Figure 4
Figure 4. An epigenetic mechanism controls the transcription of the il15rα gene in normal and leukemic CD8 T cells into negative and positive directions. (A) Presence of a DNase I HS site in IL-15Tg leukemic cells and its position in the transcription control site of the il15rα gene. Intact nuclei were isolated from purified CD8 T cells from terminally leukemic IL-15Tg mice (n = 2) and from K2 and B1 leukemic cells, and subjected to 1 U/mL of DNase I for 15 minutes at 37°C. The genomic DNA was extracted subsequently and digested by the EcoRI enzyme. The digested DNA was subjected to 0.7% Tris-acetate EDTA gel electrophoresis, blotted, and hybridized with a 32P-labeled fragment (as shown by the asterisk in panel B) from the il15rα transcription initiation site. As a control, CD8 T cells purified from WT mice were activated by plate-coated anti-CD3/anti-CD28 Abs and hIL-15 (5nM; PeproTech) for 48 hours before DNase I treatment. The sources of genomic DNAs were: normal CD8 T cells (i), CD8 T cells from leukemic IL-15Tg mouse #1 (ii), CD8 T cells from leukemic IL-15Tg mouse #2 (iii), cells from the K2 leukemic cell line (iv), and cells from the B1 leukemic cell line (v). The size of the upper band (EcoRI-EcoRI fragment, intact) was 4.5 kb, and the lower (DNase I cleaved) band migrated at 2.1 kb. (B) Determination of the positions of the HS site and CpG islands that were predicted by a computer algorithm. In the 4.5-kb EcoRI-EcoRI fragment, as shown in supplemental Figure 2, the HS (2.1 kb downstream of the 5′ EcoRI site) site was localized at the 3′ end of exon 1 (∼ 350 bp). The MethPrimer algorithm predicted the presence of 2 CpG islands: CpG1 was localized 2.1-2.25 kb downstream of the 5′ EcoRI site, and the CpG2 2.8-2.9 kb downstream of the 5′ EcoRI site. (C) ChIP revealed a potential hypermethylation of this gene region in resting normal CD8 T cells. A small aliquot of the sonicated genomic DNA (0.25 ng) from before and after the enrichment of the MBP2-bound DNA fractions was used. PCR was performed using a primer set that flanks the first CpG island. The reaction was terminated after 28 cycles of amplification in the semilog amplification phase for a quantitative comparison. Sources of DNA were: resting CD8 T cells (i), CD8 T cells from a leukemic IL-15Tg mouse (ii), and cells from the K2 leukemic cell line (iii). The black bar indicates that distant lanes from the same experiment were combined. Densitometry analysis was performed and the amount of MBP-2 bound DNA normalized to that of control DNA is indicated. (D) In vitro activation of normal CD8 T cells temporally induced IL-15Rα, and treatment with 5-Aza enabled sustained IL-15Rα expression. CD8 T cells from WT mice were incubated with plate-coated anti-CD3/anti-CD28 Abs in the presence or absence of 5-Aza (6μM) for 48 hours, and then stained with anti–IL-15Rα Ab (BAF551; R&D Systems) and FITC-streptavidin secondary Ab (eBioscience). Fractions of the cells were transferred to fresh medium with IL-2 (5nM) and cultured for an additional 4 days. At the same time, total RNA was extracted from the cells and subjected to Northern blot analysis using a 32P-labeled anti-mouse IL-15Rα fragment as the probe and to quantitative RT-PCR (supplemental Figure 4) to examine the expression of mouse IL-15Rα transcripts. Top panel shows IL-15Rα staining of activated (i) and rested (ii) CD8 T cells with and without 5-Aza. The data represent 4 independent experiments (see supplemental Table 1 for a summary of all data). Bottom panel shows Northern blotting. As a loading control, the 18S ribosomal RNA was visualized with ethidium bromide staining. Origins of RNA were the B1 (i) and K2 (ii) leukemia cell lines, ex vivo cultured DCs without (iii) and with (iv) activation by IFN-γ and LPS, CD8 T cells from WT mice cultured in the medium for 6 hours (v) or treated with 5-Aza for 6 hours (vi), TCR activated without (vii) and with (viii) 5-Aza for 48 hours, followed by IL-2 culture for 4 days without (viiii) and with (x) 5-Aza. (E) Antigenic challenge of WT mice only transiently induced IL-15Rα on specific CD8 T cells. WT mice were challenged once with OVA peptide by injecting peptide-loaded DCs subcutaneously to both inguinal regions (1.7 × 105 cells each) and IP (6.7 × 105 cells). Seven days later, IL-15Rα expression levels on the antigen-specific (tetramer-positive) activated CD8 T cells and nonactivated (tetramer-negative) CD8 T cells were analyzed by flow cytometry. The IL-15Rα expression levels were assessed on days 7 and 13. The data represent 2 independent experiments (n = 3 each; see supplemental Table 2 for a summary of all data).

