Gastrointestinal tract, its pathophysiology and in-vitro models: A “quick” reference guide to translational studies

Abstract

The gastrointestinal (GI) tract is essential for digestion, absorption, excretion, and protection, supported by a diverse microbial ecosystem. Traditional in-vitro models often fall short in capturing the physiological complexity of the GI tract, limiting their translational applications. A comprehensive approach is needed to bridge the gap between simple cell cultures and more complex systems used in translational research. This review explores the limitations of conventional two-dimensional cell cultures and emphasizes the emerging use of three-dimensional and microfluidic systems that better replicate the GI tract’s structure and functions. It highlights the importance of incorporating patient-derived cells and engineered microenvironments to enhance model relevance and support personalized medicine. The review also discusses advanced fabrication techniques such as micro-extrusion and laser-assisted bioprinting, which enable the creation of sophisticated tissue models capable of simulating critical GI processes, including molecular transport, peristalsis, and liver coupling. Advancing the complexity of in-vitro systems will help replicate the GI tract’s interactions and physiological phenomena, thus improving the translational potential of GI research. This review provides valuable insights into the advancements and challenges in GI modeling, serving as a comprehensive guide for developing models that bridge the gap between basic cell cultures and clinically relevant systems.

Key Words: Cell line; Tumour cell line; In vitro techniques; Gastrointestinal tract; Organoids; Organ-on-chip

Core Tip: This review highlights the need for advanced in-vitro models to better replicate the gastrointestinal (GI) tract’s complexity for translational research. It compares the evolution from conventional 2-dimensional cultures toward 3-dimensional and microfluidic systems. Key advancements include the use of patient-derived cells, engineered microenvironments, and bioprinting techniques like micro-extrusion and laser-assisted printing. These technologies enable modeling of essential processes such as peristalsis, molecular transport, and liver coupling. By enhancing physiological relevance, these models support personalized medicine and improve the predictive power of preclinical GI research.

INTRODUCTION

Despite significant advances in gastrointestinal (GI) research, existing in-vitro models often fail to recapitulate the full complexity of the GI tract. Functions of the digestive system include digestion, absorption, excretion, and protection. In addition to these functions, which are maintained by the respective organs individually, as well as in concert, there is a growing body of evidence that supports the importance of the microbiome, composed of a collection of bacteria, fungi, viruses and archaea. These produce a diverse ecosystem of about 10¹⁴ microorganisms[1]. Studying such a diverse and complicated system requires multidisciplinary approaches. A potential approach to studying these systems in a controlled environment are in-vitro as well as in vivo models[2]. The benefit of these models is the fact that they adhere to the “3Rs” principles, defined by Russell and Burch in 1959[3]. These 3Rs are: “Replace” animals used in experiments with non-sentient alternatives; “Reduce” the number of animals employed; and “Refine” animal experiments so that they cause minimum pain and distress[3,4]. This review aims to provide a comprehensive reference guide that bridges the gap between simple cell cultures and complex, dynamic systems used in translational studies[1,5-7]. It systematically details the anatomy and pathophysiology of specific digestive tract diseases, especially those with an inflammatory etiology, providing clear rationale for the necessity of complex model systems. Furthermore, it highlights the wide range of applications for these models and explores emerging trends in culturing techniques that better replicate the human digestive tract environment.

Following this introduction, we detail the GI tract and its pathologies. This will serve as a discussion basis. Following that we continue with in-vitro models, compare their performance, and discuss their translational relevance.

THE GI TRACT AND ITS PATHOLOGY

Oral cavity and esophagus

The digestive tract is estimated to be 8 m-9 m long and is composed of different interconnected organs (Figure 1). Malignancies in the oral cavity are, regarding the statistics, classified under the terms and codes for “lip, oral cavity (C00-C06)”. Squamous cell carcinoma (SCC) is the predominant malignancy that occurs in the oral cavity. Minor salivary gland cancers and sarcomas are less common. The incidence of this type of cancer is approximately 389.485 and accounted for 188.230 deaths in 2022[8]. The presence of human papillomavirus (HPV) is an established prognostic biomarker of favourable outcome in locally advanced oropharyngeal cancers[9]. The prevalence of HPV infection in oropharyngeal sites of head and neck cancer (e.g., the hypopharynx, larynx, and oral cavity) is less in comparison. Nevertheless, the infection apparently presents a positive prognostic marker[10].

Figure 1 Representation of the digestive system and the oesophagus. The digestive system is composed of various organs which are playing important roles in the digestion, absorption, excretion, and protection of the human body. The individual organs differ in their composition and corresponding diseases. The structure of the oesophagus is displayed as a cross section. The depiction presents multiple layers, which are important to know due to implications of physiological function as well as staging of malignant diseases (infiltration depth) and its corresponding treatment (i.e., mucosectomy or fill thickness resection); ¹Salivary glands.

The adult oesophagus is an 18- to 25-cm-long muscular tube (Figure 1). In contrast to other portions of the GI tract the oesophagus does not have a serosal covering. The most common types of cancers of the oesophagus are SCCs and adenocarcinomas (AC)[11,12]. Neuroendocrine neoplasms are very rare, with a predominance of neuroendocrine carcinomas vs neuroendocrine tumors. Another entity are undifferentiated carcinomas, which lack any of the features of the previously described cancers (squamous, glandular, neuroendocrine) and are also very rare[13,14]. Oesophageal cancer (globally) is with 510.716 new cases in 2022 the eleventh most common cancer in the world. The mortality rate (445.129) rate is indicative of the fact that the disease is difficult to treat[8]. For a long time, SCC was the prevalent subtype of cancer, however, the incidence of AC has increased drastically[11]. The highest reported rates are in countries that are part of the oesophageal cancer belt. The other common term is also “Asian Oesophageal Cancer Belt” denoting its location. It encompasses areas such as Turkey, Iran, Kazakhstan and northern and central China[15].

The stomach and small intestine

The stomach has been formerly only known as a hollow muscular structure, today, it is regarded as one of the most complex endocrine organs of the human body[16,17] (Figure 2A). The variety of hormones that are in connection with the stomach are: Gastrin, somatostatin, ghrelin and regulatory peptides are produced by cells within the stomach itself. There are also other hormones that are secreted in the more distal portions of the GI tract but also regulate gastric function[16,17].

