Theiler's murine encephalomyelitis virus (TMEV) infection of the CNS is cleared in C57BL/6 mice by a CD8 T cell response restricted by the MHC class I molecule H-2Db. The identity and function of the APC(s) involved in the priming of this T cell response is (are) poorly defined. To address this gap in knowledge, we developed an H-2Db LoxP-transgenic mouse system using otherwise MHC class I–deficient C57BL/6 mice, thereby conditionally ablating MHC class I–restricted Ag presentation in targeted APC subpopulations. We observed that CD11c+ APCs are critical for early priming of CD8 T cells against the immunodominant TMEV peptide VP2121-130. Loss of H-2Db on CD11c+ APCs mitigates the CD8 T cell response, preventing early viral clearance and immunopathology associated with CD8 T cell activity in the CNS. In contrast, animals with H-2Db–deficient LysM+ APCs retained early priming of Db:VP2121-130 epitope–specific CD8 T cells, although a modest reduction in immune cell entry into the CNS was observed. This work establishes a model enabling the critical dissection of H-2Db–restricted Ag presentation to CD8 T cells, revealing cell-specific and temporal features involved in the generation of CD8 T cell responses. Employing this novel system, we establish CD11c+ cells as pivotal to the establishment of acute antiviral CD8 T cell responses against the TMEV immunodominant epitope VP2121-130, with functional implications both for T cell–mediated viral control and immunopathology.

Infiltration of Ag-specific CD8 T cells into the brain parenchyma is a hallmark feature of CNS viral infections (1). However, the specific APC type(s) required to prime antiviral CD8 T cells in the context of brain infections has (have) not been fully defined (24). Outside the CNS, the classical pathway by which the CD8 T cell response is generated involves MHC class I (MHC-I)–restricted priming against pathogen-derived Ags by dendritic cells (DCs) (5). However, DCs are largely absent from the brain parenchyma in the steady-state (613). There is evidence that the minimal population of CNS resident DCs can be further augmented by additional DCs migrating into the brain during periods of inflammation, but whether these APCs are capable of leaving the brain to return to lymph nodes remains controversial (1416). With the availability of migratory DCs substantially diminished relative to what is found in many peripheral tissues, their relevance to the priming of naive CD8 T cells in the context of CNS infection needs further clarification. Numerically, macrophages (MΦs) in the perivascular spaces and other border areas of the CNS are a better represented population of APCs within the organ system (13). MΦs are also capable of presenting Ag to T cells in vivo, and the observation of myelin-filled MΦs both within multiple sclerosis lesions and in the draining lymph nodes of mice with experimental autoimmune encephalomyelitis raises interest in these APCS as also having a role in the priming of CD8 T cells (1720). Clarifying the roles that both DCs and MΦs play in orchestrating CD8 T cell responses against Ags within the CNS would shed light on APC dynamics within this immune-specialized organ.

The Daniel strain of Theiler's murine encephalomyelitis virus (TMEV) is a neurotropic picornavirus which has classically been studied as a murine model of multiple sclerosis (21). Strains of mice susceptible to intracranial TMEV infection, such as SJL mice, experience chronic infection and demyelination of the spinal cord. Resistant mouse strains, including C57BL/6, present with acute encephalitis and seizures before clearing TMEV infection (2224). Resistance to TMEV has been linked to the H-2D MHC-I molecule, with various alleles providing different levels of protection (2527). One such protective MHC-I allele, H-2Db, is expressed in C57BL/6 mice and is capable of presenting the immunodominant viral peptide, VP2121-130, an essential 10-aa section of the VP2 capsid protein, against which a particularly robust CD8 T cell response is normally generated (2832). Both the presence and the cytolytic capacity of the CTL response raised against this epitope are responsible for the clearance of TMEV (3133). Although the dynamics and consequences of the CD8 T cell response in this model infection of the CNS have been well characterized, the identity of the APC(s) responsible for the initial priming of this response remains undefined.

In the present work, we address the contributions of DCs and MΦs to the priming of CD8 T cell responses against viral pathogens in the CNS using a novel transgenic mouse model. We introduced a floxed H-2Db transgene into H-2Db−/− H-2Kb−/− (MHC-I–deficient) C57BL/6 mice, leaving H-2Db as its sole MHC-I molecule (34). This provides a platform by which one can engineer ablation of class I–restricted Ag presentation capabilities in specific cell types. In the current study, we have crossed this Db LoxP-transgenic mouse with mice expressing Cre-recombinase under the CD11c promoter or the LysM promoter. This enabled us to define the role of CD11c+ or LysM+ APCs in the priming of CD8 T cells specific against the Db-restricted immunodominant TMEV peptide, VP2121-130. Our findings demonstrate a critical role for CD11c+ cells in priming acute CD8 T cell responses against CNS pathogens, despite the relative rarity of these cells within that particular organ.

Animal experiments were conducted in the manner approved by the Mayo Clinic Institutional Animal Care and Use Committee. Our group created the H-2Db LoxP mouse by introducing flanking LoxP sites into the transmembrane (TM) domain of a previously cloned H2-D1 construct through site directed mutagenesis (34). Excision of the TM domain of the MHC-I molecule at the DNA level mediated by Cre-recombinase expression prevents the protein from becoming expressed on the cell surface. The Mayo Clinic Transgenic Mouse Core (Rochester, MN) was responsible for incorporating the transgene into C57BL/6 mice, who, through subsequent backcrossing with mice on an MHC-I knockout (KO) background, were bred until the transgenic Db was their only expressed MHC-I molecule. CD11c-Cre [B6.Cg-Tg(Itgax-cre)1-1Reiz/J, 008068] and CMV-cre [B6.C-Tg(CMV-cre)1Cgn/J, 006054] animals were purchased and crossed to MHC-I–deficient C57BL/6 mice for three successive generations, as previously described (34) (The Jackson Laboratory, Bar Harbor, ME). Crossing these Cre-recombinase expressing mice with the Db LoxP-transgenic animals described above resulted in Cre and Cre+ littermates, the presence of which could be determined by PCR for Cre-recombinase using primer sequences from The Jackson Laboratory. Cre-recombinase sufficient and deficient animals were not housed separately. Total MHC-I KO mice were used in place of CMV Db conditional KO (cKO) mice when sufficient numbers of CMV Db cKO mice were unavailable.

Infection and blood–brain barrier (BBB) disruption were induced as previously reported (35). Briefly, animals between 6 and 14 wk old were infected intracranially with 2 × 104 PFUs of Daniel strain TMEV in the right hemisphere of the brain in a total volume of 10 μl following anesthetization with 2% isoflurane. Mice were euthanized for flow cytometric analysis, or induced to undergo peptide-induced fatal syndrome (PIFS), at 7, 14, or 28 d postinfection (dpi). The PIFS model employed by our group involves the tail vein injection of 100 μl of a 1 mg/ml solution of VP2121-130 peptide (FHAGSLLVFM; GenScript, Nanjing, China) in PBS at either 7 or 14 d after the original TMEV infection (33, 3640). Animals were observed until at least one animal in the experiment became hunched and minimally responsive, at which point all animals were injected via tail vein with 100 μl of a 100 mg/ml FITC-albumin solution in PBS (Sigma-Aldrich, St. Louis, MO). PIFS-induced mice were euthanized 1 h after FITC-albumin injection for analysis.

Magnetic resonance imaging (MRI) capture and analysis was conducted as previously reported (36). Briefly, a random subset of animals from each genotype within PIFS experiment cohorts were subjected to T1-weighted gadolinium–enhanced MRI imaging with a Bruker Avance II 7 T vertical bore small animal MRI system (Bruker Biospin, Billerica, MA) roughly 5 h before PIFS end point–associated behaviors emerged. Mice were injected i.p. with a 100 mg/kg dose of gadolinium, anesthetized with 3% isoflurane and, following a 15 min wait for gadolinium circulation, subjected to a T1-weighted spin echo sequence under maintenance dosing of 1.5% isoflurane. Their respiratory rate was monitored throughout with an MRI compatible vitals monitoring system (model 1030; SA Instruments, Stony Brook, NY). Resultant images, taken with a field of view of 4.0 by 2.0 by 2.0 cm, were processed with the Analyze12.0 software to generate object maps of regions of gadolinium enhancement. Three-dimensional models were generated by overlaying these object maps onto 20% opaque object maps of the entire brain (Biomedical Imaging Resource; Mayo Clinic, Rochester, MN).

A FITC-albumin based permeability assay was conducted as previously reported by our group (37). In short, the left brain hemisphere taken from FITC-albumin injected animals was homogenized in RIPA buffer (50 mM Tris-HCl, 150 mM NaCl, 1% NP-40, 0.5% sodium deoxycholate, 0.1% SDS, and 5 mM EDTA) (Boston BioProducts, Ashland, MA) along with a protease inhibitor mixture using a PowerGen 125 homogenizer (Thermo Fisher Scientific, Hampton, NH). Samples were subsequently centrifuged for 10 min at 4°C at 10,000 rpm. The supernatant was subjected to a BCA protein assay (Pierce Biotechnology, Waltham, MA) and measured on a Synergy H1 Hybrid Multi-Mode Reader (BioTek, Winooski, VT). After normalizing the samples to the lowest sample concentration through the addition of PBS, the brain homogenates were measured on a fluorescent plate reader at 488-nm excitation and 525-nm emission.

