Search

Abstract: Objective: To investigate the feasibility of marking the human tumor cells with enhanced green fluorescent protein (EGFP) in vitro. Methods: The retroviral vector LGSN encoding EGFP was constructed and three human tumor cell lines were infected with LGSN amphotropic virus. Tumor cell lines that stably express EGFP were selected with G418. The integration and expression of EGFP gene were analyzed by polymerase chain reaction, and flow cytometry (FCM). Results: After gene transfection and ping-pong transduction, amphotropic producer line Am12/LGSN was generated with a stable green fluorescence signal readily detectable by FCM in up to 97% of examined cells. The viral titer in the supernatants was up to 8.2×105CFU/ml. After transduction and selection, G418-resistant leukemia K562, mammary carcinoma MCF-7, and bladder cancer 5637 cells were developed, in which the integration of both EGFP and neomycin resistance gene was confirmed by DNA amplification. In comparison with uninfected cells, FCM analysis revealed EGFP expression in up to 90% (range 85.5%–90.0%) of tumor cells containing LGSN provirus. Conclusion: The retroviral vector LGSN can effectively mark the human tumor cells with a stably EGFP expression which may be in studying tumor growth, metastasis and angiogenesis.

Objective: To investigate retroviral-mediated transfer and expression of human multidrug resistance (MDR) gene MDR1 in leukemic cells. Methods: Human myeloid cells, K562 and NB4, were infected by MDR retrovirus from the producer PA317/HaMDR, and the resistant cells were selected with cytotoxic drug. The transfer and expression of MDR1 gene was analyzed by using polymerase chain reaction (PCR), flow cytometry (FCM) and semisolid colonies cultivation. Results: The resistant cells, K562/MDR and NB4/MDR, in which integration of the exogenous MDR1 gene was confirmed by PCR analysis, displayed a typical MDR phenotype. The expression of MDR1 transgene was detected on truncated as well as full-length transcripts. Moreover, the resistant cells were P-glycoprotein postiive at 78.0% to 98.7% analyzed with FCM. The transduction efficieny in K562 cells was studied on suspension cultures and single-cell colonies. The transduction was more efficient in coculture system (67.9%-72.5%) than in supernatant system (33.1%~46.8%), while growth factors may improve the efficiency. Conclusion: Retrovirus could allow a functional transfer and expression of MDR1 gene in human leukemia cells, and MDR1 might act as a dominant selectable gene for coexpression with the genes of interest in gene therapy.

Objective: To explore whether human umbilical cord blood hematopoietic progenitor cells transduced with human O6-methylguanine-DNA-methyltransferase (MGMT) gene could increase resistance to 1,3-Bis(2-Chloroethyl)-1-Nitrosourea (BCNU). Methods: The cDNA encoding the MGMT was isolated by using RT-PCR method from total RNA of fresh human liver, the fragment was cloned into pGEM-T vector and further subcloned into G1Na retrovirus vector. Then the GlNa-MGMT was transduced into the packaging cell lines GP+E86 and PA317 by LipofectAMINE. By using the medium containing BCNU for cloning selection and ping-ponging supernatant infection between ecotropic producer clone and amphotropic producer clone, high titer amphotropic PA317 producer clone with the highest titer up to 5.8xlO5CFU/ml was obtained. Cord blood CD34+cells were transfected repeatedly with supernatant of retrovirus containing human MGMT-cDNA under stimulation of hemopoietic growth factors. Results: The retrovirus vector construction was verified by restriction endonuclease analysis and DNA sequencing. PCR, RT-PCR, Southern Blot, Western Blot and MTT analyses showed that MGMT drug resistance gene has been integrated into the genomic DNA of cord blood CD34+cells and expressed efficiently. The transgene cord blood CD34+cells conferred 4-folds stronger resistance to BCNU than untransduced cells. Conclusion: The retrovirus vector-mediated transfer of MGMT drug resistance gene into human cord blood CD34+cells and its expression provided an experimental foundation for gene therapy in clinical trial.

Objective: To establish an efficient and safe gene transfer system mediated by retrovirus for gene marking and gene therapy of human leukemia. Method: The retroviral vector LXSN, containing the neomycin resistance (NeoR) gene, was transferred into amphotropic packaging cells GP+envAm12 by liposome transfection or by ecotropic retrovirus transduction. Amphotropic retrovirus in supernatants with higher titer was used to infect human leukemic cell lines NB4, U937, and THP-1. The efficiency of gene transfer was assayed on colonies formed by transduced K562 cells. Results: The titer of DOSPER directly transfected GP+envAm12 cells determined on NIH3T3 cells was 8.0×105CFU/ml, while that of producer infected with retrovirus was 1.6×107CFU/ml. Integration of NeoRgene into all leukemia cells was confirmed by polymerase chain reaction (PCR). Absence of replication-competent virus was proved by both nested PCR for envgene and marker gene rescue assay. Gene transfer with the efficiency as high as 93.3 to 100% in K562 cells was verified by seminested PCR for integrated NeoRgene on colonies after 7 days’ culture. Conclusion: The efficiency and safety of retrovirus mediated gene transfer system might provide an optimal system in gene therapy for leukemia or genetic diseases.

Fifty patients with acute promyelocytic leukemia (APL) have been treated with all-trans retinoic acid (RA). In vitro induced differentiation of primarily cultured bone marrow cells from the patients, colony-forming unit granulocyte-macrophage (CFU-GM) and L-CFU colony-forming assays, and karyotype analysis were performed over the treatment course. The very high bone marrow complete remission (CR) rate (94%) suggested that all-trans RA was superior to conventional chemotherapeutic regimens for the treatment of APL. The leukemic clone was reduced by RA-induced terminal differentiation and loss of proliferation capacity of leukemic cells. Relapse after CR in about 40% of patients was the major reason for the failure of the RA treatment. Patients who relapsed after a chemotherapy-maintained CR could be effectively reinduced to second CR by RA. However, if relapse occurred after a CR maintained by both RA and chemotherapy, the sensitivity of newly emerged leukemic clones to RA was greatly reduced. Therefore, it is suggested that RA should be replaced by conventional chemotherapy as soon as CR is achieved. Laboratory studies proved valuable in selecting cases for RA therapy and in predicting therapeutic effects and prognosis. © 1991 by The American Society of Hematology.

All-trans retinoic acid (ATRA) has recently been recognized as the first line therapeutic agent in the treatment of patients with acute promyelocytic leukemia (APL). The extraordinary high remission rate achieved by ATRA in comparison with other chemotherapeutic agents suggested that ATRA differentiation induction therapy seemed superior to conventional chemotherapy for APL patients. However, after the great excitement aroused after the initial successes, we have to take stock and examine in detail several problems which have emerged preventing us from improving the clinical outcome in APL.Maintenance in order to prolong remission and prevention of or retreatment for the relapse are the major subjects of concern at present. Efforts should be made either to keep ATRA effective for APL patients or to resensitize the relapsing patients for repeated ATRA therapy. The administration of ATRA should be carefully adapted in accordance with the individual patient's condition. From both conceptual and practical points of view, ATRA differentiation therapy should be combined with chemotherapy, bone marrow transplantation and biomodifier treatment. Thus, a more comprehensive strategy must be planned and developed in the near future. Using molecular biological techniques, the diagnosis of APL can be more precisely made and the course of the disease more closely monitored. The central dogma, still to be revealed, is the relationship between APL pathogenesis, the chromosome translocation present with the relevant molecular alterations and the response to ATRA treatment. Current studies in all these above fields have provided us with a deeper understanding of the pathogenesis of APL and the physiological function and curative action of ATRA. This will no doubt enhance the development of the ATRA treatment regimen thereby widening the spectrum of leukemias and other neoplasias to be treated with differentiation therapy.

