Depletion of γδ⁺ T cells increases CD4⁺ FoxP3 (T regulatory) cell response in coxsackievirus B3-induced myocarditis

Abstract

Coxsackievirus B3 (CVB3) causes severe myocarditis in BALB/c mice which depends upon CD4⁺ T helper type 1 [Th1; i.e. interferon-γ⁺ (IFN-γ⁺)] and γδ⁺ cells. Depleting γδ⁺ cells using anti-γδ antibody suppresses myocarditis and CD4⁺ IFN-γ⁺ cell numbers in the spleen and heart of infected mice while increasing CD4⁺ FoxP3⁺ cells. Mice deficient in γδ⁺ cells have increased numbers of naïve (CD44^lo CD62L^hi) and fewer effector (CD44^hi CD62^lo) memory CD4⁺ cells than infected γδ⁺-cell-sufficient mice. Virus neutralizing antibody titres are not significantly different between γδ⁺ T-cell-sufficient and -deficient animals. To confirm that the memory cell response differs in acutely infected mice lacking γδ⁺ cells, CD4⁺ cells were purified and adoptively transferred into naïve recipients, which were rested for 4 weeks then infected with CVB3. Recipients given either 0·5 × 10⁶ or 1·0 × 10⁶ CD4⁺ from infected donors developed over twice the severity myocarditis and 10-fold less cardiac virus titre compared with recipients given equivalent numbers of CD4⁺ cells from infected and γδ⁺-cell-depleted donor animals. Additionally, to show that more functionally active T regulatory cells are present in γδ⁺ T-cell-depleted mice, CD4⁺ CD25⁺ and CD4⁺ CD25⁻ cells were isolated and adoptively transferred into infected recipients. Mice receiving CD4⁺ CD25⁺ cells from γδ⁺ T-cell-depleted donors developed significantly less myocarditis and CD4⁺ Th1 cell responses compared with mice receiving equal numbers of CD4⁺ CD25⁺ cells from infected γδ⁺ T-cell-sufficient animals. This study shows that γδ⁺ cells promote CD4⁺ IFN-γ⁺ acute and memory responses by limiting FoxP3⁺ T regulatory cell activation.

Introduction

T lymphocytes undergo three distinctive phases during an immune response. These are the initial activation and clonal expansion, contraction of unneeded effector cells after the acute response through cell death, and differentiation of a subpopulation of the effector cells into memory cells which will provide future protection from reinfection.^1,2 Many factors influence the quality and quantity of CD4⁺ effector cells induced during an immune response. These include both the activation state of antigen-presenting cells (costimulatory molecule expression) and proinflammatory cytokine production.^3–7 Signal transduction through Toll-like receptors effectively enhances the effector CD4⁺ cell response by both activating antigen-presenting cells and inducing a proinflammatory cytokine response.⁸ Activation of T cells alters their expression of adhesion and integrin molecules,^1,2 with naïve cells expressing low levels of CD44, CD11a and CD122 but high levels of CD62 ligand (CD62L), CCR7 and CD27 while effector cells have high levels of the former molecules and low levels of the latter. Naïve T cells primarily home to primary lymphoid organs (lymph nodes and spleen) whereas effector cells home to a wide range of non-lymphoid tissues.^9,10 Memory cells evolve from the effector population without requiring further antigenic stimulation or proliferation.^11–14 Memory cells are small resting cells that can rapidly proliferate and excrete high levels of cytokines, including interleukin-2 (IL-2), upon antigenic rechallenge, and that are resistant to apoptosis.¹⁵ Two populations of memory cells have been described. These are T effector memory (Tem) and T central memory (Tcm) cells,¹ where both express high levels of CD44, CD11a and Ly6C but Tem express low levels of CD62L and CCR7 while Tcm express high levels of CD62L and CCR7. The relative biological importance of Tem and Tcm is controversial, and may depend upon the pathogen.^16–19

