Natural regulatory T cells (nT_regs) develop in the thymus through interactions between the high-affinity T cell receptor and cognate antigens on thymic epithelial cells. Co-stimulation through CD28 and common gamma (cγ) chain cytokines, especially IL-2 and IL-7 are necessary for nT_reg development. Signal transducer and activation of transcription 5 (STAT5) signaling through the cγ chain is also required. Thymic development of nT_reg cells follows a two-step process. First, thymocytes upregulate CD25 and other IL-2 signaling molecules in response to TCR/CD28 co-stimulation and second, CD4⁺CD25^highFoxP3^- T_regs respond to IL-2 independent of the TCR, and induce FoxP3 expression in response to STAT5 activation [1]. T_reg cell-intrinsic NF-κB activation is essential for thymic T_reg development [1]. Naturally occurring T_regs cannot produce IL-2 and therefore rely on paracrine IL-2 production from conventional T cells. Mice deficient in IL-2 or CD25 have reduced numbers of FoxP3⁺ T cells and a dramatic reduction in peripheral and thymic T_regs. In humans, Hassall’s corpuscles in the thymic medulla secrete thymic stroma lymphopoietin (TSLP) which activates immature dendritic cells and upregulates the expression of co-stimulatory molecules. The activated DCs induce FoxP3 expression in CD4⁺CD8^-CD25^- thymocytes [8].

Less well-characterized are the inducible T_regs (iT_regs) that develop from conventional T cells under certain conditions. iT_regs are induced by prolonged exposure to circulating antigen, chronic inflammation, or weak co-stimulation in the periphery [2]. Soluble factors such as the cytokines IL-4, IL-10 and TGF-β, retinoic acid or neuropeptides can upregulate FoxP3 expression and generate iT_regs in the periphery. Increased expression of FoxP3 results in upregulation of other T_reg molecules, including CTLA-4, GITR, and CD127. CD4⁺ T cells that express high levels of the IL-2α receptor (CD25^high) do not respond to T cell receptor (TCR) activation or mitogen stimulation, and inhibit IL-2 transcription in CD25^- cells. Suppression of CD25^- cells is contact dependent, and requires activation of the T_regs through the TCR; however, once activated, the suppressor effector function is nonspecific [9]. CD4⁺CD25⁺ T cells suppress the immune response to some viruses, protozoa, and bacteria, and aid the survival of intracellular pathogens [10] most likely by potent suppression of proliferation and IFN-γ production of both CD4⁺ and CD8⁺ T lymphocytes [11]. Tr1 cells secrete high amounts of IL-10 and moderate amounts of TGF-β. Inhibiting IL-10 with neutralizing antibody blocks the suppressor effects of Tr1 cells [4]. Th3 cells produce high concentrations of TGF-β and moderate amounts of IL-10, and the suppressor effects are not antigen specific [4]. Interestingly, Th3 cells suppress the activation of both Th1 and Th2 cell clones while other subsets primarily inhibit Th1 cells and have no effect on Th2 cells [4]. Other cells with adaptive regulatory function include some CD4^-CD8^- and CD8⁺CD28^- T cells [8]. Early after iT_regs are stimulated, they express high levels of cell-cycle progression and T cell activation-associated genes [12], mimicking genes that are upregulated in activated effector T cells. As iT_regs mature, expression of these genes diminishes while they remain high in mature effector T cells. By 10 days after differentiation into iT_regs, most cell cycle progression and T cell activation genes are expressed at levels approximately 3 times lower than in effector T cells. In addition, genes in the FoxO family of transcription factors are over-expressed in iT_regs compared to overexpression of the FoxM1 family in effector cells [12].

Regulatory T cells typically increase late in chronic viral disease to prevent a persistent inflammatory response and viral-mediated immunopathology. In fact, tissue-protective effects of T_reg were shown in models of respiratory syncytial virus, Friend virus, and West Nile Virus infection [13,14]. Additionally, T_reg responses to viruses (and bacteria) form the basis of the “hygiene hypothesis”. Infection with influenza A in suckling mice protected these mice as adults against allergy induced airway hyperactivity due to the expansion of allergen-specific regulatory T cells [15]. Conversely, T_regs are widely accepted as a key contributor in modulating the host immune response to viral infection. T_regs are an important component in regulating the magnitude of the immune response to infection, thus preventing excessive inflammation and tissue damage. But, they can also be inappropriately induced by viruses in order to swing the balance of the immune response in favor of maintaining viral infection [16]. In this way T_regs contribute to persistent infection of many viruses including Friend virus, herpes simplex virus, hepatitis C virus, hepatitis B virus, human immunodeficiency virus, feline immunodeficiency virus, simian immunodeficiency virus, cytomegalovirus and Epstein-Barr virus [13,14,17,18,19,20,21,22,23,24,25,26].

