Soluble Flt-1 Gene Delivery in Acute Myeloid Leukemic Cells Mediating a Nonviral Gene Carrier

Vascular endothelial growth factor (VEGF) is a potent angiogenic factor involved in angiogenesis-mediated progression of acute myeloid leukemia (AML). Studies have reported the role of soluble form of fms-like tyrosine kinase (sFlT-1) delivery as an antitumor agent by inhibiting VEGF. This study investigates the outcome of delivery of a VEGF165 antagonist, soluble vascular endothelial growth factor receptor, namely sFLT-1, mediating lipofectamine 2000 in acute myeloid leukemic cells. A recombinant plasmid expressing sFLT-1 was constructed and transfected into the K562 and HL60 cells using lipofectamine 2000 transfection reagent. sFLT-1 expression/secretion in pVAX-sFLT-1 transfected cells was verified by RT-PCR and western blot. MTS assay was carried out to evaluate the effect of sFLT-1 on human umbilical vein endothelial cells and K562 and HL60 cells in vitro. Treatment with pVAX-sFLT-1 showed no association between sFLT-1 and proliferation of infected K562 and HL60 cells, while it demonstrated a significant inhibitory impact on the proliferation of HUVECs. The results of the current study imply that the combination of nonviral gene carrier and sFLT-1 possesses the potential to provide efficient tool for the antiangiogenic gene therapy of AML.


Introduction
Acute myeloid leukemia (AML) is a clonal disorder associated with acquired genetic defects that occur in early hematopoietic precursor cells. AML is categorized by rapid growth of clonal abnormal myeloid cells which fail to differentiate into functional granulocyte or monocyte and accumulate in the blood, bone marrow, or other organs [1,2]. Angiogenesis is a complex series of interdependent procedures, controlled by many different factors. e angiogenic process results from a shi in the balance of positive and negative mediators, cytokines, and growth factors [3][4][5].
Angiogenesis has been postulated to play a role in pathogenesis of AML. Research in pathogenesis of AML has identi�ed vascular endothelial growth factor (VEGF) to be among the most speci�c and essential regulators of the process of angiogenesis [6,7]. VEGF regulates angiogenesis by modulating processes such as endothelial proliferation, permeability, and survival. Recent studies show that acute myeloid cells express a rich presence of VEGF as well as its receptors which are implicated in facilitation progression and survival of acute myeloid leukemic progenitors [8,9]. Studies show that VEGF-mediated autocrine and paracrine signals are involved in in�ltration of leukemic cells in bone marrow, liver, and spleen as well as increased microvascular density in bone marrow of AML patients [10][11][12]. In addition, evidence brought forth shows a linear correlation of the expression of VEGF and its receptors by AML blasts with reported low rates remission and overall survival among AML patients [12,13].
e application of gene therapy in conjunction with other treatment modalities provides a promising novel approach in management of numerous hematological malignancies as it provides a means of circumventing setbacks suffered by current treatment modalities. e dual approach may result in improving the response by patients to already available conventional treatment [14][15][16].
Endogenous inhibitors of VEGF hold a lot of promise on approach in regulation of angiogenesis, a major factor implicated in growth, development, and survival of many hematological malignancies including AML. Soluble form of fms-like tyrosine kinase (sFLT-1), are produced endogenously by alternative splicing of the FLT-1 receptor. sFLT-1 binds with very high affinity to VEGF without initiation of signal transduction and as such results in site-speci�c inhibition of the function of VEGF leading to regulation of angiogenic process in acute myeloid cells [17,18].
Gene therapy-based sFLT-1 in antiangiogenic strategies has been studied in various types of cancers using viral and nonviral carriers. However, few studies have elaborated the role of sFLT-1 in hematologic malignancies. In the current study, we investigated the effects of sFLT-1 gene delivery using a cationic lipid on acute myeloid leukemic cells. We constructed the sFLT-1 gene that codes the 1-3 immunoglobulin-(Ig-) like domains of FLT-1 and established a recombinant plasmid expressing sFLT-1. Up to our knowledge, there is no study related to the delivery of psFLT-1 in leukemic cells by mediating lipofectamine 2000. erefore, for the �rst time, we investigated the delivery of pVAX-sFLT-1 in acute myeloid leukemic cells and explored the in vitro inhibitory effects of sFLT-1in K562 and HL60 cells as well as HUVECs.

