Substrate‐Dependent Variability in Viability and Angiogenic Marker Expression Among Three Endothelial Cell Subtypes: Insights for Artificial Tissue Vascularization

ABSTRACT

Tissue engineering faces the challenge of achieving effective vascularization within tissue constructs for sustained viability and optimal function. The success of tissue‐engineered constructs depends on selecting an optimal angiogenesis‐stimulating ECM substitute material. This study compares four substrates made from three different biomacromolecules—fibrin, fibronectin, non‐crosslinked, and crosslinked gelatin, and their effect on endothelial cells. Acknowledging the diverse range of endothelial cells that play a role in (micro)vascularization, human endothelial primary cells, human umbilical vein endothelial cells, and human microvascular endothelial cells are subjected to these materials for evaluation. Biocompatibility is assessed by measuring cell viability (Live/Dead assay), metabolic activity (alamarBlue assay), morphology (actin staining), phenotype expression (immunocytochemistry), and the production of von Willebrand factor, which promotes angiogenesis by promoting cell adhesion and migration. The results show that the use of biomaterials as culturing substrates significantly impacts the viability and morphology of the cells. While the expression of angiogenic markers is shown to rely more on the cell lineage, the use of different substrates has an impact on the expression timeline. Thus, combining cells and biomaterials in a favorable manner can be used as a powerful tool for controlled vascularization in vitro, which requires the systematic assembly of different stimuli.

The study investigates how four ECM‐mimicking biomaterials—fibrin, fibronectin, and non‐crosslinked or crosslinked gelatin—affect different types of human endothelial cells. Cell viability, morphology, and angiogenic marker expression are evaluated. Results reveal substrate‐dependent effects on cell behavior and emphasize the importance of selecting compatible cell–biomaterial combinations to enable controlled vascularization in vitro through the strategic orchestration of biological cues.

1. Introduction

Tissue engineering (TE) is a multidisciplinary field involving materials, cells, and biochemical factors to construct tissue analogs or precursors that can be used to regenerate and repair damaged or diseased tissues and organs [1, 2, 3, 4, 5]. In “scaffold‐based approaches”, where extracellular matrix (ECM) substitutes are used, the selected material needs to fulfill specific criteria, such as promoting cell attachment and simulating the native structural integrity and (bio)mechanical support [6]. Cell engagement is crucial to developing the ECM for tissue construction, with signaling molecules promoting cell proliferation and differentiation and, thus, tissue development [7]. One of TE's biggest hurdles is the efficient vascularization of engineered tissues. As tissues grow larger and more complex, they require a network of blood vessels to deliver nutrients and oxygen and remove waste products. Establishing a functional and perfusable (micro)vascular network within engineered tissues remains a significant challenge [8].

Without a strong network of blood vessels, tissue cells can suffer from hypoxia [9], necrosis [10], and decreased functionality [11]. The lack of adequate vascularization also hinders the movement of immune cells, growth factors, and regulatory molecules essential for tissue repair and regeneration. Insufficient vascularization can lead to delays in healing, difficulties integrating new tissue, and decreased overall tissue function [12]. Artificial tissues lack similar functionality, necessitating solutions in the formation of effective perfusion in vitro. Current approaches aim to incorporate pre‐formed blood vessels or stimulate the formation of new blood vessels through angiogenesis or vasculogenesis [13].

Endothelial cells (ECs) are the main building block of vascularization, as they line the inner wall of both larger vessels and microvessels. Morphologically, they are flat cells that form a continuous endothelium. In addition to their barrier function, they also regulate the release of signaling molecules and control the transport of substances and (immune) cells through the vessel wall [14]. The challenge of microvascularization is so vital that it is considered a subfield of vascular TE. It aims to optimize ECM analogs and environmental conditions to drive the autonomous organization of hierarchically structured microvessels [15]. Since there are structural and functional differences between different endothelial cell subtypes and lineages [16, 17], three were included in this study.

Selecting appropriate materials to formulate these ECM analogs plays an important role in developing in vitro tissue models, regenerative medicine, and other biomedical research and applications. While specific requirements can vary, certain features are necessary in most applications. These include biocompatibility (the material does not trigger immune responses), bioactivity (promoting cell adhesion, proliferation, and differentiation), (bio)degradability that can be matched by cellular production of intrinsic ECM components, structural properties that mimic native tissues (and in tissue engineering application allow shaping and shape fidelity after the manufacturing process), etc. [15]. In microvascular tissue engineering, various materials have been studied for their ability to stimulate angiogenesis and vasculogenesis, and some have shown promising results, providing a balance of structural support and biochemical stimuli to promote the development of functional, integrated microvascular networks [15, 18, 19]. For this purpose, the cells were tested on several materials, including fibrin, fibronectin, and gelatin, which have been proposed due to their favorable biological and mechanical properties [20, 21, 22, 23, 24].

Fibrin plays a pivotal role in wound healing and blood clotting, providing an environment suitable for cell migration, proliferation, and differentiation. Furthermore, studies show that fibrin scaffolds are suitable for angiogenesis and vascular tissue engineering [25, 26, 27, 28]. Fibrin forms a network structure that serves as a substrate for the attachment and growth of ECs [27], which colonize and spread through the fibrin network and release growth factors that stimulate the formation of new blood vessels [28].

Another frequently used material in TE is fibronectin, a glycoprotein found in the ECM of various tissues. It is a key adhesion molecule, binding to both cells and other ECM components. Fibronectin is crucial in cell adhesion, migration, and tissue organization. It helps cells attach to the ECM and provides signals for cells to organize and form tissues. While fibrin forms the initial scaffold of a blood clot, fibronectin is an important component of the ECM that helps cells adhere and migrate during tissue repair [29, 30, 31].

Gelatin is a biocompatible polypeptide compound derived from collagen [32]. It has excellent properties for promoting the adhesion and growth of ECs since it contains integrin‐binding sites recognized by cell surface receptors present on ECs, called integrins [32, 33, 34, 35]. Compared to other protein‐derived polymers, gelatin is widely available, cost‐effective, relatively simple to functionalize, and is already commonly used in food, cosmetic, and pharmaceutical applications [32, 33, 34, 35]. The gelatin surface is generally hydrophilic, allowing ECs to attach and spread easily. In addition, gelatin can release incorporated bioactive agents such as growth‐promoting factors [33].

