Reproducible extracellular matrices for tumor organoid culture: challenges and opportunities

Abstract

Tumor organoid models have emerged as valuable 3D in vitro systems to study cancer behavior in a physiologically relevant environment. The composition and architecture of the extracellular matrix (ECM) play critical roles in tumor organoid culture by influencing the tumor microenvironment and tumor behavior. Traditional matrices such as Matrigel and collagen, have been widely used, but their batch-to-batch variability and limited tunability hinder their reproducibility and broader applications. To address these challenges, researchers have turned to synthetic/engineered matrices and biopolymer-based matrices, which offer precise tunability, reproducibility, and chemically defined compositions. However, these matrices also present challenges of their own. In this review, we explore the significance of ECMs in tumor organoid culture, discuss the limitations of commonly used matrices, and highlight recent advancements in engineered/synthetic matrices for improved tumor organoid modeling.

Introduction

In cancer research, traditional models pose several challenges. in vitro 2D cell cultures tend to accumulate genetic mutations and lack stromal components, undermining their ability to reflect the true heterogeneity of tumors [1]. Similarly, although genetically engineered animal models mimic tumor structure and vasculature, they are labor-intensive and do not accurately replicate human disease processes [2, 3]. Additionally, neither conventional cell cultures nor animal models adequately reproduce the native tumor immune microenvironment [2]. Although patient-derived tumor xenografts and humanized immuno-oncology models offer improved mimicry of human tumors, they still face challenges such as high cost, lengthy preparation times, and issues with immunocompatibility [4, 5]. These limitations highlight the need for more physiologically relevant models to enhance preclinical screening and therapy development.

The tumor organoid model is a three-dimensional in vitro culture system that preserves the native cellular elements and structural organization of tissues [2, 6]. Preserving the native tumor (immune) microenvironment, as well as the genetic and histological heterogeneity, is profoundly important in tumor models. This is because the tumor microenvironment (TME) significantly influences tumor behavior, including metastasis, tumor progression, and carcinogenesis [2].

Given the ability of tumor organoids to accurately replicate interpatient, intra-, and intertumoral heterogeneity [7], it is essential to address and minimize the inherent technical variations associated with cancer organoid culture. This is crucial because establishing a reliable and reproducible model is a prerequisite for its integration into clinical practice, where it can enhance and expedite effective patient care. Despite recent advancements in organoid technology, the existing techniques for culturing cancer organoids still struggle with issues of control and reproducibility. Indeed, there are non-standardized steps in the organoid culture procedure (Fig. 1), including variations in extracellular matrices (ECMs), culture medium, and cancer tissue sources and processing.

Fig. 1

The non-standardized process, from obtaining tumor samples through the culture and maintenance phases, introduces technical variability in current organoid cultures, which affects reproducibility and hampers the faithful recapitulation of the tumor’s intrinsic heterogeneity. a The use of animal-derived extracellular matrices for 3D culture is limited by significant batch-to-batch variability and the risk of xenogenic contamination, while their complex, ill-defined, and poorly tunable composition further constrains investigations into the influence of the matrix on tumor biology. b The reliance on heterogeneous, ill-defined media components, such as conditioned medium and animal-derived serum, compromises the controlled delivery of soluble factors and unpredictably alters the organoid phenotype. c The diverse sources and collection methods for human tumor tissue, as well as the varied protocols for processing these tissues into three-dimensional organoid cultures, yield non-standardized models that may capture only a fraction of the patient’s cancer complexity

Matrices act as three-dimensional scaffolds that closely mimic the native extracellular matrix of tissues. The Engelbreth-Holm-Swarm (EHS) matrix, a basement membrane extract derived from mouse sarcoma [8], has played a pivotal role in advancing organoid research. This matrix consists of a wide range of ECM and biological components [8], collectively providing a complicated microenvironment for the encapsulated cells. This matrix contains adequate naturally occurring cell-adhesive regions that facilitate cell attachment. Additionally, it can undergo degradation and remodeling through enzymes expressed during organoid development. Nonetheless, the inherent batch-to-batch variability of the EHS matrix makes it unsuitable for clinical applications and difficult to tailor to the specific requirements of various organoid environments.

