Biofunctional Polyacrylamide Hydrogels using Tetrazole‐Methylsulfone Comonomer for Thiol Conjugation

Biofunctionalized polyacrylamide (PAAm) hydrogels are important 2D substrates for studying cell physics and mechanobiology. In this work, an arylmethylsulfone (MS) comonomer is developed that can be incorporated into PAAm gels under aqueous radical polymerization conditions. The resulting hydrogels show similar properties to unmodified PAAm gels, indicating that the comonomer is incorporated without affecting PAAm physical properties. The MS‐containing PAAm hydrogels allow efficient conjugation of thiol derivatized biomolecules and require very low comonomer content (2 mM, 0.18 mol% relative to AAm) and thiol incubation amounts (≥ 0.15 µg per gel) to achieve functional densities that elicit cell responses. Compared to carboxyl‐functionalized PAAm hydrogels, a 10‐fold lower comonomer concentration and a 10‐fold lower ligand feed concentration are sufficient to achieve comparable cell adhesion responses. The new comonomer opens up possibilities for efficient and straightforward biofunctionalization of PAAm hydrogels used in cell biophysical studies.


Introduction
Functionalized polyacrylamide (PAAm) hydrogels are widely used soft 2D platforms for studying cell biophysics and mechanobiology.PAAm gels offer optical transparency, customizable stiffness spanning the range of soft tissues (Young's modulus from 0.1 -100 kPa) and can be functionalized with specific matrix proteins to transform them from bioinert substrates [1] into bioactive ones.They can be patterned to DOI: 10.1002/admi.202301024generate well-defined 2D [2] and 2.5D topographies for cells, [3] mixed with fluorescent nanoparticles for cell traction force microscopy measurements, [4,5] or combined with conductive materials to introduce electrosensing properties. [6]AAm hydrogels are inert and interaction with cells is achieved by functionalization with cell-adhesive ligands.Grafting biomolecules to PAAm requires the introduction of reactive sites into the gel network.This can be done by activating the hydrogel after polymerization or by incorporating comonomers with reactive side groups in the gel preparation step.Activation of pre-formed PAAm networks using UV photons, [7,8] the photoactivatable reagent sulfo-SANPAH, [9,10] or hydrazine [11] introduces functional groups in the network capable of coupling amines (UV, sulfo-SANPAH) or aldehydes/ketones (hydrazine).A low specificity over the active site generation and therefore the coupling chemistry and biomolecule loading is the main drawback of these approaches. [12]Copolymerizing with functional comonomers allows a better-defined chemical process and a higher control of the coupling density, provided the comonomer tolerates and does not interfere with the radical polymerization of the acrylate groups.Copolymerization with acrylic acid (AA) [13][14][15] or its NHS-activated ester [16] has been used to mediate coupling with amine-terminated biomolecules, copolymerization with hydroxyl-bearing acrylate has been used to attach fibronectin and laminin by non-specific interactions, [17,18] and copolymerization with streptavidin-acrylamide allowed binding of biotinlabelled molecules. [19]These approaches also present some drawbacks.The NHS-activated reaction of AA comonomer is very sensitive to hydrolysis and pH.][22][23][24] Due to the ionic character of the COOH group, these AA concentrations lead to higher swelling and nonspecific protein absorption on the resulting PAAm-copolymer networks.The commercially available comonomer streptavidin-AAm is very expensive.
