Three dimensional de novo micro bone marrow and its versatile application in drug screening and regenerative medicine

Abstract

The finding that bone marrow hosts several types of multipotent stem cell has prompted extensive research aimed at regenerating organs and building models to elucidate the mechanisms of diseases. Conventional research depends on the use of two-dimensional (2D) bone marrow systems, which imposes several obstacles. The development of 3D bone marrow systems with appropriate molecules and materials however, is now showing promising results. In this review, we discuss the advantages of 3D bone marrow systems over 2D systems and then point out various factors that can enhance the 3D systems. The intensive research on 3D bone marrow systems has revealed multiple important clinical applications including disease modeling, drug screening, regenerative medicine, etc. We also discuss some possible future directions in the 3D bone marrow research field.

Introduction

The bone/bone marrow (B/BM) organ hosts several types of cells, including hematopoietic stem cells (HSCs), bone marrow mesenchymal stem cells (BMMSCs), and their progenies. The HSCs can give rise to erythrocytes, leukocytes, and platelets, while the BMMSCs are able to differentiate into osteoblasts, chondroblasts, fibroblasts, endothelia, and adipocytes.^1,2 HSCs mainly reside within the BM, whereas mesenchymal stem cells similar to BMMSCs, including adipose-derived mesenchymal cells (AMSCs), can be found in connective tissues outside the B/BM.³

The HSCs and BMMSCs play important roles in processes such as hematopoiesis and bone formation/remodeling. Microenvironments in the B/BM, called niches, temporospatially regulate the behaviors of these specialized stem cells, including self-renewal, proliferation, migration, and differentiation.⁴ Complex cell–cell interactions are important components of B/BM niches. For instance, studies show that BMMSCs help establish the hematopoietic niche and maintain HSCs in the B/BM by secreting ligands for chemokine receptors, cell adhesion molecules (CAMs), extracellular matrix (ECM), and their associated integrin receptors.^4–8 In addition, a variety of soluble factors, e.g. bone morphogenetic proteins,^9,10 Wnt,¹¹ growth factors,^12,13 and interleukins¹⁴ are involved in niche activities. Physical conditions such as mechanical or sheer forces and oxygen gradient also modulate the behaviors of these stem cells.^15–18

The delicate balance in niches is maintained by complicated cross-talk between stem cells, ECM components, soluble factors, and physical properties of the microenvironment. Consequently, severe disturbances of B/BM functions that fail to be amended by the B/BM are associated with various disease processes; for instance, osteoarthritis,¹⁹ BM failure syndromes,²⁰ multiple myeloma,²¹ and acute myeloid leukemia.²² In addition, B/BM functions can be compromised due to excessive radiation,^23,24 drug toxicity,^25–27 and traumatic accidents. The multipotencies of HSCs, MSCs, and very small embryonic-like cells²⁸ and various disease processes associated with B/BM are intriguing and have led numerous researchers to investigate the B/BM and the stem cells within it, with the aim of using HSCs and MSCs to regenerate functional organs and to use the B/BM as a model to elucidate the mechanisms of diseases.

Two main obstacles currently limit regenerative research with stem cells from the B/BM. First, HSCs and BMMSCs occur at low frequency among the cells in the B/BM²⁹ so in vitro expansions of HSCs and BMMSCs are often needed. However, conventional two-dimensional (2D) in vitro sub-culturing of HSCs and BMMSCs are associated with a reduction in the ability of BMMSCs to proliferate and differentiate.²⁹ Second, in vitro 2D culturing of stem cells does not allow the extensive cell–cell, cell–ECM, or cell–physical environment cross-talk that occurs in vivo. This increases the difficulty of creating an in vitro functional B/BM resembling its authentic organ that is suitable for basic research and clinical application purposes.

