Abstract
The extracellular matrix is not merely a scaffold but a dynamic, hierarchically organized biological-material interface through which nanoscale composition, mesoscale architecture, and bulk mechanics are jointly translated into cell fate, tissue morphogenesis, and organ-level function. For 4 decades, Matrigel, a basement-membrane extract secreted by Engelbreth-Holm-Swarm mouse sarcoma cells, has served as the default engineered interface for stem cell expansion, organoid generation, and disease modeling. Its undefined composition, however, is also its central liability: lot-to-lot variability, xenogenic origin, and an inability to tune interfacial mechanics independently of biochemistry limit both mechanistic interpretability and clinical translation. Over the past decade, this limitation has catalyzed a shift from empirical, biologically sourced interfaces toward rationally engineered biohybrid systems, including synthetic poly (ethylene glycol) networks, chemically defined recombinant-protein substrates, tunable natural biopolymers, decellularized extracellular matrix, and, increasingly, functional nanomaterials that add hierarchical, multiscale control unattainable with any single polymer network. This review synthesizes recent advances in this transition through the lens of interface engineering: the deliberate, mechanistic specification of the physicochemical properties, ligand identity and density, degradability, elastic modulus, viscoelasticity, and nanoscale topology, that govern how a biohybrid material communicates with a living cell. We trace this hierarchy from nanoscale ligand presentation and nanomaterial-mediated force transmission, through the materials-characterization methods used to quantify interfacial properties, bulk interfacial mechanics and mechanotransduction, to organoid morphogenesis and tissue-level biohybrid function, and we give particular attention to functional nanomaterials, graphene-based nanosheets, magnetic nanoparticles, aligned nanofibers, and nanocomposite reinforcement, as emerging interface-engineering modules with direct relevance to the materials-science community. We conclude by proposing a generalizable design framework and outlining the standardization and manufacturing challenges that remain before engineered interfaces can be predictively designed for next-generation biohybrid systems.
1 Introduction
Biological systems achieve their remarkable functional diversity not through any single molecular component but through the hierarchical organization and dynamic regulation of interfaces that connect cells to their surrounding environment. The extracellular matrix (ECM) is the clearest embodiment of this principle: a multiscale, biological-material interface, spanning nanoscale adhesive ligands, fibrillar and nanotopological architecture, and bulk mechanical properties, whose composition and organization are continuously remodeled to guide adhesion, migration, proliferation, and differentiation (). Recreating this interface in vitro, as an engineered biohybrid system in which isolated cells or stem-cell-derived organoids experience nanoscale, mesoscale, and bulk cues that recapitulate the relevant biology, is therefore one of the central problems in interface engineering for cell biology, tissue engineering, and regenerative medicine (Figure 1). Throughout this review, we use the term “biohybrid interface' specifically to denote a cell–matrix composite system: an engineered material scaffold, whether synthetic, natural, or nanomaterial-based, that is combined with living cells or organoids to reconstitute a defined extracellular interface. This usage is distinct from the broader sense of “biohybrid' found elsewhere in the biomaterials and robotics literature, where the term can also describe constructs integrating living cells with active synthetic components such as actuators, electronic interfaces, or biohybrid robotic systems. Our focus here is restricted to passive and stimuli-responsive matrix systems that present nanoscale, mesoscale, and bulk cues to cells, not to actuated or electromechanical biohybrid devices.
FIGURE 1
Since the early 1980s, this interface-engineering problem has been addressed largely by a single reagent: Matrigel, a reconstituted basement-membrane extract solubilized from the Engelbreth-Holm-Swarm (EHS) murine sarcoma (). Matrigel is rich in laminin, collagen IV, entactin/nidogen, and a large, poorly defined complement of growth factors, and it gels rapidly at physiological temperature into a soft, fibrous interface that supports the self-organization of an extraordinary range of cell types (). Landmark demonstrations that stem-cell-derived organoids could recapitulate crypt-villus architecture, cerebral cortical layering, and hepatobiliary branching in Matrigel-embedded culture established organoid technology as a mainstream platform for developmental biology, disease modeling, and drug discovery (; ). This biological richness is precisely what has made Matrigel indispensable as a de facto biohybrid interface and, simultaneously, what limits its value as a platform for mechanistic interface science and clinical translation: because its nanoscale composition, mesoscale architecture, and bulk mechanics are coupled, emergent, and uncontrollable, Matrigel cannot function as a system in which individual interfacial design variables are specified and studied independently.
These limitations have motivated a sustained, decade-long effort to engineer biohybrid interfaces that replace Matrigel’s empirical complexity with rational, hierarchical, and modular design, without sacrificing the biological performance that makes reconstituted basement membrane so effective. articulated the founding logic of this effort early on, proposing that fully synthetic materials, rather than biologically extracted ones, could be engineered as instructive microenvironments that present defined adhesive, proteolytic, and growth-factor cues at specified densities. This effort spans several complementary strategies operating at different points in the material hierarchy: fully synthetic poly (ethylene glycol) (PEG) networks decorated with defined adhesive peptides and protease-cleavable crosslinkers, which specify the nanoscale ligand interface directly (; ); chemically defined recombinant ECM proteins for feeder-free, xeno-free stem cell expansion (Rodin et al., 2010; 2014); tunable natural biopolymers such as hyaluronic acid, alginate, and gelatin methacryloyl (GelMA), whose bulk mechanics can be decoupled and independently controlled (; Yue et al., 2015); decellularized extracellular matrix (dECM), which preserves native tissue-specific hierarchical architecture while removing much of the immunogenic content (; ); and, increasingly, functional nanomaterials, graphene-based nanosheets, magnetic nanoparticles, aligned electrospun and self-assembling peptide nanofibers, and nanocomposite reinforcement, that introduce nanoscale and mesoscale architectural control, remote actuation, and stimuli-responsiveness that no single polymer network can provide on its own ().
This body of work is significant not simply because it produces Matrigel alternatives, but because each engineered interface functions as a controlled experiment that isolates a single hierarchical variable, ligand type, nanotopology, fiber alignment, bulk stiffness, or stress-relaxation timescale, and asks how that variable alone shapes cell behavior. In doing so, this literature has begun to convert an empirical, largely descriptive relationship between ECM and cell fate into a mechanistic, quantitative, and increasingly predictive one, and to define transferable rules for engineering the cell-material interface across biohybrid systems generally, not only in cell culture. Because that conversion depends on being able to measure, not merely specify, nanoscale and bulk interfacial properties, this review also considers the materials-characterization toolkit, atomic force microscopy, oscillatory rheology, and related surface-analytical methods, that has made such quantitative interface engineering possible, a topic of direct relevance to a materials-science readership. The purpose of this review is to synthesize this mechanistic literature specifically around Matrigel and its engineered replacements, organized explicitly around the hierarchical, multiscale structure of the biological-material interface: from nanoscale ligand presentation and functional-nanomaterial-mediated architecture, through the tools used to quantify that architecture, bulk interfacial mechanics and mechanotransduction, to organoid morphogenesis and tissue-level biohybrid function. We close by distilling a generalizable interface-engineering framework and by outlining the standardization and validation challenges that remain before engineered interfaces are broadly adopted in place of undefined biological scaffolds (Figure 1).
