Hydrogel-based experimental models of the gastrointestinal tract

Abstract

The gut microbiome plays a pivotal role in human health, yet its complexity has long eluded detailed study under physiologically relevant conditions. Hydrogel-based models are revolutionizing microbiome research by bridging the gap between traditional in vitro systems and the complexity of in vivo environments. These advanced systems replicate key physical and biochemical features of the gastrointestinal tract, offering unprecedented opportunities to study microbial behavior, adaptation, and interactions within three-dimensional, tunable architectures. Unlike suspension cultures, hydrogels provide porous, mucosa-like environments that enable the cultivation of mucosa-associated microbes, co-culturing with human cells, and mimicking healthy and disease-related states. This review explores the transformative potential of hydrogel matrices in unveiling the spatial organization, nutrient gradients, and community communication that define microbial ecosystems. By integrating the benefits of in vitro and in vivo models, hydrogel-based platforms promise to accelerate discoveries in microbiome science, with far-reaching implications for understanding human health and developing targeted therapeutics.

Download video file ^{(44.1MB, mp4)}

Video Abstract

Supplementary Information

The online version contains supplementary material available at 10.1186/s40168-025-02208-5.

Introduction

The gut microbiome exists within a highly dynamic and complex environment. Key factors such as fluctuating pH levels, fluid dynamics, nutrient availability, and inter-microbial interactions collectively shape the diverse ecosystems within the gastrointestinal (GI) tract [1, 2]. These parameters can shift significantly in disease conditions, challenging microbial stability and requiring adaptive strategies from resident microbes. Despite growing interest in the microbiome’s role in health and disease, our understanding of the native biophysical and biochemical environments that shape microbial function remains limited.

Meta-omics approaches, such as genomics, transcriptomics, and metabolomics, have significantly advanced our understanding of microbial composition and potential function. However, these methods typically overlook the local microenvironments in which microbes reside, including key gradients and mechanical constraints. This omission risks missing key insights into microbial physiology, community dynamics, and disease mechanisms.

A growing body of evidence links alterations in the gut microbiome to various diseases, including, among many others, Parkinson’s disease, inflammatory bowel disease, and multiple sclerosis [3–7]. Notably, Parkinson’s disease has been associated with an increased abundance of mucosa-associated bacteria, such as Akkermansia muciniphila, which may interact with host pathways to influence disease progression [3, 8]. However, the behavior of these bacteria within their native environment remains underexplored.

Microbial interactions with their surrounding environment, particularly in terms of spatial organization, metabolism, and physiology, differ substantially when observed in vitro versus in vivo [9–15]. Localized gradients in pH, oxygen, and nutrients create highly structured niches that strongly influence microbial behavior and evolution [16–20]. Replicating these microenvironments in experimental systems remains a key challenge for microbiome science.

This recognition has driven growing interest in hydrogel-based models designed to mimic aspects of the gut biochemical and physical landscape [21, 22]. Hydrogels, cross-linked polymeric networks of polymer compounds that retain water as their primary phase, offer tunable properties such as porosity, swelling behavior, mechanical strength, and surface chemistry [23]. These characteristics have made hydrogel-based scaffolds indispensable in fieldsranging from tissue engineering and drug delivery [24–27] to biosensing [28, 29] and microbiota modulation [30, 31]. Crucially, hydrogels provide a controlled, customizable medium for microbial culture under physiologically relevant conditions. They enable spatial structuring, co-culture, diffusion gradients, and, in some cases, long-term microbial growth. While they cannot yet fully replicate the compositional and rheological complexity of the GI tract, an ongoing challenge for the field, their modularity and biocompatibility make them powerful tools for interrogating microbial adaptation and interaction in defined contexts, bridging the gap between in vitro and in vivo (Fig. 1).

Fig. 1.

Hydrogel-based models bridge the gap between in vitro systems and complex GI environments. This schematic compares two features of traditional suspension cultures (right), physiological GI environments (left), and hydrogel-based in vitro models (center). General characteristics of each system are summarized at the bottom, indicating whether features are captured in the respective in vitro approaches. GI = gastrointestinal

Created in BioRender. Sieders, M. (2025) https://BioRender.com/5e0872x

This review focuses on hydrogel-based systems designed to study gut-associated microbes in contexts that approximate their native environment. While we highlight integrated platforms (e.g., microfluidic systems) where appropriate, our primary emphasis is on the hydrogel matrices themselves: their composition, modification strategies, and experimental utility. Our aim is to provide a multidisciplinary resource for microbiologists, bioengineers, and materials scientists interested in leveraging hydrogels to bridge the gap between traditional in vitro cultures and the spatial complexity of the gut. We begin by describing the physicochemical and structural features of the gut in health and disease that hydrogels aim to recapitulate (“Introduction” section). We then review current hydrogel applications for investigating microbial behavior, co-culture, screening, and delivery (“Physiochemical landscapes of the gut” section). Finally, we examine how specific hydrogel materials and their functionalization strategies enable modeling of particular GI microenvironments and microbial responses (“Hydrogel matrices for microbial cultivation” section). By evaluating the advantages and limitations of hydrogel-based systems, we aim to support their informed and innovative use in microbiome research.