Research projects looking into the behaviors of the B/BM system under pathological conditions (e.g. acute myeloid leukemia,³⁰ myeloma,²¹ and BM metastasis of breast cancer³¹ and the construction of diseases models for drug screening are often conducted either with in vivo animal models³² or with in vitro human B/BM stem cell lines.³³ Animal models allow for analysis of the in vivo interactions between key elements; however, the results from animal studies cannot always be translated to humans. On the other hand, in vitro studies of human B/BM stem cell lines in petri dishes fail to recreate microenvironments resembling those of the in vivo niches, making some of the results less applicable to in vivo systems. Hence, the limitations imposed by conventional animal models and culturing of B/BM stem cell lines in petri dishes restrict our investigations aimed at furthering the understanding of disease processes and hinder the development of high throughput drug screening protocols.³⁴

Recent developments including 3D printing techniques, biocompatible and biodegradable materials, together with a deeper understanding of the importance of complex dialogues between different components of B/BM, are now fueling the emergence of 3D artificial BM. A 3D BM system induced by biologically modified degradable carrier scaffolds (made of natural or synthetic biocompatible and biodegradable materials) is composed of de novo generated hematopoietic stromal cells, associated growth factors/ECM proteins, and calcified bones. The ability to establish microenvironments resembling specific niches in authentic B/BM using 3D bioengineered scaffolds provides solutions to the various limitations associated with conventionally cultured B/BM mentioned above.

The purpose of this review is to discuss the advantages of the use of 3D BM systems over 2D systems, to provide a brief survey of various factors influencing the functions and efficacies of 3D systems and current applications, and to propose possible future directions.

Why 3D over 2D?

Morphology

Conventionally, cells are grown in 2D culture systems that put geometrical and mechanical constraints upon the cells and interfere with cell morphology.³⁵ Research indicates that cell morphology has a strong influence on cell fate. For example, if a strain is placed along the axis of cell elongation, mesenchymal stem cells undergo proliferation. However, a strain placed perpendicular to the axis exerts no such proliferation.^36,37 Some research also links morphology to differentiation and gene expression.^38,39 Therefore, cell behaviors could be altered drastically in 2D culture, compared to in vivo behaviors. This emphasizes the need for well-developed 3D culture system that mimics in vivo cell–cell and cell–ECM interactions in order to develop a thorough understanding of how BM works and interacts with its surroundings.

Cell–cell interaction

Cell–cell interactions in 3D culture mimic the real BM interactions. Hematopoiesis, as an example, originates within the BM and is a dynamic process that creates all blood cellular components. In the BM microenvironment, fibroblasts and their products regulate hematopoiesis by interacting with hematopoietic cells.³⁹ Cukierman et al. discovered that fibroblasts have higher motility and divide more rapidly in 3D than in 2D culture conditions. The asymmetric shape that fibroblasts acquire in 3D culture represents the shape seen in actual BM.³³ In contrast, the mechanical constraint imposed by a 2D system causes the anchoring receptors to redistribute from the dorsal to ventral surfaces in adherent cells, creating an imbalance that deviates from the real tissue.⁴⁰

BM niche

An important component in the process of life-long blood generation is the presence of long-term repopulating HSCs (LT-HSCs) that undergo unlimited cycles of self-renewal. One problem encountered with in vitro experiments is the maintenance of this LT-HSC self-renewal ability.¹⁵ Previously studies using 2D systems showed that even the addition of facilitating factors, such as stem cell factor (SCF) and thrombopoietin (TPO), was still not sufficient to maintain the unlimited self-renewal ability of LT-HSCs.⁴¹ Much research indicates that the 3D BM model recreates a niche that resembles in vivo interactions. For example, Reagan et al. discovered that 3D myeloma BM mimics cell–cell interactions and cell–ECM interactions. Beyond this, a 3D model could be used as a clinical tool to inspect skeletal cancer biology and osteogenesis.⁴²

Optimizing 3D BM systems

Molecules

Multiple molecules are involved in regulation of BM niches. Different molecules promote proliferation or specific differentiation into different lineages. The molecules and their functions are summarized in Table 1.

Table 1.