2 Matrigel: biological utility and fundamental limitations of an undefined interface
Matrigel’s dominance in cell culture is a direct consequence of its biological completeness as an interface. As a secreted basement-membrane product, it presents cells with structural glycoproteins, laminin-111, collagen IV, nidogen/entactin, and perlecan, together with a substantial, if imprecisely characterized, cargo of growth factors, including transforming growth factor-β, epidermal growth factor, insulin-like growth factor, and platelet-derived growth factor (). This combination provides cells with adhesive ligands, a permissive fibrillar architecture, and soluble signaling cues simultaneously at a single interface, which is why Matrigel has supported such a broad range of applications, from classical three-dimensional mammary acinus assays to the generation of intestinal, cerebral, hepatic, and pancreatic organoids (Figure 2A).
FIGURE 2
The same interfacial complexity, however, undermines Matrigel’s reliability as an engineering substrate. Because it is harvested from EHS tumors implanted in mice, its protein and growth-factor composition differs measurably between production lots, and even within a single lot, local heterogeneity has been documented (). This variability has direct experimental consequences for organoid formation efficiency, differentiation trajectory, and drug response, which undermines reproducibility both within a laboratory over time and across laboratories. Mechanical characterization compounds the problem at the interfacial level specifically: Matrigel’s elastic modulus is coupled to its protein concentration and gelation conditions rather than to any independently addressable interfacial parameter, so investigators cannot hold nanoscale biochemistry constant while varying bulk mechanics, or vice versa, precluding the kind of controlled, hierarchical interface-engineering experiments that have become standard in mechanobiology ().
Beyond variability, Matrigel’s murine tumor origin raises translational concerns that are difficult to engineer around. Its xenogenic proteins pose immunogenicity risk in any therapeutic application involving in vivo delivery of Matrigel-cultured cells, and residual growth factors of unknown identity can confound the interpretation of both mechanistic interface studies and drug-screening assays (). A recent systematic appraisal of the field concludes that any viable Matrigel replacement must simultaneously satisfy four requirements, a defined and reproducible composition, sufficient biochemical complexity to support the biology under study, physiological relevance of individual components, and a traceable, non-xenogenic origin, criteria that most single-component or fully synthetic systems can meet only partially, which is why the field has pursued several parallel material strategies rather than converging on one (Wolff and Hendrix, 2025). These combined concerns, poor reproducibility, uncontrollable interfacial mechanics, and undefined composition, are the central motivations articulated across the recent Matrigel-alternative literature (Wolff and Hendrix, 2025), and they define the specific bar that engineered biohybrid interfaces must clear: not merely supporting cell viability, but doing so with a defined, reproducible, and hierarchically tunable nanoscale-to-bulk interface (Figures 2B, 3).
FIGURE 3
3 Quantifying the interface: a materials-characterization toolkit
A defining feature of interface engineering, as distinct from empirical biomaterial selection, is that every design variable introduced in Sections 4–7 must also be independently measurable; a hierarchy of interfacial properties is only actionable if each level can be quantified with a method matched to its length scale. Because the target readership of Frontiers in Materials includes materials scientists who may be less familiar with the cell-biology endpoints emphasized elsewhere in this review, this section summarizes the principal characterization techniques that have made quantitative, reproducible interface engineering possible.
At the bulk scale, oscillatory shear rheology remains the standard method for reporting the storage modulus (G′), loss modulus (G″), and, critically, the stress-relaxation half-time that distinguishes elastic from viscoelastic interfaces (; ). Because cells generate and sense forces over seconds-to-minutes timescales, reporting only an equilibrium elastic modulus, without a relaxation curve, is now understood to be an incomplete description of a hydrogel interface; parallel-plate rheometry with stress-relaxation or creep protocols is the minimum standard needed to distinguish, for example, a fast-relaxing alginate network from a covalently crosslinked, purely elastic PEG network at matched initial modulus. At finer length scales, atomic force microscopy (AFM) nanoindentation allows local elastic modulus to be mapped with sub-micron spatial resolution, revealing the kind of intra-gel and intra-tissue heterogeneity that bulk rheology averages away and that is directly relevant to Matrigel’s documented local variability (). provide a comprehensive materials-science synthesis of tissue and hydrogel stiffness measurements across length scales, and note that reported values for nominally the same tissue or material can vary by an order of magnitude depending on whether they were obtained by AFM indentation, rheometry, or macroscale tensile testing, underscoring the importance of reporting measurement method alongside any stiffness value (Figure 4).
FIGURE 4
Nanoscale ligand presentation and surface chemistry are typically characterized using a complementary set of surface-analytical tools: X-ray photoelectron spectroscopy (XPS) and time-of-flight secondary-ion mass spectrometry to confirm peptide or protein conjugation and surface stoichiometry; quartz-crystal microbalance with dissipation monitoring (QCM-D) to quantify real-time adsorption of growth factors or matrix proteins onto nanomaterial surfaces, of the kind exploited by graphene-oxide growth-factor carriers discussed in Section 7; and contact-angle goniometry to assess surface wettability, which correlates with non-specific protein adsorption and initial cell attachment. Nanoscale and mesoscale architecture, ligand clustering, nanofiber diameter and alignment, and nanomaterial dispersion within a hydrogel network, is most directly visualized using scanning and transmission electron microscopy (SEM/TEM), often combined with image-based quantification of fiber alignment (orientation order parameter) or nanoparticle spacing. Finally, degradation kinetics and network evolution are commonly followed using time-resolved rheology or swelling-ratio measurements alongside protease-cleavage assays, allowing the cell-responsive degradability engineered into PEG-4MAL and GelMA networks (Section 4) to be validated quantitatively rather than assumed from the presence of a cleavable sequence alone.
Taken together, this toolkit, rheology and AFM for mechanics, XPS/QCM-D/contact angle for surface chemistry, and electron microscopy for nanoscale and mesoscale architecture, constitutes the measurement infrastructure that converts interface engineering from a qualitative design philosophy into a quantitative, materials-science discipline, and its consistent use is what allows the findings summarized in Sections 4, 7 to be compared meaningfully across laboratories and material platforms.
