Strain differences in thymic atrophy in rats immunized for EAE correlate with the clinical outcome of immunization

Abstract

An accumulating body of evidence suggests that development of autoimmune pathologies leads to thymic dysfunction and changes in peripheral T-cell compartment, which, in turn, perpetuate their pathogenesis. To test this hypothesis, thymocyte differentiation/maturation in rats susceptible (Dark Agouti, DA) and relatively resistant (Albino Oxford, AO) to experimental autoimmune encephalomyelitis (EAE) induction was examined. Irrespective of strain, immunization for EAE (i) increased the circulating levels of IL-6, a cytokine causally linked with thymic atrophy, and (ii) led to thymic atrophy reflecting partly enhanced thymocyte apoptosis associated with downregulated thymic IL-7 expression. Additionally, immunization diminished the expression of Thy-1, a negative regulator of TCRαβ-mediated signaling and activation thresholds, on CD4+CD8+ TCRαβlo/hi thymocytes undergoing selection and thereby impaired thymocyte selection/survival. This diminished the generation of mature CD4+ and CD8+ single positive TCRαβhi thymocytes and, consequently, CD4+ and CD8+ recent thymic emigrants. In immunized rats, thymic differentiation of natural regulatory CD4+Foxp3+CD25+ T cells (nTregs) was particularly affected reflecting a diminished expression of IL-7, IL-2 and IL-15. The decline in the overall thymic T-cell output and nTreg generation was more pronounced in DA than AO rats. Additionally, differently from immunized AO rats, in DA ones the frequency of CD28- cells secreting cytolytic enzymes within peripheral blood CD4+ T lymphocytes increased, as a consequence of thymic atrophy-related replicative stress (mirrored in CD4+ cell memory pool expansion and p16INK4a accumulation). The higher circulating level of TNF-α in DA compared with AO rats could also contribute to this difference. Consistently, higher frequency of cytolytic CD4+ granzyme B+ cells (associated with greater tissue damage) was found in spinal cord of immunized DA rats compared with their AO counterparts. In conclusion, the study indicated that strain differences in immunization-induced changes in thymopoiesis and peripheral CD4+CD28- T-cell generation could contribute to rat strain-specific clinical outcomes of immunization for EAE.

Fig 1. Immunization for EAE reduced the thymic weight and thymocyte yield more effectively in DA than in AO rats.

Scatter plots indicate (A) thymic weight, thymocyte yield, normalized thymic weight (thymic weight/BW) and normalized thymocyte yield (per 100 mg thymic tissue) and (B) circulating levels of IL-6 and thymic expression of IL-6 mRNA in non-immunized and immunized for EAE DA and AO rats. (C) Representative flow cytometry contour plots show Ki-67 staining of thymocytes retrieved from non-immunized and immunized for EAE DA and AO rats (gating strategy is displayed in S2A Fig). Scatter plot indicates the frequency of Ki-67+ cells among thymocytes from non-immunized and immunized for EAE DA and AO rats. (D) Representative flow cytometry density plots show MC540 staining of thymocytes of non-immunized and immunized for EAE DA and AO rats, cultured for 18h in culture medium to assess cell apoptosis (see Materials and methods). Scatter plot indicates the frequency of MC540+ thymocytes of non-immunized and immunized for EAE DA and AO rats. (E) Scatter plot indicates IL-7 mRNA expression in thymic tissue of non-immunized and immunized for EAE DA and AO rats. Two way ANOVA showed significant interaction between the effects of strain and immunization for thymic weight (F_(1,20) = 6.954, p<0.01), normalized thymic weight (F_(1,20) = 14.63, p<0.01), thymocyte yield (F_(1,20) = 69.94, p<0.001), normalized thymocyte yield (F_(1,20) = 148.1, p<0.001), frequency of MC540+ thymocytes (F_(1,20) = 19.37, p<0.001), circulating levels of IL-6 (F_(1,20) = 19.60, p<0.001) and thymic expression of IL-6 mRNA (F_(1,20) = 35.57, p<0.001). Data points, means and ± SD are from one of two experiments with similar results (n = 6). * p<0.05; ** p<0.01; *** p<0.001.

Fig 2. Immunization for EAE affected the composition of thymocyte subsets delineated by CD4/CD8 expression.

