# Fileset

[Engineering a scaffold to control cell fates of stem cells for cartilage tissue engineering.pdf](https://mdr.nims.go.jp/filesets/3d2749f3-1e34-4991-b9da-a61bd8cd4eb3/download)

## Creator

[Bowu Peng](https://orcid.org/0009-0001-4808-2207), Huajian Chen, [Tianjiao Zeng](https://orcid.org/0000-0002-1286-0337), Mingxin Zhang, [Toru Yoshitomi](https://orcid.org/0000-0003-3847-1812), [Naoki Kawazoe](https://orcid.org/0000-0003-3916-0709), Yingnan Yang, [Guoping Chen](https://orcid.org/0000-0001-6753-3678)

## Rights

[Creative Commons BY-NC Attribution-NonCommercial 4.0 International](https://creativecommons.org/licenses/by-nc/4.0/)

## Other metadata

[Engineering a scaffold to control cell fates of stem cells for cartilage tissue engineering](https://mdr.nims.go.jp/datasets/8af82023-871e-4340-bf6b-8ea29e464525)

## Fulltext

9676 |  J. Mater. Chem. B, 2026, 14, 9676–9687 This journal is © The Royal Society of Chemistry 2026Cite this: J. Mater. Chem. B,2026, 14, 9676Engineering a scaffold to control cell fates ofstem cells for cartilage tissue engineeringBowu Peng, ab Huajian Chen,a Tianjiao Zeng,a Mingxin Zhang,abToru Yoshitomi, a Naoki Kawazoe,a Yingnan Yangc and Guoping Chen *abPrecise spatial and temporal control of the fates of mesenchymal stromal cells (MSCs) through dynamicmicroenvironmental cues is essential for MSC-based tissue engineering. Artificial niches that recapitulatethe spatiotemporal regulation of cell fates have been regarded as ideal platforms. Here, we engineeredan artificial niche scaffold by integrating growth and chondrogenic induction factors that exerted stagedeffects on MSCs to direct different phases of cartilage regeneration. The scaffold was prepared byimmobilizing heparin-binding epidermal growth factor-like growth factor and matrix metalloproteinase-2-cleavable peptide-tethered kartogenin-loaded liposomes within a gelatin/heparin/g-PGA scaffold. Theimmobilized growth factor initially promoted MSC proliferation. The upregulation of proliferation-associated signaling served as an autonomous trigger for the subsequent release of kartogenin, therebyinducing MSC chondrogenesis. The results indicated that this spatiotemporally programmed system wasable to regulate cellular fates in a spatiotemporal manner and promote cartilage regeneration.1. IntroductionCartilage injury and related diseases are considered majorcauses of physical disability worldwide, imposing a substantialsocioeconomic burden on modern society. As a connectivetissue with an avascular structure and no innervation, cartilagehas limited self-healing capacity because of its uniqueproperties.1 Traditional strategies for cartilage defect treatmentare limited and do not restore the intrinsic regenerative capa-city of cartilage.2,3 Therefore, developing new treatments thatenable effective intrinsic repair of cartilage diseases is essen-tial. In recent years, mesenchymal stromal cell (MSC)-basedtissue engineering has attracted considerable attention as apromising strategy for cartilage regeneration, owing to themultipotent properties of MSCs.3–8 This process generally pro-ceeds in a stepwise manner, encompassing sequential prolif-eration and differentiation. Hence, precise control of MSC fateis pivotal for therapeutic outcomes.9–11In the body, stem cells reside in instructive, tissue-specificmicroenvironments, or niches, that modulate stem cellfate.12,13 Niches are dynamic and complex, autonomously reg-ulating cellular behaviors during tissue repair.14,15 Wheninjuries occur, stem cells migrate to lesion sites, where theyrapidly proliferate and subsequently differentiate into specificphenotypes to facilitate tissue regeneration. Inspired by thefunctions of natural niches, artificial niches that recapitulateinstructive signals for tissue repair have attracted considerableattention in recent years.10,16–20 However, the precise design ofartificial niches to achieve their sequential functions duringtissue repair remains a major challenge, the crux of which liesin identifying a trigger or signal that induces the subsequentdifferentiation process following cell proliferation.11,21–24The extracellular matrix (ECM), as a predominant compo-nent of the niche and a reservoir of growth factors (GFs),plays important roles in providing anchoring sites, structuralsupport, and GFs for the regulation of cellular fates.14,25During the tissue repair process, ECM remodeling occursthroughout the entire process, degrading the matrix to createspace and release associated GFs that facilitate cell migra-tion, proliferation, and differentiation, thereby maintainingtissue homeostasis.26,27 Matrix metalloproteinases (MMPs)are endopeptidases that contribute to growth, development,wound healing, and pathogenesis through ECM remodeling.As major mediators of ECM remodeling, MMPs show promis-ing potential as targets for recapitulating the precise mani-pulation of cell fates in artificial niches for tissueregeneration.28,29 Among the various MMPs, MMP-2 has beenidentified as the most highly expressed during the normalproliferative behavior of stem cells.30,31 Therefore, it mayserve as a potential signaling cue to initiate the subsequentdifferentiation process and thereby promote tissue repair.a Research Center for Macromolecules and Biomaterials, National Institute forMaterials Science, Ibaraki, 305-0044, Japan. E-mail: Guoping.CHEN@nims.go.jpb Graduate School of Science and Technology, University of Tsukuba, Ibaraki,305-8577, Japanc Graduate School of Life and Environmental Science, University of Tsukuba,Ibaraki, 305-8572, JapanReceived 22nd April 2026,Accepted 12th July 2026DOI: 10.1039/d6tb00939ersc.li/materials-bJournal ofMaterials Chemistry BPAPERPublished on 13 July 2026Licensed under CC-BY-NC 4.0https://orcid.org/0009-0001-4808-2207https://orcid.org/0000-0003-3847-1812https://orcid.org/0000-0001-6753-3678http://crossmark.crossref.org/dialog/?doi=10.1039/d6tb00939e&domain=pdf&date_stamp=2026-07-27https://rsc.li/materials-bhttps://creativecommons.org/licenses/by-nc/4.0/This journal is © The Royal Society of Chemistry 2026 J. Mater. Chem. B, 2026, 14, 9676–9687 |  9677Kartogenin (KGN), a small lipophilic molecule and chon-drogenic induction factor, has been widely used in cartilageregeneration.32 Liposomes, as nanoparticle-based delivery sys-tems, have demonstrated a strong capacity for loading bothhydrophilic and lipophilic drugs. Generally, hydrophilic drugsare encapsulated within the inner aqueous core, whereaslipophilic compounds are incorporated into the lipid bilayer ofliposomes.33 However, the incorporation of lipophilic com-pounds into the bilayer