Requirement for CD40 Ligand, CD4⁺ T Cells, and B Cells in an Infectious Mononucleosis-Like Syndrome

Abstract

Respiratory challenge with the murine gammaherpesvirus 68 (γHV-68) results in productive infection of the lung, the establishment of latency in B lymphocytes and other cell types, transient splenomegaly, and prolonged clonal expansion of activated CD8⁺ CD62L^lo T cells, particularly a Vβ4⁺ CD8⁺ population that is found in mice with different major histocompatibility complex (MHC) haplotypes. Aspects of the CD8⁺-T-cell response are substantially modified in mice that lack B cells, CD4⁺ T cells, or the CD40 ligand (CD40L). The B-cell-deficient mice show no increase in Vβ4⁺ CD8⁺ T cells. Similar abrogation of the Vβ4⁺ CD8⁺ response is seen following antibody-mediated depletion of the CD4⁺ subset, through the numbers of CD8⁺ CD62L^lo cells are still significantly elevated. Virus-specific CD4⁺-T-cell frequencies are minimal in the CD40L^−/− mice, and the Vβ4⁺ CD8⁺ population remains unexpanded. Apparently B-cell–CD4⁺-T-cell interactions play a part in the γHV-68 induction of both splenomegaly and non-MHC-restricted Vβ4⁺ CD8⁺-T-cell expansion.

Infectious mononucleosis (IM) is a debilitating disease of human adolescents (14, 21) induced by the prototypic type 1 gammaherpesvirus (γHV), Epstein-Barr virus (EBV). The classical presentation is lymphoid tissue enlargement, concurrent with the presence of large numbers of activated CD8⁺ peripheral blood lymphocytes (PBL). The condition can continue for a month or more. Recent experiments have established that a substantial component of the expanded CD8⁺-T-cell population in the PBL compartment is directed against EBV peptides (5). Much of the IM phase of EBV infection thus reflects the specific host response in lymphoid tissue to this persistent virus.

Experimental dissection (28) of an apparently comparable syndrome induced by intranasal (i.n.) exposure to a type 2 γHV, the murine gammaherpesvirus 68 (γHV-68), has shown that the onset of the IM-like expansion of activated (CD62L^lo) CD8⁺ T lymphocytes in the blood follows the development of CD4⁺-T-cell-dependent splenomegaly (17, 29). Both the splenic enlargement and the massive increase in CD8⁺-T-cell numbers, particularly a prominent non-major histocompatibility complex (MHC)-restricted CD8⁺ set (28) that expresses the Vβ4 T-cell receptor (TCR), are detected subsequent to immune control (17) of the initial, lytic infection in respiratory epithelium (7). The delay in onset of the IM-like disease suggests that the driving force is persistent, latent γHV-68, which can be detected consistently in a small proportion of B lymphocytes by infectious-center assay.

Neither the splenomegaly nor the IM-like syndrome was seen in CD4⁺-T-cell-deficient mice that are homozygous for disruption (−/−) of the H-2I-A^b MHC class II gene, though the extent of viral latency detected by the infectious-center assay was at least as high as that found for the MHC class II^+/+ controls (7, 10). Also, the γHV-68 peptide-specific CD8 response (24) in these MHC class II^−/− mice was not obviously diminished (23). Early depletion of CD4⁺ T cells by treating MHC class II^+/+ mice with a subset-specific monoclonal antibody (MAb) prevented the development of splenomegaly, but the IM-like phase had not been recognized at the time of these experiments (29). Giving such a MAb later (from day 11) in the course of γHV-68 infection diminished the numbers of cycling CD8⁺ T cells in the PBL, though the frequencies of both the CD8⁺ CD62L^lo and CD8⁺ Vβ4⁺ sets were comparable to those in undepleted mice (28).

The present analysis focused on the role of the CD4⁺ subset in this IM-like disease. The part played by B lymphocytes (26) was also addressed by using immunoglobulin-deficient (Ig^−/−) μMT mice (15), which lack virus-infected cells that can readily be demonstrated by the infectious center assay (30). However, a further focus of γHV-68 latency has now been detected in the macrophage compartment by a different technique (33), and it is clear that μMT mice are indeed persistently infected with γHV-68 to the extent that they will die following simultaneous depletion of both CD4⁺ and CD8⁺ T cells (8) long after the acute phase of the infection has been controlled.

