View all hidden authors and organizations

Edited by David A. Weitz of Harvard University in Cambridge, Massachusetts, approved on August 8, 2021 (reviewed on June 15, 2021)

Hydrogels are widely used in cell culture, tissue engineering and flexible electronics. In all these applications, the mechanical properties of hydrogels play an important role. Although a lot of research has been devoted to optimizing stiffness, strain, toughness, and dynamic mechanical response, little consideration is given to the mechanical uniformity of hydrogels. By developing a general strategy for controlling the mechanical uniformity of hydrogels, we show here that nanoscale changes in matrix stiffness can significantly affect the lineage specification of human embryonic stem cells. Heterogeneous hydrogels inhibit mechanical transduction and promote dryness maintenance, while homogeneous hydrogels promote mechanical transduction and osteogenic differentiation. Therefore, engineered hydrogels with controllable and well-defined nano-level uniformity may have important implications for stem cell culture and regenerative medicine.

Due to the wide range of biological macromolecules and nano-level hierarchical assembly structures, the extracellular matrix (ECM) is mechanically inhomogeneous. Due to injury, fibrosis or tumorigenesis, mechanical inhomogeneities may be more pronounced under pathological conditions. Although considerable progress has been made in designing synthetic hydrogels to simulate ECM, the impact of the mechanical inhomogeneity of hydrogels has been widely ignored. Here, we have developed a method based on host-guest chemistry to control the uniformity of maleimide-thiol crosslinked poly(ethylene glycol) hydrogels. We show that mechanical homogeneity plays an important role in controlling the differentiation or stemness maintenance of human embryonic stem cells. Heterogeneous hydrogels can disrupt actin assembly and reduce YAP activation levels, while homogeneous hydrogels can promote mechanical transduction. Therefore, the method we developed to minimize the mechanical inhomogeneity of hydrogels may have a wide range of applications in cell culture and tissue engineering.

In tissues, cells reside in a complex extracellular microenvironment, and their mechanical properties often change in space and time during normal tissue homeostasis or disease development (1). More and more evidences show that changes in local mechanical properties can have a considerable impact on cell fate (2⇓ ⇓ ⇓ –6, 7⇓ ⇓ ⇓ ⇓ ⇓ –13). For example, the complex local mechanical environment can strongly affect wound healing and tissue regeneration (14, 15). In synthetic biomaterials (such as hydrogels) used in cell culture and tissue engineering, mechanical heterogeneity is also ubiquitous, although in many cases it is undesirable (10, 16⇓ ⇓ ⇓ ⇓ ⇓ –22 ). For hydrogels prepared by monomer polymerization, changes in local monomer concentration and heat released by chemical reactions can cause various defects in the hydrogel network. For hydrogels prepared by polymer chemical cross-linking, the wide distribution of polymer molecular weight and the uneven mixing before gelation can also cause significant changes in local mechanical properties. Although the influence of the overall mechanical properties of hydrogels on cell behavior has been widely explored, how nanoscale mechanical heterogeneity affects cell behavior is still poorly understood.

A major obstacle to solving this problem is the synthesis of hydrogels with uniformly distributed mechanical properties. By coupling a four-arm poly(ethylene glycol) (PEG) macromer with a narrow molecular weight distribution, Sakai and colleagues showed that hydrogels with minimal structural defects can be prepared (23). Chemical reactions widely used for gelation include click chemistry (24, 25), amine active ester reactions (26), maleimide-thiol conjugates (27, 28), and thiol-ene reactions (29, 30). The maleimide-thiol reaction is of particular interest because it is carried out under mild conditions, does not require catalysis, and does not produce small molecular by-products (27, 31⇓ ⇓ –34). Hydrogels prepared by the maleimide-thiol reaction have been widely used for organoid production (35), protein and cell delivery (36), and controlled release (37). However, the reaction is too fast to fully mix the macromonomer solution, resulting in uneven gelation. Due to changes in crosslink density, hydrogels usually contain microdomains with different mechanical properties (38, 39). Some methods have been developed to minimize the heterogeneity of hydrogels by slowing down this reaction, including lowering the gelation pH, changing the local pKa of thiols, or adding thiols to bind metal ions (38, 40⇓ ⇓ –43 ). However, these methods generally require non-physiological gel conditions or only result in limited improvements.

In this work, we introduced host-guest chemistry to slow down the maleimide-thiol reaction for hydrogel preparation. We found that maleimide can form a complex with β-cyclodextrin (β-CD), thereby reducing the concentration of free maleimide in the gel system. We show that the four-arm PEG hydrogel prepared by this method has fewer network defects and more uniform mechanical properties. In addition, we found that mechanical homogeneity can significantly affect the lineage specification of human embryonic stem cells (hESCs). We propose that this effect can be attributed to actin fiber assembly and subsequent destruction of the mechanical transduction pathway. We expect that this method can greatly improve the mechanical homogeneity of many cell culture systems to better regulate stem cell lineage specifications.

