DNA methylation inhibition increases T cell KIR expression through effects on both promoter methylation and transcription factors - PubMed (original) (raw)

DNA methylation inhibition increases T cell KIR expression through effects on both promoter methylation and transcription factors

Ying Liu et al. Clin Immunol. 2009 Feb.

Abstract

Killer-cell immunoglobulin-like receptor (KIR) genes are a polymorphic family expressed on NK cells, and "senescent" CD28- T cells implicated in cardiovascular disease. KIR promoters are highly homologous, and NK expression is regulated by DNA methylation. T cell KIR regulation is poorly understood. We asked if epigenetic mechanisms and/or transcription factor alterations determine T cell KIR expression. DNA methylation inhibition activated multiple KIR genes in normal T cells. KIR2DL2 and KIR2DL4 were selected for further study. Expression of both was associated with promoter demethylation, and methylation of the promoters in reporter constructs suppressed expression. KIR reporter construct expression also increased in demethylated T cells and required Ets1, Sp1 and AML sites, implying effects on transcription factors. This was confirmed for Sp1. These results indicate that KIR genes are suppressed by DNA methylation in most T cells, and DNA demethylation promotes their expression through effects on both chromatin structure and transcription factors.

PubMed Disclaimer

Figures

Figure 1

Figure 1. Effect of 5-azaC on T cell KIR2DL2expression

T cells from 5 healthy controls were stimulated with PHA for 18 hours, cultured with or without 5-azaC, and 72 hours later stained with anti-CD4-CyC, anti-CD8-FITC and anti-KIR2DL2-PE and analyzed using flow cytometry. A. Representative histograms of untreated (right column, Control) and 5 aza-C treated (left column, 5-azaC) T cells stained with anti-KIR2DL2-PE and anti-CD4-CyC (top row) or anti-CD8-FITC (bottom row). The number in the upper right hand corner represents the percent cells in that quadrant. B. Mean±SD of 5 independent experiments comparing %KIR2DL2+CD4+ or %KIR2DL2+CD8+ untreated (white bars) or 5-azaC treated (gray bars) cells. C. RNA was isolated from the CD4+ and CD8+ T cells shown in panel A and KIR2DL2 transcripts were measured relative to β-actin by real time RT-PCR. Results again represent the mean±SD of 5 determinations. (* = p<0.001, untreated vs treated).

Figure 2

Figure 2. Effect of 5-azaC on T cell KIR2DL4expression

T cells from 5 healthy controls were stimulated with PHA for 18 hours, cultured with or without 5-azaC, and 72 hours later stained with anti-CD4-CyC, anti-CD8-FITC and anti-KIR2DL4-PE then analyzed using flow cytometry as in Fig. 1. A. Representative histograms of untreated (right column, Control) and 5 aza-C treated (left column, 5-azaC) T cells stained with anti-KIR2DL4-PE and anti-CD4-CyC (top row) or anti-CD8-FITC (bottom row). The number in the upper right hand corner again represents the percent cells in that quadrant. B. Mean±SD of 5 independent experiments comparing %KIR2DL4+CD4+ or %KIR2DL4+CD8+ untreated (white bars) or 5-azaC treated (gray bars) cells. C. RNA was isolated from the CD4+ and CD8+ T cells shown in panel A and KIR2DL4 transcripts were measured relative to β-actin by real time RT-PCR. Results again represent the mean±SD of 5 determinations. (* = p<0.001, untreated vs treated).

Figure 3

Figure 3. Effect of 5-azaC on KIR2DL2 and KIR2DL4 promoter methylation

A. PBMC from 3 healthy donors were stimulated with PHA and treated with 5-azaC as in Fig 1, then CD4+ (left panel) and CD8+ (right panel) T cells were isolated and bisulfite sequencing of the indicated regions in untreated (upper panels) or 5-azaC treated cells (lower panels) performed as in Materials and Methods, cloning and sequencing 10 fragments/subject. Closed circles represent methylated dC, and open circles unmethylated dC. A map of the KIR2DL2 promoter showing the locations of CG pairs (balloons), the AML, Ets and Sp1 binding sites (arrows) and start site (bent arrow) is shown for reference. B. Bisulfite sequencing of the KIR2DL4 promoter was performed on DNA from the same subjects. The results are presented as in panel A. C. The average methylation of all CG pairs in the 10 cloned fragments from untreated and 5-azaC treated CD4+ and CD8+ cells from each donor was determined for the KIR2DL2 and KIR2DL4 promoters. Results represent the mean±SD of the percent methylation in the DNA from the 3 donors. The dark bars represent untreated T cells, and the light bars the 5-azaC treated T cells (p<0.01 for all, methylated vs unmethylated). D. PBMC from 5 additional healthy controls were stimulated with PHA, treated with 5-azaC, fractionated into CD4+ and CD8+ subsets, and KIR2DL2 and KIR2DL4 promoter sequences amplified with primers hybridizing with methylated or unmethylated sequences. Primers hybridizing with regions lacking CG pairs served as a reference. Results are presented as the methylation index [(methylated/control)/(methylated/control+unmethylated/control)] X 100 and represent the mean+SD of the 5 determinations for untreated (dark bars) or 5-azac treated (light bars) CD4+ or CD8+ cells from the 5 donors (p<0.001 for each.)

