Upregulation of multi drug resistance genes in doxorubicin resistant human acute myelogeneous leukemia cells and reversal of the resistance

The major problem in the treatment of acute myeloid leukemia (AML) patients results from multidrug resistance to administered anticancer agents. Drug resistance proteins, MDR1 and MRP1, which work as drug efflux pumps, can mediate the multidrug resistance of human leukemia cells. In this study, the mechanisms of resistance to doxorubicin-induced cell death in human HL60 AML cells were examined. Continuous exposure of cells to step-wise increasing concentrations of doxorubicin resulted in the selection of HL60/DOX cells, which expressed about 10.7-fold resistance as compared to parental sensitive cells. The expression analyses of MRP1 and MDR1 drug efflux proteins in doxorubicin-sensitive and -resistant HL60 cells revealed that there was an upregulation of MRP1 gene in HL60/DOX cells as compared to parental sensitive cells. On the other hand, while there was no expression of MDR1 gene in parental cells, the expression of MDR1 gene was upregulated in HL60/DOX cells. HL60/DOX cells also showed cross-resistance to cytosine arabinoside (Ara-c). This resistance was reversed by a combination therapy of Ara-c and cyclosporine A. However, the expression levels of CD15 and CD16 surface markers were significantly decreased in HL60/DOX cells.     

Binding of G-CSF, GM-CSF, tumor necrosis factor-alpha, and gamma- interferon to cell surface receptors on human myeloid leukemia cells triggers rapid tyrosine and serine phosphorylation of a 75-Kd protein

Granulocyte colony-stimulating factor (G-CSF), granulocyte-macrophage colony-stimulating factor (GM-CSF), gamma-interferon (gamma-IFN), or tumor necrosis factor-alpha (TNF-alpha) triggered the rapid, stable phosphorylation of a 75-Kd protein (p75) when incubated with permeabilized HL60 human myeloid leukemia cells in the presence of [gamma-32P] ATP. Among several chemical inducers of HL60 cell differentiation, dimethyl sulfoxide also triggered p75 labeling, but retinoic acid or 12-O-tetradecanoylphorbol-13-acetate did not elicit this response. Pretreatment of cells with G-CSF or GM-CSF for more than 30 seconds before permeabilization rendered the p75 labeling undetectable, suggesting that ligand-stimulated labeling was rapidly completed within this time in intact cells. Phosphorylation of p75 occurred on serine and tyrosine residues. This conclusion was confirmed by direct phosphoamino acid analysis. Immunoblot analysis of lysates of intact HL60 cells that had been incubated with G-CSF, GM-CSF, IFN, or TNF confirmed that tyrosine phosphorylation of a p75 also occurred in response to these cytokines in intact cells. Pretreatment of intact HL60 cells with one biologic agent or dimethyl sulfoxide abolished p75 labeling in response to incubation of permeabilized cells with a second agent, strongly suggesting that the same protein was phosphorylated in response to these treatments. p75 labeling was strictly dependent on expression of the appropriate ligand receptor. Data suggest that activation of a tyrosine kinase system is an early response to the binding of G-CSF, GM-CSF, TNF, or IFN to their respective cell surface receptors, or to the addition of dimethyl sulfoxide, and that the resulting phosphorylation event(s) may play a role in securing common elements in the biologic responses to these agents.     

The novel anthracycline annamycin is not affected by P-glycoprotein- related multidrug resistance: comparison with idarubicin and doxorubicin in HL-60 leukemia cell lines