Confocal analyses were conducted as previously reported (41). The right brain hemispheres taken from FITC-albumin injected animals were fresh-frozen at −80°C. For analysis, they were embedded in Tissue-Tek OCT (Sakura Finetek, Torrance, CA) and were sectioned with a cryostat. Ten micrometers coronal sections cut from near the hippocampal formation as previously described were placed on positively charged slides which were washed with PBS before fixation in 4% paraformaldehyde for 15 min at room temperature. All slides were washed thrice with 100 μl of PBS and then incubated for an additional 1 h in 100 μl of a 5% normal goat serum and 0.5% Igepal CA-630 solution (Sigma-Aldrich). Samples were subjected to a primary stain with 100 μl of a 1:200 dilution rabbit anti-mouse occludin (clone OC-3F10; Invitrogen, Carlsbad, CA) overnight at 4°C. After three washes with PBS, the Alexa Fluor 647 Goat anti-rabbit IgG secondary was applied at a 1:500 dilution in a total volume of 100 μl for 1 h (Invitrogen). After five final washes with PBS, slides were dried and covered with Vectashield medium + DAPI (Vector Laboratory, Burlingame, CA). Images were taken at room temperature using a Leica DM2500 with a 63× oil immersion objective, and were subsequently analyzed with Leica Acquisition Suite software (Wetzlar, Germany).

Flow cytometric measurements were conducted as previously published (41, 42). Spleens and thymi harvested from euthanized animals were crushed through a 70-μm filter into 10 ml of RPMI 1640 to create a single-cell suspension. Brains were harvested directly into 5 of ml RPMI 1640 before application of a Dounce homogenizer. This brain homogenate was passed through a 70-μm filter into a Percoll solution (9 ml of Percoll, 1 ml of 10× PBS, and 10 ml of RPMI 1640) and subsequently centrifuged at 7840 × g. After the myelin layer was removed, the remaining 10 ml of brain single-cell suspension was spun at 400 × g for 10 min along with the single-cell suspensions of spleens or thymi. All samples were plated in 96-well V-bottom plates and PBS washed following an ACK lysis step. If applicable, brain samples were stained with 50 μl of a 1:50 dilution of Db:VP2121-130 tetramer (National Institutes of Health Tetramer Core Facility) during a 30-min incubation step in the dark at room temperature. The relevant combination of Abs for the experimental question were subsequently applied at 1:100 dilutions during a 30-min incubation on ice in the dark in 50 μl of total volume; these Abs were against CD45 (V450, clone 30-F11; Tonbo), TCRβ (PE-Cy7, clone H57-597; Tonbo), CD8α (BV785, clone 53-6.7; BioLegend), and CD4 (PE, clone GK1.5; BD Pharmingen). To assess the usage of Vβ regions within TCRs, individual spleens were plated into 15 separate wells on a V-bottom 96-well plate and each exposed for 30 min on ice to 20 μl of separate, prediluted, FITC-conjugated mAbs in addition to a 30-μl solution of 1:100 diluted Abs against CD45 (V450, clone 30-F11; Tonbo), TCRβ (PE-Cy7, clone H57-597; Tonbo), CD8 (BV785, clone 53-6.7; BioLegend), and CD4 (PE, clone GK1.5; BD Pharmingen), as well as 1:1000 Fixable Viability Dye (eFluor780; eBioScience). Other Abs used at 1:100 dilution include anti-CD11c (PerCP-Cy5.5, clone N418; Tonbo), H-2Db (FITC, clone B22-249.R1; Accurate Chemical), F4/80 (PE, clone BM8.1; Tonbo), B220 (allophycocyanin, clone RA3-6B2; BioLegend), Gr-1 (allophycocyanin, RB6-8C5; BioLegend), NK1.1 (PE-Cy7, clone PK136; Tonbo), CD19 (PE, clone 1D3; Tonbo), I-A/I-E (BV450, clone M5/114.15.2; Tonbo), CD26 (BV605, clone H194-12; BD Biosciences), PDCA-1 (PE-Cy7, clone eBio927; eBioScience), and CD11b (BV650, clone M1/70; BioLegend). Viability dye (Ghost Red 780; Tonbo) at 1:1000 was also employed for each flow cytometric experiment. Where relevant, a lineage dump gate composed of TCRβ (allophycocyanin, clone H57-597; Tonbo), NK1.1 (allophycocyanin, clone PK136; Tonbo), F4/80 (allophycocyanin, clone BM8.1; Tonbo), B220 (allophycocyanin, clone RA3-6B2; BioLegend), and CD19 (allophycocyanin, clone 1D3; Tonbo) was used. Each sample was digitally compensated with single-stain controls and analyzed using FlowJo v10 (FlowJo LLC, Ashland, OR).

Viral plaque assays were conducted on L2 cells (grown in DMEM + 10% FBS + 1% penicillin-streptomycin) with spinal cord or brain homogenates, as previously published (43). Whole brains were homogenized in 1 ml of RPMI 1640 with a Dounce homogenizer and spinal cords in 1 ml of RPMI 1640 per 100 mg of tissue with a PowerGen 125 homogenizer (Thermo Fisher Scientific). All homogenates were sonicated for 10 s prior to application onto 12-well plates seeded with 1 × 106 L2 cells per well. A 3% agarose overlay was added after 1 h incubation with virus, and 4 d later the cells were fixed with experimental autoimmune encephalomyelitis fixative (60% formaldehyde, 30% EtOH, 10% glacial acetic acid), stained with Crystal Violet solution (1 g of Crystal Violet, 100 ml of 20% EtOH), and plaques were counted by hand.

All data presented as mean ± SEM, and all analyses were conducted with GraphPad Prism 7.0 (La Jolla, CA). Tests used include a two-sided Student t test, a one-way ANOVA with Holm–Sidak correction for multiple comparisons, and a Mann–Whitney rank sum test if the data did not follow a normal distribution.

H-2Db class I cKO mice were developed through construction of an H-2Db transgene, in which the TM domain is flanked by LoxP sites. This transgene was introduced into C57BL/6 mice deficient in H-2Db and H-2Kb, leaving the floxed Db as the only MHC-I molecule expressed by nucleated cells (Fig. 1A) (34). Crossing this transgenic mouse line with mice expressing Cre-recombinase under the CD11c, LysM, or CMV promoter, also on total MHC-I KO backgrounds, allowed for CD11c+ cell–specific, LysM+ cell–specific, or ubiquitous ablation of H-2Db expression, respectively (Fig. 1B, 1C). Resultant mice will be referred to as CD11c Db cKO, LysM Db cKO, CMV Db cKO, or Cre littermates depending on whether the F1 generation pups inherited Cre-recombinase from their parent. Both LysM and CD11c driven Cre-recombinase models have been well characterized and, although they have some off-target activity, they have been found to primarily target MΦs and classical DCs (cDCs), respectively (44). One potential concern in generating MHC-I–deficient mice is that NK cell–mediated cytotoxicity through missing-self recognition could occur. This would lead to a depletion of cells that have lost H-2Db expression. However, we did not detect a change in the percentage of CD11c+ cells within the spleens of unchallenged CD11c Db cKO mice, nor of F4/80+ cells within LysM Db cKO mice, relative to Cre littermates (Fig. 1D, 1E). We therefore did not observe NK-mediated destruction of the APCs of interest, despite the effective removal of H-2Db from the cell surface of DCs and MΦs in CD11c Db cKO and LysM Db cKO animals, respectively (Fig. 1F–H). Tightening our gating strategy further upon cDC populations (Lin CD45+ CD11c+ CD26+ PDCA-1 cells) shows that these powerful APCs most closely associated with CD8 T cell priming (45) also exhibit effective excision of H-2Db (Fig. 1G).

FIGURE 1.

Generation of a mouse model with CD11c+ or LysM+ cell specific H-2Db class I molecule excision. (A) LoxP sites (triangles) were inserted flanking the TM domain of the H-2Db gene. (B) MHC-I–deficient mice expressing Cre-recombinase under the CD11c, LysM, or CMV promoter were crossed to Db transgene expressing mice that were otherwise MHC-I deficient. (C) Representative flow plots of live splenocyte subsets from mice demonstrate reduction in surface H-2Db on the appropriate cellular subset for both LysM Db cKO (MΦs; CD11b+F4/80+MHC-II+ cells) and CD11c Db cKO animals (DCs; CD45+CD11c+CD11blowF4/80low cells). (D) Unchanged percentage (p = 0.12) of CD11c+ cells among CD45+ live events in uninfected spleens of CD11c Db cKO animals (n = 7) relative to Cre littermates (n = 4) as assessed by Mann–Whitney U test. (E) Unchanged percentage (p = 0.31) of F4/80+ cells among CD45+ live events in uninfected spleens of LysM Db cKO animals (n = 5) relative to Cre littermates (n = 5) as assessed by Mann–Whitney U test. (F) Fold change of median fluorescence intensity (MFI) of FITC-conjugated anti-H-2Db Ab on DCs isolated from uninfected spleens. Mann–Whitney U test between CD11c Db cKO (n = 12) and Cre littermates (n = 5). Data taken from two independent experiments with each data point standardized to the average of Cre littermates for that experiment. (G) Fold change of MFI of FITC-conjugated anti–H-2Db Ab on cDCs (CD45+ Lin CD11c+ CD26+ PDCA1) isolated from uninfected spleens. Mann–Whitney U test between CD11c Db cKO (n = 5) and Cre littermates (n = 6). (H) Fold change of MFI of FITC-conjugated anti–H-2Db Ab on MΦs isolated from uninfected spleens. Mann–Whitney U test between LysM Db cKO (n = 4) and Cre littermates (n = 6). Data are presented as mean ± SEM. No comparison reaches statistical significance at the chosen alpha.