Constitutive classical NF-κB activation protects chronic lymphocytic leukemia (CLL) B-cells from spontaneous cell death and plays a critical role in the chemo-resistance. RelB, representing the alternative NF-κB activity, functions specifically in lymphoid organogenesis and B-cell maturation. RelB indeed plays a tumor-supportive role and confer radiation-resistance in solid tumors. However, the involvement of RelB in CLL is unrevealed. Here, we analyzed the NF-κB activation in CLL bone marrow (BM). Both the RelA and RelB activity could be detected in CLL B-cells from BM, in spite of inevitable variability. High RelB activity, concomitant with high RelA activity, was linked to an unfavorable prognosis of CLL. CLL-derived BM stromal cells (BMSCs) were functionally similar to disease-free BMSCs in their capability to support CLL B-cells survival. The migration and adhesion ability of CLL B-cells was not affected by the RelB activity. High RelB activity, together with the RelA activity, maintained the basal survival of cells. The induction of RelA and RelB expression in the nuclear was responsible for the better survival of CLL B-cells supported by BMSCs. In addition, the presence of high RelB activity in CLL B-cells was correlated with the sensitivity to proteasome inhibitor but not fludarabine. Taken together, we provided evidences that not only RelA but also RelB, subunits of NF-κB family, plays an important role in the cellular behavior of CLL cells from BM. The strength of RelB activity influenced the prognosis of CLL patients.

No relevant conflicts of interest to declare.

No relevant conflicts of interest to declare.

S100A8 and S100A9 are two members of the S100 calcium-binding protein family, preferentially form functional heterodimers of S100A8/S100A9, and have been increasingly recognized as biomarkers in malignancies.

No relevant conflicts of interest to declare.

No relevant conflicts of interest to declare.

No relevant conflicts of interest to declare.

No relevant conflicts of interest to declare.

No relevant conflicts of interest to declare.

No relevant conflicts of interest to declare.

Abstract 4961

Abstract 2694

Abstract 3617

Abstract 2907

Abstract 4325

Abstract 2150

No relevant conflicts of interest to declare.

No relevant conflicts of interest to declare.

No relevant conflicts of interest to declare.

No relevant conflicts of interest to declare.

Abstract 3119

Abstract 4503

Abstract 1756

Abstract 2779

CD30, a member of the TNF receptor (TNFR) superfamily, is a lymphocyte-specific receptor that is originally recognized as a surface molecule overexpressed on Hodgkin’s lymphoma cells. Engagement of CD30 with its ligand CD30L activates the nuclear factor kB (NF-kB), which are mediated by interactions with TNFR-associated factor (TRAFs), and supports proliferation of Hodgkin’s cells. However, the role of individual TRAFs in the CD30 signaling pathway in Hodgkin’s cells has not been fully addressed yet. In this study, we found that except TRAF3, all other TRAFs were expressed consistently in B cell-derived Hodgkin’s cell lines (L428 and KM-H2) either in mRNA or protein level. In L428 but not KM-H2 cells, the classical (p50-RelA) and alternative (p52-RelB) NF-kB activity were constitutively activated measured by Western Blotting and EMSA. To better understand the CD30 signaling properties in Hodgkin’s lymphoma cells, we silenced the expression of TRAFs individually by means of RNAi. Successful downregulation of TRAF1 protein expression led to the apoptosis of L428 cells, which was companied by the reduction of both classical and alternative NF-kB activity. Furthermore, the expression of targeting genes of NF-kB, such as A20, c-Flip, ICAM-1, and Cyclin D1 was also decreased. Using siRNA targeting TRAF2 expression, the classical NF-kB activity was reduced while the alternative NF-kB activity moderately induced especially upon CD30L treatment, which was companied by the induction of p100 processing and RelB nuclear localization. The survival rate was decreased when TRAF2 was knockdown. TRAF5 knockdown manifested similar results as that of TRAF2. These observations were only took place in L428 cells but not KM-H2 cells. In addition, the phosphorylation of the extracellular signal-regulated kinases (ERK) 1 and 2, and subsequent activation of JunB in Hodgkin’s cells were staying unchanged in individual TRAFs knockdown experiment. Taken together, the study here further investigates the features of CD30-TRAFs-NF-kB signaling pathway in Hodgkin’s lymphoma cells. At least TRAF1, 2, and 5 are involved in CD30 signaling pathway and protecting Hodgkin’s lymphoma cells from undergoing apoptosis. The interaction between TRAFs is necessary to facilitate the signaling and proper activation of classical NF-kB, which is responsible for the antiapoptotic effect of Hodgkin’s cells. Among TRAF1, 2, and 5, the redundant function is unlikely. Furthermore, the sequential activation of ERK and JunB is unlikely dependent on TRAFs-NF-kB pathway in Hodgkin’s lymphoma cells. Thus, these data reveal that the classical NF-kB activity downstream of CD30 signaling is TRAFs-dependent in Hodgkin’s lymphoma cells, which is regulated by the relative level of individual TRAF in a cell-specific manner.

CD30, a member of the TNF receptor (TNFR) superfamily, is a lymphocyte-specific receptor that is originally recognized as a surface molecule overexpressed on Hodgkin's lymphoma cells. Engagement of CD30 with its ligand CD30L activates the nuclear factor kB (NF-kB), which are mediated by interactions with TNFR-associated factor (TRAFs), and supports proliferation of Hodgkin's cells. However, the role of individual TRAFs in the CD30 signaling pathway in Hodgkin's cells has not been fully addressed yet. In this study, we found that except TRAF3, all other TRAFs were expressed consistently in B cell-derived Hodgkin's cell lines (L428 and KM-H2) either in mRNA or protein level. In L428 but not KM-H2 cells, the classical (p50-RelA) and alternative (p52-RelB) NF-kB activity were constitutively activated measured by Western Blotting and EMSA. To better understand the CD30 signaling properties in Hodgkin's lymphoma cells, we silenced the expression of TRAFs individually by means of RNAi. Successful downregulation of TRAF1 protein expression led to the apoptosis of L428 cells, which was companied by the reduction of both classical and alternative NF-kB activity. Furthermore, the expression of targeting genes of NF-kB, such as A20, c-Flip, ICAM-1, and Cyclin D1was also decreased. Using siRNA targeting TRAF2 expression, the classical NF-kB activity was reduced while the alternative NF-kB activity moderately induced especially upon CD30L treatment, which was companied by the induction of p100 processing and RelB nuclear localization. The survival rate was decreased when TRAF2 was knockdown. TRAF5 knockdown manifested similar results as that of TRAF2. These observations were only took place in L428 cells but not KM-H2 cells. In addition, the phosphorylation of the extracellular signal-regulated kinases (ERK) 1 and 2, and subsequent activation of JunB in Hodgkin's cells were staying unchanged in individual TRAFs knockdown experiment. Taken together, the study here further investigates the features of CD30-TRAFs-NF-kB signaling pathway in Hodgkin's lymphoma cells. At least TRAF1, 2, and 5 are involved in CD30 signaling pathway and protecting Hodgkin's lymphoma cells from undergoing apoptosis. The interaction between TRAFs is necessary to facilitate the signaling and proper activation of classical NF-kB, which is responsible for the antiapoptotic effect of Hodgkin's cells. Among TRAF1, 2, and 5, the redundant function is unlikely. Furthermore, the sequential activation of ERK and JunB is unlikely dependent on TRAFs-NF-kB pathway in Hodgkin's lymphoma cells. Thus, these data reveal that the classical NF-kB activity downstream of CD30 signaling is TRAFs-dependent in Hodgkin's lymphoma cells, which is regulated by the relative level of individual TRAF in a cell-specific manner.