Understanding how cells sense and respond to their surrounding matrix is crucial for developing synthetic ECMs for organoids. So far, research on cell–matrix interactions has mainly focused on single-cell-type cultures, either seeded onto or embedded within a matrix [9]. Despite being less intricate than organoid cultures, these single-cell-type cultures offer significant insights into the development of matrices for guiding cellular behavior within a three-dimensional context. Cell culture is significantly affected by key parameters, such as the presence of cell-adhesive ligands, mechanical characteristics, scaffold structure, and the process of matrix remodeling [9]. These properties, while frequently studied separately, are inherently connected. The impact of single scaffold characteristics on a cell is partly shaped by the entire landscape of the system [10]. Moreover, evaluating the effect of changing a specific feature on cell behavior is not reliable unless other matrix characteristics are also considered. Given their clinical importance and availability, mesenchymal stem cells (MSCs) have been widely researched in single-cell cultures and on 3D engineered scaffolds. These studies have greatly contributed to our understanding of how cells interact with matrices, offering insights for designing scaffolds for organoids [11]. Notably, although most organoids originate from epithelial rather than mesenchymal cells, insights from studies on MSC-matrix interactions can be effectively utilized to develop customized matrices for epithelial organoids, such as intestinal epithelial organoids [12, 13]. Exploring the distinct interactions of epithelial and mesenchymal cells with their surrounding matrices could present an intriguing research opportunity.

Overall, the development of well-defined 3D biomaterials is a promising research area that improves reproducibility and more accurately reflects human biology. These materials can facilitate organoid formation and improve the faithful replication of the properties of both healthy and diseased tissues. For instance, synthetic biomaterials are chemically defined matrices that can be finely adjusted to affect and direct cellular behaviors. In this review, we will center our attention on ECMs and their essential attributes for 3D cell/organoid culture systems. Our emphasis will be on cutting-edge engineered ECMs designed for organoid culture, as well as addressing the present challenges and discussing recent interdisciplinary advancements for establishing standardized next-generation ECMs in organoid culture.

Organoid culture strategies

Before reviewing matrices, we first provide a summary of various organoid culture strategies. Organoid culture models are powerful tools for investigating tumor biology and the tumor immune microenvironment (TIME). Although these models can be generated from a variety of cell sources, including pluripotent stem cells (PSCs), induced pluripotent stem cells (iPSCs), and adult stem cells (AdSCs), stem cell–derived organoids often lack the full spectrum of immune, stromal, and vascular components, limiting their ability to fully recapitulate the native tumor microenvironment [14, 15].

An alternative approach involves enzymatically and/or mechanically digesting tumor tissue into small fragments or individual cells. These are then embedded into a three‐dimensional (3D) matrix such as collagen or Matrigel, which acts as a biomimetic extracellular matrix supporting cellular growth, differentiation, and spatial organization [2, 16]. To promote organoid growth and maintenance, culture media are supplemented with tailored combinations of growth factors and signaling molecules, including R-spondin, epidermal growth factor (EGF), Wnt3A, and Noggin (a BMP antagonist), that address the specific needs of different tumor types [2, 16]. Moreover, the addition of factors like the Rho-kinase inhibitor Y-27632 has been shown to enhance organoid growth and passage efficiency by promoting cell survival during the early stages of culture [17].

Several organoid culture strategies have been developed to better model the TME. One strategy involves reconstitution approaches, such as submerged Matrigel cultures, where organoids, predominantly composed of epithelial cells, are embedded in a dome or flat gel with a cocktail of growth factors [2]. In these systems, exogenous immune cells or cancer‐associated fibroblasts are later introduced to reconstruct the TME for immunotherapy studies [2]. In contrast, holistic approaches preserve the native TIME by maintaining small, intact fragments of tumor tissue. For instance, in microfluidic 3D culture, tumor spheroids are encapsulated within a collagen matrix housed in microfluidic devices [2, 18]. Here, cancer tissue is first dissociated enzymatically and mechanically into a mixture of fragments, spheroids, and single cells. This mixture is then filtered through 100 μm and 40 μm filters to yield three fractions, S1 (> 100 μm), S2 (40–100 μm), and S3 (< 40 μm). The S2 fraction, enriched with spheroids, is collected, mixed with collagen, and introduced into a microfluidic device [18]. This method enhances tumor modeling by preserving native microenvironments and endogenous cellular components such as lymphoid and myeloid cells, making it particularly useful for studying cancer progression and drug responses. Similarly, the air–liquid interface (ALI) culture method maintains native tumor–immune interactions without the need for reconstitution [2, 19]. In the ALI system, a collagen gel matrix is first prepared in an inner dish. Then minced primary tissue is mixed with a collagen solution and poured onto the prepared gel. The inner dish is then placed within an outer dish and incubated at 37°C to allow the gel to solidify. Media is added to the outer dish so that it diffuses through a permeable membrane to nourish the culture, while the top layer remains exposed to air, ensuring efficient oxygenation [19]. This setup preserves tumor cells along with their native immune and stromal counterparts, thereby maintaining their natural interactions.

