Bioactive vascular buds promote collateral vessel formation by grafting on the artificial vessel walls

Abstract

Vascular injury and some chronic ischemic lesions usually lead to insufficient blood supply to tissues, which will lead to tissue ischemia or even necrosis in severe cases. Current artificial blood vessels lack effective collateral vascularization capabilities to provide adequate blood supply in areas with restricted blood flow. Herein, inspired by the grafting of tree buds to form lateral branches, the vascular buds model was successfully constructed by inoculating HUVECs into bioactive hydrogel microspheres. Under the influence of ions dissolved from bioactive glass and three-dimensional culture environment, the cytoskeleton was remodeled, the cells showed obvious outward migration and budding trend, which significantly enhanced the angiogenesis ability. After grafted vascular buds to the lateral wall of the artificial blood vessel, a large number of collateral vessels are formed, which effectively alleviates the tissue ischemia in the region through which blood vessels pass. This study confirms the impact of bioactive ions on angiogenesis in a three-dimensional environment and offers novel insights for the development of lateral branches in artificial blood vessels.

Graphical abstract

Inspired by plant grafting techniques, vascular buds containing vascular endothelial cells were transplanted into double-layer fibroin artificial blood vessels, aiming to use the vascular buds as germinal centers for collateral vessel formation. In this environment, vascular bud can effectively promote the development of collateral vessels and form a complex and rich vascular network.

Highlights

• •l HUVECs were inoculated into BG-GelMA microspheres to establish vascular buds.
• •l Vascular buds promote collateral blood vessel formation and improve massive ischemia.
• •l Three-dimensional environment and bioactive ions work together to promote vascular budding.
• •l Inspired by plant grafting techniques, vascular buds were grafted into artificial blood vessels.

1. Introduction

Vascular injury and some chronic ischemic diseases often lead to insufficient blood supply to the tissue, such as atherosclerosis, diabetic microvascular disease, peripheral artery disease, etc [1,2]. Artificial vascular reconstruction has been an important research direction in the field of biomedical engineering in recent years [3,4]. Silk fibroin (SF), as a natural polymer material, has gradually become an ideal material for vascular regeneration and reconstruction due to its excellent biocompatibility, degradability, mechanical properties, and excellent cell adhesion [5,6]. Artificial blood vessels constructed by SF alleviated tissue ischemia to some extent and promoted functional recovery of ischaemic areas [7]. However, there are still shortcomings such as thromboembolism, thrombosis, compliance mismatch and neointimal hyperplasia, which hinder the normal integration and function of artificial blood vessels and aggravate the overall ischemic state [8]. Consequently, the rapid formation of a comprehensive vascular network is the key to the treatment of ischemia. Unfortunately, current clinical technologies for artificial blood vessels fall short of stimulating the formation of collateral vessels, which would be essential for improving blood flow throughout larger ischemic regions [[9], [10], [11]].

Lateral branches can be quickly formed by grafting vibrant tree buds on the bare trunk (Fig. 1B). Inspired by grafting techniques, we want to “grafted” vascular buds in artificial blood vessels to solve large-area ischemia by quickly forming collateral blood vessels. Similar to tree buds, the vibrant cells of vascular buds are the key to the formation of collateral blood vessels. However, direct implantation of cell clusters as vascular outgrowth centers is prone to cell necrosis. Inoculating cells on hydrogel microspheres constructed from extracellular matrix is an effective method to improve the survival rate by simulating the microenvironment in vivo and providing the nutrients.

Fig. 1.

Schematic illustration. (A) BG-GelMA microspheres were prepared by the microfluidic technique. (B) Lateral branches can be quickly formed by grafting vibrant tree buds on the bare trunk. (C) The blood vessel bud containing vascular endothelial cells is transplanted into the artificial blood vessel as the center for the formation of collateral blood vessels and the process of collateral blood vessel formation. (D) Processes of endothelial cell migration and sprouting of vascular buds.

As an ideal hydrogel scaffold material, gelatin methacryloyl (GelMA) effectively support cell attachment and growth and provide a good microenvironment for cell proliferation and differentiation [12,13]. It has been demonstrated that hydrogel scaffolds implanted subcutaneously in mice show strong angiogenesis [14]. Various types of bioactive particles can be evenly dispersed within the hydrogel network. For example, our previous work has constructed bioactive glass (BG)-hydrogels composite [15,16]. In addition, we also confirmed that the bioactive ions dissolved by BG have the function of promoting blood vessel formation [17]. These ions, such as silicon (Si), calcium (Ca), and phosphate (P), enhance vascularization processes in the local microenvironment by regulating cellular behavior and signaling pathways [18,19]. However, although BG has been shown to have the potential to promote angiogenesis in a two-dimensional environment, its vascularization effects and mechanisms in a three-dimensional (3D) environment have not been fully explored.