Figure 2 The stomach, its structure and pathogenesis of Helicobacter pylori disease. A: Comparison between gastric mucosa in the antrum and in the corpus; B: The pathogenesis of Helicobacter pylori infection-Helicobacter pylori reaches the stomach epithelium by swimming through mucus and neutralizing acid with urease. The bacterium adheres to epithelial cells and disrupts the epithelial barrier, leading to apoptosis and tissue damage. It induces interleukin (IL)-8 and other chemokines, which recruit neutrophils and macrophages. Cytokines like IL-1β, tumor necrosis factor-α, and IL-12 promote inflammation. Adaptive immunity is also engaged: T helper cells (Th1/Th2) and B cells produce antibodies (IgG, IgA), though often insufficient to clear the infection. The result is chronic inflammation, immune dysregulation, and cytokine-driven changes in gastric physiology. This may contribute to gastritis, ulcer formation, and even gastric cancer. H. pylori: Helicobacter pylori; PMN: Polymorphonuclear leukocyte; IL: Interleukin; TNF: Tumor necrosis factor; IFN: Interferon; NO: Nitric oxide.

Chronic infection (Figure 2B) with Helicobacter pylori (H. pylori) is recognized as a key factor in the development of several gastric disorders, including gastric cancer, and remains a significant global health issue due to its widespread occurrence[18,19]. After entering the stomach, H. pylori adheres to the gastric epithelium, particularly at cell junctions, establishing a protected niche that allows it to survive in the otherwise hostile acidic environment[20,21]. Once established, the bacteria can persist for life, often without causing immediate symptoms, while gradually triggering epithelial injury, inflammation, and immune activation. In many cases, the infection remains clinically silent for years before progressing to more severe outcomes such as peptic ulcers or gastric malignancies[22]. Despite extensive research, many aspects of the pathogenesis remain unclear especially the early cellular interactions and the processes driving the shift from chronic infection to cancer development.

Understanding the underlying mechanisms remains challenging, largely because commonly used model systems do not accurately mimic the complex environment of the human stomach[23]. Standard in-vitro approaches often lack key physiological features such as dynamic potential of hydrogen conditions, mucus secretion, and the full spectrum of differentiated gastric epithelial cell types. The antral region of the stomach, where H. pylori infections typically establish, features a highly specialized glandular epithelium composed of multiple cell types, including stem and early progenitor cells at the gland base, MUC6-expressing neck cells, PGC-producing chief cells, and MUC5AC-secreting pit cells toward the lumen[24]. Although parietal cells that secrete acid are present in the antrum, they appear in lower numbers compared to the gastric corpus[24]. Additionally, the antral glands include specialized cells such as various enteroendocrine cells marked by chromogranin A and Tuft cells[24]. These cell types are spatially organized, with stem and progenitor populations situated deeper in the glands alongside neck and chief cells, while the pit cells form the surface layer of the epithelium[23]. Bartfeld et al[25] have by establishing long-term, 3-dimensional culture of organoids from mouse tissues (intestine, stomach, pancreas, and liver) and human intestine and pancreas, provided a tool to generate many of those cell types and thus a potential avenue for exploring mechanisms underlying H. pylori-induced diseases. Boccellato et al[26] described gastric mucosoid cultures that reproduce the features of normal human gastric epithelium, which could enable new approaches for investigating the pathophysiology of H. pylori.

Hofer et al[27] just recently presented and patterned a homeostatic human gastric organ-on-a-chip (OoC) system with bilateral access. They claim that is capable of modeling H. pylori niche establishment and persistent colonization of the gastric epithelium[27]. The authors stated that under physiologically relevant acidic conditions at the apical surface, the OoC system supported the development of more mature gastric pit cells compared to conventional organoid cultures[27]. The differentiated pit cells displayed, after exposure to H. pylori, a distinct response that sets them apart from other epithelial cell types an aspect that had not been previously described. The authors concluded that the model could prove to be powerful tool for broader investigations into gastric epithelial dynamics, mucosal immune responses, and host-microbe interactions[27].

Diseases of the small intestine (Figure 3A and B) that are noteworthy in the context of in-vitro cell culturing are specifically coeliac disease (CeD)[28] and Crohn’s disease (Figure 3C)[29-31].

Figure 3 The small and large intestine, their anatomy and diseases overview of intestinal structure and immune responses in coeliac disease and inflammatory bowel disease. A: Anatomical overview of the small and large intestine; B: Cross-sectional structure of the small intestine, highlighting key layers (left part): Mucosa, submucosa, muscle layer, and serosa, which play roles in barrier function and immune signaling. Peyer’s patches in the small intestine are important sites for immune activation (right part). In coeliac disease, gluten peptides are incompletely digested and deamidated by tissue transglutaminase (TG2), forming neoantigens that are presented to cluster of differentiation 4 + T cells by human leucocyte antigen-DQ2/8 molecules. This leads to T cell activation, B cell maturation, and production of anti-gluten and anti-TG2 antibodies, contributing to tissue damage in the small intestine (left); C: Cross-sectional structure of the large intestine, highlighting key layers: Mucosa, submucosa, muscle layer, and serosa, which play roles in barrier function and immune signaling (middle panel). Under normal conditions, a balanced immune environment in the intestinal mucosa is maintained by regulatory immune cells and intact epithelial and mucus barriers (rightmost panel). In inflammatory bowel disease, disruption of the mucus layer and epithelial barrier allows bacterial translocation, triggering immune dysregulation, recruitment of effector immune cells, and release of proinflammatory cytokines, leading to chronic inflammation.

CeD is an autoimmune enteropathy, 1%-2% of the general population, triggered by gluten in genetically susceptible individuals risk haplotypes [human leucocyte antigen (HLA)-DQ2.5, -DQ2.2, -DQ8, and -DQ8.5][32,33], with HLA-DQ2.5 carrying the strongest risk association[34]. The immune pathophysiology is complex. It involves innate and adaptive immune responses as well as the intestinal epithelium, which may interact with environmental risk factors (e.g., gut microbiome, luminal antigens, viral infections etc.)[34]. The disease leads to atrophy of the small intestinal mucosa and consequently to malabsorption[32]. The main histological features include villous atrophy, crypt hyperplasia and intraepithelial lymphocytosis. These findings are summarized in the modified MARSH criteria, which are used for histological diagnosis[35]. Ingested gluten is partially digested into peptides, including immunogenic fragments like gliadin [particularly deamidated by tissue transglutaminase (TG2)]. Deamidated gliadin peptides activate cluster of differentiation (CD) 4 + T cells, which produce pro-inflammatory cytokines [e.g., interferon (IFN)-γ, interleukin (IL)-21], leading to villous atrophy[36,37]. Gliadin also activates innate immune pathways, including IL-15 production, which promotes intraepithelial lymphocyte proliferation and cytotoxicity[38]. The key pathophysiological mechanisms are HLA-DQ2/8-mediated antigen presentation, the IL-15 pathway and the TG2-gluten complex. The only current treatment is strict lifelong adherence to a gluten-free diet. A lack of representative human models for CeD has hindered our understanding of its pathophysiology and the development of novel treatments[39].