Experimental procedures.

The methods used here have been described previously and are appropriately referenced throughout the text. The general protocol was to infect anaesthetized, 6- to 10-week-old, female C57BL/6J(B6) and (B6 × 129)F₁ (Ig^+/+ CD40L^+/+), μMT (Ig^−/−), or CD40L^−/− mice (35) i.n. with 600 PFU of γHV-68 (7). The μMT mice (15) were bred (with permission from Werner Müller) at St. Jude Children’s Research Hospital, while all other mice were purchased from the Jackson Laboratory (Bar Harbor, Maine). The mice were anaesthetized again at the time of sampling, when PBL and spleen populations were obtained for flow cytometric analysis (28) and the lung and lymphoid compartments were assayed for the presence of lytic (lung) and latent (spleen and lymph nodes) virus (6, 7). Frequencies of virus-specific CD4⁺ T cells were determined by the gamma interferon (IFN-γ) ELISpot assay (9). The prevalence of virus-specific CD8⁺ T cells (23, 24) was assessed by stimulating cells for 6 h with γHV-68 peptide in the presence of brefeldin A and then staining for IFN-γ and analyzing in a FACScan by using CellQuest software (Becton Dickinson, San Jose, Calif.). Lymphocyte phenotypes were determined (28) by staining with phycoerythrin (PE)- or fluorescein isothiocyanate (FITC)-conjugated MAbs (all supplied by Pharmingen, San Diego, Calif.) specific for CD4 (RM4-5-PE), CD8α (53-6.72-PE), CD62L (MEL-14-FITC), and Vβ4 TCR (KT4-FITC).

Consequences of B-cell deficiency.

Previous experiments established that the Ig^−/− μMT mice (15) utilize both CD4⁺ and CD8⁺ T cells to control the acute, lytic phase of γHV-68 infection (8), though there has been some debate about the extent of subsequent viral latency (30–32). The present study with i.n. challenged μMT mice also failed to demonstrate persistent γHV-68 by the infectious center assay, but the continued presence of γHV-68 throughout the lymphoid compartment was confirmed (Table 1) by a primary culture system based on that used previously to demonstrate the presence of cytomegalovirus (6).

TABLE 1.

Virus persistence in the lymphoid tissue of Ig^+/+ and Ig^−/− mice^a

Organ	Mouse strain	Mean no. of infectious centers/107 lymphocytesb		Log10 PFU of virus/ml of culture supernatantc
		Day 15	Day 40	Day 15	Day 40
MLN	B6	150	5	4.0 × 102	2.3 × 106
	μMT	0	0	1.7 × 104	1.9 × 106
CLN	B6	200	4	3.0 × 104	1.3 × 104
	μMT	0	0	1.4 × 101	1.5 × 103
Spleen	B6	400	10	4.0 × 102	0
	μMT	0	0	2.1 × 104	4.0 × 102

^aThe Ig^+/+ (B6) and Ig^−/− (μMT) mice were infected i.n. with 600 PFU of γHV-68, and samples of the mediastinal lymph nodes (MLN), cervical lymph nodes (CLN), and spleen were taken 15 and 40 days later for assay (7). In both the infectious-center assay and the primary-cell-culture assay, no virus was detected if cells had been killed by repeated freeze-thaw cycles prior to plating of the cells. 

^bThe infectious-center assay detects virus reactivation by culturing single-cell suspensions of lymphoid tissue with NIH 3T3 fibroblast monolayers over a 6-day period (7). 

^cLymphocyte suspensions were dispensed (1 × 10⁷ and 3 × 10⁶ cells) into six-well tissue culture plates in a final volume of 5.0 ml of medium. The primary cell cultures were incubated at 37°C and 5% CO₂ for up to 6 weeks, with 4.0 ml of supernatant being removed and replaced weekly with 4.0 ml of fresh medium (6). Culture supernatants were then assayed for the presence of lytic virus by plaque assay on NIH 3T3 cells (7). 