Representative flow cytometry dot plots show CD4/CD8 staining of thymocytes retrieved from non-immunized and immunized for EAE DA and AO rats (gating strategy is displayed in S3A Fig). Scatter plots indicate the frequency of CD4-CD8- double negative (DN), CD4+CD8+ double positive (DP), and CD4+ and CD8+ single positive (SP) thymocytes from non-immunized and immunized for EAE DA and AO rats. Two way ANOVA showed significant interaction between the effects of strain and immunization for the frequency of CD4+ SP thymocytes (F_(1,20) = 9.376, p<0.001). Data points, means and ± SD are from one of two experiments with similar results (n = 6). * p<0.05; ** p<0.01; *** p<0.001.

Fig 3. Immunization for EAE affected the frequency of DN TCRαβ^- thymocytes in rats of both strains and the frequencies of the least mature CD45RC+CD2- DN and CD45RC+CD2+ DN thymocytes in DA rats.

Scatter plots indicate: (A) the frequency of CD4−CD8− double negative (DN) TCRαβ^- cells among thymocytes from non-immunized and immunized for EAE DA and AO rats (gating strategy is displayed in S3A and S3B Fig); (B) the frequency of CD45RC+CD2- and CD45RC+CD2+ DN thymocytes in non-immunized and immunized for EAE DA and AO rats (gating strategy for CD45RC+CD2-/+ cells is displayed in S4 Fig and DN thymocytes are gated as shown in Fig 2) and (C) CXCL12 mRNA expression in thymic tissue from non-immunized and immunized for EAE DA and AO rats. (D) Representative flow cytometry density plots show MC540 staining of DN thymocytes of non-immunized and immunized for EAE DA and AO rats, cultured for 18h in culture medium to assess cell apoptosis (see Materials and methods). Scatter plot indicates the frequency of apoptotic MC540+ cells within DN thymocytes of non-immunized and immunized for EAE DA and AO rats. (E) Representative flow cytometry contour plots show Ki-67 staining of DN thymocytes from non-immunized and immunized for EAE DA and AO rats (gating strategy is displayed in S2B Fig). Scatter plot indicates the frequency of Ki-67+ cells within DN thymocytes of non-immunized and immunized for EAE DA and AO rats. Two way ANOVA showed significant interaction between the effects of strain and immunization for the frequencies of DN TCRαβ^- (F_(1,20) = 73.35, p<0.001), CD45RC+CD2- DN (F_(1,20) = 63.06, p<0.001) and CD45RC+CD2+ DN (F_(1,20) = 160.3, p<0.001) cells within thymocytes, the frequency of MC540+ cells within DN thymocytes (F_(1,20) = 232.1, p<0.001), and thymic CXCL12 mRNA expression (F_(1,20) = 49.48, p<0.001). Data points, means and ± SD are from one of two experiments with similar results (n = 6). * p<0.05; ** p<0.01; *** p<0.001.

Fig 4. Effects of immunization for EAE on DP thymocyte phenotypic profile.

(A) Scatter plots indicate the frequencies of CD4+CD8+ double positive (DP) thymocytes with non-detectable (TCRαβ^-), low (TCRαβ^lo) and high (TCRαβ^hi) levels of TCRαβ surface expression among thymocytes from non-immunized and immunized for EAE DA and AO rats (gating strategy is displayed in S3A and S3B Fig). (B,C) Representative overlaid histograms indicate CD90 (Thy-1) surface expression on DP (B) TCRαβ^lo and (C) TCRαβ^hi thymocytes of (dashed line) non-immunized and (full line) immunized for EAE DA and AO rats. Data in the overlaid flow cytometry histograms are displayed as % of Max (the cell count in each bin divided by the cell count in the bin that contained the largest number of cells; http://www.flowjo.com), to allow visual comparison of samples with different event numbers collected. Scatter plots show the fold change in CD90 mean fluorescence intensity (MFI) ratio on CD90+ DP (B) TCRαβ^lo and (C) TCRαβ^hi thymocytes of non-immunized AO and immunized for EAE DA and AO rats, relative to non-immunized DA animals. Two way ANOVA showed significant interaction between the effects of strain and immunization for the frequency of DP TCRαβ^hi (F_(1,20) = 6.987, p<0.05) cells. Data points, means and ± SD are from one of two experiments with similar results (n = 6). * p<0.05; ** p<0.01; *** p<0.001.

Fig 5. Immunization for EAE decreased the number of CD4+ and CD8+ SP TCRαβ^hi thymocytes more prominently in DA than in AO rats.

