The above results reinforce that CD8 T cells play the predominant role in the pathogenesis of T1BH formation, and clarifies that the leading contribution in this process is by epitope specific cytotoxic CD8 T cells. Since a small degree of T1BH formation is observed even in control animals (infected RAG-1 −/− mice receiving sham transfer) it is likely that either direct cytopathic effects or cells of the innate immune system, likely neutrophils, also contribute to this process. Meanwhile, when CD4 cells are transferred to animals with deficient adaptive immunity, they appear to play a preventive role in T1BH formation and significantly reduce the observed baseline T1BH formation.
T1BH-s are thought to represent axonal and/or neuronal damage in MS 
. In the TMEV model, viral replication mostly takes place in neurons in the stage when our experiments were conducted 
, and at this stage and in this strain, demyelination is not observed, unless stimulated by the intravenous injection of the immunodominant VP2 peptide, as demonstrated by our team earlier 
. We therefore hypothesize that cytotoxic CD8 T cells targeting the immunodominant VP2 epitope are targeting neurons and axons, consistent with our recent published studies 
. By transferring CD8 T-cells into CD8 deficient infected hosts, we were able to restore T1BH formation. In addition epitope specific elimination of VP2 specific CD8 T-cells resulted in reduced T1BH formation. Altogether, these observations very strongly suggest that T1BH formation in this model is related to neuronal and axonal injury caused directly or indirectly by CD8 T-cells. Given that CD8 T-cells are the most numerous lymphocytes in MS lesions regardless of stage of lesion formation, and given that CD8 T-cells have been observed strongly apposed to neurons and axons in MS tissue, we propose that a similar CD8 T cell driven mechanism as observed in our model may be responsible for T1BH formation in MS 
. Another key analogy between our model and MS-related T1BH formation is that in both cases, axonal injury appears to be the main tissue damage captured by this MRI finding, which is further underlined by the fact that demyelination is absent at this stage and in this strain in our model. 
The surprising protective role of CD4 T cells may be explained by previously published findings related to CD4 T-cell biology in TMEV infection. CD4 T-cells can play both protective and pathogenic roles in TMEV infection. During the acute neuroinflammatory stage of TMEV infection, CD4+ Th1 T-cells appear to control the viral infection 
, although we did not notice significant differences in viral load in our experiment in day 7; therefore the observed protective role is unlikely to be related to this effect. Furthermore, experimental neutralization of a classic CD4 T-cell cytokine, IFN-γ, resulted in significantly accelerated disease onset 
. IFN-gamma was also demonstrated to play a critical role in protecting spinal cord neurons from persistent TMEV infection and death 
. On the other hand, these cells also play a pathogenic role in the chronic stages of TMEV infection of susceptible mice, where neutralization of IL-12, a classic CD4 Th1 cytokine, resulted in an attenuated disease phenotype with decreased demyelination 
. CD4 T-cells with a Th2 phenotype have been proposed to overall suppress inflammation and demyelination: IL-4 treatment during the early chronic phase of TMEV infection resulted in a more benign disease phenotype, with reduced anti-TMEV antibody responses 
. However, this observation is not applicable to our experimental system for two reasons: one, the utilized RAG-1 deficient mice are unable to mount an antibody response; two, we are studying the acute neuroinflammatory stage of TMEV infection. Overall, in the classic TMEV induced biphasic MS models, it remains unclear whether Th2 cytokines play a protective (suppression of pathogenic CD4 Th1 cells) or pathogenic role (by increasing antibody production) in demyelinating disease 
A recent novel observation suggests that Th17 cells, a subset of CD4 T cells prolong neuronal survival in the TMEV model in vitro 
. Such neuroprotective effect may also have contributed to the observed attenuated MRI phenotype in our study. In addition, Th17 mediated suppression of CTL function was also observed in TMEV infection, which may also theoretically lead to reduced T1BH formation, but not in our model system, as endogenous CD8 T-cells are not present in our host mice.
In the TMEV model, BALB/cByJ mice are resistent while BALB/cAnNCr mice are susceptible to the development of chronic demyelination. The susceptibility of BALB/cAnNCr mice is thought to be directly caused by a defective/absent CD4+ T-cell subset, providing another example how CD4+ T-cells may regulate the overall phenotype observed in this model, acting in the earliest, most acute stage immediately after infection 
We constantly observe low level but clearly detectable T1BH formation in RAG-1 deficient mice () 
. We therefore suspect that cells of innate immunity, likely neutrophils, macrophages or microglia may contribute to axonal/neuronal damage and resultant T1BH formation, which is an active area of research in our laboratory. We also hypothesize that the presence of CD4 T-cells in our adaptive transfer model may result in the attenuation of this innate immune response leading to the observed reduction in T1BH formation. The exact mechanism for the above is to be determined in ongoing and future experiments in our laboratory.
In conclusion, these findings demonstrate a contrasting role for CD4 vs. CD8 cells in T1BH formation in the TMEV model of MS. Further investigations in our model will clarify the role and significance of CD8 T-cells in axonal and neuronal damage, and the role of CD4 T-cells in the observed protective process.