We decided to use β-CD as a potential guest molecule to protect the maleimide from the thiol group in the pregel solution, thereby reducing the gelation rate. The concept is shown in Figure 1. β-CD is one of the most widely studied subjects in supramolecular chemistry (44⇓ ⇓ ⇓ ⇓ –49), and its inner cavity is 6.1 Å, which is slightly larger than the diameter of maleimide (4.8 ~ 5.4 Å). The size of the inner cavity of β-CD is equivalent to that of maleimide, resulting in the formation of a lock-and-key structure. In addition, the succinimide ring of maleimide is relatively hydrophobic and can bind to the inner cavity of β-CD through hydrophobic interaction. Once the host-guest complex is formed, the maleimide succinimide ring becomes unable to enter the thiol group (Figure 1A). The no-reaction energy diagram explains in detail the effect of the formation of β-CD-maleimide complex on the rate of maleimide-thiol reaction (Figure 1B). The reaction speed of maleimide and mercaptan is fast, and the activation barrier (ΔG) is low. When maleimide forms a complex with β-CD, this activation barrier is greatly increased. The increase in the free energy barrier is equal to the free energy of complex formation (ΔGeq), which is determined by the binding constant of the β-CD-maleimide complex and the β-CD concentration. Therefore, by adding an appropriate amount of β-CD, we hope to form a more uniform hydrogel by mixing thiol and maleimide-terminated 4-arm PEG (Mw: 20 kDa, named PEG-SH and PEG-Mal, respectively) As the gelation rate is reduced (Figure 1 C and D).

The use of host-guest chemistry between maleimide and β-CD reduces the gel speed and produces a uniform hydrogel. (A) Schematic diagram of β-CD-maleimide complex. (B) Schematic diagram and free energy diagram of the effect of β-CD-maleimide complex on the maleimide-thiol reaction. (C and D) Schematic diagram of the hydrogel network formed in the absence of (C) and (D) β-CD.

We next used isothermal titration calorimetry (ITC) to measure the binding free energy of β-CD and maleimide by titrating β-CD into a PEG-Mal solution (Figure 2A). The binding constant (Ka) is approximately 6.66 × 104 M-1. The binding stoichiometry is estimated to be approximately 1.0 (1/0.995), indicating that a 1:1 complex is formed. We monitored the reaction by ultraviolet-visible (UV-vis) adsorption of the maleimide group, and further studied how the complexation with β-CD affects the rate of the maleimide-thiol reaction. We mixed PEG-Mal and PEG-SH, and observed that as the reaction progressed (SI Appendix, Figure S1), the UV absorption at ~300 nm decreased, indicating that maleimide was converted into a horse that does not absorb UV light. Leimide-thiol adduct. area. The formation of maleimide-thiol conjugates at different β-CD:PEG-Mal ratios is shown in the top panel of Figure 2B. As the concentration of β-CD increases, the response slows down significantly. In addition, based on the second-order reaction rate equation, the experimentally determined β-CD-maleimide binding constant was used to theoretically predict the reaction kinetics (Figure 2B, bottom). The theoretical predictions replicate the experimental results semi-quantitatively, confirming the proposed mechanism.

The binding between β-CD and maleimide affects the kinetics of the maleimide-thiol reaction. (A) The binding affinity of β-CD and PEG-Mal is determined by titrating the β-CD solution into the four-arm PEG-Mal solution using ITC. (B) Measuring (top) and calculating (bottom) four-arm PEG-Mal and four-arm PEG-SH at different β-CD and PEG-Mal ratios (0:1, 1:1, 2: 1, and 4:1 ) Monitored by UV spectroscopy over time at 25 °C. The dashed line indicates the magnitude of the calculated reaction kinetics.

