Effect of Integrin Binding Peptide on Vascularization of Scaffold-Free Microtissue Spheroids

Abstract

Background:

Three-dimensional (3D) biomimetic models via various approaches can be used by therapeutic applications of tissue engineering. Creating an optimal vascular microenvironment in 3D model that mimics the extracellular matrix (ECM) and providing an adequate blood supply for the survival of cell transplants are major challenge that need to be overcome in tissue regeneration. However, currently available scaffolds-depended approaches fail to mimic essential functions of natural ECM. Scaffold-free microtissues (SFMs) can successfully overcome some of the major challenges caused by scaffold biomaterials such as low cell viability and high cost.

Methods:

Herein, we investigated the effect of soluble integrin binding peptide of arginine-glycine-aspartic acid (RGD) on vascularization of SFM spheroids of human umbilical vein endothelial cells. In vitro-fabricated microtissue spheroids were constructed and cultivated in 0 mM, 1 mM, 2 mM, and 4 mM of RGD peptide. The dimensions and viability of SFMs were measured.

Results:

Maximum dimension and cell viability observed in 2 mM RGD containing SFM. Vascular gene expression of 2 mM RGD containing SFM were higher than other groups, while 4 mM RGD containing SFM expressed minimum vascularization related genes. Immunofluorescent staining results indicating that platelet/endothelial cell adhesion molecule and vascular endothelial growth factor protein expression of 2 mM RGD containing SFM was higher compared to other groups.

Conclusion:

Collectively, these findings demonstrate that SFM spheroids can be successfully vascularized in determined concentration of RGD peptide containing media. Also, soluble RGD incorporated SFMs can be used as an optimal environment for successful prevascularization strategies.

Electronic supplementary material

The online version of this article (10.1007/s13770-020-00281-5) contains supplementary material, which is available to authorized users.

Introduction

Tissue Engineering (TE) is one of the most vital sciences which aims to develop biological substitutes to restore damaged tissues and organs [1, 2]. A major challenge in TE is development of complex or regeneration of damaged tissues by providing vascular structures formation which is capable of delivering oxygen and nutrients. Substantial areas of tissue become necrotic due to insufficient vascularization results in lack of blood supply and oxygen. Therefore, describing factors which influence and promote vasculogenesis on in vitro microtissue constructs should clearly be identified especially for regeneration of vascular tissues. Besides regenerative medicine studies, in vitro microtissues have been developed for drug screening, drug discovery, personalized medicine and disease modelling [3, 4]. Nevertheless, current strategies for microtissue formation are commonly limited because most of the tissue engineered structures lack microvascular networks. Vascularization of microtissues should have appropriate nutrient delivery and waste elimination, replicate angiocrine signalling to form tissue mimicking developmental processes, construct tissues of macroscopic scale, mimic blood tissue or organ barrier such as the blood–brain barrier [3, 5].

Most of the vascularization studies which aim to mimic native vascular networks and structures, is conducted using in vitro two-dimensional (2-D) culture models, though traditional monolayer approach exhibits low cell signalling, differentiation capacity and proliferation rate [6–10]. Furthermore, these models limit some different biological pathways such as cell–cell and cell–matrix interaction [9]. The additional dimensionality of 3D cultures leads to the differences in cellular responses since it induces the spatial organization of the cell surface receptors involved in interactions with surrounding cells and extracellular matrix (ECM). These physical features in 3D cultures develop the signal transduction from the outside to the inside of cells, and influence gene expression and cellular signals for the formation of 3D vascular structure. In 3D cultures, self-extracellular matrix can be produced similar to almost all cells in the in vivo environment [11]. Therefore, 3D systems can be classified as an intermediate model between 2D and in vivo systems [7, 10, 12, 13].

Methods for developing 3D microtissues are classified as scaffold-free or scaffold-based culture approaches, with the scaffold being produced by organic or synthetic substances. Scaffold-free approaches to develop microttisues can be produced by layering of cell sheets, gravity-enforced assembly in hanging drops, and non-adhesive surface culture systems for the production of spherical microtissues [14]. Scaffold free microtissue (SFM) resembles natural intercellular connection by mimicking the pre-vascularization structures in the ECM as in vivo [6, 7, 15]. SFM provides cellular homeostasis, enables cellular communication in microvascular tissues by supporting ECM production containing proteoglycans, collagen, laminin, and fibronectin [7, 9, 15].