In order to find the optimal (co‐)culture conditions that promote angiogenesis and/or vasculogenesis, growth substrates need to, in addition to structural support, trigger angiogenic/vasculogenic growth factor production, as presented in Figure 1. The aim of this work is to study the influence of fibrin, fibronectin, and gelatin coatings on the development of three different types of endothelial cells. Specifically, the potential to induce the expression of angiogenesis‐specific markers in EPCs, HUVECs, HMEC‐1s, particularly the vascular endothelial growth factor receptor 2 (VEGFR‐2), and the potential to secrete signaling molecules, such as von Willebrand factor (vWF), were extensively compared.

FIGURE 1.

The concept of our work with the scope of this article in the highlighted part of the figure. This part is critical to developing adequate tissue‐engineered substitutes with sustained viability and optimal function (created with BioRender.com; accessed on 25 October 2024).

2. Materials and Methods

2.1. Cell Sources

Three different endothelial cell types were used in this study:

1. human endothelial primary cells; passage 2 (EPCs; donated by Prof. Tomaž Velnar, MD, PhD, University Medical Center Maribor; isolated as per protocol from Baudin et al. [36]); the approval of the ethical committee for this study was obtained through the Committee for Ethical Questions of the Faculty of Medicine, University of Maribor (approval ID: 038/2023/1‐401),

2. human umbilical vein endothelial cells; passage 2 (HUVECs; PCS‐100‐010; ATCC, Manassas, VA, USA),

3. human microvascular endothelial cells; passage 2 (HMEC‐1s; CRL‐3243; ATCC, Manassas, VA, USA).

Base cultures of EPCs and HMEC‐1s were maintained in Advanced Dulbecco's Modified Eagle's Medium (ADMEM, Gibco, Thermo Scientific, Waltham, MA, USA), supplemented with 5 v/v% fetal bovine serum (FBS, Gibco, Thermo Scientific, Waltham, MA, USA), at 37°C and 5% CO₂. Base culture of HUVECs was maintained in Endothelial Cell Basal Medium + Endothelial Cell Growth Supplement (500210 and 211‐GS, respectively; Sigma–Aldrich, Darmstadt, Germany), supplemented with 2 v/v% fetal bovine serum (FBS, Gibco, Thermo Scientific, Waltham, MA, USA), at 37°C and 5% CO₂.

2.2. Preparing the Materials

2.2.1. Coating the Well Plates with the Materials

Ultra‐pure water (18.2 MΩ/cm at 25°C) for all solutions was prepared using the ELGA Purelab water purification system (Veolia Water Technologies, High Wycombe, UK).

Well, plates were coated with four materials: non‐crosslinked gelatin (gelatin NC), crosslinked gelatin (gelatin C), fibrin, and fibronectin.

Gelatin NC was coated by adding 131.6 µL/cm² gelatin solution (1 mg/mL in ultra‐pure water; translating to 131.6 µg/cm²; 04055, Sigma–Aldrich, Darmstadt, Germany—the catalog number of this exact product has since been discontinued with suggester alternative G2625; Sigma–Aldrich, Darmstadt, Germany, as this product was the same type (type A) and has the same gel strength of 175 g Bloom) per well, as per manufacturer's protocol with coating concentrations based on the protocol's guidelines of 100–200 µg/cm² [37]. The coated well plates were left to dry overnight in the laminar flow hood. We acknowledge that the gelatin starts dissolving at 37°C and does not remain on the surface after hydration. In fact, non‐crosslinked gelatin coatings melt rapidly under physiological conditions—complete loss occurs within hours to 24 h, making them unsuitable as permanent scaffolds [38]. Crucially, however, even this transient coating delivers a high density of RGD motifs that engage integrins and trigger downstream adhesion‐mediated signaling—promoting endothelial survival, proliferation, and gene‐expression changes that endure well beyond the initial contact, which has been previously shown by multiple studies [39, 40, 41]. NC gelatin was used mainly to probe these early bioactive cues, whereas our crosslinked gelatin coatings provide the long‐term stability required for the 30‐day studies.

Gelatin C was prepared using the same protocol as gelatin NC but further modified by crosslinking it with EDC/NHS (N‐Ethyl‐N′‐(3‐dimethylaminopropyl)carbodiimide hydrochloride 100 mm (E7750, Sigma–Aldrich, Darmstadt, Germany); N‐Hydroxysuccinimide 50 mm (130672, Sigma–Aldrich, Darmstadt, Germany)); dissolved in 10 mm MES buffer (2‐(N‐morpholino)ethanesulfonic acid (M3671, Sigma–Aldrich, Darmstadt, Germany) in ultra‐pure water; pH = 5.5). Lastly, EDC/NHS was washed 4 times with PBS (first quickly, then three times for 30 min), as it was toxic to the cells, to ensure thorough elimination of any residual EDC/NHS [42, 43, 44]. The coated well plates were left to dry overnight in the laminar flow hood.

Fibrin was coated using a slightly modified protocol from [45] by mixing 13.16 µL/cm² CaCl₂ solution (0.00333 g/mL in ultra‐pure water; C1016, Sigma–Aldrich, Darmstadt, Germany) and 13.16 µL/cm² thrombin solution (48 U/mL in ultra‐pure water; 605195, Sigma–Aldrich, Darmstadt, Germany) per well, and then adding 131.6 µL/cm² fibrinogen solution (1 mg/mL in phosphate‐buffered saline (PBS); translating to 131.6 µg/cm²; F3879, Sigma–Aldrich, Darmstadt, Germany) per well, followed by gentle, yet thorough shaking. The coating concentration was determined based on the wide range of concentrations used in the literature [46, 47, 48, 49]. Since 131.6 µg/cm² fits well in the range, we decided to go with the same concentration as with gelatin (discussed below). The coated well plates were left to dry overnight in the laminar flow hood.

Fibronectin was coated by adding 131.6 µL/cm² fibronectin solution (0.02 mg/mL in ultra‐pure water; translating to 2.6 µg/cm2; 341635, Sigma–Aldrich, Darmstadt, Germany) per well, as per manufacturer's protocol with coating concentrations based on the protocol's guidelines of 1–5 µg/cm² [50]. The coated well plates were left to dry overnight in the laminar flow hood.

The samples for the experiments described in Sections 2.4, 2.5, and 2.6 were prepared in the following way: 24‐well plates were coated with gelatin NC (250 µg/well – 131.6 µg/cm²), gelatin C (250 µg/well – 131.6 µg/cm²), fibrin (250 µg/well – 131.6 µg/cm²), and fibronectin (5 µg/well – 2.6 µg/cm²). After complete drying of the components, cells were distributed into the wells (50,000 EPCs, 50,000 HUVECs, 80,000 HMEC‐1s per well). These densities achieve ∼60–80% confluence within 6 h, ensuring sufficient cell‐cell contacts for early viability and marker expression studies without overcrowding long‐term cultures [51].