Then, we prepared hydrogels by mixing equimolar PEG-Mal and PEG-SH in the presence of different ratios of β-CD. The inhomogeneity of cross-linking can be studied by monitoring the remaining thiol groups with a fluorescent probe (50), the fluorescence intensity of which is increased by more than 100 times after the reaction with thiol. After removing the unreacted fluorescent probes, a laser confocal fluorescence microscope was used to scan the hydrogel to detect the spatial distribution of unreacted thiol groups (SI Appendix, Figure S2). In the hydrogel prepared in the presence of β-CD, the fluorescent spots (which can also be considered as defects in the hydrogel network) are greatly reduced. Although compared with epi-fluorescence microscope, confocal fluorescence microscope can greatly reduce the excitation and detection volume, but when imaging hydrogel samples with fluorescent dyes stacked along the z axis, it will still appear blurred, resulting in a relatively bright background. The average brightness, density and area of ​​the fluorescent spots calculated from the projection image in the z-axis direction of the hydrogel are approximately 2 to 4 times higher in the hydrogel without β-CD than in the hydrogel with β-CD ( SI appendix, Figure S3). The optimal β-CD:PEG-Mal ratio is 1:1, and the higher β-CD ratio also introduces additional defects, which may be due to competition with the maleimide-thiol reaction that reduces the reaction efficiency. We also used 5,5'-dithiobis-(2-nitrobenzoic acid) (51) to quantify the unreacted thiols remaining in the hydrogel, and the results are consistent with the fluorescence measurements (SI Appendix, Figure S4 ). It is worth mentioning that by dialysis of water, β-CD can be almost completely removed from the hydrogel. The amount of β-CD washed out is 98.2% of the amount added during the gelation process (SI Appendix, Figure S5). These measurements indicate that β-CD-maleimide host-guest chemistry can increase the homogeneity of the hydrogel network by slowing down the reaction kinetics without introducing additional chemical modifications to the hydrogel.

Then, we studied the mechanical uniformity of the hydrogel by measuring local mechanical properties using an atomic force microscope (IT-AFM)-based indentation at sub-micron spatial resolution. As shown in Figure 3A, let the hydrogel film swell to equilibrium in water, and then install it on the glass substrate before measurement. The cantilever approached the surface of the hydrogel at a constant speed of 2 μm ⋅ s-1 and then retracted (Figure 3B). The Young's modulus of the hydrogel at a given location is calculated by fitting a close trajectory using a Hertz model. The representative elastic map (40 × 40 pixels) of the hydrogel surface prepared under different β-CD:PEG-Mal ratios is shown in Figure 3 CF. Compared with the hydrogel prepared with β-CD, the spatial variation of the Young's modulus of the hydrogel prepared without β-CD is significantly more obvious. The distribution of local Young's modulus in each hydrogel is summarized in Figure 3 CF and SI appendix insets, Figure S6 AD. The Young's modulus of hydrogels with β-CD:PEG-Mal ratios of 0:1, 1:1, 2:1, and 4:1 are 79.3 ± 24.2, 92.4 ± 12.7, 86.9 ± 10.9 and 83.6 ± 15.2, respectively kPa, respectively. The addition of β-CD resulted in a slightly higher overall Young's modulus and a narrower distribution. Both of these effects can be explained by improved network homogeneity. In addition, the width of the modulus histogram obtained from the two-dimensional Young's modulus image with a ratio of β-CD:PEG-Mal of 1:1 is smaller than those under other ratios, indicating the highest mechanical uniformity (SI Appendix, Figure 2). S6E). In addition, a similar distribution of Young's modulus was observed in hydrogels prepared with different PEG concentrations, with Young's modulus in a wide range of approximately 19 to 115 kPa (SI appendix, Figure S7-S10). The macroscopic mechanical properties of the hydrogel measured by the compression mechanics test showed the same Young's modulus trend (SI appendix, Figure S11-S13). The swelling rate, porosity and microstructure of all hydrogels are almost the same, indicating that the overall physical properties are not affected by the addition of β-CD (SI appendix, Figures S14 and S15). All these results indicate that β-CD-maleimide host-guest chemistry slows down the kinetics of the maleimide-thiol reaction and successfully generates a homogeneous hydrogel. We chose the best β-CD:PEG-Mal ratio of 1:1 for subsequent cell culture studies. The heterogeneous hydrogel prepared without β-CD and the homogeneous hydrogel containing β-CD are hereinafter referred to as I-gel and H-gel, respectively.

Mechanical homogeneity of hydrogels prepared in the presence of β-CD. (A) Schematic diagram of the IT-AFM experiment of the hydrogel sample. Spread the hydrogel on a glass cover glass and immerse it in pure water for AFM measurement. The tip of the cantilever extends into the hydrogel sample and retracts to obtain the force-distance trajectory of each point, from which the elasticity is inferred using the Hertz model. (B) Representative force-distance trajectory on the hydrogel. The hysteresis between the "extended" and "retracted" trajectories is due to sample deformation and adhesion between the tip of the cantilever and the hydrogel. The red line corresponds to the use of the Hertz model to fit the contact area in the "extended" trajectory. (C--F) The two-dimensional poplar on the surface of the hydrogel determined by AFM when the ratio of β-CD and PEG-Mal is 0:1 (C), 1:1 (D), 2:1 (E), and 4. Distribution of modulus: 1 (F). (Scale bar, 1 μm.) The inset corresponds to a typical Young's modulus histogram.

