Stem cells are foundational to the success of regenerative medicine and relevant to a myriad of diseases. Their ability to differentiate into specialized cell types makes them invaluable for treating a wide range of diseases, including neurodegenerative disorders, cardiovascular diseases, and autoimmune conditions (Kolagar et al., 2020; Lerman and Lerman, 2021). Organoids, three-dimensional in vitro cultures that mimic the structure and function of native organs, represent a significant advancement in stem cell research for their ability to mimic cells in organs (Kol et al., 2014; Kretzschmar and Clevers, 2016; Zhao et al., 2022). These organoids, which can be derived from either adult stem cells (ASCs) or pluripotent stem cells, can be grown to resemble various organs with multicellular organization (Pain, 2021; Han et al., 2024; Pasca et al., 2024). This advancement provides a foundational platform for studying organ development, disease modeling, and drug testing, and for bridging the gap between traditional cell cultures and in vivo models (Jensen and Teng, 2020; Weng et al., 2023; Verstegen et al., 2025). They faithfully recapitulate structural and functional elements of the in vivo organ, allowing for studies of atypical cellular, molecular, and genetic features that underscore diseases (Lancaster and Huch, 2019; Andrews and Kriegstein, 2022; Chen et al., 2023; Silva-Pedrosa et al., 2023). Furthermore, organoids hold promise for cell replacement approaches to injury or disease (Wu et al., 2023) in humans and animals (Nakamura and Sato, 2018; Penning and van den Boom, 2023). While tissue repair is a well-known use of regenerative medicine approaches, the emergence of a microbiome in many internal tissues brings about a new challenge in understanding microbial interactions in various organs, especially in tissues that were once thought to be ‘sterile.’ The success of regenerative medicine strategies is intricately linked to understanding and managing host–pathogen interactions (Kol et al., 2014; Mohamad-Fauzi et al., 2023). In vitro models to study the association of the microbiome that adequately reflect the complexity of in vivo responses are lacking, thereby creating a gap in understanding how microbes impact various tissues that were once thought to be sterile (Yi et al., 2022; Hugon and Golos, 2023; Suarez Arbelaez et al., 2023). Infections can significantly impair tissue regeneration and compromise the function of stem cell-derived therapies (Mikulska et al., 2014), yet it is unclear what the role of the microbiome is related to therapeutic protocols. Therefore, a comprehensive understanding of how microbes interact with stem cells and organoids is crucial for developing safe and effective regenerative medicine interventions.

This review explores the evolving landscape of organoid technology, including a discussion on the importance of stem cell functions and current approaches to organoid development. This work also includes considerations of the pivotal role for cell culture models in advancing host–microbe interaction studies in low-biomass organ systems. By examining recent advancements, challenges, and future directions, we aim to provide insights into how stem cell and organoid models are reshaping our understanding of host–pathogen dynamics, disease progression, and therapeutic advancements.

Stem cells serve as the foundational cell type for organoid development, providing the regenerative capacity and cellular diversity required to model complex tissues and organ systems in vitro (Clevers et al., 2014). Their ability to self-renew and differentiate makes them indispensable for recreating the architecture and functionality of human organs. However, this intrinsic plasticity and susceptibility to environmental signals also make stem cells key targets for microbial interactions, including bacterial infections and microbiome associations to recapitulate in vivo assays (Kolb-Maurer et al., 2004; Hess and Rambukkana, 2015; Wizenty and Sigal, 2023). Unexpectedly, the association of common gut bacteria induces changes in the immune status (Hou et al., 2017) and differentiation trajectory (Mohamad-Fauzi et al., 2023) of stem cells without inducing apoptosis, suggesting that the microbiome may change the activity of stem cells in vivo and further highlights that microbes have direct access to stem cells in vivo. Understanding these host–pathogen interactions is critical, as pathogens can exploit stem cells to establish infections, alter their differentiation pathways, and potentially disrupt organoid integrity (Mohamad-Fauzi et al., 2023).

Carriage of bacteria in stem cells is of particular concern in regenerative medicine, which employs stem cell therapies as a treatment for a multitude of diseases, including digestive disorders, autoimmune liver disease, arthritis, and some cancers (Hoang et al., 2022). An emerging consideration is how microbiome members condition the developing immune system in early life via interactions with stem cells (Mishra et al., 2021). Insights into these interactions not only enhance the fidelity of organoid models for microbiome research but also provide a deeper understanding of how stem cells contribute to innate immunity and tissue resilience in both physiological and pathological conditions.