Previous studies have shown that T cells expressing the γδ T-cell receptor (TCR) are essential for induction of a potent effector CD4⁺ T helper type 1 [Th1; interferon-γ⁺ (IFN-γ⁺)] response in acute coxsackievirus B3 (CVB3) infection and for susceptibility to virus-induced myocarditis.^20,21 Studies have also demonstrated that γδ⁺ cells specifically kill CD1d⁺ cells.²¹ This study shows that depletion of γδ⁺ cells in vivo in CVB3-infected mice results in substantial increases in CD1d⁺ cells and that these are mostly CD11b⁺ macrophages. Depletion of γδ⁺ cells results in substantial increases in CD4⁺ FoxP3⁺ T regulatory cells, which probably prevent activation of a potent CD4⁺ Th1 cell response during infection. As would be expected with fewer activated effector CD4⁺ Th1 cells, there is also a reduction in CD4⁺ Th1 memory T cells in γδ⁺-cell-depleted mice as shown in a virus rechallenge model.

Materials and methods

Mice

Male BALB/cJ mice were purchased from Jackson Laboratories, Bar Harbor, ME. All mice were 5–7 weeks of age when infected. All of the studies have been reviewed and approved by the University of Vermont Institutional Animal Care and Use Committee.

Virus

The H3 variant of CVB3 was made from an infectious complementary DNA clone as described previously.²²

Anti-γδ antibody treatment of mice

Mice were injected intraperitoneally with 0·5 ml phosphate-buffered saline (PBS) or with PBS containing 100 μg anti-γδ TCR antibody (clone GL3, BD Biosciences/Pharmingen, Franklin Lakes, NJ) on days –1 and –2 relative to infection.

Infection of mice

Mice were injected intraperitoneally with 10⁴ plaque-forming units (PFU) virus in 0·5 ml PBS. Animals were killed when moribund or 7 days after infection.

Organ virus titres

Hearts were aseptically removed from the animals, weighed, homogenized in RPMI-1640 medium containing 5% fetal bovine serum (FBS) 10 mm, l-glutamine, 100 μg/ml streptomycin and 100 U/ml penicillin. Cellular debris was removed by centrifugation at 300 g for 10 min. Supernatants were diluted serially using 10-fold dilutions and titres on Hela cell monolayers were measured by plaque-forming assay.²³

Histology

Tissue was fixed in 10% buffered formalin for 48 hr, paraffin-embedded, sectioned and stained with haematoxylin and eosin. Image analysis of cardiac inflammation was performed as described previously.²²

Virus neutralizing antibody titres

After killing, mice were bled by intracardiac puncture and the blood was clotted and centrifuged. Serum was removed and heat-inactivated at 56° for 30 min. Confluent monolayers of HeLa cells were grown in 96-well tissue culture plates. The medium was removed and 1 : 10 dilution of serum in RPMI-1640 medium containing 5% FBS was added in 50 μl volume. Next, 100 PFU CVB3 was added in 50 μl medium and the plates were incubated for 24 hr in a humidified 5% CO₂ incubator. The cells were fixed for 15 min with 10% buffered formalin and stained with 0·4% crystal violet. The plates were read at 630 nm using a Biotek EL808 plate reader. The neutralizing titre was the dilution at which 50% of the HeLa cell monolayer remained. Controls were wells with HeLa cells without virus and wells with cells, virus and 1 : 10 dilutions of non-infected mouse serum.

Isolation of lymphocytes

Spleens were removed and pressed through fine-mesh screens. Lymphoid cells were isolated by centrifugation of cell suspensions on Histopaque (Sigma Chemical Co., St Louis, MO). CD4⁺ cells were purified using the BD Biosciences CD4⁺ enrichment kit according to the manufacturer’s directions. Purity of the cell population exceeded 90% CD4⁺ cells. The CD4⁺ CD25⁺ cells were isolated from the spleen using the Dynabeads FlowComp Mouse CD4⁺ CD25⁺ Treg kit (Invitrogen, Carlbad, CA) according to the manufacturer’s directions. Inflammatory cells in the heart were isolated by perfusing individual hearts with PBS, mincing the hearts finely, digesting the hearts with 0·4% collagenase II (Sigma) and 0·25% pancreatin (Sigma), then centrifuging the cell suspension on Histopaque (Sigma).

