Development of biomimetic extracellular matrix with the functions of protein sequestration and cell attachment using dual aptamer-functionalized hydrogels

Abstract

The extracellular matrix (ECM) not only has cell-binding sites for cell attachment but also protein-binding sites for molecular sequestration. Aptamers have high binding affinities and specificities against their target molecules. Thus, the purpose of this work was to develop dual aptamer-functionalized hydrogels for simultaneously recapitulating the two key features of the ECM in binding cells and sequestering proteins. We synthesized the hydrogels using free radical polymerization in a freezing procedure. As the hydrogels were macroporous with pores of 40 to 50 μm, both cells and proteins could be loaded into the hydrogels after the synthesis. Importantly, the vascular endothelial growth factor (VEGF) aptamer improved VEGF sequestration and reduced the apparent diffusivity of VEGF by over two orders of magnitude, resultantly prolonging VEGF retention and release. The c-MET aptamer promoted the attachment of endothelial cells in the hydrogel network. When two aptamers were both incorporated into the hydrogel, they could produce synergistic effects on cell survival and growth. Thus, this work has successfully demonstrated the potential of developing biomimetic ECMs with two key functions of cell attachment and protein sequestration using dual aptamer-functionalized hydrogels.

Graphical Abstract

1. Introduction

The extracellular matrix (ECM) is a dynamic and complex molecular network.^1–3 It comprises a variety of molecules including proteoglycans, fibronectin, elastin, collagen, etc.^1,4 These molecules are linked together to form a structurally stable matrix in supporting cells mechanically.¹ In addition, some of these molecules are highly charged.^1,5 For instance, sulfated proteoglycans have a high density of negative charges and interact with various growth factors for molecular sequestration via electrostatic interactions.⁵ Other molecules (e.g., fibronectin) have critical cell-binding sites and contribute to the attachment and migration of cells.^4,5 Thus, when a biomimetic ECM is developed for applications (e.g., regenerative medicine), it is necessary to recapitulate the critical functions of the ECM such as cell binding and molecular sequestration.^6,7

Hydrogels have been widely studied for the development of biomimetic ECMs.^6,8 Hydrogels are a three-dimensional crosslinked polymeric networks that contain a large amount of water.⁹ Hydrogels have many key features such as mechanical strength similar to soft tissues and porous structure supporting molecular transport and cell infiltration.^9,10 However, synthetic hydrogels often lack specific cell-binding sites for cell attachment and protein-binding moieties for protein sequestration.^6,11 Thus, great efforts have been made in functionalizing hydrogels with affinity ligands such as heparin, peptides, and aptamers.^12–15

Aptamers are single-stranded synthetic oligonucleotides selected from DNA or RNA libraries.^16,17 They have high binding specificities and affinities against their targets similar to or even superior to antibodies.¹⁸ They can be chemically synthesized and modified without losing their binding functions.^18–20 Thus, aptamers have recently received significant attention in the field of biomaterials.¹⁵ Our group and others have demonstrated that aptamers can be applied to functionalize hydrogels for sequestering growth factors.^21–26 Aptamers can also be applied to capture target cells.^27–29 However, while aptamer-functionalized hydrogels have been investigated, no study has been carried out to demonstrate that two types of aptamers can be incorporated into hydrogels to mimic the two key functions of cell binding and molecular sequestration of the ECM.

In this study, polyethylene glycol (PEG) hydrogel was used as a model to construct a biomimetic ECM with a vascular endothelial growth factor binding aptamer (VEGF aptamer) and a c-MET receptor binding aptamer (c-MET aptamer). VEGF aptamer has a high binding affinity to VEGF and has been previously demonstrated to sequester VEGF in hydrogels successfully.^30,31 VEGF is one of the most extensively studied growth factors needed for cell survival and proliferation.³² The c-MET aptamer has a high binding affinity to c-MET receptor expressing cells such as endothelial cells.^33,34 Thus, it is hypothesized that the incorporation of c-MET aptamer can facilitate cell attachment when tethered to the hydrogel network. The aptamer-functionalized PEG hydrogel was synthesized via free radical polymerization in a freezing procedure. Its physical properties such as porous structure, swelling, and mechanical strength were examined. Its capability of molecular sequestration was evaluated by the examination of VEGF retention, the calculation of apparent diffusivity of VEGF, and the assessment of VEGF bioactivity. Its capability of cell binding was evaluated by the examination of cell retention and proliferation. The synergistic effects of molecular sequestration and cell binding on cell proliferation were also studied.