Formation of vascular structures in microtissues is provided by well-orchestrated interactions between cells and the ECM with the influence of biochemical and mechanical cues. During vasculogenesis, vascular endothelial progenitors migrate and differentiate to form capillary tubes with the acquisition of arterial or venous identity [16]. Several potential regulators such as vascular endothelial growth factor (VEGF), vascular endothelial cadherin (Ve-cadherin), tyrosine protein kinase 1 (Tie-1), tyrosine kinase 2 (Tie-2) and platelet/endothelial cell adhesion molecule-1 (PECAM-1) signalling are expressed during vasculogenesis for multiple roles during vascular development. There are several techniques that could be used for vascularization of tissues in vitro [17]. Miller et al. used 3D printed hydrogel-based mold with dissolvable vascular channels and the cells were endothelialized after the molds were dissolved [18]. Since, 3D bioprinting technique allows large scale vascularized tissues, it is possible to achieve sub-100 μm capillary structures. As an alternative strategy, vascularization has been attempted by stimulation of angiogenic processes with delivery of cells with pro-vasculogenic potentials, and the use of cell spheroids co-cultured with endothelial cells or endothelial progenitor cells [19]. For instance, spheroids consisting of mesenchymal stem cells (MSCs) and endothelial cells induces both osteogenic differentiation of stem cells and vessel network formation within the engineered microtissues [20]. Additionally, co-cultured cell spheroids containing endothelial cells and osteoblasts improved the survival and proliferation of multiple cell types in vitro and 3-D microtubular networks [21]. Although the benefit of co-cultured spheroids, many challenges remain, including optimization of ratios and ideal culture conditions for different cell types [22]. In addition, the need for new approaches in pre-vascularized microtissue formation has been arising to extend the lifetime of organs and tissues for transplantation. For instance, Takebe et al. produced pre-vascularized liver bud as functional liver organoid by using cocultured iPS cell-derived hepatocytes, HUVECs and MSCs in Matrigel. The cells spontaneously aggregated due to the contractile forces of MSCs; however, the transplantation of this pre-vascularized bud into mice was failed. Another goal of making pre-vascularized 3D models is to represent the tumor microenvironment more accurately. The vasculature plays a key role in cancer metastasis to proximate the influence expression of target proteins in the tumor to investigate the effect on drug delivery and drug response. To address these concerns, Sobrino et al. developed a 3-D microfluidic tumor model with human tumor and stromal cells that surrounded by a perfusable human cell based vasculature [23]. They demonstrated that IC50 values for the effect of a chemotherapeutic drug which is called Oxaliplatin on tumor cells is higher in the microfluidic vascularized micro-tumors, compared to 2D cultures. Although many studies described different vascularization strategies, a gold standard approach for vascularization has yet to be identified. In light of this information, it is hypothesized that controlling cell-to-cell interactions during SFM formation with key ECM mimicking molecules could potentially accelerate vasculogenesis process.

Integrin binding RGD (Arg,Gly,Asp) peptide (hereafter denoted by RGD), which is one of the fundamental biomimetic unit of common ECM proteins including laminin, fibronectin, vitronectin, von Willebrand factor, osteopontin, is capable of recognising by integrin and provide cells to attach to their surrounding ECM in the microvascular matrix during physiological vasculogenesis [24, 25]. RGD supports elongation of vessels during vasculogenesis by regulating cellular growth, ECM stabilization and cell proliferation [26–31]. Since endothelial cell adhesion and migration are essential components of microvascular development, the RGD sequence regulates the interactions between endothelial cells and the surrounding ECMs during vasculogenesis [27, 32, 33]. RGD functions as natural vasculogenic proteins and contributes to express markers during vascularization period. Different vascular markers such as VEGF, vascular endothelial cadherin (Ve-cadherin), tyrosine protein kinase 1 (Tie-1), tyrosine kinase 2 (Tie-2) and PECAM-1 are affected from endothelial signals [29].