Our group reported arylmethylsulfone (MS)-bearing acrylamide monomers as a new class of comonomers that allow efficient conjugation of thiol-functionalized molecules to PAAm gels. [13,14]This work was inspired by reports of MS derivatives showing rapid and selective reaction with thiols including thiolbearing proteins. [25,26]The MS species undergo nucleophilic aromatic substitution with thiolate, releasing methanesulfinic acid as leaving group and forming a stable thioether linkage.Compared with the widely used thiol-labelling group maleimide, [27] MS derivatives react selectively with thiols, form more stable conjugates, and are unaffected by amines. [26][32] MS derivatives with thiol reaction rate coefficients spanning 9 orders of magnitude have now been reported, [31,32] with rate coefficients for the initially-reported benzothiazole (Bt-), tetrazole (Tz-), and oxadiazole (Ox-) MS derivatives spanning 3 orders of magnitude in the middle of this range. [26]Our group was the first to show that MS functions also tolerate radical polymerization conditions and can therefore be incorporated into PAAm gels to allow subsequent thiol conjugation.The inertness of MS functions to radical polymerization and their selective reactivity towards thiols in the presence of amines opened up new coupling capabilities for MS-containing PAAm gels, including the ability to orthogonally bi-functionalize a single gel with both amines and thiols. [14]Thiol-reactive PAAm hydrogels are appealing since a broad range of bioactive ligands intrinsically contain or are specifically functionalized with thiols for coupling.PAAm-co-MS hydrogels containing OxMS-and BtMS-acrylamide comonomers (Scheme 1A, molecules 2 and 3) were reacted with cell-adhesive ligand c[RGDfC] in a simple incubation step in PBS. [13]Coupling was thiol-specific and essentially quantitative (> 90%) under physiological conditions.The high efficiency of the thiol-MS reaction allowed the use of low comonomer concentration (12.1 mM, which corresponds to < 2 mol% relative to AAm) and, therefore, the physical properties of PAAm-co-MS hydrogels (stiffness, swelling ratio, protein adsorption, and network structure) did not differ from PAAm hydrogels.PAAm gel formation with 2 and AA comonomers allowed orthogonal incorporation of both thiol-and amine-terminated molecules at flexible concentrations to generate bifunctional substrates for cell culture. [14]espite offering a highly efficient pathway for PAAm biofunctionalization, the developed MS-acrylamide monomers 2 and 3 are relatively hydrophobic and show limited water solubility (≈2 mg mL −1 ). [13]PAAm-co-2/3 hydrogels had to be prepared using DMF (11% v/v) as co-solvent to ensure solubility of the MS component.This demands extensive gel washing for complete removal of DMF before cell culture.In addition, in follow up work we have shown that the linkages formed by thiol conjugation to Ox-and BtMS functions are less stable than the linkage with alternative MS derivatives. [33]PEG networks formed from TzMS precursors were stable for 15 days under cell culture conditions while Ox-and BtMS gels were stable for 6 and 10 days respectively.
The aim of the present work was to develop a new MSacrylamide monomer that offers improved water solubility to allow PAAm-co-MS gel preparation in aqueous solution and improved stability for long cell culture experiments.We describe the design and synthesis of monomer 1 (Scheme 1), which contains TzMS and acrylamide termini linked by an oligo(ethylene glycol) (OEG) of 7 repeat units.Beyond imparting water solubility, the long spacer adds flexibility to the network-bound TzMS functional groups and is expected to facilitate accessibility of the conjugated ligands to the cell membrane.The synthesis route of the new monomer avoids the need for protective group chemistries by using orthogonal chemical reactions.This allowed us to increase functionality of the molecule without increasing the synthetic effort.Here we describe the physical properties and biofunctionalization of the PAAm-co-1 hydrogels with thiol-containing cell adhesive ligands.The superior efficiency of the MS-mediated bioconjugation reaction and the consequences for cell adhesion are quantified and compared to those of PAAmco-AA hydrogels.

Synthesis of monomer 1
A MS-acrylamide monomer, referred to as monomer 1 (Scheme 1A), was designed and synthesized in this work to overcome the limited aqueous solubility of previously reported monomers 2 and 3 (Scheme 1A). [13,14]Monomer 1 contains an OEG chain between the acrylamide and the TzMS functions to promote solubility in water.The TzMS and acrylamide functions were sequentially attached to H 2 N-PEG 7 -N 3 by copper-catalyzed azide-alkyne cycloaddition (CuAAC) and acylation without needing any protection/deprotection steps.By employing these orthogonal chemistries, the synthesis pathway for 1 thereby allows introduction of the OEG spacer without increasing the number of synthesis steps versus the pathways to 2 and 3 which required protecting group strategies. [13]The synthesis of 1 therefore achieves increased functionality courtesy of the watersoluble OEG component without increasing the synthetic effort versus the previous generation of MS-acrylamide monomers.In this work we chose the TzMS function over Bt-and OxMS alternatives due to the higher stability of thiol/TzMS conjugates in cell culture conditions. [33]Moreover, the synthesis precursors for Tz are significantly cheaper than BtMS and OxMS precursors with equivalent functionality.
The synthesis begins with methylation and etherification following reported protocols (Scheme 1B). [30]These steps proceeded in good yield without the need for chromatographic purification.Subsequent oxidation gave the MS derivative c in 88% yield at 1.4-g scale.Reaction of c with N 3 -PEG-NH 2 using CuAAC gave pure compound d in 45% yield after column chromatography.The final acylation step gave the desired compound in 44% yield at 50-mg scale.The overall yield of the synthesis pathway is 14%.Compound 1 was stable in storage at −20 °C in the dry state and in water solution over a tested period of 2 months (Figure S1, Supporting Information).
The solubility of 1 was determined by analyzing the concentration in a saturated aqueous solution using HPLC, giving 6.7 mg mL −1 (Figure S2, Supporting Information).This exceeds the 2 mg mL −1 solubility limit reported for 2 and 3 in our previous work and is sufficient to prepare water solutions at the concentration required to generate functionalized PAAm hydrogels, as detailed below.