Importance of ECM in tumor organoid culture

The native tissue ECM is a complex, dynamic network of polymers that includes proteins, polysaccharides, and proteoglycans [20]. It not only provides structural support through proteins like collagen and elastin, which impart tensile strength and elasticity essential for maintaining tissue integrity and three-dimensional architecture, but also actively regulates cell behavior via glycoproteins (e.g., laminin and fibronectin) and proteoglycans that sequester and present growth factors; these components interact with cell-surface receptors such as integrins to initiate signaling cascades (e.g., focal adhesion kinase pathways) that influence adhesion, migration, proliferation, and differentiation, while the ECM’s mechanical properties (e.g., stiffness, viscoelasticity) are sensed through mechanotransduction mechanisms that further direct cellular responses [20,21,22]. Dynamic remodeling is another hallmark of the ECM. Cells continuously modulate their surrounding matrix through the secretion of enzymes such as matrix metalloproteinases (MMPs), which degrade ECM components to allow for cell migration, tissue reorganization, and even angiogenesis. This dynamic balance between ECM deposition and degradation is crucial in both tissue homeostasis and tumor progression, as aberrant ECM remodeling can lead to a microenvironment that supports malignant behavior and therapeutic resistance [23, 24].

Unlike healthy tissues, which have a controlled and balanced ECM turnover, tumors usually undergo significant and chaotic changes in the composition, structure, and mechanical properties of their surrounding ECM. This disorganized remodeling ECM can constitute a substantial portion of the total mass of a tumor, ranging from approximately 60% to 90% [25, 26]. The modified ECM extensively alters TME behavior and properties through diverse biochemical and biomechanical interactions. These interactions include mechanical properties such as viscoelasticity, stiffness, porosity, degradation rate, and microstructure, and biological factors like ligand presentation and growth factor sequestration (Fig. 2). These interactions are critical for tumor cell behavior, tumor progression, metastasis, and responses to therapeutic interventions [26, 27]. Understanding these multifaceted roles of the ECM and cell-ECM interactions provides crucial insights for the rational design of new synthetic matrices with controlled and adjustable properties to recapitulate native biochemical cues and mechanical properties, ultimately enhancing the reproducibility and clinical relevance of tumor organoid models.

Fig. 2

The ECM modulates the phenotype of cancer organoids through a dynamic interplay of biochemical and biomechanical signals. Biomechanical factors including pore size, matrix composition, architecture, scaffold (visco)elasticity, degradation, and deposition work alongside biochemical cues such as growth factor sequestration and ligand presentation, to influence drug response, metastasis, and disease progression. Moreover, engineered 3D matrices with tunable properties are now poised to address previously untestable hypotheses about these cancer–matrix interactions. These systems also enable the modeling of reciprocal interactions between the ECM and tumor microenvironment cells, such as fibroblasts and immune cells, further elucidating the complex mechanisms that drive cancer phenotype and therapeutic outcomes

Cell-ECM interactions and 3D tissue structure can significantly change the phenotype of tumor cells. These insights originate from pioneering research conducted by the Bissell laboratory, which focused on the use of naturally derived scaffolds to model both normal and tumor mammary epithelium. As an instance, the EHS matrix, obtained through the decellularization of the basement membrane of murine Engelbreth-Holm-Swarm (EHS) sarcoma, was utilized to generate 3D polarized mammary tissue fragments [28]. Notably, tissue polarity and the interaction between β4 integrin and laminin were observed to confer resistance to apoptosis following cytotoxic drug treatment. This highlights how cancer progression and treatment can be influenced by the 3D cellular organization and cell-ECM interactions. In a similar study, Kenny and colleagues underscore the substantial influence of culture dimensionality, and the resulting three-dimensional morphology of cells, on the gene expression profile in multiple breast cancer cell lines [29]. These investigations highlight the dynamic interplay between cells and their extracellular matrix (ECM) and illuminate how these interactions shape the development of biomaterials tailored to regulate cellular behavior.

Tumor organoid models have the ability to mimic the composition, structure, mechanics, and cell–matrix interactions of native tumor tissues by being cultured within hydrogel matrices in a laboratory setting. Although tumor organoids show significant potential, there has been limited research employing them to represent intra- or intertumoral extracellular matrix heterogeneity. Additionally, only a few research efforts have focused on examining the impact of distinct ECM properties on the pathogenesis and therapeutic responses of patient-derived tumor organoids. Indeed, the poorly adjustable and ill-defined nature of animal-derived ECMs poses challenges for standardization, hinders the clinical application of organoids, and limits our understanding of the mechanisms governing the interactions between organoids and their ECM. Here, we explore the constraints of widely used scaffolds, such as the Matrigel and collagen, in tumor organoid cultures. Additionally, we highlight advancements in engineered scaffolds that offer consistent control over both the physical and biochemical properties for organoid models (Table 1).

