A Microcarrier Bead-Based Three-Dimensional Angiogenesis Model Using Human Umbilical Vein Endothelial Cells

Summary

This protocol describes a microcarrier bead-based three-dimensional angiogenesis model using human umbilical vein endothelial cells. The assay allows direct visualization of endothelial sprouting and lumen-like structure formation and can be applied to studies of angiogenic mechanisms and therapeutic interventions.

Abstract

Angiogenesis is a multistep biological process involving endothelial cell activation, migration, elongation, lumen formation, branching, and anastomosis. Conventional two-dimensional culture systems are useful for the preliminary evaluation of drugs, genes, or culture conditions, but they do not adequately recapitulate the spatial organization of endothelial morphogenesis within a three-dimensional extracellular matrix. The present protocol describes a microcarrier bead-based three-dimensional angiogenesis assay using human umbilical vein endothelial cells (HUVECs). The HUVECs are first seeded onto microcarrier beads bearing chemically coupled acid-denatured porcine type I collagen and then embedded in a basement membrane matrix. Oral mucosal fibroblasts are subsequently seeded on top of the gel to provide paracrine support for endothelial sprouting and vascular-like structure formation. The culture medium is replaced every 2 days, and sprouting behavior and angiogenic phenotypes are monitored by inverted microscopy. Under appropriate culture conditions, endothelial sprouts emerge within several days, followed by the formation of lumen-like structures and inter-sprout connections. This method provides a convenient and reproducible platform for in vitro studies of endothelial morphogenesis and can be further applied to investigations of gene manipulation, pharmacological intervention, and disease modeling.

Introduction

Endothelial sprouting is a key event in angiogenesis and involves the emergence, migration, proliferation, and alignment of endothelial cells from pre-existing vascular structures, followed by lumen formation and the establishment of interconnected vascular-like networks^1,2. Although conventional two-dimensional culture systems and gel-based tube formation assays are technically straightforward, they are limited in their ability to recapitulate the spatial remodeling of endothelial cells within a three-dimensional microenvironment^3,4. The classical microcarrier bead-based three-dimensional sprouting assay reproduces several stages of angiogenesis in vitro, including endothelial sprouting, elongation, lumen formation, and network assembly, and has therefore become a widely used method for angiogenesis research.

In the present protocol, a three-dimensional microcarrier bead-based sprouting model using normal human umbilical vein endothelial cells is established. Based on previously reported endothelial microcarrier bead sprouting assays^5,6,7, this method preserves the basic framework of endothelial cell coating onto microcarrier beads, embedding in a three-dimensional matrix, and fibroblast overlay culture, while substituting a basement membrane matrix for the conventional fibrin gel to simplify the procedure and facilitate routine imaging. The basement membrane matrix was selected in this modified protocol for several practical and biological reasons. Experimentally, it provides a ready-to-use three-dimensional matrix that allows rapid gel formation, simple bead embedding, and convenient routine microscopic observation without the additional polymerization steps required for fibrin gel preparation. Biologically, the basement membrane matrix contains extracellular matrix components that support endothelial cell adhesion, migration, sprout extension, and lumen-like structure formation. Therefore, this modification is suitable for studies requiring direct visualization of endothelial sprouting behavior, comparison of angiogenic phenotypes, and evaluation of pharmacological or genetic interventions in a three-dimensional culture environment. However, because the basement membrane matrix differs from fibrin gel in matrix composition, mechanical properties, degradability, and bioactive cues, the present assay should be interpreted as a modified microcarrier bead-based sprouting model rather than a direct replacement of the classical fibrin-based assay. These matrix-related differences may influence sprouting kinetics, lumen stability, and network morphology, and should be considered when comparing results across different three-dimensional angiogenesis systems. This model is suitable for observing endothelial sprouting behavior in three dimensions and can be further applied to studies of pharmacological intervention, gene manipulation, and angiogenic mechanisms.