Hydrogel Synthesis and Biofunctionalization
PAAm-co-1 hydrogels were prepared by copolymerization of acrylamide (AAm), bisacrylamide (MBA), and comonomer 1 in aqueous conditions (Scheme 2).An 8% w/v AAm monomer concen-tration and a MBA:AAm ratio of 3.2% w/w were used to target PAAm hydrogels with a Young's modulus around 20 kPa, according to reported literature. [34]Comonomer 1 was included at a 1:AAm ratio of 0.18 mol%, which corresponds to a 2 mM concentration of comonomer 1 in the pre-gel solution.Gel compositions are described in Table S1 (Supporting Information).Transparent hydrogel thin films with swollen thickness of ca.110 μm were formed on glass coverslips by radical polymerization.Transparency is expected in PAAm gels at MBA:AAm < 4% w/w [35] since the small crosslinking clusters (2 -4 nm) that form at all crosslinker concentrations agglomerate into domains that are smaller than the wavelength of visible light, and therefore do not cause significant scattering. [36]The addition of comonomer 1 preserves this transparency.After polymerization, the hydrogels were functionalized with c[RGDfC] by incubation in a PBS solution of the peptide at pH 7.4.Nanoindentation analysis showed a Young's modulus of 17.9 ± 1.02 kPa (Figure 1; Table S2, Supporting Information).A swelling degree of 7.93 ± 0.44, defined as the ratio of the swollen to dry film thickness, was found by confocal microscopy of hydrogel films loaded with fluorescent beads (Table S2; Figure S3, Supporting Information).
To study the impact of comonomer 1 on hydrogel properties, we compared the properties of PAAm-co-1 to those of PAAm hydrogels.Comonomers with PEG spacers can generate more porous hydrogel structures and change swelling properties due to steric hindrance during PAAm polymerization. [37,38]AAm-co-1 hydrogels show similar swelling and Young's modulus to PAAm hydrogels (Figure 1; Table S2, Supporting Information).This demonstrates that the inclusion of the comonomer 1 at 2 mM concentration does not significantly affect the hydrophilicity of the PAAm network and, therefore, the mechanical and swelling properties of the hydrogel films remain unchanged.
We also compared the properties of PAAm-co-1 to those of PAAm-co-AA (AA = acrylic acid) and PAAm-co-ACPEG (ACPEG = 1 kDa acrylate-PEG-COOH).The ACPEG monomer contains a long PEG spacer for better comparison to our comonomer 1.According to reported work, PAAm-co-AA hydrogels for cell culture need to use high AA concentrations in the monomer solution (14 to 350 mM) [14,[20][21][22][23][24] to support cell attachment after biofunctionalization with RGD.These concentrations are much higher than the concentration of MS comonomers used in our work.We decided to use a 23.3 mM concentration of AA or ACPEG comonomers for the comparative experiments, which is at the low end of the reported works and a 10-fold higher concentration than comonomer 1 in our PAAm-co-1 hydrogel.The copolymer networks were functionalized with c(RGDfK) after EDC/NHS activation using the same RGD concentration for coupling as for PAAm-co-1.RGD-functionalized PAAm-co-AA and PAAm-co-ACPEG hydrogels showed similar swelling to PAAmco-1.PAAm-co-ACPEG showed a slightly higher Young's modulus of 23.4 ± 0.36 kPa (Figure 1B; Table S2, Supporting Information).The ionizable carboxyl groups (only a fraction of which are modified with RGD) introduce some charge in the network, which could lead to higher swelling.However, the carboxyl comonomer concentrations used in the networks do not seem to have a strong influence on the final properties of the hydrogel.
Thiol conjugation to TzMS groups in PAAm-co-1 hydrogels was quantified by fluorescence spectroscopy after incubation   with solutions containing increasing amounts (0.5 to 5.2 nmol) of the thiol-bearing fluorophore FITC-PEG-SH (1 kDa) for 2 h followed by washing (Figure 2A).This concentration range corresponds to the concentrations of RGD peptide used in subsequent cell experiments.Note that the amount of comonomer 1 in the precursor solution for each gel was 24 nmol.At the tested concentration range of the thiolated dye, the amount of bound dye increased linearly with the concentration of the dye in the solution (Figure 2A).The coupling efficiency, determined as the ratio between bound and feed fluorophore at the highest feed amount (5.2 nmol) was 47.7 ± 0.9% (Figure 2B; Table S3, Supporting Information).We also tested the availability of TzMS groups by incubating gels with an excess of the thiolated dye and fitting the fluorescence intensity to calibration standards spanning the investigated range (≤ 31 nmol; Figure S4; Table S4, Supporting Information).We found that 13.3 ± 2.5 nmol of TzMS groups reacted with the thiolated dye, accounting for ≈55% of the TzMS groups in the hydrogel precursor solution.Detection of fewer TzMS groups than were included in the hydrogel preparation could arise from incomplete incorporation of 1 into the PAAm network, or inability for all TzMS groups to be reacted with dye in the gel, e.g., due to slow diffusion or inaccessibility in the network structure.