A major factor in limiting the efficacy of anthracyclines is overexpression of the MDR1-encoded p-glycoprotein (p-gp). A new analogue less affected by p-gp is annamycin (ANN), an anthracycline antibiotic with high affinity for lipid membranes and significantly more activity than doxorubicin (DOX). We investigated whether ANN was affected by p-gp-mediated multidrug resistance (MDR) by comparing the cellular accumulation and retention of ANN, idarubicin (IDR), and DOX in the p-gp-negative human leukemia cell lines (HL-60S) and its DOX- selected p-gp-positive subline (HL-60/DOX) with and without verapamil (VER). As expected, HL-60/DOX cells showed lower DOX uptake than HL-60S cells; coincubation with VER (10 mmol/L) increased uptake 2.6-fold restoring it to 100% of uptake in HL-60S cells. IDR uptake increased 1.5-fold in the presence of VER, but ANN was not affected. Coincubation with VER increased DOX retention in HL-60/DOX cells 2.8-fold and IDR retention 1.4-fold; unchanged ANN retention indicated that ANN may overcome p-gp. In the cytotoxicity assay to correlate intracellular anthracycline content with antitumor activity, we found ANN to be less potent than DOX and IDR In sensitive cells, ID 50 being the drug concentration that inhibits cell growth by 50% but its resistance index (RI; ID50 resistant cells divided by ID50 sensitive cells) was lower than that of IDR and DOX (2.6 v 40 and 117.5). Coincubation in the presence of VER resulted in 4.5-fold and 2-fold RI decreases of DOX and IDR, respectively, whereas ANN did not change, further confirming ANN's ability to circumvent p-gp-mediated MDR. Confocal microscopy studies of IDR, ANN, and DOX showed higher intracellular drug compartmentalization for DOX in HL-60/DOX cells incubated in the presence of VER. This study provided evidence that, unlike DOX and IDR, ANN is not affected by p-gp- mediated MDR.     

Changes in phosphoproteins during commitment of HL60 cells to monocyte differentiation: evidence for multiple protein kinase involvement

The human promyeloid cell line HL60 differentiates toward monocytes when treated with TPA. We have analyzed, by two-dimensional gel electrophoresis, the phosphoprotein patterns within HL60 cells, labeled to equilibrium with [32P]orthophosphate when cells were treated with suboptimal (1 nM), optimal (5 and 10 nM), and supraoptimal (40 and 100 nM) concentrations of 12-O-tetradecanylphorbol-13-acetate (TPA) as regards the induction of differentiation. No change was detected in the phosphoprotein pattern at 1 nM TPA, whereas four phosphoproteins showed increased levels of phosphorylation at 5 and 10 nM TPA. When cells were treated with 40 and 100 nM TPA, in total eight and ten proteins, respectively, were phosphorylated, including the above four proteins. Two proteins were dephosphorylated when cells were treated with 40 and 100 nM TPA. A 15-kd protein, phosphorylated when HL60 cells were treated with 5 nM TPA, was observed as an intense spot in autoradiographs of total cellular phosphoproteins of two variant HL60 cell lines that are unable to differentiate toward monocytes and prior to treatment with TPA. In the case of three variant cell lines, which like HL60 differentiate toward monocytes, the phosphoprotein spot was almost absent. Thus, paradoxically, the 15-kd phosphoprotein is affected by TPA although its constitutive level of expression or increased phosphorylation state is inversely related to the potential for monocyte differentiation. This observation, together with the TPA dose-response effects on protein phosphorylation, is discussed in relation to multiple protein kinase involvement.     

Diverse resveratrol sensitization to apoptosis induced by anticancer drugs in sensitive and resistant leukemia cells

Naturally occurring dietary compound resveratrol (RES), possessing chemopreventive and cytostatic properties, has been shown as potent sensitizer for apoptosis induced by a variety of anticancer drugs. Cell cycle analysis in sensitive promyelocytic leukemia HL60 cell line and its multidrug-resistant variant HL60/VCR (P-gp positive) treated with RES resulted in cell cycle arrest in S-phase in both cell variants. Flow cytometry measurements showed diverse activities of RES in combination with anticancer drugs doxorubicin (DOX), cycloheximide (CHX), busulfan (BUS), gemcitabine (GEM) and paclitaxel (PTX), in some cases resulting in apoptosis induction, preferentially at the expense of S-phase. Thus, RES could become a candidate to enhance the efficacy of combination anticancer therapy in a variety of human cancer cells inclusive leukemias.     