Essential molecules that regulate bone marrow niches

Molecules	Actions	References
Angiopoietin (ANG)	Maintains HSC non-proliferation state	Arai et al.43
Bone morphogenetic proteins (BMPs)	Induces bone and cartilage formation; support growth and differentiation of HSC	An et al.9
		Yamaguchi et al.44
		Ogawa et al.45
		Torisawa et al.46
Chemokine ligand 12 (CXCL12), also called stromal cell-derived factor-1 (SDF-1)	Maintains HSC by binding to CXCR4 on HSC	Sugiyama et al.6
		Nagasawa et al.7
Epidermal growth factor (EGF)	Increases cell counts	Quan et al.47
Epinephrine and norepinephrine (EPI and NE)	Promotes migration, proliferation, and mobilization of CD34+ hematopoietic progenitors by activating β2-ARs	Méndez-Ferrer et al.48
Fibroblast growth factor (FGF)	Maintains MSC self-renewal	Di Maggio et al.49
Growth and differentiation factor 5 (GDF5)	Promotes intervertebral disc (IVD) phenotype.	Bucher et al.50
Intercellular adhesion molecule 1 (ICAM1)	Facilitate cell homing via cell–cell interaction	Mercier et al.51
Jagged1 (JAG-1)	Augments HSC cell counts by activating Notch1 signaling pathway	Stier et al.52
Osteopontin (OPN)	Regulates HSC proliferation	Stier et al.52
Parathyroid hormone (PTH)	Increases HSC cells counts by mediating proliferation	Calvi et al.53
Steel factor (SF), also called stem cell factor (SCF)	Regulates HSC self-renewal	Kent et al.54
Thrombopoietin (TPO)	Maintains HSC proliferation	Yoshihara et al.55
Vascular cell adhesion molecule 1 (VCAM1)	Facilitate cell homing via cell–cell interaction	Mercier et al.51
Wnt	Regulates HSC self-renewal	Kim et al.56
		Schreck et al.57

Scaffolds

Scaffolds are often used by scientists to study BM and bone or cartilage tissue engineering in a 3D system due to multiple benefits of the 3D system in supporting proliferation and differentiation, as discussed above. Multiple factors, such as scaffold composition, influence the direction of differentiation of MSC.⁵⁸ Porous scaffolds of varying pore sizes are good candidates for osteogenic differentiation.⁵⁹ Hydroxyapatite (HA) has also been investigated as an adjuvant as the bone matrix contains a mineral analog to HA. HA inclusion in scaffolds supports stem cell adherence, ECM production, and differentiation.^60,61

A good scaffold should create a supporting environment for stem cell proliferation and differentiation, leading to tissue regeneration. The properties of scaffolds, such as pore size and rigidness, also need to be considered.^62,63 Adjuvants can be added to maximize the facilitation process of cell attachment and growth. Currently, most scaffold use focuses on creation of an environment that resembles the BM matrix.⁶⁴ Various types of materials show promising results. These materials and their related features are summarized in Table 2.

Table 2.

Features of different materials that are used for the fabrication of 3D scaffolds

Material	Adjuvants	Ex vivo/In vivo	Stem cell types	Co-culture	Properties	References
Bioresorbable elastomeric polyurethane (PU) scaffolds	Hydroxyapatite (HA)	Ex vivo	MSC	Endothelial progenitor cells (EPC)	Volume: 9 mm3 (1 × 3 × 3 mm) Pore size: 200–630 µm	Duttenhoefer et al.65
Chondrogenic fibrin hydrogel	HA HA–methacrylic anhydride (HA–MA)	Ex vivo	Bone marrow-derived mesenchymal stem cells (BMMSCs)	Not reported	Sheet-like morphology Pore size: 10–100 µm	Snyder et al.66
Fibroin protein/ chitosan scaffolds	Not reported	Ex vivo	BMMSC	Not reported	Not reported	Deng et al.67
Inverted colloidal crystal (ICC) scaffolds	Not reported	Ex vivo	HSCs	CD34+	110 µm	Nichols et al.68
Macroporous collagen scaffolds	Not reported	Ex vivo	Human cord blood CD34+ cells	Murine S1/S1 stromal cells	Pore size: 50–200 µm	Noll et al.69
mPEG–PCL–mPEG (PCL) scaffolds	HA	Ex vivo	Porcine bone marrow stem cells (PBMMSCs)	Not reported	Pore size: 100 µm	Liao et al.70
Nonwoven polyethylene terephthalate (PET) fabric	Not reported	Ex vivo	Human cord blood CD34+ cells	Not reported	Pore size: 10–60 µm	Li et al.71
Photopolymerizable gelatin methacrylate (GelMA) hydrogel	Not reported	Ex vivo	BMMSC	Blood-derived endothelial colony-forming cells (ECFCs)	49.7 ± 11.8 µm for 1 mol/L gelMA; 30.13 ± 6.12 µm for 5 mol/L gelMA; 23.6 ± 5.85 µm for 10 mol/L GelMA	Chen et al.72
Plasma-medium gel	Fragmin/protamine nanoparticles and FGF-2	Ex vivo	BMMSC; adipose tissue-derived multilineage stromal cells (ASCs)	Not reported	Not reported	Kishimoto et al.73
Poly glycolic acid (PGA) and collagen scaffolds	Not reported	Ex vivo	Rat MSC	Not reported	Pore size: 8.0 ± 2.0 µm	Hosseinkhani et al.74
Poly(ɛ-caprolactone) (PCL) scaffold	Not reported	Ex vivo	MSCs	Not reported	Pore size: 330 × 260 × 100 µm	Ousema et al.75
Polyelectrolytes chitosan (CHT)/chondroitin sulfate (CS) multi-layered hierarchical scaffolds	Not reported	Ex vivo	BMMSC	Bovine chondrocytes (BCH)	Diameter size: 2 mm	Silva et al.76
Polyester nonwoven disc carriers (fibra-cel)	Not reported	Ex vivo	Mouse bone marrow cells	Not reported	Pore size: 500 µm	Tomimori et al.77
Porous cellulose microspheres	Not reported	Ex vivo	Bone marrow mononuclear cell	Not reported	Pore size: 300 µm	Mantalarisa et al.78
PuraMatrix (BD Biosciences) peptide hydrogel	Not reported	Ex vivo	MSC	Not reported	Pore size: 50–200 nm	Chen et al.79
Silk scaffolds in dexamethasone-free osteogenic media	Not reported	Ex vivo	MSC	GFP+ human multiple myeloma cell line (MM1S); endothelial co-culture	Pore size: 500–600 µm	Reagan et al.80
3D collagen scaffolds	Not reported	Ex vivo	BMMSC	Not reported	Pore size: 25 × 25 × 6 mm	Haneef et al.81