2. Materials and Methods

2.1. Chemical reagents

Poly (ethylene glycol) diacrylate (PEGDA) (Mn = 750) was purchased from Sigma-Aldrich (St. Louis, MO). Dulbecco’s phosphate buffered saline (DPBS), ammonium persulfate (APS), and tetramethylethylenediamine (TEMED) were obtained from Fisher Scientific (Suwanee, GA).

2.2. Biological reagents

Human Umbilical Vein Endothelial Cells (HUVECs) were purchased from ATCC (Manassas, VA). Medium 200 (M200), low serum growth supplement (LSGS), bovine serum albumin (BSA), fetal bovine serum (FBS), Geltrex, and calcein AM were purchased from ThermoFisher (Waltham, MA). VEGF-165 (VEGF, MW= 38.2 kDa) and VEGF enzyme-linked immunosorbent assay (ELISA) kit were obtained from Peprotech (Rocky Hill, NJ). CellTiter-Glo^® Luminescent Cell Viability Assay (ATP) were obtained from Promega (Madison, WI). Acrydite modified VEGF (MW ~10 kDa) and c-MET (MW ~17.5 kDa) aptamers, and their complementary sequence (CS) were custom ordered from Integrated DNA technologies (Coralville, PA).

2.3. Preparation of hydrogels

Macroporous hydrogels were synthesized via cryogelation method with some modification.³⁵ To synthesize the hydrogel, 10% v/v PEGDA was prepared in distilled water and cooled on an ice bath for a few minutes. APS and TEMED were added to the solution in 0.4% w/v and 0.2% v/v final concentration, respectively. The hydrogel solution was then poured into a 35 mm petri dish pre-treated with 4% P-F127 and the solution was then frozen at −20 °C overnight. The hydrogel was taken out of the freezer and allowed to fully thaw at room temperature. Cylindrical samples (6 mm diameter, 1 mm thickness) were cut out from each synthesis batch using a biopsy punch. These hydrogels were washed in distilled water overnight to remove all unreacted monomers and initiator/catalysts, sterilized in 70% ethanol for 1 h and triple washed in DPBS. To fabricate aptamer-functionalized hydrogels, acrydite modified aptamers (Table S1-Sn) were added to the 10% v/v PEGDA solution prior to adding APS and TEMED.

2.4. FT-IR analysis of hydrogels

Spectra were collected in attenuated total reflection (ATR) geometry on a vertex 70 spectrometer ( Bruker Optics, Bellerica, MA) equipped with liquid nitrogen cooled mercury cadmium telluride (MCT) detector and a Diamax diamond ATR accessory ( Harrick sci, Pleasantvile, NY) . A total of 400 scans were averaged at 6 cm⁻¹ resolution and absorbance was calculated by referencing to the clean bare diamond crystal. The ATR crystal was cleaned using 2-butanone (Methyl-ethyl-ketone) between samples and the cleanliness of the crystal spectroscopically confirmed. Hydrogel samples were measured dry to minimize interference of water peaks.

2.5. Evaluation of the hydrogel microstructure

The morphological features of the hydrogels were examined by scanning electron microscopy (SEM) and Leica SP8 DIVE multiphoton microscopy. SEM (Zeiss SIGMA VP-FESEM) was performed with an accelerating voltage of 3 kV. Prior to the imaging, hydrogels were coated with gold and mounted on carbon tape placed on aluminum stubs. The diameter of the hydrogel pores was calculated from SEM images using ImageJ software. The average of the longest and the shortest axes of the pore was calculated and presented as the pore diameter. For multiphoton imaging, the aptamer-functionalized hydrogel was stained using Cy5 labeled CS of VEGF aptamer (Table S1-Sn). Images were taken using 25x objective and generated by optical sectioning in the Z-direction. The images were then stacked in the Z-direction using ImageJ software.