Protocol

Human umbilical vein endothelial cells used in this protocol were isolated and cultured from human umbilical cord tissues collected from the Department of Obstetrics, The First People’s Hospital of Yunnan Province. Written informed consent was obtained from all donors prior to sample collection. All procedures involving human-derived materials were approved by the Medical Ethics Committee of The First People’s Hospital of Yunnan Province (approval no. KHLL2023-KY198). Human oral mucosal fibroblasts (HOMFs) were kindly provided by the research group of Dr. Jiemei Zhai at the Affiliated Stomatological Hospital of Kunming Medical University. The reagents and the equipment used are listed in the Table of Materials.

1. Cell preparation

1. Culture human umbilical vein endothelial cells (HUVECs) in endothelial cell growth medium supplemented with additional fetal bovine serum (FBS) to a final concentration of 10% at 37 °C in a humidified incubator with 5% CO₂. Cells used for the assay should be free of contamination, exhibit stable growth, and be in the logarithmic growth phase. Use cells at passage 4–6.
2. Culture human oral mucosal fibroblasts (HOMFs) in high-glucose basal medium supplemented with 10% fetal bovine serum (FBS) at 37 °C in a humidified incubator with 5% CO₂. Use cells at passage 4–6.
3. Before microcarrier bead coating, confirm that both cell types are at an appropriate confluence. Avoid overconfluent cultures or excessive enzymatic digestion, as these may adversely affect subsequent sprouting.

2. Preparation of microcarrier beads

1. Hydrate 0.5 g of dry  microcarrier beads in 50 mL of PBS for at least 3 h. After washing, resuspend the beads in 50  mL of PBS to obtain a 10 mg/mL microcarrier bead suspension, corresponding to approximately 3 × 10⁴ beads/mL.
2. Transfer the  suspension to a siliconized glass bottle, sterilize it by autoclaving at 115 °C for 15 min, and store it at 4 °C until use.
3. Before endothelial cell coating, remove the storage solution and wash the beads with sterile PBS. Equilibrate the beads in endothelial cell growth medium at 37 °C for at least 30 min.
4. Gently resuspend the equilibrated beads immediately before use to avoid sedimentation and bead aggregation.
5. Before cell coating, examine a small aliquot of the bead suspension under an inverted microscope. Proceed only when the beads are fully hydrated, evenly suspended, and free of visible aggregates or fragmented particles.

3. Coating endothelial cells onto microcarrier beads

1. Detach HUVECs using trypsin and resuspend the cells in endothelial cell growth medium.
2. Coat the microcarrier beads at a density of 1,500 cells per bead. Mix 3 × 10⁶ HUVECs with 2,000 microcarrier beads in 1.5 mL of endothelial cell growth medium.
3. Transfer the cell-bead mixture into a microcentrifuge tube and incubate for 12 h in a cell culture incubator. Throughout the full 12 h incubation period, gently invert the tube once every 20 min to maintain adequate contact between the cells and beads and to promote uniform endothelial cell attachment.
4. After 12 h, transfer the coated beads into a 25 cm² cell culture flask containing 4 mL of endothelial cell growth medium and continue incubation overnight to allow firm attachment of endothelial cells to the microcarrier bead surface.
5. Before embedding, examine representative coated beads under an inverted microscope. Proceed to matrix embedding only when more than 80% of the examined beads show a continuous and relatively uniform endothelial cell layer on the bead surface, with minimal free-floating cells, no obvious cell clumps, and no severe bead aggregation. Beads with uneven coating, large cell clusters, or extensive aggregation should not be used for embedding.
6. If the beads adhere to the bottom of the flask, gently shake the flask to release them. Avoid vigorous pipetting, which may cause bead aggregation or cell detachment.

4. Embedding the coated beads in a three-dimensional matrix

1. Transfer the endothelial cell-coated microcarrier beads into a 15 mL centrifuge tube and allow the beads to settle by gravity. Remove the excess medium and add an appropriate volume of fresh medium.
2. Mix the microcarrier bead suspension with prechilled basement membrane matrix at a 1:1 (v/v) ratio to obtain a final bead concentration of 500 beads/mL.
3. Add 100 µL of the matrix-bead suspension to the center of each well of a 24-well plate, allowing the mixture to form a single dome-shaped gel droplet without contacting the wall of the well. Gently mix the suspension immediately before dispensing and avoid introducing air bubbles.
4. Leave the 24-well plate at room temperature for 5 min to allow initial gelation and stabilization of the gel droplet. Then incubate the plate upright at 37 °C in 5% CO₂ for 15 min, followed by an additional 15 min incubation in the inverted position to ensure complete gelation and proper gel formation.
5. After the gel has fully solidified, proceed immediately to fibroblast overlay culture.