Computational modeling

Complex interactions occur between the stem cells, the growth factors, and the physical environment in the B/BM, so changes in these independent variables can significantly alter the behaviors of the B/BM—the dependent variables.⁸² Although the 3D architecture of the BM tissue in situ remains largely uncharacterized,^83–86 quantitative analysis of the interactions between these independent and dependent variables using computational modeling can aid in determining which components are more important for the specific functions of the in vivo B/BM system under normal or pathological conditions.^87–89 This in-depth understanding can guide the design of 3D de novo micro BM and advance investigations into disease processes using the 3D BM disease models.

Silva et al. developed a spatially-definite in silico model of the BM by employing TSim software and algorithm implementation.⁸⁷ This hematopoietic simulation model provides a geometric and compartmental view of the distribution, replication, differentiation, and mobilization/migration of hematopoietic cells within 3D BM. The spatial restriction of bone may prevent exponential growth and the production of blood cells within the BM, but the fractal-like structure of the trabeculae and sinuses in the BM compensated for and overcome this restriction, leading to production of a large quantity of blood cells daily upon demand.

They modeled the fine architecture of the spatial BM with an emphasis on sinusoids—a network of vessels through which mature blood cells leave the BM.⁸⁷ They used this model to simulate the replication, migration, and differentiation of HSCs and the exit of mature blood cells through the sinusoids. The behaviors of the model were assessed by measuring the percentages of HSCs, immature cells of granulocyte, erythrocyte, or platelet lineages, and mature cells of the three lineages. Data obtained by simulations were consistent with those obtained from asymptomatic patients under physiological conditions or after radiation. Their study shows that the fine architecture of the sinusoids in the B/BM is important for hematopoiesis and that maximization of hematopoiesis can be achieved by modulation of sinusoidal structures.

Colijn and Mackey modeled the dynamics in periodic chronic myelogenous leukemia (PCML) by regulating parameters such as the rate of differentiation into the leukocyte lineage.⁸⁸ Their results indicate that changes in this rate of differentiation, as well as in the rate of leukocyte amplification and the rate of HCS apoptosis, are required to generate a PCML phenotype.

Current applications

Disease modeling and drug screening

The 3D BM system is a useful tool for modeling various diseases. The use of a 3D BM system together with recently developed microfluid^46,90 or perfusion¹⁵ culture technologies can help in the development of ex vivo models of various diseases that provide accurate portraits of in vivo interactions in the niche and predict in vivo responses to different treatments. Several current studies on modeling of various diseases using 3D BM system are described below.

