| This work was supported by the Key Programs of Science and Technology of Guangzhou City (No.2006Z3-E0401) and the National Natural Science Foundation of China (No.30973421).Background and objectivesThe identification of the Philadelphia chromosome in cells from individuals with chronic myelogenous leukemia (CML) led to the recognition that the Bcr-Abl tyrosine kinase and P210protein caused CML. This in turn led to the development of imatinib mesylate, a clinically successful inhibitor of the BCR-ABL kinase. Imatinib (IM) was currently standard therapy for patients with early chronic phase CML resulting in an improved5-year survival rate and a decrease in the rate of transformation to blast crisis. Although most patients with chronic phase CML treated with imatinib had well-controlled disease, some patients relapse and/or progress to accelerated phase or blast crisis. Furthermore, most patients with advanced phases of CML initially responded to imatinib but subsequently relapse. Therefore, the mechanisms of drug resistance emerged as a major question.The most common mechanism of IM resistance was point mutations in the kinase domain of Abl that precluded the binding of IM. Other mechanisms had also been reported to contribute to IM resistance including Bcr-Abl gene amplification, Bcr-Abl protein overexpression, decreased intracellular IM concentration, and activation of other protein tyrosine kinases (PTKs).The Eph receptors were the largest family of receptor tyrosine kinases. Well-known effects of Eph receptor activation involved regulation of cell shape and movement. Although the mechanism of IM resistance in CML had been intensively investigated, few studies had investigated the role of Eph receptors in IM resistance in CML. EphB4, a member of Eph family, played an important role in tumorigenesis and regulated diverse cell functions including cell-cell contact, cell adhension, migration and repulsion. These effects required EphB4signaling and activation of the RhoA. Recent studies showed overexpression of EphB4was associated with drug resistance in Ph+acute lymphocytic leukemia (ALL). In the study, we focused on its role and mechanism in resistance to IM in CML.The Ras signaling pathway played an essential role in cancer cell growth. Ras activation led to the phosphorylation and activation of downstream signaling proteins, such as MEK and ERK. Activation of the MAPK pathway was suggested as an explanation for the incomplete effect of IM on Ph+ALL. Then, in order to determine the mechanism of EphB4receptor contributing to IM resistance in CML, we detected a serious of phosphorylation of Ras/MAPK signaling proteins.The small GTPases of the Rho family (RhoA, Rac1and Cdc42) was known regulators of the actin cytoskeleton. Cdc42and Racl were central regulators of cell motility, which contributed to Bcr-Abl-activated proliferation. In2003, Ohmine et al suggested the disturbance of RhoA protein kinase signaling, was involved in imatinib resistance. Our preliminary study (National Natural Science Foundation of China,30271463) showed RhoA was low-expression in K562-W, but was high-expression in K562-R resistance to IM by gene microarray technology. It meant RhoA played an important role in resistance to IM. In the study, we examined the RhoA-related protein activity in regulating the expression of EphB4.Homoharringtonine (HHT) was a plant alkaloid, which for the past30years had been widely used for the treatment of acute myelogenous leukemia (AML) and CML. In2008, Tong et al suggested that HHT acted as a wide spectrum tyrosine kinase inhibitor could block many associated cell signal pathways and may play an important part role in the target therapy of AML with abnormal activation of tyrosine kinase, thus supplying theoretical basis for the wider clinical use of HHT. In2011, we first also represented the multicenter study to compare the antitumor efficacy and safety profile of HHT as induction and post-induction therapy in elderly patients over60years old with newly diagnosed primary or secondary AML. We suggested that high-dose HHT as postinduction therapy not only prolonged