Cell
Lines. HL-60 (human promyelocytic leukemic cells) and K562 (myelogenous leukemia, erythroleukemia type) were obtained from American Type Culture Collection (ATCC, Manassas, VA, USA) and were maintained in complete medium composed of RPMI supplemented with 10% fetal bovine serum (Hyclone Labratories, Austria). Human umbilical vein endothelial cells (HUVECs) were obtained from ATCC and maintained in F12K supplemented with endothelial cell growth supplement (ECGS) (0.05 mg/mL) (BD Biosciences, Bedford, MA, USA) and heparin (Sigma-Aldrich, Australia) (0.1 mg/mL). e media for cells were supplemented with 1% penicillin, streptomycin (Sigma-Aldrich, St Louis, MO, USA). All of the cells were grown at 37 ∘ C in a humidi�ed atmosphere of 5% CO 2 .

Cloning of the 1-3 Ig-Like Domains of FLT-1.
Total cellular RNA was isolated from HUVECs with MasterPure RNA Puri�cation Kit (Epicenter Biotechnology, USA) according to the manufacturer's instructions. cDNA (Fermentas, Lithuania) was ampli�ed using the primers designed to code to the 1-3 Ig-like domains of FLT-1 (accession no. X51602) [19]. e forward and reverse primers were modi�ed to allow ligation into the pVAX1 plasmid at the HindIII-ECORI restriction sites as follows: forward 5 ′ CC AAG CTT ACC ATG GTC AGC TA-3 ′ and reverse 5 ′ -CG GAA TTC TTA ATA TGC ACT GAG-3 ′ . e PCR reaction was run at 94 ∘ C for 4 min; 35 cycles at 94 ∘ C for 1 min; 67.4 ∘ C (VEGF165), 55 ∘ C (VEGFR-1, and GAPDH) for 50 s; and 72 ∘ C for 50 s following an extension for 10 min at 72 ∘ C. e resulting PCR product (1000 bp) was cloned into pVAX1 plasmid (Invitrogen, San Diego, USA), following digestion, ligation, and transformation of top10 E. coli cells, positive colonies were chosen by restriction digestion. Positive colonies were then subjected to DNA sequencing analysis (NHK Bio-Science, Korea) for �nal veri�cation. subsequently the transfection medium was replaced with complete growth media and incubated for 72 h. To con�rm the expression of sFLT-1 by transfected cells, cellular RNA was isolated from the K562 and HL60 cells, cDNA was synthesized from total RNA, and then subjected to RT-PCR using the cloning primers described above. RT-PCR was performed using the BIO-X-ACT Short Mix PCR reagent (Bioline USA Inc, Taunton, MA, USA) with the same condition described earlier.

Expression of VEGF165 by Leukemic Cells.
Expression of VEGF165 and VEGFR-1 mRNA in leukemic cells was identi�ed using RT-PCR analysis. e expression of GAPDH mRNA was evaluated to verify equal mRNA levels. e presence of two fragments, 567 bp (VEGF165) and 145 bp (VEGFR-1), was identi�ed in agarose gel electrophoresis. (Figures 1 and 2).