The samples were prepared in triplicate.

2.3. Assessing Material Coatings

To confirm material presence on the surfaces and their chemical characteristics, Fourier‐transform Infrared (FTIR) spectra were recorded by measuring the attenuated total reflection (ATR) using a Cary 630 FTIR spectrometer (Agilent, Santa Clara, CA, USA). A spectral range between 4000 and 650 cm⁻¹ (with a resolution of 2 cm⁻¹) was recorded for each sample, as described previously [24, 52, 53, 54].

2.4. Material Biocompatibility Testing

2.4.1. Live/Dead Assay

The Live/Dead assay (Biotium, Inc., Fremont, CA, USA) was used to qualitatively and quantitatively assess the influence of different materials on the viability of different cell types. Viable cells were detected by the presence of intracellular esterase activity, which converts plasma membrane‐permeable calcein‐acetoxymethyl to calcein and produces green fluorescence. Dead cells were detected by ethidium homodimer III, which enters the cells through the damaged cell membrane and intercalates with the DNA double helix to emit red fluorescence. Stained cells were observed directly on the coating surface under a fluorescence microscope (EVOS, FL Cell Imaging System, Thermo Fisher Scientific Inc., Waltham, MA, USA).

After 72 h in culture, the culture medium was removed from the wells, which were washed with an FBS‐free cell‐specific medium three times. Then, the Live/Dead assay was performed as per the manufacturer's protocol. Briefly, a solution of 5 µL of calcein stock and 20 µL of ethidium homodimer stock per 10 mL was prepared in an FBS‐free cell‐specific medium and applied in 500 µL aliquots to the sample wells, followed by a 30 min incubation period at room temperature and protected from light.

The images were quantitatively analyzed using ImageJ (NIH, Bethesda, MD, USA) software.

All the material‐cell combinations for this experiment were in biological duplicates.

2.4.2. Resazurin Reduction Assay

The alamarBlue Cell Viability Reagent (Thermo Fisher Scientific Inc., Waltham, MA, USA) was used to quantify the relative metabolic activity of all three cell types grown on different materials. It was a fluorescent assay that detects cellular metabolic activity. The blue, non‐fluorescent reagent 7‐hydroxy‐10‐oxidophenoxazin‐10‐ium‐3‐one (resazurin) was reduced to highly fluorescent 7‐hydroxy‐3H‐phenoxazin‐3‐one (resorufin) by dehydrogenase enzymes in metabolically active cells. This conversion occurs only in viable cells; therefore, the amount of resorufin produced was proportional to the number of viable cells in the sample. The readily soluble resorufin was quantified using the Varioskan multiplate reader (Thermo Fisher Scientific Inc., Waltham, MA, USA) by measuring fluorescence intensity at an excitation wavelength of 530 nm and emission wavelength of 590 nm [51].

After 72 h of culture, the resazurin reduction assay was performed as per the manufacturer's protocol—the cell culture medium was removed from the wells, and a fresh cell‐specific culture medium containing alamarBlue reagent was added to the cells, followed by a 4 h incubation period at 37°C, 5% CO₂ and protected from light. Then, the fluorescence was measured.

All the material‐cell combinations for this experiment were in biological triplicate.

2.5. Morphology Assessment with Cytoskeleton Staining

Cytoskeleton staining was performed to evaluate the potential influence the materials might have had on the ECs' morphology over 30 days of culturing on the materials. This was determined through fluorescent staining of the cell cytoskeleton (F‐actin fibers) using Phalloidin—Alexa Fluor Plus 555 Phalloidin (Thermo Fisher Scientific Inc., Waltham, MA, USA). The staining was performed after 72 h of culture.

The monolayers of EC cells were fixated using a Fixation solution 5X (SIG5113, Sigma–Aldrich, Darmstadt, Germany), diluted to 1X with deionized water for 15 min at room temperature, and washed three times with PBS.

The assay was performed in a dark room. The final step was the addition of the Fluoroshield Mounting Medium with 4′,6‐diamidino‐2‐phenylindole (DAPI, Sigma–Aldrich, Darmstadt, Germany) to dye the cell nuclei. The stained cells were examined at the required wavelengths for the respective dyes (excitation/emission: DAPI  =  306/460 nm, Phalloidin  =  556/574 nm). All micrographs were acquired using the EVOS FL Cell Imaging System (Thermo Fisher Scientific Inc., Waltham, MA, USA).

At 4 time points – 0 days (6 h post‐seeding), 3 days, 7 days, and 30 days, the cells were fixated, and then the cell structure was assessed by staining F‐actin fibers with phalloidin.

All the material‐cell combinations for this experiment were in biological duplicates.

2.6. Cell Functionality Assessment

Results of the biocompatibility and morphology assessments (see Sections 3.1 and 3.3) suggest that gelatin NC does not provide a suitable culturing surface for HMEC‐1 cells, as indicated by the low viability and metabolic activity. This coating was omitted from further experiments.

2.6.1. Phenotype Retention Testing

The retention of the endothelial phenotype was observed by immunocytochemical (ICC) staining, which determined the expression of CD31, CD144, and VEGFR‐2 on the materials over 30 days of culturing.

At 4 time points – 0 days (6 h post‐seeding), 3 days, 7 days, and 30 days, the cells were fixated and immunostained for three different endothelial cell markers—CD31, CD144, and VEGFR‐2.

The monolayers of EC cells were fixated using a Fixation solution 5X (SIG5113, Sigma–Aldrich, Darmstadt, Germany), diluted to 1X with deionized water for 15 min at room temperature. The samples were washed with PBS three times for 5 min. Then, blocking solution (PBS + 1% bovine serum albumin (BSA; A2153, Sigma–Aldrich, Darmstadt, Germany) + 0.1% polysorbate 20 (TWEEN 20, P1379, Sigma–Aldrich, Darmstadt, Germany)) was added to the fixated cells for 30 min to block non‐specific binding. Again, the samples were washed with PBS three times for 5 min.

