Figure 2
Figure 2. Peritransplant administration of Fgf7 sustains a more diverse TRA transcriptome in mTEChigh during aGVHD. Acute GVHD (b→bd) was induced in 8-week-old unconditioned recipient BDF1 mice as in Figure 1A. Global gene expression was analyzed in entire residual mTEChigh cell pools isolated from individual recipients at 10, 12, and 15 weeks of age (ie, 2, 4, and 7 weeks after transplantation). The figure reflects data from 5 independent experiments, with 2 to 4 mice per group and experiment. (A) Left panel: gene expression profiling of mTEChigh in transplant recipients using Mouse Gene 1.0 ST arrays (Affymetrix). The x- and y-axes represent relative expression of individual genes (given as log2 values of signal intensity ratios, S.I.) in mice with aGVHD (b→bd) vs syngeneically transplanted recipients without disease (bd→bd). Data were plotted as a function of recipient age (10 vs 12 weeks). Genes whose expression levels were enhanced or decreased more than threefold relative to the control group (corresponding to relative expression values given by the S.I. ratios of <0.33 and >3.0, respectively, or <−1.6 and >+1.6 on the log2 scale) were considered to be significantly altered in consequence of aGVHD (threshold is indicated by dashed lines). Q2 is the lower left quadrant, denoting genes whose expression levels were depressed at both 2 and 4 weeks after allo-HSCT (10- and 12-week-old recipients). The inserted numbers denote the absolute numbers of genes in the respective quadrants. Transcripts from a total of 21 759 genes were analyzed. An annotated list of the genes most affected by aGVHD is given in supplemental Table 1. Ub (ubiquitously expressed genes, gray squares) and TRA (dark blue squares). (Right panel) O/E performance analysis (O, observed number of events; E, expected number of events) of the presence of TRA vs Ub among the genes with relative expression <0.33 and >0.3. O/E ratio >1.0 for TRA and <1.0 for Ub; *P < .001 using hypergeometric statistical analysis. A value of O/E = 1.0 represents a distribution among TRA and Ub according to expectations (that is sequestration according to the relative frequencies of TRA and Ub in adult mice without disease). (B) The linear relationship between 2 experimental groups was tested for Ub and TRA using a correlation matrix, as detailed in supplemental Figure 3. The variation of each gene was calculated using the interquartile range, a measure of statistical dispersion. Pearson’s correlation coefficient r is shown as a function of age and time after transplantation. The asterisk (*) symbol indicates a statistical P value for the quality of each measured r at any given time point for TRA and Ub, respectively; *P < .001 of Pearson’s r. (C) Tissue representation of TRA repressed during aGVHD (<0.1 and <0.33 relative expression, respectively, vs mice without GVHD at 10 and 12 weeks of age). As control, a random set of 200 TRAs was used that remained unchanged in the course of GVHD (relative expression ∼1). Many repressed TRAs were specific for tissues of the gastrointestinal tract, skin, and eye. Changes in expression of these genes in response to Fgf7 are given in supplemental Figure 4. (D) Aire dependency of TRA repressed in mTEChigh during aGVHD in 10- and 12-week-old recipients (quadrant Q2; the axes are the same as in panel A). Examples shown are the following: Scgb1a1, secretoglobin; Csn1s2a, casein γ, Mup1, major urinary protein 1; Spt1, salivary protein 1 (see supplemental Table 1). Aire-dependent TRA (yellow circles), Aire-independent TRA (dark blue circles), and Ub (gray circles). (E) Analysis of TEC proliferation in response to Fgf7. Eight-week-old BDF1 mice were left untreated or were treated with Fgf7 from days −3 to +3 after allo-HSCT. All mice simultaneously received 0.8 mg/mL BrdU in their drinking water. mTECs were analyzed for BrdU incorporation at the indicated ages after transplantation. BrdU+ cells (%) ± SD, *P < .001, ANOVA, aGVHD vs mice without aGVHD. (F) Expression levels of individual genes expressed in mTEChigh is given in a heatmap as color-coded log2 value of signal intensities. (Left panel) The map was sorted according to relative expression of genes in mTEChigh (b→bd vs bd→bd mice; 12 weeks old) and contains all TRA with a relative expression of <0.3 during aGVHD in age-matched mice. Enhancement of TRA expression in response to Fgf7 was detectable by higher signal intensities of individual genes. (Right panel) Aire-dependent TRAs were sorted as listed before.