5. Fibroblast overlay culture

1. Resuspend 3 × 10⁴ human oral mucosal fibroblasts (HOMFs) in 1 mL of endothelial cell growth medium and gently add the cell suspension along the wall of each well, allowing the cells to distribute over the gel surface without directly disrupting the gel dome.
2. Continue incubation in endothelial cell growth medium at 37 °C in a humidified incubator with 5% CO₂. Human oral mucosal fibroblasts are used as supporting cells because they provide stromal-derived paracrine signals that promote endothelial sprout extension, stabilization, and vascular-like remodeling in the three-dimensional culture system.

6. Culture maintenance and microscopic observation

1. Replace the culture medium every 2 days. During medium changes, slowly aspirate the spent medium and add fresh culture medium along the wall of the well to avoid disturbing the gel.
2. Observe the cultures daily or every other day under an inverted microscope using bright-field or phase-contrast mode. For longitudinal quantitative analysis, select multiple fields of view in each well on day 7 and photograph the same predefined fields on days 9 and 11.
3. Under optimal culture conditions, obvious endothelial sprouting is typically observed on days 4–6, lumen-like structures appear on days 5–7, and interconnected microvascular-like networks gradually form after day 7.
4. Define successful sprouting as the presence of multiple endothelial extensions growing radially from the bead surface into the surrounding three-dimensional matrix. A sprout should be counted only when it clearly extends beyond the bead surface and remains connected to the bead-associated endothelial cell layer.
5. Define lumen-like structure formation as the appearance of elongated endothelial structures with a visible central clear space or tube-like morphology. Inter-sprout connections are defined as direct contacts or continuous structures formed between adjacent sprouts.
6. Cultures with extensive cell death, severe bead aggregation, disrupted gel structure, abundant bubbles, matrix collapse, or only short, irregular cell protrusions should be considered suboptimal and excluded from quantitative analysis.

7. Image acquisition and quantitative analysis

1. Select multiple fields of view in each well on day 7 and acquire images from the same predefined fields on days 9 and 11. Images should be collected using identical magnification, exposure settings, and imaging conditions at all time points.
2. Quantify the number of sprouts per microcarrier bead and the vascularized area.
   1. For vascularized area measurement, use open-source image analysis software. Open the acquired image and manually delineate the boundary of the vascularized region of interest (ROI) using the polygon or freehand selection tool. Add the selected ROI to the ROI manager.
   2. For visualization, use a region-of-interest visualization plugin to display the outlined ROI. Enable area measurement in the measurement settings and measure the selected ROI to obtain the vascularized area. Apply the same analysis settings to all images.
   3. For angiogenic rate analysis, use the vascularized area on day 7 as the baseline, and calculate the relative change in vascularized area on days 9 and 11.
3. Count the number of individual sprouts independently by three observers, and use the average value for analysis.

Results

Angiogenesis in the three-dimensional HUVEC model

In successfully established cultures, endothelial cells are uniformly distributed on the surface of the microcarrier beads, and no obvious bead aggregation is observed. Dynamic observation showed progressive endothelial sprouting and microvascular-like network formation on days 7, 9, and 11 (Figure 1A). Quantitative analysis further confirmed the progressive angiogenic response in this model. Using day 7 as the baseline, the angiogenic rate was set to 0% on day 7 and reached 66.94% ± 18.49% on day 9 and 169.51% ± 100.83% on day 11, indicating progressive expansion of the vascularized area during culture (Figure 1B). Representative bright-field imaging showed that endothelial sprouts emerged radially from the bead surface into the surrounding three-dimensional matrix after 7 days of culture (Figure 1C). In addition, confocal microscopy provided an overview of the gross morphology and spatial organization of the three-dimensional culture system (Figure 1D). The average number of sprouts per microcarrier bead was 31.3 sprouts/bead, based on independent counts from three observers (Figure 1E). These findings indicate that the model supports endothelial sprouting, elongation, lumen-like morphology, and vascular-like remodeling in a three-dimensional environment.