Therefore, in this experiment, BG-GelMA microspheres were prepared by microfluidic technique (Fig. 1A). The angiogenic ability and mechanism of composite microspheres were studied. Further, inspired by transplantation techniques (Fig. 1B), vascular buds loaded with vascular endothelial cells were transplanted into artificial blood vessels as centers for forming collateral vessels (Fig. 1C). At the same time, the mechanism of BG-GelMA hydrogel microspheres in promoting collateral vascular formation was studied (Fig. 1D).

2. Methods

2.1. Synthesis and characterization of BG-GelMA microspheres

The fabrication of monodisperse BG particles was conducted according to our previously established protocol [20], with comprehensive methodological details provided in the Supporting Information. BG-GelMA microspheres were prepared using microfluidic technology. Summarily, 0.15 g of BG and 1 g of GelMA were dispersed in 5 g of deionized water, dissolved by heating in a water bath at 40 °C, and then 0.25 % lithium phenyl-2,4,6-trimethylbenzoylphosphinate was added to prepare solution A. 3 g Sorbitan oleate (Span 80) dissolved in 100 mL of Isopropyl myristate to prepare solution B. Subsequently, solution A was added dropwise with a syringe pump at 200 μL/min to solution B. Magnetic stirring was used during the dropwise addition. The active agent on the surface of the hydrogel microspheres was washed away with alcohol and deionized water, and the extracted powder was collected, absorbed and lyophilized to obtain the final hydrogel microspheres (GelMA, BG-GelMA).

The morphological characteristics and internal architecture of hydrogel microspheres were examined using scanning electron microscopy (SEM, Merlin, Carl Zeiss, Jena, Germany). For chemical characterization, Fourier transform infrared spectroscopy (FTIR, VERTEX 33, Bruker, Germany) was used to determine the molecular composition and phase structure of the microspheres.

2.2. Cell culture

Human umbilical vein endothelial cells (HUVECs) were purchased from Cyagen Biosciences Inc (China). The cells were cultured in Endothelial Cell Medium (ECM; Sciencell, USA) supplemented with 10 % fetal bovine serum (FBS; Sciencell, USA) and 1 % penicillin/streptomycin (Sciencell, USA) in a 5 % CO₂ incubator at 37 °C. The medium used for cell culture is recycled every 2 days. The cells used for subsequent experiments were between passages three and six.

2.3. In vitro biocompatibility test of BG-GelMA microspheres

To evaluate the adhesion and biocompatibility of microspheres. HUVECs were seeded directly onto the surface of GelMA and BG-GelMA microspheres. Adhesion and proliferation of GelMA and BG-GelMA microspheres on HUVECs were evaluated and studied using live/dead staining and cytoskeletal staining. Comprehensive experimental protocols and methodological details are available in the Supplementary Information section.

2.4. Angiogenic ability of BG-GelMA microspheres invitro

The tube-forming assay, scratch assay and immunofluorescence staining were used to demonstrate the angiogenesis role of BG-GelMA microspheres in vitro. In addition, polymerase chain reaction (qRT-PCR) and Western blot were used to verify angiogenesis-related gene and protein expression. Detailed experimental procedures are detailed in the supporting information.

2.5. Model of sprouting angiogenesis based on BG-GelMA microspheres

Drawing on experimental methods given in the reference, GelMA and BG-GelMA cell spheroids were cultured separately [21]. Matrigel and suspended spheroids were mixed in a 7:3 ratio and these spheroids were allowed to sprout for 24 h. The number of extensions and the length of buds grown from each sphere were then measured to digitally quantify in vitro sprout, with 8–10 spheres analyzed per experimental group.

2.6. Mechanism analysis of BG-GelMA microspheres promoting angiogenesisinvitro

Bioinformatics analysis and RNA sequencing (RNA-seq) were used to analyze the mechanism of pro-angiogenesis of BG-GelMA microspheres. To verify the correctness of the mechanism speculation, Western blot and immunofluorescence staining were used to analyze the protein expression of Rac1 and CDC42. Detailed experimental procedures are provided in the Supplementary Information.