Differential response of human acute myeloid leukemia cells to gemtuzumab ozogamicin in vitro: role of Chk1 and Chk2 phosphorylation and caspase 3

Gemtuzumab ozogamicin (GO) is a humanized anti-CD33 antibody conjugated to the anticancer agent calicheamicin, approved for the treatment of CD33+-relapsed acute myeloid leukemia. We have investigated the effects of GO on 4 human myeloid leukemia lines of different French-American-British (FAB) types (KG-1, THP-1, HL-60, and NB-4), observing 3 different types of response. Exposure to GO (10-1000 ng/mL) induced G2 arrest (up to 80% of the cells) followed by apoptosis (45% of the cells) in HL-60 and NB-4 cells. By contrast, in THP-1 cells we observed a strong G2 arrest (up to 75% of the cells) with little apoptosis. Finally, the KG-1 line was completely resistant to the same concentrations of GO. These different responses did not correlate with the levels of expression of either CD33 or multiple-drug resistance proteins, although the higher cyclosporin A (CsA)-inhibitable efflux activity of KG-1 cells may play a role in the resistance of this line to the drug. We could show that Chk1 and Chk2 phosphorylation, but not p53 or p21 expression, correlated with G2 arrest, implicating the ataxia-telangiectasia mutated/ataxia-telangiectasia related (ATM/ATR)-Chk1/Chk2 pathway in the cell cycle response to GO. However, apoptosis was associated with caspase 3 activation. Freshly isolated acute myeloid leukemia (AML) cells showed patterns of response to GO in vitro similar to those observed with the cell lines, including phosphorylation of Chk2 and caspase 3 activation. Our results suggest that the different molecular pathways induced by the drug in vitro may reflect, at least in part, the variable response to GO obtained in vivo.     

Involvement of SHP-1, a phosphotyrosine phosphatase, during myeloid cell differentiation in acute promyelocytic leukemia cell lines

We investigated the modulation of the expression and phosphatase activity of SHP-1 during granulocytic differentiation of human myeloid leukemia cell line, HL60 and t(15;17) positive APL cell line, HT93, in response to all-trans retinoic acid (ATRA) or dimethylsulfoxide (DMSO). ATRA induced differentiation in both cell lines which was associated with marked growth inhibition and S-phase reduction. On the other hand, DMSO induced it only in HL60 without obvious growth inhibition and S-phase reduction. The expression and phosphatase activity of SHP-1 were upregulated only when the 2 cell lines were differentiated to granulocytes. Furthermore, the changes were not dependent on the inducers or the growth inhibition. These findings suggest that SHP-1 is involved in common myeloid differentiation, and that upregulation of SHP-1 is not always related to myeloid cell growth.     

Novel 2'-substituted, 3'-deoxy-phosphatidyl-myo-inositol analogues reduce drug resistance in human leukaemia cell lines with an activated phosphoinositide 3-kinase/Akt pathway

Activation of the phosphoinositide 3-kinase (PI3-K)/Akt signalling pathway has been linked with resistance to chemotherapeutic drugs, and its down-regulation, by means of pharmacological inhibitors of PI3-K, considerably lowers resistance to various types of therapy in cell lines derived from solid tumours. Recently, a new class of Akt inhibitors, referred to as phosphatidylinositol ether lipids (PIAs), have been synthesized. We tested whether two new PIAs could lower the sensitivity threshold to chemotherapeutic drugs of human leukaemia cell lines with an activated PI3-K/Akt network. We used HL60AR (for apoptosis resistant), K562 and U937 cells. The two pharmacological inhibitors, used at 5 micromol/l, down-regulated Akt kinase activity and phosphorylation. Neither of the two chemicals affected the activity of other signalling proteins in the Akt pathway, such as phosphoinositide-dependent protein kinase-1 or PTEN. When employed at 5 micromol/l, the Akt inhibitors markedly reduced the resistance of the leukaemic cell lines to etoposide or cytarabine. Remarkably, a 5 micromol/l concentration of the inhibitors did not negatively affect the survival rate of human cord blood CD34(+) cells. Overall, our results indicate that new selective Akt pharmacological inhibitors might be used in the future for overcoming Akt-mediated resistance to therapeutic treatments of acute leukaemia cells.     

Evaluation of doxorubicin in combination with mafosfamide for in vitro elimination of myeloid and lymphoid tumor cells from human bone marrow.