Expression of sFLT-1. Transfection of K562 and HL60
cells with pVAX-sFLT-1 and pVAX-GFP by lipofectamine resulted in the expression/secretion of sFLT-1. e cell lysate and conditioned media of transfected cells were subjected to RT-PCR and western blot analysis, respectively. As the Figure  5 shown, sFLT-1mRNA at size of 1000 bp was ampli�ed from cDNA of pVAX-sFLT-1 transfected cell but not pVAX-GFP. e western blot analysis demonstrated that sFLT-1 was expressed and secret in conditioned media from pVAX-sFLT-1 at size of 37 KDa ( Figure 6). e transgene expression by recombinant plasmid was evaluated in time points 48 and 72 h posttransfection using a recombinant plasmid expressing GFP, but the results were not shown.

In Vitro Inhibitory Effect of sFLT-1 in Proliferation of K562
and HL60 Cells. Inhibitory effect of sFLT-1 expressing from pVAX-sFLT-1 transfected K562 and HL60 cells was evaluated using MTS proliferation analysis. As shown in Figure 7, sFLT-1 had no inhibitory effect on leukemic cells growth in the presence of VEGF aer 72 h as the results were compared with pVAX-GFP transfected or untransfected cells ( 0.05).

In Vitro Inhibitory Effect of sFLT-1 in Proliferation of
HUVEC. e functional con�rmation of secreted sFLT-1 mediating pVAX-sFLT-1 was evaluated by HUVEC inhibition analysis using MTS assay. As shown in Figure  8, incubation of HUVECs with conditioned media K562 cells expressing sFLT-1 aer 72 h inhibited proliferation of HUVECs by 40% when compared with HUVECs incubated with conditioned media from pVAX-GFP and untransfected cells, respectively.

In Vitro Inhibitory Effect of sFLT-1 on Migration of HUVEC.
HUVECs migration was evaluated aer treatment of cells with conditioned media of pVAX-sFLT-1 transfected. HUVECs were incubated with conditioned media from pVAX-sFLT-1, pVAX-GFP, and untransfected cells and applied to migration assay using QCM 24-well Colorimetric Cell Migration kit in presence of VEGF (40 ng/mL). As shown in Figure 9, conditioned media from sFLT-1 expressing cells inhibited migration of HUVECs by 43% compared with conditioned media from pVAX-GFP and untransfected cells, respectively.