can destabilize the liposomal structureand cause rapid drug release, whereas drugs encapsulatedwithin the aqueous core exhibit greater stability.34,35 Cyclodex-trins (CDs) are well-known for their ability to increase thesolubility, dissolution rate, and bioavailability of loadeddrugs.36,37 In particular, CDs can form inclusion complexeswith lipophilic compounds, thereby enhancing their solubilityand bioavailability. Based on these properties, drug-in-CD-in-liposome systems have been widely used in drug delivery toachieve the stable encapsulation of lipophilic drugs.36,37In this study, we designed and prepared an artificial nichescaffold that used endogenous MMP-2 as a trigger to release achondrogenic induction factor, thereby promoting subsequentdifferentiation following cell proliferation and enabling inves-tigation of the sequential proliferation and differentiation ofMSCs during cartilage regeneration (Fig. 1). In our previouswork,38 we designed a heparin-functionalized scaffold withimmobilized heparin-binding epidermal growth factor-likegrowth factor (HB-EGF) to mimic the regulatory functions ofthe ECM in controlling MSC fate. The results showed thatimmobilized HB-EGF enhanced MSC proliferation withoutinducing spontaneous differentiation. Subsequently, the sup-plement of differentiation factors is required to induce thechondrogenic differentiation of MSCs following proliferation.Based on this concept, KGN was first combined with b-cyclodextrin (b-CD) to form a KGN/b-CD inclusion complex,which was then loaded into liposomes. The KGN/b-CD-loadedliposomes were further modified with an MMP-2-responsivepeptide and subsequently incorporated into a gelatin/heparin/g-PGA (GHP) scaffold to construct an MMP-2-responsive GHP-liposome (GHP-lipo) scaffold. Finally, HB-EGF was immobilizedwithin the GHP-lipo scaffold to prepare a GHP-HB-EGF-liposome(GHP-HB-lipo) scaffold. This scaffold promoted MSC prolifera-tion through immobilized HB-EGF during the early stage andinduced chondrogenic differentiation through MMP-2-triggeredKGN release during the later stage, thereby facilitating cartilageregeneration. These scaffolds possessed multiple functions thatenabled the sequential control of MSC proliferation and chon-drogenic differentiation for cartilage regeneration.2. Results and discussion2.1 Preparation and characterization of the scaffoldsFirst, the GHP scaffold was fabricated using an ice particulatemethod.39,40 Ice particulates with diameters ranging from 150 to250 mm were used to control the pore sizes of the scaffolds. SEMresults revealed the porous structure of the GHP scaffold, whichcontained large spherical pores throughout the scaffold and smallpores on the walls of the larger pores, indicating that the GHPscaffold possessed a well-interconnected structure (Fig. 2a and b).Heparin staining showed a homogeneous distribution of heparinthroughout the GHP scaffold (Fig. 3d). Young’s modulus of the dryand hydrated scaffolds was measured using a compression test.The dry GHP scaffold exhibited a high Young’s modulus (Fig. S5).After hydration, Young’s modulus decreased significantly.Meanwhile, the KGN/b-CD inclusion complex was preparedusing a lyophilization method.41,42 FT-IR spectra showedabsorption peaks at 1537–1596 cm�1, which were assigned tothe CQC bonds of the aromatic ring of KGN, and absorptionpeaks at 1500–1600 cm�1, which were assigned to the KGN/b-CD inclusion complex (Fig. S1). These results demonstrated theFig. 1 Schematic illustration of the preparation of the spatiotemporally programmed GHP-HB-lipo scaffold and sequential proliferation andchondrogenic differentiation of MSCs within the scaffold for cartilage regeneration.Paper Journal of Materials Chemistry B9678 |  J. Mater. Chem. B, 2026, 14, 9676–9687 This journal is © The Royal Society of Chemistry 2026Fig. 2 Microstructures of the scaffolds: scanning electron microscopy (SEM) images of the (a) and (b) GHP scaffold, (c) and (d) GHP-lipo scaffold, and(e) and (f) GHP-HB-lipo scaffold. The images below are magnified views of the specific areas (dotted squares) shown in the images above.Fig. 3 Characterization of liposomes and GHP-HB-lipo scaffolds: (a) optical images of aqueous solutions of KGN and KGN/b-CD inclusion complex;hydrodynamic size distribution of (b) blank liposomes and (c) KGN/b-CD liposomes; (d) staining of heparin in GHP scaffold; (e) EDS mapping ofphosphorus (P) in GHP-lipo scaffold; (f) immunochemical staining of HB-EGF in GHP-HB-lipo scaffold; (g) release profile of KGN from the GHP-HB-liposcaffold incubated in PBS without MMP-2, where the release profile during the initial 72 h is shown in the right graph (dotted rectangle) displayed at a highmagnification; and (h) release profile of KGN from GHP-HB-lipo scaffold after 2 h and 12 h of incubation in the absence or presence of MMP-2. The GHP-HB-lipo scaffold used in this experiment was pre-incubated in PBS without MMP-2 for 24 h. Data are presented as mean � SD (n = 3). Significantdifferences: **p o 0.01; ***p o 0.001.Journal of Materials Chemistry B PaperThis journal is © The Royal Society of Chemistry 2026 J. Mater. Chem. B, 2026, 14, 9676–9687 |  9679formation of the KGN/b-CD inclusion complex. Moreover, theUV absorption spectrum of the KGN/b-CD inclusion complexshowed a strong absorption peak at approximately 276 nm,exhibiting a slight blue shift compared to that of KGN atapproximately 278 nm, suggesting the successful encapsulationof KGN within the cavity of b-CD (Fig. S2). Optical images ofaqueous solutions of KGN and the KGN/b-CD inclusioncomplex showed that the KGN/b-CD inclusion complex mark-edly increased the solubility of KGN (Fig. 3a). The KGN encap-sulation efficiency in the KGN/b-CD inclusion complex was88.4 � 2.6%.Next, the KGN/b-CD inclusion complex was encapsulatedinto liposomes. The encapsulation efficiency of the KGN/b-CDinclusion complex was evaluated by disrupting the KGN/b-CDliposomes and measuring the UV absorption at approximately276 nm (Fig. S2). The encapsulation efficiency of the inclusioncomplex within the liposomes was 52.1 � 3.2%. The hydro-dynamic size of the KGN/b-CD liposomes was evaluated bydynamic light scattering (DLS). The KGN/b-CD liposomesand blank liposomes had mean sizes of 242.9 � 120.4 nmand 231.4 � 97.6 nm, respectively (Fig. 3b and c), with nosignificant difference between the two groups. The KGN/b-CDliposomes were further modified with an MMP-2-cleavablepeptide, and the resulting MMP-2-responsive KGN/b-CD lipo-somes were incorporated into the GHP scaffold through clickchemistry.43SEM and energy-dispersive spectroscopy (EDS) mappingresults confirmed the