It was hypothesized that osteoblasts play a central role in hematopoiesis, and it has been shown that osteoblasts produce many factors essential for the survival, renewal, and maturation of hematopoietic stem cells (HSCs). By using human fetal osteoblastic cell line hFOB1.19 as a model of control, we investigated the biological characteristics of osteoblasts derived from patients with myelodysplastic syndrome (MDS) and their hematopoietic supportive function in vitro. MSCs isolated from bone marrow of MDS patients and normal donors were cultured and checked for their morphology, immunophenotype, CFU-F forming capacity and the expression of hematopoietic cytokines. A feeder layer was prepared by osteoblasts induced from 3rd generation of cultured MSCs and treated with mitomycin C. Ficoll-isolated bone marrow mononuclear cell from normal donors were then seeded on the feeder layer to co-culture in vitro without exogenous cytokines. FCM revealed that both MSCs and hFOB cells were positive for CD44, CD73(SH3), CD105(SH2) and CD90 (Thy1), but negative for CD34, CD45, HLA-DR. RT-PCR found that hFOB cells expressed mRNA of SCF, IL-6, IL-11, SDF-1, GM-CSF and G-CSF. MSCs obtained from MDS patients and normal donors were displaying fibroblastoid morphology. Their growth pattern, immunophenotype and CFU-F forming capacity were similar (P >0.05). Without exogenous cytokines, the osteoblasts derived from MDS could sustain GM-CFC survival for at least 3 weeks. The CFU-GM yield from cells in upper layer of co-culture was not different from those of control in hematopoiesis supportive experiments in vitro (P>0.05). RT-PCR clearly demonstrated that the cultured BM-MSCs from normal donor expressed mRNA of SCF, SDF-1, IL-6, and IL-11. As the MSCs differentiated toward osteoblasts, the expression of G-CSF could be detected, whereas GM-CSF remained undetectable. The same expression profile of above cytokines were also seen in osteoblasts induced from BM-MSCs of MDS patients. In conclusion, osteoblasts may play a pivotal role in hematopoiesis. The biological characteristics of osteoblasts from bone marrow of MDS patients were generally not different from those of osteoblasts in bone marrow of normal controls. Both of them could support survival of GM-CFC hematopoietic progenitor cells in vitro, according to their expression of multiple cytokines. These findings suggested that the osteoblasts derived from MDS patients may not be involved in the malignant process.

To determine the feasibility of Ig/TCR gene rearrangements as patient-specific targets for the detection of minimal residual disease in Chinese ALL patients, forty adult patients with B-lineage and T cell acute lymphoblastic leukemia(ALL) were enrolled in this study. The frequency and pattern of IGH, IGK, TCRB, TCRG and TCRD gene rearrangement were analyzed by BIOMED-2 multiplex PCR protocol. All DNA samples were obtained from the bone marrow of these patients. IGH, IGK, TCRB, TCRG and TCRD rearrangements were found in 86%,14%,27%,77%,59% of adult patients with B-lineage ALL, respectively. While in T-ALL, clonal TCRB, TCRG and TCRD rearrangements were detected in 83%,78%,39% of patients. There are several features in the pattern of Ig/TCR gene rearrangements in Chinese adult ALL patients. Firstly, IGK rearrangement in adult B-lineage ALL had a lower frequency than in children’s pre-B-ALL (14% vs 41%) in reported literature. Secondly, the frequency of TCRG Vrfl-Jr1.3, Vrfl-Jr1.1 and Vr9-Jr1.3 rearrangement was 65%,6%,29% in adult patients with B-lineage ALL, while 60%,27%,13% was the frequency of those in T-ALL, respectively. In conclusion, our data have shown that several special patterns in Ig/TCR gene rearrangements in Chinese adult ALL patients, may provide a guidance for individual treatment protocol designation and MRD-based risk stratification.

The interaction between killer cell immunoglobin-like receptor (KIR) and HLA molecule has particular relevance to the unrelated hematopoietic stem cell transplantation (HSCT). The presence of KIR on donor’s NK cells and the absence of the corresponding KIR ligand in the recipient’s HLA repertoire as a receptor-ligand model can be used to predict the outcome of HSCT. To investigate the effect of KIR-HLA receptor-ligand mismatching and the expression of inhibitory KIR (iKIR) on the unrelated Allo-HSCT of leukemia, the following study has been carried out. The KIR genotypes of 36 patients of ALL (n=18), AML (n=10), and CML (n=8) as well as their unrelated donors were obtained from the Database of China Marrow Donor Program. KIR genotyping was performed with PCR-SSP. The KIR and HLA mismatching status (receptor-ligand mismatch model) between recipients and donors was analyzed thereafter. The expression of iKIR was determined by flow cytometry on recipients after HSCT. All cases were followed up closely until July, 2007. Among all the donor/recipient pairs analyzed, 15 pairs displayed the matched pattern with respect to the KIR genotyping though, graft-versus-host (GVH) KIR ligand-mismatched in 13 pairs, and host-versus-graft (HVG) KIR ligand-mismatched in 8 pairs. 26 recipients were lack of HLA-Cw2,4,5,6,15 ligands although iKIR/2DL1 was expressed on their corresponding donors. Similarly, 35 donors with iKIR/2DL2/L3 on their NK cells could recognized HLA- Cw1,3,7,8,12,14 ligands in their recipients. iKIR/3DL1 was expressed on 9 donors although the correlated HLA-Bw4 ligand was absent on the recipients. iKIR/3DL2 was expressed on 24 donors although the correlated HLA-A11 ligand was absent on the recipients. Except for one case, 35 patients were successfully transplanted, in which seven patients were dead, giving the survival rate of 80.0%. The cause of death was either acute/chronic GVHD or relapse. The frequency of acute GVHD (60.0%, 9/15), leading to death (26.7%, 4/15), was the highest in KIR receptor-ligand matched model. The incidence of acute GVHD and death was 30.8% (4/13) and 23.1% (3/13) in the GVH KIR ligand-mismatching, respectively; while the incidence of chronic GVHD was 37.5% (3/8) and no death happened in HVG KIR ligand-mismatching pattern. The expression of iKIR/2DL1 could be detected on ALL (13.0%∼16.6%) and CML (1.8∼6.2%) patients four month after HSCT. The expression of both iKIR/2DL1 (4%) and iKIR/2DL2 (12%) was found only in one patient with CML three month after HSCT, increased to 33.4% and 32.6% accompanied by cGVHD, and dropped to 7.6% and 10.3% one year after HSCT. These results suggested that the deficiency in iKIR/2DL1 based on the KIR-HLA receptor-ligand mismatching model together with the expression of a number of activating KIR in recipients increases the occurrence of GVHD, relapse, and death in unrelated HSCT. Analysis on KIR-HLA gene expression profiling could be useful in predicting the clinical outcome of unrelated Allo-HSCT in leukemia patients. The overall and disease-free survival after HSCT could be improved by preventing the development of GVHD.