Figure 4

Figure 4. KIR promoter methylation suppresses function

The KIR2DL2 promoter (−271 to +111) and the KIR2DL4 promoter (−289 to +38) were methylated (light bars) or mock methylated (dark bars) in vitro, cloned into pGL3, then transfected into Jurkat cells and luciferase expression measured relative to a β-galactosidase control. Results represent the mean+SD of 4 independent experiments (p=0.005 for KIR2DL2 and KIR2DL4).

Figure 5

Figure 5. 5-azaC increases trans-acting factors driving KIR expression

A. The KIR2DL2 (upper row) and KIR2DL4 (lower row) promoter fragments described in Fig 4 were cloned into pmaxFP-Yellow-PRL then transfected into PHA stimulated, untreated (left column) or 5-azaC treated (right column) T cells along with pmaxGFP, then fluorescence measured using flow cytometry (open histograms) gating on the lymphocyte population. Controls include transfection with the promoterless vector (closed histograms). The percent KIR+ cells is shown in the upper right corner of each panel. B. The KIR-pmaxFP-Yellow-PRL constructs described in panel A were similarly transfected into untreated (light bars) or 5-azaC treated (dark bars) T cells and fluorescence measured relative to control pmaxGFP transfections, also as in panel A. Results represent the mean+SD of 3 independent experiments (* p<0.005). C. The KIR2DL2 and KIR2DL4 promoter constructs were then transfected into untreated T cells (black bars) or 5-azaC treated T cells (dark gray bars) as described in panels A and B. The 5-azaC treated cells were also transfected with KIR2DL2 and KIR2DL4 constructs into which mutations were induced into the promoters at the Ets1 sites (white bars), the Sp1 sites (crosshatched bars), both the Ets and Sp1 sites (light gray bars), or the AML sites (light stippled bars). Fluorescence was determined relative to pmaxGFP as before. Results represent the mean+SD of 3 independent experiments (** p < 0.001 for Ets + Sp1 combined mutations vs the intact promoter, and * p ≤ 0.03 for AML mutation vs intact promoter, in 5-azaC treated cells). D. T cells were stimulated with PHA, treated with 5-azaC and 72 hrs later nuclear extracts were isolated and Sp1 measured in equivalent amounts of nuclear protein as described in Materials and Methods. Recombinant Sp1 served as a positive control, and a “cold probe” (unconjugated oligonucleotide containing an Sp1 binding site) served as a negative control. Results represent the mean±SD of 3 independent experiments (* p =0.0016).

Figure 6

Figure 6. Increased Sp1, Ets and AML binding to KIR promoters

PBMC from 3 healthy donors were stimulated and cultured without (white bars) or with (dark bars) 5-azaC as in Fig. 1, then T cells were purified, crosslinked, sonicated, chromatin was immunoprecipitated with mAb to the indicated transcription factors or control IgG, then precipitated DNA amplified by real-time PCR. The amount of precipitated DNA is expressed relative to total input 5-azaC treated DNA, and results are presented as the mean+SD of the 3 determinations (p < 0.03 untreated vs treated for all 3 factors in both promoters).

Figure 7

Figure 7. KIR2DL4 induced by 5-azaC is functional

PHA stimulated CD4+ T cells (open bars) or CD8+ T cells (shaded bars) were treated or not with 5-azaC, then cultured alone, with plate bound IgG (Sti.+IgG) or with plate bound anti-KIR2DL4 (Sti.+2DL4) as indicated. IFN-γ was measured by ELISA 20 hours later. Results are presented as the mean+SD. ** p≤ 0.001 untreated vs 5-azaC treated, unstimulated T cells, * p<0.01 5-azaC treated KIR stimulated vs 5-azaC treated IgG stimulated T cells.

References

    1. Feinberg AP. Phenotypic plasticity and the epigenetics of human disease. Nature. 2007;447:433–40. - PubMed
    1. Lu Q, Wu A, Tesmer L, Ray D, Yousif N, Richardson B. Demethylation of CD40LG on the inactive X in T cells from women with lupus. J Immunol. 2007;179:6352–8. - PubMed
    1. Bird AP. Gene number, noise reduction and biological complexity. Trends Genet. 1995;11:94–100. - PubMed
    1. Blom B, Spits H. Development of human lymphoid cells. Annu Rev Immunol. 2006;24:287–320. - PubMed
    1. Cantrell D. T cell antigen receptor signal transduction pathways. Annu Rev Immunol. 1996;14:259–74. - PubMed

Publication types

MeSH terms

Substances

Grants and funding

LinkOut - more resources