Table 1 Types of ECMs for organoid culture: advantages and disadvantages

Natural biopolymer-based matrices

Natural biopolymer-based hydrogels can be constructed using proteins like collagen and fibrin, as well as various polysaccharides such as hyaluronic acid, alginate, chitosan, and cellulose. Various culturing systems have been established using naturally sourced and ECM-derived hydrogels, both individually and in various mixtures. The following paragraphs will provide comprehensive insights into these polymers, including their advantages and limitations in organoid culture applications.

Murine EHS matrix

Over the last decade, the EHS matrix, which is a basement membrane extract (BME), has emerged as the predominant matrix employed for the generation of both healthy and tumor organoids. It is commercially accessible under various trade names such as Corning Matrigel, Trevigen Cultrex, and Gibco Geltrex. In the reconstructed EHS matrix following mouse tumor extraction, a range of ECM proteins is preserved, primarily laminin (60%) and collagen IV (30%), along with entactin and nidogen [8, 30]. Together, these components offer crucial physical and biological support to encapsulated cells. Matrigel has been widely used in tumor organoid research because of its diverse milieu of TME-derived elements, which facilitate the expansion and maintenance of different cancerous and other niche cell types. Indeed, Matrigel provides a suitable environment for the development of organoids that can represent various tumor types, facilitating numerous in vitro studies designed to replicate patient-specific tumor biology. Moreover, the gelation of Matrigel is straightforward, primarily initiated by the self-assembly of laminin and the crosslinking of collagen and laminin by endogenous nidogen-1, occurring when the temperature exceeds 10 °C [30]. Although the EHS matrix offers several benefits, its animal origin results in significant batch-to-batch variability and the presence of ill-defined and xenogenic contaminants. These contaminations have the potential to unpredictably affect the organoid phenotype [31]. For example, Matrigel consists of nearly 2000 different proteins and more than 14,000 unique peptides [32]. Many of these components are recognized for their capacity to modify the phenotype of tumor cells. This complexity hinders the ability to conduct precise studies and limits its use in clinical applications. Despite special processing and the use of growth-factor-reduced (GFR) Matrigel, they still exhibit only around 53% similarity in protein content between batches [32]. Matrigel, in contrast to human tumor tissues, does not possess significant quantities of collagen type I and HA. Due to these limitations, the natural behavior and responses of cells may not be accurately reproduced [33].

Tumor organoids are expected to exhibit a spectrum of states, ranging from completely epithelial to entirely mesenchymal phenotypes, in contrast to normal epithelial organoids [34]. Consequently, mesenchymal tumor organoids may face challenges when establishing cultures based on Matrigel [35]. The EHS matrix is incapable of replicating patient-specific cancer ECM properties due to the non-tunability of its biochemical and mechanical properties. For instance, while the tumor ECM tends to have higher stiffness compared to healthy matrices, such as approximately 400 Pa in human healthy breast tissue, 1.08 − 68 kPa in colorectal cancer, and 3 − 42 kPa in breast cancer [36,37,38], the EHS matrix exhibits significantly lower stiffness, ranging from about 20–450 Pa [39,40,41]. In addition, the viscous nature of the EHS matrix complicates automated liquid handling, limiting its use in large-scale pharmaceutical applications. As a result, these drawbacks limit the EHS matrix's widespread use in cell-ECM interaction studies, drug screening and clinical applications. Furthermore, even if these challenges were addressed, producing EHS matrix on the necessary scale would involve a significant animal burden, raising ethical concerns.

Collagen matrices

The intense desmoplastic response in many solid tumors is typically characterized by increased deposition and remodeling of collagen, primarily types I to IV. Various aspects of the tumor are influenced by the elevated levels of collagen through a complex network of biochemical and biomechanical signaling cues, including cell adhesion sites and stiffness [42, 43]. Consequently, collagen type I matrices have emerged as a more prevalent TME biomimetic, offering a cost-effective substitute for Matrigel in in vitro tumor organoid models. In addition, by using a collagen-rich scaffold, the metastatic behavior of tumor cells can be recapitulated in 3D culture models [44]. This is due to the fact that collagen-based cell adhesion properties modulate the expression of MMPs, which are the primary enzymes responsible for tissue scaffold remodeling, consequently inducing invasive cellular behaviors [45].

It has been shown collagen I hydrogels, in an air–liquid interface (ALI) organoid culture model, facilitate long-term organoid cultures. This method allows for the culture of tumor tissue en bloc, preserving the original spatial arrangements among different cell types, including intrinsic stromal fibroblasts and immune cells [6, 46, 47]. In thick 3D cultures, where gas diffusion is significantly limited, efficient gas transport becomes critical. Employing an air–liquid interface culture, which utilizes collagen as a scaffold, rather than the conventional submerged culture that relies on Matrigel, enhances oxygen transport and results in superior organoid maturation [48]. Furthermore, a collagen-based extracellular matrix (ECM) allows researchers to fine-tune the mechanical stiffness by adjusting the collagen I concentration. In various studies, changes in intracellular responses have been correlated with variations in collagen I concentration, linking increased collagen density to a more invasive phenotype in cancer cell lines [49]. Such controllable matrices provide a valuable platform for investigating how ECM rigidity influences cellular behavior, including cell migration and signal transduction (e.g., via integrin-mediated mechanotransduction).