The absence of B-cell follicle development in the Ig^−/− μMT mice results in a spleen size that is normally about 20% of that detected in the Ig^+/+ controls (27). The relative prevalence of CD4⁺ T cells in the μMT spleen and blood is also decreased (Fig. 1, day 0). Respiratory challenge with γHV-68 fails to cause the splenomegaly found in Ig^+/+ B6 mice (30). However, the prevalence of the “activated” CD8⁺ CD62L^lo population (28) was increased in both the Ig^+/+ and Ig^−/− groups from day 14 after infection, though the IM-like phase (28) in the μMT mice was diminished in magnitude (Fig. 1B). The essential difference was that the B-cell-deficient Ig^−/− mice did not show the characteristic increase in Vβ4⁺ CD8⁺-T-cell numbers for either the spleen (Fig. 2A) or the blood (Fig. 2B).

FIG. 1.

Prevalence and activation status of splenic (A) and PBL (B) CD8⁺ T cells from γHV-68-infected B6 and μMT mice. The splenocytes were pooled, while the PBL samples were analyzed for individuals (28). The experiment was done three times; the results are from one representative experiment and are expressed as percents (spleen) or mean percents ± standard deviations (PBL).

FIG. 2.

The spectrum of TCR Vβ4 expression on CD8⁺ T cells from spleen (A) and PBL (B) populations from γHV-68 infected B6 and μMT mice. The experiment was done twice, and results of one representative experiment are shown. The values are for pooled samples from four or five mice.

γHV-68 infection in CD40L^−/− mice.

The lack of splenomegaly and Vβ4⁺ CD8⁺ T cell expansion in the μMT mice could be thought to be due to the presence of less persistently infected cells (Table 1), the decreased size of the virus-specific CD4⁺ set, or the absence of B cells. Effective T help for antibody production requires that the CD40 ligand (CD40L) expressed on the CD4⁺ T cell bind the CD40 molecule on the B cell, a recognition event that induces efficient activation of both cell types (2, 11, 13, 16, 18, 20, 22). Experiments with CD40L^−/− mice (20, 35) have established the importance of this interaction in several different virus infections (3, 4, 12, 19, 34).

Following respiratory challenge with γHV-68, the CD40L^−/− mice showed some of the changes described previously for the CD4⁺-T-cell-deficient MHC class II^−/− mice (7). Though the lytic phase of virus growth was to some extent controlled in the respiratory tract, evidence of productive infection in this site continued in the long term (Fig. 3A). Furthermore, unlike the situation for the μMT mice (Table 1), evidence of viral latency was readily demonstrated by the infectious-center assay (Fig. 3B). Also, as with the MHC class II^−/− mice (23), the magnitude of the CD8⁺-T-cell response to the p56 and p79 peptides of γHV-68 was essentially normal in the absence of the CD40-CD40L interaction (Table 2). The virus-specific CD4⁺-T-cell response detected by the ELISpot assay was, however, substantially absent from the CD40L^−/− group (Table 2).

FIG. 3.

Levels of replicating and latent γHV-68 virus in (B6 × 129)F₁ and CD40L^−/− mice. The titers (7) of infectious virus in lung (A) and the extent of viral latency in the spleen (B) are given as means ± standard deviations. The titers of lytic virus in spleen cells that were disrupted before plating were generally <1 PFU/10⁷ cells. The results given are from two separate sets of observations, with three or four mice per time point in each experiment.

TABLE 2.

Virus-specific T-cell responses in CD40L^−/− and (B6 × 129)F₁ mice^a

Day after infection	(B6 × 129)F1 mice				CD40L−/− mice
	% IFN-γ+ CD8+ cellsb		CD4+ Thp frequencyc		% IFN-γ+ CD8+ cells		CD4+ Thp frequency
	p56	p79	600 PFU	10,000 PFU	p56	p79	600 PFU	10,000 PFU
7	0.52 ± 0.18	0.42 ± 0.17	148 ± 21	ND	0.55 ± 0.07	0.49 ± 0.18	6,408 ± 4,973	ND
16	2.68 ± 0.86	4.87 ± 1.07	80 ± 68	128 ± 144	1.77 ± 1.00	2.80 ± 1.64	6,102 ± 6,982	3,234 ± 2,033
35	0.91 ± 0.15	1.67 ± 0.92	ND	ND	1.63 ± 0.62	4.03 ± 2.97	ND	ND

^aThe (B6 × 129)F₁ and CD40L^−/− mice were infected i.n. with 600 PFU of γHV-68, and single-cell spleen suspensions were analyzed for virus-specific CD8⁺ (p56 or p79) or CD4⁺ T (Thp) cells. All values are means ± standard deviations for four or five mice. ND, not done. 