In practice, however, this comparability is only partially realized. Even when two studies report the same nominal quantity, elastic modulus, for example, values are frequently obtained by different methods (AFM indentation versus bulk rheometry versus tensile testing) that are not interchangeable and, as noted above, can differ by an order of magnitude for the same material (); a modulus value reported without its measurement method attached is therefore not directly comparable to another study’s value of the same name. Testing protocols also vary in ways that are rarely standardized: rheological strain amplitude, indentation depth and tip geometry, hydration state at the time of measurement, and whether a hydrogel is tested immediately after gelation or after equilibration can each shift the reported value independent of any true difference in the material itself. Reporting practices compound this problem, as many studies report a single equilibrium modulus without an accompanying stress-relaxation curve, omit the measurement method or instrument settings entirely, or characterize only bulk mechanics while leaving nanoscale ligand density and surface chemistry unreported, so that two interfaces described as biologically or mechanically equivalent may in fact differ substantially at the length scale that determines cell response. Because these inconsistencies sit at the level of measurement and reporting rather than material design, they cannot be resolved by any single engineered interface, however well controlled, and represent one of the more tractable near-term contributions the field could make toward the broader reproducibility challenge discussed in Section 8: adopting shared minimum-reporting standards, method, instrument, and protocol alongside every quantitative interfacial value, would allow the kind of cross-study, cross-platform comparison this review has attempted to synthesize to be made on genuinely equivalent terms.
4 Engineering the cell-material interface: synthetic designer matrices
The most direct response to Matrigel’s undefined interface has been the development of fully synthetic hydrogel networks in which every structural and bioactive component of the cell-material interface is specified at the point of synthesis. The most extensively developed of these platforms is built on multi-arm poly (ethylene glycol) functionalized with maleimide end groups (PEG-4MAL), which reacts rapidly and selectively with free thiols under physiological conditions. This chemistry allows independent, modular incorporation of cysteine-terminated cell-adhesive peptides, most commonly RGD or laminin-derived sequences, and dithiol crosslinkers that can be engineered to include matrix-metalloproteinase-sensitive cleavage sites, so that the interface degrades only in response to cell-secreted proteases rather than on a fixed hydrolytic timescale (Zhu, 2010; ). In effect, PEG-4MAL chemistry converts the cell-material interface from an emergent biological property into a directly specifiable engineering parameter (Figure 5).
FIGURE 5
The explanatory power of this modularity was demonstrated most clearly by , who used a PEG-based synthetic interface to systematically dissect which nanoscale and bulk properties are required at each stage of intestinal organoid formation. By independently varying stiffness, degradability, and adhesive ligand identity, this work showed that fibronectin-derived RGD adhesion was sufficient for intestinal stem cell survival and that high interfacial stiffness, acting through a YAP-dependent mechanotransduction pathway, was required for stem cell expansion, whereas subsequent organoid differentiation required the interface to soften and to present laminin-derived adhesion (Figure 6B). reached a complementary conclusion using a similarly modular synthetic system, showing that adhesive ligand identity and crosslink density independently, rather than jointly, govern epithelial branching morphogenesis, reinforcing the principle that biochemical and biophysical cues act as separable, not redundant, design axes at a synthetic interface. This result illustrates a principle that recurs throughout the engineered-interface literature: distinct phases of tissue morphogenesis frequently require distinct, and sometimes opposing, interfacial properties, a requirement that a static reagent such as Matrigel cannot, by construction, satisfy, but that a designed and dynamically tunable synthetic interface can.
FIGURE 6
Related synthetic systems have extended this logic to additional epithelial and stem-cell contexts. used a chemically defined PEG-4MAL hydrogel, formulated using adhesive ligands identified by transcriptomic profiling of the target epithelium, to support both the generation and long-term passaging of human intestinal organoids and, in vivo, colonic wound repair, directly demonstrating that a fully synthetic, GMP-compatible interface can substitute for Matrigel in a therapeutically relevant delivery context. Subsequent work has applied the same PEG-4MAL platform, or closely related PEG-maleimide chemistries, to trophoblast-derived placental organoids, patient-derived intestinal enteroids, and endometrial epithelial-stromal co-cultures, in each case showing that synthetic interfaces support comparable, and in some respects more physiologically specific, differentiation outcomes relative to Matrigel, while eliminating lot-to-lot variability (Nikolaev et al., 2020). Collectively, this body of work establishes that the biochemical and biophysical functions historically attributed to Matrigel as an undifferentiated whole can, in most systems tested, be decomposed into a small number of independently specifiable interfacial design variables, each of which can be encoded explicitly in a synthetic network, and each of which can, per Section 3, be verified quantitatively rather than assumed.
5 Mechanotransduction at the interface: stiffness and viscoelasticity as programmable variables
A second major thread in the engineered-interface literature treats bulk interfacial mechanics not as an incidental byproduct of gelation but as a primary, independently programmable design variable at the cell-material interface. The foundational observation, that substrate stiffness alone can direct mesenchymal stem cell lineage commitment toward neurogenic, myogenic, or osteogenic fates when matched to the elastic modulus of the corresponding native tissue, was established using polyacrylamide gels with tunable, ligand-independent stiffness (; ). This work established interfacial stiffness as a bona fide morphogen-like cue and motivated the incorporation of tunable elastic modulus as a standard interface-design parameter in essentially all subsequent synthetic and natural engineered matrices (Figure 6A; representative dose-response data are shown in Figure 7). As discussed in Section 3, the reliability of this literature rests on consistent, method-matched stiffness reporting; note that native tissue stiffness itself spans nearly six orders of magnitude, from soft neural tissue to mineralized bone (Figure 4), which is the physiological range that engineered interfaces must be able to access and report accurately.
FIGURE 7
More recent work has shown that elastic modulus alone is an incomplete descriptor of interfacial mechanics. Native tissue interfaces are viscoelastic rather than purely elastic, dissipating stress over time through a process captured by stress-relaxation half-time. showed that alginate hydrogels engineered to relax stress faster, at matched initial elastic modulus, promote spreading, proliferation, and osteogenic differentiation of embedded mesenchymal stem cells, demonstrating that viscoelasticity is an interfacial mechanical variable distinct from, and not reducible to, stiffness (Figure 8; quantitative relaxation and differentiation data are shown in Figure 9). Subsequent studies have shown that matching the timescale of interfacial stress relaxation to the timescale of cellular force generation maximizes spreading (), that viscoelasticity modulates human neural progenitor maturation (Roth et al., 2021), and that interfacial viscoelasticity more broadly controls the spatiotemporal organization of multicellular tissue (). A comprehensive synthesis of this literature is provided by , who catalog how viscoelastic relaxation at the cell-material interface regulates spreading, migration, matrix remodeling, and differentiation, and by Saraswathibhatla et al. (2023), who review the underlying three-dimensional mechanotransductive signaling pathways.