the overall survival, but also had reliable drug safety for elderly patients with AML. In2010, Fang et al reported that the HHT+IM regimen reinduced hematologic responses or improved the cytogenetic responses in advanced CML patients with disease-poor response to conventional doses of imatinib.The antileukemia mechanism of HHT appeared to involve both the inhibition of protein synthesis and accordingly attained therapeutic effect in AML. However it was not still determined the mechanism of HHT treating CML with advanced/resistence to IM. At present, there was no standard therapy for blast crisis/IM resistance of CML. In2004, the members of our study group found that HHT enhanced the expression of DJ-1in K562cell by proteomics and spectrometry identification. The online string software analyzed that DJ-1down regulated the EphB4expression. It meant that HHT maybe decrease the EphB4in K562cell.In our own pre-experiment stage, the detection showed that EphB4was high-expression in K562-R cell. Moreover, the stable under-expressing EphB4cells (K562-R-EphB4-sh) were obtained (Oligonucleotides for shRNA were designed using ABI online software, Lentiviruses were generated and were cotransfected in293T cells, then directly infected K562-R cells). In the study, our purpose was to address the role of EphB4in IM resistance of CML. At the same time, it was also investigated that the mechanism of EphB4contributed to IM resistance in K562cells and xenograft. At last, we further detected whether HHT contributed to block the EphB4-mediated IM resistance pathway.Methods1. Cell cultureK562-W (K562-wild type) cells were cultured in RPMI-1640medium. For the K562IM-resistant cell line (K562-R), the above-mentioned media were further supplemented with5.45mg/L IM, as described previously. All cell lines maintained in our laboratory were harvested in the exponential growth phase.2. MTT assayK562-W, K562-R and K562-R-EphB4-sh cells were plated into96-well flat-bottomed microtiter plates at5×104cells/well in100μl media. After24hours of preincubation, gradient concentration of IM (0-8mg/L) were added.The inhibition ratio was assayed. Control cells treated with an equal concentration of DMSO (100ug/well) were included in each experiment. Untreated cells were also included as controls. The inhibition rate was quantified as [1-(ODtreated-ODblank)/(ODcontrol-ODblank]×100%. The proliferative inhibition rates of K562-R treated by IM+HHT were also analyzed by MTT assay.3. Xenograft modelBALB/C female nude mice (4-5weeks of age; SLRC laboratory animal center, China) were used in xenograft experiments. K562-W, K562-R and K562-R-EphB4-sh cells (107) were injected subcutaneously on the dorsal surface of female nude mice. The tumour volume was calculated as width2×length×1/2, as measured with a digital calliper.4. Semi-quantitative PCR analysisAll cell lines used in experiments were harvested in the exponential growth phase. Total RNA was isolated with TRIzol reagent (Invitrogen, USA). Semi-quantitative PCR (SQ-PCR) was performed on ABI9700system. Primers for gene amplification were designed with Primer5.0and sequences are as follows: EphB4-upstream primer:5’—CTCCTTCCTGCGGCTAA—3’; EphB4-downstream primer:5’—AGACGAGGTTGCTGTTGACT—3.5. Western Blot analysisTotal protein was isolated by RIPA lysis buffer and protein concentrations were determined by the BCA method. Proteins were separated by SDS-PAGE and transferred onto nitrocellulose membranes. After blocking in5%bovine serum albumin (BSA) for2hour, membranes were incubated overnight in primary antibody diluted in5%BSA at4℃. The following primary antisera were used:anti-EphB4(R&D, USA), anti-phospho-Racl+Cdc42(cst, USA), anti-phospho-RhoA (cst, USA), anti-phospho-ERK1/2(cst, USA), anti-phospho-MEK1/2(cst, USA), and anti-β-actin (Santa Cruz Biotechnology, USA). After washing, membranes were incubated for1hour with HRP-conjugated goat anti-mouse (1/5000) or with cow anti-goat (or rabbit) immunoglobulin (1/5000)(both from Santa Cruz Biotechnology, USA), diluted in5%BSA. After washing, the enhanced chemilumines-cence kit (B&M, USA) was used to visualize the secondary