A diverse set of pathogenic, commensal, and opportunistic bacteria can interact with and influence stem cell function. The mechanisms of interaction include direct contact, such as bacterial adherence and invasion, and more indirect effects mediated by secreted toxins, metabolites, and signaling molecules. The consequences of these interactions are multidimensional; they can impair or enhance the regenerative capacity of stem cells, alter differentiation trajectories, and in some cases, contribute to disease progression, including chronic inflammation and tumorigenesis (Marrazzo et al., 2019; O’Rourke and Kempf, 2019; Pani, 2025). Understanding the spectrum of microbial influence on stem cells is essential for optimizing their clinical application and for developing strategies to mitigate microbial interference in stem cell–based therapies.

Mesenchymal stem cells (MSCs), somatic multipotent stromal cells, play an emerging role in regenerative medicine due to their ability to migrate to sites of injury, differentiate into multiple lineages, and secrete a wide range of immunomodulatory factors (Wang et al., 2014; O’Rourke and Kempf, 2019). However, these same properties render MSCs highly responsive to environmental cues, including microbial signals at mucosal interfaces such as the gut lumen (Marrazzo et al., 2019; Mohamad-Fauzi et al., 2023). These interactions with pathogenic and commensal microbes influence the regenerative and immunological behavior of MSCs, ultimately affecting the clinical outcomes of MSC-based therapies.

Microbial exposure has been shown to significantly influence MSC behavior. Multiple gastrointestinal bacteria have been shown to adhere and invade MSCs without impacting cell survival, including pathogenic Salmonella enterica serovar Typhimurium and probiotic Lactobacillus acidophilus, indicating the potential for bacterial carriage in stem cells (Kol et al., 2014). Though bacterial association did not affect stem cell survival, co-incubation of canine MSCs with these gastrointestinal bacteria altered MSC immunoregulatory profiles through induction of cytokine transcription, modification of surface markers such as CD54, and enhancement of prostaglandin E2 (PGE₂)–mediated suppression of T-cell proliferation (Kol et al., 2014). Additionally, S. Typhimurium association inhibited MSC migration, which is notable given migration is a key feature for their therapeutic application (Kol et al., 2014). Such findings demonstrate that bacterial interactions can manipulate MSC function through subtle regulatory shifts rather than overt cytotoxicity and apoptosis, with implications for both therapeutic efficacy and microbial persistence in host tissues.

The ability of microbes to influence stem cell activity is further emphasized in the context of chronic infections. Hematopoietic stem cells (HSCs) are among the most common cell types used in stem cell–based therapies yet are particularly influenced by chronic infection (Li et al., 2023). Chronic inflammation and chronic infection have both been shown to deplete HSC populations, in part through an increase in the terminal differentiation pathway (Matatall et al., 2016). Mycobacterium tuberculosis has also been shown to reprogram HSCs in the bone marrow via activation of the type I interferon (IFN-I) axis (Khan et al., 2020). This reprogramming disrupts iron acquisition and metabolism, impairs myelopoiesis, and ultimately hinders the immune response to M. tuberculosis infections (Khan et al., 2020). Notably, this hindered HSC activity persisted a year after initial infection, suggesting that microbial exposure can exert prolonged immunological consequences that extend beyond acute disease (Khan et al., 2020).

Microbial interactions have also been shown to disrupt differentiation pathways. Exposure of human and goat MSCs to S. Typhimurium interfered with trilineage differentiation, particularly inhibiting osteogenic and chondrogenic differentiation (Mohamad-Fauzi et al., 2023). Moreover, S. Typhimurium activated anti-apoptotic and pro-proliferative signaling in host cells, a function that favors pathogen survival while compromising host tissue regeneration (Mohamad-Fauzi et al., 2023).

In contrast with their susceptibility to microbial influence, MSCs also exhibit intrinsic antimicrobial activity, suggesting the potential for dual therapeutic roles. For mice infected with virulent Klebsiella pneumonia, the administration of placental MSCs (PMSCs) led to a significant reduction in bacterial load via localized recruitment of polymorphonuclear neutrophils (PMNs) and concurrent dampening of potentially harmful T-cell and natural killer (NK) cell responses (Wang et al., 2020). Additionally, MSC-derived antimicrobial peptides inhibited methicillin-resistant S. aureus (MRSA) biofilm development in chronic Staphylococcus aureus infections. When combined with antibiotics, this approach evoked synergetic activity toward bacterial clearance and improved wound healing (Chow et al., 2020). These findings underscore the complex, bidirectional nature of MSC–microbe interactions and position MSCs as both potential targets of microbial manipulation and active participants in host defense.