2.6. Interconnected porosity

The interconnected porosity of the hydrogels was calculated based on previously reported method.³⁶ After synthesis, hydrogels were hydrated in PBS buffer for 24 h. The swollen hydrogels were then weighed (Ws). In order to be able to get the interconnected void volume for calculating interconnected porosity, it is important to wick away water within the interconnected pores. Thus, a Kimwipe was lightly applied to the surface of the hydrogels to remove loosely held water, and the weight of the hydrogels was recorded (W). Porosity was calculated as volume fraction (%) of pores in gel using equation 1:

$$P\text{orosity}\left(\text{\%}\right)=\left[\frac{{\text{W}}_{\text{s}}-\text{W}}{\text{W}}\right]\text{x}100$$ (1)

2.7. Swelling ratio

To calculate the equilibrium mass swelling ratio, hydrogel samples fully equilibrated in PBS buffer were first weighed (Ws). The hydrogels were lyophilized and their dry weight was recorded (Wd). The swelling ratio was then calculated according to equation 2:

$$\mathit{Swelling\; ratio}=\left[\frac{{\text{W}}_{\text{s}}}{W_d}\right]$$ (2)

2.8. Young’s modulus measurement

Hydrogels underwent axial compression to 80% strain on Instron Mechanical Testing System (Norwood, MA) fitted with a 50 N load cell after being hydrated for 24 h. The samples were then compressed with the strain rate of 1% per second. Stress-strain curves were obtained, and regression on the linear region of the graph was performed to determine the samples Young’s modulus. For cyclic compression, the hydrogel sample was loaded-unloaded at 80% strain for five consecutive cycles.

2.9. Examination of aptamer incorporation into hydrogels

To assess the aptamer incorporation, the hydrogels were incubated with FAM-labeled and Cy5 labeled CS of c-MET and VEGF aptamer (Table S1-Sn), respectively. The hydrogels were thoroughly washed and imaged with the Maestro In vivo imaging system (PerkinElmer, Waltham, MA).

2.10. Growth factor loading

Lyophilized VEGF was reconstituted by dissolving in DPBS with 0.1% bovine serum albumin (BSA). To load the VEGF to the hydrogels, the hydrogels were first gently dehydrated by blotting with sterilized Kimwipe tissue paper. After dehydrating the hydrogels, 10 μL of 10 ng/μL VEGF solution was added to the hydrogels and the hydrogels were incubated at 4 °C overnight.

2.11. Growth factor sequestration and release study

To examine growth factor sequestration, VEGF loaded hydrogels were incubated in 1 mL release medium (DPBS with 0.1% BSA) at 37 °C and 90 rpm shaking rate for 4 h. Similarly, for the release study, VEGF loaded hydrogels were incubated in 1 mL release medium (DPBS with 0.1% BSA or M200) at 37 °C at 90 rpm on a shaker. At the specific time points, the supernatant was collected and replaced with 1 mL of fresh medium. The collected supernatant was stored at −20 °C until quantification. The VEGF concentration in the supernatant was determined via a recombinant human VEGF₁₆₅ ELISA kit according to the manufacturer (Peprotech) protocol. To ensure the VEGF concentration fell within the detectable range of the assay, samples were diluted prior to the measurement. The absorbance of each sample was measured using infinite M200 Pro microplate reader (Tecan) at 405 nm and was referenced by subtracting the absorbance at 650 nm.

2.12. Calculation of diffusion coefficient

The apparent diffusion coefficient for VEGF was analyzed via the semi-empirical diffusion equation based on Fick’s second law of diffusion for a cylindrical model as shown in equation 3:³⁷

$$\frac{M_t}{M_o}=K{t}^n$$ (3)

where M_t is the amount of released protein at time t, M_o is the initial amount of the loaded protein, t is time, n is the diffusional exponent (n = 0.45), and K (kinetic constant) equal to $4{\left(\frac{D}{\pi R^2}\right)}^n$ where D is the apparent diffusivity and R is the radius of the hydrogel.

The diffusivity of VEGF in an aqueous solution at 37 °C was approximated using the Stokes-Einstein equation (4):

$$D_o=\frac{K_{B}T}{6\pi r\eta }$$ (4)

Where D_o is the diffusivity of a protein in an aqueous solution, K_B is the Boltzmann constant, T is the temperature in kelvin (K), r is protein’s hydrodynamic radius, and η is the viscosity at temperature T. The radius of VEGF was estimated as 2.8 nm based on the molecular weight.