FIGURE 8
FIGURE 9
These interfacial design principles are equally relevant to disease modeling. In breast cancer biology, interfacial stiffening, driven in vivo by collagen deposition and crosslinking in the tumor stroma, has been shown using engineered hydrogel interfaces to be sufficient to convert non-malignant mammary epithelial acini into an invasive, proliferative phenotype (Paszek et al., 2005; ), and dynamically stiffening interfaces, which mimic the temporal progression of tumor stiffening rather than presenting a single static modulus, more faithfully reproduce epithelial-to-mesenchymal transition than any static-stiffness interface (Wei et al., 2015). Together, this literature demonstrates that interfacial mechanics, both static (elastic modulus) and dynamic (stress relaxation, temporal stiffening), function as independent, quantitatively tunable design axes that engineered interfaces can access in ways that undefined biological gels such as Matrigel cannot.
6 Natural biopolymer, decellularized, and recombinant-protein interfaces
Fully synthetic interfaces offer maximal modularity but sacrifice some of the fibrillar, bioactive architecture native to tissue ECM. An alternative strategy retains a natural or naturally derived polymer backbone as the interface while engineering it for reproducibility and tunability. Hyaluronic acid (HA), a glycosaminoglycan enriched in many native stem cell niches, can be chemically modified with acrylate, thiol, or hydrazide groups to enable controlled, tunable crosslinking; showed that HA hydrogel interfaces support controlled self-renewal and differentiation of human embryonic stem cells, and subsequent work has extended this platform to independently tune interfacial stiffness for mesenchymal stem cell chondrogenic differentiation (Ren et al., 2021) and to engineer dynamic, photo-tunable HA interfaces whose modulus can be adjusted after initial gelation (Ondeck and Engler, 2016).
Gelatin methacryloyl (GelMA) represents a complementary strategy: gelatin, a denatured, non-immunogenic derivative of collagen, retains RGD and MMP-sensitive sequences natively, and methacryloyl functionalization enables rapid, dose-controlled photocrosslinking, giving investigators fine control over both interfacial stiffness and degradation rate without giving up collagen-derived adhesive motifs (Yue et al., 2015; ). This combination of native bioactivity and synthetic-style tunability has made GelMA one of the most widely adopted interfaces for both static three-dimensional culture and, increasingly, for nanomaterial-reinforced bioprinting (see Section 7).
For applications that require the fullest possible mimicry of tissue-specific interfacial architecture and composition, decellularized extracellular matrix (dECM) offers a distinct route: native tissue is processed to remove cellular and much of the immunogenic content while retaining tissue-specific structural proteins, proteoglycans, and bound growth factors, yielding an interface whose hierarchical organization is defined by its tissue of origin rather than by a tumor cell line (). Direct, tissue-matched comparisons support this approach: showed that gastric and intestinal organoids grown in gastrointestinal tissue-derived dECM hydrogels achieved organoid formation efficiency and gene-expression profiles comparable to, and in several markers superior to, matched Matrigel cultures, while also enabling long-term subculture and successful in vivo transplantation, and that dECM proteomic composition varied systematically with tissue of origin and donor age, a level of tissue-specific tunability unavailable from a single murine-tumor-derived reagent. dECM’s principal limitation mirrors that of Matrigel in miniature: because it remains a biologically sourced, batch-processed material, it retains some donor-to-donor and preparation-to-preparation variability, and its incorporation into bioinks (Section 7) typically still requires nanomaterial or crosslinker reinforcement to achieve adequate print fidelity.
Finally, for pluripotent stem cell expansion specifically, chemically defined recombinant ECM proteins have essentially displaced Matrigel as the reproducibility gold standard. Recombinant laminin-511 and laminin-521, and their integrin-binding E8 fragments, support long-term, feeder-free, xeno-free self-renewal of human embryonic and induced pluripotent stem cells with adhesion superior to full-length laminin or Matrigel itself (Rodin et al., 2010; 2014). Because these interfaces are single, sequence-defined recombinant proteins rather than complex mixtures, they eliminate lot-to-lot compositional variability at the nanoscale by construction, illustrating that, at least for two-dimensional expansion applications, a single well-chosen adhesive protein can be sufficient to replace an undefined multi-protein interface entirely.
7 Functional nanomaterials for hierarchical interface engineering
Sections 4, 6 have so far treated interface engineering primarily as a question of which bulk polymer chemistry, synthetic, natural, or recombinant, best reproduces a given set of ECM properties. Functional nanomaterials do not compete with these bulk chemistries as a further alternative in this same comparison; rather, they operate one hierarchical level beneath them, at the nanoscale and mesoscale length scales that any single bulk polymer network can occupy but cannot, on its own, independently specify or control. Functional nanomaterials constitute the newest, and arguably most consequential, layer of the interface-engineering hierarchy, because they act directly at the nanoscale and mesoscale, the size regime at which cells actually sense ligand spacing, surface topology, and local force, while remaining embeddable within any of the bulk hydrogel chemistries described in Sections 4–6. frame this as a third stage in the evolution of ECM-mimetic materials, progressing from undefined biological matrices (Matrigel), to defined synthetic or natural ECM-mimetic hydrogels, to ECM-mimetic hydrogel nanocomposites in which embedded nanomaterials confer mechanical reinforcement, electrical conductivity, or stimuli-responsive, remotely actuatable functionality that no single polymer network can provide on its own. Rather than treating nanomaterials as a separate materials catalog, it is more useful, and more consistent with the interface-engineering framing of this review, to organize them by the specific interfacial function each class contributes (Figure 10).
FIGURE 10
Two-dimensional nanosheets, most extensively graphene oxide (GO) and its reduced derivative (rGO), function primarily as high-surface-area biochemical and electrical interface modules (Zhu et al., 2024) (Figure 10A). Zhou et al. (2019) showed that GO flakes adsorb transforming growth factor-β3 (TGF-β3) with greater than 99% loading efficiency and release less than 2% over an extended culture period when embedded, together with human mesenchymal stem cells, in a three-dimensional collagen hydrogel, converting the bulk interface into a sustained, localized growth-factor reservoir that enhanced chondrogenic differentiation without repeated exogenous growth-factor supplementation (representative loading and release kinetics are shown in Figure 11). Because the adsorption step can be quantified directly by QCM-D or depletion ELISA (Section 3), the loading and release kinetics of GO-based reservoirs are among the best-characterized examples of nanomaterial-mediated biochemical interface engineering. GO and rGO incorporation additionally increases local stiffness and, where reduced, electrical conductivity, properties that have been exploited to enhance neural stem cell differentiation and osteogenic differentiation of bone marrow mesenchymal stem cells in composite hydrogels, and rGO microfibers have been shown to provide a more effective adhesive and neurogenic substrate than planar graphene films or tissue-culture plastic.