Intracellular cytokine staining

Details for intracellular cytokine staining have been published previously.²⁴ Spleen cells (10⁵ cells) were cultured for 4 hr in RPMI-1640 medium containing 10% FBS, antibiotics, 10 μg brefeldin A (BFA; Sigma), 50 ng/ml phorbol 12-myristate 13-acetate (PMA; Sigma) and 500 ng/ml ionomycin (Sigma). The cells were washed in PBS–1% bovine serum albumin (BSA; Sigma) containing BFA (PBS-BSA-BFA), incubated on ice for 30 min in PBS-BSA-BFA containing a 1 : 100 dilution of Fc Block and either peridinin chlorophyll protein (PerCP)-Cy5.5-conjugated anti-CD4 (clone GK1.5) or PerCP-Cy5.5-conjugated rat immunoglobulin G2b (IgG2b; clone A95-1) antibodies. The cells were washed once with PBS-BSA-BFA, fixed in 2% paraformaldehyde for 10 min, then resuspended in PBS-BSA containing 0·5% saponin, Fc Block and 1 : 100 dilutions of phycoerythrin (PE)-conjugated anti-IFN-γ (cloneXMG1.2) or PE-conjugated rat IgG1 (clone R3-34) antibodies and incubated for 30 min on ice. All antibodies were from BD Biosciences/Pharmingen. FoxP3 labelling was performed using the eBioscience kit (San Diego, CA) according to the manufacturer’s directions. Cells were labelled with Alexa647-conjugated anti-CD4 and PerCP-Cy5.5-conjugated anti-CD25 (clone PC61; BD Biosciences, Franklin Lakes, NJ) in PBS-1%BSA containing Fc Block, washed, fixed and permeabilized, then incubated with PE-conjugated anti-FoxP3 and Fc Block overnight at 4°. The cells were washed once in PBS-BSA-saponin and once in PBS-BSA, then resuspended in 2% paraformaldehyde.

To determine the number of individual cell populations, the total number of viable cells was determined by trypan blue exclusion. Following flow cytometry, the percentage of a subpopulation staining with a specific antibody was multiplied by the total number of cells.

Adoptive transfer of CD4⁺ cells and rechallenge

Purified CD4⁺, CD4⁺ CD25⁺ or CD4⁺ CD25⁻ cells were isolated from uninfected BALB/c mice and from BALB/c mice infected 7 days earlier with 10⁴ PFU H3 virus. Infected mice had received 0·5 ml PBS alone or PBS containing 100 μg anti-γδ antibody (GL3) on days –1 and –2 relative to infection. The cells were washed and concentrations indicated in the text were injected intravenously in 0·2 ml PBS into the tail veins of uninfected BALB/c recipients. To confirm that virus was not transferred with the cells, 10⁶ CD4⁺ cells from infected mice were homogenized and titred by the plaque-forming assay on HeLa cells; there was no virus. CD4⁺ cell recipients were rested for 28 days. No recipient mouse died or appeared infected. At day 28, the recipients were infected with 10⁴ PFU virus and killed 7 days later. Controls consisted of BALB/c mice not given CD4⁺ cells but infected with 10⁴ PFU virus.

Flow cytometry

For analysis of cell surface molecules, 10⁵ lymphocytes were incubated in PBS-BSA containing Fc Block and PerCP-Cy5.5-conjugated anti-CD4, fluorescein isothiocyanate (FITC)-conjugated anti-CD44 (clone IM7), Alexa647-conjugated anti-CD62L (clone MEL-14), and PE-conjugated anti-CD69 (clone H1.2F3) antibodies. Additional cells were labelled with FITC-conjugated anti-CD3 (clone 17A2) and PE-conjugated anti-γδ (clone GL3, BD Biosciences/Pharmingen) antibodies, then analysed using a BD LSR II flow cytometer with a single excitation wavelength (488 nm) and band filters for PerCP-Cy5.5 (695/40 nm), FITC (525 nm) and PE (575 nm). The excitation wavelength for Alexa 647 is 643 nm and a band filter of 660/20 nm. The cell population was classified for cell size (forward scatter) and complexity (side scatter). At least 10 000 cells were evaluated. Positive staining was determined relative to isotype controls.