In vascular tissue engineering, angiogenic factors, such as proteins, peptides and growth factors are involved to optimize the signalling of vascular microenvironment [34]. The goal of the present study was to investigate the effect of RGD peptide concentration on pre-vascularization of SFMs that mimic the cellular microenvironment of human umbilical vein endothelial cells (HUVECs). To the best of our knowledge, this is the first study in which HUVECs and RGD in different concentration containing environment are combined in 3D SFM spheroids to regenerate pre-vascularized tissue. First, 3D HUVEC SFMs were constructed and incubated in 0 mM, 1 mM, 2 mM, and 4 mM RGD containing media for 7 days. The diameter of SFM were measured in each time point (day 1, day 4 and day 7). Then, the viability of SFMs were evaluated. Additionally, real time PCR (qPCR) protocol was applied to analyse key vascular gene expressions including VEGF, vascular endothelial cadherin (Ve-cadherin), tyrosine protein kinase 1 (Tie-1), tyrosine kinase 2 (Tie-2) and PECAM-1 Additionally, expressions of VEGF and PECAM were identified by immunofluorescent staining.

Materials and method

Peptide synthesis

All the chemicals used for peptide synthesis were purchased from AAPPTEC (Louisville, KY, USA). Glycine-arginine-glycine-aspartic acid-serine (GRGDS) peptide were synthesized on 4-methylbenzhydrylamine (MBHA) resin (0.67 mmol/g loading capacity) as previously described [35]. Briefly, after swelling the resin in DMF (dimethylformamide), Fmoc-protected amino acids (4 eq), 2-(1H-benzotriazole- 1-yl)-1,1,3,3-tetramethyluronium hexafluorophosphate (2 eq;HBTU), hydroxybenzotri- azole (2 eq;HOBt), N,N-diisopropylethylamine (4 eq;DIEA) were added on resin and incubated for 2 h. The resin was tested for lacking unreacted amines using the Kaiser reagents after washing with DMF. 20% piperidine in DMF was used for Fmoc-protecting group removal. When all amino acids were coupled, amino acids were cleaved from resins by using trifluoroacetic acid (TFA): triisopropylsilane (TIPS): water H₂O solution at ratio of 95:2.5:2.5. The peptide was washed with cold diethyl ether three times and diethyl ether was evaporated. The peptide was freeze dried to evaporate water at − 80 °C (Biobase Biodustry Bk-FD10P, Shandong, China). Further, purification of the peptide was shown via preparative high-performance liquid chromatography (HPLC, Agilent 1200 series) system equipped with Zorbax Extend-C18 2.1 × 50 mm column. The gradient of 0.1% TFA/water and 0.1% TFA/acetonitrile were used with the mobile phase at the detection wavelength of 220 nm. Mass spectrum of GRGDS peptide were characterized via mass spectrometry (Agilent 6530 Q-TOF) with an electrospray ionization (ESI) source (see liquid chromatography and mass spectra for GRGDS in Figures S1 and 2 and Table S1, Supporting Information).

Cell culture

Human umbilical vein endothelial cells (HUVEC) was obtained from Ege University Research Group of Animal Cell Culture and Tissue Engineering Laboratory. All cell culture supplement was purchased from Gibco (Paisley, UK). Cells were cultured at 37 °C and 5% CO₂ atmosphere with DMEM (Dulbecco’s Modified Eagle Medium) which includes 1% l-glutamine and 1% penicillin, 10% FBS (fetal bovine serum). HUVECs were kept in exponential phase and used at passage three. 2 w/v% agarose solution (Sigma-Aldrich, St. Louis, MO) was prepared in PBS. Agarose solution was heated to boiling in a microwave oven and poured on 3D petri dish. After the molds were produced, 75 µl cell suspension (1 × 10⁵ cells/ml) was added on mold. GRGDS peptide in 4 mM, 2 mM, 1 mM and 0 mM concentration were mixed with EGM-2 Bullet Kit (Lonza, Walkersville, MD, USA) contained hydrocortisone, fibroblast growth factor (hFGF-B), insulin-like growth factor (R3-IGF-1), ascorbic acid, epidermal growth factor (hEGF), GA-1000 (gentamicin, amphotericin-B), heparin without FBS. The cell culture medium was replaced with fresh 4 mM, 2 mM, 1 mM and 0 mM RGD containing medium once every 2 days. 0 mM RGD groups was used as negative control group to determine the effect of EGM-2 Bullet kit media itself on vascularization without RGD. The micrographs were taken from 3 different well and 3 different agarose mold (n = 9). Micrographs of micro tissues were captured by CellSense software and light microscope (Olympus, CKX41, Waltham, MA, USA) at day 1, day 4 and day 7.