The general protocols for ICC from the respective antibody manufacturers were followed. After each step, the samples were washed with PBS three times for 5 min. The cells were stained with Anti‐CD31 antibody (A4‐273‐T100, Exbio, Vestec, Czech Republic) at a dilution of 1:20, Anti‐CD144 antibody (A6‐770‐T100, Exbio, Vestec, Czech Republic) at a dilution of 1:20 and Anti‐VEGFR‐2 antibody (MAB3571‐500, Novus Biologicals, LLC, Centennial, CO, USA) at a concentration of 25 µg/mL; CD31 and CD144 were both stained with conjugated antibodies while the VEGFR‐2 required additional secondary Anti‐Mouse antibody (ab150113, Abcam, Cambridge, UK) at a dilution of 1:200. The assays were performed in a dark room. The conjugated and primary antibodies on the samples were incubated for 18 h at 8°C protected from light, and the secondary antibody was incubated for 1 h at room temperature protected from light. The final step was the addition of the Fluoroshield Mounting Medium with 4′,6‐diamidino‐2‐phenylindole (DAPI, Sigma–Aldrich, Darmstadt, Germany) to dye the cell nuclei. Examination of the stained cells was performed at the required wavelengths for the respective dyes (excitation/emission: DAPI  =  306/460 nm, Anti‐CD31  = 488/525 nm, Anti‐CD144  = 650/671 nm, Anti‐VEGFR‐2  = 488/525 nm). All micrographs were acquired using the EVOS FL Cell Imaging System (Thermo Fisher Scientific Inc., Waltham, MA, USA).

All the material‐cell‐antibody combinations for this experiment were in biological duplicates.

Cell‐free, coated wells were also imaged to record substrate autofluorescence. No‐primary and no‐secondary antibody controls confirmed negligible nonspecific signals.

2.6.2. Quantitative Analysis of Marker Expression

Alongside qualitative assessment, CD31, CD144, and VEGFR‐2 expression in HUVECs was quantified using Python (Python Software Foundation, Wilmington, DE, USA) and ImageJ (NIH, Bethesda, MD, USA) software. Only HUVECs were selected for this step as they demonstrated the best results in previous tests.

All samples—controls and test coatings—were stained simultaneously, imaged in a single session, and captured with fixed exposure, gain, and illumination settings. Cell‐free, coated wells were imaged to record substrate autofluorescence and this value was subtracted from every image, and no‐primary and no‐secondary antibody controls confirmed negligible nonspecific signal. A serial dilution of Alexa‐Fluor reference beads imaged under identical settings confirmed a linear intensity response (R² > 0.99) across the measured range.

Since the samples were stained using the sample protocol and the images were obtained using the same settings on the microscope, they were directly comparable. The mean green/red color intensity was calculated using a simple Python script by splitting the image into RGB channels and then measuring the mean green/red intensity. The number of cells (blue‐stained nuclei) was computed using ImageJ software by setting the color threshold and then counting the areas. Then, the mean green/red color intensity was divided by the number of cells to calculate the mean color intensity per single cell, essentially indicating the magnitude of marker expression.

2.6.3. Growth Factor Release Measurements with ELISA

Von Willebrand factor (vWF) release was measured in the cell culture medium in 48‐well plates coated with each material and seeded with each cell type. The experiment had 4 planned time points – 0 days (6 h post‐seeding), 3 days, 7 days, and 30 days. The cell culture medium was always replaced precisely 72 h before taking samples, except for the first time point, where the cell culture medium was in the culture for 6 h. The medium samples were collected and frozen at −20°C until the end of sampling. Then, the samples were thawed, and an enzyme‐linked immunosorbent assay (ELISA; ab108918, Abcam, Cambridge, UK) was performed on all samples simultaneously per the manufacturer's protocol.

All the material‐cell combinations for this experiment were in biological triplicates.

2.7. Statistical Analysis

All numerical values were reported as mean with standard deviation (SD). The Shapiro–Wilk test confirmed the normal distribution of the experimental data. Levene's test was used to assess the equality of variances. As all data sets were well‐modeled by a normal distribution and homoscedasticity, a one‐way analysis of variance (ANOVA), followed by the Bonferroni post‐hoc test, was carried out accordingly. Obtained p‐values < 0.05 were considered statistically significant. Statistical analysis was performed using SPPS Statistics 27 (IBM Corp. Armonk, NY, USA).

3. Results and Discussion

3.1. Assessing Material Coatings

ATR‐FTIR was used to confirm the presence of the individual materials on the surfaces. The corresponding recordings of sample FTIR spectra are shown in Figure 2, and the characteristic vibrational frequencies are summarized in Table 1.

FIGURE 2.

FTIR spectra of gelatin NC, gelatin C, fibrin, and fibronectin compared with the spectra of the control polystyrene surface. The characteristic bands are marked in grey, with the respective frequencies listed in Table 1. The proteinaceous samples show peaks for amide I, II, and III at polymer‐specific frequencies (created with GraphPad Prism).

TABLE 1.

Characteristic vibration frequencies of polymer thin films measured using ATR‐FTIR and compared to data from previous studies.

Vibration (cm−1)	Control–polystyrene [55, 56, 65]	Gelatin NC [57, 58, 59, 66]	Gelatin C [58, 59, 61, 66]	Fibrin [62, 63, 67]	Fibronectin [60, 62, 68, 69]
Amide A band		3300	3300 (reduced)		3290
Amide B band		3080			3060
ν(CH)	2920
ν(C = O)	1720		1720
Amide I		1650	1640 (shift)	1650	1650
Amide II		1540	1530 (shift)	1535	1540
ν(COO)sym				1400
Amide III					1235

The control (polystyrene surface) spectrum exhibits weak and broad bands primarily around 2920 cm⁻¹, corresponding to aliphatic C─H stretching vibrations, and a notable peak near 1720 cm⁻¹, indicating ester carbonyl (C═O) groups, likely from the underlying polymeric substrate [55, 56]. The lack of characteristic amide bands confirms the absence of proteinaceous material.

Non‐crosslinked gelatin exhibits all characteristic protein vibrations, including Amide A ∼3300 cm⁻¹), Amide I (∼1650 cm⁻¹), and Amide II (∼1540 cm⁻¹) [57, 58, 59]. The broad amide peaks reflect a disordered secondary structure, such as random coils [60].

Crosslinked gelatin shows a more intense C═O peak around 1720 cm⁻¹, indicative of new carbonyl‐containing groups introduced during crosslinking (e.g., via EDC/NHS chemistry). Slight shifts in the Amide I and II regions suggest structural rearrangements and reduced hydrogen bonding due to crosslinking [61]. These spectral changes are typical for crosslinked hydrogels [61].

The fibrin spectrum shows prominent protein bands: Amide I (≈1650 cm⁻¹) from C═O stretching, and Amide II (≈1535 cm⁻¹) from N–H bending. These are indicative of its polypeptide backbone and confirm its typical fibrillar protein structure [62, 63, 64]. The band near 1400 cm⁻¹ may also suggest COO⁻ symmetric stretching, commonly seen in proteins [62].