FIGURE 1.

Generation of a mouse model with CD11c+ or LysM+ cell specific H-2Db class I molecule excision. (A) LoxP sites (triangles) were inserted flanking the TM domain of the H-2Db gene. (B) MHC-I–deficient mice expressing Cre-recombinase under the CD11c, LysM, or CMV promoter were crossed to Db transgene expressing mice that were otherwise MHC-I deficient. (C) Representative flow plots of live splenocyte subsets from mice demonstrate reduction in surface H-2Db on the appropriate cellular subset for both LysM Db cKO (MΦs; CD11b+F4/80+MHC-II+ cells) and CD11c Db cKO animals (DCs; CD45+CD11c+CD11blowF4/80low cells). (D) Unchanged percentage (p = 0.12) of CD11c+ cells among CD45+ live events in uninfected spleens of CD11c Db cKO animals (n = 7) relative to Cre littermates (n = 4) as assessed by Mann–Whitney U test. (E) Unchanged percentage (p = 0.31) of F4/80+ cells among CD45+ live events in uninfected spleens of LysM Db cKO animals (n = 5) relative to Cre littermates (n = 5) as assessed by Mann–Whitney U test. (F) Fold change of median fluorescence intensity (MFI) of FITC-conjugated anti-H-2Db Ab on DCs isolated from uninfected spleens. Mann–Whitney U test between CD11c Db cKO (n = 12) and Cre littermates (n = 5). Data taken from two independent experiments with each data point standardized to the average of Cre littermates for that experiment. (G) Fold change of MFI of FITC-conjugated anti–H-2Db Ab on cDCs (CD45+ Lin CD11c+ CD26+ PDCA1) isolated from uninfected spleens. Mann–Whitney U test between CD11c Db cKO (n = 5) and Cre littermates (n = 6). (H) Fold change of MFI of FITC-conjugated anti–H-2Db Ab on MΦs isolated from uninfected spleens. Mann–Whitney U test between LysM Db cKO (n = 4) and Cre littermates (n = 6). Data are presented as mean ± SEM. No comparison reaches statistical significance at the chosen alpha.

Close modal

We evaluated the potential of our Cre-recombinase strategy for the conditional ablation of H-2Db on APCs to have overt off-target effects on immune system development. DCs of both cDC1 and cDC2 lineages have been implicated in thymic selection and the development of the peripheral T cell repertoire (46, 47). Accordingly, we sought to ensure that our DC-specific H-2Db depletion strategy did not disrupt thymocyte development (48). CD11c Db cKO and Cre littermates presented equivalent levels of H-2Db on total thymic cells (Fig. 2A). Our control MHC-I total KO animals displayed the expected complete lack of CD8 T cells, but CD11c Db cKO animals had no defect in CD8 T cell development compared with Cre littermates (Fig. 2B, 2C). In secondary lymphoid organs, both MΦs and DCs are prevalent and play important roles in the fostering of immature T cells so we assessed the TCR Vβ repertoire expressed by the mature splenic T cell pool of naive animals. We found no difference in the frequency of the Vβ genes employed by CD8 T cells of either LysM Db cKO or CD11c Db cKO animals relative to Cre littermates (Fig. 3). Together, these data fail to demonstrate aberrant T cell development in the novel H-2Db cKO animals.

FIGURE 2.

Conditional ablation of H-2Db in DCs does not alter normal CD8 T cell development. (A) Unchanged median fluorescence intensity (MFI) of FITC-conjugated anti–H-2Db Ab on total live cells from uninfected thymi between CD11c Db cKO (n = 4) and Cre littermates (n = 3) (p = 0.63) suggests no overt changes in MHC-I expression. Comparison made by Mann–Whitney U test. (B) Representative flow plots gated on CD45+CD3+ events within uninfected thymus show normal T cell development in CD11c Db cKO animals. (C) Quantification of live CD45+CD3+ events, either CD4CD8 (DN), CD4+CD8+ (DP), CD4+CD8 (CD4), or CD4CD8+ (CD8), within uninfected thymi per 1000 live events. Comparisons of CD11c Db cKO (n = 3) and Cre littermates (n = 4) were made against MHC-I KO mice (n = 2) by two-way ANOVA with multiple comparisons. Data are presented as mean ± SEM. No comparison reaches statistical significance at the chosen alpha.

FIGURE 2.

Conditional ablation of H-2Db in DCs does not alter normal CD8 T cell development. (A) Unchanged median fluorescence intensity (MFI) of FITC-conjugated anti–H-2Db Ab on total live cells from uninfected thymi between CD11c Db cKO (n = 4) and Cre littermates (n = 3) (p = 0.63) suggests no overt changes in MHC-I expression. Comparison made by Mann–Whitney U test. (B) Representative flow plots gated on CD45+CD3+ events within uninfected thymus show normal T cell development in CD11c Db cKO animals. (C) Quantification of live CD45+CD3+ events, either CD4CD8 (DN), CD4+CD8+ (DP), CD4+CD8 (CD4), or CD4CD8+ (CD8), within uninfected thymi per 1000 live events. Comparisons of CD11c Db cKO (n = 3) and Cre littermates (n = 4) were made against MHC-I KO mice (n = 2) by two-way ANOVA with multiple comparisons. Data are presented as mean ± SEM. No comparison reaches statistical significance at the chosen alpha.

Close modal
FIGURE 3.

Loss of H-2Db on CD11c+ or LysM+ cells does not impact the Vβ repertoire of naive splenic T cells. (A) Comparison of the usage of variable regions of TCRβ on CD45+TCRβ+CD8+ gated events from the spleens of uninfected CD11c Db cKO mice (n = 4) shows no gross changes relative to the V region usage observed in Cre littermates (n = 4). Comparison made by two-way ANOVA (p = 0.52). (B) Comparison of the usage of variable regions of TCRβ on CD45+TCRβ+CD8+ gated events from the spleens of uninfected LysM Db cKO mice (n = 5) shows no gross changes relative to the V region usage observed in Cre- littermates (n = 4). Comparison made by two-way ANOVA (p = 0.96). Data are presented as mean ± SEM. No comparison reaches statistical significance at the chosen alpha.

FIGURE 3.

Loss of H-2Db on CD11c+ or LysM+ cells does not impact the Vβ repertoire of naive splenic T cells. (A) Comparison of the usage of variable regions of TCRβ on CD45+TCRβ+CD8+ gated events from the spleens of uninfected CD11c Db cKO mice (n = 4) shows no gross changes relative to the V region usage observed in Cre littermates (n = 4). Comparison made by two-way ANOVA (p = 0.52). (B) Comparison of the usage of variable regions of TCRβ on CD45+TCRβ+CD8+ gated events from the spleens of uninfected LysM Db cKO mice (n = 5) shows no gross changes relative to the V region usage observed in Cre- littermates (n = 4). Comparison made by two-way ANOVA (p = 0.96). Data are presented as mean ± SEM. No comparison reaches statistical significance at the chosen alpha.

Close modal

We next evaluated the effect that selectively ablating MHC-I–restricted Ag presentation ability on specific APC subsets would have on CD8 T cell priming against intracranial TMEV infection. LysM Db cKO, CD11c Db cKO, and Cre littermates were infected intracranially with TMEV and euthanized at early and late time points over the course of the infection. Using flow cytometric analysis of brain-infiltrating immune cells, we observed a reduction in CD8 T cells that are specific to the immunodominant peptide VP2121-130 at 5 and 7 dpi in CD11c Db cKO animals (Fig. 4A, 4B). CD11c Db cKO mice and Cre littermates have comparable CD45+ cell infiltration throughout the course of infection, but the ratio of total live cells in CD11c Db cKO mice that are CD8 T cells is severely reduced within the brain at early time points (Fig. 4C). Even more strikingly, of the CD8 T cells that do enter, a lower percentage are specific for the Db:VP2121-130 epitope (Fig. 4B). However, by 14 and 28 dpi, the percentage of virus-specific CD8 T cells in CD11c Db cKO and Cre brains are equivalent, with nearly 80% of CD8 T cells within the brain being specific to the immunodominant Db:VP2121-130 epitope in both genotypes (Fig. 4B). Corroborating the notion that this is due to a defect in the priming of Db:VP2121-130 epitope–specific CD8 T cells in CD11c Db cKO mice is the significant decrease in absolute cell number of these antiviral CD8 T cells in the cervical lymph nodes at 5 dpi relative to Cre littermates (Fig. 4D). This defect in priming does not seem dependent upon failure of MHC-I–deficient DCs to enter the brain, as equivalent populations between the two genotypes are observed at 5 dpi (Fig. 4E). These data suggest that loss of MHC-I on CD11c+ cells, although still permitting responsiveness to acute CNS injury, restricts the priming of a rapid CD8 T cell response to intracranial TMEV infection acutely.

FIGURE 4.