Statistics

Differences between groups were determined by Wilcoxon Ranked Score.

Results

Depletion of γδ^± T cells prevents myocarditis and increases CD4⁺ FoxP3⁺ T cells

Previous studies have shown that γδ⁺ cells are necessary for the induction of acute CVB3-induced myocarditis primarily through the induction of a effector CD4⁺ Th1 (IFN-γ⁺) cell response.^21,25 The following study determined that depletion of γδ⁺ cells significantly increased numbers of CD4⁺ FoxP3⁺ (T regulatory) cells. BALB/c mice were infected with 10⁴ PFU H3 virus and injected –1 and –2 days relative to infection with PBS or PBS containing 100 μg monoclonal anti-γδ TCR antibody. Animals were killed 7 days after infection. As expected, both myocarditis and cardiac virus titres were significantly reduced in animals depleted of γδ⁺ cells (Figs 1 and 2). Plasma was obtained from the mice and evaluated for virus neutralizing antibody titre (Fig. 2). No significant differences were observed between infected γδ⁺-cell-deficient and -sufficient groups. Spleen lymphocytes and lymphoid cells from the hearts of individual mice were isolated and evaluated for B cells (CD19⁺), T cells (CD3⁺) and macrophages (CD11b⁺ and γδ⁺ cells) (Fig. 2). No significant difference was observed in total numbers of spleen lymphocytes or in the CD19⁺ and CD3⁺ cell populations. However, PBS-treated mice had 4·2 ± 0·7%γδ⁺ cells compared with 0·3 ± 0·1% for anti-γδ antibody-treated mice, representing a 92·9% depletion of γδ⁺ cells in the spleen with anti-γδ antibody treatment. Surprisingly, numbers of CD11b⁺ cells in the spleen were significantly increased in anti-γδ antibody-treated mice. The total number of lymphoid cells isolated from the heart and the numbers of all cell subpopulations were decreased in γδ⁺-cell-depleted mice, as would be expected considering the decrease in myocarditis in these animals. Previous studies have shown that γδ⁺ cells lyse CD1d⁺ cells.²¹ CD1d⁺ cells in the spleen were determined (Fig. 3) and showed significant increases in CD1d⁺ cells in the γδ⁺-cell-depleted animals and these were mostly in the CD11b⁺ cell population (data not shown). Staining of spleen cells for IFN-γ and FoxP3 (Fig. 3) showed that although numbers of CD4⁺ IFN-γ⁺ cells were significantly reduced by γδ⁺ cell depletion, the numbers of CD4⁺ FoxP3⁺ cells were significantly increased in these animals. These cell populations were also evaluated in the lymphoid cells isolated from the heart and again showed no CD4⁺ IFN-γ⁺ cells but an increase in CD4⁺ FoxP3⁺ cells.

Figure 1.

Anti-γδ cell depletion prevents myocarditis. Representative histology of BALB/c mice infected with 10⁴ plaque-forming units H3 virus and treated with either phosphate-buffered saline (PBS) or anti-γδ T-cell receptor (TCR) antibody (100 μg intraperitoneally on days –1 and –2 relative to infection). Animals were killed 7 days after infection.

Figure 2.