2.7. In vivo angiogenesis evaluation

Surgical methods and treatment: Nude Mouse (female, 19–23 g) was obtained from the Experimental Animal Center of Southern Medical University. All animal experiments were following the guidelines approved by the Animal Care Center of Guangdong Pharmaceutical University, with measures implemented to minimize animal pain and distress. The vascular enhancement animal model was modified from a previously published study. To ensure randomness and repeatability, 15 nude mice were randomly divided into 3 groups with 5 mice in each group, which were blank control group (SF), control group (GelMA-SF) and experimental group (BG-GelMA-SF). All nude mice were generally anaesthetized by intraperitoneal injection of pentobarbital (Nembutal, 3.5 mg/100 g) and sterilized with iodine. A sagittal incision of about 3 mm was made on the dorsal skin of each nude mouse. SF, GelMA-SF and BG-GelMA-SF artificial blood vessels with a diameter of 3 mm and thickness of 2 mm were implanted into the subcutaneous skin of nude mice of blank control group, control group and experimental group, respectively, and fixed with 6-0 silk thread. After 1 week of feeding, euthanasia was performed on 15 nude mice through intraperitoneal administration of an excessive dose of sodium pentobarbital. For vascular casting, the sacrificed animals underwent a sequential perfusion protocol beginning with heparinized saline, followed by 4 % paraformaldehyde solution, and concluding with Microfil (MV-112, Flow Tech, Inc., Carver, MA). After overnight incubation at 4 °C, the implanted vascular constructs were carefully excised along with the adjacent dermal tissue for subsequent analysis. To identify the neovascularization area, Micro-CT imaging (XTV160H, X-TEK Co., UK) was used using 70 kV voltage and 100 mA current with a minimum resolution set to 20 × 20 × 20 μm. CT scans are used to reconstruct three-dimensional vision.

Histological evaluation: Following Micro-CT scan, the sample was embedded in paraffin wax and cut into continuous sections 5 μm thick for subsequent histological evaluation. Tissue sections were stained with hematoxylin and eosin (HE) according to the manufacturer's instructions. Immunofluorescence staining was performed using antibodies against CD31 (1:2000, Proteintech), VEGF (1:2000, Proteintech), CDC42 (1:2000, Proteintech), Rac1 (1:2000, Proteintech), KDR (1:2000, Proteintech), and CD34 (1:2000, Proteintech). Immunohistochemical staining was also conducted for CD31 (1:2000, Proteintech), VEGF (1:2000, Proteintech), CDC42 (1:2000, Proteintech), and Rac1 (1:2000, Proteintech). Images were then acquired using CaseViewer (Hungary) software.

2.8. Statistical analyses

All data were expressed as mean ± standard deviation (SD) (n ≥ 3). One-way analysis of variance or Student's t-test using GraphPad was used for statistical comparison. Statistical significance was determined at a threshold of P-value for all comparative analyses.

3. Results

3.1. Preparation and characterization of BG-GelMA microspheres

In this study, BG-GelMA microspheres were successfully prepared by mixing GelMA and BG nanoparticles (540 ± 100 nm in diameter) in a certain ratio using microfluidics, and the polymerization reaction was carried out in a microfluidic channel (Fig. S1). To observe the surface shape of BG-GelMA microspheres, optical microscope and confocal laser scanning microscope were used. Optical microscope observation showed that BG-GelMA microspheres had good dispersion and were similar in shape and size to GelMA microspheres (Fig. 2A). Confocal laser scanning microscope 3D scanning and reconstruction revealed that the two kinds of microspheres were three-dimensional spherical and uniform in size (Fig. 2F). SEM was used to examine the surface morphology of BG-GelMA microspheres. SEM observed uniformly dispersed BG microparticles on the surface of the BG-GelMA microspheres while the GelMA microspheres had a smooth surface (Fig. 2B). The FTIR results also showed that the BG microparticles were effectively incorporated into the hydrogel microspheres, with the peak at 1100 cm⁻¹ indicating the absorption peak of O-Si (Fig. 2G). To verify the biocompatibility and proliferation of BG-GelMA microspheres, HUVECs on the microspheres were cultured for 1, 4 and 7 days and then subjected to live/dead staining (Fig. 2C–E). Although there were some cell deaths on the 1 day, most of the cells had good viability with normal proliferation rates. This may be due to the time required for the cells to adapt to the microspheres during the initial attachment process. Over time, the cells further adapted to the microsphere surface environment and exhibited a more stable proliferation and functional state. Moreover, HUVECs were cultured on hydrogel microspheres on days 1, 4 and 7 for cytoskeleton staining (Fig. 2D), mainly to verify the proliferation rate and morphology of the cells. The staining results showed that on day 1, HUVECs were uniformly adhered to the wall on the surface of BG-GelMA microspheres, the cell morphology was pike-shaped, and some cells began to expand along the surface of the microspheres and connect. After 4 and 7 days of culture, there was no significant difference between the two groups, and the cell coverage was further improved, forming a dense cellular network structure, which showed strong proliferation ability and morphological stability. The microspheres exhibit structural stability even under prolonged cell culture conditions of up to 21 days, and the cells exhibited a well-organized and uniform distribution on the microspheres, maintaining high viability and proliferation rates throughout the culture period (Fig. S2). These results suggested that hydrogel microspheres have excellent biocompatibility and the ability to support the adhesion and proliferation of HUVECs, providing potential support for their use in vascularization applications.