Live and dead assay

Double Staining Kit (Dojindo Molecular Technologies, Inc., Kumamoto, Japan) was performed to demonstrate viability of SFM as descripted in previous study [36]. Briefly, the viable cells (Calcein-AM/DMSO, green fluorescence) and dead cells (propidium iodide/purified water, red fluorescence) were stained after 15 min of incubation and observed by using fluorescence microscope. (Olympus, CKX41). Micrographs of SFM were taken 3 different microtissues (n = 3).

Quantitative real-time PCR analysis of vascular markers

Real Time PCR (qPCR) protocol was applied to analyse vascular gene expressions as previously described [35]. At each time point (1, 4, and 7 days), total cellular RNA was isolated using Blood/Cell Total RNA Mini Kit (Geneaid, Sijhih City, Taiwan). Then, the extracted purified RNA was subjected to cDNA conversion using M-MuLV First Strand cDNA Synthesis Kit (Biomatik, Ontario, Canada). The cDNA was subjected to Step One Plus Real-time PCR system (Applied Biosystems, Foster City, CA, USA) amplification with appropriate gene-specific primers. Forward and reverse primers for RT-qPCR, shown in Table 1, including PECAM, VEGF, Ve-cadherin. Tie-1 and Tie-2 and glyceraldehyde 3-phosphate dehydrogenase (GAPDH) were purchased from Sentegen Biotechnology (Ankara, TURKEY) and used to evaluate gene expression [37–41]. Vascular gene profiles were also analyzed as compared with 0 mM RGD group. The differential expression of genes PECAM, VEGF, Ve-cadherin. Tie-1 and Tie-2 was quantified by StepOne Software v2.3 and Ct values were classified by 2^(−ΔΔCt) method described elsewhere [42]. Every group was experimented in qPCR as doublet and repeated as triplicate (n = 6).

Table 1.

Forward and reverse primers

Genes	Primers	TM (°C)
GAPDH	F: GAAATCCCATCACCATCTTCC	54.3
	R: CCAGCATCGCCCCACTT	58.8
Ve-cadherin	F: TCACCTGGTCGCCAATCC	57.8
	R: AGGCCACATCTTGGGTTCCT	58.9
PECAM	F: GCTGACCCTTCTGCTCTGTT	57.4
	R: TGAGAGGTGGTGCTGACATC	56.8
VEGF	F: ATCTTCAAGCCATCCTGTGTGC	57.6
	R: GCTCACCGCCTCGGCTTGT	64.3
Tie-1	F: AGGTCACGCTTCGCGGCTT	63.2
	R: CCAAAACGGCCCTCTCTG	56.4
Tie-2	F: TAGAGCCTGAAACAGCATACCAGG	58.5
	R: CTATTGGCAATGGCAAATGCTGGG	59.4

PECAM, VEGF, Ve-cadherin. Tie-1, Tie-2, and glyceraldehyde 3-phosphate dehydrogenase (GAPDH; housekeeping gene) (GAPDH) used to asses HUVEC differentiation in qRT-PCR amplification

Immunofluorescence staining

For immunofluorescence staining, micro tissues were rinsed twice in PBS and fixed with 4% paraformaldehyde (Sigma Aldrich, St. Louis, MO, USA) at 4 °C for 30 min followed by embedded for cutting (5 µm-thickness). Next, samples were immersed with 0.1% Triton X-100 in PBS for 1 h and blocked with 1.5% Bovine Serum Albumin (BSA) in PBS for 2 h. Then, samples were incubated with primary antibodies in PBS containing 1% BSA overnight at 4 °C according to manufacturer’s instructions. Primary antibodies (Santa Cruz Biotechnology Inc., Santa Cruz, CA, USA) were used included mouse monoclonal antibody against VEGF (147) (sc-507, 1:100 dilution) and PECAM (sc-376764, 1:100). Fluorescence secondary antibodies (Santa Cruz Biotechnology Inc., Santa Cruz, CA, USA) included m-IgG kappa BP-FITC (sc-516140; 1:200) and m-IgG kappa BP-PE (sc-21742; 1:200) were diluted with 1% BSA. The expression pattern of VEGF and PECAM with same exposure time and light intensity were characterized via capturing images using an inverted fluorescence microscope (Olympus CKX41, Tokyo, Japan). After 7 days of incubation, the formation of endothelial cell tubular structures was assessed by using immunofluorescence staining microphotographs which were taken at low magnification with a florescent microscope from three randomly selected fields for data analysis. The total length of the tubular structures was measured with the aid of ImageJ software (NIH, Bethesda, MD, USA) and calculated relative to 0 mM RGD as percentage.

