We demonstrated, using primary human immune cells, that the RSV NS1 protein induces quantitative and qualitative deficiencies in adaptive immunity. In particular, the NS1 protein (i) suppresses the CD103+ CD8+ T cell response, (ii) promotes a Th2 response, and (iii) suppresses the Th17 response (). Thus, NS1 suppresses two mechanisms that play a protective role during RSV infection (CD103+ CD8+ T cells and Th17 cells) and stimulates a mechanism contributing to enhanced disease (Th2 cells). A role for CD8+ T cells in restricting and clearing RSV infection is well established 
. Following infection with RSV or influenza virus, virus-specific CD8+ CTL accumulate in lungs at much greater concentration than in the peripheral blood 
. Mucosal CD8+CTL, but not their peripheral blood counterparts with the same antigenic specificity, are highly positive for CD103 and have the effector memory phenotype 
. The presence of CD103 on the surface of lung CD8+ CTL is important for respiratory epithelial cell-specific tropism and cytotoxicity due to expression of E-cadherin by epithelial cells, which specifically binds CD103 
. A study with human lung tumor tissues and tumor-specific CD8+ CTL demonstrated that the interaction of CD103 with E-cadherin promotes cytolytic activity by triggering lytic granule polarization and exocytosis 
. Similarly, it was found that, in human bronchoalveolar lavages, a much greater fraction of both CD4+ and CD8+ T cells expresses CD103, as compared to that in the peripheral blood, and that the CD103 molecule is required for effective lysis of E-cadherin-expressing target cells 
. Moreover, CD103 plays role in CD8+ T cell retention in human lung tumors by a CCR5-dependent mechanism 
. Thus, CD103 is a molecule expressed at high levels by mucosal and, in particular, pulmonary CTL, and is associated with their mucosal homing, cytotoxicity against respiratory tract epithelial cells, and their subsequent retention as effector memory T cells. Accumulating evidence thus indicates an important role for CD103 in pulmonary CTL-mediated lysis of virus-infected cells of the respiratory epithelium and thus protection against respiratory viruses. Therefore, the suppression of activation and proliferation of CD103+CD8+ CTL by the RSV NS1 protein demonstrated in this study identifies a new mechanism by which the virus suppresses an important protective component of the adaptive immune response.
The effects of RSV NS1 protein on T cells.
Th17 cells were initially found to have a pro-inflammatory effect and to be involved in autoimmune diseases 
. Subsequent studies elucidated their role in innate and adaptive immune response against pathogens. One of the major functions of IL-17 is the recruitment of neutrophils 
. Infection of bronchial epithelial cells with rhinovirus, a human respiratory pathogen, demonstrated that the IL-17-mediated recruitment of neutrophils is related to its ability to induce IL-8, which is a neutrophil chemoattractant 
. Neutrophils play an essential role in antibody-mediated neutralization of influenza virus in vivo 
and play a critical role in the activation of natural killer cells 
. Using Listeria monocytogenes,
it was found that Th17 cells are much more strongly induced by the respiratory route of infection than by the intravenous route 
. It has only recently been found that Th17 cells play a role in protection against viruses. IL-17 produced by Th17 cells facilitates clearance of vaccinia virus infection in mice, most likely by the attraction of neutrophils 
. Th17 cells have also been shown to protect mice from a lethal dose of influenza virus by a mechanism independent of IFNγ, T cell helper function, or perforin-mediated cytotoxicity, and which possibly involves reduction in the severity of lung damage and/or effect on the rate of lung repair 
. Th17/Th1 cells exhibit functional properties similar to those of Th17 cells 
. Thus, the suppression of proliferation of IL-17+ CD4+T cells (Th17 cells) or IL-17+IFNγ+ CD4+ T cells (Th17/Th1 cells) by the RSV NS1 protein found in this study suggests yet another novel mechanism the virus uses to counteract the anti-viral effects of both the adaptive (the Th17 and Th17/Th1 cells) and the innate (secreted IFNγ) components of the immune system.
Severe RSV disease likely is influenced by a variety of factors whose individual importance varies in the heterogeneous human population and likely also with age. One proposed scenario involves a skewing of the Th1/Th2 balance of the virus-specific response towards Th2, which includes down-regulation of the CD8+ CTL response and a B cell switch from synthesis of virus-specific protective IgA and IgG to non-protective IgE (reviewed in 
). A Th2 bias also has been implicated for a formalin-inactivated RSV vaccine that was evaluated in infants and children in the 1960s and resulted in a greatly increased frequency and severity of disease upon subsequent natural RSV infection 
. Moreover, a number of studies over the past decade have suggested that severe RSV disease may sometimes be linked to various genetic factors, including those controlling the expression of or response to Th2 cytokines 
. In particular, polymorphisms leading to an increased activity of IL-4 and IL-13 appeared to be overrepresented in infants and children with severe RSV disease 
. In the present study, NS1 expressed by wt RSV not only increased the percentage of CD4+ T cells positive for IL-4 (), but also enhanced proliferation of these cells by antagonizing the anti-proliferative effect of IFN-I (). NS1 also depressed the concentration of IFNγ in DC-T cell co-cultures (), thus favoring the Th2 response. These data suggest that the NS1 protein contributes to skewing of the Th1/Th2 balance towards Th2 during the priming of naive T cells or stimulation of memory T cells. The idea that NS1 can skew the T cell response towards Th2 was indicated in previous work. Specifically, when RSV-infected human DC were treated with small interfering RNA targeting the NS1 gene and co-cultivated with heterologous RSV-naïve cord blood CD4+ T cells, the percentage of IFNγ+ cells increased, and that of IL-4+ cells decreased 
. However, what is surprising in the present study is that these effects do not appear to be dependent on IFN-I production and signaling, as described below.
