**1. Introduction**

Angiogenesis is the process of the formation of a new capillary network. A crucial role in the angiogenesis regulation belongs to mechanisms associated with endothelial cells, which retain their ability to divide in the adult organism [1]. Angiogenesis includes all phases of new blood vessel growth: proliferation and migration of endothelial cells, the formation of a capillary tube, and the remodulation of vascular network in organs [2,3]. The balanced functioning of this system is very important, since either the excessive vessel formation or their insufficient development leads to serious diseases. For example, intensive angiogenesis contributes to tumor growth where the formation of a branched vascular network in the tumor leads to an increase in their growth and further metastasis [4]. Therefore, approaches for the down-regulation of angiogenesis are used as elements for therapy against these diseases. Also, endothelial cells, being genetically stable, are less likely to develop drug resistance

in comparison with tumor ones [5]. Taken together, antitumor agents targeting endothelial cells are supposed to be more effective than drugs targeting tumor cells. The inhibition of endothelial cell proliferation and migration may lead to the lack of structural support for tumor cells resulting in the disassembly of tumor tissues and can be used for the treatment of cancers and tumor-like diseases [6,7].

A key role in angiogenesis belongs to VEGFA (the vascular endothelial growth factor A). Its level is elevated in tissues with intensive angiogenesis and its receptors are predominantly expressed on the endothelial cells of blood vessels nearby [8,9]. VEGFA signaling has a direct impact on endothelial cells growth in vitro [10]. It prevents apoptosis of endothelial cells in vitro by inducing the expression of anti-apoptotic proteins Bcl-2 and A1 [11]. In newborn mice the suppression of VEGFA gene expression leads to apoptosis of endothelial cells in a large number of blood vessels [12]. Hypoxia is an upregulating factor for VEGFA gene expression and signaling [13,14]. For example, solid tumors under hypoxia can stimulate the production of an increased amount of VEGFA [15]. This creates conditions for the intensive development of vascular network in the growing tumors. VEGFA signaling functions through two receptor tyrosine kinases of similar structure: VEGFR1 and VEGFR2. VEGFR2 is considered to be the primary VEGFA receptor that runs angiogenesis, while VEGFA most strongly binds to the VEGFR1 receptor. VEGFR1 gene knockout mice die on the ninth day of prenatal development from disorganization and excessive growth of blood vessels. This showed that in early embryogenesis, VEGFR1 functions mainly as a decoy receptor that sequesters excess VEGFA [16]. Nevertheless, a positive modulation of angiogenesis by VEGFR-1 has been demonstrated in adults. For example, suppression of VEGFR1 led to defects in neovascularization of the eye [17]. Also, VEGFR1 expression is high in many human cancers [16,18]. VEGFR1 is suggested to serve as an alternative angiogenic pathway in the case when VEGFA is inhibited, acting in conjunction with VEGFB and PlGF ligands [19]. It gives the opportunity for VEGFR1-based inhibition of angiogenesis alternatively to the VEGFA/VEGFR2 pathway. One more alternative and promising target for the inhibition of angiogenesis is endoglin. Its expression is greatly increased in endothelial cells of blood vessels and surrounding tumors [20]. Endoglin (ENG or CD105) is a co-receptor for transforming growth factor-β (TGF-β), which participates in activating a complex signaling pathway and thus mediates the proliferation, migration, and adhesion of endothelial cells [21]. Mice with a fully inactivated endoglin die during prenatal development due to cardiac abnormalities and defects in the formation of the vascular network; their vessels stop growing and do not penetrate into the yolk sac [22]. Homozygous knockout the endoglin gene mutations in humans are also lethal. Heterozygous endoglin mutations cause hereditary hemorrhagic telangiectasia 1, which is characterized by the fragility and instability of small vessels [23]. Endoglin is preferentially expressed in the angiogenic endothelium of solid tumors, and was found to be a marker of activated endothelial cells [24]. Recently it was suggested as a promising target for antivasculogenic therapy [25,26].

A possible way for angiogenesis targeting may be the use of a gene therapy approach where small interfering RNA (siRNA) is introduced into endothelial cells with the aim to specifically inhibit the pro-angiogenic gene expression. The application of siRNA for the specific suppression of endogenous genes in cells was successfully realized by Elbashir et al. [27].