Several reports indicated a striking difference in age, chromosome abnormalities and prognosis between Western and Eastern MDS patients, even in the same subtype. The prognostic factors in Chinese MDS patients remain unclear. To investigate the prognostic factors and survival rate in Chinese cases with refractory anemia (RA) in adult MDS based on FAB classification, and to evaluate the applicability of international prognostic scoring system (IPSS) for Chinese MDS-RA patients by comparing with the clinical features of Western cases, Three hundred and seven MDS-RA cases were registered and followed-up in Shanghai, Suzhou and Beijing from 1995 to 2006. The longest follow-up duration was 103 months, and the median follow-up duration was 26.7 months. Kaplan-Meier curve, Log-rank and COX regression model were used to analyze the prognostic factors and survival rate. The median age of 307 MDS-RA cases was 52 years. The frequency of 2 or 3 lineage cytopenias was 85.6%. Abnormal karyotype occurred in 35.7% of all MDS-RA patients. By IPSS cytogenetic risk groups, 165 cases (70.2%) were in the good IPSS cytogenetic subgroup, 44 cases (18.7%) in the intermediate IPSS cytogenetic subgroup and 26 cases (11.1%) in the poor IPSS cytogenetic subgroup. According to IPSS, 20 cases (8.5%) were categorized as low risk, 195 cases (83.0%) as intermediate–I risk and 20 cases (8.5%) as intermediate–II risk. The 1–year, 2–year, 3–year, 4–year and 5–year survival rate were 90.8%, 85.7%, 82.9%, 74.9% and 71.2%, respectively. Fifteen cases (4.9%) transformed to acute myeloid leukemia and the median transformation time was 15.9 months (range 3–102 months). Lower white blood cells count (<1.5 × 109/L), platelet count (<30 × 109/L) and cytogenetic abnormalities were identified as independent prognostic factors by multivariate analysis, while the age (≥65 years), hemoglobin level(<60g/L), IPSS cytogenetic subgroup and IPSS risk subgroup were not independent prognostic factors associated with survival time by COX regression analysis. In summary, Chinese patients were younger and had lower incidence of cytogenetic abnormalities and more severe cytopenias than that of the Western patients. However, the Chinese patients have more favorable prognosis than Western patients do. WBC count, platelet count and karyotype are major prognostic factors for prediction of survival and can be helpful in identifying patients with different prognosis and consequently designing the optimal therapeutic strategies. The present IPSS scores based on conditions of western patients may not be perfectly applicable to MDS-RA patients in Asia and need to be modified.

Previous studies on either animal model or clinical settings have demonstrated that the expression of AML1/ETO9a (AE9a), a truncated version of AML1-ETO (AE) generated by alternative splicing during transcription, in patients with acute myeloid leukemia M2 subtype may have clinical significance. Moreover, the simultaneously expression of both AE and AE9a may even be more meaningful to the prognosis of the patients. To validate this hypothesis, we have measured the mRNA level of both AE and AE9a in each individual M2 patient by quantitative real-time RT-PCR, and correlated the ratio of AE/AE9a mRNA levels with the clinical outcome and prognosis in these patients. In 35 newly-diagnosed AML-M2 patients with t(8;21), expression of AE9a was detected in 30 cases (85.71%). The median value of the percentages of AE9a and AE within overall transcripts of the fusion gene in these patients were 18.16% and 81.84%, respectively (p<0.05). A positive correlation between the mRNA level of AE9a and AE was shown in 45 bone marrow samples from 19 patients during the follow-up, and the correlation coefficient was 0.900(P<0.01). The expression level of both AE9a and AE decreased in 16 patients after one course of standard conventional chemotherapy. Among them, a reduced percentage of AE9a mRNA level was prominent in 11 (68.75%) patients, while the percentage of mRNA level of AE decreased in 5 (31.25%) cases. The complete remission rate after first course of chemotherapy was 81.82%(9/11)in 11 patients, and 20.00%(1/5)in 5 patients, suggesting a higher CR rate in those patients who exhibited more conspicuous decrease in mRNA level of AE9a (P<0.05). The percentage of AE9a mRNA in the total fusion transcripts rose in all of the 5 relapsing cases. In a retrospective study on 92 patients of AML-M2 including 50 cases with t(8;21) and 42 case with either normal karyotype or other chromosome abnormalities, expression of AE9a has been found in these 50 cases by conventional RT-PCR. Between the AE9a positive and negative groups, no statistically significant difference had been found in complete remission rate, relapsing rate and the first CR period over a follow-up with medium time of 13 months (4–76 mths). However, 22 months of medium survival in AE9a positive group (38 cases) was remarkably shorter than 70 months in AE9a negative group (25 cases). Our results indicated that the mRNA of AE9a at a relatively lower level co-expressed with AE in most AML-M2 patients with t(8;21) translocation. The expression level of AE9a rather than AE may play more crucial role in development and progress of leukemia. AE9a positive patients have distinctly shorter survival time than the negative cases, predicting a worse prognosis.

The abnormal activity of DNA repair after DNA double strand breakage results in genome instability. It is also thought that the malfunction in DNA repair may be involved in the generation of variety of chromosomal translocations in human leukemia. To investigate the relationship between the activity of DNA repairing by NHEJ and the expression of protein Ku70 in CML cells, the expression of protein Ku70 deprived from the chromatin was determined by Western Blot, using TBP(TATA Box Binding Protein)as the internal control, in bone marrow cells of CML patients as well as the normal controls. Our preliminary result showed that the gray ratios of Ku70 protein band over TBP band in 24 normal BM cells were 0 to 0.235, with a mean level of 0.0983±0.0875, while which were 0.173 to 0.890 with a mean of 0.401±0.207 (p<0.001) in 27 CML cases. Furthermore, Among these 27 CML cases, the gray ratio of Ku70 versus TBP in 12 CML cases in accelerating of acute phase was significantly higher than that in 15 CML patients in chronic phase with a mean value of 0.565±0.2159 vs. 0.269±0.0381 (p<0.001). These results indicated that the activity NHEJ in CML cells was approximately 2-fold higher than that in normal controls. Ku70, as the key protein of NHEJ, was expressed significantly higher in CML than that in normal BM cells. Moreover, as the CML progressed from chronic phase to accelerating or acute phase in company with a remarkable increase of Ku70 expression, more clones with variety of chromosome abnormalities other than the typical t (9;22) translocation emerged, suggesting an error-prone repair activity of the increased NHEJ in CML. The overactive DNA repair ability of NHEJ associated Ku70 proteins found in CML cells may highly correlated to the chromosome aberration in CML, especially to the genome instability during the disease progression.

