Dihydroartemisinin Induces Apoptosis Of Multiple Myeloma RPMI8226 Cells Via Downregulating VEGF Expression And Suppressing VEGF Secretion Under Hypoxia

Multiple myeloma (MM) is characterized by the accumulation of monoclonal malignant plasma cells in the bone marrow (BM). MM is the second most frequent hematological malignancy and remains incurable to date with a median survival rate of approximately 3~4 years, even with aggressive, high-dose chemotherapy, bone marrow transplantation, and intensive supportive care. The pathophysiology of MM is complex, involving many pathways and interactions among cytokines, adhesion molecules, angiogenesis, and mechanisms of resistance. Besides the long-lived neoplastic cells with a low proliferative index, it has clearly been established that intracellular regulatory proteins (such as IL-6, VEGF and IGF-1) and interactions between malignant plasma cells and the BM microenvironment play an important role in their survival and drug resistance. Vascular endothelial growth factor (VEGF), as a principle promote factor of angiogenesis, plays a pivotal role in MM biology. In BM microenvironment, VEGF is expressed by both MM cells and BMSCs. Studies found VEGF receptor 1 (VEGFR-1) to be more commonly expressed than VEGFR-2 which usually expressed on endothelial cells. In MM, VEGFR-1 is widely expressed on both MM cell lines and patient MM cells. Data confirmed that in addition to stimulating angiogenesis, VEGF directly stimulates MM cell migration on fibronectin, proliferation, and survival via autocrine and paracrine loops and therefore confers resistance to conventional chemotherapy. Binding of VEGF to MM cellsthrough tyrosine-kinase receptors, in particular VEGFR-1 triggers important downstream signaling pathways: a PI3-kinase/protein kinase Ca (PKC0)-dependent cascade mediating MM cell migration on fibronectin;MEK-extracellular-signal-regulated protein kinase (ERK) pathway mediating cell proliferation, as well as survival signaling through upregulation of Mel-1 and survivin. These direct effects of VEGF on MM cells and BM angiogenesis suggest potiential use of novel therapies targeting VEGF. Furthermore, hypoxia is a key regulator of VEGF expression mediated via hypoxia-inducible factor-1 (HIF-1) and is an important factor of drug resistance of MM cells in the BM microvironment Dysregulation of VEGF expression and its signaling pathways therefore plays an important role in the pathogenesis and clinical features of hematologic malignancies, in particular multiple myeloma Direct and indirect targeting of VEGF and its signaling pathways under hypoxia may provide a potent novel therapeutic approach to overcome resistance to therapies and thereby improve patient outcome.Artemisinin, a sesquiterpene lactone isolated from the traditional Chinese herb Artemisia anmia Linn, is a representative of a new effective class of antimalarial drugs. Dihydroartemisinin (DHA) is the main active metabolite of artemisinin, and is more water-soluble and more effective in treating malaria than artemisinin. Our previous studies revealed that DHA possess antiangiogenic activity in vitro. Recently, we also showed that DHA inhibits the VEGF expression in solid tumor xenografts and exhibits the potent antiangiogenic effect in solid tumors in vivo. DHA also downregulates VEGF expression and induces apoptosis in chronic myeloid leukemia K562 cells. To our knowledge, the effect of DHA on VEGF expression and apoptosis in other hematological malignancies and in particular in MM under low oxygen tension, however, remains elusive.In present study, we investigated the growth inhibition effect by trypan blue exclusion and MTT method, and apoptotic inductive effect of DHA on human multiple myeloma RPMI8226 cells under 3% O2 condition by AO/EB dual staining, transmission electronic microscopy analysis, agarose gel eletrophoresis, flow cytometry