is LMP2 (5, 26, 36), and there is evidence to suggest that even this gene may not be expressed in the majority of the infected cells (3). We have referred to this state as the latency program (35) and proposed that these cells are the site of long-term persistent infection because they are not a pathogenic threat to the host and are probably not subject to immunosurveillance. If true, then EBV, in its site of persistence, is much like the other herpesviruses in that it persists in a transcriptionally quiescent state in a long-lived, resting cell. The ability of EBV to establish a latent infection in a resting B cell raises the question of what the role of the EBV growth program may be in vivo.
Cells expressing the growth program have been detected in the blood during acute infectious mononucleosis (36) but have never been found in healthy, persistently infected individuals (19, 36), even when immunosuppressed (3). It is now well established that all healthy carriers have large numbers of cytotoxic T cells (CTL) that recognize epitopes from latent proteins that are uniquely expressed during the growth program, namely, EBNA2 and the EBNA3s (reviewed in reference 13). The suggestion has been made that the lymphoblastoid form of latency is undetectable in healthy carriers because the cells are immediately destroyed by CTL. This led to the proposition that the growth program may be required to establish a latent infection before the CTL response arises, but thereafter it is not required for life-long maintenance of the persistent infection (15, 18, 28).
It is believed that EBV establishes infection through exposure of the mucosal lymphoepithelium to saliva containing infectious virus. Similarly, it is believed that the virus is shed from the mucosal lymphoepithelium into saliva (2). Recently we observed that, in tonsils, unlike the peripheral blood, there are significant numbers of latently infected, naive B cells, and their presence is associated with ongoing viral replication (4). In rare cases when we were unable to detect viral replication, infected, naive B cells were absent. This led to the suggestion that naive B cells in the tonsil are being infected with EBV to produce proliferating lymphoblasts driven by the growth program (4). These infected naive B cells would either be killed by CTL or differentiate within the tonsil to become resting memory B cells (32, 33). The latently infected cells could then leave the tonsil, accounting for the restriction of EBV to memory cells in the periphery.
In this study, we have used reverse transcription-PCR (RTPCR) analysis for EBNA2 expression to screen B-cell subsets from a panel of healthy tonsils, to test for cells expressing the growth program. We found that EBNA2 was consistently expressed but only in the naive B-cell subset, as predicted from our previous studies. Furthermore, the latently infected, naive B cells express all of the growth program latent proteins expected for B cells that have been directly infected with EBV.
MATERIALS AND METHODS
Cells and cell lines.
An in vitro-immortalized, EBV-positive lymphoblastoid cell line was used as a positive control for DNA PCR and RTPCR for EBNA1 (U-K), EBNA2, EBNA3a, EBNA3b, EBNA3c, LMP1, and LMP2a. Rael (gift of S. Speck), an EBV-positive Burkitt's lymphoma cell line, was used as a positive control for EBNA1 (Q-K) RTPCR. The cell lines were maintained in 5% CO2 in RPMI 1640 with 10% fetal bovine serum and penicillin and streptomycin.
Tonsils were obtained from patients undergoing routine tonsillectomies, primarily for obstructed-breathing disorders, at the New England Medical Center and the Massachusetts General Hospital. Tonsils were minced in PBSA (phosphate-buffered saline [PBS] with 0.5% bovine serum albumin), and the resulting suspension was passed through a silkscreen to remove any connective tissue. The cell suspension was diluted to 108 cells/ml, layered onto Ficoll-Hypaque (Pharmacia), and centrifuged at 2,000 rpm for 30 min at 25°C. Buffy coats were removed, washed twice with PBSA, and centrifuged at 1,200 rpm for 15 min. The isolated lymphocytes were then either stored frozen in aliquots for future study or fractionated into immunoglobulin D-positive (IgD+) and IgD− subsets. The frequency of virus-infected cells was estimated for each subpopulation by limiting-dilution DNA PCR as described below. Lymphocytes from EBV-negative tonsils were used as negative controls for all RTPCR experiments and were also used to bring the cell number up to 5 × 106 when necessary prior to mRNA isolation.
Magnetic bead separations.
Tonsillar mononuclear cells were resuspended to 2 × 107 cells/ml in PBSA as 1-ml aliquots. To positively select the naive (IgD+) population, biotinylated anti-IgD antibody (0.015 μg) (Southern Biotech catalog no. 2030-08) was added to each tube and incubated on a rotator at 4°C for 30 min. All tubes were washed two times with PBSA and resuspended to 180 μl in the same buffer. Then 20 μl of streptavidin-coated microbeads (Miltenyi) was added to each tube and incubated for 10 min at 4°C. Cells were again washed and resuspended in 500 μl for separation using a magnetic cell separation column (Miltenyi) and kept at 4°C at all times. The MACS column (AS or CS, depending on the cell number) was prepared by rinsing with 5 column volumes of PBSA and then inserted into a VarioMACS magnet. The flow rate of the column was adjusted by attaching a 25- (AS) or 23 (CS)-gauge needle to the base of the stopcock. Cells were then loaded onto the column, and the negative fraction was collected. The cells were washed through by applying 3 column volumes of PBSA to the column while still attached to the magnet. The retained population was then washed by removing the column from the magnet and injecting 1 column volume of PBSA from the bottom of the column using the side syringe supplied. The needle was then replaced with a 23- (AS) or 21 (CS)-gauge needle, the column was reinserted into the magnet, and cells were allowed to flow through. These cells were collected as the wash fraction and discarded. The column was again rinsed with 3 column volumes of PBSA as before to remove any remaining nonspecifically bound cells. The column was then removed from the magnet, the needle was removed, and the column was washed with 5 column volumes of PBSA to elute the retained cells. The resulting IgD+ fraction was set aside, and the IgD− B-cell fraction was isolated from the whole IgD− fraction by positive selection for the pan-B-cell marker CD19. The biotinylated anti-CD19 antibody was prepared in our own laboratory and was used at 0.072 μg/ml per tube.
FACS cell sorting.
To isolate CD80-positive and -negative IgD+ B cells, whole tonsillar lymphocyte populations were resuspended to 5 × 107 cells/ml in PBSA as 1-ml aliquots. The cells were then labeled by incubating each tube with 0.09 μg of phycoerythrin (PE)-conjugated anti-CD80 (Pharmingen) and fluorescein isothiocyanate (FITC)-coupled anti-IgD (1:100 dilution of stock [Southern Biotech]) or 0.55 μg of FITC-coupled anti-CD80 (Pharmingen) and PE-conjugated anti-IgD (Southern Biotech). After the cells were labeled and washed, they were fractionated with either a FACStar (Becton Dickinson) or MoFLo (Cytomation) cell sorter for fluorescence-activated cell sorting (FACS).