Previous studies on either animal model or clinical settings have demonstrated that the expression of AML1/ETO9a (AE9a), a truncated version of AML1-ETO (AE) generated by alternative splicing during transcription, in patients with acute myeloid leukemia M2 subtype may have clinical significance. Moreover, the simultaneously expression of both AE and AE9a may even be more meaningful to the prognosis of the patients. To validate this hypothesis, we have measured the mRNA level of both AE and AE9a in each individual M2 patient by quantitative real-time RT-PCR, and correlated the ratio of AE/AE9a mRNA levels with the clinical outcome and prognosis in these patients. In 35 newly-diagnosed AML-M2 patients with t(8;21), expression of AE9a was detected in 30 cases (85.71%). The median value of the percentages of AE9a and AE within overall transcripts of the fusion gene in these patients were 18.16% and 81.84%, respectively (p<0.05). A positive correlation between the mRNA level of AE9a and AE was shown in 45 bone marrow samples from 19 patients during the follow-up, and the correlation coefficient was 0.900(P<0.01). The expression level of both AE9a and AE decreased in 16 patients after one course of standard conventional chemotherapy. Among them, a reduced percentage of AE9a mRNA level was prominent in 11 (68.75%) patients, while the percentage of mRNA level of AE decreased in 5 (31.25%) cases. The complete remission rate after first course of chemotherapy was 81.82%(9/11)in 11 patients, and 20.00%(1/5)in 5 patients, suggesting a higher CR rate in those patients who exhibited more conspicuous decrease in mRNA level of AE9a (P<0.05). The percentage of AE9a mRNA in the total fusion transcripts rose in all of the 5 relapsing cases. In a retrospective study on 92 patients of AML-M2 including 50 cases with t(8;21) and 42 case with either normal karyotype or other chromosome abnormalities, expression of AE9a has been found in these 50 cases by conventional RT-PCR. Between the AE9a positive and negative groups, no statistically significant difference had been found in complete remission rate, relapsing rate and the first CR period over a follow-up with medium time of 13 months (4–76 mths). However, 22 months of medium survival in AE9a positive group (38 cases) was remarkably shorter than 70 months in AE9a negative group (25 cases). Our results indicated that the mRNA of AE9a at a relatively lower level co-expressed with AE in most AML-M2 patients with t(8;21) translocation. The expression level of AE9a rather than AE may play more crucial role in development and progress of leukemia. AE9a positive patients have distinctly shorter survival time than the negative cases, predicting a worse prognosis.

Several reports indicated a striking difference in age, chromosome abnormalities and prognosis between Western and Eastern MDS patients, even in the same subtype. The prognostic factors in Chinese MDS patients remain unclear. To investigate the prognostic factors and survival rate in Chinese cases with refractory anemia (RA) in adult MDS based on FAB classification, and to evaluate the applicability of international prognostic scoring system (IPSS) for Chinese MDS-RA patients by comparing with the clinical features of Western cases, Three hundred and seven MDS-RA cases were registered and followed-up in Shanghai, Suzhou and Beijing from 1995 to 2006. The longest follow-up duration was 103 months, and the median follow-up duration was 26.7 months. Kaplan-Meier curve, Log-rank and COX regression model were used to analyze the prognostic factors and survival rate. The median age of 307 MDS-RA cases was 52 years. The frequency of 2 or 3 lineage cytopenias was 85.6%. Abnormal karyotype occurred in 35.7% of all MDS-RA patients. By IPSS cytogenetic risk groups, 165 cases (70.2%) were in the good IPSS cytogenetic subgroup, 44 cases (18.7%) in the intermediate IPSS cytogenetic subgroup and 26 cases (11.1%) in the poor IPSS cytogenetic subgroup. According to IPSS, 20 cases (8.5%) were categorized as low risk, 195 cases (83.0%) as intermediate–I risk and 20 cases (8.5%) as intermediate–II risk. The 1–year, 2–year, 3–year, 4–year and 5–year survival rate were 90.8%, 85.7%, 82.9%, 74.9% and 71.2%, respectively. Fifteen cases (4.9%) transformed to acute myeloid leukemia and the median transformation time was 15.9 months (range 3–102 months). Lower white blood cells count (<1.5 × 109/L), platelet count (<30 × 109/L) and cytogenetic abnormalities were identified as independent prognostic factors by multivariate analysis, while the age (≥65 years), hemoglobin level(<60g/L), IPSS cytogenetic subgroup and IPSS risk subgroup were not independent prognostic factors associated with survival time by COX regression analysis. In summary, Chinese patients were younger and had lower incidence of cytogenetic abnormalities and more severe cytopenias than that of the Western patients. However, the Chinese patients have more favorable prognosis than Western patients do. WBC count, platelet count and karyotype are major prognostic factors for prediction of survival and can be helpful in identifying patients with different prognosis and consequently designing the optimal therapeutic strategies. The present IPSS scores based on conditions of western patients may not be perfectly applicable to MDS-RA patients in Asia and need to be modified.

The abnormal activity of DNA repair after DNA double strand breakage results in genome instability. It is also thought that the malfunction in DNA repair may be involved in the generation of variety of chromosomal translocations in human leukemia. To investigate the relationship between the activity of DNA repairing by NHEJ and the expression of protein Ku70 in CML cells, the expression of protein Ku70 deprived from the chromatin was determined by Western Blot, using TBP(TATA Box Binding Protein)as the internal control, in bone marrow cells of CML patients as well as the normal controls. Our preliminary result showed that the gray ratios of Ku70 protein band over TBP band in 24 normal BM cells were 0 to 0.235, with a mean level of 0.0983±0.0875, while which were 0.173 to 0.890 with a mean of 0.401±0.207 (p<0.001) in 27 CML cases. Furthermore, Among these 27 CML cases, the gray ratio of Ku70 versus TBP in 12 CML cases in accelerating of acute phase was significantly higher than that in 15 CML patients in chronic phase with a mean value of 0.565±0.2159 vs. 0.269±0.0381 (p<0.001). These results indicated that the activity NHEJ in CML cells was approximately 2-fold higher than that in normal controls. Ku70, as the key protein of NHEJ, was expressed significantly higher in CML than that in normal BM cells. Moreover, as the CML progressed from chronic phase to accelerating or acute phase in company with a remarkable increase of Ku70 expression, more clones with variety of chromosome abnormalities other than the typical t (9;22) translocation emerged, suggesting an error-prone repair activity of the increased NHEJ in CML. The overactive DNA repair ability of NHEJ associated Ku70 proteins found in CML cells may highly correlated to the chromosome aberration in CML, especially to the genome instability during the disease progression.