Fig. 2.

Characterization and biocompatibility of BG-GelMA microspheres. (A) Optical image of GelMA and BG-GelMA microsphere. (B) SEM images of GelMA and BG-GelMA microsphere at different magnification. (C) Live/dead staining of HUVECs on GelMA and BG-GelMA microsphere for 1, 4 and 7 days. (D) Cytoskeleton staining of HUVECs on GelMA and BG-GelMA for 1, 4 and 7 days. (E) Quantitative analysis of live/dead staining. (F) Image of GelMA and BG-GelMA microsphere under confocal laser scanning microscope. (G) The FTIR spectra of GelMA and BG-GelMA microsphere.

3.2. In vitro angiogenesis of BG-GelMA microspheres

The in vitro angiogenic potential of BG-GelMA microspheres was evaluated to investigate their ability to promote blood vessel formation. A scratch assay was first performed (Fig. 3A). Quantitative analysis of the scratch assay revealed distinct differences in cellular migration patterns between the experimental groups (Fig. 3B). Following 12- and 24-h incubation periods, microscopic evaluation and subsequent quantification demonstrated a markedly reduced residual scratch area in the BG-GelMA condition compared to the GelMA control (P < 0.05), suggesting enhanced migratory capacity of HUVECs in the presence of BG-GelMA microspheres - a crucial prerequisite for angiogenesis. To further investigate the angiogenic potential, we conducted a matrigel-based tube formation assay (Fig. 3C). Comparative analysis of the tubular networks revealed that while GelMA microspheres exhibited minimal tubulogenic activity, BG-GelMA microspheres significantly enhanced the formation of capillary-like structures. Quantitative assessment of network parameters, including the number of nodes, meshes, junctions and total length, consistently demonstrated superior angiogenic performance in the BG-GelMA group relative to the GelMA group (Fig. 3D). To further confirm the pro-angiogenic effect of BG-GelMA microspheres in vitro, immunofluorescence staining for CD31 and VEGF was performed. CD31, an endothelial cell marker, and VEGF, a key pro-angiogenic factor, were analyzed to evaluate endothelial activation and angiogenic signaling. As shown in Fig. 3E and F, the fluorescence intensity of CD31 and VEGF in the BG-GelMA group was significantly higher than that in the GelMA group, indicating that the BG-GelMA microspheres effectively up-regulated the expression of CD31 and VEGF, which is essential for promoting angiogenesis (P < 0.05). To further validate these findings at the molecular level, mRNA and protein expression levels of CD31 and VEGF were quantified using qRT-PCR and Western-blot analysis (Table S1). The qRT-PCR results demonstrated that the relative mRNA expression levels of CD31 and VEGF were increased significantly in the BG-GelMA group compared to the GelMA group (Fig. 3G) (P < 0.05). Similarly, the Western blot analysis revealed a marked increase in the protein expression levels of CD31 and VEGF in the BG-GelMA group (Fig. 3H). These results confirmed that BG-GelMA microspheres not only enhance the expression of angiogenesis-related markers at the cellular level but also upregulate the expression of these markers at the transcriptional and translational levels. In summary, BG-GelMA microspheres have an outstanding vascularization capacity.

Fig. 3.

In vitro angiogenesis of BG-GelMA microspheres. (A) Representative images of scratch wound healing showing cell migration at 0, 12, and 24 h of culture. (B) Quantitative results of wound healing of scratch wounds in each group as a percentage of the initial defect at 0 h. Wound area (%) was measured using Image J. (C) In vitro angiogenic potential of HUVECs cultured on matrigel by GelMA and BG-GelMA group following 12-h incubation. (D) Quantitative analysis of numbers of nodes, meshes and junctions and total length per high power field (HPF). (E) Immunofluorescence images of CD31 and VEGF staining. (F) Quantification of the immunostaining intensity of CD31 and VEGF. (G) Relative gene expression of CD31 and VEGF. (H) Western blot analysis and relative protein expression of CD31 and VEGF. (∗P < 0.05, ∗∗P < 0.005, ∗∗∗P < 0.001, ∗∗∗∗P < 0.0001).