Before culturing hESCs on these hydrogels, we used 7721 and HK-2 cells to study the biocompatibility of the hydrogels. Based on the use of CellTiter-Glo reagent and live/dead cell staining (SI appendix, Figure S16), after 24 hours of culture on the hydrogel, the survival rates of the two cell lines were greater than 80%. Please note that the cells on the H gel showed similar viability to the cells on the I gel, indicating that the addition of β-CD in the preparation of the hydrogel does not affect the biocompatibility. Next, we investigated the effect of hydrogel homogeneity on the differentiation of hESC cultured on two hydrogels. The hydrogel was prepared in a 12-well cell culture plate, and the well (glass surface) without hydrogel was used as a control. We also added 0.2 mg ⋅ mL-1 cRGD (Cyclo[Arg-Gly-Asp-D-Phe-Cys]) to the hydrogel mixture before gelation to promote cell attachment. Swell the hydrogel to equilibrium, and then seed hESC in mTeST1 at a density of 1 × 103 cell colonies per well according to the manufacturer's instructions. Change the medium every day. After the cells were cultured for 10 days, the cell aggregates and stemness of hESCs were identified by immunostaining with molecular markers Oct3/4 and Sox2, which are two important transcription factors that maintain the self-renewal and pluripotency of hESCs (Figure 4A). In terms of cell aggregates, hESCs adhered well to the hydrogel and control glass substrates, but showed different morphologies. The hESCs on I-gels are still cell aggregates, and only a few cells are separated and scattered on the gel. In contrast, almost all hESCs grow alone and adopt an unfolded conformation on the H gel. The percentages of cells remaining as cell aggregates on the I gel, H gel, and glass substrate were 59.4, 5.0, and 81.7%, respectively (Figure 4B). In terms of cell dryness, compared with the cells on the I-gel or glass matrix, the Oct3/4 and Sox2 in the cells on the H-gel were significantly reduced, indicating that the homogeneous hydrogel may reduce the dryness and Promote the differentiation of hESC. Quantitative fluorescence analysis also supports this finding (Figure 4C). The fluorescence intensity of hESCs Oct3/4 and Sox2 on H-gels is only 11.6% and 6.7% of the intensity of DAPI, respectively, while the fluorescence intensity on I-gels is much higher (>40%). Compared with the glass substrate, it further confirms Enhanced differentiation of hESCs on H gel. The overall Young's modulus of H gel is slightly higher than that of I gel. Then we investigated whether this difference in Young's modulus affects the dryness of hESC. By changing the polymer concentration, we obtained H gel and I gel with a higher Young's modulus of about 115 kPa and a lower Young's modulus of about 60 kPa (SI appendix, Figures S9 and S10). Regardless of small changes in Young's modulus, molecular marker immunostaining gave consistent results (SI appendix, Figures S17 and S18). These results indicate that the homogeneity of the hydrogel can significantly affect stem cell differentiation and stem cell maintenance.

Dryness maintenance and differentiation of hESCs cultured on H-gels and I-gels. (A) After culturing on different surfaces for 10 days, hESCs identified by molecular marker immunostaining. The hESCs on the cell culture wells were used as the control group. Oct3/4 (red) and Sox2 (green), specific markers for dryness maintenance, are down-regulated in H gel. The nucleus is represented by DAPI (blue). (B) Percentage of cells in cell aggregates on different substrates. **: P <0.01; ***: P <0.001. (C) The normalized fluorescence intensity of Oct3/4 and Sox2, DAPI is used as 100% of I-gel, H-gel and control cells. NS: not statistically significant, P> 0.05; *: P <0.05; and ***: P <0.001. (D) After 5 and 10 days of culture on different substrates, the corresponding mRNA expression heat maps of various stem cell lines in hESC. (E) The mRNA expression of RUNX2, OCN, ALP and collagen I in hESCs after cultured on different substrates for 5 and 10 days.

We also conducted additional experiments to rule out other possible reasons for the different differentiation behavior of hESC on I gel and H gel. Previous studies have shown that surface roughness also affects stem cell differentiation. To rule out this possibility, we measured the roughness of I gel and H gel. They show the same roughness of approximately 30 nm (SI appendix, Figure S19). According to reports, the distribution of ligands also affects stem cell lineage norms. The distribution of cRGD on the hydrogel is difficult to determine directly. As an alternative, we use a thiol-containing fluorescent dye (Cy5-PEG-SH, 2 kDa) to simulate the ligand coupling process. Our results show that the fluorescent dyes are evenly distributed on the I gel and H gel (SI appendix, Figure S20A). The change in fluorescence intensity is less than 5% (SI appendix, Figure S20B). We expect that cRGD will be evenly distributed on I gel and H gel because it is coupled to the hydrogel through the same reaction. In addition, the stress relaxation behavior of all hydrogels at different ratios of β-CD and PEG-Mal is similar, excluding the influence of stress relaxation characteristics on hESC differentiation (SI Appendix, Figure S21). Since there may still be a residual amount of β-CD (<72 μg ⋅ mL-1) in the H gel, we subsequently investigated whether β-CD affects hESC differentiation. We added additional β-CD (up to 3 mg ⋅ mL-1) to the control group, and the cell morphology and stem cells were not affected (SI appendix, Figure S22). These studies confirmed that mechanical homogeneity is the main factor determining hESC differentiation.