Suboptimal outcomes in the three-dimensional HUVEC model

In failed or suboptimal cultures, several characteristic abnormalities may be observed, including cell death (Figure 2A), rupture of the microcarrier beads (Figure 2B), abnormal cell aggregation on the bead surface (Figure 2C), extensive bubble formation within the gel (Figure 2D), local collapse of the matrix (Figure 2E), excessively high bead density (Figure 2F), uneven cell coating on the microcarrier beads (Figure 2G), bead aggregation (Figure 2H), and either no obvious sprouting or only limited sprout formation (Figure 2I,J). These abnormalities suggest that successful establishment of the model may be influenced by inadequate cell viability, insufficient cell coating, improper temperature control of the basement membrane matrix, overly vigorous resuspension, or disturbance of the gel during medium replacement.

Figure 1: Angiogenesis in the three-dimensional HUVEC model. (A) Dynamic assessment of angiogenesis in the three-dimensional HUVEC model on days 7, 9, and 11. Images were acquired using 20× magnification. Scale bar = 10 µm. (B) Quantitative analysis of the angiogenic rate. The angiogenic rate was defined as the relative change in vascularized area over time on days 7, 9, and 11, using day 7 as the baseline. It was calculated as: angiogenic rate = [(vascularized area at the indicated time point − vascularized area on day 7) / vascularized area on day 7] × 100%. Angiogenic rate data are presented as mean ± SD from three independent experiments. (C) Representative image of endothelial sprouting in HUVECs after 7 days of culture. Image was acquired using 20× magnification. Scale bar = 10 µm. (D) Confocal overview image showing the gross morphology of the three-dimensional culture system. Image was acquired using a 4× scan. Scale bar = 500 µm. (E) Quantitative analysis of sprout number in HUVECs. The number of sprouts was counted independently by three observers, and the average value was used for analysis. No statistical comparison was performed for these descriptive quantitative analyses. The endothelial sprouting structures shown in this figure are derived from HUVEC-coated microcarrier beads; HOMFs were used as overlay supporting cells and were not specifically labeled in the images. Please click here to view a larger version of this figure.

Figure 2: Suboptimal outcomes in the three-dimensional HUVEC model. (A) Cell death. As indicated by the white arrow, the cells appear as punctate debris, and the surface of the microcarrier bead becomes smooth. Scale bar = 50 µm. (B) Rupture of the microcarrier bead, as indicated by the white arrow. Scale bar = 50 µm. (C) Cell aggregation on the microcarrier bead surface. As indicated by the white arrow, the cells exhibit a bead-like clustered appearance. Scale bar = 50 µm. (D) Extensive bubble formation within the gel, as indicated by the white arrow. Scale bar = 100 µm. (E) Local collapse of the matrix, as indicated by the white arrow. Scale bar = 10 µm. (F) Excessively high microcarrier bead density. Scale bar = 10 µm. (G) Uneven cell coating on the microcarrier bead. As indicated by the white arrow, one side of the bead surface remains smooth. Scale bar = 50 µm. (H) Bead aggregation. Scale bar = 50 µm. (I,J) No obvious sprouting or only limited sprout formation. Scale bar = 10 µm. Please click here to view a larger version of this figure.

Figure 3: Schematic illustration of the experimental workflow. (A) Preparation of HUVECs, HOMFs, and microcarrier beads bearing chemically coupled acid-denatured porcine type I collagen under appropriate conditions before assay setup. (B) Mixing of HUVECs with the prepared microcarrier beads to promote endothelial cell attachment to the bead surface. (C) Bead-coating incubation with gentle inversion every 20 min for 12 h to promote uniform endothelial cell attachment. (D) Transfer of coated beads to a culture flask for further incubation to allow firm attachment of endothelial cells to the microcarrier bead surface. (E) Embedding of HUVEC-coated microcarrier beads in a basement membrane matrix, followed by seeding of HOMFs on top of the gel to provide stromal-derived paracrine support for endothelial sprouting and vascular-like remodeling. (F) Microscopic observation and quantitative analysis of endothelial sprouting, lumen-like structure formation, inter-sprout connections, sprout number, and vascularized area on days 7, 9, and 11. Please click here to view a larger version of this figure.