Anti-γδ cell depletion increases CD11b⁺ cells in the spleen. BALB/c mice were infected with H3 virus and treated with phosphate-buffered saline (PBS) or anti-γδ T-cell receptor antibody as described in Fig. 1. Results are the mean ± SEM of cardiac virus titres [logarithm base 10 plaque-forming units (PFU)/g tissue] and percentage of the myocardium inflamed as determined by image analysis. Plasma was obtained by intracardiac puncture at the time of killing and the virus neutralizing antibody titre was determined by the plaque reduction assay. Lymphocytes were isolated from spleens or hearts of individual mice by centrifugation on Histopaque, counted by trypan blue exclusion and labelled with antibodies to B cells (CD19), T cells (CD3), macrophage (CD11b) and γδ⁺ T cells. Numbers of each cell type were determined by multiplying the percentage of cells labelled with antibody to a specific cell molecule by the number of spleen cells isolated. Groups consisted of five to eight mice. *Significantly different from PBS-treated group at P < 0·05.

Figure 3.

Anti-γδ cell depletion decreases CD4⁺ interferon-γ⁺ (IFN-γ⁺) and increases CD4⁺ FoxP3⁺ cell responses. Splenic and heart infiltrating lymphocytes were isolated from BALB/c mice infected with H3 virus and treated with phosphate-buffered saline (PBS) or anti-γδ T-cell receptor antibody as described in Fig. 1. (a) Spleen cells were labelled with antibody to CD1d. (b) Spleen cells were labelled with antibody to CD4 then labelled intracellularly with antibodies to FoxP3. (c) Spleen cells were cultured for 4 hr with phorbol 12-myristate 13-acetate, ionomycin and brefeldin A, then labelled with antibody to CD4, fixed, permeabilized with saponin and labelled with antibody to IFN-γ. (a–c) Representative flow diagrams showing labelling for the indicated molecules. (d) Summary of mean cell number ± SEM CD4⁺ IFN-γ⁺ and CD4⁺ FoxP3⁺/mouse for spleen and heart infiltrating cells for five to eight mice/group. *, **Significantly different from PBS-treated group at P < 0·05 and P < 0·01, respectively.

Effects of γδ⁺ cell depletion on CD62L and CD44 expression by CD4⁺ cells

The T cells can be distinguished into naïve, Tem and Tcm cells by their relative expression of CD44 and CD62L. Tem cells are CD62L^low CD44^high; naïve cells are CD62L^high CD44^low; and Tcm cells are CD62L^high CD44^high.^1,26 To determine the effect of γδ⁺ cell depletion on T memory phenotype, spleen cells from mice 7 days after infection were labelled with antibodies to CD4, CD62L, CD69 (early activation marker) and CD44 (Fig. 4). Depletion of γδ⁺ cells resulted in significantly increased numbers of naïve and fewer Tem cells in the spleen compared with PBS-treated control mice. Activated CD4⁺ cells, as denoted by CD69 expression, were also decreased in γδ⁺-cell-depleted mice. There was no significant change in Tcm cell numbers between the groups.

Figure 4.

Splenic lymphocytes were labelled with antibodies to the indicated cell surface molecules and evaluated by flow cytometry. (a) Mean number of cells ± SEM/mouse positive for indicated molecules for five to eight mice/group. (b) Representative flow diagrams showing labelling for the indicated molecules. *Significantly different from phosphate-buffered saline (PBS)-treated group at P < 0·05.

