The leukemia cell lines, HL-60, HEL, NB4, U937 and THP-1, and lentivirus packaging cell line (293T cells) were purchased from the Shanghai Institute for Biological Sciences, Chinese Academy of Sciences, Shanghai, China. The 293T cells were grown in DMEM containing 10% fetal bovine serum (FBS) and the other cells were maintained in RPMI-1640 medium supplemented with 10% FBS. Cells were incubated at 37°C in humidified 5% CO₂ in air and those in the logarithmic growth phase were used for further experiments. The diagnosis of primary leukemia in 100 patients was confirmed by bone marrow biopsy and flow cytometry. After obtaining informed consent, 5 ml of heparin anti-coagulated bone marrow were collected and mononuclear cells were separated with Ficoll, and were then cultured under regular conditions or stored at −80°C for further use. In all samples, the proportion of leukemia cells was >80% and trypan blue excluding fraction was >98%. For the control group, peripheral venous blood was obtained from 15 healthy volunteer donors who also provided written informed consent before the samples were obtained.

Lentiviral vector for RNA interference (Shanghai Innovation Biotechnology Co., Ltd., Shanghai, China), RPMI-1640 medium, DMEM (Gibco, Carlsbad, CA, USA), FBS (HyClone, Logan, UT, USA), TRIzol for RNA extraction (Invitrogen, Carlsbad, CA, USA), the reverse transcription kit (Promega, Madison, WI, USA), the SYBR-Green Real-time PCR kit (Takara, Shiga, Japan), the cell counting kit-8 (CCK-8; Dojindo Laboratories, Kumamoto, Japan), the Annexin V-PE/7-AAD apoptosis detection kit (Nanjing KeyGen Biotech Co., Ltd., Nanjing, China), cell cycle staining solution (MultiSciences Biotech Co., Ltd., Shanghai, China), VEGFR-1 monoclonal antibody (Abcam, Cambridge, MA, USA), secondary antibodies (R&D Systems, Minneapolis, MN, USA), cytarabine (Pharmacia, Kalamazoo, MI, USA) and bevacizumab (Genentech, South San Francisco, CA, USA) were used in the present study.

The leukemia cells were first analyzed for VEGF and VEGFR expression by RT-PCR. Total cellular RNA was extracted, and cDNA was synthesized following conventional protocols. PCR was performed using a PCR thermal cycler (GeneAmp PCR System 9700; Life Technologies, Carslbad, CA, USA). The primers used were synthesized as follows: VEGF forward, 5′-AGGGAAGAGGAGGAGATGAG-3′ and reverse, 5′-GCTGGGTTTGTCGGTGTT-3′ (anticipated size, 148 bp); VEGFR-1 forward, 5′-AGACTAGATAGCGTCA CCAGC-3′ and reverse, 5′-GAAACCGTCAGAATCCTCC-3′ (anticipated size, 102 bp); VEGFR-2 forward, 5′-CCAATAAT CAGAGTGGCAGTG-3′ and reverse, 5′-CATAGACAT AAATGACCGAGGC-3′ (anticipated size, 162 bp). The cycling conditions for PCR were 38 cycles of denaturation (94°C for 20 sec), annealing (58 or 60°C for 20 sec) and extension (72°C for 30 sec). A pre-heating step at 94°C for 5 min and a final extension step consisting of 7 min at 72°C were also carried out. The primer sequences for β-actin were: forward, 5′-CCTGTACGCCAACACAGTGC-3′ and reverse, 5′-ATACTCCTGCTTGCTGATCC-3′. The temperature for annealing was 56°C and a total of 30 cycles were conducted. The other conditions were as mentioned above. The experiment was performed 3 times.

Fluorescence quantitative PCR instrument (LightCycler 1.5, Roche, Basel, Switzerland) and the SYBR-Green Master Mix kit were used for the detection of the expression of target genes and β-actin was used as an internal reference. The primers of VEGF and VEGFRs were as mentioned above. The temperature for annealing was 58 or 60°C and a total of 38 cycles were conducted. β-actin served as an internal reference and the 2^−ΔΔCT method was employed to determine the relative expression of the target genes and the experiment was repeated 3 times.

Cells were harvested and incubated with lysis buffer for protein extraction followed by the determination of protein concentration. The proteins were separated by SDS-PAGE and transferred onto polyvinylidene fluoride (PVDF) membranes. The membranes were incubated with primary antibodies (VEGFR-1, 1:500; β-actin/GAPDH, 1:5,000) followed by secondary antibodies conjugated to horseradish peroxidase. Color development was performed with the ECL kit. The experiment was performed 3 times.

The leukemia cell lines (1×10⁶) were grown in 1 ml of serum-free RPMI-1640 medium at 37°C in humidified 5% CO₂ in air for 48 h. Following centrifugation, the supernatant was obtained and the VEGF content was determined by ELISA. Each experiment was performed in duplicate according to the manufacturer’s instructions. The experiment was repeated 3 times.