2.13. Cell culture

HUVECs were cultured in 0.1% gelatin coated flasks. M200 supplemented with LSGS (2%) were used for seeding and expanding HUVECs. HUVECs were passaged at 80% confluency using 0.05% trypsin-EDTA and incubated at 37 °C, 5% CO₂, and 95% humidity. Cell passages 2-5 were used for cell experiments.

2.14. Tube formation assay

For tube formation assay, 80 μL of Geltrex solution was added to 48 well cell culture plate and incubated at 37 °C to coat the wells. After 30 minutes, 50,000 cells/well were added in basal M200 (0.5% FBS) medium. HUVECs were allowed to attach for 1 h. The medium was replaced with M200 containing 5 ng/mL VEGF from stock, and medium collected from aptamer-functionalized and control hydrogels on day 10. Cells were incubated for 6 h. HUVECs were stained with 2 μg/mL calcein AM for 30 minutes and imaged under fluorescence microscope (Olympus IX73). The total length of the tube and branching points were quantified using image J.

2.15. C-MET gene transfection

MET proto-oncogene cDNA clones were purchased from Sino Biological. HUVECs were transfected using Lipofectamine LTX transfection reagent (Thermofisher), following the manufacturer’s instructions. After 24-48 h, the medium was replaced, and cells were analyzed for c-MET receptor expression.

2.16. Flow cytometer

The c-MET aptamer was purchased with 3’-end FAM modification (Table S1-Sn). FAM labeled aptamer was added to 1 × 10⁶ cells in 1 mL of PBS with BSA (0.5% w/v) and incubated for 30 minutes at 37 °C. After incubation, the cells were washed three times with DPBS to remove unbound aptamer. The cells were then analyzed using an EMD Millipore Guava easyCyte flow cytometer (Hayward, CA) and FlowJo software, version 7.6 (FlowJo, LLC, Ashland, OR, USA).

2.17. Cell loading and imaging

To study cell attachment and proliferation on 3D hydrogels, hydrogels were dehydrated by gently blotting with sterilized Kimwipe. To assess cell attachment, hydrogels with different c-MET aptamer concentrations (0, 0.5 μM,1 μM, and 2 μM) were functionalized. The hydrogels were placed in 48-well plate. 10 μL of cell suspension (5 × 10⁴ cells) was added onto each hydrogel. The cells were allowed to be absorbed into the hydrogel for 30 minutes. The hydrogel was incubated in 200 μL of M200 supplemented with 10% FBS overnight. The cells in the hydrogel were stained with calcein AM (2 μM) for 30 minutes at 37 °C. The bulk hydrogel fluorescent intensity was quantified using Maestro In vivo imaging system. To assess cell proliferation, the concentration of c-MET aptamer was kept at 2 μM. Cells were seeded onto the hydrogel in a similar manner. At predetermined time points, the cells in the hydrogel were stained with calcein AM (2 μM) for 30 minutes at 37 °C. The bulk hydrogel was imaged under Maestro In vivo imaging system. The cell viability was assessed via CellTiter-Glo^® Luminescent Cell Viability Assay (ATP) kit. The fluorescent images of the cells in the hydrogel were acquired under a confocal microscope Zeiss LSM 880.

The synergistic effect of aptamer-mediated cell loading and protein release on cell viability and proliferation was studied by seeding cells in VEGF loaded hydrogels. In brief, 100 ng VEGF was loaded into the three control groups (blank, VEGF aptamer and c-MET aptamer-functionalized hydrogels) and dual aptamer-functionalized hydrogels. On the next day HUVECs were seeded onto hydrogels using the procedure discussed above. The cell medium was replaced with M200 + 0.5% FBS every day. At predetermined time points, the cell viability and proliferation were assessed using calcein AM staining and CellTiter-Glo^® Luminescent Cell Viability Assay (ATP) kit.