The conjugation of the dye across the gel thickness was imaged by fluorescence microscopy (Figure 2C).A ca. 20% decrease in the fluorescence intensity was observed from the top to the bottom of the hydrogel.Considering that in our experimental conditions the dye is the limiting reactant, the decay in fluorescence intensity could be explained by the relatively rapid rate of thiol-MS reaction relative to the rate of diffusion into the gel, which would have favored the reaction near the top surface at early stages of incubation, leaving progressively fewer dye molecules available to react deeper in the gel.
The conjugation efficiency of the TzMS-thiol reaction in PAAm-co-1 was compared with the efficiency of the EDC/NHSmediated coupling of amine to PAAm-co-AA and PAAm-co-ACPEG gels.The carboxyl-containing comonomers were incorporated into PAAm gels at 23.3 mM (i.e., 10-fold higher than the concentration of TzMS) which approaches the concentrations typically used in literature [14,20,21] without significantly affecting the mechanical properties.The binding efficiencies of fluorescent amine FITC-PEG-NH 2 (1 kDa) to PAAm-co-ACPEG and PAAm-co-AA were 30.8 ± 4.8% and 6.5 ± 1.7% respectively (Figure 2B).The higher grafting efficiency of the ACPEG system versus the AA system highlights the benefit of a longer spacer.Both values are still significantly lower than the binding efficiency of the thiolated fluorophore to PAAm-co-1 despite 10-fold higher concentration carboxyl comonomers versus comonomer 1.These results indicate that the thiol-TzMS coupling is much more efficient than the EDC/NHS-mediated COOH/amine coupling, in agreement with results obtained for the previous generation of MS comonomers. [14]EDC/NHS coupling proceeds via a multistep mechanism involving activation of the carboxyl group with EDC, displacement of the EDC group with NHS, and attack by the amine.The first step is the ratelimiting step, and has a second-order rate constant [39] several orders of magnitude lower than that of the thiol-TzMS reaction [30] under typical reaction conditions.Various side reactions including hydrolysis also compete with the desired reaction pathway and further decrease its efficiency. [40,41]In contrast, the thiol-TzMS reaction is a selective single-step reaction that proceeds rapidly under physiological conditions. [26]

Bioactivity of PAAm-co-1 hydrogels
The PAAm-co-1 hydrogels were tested as cell culture substrates after incubation with increasing concentrations of c[RGDfG] ligand.For comparison, PAAm-co-AA and PAAm-co-ACPEG hydrogels prepared as described above (i.e., containing 23.3 mM carboxyl comonomer) were functionalized with amine-bearing c[RDGfK] peptide.We hypothesized that the more efficient thiol-MS coupling would provide bioactivity and facilitate cell attachment at lower incubation concentrations of peptide, thus reducing the required amount of bioactive ligands necessary for cell studies.
MDCK II cells were seeded on the RGD-functionalized hydrogels and allowed to attach and spread for 24 h.Microscopy imaging showed significant cell attachment to PAAm-co-1 at RGD incubation amounts ≥ 0.15 μg (Figure 3A), with cell density increasing at higher RGD loadings (Figure 3B,C).Control PAAmco-1 hydrogels presenting the non-adhesive c[RADfC] peptide did not facilitate cell attachment, confirming the availability and specificity of cell binding to the coupled c[RGDfC] peptide (Figure S5, Supporting Information). [42,43]Gels containing the new comonomer 1 could therefore be efficiently functionalized and enabled stable cell attachment at low incubation amounts of peptide (0.15 μg, 1 mol% relative to TzMS groups).
We compared these results with cell densities on c[RDGfK]functionalized PAAm-co-AA and PAAm-co-ACPEG (Figure 3).On these substrates, stable cell attachment was required a minimum peptide amount of 1.5 μg, which is 10 times more than for PAAm-co-1 hydrogels even though both AA and ACPEG comonomers were present at 10-fold higher concentration than TzMS.Interestingly, the long PEG spacer in PAAm-co-ACPEG (ca.23 EG units per comonomer) did not improve cell attachment compared to PAAm-co-AA, contrary to previous results showing that increasing the spacing between cell-adhesive peptides and the hydrogel matrix promotes cell attachment and spreading at lower peptide concentrations. [44]The more efficient coupling of fluorescent thiol on PAAm-co-ACPEG gels (shown in Figure 2B) did not translate to improved cell attachment after biofunctionalization with the RGD peptide.Additional factors, beyond the scope of this work, may play a role in the final presentation and availability of cell adhesive ligands on the hydrogel surface.