One of the most important barriers to the application of RNAi remains the necessity to create an effective siRNA delivery system. The delivery system has to provide siRNA-targeted delivery into cells, protect it from nucleases degradation, and release siRNA for its activity in the cytosol [28]. To achieve nucleic-acid-targeted delivery to endothelial cells, we and others previously suggested a new ligand-receptor pair SDF1/CXCR4 [29–32]. SDF1 (stromal cell-derived factor-1) is a ligand for the CXCR4 (chemokine receptor type 4) expressed in the endothelium of angiogenic vessels [33]. Moreover, SDF1 plays an important role during neoangiogenesis by being a main recruiter of endothelial progenitor cells [34]. The targeted delivery via CXCR4 was achieved in our previous studies by using modular peptide carriers modified with ligand derived from the N-terminus of SDF1 [31,35–39]. The developed carriers were based on cationic cysteine-flanked cross-linking peptides that can effectively bind and protect DNA and RNA fromnuclease degradation [35–37]. The anti-VEGFA siRNA-peptide polyplexes

demonstrated an efficient inhibition of VEGFA expression in endothelial cells in vitro [37,39].Achieved VEGFA gene silencing by means of RNAi resulted in a significant decrease of VEGFA protein production and the rate of endothelial cell migration. Polyplexes were formed by anti-VEGFA siRNA and the most efficient peptide carrier L1 was tested using an in vivo treatment of endometriosis in a rat subcutaneous model [38]. Significantinhibition of endometriotic implants growth (55–60%) and a two-fold decrease in VEGFA gene expression were demonstrated. Anti-angiogenic effect of the polyplexes also was confirmed via immunohistochemical characterization of the endometriotic implants.

In the present study, we used L1 peptide-based polyplexes bearing siRNA against VEGFA, VEGFR1, and endoglin for the targeted suppression of angiogenesis in endothelial cells. Proliferation and migration of the transfected cells was evaluated. We analyzed the effects of VEGFA, VEGFR1, and endoglin gene silencing, either alone or in combination. Here, the tested hypothesis was that the combinatorial siRNA silencing of several angiogenic pathways may be more efficient than single gene knockdown and could result in synergistic anti-angiogenic effects in endothelial cells.

#### **2. Materials and Methods**

#### *2.1. Cell Lines*

GFP-expressing human breast cancer cell line MDA-MB 231 was kindly provided by Prof. Jessica Rosenholm, Abo Academy University, Turku, Finland. The cell line was maintained under mycoplasma-free conditions as described previously [40].

Endothelial cells EA.Hy926 (hybridoma of primary HUVEC (human umbilical vein endothelial cells) and A549 cells (human lung adenocarcinoma)) were kindly gifted by Dr. Cora-Jean C. Edgell from the University of North Carolina, USA. This cell line reproduces the main morphological, phenotypical and functional features of the endothelium [41]. The EA.Hy926 cells were maintained under mycoplasma-free conditions as described previously [36].

#### *2.2. Peptide Synthesis and Design*

L1 peptide carrier was synthesized using NPF Verta, LLC (SaintPetersburg, Russia), and stored desiccated at −20 ◦C. Before use, the peptide carrier was dissolved in 0.1% TFA at 2 mg/mL. The peptide purity was determined using high-performance liquid chromatography, and found to be in the range of 90–95%. L1 peptide consists of the KPVSLSYRSPSRFFESH motif connected with a DNA-binding sequence (CHRRRRRRHC) via two ε-aminocaproic acids (Ahx) [36].