FACS analysis and antibodies.
All fractionated populations were analyzed using a Becton Dickinson FACScan or FACSCalibur. After separations, all column fractions were stained with a PE-conjugated anti-IgD (Southern Biotec) as well as anti-CD20-FITC (Dako) for reanalysis to assess purity and recovery. As negative controls, MOPC21 (IgG1 isotype control; Sigma), 1a2 (IgG2a isotype control; this laboratory), and MOPC121 (IgG2b isotype control; Sigma) were used.
Limiting-dilution DNA PCR.
The absolute frequency of virus-infected cells in a given population was estimated by limiting-dilution DNA PCR analysis with a DNA PCR assay that can detect the presence of a single viral genome in as many as 106 uninfected cells. The quantitative and technical aspects of this assay have been detailed elsewhere (12). Isolated populations were serially diluted, and replicates, usually eight, of each dilution were prepared in a 96-well V-bottomed microtiter plate (Immulon). The plate was centrifuged at 1,500 rpm for 15 min at 4°C, and the supernatant was aspirated. Then, 10 μl of a lysis solution containing 0.45% Tween 20, 0.45% NP-40, 2 mM MgCl2, 50 mM KCl, 10 mM Tris (pH 8.3), and 0.5 mg of proteinase K per ml was added to each well, and the plate was incubated for at least 2 h at 55°C. After incubation, the plate was centrifuged quickly to remove condensation from the lid of the plate. PCR was performed in a final volume of 50 μl per reaction; 5 μl of cell lysate was added to each PCR. The PCR and Southern blotting conditions have been described in detail previously (20).
Poisson statistics were used to determine the absolute number of infected cells from the limiting-dilution analysis as described previously (12, 20). To exclude the possibility of external contamination of the DNA PCR, we included eight negative DNA samples per analysis. Furthermore, the lack of contamination could be confirmed since the PCR signals always fractionate consistently, for example, into the IgD− subset of peripheral blood cells, and the signals titrated out, i.e., they were weaker and less frequent with fewer cells.
cDNA synthesis.
Cells (usually 5 × 106) were pelleted in a microcentrifuge tube, resuspended in 1 ml of Trizol reagent (Gibco-BRL), and incubated at room temperature for 5 min. Then, 200 μl of chloroform was added, and the tube was shaken vigorously for 10 s. The suspension was incubated at room temperature for 5 min and centrifuged at 11,500 rpm in an Eppendorf microcentrifuge for 15 min at 4°C. The top aqueous layer was transferred to a fresh microcentrifuge tube containing 500 μl of isopropyl alcohol and incubated for 10 min at room temperature. The tube was centrifuged at 11,500 rpm for 10 min at 4°C, and the supernatant was aspirated. Then, 1 ml of 75% ethanol was added, and the tube was vigorously vortexed, and again centrifuged at 9,200 rpm for 5 min at 4°C. The supernatant was aspirated, and the pellet was allowed to dry for 10 min at room temperature. The pellet was resuspended in 7 or 14 μl of high-pressure liquid chromatography water (HPLC H2O) at 55°C for 15 min.
To synthesize cDNA from purified RNA, 7 μl of RNA suspension was transferred to a 200-μl Microamp reaction tube, and 5 μl of random primers (50 ng/μl; Gibco-BRL) was added to the RNA suspension. The mix was heated at 68°C for 8 min, followed by a 2-min incubation on ice. The tubes were rapidly centrifuged to remove condensation, and 7 μl of a mix containing 1 μl of 10 mM deoxynucleoside triphosphates (dNTPs), 4 μl of 5× Superscript II buffer (375 mM KCl, 250 mM Tris [pH 8.4], 15 mM MgCl2), and 2 μl of 100 mM dithiothreitol was added. The reaction was incubated at room temperature for 10 min, followed by the addition of 50 U of Superscript reverse transcriptase (Gibco-BRL catalog no. 18064-014) and incubation at room temperature for an additional 10 min. Next, the tube was incubated at 42°C for 50 min, and the reaction was stopped by incubation at 68°C for 15 min. The volume was made up to 100 μl by addition of 80 μl of HPLC H2O, and 20 μl was used for each PCR.
PCR of latent gene products.
PCR was performed on the synthesized cDNA for EBNA1 (U-K), EBNA1 (Q-K), EBNA2, EBNA3a, EBNA3b, EBNA3c, LMP1, and LMP2a. Reaction conditions for each were 50 mM KCl, 20 mM Tris (pH 8.4), 0.2 mM dNTPs, and 20 pM each of the amplimers. MgCl2 was 2 mM for EBNA1 (U-K), EBNA3a, EBNA3b, EBNA3c, and LMP2a, 2.5 mM for EBNA1 (Q-K), and 3.0 mM for EBNA2 and LMP1. The amplimers were as follows: EBNA1 (U-K): E1U (5′-AGCTTCCCTGGGATGAGCGT-3′) and E1K (5′-TCTTCCCCGTCCTCGTCCAT-3′) (26); EBNA1 (Q-K), RT3 (5′-TGGCCCCTCGTCAGACATGATT-3′) and Qb (5′-AGCGTGCGCTACCGGAT-3′) (gift from Sam Speck); EBNA2, E2F (5′-CATAGAAGAAGAAGAGGATGAAGA-3′) and E2R (5′-GTAGGGATTCGAGGGAATTACTGA-3′) (26); EBNA3a, L1 (5′-TCTTCCATGTTGTCATCCAGGG-3′) and U1 (5′-CTTAGGAAGCGTTTCTTGAGCTT-3′); EBNA3b, E3B-S (5′-TTCCATGTTGCAATCGGACC-3′) and E3B-AS (5′-AAAGTGACCTAGCACGACGT-3′) (gift from Robert Touitou); EBNA3c, E3C-S (5′-GGGCTGTCAAGCAATCGCAC-3′) and E3C-AS (5′-GTGGTGCATTCCACGGGTAA-3′) (gift from Robert Touitou); LMP1, L1F (5′-TTGGTGTACTCCTACTGATGATCACC-3′) and L1R (5′-AGTAGATCCAGATACCTAAGACAAGT-3′) (26); and LMP2a, L2F (5′-ATGACTCATCTCAACACATA-3′) and L2R (5′-CATGTTAGGCAAATTGCAAA) (26).