(A) Scatter plots indicate the frequency of CD4+ and CD8+ TCRαβ^hi single positive (SP) thymocytes of non-immunized and immunized for EAE DA and AO rats (gating strategy is displayed in S3A and S3B Fig). (B) Representative flow cytometry contour plots show Ki-67 staining of CD4+ and CD8+ SP TCRαβ^hi thymocytes of non-immunized and immunized for EAE DA and AO rats (gating strategy is displayed in S2C and S2D Fig). Scatter plots indicate the frequency of Ki-67+ cells within CD4+ and CD8+ SP TCRαβ^hi thymocytes from non-immunized and immunized for EAE DA and AO rats. (C) Scatter plots indicate the number of CD4+ and CD8+ SP TCRαβ^hi thymocytes in non-immunized and immunized for EAE DA and AO rats. Two way ANOVA showed significant interaction between the effects of strain and immunization for the frequency of CD4+ (F_(1,20) = 55.12, p<0.001) and CD8+ SP TCRαβ^hi (F_(1,20) = 20.97, p<0.001) thymocytes, and the number of CD4+ (F_(1,20) = 229.8, p<0.001) and CD8+ SP TCRαβ^hi (F_(1,20) = 64.32, p<0.001) thymocytes. Data points, means and ± SD are from one of two experiments with similar results (n = 6). * p<0.05; ** p<0.01; *** p<0.001.

Fig 6. Immunization for EAE diminished the frequency and number of regulatory CD4+CD25+Foxp3+ thymocytes in DA more prominently than in AO rats.

Scatter plots indicate (A) the frequency and the number of CD4+CD25+Foxp3+ thymocytes (gating strategy is displayed in S3A and S5 Figs) and (B) IL-2 and IL-15 mRNA expression in thymic tissue of non-immunized and immunized for EAE DA and AO rats. Two way ANOVA showed significant interaction between the effects of strain and immunization for the frequency (F_(1,20) = 152.2, p<0.001) and number (F_(1,20) = 7.296, p<0.05) of CD4+CD25+Foxp3+ thymocytes, and for thymic IL-2 (F_(1,20) = 28.31, p<0.001) and IL-15 (F_(1,20) = 64.0, p<0.001) mRNA expression. Data points, means and ± SD are from one of two experiments with similar results (n = 6). * p<0.05; ** p<0.01; *** p<0.001.

Fig 7. Immunization for EAE decreased the counts of CD4+ and CD8+ T-PBLs in DA rats.

Scatter plots indicate (A) the number of T-peripheral blood lymphocytes (T-PBLs)/ml, and (B) (upper) the frequency and (lower) the number of CD4+ and CD8+ T-PBLs/ml retrieved from non-immunized and immunized for EAE DA and AO rats (gating strategy for T-PBLs and CD4+ and CD8+ T-PBLs is displayed in S6 Fig). Two way ANOVA showed significant interaction between the effects of strain and immunization for the frequency of CD8+ T-PBLs (F_(1,20) = 29.50, p<0.001), and the numbers of T-PBLs (F_(1,20) = 26.16, p<0.001), CD4+ (F_(1,20) = 69.30, p<0.001) and CD8+ (F_(1,20) = 193.9, p<0.001) T-PBLs. Data points, means and ± SD are from one of two experiments with similar results (n = 6). * p<0.05; ** p<0.01; *** p<0.001.

Fig 8. Immunization for EAE decreased the frequency of CD90+CD45RC-, but increased that of CD90-CD45RC- cells among the major subpopulations of T-PBLs from DA rats.

Representative flow cytometry dot plots show CD90/CD45RC staining of (A) CD4+ and (B) CD8+ T-peripheral blood lymphocytes (T-PBLs) of non-immunized and immunized for EAE DA and AO rats (gating strategy for CD4+ and CD8+ T-PBLs is displayed in S6 Fig and for CD90+CD45RC- and CD90-CD45RC- cells in S7 Fig). Scatter plots indicate the frequency of CD90+CD45RC- cells (RTEs) and CD90-CD45RC- (memory phenotype) cells within (A) CD4+ and (B) CD8+ T-PBLs. Two way ANOVA showed a significant interaction between the effects of strain and immunization for the frequency of CD90-CD45RC- cells within CD4+ T-PBLs (F_(1,20) = 28.52, p<0.001). Data points, means and ± SD are from one of two experiments with similar results (n = 6). * p<0.05; ** p<0.01; *** p<0.001.

