Advertisement

Cord blood transplantation recapitulates fetal ontogeny with a distinct molecular signature that supports CD4+ T-cell reconstitution

Prashant Hiwarkar, Mike Hubank, Waseem Qasim, Robert Chiesa, Kimberly C. Gilmour, Aurore Saudemont, Persis J. Amrolia and Paul Veys

Data supplements

Article Figures & Data

Figures

  • Figure 1.

    Immune reconstitution after T-replete CBT and BMT. (A) Bar graph showing T-cells carried with a cord blood and a bone marrow graft. A median of 4 × 106/kg T cells are infused with a cord blood graft compared with 10 times more T cells (45 × 106/kg) infused with a bone marrow graft (P < .0001). The bar graph represents the median, and error bars represent the 25th and 75th centiles. (B) Line graph showing T-cell reconstitution after T-replete CBT and BMT. Despite a 10 times lower number of T cells infused with the cord blood graft, a significantly higher CD3+ T-cell recovery is observed 2 months post-CBT compared with after BMT. (C-D) Line graph showing CD4+ and CD8+ T-cell recovery after CBT and BMT, respectively. The T-cell recovery observed after T-replete CBT was asymmetrically CD4+ T-cell biased in contrast to CD8+ T-cell biased immune reconstitution after T-replete BMT. The dots represent the median, and the error bars represent the 25th and 75th centile. The green line represents CD4:CD8 ratio plotted on the right Y-axis.

  • Figure 2.

    Exploratory analysis of gene expression profiles. (A) Three-dimensional principal component analysis and (B) unsupervised hierarchical clustering of gene expression profile of naive CD4+ T cells from cord blood, peripheral blood, fetal mesenteric lymph nodes and 2 months after CBT and BMT. Naive cord blood CD4+ T cells have a distinct transcription profile to naive peripheral blood CD4+ T cells, but similar to fetal T cells. Cord blood T cells during early reconstitution after CBT retain the fetal-like transcription profile, and thus recapitulate fetal ontogeny. (C) Two-dimensional principal component analysis showing a relationship among naive CD4+ T cells from cord blood, peripheral blood, and fetal mesenteric lymph nodes, and 2 months after CBT and BMT, vs T-regulatory cells from fetal mesenteric lymph nodes and peripheral blood. T cells segregate based on developmental stage and T-cell type. Thus, confirming the distinct transcription profile of naive CD4+ T cells after CBT is not a result of adoption of T-regulatory function.

  • Figure 3.

    Transcriptional signature of naive cord blood CD4+ T cells. (A) Venn diagram of differentially expressed genes in 3 microarray experiments comparing the naive CD4+ T cells from normal donor cord blood and peripheral blood. Sixty genes overlapped in the 3 experiments. (B) Sixty genes that represent the signature of naive cord blood CD4+ T cells. (C) Scatterplot of pairwise global gene expression comparison of naive cord blood CD4+ T cells and naive peripheral blood CD4+ T cells. Gene expression values are plotted on a log scale. Genes that were differentially expressed between groups (determined using P < .05 and fold-change ≥2) are indicated in red and blue. Specific genes that were differentially expressed are highlighted with arrows.

  • Figure 4.

    Transcriptional signature of naive cord blood CD4+ T cells is rich in genes induced in the lymphopenic environment. (A) Scatterplot of pairwise global gene expression comparison comparing gene expression in the 2 posttransplant environments (ie, cord blood transplantation and bone marrow transplantation). Gene expression values are plotted on a log scale, and specific genes upregulated in the lymphopenic environment are highlighted with arrows. (B) Nineteen genes induced in the lymphopenic signature and 41 genes of fetal signature. (C) Bar plot showing mean (and where possible, standard deviation) transcript values of 19 upregulated/downregulated genes representing genes induced in the lymphopenic environment. Interestingly, the differential regulation of 19 genes was higher in the naive CD4+ T cells from the cord blood and during early reconstitution after CBT than compared with bone marrow transplantation.

  • Figure 5.

    Plots showing upregulation of TCR-MAPK-AP1 signaling in the cord blood CD4+ T cells. (A) Gene expression profile of naive CD4+ T cells from cord blood and peripheral blood donors were compared to derive enrichment plots of TCR and MAPK signaling and transcript values of the 2 important transcription factors FOS and JUN (AP-1 complex). FOS and JUN upregulation is expressed as mean (and, where possible, as standard deviation). (B) When naive CD4+ T cells from the 2 posttransplant lymphopenic conditions (ie, CBT vs BMT) were compared, reconstituting CD4+ T cells after CBT were observed to have an upregulated TCR and MAPK signaling.