Fibronectin presents a similar protein signature to fibrin with characteristic Amide I (≈1650 cm⁻¹), Amide II (≈1540 cm⁻¹), and Amide III (≈1235 cm⁻¹) bands. The Amide A (≈3290 cm⁻¹) and Amide B (≈3060 cm⁻¹) peaks further support the presence of N–H and CH stretching modes associated with folded proteins [60, 62]. These features confirm the adsorption of the protein onto the surface.

Overall, all coatings showed characteristic vibrations, as described in previous studies, with the individual material references listed in Table 1. These results confirm the successful coating of all tested solutions.

3.2. Material Biocompatibility

Evaluating the biocompatibility of materials is a critical step in scaffold‐based TE, as it greatly influences the success of the intended biological function. This study assessed biocompatibility using two methods –Live/Dead cell viability assay and alamarBlue quantitative cell metabolic activity assay. Both tests provide a robust understanding of the interaction between the materials and EC types and elucidate how these materials can support long‐term cell viability. The results of these tests can also provide valuable information about the potential of these materials to support blood vessel formation. Therefore, selecting materials that exhibit high biocompatibility with ECs is critical for successfully engineering various tissues in vitro.

The Live/Dead assay offers a qualitative assessment of cell survival and indicates the ability of the material to support cell growth without causing cytotoxicity. Frequencies of viable cells (green) compared to non‐viable cells (red) on each material also provide quantitative evidence of biocompatibility. Figure 3A shows micrographs of cells cultured on the respective coatings and analyzed with Live/Dead assay. From the images, it is clear that most of the cells were alive except for HMEC‐1s on gelatin NC, where few cells are present, many of which are dead. This is supported by measurements with less than 5% dead cells in all the images, except for HMEC‐1s on gelatin NC, with 26% dead cells. In this sense, the results are in accordance with the results from the resazurin reduction assay (shown in Figure 3B–D), where HMEC‐1s on gelatin NC exhibited the comparatively lowest metabolic activity. HMEC‐1s cultured on the fibrin coating (that produced lower metabolic activity compared to the control) look similar to the control sample. What also stands out from the Live/Dead assay results are the EPCs cultured on gelatin C, which exhibit empty patches where neither live nor dead cells are visible. The reason for this growth is unclear. However, it could be due to gelatin not coating homogeneously over the surface when drying and then staying non‐homogeneously coated after crosslinking, similar to the findings of Yakhno et al. [70].

FIGURE 3.

(A) Live/dead images of EC cells on coatings of different materials. The green color represents live cells and the red color represents dead cells. Scale bars represent 400 µm. (B) Relative metabolic activity of EPCs on coatings of different materials, (C) Relative metabolic activity of HUVECs on coatings of different materials, (D) Relative metabolic activity of HMEC‐1s on coatings of different materials. Triple asterisk denotes strong statistical significance (p < 0.001) (created with GraphPad Prism and Adobe Illustrator).

The resazurin reduction assay (alamarBlue) assay, on the other hand, provides a quantitative assessment of cellular metabolic activity. By assessing the metabolic activity of cells, we gain insight into how cells function when they come into contact with materials. High metabolic activity indicates good cell health and viability, suggesting the cells are metabolically active and proliferating [71]. Figure 3B–D show the metabolic activity of the cells on different material coatings. All the materials affected the metabolic activity of the EPCs and HUVECs positively, while for HMEC‐1s, gelatin NC and fibrin had a negative impact on their metabolic activity, where gelatin NC and fibrin reduced the metabolic activity of HMEC‐1s by 33.8% and 6.1%, respectively, compared to the control.

In EPCs, all the materials significantly increased cells' metabolic activity: fibrin—+22.2% ± 0.8% (p < 0.001; n = 4), followed by fibronectin—+11.1% ± 2.4% (p < 0.001; n = 4), gelatin C—+6.7% ± 0.7% (p < 0.001; n = 4), and gelatin NC—+4.3% ± 0.8% (p < 0.001; n = 4).

In HUVECs, all the materials had similar results, increasing the cells' metabolic activity by approximately 10%: gelatin C—+13.6% ± 0.4% (p < 0.001; n = 4), followed by gelatin NC—+11.2% ± 0.2% (p < 0.001; n = 4), fibronectin—+10.7% ± 0.6% (p < 0.001; n = 4), and fibrin—+8.6% ± 0.2% (p < 0.001; n = 4).

In HMEC‐1s, gelatin C and fibronectin significantly increased cells' metabolic activity: gelatin C—+11.7% ± 1.3% (p < 0.001; n = 4), fibronectin +13.7% ± 1.1% (p < 0.001; n = 4), while gelatin NC and fibrin significantly decreased cells' metabolic activity: gelatin NC—‐33.8% ± 0.7% (p < 0.001; n = 4), fibrin—‐6.1% ± 4.6% (p < 0.001; n = 4).

Gelatin NC performed worse than gelatin C on all the cells (by 2.2%, 1.8%, and 40.7% on EPCs, HUVECs, and HMEC‐1s, respectively)—the difference was significant on HMEC‐1s (p < 0.001; n = 4), which led to the decision that gelatin NC would not be used in subsequent longer‐term experiments.

All the wells were seeded with cells at identical densities and verified comparable attachment after 6 h. Overall metabolic activity per culture well was assessed—i.e., a composite indicator of both viability and per‐cell metabolism—rather than to derivation of per‐cell rates, as measured viability separately (Live/Dead imaging) found that trends in resazurin reduction paralleled viability, indicating that the signal was not driven solely by cell number.

3.3. Morphology Assessment of Cells Grown on Materials

Material characterization is an important step in determining parameters that influence cell attachment, alignment, and overall cell behavior. The materials were selected based on the abundance of data (surface/topography evaluation using AFM, etc.) for gelatin, fibrin, and fibronectin, which indicate properties suitable for use with endothelial cells [22, 24, 72, 73, 74]. The following assessment methods, however, were crucial in determining cell morphology on these materials due to specific combinations and settings used.

3.3.1. Preliminary Morphology Assessment

Assessing cell morphology after exposure is another critical step in evaluating the performance of materials intended for TE. Cell morphology can provide deep insights into cell viability, differentiation, and functionality because it is closely related to cell behavior [75, 76]. By observing cells' shape, size, and arrangement, we can discern clues to their mechanical, migratory, proliferative, and differentiating capabilities [76, 77, 78]. One of the crucial elements affecting cellular morphology is the cytoskeleton, which is responsible for maintaining cell shape, enabling cell motility, and facilitating intracellular transport and cell division [79]. Therefore, changes in the cytoskeleton's organization may indicate alterations in cell behavior. In this study, we performed fluorescent staining of F‐actin fibers with phalloidin to investigate the possible effects of the materials on cell morphology.