Physiochemical landscapes of the gut

The GI tract hosts a diverse array of microbial niches, shaped by intertwined axial and radial gradients in rheology, fluid dynamics, chemical composition, pH, water content, and microbial community structure (Table 1) [13, 57]. These gradients influence microbial colonization and activity throughout the lumen and mucosa, while being further modulated by factors such as diet and disease. As such, physiochemical conditions vary not only between anatomical sites but also between healthy and diseased individuals, posing a major challenge to in vitro modeling. Axially, the tract transitions from acidic and enzyme-rich proximal regions to more buffered, fermentation-active distal environments. Radially, oxygen, mucin structure, pH, and microbial density all shift dramatically between the epithelium and the lumen [56, 58–61]. Mucus, an endogenous hydrogel, provides both a protective barrier (inner layer) and a specialized microbial niche (outer layer). It consists primarily of water and large glycoproteins (notably MUC2), along with proteins, lipids, and DNA (Table 2) [21, 77–79]. These components determine mucus viscosity and diffusion behavior, which also change along the gut, being thinner and more elastic proximally, and thicker and more porous distally (Table 2, Fig. 2) [66, 72]. Although the inner mucus layer is typically considered sterile, recent investigations do show the ability of some microbes to penetrate the inner mucus and reach the epithelium. Disease or inflammation may additionally permit penetration by mucolytic or invasive bacteria [17, 18, 80, 81]. Rheological and gel properties can vary severely within and between populations, together with other biophysical and chemical parameters found in GI mucus [21, 62, 71, 78, 82].

Table 1.