We used qRT-PCR to further study the pedigree specification of hESC. As shown in Table S1 of the SI Appendix, 25 well-established markers (ESC: Oct-4, Sox-2, Nanog and Sall4) were used in the experiment; myogenic precursor cells: CD56 and CD146; VE-cadherin; Osteoblasts: ALPP, collagen I, RUNX2 and OCN; neural stem cells: CD133, nestin and CD15; mesenchymal stem cells (MSC): CD10, CD13 and CD73; skin stem cells: K15 and follistatin; adipose stem cells: CD44 and ICAM-1/CD54; intestinal stem cells: gremlin 1, Lrig 1 and Lgr 5). The relative gene expression levels are summarized in a two-dimensional graph (Figure 4D). Obviously, the expression level of osteoblast marker messenger ribonucleic acid (mRNA) of the cells seeded on the H gel was significantly higher than that of the cells on the I gel and the control glass substrate. The relative mRNA levels of RUNX2, OCN, ALPP and collagen I in cells cultured on the H gel for 5 or 10 days were significantly higher than those of the other two groups (Figure 4E). These results indicate that the homogeneity of the hydrogel can promote osteogenic differentiation.

In order to understand the underlying mechanism of the different cell fate commitments on the homogeneous gel and the I gel, we used molecular marker immunostaining to study the morphology, cytoskeleton structure and adhesion plaques of the cells on the I gel and H gel (Figure 5A) ). The average diffusion area of ​​hESC nuclei is significantly different between the two types of hydrogels (Figure 5A). The average core area on H-gel and I-gel are ~122 and ~38 μm2, respectively (SI appendix, Figure S23A). The compressed actin bundles of the cells on the H gel are longer and more organized than the cells on the I gel. As a quantitative indicator of cytoskeletal tension, the relative strength of actin was checked by normalizing the area of ​​actin to the nuclear area. The relative actin strength of H gel was 17.3, while that of I gel was only 2.1 (SI Appendix, Figure S23B). In addition, punctate paxillin was observed on the H gel, indicating that they aggregated on the cell surface to form mature adhesion spots. In contrast, less and more diffuse paxillin was observed on the I gel than on the H gel (Figure 5A). The relative paxillin intensity (paxillin area/nucleus area) is 3.5 on H gel and as low as 0.52 on I gel (SI appendix, Figure S23C). All these results indicate that hESCs grown on H gel show more efficient mechanical transduction than hESCs grown on I gel.

The mechanical uniformity of the hydrogel affects the mechanical transduction pathway of hESC. (A) Immunostaining of hESC nuclei (DAPI, blue), F-actin (phalloidin, green) and adhesion plaques (paxillin, red) on gel I and gel H. (Scale bar, 20 μm.) (B) Immunostaining of hESC nuclei (DAPI, blue) and YAP/TAZ (green) cultured on gel I and gel H. (Scale bar, 40 μm.) (C and D) The intensity distributions of DPAI and YAP/TAZ (the line area in B) are more relevant on the H gel (D) than on the I gel (C).

We further investigated whether the mechanically sensitive transcriptional co-activator YAP/ (a related protein)/TAZ (transcriptional co-activator with PDZ binding motif) is involved in the measurement of mechanical homogeneity of genetic events and ultimately affects hESC differentiation. YAP/TAZ was mainly observed in cell nuclei cultured on H gel (Figure 5B, bottom). In contrast, YAP/TAZ remained mostly inactive and located in the cytoplasm of the cells on the I gel (Figure 5B, top). The fluorescence intensity distribution of YAP/TAZ and the nuclei of the cross-sections of cells cultured on the two hydrogels further confirmed these findings (Figure 5C and D). The cells seeded on the H gel showed 88.5 ± 6.1% nuclear YAP/TAZ activation, while the cells seeded on the I gel were mostly inactivated in the cytoplasm, with only 16.5 ± 5.9% nuclear localization (SI Appendix, Figure S23D). All these results indicate that the homogeneity of the hydrogel promotes the activation of YAP/TAZ, thereby further regulating the differentiation of hESC.