Priming of virus-specific CD8 T cells is impaired in mice with H-2Db–deficient CD11c+ cells. (A) Representative flow cytometry data for Db:VP2121-130 tetramer+ CD8+ events gated on live CD45+TCRβ+ events from brain tissue across the time course of TMEV infection shows profound defect in early T cell priming in CD11c Db cKO animals. (B) Quantitation of flow cytometry data as a percent of the parental gate or (C) as a number of cells per million live cells across a time course of brains following intracranial TMEV infection in CD11c Db cKO animals reveals defect in early antiviral T cell priming. n > 3 for each genotype at each timepoint. Groups were of consistent gender and age within the same timepoint. Comparisons made with Mann–Whitney U test between CD11c Db cKO and Cre littermates at each time point. (D) Total number of cervical lymph nodes (cLN) cells is not reduced to a statistically significant degree (p = 0.202) in CD11c Db cKO animals relative to Cre littermates at 5 dpi, but the total number of Db:VP2121-130 epitope–specific CD8 T cells is reduced to a statistically significant degree (p = 0.005). (E) Total number of CD11c+TCRβCD45+ cells from brains of CD11c Db cKO (n = 5) and Cre littermates (n = 7) at 5 dpi shows no difference (p = 0.343). Analyses conducted with Mann–Whitney U test. Data are presented as mean ± SEM. *p < 0.05.

FIGURE 4.

Priming of virus-specific CD8 T cells is impaired in mice with H-2Db–deficient CD11c+ cells. (A) Representative flow cytometry data for Db:VP2121-130 tetramer+ CD8+ events gated on live CD45+TCRβ+ events from brain tissue across the time course of TMEV infection shows profound defect in early T cell priming in CD11c Db cKO animals. (B) Quantitation of flow cytometry data as a percent of the parental gate or (C) as a number of cells per million live cells across a time course of brains following intracranial TMEV infection in CD11c Db cKO animals reveals defect in early antiviral T cell priming. n > 3 for each genotype at each timepoint. Groups were of consistent gender and age within the same timepoint. Comparisons made with Mann–Whitney U test between CD11c Db cKO and Cre littermates at each time point. (D) Total number of cervical lymph nodes (cLN) cells is not reduced to a statistically significant degree (p = 0.202) in CD11c Db cKO animals relative to Cre littermates at 5 dpi, but the total number of Db:VP2121-130 epitope–specific CD8 T cells is reduced to a statistically significant degree (p = 0.005). (E) Total number of CD11c+TCRβCD45+ cells from brains of CD11c Db cKO (n = 5) and Cre littermates (n = 7) at 5 dpi shows no difference (p = 0.343). Analyses conducted with Mann–Whitney U test. Data are presented as mean ± SEM. *p < 0.05.

Close modal

Where CD11c Db cKO animals mounted a temporally delayed antiviral CD8 T cell response against TMEV, LysM Db cKO animals displayed an antiviral response that was moderately attenuated compared with Cre littermates. The abundance of tetramer+ CD8 T cells relative to the CD8 T cell response in general was unaffected by loss of H-2Db on MΦs (Fig. 5A). However, there was a widespread attenuation of the CD45+ compartment as a whole (Fig. 5A, 5B). At 5 dpi, flow cytometric analysis revealed a statistically significant reduction in total number of infiltrating NK cells, CD4 T cells, and GR1+ cells with minor reductions in B cells (Fig. 5C). Together, these data suggest that although the priming of Db:VP2121-130 epitope–specific CD8 T cells was not overtly affected by the loss of H-2Db on LysM+ APCs, the broad contours of the immune response in the CNS were impacted by the loss of this MHC-I molecule.

FIGURE 5.

Multiple hematopoietic cells display reduced CNS recruitment in absence of MHC-I on LysM+ APCs. (A) Quantitation of flow cytometry data as a percent of the parental gate or (B) as a number of cells per million live cells across a time course of intracranial TMEV infection in LysM Db cKO animals reveals overall deficiency in immune cell infiltration of CNS. n > 3 for each genotype at each timepoint. Groups were of a consistent gender and age within the same timepoint. Comparisons were made between LysM Db cKO and Cre littermates at each time point with Mann–Whitney U test. (C) Total hemocytometer counts of brain immune cells from 5 dpi LysM Db cKO and Cre littermates shows statistically significant reduction across several subtypes by Mann–Whitney U test. T cells are defined as CD4/8+ TCRβ+ CD19 CD45+ events, NK cells as NK1.1+ TCRβ CD19 CD45+ events, and B cells as CD19+ TCRβ NK1.1 CD45+ events. GR1+ cells are defined as TCRβ CD19 NK1.1 GR1+ CD45+ events, thus loosely encompassing a variety of myeloid cells. Data are presented as mean ± SEM. *p < 0.05.

FIGURE 5.

Multiple hematopoietic cells display reduced CNS recruitment in absence of MHC-I on LysM+ APCs. (A) Quantitation of flow cytometry data as a percent of the parental gate or (B) as a number of cells per million live cells across a time course of intracranial TMEV infection in LysM Db cKO animals reveals overall deficiency in immune cell infiltration of CNS. n > 3 for each genotype at each timepoint. Groups were of a consistent gender and age within the same timepoint. Comparisons were made between LysM Db cKO and Cre littermates at each time point with Mann–Whitney U test. (C) Total hemocytometer counts of brain immune cells from 5 dpi LysM Db cKO and Cre littermates shows statistically significant reduction across several subtypes by Mann–Whitney U test. T cells are defined as CD4/8+ TCRβ+ CD19 CD45+ events, NK cells as NK1.1+ TCRβ CD19 CD45+ events, and B cells as CD19+ TCRβ NK1.1 CD45+ events. GR1+ cells are defined as TCRβ CD19 NK1.1 GR1+ CD45+ events, thus loosely encompassing a variety of myeloid cells. Data are presented as mean ± SEM. *p < 0.05.

Close modal

Because the restoration of brain-infiltrating Db:VP2121-130 epitope–specific CD8 T cells by 14 dpi in CD11c Db cKO animals was compelling, we next assessed how effectively this population cleared virus from the CNS. Because intracranial TMEV infection spreads through neuronal connections, we evaluated viral load in brains isolated at 7 dpi and spinal cords at 14 dpi (49, 50). We observed significantly higher viral loads in CD11c Db cKO brain tissue than in Cre littermates at the 7 dpi timepoint (Fig. 6A). As the spread of TMEV into the spinal cord is prevented in wild-type C57BL/6 mice, a low viral load in that region late in the timecourse of infection is indicative of effective antiviral immunity (24, 5154). Based on this, our observation that, by 14 dpi, the CD11c Db cKO animals had equivalent viral load in the spinal cord as Cre littermates shows a restoration of functional CNS immune-surveillance (Fig. 6B). These data demonstrate that both a quantitative and functional restoration of Db:VP2121-130 epitope–specific CD8 T cells had occurred in CD11c Db cKO animals at a late timepoint.

FIGURE 6.

Loss of viral clearance observed in CD11c Db cKO animals is regained by 14 dpi. Analysis of CNS viral load through L2 cell plaque assay conducted with brain tissue homogenate at (A) 7 dpi and with spinal cord homogenate at (B) 14 dpi. Mice were age and gender matched within each experiment and n ≥ 3 for each. Analyses conducted with Kruskal–Wallis test. Data are presented as mean ± SEM. *p < 0.05.

FIGURE 6.

Loss of viral clearance observed in CD11c Db cKO animals is regained by 14 dpi. Analysis of CNS viral load through L2 cell plaque assay conducted with brain tissue homogenate at (A) 7 dpi and with spinal cord homogenate at (B) 14 dpi. Mice were age and gender matched within each experiment and n ≥ 3 for each. Analyses conducted with Kruskal–Wallis test. Data are presented as mean ± SEM. *p < 0.05.

Close modal

In addition to mediating viral clearance, we next assessed the capacity of early versus late-primed Db:VP2121-130 epitope–specific CD8 T cells to promote neuropathology in a model of BBB disruption developed by our research program (33, 3640). This PIFS model is mediated by Db:VP2121-130 epitope–specific CD8 T cells following i.v. administration of VP2121-130 peptide during acute TMEV infection. Administration of VP2121-130 peptide results in extensive CNS vascular permeability, characterized by significant cerebral endothelial cell tight junction remodeling, microhemorrhage formation, neurologic deficits, and death within 48 h (33, 3640). We initiated PIFS through VP2121-130 peptide administration in 7 dpi CMV Db cKO and CD11c Db cKO animals, as well as Cre littermate controls. Interestingly, by 24 h postinduction of PIFS, the CD11c Db cKO animals remained healthy whereas Cre littermates were becoming moribund. At onset of visible symptoms in Cre littermates, animals randomly selected from each group were subjected to gadolinium-enhanced T1-weighted MRI. Following MRI imaging, all animals were injected i.v. with a FITC-albumin conjugate and then euthanized 1 h later, permitting analysis of CNS vascular permeability with a second methodology (55). By immunofluorescence, the BBB in CMV Db cKO and CD11c Db cKO animals was preserved, with FITC-albumin (green) colocalizing with the tight junction protein occludin (red) in coronal brain sections (Fig. 7A). Cre animals, however, displayed significant disruption of tight junction integrity along the blood vessels and a diffuse spread of FITC+ regions throughout the brain parenchyma (Fig. 7A). MRI data gathered was in accordance with the immunofluorescent imaging, revealing considerable gadolinium enhancement, consistent with CNS vascular permeability, in the brains of Cre littermates. In contrast, CD11c Db cKO littermates had CNS vascular permeability indistinguishable from CMV Db cKO animals, which are protected from the CD8 T cell–mediated BBB disruption because of their ubiquitous MHC-I deficiency (Fig. 7B). As expected, measurement of fluorescence in brain homogenates after euthanasia revealed increased leakage of FITC-albumin into the brain parenchyma in Cre littermates relative to CD11c Db cKO mice (Fig. 7B). Assessment of CNS vascular permeability in the PIFS model was repeated in mice 14 dpi. At this time point, the CD8 T cell response against the immunodominant Db:VP2121-130 viral epitope is restored (Fig. 4C). Restoration of the Db:VP2121-130 epitope–specific CD8 T cell response resulted in renewed capacity to induce BBB disruption in CD11c Db cKO mice, with regions of gadolinium enhancement and FITC-albumin accumulation both becoming equivalent to patterns observed in Cre littermates (Fig. 7C).This late-arising Db:VP2121-130 epitope–specific CD8 T cell response therefore restored its potency to induce BBB disruption in the PIFS model. These data not only establish the functionality of T cells primed later in the course of infection but also, importantly, demonstrate a minimum quantitative threshold of CD8 T cells required to cause pathology.

