Abstract

ZAP-70 expression is a strong prognostic marker in chronic lymphocytic leukemia (CLL). However, standardization of ZAP-70 using flow cytometry is problematic. In previous study, ZAP-70 methylation status, assessed by COBRA methodology, was found to be nicely correlated with ZAP-70 expression in CLL (

Corcoran et al.
Haematologica
2005
;
90
:
1078
–1088
). However, this technique is poorly suitable for routine use. We developed a new quantitative method for ZAP-70 methylation that simplifies this approach. Pyrosequencing is a sequencing-by-synthesis method that uses the release of pyrophosphate (Ppi) molecule during incorporation of nucleotides. Released Ppi are quantitatively converted into a light flash by enzymatic cascade. These light signals are measured by a charge-coupled device camera and are displayed as peaks in a pyrogram. Pyrosequencing combines the ability of direct quantitative sequencing, reproducibility, speed, ease-of-use. It has many advantages compared to other techniques for studies of methylation as bisulfite sequencing, COBRA, methylspecific PCR (MSP) and quantitative-MSP (Q-MSP). Samples were obtained from blood for 115 CLL patients and from paraffin embedded tissue after lymph node biopsy for 8 patients with 2 CLL, 2 MCL and 4 DLBCL. All patients were untreated at the time of sampling. Binet stage at diagnosis for the 117 CLL were stage A (114 pts), stage B (2pts), stage C (1pt). DNA from peripheral blood cells and from paraffin-tissue samples was treated by bisulfite and the ZAP-70 intron1-exon2 region was amplified using ZAPDN1-F and ZAPDN1-R primers. The percentage methylation was calculated as the average CpG methylation per clone and of all the clones sequenced for each sample. We found that methylation in four CpG pairs (C-223, C-243, C-254, C267) in the first intron of ZAP-70 highly correlated with repression of ZAP-70, as determined by western blot and real time PCR. Calibration mixture of DNAs with known methylation were analyzed. As expected, results showed that T-lymphocytes DNA is umethylated with a mean of 4% [range 2–12%, mean (SD) 0.41%] whereas M.SssI treated DNA was nearly fully methylated in the 4 CpG sites with a mean of 89% [range: 73 to 100%, mean (SD): 1,54%]. For each patient, biotinylated single-strand DNA fragments were generated, amplified by PCR and sequenced using an automated pyrosequencing instrument, PSQTM 96 MA (Biotage). A 50% methylation level cut-off easily discriminates ZAP-70 positive from negative CLL cells. There was an excellent correlation between methylation quantification results observed with bisulfite sequencing and pyrosequencing for the 4 CpG sites (r2>0,95). DNA extracted from tissue paraffin embedded tissue (2 CLL, 2 MCL, 4 DLBCL) show the same quality of pyrograms. Sensibility of pyrosequencing is about 5%. Additionally, we found an excellent correlation of ZAP-70 hypermethylation status with IgVH mutational status (p<0.0001) and CD38 expression (p<0.0001). Patients with unmethylated ZAP-70 had shorter median overall survival (80 months/not reached, p=0.0014) and time to treatment (24 months/not reached, p<0.0001). The pyrosequencing assay has several advantages: it is straight-forward to perform, provides high reproducibility, is independent of T-lymphocytes contamination, needs small quantity of DNA, is usable from tissue paraffin or frozen cells and do not express the uncertainties associated with the use of ZAP-70 protein expression. In addition, pyrosequencing delivers results faster and at lower cost than IgVH sequencing and is more suitable for routine use.

Disclosures: Leporrier:AMGEN: Consultancy, Honoraria, Research Funding; BMS: Consultancy, Honoraria, Research Funding; GENZYME: Consultancy, Honoraria; CHUGAI: Research Funding; ROCHE: Research Funding; SCHERING: Research Funding; NOVARTIS: Research Funding; JANSEN-CILAG: Research Funding.

Author notes

Corresponding author