FIGURE 11
Magnetic nanoparticles introduce a fundamentally different interfacial capability: remote, wireless actuation of mechanotransduction without perturbing the bulk mechanics of the surrounding construct (Figure 10B). Superparamagnetic iron oxide nanoparticles conjugated to receptor-targeting ligands can be used to apply piconewton-scale torque directly to specific mechanosensitive receptors, including the TREK1 ion channel and integrin-α5, driving osteogenic differentiation of mesenchymal stem cells and, in combination with BMP2, synergistic mineralization in injectable hydrogel constructs and in an organotypic ex vivo fetal bone model (). Subsequent work extended magnetic ion-channel activation of TREK1 to promote osteogenic paracrine signaling not only in the directly targeted cells but also in neighboring, untargeted cells within the same construct, indicating that nanoparticle-mediated mechanotransduction can propagate through a tissue-engineered interface rather than remaining confined to individual labeled cells (). Because the applied force is delivered through an oscillating external field rather than through the bulk hydrogel, magnetic nanoparticles decouple mechanotransductive stimulation from bulk stiffness entirely, an interfacial degree of freedom unavailable to any purely chemical hydrogel design. The same nanoparticles can additionally be used architecturally: embedding superparamagnetic iron oxide nanoparticles within collagen hydrogels allows an external magnetic field to align collagen fibers during gelation, producing anisotropic scaffold architecture that supports directed neuronal process outgrowth (), and more recent magnetoresponsive fiber-integrating hydrogels, in which pre-formed, nanoparticle-loaded electrospun fiber segments are dispersed in a collagen precursor and aligned by a 20 mT field before crosslinking, have been shown to guide neural stem cell differentiation and produce interconnected, aligned neuronal networks substantially more effectively than randomly oriented fibers of the same chemistry (Sousa et al., 2024).
Aligned nanofibers, whether produced by electrospinning, magnetic templating, or peptide self-assembly (Zhu et al., 2019), address a distinct interfacial property: anisotropic architecture and directional force transmission, which isotropic bulk hydrogels cannot reproduce regardless of how their ligand density or stiffness is tuned (Figure 10C). Multiscale nanofiber-composite hydrogels, in which short, magnetically alignable gelatin nanofibers are dispersed within a gelatin-hyaluronic acid bulk network, allow nanofiber alignment and bulk hydrogel composition to be tuned as two independent interfacial variables; aligned nanofibers more effectively guide anisotropic dermal fibroblast elongation, while random nanofiber architecture at intermediate HA content favors myofibroblastic differentiation, showing that fiber-level topology and bulk biochemistry act as separable, combinable design axes. At the smallest length scale, self-assembling peptide nanofibers such as RADA16 form minimal, fully synthetic nanofibrous interfaces from short amphipathic peptide sequences that spontaneously assemble into β-sheet nanofibers under physiological ionic strength, offering a route to ECM-mimetic nanoarchitecture without any synthetic polymer backbone at all (Zhang, 2003).
Finally, nanomaterial reinforcement more broadly, incorporating rigid nanoparticles, nanoclay, or nanosheets into an otherwise soft polymer network, provides a route to decouple bulk mechanical properties from ligand and crosslink density, and to engineer stimuli-responsive interfaces whose stiffness or degradation can be switched by light, magnetic field, or pH after initial fabrication (Figure 10D). This same reinforcement strategy underlies most current nanomaterial-containing bioinks: dECM- and GelMA-based bioinks typically require nanomaterial or nanoparticle reinforcement to achieve adequate print fidelity and post-print mechanical stability, so that hierarchical, multiscale interface engineering and additive-manufacturing engineering converge on a common set of nanomaterial design tools. Taken together, this literature indicates that functional nanomaterials should be understood not as an optional additive to hydrogel design but as the nanoscale and mesoscale tier of the interface-engineering hierarchy, the level at which ligand presentation, local topology, and force transmission are actually specified, sitting directly beneath, and mechanistically coupled to, the bulk stiffness and viscoelasticity discussed in Section 5.
8 Toward a generalizable interface-engineering framework for biohybrid systems
Across the synthetic, natural-polymer, dECM, recombinant-protein, and nanomaterial literatures reviewed above, a small number of interface-engineering principles recur with sufficient consistency to be treated as generalizable design rules for engineered biohybrid systems, rather than as findings specific to any one material class. Figure 12A summarizes these principles as a linear framework, from material design through interfacial properties, mechanotransduction, and cell fate, to organoid architecture and tissue-level biohybrid function, alongside the design-space each material class currently occupies.
FIGURE 12
First, nanoscale ligand identity and density should be specified explicitly and independently of bulk mechanics. Multiple systems demonstrate that different integrin-binding sequences (e.g., RGD versus laminin-derived IKVAV) can direct qualitatively different cell behaviors even at matched elastic modulus, indicating that ligand chemistry is instructive at the interface, not merely permissive (; ).
Second, interfacial degradability should be cell-responsive rather than fixed. Incorporating protease-sensitive (typically MMP-cleavable) crosslinks, in either synthetic PEG or natural HA and GelMA backbones, allows the interface to be remodeled on a timescale set by cellular activity rather than by bulk hydrolysis, which is essential for processes such as organoid budding, lumen formation, and cell migration that require local, cell-directed interfacial opening.
Third, bulk elastic modulus should be matched to the developmental or tissue-specific target state, and, where morphogenesis proceeds through distinct stages, stiffness should be capable of changing over time within a single experiment or construct. Interfaces that allow post-gelation softening or stiffening more faithfully recapitulate multi-stage processes, such as stem cell expansion followed by differentiation, or tumor progression, than any static-modulus interface can.
Fourth, viscoelasticity and stress-relaxation timescale constitute an independent mechanical design axis from elastic modulus and should be characterized, using the rheological methods summarized in Section 3, and reported alongside it, since two interfaces with identical reported stiffness can produce substantially different cell behavior if their relaxation timescales differ.