3.3. Characteristics and mechanism of BG-GelMA microspheres promoting sprouting

RNA-seq was used to detect the differences in HUVECs gene expression after 3 days of BG-GelMA culture to determine the molecular mechanism of BG-GelMA and cell-promoting angiogenesis. The principal component analysis (PCA) of gene expression profiling revealed significant clustering between the GelMA and BG-GelMA group, indicating significant differences in transcription (Fig. 4A). The volcano plot analysis revealed that 3155 genes were up-regulated, and 296 genes were down-regulated in the BG-GelMA group compared to the GelMA group (Fig. 4B). To further investigate the biological processes and pathways involved in the pro-angiogenic effects of BG-GelMA microspheres, functional annotation and pathway analysis were conducted through comprehensive Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways and gene ontology (GO) enrichment analyses. The results of KEGG and GO enrichment analysis were presented as a scatter plot and bar plot. KEGG pathways enrichment analysis confirmed that BG-GelMA microspheres activated several angiogenesis-related signaling pathways including the MAPK signaling pathway, focal adhesion and Rap1 signaling pathway (Fig. 4D). In addition, GO enrichment analysis also confirmed that the BG-GelMA group differed in 15 terms related to angiogenesis compared to the GelMA group. The up-regulated genes were significantly enriched for biological processes associated with angiogenesis, endothelial cell proliferation and extracellular matrix organization (Fig. 4E).

Fig. 4.

Evaluation and mechanism of blood vessel formation promoted by BG-GelMA microspheres. (A) PCA of gene expression. (B) Volcano plot of the differentially expressed genes between GelMA and BG-GelMA group (≥2-fold difference, P value < 0.05; red: upregulated genes; blue: downregulated genes). (C) Spherically budding tests of GelMA and BG-GelMA microspheres. (D) Angiogenesis-related KEGG enrichment pathways analysis of GelMA and BG-GelMA microspheres. (E) Angiogenesis-related GO enrichment analysis of GelMA and BG-GelMA microspheres. (F) Quantitative statistical analysis of sprouting length and number of spheroids. (G) Heatmap of the expression of angiogenesis-related genes in GelMA and BG-GelMA microspheres. (H, I) Relative gene expression and Western-blot analysis of KDR and DLL4. (∗P < 0.05, ∗∗P < 0.005, ∗∗∗P < 0.001, ∗∗∗∗P < 0.0001).

To further validate the angiogenic gene expression profiles, a heatmap analysis was performed to visualize the differential expression patterns between the GelMA and BG-GelMA groups (Fig. 4G). Among the 100 differentially expressed genes related to angiogenesis, 20 common and representative genes were selected for demonstration, of which 16 were up-regulated and 4 were down-regulated. It was worth noting that in the BG-GelMA group, the expression levels of key regulatory factors VEGFA, involved in the proliferation and migration of endothelial cells, as well as Delta Like Canonical Notch Ligand 4 (DLL4), a key regulator involved in angiogenic sprouting, were upregulated. Based on this result, further spheroidal budding tests were conducted to verify these findings at a functional level. Refer to the experimental method in the paper and improve it [21]. The results showed that HUVECs cultured with BG-GelMA microspheres showed significantly enhanced budding formation compared with the GelMA group (Fig. 4C). Quantitative measurement showed that the length and number of budding increased, which was consistent with sequencing data (Fig. 4F). Kinase Insert Domain Receptor (KDR) and DLL4 play key regulatory roles in the vascular sprouting process [22,23]. The expression levels of KDR and DLL4 were analyzed by qRT-PCR and Western-blot (Fig. 4H and I, Table S1). qRT-PCR results showed that KDR and DLL4 in the BG-GelMA group were significantly up-regulated compared with the GelMA group. Again, Westernblot results showed consistent results.

3.4. BG-GelMA microspheres promote angiogenesis by modulating the cytoskeleton

In further analysis, it was found that there were significant differences in the expression of cytoskeleton-related genes in the BG-GelMA group compared with GelMA group. Specifically, the BG-GelMA group showed up-regulation of genes involved in cell morphological change, migration, and microtubule remodeling. To further understand the potential role of these genes in angiogenesis, GO enrichment analysis of cytoskeleton-related genes was performed (Fig. 5A). All GO terms were categorized into three main domains: biological process, cellular component, and molecular function (Fig. 5B). Cytoskeletal staining and SEM were performed to confirm these molecular changes (Fig. 5C) more visually. Compared with GelMA, HUVECs in the BG-GelMA group exhibited enhanced stretching ability, the F-actin fibers of the cells were arranged in a more orderly manner, and the microtubule network showed a more complex extended structure. The SEM results also showed that the cells extended more filopodia on the surface of the BG-GelMA microspheres, suggesting that the cells had good adhesion and stretching ability on the material surface.

Fig. 5.