To date, there is no model system that fully recapitulates the complexity of CeD[40]. Current in-vitro models include immortalized cell lines (CLs) and mucosal biopsies. The immune system has been investigated using CLs of monocytes, such as THP-1, or intestinally derived T cells[41,42]. Existing data on epithelial barrier function are largely based on the CLs Caco2, T84, and HT-29[43-45]. Immortalized CLs do not represent the genetics of CeD and have poor genomic integrity[40]. Patient-derived intestinal biopsy material does contain the CeD-associated genetic background and directly reflects the disease phenotype, but is scarce because of its invasive nature. Another option are murine models. In terms of completeness they present a living model with working inter-organ communication[46]. However, the model requires thorough understanding of induction of disease and is due to interspecies differences in physiology, pharmacology and cellular processes difficult to extrapolate to humans.

OoC technology may solve many of these drawbacks[47]. A recent paper[48] showcased the potential of an induced pluripotent stem cells (iPSCs)-derived small intestine-on-chip with a self-organised tight epithelial layer, including villus-like structures and a cell type composition that resembles the human small intestine. Just recently, an in-depth review on human organoids and OoC in CeD has been published[39].

Although significant progress has been made in managing inflammatory bowel disease (IBD)[49], a definitive cure remains out of reach. This is largely due to an incomplete understanding of the disease’s complex origins and biological mechanisms[31,50]. Over recent decades, a range of experimental models spanning in-vitro, in vivo, and ex vivo systems has been introduced to help close these knowledge gaps[31,40,46,51].

Crohn’s disease and ulcerative colitis (UC) are two main forms of IBD, but they differ significantly[52]. Crohn’s can impact any segment of the GI tract in a patchy, discontinuous pattern, often involving all layers of the gut wall[53]. UC, in contrast, is confined to the colon and rectum and affects only the inner lining in a continuous manner[54].

Current thinking emphasizes an abnormal immune response to intestinal microbes[55], likely triggered by environmental factors in genetically susceptible individuals[56]. In IBD, several key defense mechanisms in the gut are impaired. These include weakened tight junctions, changes in the mucus layer, and microbial imbalances[52]. Dysfunctional goblet and Paneth cells contribute to this, as they produce less protective mucus and antimicrobial substances, respectively. Defects in autophagy-related genes such as NOD2 and ATG16 L1 are also associated with increased disease risk[52].

The weakened barrier allows gut microbes to invade the intestinal wall, prompting immune cells like macrophages and dendritic cells to release pro-inflammatory cytokines [e.g., tumor necrosis factor (TNF)-α, IL-6, IL-23], which attract more immune cells and fuel ongoing inflammation[57-59]. This results in increased gut permeability and a self-sustaining inflammatory cycle[59]. A key feature is the imbalance between regulatory T cells (Tregs) and inflammatory Th17 cells, which contributes to immune overactivation[52].

Furthermore, mitochondrial dysfunction has emerged as a factor in IBD pathology[60]. Genes linked to mitochondrial stability (e.g., MDR1, HNF4A) are disrupted, leading to oxidative stress and impaired energy production in intestinal cells[52,61]. This mitochondrial damage affects cell renewal and further weakens the gut barrier. Such findings suggest that targeting mitochondrial health could be a promising strategy for restoring intestinal function and controlling inflammation in IBD[62].

There are various three-dimensional (3D) intestinal inflammation models that can at least partially, recapitulate IBD features. These include models based on scaffolds or hydrogels and those based on decellularized tissue models, as well as more complex intestine-on-a-chip systems and organoids.

Some of these include: Leonard et al[63] developed a 3D co-culture model by embedding human blood monocyte-derived macrophages and DCs in a collagen matrix on a semi-permeable Transwell^® filter insert, with Caco-2 cells seeded on top. This model was subjected to different types of proinflammatory stimuli [lipopolysaccharides (LPS) from Escherichia coli and Salmonella typhimurium, IL-1β, IFN-γ], IL-1β presented the strongest induction of inflammation[63].

Later on, the model was further improved by replacing the primary immune cells with a macrophage-derived CLs and dendritic-like cells (MUTZ-3), in an effort to enhance reproducibility and facilitate a more comprehensive assessment of cytotoxicity[64].

Another study described a triple co-culture intestinal model consisting of an intestinal epithelial layer (Caco-2/HT29-MTX cells) and immunocompetent cells. The main goal was to represent a healthy intestine characterized by a stable intestinal barrier and to evaluate the efficacy of anti-inflammatory drugs[65].

Le et al[66] described a complex in-vitro triple-culture model aiming to develop an inflammation-triggered in-vitro leaky gut model using Caco-2/HT29-MTX-E12 combined with macrophage-like THP-1 cells or primary human-derived macrophages.

The large intestine

The large intestine is approximately 1.5 m long and its primary functions include desiccation, compaction of waste and storage in the sigmoid colon and rectum (Figure 3A). The colon is inhabited by a multitude of different bacteria, which produce vitamins (vitamin K and B), other metabolic by-products [e.g., short-chain fatty acids (SCFAs)] and help in regulating other important organic systems (e.g., gut-brain axis).