Physiochemical composition along the healthy human GI tract in fasting state

				GI region
				Stomach	Small intestine						Cecum	Colon					Rectum
Factor		Study	Metric		Duodenum	Proximal (jejunum)	Traverse			Distal (ileum)		Proximal			Traverse	Distal
pH		[32]	μ1/2 (IQR)	1.7 (1.4–2.1)	6.1 (5.8–6.5)		**			**	**	**			**	**	**
		[33]	μ ± σ	**	6.6 ± 0.5		7.4 ± 0.4			7.5 ± 0.5		6.4 ± 0.6			6.6 ± 0.8	7.0 ± 0.7
		[34]	μ1/2 (IQR)	1.5 (1.2–1.9)	6.4 (5.7–6.9)	6.6 (6.1–7.1)	7.00 (6.6–7.4)			7.3 (7.0–7.5)	5.6 (5.1–6.3)	5.6 (5.0–6.4)			5.7 (5.3–6.4)	6.6 (6.2–7.0)	6.6 (5.8–7.1)
		[35]	μ1/2 (IQR)	1.3 (1.1–1.6)	6.5 (6.2–6.7)		**			**	**	**			**	**	**
		[36]	μ1/2 (IQR)	2.0 (1.8–3.1)	7.0 (6.8–7.2)						6.1 (5.9–6.3)	6.5 (6.3–6.6)					**
		[37]	μ1/2 (IQR)	**	5.4 (4.8–6.3)	5.8 (5.3–6.7)	7.3 (5.9–7.6)			6.9 (5.8–7.5)	**	**			**	**	**
		[38]	μ1/2 (IQR)	2.5 (1.8–5.0)	6.2 (4.6–6.9)	**	**			**	**	**			**	**	**
Buffer capacity (mmol l−1 ΔpH−1)		[39]	μ ± σ	**	**	**	**			**	**	21.4 ± 7.9			**	**	**
		[40]	μ ± σ	14.3 ± 9.5	**	**	**			**	**	**			**	**	**
		[41]	μ ± σ	**	**	3.2 ± 1.3	**			22.9 ± 17.4	**	32.1 ± 4.9			44.4		**
		[42]	μ1/2 (range)	**	**	**	**			8.9 (3.6)	21.5 (7.9)	**			**	**	**
		[43]	μ	**	**	2.8	**			**	**	**			**	**	**
		[44]	range	**	**	4–13	**			**	**	**			**	**	**
		[45]	μ ± σ	**	**	**	**			**	**	45 ± 17			**	**	**
		[38]	μ1/2 (IQR)	7.7 (2.7–16.1)	5.8 (3.5–9.7)	**	**			**	**	**			**	**	**
Surface tension (mN m−1)		[39]	μ	**	**	**	**			**	**	39.2			**	**	**
		[46]	μ1/2 (IQR)	38.5 (37–41.3)	**	**	**			**	**	**			**	**	**
		[40]	μ ± σ	34.8 ± 5.2	**	**	**			**	**	**			**	**	**
		[47]	μ ± σ	33.6 ± 5.9	**	33.7 ± 2.8	**			**	**	**			**	**	**
		[43]	μ ± σ	**	**	28 ± 1.0	**			**	**	**			**	**	**
		[38]	μ1/2 (IQR)	44.9 (40.7–47.2)	32.6 (30.9–34.7)	**	**			**	**	**			**	**	**
Compositional profile	Protein (mg mL−1)	[48]	μ ± σ	**	**	**	**			**	**	12.0 ± 9.4			**	**	**
		[42]	μ ± σ	**	**	**	**			5.1 ± 3.3	10.2 ± 2.2	**			**	**	**
		[39]	μ ± σ	**	**	**	**			**	**	9.7 ± 4.6			**	**	**
		[40]	μ ± σ	4.9 ± 1.0	**	**	**			**	**	**			**	**	**
		[49]	μ ± σ	1.8 ± 0.7	**	2.1 ± 1.2	**			**	**	**			**	**	**
		[43]	μ ± σ	**	**	1 ± 0.1	**			**	**	**			**	**	**
		[38]	μ1/2 (IQR)	**	3.2 (1.4–7.3)	**	**			**	**	**			**	**	**
	SCFA (mM)	[48]	μ ± σ	**	**	**	**			**	**	78.5 ± 42.2			**	**	**
		[42]	μ ± σ	**	**	**	**			8.6 ± 6.6	32.2 ± 17.6	**			**	**	**
		[39]	μ ± σ	**	**	**	**			**	**	30.9 ± 15.4			**	**	**
		[45]	μ ± σ	**	**	**	**			22 ± 22	**	88 ± 36			**	**	**
	Lipids (mg mL−1)	[48]	μ1/2	**	**	**	**			**	**	0.58			**	**	**
		[50]	μ1/2	**	0.45	**	**			**	**	**			**	**	**
		[51]	μ	0.56	0.60	**	**			**	**	**			**	**	**
		[52]	μ ± σ	**	0.43 ± 0.45	**	**			**	**	**			**	**	**
	Phospholipids (µM)	[48]	μ ± σ	**	**	**	**			**	**	266 ± 110			**	**	**
		[42]	μ ± σ	**	**	**	**			73 ± 41	166 ± 110	**			**	**	**
		[50]	μ1/2	**	580	**	**			**	**	**			**	**	**
		[39]	μ ± σ	**	**	**	**			**	**	362 ± 210			**	**	**
		[53]	range	**	55–90	**	**			**	**	**			**	**	**
		[52]	μ ± σ	**	433.3 ± 230.4	**	**			**	**	**			**	**	**
		[51]	μ ± σ	2900 ± 500 mM	6300 ± 1000	**	**			**	**	**			**	**	**
		[43]	μ ± σ	**	**	200 ± 70	**			**	**	**			**	**	**
	Bile acid (mM)	[42]	μ ± σ	**	**	**	**			0.07 ± 0.15	0.12 ± 0.12	**			**	**	**
		[39]	μ ± σ	**	**	**	**			**	**	0.12 ± 0.12			**	**	**
		[45]	μ1/2 (μ)	**	**	**	**			0.03 (0.1)	**	0.26 (0.38)			**	**	**
		[50]	μ1/2	**	3.3	**	**			**	**	**			**	**	**
		[48]	μ ± σ	**	**	**	**			**	**	0.28 ± 0.23			**	**	**
		[49]	μ ± σ	0.2 ± 0.5	**	2.9 ± 2.9	**			**	**	**			**	**	**
		[51]	Range/μ ± σ	0.21–0.41	5.9 ± 1.8	**	**			**	**	**			**	**	**
		[47]	μ ± σ	0.82 ± 0.57	**	1.52 ± 1.77	**			**	**	**			**	**	**
		[40]	μ ± σ	0.33 ± 0.31	**	**	**			**	**	**			**	**	**
		[44]	μ ± σ	**	2.7 ± 1.7	3.4 ± 1.6	**			**	**	**			**	**	**
		[46]	μ1/2 (IQR)	0 (0–0.04)	**	**	**			**	**	**			**	**	**
		[37]	μ1/2 (IQR)	**	1.4 (0.4–2.5) μg mL−1	1.1 (0.5–1.5) μg mL−1	0.9 (0.4–1.9) μg mL−1			0.7 (0.2–1.5) μg mL−1	**	**			**	**	**
		[52]	μ ± σ	**	1.4 ± 0.9	**	**			**	**	**			**	**	**
		[53]	μ ± σ	**	3.5 ± 1.8	**	**			**	**	**			**	**	**
		[43]	μ ± σ	**	**	2.0 ± 0.2	**			**	**	**			**	**	**
		[38]	μ1/2 (IQR)	**	2.7 (1.9–3.5)	**	**			**	**	**			**	**	**
	Carbohydrates (mg mL−1)	[39]	μ ± σ	**	**	**	**			**	**	8.1 ± 8.6			**	**	**
		[42]	μ ± σ	**	**	**	**			1.6 ± 1.0	2.3 ± 1.0	**			**	**	**
Osmolality (mOsm kg−1)		[48]	μ ± σ	**	**	**	**			**	**	218 ± 79			**	**	**
		[42]	μ ± σ	**	**	**	**			60 ± 50	144 ± 65	**			**	**	**
		[39]	μ ± σ	**	**	**	**			**	**	81 ± 102			**	**	**
		[54]	μ ± σ	**	226 ± 35	**	**			**	**	**			**	**	**
		[54]	μ ± σ	**	215 ± 37	**	**			**	**	**			**	**	**
		[49]	μ ± σ	191 ± 39	**	271 ± 15	**			**	**	**			**	**	**
		[40]	μ ± σ	220 ± 58	**	**	**			**	**	**			**	**	**
		[44]	μ ± σ	**	137 ± 54	200 ± 68	**			**	**	**			**	**	**
		[47]	μ ± σ	221 ± 15	**	278 ± 16	**			**	**	**			**	**	**
		[52]	μ ± σ	**	181 ± 41	**	**			**	**	**			**	**	**
		[45]	μ ± σ	**	**	**	**			128 ± 56)	**	299 ± 49			**	**	**
		[38]	μ1/2 (IQR)	87.4 (71.0–125.7)	181.2 (150.6–214.1)	**	**			**	**	**			**	**	**
Water content (%)		[55] a	μ	80.3	80.7	79.4	78.0	84.4	86.5	88.5	86.7	81.0	80.2	80.1	77.4	78.3	76.3
		[39]	μ ± σ	**	**	**	**			**	**	70.3 ± 17.0			**	**	**
		[45]	μ ± σ	**	**	**	**			80.3 ± 9.9	**	39 ± 18			**	**	**
pO2 (mm Hg.)		[56]	μ ± σ	58 ± 15	32 ± 8		11 ± 3									3 ± 1

