Table 3.

Advantages and limitations of different in vitro intestinal models for evaluating microbiota-directed therapeutics.

In Vitro Cellular Model	Description	Example(s) and References	Advantages	Limitations
Immortalized cell lines	Cell cultures that have been modified to divide indefinitely	Caco-2 cells (undergoes spontaneous enterocytic differentiation to adopt morphological and functional similarities to a human intestine) 126	Easy model to work with and maintain Inexpensive and widely available for use Well-characterized for many cell types High throughput capacity No need for overly complex or specialized equipment	Do not fully mimic the behavior and function of primary cells isolated directly from the issue of interest Misses many factors that are present in vivo (ex: intestinal mucus layer, interactions between various cell types, immune functions) Only host cells and host response is observed Does not model the complex interactions between different cell types found in vivo
Co-culture of cell lines	Culture of a combination of cell lines in the same culture dish or system	Co-culture of Caco-2 cells, HT29-MTX, and Raji B cells, adding mucus production and immune response elements to generate a complex morphological and functional cell culture model of the human intestine127	Allows for the study of cell-cell interactions Efficient and straight forward model to test host responses More closely mimics the in vivo microenvironment (can incorporate mucus layer, immune functions) than single immortalized cell culture leading to more biologically relevant results No need for overly complex or specialized equipment High throughput capacity	More technically challenging and time-consuming than using immortalized cell lines Only host cells and host response is observed Does not fully reproduce the complex microenvironment found in vivo
3D cell culture	Culture of cells in a 3D environment, can be achieved using various scaffolds or gels	Caco-2 cultured with electrospun scaffolds, hydrogels, 3D-printed scaffolds, or subepithelial-like tissue constructs containing fibroblasts 128–130	More closely mimics the in vivo microenvironment than standard monolayer culture Higher morphological and physiological model accuracy Improved epithelial cell polarization Includes effects of extracellular matrix molecules More representative paracellular permeability	Low throughput compared to monolayer cell culture More technically challenging and more costly than monolayer cell culture Only host cells and host response is observed Does not fully reproduce the complex microenvironment found in vivo
Microfluidics-based co-culture systems	Co-culture of differentiated human epithelial cells with facultative anaerobic or anaerobic bacterium under both aerobic or anaerobic conditions	HuMiX,131 MIMICS132	Provides the ability to study host-microbe interactions Allows for precise control of the microenvironment, including the composition, flow of media and oxygen levels Scalable and automatable for high-throughput	Can be costly to set up and maintain May not be suitable to all cell or bacteria types More technically challenging to set-up than regular monolayer cell culture Does not fully reproduce the complex interactions between different cell types found in vivo
3D cell culture in microfluidics co-culture system	Morphological features of an intestine, such as villi and crypt structures, which can be colonized by living bacteria and allows the study of host-microbiome interactions in an immunocompetent environment from a readily accessible immortalized cell line	Organ-on-chip133	Higher morphological and physiological model accuracy Provides the ability to study host-microbe interactions Allows for precise control of the microenvironment, including the composition, flow of media and oxygen levels	Can be costly to set up and maintain May not be suitable to all cell or bacteria types More technically challenging to set-up than regular monolayer cell culture Does not fully reproduce the complex interactions between different cell types found in vivo
Organoids	Stem cell derived, self-organizing, 3D tissue-like structures, which closely mimic a tissue of interest	iPSC- or patient-derived intestinal organoids134–137, iHACS (Intestinal Hemi-Anaerobic Co-culture System)138	More closely mimic the microenvironment and cell-cell interactions found in vivo Higher morphological and physiological model accuracy Can be used to study the development and function of organs over time, as well as to model diseases and test potential therapies Provides the ability to study host-microbe interactions	Time-consuming and costly to culture and expand organoids More technically challenging to set-up experiment and the need of micro-injection to study host-microbe interactions Difficult to scale up for use in high-throughput studies