In an attempt to improve in vitro pharmacological purging of autologous grafts, the ability of doxorubicin (DOX), alone and in combination with mafosfamide (AZ), to eliminate tumor cells from human bone marrow was assessed. HL60 and Raji cells were mixed with a 20-fold excess of normal marrow cells and were incubated either with DOX (0.4-3.2 micrograms/ml) for 1 h or AZ (20-140 micrograms/ml) for 30 min or both drugs sequentially. Cytotoxicity was evaluated on tumor cells and GM-CFU by clonogenic assays and on earlier hemopoietic progenitors by liquid long-term marrow cultures (LTMC) for 5 weeks. DOX at 3.2 micrograms/ml and AZ at 140 micrograms/ml spared 1.08 and 1.23% of GM-CFU respectively, and yielded similar tumor cell log-kills for HL60 cells (3.04 and 2.95) and Raji cells (3.24 and 3.40). With the combination of AZ and DOX, the best therapeutic index was observed when the cells were incubated with AZ prior to DOX. Under these conditions, AZ at 80 micrograms/ml together with DOX at 1.6 micrograms/ml significantly increased log-kill values for HL60 cells to 3.96 by a synergistic effect and for Raji cells to 3.85 by an additive effect. In LTMC, GM-CFU recovery after treatment with AZ alone and with the combination of AZ and DOX was 59.9 and 20.0%, respectively, while it was 7.9 and 2.9% at culture initiation. These results suggest that the purging efficiency of DOX is comparable to that of AZ and may be enhanced by combination with AZ.     

In vitro chemoresistance profile and expression/function of MDR associated proteins in resistant cell lines derived from CCRF-CEM,K562, A549 and MDA MB 231 parental cells

Although cellular experiments have elucidated a number of active principles in the study of the multidrug resistance (MDR) phenomena, most of the drug resistant tumor cells were derived from different parental cell lines. This fact limits generalization of some experimental data and conclusions, and therefore we selected and characterized cell lines resistant to various anti-cancer agents derived from four parental cell lines: CEM (human T-lymphoblastic leukemia), K562 (human myeloid leukemia), A549 (human lung adenocarcinoma) and MDAMB 231 (human breast adenocarcinoma). In total we obtained a set of 42 resistant sublines, which is an excellent tool for the future studies of different aspects of MDR. In this study we report on some basic characteristics of these sublines, namely, cross-resistance to other anti-cancer drugs investigated by in vitro MTT assay, expression of MDR associated proteins (Pgp, MRP1, LRP, GST-pi and Topo IIalpha) as well as the functional activity of Pgp and MRP.     

Modulation of etoposide (VP-16) cytotoxicity by verapamil or cyclosporine in multidrug-resistant human leukemic cell lines and normal bone marrow

We studied the effects of two modulators of multidrug resistance (MDR), cyclosporine and verapamil, on the cytotoxicity of etoposide (VP-16) in normal human bone marrow; two human leukemia cell lines, K562 and CEM; their MDR variants, K562/DOX and CEM/VLB; and mixtures of normal marrow and leukemic cells. VP-16 was selectivity toxic to the parental leukemic cells, with IC-50 values of 2 microM for CEM cells, 1.5 microM for K562 cells, and 12 microM for normal marrow CFU-GM. This selectivity was lost in the MDR variant leukemia cells, with IC-50s of 20 microM in K562/DOX and 8 microMs in CEM/VLB. Cyclosporine, 6 microMs, and verapamil, 20 microM, alone were nontoxic to bone marrow CFU-GM, and did not significantly increase the toxicity of VP-16 to normal marrow cells or to the two drug-sensitive leukemic cell lines. However, cyclosporine specifically enhanced the cytotoxicity of VP-16 in the MDR leukemia cells, reducing the IC-50 to the same level as the parental sensitive cells. Verapamil was considerably less effective. In a mixing experiment that included K562/DOX cells and normal bone marrow, cyclosporine increased the toxicity of VP-16 to the resistant leukemic cells by nearly 20-fold. Because the cytotoxic effect of cyclosporine is additive for resistant tumor cells, its combination with VP-16 may be useful in the purging of contaminating tumor cells prior to autologous bone marrow transplantation.     