FIGURE 7.

Late-arising virus-specific CD8 T cells in CD11c Db cKO animals possess the capacity to elicit BBB disruption. (A) BBB disruption was induced in mice 7 or 14 d post-TMEV infection through the administration of VP2121-130 peptide via tail vein injection. FITC-albumin injected prior to euthanasia the following day (i.e., day 8 or day 15) was used along with immunostaining for Occludin to visualize vessel integrity 8 dpi, and also to measure vascular leakage by assessing fluorescence intensity from brain homogenates of standardized concentration with a plate reader. Scale bar, 250 μm. Representative T1-weighted gadolinium–enhanced MRI imaging of these animals prior to euthanasia at (B) 8 dpi and (C) 15 dpi (p = 0.13) is presented next to the FITC-albumin brain homogenate fluorescence data from the appropriate time point. Each experiment is conducted n ≥ 3 for each genotype and analyses are conducted with one-way ANOVA. Data are presented as mean ± SEM. *p < 0.05.

FIGURE 7.

Late-arising virus-specific CD8 T cells in CD11c Db cKO animals possess the capacity to elicit BBB disruption. (A) BBB disruption was induced in mice 7 or 14 d post-TMEV infection through the administration of VP2121-130 peptide via tail vein injection. FITC-albumin injected prior to euthanasia the following day (i.e., day 8 or day 15) was used along with immunostaining for Occludin to visualize vessel integrity 8 dpi, and also to measure vascular leakage by assessing fluorescence intensity from brain homogenates of standardized concentration with a plate reader. Scale bar, 250 μm. Representative T1-weighted gadolinium–enhanced MRI imaging of these animals prior to euthanasia at (B) 8 dpi and (C) 15 dpi (p = 0.13) is presented next to the FITC-albumin brain homogenate fluorescence data from the appropriate time point. Each experiment is conducted n ≥ 3 for each genotype and analyses are conducted with one-way ANOVA. Data are presented as mean ± SEM. *p < 0.05.

Close modal

Studying the nature of Ag presentation in the context of CNS pathogens has traditionally required ablation of whole APC subpopulations or bone marrow transfers (5659). The novel H-2Db cKO–transgenic mouse system put forward in this study lays the groundwork to dissect the relative contributions of DCs and other APC subsets in CD8 T cell priming against CNS pathogens without disrupting other functions of these cell types, such as CD4 T cell priming or cytokine secretion. The presently described H-2Db cKO mouse, in conjunction with our previously published H-2Kb cKO mouse, enables assessment of Ag presentation in disease models (41). This work builds upon the previous by demonstrating that loss of the CD8 T cell response restricted toward an endogenous viral epitope, VP2121-130, severely hampered viral control where loss of the H-2Kb:OVA257–264 restricted T cell response did not (41). Although both demonstrate a pivotal role for DCs in the priming of CD8 T cell responses against viruses of the CNS, they add distinct contributions to the decades long investigation into the nonequivalency of the H-2K and H-2D molecule in mediating immune-surveillance against pathogens of the CNS (6063). In this study, we have articulated the requirement for CD11c+ cells in priming acute CD8 T cell responses against intracranial infection by the picornavirus TMEV and shown that LysM+ APCs play a subordinate role in this process. The loss of H-2Db among this latter cellular population mildly reduced infiltration of some immune cell compartments while not overtly impacting the antiviral CD8 T cell response of interest. Furthermore, our data could be consistent with a potential compensatory APC population capable of priming antiviral T cell responses, albeit much more slowly, in the absence of the Ag presentation ability of CD11c+ cells.

The importance of CD11c+ cells in the generation of a robust antiviral CD8 T cell response has been previously put forward and is further confirmed in vivo with the results of this study employing our CD11c Db cKO model (59, 64). Using a series of bone marrow transfer experiments, Mendez-Fernandez et al. (57) have shown that the APC responsible for priming CD8 T cells against the VP2121-130 epitope in an H-2Db–restricted manner must be a cell of hematopoietic identity. Their study supported a working model in which the initial priming of naive CD8 T cells occurred in secondary lymphoid organs (57), but data from other groups leave open the possibility that naive T cells could gain entry to the inflamed brain parenchyma and be primed in situ (56, 6568). The tools that we put forth in this current manuscript, however, are insufficient alone to elucidate the location of the priming of the eventual CNS-occupying antiviral CD8 T cell population. Considering established literature related to T cell priming, we envision four, nonmutually exclusive, contexts in which the clearly pivotal CD11c+ cells may be cross-presenting Ag 1) DCs in secondary lymphoid organs acquire viral Ags draining out of the CNS via lymphatics, 2) rare DCs from the CNS itself are acquiring viral Ags and trafficking to draining lymph nodes, 3) unidentified migratory APCs acquire Ag and traffic to draining lymph nodes where they transfer viral Ag to CD11c+ cells, and, finally 4) rare CD11c+ cells within the CNS are capable of stimulating naive CD8 T cells within the CNS, sidestepping the need for secondary lymphoid tissues for the priming process (56, 6971). Although our novel, to our knowledge, mouse model has further established the significance of CD11c+ cells in this model of intracranial viral challenge, the migratory dynamics of involved APCs remain an area in which further investigation is still required, although our data at 5 dpi suggests that a difference in DC migration into the CNS is not the root cause of the phenotypic differences in T cell priming between CD11c Db cKO and Cre littermate animals (Fig. 4E).

A striking result from the current study is the restoration of Db:VP2121-130 epitope–specific, functionally antiviral, CD8 T cells in the brains of CD11c Db cKO animals at 14 and 28 dpi. This rise in CD8 T cell numbers in CD11c Db cKO mice, at a time point when the T cell response in control Cre littermates is contracting, could be explained by a compensatory APC population that is capable of initiating a CD8 T cell response against pathogens within the CNS when CD11c+ cells are unable to effectively present Ag. The APC responsible for this delayed CD8 T cell priming remains sufficient to induce an adaptive response capable of clearing TMEV infection. Furthermore, the Db:VP2121-130 epitope–specific CD8 T cells present at 14 dpi have the capacity to induce BBB disruption in an Ag-specific manner, as demonstrated in the PIFS model (35). These observations alone do not necessarily imply that CD8 T cell populations raised by different APCs use the same Db:VP2121-130 epitope–specific TCR. It remains possible that different clonally expanded populations of antiviral T cells exist between CD11c Db cKO and Cre animals at the 14 d time point, as it has been suggested before that T cell responses driven by different APCs might gravitate toward different TCR repertoires (17). Analysis of these two seemingly equally functionally antiviral T cell populations for differential Vβ TCR usage is an area of future research. Finally, the identity of the APC-mediating compensatory CD8 T cell priming activity in the CD11c Db cKO animals remains under investigation. It remains to be demonstrated that such T cell priming is not a result of imperfect excision of the H-2Kb LoxP transgene from CD11c+ cells.

MΦs are a potential compensatory APC that could participate in the priming of Db:VP2121-130–specific CD8 T cells along with DCs. In multiple organ systems, MΦs have been suggested to play a role in naive T cell priming, and the activity of these cells could account for the T cell priming occurring in the absence of DC-expressed MHC-I in the CD11c Db cKO mice (17, 72). The kinetically delayed nature of this CD8 T cell priming (Fig. 4A–C), could well be explained by the documented inferiority of MΦs as APCS. In one such study that arrived at that conclusion, bone marrow–derived, peptide-pulsed APCs injected s.c. were shown to prime CD8 T cells against a Db-restricted lymphocytic choriomeningitis virus Ag, but injected MΦs needed to be exposed to three times the concentration of peptide than what was necessary for injected DCs to prime a numerically equivalent T cell response (73). If MΦs were the APCs mediating the priming of Db:VP2121-130–specific CD8 T cells in CD11c Db cKO mice, it is possible that the necessity of higher Ag concentration/viral load drove the delayed kinetics of the T cell priming event. In a simple wild-type mouse, it is possible that the contribution of any potential MΦ populations to priming this antiviral CD8 T cell response are minimal in comparison with those of CD11c+ cells and are therefore masked as long as DCs remain Ag presentation competent. Although speculative, this would explain why neither the LysM Db cKO mouse explored in the current study, nor the LysM Kb cKO mouse previously published, were sufficient to impact viral specific CD8 T cell populations relative to appropriate Cre− littermate controls (41).