assay in vitro. The effect of DHA on VEGF expression of RPMI8226 cells under the hypoxia was assayed by RT-PCR and western blot analysis, and VEGF secretion was determined by ELISA assay. We also evaluated the effect of DHA on expression and activation of signaling transduction protein: ERK1/2, Mcl-l, caspase-3 and PARP under hypoxia by western blot and measured the disruption of mitochondrial transmembrane potential (A*Fm) on flow cytometry after JC-1 staining. Besides, the conditioned medium (CM) of RPMI8226 cells pretreated with DHA wasassessed for its inductive effect on angiogenesis using chorioallantoic membrane (CAM) model.Results1. The growth inhibition effect on RPMI 8226 cells under hypoxiaUnder hypoxic condition, when treated with DHA at concentrations ranging from 0 ~ 40nmol-L"1 for 12, 24, 36, 48 and 72 hrs, human multiple myeloma RPMI 8226 cells growth was effectively inhibited in a concentration-dependent and time-depentent manner by the trypan blue exclusive assay. After incubated with 10 umol-L"1 DHA, the number of viable cells compared with that of control group was decreased to 57.6% (24 hrs), 36.1% (48 hrs) and 26.9% (72 hrs). The values of 50% inhibition concentration (IC50) of DHA for RPMI 8226 cells under hypoxia measured by the MTT assay were 30.24 + 6.71 umol-L"1 and 16.76+5.82 umol-L"1 at 24 hrs and 48 hrs, respectively (Fig. 1-2).2. Inductive effect of apoptosis on RPMI 8226 cells under hypoxia2.1 Cytomorphological changes of apoptosisAfter dually stained with AO/EB, fluorescent assay revealed that untreated RPMI 8226 cells showed uniform, diffused chromatin stained with AO. In contrast, after exposure to 3 ~ 12 umol-L"1 DHA in hypoxia for 48 hrs, cells showed characteristic apoptotic alterations: condensed, marginated chromatin clearly stained with AO;these cells were further classified as either viable (EB -) or inviable (EB +), depending on the integrity of the cytoplasmic membrane. In addition, there were some late apoptotic cells undergoing degradation;these cells were invariably EB (+), with obviously dispersed nuclear chromatin. (Fig. 3)Transmission electronic micrographs revealed that control RPMI 8226 cells had intact cytoplasm membranes, clear cytoplasm, intact subcellular organelles, intact nuclear membrane and normal nucleoli. In contrast, after exposure to 12 umol-L"1 DHA for 48 hrs, cells showed characteristic apoptotic alterations: shrinking cellular figure, decreasing cell surface microvilli, cytoplasmic vacuoles, chromatin condensation and margination, DNA fragmentation and formation of apoptotic bodies (Fig. 4). These morphological changes suggested that DHA induces typical cytomorphological features of apoptosis in RPMI 8226 cells under hypoxia.2.2 Analysis of DNA fragmentationWhen RPMI 8226 cells were treated with DHA in hypoxia for 48 hrs, the appearance of a typical "DNA ladder" was observed on a DNA agarose eletrophoresis gel especially in 6 and 12 umol-L"1 DHA groups. The result indicated endogenous restricted enzymes were activated to cut the genomic DNA into multiples of 180 ~ 200 bp in length, forming the "ladder" by gel electrophoresis (Fig. 5).2.3 Cell apoptotic percentage and distribution of cell cycle phasesAs shown in Fig. 6, there were only two peaks in the control group in DNA conent histogram by flow cytometric analysis. In contrast, the exposure of RPMI 8226 cells to DHA under hypoxia resulted in the occurrence of a sub-G] peak, which was representative of cells undergoing apoptosis. Futhermore, the percentage of apoptotic cells significantly increased with the DHA concentration escalating (4.59% in control group versus 23.28%, in 12 umol-L"1 DHA group, PO.01) (Fig. 6, Table 1). Meanwhile, treatment with DHA significantly decreased cells in S phase, 42.86% versus 14.58%, with corresponding increases of cells in G2/M phase (8.06% versus 30.37%). (Fig. 7) DHA induced growth arrest in