BG-GelMA microspheres promote angiogenesis by modulating the cytoskeleton. (A) Cytoskeleton-related GO scatter plot of GelMA and BG-GelMA microspheres. (B) Cytoskeleton-related GO barplot of GelMA and BG-GelMA microspheres. (C) Cytoskeleton staining and SEM of GelMA and BG-GelMA microspheres. (D) Immunofluorescence images of Rac1 and CDC42 staining. Rac1 (red), CDC42 (green), Dapi (blue). (E) Western-blot analysis of RhoA, Rac1 and CDC42. (F) Quantitative analysis of Rac1, CDC42 and RhoA relative protein expression. (G) Quantitative analysis of Rac1 and CDC42 relative gene expression. (∗P < 0.05, ∗∗P < 0.005, ∗∗∗P < 0.001, ∗∗∗∗P < 0.0001).

Analysis of cytoskeleton-related genes showed a high concentration of genes in the Rho family of small GTPases (Fig. S3). The Rho family of small GTPases Cell division cycle 42 (CDC42) and Ras-related C3 botulinum toxin substrate 1 (Rac1) are key regulators of cytoskeletal dynamics. CDC42 regulates morphological changes in the cytoskeleton and regulates membrane transport functions involved in physiological processes such as cell growth and proliferation, cell viability, and cell polarity. Rac1 plays an important role in cytoskeletal remodeling. Rac1-GTP can promote new actin polymerization activity and the formation of lamellipodia, which drives cell frontend expansion and migration. Immunofluorescence results showed that BG-GelMA microspheres promoted the expression of CDC42 and Rac1 compared with GelMA group (Fig. 5D–S4). At the same time, a brighter fluorescence signal was observed at the front edge of the cells, indicating the accumulation of CDC42 in the front edge of the cells. This result suggested that CDC42 may be involved in dynamic cellular morphological changes and migration processes. qRT-PCR confirmed that the expressions of CDC42 and Rac1 were increased in BG-GelMA group compared with GelMA group (Fig. 5G–Table S1). Further, HUVECs were treated with CDC42 or Rac1 inhibitors that partially inhibit BG-GelMA-induced angiogenesis (Fig. S5). Immunofluorescence staining revealed reduced CD31 and VEGF levels upon CDC42 or Rac1 inhibition (Figs. S6A and C; Figs. S7A and C). Western-blot and qRT-PCR confirmed decreased protein and mRNA levels of angiogenesis-related factors (Figs. S6B and 7B; Fig. S8). These results further validate the critical roles of CDC42 and Rac1 in angiogenesis. Additionally, RhoA, a key member of the Rho GTPase family, is essential in regulating cytoskeletal dynamics, as well as cell adhesion and motility. The increased expression of CDC42 and Rac1 in BG-GelMA group may be related to RhoA signaling pathway. Western-blot confirmed that the expression and activation of RhoA, CDC42 and Rac1 were increased in the BG-GelMA group compared with the GelMA group (Fig. 5E and F).

3.5. Evaluation in vivo collateral vessel formation of BG-GelMA-SF

Double-layered silk fibroin (SF) artificial blood vessels can effectively mimic the structure and function of blood vessels. Viability assessment through fluorescent live/dead staining showed that the cells on the SF were evenly distributed, and most cells remained active, indicating that the SF artificial blood vessels had excellent biocompatibility and could effectively support cell attachment and growth (Fig. 6F). The HUVECs were cultured on microspheres for 3 days to form vascular buds. Then the vascular buds spheres were grafted into SF artificial blood vessels, which served as vasculogenic centers (Fig. 6A). SEM images showed the uniform porous network structure of the SF artificial blood vessels, which facilitated cell attachment and migration. In addition, GelMA and BG-GelMA vascular buds were uniformly distributed in the SF artificial blood vessels (Fig. 6B). The SF artificial blood vessels grafted with vascular buds were continued to be cultured for 24 h and then analyzed to CD31 immunofluorescence staining (Fig. 6E). The results showed that the cells in the vascular buds successfully migrated to the surface of the SF artificial blood vessels and formed vascular-like structures. In particular, the BG-GelMA vascular buds exhibited a stronger vasculogenic capacity compared with the GelMA vascular buds. SF, GelMA-SF, and BG-GelMA-SF artificial blood vessels were implanted subcutaneously in nude mice for 7 days to validate the ability of BG-GelMA vascular buds to act as vascular generative centers for vessel formation in vivo. The results showed significant differences in blood vessel formation among the three groups. As is shown in Fig. 6C, compared to SF and GelMA-SF artificial blood vessels, BG-GelMA-SF demonstrated a significant ability to support collateral vessel formation, which was reflected in enhanced vascular germination and vascular network development from BG-GelMA vascular buds. The angiographic results showed that more collateral vessels in the BG-GelMA-SF group were filled with contrast agents. Three-dimensional vascular architecture was analyzed using Micro-CT imaging for comprehensive evaluation of vascular network distribution patterns. The collateral vessel formed around the BG-GelMA-SF artificial blood vessels show better structural integrity and higher density, which can be quantitatively evaluated by analyzing blood vessel volume, blood vessel thickness, blood vessel number and blood vessel density (Fig. 6D–S10).

