Abstract
ZAP70 expression has been shown to be involved in enhanced signalling and more aggressive disease in a subset of CLL. Mechanisms regulating ZAP70 expression are unknown. We have shown previously that despite the absence of a 5’ CpG island, the methylation status of a small region of CpG dinucleotides (CpGs) correlates with the transcriptional state of the gene in both normal lymphocytes and B cell leukemias. Quantitative methylation analysis of 605 CpGs across the 28kb genomic region spanning ZAP70 was performed by MassARRAY on a panel of 17 CLL tumor cell samples, 4 lymphoid cell lines and B cell, T cell and myeloid cell samples pooled from 3 normal individuals. All samples showed hypermethylation of the gene body and of the gene’s two 3’ CpG islands. However, there was variability between samples in the methylation of 12 consecutive CpGs within a 1kb predicted promoter region (PPR), spanning the transcription start site (TSS) and in the methylation of 24 consecutive CpGs in an adjacent 1kb differentially methylated region (DMR), downstream of the TSS. The methylation of the PPR and DMR, together with the expression status of the samples, suggested four different states for the gene (Table 1).
Table 1 - ZAP70 gene states defined by ZAP70 expression status and methylation of the PPR and DMR.
. | . | MEAN CpG METHYLATION (%) . | . | |
---|---|---|---|---|
SAMPLE . | ZAP70 EXPRESSION STATUS . | PPR . | DMR . | GENE STATE . |
NAMALWA | − | 65 | 82 | I |
B CELLS | − | 48 | 82 | I |
MYELOID | − | 53 | 80 | I |
CLL6 | − | 4 | 86 | II |
CLL7 | − | 5 | 75 | II |
CLL8 | − | 12 | 78 | II |
CLL10 | − | 12 | 62 | II |
CLL11 | − | 4 | 62 | II |
CLL12 | − | 21 | 77 | II |
HBL2 | − | 18 | 60 | II |
CLL13 | + | 4 | 40 | III |
CLL14 | + | 5 | 46 | III |
CLL15 | + | 7 | 45 | III |
CLL16 | + | 9 | 43 | III |
CLL17 | + | 5 | 56 | III |
NALM6 | + | 8 | 52 | III |
CLL1 | + | 3 | 4 | IV |
CLL2 | + | 3 | 4 | IV |
CLL9 | + | 4 | 6 | IV |
CLL4 | + | 3 | 8 | IV |
CLL5 | + | 4 | 16 | IV |
CLL3 | + | 4 | 17 | IV |
JURKAT | + | 3 | 4 | IV |
T CELLS | + | 9 | 13 | IV |
. | . | MEAN CpG METHYLATION (%) . | . | |
---|---|---|---|---|
SAMPLE . | ZAP70 EXPRESSION STATUS . | PPR . | DMR . | GENE STATE . |
NAMALWA | − | 65 | 82 | I |
B CELLS | − | 48 | 82 | I |
MYELOID | − | 53 | 80 | I |
CLL6 | − | 4 | 86 | II |
CLL7 | − | 5 | 75 | II |
CLL8 | − | 12 | 78 | II |
CLL10 | − | 12 | 62 | II |
CLL11 | − | 4 | 62 | II |
CLL12 | − | 21 | 77 | II |
HBL2 | − | 18 | 60 | II |
CLL13 | + | 4 | 40 | III |
CLL14 | + | 5 | 46 | III |
CLL15 | + | 7 | 45 | III |
CLL16 | + | 9 | 43 | III |
CLL17 | + | 5 | 56 | III |
NALM6 | + | 8 | 52 | III |
CLL1 | + | 3 | 4 | IV |
CLL2 | + | 3 | 4 | IV |
CLL9 | + | 4 | 6 | IV |
CLL4 | + | 3 | 8 | IV |
CLL5 | + | 4 | 16 | IV |
CLL3 | + | 4 | 17 | IV |
JURKAT | + | 3 | 4 | IV |
T CELLS | + | 9 | 13 | IV |
Bisulphite cloning and sequencing of a PCR amplicon spanning an exon1 C/A SNP (rs2276645) and the PPR/DMR junction was performed together with cDNA pyrosequencing of rs2276645 on the five CLL tumor samples identified with gene state III. All samples showed allele specific methylation (ASM) of the A allele within the DMR and almost complete restriction of ZAP70 expression to the hypomethylated C allele. Bisulphite pyrosequencing of two DMR CpGs in purified leukocyte populations from these cases showed that ASM appears restricted to CLL cells, with hypermethylation and hypomethylation of the myeloid and T cells respectively (Table2). This suggests that while methylation of the DMR is sufficient for allele restriction, ASM does not result from imprinting and may be restricted to CLL tumor cells.