Statistical analysis

All data are statistically analyzed by two-way ANOVA (SPSS 12.0, SPSS GmbH, Germany) and the Student–Newman–Keuls method as a post hoc test. Significant differences between groups were determined using p values less than 0.05 (p < 0.05, p < 0.01, p < 0.001).

Results

The SFMs with different RGD concentration were constructed (Fig. 1). As shown in Fig. 2, SFMs were successfully formed and SFM images were taken by light microscope. The diameters of microtissues were calculated in ImageJ (Fig. 3). The diameters of microtissue in. 0 mM, 1 mM, 2 mM and 4 mM RGD were 274.13 ± 10.48 µm, 279.69 ± 7.67 µm, 275.04 ± 6.51 µm, and 568.23 ± 33.02 µm, respectively at day 1. 4 mM RGD containing group was larger compared to other groups (p < 0,001). At day 4, the diameter of 0 mM, 1 mM, 2 mM and 4 mM RGD containing microtissues were calculated as 243.23 ± 7.92 µm, 225.60 ± 4.55 µm, 282.10 ± 4.57 µm and 326.79 ± 15.34 µm, respectively. Diameters of 0 mM, 1 mM, 2 mM and 4 mM RGD groups were 228.26 ± 6.45 µm, 218.18 ± 5.69 µm, 294.41 ± 8.55 µm and 310.15 ± 11.07 µm at day 7.

Fig. 1.

Schematic diagram of SFM fabrication

Fig. 2.

The micrographs of HUVEC SFM in A 0 mM, B 1 mM, C 2 mM, and D 4 mM RGD at day 1, day 4 and day 7 (Scale bar represents 100 μm)

Fig. 3.

Spheroid diameter values of HUVEC SFM in 0 mM, 1 mM, 2 mM and 4 mM RGD at day 1, day 4 and day 7. *p < 0.05, **p < 0.01, ***p < 0.001

Fluorescent microscopy images of live/dead staining analysis for 0 mM, 1 mM, 2 mM and 4 mM groups at day 7 were shown in Fig. 4. Green color and red color in the images represent live cells and dead cells, respectively. 4 mM RGD group has the highest red intensity between experimental groups while 2 mM RGD group has the higher green color intensity. Red color intensity of 4 mM RGD was higher than 0 mM RGD peptide group which was used as negative control. In all groups, viable cells are located inside of the microtissue, while dead cells are located in outside of microtissues.

Fig. 4.

A-D Viability analysis of SFM in 0 mM, 1 mM, and 2 mM, and 4 mM RGD at day 7. Live cells and dead cells are stained with calcein-AM (green) and EthD-1 (red-dead) (scale bar represents 50 µm)

Expression of vasculogenic markers VEGF, Tie-1, Tie-2, Ve-Cadherin, and PECAM with incubation time for SFM is shown in Fig. 5A–E, respectively. VEGF, Tie-1, Tie-2, Ve-Cadherin expression of 2 mM RGD were increased gradually with incubation time and higher than other groups in all time points. In day 7, VEGF expression of 2 mM RGD was higher (57.56 ± 7.30, p < 0.001) than 0 mM RGD (3.01 ± 0.36), 1 mM RGD (41.20 ± 12.05, p < 0.01) and 4 mM RGD (2.62 ± 0.46, p < 0.001). Similarly, Tie-1 (115.85 ± 11.94, p < 0.001), Tie-2 (68.21 ± 11.99, p < 0.01) and Ve-Cadherin (26.46 ± 2.23, p < 0.001) was highest in 2 mM RGD group at day 7. PECAM expression peaking at day 4 and returning to the baseline level at day 7 in all groups was highest in 2 mM RGD. While PECAM expression of 1 mM and 2 mM RGD were calculated as 53.37 ± 4.11 (p < 0.001) and 49.54 ± 12.29 (p < 0.01) at day 1, they were 68.16 ± 11.46 (p < 0.01) and 116.08 ± 16.80 (p < 0.001) at day 4, respectively.

Fig. 5.