Master mixes with the indicated conditions were aliquoted to 200-μl Microamp reaction tubes, and 20 μl of the cDNA suspension described above was added to give a final volume of 50 μl. Therefore it was possible to perform RTPCR for up to five different latent genes from one cDNA pot. Reactions were incubated at 95°C for 5 min, and 1 U of Taq DNA polymerase (Perkin Elmer) was added to each tube. The tubes were loaded in a Geneamp 9600 thermocycler and the following conditions were run: for EBNA1 (Q-K), EBNA3a, EBNA3b, EBNA3c, and EBNA2, 95°C for 15 s, 62°C for 30 s, and 72°C for 30 s, repeated for 40 cycles; for EBNA1 (U-K) and LMP1, 95°C for 15 s, 65°C for 30 s, and 72°C for 30 s, repeated for 40 cycles; for LMP2a, 95°C for 15 s, 55°C for 30 s, and 72°C for 1 min, repeated for 40 cycles.
All reactions were concluded with a 5-min incubation at 72°C to complete the extension of all synthesized products. PCR products were electrophoresed on agarose gels, blotted, and probed as described above.
RTPCR products were analyzed in the same manner as the DNA PCRs except the PCR products were detected by probing Southern blots with 32P-random-primed PCR products isolated from an EBV-positive lymphoblastoid or Rael cell line. An EBV-positive lymphoblastoid cell line was used as a positive control for RTPCR for EBNA1 (U-K), EBNA2, EBNA3a, EBNA3b, EBNA3c, LMP1, and LMP2a, and the Rael Burkitt's lymphoma cell line was used as a positive control for EBNA1 (Q-K) RTPCR. These protocols detect the mRNA from 1 EBV-positive cell in the presence of 5 × 106 EBV-negative tonsil cells (see Fig. 2).
FIG. 2.
EBNA2 is expressed by latently infected naive but not IgD− B cells in the tonsil. IgD+ (naive) and IgD− B cells were isolated from tonsils using biotinylated antibodies and MACS columns as described in the legend to Fig. 1. RTPCR was then performed on 5 × 106 cells from each population. The PCR products were fractionated on agarose gels, Southern blotted, and probed with a sequence-specific probe. For details see the text. The expected position of the PCR product is indicated by the arrow. (A) Examples from five different EBV-positive tonsils. D+, surface IgD+ B-cell population; D−, surface IgD− B-cell population. (B) Results from a single tonsil with spiking and negative controls. Either five or one cell from an EBNA2-positive lymphoblastoid cell line (LCL) was spiked into the IgD− population prior to mRNA extraction for EBNA2 RTPCR. The negative controls are B cells from tonsils that were EBV negative in the DNA PCR assays used to derive the frequencies shown in Table 1.
RESULTS
Naive (IgD+) but not IgD− B cells from the tonsil express EBNA2, the latent gene characteristic of the lymphoblastoid growth program.
When EBV newly infects resting B cells, EBNA2 is the first gene to be expressed (1, 29), and it regulates the expression of all the other latent genes (38, 39, 42). The result is an activated lymphoblast proliferating under the control of the growth program. Thus, EBNA2 expression is a specific marker for the lymphoblastoid growth program. To test if the infected IgD+ population from the tonsil bore the hallmarks of direct infection, we have screened a panel of tonsils for EBNA2 expression.
Tonsillar B cells were fractionated into IgD+ (naive) and IgD− (memory plus germinal center) B cells using biotinylated antibodies and the MACS system (Fig. 1). The cells were then split into two aliquots. The first aliquot was used to generate a limiting-dilution series for DNA PCR analysis to estimate the absolute frequency of virus-infected cells in the two populations. The second aliquot was used to isolate mRNA from the two populations of cells for cDNA preparation and PCR analysis for EBNA2 expression.
FIG. 1.
FACS analysis of tonsillar lymphocytes before and after purification of the naive (IgD+) and IgD− populations. Naive (IgD+) and IgD− B cell subsets were isolated from whole tonsils using the MACS system as described in Materials and Methods. IgD+ cells were first isolated by positive selection with biotinylated antibodies to IgD. The remaining cells were then positively selected for the IgD− B cells using a biotinylated antibody to CD19. The resulting populations were stained with a FITC-coupled antibody to the pan-B-cell marker CD20 and a PE-coupled antibody to IgD. The left-hand panel shows whole tonsillar lymphocyte staining prior to separation; the middle panel shows IgD+ B cells (96% pure); and the right panel shows IgD− B cells (97% pure).
The results of the RTPCR analysis from five such tonsils are shown in Fig. 2A, and a summary of results from 16 tonsils are given in Table 1. EBNA2 was detected in the IgD+ population (13 of 16) but never in the IgD− population (0 of 16). For each tonsil, multiple negative controls and a sensitivity control of the type shown in Fig. 3 were performed. The failure to detect EBNA2 in the IgD+ cells from 3 of the 16 tonsils could genuinely reflect a lack of EBNA2 expression; however, we suspect it is more likely a consequence of a technical failure. Although the overall quality of the cDNA in these samples appeared good, the number of infected cells is so small that relatively minor variations in the cell fractionation or cDNA synthesis protocol could cause a false-negative result. Due to limited amounts of material, it was not possible to reanalyze the cells from these three tonsils. The failure to detect EBNA2 in any of the IgD− population, however, was consistent even when sufficient cells were available to confirm the negative result. Furthermore, for tonsils for which sufficient cells were available, we have preformed spiking experiments in which 5 or 1 infected cell from an in vitro-transformed lymphoblastoid cell line was added to the purified IgD− B cells prior to mRNA extraction. An example of one such experiment is shown in Fig. 2B. This result confirms that the EBNA2 transcript was detectable in the IgD+ but not the IgD− cells. However, EBNA2 was detected when the EBV lymphoblasts were spiked into the IgD− sample.
TABLE 1.
EBNA2 latent gene expression in latently infected naive and IgD− B cells from tonsilsa
| Tonsil no. | No. of infected cells/107 B cells
|
EBNA2 expression
|
||
|---|---|---|---|---|
| IgD+ | IgD− | IgD+ | IgD− | |
| 1 | >400 | >800 | − | − |
| 2 | >200 | >550 | + | − |
| 3 | 200 | 2,300 | + | − |
| 4 | 200 | 400 | + | − |
| 5 | 200 | 333 | + | − |
| 6 | 200 | 200 | + | − |
| 7 | 150 | >800 | + | − |
| 8 | 140 | 80 | + | − |
| 9 | 140 | 70 | + | − |
| 10 | 100 | 1,000 | + | − |
| 11 | 75 | >400 | + | − |
| 12 | 75 | 330 | − | − |
| 13 | 75 | 200 | − | − |
| 14 | 50 | >400 | + | − |
| 15 | 40 | 2,600 | + | − |
| 16 | 40 | 100 | + | − |
Frequencies are expressed as the absolute number of virus-infected cells per 107 B cells in the population. The frequencies were measured as detailed previously (12). Briefly, purified cell populations were serially diluted, and multiple aliquots were made of each diluted sample. DNA PCR was then performed on each aliquot, the fraction of negative aliquots for each dilution was estimated, and Poisson statistics were used to calculate the frequencies. Since the DNA PCR will detect a single genome of viral DNA, this assay provides an absolute quantitative measure of the frequency of infected cells.