successful incorporation of MMP-2-responsive KGN/b-CD liposomes into the scaffold (Fig. 3e).The incorporation efficiency of the MMP-2-responsive KGN/b-CD liposomes was 45.1 � 2.2%. Subsequently, HB-EGF wasimmobilized within the GHP-lipo scaffold to construct theGHP-HB-lipo scaffold. Immunochemical staining for HB-EGFdemonstrated the successful immobilization of HB-EGF withinthe GHP-lipo scaffold (Fig. 3f and Fig. S3), with an immobiliza-tion efficiency of 85.2 � 2.0%. SEM observations showed thatthe GHP-lipo and GHP-HB-lipo scaffolds possessed the sameinterconnected structure as the GHP scaffold (Fig. 2). Theincorporation of MMP-2-responsive KGN/b-CD liposomes andthe immobilization of HB-EGF did not affect the porous struc-ture of the scaffolds (Fig. 2c–f).The release profile of KGN from the GHP-HB-lipo scaffold inthe absence and presence of MMP-2 was investigated (Fig. 3gand h). In PBS without MMP-2, the GHP-HB-lipo scaffoldexhibited a slight initial burst release, followed by a very slowrelease of KGN during the 21-day incubation period (Fig. 3g).Only approximately 12% of KGN was released from the scaffoldafter 21 days of incubation, demonstrating the long-termstability of the KGN/b-CD inclusion complex loaded withinthe GHP-HB-lipo scaffold. After incubation in PBS withoutMMP-2 for 24 h, the GHP-HB-lipo scaffold was transferred toPBS containing MMP-2 to evaluate the MMP-2-responsiverelease of KGN (Fig. 3h). When incubation was continued inPBS without MMP-2, only a small amount of KGN was released.In contrast, in the presence of MMP-2, 3.1 � 0.9% and21.4 � 2.8% of KGN were released from the GHP-HB-liposcaffold after 2 h and 12 h of incubation, respectively. Theseresults demonstrated the MMP-2-responsive release of KGNfrom the GHP-HB-lipo scaffold, which was attributed to theincorporation of an MMP-2-cleavable peptide serving as a linkerbetween the KGN/b-CD liposomes and the scaffold.2.2 Cell adhesion, distribution, and viability in the scaffoldsSubcultured MSCs were harvested and seeded into the scaffoldsfor three-dimensional (3D) culture. Cell adhesion, distribution,and viability were evaluated. As shown in Fig. 4a, the seededMSCs adhered to the pores of the GHP-HB-lipo scaffold. MSCsin the GHP-HB-lipo scaffold exhibited adhesion behavior simi-lar to that observed in the GHP scaffold (Fig. S4a). Nuclearstaining of the vertical cross-section of the scaffold showed thatMSCs were homogeneously distributed throughout the entirestructure of both the GHP-HB-lipo and GHP scaffolds (Fig. 4band Fig. S4b). Live/dead staining indicated that all cellsremained viable, and no dead cells were observed in eitherthe GHP-HB-lipo or GHP scaffold after 1 day of culture (Fig. 4cand Fig. S4c). These results indicated that both the GHP-HB-lipo and GHP scaffolds supported cell adhesion, facilitatedhomogeneous cell distribution, and maintained high cell via-bility. The incorporation of liposomes and immobilized HB-EGF did not affect cell adhesion, distribution, or viabilitywithin the scaffolds.2.3 Sequential proliferation and chondrogenic differentiationof MSCs during in vitro culture in the GHP-HB-lipo scaffoldTo investigate the effects of the GHP-HB-lipo scaffold on MSCfate regulation, the proliferation and chondrogenic differentia-tion of MSCs cultured in the GHP-HB-lipo scaffold were eval-uated. MSCs were seeded into the GHP-HB-lipo scaffold andcultured in DMEM without the chondrogenic induction factor,KGN, for 28 days (Fig. 5a). DNA quantification showed thatMSCs in the GHP-HB-lipo scaffold proliferated significantlyafter 7 days of culture (Fig. 5b). However, the DNA contentdecreased after 14 days of culture and then gradually increasedagain after 21 and 28 days of culture. Quantification of sGAGshowed that the sGAG/DNA ratio did not change significantlyafter 7 days of culture but increased significantly after 14 daysof culture (Fig. 5c). The sGAG/DNA ratio remained at a similarlevel after further culture for 21 and 28 days. Gene expressionanalysis showed that the expression levels of type I collagen(COL I), type II collagen (COL II), SRY-box transcription factor 9(SOX9), and aggrecan (ACAN) did not change significantly after7 days of culture. However, the expression levels of these genesincreased significantly after 14 days of culture (Fig. 5d–g). Theexpression levels of COL II, SOX9, and ACAN continued toincrease after 21 and 28 days of culture. In contrast, theexpression level of COL I decreased after 21 and 28 days ofculture. The elevated sGAG/DNA ratio and increased expressionlevels of COL II, SOX9, and ACAN demonstrated the chondro-genic differentiation of MSCs. These results indicated thatMSCs cultured in the GHP-HB-lipo scaffold proliferated duringthe initial 7 days of culture and subsequently underwentPaper Journal of Materials Chemistry B9680 |  J. Mater. Chem. B, 2026, 14, 9676–9687 This journal is © The Royal Society of Chemistry 2026chondrogenic differentiation during the following 14, 21, and28 days of culture.To further demonstrate the effects of the GHP-HB-liposcaffold on MSC proliferation and chondrogenic differentia-tion, the GHP-HB scaffold was used as a comparison group andseeded with MSCs in the same manner as the GHP-HB-liposcaffold described above (Fig. 5a). MSCs seeded in the GHP-HBscaffold were cultured in KGN-free DMEM for 14 days andsubsequently cultured in KGN-containing DMEM for an addi-tional 14 days. MSCs proliferated significantly during culturing inthe KGN-free medium, whereas the DNA content decreasedduring culturing in the KGN-containing medium (Fig. 5b). Theaddition of KGN to the culture medium significantly increasedthe sGAG/DNA ratio and the expression levels of COL I, COL II,SOX9, and ACAN (Fig. 5c–f). The immobilized HB-EGF promotedMSC proliferation, whereas the supplemented KGN in the culturemedium induced chondrogenic differentiation. Compared withthe GHP-HB scaffold, the GHP-HB-lipo scaffold could releaseKGN to induce chondrogenic differentiation without requiringadditional supplementation of KGN in the culture medium.Furthermore, immunobiological staining of chondrogen-esis-related proteins (ACAN and COL II) was performed afterMSCs were cultured in the GHP-HB-lipo scaffold for 7, 14, 21,and 28 days (Fig. 6b). After 7 days of culture, ACAN and COL IIstaining was negative. However, after 14 days of culture, weakpositive staining for ACAN and COL II was observed. Thestaining intensity increased markedly after 21 and 28 days ofculture. ACAN and COL II were not positively stained afterMSCs were cultured in the GHP-HB scaffold in the KGN-freemedium for 14 days. In contrast, when the cells were subse-quently