Within a susceptible herd, reproductive failure due to PRRSV infection can range from sporadic abortions to abortion storms that may persist within the herd for up to 6 months [27]. Exposure late in gestation may manifest as late-term abortion, or stillborn, partially autolyzed, or mummified fetuses [27]. Neonatal infection results in severe dyspnea and tachypnea, and mortality of up to 100%, while disease in weaned pigs is primarily due to pneumonia and secondary bacterial or viral infections [27]. PRRSV infection is also associated with decreased local cellular immunity, resulting in increased susceptibility to secondary bacterial and viral infections [28,29,30]. Piglets infected with PRRSV in utero have a decreased innate immune response to bacterial pathogens [28]. In utero infection with PRRSV inhibits macrophage phagocytosis of Salmonella spp., and inhibits alveolar macrophage oxidative burst [28]. Additionally, both PRRSV infection and vaccination decreases the efficiency of vaccines against Mycoplasma hyopneumoniae [31] and porcine pestivirus [32]. Infection and vaccination with PRRSV induces a rapid, non-neutralizing antibody response, and an early, weak, non-specific gamma interferon (IFN-γ) response [33,34]. A PRRSV-specific T lymphocyte IFN-γ response does not appear until at least 2 weeks after infection, [35] gradually increases, and plateaus at 6 months postinfection. This IFN-γ is associated with a slow increase in neutralizing antibody [33,36]. Peak viremia and shedding occur before development of a protective neutralizing antibody and IFN-γ [36]. Acute infection is followed by persistence of lower levels of virus in lymphoid tissue and then clearance after several months [36]. The cause of the delayed IFN-γ and neutralizing antibody response resulting in persistent infection is unknown. The ability to induce a rapid IFN-γ response is important both for viral clearance and heterologous protection by vaccination [37,38]. Vaccine strains currently in use in the United States do not provide adequate heterologous protection, perhaps because of their inability to stimulate an adequate IFN-γ response. We hypothesized that one reason for the inadequate IFN-γ may be due to the ability of PRRSV to stimulate regulatory T cellsin vitro [39].

PRRSV infection results in a significant upregulation of IL-10 expression in peripheral blood mononuclear cells and pulmonary alveolar macrophages in vivo and in vitro [40,41,42,43]. However, the ability of PRRSV to induce IL-10 appears to vary depending on strain and it was recently demonstrated that a highly virulent PRRSV strain did not induce IL-10 in vitro or in vivo [44,45,46]. The primary action of IL-10 is to inhibit inflammatory cytokines including IL-1β, IL-6, TNF, and IL-12, and antagonize the function of antigen presenting cells by decreased surface expression of MHC class I and II molecules and reduction of costimulatory molecules [47]. Additionally, IL-10 inhibits the production of nitric oxide by macrophages [48,49]. Since IL-12-mediated production of IFN-γ by pulmonary alveolar macrophages reduces PRRS viral titers in the lungs and serum [50], inhibition of IL-12 synthesis by IL-10 likely enhances the occurrence of natural disease. IL-10 also affects innate immunity by inhibiting the response to Toll-like receptors (TLR), including TLR7 and TLR8 that recognize single-stranded viral RNA [51]. IL-10 treated DCs induce the proliferation of regulatory T cells (T_regs) as well as antigen-specific anergy in CD4⁺ and CD8⁺ T lymphocytes [47]. IL-10 not only plays a role in the development of type 1 T_regs (Tr1), but is also one of the primary mechanisms by which T_regs inhibit effector T lymphocyte function [51].