While stem cells hold significant clinical promise, their vulnerability to bacterial influence underscores a key gap in our understanding of microbiome–stem cell interactions across different organ systems. The complexity of these interactions necessitates model systems that can isolate and dissect these processes in controlled conditions, as traditional in vivo models are often too complex to resolve these interactions at a mechanistic level. In vitro work aimed at disentangling these dynamics, such as that using organoid models, can bridge the experimental gap and support translational findings.

Often described as “mini-organs in a dish,” organoids are defined by their three-dimensional (3D) multicellular structure and ability to self-organize, self-renew, and recapitulate key structural, functional, and molecular features of native tissues and organs (Clevers, 2016). They offer significant advantages over traditional 2D cell cultures, including long-term culture viability, enhanced cellular complexity, and preserved cell–cell and cell–matrix interactions (Kim et al., 2020). Organoids have been successfully generated for a wide range of mammalian species and multiple organs, including the skin, brain, liver, stomach, gastrointestinal tract, heart, pancreas, testis, endometrium, and placenta (Turco et al., 2018; Brooks et al., 2021; Puschhof et al., 2021b; Cho et al., 2022; Gnecco et al., 2023; Sreenivasamurthy et al., 2023; Tian et al., 2023; Stopel et al., 2024). The development of an organoid model includes deciding on multiple key components, including stem cell origin, scaffold, and media. These elements collectively influence the variability, heterogeneity, and functionality of the organoids, ultimately determining their suitability for specific applications.

Organoids can be derived from three cell sources: ASCs, embryonic stem cells (ESCs), or induced pluripotent stem cells (iPSCs) ( Figure 1 ) (Dutta et al., 2017; Ouchi et al., 2019; Pleguezuelos-Manzano et al., 2020) ( Figure 1A ). ASC-derived organoids, which are predominately epithelial and lack vasculature, closely resemble mature tissues and retain the genetic and functional characteristics of the donor. This makes them particularly suitable for disease modelling, personalized medicine approaches, and microbiome association studies with mature tissue structures (Yang et al., 2023). However, ASC-derived organoids lack the ability to produce diverse cell types in one model and are challenging to culture long term or at large scales (Yang et al., 2023). ESCs, derived from the inner cell mass of blastocysts, and iPSCs, generated by reprogramming of somatic cells, offer the advantage of producing organoids with greater cellular diversity, including mesenchymal, epithelial, and endothelial cells. Compared to ASC-derived organoids, which are typically unified in cell type, iPSCs can simultaneously differentiate into multiple cell types—such as spinal cord neurons and skeletal muscle cells—creating complex hybrid structures representative of in vivo tissue arrangements (Yang et al., 2023). While both the diversity and differentiation capacity enhance their value for tissue developmental studies, ESC- and iPSC-derived organoids often do not reach full maturity, failing to fully replicate the functionality of adult tissues with mature and differentiated cell types. However, organoids that contain a mixture of differentiated and non-differentiated stem cells may be a faithful model for in vivo inflammation, where epithelial degradation leads to the exposure of stem cells, as seen in gut inflammation and intestinal stem cells (Quan et al., 2025).

Beyond the selection of cell type, there are also multiple scaffolding materials and media components to choose from. For some cell sources, like iPSCs, cells are embedded in an extracellular matrix (ECM) that mimics the tissue-specific microenvironment. An ECM provides structural support and biochemical cues for organoid growth and maintenance, enabling the spherical structure inherent to organoids. Matrigel, derived from mouse Engelbreth-Holm-Swarm sarcoma, is a widely used basement membrane ECM, mainly composed of laminin, collagen IV, entactin, perlecan, and growth factors (Hughes et al., 2010). Despite its effectiveness, Matrigel has notable limitations, including its animal origin, batch-to-batch variability, and poorly defined composition, and the ability of bacteria to digest Matrigel complicates its use in host–microbe interaction studies (Hughes et al., 2010; Kozlowski et al., 2021; Kim et al., 2022). Alternatives to Matrigel are emerging, including natural hydrogels, synthetic hydrogels, and hybrid hydrogels (Kozlowski et al., 2021; Kim et al., 2022). Additionally, non-hydrogel systems such as silk microspheres and suspension cultures, where organoids are cultured in microwells designed and coated for low adhesion, have emerged (Matkovic Leko et al., 2023). Last, the cell culture media delivers essential signals that guide organoid differentiation, growth, and maintenance. Organoid culture media consist of basal media, antibiotics/antimycotics, and soluble factors that mimic the native tissue microenvironment ( Figure 1 ) (Pleguezuelos-Manzano et al., 2020). This combination of cell source, scaffold, and culture media together dictates the success of an organoid model and contributes to its replicability, making them crucial considerations during the development stage.