2.18. Statistical analysis

Unless otherwise stated, all quantitative evaluations were performed on measurements made in triplicates. Statistical analysis was performed with the graphic software Prism v.9 ( GraphPad Software Inc, La Jolla, CA) Two-tailed, nonparametric t tests were carried out with the default parameters when comparing two groups. One-way analysis of variance (ANOVA) followed by Bonferroni post-test was used to evaluate statistical significance between datasets of multiple groups. Confidence intervals equal to or less than 0.05 were considered statistically significant.

3. Results and Discussion

3.1. Synthesis and characterization of dual aptamer-functionalized hydrogels

We fabricated aptamer-functionalized hydrogels using free radical polymerization coupled with a freezing procedure. PEGDA and Acrydite-modified aptamers were mixed with APS/TEMED, and the reaction mixture was immediately transferred to a −20 °C environment (Figure 1A). After overnight incubation, the hydrogels were thawed at room temperature to melt the ice crystals inside the hydrogels. As ice crystals function as porogens, this method would enable us to form macroporous structures. Macroporous structures can be acquired using other methods.^38–42 For instance, gas formation can be applied to generate porous structure in the hydrogels.⁴² Gas foaming is fast.⁴³ It does not need specific treatment. However, as the reaction is too fast, it is difficult to control the pore formation.⁴³

Figure 1.

Fabrication and characterization of dual aptamer-functionalized hydrogel. A. Illustration of synthesis of aptamer-functionalized hydrogels via free radical polymerization coupled with a freezing-thawing procedure. Both aptamers (VEGF aptamer and c-MET aptamer) are incorporated into the pre-gel solution containing poly(ethylene)glycol diacrylate. After the addition of APS and TEMED, the pre-gel solution is allowed to freeze. The frozen hydrogel is thawed at room temperature. B) Image of the bulk hydrogel. (BF: bright-field image) The c-MET and VEGF aptamers were stained with their complementary sequences labeled with FAM and Cy5, respectively. C) Multiphoton image of the surface and z-stack image of the hydrogel stained with Cy5 labeled complementary sequence. D) Scanning electron microscope image of the hydrogel under different magnifications. E) Pore distribution of the hydrogel. F) Porosity of the aptamer-functionalized hydrogel (Apt (+)) and control (Apt (−)) hydrogel. G) The swelling ratios of Apt (+) and Apt (−) hydrogels. H) The stress vs. strain profile of Apt (+) and Apt (−) hydrogels. I) The Young’s modulus was calculated from the stress vs. strain curve of the hydrogels.

In addition, the removal of gas bubbles from hydrogels may be difficult. The freezing/thawing procedure does not have the issues of gas formation. Thus, we used this polymerization-freezing-thawing method in the current work. This method has not been used to synthesize aptamer-functionalized hydrogels previously.

We first examined if the two aptamers could be incorporated into the hydrogel with this method. The hydrogel was treated with FAM labeled CS of the c-MET aptamer and Cy5 labeled CS of the VEGF aptamer and then thoroughly washed. If the two aptamers were incorporated into the hydrogel network, they would hybridize with their CSs. Otherwise, CSs would be washed from the hydrogel. The aptamer-functionalized hydrogel (Apt (+)) exhibited a much stronger fluorescence intensity of both FAM and Cy5 than that of control hydrogel without the aptamer (Apt (−)) (Figure 1B). This result shows that the two aptamers were incorporated into the hydrogel network during the polymerization coupled with freezing.

We examined the pore structure and distribution via multiphoton microscopy and SEM. The hydrogel had an interconnected pore structure with sturdy walls. The pores were overall uniform throughout the hydrogel (Figure 1C). The average pore diameter ranged from 40-50 μm (Figures 1D and 1E). We further assessed the effects of aptamer incorporation on porosity, swelling and mechanical properties. As shown in Figures 1F and 1G, the porosity and swelling ratio had no significant difference between the Apt (+) and Apt (−). These two hydrogels also exhibited virtually the same stress-strain relationship (Figure 1H) and Young’s modulus (Figure 1I). When the PEG hydrogels were subjected to cyclic compression at 80% strain, they were able to retain the elasticity for consecutive loading-unloading cycles (Figure S1-Sn). These measurements showed that the aptamer incorporation did not significantly affect the physical properties of the hydrogels. This negligible effect of aptamers on the physical properties of the hydrogels could be attributed to the low concentration of aptamer in comparison to the concentration of monomers used to synthesize the hydrogels. The total concentration of aptamers per hydrogel was 4 μM (VEGF aptamer (2 μM) and c-MET aptamer (2 μM)) whereas PEGDA concentration was 160 mM. Alternatively, the weight ratio (w/w) of aptamer to monomer was 1/2000.