Recall challenge of CD4⁺ T cells to virus challenge

Although changes in CD62L and CD44 expression consistent with differences in memory response were observed as early as day 7 after infection, this study does not prove that fewer T memory cells are produced in γδ⁺-cell-depleted mice. Observation times longer than 7 days are not practical because animal mortality is 50% at day 7 but rapidly increases to 100% by day 10 after infection. However, one can determine if fewer memory cells are generated by adoptive transfer of purified CD4⁺ cells into naïve recipients followed by virus challenge 28 days later. A schematic of the experimental design is given in Fig. 5. As a control, BALB/c mice that had not received CD4⁺ cells were also infected with virus. Seven days after infection, the mice were killed and evaluated for myocarditis, cardiac virus titre and CD4⁺ cell response by measuring CD69 and IFN-γ expression. Aliquots of the purified CD4⁺ cells from H3 virus-infected donors were also homogenized and titred on HeLa cells; this showed that no infectious virus was transferred with the cells. Figure 6 shows representative histology for each group. Mice given CD4⁺ cells from infected donors developed significantly more myocarditis than control mice not given CD4⁺ cells or given naïve CD4⁺ cells. CD4⁺ cells obtained from infected and γδ⁺-cell-depleted donors gave only a minimal memory response. Myocarditis was increased slightly in these mice, but not to the levels observed in mice given CD4⁺ cells from H3-infected donors with γδ⁺ cells (Fig. 7). Cardiac virus titres were also significantly reduced in mice receiving CD4⁺ cells from H3-infected donors, indicating that the memory CD4⁺ cells participate in virus control. No change in virus titres was observed in mice given naïve CD4⁺ cells or CD4⁺ cells from infected and γδ⁺-cell-depleted donors. Recipients of CD4⁺ cells from H3-infected donors consistently have increased numbers of activated (CD69⁺) (Fig. 7, bottom left) and IFN-γ⁺ (Fig. 7, bottom right) cells than control mice or mice given CD4⁺ cells from naïve donors. A cell dose–response study was performed using the CD4⁺ cells from naïve, infected and infected/γδ⁺-cell-depleted donors (Fig. 8). Giving as few as 5 × 10⁵ CD4⁺ cells from infected donors increased subsequent myocarditis in recipient animals but none of the cell concentrations of naïve CD4⁺ cells altered myocarditis. Only the highest concentration of CD4⁺ cells from infected/γδ⁺-cell-depleted donors produced more disease.

Figure 5.

Experimental design for determining CD4⁺ memory cells. BALB/c mice were uninfected (phosphate-buffered saline; PBS) or infected with H3 virus (10⁴ plaque-forming units; PFU). Half of the H3-infected mice were also treated intraperitoneally with 100 μg anti-γδ T-cell receptor antibody on days –1 and –;2 relative to infection. After 7 days, spleens were removed, CD4⁺ cells were isolated and 1 × 10⁶ CD4⁺ cells were injected intravenously (i.v.) through the tail vein into BALB/c mice. The animals were rested for 28 days, then infected with 10⁴ PFU H3 virus. Control mice were BALB/c animals infected with virus but not receiving donor CD4⁺ cells. After 7 days, the mice were killed.

Figure 6.

Representative histology of mice described in Fig. 5. BALB/c mice were injected intravenously with 10⁶ purified CD4⁺ cells from uninfected (naïve), H3-infected or H3-infected and anti-γδ T-cell receptor antibody-treated donor mice. The recipients were rested for 28 days after receiving CD4⁺ cells. Control mice did not receive CD4⁺ cells. All mice were injected intraperitoneally with 10⁴ plaque-forming units CVB3 and killed 7 days after infection for evaluation of myocarditis.

Figure 7.

Cardiac virus titres (logarithm base 10 plaque-forming units/g tissue) and % myocardium inflamed for mice described in Fig. 5. Results are the mean ± SEM of six to eight mice. Spleen lymphocytes from individual mice were labelled for CD4, CD69 and intracellularly for interferon-γ (IFN-γ). *,**Significantly different from control (no CD4⁺ cells) at P < 0·05 and P ≤ 0·01. ^#Significantly different from H3 CD4⁺ cell recipient at P < 0·01.

Figure 8.

Dose response of CD4⁺ cells. Recipients were injected intravenously with between 0 and 10⁶ purified CD4⁺ cells from naïve, H3-infected or H3-infected and anti-γδ antibody-treated donors. Recipients were rested for 28 days then infected with H3 virus. Recipients were killed 7 days after infection for analysis of myocarditis. *,**Significantly different from 0 CD4⁺ cells transferred at P ≤ 0·05 and P < 0·01, respectively.