Figure 4 shows F‐actin‐stained cells. All the cells on all the materials look very similar to controls except for EPCs on fibrin (exhibiting increased cell density) and HMEC‐1s on gelatin NC (exhibiting decreased cell density and irregular shape compared to the control sample). This is coherent with the results of the resazurin reduction assay, where fibrin had a very positive impact on EPCs, and gelatin NC had a very negative impact on HMEC‐1s (Figure 3D).

FIGURE 4.

Images of EC cells on coatings of different materials with cytoskeleton (F‐actin) stained. Red color represents F‐actin fibers, and blue color represents cell nuclei. Scale bars represent 400 µm (created with Adobe Illustrator).

Based on resazurin reduction assay results, Live/Dead, and cytoskeleton staining, 3 of the 4 materials were selected for further, longer‐term experiments: gelatin C, fibrin, and fibronectin.

3.3.2. Morphology Assessment Over a 30‐Day Period

Assessing cellular morphology over time is important, as it allows the monitoring of changes in cellular behavior over time and the observation of long‐term effects of the materials on cells. This is particularly important in TE applications where the scaffold is intended to support cell growth and function over an extended period [80].

Figures 5, 6, 7 show how the cytoskeleton changed through 30 days of culture on different materials compared to the control. What is more, cell density can be assessed visually. It can be observed that most of the cell‐material combinations fared similarly to the control, with some notable variations in cell density and cytoskeleton arrangement.

FIGURE 5.

Images of EPCs with cytoskeleton (F‐actin) stained; 0 days, 3 days, 7 days, 30 days. Red color represents F‐actin fibers and blue color represents cell nuclei. Scale bars represent 400 µm (created with Adobe Illustrator).

FIGURE 6.

Images of HUVECs with cytoskeleton (F‐actin) stained; 0 days, 3 days, 7 days, 30 days. Red color represents F‐actin fibers and blue color represents cell nuclei. Scale bars represent 400 µm (created with Adobe Illustrator).

FIGURE 7.

Images of HMEC‐1s with cytoskeleton (F‐actin) stained; 0 days, 3 days, 7 days, 30 days. Red color represents F‐actin fibers and blue color represents cell nuclei. Scale bars represent 400 µm (created with Adobe Illustrator).

Over a 30‐day period, EPCs cultured on fibrin and fibronectin take on elongated morphologies that run parallel to each other, suggesting an induction of a spatial directionality of the culture on these surfaces. This was less prominent on gelatin C and the control. Collective alignment of the cells suggests a potential for cells to connect into a vessel‐like structure (a key characteristic required in endothelialization and vascular homeostasis) as also observed by Whited et al. and Steward Jr. et al. [81, 82], provided with other necessary stimuli, such as vascular growth factors and shear stress [83]. The results coincide well with the results of the resazurin reduction assay (Figure 3B), where EPCs exhibited the highest metabolic activity when cultured on fibrin and fibronectin.

In HUVEC cultures, elongated yet less collectively aligned morphologies were observed after 30 days on gelatin C substrates, where the highest metabolic activity was observed for this cell type (Figure 3C). They stayed more circular on other surfaces, possibly due to differences in substrate stiffness [84, 85].

HMEC‐1s exhibit a circular morphology across all surfaces. Furthermore, a significantly higher cell density was observed when cultured on the testing materials compared to the control, already after a few days of culture, and more so after 30 days of culture. This, however, does not coincide with the resazurin reduction assay results (Figure 3D), where metabolic activity is only slightly higher or, in some cases, even lower than in control, indicating that while these cells may adhere and spread on the material surfaces, their metabolic activity could be reduced, possibly due to different substrate stiffness, surface roughness, morphology and chemistry of the materials, which have been observed to affect cell metabolism [86].

3.4. Cell Functionality Assessment

3.4.1. Cell Phenotype Retention Assessment

In TE, the ability of cells to maintain their intended phenotype is critical to the functional integrity and success of the construct. Phenotype retention refers to maintaining cell‐specific characteristics, such as protein expression, production, secretion, and biological behavior, that define a particular cell type. This is critical for cells to perform their intended biological functions. Therefore, evaluating the phenotype of cells cultured on different materials is essential for assessing material suitability for TE applications.

This study investigated the expression of specific CD31, CD144, and VEGFR‐2 markers because they are characteristic features of the EC phenotype [87, 88, 89]. CD31, also known as platelet endothelial cell adhesion molecule (PECAM‐1), plays an important role in angiogenesis and adhesion between endothelial cells and immune cells [89]. CD144 or vascular endothelial (VE)‐cadherin is an endothelium‐specific adhesion molecule that forms and stabilizes intercellular junctions, contributing to vascular integrity [88]. VEGFR‐2 is the primary receptor for vascular endothelial growth factor (VEGF) in ECs, and its activation triggers several cellular responses, including proliferation, migration, and increased vascular permeability, all of which are critical for angiogenesis [87].

In the experiments, each cell type was exposed to different materials (gelatin C, fibrin, fibronectin, and control), as in the previous tests, followed by ICC analysis to evaluate how different materials influence cell phenotype retention. By performing this assessment at different time points (0, 3, 7, and 30 days), we could observe changes in phenotype expression, which is critical for understanding the long‐term effects of materials on cell behavior.

Table 2 below presents a qualitative assessment of the positive/negative results of staining the cells with different markers. The representative images are shown in Figure 8 below, and the remaining images are found in the supporting information document (Tables S1–S12). The imperative results were subsequently quantitatively analyzed using Python and ImageJ; the results are presented in Section 3.4.2.

TABLE 2.

Representing subjective assessment of the presence of marker expression on different combinations of materials and cell subtypes. The plus sign denotes that the sample did express the cell marker (subjective evaluation), and the minus sign denotes that the sample did not express the cell marker (subjective evaluation). Superscript numbers denote the samples which are presented in Figure 8 below, as these combinations produced a change of expression during the 1‐month culture period.