An epigenetic mechanism controls the transcription of the il15rα gene in normal and leukemic CD8 T cells into negative and positive directions. (A) Presence of a DNase I HS site in IL-15Tg leukemic cells and its position in the transcription control site of the il15rα gene. Intact nuclei were isolated from purified CD8 T cells from terminally leukemic IL-15Tg mice (n = 2) and from K2 and B1 leukemic cells, and subjected to 1 U/mL of DNase I for 15 minutes at 37°C. The genomic DNA was extracted subsequently and digested by the EcoRI enzyme. The digested DNA was subjected to 0.7% Tris-acetate EDTA gel electrophoresis, blotted, and hybridized with a 32P-labeled fragment (as shown by the asterisk in panel B) from the il15rα transcription initiation site. As a control, CD8 T cells purified from WT mice were activated by plate-coated anti-CD3/anti-CD28 Abs and hIL-15 (5nM; PeproTech) for 48 hours before DNase I treatment. The sources of genomic DNAs were: normal CD8 T cells (i), CD8 T cells from leukemic IL-15Tg mouse #1 (ii), CD8 T cells from leukemic IL-15Tg mouse #2 (iii), cells from the K2 leukemic cell line (iv), and cells from the B1 leukemic cell line (v). The size of the upper band (EcoRI-EcoRI fragment, intact) was 4.5 kb, and the lower (DNase I cleaved) band migrated at 2.1 kb. (B) Determination of the positions of the HS site and CpG islands that were predicted by a computer algorithm. In the 4.5-kb EcoRI-EcoRI fragment, as shown in supplemental Figure 2, the HS (2.1 kb downstream of the 5′ EcoRI site) site was localized at the 3′ end of exon 1 (∼ 350 bp). The MethPrimer algorithm predicted the presence of 2 CpG islands: CpG1 was localized 2.1-2.25 kb downstream of the 5′ EcoRI site, and the CpG2 2.8-2.9 kb downstream of the 5′ EcoRI site. (C) ChIP revealed a potential hypermethylation of this gene region in resting normal CD8 T cells. A small aliquot of the sonicated genomic DNA (0.25 ng) from before and after the enrichment of the MBP2-bound DNA fractions was used. PCR was performed using a primer set that flanks the first CpG island. The reaction was terminated after 28 cycles of amplification in the semilog amplification phase for a quantitative comparison. Sources of DNA were: resting CD8 T cells (i), CD8 T cells from a leukemic IL-15Tg mouse (ii), and cells from the K2 leukemic cell line (iii). The black bar indicates that distant lanes from the same experiment were combined. Densitometry analysis was performed and the amount of MBP-2 bound DNA normalized to that of control DNA is indicated. (D) In vitro activation of normal CD8 T cells temporally induced IL-15Rα, and treatment with 5-Aza enabled sustained IL-15Rα expression. CD8 T cells from WT mice were incubated with plate-coated anti-CD3/anti-CD28 Abs in the presence or absence of 5-Aza (6μM) for 48 hours, and then stained with anti–IL-15Rα Ab (BAF551; R&D Systems) and FITC-streptavidin secondary Ab (eBioscience). Fractions of the cells were transferred to fresh medium with IL-2 (5nM) and cultured for an additional 4 days. At the same time, total RNA was extracted from the cells and subjected to Northern blot analysis using a 32P-labeled anti-mouse IL-15Rα fragment as the probe and to quantitative RT-PCR (supplemental Figure 4) to examine the expression of mouse IL-15Rα transcripts. Top panel shows IL-15Rα staining of activated (i) and rested (ii) CD8 T cells with and without 5-Aza. The data represent 4 independent experiments (see supplemental Table 1 for a summary of all data). Bottom panel shows Northern blotting. As a loading control, the 18S ribosomal RNA was visualized with ethidium bromide staining. Origins of RNA were the B1 (i) and K2 (ii) leukemia cell lines, ex vivo cultured DCs without (iii) and with (iv) activation by IFN-γ and LPS, CD8 T cells from WT mice cultured in the medium for 6 hours (v) or treated with 5-Aza for 6 hours (vi), TCR activated without (vii) and with (viii) 5-Aza for 48 hours, followed by IL-2 culture for 4 days without (viiii) and with (x) 5-Aza. (E) Antigenic challenge of WT mice only transiently induced IL-15Rα on specific CD8 T cells. WT mice were challenged once with OVA peptide by injecting peptide-loaded DCs subcutaneously to both inguinal regions (1.7 × 105 cells each) and IP (6.7 × 105 cells). Seven days later, IL-15Rα expression levels on the antigen-specific (tetramer-positive) activated CD8 T cells and nonactivated (tetramer-negative) CD8 T cells were analyzed by flow cytometry. The IL-15Rα expression levels were assessed on days 7 and 13. The data represent 2 independent experiments (n = 3 each; see supplemental Table 2 for a summary of all data).

Close Modal
This Feature Is Available To Subscribers Only

or Create an Account

Close Modal
Close Modal