Discussion

The present protocol describes a modified three-dimensional endothelial sprouting model based on microcarrier beads. Figure 3 provides a schematic overview of the experimental workflow. The protocol was adapted from the classical angiogenesis assay reported by Nakatsu and Hughes and optimized to meet the objectives of the present study^8,9. In the classical system, endothelial cells are coated onto collagen-coated microcarrier beads, embedded in a three-dimensional matrix, and co-cultured with fibroblasts to support sprouting, elongation, lumen formation, and vascular-like anastomosis^7,10. In the present protocol, a basement membrane matrix was used in place of fibrin gel, and human oral mucosal fibroblasts were used as supporting cells, thereby establishing a convenient three-dimensional culture system for in vitro observation of endothelial angiogenic behavior. The sprouting structures shown in the representative images are derived from HUVEC-coated microcarrier beads, whereas human oral mucosal fibroblasts are seeded as overlay supporting cells and are not specifically labeled in these images.

An important feature of this protocol is the use of a basement membrane matrix instead of the classical fibrin gel system. Experimentally, the basement membrane matrix simplifies assay setup because it is ready to use, undergoes rapid temperature-dependent gelation, and allows convenient embedding of endothelial cell-coated microcarrier beads without the need for separate fibrinogen, thrombin, or aprotinin preparation. This feature makes the protocol relatively simple and suitable for routine microscopic observation. However, this modification also changes the biological microenvironment of the assay. Unlike fibrin gel, the basement membrane matrix contains basement membrane-associated extracellular matrix components and bioactive cues that may influence endothelial adhesion, migration, sprout extension, lumen-like morphology, and network remodeling. In addition, differences in matrix stiffness, degradability, and ligand composition may affect sprouting kinetics and network morphology. Therefore, results obtained using this modified assay should be interpreted as outcomes of a basement membrane matrix-based microcarrier bead sprouting model rather than as a direct equivalent of the original fibrin-based assay.

Successful establishment of this model depends on several critical steps. Uniform endothelial cell coating on the microcarrier bead surface is essential for consistent sprout initiation and extension. Inadequate or uneven coating may result in reduced sprout number, poor directional growth, and decreased reproducibility. Stable gel formation is another key determinant of assay performance. During matrix preparation and embedding, improper temperature control may lead to premature gelation, uneven bead distribution, or bubble formation, all of which can interfere with endothelial sprouting. In addition, supporting fibroblasts are required to provide sustained paracrine signals that promote sprout growth and stabilization. Careful control of these steps is therefore essential for obtaining stable and reproducible results.

Several measures may improve assay consistency. The representative suboptimal outcomes shown in Figure 2 can be used as practical troubleshooting indicators. Extensive cell death may result from poor cell viability, overdigestion during cell detachment, inappropriate passage number, or delayed medium replacement; therefore, healthy logarithmic-phase endothelial cells and gentle enzymatic digestion are recommended. Rupture of microcarrier beads or bead aggregation may be caused by vigorous pipetting, excessive mechanical agitation, or inadequate bead resuspension, and can be reduced by gentle mixing and careful handling during bead preparation and coating. Uneven endothelial cell coating or large cell clusters on the bead surface usually indicate insufficient mixing, excessive cell density, or inadequate bead-cell contact during the coating period; in this case, the cell-bead suspension should be gently inverted at regular intervals and examined before embedding. Bubble formation and local matrix collapse are commonly related to improper matrix handling, premature gelation, or disturbance of the gel during medium addition or replacement; keeping the matrix on ice before use, avoiding air bubbles during dispensing, allowing complete gelation, and adding medium slowly along the well wall can reduce these problems. Cultures showing no obvious sprouting or only limited sprout formation may reflect poor endothelial cell attachment, low cell viability, insufficient fibroblast-derived support, or unstable matrix formation, and should be excluded from quantitative analysis.