Nanoscale organization and clustering of bioactive ligands for example have been shown to affect cell-substratum adhesion independent from the overall ligand density. [45,46]In summary, these results confirm that thiolated RGD conjugation to PAAm-co-1 is far more efficient than EDC/NHS coupling of amine-bearing RGD to carboxyl-containing PAAm gels, requiring 10-fold less ligand and 10-fold lower comonomer concentration to achieve a similar cell density.

Conclusion
We have developed a new water-soluble TzMS-acrylamide comonomer (1) that can be incorporated into PAAm hydrogels and mediate functionalization of the hydrogels with thiolated ligands under physiological conditions.The efficient thiol/TzMS reaction allows the preparation of biofunctionalized PAAm hydrogels for cell studies using a very low comonomer concentration that does not alter the swelling or mechanical properties of the PAAm network.In comparison with PAAm hydrogels containing carboxylated comonomers, a 10-fold lower concentration of both comonomer and cell adhesive ligand during the functionalization step are needed to achieve comparable cell responses.Incorporating comonomer 1 into PAAm hydrogels is positioned as the leading option for PAAm (bio)functionalization based on efficiency, orthogonality, cost, and maintenance of PAAm properties.
Equipment: pH measurements were performed with a Eutech Elite pH Spear (Thermo Scientific).Nuclear magnetic resonance (NMR) spectroscopy was performed with a Bruker Avance 300 MHz equipped with a He-cooled 5 mm TCI-CryoProbe, i.e., a proton-optimized triple resonance NMR "inverse" probe with external water-cooling unit (CP TCI 500S2, H-C/N-D-05 Z) from Bruker (Massachusetts, USA).All measurements were done at 298 K.The chemical shifts were recorded in parts per million ( ppm) with the NMR solvent peak used as reference.Spectra were analyzed using Bruker's TopSpin or Mestrelab Research's Mnova software.HPLC analysis and purification of the compounds were performed with a HPLC JASCO 4000 (Japan) equipped with a diode array, UV-vis detector and fraction collector.Reprosil C18 columns were used for semi-preparative (250 × 25 mm, flow 10 mL mi −1 n) and analytical (250 × 5 mm, flow 1 mL mi −1 n) runs.Solvent gradients using a combination of the following eluents were used: solvent A (MilliQ water + 0.1% TFA) and solvent B (95% ACN/5% MilliQ water + 0.1% TFA), with runs typically over 45 min Figure 3. Comparative analysis of MDCK-II attachment on RGD-functionalized hydrogels.A) Representative images of MDCK-II cells on PAAm gels after 24 h in vitro.Cells attach and spread on PAAm-co-1 hydrogels presenting integrin-binding peptide RGD even at 10-fold lower peptide amounts.In contrast, cell attachment on RGD-functionalized PAAm-co-AA and PAAm-co-ACPEG is only observed with higher amounts of the coating RGD peptide gels.Scale bar: 300 μm.B) Cell density at different coating amounts of RGD peptide.C) Cell density quantification on hydrogels coated with 3 μg of peptide, including RGD-coated PAAm homopolymer gels, and PAAm-co-1 gels coated with the non-adhesive RAD peptide serving as negative controls of cell attachment.Note that the concentration of carboxyl groups in PAAm-co-AA and PAAm-co-ACPEG for amine binding was 10-fold higher than the concentration of MS groups for thiol binding in PAAm-co-1.Data in B) and C) are presented as mean ± SE (standard error).Statistical significance was determined using the ANOVA test with Tukey multiple pairwise-comparisons (0.05 significance level).*p < 0.05, ns-not significant.N ≥ 3 per condition.
duration.Analytical runs were done at a flowrate of 1 mL min −1 and preparative runs were done at 10 mL min −1 .Mass spectrometry was performed with a 6545 AccurateMass Quadrupole Time-of-Flight (LC/Q-TOF-MS) with electrospray ionization from Agilent (California, USA).