The development of hydrogel technology has greatly expanded the scope of many biomedical fields, including cell culture, tissue engineering, flexible electronics, and even artificial skin. In all these applications, the mechanical properties of hydrogels play an important role. Although a lot of research has been devoted to optimizing stiffness, strain, toughness, and dynamic mechanical response, few researches have been done to improve the mechanical uniformity of hydrogels. In this work, we provide a method to reduce the rate of the Michael-type addition reaction between maleimide and thiol. We show that the resulting hydrogel is mechanically uniform on the nanometer scale. Since this method can reduce the number of defects in the hydrogel, it can also improve other mechanical properties such as stretchability and toughness. In addition, the concept of using kinetic traps to slow down gelation is not limited to maleimide-thiol conjugation, but can be easily extended to other reactions. We envision that this technology can be widely used in hydrogel technology.

Although synthetic hydrogels have been widely used in stem cell culture and tissue engineering, their mechanical inhomogeneity is often overlooked. Here, we show that heterogeneous mechanical properties can greatly affect the behavior of hESC and increase the complexity of its mechanical cues. Unexpectedly, we found that I-gels are good for maintaining dryness. Even in a relatively hard hydrogel (~80 kPa), hESCs remained as cell aggregates and did not differentiate within 10 days. This proposes a method to expand hESC in vitro using a hydrogel matrix with heterogeneous mechanical properties. On the other hand, we found that H-gels can promote osteogenic differentiation. Therefore, to use hydrogel scaffolds for cartilage and bone regeneration, it is very important to improve the uniformity of the hydrogel. In many basic studies of stem cells that respond to mechanical signals, the homogeneity of the hydrogel should also be considered, as it may complicate the interpretation of the results.

It has also been studied how surfaces with different mechanical and geometric patterns affect the differentiation of stem cells. Anseth and colleagues (52) used photodegradation reactions to create micron-scale patterns of soft (~2 kPa) and hard (~10 kPa) regions in PEG hydrogels. They found that evenly spaced patterns promote mechanical transduction and lead to enhanced osteogenesis, while random patterns lead to inefficient mechanical transduction. Similarly, in our research, we found that nanoscale changes in the mechanical properties of hydrogels can also significantly affect the focal adhesion structure and the formation of stretched actin bundles, which enables hESCs to retain stem cell characteristics. On the other hand, homogeneous hydrogels lead to more effective mechanical transduction and promote osteogenesis. Given that the size of mechanoreceptors (ie integrins) is less than 100 nm (53), it is reasonable to infer that nano-scale changes in mechanical characteristics are sufficient to interfere with mechanosensing pathways. Interestingly, in the study of Dalby and colleagues (54) using random pattern ligands, they found that the disordered pattern promoted cell adhesion and subsequent osteogenic differentiation, while the ordered pattern did not. In another work by Chien, Jin and colleagues (55), it was found that the size of the pattern is the key to inducing differentiation into osteoblast-like cells. Later, Ding and colleagues showed that both matrix stiffness and nanoscale characteristics affect stem cell differentiation (56). Obviously, the effect of different surface mechanical patterns is different from the effect of different surface nanoscale patterns. However, they can be understood under the same canvas; that is to say, ECM organization can affect the mechanical signal transduction of stem cells, regulate the cytoskeleton structure, and ultimately lead to differences in gene transcription.

In summary, in this work, we report a method for minimizing the mechanical heterogeneity in the synthesis of pollution-free maleimide-thiol cross-linked PEG hydrogels by using host-guest chemistry to reduce the gel speed. This method is universal and can improve many mechanical properties of the resulting hydrogel. In addition, we show that nano-scale changes in matrix stiffness can significantly affect the lineage specification of hESC, which has not been seriously considered in many current stem cell studies. Given that the mechanical mode of tissue in living tissue is greatly affected by injury and disease, our results may improve our understanding of disease progression and mechanical transduction during wound healing. In addition, engineered H gel is particularly useful for promoting osteogenic differentiation and should be widely used in bone tissue regeneration.

All titrations were carried out at 298 K using a Microcal ITC200 device. The four-arm PEG-Mal and β-CD were dissolved in phosphate buffered saline (PBS, 10 mM, pH = 7.4) at concentrations of 0.66 mM and 0.06 mM, respectively. Then, the four-arm PEG-Mal solution was added to the β-CD solution during the titration. Perform a blank titration in PBS (10 mM, pH = 7.4) and subtract the result from the corresponding titration to account for the effect of dilution. The fitting was performed using Origin software provided by Microcal.

The reaction of the four-arm PEG-SH and the four-arm PEG-Mal was monitored by UV spectroscopy. A V550 (JASCO Inc.) spectrophotometer was used to record the UV absorbance of a mixture of four-arm PEG-SH (400 μM) and four-arm PEG-Mal (400 μM) at 300 nm to monitor the four-arm gelation process. Arm PEG-Mal. For the reaction rate of PEG-SH/PEG-Mal/β-CD hydrogel, the UV absorbance of a mixture containing four-arm PEG-SH, four-arm PEG-Mal and β-CD, where the molar ratio is β-CD and four The arm PEG-Mal is 0:1, 1:1, 2:1, and 4:1, respectively, recorded at 300 nm. Then, the concentration of the four-arm PEG-Mal and maleimide-thiol adduct was calculated according to the calibration curve based on the ultraviolet absorbance. The cuvette width is 1 cm and the bandwidth is set to 0.2 nm.