According to the principles for the design of the RNA interference sequence, 3 sequences targeting the VEGFR-1 gene (NM_002019) were designed: -sh1: 704–723 bp, 5′-CACACCACGCCCAGTCAAA-3′;-sh2: 3221–3240 bp, 5′-AATCTACAGCACCAAGAGC-3′; and -sh3: 2639–2658 bp, 5′-AATCTTGACCCACATTGGC-3′. The shRNAs targeting the VEGFR-1 gene were then connected to the lentiviral vector, pRNAT-U6.2/Lenti, yielding recombinant vectors. Following PCR and sequencing, the recombinant vectors together with lentiviral packaging plasmids (at a ratio of 1:1:1) was transfected into the 293T cells using Lipofectamine 2000. The viral solution was prepared and named sh-VEGFR-1 KD. In addition, the siRNA sequence unrelated to VEGFR-1 (5′-TTCTCCGAAC GTGTCACGT-3′) was synthesized and served as the negative control (shRNA-NC). Lentiviral particles were conjugated to green fluorescent protein (GFP). The serial dilution method and flow cytometry were performed to determine the lentivirus titer. The lentivirus in the sh-VEGFR-1 KD and shRNA-NC group was used to transfect the U937 cells at the appropriate titer and named sh-KD-U937 (KD group) and sh-NC-U937 (NC group), respectively. In addition, untreated U937 cells were used as the controls (CON group). The proportion of GFP-positive cells was observed under a fluorescence microscope 72 h later. When the proportion was >90%, the cells were used for the following experiments. These cells were harvested at 120 h after being transfected, real-time PCR and western blot analysis were performed to determine the efficiency of RNA interference at the mRNA level and protein level. The experiment was repeated 3 times.

Twenty-four hours after transfection, 3 groups of cells (5×10³) in the logarithmic growth phase were seeded in 96-well plates (100 μl/well) and each experiment was performed in sextuplicate followed by incubation for 7 days. The medium was refreshed every 48 h. The cells were then incubated with CCK-8 solution (5 mg/ml, 10 μl/well) for 4 h followed by blending. The absorbance (A) was determined at 450 nm with a microplate reader, which represents the cell proliferation rate. The experiment was repeated 3 times.

Migration assays were performed in Transwell plates (Costar, Cambridge, MA, USA) of 6.5 mm in diameter, with 5-μm pore filters, as previously described by Fragoso et al(5). In brief, cell lines or different groups of U937 cells (1×10⁶ cells/ml) were placed in serum-free medium with or without bevacizumab (1 μg/ml) for 1 h. Cell aliquots (100 μl) were subsequently added to the upper compartments, and 600-μl of serum-free medium with or without VEGF (50 ng/ml) were added to the lower compartments. The transwell plates were incubated at 37°C, 5% CO₂ for 4 h. Cell migration was determined after 4 h by counting the number of migrated cells in 6 high-power fields (x400 magnification) under an optical microscope.

Seventy-two hours after transfection, 3 groups of U937 cells in the logarithmic growth phase were harvested and seeded in 96-well plates at a density of 1×10⁵ cells/ml (100 μl/well). These cells were then treated with cytarabine or bevacizumab (10 μl) at various concentrations in RPMI-1640 medium containing 10% FBS. For each concentration, the experiment was performed in sextuplicate. Additionally, a normal control and blank group were also prepared. After 48 h of incubation, cells were incubated with CCK-8 solution (10 μl) for 4 h followed by blending (5 min). The absorbance (A) was determined at 450 nm with a microplate reader and the cell proliferation and inhibition rate were calculated as follows: proliferation rate/inhibition rate (%) = [(A_control − A_experiment)/A_control − A_blank)] ×100%. The experiment was repeated 3 times.

After 48 h of treatment with different drugs, 3 groups of U937 cells were collected followed by washing with PBS twice, then resuspended in PBS (1X) at a density of 1×10⁶ cells/ml. Subsequently, 1 μl of Annexin V-PE and 5 μl of 7-AAD were added followed by blending. A blank group, Annexin V-PE group and 7-AAD group were also prepared. Cells were incubated for 15 min followed by flow cytometry for the apoptotic rate.

Three groups of U937 cells were seeded in 24-well plates at a density of 1×10⁶ cells/ml and maintained in medium containing different drugs for 48 h. Cells (5×10⁵) were then collected and washed with PBS (4°C) twice. The cells were then resuspended in 1 ml of PBS and added to 4 ml of cold absolute ethanol, which was blended and fixed at −20°C for 15 min followed by centrifugation. The supernatant was removed and 5 ml of PBS were added to the sediment followed by incubation for 15 min. Centrifugation was then performed and the supernatant was removed. The sediment was stained with DNA staining solution [propidium iodide (PI), 50 mg/l and RNase A, 10 mg/ml] in the dark for 30 min. In addition, a blank group and PI group were also prepared. Flow cytometry was performed within 1 h.