The most common disorders in the colon are different types of inflammations (“colitis”), ischaemic changes, disturbances in the microbiota and neoplasms[1,67,68]. According to the newest statistics, more than 1.9 million new colorectal cancer (CRC) (including anus) cases and 903.859 deaths were estimated to occur in 2022[8]. Overall, CRC ranks third in terms of incidence and second in terms of mortality. Broadly speaking, CRC is responsible for about one in 10 cancer cases and deaths and is the most common type of GI cancer[8]. Most of the tumours of the colon and rectum are carcinomas. Other tumor types (neuroendocrine neoplasms, hamartomas, mesenchymal tumours, lymphomas) are relatively rare. Of the carcinomas, more than 90% are AC[69]. It has become apparent that CRC is a heterogeneous tumor with a variety of risk factors[70], different histological subtypes[7], molecular alterations[71], and a complex tumor immune microenvironment[72]. The pathobiological basis for carcinomas can be nowadays explained by multiple mechanisms, which mostly revolve around progression and subsequent accumulation of specific mutations that lead to the development of cancer[68]. There are currently two better known pathways that explain the formation of CRC. The first, and best known, is the adenoma-carcinoma sequence, which was proposed by Vogelstein et al[73] in 1988. This sequence involves the stepwise transformation of normal mucosa to a benign precursor (adenoma), which in turn undergoes subsequent mutations that disrupt mechanisms regulating epithelial renewal and transforms into carcinoma[73,74]. The second, is the serrated neoplasia pathway. It plays an important role in about 30% of all CRC cases. Precursor lesions are sessile serrated lesions and traditionally serrated adenomas. The initiating mutations for dysplastic changes are BRAF or KRAS mutations. Other distinctive critical events are methylation of various CpG islands, MLH1 promoter methylation, TP53 mutations and alterations in the WNT signalling pathways[75,76]. Nevertheless, apart from the above traditional carcinogenesis models, non-conventional pathways exist with complex means of genetic progression[77] comprising a plethora of different driver genes[78], significant plasticity in intra-tumoral gene expression[79] and epigenetic reprogramming[80].

What is more, the role of the microbiota has gained in importance even in oncological therapy[81-83]. Emerging evidence suggests that probiotics, particularly Bifidobacterium, play a pivotal role in gut microbiota modulation, which possess both intestinal protective and anti-cancer properties. It has been reported that probiotic strains such as Bifidobacterium longum subsp. infantis, Lactobacillus rhamnosus, Lactobacillus acidophilus and Lactobacillus casei can help in the treatment of CRC through cancer cell immunomodulation and chemoprotective effects[84]. Ramesh et al[81] demonstrated strong binding affinity for Bifidobacterium and toll-like receptor 2 thereby suggesting the activation of anti-inflammatory pathways. Reportedly, this interaction enhances IL-10 production while reducing pro-inflammatory cytokines (e.g., IL-6 and TNF-α). These processes improve gut homeostasis and mitigate chronic inflammation, a key driver of CRC progression. The authors conclude that future research should focus on personalized probiotics and validating their synergy with chemotherapy.

Following this chapter, which was dedicated to the structural properties of the GI tract and their malignancies, we now explore the inhabitants, their impact on the physiology, individual cells and their translational value in modelling.

THE GUT, ITS INHABITANTS AND ITS EFFECTS ON NORMAL AND PATHOLOGICAL GI TRACT FUNCTION

The GI tract gives home to a microbiome, composed of a collection of bacteria, fungi, viruses and archaea. These produce a diverse ecosystem of about 10¹⁴ microorganisms[1]. The term dysbiosis presents a change in the composition of the gut microbiota[1,85-87]. Over the last decade, knowledge about the relationship between dysbiosis and the pathogenesis of various diseases (especially cardiovascular disease) has rapidly accumulated[88-90]. It comes as no surprise that some of these potential diseases include cardiovascular disease, chronic kidney disease, type 2 diabetes mellitus, non-alcoholic fatty liver disease, and even certain types of cancer[85,91-94] (Figure 4). On the pathophysiological level it has been proposed that the (gut dysbiosis-derived) LPS and SCFAs, secondary bile acids, amines methylamines, trimethylamine N-oxide are of major importance in that regard.

Figure 4 Depiction of the effects of dysbiosis on the human body. This diagram illustrates the interconnected relationship between gut microbiota and various organ systems through distinct gut-organ axes. Central to this interaction is dysbiosis, an imbalance in gut microbiota, which contributes to multiple diseases. It is evident that changes of the microbiota have a systemic impact beyond the digestive system, demonstrating its role in health and disease through multiple organ interactions. Microbial metabolites act as an inflammatory stimulus which elicits a generalized inflammation. Furthermore, these metabolites (lipopolysaccharides, fatty acids, bile acids etc.) have to be digested finally arrive to the liver. This elicits further inflammation via activation of Kupffer cells.

A summary of the major mechanisms

A reduced expression of tight junction proteins (e.g., zonula occludens-1, claudin-1 and occluding) and an imbalance between epithelial cell death and regeneration can lead to a leaky-gut[85,95,96]. What follows is the translocation of bacteria, which stimulate, via the recognition of their pathogen associated molecular patterns, an immune response and general inflammatory reaction [secretion of pro-inflammatory cytokines (like IL-18, IL-1, IL-6, and TNF-α)]. This affects the whole organism (e.g., damaging the integrity of the blood brain barrier).

Gut dysbiosis increases systemic inflammation (via LPS, IL-6, TNF-α), contributing to endothelial dysfunction, hypertension, and heart failure. LPS can act as a modulator of toll-like receptors. The upregulation of these proteins has been associated with an inflammatory activation which in turn promoted the process of atherosclerosis[87,97].

Microbial metabolites (e.g., SCFAs, tryptophan metabolites) and neuroactive compounds (e.g., gamma-aminobutyric acid, serotonin, dopamine) modulate brain function and behavior. Dysbiosis can also lead to an altered synthesis of neurotransmitters (5-hydroxytryptamine, dopamine, noradrenaline, and glutamate). This consequently leads to deregulated microbiota-gut-brain signaling[98,99].

The portal vein transports gut-derived bacterial metabolites and endotoxins directly to the liver. Dysbiosis and leaky gut lead to endotoxemia and hepatic inflammation via activation of Kupffer cells, which in turn release pro-inflammatory mediators, such as TNF-α, ILs (IL-1 and IL-10), lysosomal enzymes (protease and phosphatase)[100,101]. It has been shown that there are distinctive gut-liver axis disruption patterns in the prevalent chronic liver diseases, adrenoleukodystrophy and metabolic dysfunction-associated steatotic liver disease[102].

Gut microbiota shape systemic and pulmonary immunity through SCFAs and Treg induction. Gut dysbiosis increases pro-inflammatory cytokines and compromises lung mucosal immunity[103].

Microbiota regulate systemic and cutaneous inflammation, influence Treg/Th17 balance, and affect skin barrier integrity. SCFAs and tryptophan-derived metabolites promote anti-inflammatory pathways. Dysbiosis leads to skin flares via immune dysregulation and altered lipid metabolism. This has been linked to diseases such as psoriasis, atopic dermatitis and acne[104].

As an illustration, it has been shown that there is a potential fourfold increase in obesity risk within 15 years of emigrating to the United States, compared to populations remaining in their birth country. This fact is accompanied with a decrease in their gut microbial diversity and function[86,105].