CD74基因表达对急性髓系白血病细胞增殖与侵袭的抑制作用

目的:探讨白细胞分化抗原74（CD74）基因表达对急性髓系白血病（AML）细胞增殖、侵袭及巨噬细胞移动抑制因子（MIF）/细胞外信号调节激酶（ERK）信号通路的影响。方法:收集83例AML患者的骨髓样本（AML组）、25例正常供者的骨髓样本（对照组）及人AML细胞系HL60、MOLM-13、THP-1、U-937、人原代正常骨髓单个核细胞（BMMC）为研究对象，Western blot检测AML骨髓样本及各细胞系中CD74蛋白表达；利用LipofectamineTM2000转染试剂盒对MOLM-13细胞进行转染，并分为空白组（细胞不进行转染）、si-NC组（将si-NC转染于细胞）、si-CD74组（将si-CD74转染于细胞）；Western blot检测各组细胞中CD74、基质金属蛋白酶9（MMP-9）及MIF/ERK信号通路相关蛋白MIF、磷酸化细胞外信号调控蛋白激酶（p-ERK）、ERK表达水平；CCK-8法检测各组MOLM-13细胞在0、24、48、72h的增殖情况；Transwell实验检测各组MOLM-13细胞侵袭情况。结果:与对照组比较，AML组骨髓细胞中CD74蛋白表达水平显著升高（P<0.05），CD74蛋白在HL60、MOLM-13、THP-1、U-937细胞系中均高表达，且在MOLM-13细胞系中表达量最高。与空白组和si-NC组比较，si-CD74组MOLM-13细胞的OD450值（在24、48、72h时）、细胞侵袭数目、CD74、MMP-9、MIF和p-ERK蛋白表达显著降低（P均<0.05），但空白组与si-NC组比较, 差异无统计学意义（P均＞0.05）。结论:干扰AML细胞中CD74表达，可抑制细胞增殖与侵袭，其作用机制可能与抑制MIF/ERK信号通路有关。     

Inhibition of serine/threonine phosphatase PP2A enhances cancer chemotherapy by blocking DNA damage induced defense mechanisms

A variety of mechanisms maintain the integrity of the genome in the face of cell stress. Cancer cell response to chemotherapeutic and radiation-induced DNA damage is mediated by multiple defense mechanisms including polo-like kinase 1 (Plk-1), protein kinase B (Akt-1), and/or p53 pathways leading to either apoptosis or cell cycle arrest. Subsequently, a subpopulation of arrested viable cancer cells may remain and recur despite aggressive and repetitive therapy. Here, we show that modulation (activation of Akt-1 and Plk-1 and repression of p53) of these pathways simultaneously results in paradoxical enhancement of the effectiveness of cytotoxic chemotherapy. We demonstrate that a small molecule inhibitor, LB-1.2, of protein phosphatase 2A (PP2A) activates Plk-1 and Akt-1 and decreases p53 abundance in tumor cells. Combined with temozolomide (TMZ; a DNA-methylating chemotherapeutic drug), LB-1.2 causes complete regression of glioblastoma multiforme (GBM) xenografts without recurrence in 50% of animals (up to 28 weeks) and complete inhibition of growth of neuroblastoma (NB) xenografts. Treatment with either drug alone results in only short-term inhibition/regression with all xenografts resuming rapid growth. Combined with another widely used anticancer drug, Doxorubicin (DOX, a DNA intercalating agent), LB-1.2 also causes marked GBM xenograft regression, whereas DOX alone only slows growth. Inhibition of PP2A by LB-1.2 blocks cell-cycle arrest and increases progression of cell cycle in the presence of TMZ or DOX. Pharmacologic inhibition of PP2A may be a general method for enhancing the effectiveness of cancer treatments that damage DNA or disrupt components of cell replication.     

