CD8+ T cells directed against a viral peptide contribute to loss of motor function by disrupting axonal transport in a viral model of fulminant demyelination

Abstract

Demyelination, a pathological hallmark of multiple sclerosis, may be a necessary but not a sufficient condition for motor dysfunction associated with this disease. We favor a neurodegenerative model of multiple sclerosis and suggest that demyelination creates a permissive environment wherein the denuded axon becomes susceptible to immune-mediated injury. Unfortunately, the cellular effectors responsible for eliciting such axonal injury are currently unknown. Based on previous observations implicating cytotoxic T cells in this injury, we assessed motor function, axon dropout, and axon injury following peptide depletion of the immunodominant CD8+ antiviral T cell response in the IFNgamma receptor-deficient mouse model of acute demyelination. We found that the targeted removal of this population of cytotoxic effector cells prior to infection with the Theiler's murine encephalomyelitis virus caused a substantial preservation of motor function at 45 days postinfection that was associated with preservation of retrograde axonal transport in a subpopulation of surviving axons within the spinal cord. We conclude that cytotoxic T cells may be responsible for the initiation of axon injury following demyelination.

Figure 1

VP2_121–130 peptide treatment depletes virus-specific CD8+ T cells from the CNS. IFNγR^−/− mice were treated prior to infection and throughout the course of infection with a VP2_121–130 peptide to deplete the immunodominant population of CD8+ T cells or with an irrelevant peptide derived from the E7 protein of HPV. Brain infiltrating lymphocytes were collected at 7 days postinfection, stained with anti-CD8 and a tetramer that recognizes VP2_121–130-specific T cells, and analyzed by flow cytometry. As shown in panel A and quantified in panel D, approximately 40% of CD8+ T cells were VP2 tetramer-positive in E7-treated mice. In contrast, as shown in panel B and quanitfied in panel D, the population of VP2 tetramer-positive T cells was almost completely depleted in VP2 peptide-treated mice (P<0.001 vs. E7-treated by t-test). Likewise, as shown in panel C, since VP2_121–130-specific T cells represent the immunodominant population of T cells in the brain at 7 days postinfection, VP2 peptide depletion effectively reduced the overall population of CD8+ CD45^hi cells present in the brain. The polygon in panel A and panel B indicates CD8+ VP2 tetramer-positive cells used for the analysis.

Figure 2

The absence of VP2-specific CD8+ T cells confers preservation of motor function. Function was assessed in VP2- or E7-treated mice prior to infection and at 21 and 45 days postinfection by rotarod analysis. At 45 days postinfection E7-treated mice performed at less than 13% of baseline function, while VP2 peptide-treated mice functioned at almost 50% of baseline (P=0.035 E7 vs VP2 by t-test).

Figure 3

Peptide depletion of VP2-specific CD8+ T cells does not prevent the loss of spinal cord axons. Spinal axons were counted and binned by cross-sectional area in uninfected mice (B6) and in E7 or VP2 peptide-treated mice at 45 days postinfection. Both large (>10 μm²) and small (1–4 μm²) axons were lost in infected mice and the extent of loss did not differ between treatment groups. While medium-sized axons also trended toward reduction in the infected mice this difference was not statistically significant.

FIgure 4

Peptide depletion of VP2-specific CD8+ T cells does not alter the extent or quality of demyelination in the spinal cord. Photomontages of 10x fields collected from uninfected (A) and E7-treated (C) or VP2-treated (E) mice at 45 days postinfection show that the amount and distribution of demyelination in the spinal cord was not grossly changed by peptide treatment. Likewise, higher magnification images of normal myelin in an uninfected mouse (B) and border areas at the junction of demyelinated and normally myelinated areas in E7-treated (D) and VP2-treated (F) mice show that the quality of demyelination in the infected mice was not changed by peptide treatment. The scale bar in E is 500 μm and refers to A, C, and E. The scale bar in F is 100 μm and refers to B, D, and F.

Figure 5

Peptide depletion of VP2-specific CD8+ T cells does not alter gross axonal pathology in the spinal cord. Analysis of Bielschowsky stained (A, C, E, and G) or hematoxylin and eosin stained (B, D, F, and H) spinal cord sections at low magnification (A, B, C, and D) or higher magnification (E, F, G, and H) did not reveal any obvious differences in the extent or pattern of axonal injury between E7-treated mice (A, B, E and F) and VP2-treated mice (C, D, G and H). Images were collected in the ventrolateral region of cervical spinal cord in both treatment groups. The scale bar in D is 500 μm and refers to A–D; the scale bar in H is 100 μm and refers to E–H.

Figure 6

Peptide depletion of VP2-specific CD8+ T cells confers preservation of retrograde axonal transport in reticulospinal and raphespinal axons. Retrograde transport of FluoroGold was assessed in uninfected mice and at 45 days postinfection in VP2 and E7 peptide-treated mice. Representative sections analyzed under UV illumination are shown for VP2-treated (A) and E7-treated (B) mice. As shown in panel C, the number of retrogradely labeled reticular nucleus and raphe nucleus neurons was significantly increased in mice depleted of VP2-specific T cells (P=0.014 by Holm-Sidak one way ANOVA).