Previous Article | Next Article ![]()
Journal of Virology, April 2005, p. 5227-5231, Vol. 79, No. 8
0022-538X/05/$08.00+0 doi:10.1128/JVI.79.8.5227-5231.2005
Copyright © 2005, American Society for Microbiology. All Rights Reserved.
Center for Emerging Infectious Diseases, Yerkes National Primate Research Center, Emory University School of Medicine, Atlanta, Georgia,1 Graduate Program in Molecular Cell Biology, Washington University School of Medicine, St. Louis, Missouri2
Received 10 September 2004/ Accepted 23 November 2004
|
|
|---|
HV68) infection of mice results in the establishment of a chronic infection, which is largely maintained through latent infection of B lymphocytes. Acute virus replication is almost entirely cleared by 2 weeks postinfection. Spontaneous reactivation of
HV68 from latently infected splenocytes upon ex vivo culture can readily be detected at the early stages of infection (e.g., day 16). However, by 6 weeks postinfection, very little spontaneous reactivation is detected upon explant into tissue culture. Here we report that stimulation of latently infected splenic B cells harvested at late times postinfection with cross-linking surface immunoglobulin (Ig), in conjunction with anti-CD40 antibody treatment, triggers virus reactivation. As expected, this treatment resulted in B-cell activation, as assessed by upregulation of CD69 on B cells, and ultimately B-cell proliferation. Since anti-Ig/anti-CD40 stimulation resulted in splenic B-cell proliferation, we assessed whether this reactivation stimulus could overcome the previously characterized defect in virus reactivation of a v-cyclin null
HV68 mutant. This analysis demonstrated that anti-Ig/anti-CD40 stimulation could drive reactivation of the v-cyclin null mutant virus in latently infected splenocytes, but not to the levels observed with wild-type
HV68. Thus, there appears to be a role for the v-cyclin in B cells following anti-Ig/anti-CD40 stimulation independent of the induction of the cell cycle. Finally, to assess signals that are not mediated through the B-cell receptor, we demonstrate that addition of lipopolysaccharide to explanted splenocyte cultures also enhanced virus reactivation. These studies complement and extend previous analyses of Epstein-Barr virus and Kaposi's sarcoma-associated virus reactivation from latently infected cell lines by investigating reactivation of
HV68 from latently infected primary B cells recovered from infected hosts. Background. Gammaherpesviruses are characterized by their ability to establish latency in lymphocytes. The establishment and maintenance of latency is important for persistence in the infected host, while virus reactivation is needed for transmission of the virus to new hosts and may also be required to maintain reservoirs of latently infected cells in the chronically infected host (2, 12, 15, 33, 34). The switch between latency and the lytic cycle for the human gammaherpesviruses, Epstein-Barr virus (EBV) and Kaposi's sarcoma-associated herpesvirus (KSHV), has been extensively characterized in established latently infected cell lines in vitro (11, 27, 42). Initiation of the EBV lytic cycle can be stimulated by several different reagents, including anti-immunoglobulin (anti-Ig), calcium ionophore, sodium butyrate, and the phorbol ester tetradecanoyl phorbol acetate (TPA) (27). KSHV reactivation can also be induced by stimulation with phorbol esters and sodium butyrate (3, 19).
Murine gammaherpesvirus 68 (
HV68) is closely related to EBV and KSHV, and infection of mice with
HV68 provides a tractable animal model in which to study the pathobiology of gammaherpesviruses in vivo (28). Ectopic expression of the
HV68 ORF50 gene product, Rta, in a cell line (S11) latently infected with
HV68 has been shown to drive expression of both early and late viral genes and production of viral particles (36, 44). However, the initial stimulus required for the expression of Rta, leading to the switch from latent to lytic infection, both in vitro and in vivo is still not known.
B-cell stimulation drives virus reactivation from the S11 lymphoma cell line latently infected with
HV68.