While organoids have transformed our ability to model tissue biology in vitro, their initial simplicity left gaps in physiological relevance. A critical leap in this field is the development of assembloids: advanced models that incorporate multiple cell types (e.g., stromal, vascular, and immune components), mimicking the cellular cross talk seen in native tissues (Kanton and Pasca, 2022). This integration enables studies of more complex tissue environments, from tumor-stroma interactions to the influence of bacterial interactions, providing insights that were previously inaccessible. The use of assembloids in host–microbe interactions allows precise experimentation; however, their geometry and orientation must be considered prior to use. In traditional organoid cultures, the cells’ apical surface faces inward (apical-in), forming a closed lumen (Puschhof et al., 2021b). While ideal for some studies, this orientation obstructs direct access to the epithelial surface, a challenge for research on host–pathogen interactions, where the apical surface is the primary site of contact. To address this, polarity reversal techniques have been developed, flipping the apical surface outward (apical-out) (Co et al., 2019). This innovation eliminates the need for micromanipulation techniques, allowing organoids to be directly exposed to infectious agents or therapeutic compounds in the biologically relevant orientation.

Organoids are a promising platform, but their utility is not without limitations. Reproducibility remains a significant challenge, with variability stemming from both inter- and intra-organoid heterogeneity, as well as temporal changes during development and across passages. Current protocols lack standardization, with batch-to-batch variation often exacerbated by the reliance on inconsistent animal-derived matrices (Kozlowski et al., 2021; Kim et al., 2022). Another shortcoming lies in the absence of vascularization and immune components in many current organoid models. Without blood vessels, nutrient and oxygen delivery becomes diffusion-limited, leading to within-organoid heterogeneity, especially in larger organoids (Grebenyuk et al., 2023; Tan et al., 2024). Similarly, the lack of immune cells within the organoid diminishes the ability to study immune responses or inflammation-driven diseases (Jeong and Kang, 2023).

Thorough validation of the organoids through characterization of the structural, molecular, and functional level is also an essential step to validate their relevance, reproducibility, and suitability for specific applications ( Figure 1B ). To better reflect in vivo conditions, organoid models also need to capture contributions of the local microbiome, but to date the incorporation of a microbiome in cell culture models is not standard practice and remains difficult to do (Allen and Sears, 2019; Zhong et al., 2023). Though organoids are not a research panacea and still have many shortcomings, they present a promising method for reducing the need for animal models and for increasing the repeatability of study findings. By leveraging the benefits of stem cell and organoid models, researchers can discern facets of pathogen behavior, track host cell responses, and better understand disease progression with unprecedented precision.

One area where early stem cell models and complex organoids have proven particularly valuable is the study of host–microbe interactions in low-biomass situations—often intricate and dynamic relationships with profound implications for health and disease. Commensal microbes are increasingly linked to improved immune function, conditioning the immune system with exposure to bone marrow, the promotion of systemic health, and resistance to infections. Contrastingly, pathogenic microbes remain a global health challenge, driving outbreaks and undermining regenerative cell therapies, underscoring the urgent need for effective treatment strategies. Understanding these dual dynamics is paramount, and organoids have emerged as a critical platform for this purpose. Because organoids provide a controlled and largely reproducible in vitro environment, these models enable precise examination of microbial interactions and host responses. Additionally, organoids allow for time series experiments, which provide valuable insights that are largely unattainable in vivo.

Importantly, organoid-based studies enable the modelling of both high-biomass systems, characterized by abundant microbial populations, and typically understudied low-biomass systems, where microbial presence is sparse and thus challenging to model ( Figure 2 ). While organoid models have proven particularly useful for modeling host–microbe interactions in microbially dense organs, the scope of this review is limited to body locales presumed to have lower microbial density and those that are understudied (Puschhof et al., 2021a; Xiang et al., 2024).