^aStudy was performed in pigs

^** Data not available

Table 2.

Physiological conditions and composition along a normal GI tract mucosal layer

				GI mucosal region
				Stomach	Small intestine				Cecum	Colon			Rectum
Factor		Study	Metric		Duodenum	Proximal (Jejunum)	Traverse	Distal (Ileum)		Proximal	Traverse	Distal
Mucin glycoproteins (mg mL−1)		[62]a*	Range		7.0–8.5
		[63]	μ (range)	47 (35–49)	**	**	**	**	**	**	**	**	**
		[64]b*	μ ± σ	**	48.8 ± 2.9				**	**	**	**	**
		[65]*	μ ± σ	**	**	**	**	**	**	54.0 ± 4.5			**
pH		[59]a	μ	5.6	6.7	7.1	**	7.1	7.3	7.5	**	7.5	**
		[66]b	μ ± SEM	**	6.5 ± 0.1	**	**	7.5 ± 0.1	**	**	**	**	**
		[67]	μ	**	**	6.0	**	**	**	**	**	**	**
		[68]b	μ ± σ	**	**	7.3 ± 0.1	**	**	**	**	**	**	**
		[69]b	μ ± SEM	**	6.9 ± 0.1	**	**	7.9 ± 0.1	**	**	**	**	**
		[70]	μ	**	**	**	**	7.6	**	**	**	**	**
Protein (mg mL−1)		[62]a*	Range	**	54.6–66.3
		[71]a	μ	17.4	13.3	72.8			**	22.0			**
		[64]b*	μ ± σ	**	60.6 ± 3.6				**	**	**	**	**
		[66]b	μ ± SEM	**	56.4 ± 4.1	**	**	24.6 ± 2.4	**	**	**	**	**
		[70]	μ	**	**	**	**	3.42	**	**	**	**	**
		[63]	μ (range)	17.4 (12.3–24.7)	**	**	**	**	**	**	**	**	**
DNA (mg mL−1)		[62]a*	Range	**	8.4–10.2
		[64]*	μ ± σ	**	1.8 ± 0.2				**	**	**	**	**
		[63]*	μ ± σ	**	**	**	**	**	**	2.0 ± 0.5			**
Lipids (mg mL−1)		[62]a*	Range	**	49.2–59.7
Phospholipids (mg mL−1)		[62]a*	Range	**	2.6–3.2
Thickness (μm)	Inner layer	[72]b	μ ± σ	154 ± 16	16 ± 3	15 ± 2	**	29 ± 8	**	116 ± 51			**
	Outer layer	[72]b	μ ± σ	120 ± 38	154 ± 39	108 ± 5	**	447 ± 47	**	714 ± 109			**
Mesh size (nm)		[73]a	μ ± σ	**	211 ± 7				**	**	**	**	**
		[70]	μ	**	**	**	**	63	**	**	**	**	**
		[59]a	μ + SEM (Ferit’s diameter)	4000 ± 2000	5300 ± 2900	4000 ± 1700	**	4100 ± 2100	5700 ± 3700	5600 ± 3500	4200 ± 2800	**	**
Water/Wet weight (% w/w)		[59]a	Range	89.0–92.8	77.7–91.0				87.7–94.5				**
		[62]*	Range	**	83–86
		[64]b *	μ ± σ	**	83.8 ± 0.3				**	**	**	**	**
		[63]*	μ ± σ	**	**	**	**	**	**	82.0 ± 17.3			**
pO2 (mm Hg.)	Tissue	[74]	μ ± σ	46.3 ± 15.4	**	**	36.0 ± 9.7	33.5 ± 11.5	30.3 ± 7.4	38.5 ± 10.0	29.3 ± 11.0	39.2 ± 7.2	**
		[75]	μ ± 95% CI	77 (49–105)	**	**	**	**	**	**	**	**	**
	Mucosa	[76]	μ ± SEM	**	**	**	**	**	**	**	**	**	11.25 ± 3