Synthesis of Compound a: Adapted from Motiwala et al. [30] 4-(5mercapto-1H-tetrazol-1-yl)phenol (3.0 g, 15.4 mmol, 1 equiv.)was dissolved in anhydrous THF (20 mL) under nitrogen atmosphere and the solution was cooled to 0 °C in an ice bath.N,N-Diisopropylethylamine (DIPEA, 5.51 mL, 30.9 mmol, 2.0 equiv.) was added with a nitrogenpurged syringe, then iodomethane (1.92 mL, 30.9 mmol, 2.0 equiv.) was added dropwise over a period of 10 min.The ice bath was removed, and the reaction mixture was stirred at room temperature for 24 h, at which point thin layer chromatography (TLC, EtOAc/n-hexane 1:1 v/v) indicated complete consumption of starting material Rf = 0.3 and formation of a new spot at Rf = 0.4.The solvent was removed under vacuum, the residue was redissolved in EtOAc, and the organic solution was washed first with brine and then with water.The organic layer was dried over anhydrous magnesium sulfate and the solvent was removed under vacuum to give the product (

Synthesis of Compound b:
Adapted from Motiwala et al. [30] 4-(5-Methylthio)-1H-tetrazol-1-yl)phenol (compound a, 2.8 g, 13.2 mmol, 1 equiv.)and potassium carbonate (4.6 g, 33.1 mmol, 2.5 equiv.)were added into anhydrous DMF (7 mL) under nitrogen atmosphere and the resultant mixture was stirred at room temperature for 30 min.Propargyl bromide (80% w/w in toluene, 2.3 mL, 26.5 mmol, 2 equiv.)was added dropwise with a nitrogen-purged syringe over a period of 5 min and the reaction mixture was stirred at room temperature for 24 h, at which point TLC (EtOAc/n-hexane 1:1 v/v) indicated complete consumption of starting material Rf = 0.4 and formation of a new spot at Rf = 0.6.Ice cold water (ca. 100 mL) was added to precipitate out the product, which was isolated by vacuum filtration, washed several times with water (ca.20 mL) and dried.The crude solid was resuspended in n-hexane (ca. 100 mL), sonicated for 5 min and filtered.The recovered white solid was washed with n-hexane (ca.20 mL) and dried under vacuum to give the desired product (  S8 and S9 (Supporting Information).
Quantification of 1 Solubility in Water: To prepare a standard curve, known masses of 1 (0.15 -0.21 mg) were dissolved in 210 μL, 85 μL and 50 μL volumes of ACN respectively to give stock solutions of 1.0, 2.0, and 3.0 mg mL −1 .A 20 mL of each stock solution was injected into the analytical HPLC, and the peak area at retention time of 27.2 min (254 nm channel) was found by integration (Figure S1, Supporting Information).To determine the aqueous saturation concentration, water (100 uL) was added to 1 (1.53 mg) and the solution was sonicated for 45 min at room temperature.An aliquot (50 uL) was taken with a micropipette, transferred into an Epi, diluted with 50 uL of water, and 20 uL was injected into the HPLC.The peak area was converted into the concentration in the saturated solution using the standard curve, giving 6.67 mg mL −1 .
Synthesis of PAAm Hydrogels: PAAm-co-1 gels were prepared by adapting a reported protocol targeting a PAAm hydrogel with a 8.26% polymer content (%T), 3.19% crosslinker amount (%C), and a stiffness of 20 kPa (Table S1, Supporting Information). [34]Acrylamide (AAm, 29.81 mmol) and N,N'-methylenebisacrylamide (MBA, 0.46 mmol) were dissolved in phosphate-buffered saline (PBS, pH 7.4) to prepare a stock solution containing 21.5% w/v (2.98 M) AAm and 0.71% w/v (46 mM) MBA.The stock solution and 1 comonomer solution (3.2 mM, in deionized water) were mixed to obtain a final solution of 8% w/v (1.1 M) AAm, 0.264% w/v (17.1 mM) MBA, and 0.145% w/v (2 mM) comonomer 1. Oxygen was removed from the solution by degassing with argon for 5 min.Additionally, PAAm, PAAm-co-AA containing acrylic acid (AA, 23.3 mM) and PAAm-co-ACPEG containing acrylate-PEG-COOH (ACPEG-1 kDa, 23.3 mM) hydrogels were prepared by replacing 1 with deionized water and the respective comonomer, to achieve the same final concentrations of AAm and MBA.Before polymerization, the pH of the PAAm-co-AA precursor solution was adjusted to pH 7.4 with an aqueous solution of NaOH (1 M).Hydrogel films of 13 mm diameter were prepared between two glass substrates.The top coverslip (13 mm, VWR) was functionalized with 3-acryl-propyltrimethoxysilane (APTMS) to anchor the hydrogel films to the glass coverslips.For this purpose, coverslips were immersed in 0.5% APTMS solution in absolute ethanol for 1 h at room temperature (RT), washed twice with absolute ethanol, and air dried.The bottom substrate (glass slide, VWR) was coated with Sigmacote reagent to obtain a hydrophobic surface and enable hydrogel detachment after polymerization.Polymerization was started by adding the radical initiator ammonium persulfate (APS, 10% w/w in deionized water, 1/100 of total volume), and catalyst N,N,N′,N′tetramethylethylenediamine (TEMED, 1/1000 of total volume) to gel precursor solutions with thorough mixing.Precursor aliquots of 12 μL were quickly pipetted onto a Sigmacote-treated glass slide and covered with APTMS-treated coverslips.After 10 min, the entire glass slide housing 8 gels was immersed in PBS for 1 h, and the hydrogels bound to the coverslips were detached from the glass slide.Gels were thoroughly washed by three buffer changes to remove unreacted monomers and kept in PBS at 4 °C until further use.