Donor tissue availability does not restrict collagen-based culture methods, as biomedical-grade collagen can be industrially sourced from cows and pigs. However, due to its common animal origin, the resulting matrices share akin limitations with Matrigel, including batch-to-batch variability, constrained biophysical and biochemical adjustability, and potential contamination with unclear and xenogenic impurities. Furthermore, certain collagen-based culture techniques involve co-culturing with support cells [50], thereby introducing undefined elements into the organoid culture. Additionally, the microstructural properties of collagen hydrogels, such as fibril diameter and alignment, are significantly affected by fluctuations in pH and temperature during the gelation process [51]. Consequently, collagen gelation performed under different environmental conditions can lead to architectural variability and a broad spectrum of collagen fibril sizes among samples, which may significantly influence cell–matrix interactions [42]. Thus, achieving reproducible and reliable collagen‐based culture systems will require rigorous standardization and precise control of gelation conditions. Additionally, alternatives such as synthetic hydrogels (e.g., PEG‐based systems), and biopolymer-ased engineered matrices can be employed to overcome these limitations and provide more tunable, reproducible, and defined culture environments.

Decellularized tissue-derived matrices

Recent advancements in decellularization techniques have made it possible to generate decellularized extracellular matrices (dECMs) from a variety of tumors, organs, tissues, and cell sheets, offering numerous potential applications [52, 53]. In this process, allogenic/xenogenic cells are removed from their scaffolds. In this type of ECM, numerous biochemical cues necessary for the creation of a spatially organized tissues, including the challenging incorporation of glycoproteins, are inherently present. As a result, the requirement for further chemical modification of the scaffolds is greatly reduced. Furthermore, the decellularized ECM preserves protein profile, the native architecture of ECM, and the apical-basal polarity [54,55,56]. A dECM preserves tumor/tissue-specific biochemical properties of the native scaffold. In a groundbreaking study, human tissue-derived ECM that accurately mirrored the brain-specific scaffold niche was employed to establish an in vitro glioblastoma multiforme model [57, 58]. Brain tissues' ECMs are rich in hyaluronic acid, glycosaminoglycans, and proteoglycans, while lacking fibrous collagen. As a result, patient tissue-derived ECMs offer distinct biochemical signals compared to other tissues and collagen-based hydrogels. Embedding patient-derived GBM cells in pdECM, rather than collagen, resulted in heightened invasiveness, leading to increased dissemination of cells and expansion of the tumor sphere core. Similarly, organ-specific metastases were successfully recapitulated using dECMs from various organs [59].

Miyauchi et al. employed a perfusion decellularization method to obtain liver dECM, with the aim of modeling hepatocellular carcinoma (HCC) through the recellularization with Huh7 and HLF cells through the bile duct [60]. The liver ECM preserved its original components and intact vascular networks, offering an optimal scaffold for HCC cells. Compared to normal liver scaffolds, the fibrotic model liver ECM exhibited notably elevated expression levels of multidrug resistance genes, oncogenes, and adhesion-related genes. Giobbe et al. introduced a dECM derived from porcine small intestine (pSI ECM) to culture endoderm-derived organoids [61]. Both biochemical cues and biomechanical cues were conserved, providing substantial support for organoid growth. Organoids cultured in pSI ECM exhibited sustained expression profiles akin to their natural state, in contrast to Matrigel-cultured organoids. These studies highlight the potential of utilizing patient-derived decellularized ECM in the development of cancer 3D culture models.

Despite the noted progress, decellularized ECM-based scaffolds encounter various challenges. Foremost, the quantity and quality of available ECM are limited by the number of donors and their health status. For instance, structural alterations that occur in emphysematous and fibrotic lung tissue can lead to reduced cell survival of less than 1 week in culture [62], or result in significant phenotypic changes [63]. On the contrary, a myocardial infarction is recognized for initiating remodeling processes that increase the stiffness of the extracellular matrix and modify its molecular composition. Interestingly, infarcted myocardium ECM is recellularized with MSCs, they exhibit an elevated secretion of pro-survival and immunomodulatory factors, promoting the survival of seeded cells [64]. Overall, the unfavorable impacts of dECM from unrelated healthy or diseased tissues on organoid generation should not be overlooked. Adjusting the biophysical properties of decellularized ECM-based scaffolds can be achieved through techniques such as cross-linking or mixing with other polymers, but developing decellularized ECM with suitable mechanical properties remains a challenge [65]. Furthermore, dECM-based hydrogels can vary chemically depending on the source of the ECM and the processing method. For instance, aggressive decellularization may remove crucial surface proteoglycans needed for successful organoid formation [66]. This variability in dECM-based hydrogels can lead to less reliable and reproducible results. Hence, it is essential to thoroughly characterize the dECMs before constructing an in vitro cancer model. Alternatively, researchers can utilize synthetic extracellular matrices, which offer more defined and reproducible properties, thereby providing a controlled environment for organoid development.