Fig. 6.

Characterization of BG-GelMA-SF and collateral vessel formation in vivo. (A) SF artificial blood vessel grafted vascular buds and in vivo implantation process. (B) SEM of SF, GelMA-SF and BG-GelMA-SF artificial blood vessels. (C) Macroscopic examination of vascular constructs with adjacent host tissue integration (delineated area within black dashed circle indicates the implanted vascular graft region; Micro-CT analysis of 3D reconstruction images of artificial blood vessels and surrounding tissues (the yellow dotted circle is the artificial blood vessel area). (D) Quantitative analysis of volume and thickness of blood vessel formation. (E) CD31 immunofluorescence staining of vascular buds as germinal center. CD31 (red), Dapi (blue). (F) Live/dead staining of SF artificial blood vessels. (∗P < 0.05, ∗∗P < 0.005, ∗∗∗P < 0.001, ∗∗∗∗P < 0.0001).

To further investigate the vascular microstructure on the surface of the artificial blood vessel, tissue sections were examined. HE staining of the resected tissue 7 days after implantation observed a circular structure of monolayer cells, indicating neovascularization (Fig. 7A). In addition, the number of functional vessels in the BG-GelMA-SF group was significantly increased, the endothelial membrane was intact, and the lumen structure was clear. The key markers of endothelial cell function and angiogenesis, CD31 and VEGF, were further stained by immunofluorescence. The results demonstrated that CD31 and VEGF were significantly upregulated in the BG-GelMA-SF group, with tubular patterns of green fluorescence distribution, indicating enhanced angiogenesis signals within the tissue. To investigate the role of small GTPase in endothelial cell dynamics, expressions of CDC42 and Rac1 were detected. Immunofluorescence results showed that CDC42 and Rac1 levels were increased in the BG-GelMA-SF group, indicating that BG-GelMA angiogenesis globule has been acting as an angiogenic center, activating endothelial cell proliferation, angiogenic signaling, and cytoskeletal remodeling pathways to stimulate neovascularization (Fig. 7B). The results of immunofluorescence were further quantitatively analyzed (Fig. 7C). In addition, KDR and CD34 staining showed strong positive signals for neovascularization endothelial cells, further supporting the vascularization role of BG-GelMA (Fig. S9). Immunohistochemical (IHC) staining results further corroborated the immunofluorescence findings (Fig. 7D). In the BG-GelMA-SF group, CD31 and VEGF IHC staining showed many positive cells around the newly formed blood vessels, which is consistent with the observed enhanced angiogenesis. For small GTPases, the IHC results for CDC42 and Rac1 were also aligned with the immunofluorescence data.

Fig. 7.

BG-GelMA-SF promotes collateral vessel formation in vivo. (A) HE staining of histological images of the perivascular tissue (general images of transverse section and local enlarged images). (B) Immunofluorescence staining for CD31, VEGF, Rac1 and CDC42 in vivo. (C) Immunofluorescence quantitative analysis of CD31, VEGF, Rac1 and CDC42. (D) IHC staining of CD31, VEGF, Rac1 and CDC42. (∗P < 0.05, ∗∗P < 0.005, ∗∗∗P < 0.001, ∗∗∗∗P < 0.0001).

4. Discussion

In this study, inspired by grafting techniques commonly used in plant biology, a new strategy of forming collateral vessels with artificial blood vessels was introduced. As grafting vibrant tree buds onto bare trunks can rapidly form side branches, we successfully constructed a vascular buds model by inoculating HUVECs into bioactive hydrogel microspheres. Vibrant cells in vascular buds play a central role in initiating and promoting collateral vessel growth. The grafting of vascular buds onto artificial blood vessels creates a dynamic environment for the rapid growth of collateral vessels to induce collateral vessel formation during massive ischemia thereby improving the overall ischemic state.

Ischemic disease is characterized by insufficient blood supply to the tissue. At the same time, the ability of tissue to form new blood vessels and revascularization is often impaired [[24], [25], [26]]. The key to treatment is vascular regeneration and perfusion recovery. Although SF is an ideal material for artificial blood vessels, there are still some problems such as thromboembolism and thrombosis, compliance mismatch, and neointimal hyperplasia [27,28]. Introducing vascular buds into the SF artificial blood vessel wall just solves this drawback. The study innovatively focuses on the potential of collateral vascularization in artificial blood vessels to promote angiogenesis in the local microenvironment. Vascular buds are supported by hydrogel microspheres. In addition to serving as excellent carriers for cells, hydrogel microspheres can mimic the natural extracellular matrix microenvironment, offering favorable conditions for cell growth and differentiation [[29], [30], [31]]. The angiogenic properties of BG have been widely recognized. However, BG extracts have been used in vitro for cell studies in most previous studies [32]. In contrast, 3D environment better simulate in vivo conditions but have not been widely explored. BG-GelMA microspheres provide a supportive 3D matrix for endothelial cells, with BG's sustained ion release activating signaling pathways to promote proliferation, migration, and lumen formation, enhancing vascularization.