#### *2.3. siRNA Preparation of Peptide*/*siRNA Complexes*

The sense strand of anti-VEGFA siRNA 5 -GCG GAU CAA ACC UCA CCA Att-3 targets human VEGFA mRNA [42]. The sense strand of anti-VEGFR1 siRNA 5 -GGC CAA GAU UUG CAG AAC Utt-3 targets human VEGFR1 mRNA [43]. The sense strand of anti-endoglin siRNA 5 -CGG UGA CGG UGA AGG UGG AAC UGA G-3 targets human endoglin mRNA [44]. The sense strand of anti-GFP siRNA 5 -CAA GCU GAC CCU GAA GUU Ctt-3 targets GFP mRNA [45]. A non-silencing siRNA 5 -UUC UCC GAA CGU GUC ACG U- 3 served as a mock siRNA [46]. siRNAs were purchased from Syntol JSC, Moscow, Russia. siRNA/peptide complexes were prepared at 8 to 1 and 16 to 1 N/P ratios (peptide nitrogen/RNA phosphorus ratio). All positively charged amino acids were taken into account for the calculation of N/P charge ratios. The appropriate volume of the peptide carrier (2 mg/mL) was added to the siRNA solution (100 μg/mL) in Hepes-buffered mannitol (HBM) (5% *w*/*v* mannitol, 5 mM Hepes, pH 7.5) and vortexed. Then, thepolyplexes were allowed to stand at room temperature for 2 h.

X-tremeGENE liposomal transfection reagent (Roche, Mannheim, Germany) was used as a control siRNA carrier according to the manufacturer recommendations.

#### *2.4. Cytotoxicity Assay*

A total of 0.6 <sup>×</sup> <sup>10</sup><sup>4</sup> MDA-MB-231 and <sup>Е</sup>A.Hy926 cells (at the low density) and 2.5 <sup>×</sup> <sup>10</sup><sup>4</sup> <sup>Е</sup>A.Hy926 cells (at the high density) were seeded in 96-well plates and incubated overnight. The cytotoxicity of peptide/siRNA complexes was evaluated using Alamar blue assay (Invitrogen, Eugene, OR, USA) for cell viability after 16 h of incubation. The fluorescence was recorded on a Wallac 1420D scanning multilabel counter (Thermo Fisher Scientific Oy, Vantaa, Finland) with an excitation wavelength at 544 nm and emission wavelength at 590 nm. The relative fluorescence intensity was counted according to (F−Ff)/(Fb−Ff) × 100%, where Fb is the fluorescence intensity in untreated control and Ff is the fluorescence intensity without cells. The results are presented as mean± S.E.M of the means obtained from three independent experiments with three samples.

#### *2.5. siRNA Transfer to MDA-MB-231 Cells*

Transfection experiments were performed in triplicate. A total of 2.5 <sup>×</sup> 104 cells was seeded in 24-well plates and incubated overnight. Before transfection, the cell culture medium was replaced with medium without FBS.Anti-GFP siRNA and mock siRNA complexes were added and incubated with cells for 2.5 h. The final concentration of siRNA was 200 nM in each well and the volume of medium was 250 μL. After incubation in 1000 μL of normal culture medium for the next 48 h, cells were washed cells by 1× PBS (pH 7.2) and permeabilized with the reporter cell lysis buffer (25 mM Gly-Gly, 15 mM MgSO4, 4 mM EGTA, 1 mM DTT, 1 mM PMSF; pH 7.8). GFP fluorescence in the cell extracts was measured with a Wallac 1420D scanning multilabel counter (Thermo Fisher Scientific Oy, Vantaa, Finland) at excitation wavelength of 485 nm and emission wavelength of 535 nm. The GFP fluorescence level was normalized by the total protein concentration of the cell extracts, measured using Bradford reagent (Helicon, Moscow, Russia). The data are shown as mean ± S.E.M of the means obtained from three independent experiments with three samples. Visual appearance of MDA-MB-231 cells after the transfection was registered using a Leica DM 2500 microscope (Wetzlar, Germany) with a Leica DFC345 FX camera.

#### *2.6. siRNA Transfer to* Е*A.Hy926 Cells*

Transfection experiments in ЕA.Hy926 cells were performed in duplicates. The cells (15 <sup>×</sup> 104) were seeded in 24-well plates and incubated overnight. A fully supplemented cell culture medium was aspirated and replaced with medium without FBS just before the addition of siRNA complexes, followed by incubation for 4 h. The final concentration of siRNA was 200 nM per well in 1000 μL of medium. After incubation in a fully supplemented cell culture medium for the next 48 h, cells were taken for RNA extraction.
