Expression Of WT1 Gene And Its Isoforms In Leukemia Cells And Clinical Application

Objective: To investigate expression levels of Wilms'tumor gene (WT1) in bone marrow (BM) cells of acute leukemia patients (Als) with different subtype and chronic myelogenous leukemia (CML) patients at different phase, as well as the relationship between the WT1 gene expression level and clinical prognosis, emphasizing on the significance of WT1 gene expression level as a biomarker during the follow-up after allogeneic bone marrow transplantation (allo-BMT). In the meantime, the expression levels and the ratio of WT1(+17AA)/(-17AA) isoforms in both variety of leukemia cell lines and primary bone marrow cells of acute leukemia patients were also determined. Besides, the relationship between WT1 isoforms expression and induced differentiation of NB4 leukemia cells was explored. Finally the targeting immunotherapy for leukemia cells in vitro by using cytotoxic T lymphocyte (CTL) specific against WT1-derived peptide was also performed. Methods: (1) Real-time quantitative reverse transcription polymerase chain reaction (RQ-RT-PCR) method was established for detecting WT1 and GAPDH expression levels in BM cells of 15 cases of CML at first diagnosis , 108 ALs with different subtype and a total of 111 specimens from 15 leukemia patients during their follow-up after allo-BMT as well as 23 non-leukemia patients by using LightCycler. Normalized WT1 expression level (WT1_N ) was determined as a ratio of WT1 copies over GAPDH copies times 10~4. The clinical data of patients were well documented. WT1 , WT1(+17AA) isoforms and GAPDH expression levels in NB4 leukemia cells induced by retinoic acid (ATRA) were also detected by LightCycler and the value of WT1(+17AA) isoforms was normalized as WT1_N (+17AA). The expression of CD11b in these cells was simultaneously detected with a Becton Dickinson FACSCalibur flow cytometry. The expression level of both WT1 gene and WT1(+17AA) isoforms was determined by RQ-RT-PCR in leukemia cell lines including K562, SHI, Jurkat, NB4, NB4/WT1A and NB4/WT1D as well as 79 initially diagnosed acute leukemias on a MJ Opticon TM 2, taking ABL gene as internal quantitative control. The ratios of WT1(+17AA) isoforms expression value over total WT1 gene expression were then calculated. (2) All primers and the TaqMan probes for WT1 gene, WT1(+17AA) isoforms and ABL gene were designed by Primer Priemer software (version 5.0) . (3) One 9-mer WT1 peptide (CMTWNQMNL) that contains HLA-A*0201-binding anchor motifs was synthesized. The dendritic cells (DCs) generated from peripheral blood nononuclear cells from an HLA-A*0201-positive healthy donor were repeatedly loaded with WT1 peptide so as to elicit CTLs specific for WT1 peptide, restricted by HLA-A*0201. The specific lysis effect of WT1 peptide specific CTLs upon leukemia cell lines and leukemia cells were measured by MTT assay. (4) Statistical analysis were performed with SPSS software (version 10.0). Nonparameter analysis (Mann-Whitey U test or Kruskal Wallis Test) and Oneway ANOVA were used to compare the means among the different groups. Both Spearman and Pearson Coefficient correlations were used to analyze the relationship between two parameters. P-values <0.05 were considered to be significant. K562 cell line with high WT1 expression was regarded as a positive control. Results: (1) The absolute WT1 expression levels in 23 non-leukemia patients were ranged from 0 to 102 copies per 100 ng RNA and the corresponding GAPDH expression levels were ranged from 106 to 107 copies per 100 ng RNA , thus normalized WT1 (WT1N) expression levels were ranged from 0 to 8.56 and the Median WT1N expression levels were 1.47. Absolute WT1 expression levels in K562 cell line was 2.5×105 copies per 100 ng RNA and the corresponding WT1N expression level were more than 100. (2) The absolute WT1 expression levels in 15 CML patients in chronic phase were ranged from 10 to 103 copies per 100 ng RNA and the WT1N expression level were ranged from 1.16 to 39.38 with a Median WT1N expression level of 6.83;The Median WT1N expression level of 63 newly diagnosed AML and 7 newly diagnosed ALL patients was 70.61(ranged from 3.40 to 527.51) and 119.04 (ranged from 10.73 to 283.6) respectievely. Nonparameter statistic analysis revealed that the WT1N expression level in CML was significantlylower than that in AML and ALL patients, while slightly higher than that in control groups (p=0.046). (3) The Median expression levels of WT1N in 70 newly diagnosed ALs and 11 relapsed Als were statistically higher than those of 23 ALs in complete remission(CR) and 23 non-leukemic controls (75.10 and 89.56 vs 2.07 and 1.47 ). No statistic differences were found between the CR group and control group, nor