Fifth, nanoscale and mesoscale architecture, ligand clustering, surface topology, nanofiber alignment, and nanomaterial-mediated force transmission or remote actuation, constitute a distinct design tier beneath bulk mechanics, one that functional nanomaterials are uniquely positioned to control. Where isotropic bulk hydrogels cannot reproduce anisotropic native tissue architecture or decouple mechanotransductive stimulation from bulk stiffness, embedded nanomaterials (aligned nanofibers, magnetic nanoparticles, reinforcing nanosheets) can, as demonstrated across the graphene-oxide, magnetic-nanoparticle, and nanofiber systems reviewed in Section 7.
Sixth, biological completeness and engineering control trade off against one another in a roughly monotonic way across material classes, from fully synthetic PEG networks (maximal control, minimal native complexity) through recombinant proteins, nanocomposites, and natural polymers (intermediate) to dECM and Matrigel itself (maximal native complexity, minimal control), as summarized in Figure 12B. Matrix, or interface, selection should therefore be treated as an explicit design decision matched to the scientific question at hand, rather than a default inherited from prior literature.
Finally, reproducibility and manufacturability should be treated as first-order interface-design criteria, not afterthoughts. Interfaces built from recombinant, single-sequence-defined components (e.g., laminin-511 E8, PEG-4MAL) or well-characterized nanomaterial dispersions inherently support batch-to-batch consistency and GMP-compatible scale-up in a way that biologically sourced or dECM-based materials cannot fully achieve, a distinction that becomes decisive for any application intended to progress from basic research toward clinical translation.
Beyond GMP-compliant production itself, several interrelated economic and regulatory barriers determine whether an engineered matrix can move from a research reagent to a clinical or commercial product.
Cost and scalability. Recombinant protein- and peptide-functionalized interfaces require expression, purification, and lot characterization steps that scale poorly relative to bulk animal-derived matrices; per-batch costs can be an order of magnitude higher than Matrigel, particularly when multiple recombinant components must be co-produced and independently validated. Synthetic and nanomaterial-containing systems face parallel constraints: multi-step functionalization chemistries and the specialized equipment required for reproducible nanoscale patterning (e.g., electrospinning, lithography, controlled self-assembly) are difficult to run at the throughput needed for commercial cell-therapy or organoid-biobanking applications.
Batch-to-batch consistency. Lot variability is often cited as a weakness of Matrigel, but engineered systems are not inherently immune to it. Recombinant proteins can vary in post-translational modification, folding, and aggregation state across expression runs; nanomaterial syntheses are sensitive to small changes in temperature, pH, or reaction time that propagate into mesoscale architecture and, in turn, bulk mechanical properties. Without standardized in-process controls, engineered platforms risk reproducing the same variability problem they were designed to solve.
Raw-material sourcing. Several engineered platforms depend on specialty polymers, custom peptide sequences, or nanomaterials available from a small number of suppliers. This narrow supplier base creates sourcing risk and can complicate long-term supply agreements in a way that is less familiar to laboratories and manufacturers accustomed to commercially available Matrigel.
Quality control. Matrigel’s release criteria rest on decades of empirical benchmarking against known biological performance. Engineered interfaces generally lack equivalent validated assays linking the nanoscale, viscoelastic, and surface-chemical properties discussed in Section 3 to functional lot-release specifications, making it difficult to define acceptance criteria that a manufacturer or regulator can audit consistently across production sites.
Regulatory pathway. Because engineered matrices frequently combine synthetic, biologic, and nanomaterial components, they may not map cleanly onto a single existing regulatory category (e.g., device versus biologic versus combination product), and sponsors typically need earlier, more iterative engagement with regulators than would be required for a traditional single-component biomaterial. This ambiguity adds time and cost to development and is a significant, underappreciated barrier to clinical translation.
Taken together, these factors indicate that the translation of engineered interfaces from research to clinical or commercial use will depend as much on parallel investment in scalable manufacturing processes, harmonized characterization standards, and early regulatory engagement as on continued innovation in materials design itself.
9 Conclusion
The transition from Matrigel to engineered biohybrid interfaces reflects a broader shift in biomaterials science, from empirically effective but mechanistically opaque biological reagents toward rationally designed, hierarchically organized systems in which nanoscale, mesoscale, and bulk properties are each independently specified, quantitatively measured, and, ideally, independently tunable. This shift has already yielded substantial mechanistic insight, from the discovery that stem cell niches require distinct, stage-specific mechanical and adhesive interfaces, to the recognition of viscoelasticity and nanomaterial-mediated architecture as independent regulators of cell fate, insight that would have been difficult or impossible to obtain using an undefined reagent such as Matrigel alone.
Collectively, these studies indicate that future biohybrid systems should no longer be viewed as collections of biomaterials, but as engineered biological interfaces whose hierarchical organization, nanoscale architecture, and dynamic physicochemical properties collectively determine cellular function. Establishing quantitative design rules that link material architecture to biological performance, of the kind summarized in Figure 12, and grounding those rules in the standardized materials-characterization methods discussed in Section 3, will enable predictive engineering of next-generation biohybrid systems, rather than iterative, empirical optimization of any single material class. These design principles are intended as a general conceptual framework rather than a universal design guideline: the relative importance of any given interface parameter, ligand density, degradability, elastic modulus, viscoelasticity, or nanoscale architecture, is application-dependent, and organoid morphogenesis, regenerative-medicine scaffolds, and immunoengineering constructs are each likely to prioritize a different subset of these parameters. The framework’s value therefore lies in specifying which properties should be measured and reported, not in prescribing a fixed hierarchy or a single optimal parameter set for every application.
Substantial challenges remain before engineered interfaces can be considered a full, rather than partial, replacement for biologically derived scaffolds. Reproducibility across the field would benefit from standardized reporting of nanoscale, viscoelastic, and surface-chemical, not only bulk elastic, interfacial properties, ideally following the shared measurement conventions outlined in Section 3; direct, side-by-side benchmarking of engineered interfaces against Matrigel across a wider range of primary and stem-cell-derived organoid systems is still limited; and the manufacturing scalability, cost, and regulatory pathway for synthetic, recombinant, and nanomaterial-containing interfaces, while improving, still lag behind the decades of accumulated use that support Matrigel’s continued default status in many laboratories. These scientific advantages must be weighed against practical barriers to manufacturing at scale, including GMP production costs, batch-to-batch consistency, raw-material sourcing, and quality-control standards, discussed in Section 8, as well as a regulatory pathway that remains less well-defined than the decades of precedent supporting Matrigel. Closing this gap will require not only continued materials innovation but parallel investment in scalable production, harmonized characterization standards, and early regulatory engagement, an infrastructure that engineered alternatives have only begun to build. Addressing these challenges by validating shared interface-engineering design principles, and the materials-characterization standards that underpin them, across tissue types and applications represents the clearest path toward a new generation of biohybrid culture systems that combine the biological performance of native ECM with the reproducibility and mechanistic transparency required for both rigorous materials science and clinical translation.