FIG. 3.
Sensitivity of the RTPCR assays used. cDNA prepared from EBV-positive cell lines was serially diluted into cDNA from EBV-negative tonsils to generate 100, 10, and 1 cell equivalents. The negative controls are B cells from EBV-negative tonsils alone. The expected positions of the PCR products are indicated by the arrows. Similar sensitivity and negative controls were performed for all experiments. An EBV-positive lymphoblastoid line was used for all RTPCRs except EBNA1 (Q-K), for which Rael cells were used. All of the assays could detect a single cell equivalent, although the EBNA3a assay was significantly less sensitive than the others. D+, surface IgD+ B-cell population; D−, surface IgD− B-cell population.
The failure to detect an EBNA2 transcript in the IgD− population was not a consequence of lower levels of virus infection in this subset. The estimated frequencies of virus-infected cells in the naive (IgD+) and IgD− populations, shown in Table 1, demonstrate that for any given tonsil there were as many, if not more, infected cells in the IgD− population as in the IgD+ population. We conclude, therefore, that EBNA2 is reproducibly expressed in the naive (IgD+) but not the IgD− population from tonsils.
Naive but not IgD− B cells from the tonsil also express the EBNA3 family of latent genes.
In addition to EBNA2, the other latent genes that are restricted in their expression to the growth program are EBNA3a, -b, and -c (reviewed in reference 27). Therefore, we analyzed EBNA2, EBNA3a, EBNA3b, and EBNA3c expression in IgD+ and IgD− cells from an additional five tonsils. The results for two representative tonsils are shown in Fig. 4, and a summary of the results for all five tonsils are presented in Table 2. We found that EBNA2, EBNA3a, EBNA3b, and EBNA3c expression was restricted to the IgD+ subset; none were found in the IgD− population (Fig. 4A). However, the EBNA3s were not detected in two of five of the tonsils tested (Fig. 4B) even though EBNA2 expression was reproducibly detected in all five. The reason for these negative results is unclear. They were not due to technical failures, since both tonsils were retested for EBNA3 expression in at least three independent experiments and were uniformly negative, while control and spiking experiments were positive, and the same cDNA preparations were positive for EBNA2. It is possible that a subset of tonsils contain infected IgD+ cells that express EBNA2 without the EBNA3s, but it is unclear why this would happen in some tonsils and not others, unless it was caused by differential and transient gene expression in subsets of IgD+ cells. This seems unlikely, given the tight linkage between expression of EBNA2 and the EBNA3s. Technical explanations seem more likely. All of the RTPCR assays we have used readily detect gene expression in a single cell infected in vitro (Fig. 3). The exception is EBNA3a, for which 10 or more cells are required to obtain a strong signal. Therefore, failure to detect EBNA3a may simply reflect the lack of sensitivity of the assay. A more likely explanation for the discrepancy between EBNA2 and EBNA3 detection is that the transcript copy number for the EBNA3s may be significantly lower in the in vivo-infected cells than in the cell lines, whereas the EBNA2 transcript level may be similar. It is also conceivable that our EBNA3 primers do not detect all of the viral isolates due to sequence variation in the EBNA3 genes.
FIG. 4.
EBNA3 proteins are expressed by latently infected naive but not IgD− B cells in the tonsil. IgD+ (naive) and IgD− B cells were isolated from tonsils using biotinylated antibodies and MACS columns as described in the legend to Fig. 1. cDNA was prepared from 5 × 106 cells of each population and divided for DNA PCR detection of EBNA2, EBNA3a, EBNA3b, and EBNA3c. The PCR products were fractionated on agarose gels, Southern blotted, and probed with sequence-specific probes. For details, see the text. The expected positions of the PCR products are indicated by the arrows. (A) Representative example of tonsils that express EBNA2 and all of the EBNA3 family. (B) Representative example of tonsils that express EBNA2 but lack detectable EBNA3s.
TABLE 2.
EBV latent gene expression in latently infected naive and IgD− B cells from tonsilsa
| Tonsil no. | Expression
|
|||||||
|---|---|---|---|---|---|---|---|---|
| EBNA2
|
EBNA3a
|
EBNA3b
|
EBNA3c
|
|||||
| IgD+ | IgD− | IgD+ | IgD− | IgD+ | IgD− | IgD+ | IgD− | |
| 3 | + | − | − | − | − | − | − | − |
| 8 | + | − | − | − | − | − | − | − |
| 9 | + | − | + | − | + | − | + | − |
| 11 | + | − | + | − | + | − | + | − |
| 15 | + | − | + | − | + | − | + | − |
Tonsil numbers are the same as in Table 1.
Combining the results in Tables 1 and 2, we found EBNA2 expressed in IgD+ B cells from 13 of 16 tonsils and in the IgD− population from 0 of 16. EBNA3a, -b, and -c were found in IgD+ cells from three of five tonsils and in the IgD− population from zero of five.
Expression of other latent genes in the IgD+ population.
The growth program is characterized by expression of the EBNA2 and EBNA3 genes. EBNA1 and the latent membrane proteins (LMPs) are also expressed, but they are not unique to the growth program. They are also present in the more restricted forms of latency found in tumors such as Hodgkin's lymphoma (10, 21, 23, 24) and nasopharyngeal carcinoma (6, 8, 41). In these tumors, EBNA2 and the EBNA3s are not expressed. The only EBNA found is EBNA1, and it is expressed from a unique promoter (Qp) (22, 30, 37) that is not used in the growth program. To assess if the overall pattern of latent gene expression in latently infected, naive B cells was consistent with the growth program, we screened IgD+ B cells from a panel of six tonsils for expression of the EBNA1 (U-K) and EBNA1 (Q-K) transcripts and the LMP1 and LMP2 genes. An example of the results obtained with three tonsils is shown in Fig. 5A, and a summary of all the results obtained is shown in Table 3. Typical sensitivity and negative controls are shown in Fig. 3. As expected, from the studies already described, EBNA2 was detected in all of the samples of IgD+ cells tested. In addition, LMP1 and LMP2 were also expressed in the IgD+ cells from six of six tonsils. We were unable to detect the presence of the Qp-derived form of the EBNA1 transcript in six of six tonsils, although EBNA1 itself was being produced, since three of three tonsils negative for EBNA1 (Q-K) were positive for EBNA1 (U-K). This RTPCR detects all of the known splice variants of the EBNA1 transcript. To test if the failure to detect EBNA1 (Q-K) was due to technical difficulties with our assay, we spiked different numbers of Rael cells into the tonsillar B-cell preparations. Rael is a Burkitt's lymphoma line that expresses the EBNA1 (Q-K) transcript. As shown in Fig. 5B, we readily detected EBNA1 (Q-K) when as few as 1 Rael cell was spiked into the tonsillar B cell. Thus, the pattern of latent genes found in the IgD+ population is EBNA1+, EBNA2+, EBNA3a+, EBNA3b+, EBNA3c+, LMP1+, and LMP2+ but EBNA1 (Q-K) negative. This is precisely the expected pattern for the growth program associated with the lymphoblastoid form of latency found in directly infected cells in vitro.
