Identification and characterization of T reg-like cells in zebrafish

Abstract

Regulatory T (T reg) cells are a specialized sublineage of T lymphocytes that suppress autoreactive T cells. Functional studies of T reg cells in vitro have defined multiple suppression mechanisms, and studies of T reg-deficient humans and mice have made clear the important role that these cells play in preventing autoimmunity. However, many questions remain about how T reg cells act in vivo. Specifically, it is not clear which suppression mechanisms are most important, where T reg cells act, and how they get there. To begin to address these issues, we sought to identify T reg cells in zebrafish, a model system that provides unparalleled advantages in live-cell imaging and high-throughput genetic analyses. Using a FOXP3 orthologue as a marker, we identified CD4-enriched, mature T lymphocytes with properties of T reg cells. Zebrafish mutant for foxp3a displayed excess T lymphocytes, splenomegaly, and a profound inflammatory phenotype that was suppressed by genetic ablation of lymphocytes. This study identifies T reg-like cells in zebrafish, providing both a model to study the normal functions of these cells in vivo and mutants to explore the consequences of their loss.

Figure 1.

Foxp3 phylogeny and identification of foxp3a:EGFP-positive lymphocytes. (A) Phylogenetic relationship of zebrafish Foxp3 proteins to Foxp3 proteins from other species (human, Homo sapiens, H.s.; mouse, Mus musculus, M.m.; frog, Xenopus laevis, X.l.; pufferfish, Tetraodon nigroviridis, T.n.) and other human Foxp family proteins. Unrooted neighbor-joining best trees were produced using MacVector v12.5.1. Sequences were aligned by ClustalW. Distance was absolute, and gaps were distributed proportionally. GenBank accession nos. for sequences used were D.r. Foxp3a (NP_001316496), D.r. Foxp3b (XM_021478427), H.s. Foxp1 (NP_001231745), H.s. Foxp2 (NP_055306), H.s. Foxp3 (ABQ15210), H.s. Foxp4 (NP_001012426), M.m. Foxp3 (NP_001186277), X.l. Foxp3 (NP_001121199), and T.n. Foxp3 (ADD91631). (B) qRT-PCR of foxp3 paralogs in wild-type AB WKM relative to nonhematopoietic fin tissue. Error bars indicate SEM; n = 3. (C) Flow cytometry analysis of WKM from representative AB and Tg(foxp3a:EGFP) animals. Gates of major hematopoietic lineages are indicated at right. (D) Percentages of EGFP-positive cells in each gate are shown, and data were used to calculate fold differences of EGFP-positive cells in Tg(foxp3a:EGFP; n = 5) and Tg(foxp3a:EGFP); rag1(lf) (n = 5) versus intrinsically fluorescent cells in wild-type AB animals (n = 5). Two-tailed Student’s t test, Tg(foxp3a:EGFP) versus Tg(foxp3a:EGFP); rag1(lf), **, P < 0.001. (E) Cytological stains of sorted foxp3a:EGFP-positive cells compared with sorted lck:EGFP-positive lymphocytes. Individual cells were extracted and aligned for comparative purposes. (F) Flow cytometry of WKM from a representative Tg(foxp3a:EGFP); rag1(lf) animal showing absence of the EGFP-positive population. (G) qRT-PCR of foxp3 paralogs in foxp3a:EGFP-positive cells from WKM compared with bulk WKM lymphocytes. †, Expression of foxp3b was below the limit of detection in foxp3a:EGFP-positive and bulk lymphocytes. Error bar indicates SEM; n = 3.

Figure 2.

Identification and analysis of foxp3a:EGFP-positive thymocytes. (A) Flow cytometry analysis of thymus from representative AB and Tg(foxp3a:EGFP) animals. (B) qRT-PCR of selected genes in foxp3a:EGFP-positive thymocytes relative to bulk thymocytes. Relative CD4 and CD8 expression is also shown and represents zebrafish cd4-1 and cd8a genes. †, Expression of cd8a was below the limit of detection in foxp3a:EGFP-positive thymocytes. Error bars indicate SEM.; n = 3. (C) Cells from dissected thymus of a Tg(foxp3a:EGFP); Tg(rag2:mCherry) animal showing a double-positive thymocyte (arrowheads). Bars, 10 µm.

Figure 3.