Peritransplant administration of Fgf7 sustains a more diverse TRA transcriptome in mTEChigh during aGVHD. Acute GVHD (b→bd) was induced in 8-week-old unconditioned recipient BDF1 mice as in Figure 1A. Global gene expression was analyzed in entire residual mTEChigh cell pools isolated from individual recipients at 10, 12, and 15 weeks of age (ie, 2, 4, and 7 weeks after transplantation). The figure reflects data from 5 independent experiments, with 2 to 4 mice per group and experiment. (A) Left panel: gene expression profiling of mTEChigh in transplant recipients using Mouse Gene 1.0 ST arrays (Affymetrix). The x- and y-axes represent relative expression of individual genes (given as log2 values of signal intensity ratios, S.I.) in mice with aGVHD (b→bd) vs syngeneically transplanted recipients without disease (bd→bd). Data were plotted as a function of recipient age (10 vs 12 weeks). Genes whose expression levels were enhanced or decreased more than threefold relative to the control group (corresponding to relative expression values given by the S.I. ratios of <0.33 and >3.0, respectively, or <−1.6 and >+1.6 on the log2 scale) were considered to be significantly altered in consequence of aGVHD (threshold is indicated by dashed lines). Q2 is the lower left quadrant, denoting genes whose expression levels were depressed at both 2 and 4 weeks after allo-HSCT (10- and 12-week-old recipients). The inserted numbers denote the absolute numbers of genes in the respective quadrants. Transcripts from a total of 21 759 genes were analyzed. An annotated list of the genes most affected by aGVHD is given in supplemental Table 1. Ub (ubiquitously expressed genes, gray squares) and TRA (dark blue squares). (Right panel) O/E performance analysis (O, observed number of events; E, expected number of events) of the presence of TRA vs Ub among the genes with relative expression <0.33 and >0.3. O/E ratio >1.0 for TRA and <1.0 for Ub; *P < .001 using hypergeometric statistical analysis. A value of O/E = 1.0 represents a distribution among TRA and Ub according to expectations (that is sequestration according to the relative frequencies of TRA and Ub in adult mice without disease). (B) The linear relationship between 2 experimental groups was tested for Ub and TRA using a correlation matrix, as detailed in supplemental Figure 3. The variation of each gene was calculated using the interquartile range, a measure of statistical dispersion. Pearson’s correlation coefficient r is shown as a function of age and time after transplantation. The asterisk (*) symbol indicates a statistical P value for the quality of each measured r at any given time point for TRA and Ub, respectively; *P < .001 of Pearson’s r. (C) Tissue representation of TRA repressed during aGVHD (<0.1 and <0.33 relative expression, respectively, vs mice without GVHD at 10 and 12 weeks of age). As control, a random set of 200 TRAs was used that remained unchanged in the course of GVHD (relative expression ∼1). Many repressed TRAs were specific for tissues of the gastrointestinal tract, skin, and eye. Changes in expression of these genes in response to Fgf7 are given in supplemental Figure 4. (D) Aire dependency of TRA repressed in mTEChigh during aGVHD in 10- and 12-week-old recipients (quadrant Q2; the axes are the same as in panel A). Examples shown are the following: Scgb1a1, secretoglobin; Csn1s2a, casein γ, Mup1, major urinary protein 1; Spt1, salivary protein 1 (see supplemental Table 1). Aire-dependent TRA (yellow circles), Aire-independent TRA (dark blue circles), and Ub (gray circles). (E) Analysis of TEC proliferation in response to Fgf7. Eight-week-old BDF1 mice were left untreated or were treated with Fgf7 from days −3 to +3 after allo-HSCT. All mice simultaneously received 0.8 mg/mL BrdU in their drinking water. mTECs were analyzed for BrdU incorporation at the indicated ages after transplantation. BrdU+ cells (%) ± SD, *P < .001, ANOVA, aGVHD vs mice without aGVHD. (F) Expression levels of individual genes expressed in mTEChigh is given in a heatmap as color-coded log2 value of signal intensities. (Left panel) The map was sorted according to relative expression of genes in mTEChigh (b→bd vs bd→bd mice; 12 weeks old) and contains all TRA with a relative expression of <0.3 during aGVHD in age-matched mice. Enhancement of TRA expression in response to Fgf7 was detectable by higher signal intensities of individual genes. (Right panel) Aire-dependent TRAs were sorted as listed before.

Close Modal

or Create an Account

Close Modal
Close Modal