Fig 9. Immunization for EAE increased the frequency of activated and proliferating cells among the major subpopulations of T-PBLs.

(A) Representative flow cytometry dot plots show CD25 expression on CD4+ and CD8+ T-peripheral blood lymphocytes (T-PBLs) from non-immunized and immunized for EAE DA and AO rats (gating strategy for CD4+ and CD8+ T-PBLs is displayed in S6 Fig). Scatter plots indicate the frequency of activated CD25+ cells among CD4+ and CD8+ T-PBLs from non-immunized and immunized for EAE DA and AO rats. (B) Representative flow cytometry dot plots show Ki-67 expression in CD4+CD28+ (R2) and CD8+CD28+ T-PBLs (R2) from non-immunized and immunized for EAE DA and AO rats. Please note that virtually there are no Ki-67+ cells among either CD4+CD28- (R1) or CD8+CD28- (R1) cells. Dashed line indicates cut off between Ki-67+ and Ki-67- cells. Scatter plots show the frequency of proliferating Ki-67+ cells among CD4+CD28+ and CD8+CD28+ T-PBLs from non-immunized and immunized for EAE DA and AO rats. Two way ANOVA showed significant interaction between the effects of strain and immunization for the frequency of CD25+ cells among CD4+ T-PBLs (F_(1,20) = 22.73, p<0.001), Ki-67+ cells among CD4+CD28+ (F_(1,20) = 30.04, p<0.001) and CD8+CD28+ (F_(1,20) = 7.182, p<0.05) T-PBLs. Data points, means and ± SD are from one of two experiments with similar results (n = 6). * p<0.05; *** p<0.001.

Fig 10. Immunization for EAE increased the frequency of CD28- cells among the major subpopulations of T-PBLs and elevated the circulating levels of TNF-α in DA rats.

(A) Representative flow cytometry dot plots show CD28 staining of CD4+ and CD8+ T-peripheral blood lymphocytes (T-PBLs) from non-immunized and immunized for EAE DA and AO rats (gating strategy for CD4+ and CD8+ T-PBLs is displayed in S6 Fig and for CD28- cells among them in S9 Fig). Scatter plots indicate the frequency of CD28- cells among CD4+ and CD8+ T-PBLs from non-immunized and immunized for EAE DA and AO rats. (B) Representative overlaid flow cytometry histograms show p16^INK4a expression in CD4+ and CD8+ T-PBLs from non-immunized (dark gray line) and immunized for EAE (black line) DA and AO rats. Scatter plots indicate the fold change in p16^INK4a mean fluorescence intensity (MFI) in CD4+ and CD8+ T-PBLs of non-immunized AO and immunized for EAE DA and AO rats, relative to non-immunized DA animals. Data in the overlaid flow cytometry histograms are displayed as % of Max (the cell count in each bin divided by the cell count in the bin that contained the largest number of cells; http://www.flowjo.com), to allow visual comparison of samples with different event numbers collected. (C) Scatter plot indicates the concentration of TNF-α in plasma of non-immunized and immunized for EAE DA and AO rats. Two way ANOVA showed significant interaction between the effects of strain and immunization for the frequency of CD28- among CD4+ (F_(1,20) = 125.8, p<0.001) and CD8+ (F_(1,20) = 44.12, p<0.001) T-PBLs. Data points, means and ± SD are from one of two experiments with similar results (n = 6). * p<0.05; ** p<0.001; *** p<0.001.

Fig 11. Greater frequency of granzyme B+ CD4+ T cells in SC mononuclear infiltrate from DA rats immunized for EAE than in their AO counterparts.

Flow cytometry dot plots show (A) gating strategy for CD4+TCRαβ+ lymphocytes, (B) fluorescence minus one (FMO) control staining for setting cut off boundary for analysis of granzyme B expression in CD4+TCRαβ+ lymphocytes gated as shown in (A), and granzyme B expression in CD4+TCRαβ+ lymphocytes (fully stained sample). (C) Scatter plot indicates the frequency of granzyme B+ CD4+ T cells in mononuclear cell infiltrate retrieved from spinal cord (SC) of DA and AO rats immunized for EAE. Student’s t-test was used to assess the statistical significance of differences between DA and AO rats. Data points, means and ± SD are from one of two experiments with similar results (n = 6). *** p<0.001.