  • Figure 6.

    AP-1 transcription factor complex mediates rapid proliferation of cord blood CD4+ T cells. (A) Line graph showing increased proliferation of cord blood CD4+ T cells in response to self-APCs. Cord blood CD4+ T-cell proliferation increased with an increasing APC:T-cell ratio of 1:1, 2:1, and 4:1. However, no such effect was observed on peripheral blood CD4+ T cells. (B) Line graph showing inhibition of cord blood CD4+ T-cell proliferation at different concentrations of AP-1 inhibitor. The inhibitory effect was proportional to the increasing concentration of AP-1 inhibitor. The dots represent mean and error bars represent standard deviation.

Tables

  • Table 1.

    Demographics of cord blood and bone marrow recipients that contributed to the T-cell reconstitution study

    CBT (n = 30; 26 single-cord and 4 double-cord)BMT (n = 40)
    Age at transplant
     1 y (0.1-12)4.3 y (0.6-12)
    Diagnosis
     Acute leukemia, 12 (40%); myelodysplastic syndrome/chronic myeloid leukemia, 2 (6%); immunodeficiency, 12 (40%); hemophagocytic lymphohistiocytosis 4 (14%)Acute leukemia, 17 (42%); myelodysplastic syndrome/chronic myeloid leukemia, 3 (8%); immunodeficiency, 13 (32%); hemophagocytic lymphohistiocytosis, 5 (13%); metabolic, 2 (5%)
    Conditioning
     Myeloablative conditioning, 29 (97%; TBI or Bu-based or treosulfan-based); no conditioning, 1 (3%)Myeloablative conditioning, 40 (100%; TBI or Bu-based or treosulfan-based)
    HLA matching
     ≤8/10: 1710/10: 40
     9/10: 14
     10/10: 03
    Acute GVHD
     Grade II: 10 (33%)Grade II: 7 (18%)
     Grade III-IV: 5 (16%)Grade III-IV: 3 (8%)
    Viral reactivations
     CMV: 4 (13%)CMV: 7 (17%)
     ADV: 4 (13%)ADV: 4 (10%)
  • Table 2.

    Significant canonical pathways in naive cord blood CD4+ T cells after enrichment mapping (at P < .005; FDR q value < 0.1) of gene set enrichment analysis pathways

    NameSizeESNESNOM P valueFDR q value
    BIOCARTA_STRESS_PATHWAY250.7478522.15303700.002696
    BIOCARTA_TNFR2_PATHWAY180.7470832.00650100.005757
    BIOCARTA_GPCR_PATHWAY320.6492871.97376900.008621
    KEGG_T_CELL_RECEPTOR_SIGNALING_PATHWAY1070.5119061.95690100.009042
    BIOCARTA_MET_PATHWAY350.630851.94021500.010134
    BIOCARTA_TCR_PATHWAY420.5902071.89187800.014213
    BIOCARTA_MAPK_PATHWAY860.4900171.78545500.023409
    KEGG_MAPK_SIGNALING_PATHWAY2570.3632481.548704.00158730.096791
    BIOCARTA_PPARA_PATHWAY540.5071071.717286.001742160.037614
    BIOCARTA_PYK2_PATHWAY260.6356271.843618.001748250.019434
    BIOCARTA_INTEGRIN_PATHWAY350.5686511.794564.001904760.023841
    BIOCARTA_HSP27_PATHWAY150.6988281.79994.001953130.023302
    BIOCARTA_CDMAC_PATHWAY160.7686641.966973.001996010.008
    BIOCARTA_INSULIN_PATHWAY210.672221.866742.002053390.018618
    BIOCARTA_IL1R_PATHWAY330.5822271.803144.003616640.024056
    BIOCARTA_IGF1_PATHWAY210.6481031.779082.003656310.024106
    BIOCARTA_RELA_PATHWAY160.7153021.845727.00375940.020061
    BIOCARTA_RACCYCD_PATHWAY260.6285281.833218.0038610.02129
    BIOCARTA_IL6_PATHWAY220.6847491.897918.003913890.013879
    • Bold text indicates pathways that are discussed in the “Results” and “Discussion.”

    • ES, enrichment score; FDR, false discovery rate; NES, normalized enrichment score; NOM, nominal.