Furthermore, environmental factors (e.g. diet, household cohabitation) greatly outweigh heritable genetic contributions to the composition and function of gut microbiota[106]. Rothschild et al[107] showed with their microbiome-association index, mimicking heritability statistics, that most significant associations were between the gut microbiome and host phenotypes for body mass index, waist-to-hip ratio, fasting glucose levels, glycemic status, high-density lipoprotein cholesterol levels, and monthly lactose consumption[106,107].

Antibiotic use for microbiota manipulation remains debatable due to potential side-effects (Figure 5). It represents an aggressive approach with drawbacks such as reduced bacterial diversity, altered gene expression, and selection for resistant bacteria[108,109], earning antibiotics the label of “deep modulators” of gut microbiota[109,110]. Studies link antibiotics to various outcomes, including effects on obesity, insulin resistance, diabetes, and myocardial infarction[111-113]. Faecal microbiota transplantation has already therapeutically confirmed the importance of a healthy gut microbiota in certain patients. This form of treatment is several decades old and still presents an important intervention[114].

Figure 5 Schematic representations of the effects of antibiotics on gut microbiota. This figure highlights the profound effects of antibiotics on gut microbiota composition and diversity. It emphasizes that antibiotic use disrupts the gut microbiome, potentially leading to long-term health consequences such as inflammation, metabolic disorders, and antibiotic resistance. In detail, most antibiotics reduce total bacterial diversity, often leading to microbial imbalances (dysbiosis). Several antibiotic classes, including macrolides, beta-lactams, and glycopeptides, promote an increase in Proteobacteria, which is associated with gut inflammation and dysbiosis. Many antibiotics, such as macrolides, beta-lactams, and glycopeptides, significantly reduce Firmicutes and Actinobacteria, which play a crucial role in gut homeostasis. While some antibiotics, like carbapenems and lincosamides, reduce Bacteroidetes diversity, others like macrolides promote their growth. While carbapenems decrease Enterobacteriaceae, they increase Lactobacilli, showing a mixed impact on gut flora.

Regarding the potential for modelling some of the gut axes, significant advances have been made. Trapecar et al[115] utilized interconnected human microphysiological systems of the gut, liver, and circulating Treg and Th17 cells to model UC ex vivo. The research revealed that microbiome-derived SCFAs can either alleviate or exacerbate UC severity, depending on the involvement of effector CD4 T cells. Their findings highlight the complex role of SCFAs in modulating inflammation within the gut-liver axis[115].

Also, in another study Trapecar et al[116] developed a human multi-organ OoC microphysiological system integrating the gut, liver, and brain to investigate how microbial metabolites influence neurodegenerative diseases. The study demonstrated that microbial metabolites, particularly SCFAs, can modulate neuroinflammation and neuronal health. The platform provides insights into the gut-liver-brain axis and its role in the pathogenesis of neurodegenerative conditions, offering a novel approach to study complex inter-organ interactions.

Furthermore, Zhang et al[117] presented a protocol for co-culturing primary human colon epithelial cells with human gut bacteria under controlled oxygen conditions, including anaerobic environments. The method enables studying host-microbe interactions in a physiologically relevant setting, facilitating research into how gut bacteria influence colon health and disease. Finally, Zhang et al[118] developed in 2024 an immune-competent human gut microphysiological system to examine the anti-inflammatory effects of the commensal bacterium Faecalibacterium prausnitzii. The findings indicate that Faecalibacterium prausnitzii can modulate inflammation within the human gut environment, highlighting its potential therapeutic role in treating IBD[118].

Collectively, these studies utilize advanced human microphysiological systems to explore complex interactions between the gut microbiota, host tissues, and the immune system, providing valuable insights into inflammatory and neurodegenerative diseases and show the great potential of these methods. These studies directly link us to our next section on the specifics of cell culturing.

CELL CULTURES AND MODELLING

Key elements of the intestinal microenvironment are the biochemical interactions, the cells, which constitute the 3D architecture, flow dynamics and motility. Traditionally used two-dimensional (2D) immortalised (tumour) CLs survive long-term but are not genetically stable nor represent any human in particular. In contrast, primary cultures are patient-unique, but short-lived. Some other approaches, which have shown promise, include organoids[119-122] (spheroids and tumoroids)[123], multi-well systems[119,124], as well as microphysiological OoC models[125]. Their hierarchy, regarding complexity and usage applicability, can be seen in Figure 6. 3D organoid cultures resemble the crypt-villus domain and contain all cell lineages, are long-lived and genetically stable. Unfortunately, manipulation of the 3D organoid system is more challenging[126-128]. The applications of these models are manifold (e.g., functional test with drug and toxicity testing, tissue engineering and simulation of absorption, metabolism etc.)[2]. A comparison of the individual methods in relation to modelling human disease can be seen in Figure 7.

Figure 6 Representation of the milestones and different techniques of tissue culturing through time. This diagram illustrates the evolution of intestine tissue culture and engineering platforms leading to the development of intestine-on-a-chip technology for personalized medicine. Pre-clinical background (top panel) showcases a progression from two-dimensional monocultures, three-dimensional co-cultures, bioreactors, three-dimensional printing, organoids, and tissue samples to intestine-on-a-chip and animal models, demonstrating a shift from high-throughput methods to increased physiological relevance. The bottom panel shows the intestine-on-a-chip, which integrates the microbiome, organoids, and extracellular matrix within a controlled flow channel, enabling intestine disease modelling. This facilitates therapy development, ultimately leading to personalized treatments tailored to individual patients. 2D: Two-dimensional; 3D: Three-dimensional.

Figure 7 Overview of different in-vitro models to mimic diseases. Four widely used in-vitro platforms (two-dimensional cell cultures, three-dimensional cell cultures, organoids, and organ-on-a-chip systems) are being compared based on their structural characteristics, vessel types, experimental duration, measurement techniques, and automation potential. The key features of each system are presented and the gradient bar at the bottom indicates the trade-off between throughput (highest in two-dimensional cultures) and physiological relevance (highest in organ-on-a-chip models). These in-vitro platforms vary in their ability to replicate tissue-specific microenvironments and are selected based on the desired biological complexity, readout needs, and disease modeling objectives. 2D: Two-dimensional; 3D: Three-dimensional.

2D in-vitro models

Many in-vitro intestinal models exist in the form of simple 2D systems, which rely on culturing an intestinal epithelial cell monolayer (e.g., Caco-2 cells) or co-cultured mixtures of intestinal cells on static micro-porous Transwell supports[129-131]. For modelling different disease states, immune cells and microbes can be added to the apical and basolateral side of the culture well, respectively[132].