In previous experiments, we have shown that PRRSV infection increases the number of CD4⁺CD25⁺FoxP3⁺ regulatory T cells in the lungs and PBMCs of pigs (LeRoith, unpublished). Silva-Campa demonstrated that the CD25⁺FoxP3⁺ cells induced by PRRSV produce TGF-β, suggesting that the cells are not only increased, but are also functional [39]. These CD25⁺FoxP3⁺ T_regs exhibited suppressor activity in vitro. This same group reported that while American genotype PRRSV strains were capable of inducing T_reg and upregulating TGF-β production, DCs infected with European genotype PRRSV induced neither TGF-β nor T_regs [52]. Wongyanin and colleagues demonstrated that peripheral blood mononuclear cells (PBMCs) cultured with American genotype PRRSV in vitro induced virus-specific CD4⁺CD25⁺FoxP3⁺ T_regs and the addition of monocyte-derived dendritic cells (MoDC) to the cell culture enhanced T_reg induction [53]. Not only was there a significant increase in the numbers of CD4⁺CD25⁺FoxP3⁺ cells, but these PRRSV-specific T_regs exhibited suppressive activity when co-cultured with PHA-stimulated autologous peripheral leukocytes. The authors reported in this study that PBMCs collected from PRRSV-infected pigs 10 days post-inoculation exhibited significantly higher numbers of CD4⁺CD25⁺FoxP3⁺ lymphocytes when cultured in the presence of PRRSV compared to mock-infected cell lysate or PBMCs alone. The role of PRRSV nucleocapsid protein (N) in T_reg induction was recently described, and it was found that N-protein induced IL-10 producing cells and CD4⁺CD25⁺FoxP3⁺ T_regs in a DC in vitro system [54]. In this study, T_reg induction was found to be dependent, at least in part, on IL-10, as neutralization of IL-10 by anti-IL-10 antibody drastically reduced the PRRSV-induced T_regs.

In order to assess the potential benefit of using a mucosal rather than parenteral immunization approach, Dwivedi et al. investigated the ability of an intranasally delivered PRRS-modified live vaccine (MLV) augmented with a potent Mycobacterium tuberculosis whole cell lysate adjuvant to induce cross protective immunity against a heterologous PRRSV strain. Vaccinated pigs had a reduced frequency of T_regs in respiratory mucosal and systemic sites compared to unvaccinated pigs, and this correlated with decreased secretion of immunosuppressive cytokines IL-10 and TGF-β, diminished lung pathology, and increased PRRSV neutralizing antibody titers and IFN-γ secretion [55,56]. These findings suggest that the route of immunization and adjuvant-mediated immunomodulation may influence T_reg dynamics, thereby facilitating or negating efficient viral clearance. This same group recently reported on pigs that that were maintained on a commercial farm and experimentally infected with PRRSV.

At two days post-inoculation, infected pigs had an increased frequency of circulating CD4⁺CD25⁺Foxp3⁺ T_regs, reduced frequency of CD4^-CD8⁺ and CD4⁺CD8⁺ T cells, and enhanced IL-10, IL-4 and IL-12 secretion [57].

It is unknown which components of the virus are responsible for T_reg proliferation, but this is currently under investigation by our group. We demonstrated that T_reg induction by PRRSV results in increased susceptibility to natural Mycoplasma hyopneumoniae infection [6]. In this study, pigs were inoculated with a virulent strain of PRRSV and a modified live vaccine derived from the same strain. The attenuated strain contains silent mutations in the replicase region (ORF 1) [58] and conservative or non-conservative amino acid changes in the structural proteins [58]. We hypothesized that even with changes that decrease pathogenicity, the mutations would not alter the virus’s ability to stimulate T_regs. Consistent with our hypothesis, we found that, although attenuated, the vaccine strain did not differ from the parent strain in its ability to activate T_regs. The animals inoculated with the attenuated vaccine did not differ from animals inoculated with the parent strain in the severity of M. hyopneumonia-mediated disease [6]. Similar to previous findings that infection with wild type PRRSV or vaccination with PRRS MLV vaccines has been shown to decrease the efficacy of M. hyopneumoniae vaccines [31], inoculation with each of the three PRRS viruses in this study resulted in activation of regulatory T cells and likely decreased the ability of the pigs to mount an effective anti-bacterial immune response.