Table 2 – Mean methylation of 2 DMR CpGs in leukocyte populations from CLL patients with known ASM of the DMR in their tumor cells.
. | MYELOID CELLS . | CLL CELLS . | T CELLS . | |||
---|---|---|---|---|---|---|
PATIENT . | CD15 (%) . | METHYLATION(%) . | CD19 (%) . | METHYLATION (%) . | CD2 (%) . | METHYLATION (%) . |
CLL13 | 83 | 90 | 98 | 59 | 71 | 21 |
CLL14 | 98 | 99 | 98 | 50 | 88 | 10 |
CLL15 | 88 | 84 | 93 | 45 | 85 | 24 |
CLL16 | 99 | 96 | 99 | 52 | 82 | 18 |
CLL17 | 92 | 89 | 98 | 49 | 90 | 22 |
. | MYELOID CELLS . | CLL CELLS . | T CELLS . | |||
---|---|---|---|---|---|---|
PATIENT . | CD15 (%) . | METHYLATION(%) . | CD19 (%) . | METHYLATION (%) . | CD2 (%) . | METHYLATION (%) . |
CLL13 | 83 | 90 | 98 | 59 | 71 | 21 |
CLL14 | 98 | 99 | 98 | 50 | 88 | 10 |
CLL15 | 88 | 84 | 93 | 45 | 85 | 24 |
CLL16 | 99 | 96 | 99 | 52 | 82 | 18 |
CLL17 | 92 | 89 | 98 | 49 | 90 | 22 |
Native chromatin immunoprecipitation (N-ChIP) using anti-AcH3, H3K14Ac and H3K14Me2 antibodies was performed on the 4 cell lines and tumor cells from CLLs 1, 2, 6, 7, 13 and 14 from the MassARRAY series. PCR for amplicons across the PPR and DMR showed the presence of all 3 histone modifications in ZAP70 expressing JURKAT and NALM6 cells but these modifications were absent in the ZAP70 negative NAMALWA and HBL2 cells. In contrast, all 6 CLL samples showed enrichment for all 3 modifications, regardless of gene state, suggesting an open, active/permissive chromatin structure, despite clear differences in methylation of the DMR.
Further bisulphite pyrosequencing and N-ChIP of NAMALWA and HBL2 cells cultured for 6 days in the presence of 0.5μM Decitibine showed concomitant DMR demethyaltion, increased AcH3 within the DMR and up regulation of ZAP70 expression, all of which were reversed when the drug stimulus was removed.
Taken together this data suggests that ZAP70 is regulated by epigenetic mechanisms, with the methylation status of a small DMR playing a key role, sufficient to differentiate the transcriptional activity of two alleles within a single cell. It is apparent that the gene is primed for expression in all CLL cells and that methylation of the DMR is part of the key switching process between active transcription and silencing. The differences in DMR methylation between an individual’s expressing T cells and CLL cells, suggests that differences may exist in the mechanism of regulation between T and B cells and raises the possibility that such differences could be exploited as targets for therapy.
Disclosures: No relevant conflicts of interest to declare
Author notes
Corresponding author
This feature is available to Subscribers Only
Sign In or Create an Account Close Modal