a concentration-dependent fashion (3 ~ 12 umol-L"1) and this effect exhibited a correlation with the inhibition of cell growth. These results indicated that DHA-induced growth arrest of RPMI 8226 cells was partially due to cell cycle block, which caused cells to accumulate in the G2/M-phase and partially to apoptotic sub-Gi phase.3. Mechanism of apoptotic induction3.1 Inhibition of activation of extracellular-signal-regulated protein kinasei 12ERKl/2 expression did not change significantly in RPMI 8226 cells with or without DHA incubation under hypoxia. A lower level of phospho-ERK2, however, was observed. In comparison with control, the expression intensity of phospho-ERK2 protein with 3, 6 and 12 umol-L"1 of DHA incubation under hypoxia for 6 hrs was significantly lower by 40.6% (P<0.05), 85.4% and 98.2%, respectively (Fig.8). The results showed that the activation of ERK2 protein in RPMI 8226 cells was rapidly inhibited by DHA treatment. Coupled with the growth inhibition effect of DHA on RPMI 8226 cells, above data implied that ERKl/2 signaling pathway may play an important role in VEGF-mediated MM cells proliferation and MM cells-induced apoptosis.3.2 Disruption of mitochondrial transmembrane potentialThe changes in mitochondrial membrane potential (A*Pm) were examed by measuring the uptakeof the mitochondrial specific fluorescent probe JC-1 in RPMI 8226 cells undergoing apoptosis. As shown in Fig. 9, RPMI 8226 cells showed a collapse in the A^Pm after 48 hrs of treatment with DHA. The collapse increased with the increase of the concentration of DHA from 6 to 12 umol-L"1, exhibiting a concentration-dependent manner.3.3 Downregulation of expression of antiapoptotic Mcl-1As shown in Fig. 10, the downregulation of antiapoptotic Mcl-1L (active form, 40 kDa) expression was observed in RPMI 8226 cells following DHA incubation for 24 hrs under hypoxia, with corresponding increases of Mcl-ls (inactive form, 32 kDa) expression. The results implied that this downregulation was related to its sensitivity to DHA-induced apoptosis via A*Pm collapse.3.4 Activation of caspase-3 and PARPBoth control and 6, 12 umol-L*1 DHA treated groups expressed pro-caspase-3 (inactive form, a 32 kDa band) by western blot analysis. The cleaved form of caspase-3 (active), a 17 kDa band could only, however, be detected after 12 hrs incubation with DHA. Meanwhile, the expression of poly (ADP-ribose) polymerases (PARP, 112 kDa) was also observed to be downregulated by DHA, with corresponding upregulation of cleaved PARP (active form, 85 kDa) in a concentration-dependent (6 -12 umol-L'1) and time-dependent (12 ~ 48 hrs) fashion (Fig. 11). These data indicated that caspase-3 and PARP was activated in apoptotic RPMI 8226 cells during DHA treatment under hypoxia.4. Downregulation of VEGF expression and suppression of VEGF secretion of RPMI 8226 cells under hypoxia4.1 Downregulation of VEGF mRNA Expression in RPMI 8226 cellsRT-PCR analysis showed that two forms of transcripts (VEGF12i and VEGFi65 mRNA) were detected. DHA could significantly reduced VEGF mRNA expression in a dose-dependent manner (3 ~ 12 umol-L"1). Compared with the control group, even at the lower concentration of 3 umol-L"1 DHA, the levels of VEGFi65 and VEGF121 mRNA were decreased by 12.7% and 15.4% (P<0.05,). The results revealed that DHA could downregulate VEGF expression at gene level (Fig. 12).4.2 Downregulation of VEGF protein expression in RPMI8226 cellsAs shown in Fig. 13, DHA was found to lead to a reduction of VEGF expression in RPMI 8226 cells in a dose (3 ~12 umol-L"1) and time (6 ~ 48 hrs) -dependent fashion under hypoxia by westernblot analysis. Noticeably, when incubated in hypoxic condition for 48 hrs, VEGF protein expression was decreased by 24.4+5.5% (3 umol-L'1, P<0.05 ) and 90.7+2.2% (12 umol-L"1 P<0.001) versus the control group. And with the exposure to hypoxia for 6, 12, 24 and 48 hrs, the