Other naturally-derived ECMs

Matrices for tumor organoid culturing can be provided by other naturally-derived ECMs such as alginate, gelatin, hyaluronic acid, and fibrinogen. While these materials are biocompatible, it should be noted that their biological and biophysical characteristics vary in comparison to collagen.

Some studies have shown that minimally supportive hydrogels can yield remarkable results under specific conditions. For example, Capeling and colleagues hypothesized that adhesive cues from the extracellular matrix might be unnecessary for culturing human iPSC‐derived intestinal organoids (HIOs), given that these organoids naturally develop an inner epithelium, an outer mesenchyme. They observed that pluripotent stem cells could differentiate into HIOs even without chemical signals from the alginate scaffold, suggesting that the cells may create their own microenvironment solely through mechanical support [67]. In a non-functionalized alginate environment, the generation of human and murine vascular organoids was also demonstrated by Rossen and colleagues [68]. Alginate is well-defined chemically compared to an EHS matrix [69]. Furthermore, the physical properties of alginate hydrogels, including parameters such as elastic modulus and toughness, can be readily adjusted [70]. For these reasons, along with its affordability, ease of modification and functionalization, biocompatibility, and extensive use in various biological and materials applications [71, 72], alginate is considered an attractive material for further exploration. Although alginate, being biologically derived, can exhibit some batch-to-batch variability in its mechanical properties [73], this variability is less pronounced compared to EHS matrices [32]. Similarly, Hyaluronic acid, both in isolation and in combination with chitosan, has been widely employed in the growth and generation of neural organoids [74,75,76]. It's worth noting that materials based on collagen and alginate have received FDA approval for a wide range of applications [77], facilitating rapid clinical implementation.

Synthetic/engineered polymer-based matrices

Numerous biomaterial platforms have been developed to enable the 3D culture of cancer cell lines, spheroids, as well as tumor and healthy tissues, providing significant insights into cancer biology and advancing therapeutic approaches for malignancies. However, synthetic/engineered matrices have not yet become standard practice in human cancer organoid cultures. Their integration in this context offers a promising opportunity to better understand the role of scaffolds in governing patient-specific tumors, and due to their reproducibility and well-defined properties, they hold significant potential for clinical applications. Here, we will review current advances in developing synthetic/engineered scaffolds for organoid models. We also will survey the advantages and disadvantages of these matrices in comparison with naturally-derived ECMs (Fig. 3).

Fig. 3

Significance of standardizing extracellular matrices for tumor organoid models. a Engineered scaffolds provide high batch-to-batch reproducibility and facilitate the standardization of organoid generation and maintenance, in contrast to animal-derived matrices. b Custom-designed engineered scaffolds can be tailored to replicate the original tumor scaffold's architecture and composition in a patient- and disease-specific manner

Synthetic polymer-based matrices

In contrast to the native biomolecules present in the in vivo ECMs of the TME, preclinical cancer models have utilized simple scaffolds constructed from biocompatible synthetic polymers, including polyethylene glycol (PEG) [78], polycaprolactone (PCL) [79], and poly(lactic-co-glycolic) acid (PLGA) [80].

Many studies using developed synthetic matrices for organoids culture. For example, Gjorevski and colleagues developed synthetic PEG-based and mechanically dynamic scaffolds for stem cells-derived organoids [12]. Interestingly, they revealed that diverse mechanical environments and ECM elements, which possess key matrix characteristics, were necessary to regulate different stages of intestinal stem cell (ISC) organoid generation. At the early stages of culture, intermediate stiffness (~ 1.3 kPa) of matrix decorated with the integrin-binding RGD peptide promoted ISC expansion via a mechanism that relies on yes-associated protein 1 (YAP). On the other hand, later intestinal stem cell differentiation and organoid formation required a soft matrix with lower stiffness (~ 190 Pa) and laminin-based adhesion. Therefore, hydrogel systems with dynamic biochemical and mechanical properties may be needed to provide the dynamic niche requirements for the different stages of organoid development. In another study, Hernandez-Gordillo and colleagues [81], designed a fully synthetic PEG-based ECM with tunable ligand concentration and physical characteristics, enabling the growth of enteroids and organoids from single cells. They revealed that PEG hydrogels with low stiffness (~ 100 Pa), crosslinked with MMP-degradable peptides, and in combination with the α2β1 integrin-binding peptide (GFOGER), provided the most efficient support for organoid development and growth, comparable to Matrigel. Cruz-Acuña et al. [13] designed a well-defined synthetic scaffold using a four-armed, maleimide-terminated PEG functionalized with RGD. This scaffold supports the highly reproducible generation and expansion of human intestinal organoids and comparable survival rates as in Matrigel. When RGD is replaced with other cell-binding ligands, including GFOGER, AG73, and IKVAV, significant reductions in organoid survival rates are observed.