FIG. 5.
Naive (IgD+) B cells express all of the latent genes expected for the growth program. IgD+ (naive) B cells were isolated from tonsils using biotinylated antibodies and MACS columns as described in the legend to Fig. 1. cDNA was prepared from 5 × 106 cells and divided for DNA PCR detection of EBNA1 (U-K), EBNA1 (Q-K), EBNA2, LMP1, and LMP2. The PCR products were fractionated on agarose gels, Southern blotted, and probed with sequence-specific probes. For details, see the text. The expected positions of the PCR products are indicated by the arrows. (A) Results from three tonsils. (B) Spiking control, in which cells from the EBNA1 (Q-K)-positive Rael cell line were diluted into naive tonsillar B cells prior to mRNA extraction for RTPCR detection of EBNA1 (Q-K).
TABLE 3.
EBV latent gene expression in latently infected naive B cells from tonsils
| Tonsil no. | Expression
|
||||
|---|---|---|---|---|---|
| EBNA2 | LMP2a | LMP1 | EBNA1 (Q-K) | EBNA1 (U-K) | |
| 1 | + | + | + | − | + |
| 5 | + | + | + | − | + |
| 6 | + | + | + | − | + |
| 2 | + | + | + | − | NDa |
| 9 | + | + | + | − | ND |
| 16 | + | + | + | − | ND |
ND, not done.
Latently infected naive (IgD+) tonsillar B cells express an activated surface phenotype.
Resting naive B cells are negative for expression of the costimulatory molecule B7.1 (CD80). This molecule is expressed at high levels when naive B cells become activated lymphoblasts through either antigen activation or EBV infection (9). To test if EBV-infected, naive (IgD+) B cells were phenotypically activated, we isolated tonsilar lymphocytes and stained for IgD and CD80 expression. The cells were then fractionated into IgD+ CD80+ (activated naive) and IgD+ CD80− (resting naive) B cells using MoFlo FACS. An example of the staining profile and sort gates is shown in the left-hand panel of Fig. 6A, and reanalysis of the purified populations is shown in the middle and right panels. The frequency of virus-infected cells in each population was then estimated using the limiting-dilution DNA PCR assay. The PCR results for one experiment are shown in Fig. 6, and the quantitation from two such experiments is summarized in Table 4. It can readily be seen that separation on the basis of CD80 expression resulted in a marked enrichment of the virus-infected cells into the CD80+ fraction. The percentage of IgD+ cells expressing CD80 could be estimated from the FACS analysis prior to sorting. From these numbers, it is possible to backcalculate the absolute numbers of virus-infected cells residing in each fraction. For experiment number 1, >90% of the infected cells could be accounted for in the CD80+ fraction, whereas in experiment 2, approximately 80% were in the CD80+ fraction. These represent rough estimates; nevertheless, overall the experiments are consistent with most if not all of the infected IgD+ naive B cells expressing an activated cell surface phenotype.
FIG. 6.
Limiting-dilution DNA PCR analysis of activated (CD80+) and resting (CD80−) naive (IgD+) tonsillar B cells. (A) Whole tonsils were stained with a FITC-coupled antibody to CD80 and a PE-coupled antibody to IgD. The presort staining and the sort gates are shown in the dot plot to the left, taken from a Cytomation MoFlo cell sorter. Reanalysis of the isolated population is shown in the middle (93% pure CD80+ IgD+ cells) and right (99% pure CD80− IgD+ cells) panels analyzed on a FACSCalibur. (B) The cells were serially diluted, and EBV-specific DNA PCR was performed on replicates of each cell dilution. The PCR products were fractionated on an agarose gel, Southern blotted, and probed with a sequence-specific probe. The fraction of negative samples at each dilution was then estimated, and the fractions were used to calculate an absolute frequency of infected cells using Poisson statistics. Only a limited amount of the whole dilution series is shown. The expected positions of the PCR products are indicated with arrowheads.
TABLE 4.
EBV-infected naive B cells from tonsils are CD80 positivea
| Expt | No. of infected cells/107 B cells
|
||
|---|---|---|---|
| IgD+ | IgD+ CD80− | IgD+ CD80+ | |
| 1 | 75 | 4b | 400 |
| 2 | 420 | 170 | 5,600 |
Frequencies are expressed as the absolute number of virus-infected cells per 107 B cells in the population. The frequencies were measured as detailed previously (12). Briefly, purified cell populations were serially diluted, and multiple aliquots were made of each diluted sample. DNA PCR was then performed on each aliquot, the fraction of negative aliquots for each dilution was estimated, and Poisson statistics were used to calculate the frequencies. Since the DNA PCR will detect a single genome of viral DNA, this assay provides an absolute quantitative measure of the frequency of infected cells. There were 31% and 10% CD80+ cells in the IgD+ fraction for experiments 1 and 2, respectively.
Based on a single positive cell among 2.2 × 106 cells tested.
EBV-infected naive (IgD+) B cells are not always detected in EBV-positive tonsils.
We have developed a quantitative, limiting-dilution assay that allows the measurement of absolute numbers of EBV-infected cells within a given population of cells (12, 20). With this assay, we have shown previously that peripheral blood, a site of persistent infection, contains infected IgD− B cells but completely lacks infected IgD+ (naive) B cells (4). In comparison, tonsils contain both infected IgD+ and IgD− B cells. We noted, however, that there were rare tonsils that lacked infected IgD+ cells. Infected IgD− B cells were found in 100% of the samples tested for both (59 of 59) tonsils and peripheral blood (16 of 16) (Table 5). However, infected IgD+ (naive) cells were only found in 90% of the tonsils (52 of 59) and were never found in the peripheral blood (0 of 16). None of the tonsils contained infected IgD+ B cells in the absence of infected IgD− B cells. In an analysis of a limited number of tonsils, we showed previously (4) that five of five tonsils with infected IgD+ B cells contained linear viral DNA characteristic of infectious virus, whereas two of two tonsils that lacked infected IgD+ cells lacked linear DNA. We interpret these results to mean that the virus persists within IgD− B cells throughout the lymphoid system, but the IgD+ cells are being directly infected by the virus in the tonsils and are either killed or differentiate into IgD− B cells before they can leave the tonsil.