Vaccine efficacy appears to be related to the ability to stimulate IFN-γ production and efficacy against heterologous virus challenge seems to correlate more with the ability to stimulate IFN-γ than homology of the vaccine strain to the infective strain. Current vaccines fail to protect against other strains [59], which may be due to their inability to stimulate IFN-γ production. Our study was the first to show the correlation between vaccine induction of T_regs and increased susceptibility to bacterial infection. Although T_regs that are induced by PRRS in vitro produce TGF-β [39], the induction of T_regs may indirectly result in IL-10 production, a phenomenon that is well established in the PRRS literature [41,42]. Production of IL-10 instead of IFN-γ by the MLV vaccine strain would lead to a lack of heterologous protection, and decreased efficacy of other vaccines, as seen by other authors [31]. Our results suggest that mutations in the vaccine strain that result in attenuation of the virus do not alter the virus’s ability to stimulate T_regs [6]. This information can help us design vaccines in which the T_reg-stimulating epitopes are mutated or deleted in order to stimulate a robust virus-specific IFN-γ response, and provide protection against heterologous strains [6].

Lactate dehydrogenase-elevating virus (LDV) has been described as an “ideal” persistent virus since it is associated with life-long viral infection in mice in the absence of clinical disease while escaping the host immune response [60,61]. Cytotoxicity is limited to resident tissue macrophages, in which viral replication is maintained. Lifelong immunotolerance is maintained and has been shown to be due in part to continuous generation of LDV antigens in the thymus [62]. In addition, LDV infection and replication within macrophages appears to be robustly resistant to typical antiviral immune responses including interferon-α/β, antiviral antibodies and virus-specific cytotoxic T lymphocytes [60]. Collectively, these features suggest a role for T_regs in the pathogenesis of LDV infection.

Inada and colleagues demonstrated that mice challenged with heat-inactivated or killed LDV induced a strong virus-specific delayed-type hypersensitivity (DTH) reaction, whereas a delayed hypersensitivity reaction was undetectable in young (1–3 month-old) mice challenged with live virus [63]. In these experiments pretreatment with cyclophosphamide partially restored the DTH response, which was attributed to cyclophosphamide mediated elimination of suppressor T cells [64,65]. Live virus also elicited a DTH response in old mice (> 8 months) without cyclophosphamide treatment, which is not surprising in light of the fact that older mice have reduced suppressor T cell activity. The cumulative findings from this study suggest that inoculation with live LDV induces suppressor T cells, and that this dampens the virus-specific delayed hypersensitivity response [63].

Further demonstrating the immunosuppressive ability of LDV, Robertson, et al. reported that co-infection with LDV and Friend virus (FV) delayed the FV specific CD8⁺ T cell response and that this resulted in an increase in duration and severity of the acute phase of FV infection [66]. The suppressed FV-specific CD8⁺ T cell response occurred in mice acutely co-infected with LDV and FV as well as in mice inoculated with LDV 8 weeks prior to FV infection. In addition, mice infected with LDV exhibited significant regulatory T cell-mediated suppression of IFN-γ production by FV-specific CD8⁺ T cells that peaked at day 3 post-infection and was diminished by day 7 post-infection. However, failure of FV/LDV co-infected mice to mount a strong CD8⁺ T cell response was not attributed solely to T_reg-mediated suppression, because neither depletion of CD4⁺ cells nor pre-treatment with anti-CD25 antibody restored the normal CD8⁺ response [66].

The antigen presenting ability of spleen, lymph node and peritoneal macrophages from LDV infected mice was shown to be diminished as measured in vitro by reactivation of memory T cells. The experiments showed that LDV-mediated impairment of antigen presentation was not due to diminished uptake of antigen by macrophages and LDV-infected peritoneal macrophages were not immunosuppressive in cell culture. Rather, the authors concluded that the decrease in antigen presenting ability of LDV-infected macrophages was related to reduced expression of Ia antigen or virus-mediated elimination of Ia positive macrophages from the peritoneum [67].

Much attention on the interaction of T_regs and viral pathogens has focused on chronic disease, demonstrating the ability of T_regs to delay or prevent viral clearance leading to persistent infection [16,68]. From the host point-of-view, virus-induced T_regs represent a detrimental factor in the context of chronic or persistent infection. However, coronavirus infection in the central nervous system of mice highlights a beneficial role of T_regs in reducing bystander damage as a consequence of acute infection. Coronavirus infection of the central nervous system exemplifies the necessity of a delicately balanced and finely orchestrated immune response. While a rapid and strong pro-inflammatory response aids in viral clearance, host tissue destruction secondary to immunopathology can be a deleterious side effect. An appropriate anti-inflammatory response is necessary for minimizing collateral damage while still allowing clearance of the invading pathogen.