Microglia are another potential compensatory APC that might be responsible for the activation of Db:VP2121-130–specific CD8 T cells in our cKO system. It has been previously established that activated microglia in C57BL/6 mice can cross-present soluble OVA Ag to naive CD8 T cells in vivo, and other works have confirmed that TMEV infection is sufficient to drive microglia toward an activated, APC-like, phenotype (74, 75). Together, these data are permissive of a role for microglia in the priming of naive CD8 T cells against the TMEV immunodominant VP2121-130 epitope. Nevertheless, there are conflicting data which cast doubt on the likelihood of this scenario, such as the observed abrogation of the Db:VP2121-130 epitope–specific CD8 T cell response in TMEV-infected splenectomized LTα−/− animals (57). Those animals, completely deficient in secondary lymphoid organs, were unable to mount Db:VP2121-130 epitope–specific CD8 T cell responses against TMEV, but that study looked only at early time points, so it remains unknown whether CD8 T cell priming could occur physically within the CNS at later time points during TMEV infection (57). The ability of microglia to mediate naive CD8 T cell priming is of interest and is currently under investigation by our research program.

In summary, adaptive immune responses, and particularly those of CD8 T cells, are pivotal to host defense against viruses. In addition, these lymphocytes, themselves, can be drivers of neuropathology. Therefore a more robust understanding of how APCs drive this form of immunity could have a profound impact on the development of effective therapies as well as limiting toxicities that arise from mobilizing activated effector CD8 T cells into the brain (76, 77). In furtherance of these goals, we have described in this study a Cre-recombinase system for the cell-specific deletion of H-2Db on CD11c+ cell populations in C57BL/6 mice. In the process, we have shown these APCs to be the significant driver of the acute CD8 T cell responses against TMEV infection. This work also provokes new questions about possible additional APC(s) responsible for CTL priming. Finally, we introduce a novel transgenic mouse model that can serve as a critical reagent for more detailed investigations of CTL priming both within and outside the CNS moving forward.

We thank the National Institutes of Health Tetramer Core Facility for providing the H-2Db:VP2121-130 tetramer.

This work was funded by National Institute of Neurological Disorders and Stroke Grant R01 NS103212 awarded to A.J.J. Z.P.T. and C.E.F. were supported by National Institutes of Health National Research Service Award Institutional Research Training Grant T32 AI07425-24.

Abbreviations used in this article:

BBB

blood–brain barrier

cDC

classical DC

cKO

conditional KO

DC

dendritic cell

dpi

day postinfection

KO

knockout

macrophage

MHC-I

MHC class I

PIFS

peptide-induced fatal syndrome

TM

transmembrane

TMEV

Theiler's murine encephalomyelitis virus.