^aStudy was performed in pigs

^bStudy was performed in mice/rats

^*Calculated assuming wet fraction with a density of water

^**Data not available

Fig. 2.

Spatial variation of physiochemical parameters. The schematic summarizes key biophysical and chemical features across spatial gradients in the GI tract that influence microbial colonization and function. It consists of three main components (green, blue, yellow). The green panel categorizes the different GI microenvironments occurring throughout the architecture of the GI tract. Distinct mucosal and luminal environments exhibit axial (along the length of the GI tract) and radial (from lumen to epithelium) variation giving rise to four distinct gradients (labelled 1 through 4). In the blue panel, two categories of environmental parameters are categorized: biophysical conditions in the top left and chemical parameters denoted in the center wheel. All parameters known to fluctuate along particular gradients highlighted in the green panel are labelled accordingly. The legend categorizes multiple internal and external perturbations including disease (red), lifestyle factors (blue), diet (green), and inter/intra-individual variability (yellow). Biophysical and chemical parameters which are known to be directly influenced by these factors are annotated using the respective color codes of these factors. Several examples of such interactions are highlighted for a few of the biophysical and chemical parameters in the yellow panels. Arrows between the chemical and biophysical conditions highlight their interconnected nature, where influences on one might induce changes in the other. GI = gastrointestinal, SCFA = short-chain fatty acid

Created in BioRender. Sieders, M. (2025) https://BioRender.com/ennhyq8

Healthy GI homeostasis is easily perturbed, and defining ‘baseline’ gut conditions is increasingly elusive. We categorize modulators into four groups: (i) disease-related, (ii) lifestyle-related, (iii) dietary-related, and (iv) endogenous inter-individual variation (Fig. 2). Meals and diet can alter gastric pH, fluid osmolality, surface tension, and water content [32, 35, 42, 45, 51, 83, 84], while fibers reshape digesta rheology and flow [85]. Lifestyle factors, such as exercise, sleep, and alcohol, affect motility and transit time [86–90]. Age and diseases like Crohn’s, Parkinson’s, ulcerative colitis, and cystic fibrosis are associated with altered mucus properties, pH profiles, or motility patterns [70, 87, 91–99]. Mucin expression, glycosylation, and thickness can also change under stress, inflammation, or through microbial interactions [100–105]. These alterations exist on top of pre-existing endogenous variation in such parameters found within populations [106–108].

The mucosal microbiota, distinct from luminal communities, includes taxa such as Akkermansia, Bacteroides, Ruminococcus, and Mucispirillum, which metabolize host mucins. Microbial densities typically decrease toward the epithelium, and even subtle shifts in mucus composition can drastically affect colonization patterns [18]. Notably, luminal and mucosal microbial responses to physiochemical cues are not merely passive; bacteria can also modulate gut rheology, pH, and chemical gradients, influencing host environments and subsequently microbial abundance through complex interplays [14, 37, 109–114].