However, to date, 2D in-vitro models have not readily provided an intestinal model that accurately recapitulates the architecture, segment specificity, paracrine and autocrine molecular signalling of healthy tissue. Additionally, various drug transporters are either mis-localized or have inaccurate expression in transformed cells when compared to the human intestine[133].

The biggest drawback, however, is the lack of cellular diversity in a single cell type system. Normal intestinal epithelium consists of several cell types including enterocytes, goblet cells, stem cells, enteroendocrine cells, Tuft cells, M cells and Paneth cells that are not accurately represented in 2D models. Intestinal cells cultured in Transwell plates also often fail to produce mucus (e.g., Caco-2 cells) or undergo differentiation of intestinal villi that become injured; therefore, these models do not effectively recapitulate the physiology and pathophysiology of the native tissue. Translatability is also a concern in terms of species differences as well as cancerous vs non-cancerous. They also lack the ability to culture bacterial community dynamics and are only able to study single bacterium-host interactions[134]. Some widely used CLs in this field are Caco-2, HT-29, T84, HuTu-80 etc. (e.g., IEC-6, IEC-18, IPEC-J2, IPEC-1).

Overall, despite the drawbacks of 2D culturing, this method has significantly impacted our understanding of spatially organized structures, with cell identities resembling those found in tissues in vivo[135].

Caco-2 was developed in the 1970’s[136]. It has been widely studied in various domains (e.g., pharmacological, nutritional, microbiological). Caco-2 cells are the most widely used cell model to study the permeability of drugs over the last 20 years[128,137]. The CLs is of human origin originated from a 72-year-old Caucasian male with colorectal adenocarcinoma[136]. It has 23 derived CLs (“children”), its doubling time is from 51.41 hours[138] to multiple days and is microsatellite stable (MSS)[139]. It can either function as undifferentiated large intestinal cells or can spontaneously differentiate to resemble a small intestine-like phenotype with enterocyte-like absorptive properties. While challenging to control and reproduce, this CL is valuable for studying transport kinetics and can simulate both small and large intestines[140]. A multitude of articles (> 200) have utilized the Caco-2 model to explore bacteria, bacterial metabolites, intestinal barrier function, bacterial adhesion/invasion, and innate immune responses[140]. The model can also be, to further improve certain aspects of transport mechanisms, combined with a co-culture of HT-29 cells. The model apparently mimics both enterocytes and goblet cells, and it was reported to generate more predictable experimental results, due to the role of mucus on drug transport[141].

The HT-29 CL was isolated from a 44-year-old Caucasian female with colorectal adenocarcinoma in the 1970s’ by Fogh and Trempe[142]. The CL has a doubling time from 24 hours to 60 hours, is MSS and has 69 derived CLs[143]. It was initially used for cancer biology research but later transitioned to other studies due to its versatile phenotype. HT-29 cells can be cultured in an undifferentiated state or differentiated to form a polarized membrane, providing flexibility for various research objectives. It has been shown that these cells produce cytokines, including ILs and TNF-α, making them valuable for nutrition and host-microbiome interaction investigations[144,145]. Using a Transwell system with apical and basolateral polarity, HT-29 cells are suitable for studying bacterial adhesion and transport. What sets them apart from Caco-2 is the presence of mucus-producing goblet cells, making them particularly useful for immune function and bacterial-host interaction studies[146,147]. According to literature, mucins expressed in this CL include both secretory (MUC2, MUC5AC) as well as membrane bound (MUC1, MUC3, MUC4) types[148].

The T84 CL was established from a xenograft produced by subcutaneous injection of tumour cells into BALB/c nude mice[149]. The tumour cells were derived from a lung metastasis of a colorectal adenocarcinoma in a 72-year-old Caucasian male. The CL has a doubling time of 33.90 hours, is MSS and exhibits brush border properties, including enzymes and transporters, making them a valuable model for studying epithelial barrier function and electrolyte transport[149].

The HuTu 80 CL was isolated from a 53-year-old male of allegedly, Caucasian ethnicity, but exome analysis finds it to be mostly of African lineage, with a duodenal adenocarcinoma. It has a doubling time of approximately 26 hours, is MSS and has one subline. The literature on its origin is conflicting, since it is listed as colon or small intestine[138,150]. Th CL was included in multiple studies trying to understand the molecular landscape of intestinal cancers[150] and the landscape of pharmacogenomic interactions in cancer[151].

Some non-neoplastic CLs that are worth mentioning are JFCF-6 (derived from the small intestine, jejunum; also called jejunal fibroblast cystic fibrosis-6; 14 derived CLs), H-4 (enterocytes from a fetal small intestine; spontaneously immortalized CL)[152] and HUIEC[153]. HUIEC is a stable, fast-growing untransformed, adult human epithelial intestinal CL.

3D in-vitro models and spheroids

An intestinal organ culture was first described in 1969[154]. This method utilized biopsy tissue and a traditional culture-dish system but was limited in the amount of time tissue could be cultured[140]. In 2009, a significant breakthrough occurred with the development of intestinal organoids[155]. Furthermore, in 2014 a reproducible method to direct the differentiation of LGR5 + stem cells to become a specific cell type was published (e.g., enterocytes, goblet cells, stem cells, enteroendocrine cells)[156]. GI organoids can be derived from various portions of the GI tract, including the oesophagus[157], stomach[23], pancreas[158], small intestine[159,160], and colon[161,162].

3D organotypic cultures, derived from primary tissue, embryonic stem cells, or iPSCs[163], exhibit self-renewal, self-organization, and mimic the functionality of the original tissue[164]. They hold great promise for various applications, from basic research to translational uses like disease modelling, drug testing, and host-microbe interactions[165]. Moreover, induced Human Intestinal Organoids can be created from iPSCs. Organoids are mostly grown in 3D, however, they can also be seeded and grown as monolayers that allow easier access to the apical side and studies towards the intestinal barrier function[166].

Noel et al[167] used a macrophage-enteroid co-culture model to investigate mucosal gut physiology. They found out that macrophages enhanced barrier function and maturity of enteroid monolayers by measuring an increased transepithelial electrical resistance and cell height. Bardenbacher et al[168] wanted to investigate the intestinal barrier breakdown in small mouse intestinal organoids to evaluate the impact of IFN-γ induced disruption of tight junction proteins and determined the impact of INF-γ on the intestinal barrier integrity. Some applications of organoid in disease models are cystic fibrosis[169], IBD[170], mechanisms in CRC[171], tissue repair[172] etc. Organoids are also used for studies examining host-microbe interactions[173].