Compared with conventional two-dimensional tube formation assays, this three-dimensional microcarrier bead-based model more effectively recapitulates the spatial behavior of endothelial cells within an extracellular matrix and allows visualization of endothelial sprouting, elongation, lumen-like structure formation, and network assembly. In addition, the three-dimensional culture environment provides conditions that more closely resemble in vivo nutrient and signaling distribution^11,12. Compared with microfluidic vascular chip systems, which are often technically demanding and costly, this method is relatively simple, reproducible, and cost-effective, making it well-suited for routine in vitro studies of angiogenesis.

One important limitation of this protocol is that a basement membrane matrix was used as a substitute for the classical fibrin gel system. For this reason, the present assay should be regarded as a modified version of the previously reported endothelial microcarrier bead sprouting model rather than a direct replication of the original fibrin-based method. Because different matrices differ in composition, mechanical properties, and degradation behavior, they may influence sprouting kinetics, lumen stability, and vascular network morphology^13,14,15. Therefore, caution is needed when comparing results obtained with this model to those generated using conventional fibrin-based systems.

This model has potential applications beyond the observation of basal endothelial sprouting. It may also be adapted for pharmacological intervention studies, gene overexpression or knockdown experiments, signaling pathway analysis, and investigations of disease-associated endothelial phenotypes^16,17. With further optimization, this three-dimensional sprouting assay may serve as a reliable in vitro platform for studying mechanisms of vascular development and therapeutic strategies targeting angiogenesis.

Materials

List of materials used in this article

Name	Company	Catalog Number	Comments
1.5 mL microcentrifuge tube	Beyotime	FTUB306	Clear, nuclease-free
15 mL centrifuge tube	Haier	311005	Conical-bottom, sterile, bag-packed
24-well cell culture plate	Haier	612003	Flat-bottom, tissue culture-treated, individually wrapped, for adherent cell culture
Automated cell counting slide	Countstar	12-0005-50	Disposable slide for automated cell counter, 50 slides/box
CO2 cell culture incubator	Thermo Fisher Scientific	3110	Used for cell culture at 37 °C with 5% CO2
Cytodex 3 microcarriers	Cytiva	17048503	Dry powder; collagen-coated dextran microcarrier beads
Dotted Line plugin	N/A	N/A	Region-of-interest visualization plugin used for displaying outlined vascularized areas
Dulbecco’s Modified Eagle Medium (DMEM), high glucose	EvaCell (EVA Life Sciences)	E2102	DMEM basic (1×), containing 4.5 g/L D-glucose, L-glutamine, and 110 mg/L sodium pyruvate, 500 mL
Endothelial Cell Medium (ECM)	ScienCell Research Laboratories	1001	Endothelial cell culture medium supplemented with additional FBS to a final concentration of 10% before use
Fetal bovine serum (FBS)	Gibco	10099141C	Qualified, single-use format, Australia origin, 500 mL
Human oral mucosal fibroblasts (HOMFs)	Jiemei Zhai research group, The Affiliated Stomatological Hospital of Kunming Medical University	N/A	Primary fibroblasts isolated from human oral mucosa; kindly provided by another laboratory
Human umbilical vein endothelial cells (HUVECs)	In-house	N/A	Primary endothelial cells isolated from human umbilical veins; no catalog number
ImageJ software	National Institutes of Health (NIH), USA	N/A	Open-source software; version 1.54p
Inverted microscope	Olympus, Japan	N/A	For bright-field, phase-contrast, and fluorescence imaging
Laser scanning confocal microscope	Nikon, Japan	N/A	For confocal imaging;model not specified
Matrigel basement membrane matrix	Corning	354262	High concentration (HC), phenol red-free, LDEV-free, 10 mL
Penicillin-Streptomycin (Pen-Strep)	Gibco	15140-122	Penicillin-streptomycin solution for cell culture, 100×, 100 mL
Phosphate-buffered saline (PBS)	LiJi Bio	AC08L011	1×, sterile, 500 mL
T25 cell culture flask	ABCBIO	ABC707008	T25, vented cap, tissue culture-treated, for adherent cell culture
Trypsin-EDTA (0.25%)	Gibco	25200-056	0.25% trypsin-EDTA solution, 100 mL