Aljitawi et al. created a scaffold of BMMSCs co-cultured with leukemia cells.³⁴ Not only does this 3D system mimic the structure, it also offers the supporting microenvironment required by the leukemia cells. In this 3D system, leukemia cells are more chemo-resistant and proliferative due to this supporting niche. The cells in this 3D system also shows distinctive physiological features not seen when they are cultured in a petri dish (2D). For example, leukemia cells cultured in a 2D system undergo a short proliferation stage followed by inhibition.^34,91 Cells grown in a 3D system do not show this deviation in proliferation.³⁴

Ferrarini et al. used a 3D model to study multiple myeloma cell growth. They noted good preservation of myeloma cells in the supporting niche, complete with the development of bone lamellae and blood vessels, thereby achieving optimal nutritional support. They also screened the toxicity of the drug Bortezomib on myeloma cells. Importantly, the Bortezomib treatments in the 3D model showed a promising response consistent with that observed in vivo. Overall, the research indicates that 3D models are good tools for the study of multiple myeloma and for investigation of pre-clinical treatments.²¹

Mukhopadhyay et al. used 3D cultures of macrophages, derived from BM to study epithelioid granuloma, an inflammatory disease caused by macrophage accumulation as an inflammatory response.⁹² The addition of carbon nanomaterial to cells cultured in the 3D system was positively correlated with the induction of multinucleated epithelioid granulomas by asbestos fibers or multiwalled carbon nanotubes. This 3D system can serve as a tool to evaluate the potential risk of some engineered nanomaterials in inducing granulomas in lungs or mesothelium.⁹³

Osteoarthritis is a degenerative disease of bone and cartilage.⁹⁴ Over the years, scientists have used MSCs as the main material for the constructs used to study osteoarthritis because of their ability to differentiate into osteocytes and chondrocytes.^95,96 Similar to some scaffolds described in Table 2, HA is often used in bone constructs.⁹⁷ Chitosan with TGF-β3 is used in cartilage constructs.⁹⁸ A study by Snyder et al. discussed the application of 3D systems in cartilage repair. The 3D system is composed of BMMSCs incubated in a hydrogel fabricated from fibrin/hyaluronic acid and methacrylic anhydride. The system shows induced BMMSC proliferation, early chondrogenesis, and increased mechanical strength, thereby supporting its application in the delivery of BMMSCs for potential repair of injured cartilage.⁶⁶

Stem cell expansion

One barrier in much of the current BM research is the limited number of stem and nucleated cells in the BM. Therefore, expanding their number is desirable. A 2D system imposes multiple barriers to this expansion, such as reduction in clonogenicity and differentiation capacity; thereby posing a challenge to physiological and pathological studies.⁹⁹

Papadimitropoulos et al. demonstrated a promising 3D system using a porous scaffold. Unlike the case in a 2D system, the proliferated and differentiated cells in their 3D system maintain their progenitor properties with a 4.3 fold higher clonogenicity and a higher efficiency of differentiation. The multipotency-related gene clusters are significantly upregulated in the 3D system. Overall, the 3D system creates a supportive niche that allows the expansion of MSCs while maintaining their normal functionality; thereby creating a system that can be used pre-clinically or clinically in the field of regenerative medicine.²⁹

Other research done by Sadat Hashemi et al. suggested that ex vivo expansion of HSCs is a possible way to increase umbilical cord blood CD34⁺ cells (UCB-CD34⁺), which are useful in BM transplantation. UCB-CD34⁺ cells were co-cultured in the 3D scaffolds, which mimic the BM microenvironment and supports stem cell differentiation and expansion, leading to a significant increase in UCB-CD34⁺ cell counts.¹⁰⁰

Regeneration

Regenerative medicine is a field dedicated to the regeneration of damaged tissues or organs by the body’s own repair mechanisms or by implantation of in vitro grown tissues and organs. Stem cells are commonly used in this field due to their ability to differentiate into many types of tissue and proliferate without losing their differentiation ability. Many types of stem cells are used in numerous studies, but MSCs or BMMSCs have received the most attention. Recognition of the benefits of 3D models had led to more researches focused on incorporating MSCs or BMMSCs into 3D scaffolds to optimize the B/BM formation results. The following recent studies all exploited the benefits of 3D scaffold models.¹⁰¹

For years, researchers have used dental pulp stem cells (DPSCs) in the hope of regaining tooth sensitivity and vitality after root canal therapy.¹⁰² However, the limitation is the availability of DPSCs. Recently, Ravindran et al. discovered that incorporating BMMSCs into 3D scaffolds could lead to odontogenic lineage differentiation into DPSCs both in vitro and in vivo, resulting in the formation of vascularized tissue resembling dental pulp. The 3D scaffold provides insights into possible applications of dental pulp regeneration.¹⁰³