1
Mockus
,
T. E.
,
H. M.
Ren
,
Shwetank
,
A. E.
Lukacher
.
2019
.
To go or stay: the development, benefit, and detriment of tissue-resident memory CD8 T cells during central nervous system viral infections.
Viruses
11
:
842
2
Aguilar-Valenzuela
,
R.
,
J.
Netland
,
Y. J.
Seo
,
M. J.
Bevan
,
A.
Grakoui
,
M. S.
Suthar
.
2018
.
Dynamics of tissue-specific CD8+ T cell responses during west nile virus infection.
J. Virol.
92
: e00014–18.
3
Galea
,
I.
,
I.
Bechmann
,
V. H.
Perry
.
2007
.
What is immune privilege (not)?
Trends Immunol.
28
:
12
18
.
4
Wang
,
Y.
,
M.
Lobigs
,
E.
Lee
,
A.
Müllbacher
.
2003
.
CD8+ T cells mediate recovery and immunopathology in West Nile virus encephalitis.
J. Virol.
77
:
13323
13334
.
5
Guermonprez
,
P.
,
J.
Valladeau
,
L.
Zitvogel
,
C.
Théry
,
S.
Amigorena
.
2002
.
Antigen presentation and T cell stimulation by dendritic cells.
Annu. Rev. Immunol.
20
:
621
667
.
6
Bulloch
,
K.
,
M. M.
Miller
,
J.
Gal-Toth
,
T. A.
Milner
,
A.
Gottfried-Blackmore
,
E. M.
Waters
,
U. W.
Kaunzner
,
K.
Liu
,
R.
Lindquist
,
M. C.
Nussenzweig
, et al
.
2008
.
CD11c/EYFP transgene illuminates a discrete network of dendritic cells within the embryonic, neonatal, adult, and injured mouse brain.
J. Comp. Neurol.
508
:
687
710
.
7
Greter
,
M.
,
F. L.
Heppner
,
M. P.
Lemos
,
B. M.
Odermatt
,
N.
Goebels
,
T.
Laufer
,
R. J.
Noelle
,
B.
Becher
.
2005
.
Dendritic cells permit immune invasion of the CNS in an animal model of multiple sclerosis.
Nat. Med.
11
:
328
334
.
8
Jordão
,
M. J. C.
,
R.
Sankowski
,
S. M.
Brendecke
,
G.
Sagar
,
G.
Locatelli
,
Y. H.
Tai
,
T. L.
Tay
,
E.
Schramm
,
S.
Armbruster
,
N.
Hagemeyer
, et al
.
2019
.
Single-cell profiling identifies myeloid cell subsets with distinct fates during neuroinflammation.
Science
363
: eaat7554.
9
McMenamin
,
P. G.
1999
.
Distribution and phenotype of dendritic cells and resident tissue macrophages in the dura mater, leptomeninges, and choroid plexus of the rat brain as demonstrated in wholemount preparations.
J. Comp. Neurol.
405
:
553
562
.
10
Mundt
,
S.
,
D.
Mrdjen
,
S. G.
Utz
,
M.
Greter
,
B.
Schreiner
,
B.
Becher
.
2019
.
Conventional DCs sample and present myelin antigens in the healthy CNS and allow parenchymal T cell entry to initiate neuroinflammation.
Sci. Immunol.
4
: eaau8380.
11
Nayak
,
D.
,
B. H.
Zinselmeyer
,
K. N.
Corps
,
D. B.
McGavern
.
2012
.
In vivo dynamics of innate immune sentinels in the CNS.
Intravital
1
:
95
106
.
12
Prodinger
,
C.
,
J.
Bunse
,
M.
Krüger
,
F.
Schiefenhövel
,
C.
Brandt
,
J. D.
Laman
,
M.
Greter
,
K.
Immig
,
F.
Heppner
,
B.
Becher
,
I.
Bechmann
.
2011
.
CD11c-expressing cells reside in the juxtavascular parenchyma and extend processes into the glia limitans of the mouse nervous system.
Acta Neuropathol.
121
:
445
458
.
13
Mrdjen
,
D.
,
A.
Pavlovic
,
F. J.
Hartmann
,
B.
Schreiner
,
S. G.
Utz
,
B. P.
Leung
,
I.
Lelios
,
F. L.
Heppner
,
J.
Kipnis
,
D.
Merkler
,
M.
Greter
,
B.
Becher
.
2018
.
High-dimensional single-cell mapping of central nervous system immune cells reveals distinct myeloid subsets in health, aging, and disease [Published erratum appears in 2018 Immunity 48: 599.].
Immunity.
48
:
380
395.e6
14
Clarkson
,
B. D.
,
A.
Walker
,
M.
Harris
,
A.
Rayasam
,
M.
Sandor
,
Z.
Fabry
.
2014
.
Mapping the accumulation of co-infiltrating CNS dendritic cells and encephalitogenic T cells during EAE.
J. Neuroimmunol.
277
:
39
49
.
15
Clarkson
,
B. D.
,
A.
Walker
,
M. G.
Harris
,
A.
Rayasam
,
M.
Sandor
,
Z.
Fabry
.
2015
.
CCR2-dependent dendritic cell accumulation in the central nervous system during early effector experimental autoimmune encephalomyelitis is essential for effector T cell restimulation in situ and disease progression.
J. Immunol.
194
:
531
541
.
16
De Laere
,
M.
,
Z. N.
Berneman
,
N.
Cools
.
2018
.
To the brain and back: migratory paths of dendritic cells in multiple sclerosis.
J. Neuropathol. Exp. Neurol.
77
:
178
192
.
17
Bernhard
,
C. A.
,
C.
Ried
,
S.
Kochanek
,
T.
Brocker
.
2015
.
CD169+ macrophages are sufficient for priming of CTLs with specificities left out by cross-priming dendritic cells.
Proc. Natl. Acad. Sci. USA
112
:
5461
5466
.
18
Kivisäkk
,
P.
,
J.
Imitola
,
S.
Rasmussen
,
W.
Elyaman
,
B.
Zhu
,
R. M.
Ransohoff
,
S. J.
Khoury
.
2009
.
Localizing central nervous system immune surveillance: meningeal antigen-presenting cells activate T cells during experimental autoimmune encephalomyelitis.
Ann. Neurol.
65
:
457
469
.
19
Pope
,
J. G.
,
C. L.
Vanderlugt
,
S. M.
Rahbe
,
H. L.
Lipton
,
S. D.
Miller
.
1998
.
Characterization of and functional antigen presentation by central nervous system mononuclear cells from mice infected with Theiler’s murine encephalomyelitis virus.
J. Virol.
72
:
7762
7771
.
20
van Zwam
,
M.
,
J. N.
Samsom
,
E. E.
Nieuwenhuis
,
M. J.
Melief
,
A. F.
Wierenga-Wolf
,
I. E.
Dijke
,
S.
Talens
,
M.
van Meurs
,
J. S.
Voerman
,
L. A.
Boven
,
J. D.
Laman
.
2011
.
Myelin ingestion alters macrophage antigen-presenting function in vitro and in vivo.
J. Leukoc. Biol.
90
:
123
132
.
21
Libbey
,
J. E.
,
T. E.
Lane
,
R. S.
Fujinami
.
2014
.
Axonal pathology and demyelination in viral models of multiple sclerosis.
Discov. Med.
18
:
79
89
.
22
DePaula-Silva
,
A. B.
,
T. J.
Hanak
,
J. E.
Libbey
,
R. S.
Fujinami
.
2017
.
Theiler’s murine encephalomyelitis virus infection of SJL/J and C57BL/6J mice: models for multiple sclerosis and epilepsy.
J. Neuroimmunol.
308
:
30
42
.
23
Drescher
,
K. M.
,
D.
Sosnowska
.
2008
.
Being a mouse in a man’s world: what TMEV has taught us about human disease.
Front. Biosci.
13
:
3775
3785
.
24
Tsunoda
,
I.
,
R. S.
Fujinami
.
2010
.
Neuropathogenesis of Theiler’s murine encephalomyelitis virus infection, an animal model for multiple sclerosis.
J. Neuroimmune Pharmacol.
5
:
355
369
.
25
Azoulay
,
A.
,
M.
Brahic
,
J. F.
Bureau
.
1994
.
FVB mice transgenic for the H-2Db gene become resistant to persistent infection by Theiler’s virus.
J. Virol.
68
:
4049
4052
.
26
Rodriguez
,
M.
,
C. S.
David
.
1985
.
Demyelination induced by Theiler’s virus: influence of the H-2 haplotype.
J. Immunol.
135
:
2145
2148
.
27
Rodriguez
,
M.
,
C. S.
David
.
1995
.
H-2 Dd transgene suppresses Theiler’s virus-induced demyelination in susceptible strains of mice.
J. Neurovirol.
1
:
111
117
.
28
Bell
,
M. P.
,
D. N.
Renner
,
A. J.
Johnson
,
K. D.
Pavelko
.
2014
.
An elite controller of picornavirus infection targets an epitope that is resistant to immune escape.
PLoS One
9
: e94332.
29
Borson
,
N. D.
,
C.
Paul
,
X.
Lin
,
W. K.
Nevala
,
M. A.
Strausbauch
,
M.
Rodriguez
,
P. J.
Wettstein
.
1997
.
Brain-infiltrating cytolytic T lymphocytes specific for Theiler’s virus recognize H2Db molecules complexed with a viral VP2 peptide lacking a consensus anchor residue.
J. Virol.
71
:
5244
5250
.
30
Dethlefs
,
S.
,
N.
Escriou
,
M.
Brahic
,
S.
van der Werf
,
E. L.
Larsson-Sciard
.
1997
.
Theiler’s virus and Mengo virus induce cross-reactive cytotoxic T lymphocytes restricted to the same immunodominant VP2 epitope in C57BL/6 mice.
J. Virol.
71
:
5361
5365
.
31
Johnson
,
A. J.
,
M. K.
Njenga
,
M. J.
Hansen
,
S. T.
Kuhns
,
L.
Chen
,
M.
Rodriguez
,
L. R.
Pease
.
1999
.
Prevalent class I-restricted T-cell response to the Theiler’s virus epitope Db:VP2121-130 in the absence of endogenous CD4 help, tumor necrosis factor alpha, gamma interferon, perforin, or costimulation through CD28.
J. Virol.
73
:
3702
3708
.
32
Johnson
,
A. J.
,
J.
Upshaw
,
K. D.
Pavelko
,
M.
Rodriguez
,
L. R.
Pease
.
2001
.
Preservation of motor function by inhibition of CD8+ virus peptide-specific T cells in Theiler’s virus infection.
FASEB J.
15
:
2760
2762
.
33
Willenbring
,
R. C.
,
F.
Jin
,
D. J.
Hinton
,
M.
Hansen
,
D. S.
Choi
,
K. D.
Pavelko
,
A. J.
Johnson
.
2016
.
Modulatory effects of perforin gene dosage on pathogen-associated blood-brain barrier (BBB) disruption.
J. Neuroinflammation
13
:
222
.
34
Mendez-Fernandez
,
Y. V.
,
A. J.
Johnson
,
M.
Rodriguez
,
L. R.
Pease
.
2003
.
Clearance of Theiler’s virus infection depends on the ability to generate a CD8+ T cell response against a single immunodominant viral peptide.
Eur. J. Immunol.
33
:
2501
2510
.
35
Johnson
,
A. J.
,
Y.
Mendez-Fernandez
,
A. M.
Moyer
,
C. R.
Sloma
,
I.
Pirko
,
M. S.
Block
,
M.
Rodriguez
,
L. R.
Pease
.
2005
.
Antigen-specific CD8+ T cells mediate a peptide-induced fatal syndrome.
J. Immunol.
174
:
6854
6862
.
36
Johnson
,
H. L.
,
Y.
Chen
,
F.
Jin
,
L. M.
Hanson
,
J. D.
Gamez
,
I.
Pirko
,
A. J.
Johnson
.
2012
.
CD8 T cell-initiated blood-brain barrier disruption is independent of neutrophil support.
J. Immunol.
189
:
1937
1945
.
37
Johnson
,
H. L.
,
Y.
Chen
,
G. L.
Suidan
,
J. R.
McDole
,
A. K.
Lohrey
,
L. M.
Hanson
,
F.
Jin
,
I.
Pirko
,
A. J.
Johnson
.
2012
.
A hematopoietic contribution to microhemorrhage formation during antiviral CD8 T cell-initiated blood-brain barrier disruption.