between the newly diagnosed group and relapsed group. Of the 70 newly diagnosed Als ,Median WT1N expression level of acute granulocytic leukemias was significantly higher than that of acute monocytic leukemias (M5),but there were no statistic differences among the M1 , M2 , M3 and ALL subtypes. Furthermore the WT1N levels were not correlated to peripheral WBC counts, blast percentage in BM and the multidrug resistant gene(MDR1) expression at presentation, while were correlated to chromosome karyotypes. Dynamics of WT1N levels in 2 patients during treatment showed that WT1 expression levels may predict relapse . (4) In general, the dynamic curves of WT1N levels following allo-BMT were consistent with the trends in expression levels of corresponding fusion genes for minimal residual disease (MRD) monitoring. Spearman Rho correlation analysis revealed that the correlation coefficient between the WT1N expression levels and expression of BCR/ABL, AML/ETO, PML/ RARαand MLL/AF17 fusion genes were 0.678 (P=0.000), 0.677 (P=0.000), 0.806 (P=0.000) and 0.553 (P=0.049) respectively. Three patients replasped after allo-BMT and one patient replased before allo-BMT during the follow-up. A re-increment of WT1N expression level during follow-up could be detected 40 to 180 days earlier to hematological relapse. (5) The expression of both WT1 and WT1(+17AA) rapidly decreased during the dfferentiation of NB4 cells induced by ATRA. The WT1N level was 191.11, 121.17, 66.72, 43.47 ,18.29 ,4.04 and 3.79 respectively, prior to and at 6, 12, 18 ,24,48 and 72 hours after exposure to ATRA; and they were in adversely accordance with the dynamic changes of CD11b positive rates(γ=-0.65, p<0.01); The WT1 N (+17AA) level was 105.12,46.89,20.50,10.38,8.85,2.16 and 1.92,respectively, at the above time points, these values also in adversely accordance with the CD11b changes (γ=-0.77,p<0.01); Furthermore, the ratio of WT1 N(+17AA) v.s WT1N(WT1 N (+17AA) / WT1N) was significantly decreased during the first 18 hours after exposure to ATRA and re-increased 24 hours later , indicating a remarkable decrease of WT1(+17AA) level comparing with WT1 during the early stage of NB4 cell differentiation. (6) The expression of WT1 gene (WT1/ABL) in K562, SHI, NB4, BN4/WT1A and NB4/WT1D cell lines were 2 to 3 logarithms higher than that in Jurkat and U937 cell lines. The ratios of WT1(+17AA)/WT1 in K562, SHI, Jurkat, U937, NB4, NB4/WT1A and NB4/WT1D were 0.51, 0.41, 0.25, 0,07, 0.53,0.25 and 0.62 respectively, suggesting a dominant expression of WT1(+17AA) isoforms in K562 and NB4 cells. The median ratio of WT1(+17AA)/WT1 in 79 newly diagnosed acute leukemia patients was 0.64, ranging from 0.30 to 0.93, and there were no statistical differences in the ratios of WT1(+17AA)/WT1 among different subtypes of acute leukemias. (7) The CTLs specific for WT1 peptide and restricted by HLA-A*0201 can exert a specific lysis upon WT1-expressing, HLA-A2.1-positive leukemia cell line (NB4) and primary leukemia cells. No such effect of these CTLs could be found upon WT1+, HLA-A2.1-leukemia cell line K562, the WT1-, HLA-A2+ leukemia cell lines (U937) and the primary leukemia cells with corresponding immunotype. There were also no effects upon the WT1-normal blood cells. (8) The specific lysis effect of WT1-CTLs upon NB4, NB4/WT1A,and NB4/WT1D cell lines were of no statistical significant differences, suggesting that the specific lysis of WT1-CTLs against leukemias was of no relevance to the ratios of WT1(+17AA)/WT1 in leukemia cells. Conclusions: (1) WT1 expression levels in patients with AL were remarkablely higher than that in non-leukemia patients who expressed a marginal level. WT1 can be a biomarker for detecting MRD and evaluating therapy efficacy in leukemias. (2) WT1 expression levels in leukemia patients following allo-BMT measured by real time RT-PCR can be a useful tool for monitoring MRD and warning the clinical relapse during follow-up. (3) These data suggest that the abnormally high expression of WT1 and WT1(+17AA) isoforms were associated with the block of cell differentiation. Of the four main isoforms WT1(+17AA) may dominantly contribute to blocking cell differention. (4) The WT1 gene is expressed in leukemia cell lines and primaryleukemia cells with a significantly higher level of the WT1(+17AA) isoforms, and there were no statistical differences in the ratios of WT1(+17AA)/WT1 among different subtypes of acute leukemias. (5) The CTLs which is specific for WT1 and restricted by HLA-A*0201 can exert specific lysis upon leukemia cell lines and primary leukemia cells, but not normal hematopoietic cells, providing a rationale for developing the strategy of WT1 peptide-based adoptive T-cell therapy and vaccination for human leukemia and solid tumors.