Elimination of drug-resistant myeloma tumor cell lines by monoclonal anti-P-glycoprotein antibody and rabbit complement

The effectiveness of ex vivo chemotherapy with drugs, such as vincristine, etoposide, and Adriamycin (doxorubicin, Adria Labs, Columbus, OH) for elimination of residual tumor cells from human bone marrow grafts could be undermined by the presence of multidrug- resistant tumor cells in the bone marrow. Therefore, to supplement chemoseparation, we investigated whether MRK-16, a monoclonal antibody (MoAb) to the surface moiety of multidrug resistance-associated P- glycoprotein antigen, can eliminate drug-resistant tumor cells in the presence of rabbit complement (RC). Two doxorubicin (DOX)-resistant human myeloma tumor cell line, 8226/DOX40 (resistant to 4 x 10(-7) mol/L DOX) and 8226/DOX6 (6 x 10(-8) mol/L DOX) with high and low amounts of cell surface P-glycoprotein, respectively, and the drug- sensitive parent cell line 8226/S were used as tumor models in this study. Using the limiting dilution assay, we have shown that three cycles of treatment with 25 micrograms/mL of MRK-16 MoAb and a 1:4 final dilution of RC eliminated 2.90 +/- 0.10 logs of 8226/DOX40 cells and 1.94 +/- 0.18 logs of 8226/DOX6 cells. One and two cycles of treatment were less effective, eliminating 0.47 +/- 0.40 and 1.94 +/- 0.36 logs of 8226/DOX40 and 0.12 +/- 0.20 and 1.63 +/- 0.58 logs of 8226/DOX6 cells, respectively. The 8226/S cell growth was unaffected by one to three cycles of treatment. The cell kill was not impaired when the antibody plus complement treatment was carried out on a mixture of 8226/DOX40 or 8226/DOX6 cells with a ninefold excess of irradiated bone marrow mononuclear cells (MNCs). The three cycles of treatment with antibody plus complement did not adversely affect granulocyte- macrophage colony-forming unit (GM-CFU) survival in hematologically normal marrows (92.5% to 104% survival) or in myeloma patient marrows (85% to 100%). These results show that it is possible to eliminate drug- resistant myeloma tumor cell lines from the admixed human bone marrow by treatment with MRK-16 MoAb plus RC. This method could prove to be effective for elimination of other drug-resistant tumor cell lines including those of leukemia and solid tumors, and will be further useful for supplementing chemopurging, and immunopurging of bone marrow with other antitumor cell antibodies.     

Differential expression of the amv gene in human hematopoietic cells.

Total cellular RNAs from a variety of fresh and culture-derived human hematopoietic neoplastic cell types at various stages of differentiation and human sarcoma, carcinoma, melanoma, and glioblastoma cell lines were enriched for poly(A)- containing sequences, fractionated by gel electrophoresis, and blot hybridized to a cloned DNA probe containing the transforming sequences (v-amv) of avian myeloblastosis virus (AMV), a virus known to cause myeloid leukemias in chickens. Expression of RNA sequences homologous to AMV was detected in all immature myeloid and lymphoid T cells in addition to the single erythroid cell line examined, but not in mature T cells or in B cells, including lymphoblast cell lines derived from patients with Burkitt lymphoma. In addition, induction of the cell line HL60, a promyelocytic leukemia line, to differentiate with dimethyl sulfoxide or retinoic acid resulted in a reduction of the level of expression of the human cellular gene c-amv homologous to v-amv. There was no detectable c-amv mRNA in any of the solid tumor cell lines examined. Thus, expression of the human c-amv gene could be correlated with the stage of differentiation of different hematopoietic cell types determined by morphologic and marker studies. Expression of c-amv could not be correlated with the extent of methylation in HL60 and in HL60 induced to differentiate with dimethyl sulfoxide.     

Effect of arsenic trioxide on viability, proliferation, and apoptosis in human megakaryocytic leukemia cell lines.