expression of VEGF in RPMI 8226 cells treated with 12 umol-L'1 DHA was significantly downregulated by 41.0+11.1% (12 hrs, PO.05), 84.1 + 10.8% (24 hrs), 91.5 + 11.8% (48 hrs) as compared with the control group. These results demonstrated that DHA could significantly downregulate VEGF protein expression.4.3 Decrease of VEGF secretion by RPMI 8226 cellsAs Fig. 14 showed, a significant dose-dependent decrease in VEGF secretion by RPMI 8226 cells was induced by DHA treatment [mean ± SD versus control: 495 ± 27 (3 umol-L"1), 218 ± 42 (6 umol-L'1) and 123 ± 14 (12 umol-L"1) vs 587 ± 22 pg/ml per 106 cells]. Compared with control, RPMI 8226 cells pretreated with 3, 6,12 umol-L"1 DHA showed significant lower secretion of VEGF protein by 15.7 ± 4.7% (P<0.05), 62.9 ± 2.0% (PO.001) and 79.0 ± 2.3% (PO.001), respectively.5. Inhibition of MM-induced angiogenesis on CAMAngiogenic potentials of myeloma RPMI 8226 cells after pretreatment with DHA in hypoxia were examed using a modified CAM assay. CM-loaded sponges displayed an intense capillary growth with numerous microvessels converging towards the sponges in 'spoked wheel' pattern on day 12 (Fig. 15b-d). The angiogenic activity was decreased in response to the CM from RPMI 8226 cells pretreated with DHA in a dose-dependent manner. Compared with CM from normal control RPMI 8226 cells, the number of microvessels induced by CM from RPMI 8226 cells pretreated with 3, 6 and Humol-L"1 of DHA was approximately reduced by 28.6% (P<0.05), 41.3% and 61.4%, respectively. In addition, there was no significant difference in the number of microvessels between negative control group (RPMI 1640 medium alone, Fig. 15a) and CM group from 12 umol-L'1 DHA pretreated RPMI 8226 cells (Fig. 15e) (P>0.05). The results demonstrated the efficacy of DHA to inhibit the angiogenic potentials of RPMI 8226 cells.SummaryThe data presented here demonstrated that: (1) Dihydroartemisinin could effectively inhibit the growth of multiple myeloma RPMI 8226 cells under hypoxic condition in a concentration-dependentand time-depentent manner. (2) Dihydroartemisinin at the concentration of 3 ~ 12 ixmol-L'1 could induce apoptosis of RPMI 8226 cells in hypoxia in a concentration-dependent and/or time-depentent manner. (3) Dihydroartemisinin could inhibit activation of ERK2 protein, cause a collapse in the Afm, downregulate expression of VEGF and antiapoptotic Mcl-1, activate caspase-3 and PARP in RPMI 8226 cells, and inhibit secretion of VEGF of RPMI 8226 cells under hypoxia in a concentration-dependent and time-dependent fashion. (4) Dihydroartemisinin could effectively inhibit the angiogenic potential of hypoxic RPMI 8226 cells.The above data demonstrated that the effects of cell growth inhibition and apoptotic induction by dihydroartemisinin are probably associated with the occurrence of downregulation of VEGF expression and suppression of VEGF secretion, which consequently results in inhibition of activation of ERK1/2 signaling pathway, downregulation of antiapoptotic Mcl-1 expression, disruption of mitochondrial membrane potential, activation of executioner caspase-3 and PARP, and induction of apoptosis. Furthermore, inhibition of the angiogenic potential of RPMI 8226 cells by dihydroartemisinin probably resulted from the decrease of VEGF secretion via paracrine mechanism.ConclusionIn conclusion, the present study has demonstrated that dihydroartemisinin possesses the ability of inhibition of proliferation, induction of apoptosis and inhibition of the angiogenic potential on MM RPMI 8226 cells via targeting of VEGF and its signaling pathways under mimic hypoxic BM rnicroenvironment. Coupled with its well toleration and almost no resistence in malaria treatment, dihydroartemisinin may be a promising VEGF-targeted drug in hematological malignant tumors, in particular in MM therapy and thereby overcome resistance and improve patient outcome.