TABLE 5.
Tonsils always contain infected IgD− B cells, but some tonsils lack infected naive (IgD+) B cellsa
| Infected population | Frequency of infection as assayed by limiting dilution DNA PCRb
|
|
|---|---|---|
| Tonsil | Peripheral blood | |
| IgD− | 59/59 | 16/16 |
| IgD+ | 52/59 | 0/16 |
The data include results from five tonsil and five peripheral blood samples that were previously published (4). The additional data have not been previously published.
Only the results from EBV-positive tonsils are included.
DISCUSSION
In this paper, we report that the tonsillar lymph nodes of healthy carriers of EBV contain infected lymphoblastoid cells that express the growth latency program. We further show that these cells are all IgD+ (naive) B cells. In a previous study, we demonstrated a direct correlation between the detection of these infected IgD+ cells and the presence of viral replication. Since the naive, infected lymphoblasts are not found when infectious virus is absent (4), we conclude that they are not a self-sustaining compartment of viral persistence. Instead, they are being continuously generated by direct infection with the virus and then removed, either through differentiation (see below) or through the actions of CTL.
The lymphoblastoid growth program has already been characterized extensively because it is the program used when the virus establishes latency in vitro. However, lymphoblastoid cells expressing the growth latency program have only been detected before in vivo during acute infectious mononucleosis (36). These cells stimulate a rapid and potent CTL response that is particularly focused on the EBNA2 and EBNA3 latent proteins (13), which are uniquely expressed by the growth program. The CTL response is thought to eliminate infected cells, driven to proliferate by the growth program, before they develop into a pathogenic threat. This threat is revealed in immunosuppressed individuals who are at risk for developing life-threatening lymphomas derived from B cells expressing the growth program (31, 40).
The oncogenic risk to the host posed by proliferating, EBV-infected lymphoblasts has led to the suggestion that the EBV-encoded growth program is essential for the establishment of latency during acute infection, before the CTL response arises. Thereafter, the proliferating lymphoblasts are killed as soon as they are produced and play no role in the maintenance of persistent latent infection. This view now appears too simple. The frequency of infected IgD+ cells in the tonsil is comparable to that of infected IgD− cells in the periphery—a population that is presumably under less immunosurveillance because the major CTL target antigens, EBNA2 and the EBNA3s, are not expressed (5, 26, 36). It appears, therefore, that the CTL response is not particularly effective in eliminating the naive, infected lymphoblasts. Furthermore, it is noteworthy that the lymphomas that arise in allograft patients, in whom the CTL response is suppressed iatrogenically, are not IgD+ (11). This implies that CTL are not required to regulate the IgD+ cells expressing the growth program in the tonsil.
The key to understanding why latently infected, naive B cells in the tonsil express the growth program but may not require immunosurveillance lies, we believe, in the specificity of the infection process for the naive compartment and in our model of EBV persistence. It is well known that EBV has no specificity for naive B cells in vitro; it can equally well infect and growth-transform memory B cells. Yet we have shown in this study that only naive IgD+ B cells are being infected by EBV and driven to express the growth program. The explanation must be that infectious virus is restricted to regions of the tonsil that contain only naive B cells. The only region of the tonsil that is highly enriched for naive IgD+ B cells and lacks other B cell types is the mantle zone that surrounds the follicles (16, 25). It would be a reasonable scenario to suggest that latently infected memory B cells from the periphery extravasate in the marginal zones of the tonsil and reactivate the virus in response to signals that they receive as they migrate to the mantle zone. Thus, infectious virus would only be produced in regions where IgD+, naive B cells were present.
We have proposed that EBV-infected, naive B cells in vivo are able to recapitulate B-cell differentiation driven by antigen (4, 32, 33). They do not remain lymphoblasts, but rather differentiate through a germinal center-type reaction (16, 17) to enter the peripheral B-cell pool as resting memory cells. In this scenario, expression of the growth program in a naive B cell in a lymph node would be transient and not a pathogenic threat even to the immunosuppressed host. If true, then the CTL exist not to kill infected naive cells in the mantle zone, but to kill cells that express the growth program in inappropriate locations where they cannot differentiate into a resting memory cell. This would include all nonnaive B cells in lymph nodes and all extranodal B cells that would not have access to follicles to allow differentiation. These ideas are especially interesting in light of recent studies from acute infectious mononucleosis patients (15a). These authors microdissected single cells from the tonsils of infectious mononucleosis patients. They observed that the proliferating EBNA2+ clones in the tonsils are memory cells. Since we have never detected memory lymphoblastoid cells in the tonsils of healthy carriers, we assume that the situation in infectious mononucleosis is a deregulated one. Here the virus has escaped the usual anatomical confines that allow only infection of naive B cells and has infected memory cells. These cells grow out of control because they cannot differentiate. Eventually these clones of memory cells are destroyed when the CTL response arises. The situation in infectious mononucleosis is so atypical that proliferating, infected lymphoblastoid cells expressing the growth program are even present in the peripheral blood (36). Such cells have never been found in the blood of healthy carriers of the virus (5, 26, 36) even when immunosuppressed (3). It is our presumption that it is the role of the CTL to clear these types of aberrant and uncontrolled infections of B cells in the lymph nodes and in the blood. The infection of naive B cells in the tonsils is, by comparison, regulated by differentiation and may not require surveillance by CTL.
One caveat to our experiments is that we do not know what fraction of the infected IgD+ naive cells are actually expressing the growth program. In previous studies, we have shown that all of the EBV-positive naive cells are latently infected (7), but it is conceivable that a large fraction of the infected cells are resting and expressing none of the growth-promoting latent genes. However, it is difficult to conceive how this could occur. When EBV infects B cells, they automatically go to the growth program and start to proliferate; there is no evidence that other forms of latency are adopted. Such infected, naive B cells in vivo would then have to leave the cell cycle and be maintained as resting naive B cells that stay in the tonsil, because there are no infected naive B cells in the peripheral blood. Such behavior has never been documented for a normal naive B cell. Activated naive B cells can only exit the cell cycle through differentiation or death.
In conclusion, cells in the tonsil expressing the growth program, characteristic of direct infection, are restricted to the naive subset. It will be important now to characterize the type and location of cells that replicate the virus in the tonsils of healthy carriers and identify the signals that are required in vivo to initiate viral replication.