cultured in the KGN-containing medium for anadditional 14 days, weak positive staining for ACAN andCOL II was observed (Fig. 6a).These results indicated that the GHP-HB-lipo scaffold exhib-ited multiple functions that promoted MSC proliferation at theearly stage and induced chondrogenic differentiation at a laterstage. The proliferation of MSCs during the initial 7 days ofculture was likely promoted by the immobilized HB-EGF. Thesubsequent chondrogenic differentiation of the proliferatedMSCs was likely induced by the MMP-2-triggered release ofKGN. Previous studies have reported that MMP-2 expression isupregulated during MSC proliferation.31,44,45 The upregulatedMMP-2 could serve as an endogenous trigger to induce thecleavage of the MMP-2-responsive peptide, thereby releasingKGN-encapsulated liposomes from the GHP-HB-lipo scaffold(Fig. 3h). The released KGN could subsequently induce thechondrogenic differentiation of MSCs.2.4. In vivo evaluation of the GHP-HB-lipo scaffoldTo evaluate the in vivo efficacy of the GHP-HB-lipo scaffold,subcutaneous implantation was performed. The GHP-HB scaf-fold was also implanted as a comparison group. MSCs wereseeded into the GHP-HB and GHP-HB-lipo scaffolds and cul-tured in DMEM without a chondrogenic induction factor for12 h. The scaffolds were then subcutaneously implanted intothe dorsal region of nude mice for 28 days (Fig. 7a). Theimplants after 28 days of implantation displayed the sameround shapes as those before implantation (Fig. S6), indicatingthat the structure of the scaffolds was maintained during the28 days of implantation.DNA quantification showed that the DNA content in boththe GHP-HB and GHP-HB-lipo scaffolds was significantlyFig. 4 Evaluation of cell adhesion, distribution, and viability after MSCs were seeded and cultured in GHP-HB-lipo scaffold for 1 day: (a) SEM image:arrows indicate MSCs; (b) nuclear staining of MSCs in the vertical cross-section of the scaffold; (c) live/dead staining of the horizontal cross-section of thescaffold. Green fluorescence indicates living cells, whereas red fluorescence indicates dead cells.Journal of Materials Chemistry B PaperThis journal is © The Royal Society of Chemistry 2026 J. Mater. Chem. B, 2026, 14, 9676–9687 |  9681Fig. 6 Immunobiological staining of ACAN and COL II after MSCs were cultured in (a) GHP-HB scaffold after 14 day culture in absence of KGN, followedby another 14 day culture in the presence of KGN, and (b) GHP-HB-lipo scaffold for 7, 14, 21, and 28 days. Brown staining indicates positive expression ofACAN and COL II.Fig. 5 Sequential proliferation and chondrogenic differentiation of MSCs in GHP-HB and GHP-HB-lipo scaffolds: (a) schematic illustration of theprocedures used during 28 days of in vitro culture; (b) quantification of DNA content; (c) quantification of sGAG content; (d)–(g) gene expression levels ofCOL I, COL II, SOX9, and ACAN. The same number of MSCs used for cell seeding were used for the control. Data are presented as mean � SD (n = 3).Significant differences: *p o 0.05; **p o 0.01; ***p o 0.001; ns, not significant.Paper Journal of Materials Chemistry B9682 |  J. Mater. Chem. B, 2026, 14, 9676–9687 This journal is © The Royal Society of Chemistry 2026higher than that of the MSCs used for seeding, indicating cellproliferation within the scaffolds (Fig. 7b). The DNA content inthe GHP-HB-lipo scaffold was significantly lower than that inthe GHP-HB scaffold, indicating that cell proliferation in theGHP-HB scaffold was greater than that in the GHP-HB-liposcaffold.Quantification of the sGAG content showed that the sGAG/DNA ratio in the GHP-HB scaffold was not significantlydifferent from that of the MSCs used for seeding (Fig. 7c).However, the sGAG/DNA ratio in the GHP-HB-lipo scaffoldwas significantly higher than that in the GHP-HB scaffold.Gene expression analysis showed that the expression levelsof COL I, COL II, SOX9, and ACAN in the GHP-HB-lipo scaffoldwere significantly higher than those in the GHP-HB scaffold(Fig. 7d). HE staining showed homogeneous cell and ECMdistribution in both GHP-HB and GHP-HB-lipo scaffolds(Fig. 7e). Immunohistochemical staining indicated thatACAN and COL II were more strongly stained in the GHP-HB-lipo scaffold than in the GHP-HB scaffold. These resultsdemonstrated that the GHP-HB-lipo scaffold was more bene-ficial for cartilage regeneration than the GHP-HB scaffold.This outcome was likely because the GHP-HB scaffold couldonly promote MSC proliferation, whereas the GHP-HB-liposcaffold could not only enhance MSC proliferation butalso promote the chondrogenic differentiation of theproliferated MSCs.3. DiscussionPrecise control of MSC proliferation and chondrogenic differ-entiation is critical for MSC-based cartilage tissue engineering.Here, we developed an artificial niche with spatiotemporalcontrol over MSC fates by incorporating HB-EGF and KGN intoa GHP scaffold. HB-EGF was immobilized through its inter-action with heparin incorporated within the GHP scaffold toenhance cell proliferation, whereas KGN/b-CD liposomes weretethered through an MMP-2-cleavable peptide to induce chon-drogenic differentiation. MSCs cultured in the GHP-HB-liposcaffold exhibited initial proliferation, followed by chondro-genic differentiation. Other MMPs, such as MMP-9 and MMP-13, have also been used to achieve MMP-responsive function-ality in hydrogels and scaffolds to control the release oftherapeutic factors.46,47 For example, MMP-9 has been usedto control the release of anti-inflammatory and anti-rheumaticdrugs to treat rheumatoid arthritis (RA) because of the elevatedsecretion of MMP-9 in the RA microenvironment.46 Likewise,MMP-13 has been used to trigger the release of multipleFig. 