Understanding how microbial communities adapt to the heterogeneous and dynamic microenvironments of the gut, under both healthy and perturbed conditions, is essential for designing relevant in vitro models. Rather than viewing this complexity as a barrier, we propose that the interplay of gradients and feedbacks in the gut offers a blueprint for engineering responsive, tunable in vitro environments using hydrogels as foundational scaffolds. Tables 1 and 2 offer literature-based reference values for key GI physiochemical parameters. These summaries contextualize the design targets for hydrogel-based modeling, which must be tailored to recapitulate region-specific and disease-specific gut microenvironments with high fidelity. These summaries offer practical guidance for designing gut-relevant hydrogel systems, e.g., tuning pH, viscosity, or mucin content, to mimic region-specific conditions in vitro.

Hydrogel matrices for microbial cultivation

A wide variety of polymers are available for hydrogel preparation, each with distinct advantages and limitations (Table 3). When designing hydrogels for microbial culturing, certain characteristics are particularly important: (i) biocompatibility, (ii) resistance to microbial degradation and metabolism, and (iii) the ability to maintain a gelled state under physiological conditions [150]. Additional considerations include the ability to extract viable bacteria from the hydrogel matrix, modulate pH, support nutrient flow, and control rheology. The ability to retrieve live microbes from the matrix is especially relevant for microbiologists performing cultivation within hydrogels and wishing to recover viable cells without damage. This is common in assays, such as isolation and enumeration, or studies of microbial interactions and adaptive responses under defined environmental conditions. For example, isolating anaerobic gut bacteria requires careful retrieval of live cells from encapsulating matrices for downstream culturing or analysis. Researchers seeking versatile, all-in-one polymers for microbial culturing face a narrower selection, simplifying their choice.

Table 3.

Common polymers in hydrogel-based models benchmarked for microbial growth support

	Polymer	Crosslinking			Working Concentration range (% (w/v))	Microbial encapsulation	Live microbial release	Applications and key characteristics	Literature
		Thermal (Tmelt/Tgel °C)	Photo (photo-initiator)	Chemical (crosslinking agent)
Bio-derived	Agarose	✔ (~ 86/~ 37)	✖	✖	0.2–3	✔	✖	• Well-characterized polymer and perhaps the most accessible to labs across the world • Thermal crosslinking limits the user to encapsulation and not retrieval of live microbes from matrix • Low gelling point agarose further enhances usability due to more biocompatible temperatures for gelling, though more expensive and less accessible	[17, 115, 116]
	Low gelling point agarose	✔ (~ 65/~ 26)					✔ (agarase addition to beads)		[117, 118]
	Alginate	✖	✖	✔ (Ionic: Divalent Cations, primarily Ca2)	0.75–2	✔	✔ (Sodium citrate/EDTA buffer chaperoning cations out of matrix)	• Easy to use, cheap, and well-known biocompatibility particularly well established for use with microbes • A great functionality benefit is the ability to retrieve live cells from the gelled matrix by decomposing the hydrogel at neutral pH and temperature • Does require the user to supply calcium ions continuously to prevent gel degradation • Brittle and sensitive, which may limit its usability in long-term cultivation	[119–123]
	GelMA	✖	✔ (Radical based, 365 nm)	✖	1.5–6	✔	✖	• Easy to use, cheap, and well-known biocompatibility particularly well established for human cell lines • Mimics the extracellular matrix but may be limited in modelling microbial niches in the GI tract	[124, 125]
	Collagen (type I)	✖	✖	✔ (EDC + NHS)	0.2 (coat)–0.4 (gel)	✖	✖	• Used for coating and generating matrices for microbes to adhere to, but not to encapsulate • Similar features to GelMA	[126, 127]
	Xanthan gum	✖	✖	✔ (PDA)	2.5–4.2	✔	✖	• Similar features to alginate • Less straight forward gelation process	[128]
Synthetic	Acrylamide, N,N-Dimethylacrylamide/N-Methylolacrylamide/HEAA	✖	✖	✔ (N,N- Methylenebisacrylamide/N,N-Bis(acryloyl)cystamine/N,N-dihydroxyethylene-bis-acrylamide + TEMED)	10–20	✖	✔ (TCEP solution when dissolvable crosslinker N,N'-Bis(acryloyl)cystamine is used)	• Tunable and well controlled properties for microbial growth • Wide varieties of options to choose from and combine • The different polymers will impact the growth of microbes	[129, 130]
	PEG, PEGDA, PEGDMA	✖	✔ (2-hydroxy-2-methylpropiophenone or Lithium phenyl-2,4,6-trimethybenzoyphospinate or 2,4,6-trimethylbenzoylphenyl phosphinic acid ethyl ester, 365 nm)	✔ (APS + TEMED, thiolated co-polymers, i.e., thiolated PEG)	3.5–20	✔	✔ (365 nm patterned light)	• Well established hydrogel matrix for use within the field of biomanufacturing using encapsulated microbes • Not benchmarked for GI mimicking	[131–137]
	Pluronic F-127(-DMA/-BUM)	✖	✔ (2-hydroxy-2-methylpropiophenone, 365 nm)	✖	30	✔	✖	• Well established hydrogel matrix for use within the field of biomanufacturing using encapsulated microbes • Not benchmarked for GI mimicking	[136, 138, 139]
	PVA	✔ (~ 121/~ − 20)	✖	✔ (H3BO3/Na2SO4)	7.5–12.5	✔	✖	• Has to be gelled through freeze–thaw cycles with glycerol in the matrix which is suboptimal	[79, 140, 141]
Composite	Pectin/Alginate	✖	✖	✔ (Ionic: Divalent Cations, primarily Ca2)	2	✔	✔ (Sodium citrate/EDTA buffer chaperoning cations out of matrix)	• Supports stable microbial encapsulation with minimal disintegration	[142]
	PVA(-stilbazolium)/Alginate	✖	✖	✔ (H₃BO₃, Ba2+, Ca2+, Na2SO4)	1–10	✔	✖	• Combines mechanical properties of PVA with ionically crosslinked alginate • Widely used across various applications highlighting its popularity and usability	[140, 143–147]
	Gellan/Xanthan Gum	✔ (~ 121/~ − 80)	✖	✖	0.25–2.5	✔	✖	• Stability across wide range of temperatures and pH • Limited use for temperature sensitive microbes	[148, 149]