ACKNOWLEDGMENTS
We thank Allen Parmalee for excellent flow cytometry and Cheryl Greene for providing the tonsils.
The authors' work is supported by Public Health Service grants AI 18757 and CA 65883.
REFERENCES
- 1.Allday M J, Crawford D H, Griffin B E. Epstein-Barr virus latent gene expression during the initiation of B cell immortalization. J Gen Virol. 1989;70:1755–1764. doi: 10.1099/0022-1317-70-7-1755. [DOI] [PubMed] [Google Scholar]
- 2.Anagnostopoulos I, Hummel M, Kreschel C, Stein H. Morphology, immunophenotype, and distribution of latently and/or productively Epstein-Barr virus-infected cells in acute infectious mononucleosis: implications for the interindividual infection route of Epstein-Barr virus. Blood. 1995;85:744–750. [PubMed] [Google Scholar]
- 3.Babcock G J, Decker L L, Freeman R B, Thorley-Lawson D A. Epstein-Barr virus-infected resting memory B cells, not proliferating lymphoblasts, accumulate in the peripheral blood of immunosuppressed patients. J Exp Med. 1999;190:567–576. doi: 10.1084/jem.190.4.567. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Babcock G J, Decker L L, Volk M, Thorley-Lawson D A. EBV persistence in memory B cells in vivo. Immunity. 1998;9:395–404. doi: 10.1016/s1074-7613(00)80622-6. [DOI] [PubMed] [Google Scholar]
- 5.Chen F, Zou J Z, DiRenzo R L, Winberg G, Hu L F, Klein E, Klein G, Ernberg I. A subpopulation of normal B cells latently infected with Epstein-Barr virus resembles Burkitt lymphoma cells in expressing EBNA-1 but not EBNA-2 or LMP1. J Virol. 1995;69:3752–3758. doi: 10.1128/jvi.69.6.3752-3758.1995. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Deacon E M, Pallesen G, Niedobitek G, Crocker J, Brooks L, Rickinson A B, Young L S. Epstein-Barr virus and Hodgkin's disease: transcriptional analysis of virus latency in the malignant cells. J Exp Med. 1993;177:339–349. doi: 10.1084/jem.177.2.339. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Decker L L, Klaman L D, Thorley-Lawson D A. Detection of the latent form of Epstein-Barr virus DNA in the peripheral blood of healthy individuals. J Virol. 1996;70:3286–3289. doi: 10.1128/jvi.70.5.3286-3289.1996. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Fahraeus R, Fu H L, Ernberg I, Finke J, Rowe M, Klein G, Falk K, Nilsson E, Yadav M, Busson P, et al. Expression of Epstein-Barr virus-encoded proteins in nasopharyngeal carcinoma. Int J Cancer. 1988;42:329–338. doi: 10.1002/ijc.2910420305. [DOI] [PubMed] [Google Scholar]
- 9.Freedman A S, Freeman G, Horowitz J C, Daley J, Nadler L M. B7, a B-cell-restricted antigen that identifies preactivated B cells. J Immunol. 1987;139:3260–3267. [PubMed] [Google Scholar]
- 10.Herbst H, Dallenbach F, Hummel M, Niedobitek G, Pileri S, Muller L N, Stein H. Epstein-Barr virus latent membrane protein expression in Hodgkin and Reed-Sternberg cells. Proc Natl Acad Sci USA. 1991;88:4766–4770. doi: 10.1073/pnas.88.11.4766. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Kaplan M A, Ferry J A, Harris N L, Jacobson J O. Clonal analysis of posttransplant lymphoproliferative disorders, using both episomal Epstein-Barr virus and immunoglobulin genes as markers. Am J Clin Pathol. 1994;101:590–596. doi: 10.1093/ajcp/101.5.590. [DOI] [PubMed] [Google Scholar]
- 12.Khan G, Miyashita E M, Yang B, Babcock G J, Thorley-Lawson D A. Is EBV persistence in vivo a model for B cell homeostasis? Immunity. 1996;5:173–179. doi: 10.1016/s1074-7613(00)80493-8. [DOI] [PubMed] [Google Scholar]
- 13.Khanna R, Burrows S R, Moss D J. Immune regulation in Epstein-Barr virus-associated diseases. Microbiol Rev. 1995;59:387–405. doi: 10.1128/mr.59.3.387-405.1995. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Kieff E. Epstein-Barr virus and its replication. In: Fields B N, Knipe D M, Howley P M, editors. Fields virology. 2nd ed. Vol. 2. New York, N.Y: Raven Press; 1996. pp. 2343–2396. [Google Scholar]
- 15.Klein G. Epstein-Barr virus strategy in normal and neoplastic B-cells. Cell. 1994;77:791–793. doi: 10.1016/0092-8674(94)90125-2. [DOI] [PubMed] [Google Scholar]
- 15a.Kurth, J., T. Spieker, J. Wustrow, J. G. Strickler, M. L. Hansmann, K. Rajewsky, and R. Küppers. Tracing EBV-infected B cells in infectious mononucleosis: viral strategies for spreading in the B cell compartment and establishment of latent infection. Immunity, in press. [DOI] [PubMed]
- 16.Liu Y J, Arpin C. Germinal center development. Immunol Rev. 1997;156:111–126. doi: 10.1111/j.1600-065x.1997.tb00963.x. [DOI] [PubMed] [Google Scholar]
- 17.MacLennan I C. Germinal centers. Annu Rev Immunol. 1994;12:117–139. doi: 10.1146/annurev.iy.12.040194.001001. [DOI] [PubMed] [Google Scholar]
- 18.Masucci M G, Ernberg I. Epstein-Barr virus: adaptation to a life within the immune system. Trends Microbiol. 1994;2:125–130. doi: 10.1016/0966-842x(94)90599-1. [DOI] [PubMed] [Google Scholar]
- 19.Miyashita E M, Yang B, Babcock G J, Thorley-Lawson D A. Identification of the site of Epstein-Barr virus persistence in vivo as a resting B cell. J Virol. 1997;71:4882–4891. doi: 10.1128/jvi.71.7.4882-4891.1997. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Miyashita E M, Yang B, Lam K M, Crawford D H, Thorley-Lawson D A. A novel form of Epstein-Barr virus latency in normal B cells in vivo. Cell. 1995;80:593–601. doi: 10.1016/0092-8674(95)90513-8. [DOI] [PubMed] [Google Scholar]
- 21.Niedobitek G, Kremmer E, Herbst H, Whitehead L, Dawson C W, Niedobitek E, von Ostau C, Rooney N, Grasser F A, Young L S. Immunohistochemical detection of the Epstein-Barr virus-encoded latent membrane protein 2A in Hodgkin's disease and infectious mononucleosis. Blood. 1997;90:1664–1672. [PubMed] [Google Scholar]