7 In vivo evaluation of GHP-HB and GHP-HB-lipo scaffolds for cartilage regeneration after implantation for 28 days: (a) schematic illustration of thein vivo implantation procedure; (b) quantification of DNA content; (c) quantification of sGAG content; (d) gene expression levels of COL I, COL II, SOX9,and ACAN; and (e) HE staining and immunobiological staining of ACAN, COL II, and COL I. The same number of MSCs used for cell seeding were used forthe control. Data are presented as mean � SD (n = 4). Significant differences: *p o 0.05; **p o 0.01; ***p o 0.001; ns, not significant.Journal of Materials Chemistry B PaperThis journal is © The Royal Society of Chemistry 2026 J. Mater. Chem. B, 2026, 14, 9676–9687 |  9683therapeutic factors in the arthritic microenvironment.47 In thisstudy, we used MMP-2 to trigger the release of KGN after MSCproliferation because MMP-2 has been reported to be highlyexpressed in proliferating stem cells and its expressionincreases with cell number.30,31 The KGN/b-CD liposomes wereintroduced in the scaffold through tethering with an MMP-2cleavable peptide by click chemistry.When MSCs were cultured in the GHP-HB-lipo scaffold, theimmobilized HB-EGF enhanced MSC proliferation during theinitial stage without inducing chondrogenesis, which was con-sistent with our previous findings.38 Following cell prolifera-tion, MMP-2 released by the proliferated MSCs could cleave theMMP-2-cleavable peptide, thereby releasing KGN. MMP-2secreted by MSCs is freely diffusible and can distribute through-out the culture medium. Therefore, the secreted MMP-2 shouldbe homogenously distributed throughout the scaffold to inducechondrogenic differentiation of the proliferated MSCs. Cellculture results demonstrated the sequential proliferation andchondrogenic differentiation of MSCs during 3D culture in theGHP-HB-lipo scaffold. Animal experiments also demonstratedthe sequential proliferation and chondrogenic differentiation ofMSCs within the GHP-HB-lipo scaffold. These results suggestedthat the GHP-HB-lipo scaffold could serve as an artificial niche todirect MSC fate and holds potential for cartilage regeneration,particularly when only a limited number of donor-derived MSCsare available.Subcutaneous ectopic implantation was used in this study.Although this model has been widely used to examine thepotential of scaffolds for tissue engineering applications, itdoes not fully reflect the native cartilage microenvironment.Therefore, orthotopic implantation should be considered infuture studies to further verify the effectiveness of the scaffoldfor cartilage tissue engineering.4. ConclusionIn this study, we fabricated a multifunctional GHP-HB-liposcaffold that served as an artificial niche to recapitulate thespatiotemporal control of MSC fate by immobilizing HB-EGF andMMP-2-cleavable peptide-tethered KGN-loaded liposomes withina GHP scaffold. The scaffold supported cell adhesion andpromoted MSC proliferation. The KGN-loaded liposomes couldbe released in the presence of MMP-2, thereby inducing thechondrogenic differentiation of MSCs. In vitro cell culture andin vivo animal experiments demonstrated that the scaffold couldinitially promote MSC proliferation and subsequently induce thechondrogenic differentiation of the proliferated MSCs. The scaf-fold could serve as an artificial niche to direct MSC fate andshowed promising potential for cartilage regeneration.5. Experimental section5.1 Preparation of the gelatin/heparin/c-PGA (GHP) scaffoldThe GHP porous scaffold was prepared using an ice particulate-based method according to our previous work, with minormodifications.38 Briefly, ice particulates were obtained byspraying pure water into liquid nitrogen, and ice particulateswith diameters of 150–250 mm were collected by sieving in alow-temperature chamber. Gelatin, heparin sodium, and g-PGAwere then dissolved separately in a 35% (v/v) acetic acidsolution, and the resulting solutions were mixed to form ahomogeneous solution containing 8% (m/v) gelatin, 1% (m/v)heparin sodium, and 0.2% (m/v) g-PGA. Subsequently, thetemperature of the mixed solution and the ice particulateswas equilibrated at �4 1C for 6 h. The ice particulates werethen added to the mixed solution at an ice-to-solution ratio of7 : 3 (m/v), and the final mixture was transferred into a siliconemold. The construct was placed in a �80 1C freezer for 12 h.Finally, the frozen construct was lyophilized and crosslinkedusing a series of ethanol solutions containing EDC and NHS.After crosslinking, the scaffold was washed with Milli-Q waterto obtain the GHP scaffold.5.2 Microstructure characterization and heparin stainingThe microstructure of the scaffold was evaluated by scanningelectron microscopy (SEM; JSM-IT800, Tokyo, Japan). The dis-tribution of heparin within the scaffold was examined bystaining according to our previous work.38 Briefly, the GHPscaffold was stained with Alcian Blue 8GX (0.5%, w/v) in 3%acetic acid for 30 min and washed three times with Milli-Qwater. The stained samples were then lyophilized and exam-ined using an optical microscope.5.3 Characterization of the mechanical properties of thescaffoldsThe mechanical properties of the scaffolds were evaluated bymeasuring their Young’s modulus. The scaffolds werepunched into cylindrical specimens (j 6 � 6 mm). Young’smodulus was measured under both dry and hydrated conditions.The hydrated scaffolds were prepared by immersing the scaffoldsin PBS for 2 h at room temperature. The samples were com-pressed at a rate of 0.1 mm s�1 using a texture analyzer (TA-XTPlus; Texture Technologies, Hamilton, MA, USA). The initiallinear region of the stress–strain curves was used to calculateYoung’s modulus, and three samples from each group wereevaluated.5.4 Preparation of the KGN/b-CD inclusion complexThe KGN/b-CD inclusion complex was prepared according toprevious studies.41,42 Briefly, 1 mM b-CD solution and 1 mMKGN solution were mixed in water and stirred at 500 rpm for24 h in the dark at 4 1C. The solution was then filtered througha 0.45 mm filter, and the transparent filtrate was collected andlyophilized to obtain the KGN/b-CD inclusion complex. Thelyophilized powder was stored in the dark at 4 1C until furtheruse. The filter residue was collected and dissolved in DMSO,and the absorbance at 278 nm was measured by UV-vis spectro-metry to calculate the encapsulation efficiency.Paper Journal of Materials Chemistry B9684 |  J. Mater. Chem. B, 2026, 14, 9676–9687 This journal is © The Royal Society of Chemistry 20265.5 Preparation of KGN/b-CD liposomes and modificationwith the MMP-2-responsive peptide1,2-Dipalmitoyl-sn-glycero-3-phosphocholine (DPPC), 1,2-distearoyl-sn-glycero-3-phosphoethanolamine-N-[amino-(poly-ethylene glycol)-2000] (ammonium salt) (DSPE–NH2), and cho-lesterol (Chol) were used to prepare KGN/b-CD-encapsulatedDPPC/Chol/DSPE liposomes (KGN/b-CD-lipo) using a thin-filmhydration method. Briefly, 600 mL of DPPC (5 mg mL�1),320 mL of DSPE–NH2 (5 mg mL�1), and 80 mL of cholesterol(5 mg mL�1) (molar ratio of DPPC : DSPE–NH2 : Chol = 7 : 1 : 2)were mixed in a chloroform/methanol solvent mixture, and thesolvent was removed by rotary evaporation under reducedpressure to form a thin lipid film. Subsequently, 1 mL ofKGN/b-CD solution in PBS (1 mg mL�1) was added to the flaskfor hydration, and the flask was sonicated at 45 1C for 30 min.After sonication, the suspension was extruded at 45 1C througha polycarbonate membrane with a pore size of 200 nm. Theextruded suspension was centrifuged at 20 000 rpm for 15 min.The resulting sediment