Moreover, the FT-IR spectra of hydrogels were acquired to monitor the conversion of C=C to C-C during free radical polymerization. The symmetrical C=C stretch at 1620 cm⁻¹ and vinyl C-H out of plane bending 815 cm⁻¹ of PEGDA hydrogel spectrum disappeared from the PEGDA spectrum after reaction, indicating the consumption of C=C bonds for crosslinking (Figure S2-Sn).

Hydrogels developed for mimicking the ECM are usually loaded with proteins and/or cells during their formation.^44,45 In contrast, the synthesis of the dual aptamer-functionalized hydrogels does not involve protein or cell loading. Proteins and cells are fragile and can easily lose their bioactivity if the formation of hydrogels involves harsh conditions.⁴⁶ For instance, the method we used in this study involves free radicals and ice crystal formation that can easily cause cell death. Thus, we decoupled the synthesis of the off-the-shelf hydrogels from the loading of proteins and cells. Next, we examined if we could load proteins and cells into the hydrogels and if the aptamers could sequester proteins and bind to cells in the hydrogels.

3.2. Examination of aptamer-mediated protein loading and release

Hydrogels without specific protein-binding sites sequester proteins only based on their porous structures. Diffusion is the major mechanism in determining protein release.^14,47 In contrast, aptamer-functionalized hydrogels sequester proteins for sustained protein release based on both aptamer-protein binding and diffusion (Figure 2A).¹⁴

Figure 2.

Assessment of VEGF sequestration, release, and bioactivity. A) Illustration of VEGF sequestration and release. Diffusion is the main mechanism for VEGF release from the Blank hydrogel. Diffusion coupled with a binding reaction is the main mechanism for VEGF release from the VEGF-A hydrogel. B) Retention of VEGF in Blank and VEGF-A hydrogels. The hydrogels were incubated in a washing buffer for 4 hours and the washing buffer was examined to quantify VEGF retention. (**** P<0.0001) C) Cumulative release of VEGF from hydrogels. D) Normalized apparent diffusivity (De/Do) of VEGF in the hydrogels. E) Bioactivity of VEGF released on day 10 from Blank and VEGF-A compared to stock VEGF in stimulating tube formation of HUVECs. F) Quantification of the tube length. G) Quantification of tube branching points. (ns: non-significant, * P< 0.05)

A VEGF aptamer was used as a model to functionalize PEG hydrogels for sequestering and releasing VEGF.⁴⁸ VEGF was loaded to the aptamer-functionalized hydrogels in 1 to 5 ratio to assess VEGF retention and release. The result shows that VEGF aptamer-functionalized hydrogel (VEGF-A) retained more than 80% of the loaded VEGF while Blank retained less than 20% (Figure 2B), suggesting that the aptamer could more stably sequester VEGF in macroporous PEG hydrogels. Next, the release of VEGF from aptamer-functionalized hydrogels was examined. The blank hydrogel (Blank) released VEGF abruptly (Figure 2C). By contrast, the VEGF-A was able to prolong the release of VEGF.

We calculated the apparent diffusivity by considering the overall effects of diffusion and binding reactions. The release data were directly fitted to a semi-empirical equation established with Fick’s second law of diffusion.³⁷ The equation is effective to simulate the first ~60% release of molecules.³⁷ The calculated apparent diffusivity of VEGF from the VEGF-A was (4.22 ± 0.63) x 10⁻⁹ cm²/sec while for the Blank, the apparent diffusivity was (7.42 ± 1.95) x 10⁻⁷ cm²/sec. The diffusivity of VEGF in aqueous solution is 12 × 10⁻⁷ cm²/sec according to the calculation from the Einstein-Stokes equation.

The aptamer-functionalized hydrogel can reduce the diffusivity of VEGF by two orders of magnitude (Figure 2D). The experimental results and theoretical calculation demonstrate that the presence of aptamers in hydrogels can significantly reduce the diffusion and release rates of proteins.