Discussion
Several approaches were investigated for the effectiveness in delivering sFLT-1 gene therapy, including naked plasmid DNA [20], HVJ-cationic liposome-formulated plasmid DNA, and adenoviral-mediated approaches [21]. Most of these studies have evaluated delivery of sFLT-1 in solid tumours, including a peritoneal metastasis model using human �brosarcoma cells [21], ovarian cancer [22], follicular thyroid carcinoma [19], and multiple myeloma [23]. ese studies indicated that sFLT-1 expression caused a signi�cant inhibition in endothelial cell proliferation in vitro and the suppression of tumour growth and metastasis, as well as longer survival period. Also, it is determined that the systemic delivery of the sFLT-1 mediated by mammalian cells could be effective to inhibit tumour growth and angiogenesis [19]. ese studies were focused on solid tumours. Only one study was performed on haematologic malignancies, which is multiple myeloma (MM), demonstrating that the systemic delivery of ADV-sFLT effectively inhibited MM growth through decreasing the microvessel density of tumours. It is worth mentioning that the authors of that study found that the biological effect of sFLT-1 in multiple myeloma was mainly through inhibiting the angiogenesis of the tumour without affecting its growth. ese �ndings present the effect of sFLT-1 on VEGF-mediated paracrine loop [23]. Most of studies have delivered sFLT-1 using viral vectors; however, due to immune response concerns related to viral vectors, using nonviral gene carriers, which are not immunogenic, can be effective approach for delivery of such inhibitors. erefore, we have developed an approach to inhibit angiogenesis through inhibition of VEGF with its natural inhibitor, sFLT-1. To this end, one of the most widely used cationic gene carriers, Lipofectamine 2000, was used for delivering therapeutic gene into angiogenic endothelial cells. e human sFLT-1 molecule constructed in the current study contained the extracellular domains 1-3 of the FLT-1 receptor and was able to effectively block angiogenesis and tumor growth. Barleon  between the affinity of full-length extracellular domains and the 1-3 extracellular domains of sFLT-1 mutants in binding VEGF, and the �rst three Ig-like loops are able to compete efficiently with VEGF165 [24]. To our knowledge, to date, there is no study related to delivery of psFLT-1 in leukemic cells mediating lipofectamine 2000. Here, we investigated inhibitory effect of sFLT-1 myeloid leukemic cells and human umbilical endothelial cells. e results demonstrated that although sFLT-1 did not inhibit cell proliferation of the leukemic cells, it could signi�cantly suppress the proliferation and migration of HUVEC aer incubation with conditioned media K562 cells expressing sFLT-1. Interestingly, the same results were observed in the study by Liu et al. ey determined that the proliferation of human multiple myeloma cells (KM3) transduced with ADV-sFLT was not inhibited in the presence or absence of VEGF, while the conditioned media of transduced cells could signi�cantly inhibit the growth of HUVECs. Moreover, they found that ADV-sFLT-1 exerted its inhibitory effect on human MM tumours mainly through inhibiting the angiogenesis of the tumour and decreasing the microvessel density of tumours, without affecting the tumour cell proliferation directly [23]. Also, Kim et al. showed that PEI-g-PEG-1.3RGD/pCMV-sFLT-1 delivery in colon adenocarcinoma cells had no inhibitory effect on the growth of colon adenocarcinoma cells, whereas the transfection of primary endothelial cells (CADMEC) with PEIg-PEG-1.3RGD/pCMV-sFLT-1 resulted in the suppression of VEGF-driven proliferation of endothelial cells. erefore, they exhibited signi�cant selectivity and effectiveness of PEIg-PEG-RGD/pCMV-sFLT-1 complexes in an endothelial cell inhibition assay by abrogating VEGF effects through the binding sFLT-1 [25]. On the other hand, other studies which focused on the sFLT-1 gene delivery in solid tumours reported only the inhibitory effect of sFLT-1 on the proliferation of endothelial cells in vitro, while the inhibitory effect of sFLT-1 on the VEGF-mediated proliferation of tumour cells in vitro was not studied [19,22].
Studies showed that expression of VEGEF and its receptors on myeloid leukemic cells contributed to the growth and survival of leukemic blasts through the autocrine loop [7]. Also, the role of the VEGF-mediated loop in the progression of myeloid leukemic cells was investigated by the application of an antisense-VEGF in K562 cells. e silencing of VEGF by an antisense-VEGF in K562 cells inhibited proliferation of K562 cell through increasing cell apoptosis [26]. To the contrary, in the current study, although the expression of VEGF and VEGFR-1 in K562 and HL60 cells was identi�ed, inhibition of the VEGF-mediated autocrine loop was not observed by sFLT-1 in these cells. Consequently, these results are in agreement with previous studies that propose VEGFmediated inhibitory effect of sFLT-1 trough paracrine loop. is means that sFLT-1 is probably involved in the survival and progression of leukemic blasts by its inhibitory effects on marrow endothelial cells (paracrine loop). is statement was concluded in the study by Liu et al. on multiple myeloma. To explain this behaviour, it is mentioned that the solubility of sFLT-1 probably allows it to act in a paracrine manner on vascular endothelial cell receptors [23]. Taken together, more investigations may be needed to explain and understand the effect of sFLT-1 gene delivery on VEGF-derived proliferation and the survival of leukemic blasts.

Conclusion
In agreement with other studies, no link between the proliferation of K562 and HL60 cells and the inhibitory effect of sFLT-1was detected. However, it was found that sFLT-1 expression from K562 cells can inhibit signi�cantly the VEGF-driven proliferation and migration of HUVECs in vitro. We suggest that although sFLT-1 could not affect the proliferation of K562 and HL60 cells in a VEGF-mediated autocrine loop, it may play an inhibitory role by affecting VEGF-mediating paracrine loop.