was washed, resuspended in 1 mL ofPBS, and stored at 4 1C until further use. Dynamic lightscattering (DLS; Beckman Coulter) was used to characterizethe hydrodynamic size of the liposomes before and after KGN/b-CD loading. The amount of KGN/b-CD encapsulated withinthe liposomes was determined by disrupting the liposomeswith a chloroform/methanol solvent mixture and measuringthe absorbance at 276 nm using UV-vis spectrometry. Theencapsulation efficiency was calculated by dividing the encap-sulated amount of KGN/b-CD by the initial loading amount.MMP-2-cleavable peptide-modified KGN/b-CD liposomeswere prepared by conjugating the MMP-2-cleavable peptide tothe KGN/b-CD liposomes. The peptide sequence was N3-GGPVGLIGGG-CO-NHS. Briefly, 15.6 mg of the peptide wasadded to 1 mL of KGN/b-CD liposome suspension (KGN,10 mM) and allowed to react at 4 1C for 6 h. The reactionmixture was then centrifuged at 20 000 rpm for 15 min toobtain the MMP-2-cleavable peptide-modified KGN/b-CD lipo-somes. The prepared liposomes were further washed, resus-pended in PBS, and stored at 4 1C for further use.5.6 Preparation of the GHP-HB-lipo scaffoldThe GHP scaffold was punched into cylindrical discs (j 6 �2 mm), and the discs were immersed in an EDC/NHS solutionat room temperature overnight to activate the carboxyl groups.Subsequently, 57 mL of N-(1R,8S,9S)-bicyclo[6.1.0]non-4-yn-9-ylmethyloxycarbonyl-1,8-diamino-3,6-dioxaoctane (BCN–NH2;10.23 mg mL�1) solution was added to the discs and allowedto react at room temperature for 12 h. Unbound BCN–NH2 wasremoved by washing with PBS. Subsequently, 57 mL of MMP-2-responsive KGN/b-CD-lipo solution (KGN, 0.1 mg) was added tothe discs and allowed to react at 4 1C for 12 h with shaking. Theresulting GHP-lipo scaffold was washed with PBS, and theincorporation efficiency of the MMP-2-responsive KGN/b-CD-lipo was evaluated by immersing the GHP-lipo scaffolds in achloroform/methanol solvent mixture and measuring theabsorbance at 276 nm using UV-vis spectrometry. Theincorporation efficiency was calculated by dividing the incor-porated amount of MMP-2-responsive KGN/b-CD-lipo by theinitial loading amount. Subsequently, the scaffold wasimmersed in 1 mL of HB-EGF solution (100 ng mL�1) undershaking at 4 1C for 24 h to prepare the GHP-HB-lipo scaffold.The prepared GHP-HB-lipo scaffold was washed with PBS andstored at 4 1C for further use. The residual HB-EGF solution andPBS eluates were collected to evaluate the loading efficiency.The GHP-HB scaffold was prepared by directly immersing theGHP scaffold in 1 mL of HB-EGF solution (90 ng mL�1) usingthe same procedure described above.5.7 Immunochemical staining of HB-EGFThe GHP-HB-lipo scaffold was incubated overnight in PBScontaining 2% bovine serum albumin at 4 1C to block non-specific antigen binding. Subsequently, the scaffold was incu-bated overnight at 4 1C with an anti-HB-EGF primary antibody(human HB-EGF antibody, AF-259-NA, Bio-Techne R&D Sys-tems). After washing three times with PBS, the sample wasincubated with Donkey Anti-Goat IgG H&L (Abcam) at roomtemperature for 2 h, followed by three washes with PBS. Colordevelopment was performed using 3,30-diaminobenzidine(DAB) for 5 min. The stained sample was washed three timeswith Milli-Q water and examined by optical microscopy.5.8 Immobilization efficiency of HB-EGFThe HB-EGF immobilization efficiency in the GHP-HB-liposcaffold was measured using an enzyme-linked immunosor-bent assay (ELISA) kit (human HB-EGF quantikine ELISA kit,Bio-Techne R&D Systems). The immobilized amount of HB-EGFin the scaffold was calculated by subtracting the residualamount of HB-EGF from the initial loading amount. Theimmobilization efficiency was calculated by dividing the immo-bilized amount of HB-EGF in the scaffold by the initial loadingamount. Three independent samples were analyzed to calculatethe mean and standard deviation.5.9 KGN release from GHP-HB-lipo scaffoldsThe GHP-HB-lipo scaffolds were immersed in 30 mL PBS at37 1C with shaking. At predetermined time points, 1 mL of PBSwas collected, and 1 mL of fresh PBS was added to replace thecollected volume. The amount of KGN in each collected samplewas measured by UV-vis spectrometry at 278 nm. The cumula-tive release of KGN was calculated by dividing the cumulativeamount of released KGN by the initial amount of KGN in thescaffolds.5.10 MMP-2-responsive release of KGN from the GHP-HB-liposcaffoldThe MMP-2-responsive release behavior of KGN/b-CD liposomeswas characterized by treating the GHP-HB-lipo scaffold withMMP-2. Previous studies have reported that the upregulationof MMP-2 expression is associated with MSC proliferation.31,44,45The concentration of MMP-2 in the culture medium supernatanthas been reported to be approximately 337.9 ng per 106 cells.48 Innormal human tissues, the proportion of active MMP-2 relativeJournal of Materials Chemistry B PaperThis journal is © The Royal Society of Chemistry 2026 J. Mater. Chem. B, 2026, 14, 9676–9687 |  9685to total MMP-2 is approximately 20%.49 Based on these reportsand the DNA quantification results, a concentration of100 ng mL�1 was selected for the MMP-2-responsive releaseassay. The GHP-HB-lipo scaffold was equilibrated in 1 mL ofPBS for 24 h and then transferred to a 24-well plate containing1 mL of MMP-2 solution (100 ng mL�1) and incubated withshaking at 37 1C. After 2 h and 12 h of incubation, the GHP-HB-lipo scaffold was removed, and methanol was added to each well.The absorbance of the resulting solution at 276 nm was thenmeasured using UV-vis spectrometry. The release efficiency wascalculated by dividing the released amount of KGN by the initialamount of KGN in the scaffold.5.11 Culture of MSCs in the GHP-HB-lipo scaffoldHuman bone marrow-derived MSCs at passage 2, obtainedfrom Lonza (Walkersville, MD, USA), were subcultured inMSCGMt medium. MSCs after two additional passages werecollected by trypsinization and resuspended in culture mediumat a concentration of 5 � 106 cells mL �1. 