Expression analyses of foxp3a:EGFP-positive cells. (A) Heat map of T reg signature genes as defined by Sugimoto et al. (2006). Mouse gene names are indicated. Only genes with a normalized read count >10 are shown, as are genes with unambiguous zebrafish-to-mouse orthology, except for Foxp3, for which expression values of foxp3a are represented. (B) Top, GSEA showing that expression of a T reg signature, as defined by Sugimoto et al. (2006), is enriched in zebrafish foxp3a:EGFP cells. Bottom, GSEA showing enrichment of a broader T reg signature, as defined by Hill et al. (2007). (C) Violin plots from single-cell analyses showing that the expression profile of foxp3a:EGFP-positive cells is biased to that of mature (TCR⁺) cd4-1⁺cd8a^– lymphocytes. Cell populations were assigned from single-cell analyses of sorted lck:EGFP-positive (Lck) and thymic (double-negative [DN], double-positive [DP], CD4, and CD8) cells. By ANOVA, foxp3a (P = 1.61 × 10^–22), cd4-1 (P = 1.02 × 10^–36), cd8a (P = 1.35 × 10^–58), TCRα (P = 2.60 × 10^–6), and TCRβ (P = 5.18 × 10^–8).

Figure 4.

foxp3a mutations and effects on survival. (A) Gene structure of foxp3a with sites and sequences of mutations indicated. (B) qRT-PCR of foxp3a in WKM relative to nonhematopoietic fin tissue. foxp3a(um252) animals show a reduction of foxp3a transcripts. Error bars indicate SEM; n = 3. (C) Kaplan–Meier survival analysis of foxp3a mutants compared with wild-type animals. Log-rank (Mantel–Cox) test, *, P < 0.0001 for pairwise comparisons of foxp3a(um252/um252) (n = 37), foxp3a(um253/um253) (n = 33), and foxp3a(um252/um253) (n = 54) versus wild type (n = 59).

Figure 5.

Lymphocyte expansion and lymphocyte-dependent inflammation in foxp3a mutants. (A) Flow cytometry analyses of WKM from representative foxp3a mutant animals. (B) Differential counts of nonerythroid cells from WKM. Counts were obtained from at least 200 cells per WKM cytospin with May-Grünwald Giemsa staining. Two-tailed Student’s t test (foxp3a mutant vs. AB), *, P < 0.05; **, P < 0.001. (C) Representative images of hematoxylin and eosin–stained sagittal sections of spleens from wild-type AB and foxp3a(um252) mutants. Images showing largest extent of splenic size are shown. Spleens are shown bordered by dashed lines. Bars, 250 µm. (D) Representative images of dissected spleens from Tg(lck:EGFP) and Tg(lck:EGFP); foxp3a(um252) animals. Spleens are shown bordered by dotted lines. Bars, 250 µm. (E) Flow cytometry analysis of spleens from AB (n = 5), Tg(lck:EGFP) (n = 5) and Tg(lck:EGFP); foxp3a(um252) (n = 5) animals. Percentages of EGFP-positive cells in the lymphocyte gate are shown. Two-tailed Student’s t test, *, P < 0.05. (F) Representative images of hematoxylin and eosin–stained sagittal sections of urogenital pores from wild-type AB, foxp3a(um252), and foxp3a(um252); rag1(lf) animals. In total, 0/4 wild-type AB, 6/6 foxp3a(um252), and 0/3 foxp3a(um252); rag1(lf) animals displayed connective tissue inflammation. Insets, increased cellularity characterized by marked influx of lymphocyte-predominant inflammatory cells (arrowheads) in foxp3a(um252) mutants compared with the wild-type AB and foxp3a(um252); rag1(lf) double mutants. Bars: 100 µm; (inset) 10 µm. (G) Representative images of hematoxylin and eosin–stained sagittal sections of skin from wild-type AB, foxp3a(um252), and foxp3a(um252); rag1(lf) animals. In total, 0/4 wild-type AB, 5/6 foxp3a(um252), and 0/3 foxp3a(um252); rag1(lf) animals displayed skin inflammation. Bar, 25 µm. (H) qRT-PCR analysis of inflammation marker genes ifng1-2 and il1b. ΔCt values were calculated relative to a β-actin control. Two-tailed Student’s t test, *, P < 0.05; ns, not significant. Error bar indicates SEM; n = 3.