A-E The mRNA expression levels (as fold difference) of VEGF, Tie-1, Tie-2, VE-cadherin, PECAM-1 for HUVECs encapsulated on SFMs in 0 mM, 1 mM, 2 mM and 4 mM RGD and incubated in vasculogenic medium for up to 7 days. Error bars represent mean ± SE (n = 5) [significant differences were determined by one-way ANOVA] Newman–Keuls multiple comparison test. *p < 0.05, **p < 0.01, ***p < 0.001

Immunofluorescence staining of differentiated SFMs assessed as previously described [24]. Briefly, SFM was incubated in vasculogenic medium for 7 days and then chemically fixed for immunostaining. Lastly, proteins were stained using first primary antibodies against PECAM (red), VEGF (green) and then labelled with secondary antibodies to be visualized using an inverted fluorescence microscopy. As demonstrated in Fig. 6A–D, the expression of these vasculogenic maturation related proteins was drastically higher in 2 mM RGD containing SFM than 4 mM, 0 mM and 1 mM RGD groups, respectively. In consistent with real-time PCR results, cells on 2 mM RGD groups showed an elevated expression of these markers compared to 4 mM, 1 mM and 0 mM RGD, highlighting the vasculogenic inductive effect of 2 mM RGD. The formation of endothelial cell tubular structures was increased to 220 ± 12% and 150 ± 10% in 2 mM and 1 mM RGD group, respectively (p < 0.05), whereas the total tube length was reduced to 85% in 4 mM RGD group.

Fig. 6.

A-D Expression pattern of vasculogenic markers PECAM (green), VEGF (red) for HUVECs seeded on CAP treated 0 mM, 1 mM, 2 mM and 4 mM RGD. E The total length of the tubular structures after 7 days’ incubation in vasculogenic medium (scale bar represents 100 µm). *p < 0.05

Discussion

The ability to achieve mass transfer in engineered microtissue is an important step to develop cell-based therapies to enhance availability of oxygen and nutrient supply. Well-orchestrated interactions between cells and the extracellular matrix involve formation of vascular structure in tissues. This is also supported by biochemical and mechanical activities. The approach in this study incorporates HUVECs and RGD containing environment to create modular SFMs that mimic the native microenvironment for improvable strategy to vascularize tissues.

According to the differential adhesion theory, which was attempt to explain tissue self-assembly, cells aggregate as a spheroid to maximize adhesion and minimize energy [43]. The inner core is formed by the cells of highest cohesion, and the periphery of the spheroid those of lower cohesion. In the present study, RGD in different concentration influenced the organization of HUVECs and the network within the SFM. 2 mM RGD containing group had larger SFM formation at day 7, while the larger microtissue was formed in 4 mM RGD group at day 1. The integrity of 4 mM RGD group was considered to be lost due to the blockage of α and β sub-units of integrin in higher RGD concentrations. Previously, it has been compared the rates of cell migration and viability in the presence of soluble peptide concentrations. It was reported that the rate of disconnection of focal adhesions at the back of the cell, which may induce higher migration levels increased with the addition of soluble peptide and the presence of a soluble peptide helps to block individual communications from reassembling and allows an unzipping of the polyvalent complex [44].

The viability of cells after formation of SFMs was higher in 2 mM RGD group and subsequent to formation of SFMs cells continued to proliferate, as shown by a steady increase in cell number of SFM. Guo et al. concluded that RGD peptide promotes lifespan and density of HUVECs [45]. On the other hand, the cell viability of HUVECs in 4 mM RGD was lowest compared to other groups. In 4 mM RGD group, integrin attachment points of HUVEC could be triggered to choose apoptosis pathway [46, 47]. However, cell death in 4 mM group may be caused by high acidic conditions due to aspartic acid in RGD sequence resulting in hydrolyzation after liposomal activity [46]. TFA which is used for cleavage step can bind to the free amino termini and side chains of positively charged amino acids in GRGDS peptide could be the reason of the acidic condition. The viability of 2 mM RGD containing group comparable to cells embedded in bulk materials such as hydrogels, suggesting that the microtissue fabrication process was not harmful to the cellular component [48, 49].