57 mL of cellsuspension was added to the upper surface of each GHP-HB-lipo scaffold disc and cultured for 6 h. After 6 h of culture, thescaffold discs were inverted, and the opposite side was seededwith an additional 57 mL of cell suspension. After a further 6 h,the samples were transferred to flasks containing DMEMsupplemented with 2.5% FBS, 4 mM L-glutamine, and100 U mL�1 penicillin–streptomycin and cultured under shak-ing at 60 rpm. The culture was maintained for 28 days, and themedium was changed every 3 days.5.12 Culture of MSCs in the GHP-HB scaffoldMSCs were seeded into GHP-HB scaffold discs using the sameprocedure as that described for the GHP-HB-lipo scaffold discs.MSCs in the GHP-HB scaffold were cultured in DMEM supple-mented with 2.5% FBS, 4 mM L-glutamine, and 100 U mL�1penicillin–streptomycin under shaking at 60 rpm for 14 days.Subsequently, the culture medium was replaced with DMEMsupplemented with KGN/b-CD, and the culture was continuedfor an additional 14 days. The medium was changed every3 days.5.13 Evaluations of cell adhesion, distribution, and viabilityAfter MSCs were cultured in the scaffolds for 1 day, nuclearstaining was performed to examine cell distribution, and live/dead staining was performed to evaluate cell viability. Celladhesion was observed by SEM. The cell-seeded scaffolds werewashed with PBS and fixed with 2.5% glutaraldehyde for 1 h.After fixation, the scaffolds were washed with Milli-Q water andlyophilized. The lyophilized scaffolds were coated with plati-num and examined by SEM.5.14 Quantification of DNA and sulfated glycosaminoglycan(sGAG)MSC proliferation in the scaffolds was evaluated by quantifyingDNA content. The cell/scaffold constructs were harvested atpredetermined time points for quantification of the DNA andsGAG content. The harvested samples were washed, freeze-dried, and digested with papain solution at 60 1C undershaking for 12 h. The papain solution was prepared by dissolv-ing papain in 0.1 M PBS buffer (400 mg mL�1) containingL-cysteine hydrochloride monohydrate (5 mM) and EDTA(5 mM) at pH 6. The amounts of DNA and sGAG in the digestionsolution were quantified using Hoechst 33258 and a BlyscantGlycosaminoglycan Assay Kit, respectively. The same number ofMSCs used for cell seeding were used for the control. Threeindependent samples were analyzed to calculate the mean andstandard deviation.5.15 Real-time PCRSamples were collected at predetermined time points for geneexpression analysis. The samples were washed three times withPBS and frozen in liquid nitrogen. The frozen samples werepulverized using a crusher and lysed with 1 mL of Sepasol-RNAI Super G solution. Total RNA was extracted according to apreviously reported protocol.3 Following reverse transcriptionusing a High-Capacity cDNA Reverse Transcription Kit, the expres-sion of glyceraldehyde-3-phosphate dehydrogenase (GAPDH;housekeeping gene), COL I, COL II, ACAN, and SOX9 was analyzedusing a QuantStudios 3 Real-Time PCR System (Thermo FisherScientific). The primer and probe sequences as shown belowwere used. GAPDH: (forward) 50-ATGGGGAAGGTGAAGGTCG-30,(reverse) 50-TAAAAGCAGCCCTGGTGACC-30, (probe) 50-CGCCCAATACGACCAAATCCGTTGAC-30; COL I: (forward) 50-CAGCCGCTTCACCTACAGC-30, (reverse): 50-TTTTGTATTCAATCACTGTCTTGCC-3 0, (probe): 5 0�CCGGTGTGACTCGTGCAGCCATC-3 0;COL II: (forward) 50GGCAATAGCAGGTTCACGTACA-30, (reverse)50-CGATAACAGTCTTGCCCCACTT-30, (probe) 50�CCGGTATGTTTCGTGCAGCCATCCT-3 0; ACAN: (forward) 5 0-TCGAGGACAGCGAGGCC-3 0, (reverse) 5 0-TCGAGGGTGTAGCGTGTAGAGA-3 0, (probe) 50-ATGGAACACGATGCCTTTCACCACGA-30; SOX9: (for-ward) 5 0-CACACAGCTCACTCGACCTTG-3 0, (reverse) 5 0-TTCGGTTATTTTTAGGATCATCTCG-3 0, (probe) 5 0-CCCACGAAGGGCGACGATGG-30. The 2�DDCt method was used to calculate therelative expression of each gene using GAPDH as the endogenouscontrol. MSCs used for cell seeding were used as the control. Theexpression level of each gene was normalized to that of thecontrol. Three independent samples were analyzed to calculatethe mean and standard deviation.5.16 Immunochemical stainingAfter MSCs were cultured in the GHP-HB-lipo scaffold for 7, 14,21, and 28 days, the samples were collected for immunochem-ical staining of ACAN and COL II. The samples were washedthree times with PBS and fixed with 4% paraformaldehyde for24 h at room temperature. The fixed samples were dehydrated,embedded in paraffin, and sectioned using a microtome toobtain cross-sections with a thickness of 7 mm. The paraffinsections were deparaffinized and treated with proteinase K for10 min for antigen retrieval, followed by blocking with aperoxidase-blocking solution for 5 min and incubation in10% goat serum for 30 min. The sections were incubated withrabbit polyclonal anti-aggrecan antibody (1 : 100; PA1-1746,Thermo Scientific, USA) and rabbit polyclonal anti-collagenPaper Journal of Materials Chemistry B9686 |  J. Mater. Chem. B, 2026, 14, 9676–9687 This journal is © The Royal Society of Chemistry 2026type II antibody (1 : 100; AB746, Merck Millipore, USA) at roomtemperature for 2 h. After washing three times with PBS, thesections were incubated with a peroxidase-labeled polymer-conjugated secondary antibody (EnVision+ System-HRP, Dako,Japan) at room temperature for 30 min, followed by colordevelopment using 3,30-diaminobenzidine (DAB; Dako, Japan)for 5 min. All procedures were performed at room temperature.The stained sections were examined by optical microscopy.5.17 In vivo animal experimentAll in vivo experiments were conducted with approval from theAnimal Experiment Ethics Committee of NIMS (accreditationno. 76-2023-10) and in accordance with the committee guide-lines. GHP-HB and GHP-HB-lipo scaffold discs (j 6 � 2 mm)were seeded with MSCs and cultured in vitro for 12 h beforeimplantation. Nude mice obtained from Charles River Labora-tories (Yokohama, Japan) were used for subcutaneous implan-tation. After 28 days of implantation, the mice were euthanized,and the samples were collected. Quantification of the DNAcontent and sGAG, as well as real-time PCR analysis wereconducted using the same methods described above. Fourindependent samples were analyzed to calculate the meanand standard deviation.The collected samples were also used for hematoxylin andeosin (HE) staining and immunochemical staining. Paraffinsections of each sample were prepared using the same proce-dure described above. The paraffin sections were deparaffinizedand stained with HE. Immunochemical staining of ACAN, COLII, and COL I was performed as described above using rabbitpolyclonal anti-aggrecan antibody, rabbit polyclonal anti-collagen type II antibody, and rabbit polyclonal anti-collagentype I antibody (1 : 100; PA1-26204, Thermo Scientific, USA).5.18 Statistical analysisQuantitative data from multiple groups were statistically ana-lyzed using one-way analysis of variance followed by Tukey’spost hoc test in GraphPad Prism software (GraphPad Software,Boston, MA, USA; https://www.graphpad.com). Two-tailed t-tests were used to determine the significance of differencesbetween two groups. The data are presented as the mean �standard deviation (SD), and statistical significance is indicatedas *(p o 0.05), **(p o 0.01), ***(p o 0.001), and ns (notsignificant).Conflicts of interestThere are no conflicts to declare.Data availabilityThe