In order to calculate the reaction rate of maleimide and thiol based on the binding constant measured by ITC, the reaction rate can be described by an equation. 1-3, where [SH] and [Mal] correspond to the concentration of maleimide and thiol, respectively. [Mal·SH] corresponds to the concentration of maleimide and thiol compounds. V=d[Mal·SH]dt=k1([SH]-[Mal·SH])([Mal]-[Mal·SH]), [1] ∫0[Mal·SH]d[Mal·SH] ([SH]−[Mal·SH])([Mal]−[Mal·SH])=∫0tk1dt, [2] 1[SH]− [Mal]ln[Mal]([SH]-[Mal·SH ])[SH]([Mal]-[Mal·SH])=k1t. [3]

The concentration of maleimide, β-CD and the complex maleimide and β-CD (Mal CD) in the mixture of maleimide and β-CD in the thermodynamic equilibrium is determined as [CD][ Mal]=[Mal·CD], [4]

Where ka is the association constant of maleimide and β-CD. The integral molar mass of maleimide is equal to the integral molar mass of thiol (5), and the integral molar mass of β-CD is equal to the integral molar mass of thiol. K2 corresponds to the molar ratio of β-CD to mercaptan: [Mal·CD] [Mal]=[SH] [5] [Mal·CD] [CD]=k2[SH] [6] kak2[Mal] [SH] [Mal](ka[Mal] 1)=[SH] [7] [Mal]=−((k2−1)ka[SH] 1) ((k2−1)ka[SH]] 1)2 4ka[ SH]2ka=Ψ([SH]). [8]

In summary, the concentration of [Mal·SH] in the system at time t can be described by the following formula: 1[SH]-Ψ([SH])lnΨ([SH])([SH]-[Mal·SH])[ SH](Ψ([SH])−[Mal·SH])=k1t, [9] [Mal·SH]=Ψ([SH])−Ψ([SH])([SH]− Ψ([SH ]))[SH]ek1t([SH]−Ψ([SH]))−Ψ([SH]). [10]

[Mal·SH] The relationship between generation and time under different ratios of β-CD and four-arm PEG-Mal (0:1, 1:1, 2:1, and 4:1) is calculated and displayed​​​ Figure 2B at the bottom.

For this purpose, both the four-arm PEG-Mal (20 K) and the four-arm PEG-SH (20 K) were dissolved in ddH2O to a concentration of 7 mM. Then, the two solutions were quickly mixed at a volume ratio of 1:1. The transparent PEG-SH/PEG-Mal hydrogel is formed within a few seconds after mixing. To prepare PEG-SH/PEG-Mal/β-CD hydrogels, β-CD was dissolved in a four-arm PEG-Mal solution (7 mM). Then, the four-arm PEG-Mal/CD solution and the four-arm PEG-SH solution were mixed at a volume ratio of 1:1 to obtain a PEG-SH/PEG-Mal/β-CD hydrogel. All the resulting hydrogels were dialyzed in ddH2O for 24 hours to remove β-CD and unreacted PEG.

The AFM experiment was performed on a commercial AFM (JPK, Nawizard II). Use commercial software from JPK to obtain and analyze force curves. The D-type MLCT (Germany Bruker, cone half-opening angle α <20°, tip radius: 20 nm) cantilever beam was selected for the experiment. The spring constant of the cantilever is 50 to 60 pN ⋅ nm-1, and the equipartition theorem is used in the solvent of each experiment to calibrate it before measurement. The maximum load force is set to 300 pN. All AFM experiments are performed at room temperature. In a typical experiment, the hydrogel is dispersed on the surface of the glass substrate in deionized water. Then, the cantilever is moved above the hydrogel with the help of the positioning system. The cantilever is brought to the sample at a constant speed of 2 μm ⋅ s-1 until a load force of 300 pN is reached. Then, the cantilever retracts and moves to another position in the next cycle. By fitting the approximation curve to the Hertz model (11), we obtained the Young's modulus value of the hydrogel. Use the commercial data processing software provided by JPK for indentation fitting. Usually, three to four such areas (5 μm × 5 μm, 1,600 pixels) are randomly selected on each sample to determine the average Young's modulus. F(h)=2πtan⁡αE1−v2h2, [11]

Where F is the stress of the cantilever, h is the depth at which the tip of the cantilever squeezes the hydrogel, α is the half-angle of the tip (15°), E is the Young’s modulus of the tissue, and v is the Poisson’s ratio. We have chosen this in the calculation v = 0.5.

hESCs are provided by the Stem Cell Bank of the Chinese Academy of Sciences and are labeled ESC H9. Amplify hESCs in mTeSR1 (STEMCELL Technologies) on Matrigel-coated (BD Biosciences) tissue culture plates, and replace mTeSR1 daily. All hMSCs were cultured for 2 days before the experiment.