VEGF is a key angiogenesis regulator that binds to KDR (VEGFR2), triggering receptor dimerization and autophosphorylation, thereby driving endothelial cell proliferation, survival, migration, and vascular permeability, ultimately promoting angiogenesis [33,34]. Sprouting is a form of angiogenesis that involves the proliferation, migration of vascular endothelial cells and the formation of new blood vessels [35]. In this experiment, BG-GelMA microspheres enhanced vascular buds formation, producing more and longer buds in vitro. Molecular analysis revealed increased expression of KDR and DLL4, indicating their synergistic role in vascular budding regulation [36]. BG-GelMA microspheres activated VEGF/KDR signaling to promote cell proliferation and migration, while DLL4-Notch interaction inhibited excessive endothelial cell activity, enhancing vessel differentiation and maturity. These results demonstrate that BG-GelMA microspheres create a favorable microenvironment for initiating and stabilizing angiogenic buds.

The cytoskeleton promotes angiogenesis by mediating cell polarity, membrane protrusion, and cell-matrix interactions [37]. Small GTPases play a critical role in cytoskeletal reorganization and are essential for the formation of cellular structures that regulate endothelial cell motility and vascular sprouting [38]. Among them, Rac1 and CDC42 synergistically regulate cytoskeletal reorganization and cell polarity to drive leading edge extension and directional sensing. Rac1 promotes actin polymerization and lamellipodia formation, while CDC42 induces filopodia formation and establishes cell polarity, ensuring directional migration [39]. The roles of CDC42 and Rac1 in endothelial cell migration and angiogenesis reveal their molecular mechanisms in the pro-angiogenic effects of BG-GelMA microspheres. BG-GelMA treatment significantly enhanced CDC42 and Rac1 activities to promote cell migration and lumen formation, suggesting that cytoskeletal remodeling drives microsphere angiogenic potential. Combined with bioinformatics analyses, we hypothesised that in the BG-GelMA microsphere environment, VEGF-KDR interactions activate key pathways (PI3K/Akt and MAPK/ERK, among others) that drive endothelial cell morphology changes, migration and angiogenesis.

In this experiment, BG-GelMA vascular buds were transplanted onto SF artificial blood vessels to provide a dynamic microenvironment for the rapid growth of collateral vessels, thereby promoting collateral vessel formation during massive ischemia and thus improving the overall ischemic state. In our study, we observed the ability of vascular buds to form vascular collateral branches early in the body. Considering that grafting vascular buds may damage the integrity of the artificial vessels, we did not infuse the blood flow immediately. By demonstrating robust collateral vessel formation without perfusion, we established a foundation for safely introducing blood flow after collateral circulation stabilizes. In subsequent studies, the maturation, function and stability of neovascularization and the degradation of microspheres in vivo will be focused on.

5. Conclusion

In this study, inspired by grafting techniques of tree, a new strategy was introduced to promote collateral vessel formation. Vascular buds were prepared by inoculating HUVECs into BG-GelMA microspheres. The vascular buds were grafted onto SF artificial blood vessels, which provided a dynamic environment for rapid growth of collateral vessels during ischemia. In addition, BG-GelMA microspheres further promoted angiogenesis by regulating cytoskeletal reorganization and dynamic changes, promoting cell migration and morphological changes. Therefore, BG-GelMA vascular buds is an effective method to construct the collateral circulation of artificial vessels, which has the potential to be used in tissue ischemia.

CRediT authorship contribution statement

Yulian Yang: Writing – original draft, Methodology, Formal analysis, Data curation. Yonghao Qiu: Visualization, Methodology, Investigation, Formal analysis. Shijing Xu: Formal analysis, Data curation. Huichang Gao: Visualization, Software. Chunhui Wang: Resources, Formal analysis. Haohui Huang: Formal analysis, Data curation. Zhengyu Yang: Software, Formal analysis. Xiaofeng Chen: Supervision, Project administration. Fujian Zhao: Writing – review & editing, Project administration.

Ethics approval and consent to participate

The study was approved by the ethics committee (Full name: GuangDong Pharmaceutical University Experimental Animal Ethics Committee Inspection) (Reference number: gdpulacspf2022619), GuangDong Pharmaceutical University, China.

Declaration of competing interest

The authors declared that they have no conflicts of interest to this work.

Appendix A. Supplementary data

The following is the Supplementary data to this article: