In a systematic, step-wise approach, we identified a paradigm that leads to reliable and reproducible induction of EAE by adoptive transfer in C57BL/6 mice, a mouse strain that is widely considered relatively disease resistant to this method of EAE induction. We made several interesting observations that led to the development of this experimental approach.
We observed that PT injections of the donor mice were required for successful adoptive transfer of EAE into C57BL/6 mice. PT treatment of recipient mice had no apparent effect on disease susceptibility. PT is predominantly used as an adjuvant for EAE disease induction by active immunization. Traditionally it was thought that PT disrupts the blood-brain barrier (BBB) by amplifying an inflammatory antigen-specific autoimmune response [22
]. Other proposed mechanisms of PT in EAE disease induction include; the prevention of anergy, the induction of autoantigen-specific T cells [23
], the abrogation of T-cell tolerance [24
], and the irreversible inhibition of G proteins and other second messengers than may affect signaling pathways in the T cell or the APC that regulates T-cell differentiation [25
]. In this regard, Shive and coworkers demonstrated that PT stimulates APCs resulting in the differentiation of T helper cells towards a Th1 phenotype [26
]. More recently, Hofstetter and colleagues showed that co-injection with PT during active immunization activates APCs in the peripheral lymphoid organs and the CNS by inhibiting signaling through G proteins rather than signaling through Toll-like receptors [27
]. In the adoptive transfer model of EAE, the transferred antigen-specific encephalitogenic T cell is less dependent on reactivation in CNS than in the active immunization model but interaction with DCs in the perivascular spaces and meninges is necessary for the entry of encephalitogenic T cells into the CNS parenchyma [28
]. Our data suggest that PT treatment of the donor alone is sufficient to promote an encephalitogenic T cell phenotype that when transferred into healthy C57BL/6 recipients can result in the development of EAE.
With regard to cytokine expression by T cells to be adoptively transferred we focused on IFNγ, IL-17, and GM-CSF. In the past two decades, TH1 cells that express IFNγ, tumor necrosis factor beta (TNF)-ß, IL-2, and nitric oxide [29
] have been implicated in EAE pathogenesis. IFNγ is considered the signature cytokine of TH1 cells, which activate myeloid cells to promote cell-mediated immunity. Perhaps the most convincing evidence to support a pathogenic role of IFNγ in CNS autoimmune disease was derived from an open-label, randomized clinical trial, in which 18 patients with MS received 1 μg, 30 μg, or 1000 μg of recombinant IFNγ i.v. twice a week for four weeks [30
]. During treatment, 7 patients experienced an exacerbation. The trial investigators also detected an increase in circulating monocytes bearing major histocompatibility complex (MHC) II surface antigen, suggesting that clinical attacks during treatment were immunologically mediated. Interestingly, studies performed in IFNγ deficient mice or in mice treated with anti-IFNγ mAb developed more severe EAE [31
]. Therefore the contribution of IFNγ to EAE pathogenesis remains unclear. We were able to detect both IFNγ in cell culture supernatants in all experimental paradigms, suggesting that TH1 cells were present in the adoptively transferred cells. Our data suggest that IFNγ expression by T cells is not sufficient for encephalitogenicity in this EAE model.
More recently, TH17 cells were identified as a distinct lineage of CD4+
T helper cells that may facilitate the initiation and perpetuation of CNS autoimmune diseases [35
]. TH17 cells synthesize IL-17, which mediates proinflammatory and allergic responses. The role of IL-17 as a causal factor in the etiology of EAE has remained somewhat controversial as mice deficient in IL-17 can develop EAE [32
]. In our model of adoptive transfer and in the C57BL/6 mouse strain it appears that high expression of IL-17 by transferred donor cells is not absolutely required for disease induction. Furthermore, high expression of IL-17 by adoptively transferred day 6 LNC is not sufficient to induce CNS autoimmune disease. However, adoptive transfer EAE may be a multi-phasic disease. O'Connor et al recently showed in a C57BL/6 adoptive transfer model that purified antigen-specific TH1 cells are highly pathogenic, whereas transfer of purified TH17 cells does not cause disease [39
]. Activated TH1 cells could readily access the non-inflamed CNS of recipient mice, whereas TH17 cells only appear in the CNS of mice with established EAE. These data and our own observations suggest that a high number of TH17 donor cells may not be required for disease induction, but may be critical to sustain CNS inflammation. TH17 cells may be recruited from the host during the amplification of the initial inflammatory response.
Other investigators recently showed that GM-CSF may play critical role in different models of active and passive EAE [41
]. Our own results indicated that GM-CSF is secreted by LNC and SPC after antigen re-stimulation in the presence of IL-12. While GM-CSF was highly expressed by day 12 encephalitogenic LNC in our EAE model, it was also highly expressed by day 12 non-encephalitogenic SPC.
The transcription factor T-bet has recently been shown to be critical for encephalitogenicity [20
]. Its precise role in EAE and MS pathogenesis remains to be fully elucidated. T-bet has pleiotropic biological effects, among one of which is the regulation of expression of the IL-23 receptor [21
]. In turn, IL-23 was recently shown to regulate the expression of GM-CSF by CD4+
]. When we analyzed both splenocytes and LNC for expression of T-bet immediately before transfer into recipient mice, we found that T-bet+
cells were maintained in the day 12 LNC, the only cell population able to transfer EAE. While T-bet+
cells were present in the CD4+
populations of the day 6 splenocytes and LNC, as well as day 12 splenocytes, these cells failed to transfer EAE.
In summary, we have identified an algorithm that leads to a reproducible induction of adoptive transfer EAE in C57BL/6 mice. There was no single biomarker associated with donor T cell encephalitogenicity. Instead, adoptively transferred day 12 LNC have the potential of reproducibly inducing EAE in the C57BL/6 mice are characterized by a composite of inflammatory markers that include the expression of GM-CSF and T bet. Our data further suggest that GM-CSF expression by CD4+ T cells regulated by IL-23 contributes to their encephalitogenicity in our EAE model. Finally, there was an increased prevalence of MOGp35-55-specific CD4+ T cells in day 12 LNC after in vitro re-stimulation. Our protocol is currently being verified in genetically-modified mice on the C57BL/6 background.