Mice infected with the neurovirulent strain of mouse hepatitis virus JHM (JHMV) develop a rapidly progressive, fatal disease that has been shown to be mediated, in part, by CD4⁺ T cells. This was evidenced by the fact that infection with a recombinant JHMV strain containing a single mutation in an immunodominant CD4⁺ T cell epitope (rJ.M_Y135Q) resulted in nonlethal mild encephalitis, and the decrease in mortality correlated with decreased numbers of virus-specific CD4⁺ T cells in the brain [69]. Further investigations found that significantly higher levels of the pro-inflammatory cytokines/chemokines IL-6, CCL2, CCL5 and IFN-γ were detected in the brains of mice infected with the non-mutated recombinant JHMV strain (rJ) compared to rJ.M_Y135Q-infected mice [70]. CD4⁺ T_regs were shown to be critical in ameliorating disease because (1) greater numbers of T_regs were detected in the brains of rJ.M_Y135Q-infected mice compared to rJ-infected mice, (2) their depletion in rJ.M_Y135Q-infected mice increased morbidity and mortality, and (3) adoptive transfer of T_regs into rJ-infected mice increased survival from 0% to 50% [70]. The authors concluded that in the setting of acute encephalitis, T_regs may aid in limiting immunopathology, thus decreasing clinical disease without delaying viral clearance.

In contrast to the neurovirulent JHMV strain, mice infected with an attenuated JHMV variant (J2.2-V-1) develop chronic demyelinating encephalomyelitis and the demyelination is largely due to immunopathology associated with viral clearance [71,72,73]. Trandem et al. demonstrated that adoptive transfer of T_regs ameliorated clinical disease and demyelination in J2.2-V-1-infected mice and did not delay viral clearance in immunocompetent C57BL/6 mice [74]. The authors provided evidence that this improved clinical outcome was due in part to T_regs functioning in the draining cervical lymph nodes to suppress T cell proliferation, dendritic cell activation and expression of pro-inflammatory mediators. These findings further support the notion that T_regs play an important role in limiting neuropathology associated with mouse hepatitis virus (MHV) infection while still allowing for viral clearance.

Identification and characterization of pathogen-specific epitopes targeted by T_regs has received considerable interest due to possible therapeutic interventions aimed at diminishing inflammation-induced tissue damage in a pathogen-specific fashion. Zhao and colleagues identified T_regs that specifically recognize two mouse hepatitis virus-specific epitopes using the neurotropic rJ2.2 strain of mouse hepatitis virus, which is known to cause mild acute encephalitis and chronic demyelination [7,75]. These virus-specific T_regs were present in the virus-infected central nervous system and, based on concurrent detection with virus-specific effector CD4⁺ T cells and identification within the naïve T cell precursor pools in the spleen and lymph nodes, are presumed to arise from the natural T_reg pool. In addition, the virus-specific T_regs expressed IL-10 and IFN-γ upon stimulation with viral peptide and suppressed proliferation of cognate-epitope specific effector CD4⁺ T cells. While T_regs are known to play a critical role in reducing immunopathology and clinical disease in rJ2.2-infected mice, the results from this study suggest that virus-specific T_regs may be significantly more potent in diminishing immunopathology associated with encephalomyelitis compared to adoptive transfer of natural T_regs, particularly during acute infection when maximum viral antigen is present [7,74,76].

In contrast to the protective effect of T_regs in the pathogenesis of neurotropic mouse hepatitis virus infection, Shalev and colleagues demonstrated that T_regs contributed to more severe fulminant viral hepatitis in susceptible mice infected with murine hepatitis virus strain 3. This phenomenon was mediated in part by increased T_reg expression of the immunosuppressive cytokine fibrinogen-like protein 2 [77]. These findings suggest that T_reg activation following infection with mouse hepatitis virus may culminate in different outcomes depending on the anatomic location of disease, such that they are paradoxically beneficial to the host in limiting CNS disease but harmful in potentiating fulminant hepatitis.