antibody.6. PE Annexin V Apoptosis DetectionThe apoptosis rate of K562-R cell (HHT, IM or HHT+IM incubated) was detected with PE Annexin V in a buffer containing7-Amino-Actinomycin (7-AAD) and analyzed by flow cytometry.7. Statistical analysisStatistical analysis was performed with SPSS13.0statistical software. Values represented the mean of3independent experiments and were expressed as mean±SD. The comparison of xenograft volumes, expressions of phosphorylated protein and mRNA were evaluated by ANOVA analysis. Paired t-test was used in the comparison of phosphoralated proteins of cell and xenogaft tissues by IM or HHT intervention. P value<0.05was considered statistically significant.Results1. The low-expression of EphB4in K562-R-EphB4-sh cell SQ-PCR and Western blot was used to demonstrate decreased expression of the EphB4mRNA and protein in knockdown. The EphB4receptor was overexpressed in IM-resistant K562-R cell line comparing with K562-W cell line. But the expression of EphB4mRNA significant decreased in knockdown K562-R-EphB4-sh cell comparing with K562-R (multiple comparisons, P<0.001) and K562-W (multiple comparisons, P<0.001) cell lines. Meanwhile, the expression level of EphB4also significantly decreased in K562-R-EphB4-sh cell lines (multiple comparisons, P<0.001) than K562-R (multiple comparisons, P<0.001) and K562-W (multiple comparisons, P<0.001) cell lines. This meant K562-R-EphB4-sh cell line was established well.2. K562-R-EphB4-sh cell was sensitivity to IMMTT was used to analyze the tolerance dose of IM in K562-R-EphB4-sh cells. K562-R-EphB4-sh was sensitive to IM (IC500.93mg/L) and K562-R showed IM resistance (IC505.45mg/L)(P<0.001). These suggested K562-R-EphB4-sh cell was sensitive to IM when the expression of EphB4was down regulated. However, they were still less sensitive than K562-W cells (IC500.16mg/L, multiple comparisons, P <0.001).3. Establishing the K562xenograft modelsOn day30after implantation, tumours reached an average size of over1000mm3. There were no differences (F=0.249, P=0.782) in average volume among K562-W (1588.78±210.89mm3), K562-R (1631.13±454.57mm3) and K562-R-EphB4-sh (1710.60±266.41mm3) xenografts. Further the xenograft models were determined by histopathology.4. K562-R-EphB4-sh xenograft tumors was sensitivity to IMXenograft models using K562-W, K562-R and K562-R-EphB4-sh cells were generated to compare tumor volume and overall survival.3groups (N=6/group) of nude mice all received oral standard-dose IM (91ug/g of body weight daily for30days) treatment after xenograft models established successfully. K562-W tumor volumes significantly decreased with IM treatment (806.15±202.31mm3versus1586.73±230.94mm3, t=12.539, P<0.001). K562-R-EphB4-sh tumor volumes did not change significantly (1630.16±412.01mm3versus1720.06±290.54mm3, t=0.922, P=0.399). K562-R tumor volumes continued to increase after receiving IM treatment (2301.25±555.76mm3versus1733.82±399.22mm3mm3, t=-6.452, P=0.001). Meanwhile, all nude mice with K562-R xenograft tumor are died.2and4nude mice were dead in the K562-W and K562-R-EphB4-sh xenograft groups during the period of treatment (30days). Overall, the K562-R-EphB4-sh xenograft group had good efficacy for receiving oral IM treatment comparing with K562-R xenograft nude mice receiving the same dose of IM.5. Phosphorylation of molecules involved in K562-R-EphB4-sh cell and xenograft tissueWe examined the phosphorylation of molecules (MEK/ERK, Racl/Cdc42/RhoA) in K562-W, K562-R and K562-R-EphB4-sh both cell lines and xenograft tissue. Expression of Ras-GTPase was similar in the three cell lines (F=0.948, P=0.439). Phosphorylation of MEK/ERK was higher in the K562-IM resistant and K562-R-EphB4-sh cells compared to K562-W type cells (multiple comparisons, P<0.001). In contrast, phosphorylation of Rac1+Cdc42and RhoA was significantly increased in K562-R cells compared to K562-R-EphB4-sh and K562-W cell lines (multiple comparisons, P<0.001).Analysis of xenograft tissue demonstrated that the parallel results with the highest EphB4receptor expression in K562-R xenograft tissue, and the lowest in K562-R-EphB4-sh xenograft