Many connections between microbes and diseases remain merely correlative, largely due to the challenges of unraveling the multitude of changes during host–microbe interactions in a reductionist yet meaningful manner. This highlights the need for advanced in vitro model systems to provide mechanistic insights into microbial impacts on the host tissue that have translational importance for therapeutic value. Despite their limitations, animal models and conventional 2D cell cultures are fundamental tools in deciphering host–pathogen mechanisms during bacterial infections. Organoid models present a valuable alternative to animal models, which demand significant resources such as funding, labor, and housing facilities. In addition to reducing costs, organoids offer new insights into disease development mechanisms and the impact of virulence factors on host epithelium by closely replicating in vivo tissue environments. Furthermore, organoids provide ample opportunity to collect material for various downstream analyses, including that for metagenomics, metabolomics, and visualization ( Figure 1 ), that can be used to provide new insights into mechanistic causes of infection that may serve as new therapeutic targets.

Notable progress has been made in understanding interactions in certain host contexts, such as gut microbes and gastrointestinal organoids derived from human (Chakrabarti et al., 2021; Puschhof et al., 2021b) and animal tissues (Beaumont et al., 2021). Nevertheless, studies focusing on low-biomass tissues remain relatively underexplored. The section below highlights organoids representing low-biomass tissue and details the work of their use in understanding microbial interactions in these contexts ( Table 1 ).

Research on co-infections involving organoids derived from low-biomass tissues with pathogenic and/or commensal bacteria has expanded our understanding of microbial-host interactions. Though an important area of research, the number of studies implementing organoids from low-biomass tissues remains limited, leaving additional questions for the emerging area of low-biomass microbiome association. Existing work primarily focuses on human- and mouse-derived organoids, while in vitro models from livestock and companion animals have been largely restricted to investigating intestinal host–microbe interactions (Kawasaki et al., 2022). Moreover, to the best of our knowledge, no studies have examined the interactions between commensal microbiota and organoids derived from low-biomass associations from tissues that are beyond the GIT. Further research is essential to uncover the dynamics and therapeutic implications of such co-infections within these organoid models.

The convergence of regenerative medicine and microbiology highlights the critical need to understand the intricate interactions between bacteria and stem cells, a symbiosis that can either enhance or hinder therapeutic outcomes. MSCs, pivotal players in regenerative therapies, are known to interact with microbes present in the body, which can significantly alter their immunomodulatory and regenerative capacities (Kol et al., 2014). Similarly, organoid research has propelled biomedical sciences into a new era, deepening our insights into life, disease, and potential therapies (Wang et al., 2017; Yi et al., 2021). While the initial studies provide hope that organoid models are useful for host/microbiome interaction studies, the reliability and maturity of organoid technology need to be improved for more consistency and expanded to prove their ultimate usefulness.

The existing paradigm linking stem cell properties to regenerative medicine is evolving, necessitating a deeper understanding of the multifaceted role of MSCs as niche cells and tissue organizers (Bianco et al., 2013). Understanding the interplay between microbiome and stem cells is pivotal, as it may hold the key to unlocking more efficacious and predictable regenerative therapies, especially when coupled with advancements in stem cell technologies, tissue engineering, and biomaterials. Understanding these interactions is particularly crucial in the context of tissue engineering, where the presence of bacteria can dramatically influence the success of regenerative approaches (Mhashilkar and Atala, 2012). The complexity of these interactions necessitates thoughtful modeling approaches and careful consideration of in vivo recapitulation to understand multifaceted microbial impacts with precision.

This precision offered by organoids also presents an exciting future for host–microbe interaction studies. Well-characterized organoid platforms provide an unprecedented opportunity to evaluate emerging host–microbiome interaction theories. For instance, the oral microbe Fusobacterium nucleatum has been found in colorectal tumors from patients, and an increasing number of studies suggest F. nucleatum is a driver of colorectal tumor development (Wang and Fang, 2023; Zepeda-Rivera et al., 2024). Though associations between F. nucleatum and cancer have been posited, the mechanistic nature of this host–microbe relationship remains understudied in part due to the notable complexity of studying cancer progression in vivo. Organoids, which minimize confounding factors while recapitulating many in vivo responses, present as a model platform for further studies of oncogenic bacteria. Compounding their usefulness, hybrid organoids provide the additional opportunity to study theorized bacterial-mediated organ axes, such as the gut–brain (Carabotti et al., 2015), mouth–brain (Bowland and Weyrich, 2022), and mouth–urinary connections (Yuan et al., 2021). These organ–organ–microbe connections remain difficult to study in animal models, while other in vitro platforms lack the ability to faithfully model multiple systems simultaneously. Organoids thus serve as a middle-ground in vitro model, reducing complexity to allow for mechanistic observation while replicating many in vivo responses. Further characterization and validation of organoid models will continue to be necessary, especially as more complex organoid experiments are published, but the current utility of organoids in host–microbe interaction work is a promising start.