- 22.Nonkwelo C, Skinner J, Bell A, Rickinson A, Sample J. Transcription start sites downstream of the Epstein-Barr virus (EBV) Fp promoter in early-passage Burkitt lymphoma cells define a fourth promoter for expression of the EBV EBNA-1 protein. J Virol. 1996;70:623–627. doi: 10.1128/jvi.70.1.623-627.1996. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Oudejans J J, Dukers D F, Jiwa N M, van den Brule A J, Grasser F A, de Bruin P C, Horstman A, Vos W, van Gorp J, Middeldorp J M, Meijer C J. Expression of Epstein-Barr virus encoded nuclear antigen 1 in benign and malignant tissues harbouring EBV. J Clin Pathol. 1996;49:897–902. doi: 10.1136/jcp.49.11.897. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Pallesen G, Hamilton-Dutoit S J, Rowe M, Young L S. Expression of Epstein-Barr virus latent gene products in tumour cells of Hodgkin's disease. Lancet. 1991;337:320–322. doi: 10.1016/0140-6736(91)90943-j. [DOI] [PubMed] [Google Scholar]
- 25.Perry M, Whyte A. Immunology of the tonsils. Immunol Today. 1998;19:414–421. doi: 10.1016/s0167-5699(98)01307-3. [DOI] [PubMed] [Google Scholar]
- 26.Qu L, Rowe D T. Epstein-Barr virus latent gene expression in uncultured peripheral blood lymphocytes. J Virol. 1992;66:3715–3724. doi: 10.1128/jvi.66.6.3715-3724.1992. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Rickinson A B, Kieff E. Epstein-Barr virus. In: Fields B N, Knipe D M, Howley P M, editors. Fields virology. 3rd ed. Vol. 2. New York, N.Y: Raven Press; 1996. pp. 2397–2446. [Google Scholar]
- 28.Rickinson A B, Yao Q Y, Wallace L E. The Epstein-Barr virus as a model of virus-host interactions. Br Med Bull. 1985;41:75–79. doi: 10.1093/oxfordjournals.bmb.a072030. [DOI] [PubMed] [Google Scholar]
- 29.Robertson E, Kieff E. Reducing the complexity of the transforming Epstein-Barr virus genome to 64 kilobase pairs. J Virol. 1995;69:983–993. doi: 10.1128/jvi.69.2.983-993.1995. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Schaefer B C, Strominger J L, Speck S H. Redefining the Epstein-Barr virus-encoded nuclear antigen EBNA-1 gene promoter and transcription initiation site in group I Burkitt lymphoma cell lines. Proc Natl Acad Sci USA. 1995;92:10565–10569. doi: 10.1073/pnas.92.23.10565. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Thomas J A, Hotchin N A, Allday M J, Amlot P, Rose M, Yacoub M, Crawford D H. Immunohistology of Epstein-Barr virus-associated antigens in B cell disorders from immunocompromised individuals. Transplantation. 1990;49:944–953. doi: 10.1097/00007890-199005000-00022. [DOI] [PubMed] [Google Scholar]
- 32.Thorley-Lawson D A. EBV—schizophrenic but not dysfunctional. Epstein-Barr Virus Rep. 1999;6:158–163. [Google Scholar]
- 33.Thorley-Lawson D A, Babcock G J. A model for persistent infection with Epstein-Barr virus: the stealth virus of human B cells. Life Sci. 1999;65:1433–1453. doi: 10.1016/s0024-3205(99)00214-3. [DOI] [PubMed] [Google Scholar]
- 34.Thorley-Lawson D A, Mann K P. Early events in Epstein-Barr virus infection provide a model for B cell activation. J Exp Med. 1985;162:45–59. doi: 10.1084/jem.162.1.45. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Thorley-Lawson D A, Miyashita E M. Epstein-Barr virus and the B cell: that's all it takes. Trends Microbiol. 1996;4:204–208. doi: 10.1016/s0966-842x(96)90020-7. [DOI] [PubMed] [Google Scholar]
- 36.Tierney R J, Steven N, Young L S, Rickinson A B. Epstein-Barr virus latency in blood mononuclear cells: analysis of viral gene transcription during primary infection and in the carrier state. J Virol. 1994;68:7374–7385. doi: 10.1128/jvi.68.11.7374-7385.1994. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Tsai C N, Liu S T, Chang Y S. Identification of a novel promoter located within the Bam HI Q region of the Epstein-Barr virus genome for the EBNA 1 gene. DNA Cell Biol. 1995;14:767–776. doi: 10.1089/dna.1995.14.767. [DOI] [PubMed] [Google Scholar]
- 38.Wang F, Tsang S F, Kurilla M G, Cohen J I, Kieff E. Epstein-Barr virus nuclear antigen 2 transactivates latent membrane protein LMP1. J Virol. 1990;64:3407–3416. doi: 10.1128/jvi.64.7.3407-3416.1990. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Woisetschlaeger M, Jin X W, Yandava C N, Furmanski L A, Strominger J L, Speck S H. Role for the Epstein-Barr virus nuclear antigen 2 in viral promoter switching during initial stages of infection. Proc Natl Acad Sci USA. 1991;88:3942–3946. doi: 10.1073/pnas.88.9.3942. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40.Young L, Alfieri C, Hennessy K, Evans H, O'Hara C, Anderson K C, Ritz J, Shapiro R S, Rickinson A, Kieff E, et al. Expression of Epstein-Barr virus transformation-associated genes in tissues of patients with EBV lymphoproliferative disease. N Engl J Med. 1989;321:1080–1085. doi: 10.1056/NEJM198910193211604. [DOI] [PubMed] [Google Scholar]
- 41.Young L S, Dawson C W, Clark D, Rupani H, Busson P, Tursz T, Johnson A, Rickinson A B. Epstein-Barr virus gene expression in nasopharyngeal carcinoma. J Gen Virol. 1988;69:1051–1065. doi: 10.1099/0022-1317-69-5-1051. [DOI] [PubMed] [Google Scholar]
- 42.Zimber-Strobl U, Suentzenich K O, Laux G, Eick D, Cordier M, Calender A, Billaud M, Lenoir G M, Bornkamm G W. Epstein-Barr virus nuclear antigen 2 activates transcription of the terminal protein gene. J Virol. 1991;65:415–423. doi: 10.1128/jvi.65.1.415-423.1991. [DOI] [PMC free article] [PubMed] [Google Scholar]