GelMA = methyacrylate modified gelatin, PEGDA = polyethylene glycol diacrylate, PEGDMA = polyethylene glycol dimethacrylate, PEG = polyethylene glycol, PDA = polydopamine, PVA = polyvinyl alcohol, F-127 = 2-[2-(2-hydroxyethoxy)propoxy]ethanol, DMA = dimethylacetamide, BUM = bis-urethane methacrylate, EDC = 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide hydrochloride, NHS = N-hydroxysuccinimide, HEAA = (2-Hydroxyethyl)acrylamide, TCEP = tris(2-carboxyethyl)phosphine, TEMED = tetramethylethylenediamine, APS = ammonium persulfate, GI = gastrointestinal

Hydrogels are generally categorized into three main types: bio-derived, synthetic, and composite hydrogels (Table 3). Bio-derived hydrogels, such as alginate or gelatin, are sourced from natural materials and typically have excellent biocompatibility. However, they may suffer from mechanical weakness and batch-to-batch variability depending on sourcing. Synthetic hydrogels, including polyethylene glycol (PEG) and polyacrylamide, offer precise control over stiffness and degradation rates but generally lack the inherent bioactivity of natural polymers. Composite hydrogels combine elements from both categories, using synthetic and/or natural crosslinkers to enhance their versatility and functionality compared to single-component systems.

Agarose and alginate

Agarose, one of the most commonly used bio-derived crosslinkers, finds applications ranging from DNA separation to specialized 3D microbial culturing, although agar or phytagel remain more common for traditional plate culturing. Low-melting-point agarose is particularly advantageous for microbial encapsulation, as it remains liquid at temperatures of 37 °C, enabling the encapsulation of bacteria at physiological temperatures (Table 3). However, retrieving live bacteria from agarose is challenging. Typically, biomass must be extracted by melting the agarose at temperatures above 37 °C, which harms biological samples. This limitation could be overcome in small volume platforms, where agarose can be degraded enzymatically by agarase at physiological temperatures [117].

Ionically crosslinked alginate and calcium hydrogels are widely used for live microbial cultures. Despite sensitivity to ion chelators, pH variations, and their brittleness, which pose challenges in long-term cultivation, alginate hydrogels are favored due to their simplicity and key advantages. They are easier to handle than radical- or photo-crosslinking-based hydrogels (e.g., PEG and other synthetics). Gelation can be finely tuned by the choice of divalent ion donor and environmental conditions, and may even be crosslinked with UV light [151]. Incorporating glucono-delta-lactone allows gradual release of Ca²⁺, facilitating uniform crosslinking of alginate gels at neutral pH [119]. Though alginate is generally considered biologically inert material, complex microbiomes can degrade the alginate backbone, compromising long-term stability, especially in marine microbes of origin [152] and human GI isolates [153]. In gut microbiome studies, especially those mimicking mucus layers, alginate gels may be less relevant due to the long-term inoculation needed to achieve stable communities [154]. A unique feature of alginate hydrogels is their efficient degradation via calcium ion scavenging chelators such as sodium citrate and EDTA at physiological pH and temperature, allowing live microbial recovery after experiments. This has been applied to plate out the broken-down matrix and retrieve live cell counts [120]. Furthermore, alginate-based gels are popular for mimicking the mucosal layer, reportedly reproducing several key properties of GI mucosa, including rheology, with reasonable accuracy [120, 121].