Tang et al. focused on large bone reconstruction in an animal model. Treatment with biodegradable polydioxanone (PDO) alone is not sufficient as a long-term treatment as the PDO material was completely degraded and substituted with fibrous tissue after 24 weeks. More promising results (better shape and radian) were obtained using a 3D scaffold of BMMSCs. This research indicates a possible platform for long bone repair research and clinical applications using 3D BMMSC scaffolds.¹⁰⁴

Haneef et al. evaluated the proliferation and differentiation of BMMSCs into cardiomyocytes in a 3D collagen scaffold. The 3D system supported proliferation of more cells following electrostimulation. Immunocytochemistry confirmed the differentiation of BMMSCs into cardiomyocytes, as indicated by the presence of the cardiac markers troponin I, myosin, and desmin. This research points out that 3D scaffolds could be valuable in regeneration of cardiomyocytes from patients suffering from myocardial infarction.⁸¹

Perspective

The development of functional de novo 3D B/BM in vitro has long been considered an important and challenging task. In adult mammals, BM is the only organ that is able to generate the full spectrum of hematopoiesis under normal conditions.¹⁰⁵ This exceptional organ repairs itself in the same way that it originally forms in an embryo. The combination of in vivo, ectopic, de novo B/BM regeneration and in vitro bioengineering by utilizing a combination of chemical, biological, and molecular approaches will allow researchers to develop cost-effective and useful 3D B/BM.¹⁰⁶ These manageable humanized 3D B/BM models will facilitate in-depth, dissectible, and integrated investigation into the temporospatial microenvironmental factors that regulate the fates of HSCs and MSCs. Stage- and lineage-specific niches will be established and identified by grafting specific types of stromal cells, including isogenic stromal cells, and by modifying these cells and their associated microenvironments with specific chemical and biological regulators (Figure 1). Human CD34⁺ as well as BMMSC cells will be appreciably expanded in ossicles containing human stromal cells, rather than animal stromal cells, because the human stromal cells may provide more compatible growth factors, interleukins, and CAMs than can be provided by mouse stromal cells. Bioengineered human stromal cells, such as paired isogenic gene knock-in and knock-out cells could be stably transduced with GFP, which will allow researchers to track them continuously in order to monitor their fate and geometrical contribution with time. With the help of 3D multiphoton microscopy, qRT-PCR and microarray techniques, researchers will be able to establish the tissue architecture, heterotypic cell–cell interactions, cell replication/differentiation, dedifferentiation/transdifferentiation,^107,108 and differential gene expression profiles. The establishment of efficient and reproducible 3D B/BM culture and stem cell expansion protocols suitable for industry and automated high throughput platforms will enable the development of time-saving strategies for clinical applications. The outcomes of these efforts and studies will provide useful and important information for carrying out microenvironmental and HSC research to develop experimental approaches that more realistically create 3D B/BM systems. In addition, this research will increase the understanding of the sophisticated cellular and molecular mechanisms that govern the microenvironmental regulation of hematopoiesis and de novo tissue regeneration.

Figure 1.

A cartoon representation of a bioengineered 3D micro bone marrow system and its applications. a,¹⁰⁶; b, Biotek; c,⁴⁶

The low frequency of BMMSCs motivates the use of stem cells from other sources. For example, adipose mesenchymal stem cells are promising due to their large quantities and the procedure for obtaining adipose MSCs is relatively noninvasive and causes little discomfort.^73,109 In addition, the in vivo microenvironment can be optimally mimicked by incorporating, co-culture into research projects. Co-culture techniques play an important role in enabling complex cell–cell interaction, allowing better elucidation of in vivo activities. Furthermore, co-culture technique can create a platform for clinical applications in regenerative medicine. In vitro perfusion can also substitute for static media in order to mimic the dynamic microenvironment in vivo.

Concluding remarks

Much evidence indicates that 3D BM models have multiple advantages over 2D systems. Different 3D scaffold materials and molecules can be used to achieve different functionalities by optimizing microenvironments for specific stem cell sources. 3D models offer insightful platforms for researchers and clinicians in multiple fields, including disease modeling, drug screening, regenerative medicine, etc. The goal of further research should be the optimization of 3D scaffold efficacy by modulating cellular and molecular components and properties.

Author contributions

All authors participated in writing and reviewing the manuscript.