Proteins are fragile biomolecules. They can easily lose bioactivity when encapsulated in and released from biomaterials.⁴⁶ For example, the bioactivity of released VEGF was found to decrease by 80 to 95% within 7 to 10 days.^49,50 As the maintenance of protein bioactivity is important to the function of a biomimetic ECM construct, the bioactivity of released VEGF at the last day was examined using the tube formation assay. When treated with medium collected from the Blank, the HUVECs remained dispersed without forming tubes. In contrast, when treated with medium collected from the VEGF-A, the cells formed spindles and cell networks (Figure 2E). The total tube length of HUVECs cultured with the VEGF extracted from the VEGF-A was 7.9 mm/mm², whereas that with the stock VEGF of the same concentration was 9.1 mm/mm² (Figure 2F). Moreover, the total measured branching points were similar to the stock VEGF treated samples (Figure 2G). These results show that VEGF in the aptamer-functionalized porous hydrogel could maintain high bioactivity.

The natural ECM has components binding to signaling molecules.⁵ For instance, heparan sulfate is a glycosaminoglycan in the ECM and it can sequester and release VEGF.^51,52 Synthetic hydrogels without specific functionalization do not have those components. Thus, synthetic hydrogels, particularly macroporous hydrogels, cannot stably sequester proteins.⁵³ Our data show that aptamer-mediated protein binding can significantly reduce the apparent diffusivity of proteins in the hydrogels. Thus, aptamers are promising for mimicking the protein-binding sites of the natural ECM in sequestering proteins for controlled release.

3.3. Examination of aptamer-mediated cell adhesion and growth

The c-MET aptamer was selected using the cell-SELEX method with a K_D value in the range of tens of nM.^54,55 The c-MET aptamer was incorporated into the PEG hydrogels as binding sites for the attachment of HUVECs (Figure 3A). The aptamer can bind to HUVECs expressing the c-MET receptor (Figure 3B).

Figure 3.

Examination of c-MET aptamer-mediated cell attachment. A. Schematics illustration of cell attachment. B. Flow cytometry analysis of c-MET aptamer binding to c-MET receptor expressed on the surface of HUVECs. The cells were incubated with FAM labeled c-MET aptamer (200 nM) for 15 minutes at 37 °C. C. Effect of the aptamer concentration on cell attachment onto the aptamer-functionalized hydrogels. The values were normalized to that of Blank hydrogel (0 μM). D. Representative fluorescence image of cell loaded hydrogels. The hydrogels were incubated in complete medium (M200 + 10% FBS). The cells were stained with calcein AM, and hydrogels were imaged under the Maestro In vivo imaging system. Scale bar: 1mm. E. Examination of cell survival and proliferation via the ATP proliferation assay. (ns: nonsignificant, * P< 0.05, ** P< 0.01) F. Confocal microscopy images of live cells in the hydrogel. Scale bar: 200 μm.

We first examined the effect of the concentration of the aptamer on cell binding. Cell loading was concentration dependent (Figure 3C). Next, we studied whether aptamer-mediated cell binding promoted cell proliferation in the hydrogels. We loaded HUVECs in the hydrogels and collected the samples for the measurement of cell proliferation at day 1, day 4, and day 7. The result shows that cells could survive or even proliferate in the c-MET aptamer-functionalized hydrogel (c-MET-A) whereas they could not in the Blank hydrogel (Figures 3D and 3E). To confirm this result, we also used a confocal microscope to examine the cells in the hydrogels (Figures 3F). The confocal analysis is consistent with the bulk hydrogel image and molecular analysis, suggesting that HUVECs could attach and survive in the aptamer-functionalized hydrogels.