Initially, we addressed whether stimuli that drive EBV reactivation from latently infected B-lymphoblastoid and Burkitt's lymphoma cell lines also trigger
HV68 reactivation, using the B-lymphoma cell line S11 latently infected with
HV68. S11 cells were treated with 50 ng of TPA in dimethyl sulfoxide, with F(ab')2 fragment goat anti-mouse IgM-IgG (Jackson ImmunoResearch, West Grove, Pa.) at 5 µg/ml and anti-mouse CD40 (BD Biosciences Pharmingen, San Diego, Calif.) at 2.5 µg/ml, or with dimethyl sulfoxide alone as a control. The supernatants from untreated and treated S11 cells were recovered at 48 h posttreatment and plated onto murine embryonic fibroblasts (MEFs) to score for the presence of virus as determined by cytopathic effect (CPE) (Fig. 1). To control for possible toxic effects of these treatments on the MEF monolayers, which might score as CPE, a parallel series of treatments of
HV68-negative murine B-cell line A20 cells was carried out and supernatants from these cells were plated onto MEF monolayers. Notably, no CPE was detected with A20 supernatants for any of the treatment conditions (data not shown). However, stimulation of S11 cells with TPA or cross-linking surface Ig (with or without anti-CD40) resulted in virus reactivation (Fig. 1). We observed the best stimulation of virus reactivation with a combination of anti-Ig and anti-CD40. This is notable since anti-CD40 stimulation of cells latently infected with EBV inhibits virus reactivation (9, 21).
![]() View larger version (40K): [in a new window] |
FIG. 1. Induction of HV68 reactivation from the S11 lymphoma cell line. S11 cells were treated with the indicated reagents as described in the text. Tissue culture supernatant harvested from the cells at 48 h posttreatment was plated onto confluent monolayers of MEFs, and the appearance of viral CPE was monitored (24). Twenty-four wells per supernatant dilution were analyzed. Data shown are representative of results from three independent experiments. No Tx, no treatment; DMSO, dimethyl sulfoxide.
|
HV68-infected mice.
Acute
HV68 infection is cleared from the spleen by days 9 to 15 post-intranasal infection of a mouse (30, 39). Early and late
HV68 latency was analyzed at days 16 and 42 postinfection, respectively, to compare the differences between the establishment and maintenance of latency (35, 43) (Fig. 2A and B). Latency was monitored by measuring the frequency of cells reactivating from latency ex vivo by employing a limiting dilution reactivation assay and determining the frequency of cells containing viral genome as detected by limiting dilution nested PCR analysis, as previously described (39, 40). Following intranasal inoculation of 1,000 PFU of
HV68 into C57BL/6 mice, the frequency of spleen cells that could reactivate virus ex vivo was approximately 1 in 7,000 at day 16 postinfection (Fig. 2A) and the frequency of spleen cells harboring viral genome was approximately 1 in 400 (Fig. 2B). At day 42 postinfection, the frequency of spleen cells harboring viral genome was 1 in 4,000 (Fig. 2A) but the level of reactivation from spleen cells was too low to be accurately estimated using our standard reactivation assay protocol (Fig. 2B). This dramatic drop in the frequency of splenocytes which reactivate virus upon explant into tissue culture has been shown previously (35). The discrepancy between the frequency of latently infected cells and the ability of the virus to reactivate from these cells is greater at later time points postinfection (e.g., day 42) than at earlier time points (e.g., day 16).
![]() View larger version (33K): [in a new window] |
FIG. 2. Stimulation of HV68 reactivation from explanted splenocytes with anti-Ig and anti-CD40 treatment. (A and B) Bulk spleen cells were obtained from HV68-infected C57BL/6J mice at days 16 and 42 postinfection and analyzed by limiting dilution viral genome PCR (A) or limiting dilution ex vivo reactivation (B) as previously described (39-41). (C to E) Bulk spleen cells were analyzed for the frequency of cells reactivating virus with and without anti-IgM-IgG and anti-CD40 stimulation following intranasal inoculation at day 16 (C), day 42 (D), and 3 or 6 months (E) postinfection. For all assays, curve fit lines were derived from nonlinear regression analysis, and symbols represent mean percentages of wells positive for virus (viral DNA or CPE) ± the standard errors of the mean. The dashed lines are at 63.2%, from which the frequencies of viral genome-positive cells or the frequencies of reactivating virus were calculated based on the Poisson distribution. The data shown represent results from at least three independent experiments, with cells pooled from five mice per experimental group.