Dye Coupling Assay: The thiol-bearing FITC-PEG-SH or the aminebearing FITC-PEG-NH 2 fluorescent dyes were used to assess the coupling efficiency of the thiol/MS and amine/COOH reactions, respectively.The hydrogels were incubated with 30 μL solutions containing different concentrations of fluorescent dyes, corresponding to 0.26 nmol, 2.6 nmol, 5.2 nmol, and 31.1 nmol in PBS, for 2 h at RT in a humid chamber, followed by washing with PBS (3 × 1 mL for 5 min each) to remove the unreacted dye.For dye coupling to PAAm-co-AA and PAAm-co-ACPEG gels, carboxylic acid groups were first activated by incubating each gel in a solution (100 μL) containing N-(3-(dimethylamino)propyl)-N′-ethyl-carbodiimide hydrochloride (EDC, 0.2 M), N-hydroxysuccinimide (NHS, 0.1 M), 2-(Nmorpho)-ethanesulfonic acid (MES, 0.1 M), and NaCl (0.5 M) for 15 min at RT, followed by washing once in 0.1 M MES buffer containing 0.5 M NaCl and three times in PBS (5 min each).Dye coupling and subsequent washing were then performed at the same concentrations and following the same protocol as described for PAAm-co-1 gels.Hydrogels were placed on a 30 μL droplet of PBS on a Parafilm-covered plate and the fluorescent intensity was measured using the TECAN Infinite M200Pro well plate reader (excitation wavelength: 488 nm; emission wavelength: 520 nm).For obtaining the calibration curve, the fluorescence intensity of solutions containing 0.26 nmol, 2.6 nmol, 5.2 nmol, 15 nmol, 21 nmol, and 31.1 nmol of fluorescent dyes in PBS, (30 μL/gel) was measured in the same way as specified for the samples above (Figure S4, Supporting Information).Data were then fitted with a linear model in the range ≤ 5.2 nmol (Figure S4A, Supporting Information).For higher dye concentrations (≤ 31 nmol), log transformations of fluorescent intensity and dye amount were used to obtain a linear fit (Figure S4B, Supporting Information).
Mechanical Characterization of Hydrogels: The mechanical properties of hydrogels were determined by nanoindentation using the Pavone Nanoindenter (Optics11, Amsterdam, Netherlands), equipped with a spherical probe (48 μm radius) attached to a cantilever with a 0.41 N m −1 spring constant.Prior to nanoindentation measurements, all hydrogels were functionalized with 3 μg of RGD peptide.The probe was calibrated in PBS at RT, according to the manufacturer's instructions.Indentation measurements were performed at 1 μm −1 s and each hydrogel was indented 25 times at different positions 200 μm apart.The Young's modulus was derived by fitting the resulting force/indentation curves up to 1 μm indentation depth to the Hertz model: where E is the Young's modulus of the hydrogel,  is the Poisson ratio (0.5 for PAAm hydrogels), 36 R is the radius of the spherical probe, and h c the indentation depth.Characterization of Swelling Properties: Freshly prepared hydrogel films containing 0.001% FluoSpheres fluorescent beads (200 nm diameter, 660/680 nm, Thermo Fisher) were equilibrated in PBS overnight.Hydrogels were functionalized with 3 μg RGD peptide to account for possible changes in swelling properties due to peptide coating.Z-axis confocal line scans were taken with a LD C-Apochromat 40× water immersion objective (1.1 NA) at random positions on the hydrogel using the LSM880 laser scanning confocal microscope (Zeiss).Images were acquired by exciting the fluorescent beads with a 633 nm diode laser (5.0% power) and emission was detected by an internal PMT detector (660-735 nm) with the gain set at 700.The line scans were converted into intensity profiles to determine the thickness of the hydrogel film in the swollen state (h s ).The hydrogel films were dried overnight at RT in the chemical hood with constant airflow and imaged using the same procedure to measure the thickness in the dry state (h d ).The swelling ratio (SR) was calculated as: Hydrogel Functionalization: c[RGDfC] (Genecust) solutions at six different concentrations (0.0001, 0.001, 0.005, 0.01, 0.05, and 0.1 mg mL −1 ) in PBS were prepared.Solutions of c[RADfC] (Genecust) were used for the negative controls.Hydrogels were incubated with 30 μL of peptide solution resulting in 0.003, 0.03, 0.15, 0.3, 1.5, and 3 μg of peptide per hydrogel.The incubation proceeded for 2 h at RT in a humid chamber to prevent drying, followed by washing with PBS (3 × 1 mL for 5 min each).Unreacted MS groups were blocked by following the same incubation and washing protocol using 2-mercaptoethanol (284 mM, 30 μL).Before incubation, the carboxylic acid groups of PAAm-co-AA and PAAm-co-ACPEG were activated as described above (Dye coupling assay).Peptide coupling and subsequent washing were then performed at the same concentrations and following the same protocol as described for PAAm-co-1 gels.Unreacted carboxylic acid groups were blocked by following the same incubation and washing protocol using ethanolamine (81.8 mM, 30 μL).