Statements
Author contributions
DZ: Conceptualization, Data curation, Formal Analysis, Investigation, Methodology, Resources, Software, Supervision, Validation, Visualization, Writing – original draft, Writing – review and editing. TM: Writing – review and editing.
Funding
The author(s) declared that financial support was not received for this work and/or its publication.
Conflict of interest
The author(s) declared that this work was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
Generative AI statement
The author(s) declared that generative AI was used in the creation of this manuscript. Specifically, generative AI assisted with language editing and the creation of conceptual illustrations, which were subsequently reviewed, refined, and verified by the authors to ensure scientific accuracy.
Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.
Publisher’s note
All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.
References
1
AisenbreyE. A.MurphyW. L. (2020). Synthetic alternatives to matrigel. Nat. Rev. Mater5 (7), 539–551. 10.1038/s41578-020-0199-8
2
Antman-PassigM.ShefiO. (2016). Remote magnetic orientation of 3D collagen hydrogels for directed neuronal regeneration. Nano Lett.16 (4), 2567–2573. 10.1021/acs.nanolett.6b00131
3
ChaudhuriO.KoshyS. T.Branco da CunhaC.ShinJ. W.VerbekeC. S.AllisonK. H.et al (2014). Extracellular matrix stiffness and composition jointly regulate the induction of malignant phenotypes in mammary epithelium. Nat. Mater13 (10), 970–978. 10.1038/nmat4009
4
ChaudhuriO.GuL.KlumpersD.DarnellM.BencherifS. A.WeaverJ. C.et al (2016). Hydrogels with tunable stress relaxation regulate stem cell fate and activity. Nat. Mater15 (3), 326–334. 10.1038/nmat4489
5
ChaudhuriO.Cooper-WhiteJ.JanmeyP. A.MooneyD. J.ShenoyV. B. (2020). Effects of extracellular matrix viscoelasticity on cellular behaviour. Nature584 (7822), 535–546. 10.1038/s41586-020-2612-2
6
CleversH. (2016). Modeling development and disease with organoids. Cell165 (7), 1586–1597. 10.1016/j.cell.2016.05.082
7
Cruz-AcuñaR.GarcíaA. J. (2017). Synthetic hydrogels mimicking basement membrane matrices to promote cell-matrix interactions. Matrix Biol.57-58, 324–333. 10.1016/j.matbio.2016.06.002
8
Cruz-AcuñaR.QuirósM.FarkasA. E.DedhiaP. H.HuangS.SiudaD.et al (2017). Synthetic hydrogels for human intestinal organoid generation and colonic wound repair. Nat. Cell Biol.19 (11), 1326–1335. 10.1038/ncb3632
9
DischerD. E.JanmeyP.WangY. L. (2005). Tissue cells feel and respond to the stiffness of their substrate. Science310 (5751), 1139–1143. 10.1126/science.1116995
10
Elosegui-ArtolaA.GuptaA.NajibiA. J.SeoB. R.GarryR.TringidesC. M.et al (2023). Matrix viscoelasticity controls spatiotemporal tissue organization. Nat. Mater22 (1), 117–127. 10.1038/s41563-022-01400-4
11
EnemchukwuN. O.Cruz-AcuñaR.BongiornoT.JohnsonC. T.GarcíaJ. R.SulchekT.et al (2016). Synthetic matrices reveal contributions of ECM biophysical and biochemical properties to epithelial morphogenesis. J. Cell Biol.212 (1), 113–124. 10.1083/jcb.201506055
12
EnglerA. J.SenS.SweeneyH. L.DischerD. E. (2006). Matrix elasticity directs stem cell lineage specification. Cell126 (4), 677–689. 10.1016/j.cell.2006.06.044
13
GerechtS.BurdickJ. A.FerreiraL. S.TownsendS. A.LangerR.Vunjak-NovakovicG. (2007). Hyaluronic acid hydrogel for controlled self-renewal and differentiation of human embryonic stem cells. Proc. Natl. Acad. Sci. U. S. A.104 (27), 11298–11303. 10.1073/pnas.0703723104
14
GjorevskiN.SachsN.ManfrinA.GigerS.BraginaM. E.Ordóñez-MoránP.et al (2016). Designer matrices for intestinal stem cell and organoid culture. Nature539 (7630), 560–564. 10.1038/nature20168
15
GongZ.SzczesnyS. E.CaliariS. R.CharrierE. E.ChaudhuriO.CaoX.et al (2018). Matching material and cellular timescales maximizes cell spreading on viscoelastic substrates. Proc. Natl. Acad. Sci. U. S. A.115 (12), E2686–E2695. 10.1073/pnas.1716620115
16
GuimarãesC. F.GasperiniL.MarquesA. P.ReisR. L. (2020). The stiffness of living tissues and its implications for tissue engineering. Nat. Rev. Mat.5, 351–370. 10.1038/s41578-019-0169-1
17
GuoX.LiuB.ZhangY.CheongS.XuT.LuF.et al (2024). Decellularized extracellular matrix for organoid and engineered organ culture. J. Tissue Eng.15, 20417314241300386. 10.1177/20417314241300386
18
HenstockJ. R.RotherhamM.RashidiH.ShakesheffK. M.El HajA. J. (2014). Remotely activated mechanotransduction via magnetic nanoparticles promotes mineralization synergistically with bone morphogenetic protein 2: applications for injectable cell therapy. Stem Cells Transl. Med.3 (11), 1363–1374. 10.5966/sctm.2014-0017
19
HenstockJ. R.RotherhamM.El HajA. J. (2018). Magnetic ion channel activation of TREK1 in human mesenchymal stem cells using nanoparticles promotes osteogenesis in surrounding cells. J. Tissue Eng.9, 2041731418808695. 10.1177/2041731418808695
20
HughesC. S.PostovitL. M.LajoieG. A. (2010). Matrigel: a complex protein mixture required for optimal growth of cell culture. Proteomics10 (9), 1886–1890. 10.1002/pmic.200900758
21
KimD. G.ChaG. D. (2025). Extracellular-matrix-mimetic hydrogels by using nanomaterials. Int. J. Mol. Sci.26 (11), 4987. 10.3390/ijms26114987
22
KimS.MinS.ChoiY. S.JoS. H.JungJ. H.HanK.et al (2022). Tissue extracellular matrix hydrogels as alternatives to matrigel for culturing gastrointestinal organoids. Nat. Commun.13 (1), 1692. 10.1038/s41467-022-29279-4