Since the RSV NS genes are IFN-I antagonists, the most obvious explanation for the effects of NS1 deletion observed in the present study would be that they are mediated by increased IFN-I production and signaling, which are known to affect DC maturation and T cell activation, polarization and proliferation. For example, as noted, NS1 suppresses DC maturation, due in part to inhibition of IFN-I production and signaling 
. IFN-I has been reported to have stimulatory or inhibitory effects on CD8+ T cell proliferation, depending on the situation (Introduction
). Type I and type III IFN have been reported to suppress activation and proliferation of CD4+ T cells during co-culture with RSV-infected DC 
. IFN-I has been shown to promote a Th1 response by increasing the frequency of human IFNγ-secreting CD4+ Th cells and antagonizing the suppressive effect of IL-4 on IFNγ production 
. In the present study, one effect indeed did appear to be mediated directly by changes in IFN-I production and IFN-I signaling: specifically, proliferation of total CD4+ T cells was inhibited by IFN-I. However, the other observed effects, namely the suppression of CD103+CD8+ T cells and Th17 cells and enhancement of Th2 cells by NS1, did not appear to be directly mediated by changes in IFN-I production and signaling (). This was suggested by several observations. For example, while both NS1 and NS2, individually and in combination, antagonize IFN-I production and signaling in epithelial cells (Introduction
), the effects in the present study were seen only in response to deletion of NS1 and not the NS2 protein. The inclusion of IFNAR2-blocking antibody in the co-cultures provided evidence that the NS1-specific effects on CD103+CD8+ cells, Th17 cells, and Th2 cells were not due to changes in IFN-I production or signaling. This also was indicated in the experiments in which exogenous IFN-I was added to the co-cultures.
An alternative possibility is that these NS1-mediated effects involve suppression of pathways that do not depend on IFN-I production and signaling. As noted, maturation of DCs in response to RSV is partly independent of IFN-I production and signaling 
. Similarly, Lopez et al. previously demonstrated that the maturation of DC in response to negative strand RNA viruses involves intracellular IFN-I induction pathways, including activation of nuclear factor-kB, but not the released IFN-I or subsequent IFN-I signaling 
. In the present study, infection of DC with the ΔNS1/2 mutant induced a strong up-regulation in the transcription of multiple genes involved in DC maturation and T cell activation (). Much of this up-regulation was suppressed by wt RSV. The role of IFN-I in this up-regulation was assessed with IFNAR2-blocking antibody. The blockade strongly reduced expression of CXCL9, CXCL10 and RIG-I, which are known to be induced mainly by IFN-I 
. In contrast, the blockade only partly reduced the up-regulation of several maturation-related markers including CD38, CD40, and CD80 among others. Thus, we suggest that the effects on the T cells observed in the present study result from NS1-mediated suppression of signaling pathways leading to DC maturation, which likely include signaling pathways leading to induction of IFN-I, but are independent of secreted IFN-I and IFN-I signaling. IL-12 and IL-23 also may be involved. IL-12 and IL-23 are involved in polarization of T cell response toward Th1 
and maintenance of the Th17 cell population 
, respectively. The expression of each by RSV-infected DC was increased by deletion of NS1/2. This did not appear to be a result of increased IFN-I production and signaling because the IFNAR2 blockade increased rather than decreased expression, indicative of suppression rather than stimulation by IFN-I (). Suppression of IL-12 expression by IFN-I also has previously been reported in humans and mice 
. NS1 also reduced secretion of IFNγ by both CD8+ and CD4+ T cells. IFNγ promotes the differentiation of Th cells to Th1 and suppresses their development to Th2 phenotype, and IFNγ secreted by CD8+ T cells prevents Th2-driven pathology during RSV infections 
. Thus, suppression of IFNγ secretion by CD4+ T cells, which was not reversed by exogenous IFN-I, may be involved in the Th2-promoting effect of the NS1 protein demonstrated in this study.
In conclusion, the data presented in this study suggest that the RSV NS1 protein has quantitative and qualitative effects on the adaptive immune response, thus promoting infection and disease. The experimental system used in this study involved human primary cells from adult donors and thus represents a model of the immune system of RSV-immune children and adults for investigation of the effects of immunomodulating RSV proteins on the virus-specific memory T cells. However, the pattern of immune response to RSV depends on age and is biased towards Th2 in neonates 
. Therefore, it is possible that, in RSV-naïve individuals, the three major effects of the NS1 protein, especially that of enhanced Th2 activity, might be even more pronounced than in adults.