Control

Gelatin C

Fibrin

Fibronectin

0d

3d

7d

30d

0d

3d

7d

30d

0d

3d

7d

30d

0d

3d

7d

30d

EPC

CD31

–

–

–

–

–

–

–

–

–

–

–

–

–

–

–

–

CD144

–

–

–

–

–

–

–

–

–

–

–

–

–

–

–

–

VEGFR‐2

+

+

+

+

+

+

+

+

+

+

+

+

+

+

+

+

HUVEC

CD31

+

+

+

+

+

+

+

+

−1

+

+

+2

+

+

+

+

CD144

−3

–

+

+4

−5

+

+

+6

+

+

+

+

−7

+

+

+8

VEGFR‐2

+

+

+

+

+

+

+

+

+

+

+

+

+

+

+

+

HMEC‐1

CD31

−9

–

–

+10

−11

–

+

+12

−13

–

+

+14

−15

–

+

+16

CD144

−17

+

+

+18

−19

+

+

+20

+

+

+

+

−21

+

+

+22

VEGFR‐2

+

+

+

+

+

+

+

+

+

+

+

+

+

+

+

+

FIGURE 8.

Images of combinations that produced a change of expression during the 1‐month culture period. Green represents CD31 expression, red color represents CD144 expression and blue color represents cell nuclei. Scale bars represent 400 µm (created with Adobe Illustrator).

ICC staining data suggest that the choice of material influenced endothelial marker expression. The final phenotype expression of tested markers (CD31, CD144, and VEGFR‐2) was comparable between the control and tested materials on respective cell types; however, it was in significantly different time frames. For example, EPCs only showed expression of VEGFR‐2 and in the same manner on all materials, while CD31 and CD144 expressions were not detected in any of the samples. This could possibly be due to the fact that early‐outgrowth ECs typically express progenitor markers (CD34, VEGFR‐2) but display little or no junctional CD31 until they mature into late‐outgrowth endothelial colony‐forming cells after 2–3 weeks in culture and additional passages [90, 91]. Donor‐to‐donor variability can further shift the timing of this transition. Thus, the absent CD31/CD144 signals most likely reflect the developmental stage of our EPCs rather than an effect of the coatings. In contrast, HUVECs and HMEC‐1s expressed all tested markers on all surfaces. Interestingly, however, most markers were expressed sooner on the tested materials compared to the control surface. This suggests that the selected materials could stimulate the expression of crucial endothelial markers, which would enhance the angiogenic potential of these cells.

It is worth noting that the lack of expression of CD31 and CD144 in EPCs led to the decision to exclude these cells from further experiments. CD31 and CD144 are essential for angiogenesis [92] and stabilization of formed vessels [93], respectively, and their absence could, therefore, limit the ability of EPCs to contribute to vascularization in TE applications effectively.

3.4.2. Quantitative Analysis of Marker Expression

To shed more light on the angiogenic potential of fibrin, fibronectin, and crosslinked gelatin, HUVEC were cultured on these substrates, and the expression of CD31, CD144, and VEGFR‐2 was quantified after a few hours and after 30 days. The results are summarized in Figure 9. CD31 expression is initially high in the control samples but decreases significantly after 30 days. When cultured on fibrin, the initial expression is also high and remains high after 30 days. In contrast, CD31 expression on fibronectin and gelatin C is initially low and increases significantly after 30 days toward moderate levels. CD144 expression in HUVECs is initially low on all tested surfaces and remains at the same levels in the control samples. Gelatin C and fibronectin show slight increases in expression, while fibrin exhibits significant increases in CD144 expression. VEGFR‐2 expression is initially high in the control samples, fibrin, and fibronectin, but it decreases over time. After 30 days, the expression decreases to moderate levels in fibronectin and to very low levels in control and fibrin. In contrast, HUVECs cultured on gelatin C initially show moderate levels of VEGFR‐2 expression, which increases to high levels after 30 days.

FIGURE 9.

Relative fluorescence intensity per single HUVEC on different coating materials, measured for (A) CD31, (B) CD144, and (C) VEGFR‐2 staining. The results are relative to the lowest result from the dataset, which was set to a value of 1.00 (created with GraphPad Prism and Adobe Illustrator).

These findings underscore the unique properties of each substrate and suggest that both chemistry and structural properties play a role in in vitro vascularization. Furthermore, the results indicate enormous potential for blending different components and tuning the properties of culture substrates.

As a derivative of collagen, gelatin is rich in Arg‐Gly‐Asp (RGD) sequences that promote cell adhesion and cell‐substrate, as well as cell‐cell interactions [24, 32, 94], which could be a reason for the accelerated expression of angiogenic markers. The polymer mesh is stabilized and its mechanical strength is increased through additional crosslinking. While the initial expression of the evaluated markers is relatively low compared to other substrates, gelatin C is the only material on which the expression of all markers increased after 30 days. The enhanced performance of crosslinked gelatin (gelatin C) over other protein coatings can be attributed to its structural stability and bioactivity, which provide a conducive environment for prolonged endothelial cell activity and marker expression [95, 96].

Despite being a key ECM protein known for promoting cell adhesion, migration, and tissue organization, fibronectin may exhibit inferior performance compared to crosslinked gelatin due to its natural propensity to form less stable fibrillar networks. Although beneficial for initial cell attachment and spreading, these networks might not sustain long‐term cellular activities as effectively as the more mechanically robust crosslinked gelatin. Additionally, fibronectin's interaction with various ECM components and growth factors can lead to complex signaling environments that may not consistently favor prolonged angiogenic markers [95].

The specific structural characteristics of fibrin facilitate cell migration and tissue ingrowth for wound healing. However, they also exhibit poor mechanical strength and stability [25, 26, 27, 28]. Studies further show that fibrin (as well as fibronectin) binds and interacts with several growth factors and facilitates localized and spatially regulated signaling [97]. As a substrate for HUVECs in vitro, fibrin seems to effectively maintain endothelial cell identity (CD31) and junction formation (CD144) but shows a decline in angiogenic potential (VEGFR‐2) over time. Several studies have indicated that fibronectin improves endothelialization and cell morphogenesis in artificial tissue scaffolds [29, 31]. While fibronectin typically exists as a dimer it can also form fibrils that interact with several ECM components, as well as growth factors, glycosaminoglycans, and other molecules [98]. As a substrate for HUVEC culture, fibrinogen seems to impact differentiation and shows angiogenic potential, however, with some decline over time.

3.4.3. Cell Function Testing via Functional Protein Release Measurements using ELISA

Assessing cell functionality is a critical step in verifying the suitability of materials for TE. While understanding cell survival and phenotype retention is essential, these aspects do not necessarily reflect whether cells function correctly. Therefore, functional assays are particularly important, such as measuring specific proteins that a cell produces. These tests provide valuable insight into how the material affects cellular processes and whether the cells can perform their intended biological functions.