Arsenic trioxide (As2O3) has been demonstrated to be effective for the treatment of acute promyelocytic leukemia (APL) and to inhibit proliferation and produce apoptosis in the APL cell line NB4. To determine if As2O3 might be useful for the treatment of other lineages, we investigated the effects of As2O3 on viability, proliferation, and induction of apoptosis in the megakaryocytic leukemia cell lines HEL, Meg-01, UT7, and M07e. Our results showed that As2O3, at concentrations of 0.1-2.0 microM, causes a dose- and time-dependent inhibition of survival and growth in all four megakaryocytic leukemia cell lines studied. In contrast, As2O3 at similar concentrations had no effects on either viability or growth of the nonmegakaryocytic leukemia cell line HL60 and two human breast cancer cell lines, ZR75 and MCF7. In situ end-labeling of DNA fragments (TUNEL assay) indicated that As2O3, at concentrations of 0.5-2 microM, could significantly induce apoptosis in the aforementioned four megakaryocytic leukemia cell lines, but not in the nonmegakaryocytic HL60, ZR75, and MCF7 cell lines. These results were confirmed using conventional morphologic assessment and the DNA ladder assay. Induction of apoptosis in arsenic-treated Meg-01 and UT7 cells was accompanied by a dose-response decrease of Bcl-2 protein, whereas As2O3 had no effect on this measurement in HL60, ZR75, and MCF7 cell lines. Pertinently, these concentrations of As2O3 produced identical changes in the characteristics of the APL cell line NB4. Collectively, these data demonstrate that As2O3 can selectively inhibit growth and induce apoptosis in megakaryocytic leukemia cell lines. The use of As2O3 for the treatment of malignant megakaryocytic disorders should be considered.     

MDR1 gene overexpression confers resistance to imatinib mesylate in leukemia cell line models

Inappropriate expression of the multidrug resistance (MDR1) gene encoding the P-glycoprotein (Pgp) has been frequently implicated in resistance to different chemotherapeutic drugs. We have previously generated chronic myeloid leukemia (CML) cell lines resistant to the tyrosine kinase inhibitor imatinib mesylate (STI571), and one line (LAMA84-r) showed overexpression not only of the Bcr-Abl protein but also of Pgp. In the present study, we investigated this phenomenon in other cell lines overexpressing exclusively Pgp. Thus, cells from the K562/DOX line, described as resistant to doxorubicin due to MDR1 gene overexpression, grew continuously in the presence of 1 microM imatinib, but died in 4 to 5 days if the Pgp pump modulators verapamil or PSC833 were added to the imatinib-treated culture. Analysis of cell proliferation by the MTS (3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium) assay confirmed the differential sensitivity of K562/DOX to imatinib, which was also reversed by verapamil or PSC833. Flow cytometric analysis of the total phosphotyrosine content by intracytoplasmic staining after a 2-hour incubation with escalating doses of imatinib showed that the inhibitory concentrations of 50% (IC(50)) for inhibition of cellular protein tyrosine phosphorylation were 15, 10, and 5 microM for K562/DOX, K562/DOX plus verapamil, and K562, respectively. Retroviral-mediated transfection of the BCR-ABL(+) AR230 cell line with the MDR1 gene decreased its sensitivity to imatinib, an effect that was also reversed by verapamil. The possible role of MDR overexpression in clinical resistance to imatinib remains to be defined. We therefore confirm that imatinib should be added to the extensive list of drugs that can be affected by the MDR phenomenon.     

Synergy between transforming growth factor-beta and tumor necrosis factor-alpha in the induction of monocytic differentiation of human leukemic cell lines

We examined the effect of transforming growth factor-beta (TGF-beta) alone and in combinations with other factors on the growth and differentiation of the human promyelocytic cell line HL60 and the human monoblastic cell line U937. Treatment with TGF-beta alone did not significantly affect growth or differentiation of HL60 cells, while it significantly inhibited proliferation and induced monocytic differentiation of a small percentage of U937 cells. Combinations of TGF-beta and tumor necrosis factor-alpha (TNF-alpha) acted in synergy to inhibit cell proliferation and to induce monocytic differentiation of both HL60 and U937 cells. In contrast, no synergy was observed when HL60 cells were treated with TGF-beta in various combinations with interferon-alpha (IFN-alpha), interferon-gamma (IFN-gamma), and retinoic acid. Examination of TNF-alpha receptor expression on HL60 and U937 cells showed that these cell lines expressed comparable levels of high-affinity TNF-alpha binding sites. Treatment of HL60 and U937 cells with TGF-beta did not induce significant changes in TNF-alpha receptor expression in either cell line. In contrast, HL60 cells expressed much lower levels of TGF-beta receptors than did U937 cells. Treatment of both HL60 and U937 cells with TNF-alpha induced a dose-dependent increase in expression of TGF-beta receptors, suggesting that the synergy between TNF-alpha and TGF-beta may result, at least in part, from upregulation of TGF-beta receptor expression by TNF-alpha.