J. Neuroinflammation
9
:
60
.
38
Johnson
,
H. L.
,
F.
Jin
,
I.
Pirko
,
A. J.
Johnson
.
2014
.
Theiler’s murine encephalomyelitis virus as an experimental model system to study the mechanism of blood-brain barrier disruption.
J. Neurovirol.
20
:
107
112
.
39
Johnson
,
H. L.
,
R. C.
Willenbring
,
F.
Jin
,
W. A.
Manhart
,
S. J.
LaFrance
,
I.
Pirko
,
A. J.
Johnson
.
2014
.
Perforin competent CD8 T cells are sufficient to cause immune-mediated blood-brain barrier disruption.
PLoS One
9
: e111401.
40
Suidan
,
G. L.
,
J. W.
Dickerson
,
H. L.
Johnson
,
T. W.
Chan
,
K. D.
Pavelko
,
I.
Pirko
,
K. B.
Seroogy
,
A. J.
Johnson
.
2012
.
Preserved vascular integrity and enhanced survival following neuropilin-1 inhibition in a mouse model of CD8 T cell-initiated CNS vascular permeability.
J. Neuroinflammation
9
:
218
.
41
Malo
,
C. S.
,
M. A.
Huggins
,
E. N.
Goddery
,
H. M. A.
Tolcher
,
D. N.
Renner
,
F.
Jin
,
M. J.
Hansen
,
L. R.
Pease
,
K. D.
Pavelko
,
A. J.
Johnson
.
2018
.
Non-equivalent antigen presenting capabilities of dendritic cells and macrophages in generating brain-infiltrating CD8 + T cell responses.
Nat. Commun.
9
:
633
.
42
Cumba Garcia
,
L. M.
,
A. M.
Huseby Kelcher
,
C. S.
Malo
,
A. J.
Johnson
.
2016
.
Superior isolation of antigen-specific brain infiltrating T cells using manual homogenization technique.
J. Immunol. Methods
439
:
23
28
.
43
Rodriguez
,
M.
,
W. P.
Lafuse
,
J.
Leibowitz
,
C. S.
David
.
1986
.
Partial suppression of Theiler’s virus-induced demyelination in vivo by administration of monoclonal antibodies to immune-response gene products (Ia antigens).
Neurology
36
:
964
970
.
44
Abram
,
C. L.
,
G. L.
Roberge
,
Y.
Hu
,
C. A.
Lowell
.
2014
.
Comparative analysis of the efficiency and specificity of myeloid-Cre deleting strains using ROSA-EYFP reporter mice.
J. Immunol. Methods
408
:
89
100
.
45
Noubade
,
R.
,
S.
Majri-Morrison
,
K. V.
Tarbell
.
2019
.
Beyond cDC1: emerging roles of DC crosstalk in cancer immunity.
Front. Immunol.
10
:
1014
.
46
McCaughtry
,
T. M.
,
T. A.
Baldwin
,
M. S.
Wilken
,
K. A.
Hogquist
.
2008
.
Clonal deletion of thymocytes can occur in the cortex with no involvement of the medulla.
J. Exp. Med.
205
:
2575
2584
.
47
Klein
,
L.
,
B.
Kyewski
,
P. M.
Allen
,
K. A.
Hogquist
.
2014
.
Positive and negative selection of the T cell repertoire: what thymocytes see (and don’t see).
Nat. Rev. Immunol.
14
:
377
391
.
48
Oh
,
J.
,
J. S.
Shin
.
2015
.
The role of dendritic cells in central tolerance.
Immune Netw.
15
:
111
120
.
49
Tsunoda
,
I.
2008
.
Axonal degeneration as a self-destructive defense mechanism against neurotropic virus infection.
Future Virol.
3
:
579
593
.
50
Lipton
,
H. L.
1975
.
Theiler’s virus infection in mice: an unusual biphasic disease process leading to demyelination.
Infect. Immun.
11
:
1147
1155
.
51
Palma
,
J. P.
,
H. G.
Lee
,
M.
Mohindru
,
B. S.
Kang
,
M.
Dal Canto
,
S. D.
Miller
,
B. S.
Kim
.
2001
.
Enhanced susceptibility to Theiler’s virus-induced demyelinating disease in perforin-deficient mice.
J. Neuroimmunol.
116
:
125
135
.
52
Fiette
,
L.
,
C.
Aubert
,
M.
Brahic
,
C. P.
Rossi
.
1993
.
Theiler’s virus infection of beta 2-microglobulin-deficient mice.
J. Virol.
67
:
589
592
.
53
Rodriguez
,
M.
,
A. J.
Dunkel
,
R. L.
Thiemann
,
J.
Leibowitz
,
M.
Zijlstra
,
R.
Jaenisch
.
1993
.
Abrogation of resistance to Theiler’s virus-induced demyelination in H-2b mice deficient in beta 2-microglobulin.
J. Immunol.
151
:
266
276
.
54
Rodriguez
,
M.
,
M. D.
Lindsley
,
M. L.
Pierce
.
1991
.
Role of T cells in resistance to Theiler’s virus infection.
Microb. Pathog.
11
:
269
281
.
55
Suidan
,
G. L.
,
J. R.
Mcdole
,
Y.
Chen
,
I.
Pirko
,
A. J.
Johnson
.
2008
.
Induction of blood brain barrier tight junction protein alterations by CD8 T cells.
PLoS One
3
: e3037.
56
Ling
,
C.
,
Y. I.
Verbny
,
M. I.
Banks
,
M.
Sandor
,
Z.
Fabry
.
2008
.
In situ activation of antigen-specific CD8+ T cells in the presence of antigen in organotypic brain slices.
J. Immunol.
180
:
8393
8399
.
57
Mendez-Fernandez
,
Y. V.
,
M. J.
Hansen
,
M.
Rodriguez
,
L. R.
Pease
.
2005
.
Anatomical and cellular requirements for the activation and migration of virus-specific CD8+ T cells to the brain during Theiler’s virus infection.
J. Virol.
79
:
3063
3070
.
58
Probst
,
H. C.
,
K.
Tschannen
,
B.
Odermatt
,
R.
Schwendener
,
R. M.
Zinkernagel
,
M.
Van Den Broek
.
2005
.
Histological analysis of CD11c-DTR/GFP mice after in vivo depletion of dendritic cells.
Clin. Exp. Immunol.
141
:
398
404
.
59
Steel
,
C. D.
,
S. M.
Hahto
,
R. P.
Ciavarra
.
2009
.
Peripheral dendritic cells are essential for both the innate and adaptive antiviral immune responses in the central nervous system.
Virology
387
:
117
126
.
60
Altintas
,
A.
,
Z.
Cai
,
L. R.
Pease
,
M.
Rodriguez
.
1993
.
Differential expression of H-2K and H-2D in the central nervous system of mice infected with Theiler’s virus.
J. Immunol.
151
:
2803
2812
.
61
Lin
,
X.
,
L. R.
Pease
,
M.
Rodriguez
.
1997
.
Differential generation of class I H-2D- versus H-2K-restricted cytotoxicity against a demyelinating virus following central nervous system infection.
Eur. J. Immunol.
27
:
963
970
.
62
Njenga
,
M. K.
,
L. R.
Pease
,
P.
Wettstein
,
T.
Mak
,
M.
Rodriguez
.
1997
.
Interferon alpha/beta mediates early virus-induced expression of H-2D and H-2K in the central nervous system.
Lab. Invest.
77
:
71
84
.
63
Pavelko
,
K. D.
,
Y.
Mendez-Fernandez
,
M. P.
Bell
,
M. J.
Hansen
,
A. J.
Johnson
,
C. S.
David
,
M.
Rodriguez
,
L. R.
Pease
.
2012
.
Nonequivalence of classical MHC class I loci in ability to direct effective antiviral immunity.
PLoS Pathog.
8
: e1002541.
64
MacNamara
,
K. C.
,
S. J.
Bender
,
M. M.
Chua
,
R.
Watson
,
S. R.
Weiss
.
2008
.
Priming of CD8+ T cells during central nervous system infection with a murine coronavirus is strain dependent.
J. Virol.
82
:
6150
6160
.
65
Herz
,
J.
,
M.
Paterka
,
R. A.
Niesner
,
A. U.
Brandt
,
V.
Siffrin
,
T.
Leuenberger
,
J.
Birkenstock
,
A.
Mossakowski
,
R.
Glumm
,
F.
Zipp
,
H.
Radbruch
.
2011
.
In vivo imaging of lymphocytes in the CNS reveals different behaviour of naïve T cells in health and autoimmunity.
J. Neuroinflammation
8
:
131
.
66
Hughes
,
C. E.
,
R. A.
Benson
,
M.
Bedaj
,
P.
Maffia
.
2016
.
Antigen-presenting cells and antigen presentation in tertiary lymphoid organs.
Front. Immunol.
7
:
481
.
67
Mitsdoerffer
,
M.
,
A.
Peters
.
2016
.
Tertiary lymphoid organs in central nervous system autoimmunity.
Front. Immunol.
7
:
451
.
68
Cose
,
S.
,
C.
Brammer
,
K. M.
Khanna
,
D.
Masopust
,
L.
Lefrançois
.
2006
.
Evidence that a significant number of naive T cells enter non-lymphoid organs as part of a normal migratory pathway.
Eur. J. Immunol.
36
:
1423
1433
.
69
Allan
,
R. S.
,
J.
Waithman
,
S.
Bedoui
,
C. M.
Jones
,
J. A.
Villadangos
,
Y.
Zhan
,
A. M.
Lew
,
K.
Shortman
,
W. R.
Heath
,
F. R.
Carbone
.
2006
.
Migratory dendritic cells transfer antigen to a lymph node-resident dendritic cell population for efficient CTL priming.
Immunity
25
:
153
162
.
70
Brabb
,
T.
,
P.
von Dassow
,
N.
Ordonez
,
B.
Schnabel
,
B.
Duke
,
J.
Goverman
.
2000
.
In situ tolerance within the central nervous system as a mechanism for preventing autoimmunity.
J. Exp. Med.
192
:
871
880
.
71
Reynoso
,
G. V.
,
A. S.
Weisberg
,
J. P.
Shannon
,
D. T.
McManus
,
L.
Shores
,
J. L.
Americo
,
R. V.
Stan
,
J. W.
Yewdell
,
H. D.
Hickman
.
2019
.
Lymph node conduits transport virions for rapid T cell activation.
Nat. Immunol.
20
:
602
612
.
72
Milo
,
I.
,
A.
Sapoznikov
,
V.
Kalchenko
,
O.
Tal
,
R.
Krauthgamer
,
N.
van Rooijen
,
D.
Dudziak
,
S.
Jung
,
G.
Shakhar
.
2013
.
Dynamic imaging reveals promiscuous crosspresentation of blood-borne antigens to naive CD8+ T cells in the bone marrow.
Blood
122
:
193
208
.
73
Pozzi
,
L. A.
,
J. W.
Maciaszek
,
K. L.
Rock
.
2005
.
Both dendritic cells and macrophages can stimulate naive CD8 T cells in vivo to proliferate, develop effector function, and differentiate into memory cells.
J. Immunol.
175
:
2071
2081
.
74
Jarry
,
U.
,
P.
Jeannin
,
L.
Pineau
,
S.
Donnou
,
Y.
Delneste
,
D.
Couez
.
2013
.
Efficiently stimulated adult microglia cross-prime naive CD8+ T cells injected in the brain.
Eur. J. Immunol.
43
:
1173
1184
.
75
Olson
,
J. K.
,
A. M.
Girvin
,
S. D.
Miller
.
2001
.
Direct activation of innate and antigen-presenting functions of microglia following infection with Theiler’s virus.
J. Virol.
75
:
9780
9789
.
76
Coder
,
B.
,
W.
Wang
,
L.
Wang
,
Z.
Wu
,
Q.
Zhuge
,
D. M.
Su
.
2017
.
Friend or foe: the dichotomous impact of T cells on neuro-de/re-generation during aging.
Oncotarget
8
:
7116
7137
.
77
McGavern
,
D. B.
,
D.
Homann
,
M. B.
Oldstone
.
2002
.
T cells in the central nervous system: the delicate balance between viral clearance and disease.
J. Infect. Dis.
186
(
Suppl. 2
):
S145
S151
.

The authors have no financial conflicts of interest.