|
HV68. Recovered splenocytes were plated for the ex vivo reactivation assay with either medium alone or with the addition of anti-IgM-IgG and anti-CD40 as described above. Notably, no CPE was observed on MEF monolayers when the monolayers were incubated with the antibodies alone or with the antibodies plus naïve spleen cells (data not shown). B-cell stimulation led to a modest increase (ca. threefold) in the frequency of splenocytes reactivating virus at day 16 postinfection (Fig. 2C). At day 42 postinfection, stimulation with anti-IgM-IgG and anti-CD40 led to a substantial increase in the frequency of cells reactivating virus, from an estimated 1 in 106 to 1 in 80,000 cells (Fig. 2D).
Splenocytes harboring
HV68 are still readily detectable at late times postinfection (approximately 1 in 20,000 at 6 months postinfection) and at a much higher frequency in sorted memory B cells in the spleen (1 in 2,000 at 6 months postinfection) (43). To increase our ability to detect reactivation at 3 and 6 months postinfection, it was necessary to scale up the reactivation assay by plating 10-fold more cells per well in a proportionately larger surface area (Fig. 2E). Using this scaled-up ex vivo reactivation assay, we were able at 3 months postinfection to detect reactivation of virus by ca. 1 in 150,000 spleen cells when the cells were stimulated with anti-Ig and anti-CD40, and 1 in 300,000 bulk spleen cells reactivated virus when the cells were stimulated at 6 months postinfection (Fig. 2E). In comparison, approximately 1 in 17,000 and 1 in 30,000 cells, respectively, harbored viral genome at 3 and 6 months postinfection as determined by limiting dilution PCR. In the absence of B-cell stimulation, fewer than 1 in 106 cells spontaneously reactivated virus (Fig. 2E). Thus, B-cell stimulation is able to drive ca. 10% of viral genome-positive splenocytes to reactivate virus, whereas in the absence of stimulation,
1% of the viral genome-positive splenocytes spontaneously reactivate virus upon explant into tissue culture.
B-cell stimulation does not completely overcome the reactivation defect observed with
HV68 lacking a functional v-cyclin gene.
To demonstrate that the anti-IgM-IgG and anti-CD40 stimulation protocol we employed drives B-cell activation in culture, we assessed cellular proliferation as indicated by the loss of carboxyfluorescein diacetate succinimidyl ester (CFSE) staining (14) and cellular activation as indicated by the induction of CD69, a marker of early activation on B and T cells. We labeled 106 spleen cells harvested at day 42 postinfection with 1 µM CFSE. The dye-labeled cells were then stimulated and analyzed by flow cytometry for loss of CFSE staining at several time points poststimulation (Fig. 3A). At 24 h poststimulation, there was very little evidence of B-cell proliferation (Fig. 3A, upper panel). However, within 24 h poststimulation, the majority of the B cells were CD69hi compared to unstimulated control cells, which remained CD69lo (Fig. 3B). By 120 h poststimulation, all the B cells present had proliferated (Fig. 3A, lower panel). These data indicate that the anti-IgM-IgG and anti-CD40 stimulation protocol described here is effective in activating B cells and driving them to proliferate in vitro.
![]() View larger version (19K): [in a new window] |
FIG. 3. Effects of ex vivo stimulation on splenic B cells. (A) Bulk spleen cells from day-42-postinfection mice were labeled with CFSE and stimulated with anti-IgG-IgM and anti-CD40. At several time points poststimulation, the cells were stained with anti-CD19 monoclonal antibody and analyzed by flow cytometry for loss of CFSE. Plots were gated on CD19+ cells. (B) Bulk spleen cells from day-42-postinfection mice were either stimulated with anti-IgG-IgM and anti-CD40 or left unstimulated in medium for 24 h. The spleen cells were then stained with anti-CD69 and anti-CD19 and analyzed by flow cytometry. The plot was gated on CD19+ B cells. The black line represents the unstimulated B-cell population, and the light gray line represents the stimulated B-cell population. Results shown are representative of results from two independent experiments.