3D bio-printed models

To confront the above-mentioned limitations of traditional 2D and 3D in-vitro models, biomimetic systems (OoC or microfluidic devices), which are based on recent advances in microfabrication, microchip technology, microfluidics, and tissue engineering approaches, have emerged.

It is well established that the 3D physical environment plays an important role in the morphology development, biochemistry, and metabolism of cells[174,175] and has been shown specifically for the small intestine as well[176,177]. Existing microfluidic platforms are focusing on introducing physical cues such as fluid sheer stress, flow rate and direction, shear stress and chemical gradients that are important at both the tissue and organ level[178]. Often, well-characterized intestinal transformed CLs are used to examine changes in cellular microenvironment, polarization, spatial organization, and/or differentiation of the cultures better correlate with those found in vivo.

There are examples of 3D models of the intestine: IPSC-derived intestinal organoid cultures[177,179,180] and ex vivo intestinal enteroid cultures derived from either a single LGR5 + stem cell/Paneth cell unit or multi-cell human intestinal crypts[155] obtained from human intestinal tissue. The iPSC-derived cultures have a more immature, fetal tissue phenotype making them valuable to study developmental biology. These ex vivo organoids self-organize into a 3D “mini-intestine” and contain all intestinal epithelial cell types that are found in vivo, and can be expanded for years without genetic instabilities[155]. Epithelial culture systems in the form of organoids and enteroids recapitulate the central features of normal intestinal epithelial architecture and function including cell polarization, the presence of the brush border with microvilli and appropriate physiologic responsiveness in-vitro. They are often cultured in Matrigel because it resembles the extracellular matrix composition of the native basement membrane. The extracellular matrix has a crucial role for epithelial cells as it helps them to maintain their apical to basal cell polarity when cultured both in Matrigel or on Transwell inserts[162]. As mentioned, organoids produce higher levels of intestinal differentiation; however, the cells in organoids do not experience physiological peristalsis-like motions and cannot be cultured with a living microbiome under such static conditions because bacterial overgrowth will result in death of the epithelium[181]. This is a critical limitation because mechanical deformations resulting from peristalsis influence both normal epithelial cell differentiation[182] and restrain microbial overgrowth in vivo and in-vitro[183].

OoC

Creating effective OoC models poses the challenge of reproducing physiologically relevant biology akin to complex in vivo tissues. OoC is a device where biology is coupled with microtechnology. The chip takes the form of a microfluidic device containing networks of hair-fine microchannels for guiding and manipulating minute volumes (picolitres up to millilitres) of solution[125]. The so-called organs present miniaturized tissues in microfluidic chips, which can replicate specific tissue functions, proving to be simplified yet effective models for human physiology and disease. As of the hierarchy of cultures, 2D cell cultures are regarding the simulation of complex physiological processes the least relevant, followed by 3D cell cultures, organoids, and OoCs in increasing order.

The typical workflow of such a system is as follows. Intestinal epithelial cells are introduced into the device and allowed to adhere briefly[153]. Subsequently, a media solution is pumped through channels, either using syringes or peristaltic pumps to establish fluid flow[184-186]. Alternatively, gravity can drive flow in platforms like multi-well plates, which require incubation on a rocking platform[187,188]. This flow induces the spontaneous differentiation of intestinal cells, leading to the formation of 3D villi-like structures[189]. These villi structures are covered with brush borders and mucus, connected by tight junctions, closely mimicking the structure and function of human intestinal villi. In some gut-on-chip models, peristaltic motions can be generated by applying a vacuum to hollow side chambers[184,190]. Additionally, advanced chip devices allow for the inclusion of other cells (e.g., endothelial cells on the basolateral side, immune cells[191,192].

Interestingly, when exposing Caco-2 cells embedded in a gut-on-a-chip microfluidic device to cyclic mechanical distortion that mimics peristalsis-like motions of the living intestine and sheer flow analogous to that in the gut lumen, they can (given enough time extended culture periods) potentially reorganize into 3D intestinal villi lined by columnar epithelial cells[184,189]. These villi closely mimic the native architecture[193] and exhibit multiple differentiated features, including reestablishment of basal proliferative cell crypts, differentiation of different small intestine cells (absorptive, mucus-secretory, enteroendocrine, and Paneth), production of mucus, and creation of a highly resistance epithelial barrier. Several other attempts have been made to study epithelial intestinal cells in a 3D environment, by utilizing different microfabrication techniques to provide supports for cells that contain grooves, ridges, crypts and villi[194-196]. Numerous studies also reported microfluidic models to culture intestinal epithelial cells. The so-called rotating wall vessels enabled prolonged culture of both mammalian cells and bacterial populations[197-199]. These devices were designed to produce laminar flow facilitating growth of intestinal organoids in suspension culture in conjunction with bacteria to simulate an enteric infection in a fluidic setting[197,200]. Microanatomy of the GI tract, especially the villi and crypts further affect intestinal fluid dynamics, pressure and wall stiffness[201]. Recent studies have shown that recreating the micro-topography of the small intestine with similarly shaped villi, Caco-2 monolayers exhibit differentiation morphology along the crypt-villus axis and showed different transepithelial resistances and permeability for small molecules than in a 2D model[193,202].

These systems often use transformed CLs raising concerns as to whether the findings have biological relevance. The co-development of microfluidic devices with integrated stem cell-derived organoids and enteroids will increase the complexity of current systems. Additionally, the ability to disperse 3D cultures and then grow them as monolayers, which appear to be identical in composition and physiology to their 3D counterparts, on extracellular matrices represents an important advance by providing separate access to apical and basolateral surfaces of the epithelium[203]. However, even advanced microfluidic models have some limitations that need to be addressed. Current devices do not recapitulate the full geometrical and physiological complexity of the intestine and to date, there has not yet been a realistic 3D model available to study primary intestinal behavior directly in-vitro with an implemented vascular system[204].

In 2017 Alimperti et al[205] developed a biomimetic vascular model on a chip, creating a perfusable channel with an endothelial lining in a collagen matrix to evaluate mural cell-endothelial cell regulated barrier function and the impact of inflammatory factors, N-cadherin expression etc. They were able to provide a model for drug testing and diseases.

Zhang et al[206], 2016 developed a biodegradable scaffold with build in vasculature for OoC research and surgical anastomosis. Vascular networks can be built from natural hydrogels, which are, however, soft and only provide temporary support. Synthetic biodegradable polymers provide better support, however have low permeability. AngioChip is a stable, biodegradable scaffold which accommodates both criteria, using poly [octamethylene maleate (anhydride) citrate] and a 3D stamping technique.