The ECM components such as collagen and fibronectin can interact with cell receptors for cell attachment.⁵ Cell attachment is vital for the survival and proliferation of adherent cells.⁵⁶ The loss of adhesion sites lowers cells viability due to anoikis, programmed cell death induced by the loss of cell-ECM interaction.^57,58 Our data show that the cells loaded in the blank PEG hydrogels underwent death with time as PEG itself does not have functional moieties for cells to attach. In contrast, the cells in the c-MET-A could survive. Aptamers have several advantages compared to natural ECM components and other affinity ligands for the functionalization of hydrogels. First, as aptamers have high affinity, they only need small quantities for the functionalization of hydrogels.^18,23 Other ligands such as peptides may be used in larger quantities (millimolar).^31,59 Second, while we used an unmodified c-MET aptamer in this work, aptamers can in principle be chemically modified to resist degradation and prolong the duration of cell binding.^18,60 Third, aptamers can be selected to bind to a specific receptor on cells as they have high specificity for their targets.^18,61 This feature allows for cell specific attachment to hydrogels and avoids the attachment of undesired cells. It has the potential to incorporate multiple aptamers into hydrogels for the loading and attachment of different types of cells for constructing a more dynamic and complex tissue mimic.

3.4. Examination of cell survival and growth in the dual aptamer-functionalized hydrogel

After individually demonstrating the two functions of VEGF sequestration and HUVEC attachment, we integrated the VEGF and c-MET aptamers into the PEG hydrogel and evaluated HUVEC survival and growth in the dual aptamer-functionalized hydrogel (Figure 4A). The control hydrogels were single aptamer-functionalized hydrogels, i.e., hydrogels functionalized with either the VEGF aptamer or the c-MET aptamer. The c-MET aptamer was used to serve as an anchor for cell attachment while VEGF aptamer was used to sustain the release of VEGF. As both cell attachment and VEGF stimulation are important for HUVEC survival and growth, we anticipated to observe synergistic effects in the presence of both aptamers in comparison to hydrogels where only one aptamer is present.

Figure 4.

Examination of the synergistic effect of two aptamers on cell survival and proliferation. A. Schematics illustration of protein sequestration and cell attachment in a dual aptamer-functionalized hydrogel. B. Quantification of the cell survival and proliferation via ATP proliferation assay. The hydrogels were loaded with 100 ng of VEGF and incubated in a low-serum (M200 + 0.5% FBS) medium. It is important to note that the culture medium condition was different from Figure 3. (ns: nonsignificant, * P< 0.05, ** P< 0.01) C. Confocal microscopy images of live cells in the hydrogels. Scale bar: 200 μm.

The HUVECs were examined using the ATP proliferation assay and confocal imaging. The ATP proliferation data show that HUVECs could survive and proliferate in the dual aptamer-functionalized hydrogels with time (Figure 4B). By contrast, the cells in the control hydrogels did not. It is important to note that the experimental conditions for this evaluation were different from those used for examining the function of the individual c-MET aptamer-functionalized hydrogel as shown in Figure 3. In the Figure 3 test, the cells were cultured using a normal cell culture medium. In the Figure 4 test, the cells were cultured using a low-serum cell culture medium in order to clearly examine the function of VEGF sequestration. The confocal imaging analysis is consistent with the ATP proliferation assay (Figure 4C). Taken together, the data demonstrate that the presence of two aptamers has a synergistic effect on HUVEC survival and growth, suggesting that dual aptamer-functionalized hydrogels are a promising ECM mimic with two critical functions of cell attachment and protein sequestration.

Aptamers can bind not only their protein targets but also their complementary sequences (CSs).^62,63 In the presence of CSs, aptamers can hybridize with CSs for self-inactivation and the inactivated aptamers can further release their target proteins or cell receptors.^22,64–66 Thus, while it is not the focus of the current study, future efforts can be made to use CSs to dynamically regulate the binding states of proteins and cells.

4. Conclusions

Dual aptamer-functionalized hydrogels can be synthesized using free radical polymerization coupled with a freezing-thawing procedure. The hydrogels have a macroporous structure with the ability to sequester bioactive proteins and living cells. The presence of the aptamers does not significantly change the physical properties of the hydrogels. However, the aptamers can significantly reduce the apparent diffusivity of loaded proteins and prolong protein release. Moreover, the aptamers can promote cell attachment within the macroporous hydrogel network. The presence of both the protein-binding aptamer and the cell-binding aptamer can lead to a synergistic effect on cell survival and proliferation. Thus, this work has successfully demonstrated the potential of developing a bi-functional biomimetic ECM using dual aptamer-functionalized hydrogels with two key functions of cell attachment and protein sequestration.