|
HV68 is important for efficient reactivation from latency (13, 38). It has been hypothesized that the ability of the v-cyclin to drive cells into the cell cycle is important in virus reactivation (20, 37). Since anti-Ig/anti-CD40 is able to drive B cells into the cell cycle, we assessed whether this treatment would overcome the reactivation defect observed with the
HV68-v-cyclin.stop mutant. Since the previously published data on the v-cyclin.stop virus were collected following intraperitoneal inoculation (38), we examined splenocytes from mice harvested at day 42 post-intraperitoneal inoculation. As expected, with wild-type
HV68 we observed an increase in the frequency of
HV68-infected splenocytes reactivating virus upon stimulation with anti-IgG-IgM and anti-CD40 (Fig. 4A). In addition, we observed an increase in the frequency of splenocytes reactivating the v-cyclin.stop mutant virus upon stimulation (Fig. 4B). However, anti-Ig/anti-CD40 stimulation did not restore the frequency of v-cyclin.stop-reactivating cells to that of wild-type virus-reactivating cells. Importantly, when we assessed the frequency of latently infected cells, we found that there was no difference in the efficiencies of the establishment of latency between wild-type and v-cyclin.stop viruses (Fig. 4C). Thus, taken together, these findings indicate that in addition to inducing the cell cycle the v-cyclin likely contributes another function(s) that plays a role in virus reactivation.
![]() View larger version (31K): [in a new window] |
FIG. 4. Effects of stimulation on spleen cells infected with v-cyclin mutant virus, and LPS stimulation of splenocytes infected with wild-type (WT) HV68. (A to C) Bulk spleen cells were harvested at day 42 postinfection from mice inoculated via intraperitoneal injection with either wild-type HV68 (A and C) or v-cyclin mutant virus (B and C) (5, 38). Splenocytes were analyzed for the frequency of cells reactivating virus with and without anti-IgG-IgM and anti-CD40 stimulation (A and B) and the frequency of cells harboring viral genome (C). (D) Bulk spleen cells were harvested at 42 days postinfection from mice inoculated intranasally and were analyzed for the frequency of cells reactivating virus following no stimulation or stimulation with LPS (Sigma-Aldrich) at a final concentration of 25 µg/ml. For all assays shown, curve fit lines were derived from nonlinear regression analysis. Symbols represent the mean percentages of positive wells, with the standard errors of the mean shown. Frequency analysis was based on Poisson distribution, with the horizontal dashed lines at 63.2%. The data shown represent results from two to three independent experiments, with cells pooled from five mice per experimental group.
|
HV68 reactivation.
To assess whether
HV68 reactivation requires signals mediated through the B-cell receptor, or that mimic B-cell receptor signaling, we assessed a B-cell activation stimulus that is B-cell receptor independent. Lipopolysaccharide (LPS) stimulation of B cells is mediated through Toll-like receptor 4. Splenocytes recovered from mice at day 42 postinfection were plated for the standard reactivation assay with medium alone or were stimulated with LPS. Importantly, LPS alone or LPS stimulation of naïve spleen cells did not cause any detectable CPE on MEF monolayers (data not shown). LPS stimulation of splenocytes resulted in a significant increase in the frequency of cells reactivating virus from latency (1 in 56,000 cells) (Fig. 4E). Notably, the frequency of day-42-postinfection splenocytes reactivating virus in response to LPS was very similar to that observed when the anti-IgM-IgG-plus-anti-CD40 stimulation protocol was used (Fig. 2D).