data that support the findings of this study are availablefrom the corresponding author upon reasonable request. Sup-plementary information (SI) includes the experimental data,such as FTIR and UV-vis spectra of KGN/b-CD inclusioncomplex; immunochemical staining of HB-EGF in GHP-liposcaffold; mechanical properties of GHP scaffolds; cell adhesion,cell distribution and cytocompatibility of GHP scaffold, and thegross appearance of the explanted scaffolds following in vivoimplantation. See DOI: https://doi.org/10.1039/d6tb00939e.AcknowledgementsThis work was supported by JSPS KAKENHI Grant Number24K03289 and JST SPRING Grant Number JPMJSP2124.References1 Articular Cartilage of the Knee: Health, Disease and Therapy,ed. H. K. Gahunia, A. E. Gross, K. P. H. Pritzker, P. S. Babynand L. Murnaghan, Springer New York, New York, NY, 2020.2 C. Chung and J. A. Burdick, Adv. Drug Delivery Rev., 2008, 60,243–262.3 E. A. Makris, A. H. Gomoll, K. N. Malizos, J. C. Hu andK. A. Athanasiou, Nat. Rev. Rheumatol., 2015, 11, 21–34.4 K. Johnson, S. Zhu, M. S. Tremblay, J. N. Payette, J. Wang,L. C. Bouchez, S. Meeusen, A. Althage, C. Y. Cho, X. Wu andP. G. Schultz, Science, 2012, 336, 717–721.5 T. D. Bornes, A. B. Adesida and N. M. Jomha, Arthritis Res.Ther., 2014, 16, 432.6 U. Nöth, A. F. Steinert and R. S. Tuan, Nat. Clin. Pract.Rheumatol., 2008, 4, 371–380.7 G. Filardo, F. Perdisa, A. Roffi, M. Marcacci and E. Kon,J. Orthop. Surg. Res., 2016, 11, 42.8 A. Goldberg, K. Mitchell, J. Soans, L. Kim and R. Zaidi,J. Orthop. Surg. Res., 2017, 12, 39.9 M. P. Lutolf, P. M. Gilbert and H. M. Blau, Nature, 2009, 462,433–441.10 O. Z. Fisher, A. Khademhosseini, R. Langer and N. A.Peppas, Acc. Chem. Res., 2010, 43, 419–428.11 C. Cha, W. B. Liechty, A. Khademhosseini and N. A. Peppas,ACS Nano, 2012, 6, 9353–9358.12 R. Schofield, Blood Cells, 1978, 4, 7–25.13 F. M. Watt and A. B. L. M. Hogan, Science, 2000, 287,1427–1430.14 S. W. Lane, D. A. Williams and F. M. Watt, Nat. Biotechnol.,2014, 32, 795–803.15 E. Alsberg, H. A. Von Recum and M. J. Mahoney, ExpertOpin. Biol. Ther., 2006, 6, 847–866.16 S. Tan and N. Barker, Growth Factors, 2013, 31, 175–184.17 A. K. Gaharwar, I. Singh and A. Khademhosseini, Nat. Rev.Mater., 2020, 5, 686–705.18 B. W. H. Mui, J. J. Y. Wong, C. E. Dumas, J. H. Wang, T. Bray,K. Hirose, L. Connolly, A. Winkel, S. Timmler, N. A. Bright,E. Sliauteryte, R. T. Káradóttir, P. G. Robey, K. Franze,K. J. Chalut and M. A. Storer, Science, 2026, 392, eady3136.19 A. J. Graham, M. W. L. Khoo, V. Srivastava, S. Viragova,H. Kim, K. Parekh, K. M. Hennick, M. Bird, N. Goldhammer,J. Z. Yu, G. Hu, N. T. Brinkley, L. Pardo, J. S. Amaya,C. D. Morley, N. Chadha, P. Lebel, S. Kumar, J. M.Rosenbluth, T. J. Nowakowski, O. Chaudhuri, O. Klein,Journal of Materials Chemistry B Paperhttps://www.graphpad.comhttps://doi.org/10.1039/d6tb00939eThis journal is © The Royal Society of Chemistry 2026 J. Mater. Chem. B, 2026, 14, 9676–9687 |  9687R. Gómez-Sjöberg and Z. J. Gartner, Nat. Mater., 2026, 25,1239–1251.20 Z. Zhou, Y. Hou, Z. Yao, Q. Han, S. Wang, H. Mei, H. Chen,B. Tian, M. Li, L. Cheng, C. Zeng and Z. Dong, Biomaterials,2026, 332, 124174.21 M. P. Lutolf and H. M. Blau, Adv. Mater., 2009, 21,3255–3268.22 A. J. Wagers, Cell Stem Cell, 2012, 10, 362–369.23 D. Wu, I. Eugenis, C. Hu, S. Kim, A. Kanugovi, S. Yue,J. R. Wheeler, I. Fathali, S. Feeley, J. B. Shrager, N. F. Huangand T. A. Rando, Nat. Mater., 2025, 24, 1364–1374.24 Y. Huang, Y. Wang, G. Yang, Y. Jiang, M. Sun, S. Cheng,L. Zhong, X. Pu, B. Yang, J. Chen, Z. Chen, C. Li, L. Yang,C. Chen and J. Zhang, Bioact. Mater., 2026, 57, 660–675.25 M. R. Hicks and A. D. Pyle, Trends Cell Biol., 2023, 33,112–123.26 K. S. Midwood, L. V. Williams and J. E. Schwarzbauer, Int.J. Biochem. Cell Biol., 2004, 36, 1031–1037.27 A. E. Mayorca-Guiliani, D. J. Leeming, K. Henriksen,J. H. Mortensen, S. H. Nielsen, Q. M. Anstee, A. J. Sanyal,M. A. Karsdal and D. Schuppan, npj Metab. Health Dis., 2025,3, 25.28 A. Page-McCaw, A. J. Ewald and Z. Werb, Nat. Rev. Mol. CellBiol., 2007, 8, 221–233.29 S. G. Almalki and D. K. Agrawal, Stem Cell Res. Ther., 2016,7, 129.30 N. Tremain, J. Korkko, D. Ibberson, G. C. Kopen,C. DiGirolamo and D. G. Phinney, Stem Cells, 2001, 19,408–418.31 G. Kasper, J. D. Glaeser, S. Geissler, A. Ode, J. Tuischer,G. Matziolis, C. Perka and G. N. Duda, Stem Cells, 2007, 25,1985–1994.32 K. Johnson, S. Zhu, M. S. Tremblay, J. N. Payette, J. Wang,L. C. Bouchez, S. Meeusen, A. Althage, C. Y. Cho, X. Wu andP. G. Schultz, Science, 2012, 336, 717–721.33 N. Monteiro, A. Martins, R. L. Reis and N. M. Neves,J. R.Soc., Interface, 2014, 11, 20140459.34 S. Maritim, P. Boulas and Y. Lin, Int. J. Pharm., 2021,592, 120051.35 P. Bhatt, R. Lalani, I. Vhora, S. Patil, J. Amrutiya, A. Misraand R. Mashru, Int. J. Pharm., 2018, 536, 95–107.36 L. Zhang, Q. Zhang, X. Wang, W. Zhang, C. Lin, F. Chen,X. Yang and W. Pan, Int. J. Pharm., 2015, 492, 40–45.37 R. Gharib, H. Greige-Gerges, S. Fourmentin, C. Charcossetand L. Auezova, Carbohydr. Polym., 2015, 129, 175–186.38 B. Peng, H. Chen, C. Lu, T. Zeng, M. Wang, T. Yoshitomi,N. Kawazoe, Y. Yang and G. Chen, Adv. Healthcare Mater.,2025, e02771.39 R. Sun, H. Chen, M. Wang, T. Yoshitomi, M. Takeguchi,N. Kawazoe, Y. Yang and G. Chen, Biomaterials, 2024,307, 122511.40 X. Liu, H. Chen, M. Wang, T. Zeng, T. Yoshitomi,N. Kawazoe, Y. Yang and G. Chen, Mater. Horiz., 2025, 12,8654–8666.41 W. Wei, W. Liu, H. Kang, X. Zhang, R. Yu, J. Liu, K. Huang,Y. Zhang, M. Xie, Y. Hu and H. Dai, Adv. Healthcare Mater.,2023, 12, 2300108.42 X. Liu, Y. Chen, A. S. Mao, C. Xuan, Z. Wang, H. Gao, G. An,Y. Zhu, X. Shi and C. Mao, Biomaterials, 2020, 232, 119644.43 R. Sun, M. Wang, T. Zeng, H. Chen, T. Yoshitomi,M. Takeguchi, N. Kawazoe, Y. Yang and G. Chen, Bioact.Mater., 2025, 44, 205–219.44 T. Tondreau, N. Meuleman, B. Stamatopoulos, C. De Bruyn,A. Delforge, M. Dejeneffe, P. Martiat, D. Bron andL. Lagneaux, Cytotherapy, 2009, 11, 559–569.45 K. B. Fonseca, F. R. Maia, F. A. Cruz, D. Andrade, M. A. Juliano,P. L. Granja and C. C. Barrias, Soft Matter, 2013, 9, 3283.46 X. Yan, J. Zhao, M. Yan, Y. Zhu, H. Song, Y. Tian, J. Gou,H. He, T. Yin, X. Tang, X. Jiang and Y. Zhang, J. Drug DeliverySci. Technol., 2026, 115, 107746.47 Z. Tianyuan, D. Haoyuan, L. Jianwei, H. Songlin, L. Xu,L. Hao, Y. Zhen, D. Haotian, L. Peiqi, S. Xiang,J. Shuangpeng, G. Quanyi and L. Shuyun, Adv. Funct. Mater.,2023, 33, 2213019.48 P. Kuchroo, V. Dave, A. Vijayan, C. Viswanathan andD. Ghosh, Stem Cells Dev., 2015, 24, 437–450.49 Y. Weng, M. Cai, J. Zhu, J. Geng, K. Zhu, X. Jin and W. Ding,Oncol. Res. Treat., 2013, 36, 256–259.Paper Journal of Materials Chemistry B