Statistical analysis was performed using SPSS 14.0 statistical software, and the t-test was used for comparisons between two means. The χ² test was conducted for comparisons between apoptotic rates. A value of P<0.05 was considered to indicate a statistically significant difference.

The expression of VEGF, VEGFR-1 and VEGFR-2 was detected by RT-PCR in 5 AML cell lines. As shown in Table I, VEGF and VEGFR-1 were expressed in all 5 AML cell lines, whereas only HEL cells expressed VEGFR-2. We also detected the VEGF protein levels in the cell supernatants by ELISA (Table I). The U937 cell line had the highest VEGF protein level and the THP-1 cell line the second highest (P<0.05 for both). Furthermore, quantitative PCR and western blot analysis were used to detect VEGFR-1 expression in the 5 AML cell lines. As shown in Fig. 1A and B, the expression of VEGFR-1 was the highest in the U937 cell line and the lowest in the NB4 cell line.

In order to determine whether the migratory behavior of the leukemia cells is associated with the high expression of VEGF/VEGFR-1, VEGF-induced cell migration was performed in transwell migration assays with the 5 AML cell lines. As shown in Fig. 1C, the U937 cells showed the most potent chemotactic response to VEGF (with a 2.35-fold ratio of VEGF-migrated cells/control; P<0.05), the THP-1 cell line showed the second most potent chemotactic response to VEGF (with a 1.75-fold ratio of VEGF-migrated cells/control; P<0.05), with the remaining 3 cell lines showing comparable migration capacity.

As regards the primary leukemia samples, we detected the mRNA expression of VEGF, VEGFR-1 and VEGFR-2 in 100 patients with AML. VEGF was expressed in 92% of the AML samples, and VEGFR-1 and VEGFR-2 were expressed in 87 and 37% of the samples, respectively. The patients were divided into 2 groups according to the French-American-British (FAB) classification system of acute leukemia (Table II). The quantification of VEGF and VEGFR-1 mRNA levels demonstrated significantly higher levels in the M4/M5 vs. Ml, M2 and M3 FAB subtypes (P<0.05 for both). In addition, the significantly higher expression of VEGF and VEGFR-1 mRNA was observed in patients with extramedullary infiltration vs. those with no extramedullary infiltration, whereas VEGFR-2 levels did not significantly differ between the 2 groups.

Given the high expression level of VEGF and VEGFR-1 in the U937 cells and the M4 and M5 subtypes of AML samples characteristic of a higher migration rate or extramedullary infiltration, we hypothesized that VEGF-VEGFR-1 may play significant roles in the regulation of leukemia cell migration and other biological characteristics. In the following experiments, the U937 cells were used as target cells, and lentivirus-mediated RNA interference was performed to suppress the VEGFR-1 expression in the U937 cell line.

The shRNA-NC and sh-VEGFR-1 KD lentiviral vectors were used to transfect the U937 cells at a multiplicity of infection (MOI) of 10. Seventy-two hours after transfection, the transfection efficiency of the U937 cells was >90%, as observed under an inverted fluorescence microscope. One hundred and twenty hours after transfection, we collected the cells and used quantitative PCR to detect VEGFR-1 mRNA expression. VEGFR-1 mRNA expression decreased by 73.53±0.026% in the KD group when compared with the NC group (Fig. 2A). Western blot analysis showed that the VEGFR-1 protein expression in the KD group decreased by 94.14% (P<0.05) when compared with the NC group (Fig. 2B). No significant difference in the mRNA and protein expression of VEGFR-1 was observed between the CON and NC group.

Furthermore, we performed ELISA to demonstrate that there was no significant difference in VEGF production in the U937 cell supernatants in the CON (1.69±0.17 ng/ml) and NC group (1.65±0.04 ng/ml), while VEGF production in the U937 cell supernatants in the KD group (0.41±0.06 ng/ml) decreased dramatically by >75% (P<0.05) when compared to the CON and NC group (Fig. 2C). These results indicated that lentivirus-mediated RNA interference suppressed the autocrine loop of VEGF/VEGFR-1 in the transfected U937 cells.

These results demonstrate the feasibility of suppressing VEGFR-1 expression by lentivirus-mediated RNA interference, which inhibits the mRNA and protein expression of VEGFR-1 and blocks the VEGF/VEGFR-1 autocrine loop in U937 cells.

To determine the effects of VEGFR-1 gene silencing on the biological characteristics of U937 cells, a series of in vitro experiments were conducted, including cell proliferation and migration assays, cell cycle analysis and determination of the apoptotic rate.