It was hypothesized that osteoblasts play a central role in hematopoiesis, and it has been shown that osteoblasts produce many factors essential for the survival, renewal, and maturation of hematopoietic stem cells (HSCs). By using human fetal osteoblastic cell line hFOB1.19 as a model of control, we investigated the biological characteristics of osteoblasts derived from patients with myelodysplastic syndrome (MDS) and their hematopoietic supportive function in vitro. MSCs isolated from bone marrow of MDS patients and normal donors were cultured and checked for their morphology, immunophenotype, CFU-F forming capacity and the expression of hematopoietic cytokines. A feeder layer was prepared by osteoblasts induced from 3rd generation of cultured MSCs and treated with mitomycin C. Ficoll-isolated bone marrow mononuclear cell from normal donors were then seeded on the feeder layer to co-culture in vitrowithout exogenous cytokines. FCM revealed that both MSCs and hFOB cells were positive for CD44, CD73(SH3), CD105(SH2) and CD90 (Thy1), but negative for CD34, CD45, HLA-DR. RT-PCR found that hFOB cells expressed mRNA of SCF, IL-6, IL-11, SDF-1, GM-CSF and G-CSF. MSCs obtained from MDS patients and normal donors were displaying fibroblastoid morphology. Their growth pattern, immunophenotype and CFU-F forming capacity were similar (P>0.05). Without exogenous cytokines, the osteoblasts derived from MDS could sustain GM-CFC survival for at least 3 weeks. The CFU-GM yield from cells in upper layer of co-culture was not different from those of control in hematopoiesis supportive experiments in vitro(P>0.05). RT-PCR clearly demonstrated that the cultured BM-MSCs from normal donor expressed mRNA of SCF, SDF-1, IL-6, and IL-11. As the MSCs differentiated toward osteoblasts, the expression of G-CSF could be detected, whereas GM-CSF remained undetectable. The same expression profile of above cytokines were also seen in osteoblasts induced from BM-MSCs of MDS patients. In conclusion, osteoblasts may play a pivotal role in hematopoiesis. The biological characteristics of osteoblasts from bone marrow of MDS patients were generally not different from those of osteoblasts in bone marrow of normal controls. Both of them could support survival of GM-CFC hematopoietic progenitor cells in vitro, according to their expression of multiple cytokines. These findings suggested that the osteoblasts derived from MDS patients may not be involved in the malignant process.

To determine the feasibility of Ig/TCR gene rearrangements as patient-specific targets for the detection of minimal residual disease in Chinese ALL patients, forty adult patients with B-lineage and T cell acute lymphoblastic leukemia(ALL) were enrolled in this study. The frequency and pattern of IGH, IGK, TCRB, TCRG and TCRD gene rearrangement were analyzed by BIOMED-2 multiplex PCR protocol. All DNA samples were obtained from the bone marrow of these patients. IGH, IGK, TCRB, TCRG and TCRD rearrangements were found in 86%,14%,27%,77%,59% of adult patients with B-lineage ALL, respectively. While in T-ALL, clonal TCRB, TCRG and TCRD rearrangements were detected in 83%,78%,39% of patients. There are several features in the pattern of Ig/TCR gene rearrangements in Chinese adult ALL patients. Firstly, IGK rearrangement in adult B-lineage ALL had a lower frequency than in children's pre-B-ALL (14% vs41%) in reported literature. Secondly, the frequency of TCRG Vrfl-Jr1.3, Vrfl-Jr1.1 and Vr9-Jr1.3 rearrangement was 65%,6%,29% in adult patients with B-lineage ALL, while 60%,27%,13% was the frequency of those in T-ALL, respectively. In conclusion, our data have shown that several special patterns in Ig/TCR gene rearrangements in Chinese adult ALL patients, may provide a guidance for individual treatment protocol designation and MRD-based risk stratification.

Non-homologous end joining (NHEJ) is a major mechanism by which eukaryote cells can repair the DNA double-strand break (DSB) and protect the cell from further damages. Evidences have suggested that the genomic instability caused by deficient DNA repair function may contribute to the development of solid tumor, while its role in leukemogenesis has not been adequately studied. To study the NHEJ function in myeloid leukemia, an in vitro system was developed for clinical samples by using the linear plasmid pUC18 DNA digested with EcoR I as assay system.. Nuclear protein extracted from leukemic cells and mononuclear cells (MNCs) from normal BM or PB, was incubated with linear plasmid pUC18 DNA under certain conditions. Plasmid DNA was separated by agarose gel electrophoresis and imaged by SYBR greenIstaining. End-ligation efficiency was assessed by dividing the densitometry readings for the sum of all converted plasmid products by the sum of all products. This assay was performed on 7 myeloid leukemia cell lines, 16 samples of normal BM or PB cells, 20 cases of CML cells and 19 cases of de novo AML cells. E. coli strain DH5α was electrotransformed with pUC18 DNA end-joined by nuclear proteins from normal BM or leukemia cells, and was plated on LB agar, containing X-gal and IPTG. Correct ligation of cut plasmid DNA resulted in blue colonies while faulty repair would result in white colonies. The percentage of white colonies over total colonies gave the frequency of misrepair. Primers around the EcoR I site were designed and colony PCR was performed on blue and white colonies. Sequencing of PCR products was performed. Antibodies against the nuclear repair proteins Ku70, Ku86 and DNA-PKcs were used for antibody abrogation studies. We have found that the mean repair efficiency of normal MNCs was 18.6±13.1% (0~46.6%),. The ligation efficiencies in myeloid leukemic cell lines ranged 10.6%~31.0% (mean 22.4%, P>0.05). The mean ligation efficiency was significantly higher in CML cells as compared with normal MNCs (24.8±14.9% v.s 18.6%, p=0.024). The DNA repair capacity of de novo acute myeloid leukemia cells(7.2%~76.9%)was markedly increased as compared with normal MNCs (mean 41.1±15.4% v.s 18.6%, p<0.0001 ). The mean frequency of misrepair from AML and CML cells were considerable higher (AML, 8.17% and CML, 2.10%) than that of normal BM cells (0.91%). Most misrepair were small deletions near EcoR I site. Large deletions (>100bp) were found from assays with AML cells. Three abnormalities of sequence deletion, inversion and insertion were found in a rare misrepair white colony with AML cells. The DNA end-ligation efficiencies of AML cells could be dramatically inhibited by antibodies against proteins Ku70, Ku86 and DNA-PK. We concluded that the Cell-free NHEJ assay system can be used to examine the clinical leukemic samples. NHEJ efficiencies were slightly enhanced in most myeloid leukemia cell lines compared with normal MNCs. The end-ligation efficiency was significantly higher in primary myeloid leukemia cells. The overactive but more error-prone NHEJ function relied on the activity of Ku and DNA-PK proteins for DSB repair may contribute to genomic instability in AML cells.