tissue (multiple comparisons, P<0.001). MEK/ERK phosphorylation was also higher in K562-R-EphB4-sh and K562-R xenograft tissue compared to K562-W xenograft tissue (multiple comparisons, P<0.001). Phosphorylation of Rac1+Cdc42and RhoA was only highest in K562-R xenograft tissue (multiple comparisons, P<0.001).6. HHT enhanced K562-R cell to the IM sensitivityThe result of PE Annexin V Apoptosis Detection showed the apoptosis rate of K562-R was only2.35±0.11%, which were significantly lower than (multiple comparisons, P=0.001) K562-W (71.82±1.94%) when they co-cultured with IM (1.2mg/L). But the apoptosis rate reached58.71±2.39%with K562-R incubated with HHT (20ug/L)+IM, which was higher to K562-R incubated with IM (multiple comparisons, P=0.002). MTT assay showed that IC50of K562-R treated by IM was5.45mg/L). But under the stimulation of HHT, IC50of K562-R decreased from5.45to1.17mg/L (t=15.271,P<0.001).7. It was more effective for K562-R xenograph nude mice received HHT+IM regimen3groups of K562-R (N=6/group) xenograph nude mice respectively received the HHT, IM and HHT+IM regimens. K562-R tumor volumes significantly decreased with IM+HHT treatment comparing with before treatment (1692.82±317.14mm3versus975.56±132.42mm3, t=8.637, P<0.001). On the contrary, K562-R tumor volumes significantly increased with simple IM treatment comparing with before treatment (1733.82±399.22mm3versus2301.25±555.76mm3, t=-6.452, P=0.001).6,5and1nude mice were dead in IM, HHT and IM+HHT arms respectively (within4weeks). As a result, the HHT plus IM regimen had better efficacy comparing with simple HHT or IM treatment.8. HHT blocked the EphB4/Racl/Cdc42/RhoA pathwayWe examined the expressions of EphB4/Racl/Cdc42/RhoA and MEK/ERK pathway protein. The EphB4, phosphorylation of Racl+Cdc42and RhoA was significantly decreased after HHT incubate in K562-R cells (t=49.730, P<0.001and t=23.507, P=0.002and t=86.554, P<0.001), but there were no differences of MEK/ERK (t=1.296, P=0.324and t=-1.147, P=0.370). Differing from HHT, the phosphorylation levels of EphB4(t=1.644, P=0.242), Racl+Cdc42(t=2.203, P=0.158) and RhoA (t=-3.538, P=0.071) were still high after doxorubicin stimulation. Analysis of xenograft tissue demonstrated that the parallel results. The expressions of EphB4/Racl/Cdc42/RhoA began to decrease after HHT treatment.Conclusions1. Xenograft models of K562-W, K562-R and K562-R-EphB4-sh were established well using BALB/C female nude mice.2. EphB4was over-expressed in IM-resistant K562-R cell. However, the stable under-expressing EphB4cell of K562-R-EphB4-sh was established and its sensitivity to IM was recovered definitely. It was also effective for oral IM in K562-R-EphB4-sh xenograft nude mice, but not reached the treatment response in K562-R xenograft nude mice. We first presented here a new important marker of IM resistance-EphB4.3. We found that MEK/ERK activity was not change with down-regulated EphB4. There existed high expressions of the phosphorylation of MEK/ERK between K562-R and K562-R-EphB4-sh cell/xenograft tissue. The phosphorylations of RhoA and Racl+Cdc42were simultaneously decreased in K562-R-EphB4-sh cell/xenograft tissue comparing with K562-R cell/xenograft tissue. Therefore, we first found a new pathway about IM resistance in myeloid leukemia: the activity of Rac1/Cdc42/RhoA was determined by the overexpression of EphB4.4. HHT significantly enhanced the apoptosis rate and proliferative inhibition of K562-R cells treated by IM. HHT+IM therapy reduced the volume of K562-R xenograft and prolonged the survival of nude mice, comparing with simple IM or HHT. It was more effective for HHT+IM regimen in the treatment response of K562-R xenograft nude mice.5. The results showed that the EphB4/Racl/Cdc42/RhoA was simultaneously decreased after HHT intervention in K562-R cell and xenograft tissue. The EphB4/Rac1/Cdc42/RhoA was blocked by HHT treatment, but doxorubicin did not affect the activity of pathway. As a result, we can make a theoretical explain why the efficacy of HHT+IM regimen gained more treatment profits than simple HHT or IM treating myeloid leukemia. |
PDF Full Text Request