Cells from 3 different groups were grown for 7 consecutive days and the growth curve was delineated according to the proliferation inhibition rate. As shown in Fig. 3A, the proliferation rate of the U937 cells in the KD group was markedly slower than that of the cells in the NC and CON group after 3 days of culture (P<0.05), and this tendency continued for at least 7 days.

Seventy two hours after transfection, 3 groups of U937 cells were seeded in transwell plates for cell migration assay. As shown in Fig. 3B, before drug treatment, the number of migrated cells in the KD group was markedly lower than that in the NC and CON group (with a 2.5-fold ratio of VEGF-migrated cells/control; P<0.05). Following treatment with VEGF, the number of migrated cells in the NC and CON group was markedly increased, while this phemomenon was not observed in the KD group.

The apoptotic rate of the U937 cells in the KD group was slightly higher than that in the CON and NC group (Fig. 4C). No significant difference in cell cycle distribution was observed among the 3 groups of U937 cells by cell cycle analysis prior to drug treatment (Fig. 5).

Following treatment with VEGF (Fig. 5), the proportion of cells in the S + G₂ phase and G₁ phase increased and decreased, respectively, in the CON and NC group. However, in the KD group, no marked changes in the proportion of cells in the S + G₂ phase or G₁ phase were observed following treatment with VEGF. These results demonstrate that VEGF enhances the colony-forming ability of the U937 cells by increasing the proportion of cells in the S + G₂ phase. However, after VEGFR-1 gene silencing, these effects conferred by VEGF were blocked, accompanied by a suppressed colony-forming ability. These findings suggest that VEGFR-1 plays a pivotal role in the proliferation of leukemia cells.

As demonstrated above, VEGFR-1 gene silencing profoundly inhibits the in vitro proliferation and migration of the U937 cells, but does not directly increase apoptotic rate of the U937 cells. In the following experiments, VEGFR-1 gene silencing in combination with bevacizumab or cytarabine were used to explore their synergistic effects on the inhibition of cell proliferation and apoptotic effects. As shown in Fig. 4A, bevacizumab at various concentrations did not profoundly affect the proliferation of U937 cells in the CON, NC and KD group. Flow cytometry also showed that the apoptotic rate in the 3 groups of U937 cells was not affected by bevacizumab (data not shown). These findings showed that in vitro VEGFR-1 gene silencing in combination with bevacizumab did not confer suppressive effects on the proliferation and did not increase the apoptosis of U937 cells. Following treatment with various concentrations of cytarabine, a higher number of shrunken and ruptured cells was observed in the KD group. As indicated by the growth curve (Fig. 4B), the inhibition rate increased with the increase in the cytarabine concentration in each group. When the cytarabine concentration reached ≥25 ng/ml, the inhibition rate in the KD group dramatically increased compared to the CON and NC group (P<0.05). The IC₅₀ of cytarabine in the U937 cells in the KD, CON and NC group was 50, 200 and 200 ng/ml, respectively. These results revealed that the sensitivity of U937 cells to cytarabine increased by VEGFR-1 gene silencing.

The apoptotic rate in the different groups was detected with a flow cytometer following treatment with cytarabine. As shown in Fig. 4C, the apoptotic rate in the KD group was higher than that in the CON and NC group. When the cytarabine concentration reached 25 or 50 ng/ml, a significant difference in the apoptotic rate was observed between the KD group and the other 2 groups (CON and NC group) (P<0.05 for both). Furthermore, the apoptotic rate increased with the increase in the cytarabine concentration.

To elucidate the mechanisms underlying the inhibitory effects of VEGFR-1 gene silencing in combination with cytarabine on the proliferation and increased apoptosis of U937 cells, we further examined the cell cycle distribution in the 3 groups of U937 cells following treatment with cytarabine (50 ng/m). As shown in Fig. 5, the proportion of G₁ phase cells in the KD group increased from 47 to 62.2%, and that in the S + G₂ phase decreased from 53 to 37.8%. However, the proportion of cells in the G₁ phase increased from 46.4 to 51.7% and from 44.6 to 51.9% in the CON and NC group, respectively, and that in the S + G₂ phase decreased from 53.6 to 48.3% and from 55.4 to 48.1%, respectively. These results suggested that, after VEGFR-1 gene silencing, cytarabine treatment markedly decreased the number of cells in the S + G₂ phase and that these cells were more sensitive to cytarabine when compared with those in the CON and NC group.

We further analyzed the expression of anti-apoptotic genes in the 3 groups of U937 cells following treatment with cytarabine. As shown in Fig. 6, following treatment with cytarabine (50 ng/ml), the expression of survivin in the KD group was dramatically lower than that in the CON and NC group. These findings suggest that VEGFR-1 gene silencing downregulates the expression of anti-apoptotic genes.