^bDetermined by flow cytometric analysis of spleen populations following 6 h of stimulation with the H-2D^b-restricted p56 peptide or the H-2K^b-restricted p79 peptide in the presence of brefeldin A. The lymphocytes were fixed and stained for the presence of IFN-γ. 

^cReciprocal of the CD4⁺ Thp frequency, determined in a 48-h ELISpot assay. The Thp frequencies were not obviously modified by infecting the mice with a higher dose of virus. 

The prevalence of activated CD8⁺ CD62L^lo cells tended to be lower but, in the groups of three to six mice used in these experiments, was not significantly different from that found for the CD40L^+/+ controls (data not shown). However, the prominent Vβ4⁺ CD8⁺-T-cell response that occurs in conventional mice (28) was completely abrogated by the absence of the CD40L (Fig. 4). Furthermore, the elimination of the CD4⁺ subset by treating the (B6 × 129)F₁ mice with a MAb to CD4 from the time of infection (1) also prevented the expansion of the Vβ4⁺ CD8⁺ set (Fig. 4), though the prevalence of CD8⁺ CD62L^lo cells in the spleen and PBL compartments of such mice was consistently above the levels found in the naive controls (Fig. 5).

FIG. 4.

Expression of the Vβ4 TCR on CD8⁺ T cells in the PBL population. Some of the γHV-68-infected (B6 × 129)F₁ and CD40L^−/− mice were treated from 2 days before virus challenge with successive doses of the GK1.5 MAb, a procedure that effectively eliminates the CD4⁺ subset (1). The experiment was done twice, with results of one representative experiment being shown. The results are means ± standard deviations for three or four individuals.

FIG. 5.

Activation status of splenic (A) and PBL (B) CD8⁺ T cells in γHV-68-infected intact or CD4-depleted (Fig. 4) B6 mice. The results are means ± standard deviations for a representative experiment (28). The total number of activated CD8⁺ T cells (A) was derived by multiplying the cell count for the spleen by the percent CD8⁺ CD62L^lo cells. With the exception of the findings for the CD4-depleted PBL population assayed on day 12, all values shown in both panels for the γHV-68-infected mice are significantly greater (P < 0.05) than those for the uninfected controls (day 0). The experiment was done three times, with results of one representative experiment being shown.

Conclusions.

The experiments with the CD4-depleted and CD40L^−/− mice establish that CD4⁺ T cells are required to promote the expansion of Vβ4⁺ CD8⁺ T cells that is so characteristic of γHV-68 infection (28). The virus-specific CD8⁺-T-cell response does not, however, seem to depend on CD4⁺ T help, and the prevalence of CD8⁺ CD62L^lo T cells in the spleen and PBL is still increased in the absence of the CD4⁺ subset. The same profile is seen in the absence of B cells, though the Ig^−/− μMT mice make an effective CD4⁺-T-cell response that can control persistent γHV-68 infection by an IFN-γ-dependent process (8).

The obvious conclusion is that the CD4⁺ helpers induce some modification of the B-cell surface that stimulates the Vβ4⁺ CD8⁺ T cells. The CD4⁺-T-cell depletion experiments indicate that this event must occur during the acute phase of the host response (Fig. 4 and 5), prior to day 11 (28). It is not known whether the entity recognized by this unusual non-MHC-restricted Vβ4⁺ CD8⁺ set is encoded by the virus or is some aberrantly expressed self component. Apart from the fact that T-cell help is required for both the massive γHV-68-induced, nonspecific IgG response and for the production of virus-specific antibody (25), we currently know very little about the interaction between CD4⁺ T cells and B cells in this infection.