Cell Culture: MDCK-II cells (Madin-Darby Canine Kidney, Sigma-Aldrich) were grown at 37 °C and 5% CO 2 in DMEM medium (Gibco), containing 10% v/v fetal bovine serum (FBS, PAN Biotech) and 1% v/v penicillin-streptomycin (Thermo Fisher).Cells were dissociated enzymatically with 1 mL of TrypLE Express Enzyme (Thermo Fisher) for 5 min.Fresh medium was added to inactivate the enzyme followed by centrifugation at 250 g for 5 min.The cells were resuspended in fresh medium, and the number of live cells was determined by erythrosine B staining (Logos Biosystems).
Study of Cell Attachment: RGD functionalized hydrogels were transferred to sterile 24 well-plates (Greiner) and sterilized with 50% v/v aqueous ethanol for 5 min.The gels were washed three times with PBS (pH 7.4) for 5 min each.Each hydrogel was seeded with 500 μL of cell suspension at a density of 250 cells/mm 2 .Cells were grown for 24 h on functionalized hydrogels and imaged using the Cell Discoverer 7 microscope (Zeiss).Brightfield images were taken with a Plan-Apochromat 5× objective (0.35 NA) at 2× optical magnification.
Fluorescence Imaging and Image Analysis: The fixed samples were mounted with a fluorescence mounting medium (VECTASHIELD) and imaged using the Cell Discoverer 7 microscope (Zeiss).Epifluorescent images spanning an area of 5.24 × 4.8 mm (20 frames) in the center of each hydrogel were taken with a Plan-Apochromat 5× objective (0.35 NA) at 2× optical magnification.Cell density was determined using ImageJ by quantifying the number of cells per image area.The DAPI channel of each image was processed using a Median filter (radius = 8), followed by Otsu thresholding, and watershed segmentation to produce a binary mask and count the cell nuclei.At least 3 hydrogels for each hydrogel type and peptide coating from two independent experiments were analyzed.
Statistical Analysis: Data were analyzed and graphical plots were produced using R software.Quantitative measurements were analyzed via a Shapiro-Wilk test to assess normality and compared using the ANOVA test with Tukey multiple pairwise-comparisons.A p-value less than 0.05 was considered statistically significant.

Scheme 2 .
Scheme 2. Workflow for hydrogel preparation showing A) the comonomer structures incorporated into PAAm hydrogels, B) the resulting hydrogel structures, C) the conjugation conditions for attaching cell-adhesive peptides, and D) the resulting cell-adhesive hydrogels.

Figure 1 .
Figure 1.Swelling and mechanical properties of RGD-functionalized PAAm copolymers (co-AA, co-ACPEG, and co-1) containing 23.3 mM AA/ACPEG and 2 mM comonomer 1 compared to PAAm homopolymer hydrogels.The copolymer hydrogels were functionalized with 5.2 nmol (3 μg) of RGD peptide and the PAAm hydrogel was also incubated with the peptide to keep identical protocols.A) Swelling ratio as quantified by the ratio of the swollen and dry hydrogel film thickness obtained by confocal imaging of hydrogels loaded with fluorescent beads.B) Young's modulus obtained by nanoindentation measurements.Data are presented as mean ± SE (standard error).Statistical significance was determined using the ANOVA test with Tukey multiple pairwise-comparisons (0.05 significance level).*p < 0.05, ns-not significant.N = 4 per hydrogel type.

Figure 2 .
Figure 2. Coupling efficiency of the MS/thiol reaction.A) Quantification of the amount of fluorophore bound to the hydrogels after incubation and washing compared to the initial amount of incubated fluorophore.The initial amounts of 5.2 nmol, 2.6 nmol, and 0.5 nmol fluorophore correspond to the three highest amounts of RGD peptide used in cell experiments.B) Coupling efficiency was calculated as the ratio of the bound fluorophore and the initial amount of incubated fluorophore (5.2 nmol).Note that the concentration of carboxyl groups in PAAm-co-AA and PAAm-co-ACPEG for amine binding was 10-fold higher than the concentration of MS groups for thiol binding in PAAm-co-1.Data are presented as mean ± SE (standard error).Statistical significance was determined using the ANOVA test with Tukey multiple pairwise-comparisons (0.05 significance level).**p < 0.01, ***p < 0.0001.C) Orthogonal projections of hydrogel cross-sections after incubation with 5.2 nmol FITC-PEG-SH (PAAM-co-1) or FITC-PEG-NH 2 (PAAm-co-ACPEG and PAAm-co-AA).