Central nervous system leukemia (CNSL) is one of the major complications of acute leukemia which always lead to relapse and imply a poor prognosis. Our study aimed to establish a human monocytic leukemia model with CNS infiltration in Balb/C nude mice, so as to help investigation on pathogenesis of CNSL. 11 and 16 of four and six-week-old Balb/C nu/nu mice, respectively, were treated by splenectomy, cytoxan intraperitoneal injection, and sublethal irradiation, and referred to as SCI-nu/nu mice. SHI-1 monoblasts were derived from an acute monocytic leukemia patient, featured by t(6;11)(q21;q23), P53 dysfunction and a high tumorogenecity. SHI-1 cell were injected into tail vein of mouse. One of the four-week-old mouse was sacrificed in one week’s interval until the mice die of leukemia. 6 of six-week-old mice were also killed 30 days after inoculation. The leukemic cells engrafted in the mice were tracked by RT-PCR, histopathological examination, immunohistochemistry and FCM. All mice were inspected carefully and all tissues from different organs including head and spine were properly processed and sectioned. CNSL was graded as follows: grade 0: no leukemic cell infiltrated in brain; grade I: mild and focally infiltration in subdural space and pia-arachnoid; grade II: moderate to dense infiltration in subdural space and pia-arachnoid without penetrating to brain parenchyma; grade III: extensive infiltration in brain parenchyma. 25–30 days after injection, SCI-nu/nu mice presented anorexia, hunched posture, lethargy, weight loss and died after a median of 41 days (range, 33 to 46 days). All the 4-week-old mice were paralysed in rear legs, while only 30% in six-week-old mice,. Autopsy showed that green solid tumor were grossly formed in stomach, kidney, thoracic and lumbar vertebrae, head, neck, mediastinum, heart, scapula. MLL-AF6 fusion gene products could be amplified in multiple organs by RT-PCR including in the brain. Leukemic infiltrated foci were seen in liver, lung, stomach, kidney, heart, testis. In paralyzed mice the destructed thoracic and lumbar vertebral bone marrow was fully occupied by leukemic cells. Leukemic cells penetrated to the surface of vertebrae, forming neoplasm, and traveled to the subdural space, but seldom involving the parenchyma, indicating that paralysis was most likely caused by the destruction of bone and compression by leukemia cells. Central nervous system involvement were found in all SCI-nu/nu mice: grade I 71.4% (5/7); grade II 28.6 % (2/7) after 30 days, and becoming more seriously [grade II 35.3%(6/17), grade III 64.7%(11/17)] upon death:. Leukemia cells filled in the subdural space and pia-arachnoid, covered the surface of cerebrum, cerebellum and along the virchow-robin space of pia mater. For grade III, leukemic cells focally scattering around the peri-vascula could be seen in deep parenchyma. Brain parenchyma infiltration always accompanied by massive infiltration of adjacent pia-arachnoid and skull. The series of sections from our successful and reproducible experimental model of CNSL and systemic leukemia in SCI-nu/nu mice clearly demonstrated that the leukemia cells in central nervous system came from bone marrow of skull and vertebrae, via subdural space, pia-arachnoid, and the Virchow-Robin spaces, finally penetrated into the nervous parenchyma. This experimental model may provide an useful tool for studies on the pathogenesis of CNSL.

The chemokine, stromal cell-derived factor-1 (SDF-1), which is mostly produced by marrow stromal cells, has various effects on hematopoietic cell functions. Its crucial role in the survival, trafficking, and homing of hematopoietic stem cells and progenitor cells is currently established. Here we described the potential significance of bone marrow plasma levels of SDF-1α in patients with hematopoietic malignancies. The bone marrow plasma was obtained from 74 aspirate samples from 38 patients with acute myeloid leukemia (AML), 12 acute lymphoblastic leukemia (ALL) and 24 myelodysplastic syndrome (MDS). The concentration of SDF-1α was determined by a sandwich enzyme-linked immunosorbent assay (ELISA). The results revealed a significant negative correlation of marrow plasma levels of SDF-1 with peripheral white blood cells counts (r=−0.304, P<0.05). The patients with low or normal WBC count (<=10×109/L) had significantly higher SDF-1 levels than those with high WBC count (>10×109/L) (358.88±251.64 pg/mL vs 149.33±129.16 pg/mL, P<0.001). The correlation of SDF-1α levels with platelet counts was also observed (r=0.279, P=0.002). No statistically significant differences were observed in the marrow plasma levels of SDF-1α between normal and abnormal karyotypic groups. There was no significant difference between patients with AML and ALL (279.24±318.50 pg/mL vs 209.78±120.96 pg/mL, P>0.05), however, MDS patients had significantly higher SDF-1α levels (528.65±351.57 pg/mL) than AML and ALL patients (P<0.001). In the subtypes of AMLs, granulocytic leukemia has higher marrow plasma levels of SDF-1α (337.81±350.92 pg/mL, n=28) than mononuclear leukemia (104.94±96.12 pg/mL, n=7). Our results suggest that the migration of malignant leukemic hematopoietic cells, especially of mononuclear leukemic cells, might be associated with the marrow SDF-1α level.

The rearrangement of MLL gene is reported in 70%~80% of infant and in 5%~10% of older children (under the age of 15) with acute leukemia (AL). The biological features associated with alterations in MLL gene are hyperleukocytosis, CD10−/CD19+ phenotype and very poor prognosis. To explore the MLL rearrangement in details in our AL children patients and obtain more information on the relationship between the MLL gene abnormality and clinical outcomes. The following study has been conducted. A total of 298 patients with AL attended The Affiliated Children’s Hospital of Soochow University, including 16 cases with MLL rearrangements, were recruited in this study. Of the cohort, 11 were diagnosed as ALL, 5 were AML. 9 of 16 patients were in infant age (up to 2 year) and the rest were between the age of 2 to 13 years. Fluorescence in situ hybridization (FISH) analysis using LSI MLL dual color probe. Multiplex reverse transcriptase- polymerase chain reaction (multiplex RT-PCR) were used to discriminate 13 different fusion transcripts. These results were analyzed together with R banding karyotyping and immunolphenotyping determined by flow cytometry. We have found MLL rearrangements in 16 cases of childhood AL which were accounted for 5.4% of 298 AL patients, and 56.3% of infant AL. Among 106 cases analyzed by multiplex RT-PCR, MLL gene rearrangement was found in 11 cases, including MLL/AF4 fusion gene in 2, MLL/AF6 fusion gene in 1, MLL/AF6, MLL/ELL combined with MLL/AFX or HOX11 in one of each, MLL/AF9 in 2, MLL/AF10 in 1, MLL/ELL in 2. MLL partial tandem duplication in 1. In addition an activated HOX11 gene was found in 1 case.. In 27 cases assayed by FISH, MLL gene rearrangements have been detected in 9 cases (36.0%). In 16 patients with MLL gene rearrangements, 14 (87.5%) exhibited clonal chromosome abnormalities involved chromosome 11 in 11 cases, presenting as t(4;11) in 2, t(6;11), t(8;11), t(7;8;11), and t(9;11) in one of each, respectively, trisomy 11 in 2 and 11q- in 3 cases. Among these 16 patients, 11 were B-ALL, including Pro-B and Pre-B ALL; 5 of AML-M5, 3 of these 5 M5 patients were CD7+ and CD2+. Of these 16 patients 8 received chemotherapy and 7 of them achieved complete remission, while the other 8 patients eventually gave up treatment. Our results demonstrated that multiplex RT-PCR combined with FISH provided a more accurate and sensitive method for detection of MLL gene rearrangements, including chromosome translocation, deletion and duplication. Our findings lead to the detection of novel rearrangements at molecular genetic level. These findings regarding the MLL rearrangement provide most important information in guiding therapy and predicting prognosis in childhood AL. Besides our results also provide evidence in support of the value of 11q23/MLL in WHO classification categories.