Effects of transfer of CD4⁺ CD25⁺ and CD4⁺ CD25⁻ cells into recipients

The CD4⁺ cells transferred in the experiment above would contain mixtures of T regulatory (FoxP3⁺) and T effector cells. To evaluate the relative functional state of the T regulatory cells in γδ⁺-cell-depleted compared with γδ⁺-cell-sufficient mice, CD4⁺ CD25⁺ and CD4⁺ CD25⁻ cells were isolated from the spleens of anti-γδ and PBS-treated mice 7 days after infection. Then either 10⁵ and 10³ CD4⁺ CD25⁺ cells or 10⁶ and 10³ CD4⁺ CD25⁻ cells were injected intravenously into BALB/c mice on the same day as infection. Recipients were killed 7 days later and evaluated for myocarditis (Fig. 9). Infected BALB/c mice not given cells acted as controls. Transfer of 10⁵ CD4⁺ CD25⁺ cells from anti-γδ-treated mice was significantly more protective than giving the same number of CD4⁺ CD25⁺ cells from infected PBS-treated donors. Giving 10⁶ CD4⁺ CD25⁻ cells from the infected PBS-treated donors also significantly increased myocarditis in recipients compared with the same number of cells from anti-γδ-treated donors. This shows that on an equal cell basis, CD4⁺ CD25⁺ (T regulatory) cells are more active when derived from anti-γδ-treated animals.

Figure 9.

CD4⁺ CD25⁺ T regulatory cells from mice deficient in γδ⁺ T cells are more suppressive of myocarditis on a per cell basis than CD4⁺ CD25⁺ cells from γδ⁺-T-cell-sufficient mice. BALB/c mice were treated with phosphate-buffered saline (PBS) or anti-γδ antibody and killed 7 days later. Splenocytes were purified into CD4⁺ CD25⁺ and CD4⁺ CD25⁻ cell populations and transferred intravenously through the tail vein into BALB/c recipient mice which were infected the same day with 10⁴ plaque-forming units virus. Recipients were killed 7 days after infection and evaluated for myocarditis. The control group (No Cells) was infected and injected with PBS but no cells. Results are the mean ± SEM of four mice/group. *Significantly different from the ‘No Cells’ group at P < 0·05.

Discussion

We believe that this communication is the first published report that γδ⁺ T cells not only enhance CD4⁺ T effector cell responses subsequent to virus infection but also enhance T memory cell generation and that the memory CD4⁺ cells substantially aggravate myocarditis upon virus rechallenge in vivo. Most importantly, depletion of γδ⁺ cells caused significant increases in CD11b⁺ and CD4⁺ FoxP3⁺ cell numbers in the spleen, providing indirect evidence that γδ⁺ cells mediate their effects on CD4⁺ Th1 cell responses by preventing T regulatory cell activation. Although not shown in this communication, the modulation of T regulatory cells may be through γδ⁺ cell effects on the antigen-presenting cells because these cells increase substantially in the spleen whereas numbers of B cells and T cells remain constant in control and γδ⁺-cell-depleted mice. Depletion of γδ⁺ cells reduced activated CD4⁺ effector cells, as shown both by the lower numbers of CD4⁺ IFN-γ⁺ cells and CD4⁺ CD69⁺ cells at 7 days after infection. This correlated to a significant increase in CD4⁺ cells with a naïve phenotype (CD44^lo CD62^hi) and a significant decrease in cells with a Tem phenotype (CD44^hi CD62L^lo), which indicated that the presence of the γδ⁺ cells promotes a memory cell response as well as the acute effector cell response. One concern would be that day 7 after infection is considered to be too soon to see memory T cells and raises the question whether memory CD4⁺ cells really are decreased in the absence of γδ⁺ cells. Measuring memory phenotype in spleen cells from mice infected for longer than 7 days is not practical because animal mortality rapidly increases to 100% of infected mice after day 7 of infection. However, it was possible to confirm that memory CD4⁺ cells were reduced in γδ⁺-cell-depleted mice by adoptive transfer of purified CD4⁺ cells from CVB3-infected donors and rechallenge with virus. This produced a dramatically enhanced myocarditis when CD4⁺ cells were isolated from γδ⁺ cell intact, virus-infected donors. However, when CD4⁺ cells were isolated from infected donors lacking γδ⁺ cells, there was no increase in myocarditis or CD4⁺ cell activation (by CD69 or IFN-γ expression) over control infected mice not given CD4⁺ cells. The reduction in memory T cells is not surprising because one would expect fewer memory cells when the effector CD4⁺ cell response is decreased. This study is the first to show that memory CD4⁺ cells can themselves enhance myocarditis with virus rechallenge. Consequently, CD4⁺ cells play an important role in myocarditis susceptibility. The cell dose–response data indicate that transfer of even 5 × 10⁵ purified CD4⁺ cells from infected donors produced an increase in myocarditis in recipients. However, no increase in myocarditis was observed when fewer than 1 × 10⁶ CD4⁺ cells were transferred from γδ⁺-cell-depleted donors. One potential concern could have been that virus was transferred with the CD4⁺ cells from infected mice. To prove that virus transfer was not a problem, separate aliquots of the purified CD4⁺ cells were titred for virus and none was found (data not shown).