For the cell culture of hESC on the hydrogel, 150 μL of the hydrogel was prepared on a round cover glass (20 mm, WHB) in a 12-well cell culture plate (Thermo). The cells seeded on the glass substrate were used as a control group. Then, transfer 2 mL of PBS (10 mM, pH = 7.4) into each well and allow the hydrogel to swell for 24 hours before removing the PBS. The swelling cycle is repeated four times. Then, hESCs were seeded on the hydrogel in mTeSR1 at a density of 1 × 103 cell colonies per well, and mTeSR1 was replaced daily. The qPCR analysis was performed after 5 and 10 days. Immunocytochemical staining was also performed after 10 days.

For immunofluorescence, cells were fixed in 2% (vol/vol) paraformaldehyde for 30 minutes, and then treated with 0.1% Triton X-100 for 15 minutes. After blocking in 5% (weight/volume) bovine serum albumin (BSA) for 1 hour to minimize non-specific binding, anti-Oct 3/4, anti-Sox2, and anti-YAP primary antibodies diluted in antibody dilution buffer Add to fixed cells and incubate overnight at 4°C. Then, pour out the primary antibody solution, and immediately wash the petri dish with PBS for 3 × 5 minutes. After rinsing with PBST (0.5wt% Tween-20 in PBS) 3 times, add the secondary antibody (goat anti-mouse Alexa Fluor 488; goat anti-mouse Alexa Fluor 594) in PBS, and add the cells to the culture dish. Incubate at room temperature for 60 minutes. Then, pour out the secondary antibody and wash the cells with PBST for 3 × 5 minutes. All images were obtained using a laser confocal fluorescence microscope (Olympus FV3000). For high (CPEG-Mal = CPEG-SH = 105 mg ⋅ mL-1)- and low PEG (CPEG-Mal = CPEG-SH = 52.5 mg ⋅ mL-1) images of hESC on hydrogels, use Obtained by OLYMPUS-IX81 (OLYMPUS) fluorescence microscope. Use ImageJ for quantitative image analysis.

Use TRIzol reagent to extract total RNA from cells, and then use 5× PrimeScript RT Master Mix to reverse transcribe 1,000 ng of total RNA into complementary ribonucleic acid (cDNA) according to the manufacturer's instructions. Real-time PCR was performed on the LightCycler 96 system (Roche) using SYBR Premix Ex TaqTM II kit (DRR081A, Takara Bio Inc.). The sense and antisense sequences of the primers used for quantitative RT-PCR are summarized in Table S1 in the SI Appendix. The reaction was carried out under the following conditions: incubation at 95°C for 2 minutes; denaturation at 94°C for 15s, annealing at 58°C for 15s, polymerization at 72°C for 60s, 40 cycles; and finally extension at 72°C for 5 minutes. All cycle threshold (Ct) values ​​are quantitatively analyzed and normalized to their respective GAPDH values. Use the 2-ΔΔCt method to calculate the relative level.

Accordingly, use Student's t test or one-way analysis of variance to determine statistical significance. The statistical significance was set as P value<0.05.

All research data is included in the article and/or SI appendix.

This work is mainly funded by the National Key Research and Development Program (2020YFA0908100), the National Natural Science Foundation of China (11934008, 11804148, 21902075), the Natural Science Foundation of Jiangsu Province (BK201803), and the special funds for basic scientific research operations of central universities (020414380187, 020414380148, 020414380138), Science and Technology Innovation Fund of Nanjing University (020414913413).

↵1B.X. and DT made the same contribution to this work.

Author contributions: BX, WW and YC design research; BX, DT, XW, ZX, JG, and YH conducted research; ZZ contributed new reagents/analysis tools; BX analysis data; BX, MQ, XZ, WW and YC Wrote this paper.

The author declares no competing interests.

This article is directly contributed by PNAS.

This article contains online support information at https://www.pnas.org/lookup/suppl/doi:10.1073/pnas.2110961118/-/DCSupplemental.

This open access article is distributed under the Creative Commons Attribution-Non-Commercial-No Derivative License 4.0 (CC BY-NC-ND).

Thank you for your interest in advertising on PNAS.

Note: We only ask you to provide your email address so that the people you recommend the page to know that you want them to see it, and that it is not spam. We do not capture any email addresses.

Feedback privacy/legal

Copyright © 2021 National Academy of Sciences. Online ISSN 1091-6490. PNAS is a partner of CHORUS, COPE, CrossRef, ORCID and Research4Life.