The methylation at CpG island in the promoter region of a gene is one of the important epigenetic mechanism which regulates the gene activity. To study the DNA methylation pattern of WT1 gene promoter region within hematologic neoplastic cell lines and its correlation with WT1 gene expression by using the PCR-based methods. RT-PCR and Methylation-specific PCR were performed to study the WT1 gene expression in 8226, HL-60, Jurkat, K562, KG-1, NB4, Raji, SHI-1, U266 and U937 cell lines and the DNA methylation status in promoter region of WT1 gene. After treatment of U937 cell line by 5-aza-CdR, a demethylation inducing agent, the changes of WT1 gene expression level and the methylation status in its promter region in U937 cells was determined. Our Results showed that HL-60, K562, KG-1, NB4, SHI-1 cell lines demonstrated higher level of WT1 expression, while extremely low level was found in 8226, Jurkat, Raji, U266 and U937. The DNA hypermethylation in WT1 gene promoter region was identified in 8226, Jurkat, Raji, U266 and U937 cell lines. The WT1 gene expression in U937 was markedly enhanced after treatment with 5-aza-CdR in company with the decrease of methylated level and the increase of unmethylated level in its promoter region. These results indicate that modulation of the DNA methylation in WT1 promoter region is one of the epigenetic mechanisms to regulate its expression.

IGFBP7 (insulin-like growth factor binding protein 7) is a newly-identified protein which can bind with insulin-like growth factors(IGFs), with 45%~65% homology to other IGF binding proteins. Several studies reported that IGFBP7 may be involved in the development of meningioma, breast cancer and prostate cancer, while its correlation with human leukemia had not been reported yet. We have observed an induced expression of IGFBP7 in human leukemia K562 cells which were transfected with and overexpressed RbAp46 gene. Besides, the expression level of WT1, RbAp46 and IGFBP7 was coordinately regulated during the induced differentiation of NB4 acute promyelocytic leukemia cells (data not shown), suggesting a role of IGFBP7 gene in human leukemia.. To investigate the expression level of IGFBP7(previous known as Insulin-like growth factor binding protein related protein 1)in bone marrow (BM) of leukemia patients and its correlation with other leukemia biomarkers, Real-time quantitative reverse transcription polymerase chain reaction (QRT-PCR) method was established for detecting IGFBP7 expression levels in BM of 127 patients with acute leukemia (AL), 24 with chronic myelogenous leukemia in chronic phase (CML-CP), 9 with CML in blast crisis (CML-BC) and 27 with non-leukemias. The M-Estimators of RbAp46N in 63 patients with newly diagnosed AMLs were higher than those of 27 with ALLs and 27 non-leukemic controls and 10 with AMLs in complete remission(CR)(193.80 v.s 88.84, 81.30 and 67.82, respectively). Although IGFBP7N was 101.13 in CML-BC and 168.77 in CML-CP, difference between them was not statistically significant. Difference in IGFBP7N values among CML-BC, AML, ALL and control groups was also not significant. The M-Estimators of IGFBP7N were lowest in M4 and highest in M5 among initial acute leukemia subtypes, and there was statistical differences between M4 and M5, and between M4 and M2. IGFBP7N expression level was not correlated to the expression of fusion genes BCR/ABL and AML1/ETO, but it was positively correlated with the expression level of PML/RARαfusion gene, the WT1 and RbAp46 gene. The correlation coefficiency was 0.71, 0.31 and 0.52, respectively. Our results suggested that IGFBP7 might act in WT1 mediating pathway to participate in acute myelocytic leukemias.

The notch family of transmembrane receptors play crucial roles in cell fate determination through interaction with the Delta/Serrate /Lag-2 family of ligands. The transcription of the hairyl enhancer of split (HES-1) gene is one of the known target of notch signaling. Notch1 is involved in the pathogenesis of T-acute lymphoblastic leukemia (T-ALL) carrying the very rare translocation t (7;9)(q34;q34.3). However, the function of Notch signaling in leukemogenesis is currently still unclear. To investigate the possible role of notch signaling in human acute leukemia,A quantitative real-time reverse transcriptase polymerase chain reaction (RQ-RT-PCR) method has been established for detecting expression levels of Notch1~2, Jagged1, Delta1, Hes1 and internal reference GAPDH in primary BM samples of 34 acute leukemia patients and 11 normal controls by LightCycler using TaqMan probe; 34 pretreated acute leukemia patients includes 16 acute lymphoblastic leukemia (7 T-ALL, 9 B-ALL) and 18 acute myeloblastic leukemia (10 M1,3 M2,3 M3 and 2 M5) In all of these cases the leukemia blast cells are more than 80%. The results demonstrated as the follows: The average expression levels of Notch1 in 16 ALL were statistically higher than that in 11 normal and 18 AML [ 13225.0±3306.3 (16) vs 1725.6±1237.8 (11) vs 2354.8±2985.6(18), F=7.85, P<0.01, ] There was no statistically difference between normal and AML group( P>0.05); the average expression levels of Notch2 did not have statistically difference among ALL, normal and AML.[143671.7±372814.9 (16) vs 52178.2±36747.5 (11) vs 66520.2±62643.2 (18) F=0.15, P>0.05 ]. The average expression levels of Jagged1 in 16 ALL were statistically less than that in 18 AML and 11 normal [129.9±353.8(16) vs 375.4±697.5(18) vs 2212.5±3371.7(11), F=20.23, P<0.01] while the jagged1 level was significantly lower in AML than in normal. The average levels of Delta1 in 16 ALL were statistically higher than that in normal and AML[30884.6±104097.1(16) vs 2299.8±4301.4(11) vs 579.7±641.7(18), F=5.78, P<0.01], and no statistically difference between AML and normal has been found. The average levels of Hes1 in AML were statistically less than normal and ALL[20.1±20.1(18) vs 1013.2±2507.0 (11) vs 244.5±347.3(16), F=6.62, P<0.01, there were no statistically difference between normal and ALL,(P.>0.05)]. In addition, the average levels of Delta1 in 9 B-ALL were statistically higher than those in 7 T-ALL[60316.6±145713.4(9) vs 378.0±238.6 (7), t=4.356, p=0.001]., the average expression levels of Notch1~2, Jagged1 and Hes1 in B-ALL were no statistically difference compare to T-ALL. In conclusion, Notch1 and Delta1 expression levels in ALL were statistically higher than that in normal and AML group, there were no statistically difference between normal and AML. Jagged1 expression levels in ALL were statistically less than that in AML and least in normal. Hes1 expression levels in AML were statistically less than that in normal and ALL, while no statistically difference could be found between normal and ALL groups. Taking altogether, in AL, the expression of Hes1 had not been found upregulated, instead was downregulated in AML. No statistically difference in Notch2 expression has been found among AML, ALL and normal group. In addition, Delta1 expression levels in B-ALL were statistically higher than that in T-ALL.