The importance of supporting endothelial cells in promoting and maintaining new vessel formation has become increasingly clear [50]. Endothelial cells secrete angiogenic growth factors, vascular endothelial growth factors, and angiopoietins in significant levels and enhance early capillary network formation such as VEGF, Tie-1, Tie-2, PECAM and Ve-cadherin [51–53]. Herein, HUVECs in SFM not only survived but also secreted angiogenic growth factors such as VEGF, Tie-1, Tie-2, PECAM and Ve-cadherin in higher RGD concentration up to 2 mM at day 1, 4, and 7 day. Although the vasculogenic markers were expressed in 0 mM RGD group which was used to eliminate the influence of EGM-2 Bullet kit media on vascularization, the expression of vasculogenic gene expressions were significantly higher in 1 mM and 2 mM RGD groups. VEGF tend to provide connections of HUVEC during vasculogenesis [54] so it is possible to survival and vasculogenesis of HUVEC SFM. Zheng et al. reported that increasing amount of RGD peptide, VEGF expression was increased significantly [55]. However, in another IIlıan et al. speculated that VEGF may might also initiate apoptosis process [54]. In our study, 4 mM RGD group may be directed to apoptosis by increased VEGF. The positive correlation between increased proliferation and enhanced differentiation might be due to increased integrin binding. In 2 mM RGD group, Tie-1:Tie-2 heterocomplex may start constructs vascular network formation because of both genes overexpressed. HUVEC communications and new vascular formation may be provided by Ve-cadherin which has been expressed in 2 mM RGD group. The activation of the integrin-mediated signal pathways triggers the expression of multiple genes and acts in synergy with other receptor pathways, such as growth factor receptors, to regulate cell growth and differentiation [56, 57]. The results clearly indicate that HUVEC SFMs could be reconfigured into optimum amount of RGD, retaining their native cellular reorganizing capacity as well as the ability to secrete proangiogenic factors. PECAM and VEGF expression profile was parallel to immunofluorescent results in our study. Genes of PECAM was overexpressed as from first day in 1 mM and 2 mM groups. The role of PECAM is to provide cellular adhesion and migration of HUVEC during vasculogenesis [58]. Optimal RGD peptide may direct overexpression and protein production of PECAM. PECAM and VEGF proteins showed the vascular network was maintained in 7 days. 2 mM RGD group obviously showed higher PECAM and VEGF protein levels, compared with other groups. To eliminate EGM-2 Bullet kit media on vascularization, 0 mM RGD was used.

SFM fabrication techniques provide cellular homeostasis, enables cellular communication in microvascular tissues by supporting ECM production. Even though, agarose gel supported the 3-D microtissue formation, this method is considered as scaffold-free method because HUVECs were not adhered to a surface of agarose gel and cells spontaneously self-assembled and cell-to-cell interactions was provided. Similar results can be obtained with other scaffold free microtissue fabrication technique such as gravity-enforced assembly in hanging drops. Nevertheless, agarose gel as a mold provides large-scale and safe microtissue formation compared to hanging drop method.

In conclusion, the present study showed that soluble RGD incorporated HUVEC SFMs can give rise to pre-vascularized tissue. Maximum microtissue dimension and cell viability observed in 2 mM RGD containing SFM. Vascular gene expressions such as VEGF, Tie-1, Tie-2, Ve-Cadherin, and PECAM, in 2 mM RGD containing SFM were higher than other groups, while 4 mM RGD containing SFM expressed minimum vascularization related genes. 3-D pre-vascularized spheroids were constructed with optimum RGD concentration. HUVECs spontaneously self-assembled, reached a structural equilibrium that is obtained by cell-to-cell interactions and vascularized without any scaffold which cells adhere and any need for co-culture. Even though integrin binding peptide RGD has been used for scaffold functionalization to support pre-vascularization, there is no study in the literature that shows that soluble RGD in optimum concentration supported cell-to-cell interactions in SFMs and vasculogenesis to the best of our knowledge. 3D biomimetic models via various approaches can be used in drug delivery and tissue engineering applications instead of in vivo models. An optimal vascular microenvironment in 3D model that mimics the cell ECM and an adequate blood supply for the survival of cell transplants are major challenge that need to be overcome in vascular tissue engineering. Findings of the current study highlight the significance of developing a 3D microenvironment that supports cell–cell interactions which can in turn contribute toward an optimal condition for successful 3D tissue modelling strategies.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Compliance with ethical standards

Conflict of interest

Authors have declared that there is no conflict of interest. The authors alone are responsible for the content and writing of this article.

Ethical statement

There are no animal experiments carried out for this article.