PEG-based hydrogels are inexpensive, commercially available in various molecular weights and structures, and can be adjusted by regulating the proportions of the polymers and cross-linkers. They are also amenable to straightforward chemical functionalization, enabling the incorporation of biological ligands, cell-binding sites, signaling molecules, and crosslinking points. This facilitates the development of diverse matrices with precise tuning of both physical and biochemical properties [82,83,84]. Synthetic ECMs also possess precisely defined and highly reproducible characteristics, leading to reliable outcomes in preclinical models. Furthermore, several of these materials, such as PEG and PLGA, have obtained FDA approval for their clinical application. Nguyen et al. [85] studied over 1200 distinct synthetic hydrogels and identified materials that surpass Matrigel in sensitivity, reproducibility, and consistency in drug screening. This makes them viable alternatives for various cell-culture applications, particularly in contexts related to organoid vascularization.

These hydrogels have their own limitations. Synthetic hydrogels require both biophysical and biochemical adjustments for proper functionality. Without these modifications, such as cell-binding peptides, cells may struggle to adhere to the scaffolds, resulting in anoikis, a type of programmed cell death [86]. Inappropriate cue spacing can also result in cell death [87]. Moreover, customizing these hydrogels with precisely positioned peptides is costly and requires specialized knowledge, which makes them less appealing to researchers. Furthermore, the dependence of synthetic hydrogels on cytotoxic initiators and the cytotoxicity of their by-products limits the choice of polymers for cell culture applications [88,89,90]. Synthetic ECMs may contain unreacted groups, which can be cytotoxic [91]. Additionally, they may trigger immunological reactions in the body when used in medical implants or in organoids that possess immune components, potentially altering the results of immunotherapy research [92]. Synthetic polymers often experience high swelling and do not possess the cellular-level structural characteristics present in the native scaffolds. Given these considerations, it could be advantageous to design matrices based on biopolymers.

Biopolymer-based engineered matrices

To address some of the limitations of natural biopolymer-based and synthetic polymer-based matrices mentioned earlier, researchers are attempting to create biopolymer-based engineered scaffolds for 3D culture models. For instance, scaffolds based on alginate polysaccharide [67] and purified silk protein [93] have been developed as engineered matrices for intestinal organoids. These alternatives offer distinct biochemical, biophysical, and structural properties, all the while maintaining higher levels of chemically well-defined and reproducibility when compared to animal-derived scaffolds. Broguiere et al. [94] developed a well-defined soft fibrin matrix using purified human plasma fibrinogen, supplemented with laminin-111, and incorporating native Arg-Gly-Asp (RGD) adhesion domains. This tailored fibrin/laminin matrix facilitated the sustained long-term culture of epithelial organoids, providing a straightforward substitute for BME.

Another promising ECM for organoid models is developed using recombination-engineered proteins. Protein-based hydrogels can be independently adjusted for chemical functionality by introducing unnatural amino acids [95, 96], stiffness and viscoelastic characteristics [97,98,99]. Recombination-engineered proteins can be directed to undergo controlled degradation and remodeling processes by incorporating protease recognition sites [100] or altering crosslinking chemistry [101]. They can also be customized for various applications in the biomedical field [102] and rendered responsive to temperature changes [103]. The ability to program recombinant proteins has sparked growing enthusiasm in formulating protein-based scaffolds for 3D cultures. For instance, DiMarco et al. [104]. designed a recombinant engineered scaffold with an elastin-like structural backbone and an RGD amino acid sequence for primary organoids culture. The highest organoid formation efficiency was observed in soft (~ 200 Pa) engineered ECMs with a high concentration of RGD ligands, compared to stiff matrices with low RGD concentrations. This efficiency matched that observed in collagen I hydrogel controls. The inhibition of MMP, regardless of the mechanics of the elastin-like protein matrix, hinders organoid growth [104]. This effect is particularly evident in stiffer engineered scaffolds, emphasizing the need to define the degradation and mechanical criteria for matrix-organoid platforms. In another study, Hunt and colleagues [105], developed a hybrid, tunable, fully defined ECM called hyaluronan, elastin-like protein (HELP) for epithelial-only intestinal organoid culture. The HELP scaffold facilitates organoid formation, differentiation, and passaging, demonstrating comparable performance to animal-derived matrices even after multiple passages. Additionally, HELP promotes the formation and the differentiation of enteroids into various mature intestinal cell types. Its properties, including integrin-ligand concentration, stress relaxation rate, and stiffness, are easily customizable, enabling detailed investigations into organoid-matrix interactions and possible personalized optimization.