Our study assessed EC functionality and the impact of different material formulations on it by measuring the release of vWF using ELISA. The vWF is a crucial glycoprotein in the human body that plays a critical role in EC function and hemostasis (blood clot formation) [99]. It is mainly synthesized and secreted by ECs and is essential for various processes of EC function, including angiogenesis [99, 100, 101, 102]. Even though lower levels of vWF can stimulate angiogenesis and result in a larger vascular network through excessive VEGFR‐2 signaling [99, 100], studies have shown that such vessels are malformed and dysfunctional [100], suggesting that vWF controls angiogenesis by inhibiting VEGFR‐2 signaling and thus enabling the formation of “high‐quality blood vessels”.

Its production indicates functional ECs; thus, it was used in this study as a marker of endothelial functionality. In addition, disruptions in vWF production may impair angiogenesis, leading to dysfunctional vessel formation [100, 101]. Therefore, vWF production measurement may help determine whether the materials used in this study support the proper function of ECs over time.

By coupling vWF ELISA with the panel of membrane‐bound markers, we obtained complementary insight: ICC confirmed that cells retained an endothelial phenotype, while vWF secretion demonstrated that they remained functionally active on the different coatings throughout the 30‐day culture.

As mentioned, EPCs performed poorly in the ICC analysis; therefore, they were omitted from vWF release measurements. Instead, only HUVECs and HMEC‐1s were evaluated, and the results were intriguing. HUVECs, for instance, produced detectable vWF at all tested time points on all materials, while HMEC‐1s only produced detectable vWF at specific time points on certain materials. This variability suggests that the cell type may influence vWF production and potentially other functional attributes. The concentrations of vWF produced by different cell types over four time points of 0, 3, 7, and 30 days are plotted on graphs in Figure 10.

FIGURE 10.

The concentration of vWF in cell culture medium on samples of (A) HUVECs on different material coating; and (B) HMEC‐1s on different material coatings (created with GraphPad Prism and Adobe Illustrator).

Interestingly, the vWF concentration tended to increase over time on most materials, except fibrin, where the measured vWF concentration was highest directly after cell seeding, dropped to its lowest point at day 3, and then gradually increased until day 30. The initial decrease in vWF production observed with fibrin can be attributed to its preparation process in which fibrinogen is crosslinked with thrombin. Excess thrombin remaining in the material may trigger the initial activation of the receptors PAR‐1, PAR‐2, and PAR‐4 on endothelial cells [103]. This activation leads to exocytosis of Weibel‐Palade bodies [104], causing a subsequent burst of vWF release [105]. As thrombin is gradually washed out over time, a decrease in vWF production is observed. Overall, the increasing trend in vWF release suggests that the materials might be supporting the continued functionality of the ECs, which is a positive sign for their potential application in TE.

Another noteworthy observation is the difference in vWF concentration ranges between the HUVECs and HMEC‐1s. After 30 days, HMEC‐1s produced approximately 1 mIU/mL of vWF on all surfaces. In contrast, HUVECs produced vWF in a wide range of concentrations, from 6.75 mIU/mL (fibrin) to 32.14 mIU/mL (gelatin C). The normal plasma vWF concentration equals 10 µg/mL [106], corresponding to 1 IU/mL [107]. The concentrations observed in our measurements are thus 10–100 times lower than those typically found in human plasma. This discrepancy can be attributed to the substantially higher volume of the medium relative to the number of cells compared to plasma, a factor that was anticipated. Additionally, it should be noted that our current model's representation of endothelium lacks the necessary complexity, likely resulting in the absence of physiological protein concentrations. Instead, our primary focus has been on conducting a qualitative assessment of the outcomes and the underlying trends they reveal. The wider spectrum of vWF production by HUVECs suggests potential variations in their sensitivity and response to diverse materials, underscoring the critical significance of interactions between cells and materials.

Furthermore, it is noteworthy that the levels of vWF generated by HUVECs were approximately 10–20 times more akin to those observed in human plasma than those generated by HMEC‐1s. This proximity implies a greater potential for these cells to better mimic in vivo conditions, rendering them a more viable option for applications in microvascular tissue engineering.

4. Conclusions

TE presents an intricate interplay between materials, cells, and biochemical factors, aiming to construct tissue analogs that can regenerate and repair damaged or diseased tissues and organs, where vascularization remains a key challenge. Our study evaluated the interactions between different materials and endothelial cell (EC) types to enhance angiogenesis and support EC functionality in TE applications.

We tested gelatin, fibrin, and fibronectin for their ability to promote angiogenesis and EC functionality. We assessed the biocompatibility, morphology, phenotype retention, and functionality of ECs cultured on these materials. Our findings reveal promising insights into the behavior of different EC types on these materials.

Biocompatibility assessments indicated that gelatin C, fibrin, and fibronectin supported EC viability and metabolic activity, with some variations in different cell types, suggesting that all materials are suitable components for in vitro EC culture and potentially broader biomedical applications.

Morphology assessment revealed that most cells exhibited similar morphology with some variations in cell density on the tested materials compared to controls, and cell elongation was observed on specific materials, especially fibrin and fibronectin. Long‐term morphology studies showed potential differences in growth behavior on different materials over 30 days, suggesting that materials might influence how cells spread and interact over time.

Phenotype expression assessments highlighted that the selected materials influenced endothelial marker expression in both HUVECs and HMEC‐1s. The results show varying phenotype expression as a function of culture substrate and culturing time over a 30‐day period.

Functionally, vWF release measurements suggested material‐dependent variations in EC biological performance. Furthermore, HUVECs produced vWF levels closer to physiological concentrations compared to HMEC‐1s, indicating better functional mimicry under in vitro conditions.

Assessing tubulogenesis would guarantee a more angiogenic characterization; however, our aim for this study was an early‐stage materials screen rather than a complete endothelial differentiation assessment.

The results highlight the importance of substrate selection for in vitro vascularization. Further studies on the long‐term performance and mechanical properties of the materials, especially in 3D, are warranted to optimize their use in constructing functional vascular networks in engineered tissues. Based on availability, low cost, and performance in the described experiments, gelatin especially merits further investigation for biomedical applications.

Conflicts of Interest

The authors declare no conflict of interest.

Supporting information

Supporting File 1: mabi70041‐sup‐0001‐SuppMat.docx.

Vajda J., Vihar B., Milojević M., Bjelić D., Brečko A., and Maver U., “Substrate‐Dependent Variability in Viability and Angiogenic Marker Expression Among Three Endothelial Cell Subtypes: Insights for Artificial Tissue Vascularization.” Macromol. Biosci. 25, no. 11 (2025): e00333. 10.1002/mabi.202500333

Data Availability Statement

The data that support the findings of this study are available in the Supporting Information of this article.