OoC devices can also include microbes and mimic the gut microbiome, as shown is chapter 3, Trapecar et al[115,116], Zhang et al[117,118] and host interactions[207,208]. Some examples are a co-culture of Caco-2 cells with Lactobacillus rhamnosus[184], enterohemorrhagic Escherichia coli[209], or obligate anaerobes[210]. What is more, when looking at disease models, there are a number of articles on studying IBD[211-214], virus pathogenesis and infections[215-217], metabolic diseases[218] etc., Yoon et al[213] cultured IBD patient cells to validate the synergistic actions of peptides and hydrogels used to treat IBD, Khan et al[214] propose synthetic- and engineered community-based microbiota transplants as a potential therapy approach.

However, as depicted in Figure 7, some of the major challenges in the development and application of advanced in-vitro models, such as organoids and OoC platforms, compared to simpler in-vitro culture methods include scaling up production, integrating multiple organ systems, and ensuring long-term physiological stability. These limitations currently hinder their broad translational application[125,219].

CONCLUSION

To address the number, or rather the lack of, CLs. A very useful resource in the field of CL research is the Cellosaurus database[143]. The authors have, to gain an overview of the topic at hand, looked through the database by searching for “small intestine”. In November 2023 there were 124 hits for this search term. Now, in January 2025 there are 135 hits. From the old record 2 CL were removed (HT-29 since it is a colon adenocarcinoma; and CVCL A5HM since it is also of colonic origin) and 13 new ones were included (6 of human origin from the same patient-iPSCs; the remaining 7 were of animal origin). These consist of 63 human (46.7% of 135) and 7 animal CLs (53.3% of 135). From these 63 human CLs: 3 CLs are Hutu-80 derivatives, one CL (JFCF-6) is the parent of 14 CLs, 1 CL has been discontinued, 3 CLs are reported to be contaminated, 4 CLs possibly misidentified. Furthermore, from these 63 CLs, 17 CLs are either metastases or do not belong in the category “small intestine”. By excluding metastases, derived/children CLs, induced pluripotent CLs this leaves us with merely 18 CLs of which 6 are benign and 12 malignant. As a comparison, a Cellosaurus search for “Triple-negative breast cancer CL” on 20 September 2021 resulted in 144 entries[220]. On November 2023 the search yielded 163 hits with only 2 being of animal origin (November 2023) and now (January 2025) the search yields 204 hits with only 4 being of animal origin. This shows that the number and variety of available CLs in this field is profoundly lacking. Regarding the fabrication methods, it is highly desirable these can use a broad spectrum of available materials that facilitate fast printing of several mm large structures while keeping sub-micron resolution to create nano-patterns and localised mechanical improvements. Joining different bioprinting techniques, such as micro-extrusion and laser-assisted lithography for simultaneous use, could be a major stepping stone in this direction and enabling the fabrication of complex tissue models such as a vascularised interstitium underlying a parenchymal epithelium. The ultimate goal of these techniques will therefore be the construction of advanced tissue models for simulations of molecular transport and other physiological phenomena occurring on epithelia lined on vascularised connective tissues, focusing on the gut. To successfully fabricate tissues of such complexity, micro-extrusion bioprinting for constructing several mm large scaffolds will be combined with laser-assisted bioprinting for local nano-patterning and structural modification. This will allow the fabrication of a branched microfluidic system in a biopolymeric matrix (vascularised connective tissue), which will induce polarisation on an attached epithelial layer of enterocytes. The intestinal epithelium boasts a wide range of gradients, offering significant potential when combining primary intestinal culture systems with chemical and matrix gradients. These gradients can be integrated into surfaces, solutions, scaffolds, or generated through co-cultured cells, including bacteria or stroma. In addition to the 3D environment, it is important to consider several other critical features when developing a complex physiologically relevant in-vitro model. The epithelium must be supported by matrices that mimic the lamina propria and the muscular layers that form a concentric tube along with the epithelium. Further, it has to contain a heterogeneous population of cells that is spatially organized to mimic the crypt-villus axis. The luminal unidirectional shear stress on epithelium’s apical and blood flow on the basolateral surface has to be simulated. The intestine also periodically squeezes by the muscular layers (peristalsis). More complex models need to consider the unique coupling of the intestine to the liver. Specifically, there is basolateral outflow from the intestinal epithelium to the liver through the portal system, whereas liver products are secreted into the lumen of the duodenum acting on the apical side of the epithelium. Some key challenges and dilemmas that remain are ethical considerations surrounding the sourcing and further use of these models. Namely, older informed consents do not sufficiently address many of the issues that include chimeric research, gene-editing technologies etc.[221]. Furthermore, there is the urgent need for improved standardization and rigorous validation of in-vitro models to ensure reproducibility and reliability[221-223]. Establishing standardized protocols for organoid culture, differentiation, and functional assessment will facilitate broader adoption and regulatory acceptance. Moreover, the integration of more sophisticated co-culture systems that incorporate immune cells, enteric neurons, and other relevant cell types is crucial to better recapitulate tissue complexity and physiological interactions[224]. Such multi-cellular models can provide deeper insights into tissue homeostasis, inflammation, and disease pathogenesis. A next-generation approach would integrate OoC platforms with embedded biosensors, such as impedance spectroscopy or metabolite-sensitive fluorescent probes, to enable real-time monitoring of both microbial metabolite dynamics and epithelial barrier function. This type of dynamic model could offer valuable insights into the bidirectional communication between host tissues and the microbiome. For example, it could reveal how SCFAs influence tight junction regulation and, conversely, how epithelial responses shape microbial composition. In parallel, advancements in real-time imaging techniques are necessary to monitor cellular behavior and tissue function dynamically within these complex models[225-227]. Such technologies will enable visualization of cellular interactions, differentiation, and responses to stimuli without disrupting the system[227]. These innovations will enhance our understanding of organoid biology and improve the predictive power of these models for translational applications. In case of patient specific cultures, these methods would enable patient specific treatments and each patient, thereby tailoring treatment choices and improving response rates. Finally, as shown, there is a variety of systems from which a researcher can choose. The choice depends heavily on its application. This review highlights that while traditional 2D cultures provide foundational insights, emerging 3D and microfluidic systems better replicate the physiological complexity of the GI tract. Future research should focus on hybrid models that integrate patient-derived cells with engineered microenvironments to enhance both translational relevance and personalized medicine applications.