Protein-based hydrogels present several inherent limitations. Primarily, achieving successful recombinant expression for all proteins is not always possible, and the process of reaching their appropriate refolding and functionality can present significant challenges in practice. Furthermore, specific engineered proteins and self-assembling peptides may trigger immune responses [106,107,108]. Simply being human-derived does not guarantee the non-immunogenicity of engineered proteins [109]. It is imperative to be cautious in order to prevent the introduction of other immunogenic elements, such as bacterial endotoxins. Consequently, protein production for medical applications is preferably carried out in mammalian or yeast expression systems.

Hybrid matrices

Hybrid polymers can be formed by combining natural biopolymers with synthetic/engineered polymers, resulting in a diverse range of physicochemical and biological properties. This may be attributed to the synergistic interactions among the constituents of these materials. Numerous studies developed hybrid matrices that we mentioned several of them in previous sections [75, 105]. Studies have shown that the stiffness of collagen-based scaffolds can be adjusted through a range of synthetic approaches, including cross-linking with synthetic polymers [110], applying physical stimuli such as ultraviolet radiation or thermal conditions, or establishing an interpenetrating polymer network (IPN) [111]. Hence, collagen-based matrices can serve as matrices that regulate phenotypes by enhancing complexity to mimic the TME. Although numerous studies have explored techniques for manipulating the mechanical characteristics and structure of collagen matrices [51], altering collagen matrices often involves adding potentially toxic agents or making specific chemical modifications to the collagen, which can interfere with its natural crosslinking and ligand availability. Thus, currently, synthetic hydrogels functionalized with cell-binding cues are preferred for investigating how biomechanical properties impact organoid development. A study conducted by Bejoy et al. evaluated the impact of hyaluronic acid, a significant component of the brain's ECM, with or without heparin, on neuronal patterning of hiPSC-derived brain organoids. The study demonstrates that stiff Hep-HA hydrogels have a caudalizing impact on neural spheroids. They showed that the stiffness of this hybrid material can determine stem cells’ fate. Forebrain development was favored with lower modulus (300–400 Pa), while higher modulus (1000–1300 Pa) resulted in hindbrain development [112]. This research contributes valuable insights into the role of biomimetic extracellular matrix components in organoid development. In a separate study conducted by Xiao et al. [113], a hybrid scaffold was engineered for the 3D culture of primary GBM cells. This matrix was created through the conjugation of RGD to PEG, followed by the crosslinking of hydrogels by combining PEG and thiolated HA. They showed that elevated HA content was associated with increased expression of CD44, a cell-surface receptor binding to HA and a marker for cancer stem cells, compared to gliomasphere cultures. 3D cultures grown in the mentioned scaffolds with reduced hyaluronic acid content displayed approximately three times more drug sensitivity compared to scaffolds with same stiffness but higher hyaluronic acid. Moreover, the CD44 knockdown nullified this drug-resistance trait. These findings underscore the capacity of adaptable hybrid matrix platforms to offer distinct insights into the matrix-induced mechanisms governing cancer organoid phenotypes and their linked drug responses.

Conclusion and perspective

Specific properties of engineered/synthetic matrices hold promise for enhancing the reproducibility and efficiency of organoid models compared to natural biopolymers such as Matrigel and collagen. However, despite significant progress, organoid culture efficiency tends to be lower in engineered/ synthetic scaffolds than in Matrigel. Additionally, synthetic hydrogels facilitating organoid culture in a specific tissue often lack direct applicability to other tissues. The restrictions of engineered/synthetic matrices may stem from their restricted biodegradability and remodeling capabilities, as well as their comparatively lower incorporation of scaffold elements and cell-adhesive factors compared to the more intricate animal-derived matrices. These pivotal limitations of current synthetic scaffolds need to be tackled by future developments in polymer science and biomaterials engineering while ensuring that operational simplicity and accessibility for most researchers are maintained. A further restriction of recently engineered matrices is the limited spatiotemporal regulation over biophysical and biochemical properties, which is crucial for accurately modeling the dynamic TME. To overcome this restriction, some studies have engineered platforms for the reversible and irreversible modification of both biochemical and biomechanical properties of scaffolds, both spatially and temporally [114,115,116]. While most studies focus on using novel matrices for various cell types, the impact of these materials on organoid models has not been extensively studied. In the future, we expect a gradual shift from using EHS in organoid culture towards the development of new well-defined materials that allow for more precise control of the cells' mechanical and chemical microenvironments.