Interpreting data from the transfer of whole CD4⁺ cell populations might be misleading because we show that T regulatory cells are increased in mice depleted of γδ⁺ cells. To better control for the complex response in γδ⁺-cell-sufficient and γδ⁺-cell-deficient mice, purified populations of CD4⁺ CD25⁺ (T regulatory cell enriched) and CD4⁺ CD25⁻ (effector cell enriched) cells were transferred and showed that CD4⁺ CD25⁺ cells from γδ⁺-cell-depleted mice were significantly more suppressive than an equal number of CD4⁺ CD25⁺ cells from γδ⁺-cell-sufficient donors. Similarly, while CD4⁺ CD25⁻ effector cells from PBS-treated donors enhanced myocarditis, this population did not enhance disease when isolated from infected γδ⁺-cell-deficient donors. The data indicate that γδ⁺ cell depletion not only increases the number of T regulatory cells, but also substantially enhances their immunoregulatory activity on a per cell basis.

Effectors of the innate immune system can determine whether T regulatory cells are activated. Invariant natural killer T (iNKT) cells are essential for T regulatory cell generation,³⁷ possibly because of high transforming growth factor-β and IL-10 production by the activated iNKT cells.^38,39 Activation of iNKT cells is primarily dependent on the recognition of lipid ligands presented by CD1d, a major histocompatibility complex class I-like molecule, and T regulatory cells fail to occur in CD1d^−/− mice.³⁷ As with iNKT cells, the γδ⁺ cells that were important in CVB3-induced myocarditis also recognize CD1d^27–31 and lyse CD1d⁺ target cells.³² In this light, the ability of γδ⁺ T cells to significantly reduce the numbers of CD1d⁺ cells could inhibit the activation of the CD1d-restricted iNKT cells. The reduction in iNKT cell activation subsequently prevents T regulatory cell differentiation. Once γδ⁺ cells are removed, the increase in CD1d⁺ cells would then aid T regulatory cell activation. Consequently, although both γδ⁺ and iNKT cells are both CD1d restricted, their roles in immunity might be conflicting. Another mechanism by which γδ⁺ T cells might influence T regulatory cell activation is through CD1d⁺ antigen-presenting cells (dendritic cells or macrophages). Antigen-presenting cells clearly impact the activation of T regulatory cells. Dendritic cells stimulated to produce high levels of IL-10 control the differentiation of T regulatory cells³⁴ through modulation of costimulatory factors, CD40, CD80 and CD86.^35,36 Interaction of CD1d-restricted T cells with macrophage can alter macrophage function.^31,33 Interaction of γδ⁺ cells with dendritic cells through CD1d on the latter cells might decrease expression of either immunosuppressive cytokines or accessory molecules that are needed for T regulatory cell activation. The precise mechanisms by which γδ⁺ cells suppress T regulatory cell activation in viral myocarditis require further investigation.

Disclosures

The author declares having no financial or commercial conflict of interest.