23
KleinmanH. K.MartinG. R. (2005). Matrigel: basement membrane matrix with biological activity. Semin. Cancer Biol.15 (5), 378–386. 10.1016/j.semcancer.2005.05.004
24
LancasterM. A.KnoblichJ. A. (2014). Organogenesis in a dish: modeling development and disease using organoid technologies. Science345 (6194), 1247125. 10.1126/science.1247125
25
LutolfM. P.HubbellJ. A. (2005). Synthetic biomaterials as instructive extracellular microenvironments for morphogenesis in tissue engineering. Nat. Biotechnol.23 (1), 47–55. 10.1038/nbt1055
26
NicholJ. W.KoshyS. T.BaeH.HwangC. M.YamanlarS.KhademhosseiniA. (2010). Cell-laden microengineered gelatin methacrylate hydrogels. Biomaterials31 (21), 5536–5544. 10.1016/j.biomaterials.2010.03.064
27
NikolaevM.MitrofanovaO.BroguiereN.GeraldoS.DuttaD.TabataY.et al (2020). Homeostatic mini-intestines through scaffold-guided organoid morphogenesis. Nature585 (7826), 574–578. 10.1038/s41586-020-2724-8
28
OndeckM. G.EnglerA. J. (2016). Mechanical characterization of a dynamic and tunable methacrylated hyaluronic acid hydrogel. J. Biomech. Eng.138 (2), 021003. 10.1115/1.4032429
29
PaszekM. J.ZahirN.JohnsonK. R.LakinsJ. N.RozenbergG. I.GefenA.et al (2005). Tensional homeostasis and the malignant phenotype. Cancer Cell8 (3), 241–254. 10.1016/j.ccr.2005.08.010
30
RenY.ZhangH.WangY.DuB.YangJ.LiuL.et al (2021). Hyaluronic acid hydrogel with adjustable stiffness for mesenchymal stem cell 3D culture via related molecular mechanisms to maintain stemness and induce cartilage differentiation. ACS Appl. Bio Mater4 (3), 2601–2613. 10.1021/acsabm.0c01591
31
RodinS.DomogatskayaA.StrömS.HanssonE. M.ChienK. R.InzunzaJ.et al (2010). Long-term self-renewal of human pluripotent stem cells on human recombinant laminin-511. Nat. Biotechnol.28 (6), 611–615. 10.1038/nbt.1620
32
RodinS.AntonssonL.NiaudetC.SimonsonO. E.SalmelaE.HanssonE. M.et al (2014). Clonal culturing of human embryonic stem cells on laminin-521/E-cadherin matrix in defined and xeno-free environment. Nat. Commun.5, 3195. 10.1038/ncomms4195
33
RothJ. G.HuangM. S.LiT. L.FeigV. R.JiangY.CuiB.et al (2021). Advancing models of neural development with biomaterials. Nat. Rev. Neurosci.22, 593–615. 10.1038/s41583-021-00496-y
34
SaraswathibhatlaA.IndanaD.ChaudhuriO. (2023). Cell-extracellular matrix mechanotransduction in 3D. Nat. Rev. Mol. Cell Biol.24 (7), 495–516. 10.1038/s41580-023-00583-1
35
SousaJ. P. M.MonteiroC. F.DeusI. A.CompletoA.StratakisE.ManoJ. F.et al (2024). Magnetoresponsive anisotropic fiber-integrating hydrogels for neural tissue regeneration. Small Struct.5, 2400213. 10.1002/sstr.202400213
36
WeiS. C.FattetL.TsaiJ. H.GuoY.PaiV. H.MajeskiH. E.et al (2015). Matrix stiffness drives epithelial-mesenchymal transition and tumour metastasis through a TWIST1-G3BP2 mechanotransduction pathway. Nat. Cell Biol.17 (5), 678–688. 10.1038/ncb3157
37
WolffL.HendrixS. (2025). Rethinking matrigel: the complex journey to matrix alternatives in organoid culture. Adv. Sci. (Weinh)12 (47), e08734. 10.1002/advs.202508734
38
YueK.Trujillo-de SantiagoG.AlvarezM. M.TamayolA.AnnabiN.KhademhosseiniA. (2015). Synthesis, properties, and biomedical applications of gelatin methacryloyl (GelMA) hydrogels. Biomaterials73, 254–271. 10.1016/j.biomaterials.2015.08.045
39
ZhangS. (2003). Fabrication of novel biomaterials through molecular self-assembly. Nat. Biotechnol.21 (10), 1171–1178. 10.1038/nbt874
40
ZhouM.LozanoN.WychowaniecJ. K.HodgkinsonT.RichardsonS. M.KostarelosK.et al (2019). Graphene oxide: a growth factor delivery carrier to enhance chondrogenic differentiation of human mesenchymal stem cells in 3D hydrogels. Acta Biomater.96, 271–280. 10.1016/j.actbio.2019.07.027
41
ZhuJ. (2010). Bioactive modification of poly(ethylene glycol) hydrogels for tissue engineering. Biomaterials31 (17), 4639–4656. 10.1016/j.biomaterials.2010.02.044
42
ZhuE.WangS.YanX.SobaniM.RuanL.WangC.et al (2019). Long-range hierarchical nanocrystal assembly driven by molecular structural transformation. J. Am. Chem. Soc.141, 1498–1505. 10.1021/jacs.8b08023
43
ZhuE.YuJ.LiY.-R.MaF.WangY.-C.LiuY.et al (2024). Biomimetic cell stimulation with a graphene oxide antigen-presenting platform for developing T cell-based therapies. Nat. Nanotechnol.19, 1914–1922. 10.1038/s41565-024-01762-6
Summary
Keywords
biohybrid systems, decellularized extracellular matrix, engineered hydrogels, extracellular matrix, functional nanomaterials, interface engineering, mechanotransduction, viscoelasticity
Citation
Zhao D and Mesquita TR (2026) From matrigel to engineered interfaces: mechanistic design principles for reproducible biohybrid cell culture systems. Front. Mater. 13:1954224. doi: 10.3389/fmats.2026.1954224
Received
31 July 2026
Revised
16 August 2026
Accepted
20 August 2026
Published
10 September 2026
Volume
13 - 2026
Edited by
Enbo Zhu, University of California, United States
Reviewed by
Jiaming Su, Zhejiang University, China
Yuqing Wang, University of California, Los Angeles, United States
Updates
Copyright
© 2026 Zhao and Mesquita.
This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
*Correspondence: Thassio Ribeiro Mesquita, thassioricard.ribeiromesquita@csmc.edu
Disclaimer
All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article or claim that may be made by its manufacturer is not guaranteed or endorsed by the publisher.