We did not observe any significant difference in the levels of reactivation between spleen cells stimulated with anti-Ig/anti-CD40 and those stimulated with LPS at day 42 postinfection. Both treatments led to upregulation of CD69 within 24 h, as well as similar levels of B-cell proliferation, as determined by loss of CFSE over several days following stimulation in vitro (data not shown). Although the immediate downstream pathway following LPS stimulation of B cells is different from that following antibody stimulation, both pathways eventually activate similar transcription factors further downstream, notably NF-
B and AP-1 (32). Since the stimulation pathways lead to the activation of similar downstream transcription factors, one cannot clearly decipher what cellular factors and their expression patterns are necessary to drive reactivation of
HV68. Further characterization of the role that specific cellular transcription factors, such as NF-
B and AP-1, play in establishment of and reactivation from latency is needed. However, it is worth noting that activation of NF-
B has been shown to inhibit
HV68, EBV, and KSHV reactivation in tissue culture models (1). Thus, it seems unlikely that activation of NF-
B, either through B-cell receptor stimulation or signaling through Toll-like receptor 4, plays a role in the observed induction of
HV68 reactivation.
Conclusions.
Here we have shown that reactivation of
HV68 ex vivo can be stimulated by treatment of latently infected splenocytes with either anti-IgG-IgM and anti-CD40 or LPS. A number of different reagents have been shown to disrupt EBV or KSHV latency in vitro, including phorbol esters, sodium butyrate, and cross-linking surface Ig (19, 27, 31). Synchronous anti-Ig stimulation of several lymphoma-derived cell lines latently infected with EBV leads to the transcription of the BZLF1 gene, which encodes Zta (6, 8, 26, 27). Zta and another immediate-early viral gene product, Rta, are the major transactivators of EBV lytic gene expression. Both Zta and Rta have been shown to drive lytic replication of EBV following transfection of latently infected cell lines (16, 25, 45). Together they act in a synergistic manner to promote early lytic viral gene expression, and their expression has been shown to be critical in the disruption of latency and the induction of the viral lytic cycle (17). Overexpression of the KSHV Rta gene has also been shown to be capable of initiating expression of lytic genes in B-cell lines latently infected with KSHV, although the KSHV homologue of Zta is unable to induce lytic viral gene expression (29, 42).
HV68 encodes an Rta homologue, but a Zta homologue has not been identified. Overexpression of
HV68 Rta in a cell line latently infected with
HV68, S11, leads to the induction of early and late viral genes and the production of viral particles (44). Stimulation and hence activation of latently infected cells appears to be a critical component of the reactivation of gammaherpesviruses. This hypothesis has been further shown by the inhibition of the signaling cascade following stimulation with anti-Ig antibodies with the immunosuppressive drugs cyclosporine A and FK506, which prevents induction of EBV lytic replication (10). These results suggest that the full downstream signaling cascade is necessary to drive reactivation.
Several studies have shown that there is a potential role for epigenetic regulation of the key viral genes involved in reactivation. Both methylation and chromatin structure of the Rta promoter in KSHV appear to play a role in the silencing of this gene during latency (4, 18). Culturing cell lines latently infected with KSHV with either 5'azacytidine, a DNA methyltransferase inhibitor, or Trichostatin A, a histone deacetylase inhibitor, induces KSHV reactivation (4, 18). In addition, methylation has been shown to be critical in the regulation of latency promoters in EBV (22, 23), although whether DNA methylation plays a role in regulating EBV reactivation is unclear. However, previous studies have shown a strong correlation between methylation of the viral genome and the switch from the EBV growth-promoting latency program to a more restricted pattern of viral gene expression (22, 23).
In conclusion, we have shown that reactivation of
HV68 from latently infected splenic B cells at late times postinfection can be triggered by either
-Ig/
-CD40 or LPS stimulation. This underscores the idea that there are common intracellular signaling pathways that, in general, lead to gammaherpesvirus reactivation from latently infected B cells. In addition, these studies confirm that at least a subset of the viral genome-positive splenocytes can reactivate virus, confirming that they represent bona fide latently infected cells. Why only ca. 10% of viral genome-positive cells are stimulated to reactivate virus remains an open question. However, this is reminiscent of what is observed with most cell lines latently infected with EBV, which are inefficiently triggered to reactivate virus upon treatment with appropriate reactivation stimuli. Further studies will be required to determine whether alternative induction protocols can enhance the percentage of viral genome positive-splenocytes reactivating virus.
|
|
|---|
This article has been cited by other articles:
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Copyright © 2009 by the American Society for Microbiology. For an alternate route to Journals.ASM.org, visit: http://intl-journals.asm.org | More Info»