Other bio-derived materials

Several other bio-derived polymers have been explored for microbial encapsulation, each with application-dependent pros and cons. Among these, gelatin methylacrylate (GelMA) and collagen, notably for their ability to simulate the human extracellular matrix, make them useful in human cell culture but less for microbial systems [155]. GelMA is prized for its bioactive extracellular-matrix-like properties that promote cell proliferation and spreading. While these traits are less relevant for microbes in non-host-associated environments (e.g., mucosal or luminal spaces), GelMA rapid crosslinking (~ 1–10 min) via exposure to light (Table 3) makes it appealing for specific 3D microbial culturing contexts.

Xanthan gum, crosslinked with polydopamine, has emerged as a robust alternative to polymers like alginate. Xanthan-based hydrogels exhibit strong mechanical stability, wide pH tolerance, and support long-term culturing of microbes such as Bacillus thuringiensis and Planococcus sp. [128]. These systems maintain microbial encapsulation and metabolic activity, making xanthan gum suitable for applications requiring extended stability.

Compared to alginate, xanthan gum offers superior chemical stability and dual sorption capacity for cationic and anionic molecules, enabling functionality across diverse environmental conditions. While the sensitivity of alginate to ion chelators limits its use, the robustness of xanthan gum suits biotechnological and environmental applications. Unlike GelMA and collagen, which mimic extracellular matrices in human cell culture, xanthan gum prioritizes mechanical and chemical stability, making it more practical choice for microbial systems. However, some gut microbes can enzymatically degrade xanthan gum as well [156].

Synthetic polymers and composite materials

A diverse array of synthetic polymers is used to encapsulate live microbes, including polyacrylamide, polyethylene glycol-based molecules, polyvinyl alcohol, and Pluronic F127, which features triblock copolymers with hydrophobic poly (propylene oxide) cores and hydrophilic poly(ethylene oxide) chains (Table 3). These synthetic hydrogels are primarily employed where mimicking GI environments is not the primary objective, facilitating assays or production methods challenging in conventional suspension or plate cultures [131, 132, 138].

A promising frontier in GI research involves novel, underexplored materials for microbial cultivation. For example, Han et al. (2022) developed a photo-crosslinkable colon-derived decellularized extracellular matrix for co-culturing gut microbes. While promising, such materials may not fully replicate the native GI microenvironment, as microbes in vivo are typically separated from epithelial cells by mucus layers [157]. Other approaches use mucins as a polymer backbone for hydrogels. For example, Duffy et al. (2015) developed a photo-crosslinkable gel from methacrylated bovine submaxillary mucin, formed by 10 min exposure to 365 nm UV light [158]. Meanwhile, Joyner et al. (2019) used PEG-thiol-based crosslinkers with disulfide bonds for porcine gastric mucin gels mimicking native mucus-like rheology [159]. Neither method has been extensively verified for microbial culturing. UV-based gelation poses practical challenges due to prolonged exposure times, and the generation of free radicals during the crosslinking process (via chemical agents or UV) can damage cells. Careful titration of UV exposure and photo crosslinker dosage is essential to balance crosslinking efficiency with cell viability, especially for sensitive strains or co-cultures.

Modifications and functionalization

Hydrogel functionalization strategies tailor matrix properties to mimic specific GI tract conditions. Modifications can replicate regional characteristics, such as pH, nutrient gradients, or mechanical properties, to better study microbe-gut interactions. The base polymer (Table 3) strongly influences hydrogel properties, but modifications offer considerable flexibility to model native luminal or mucosal microenvironments (Fig. 3).

Fig. 3.

Customizable hydrogels feature for supporting microbial growth. The schematic highlights tunable hydrogel properties demonstrated in the literature, grouped by their functional roles. These include control over environmental parameters (e.g., pH, oxygen, and nutrients), mesh size (porosity affecting microbial mobility and growth), surface coatings (e.g., mucus functionalization or hydrogel surface applications), and rheological properties (solid- or liquid-like behavior). Structural features such as spatial hierarchy, custom shapes, and 3D features like synthetic villi or crypts can also be incorporated. Liquid-filled regions can support linear gradients while maintaining gel hydration. While most features are broadly applicable, compatibility depends on the hydrogel polymer and other modifications. ECM = extracellular matrix

Created in BioRender. Sieders, M. (2025) https://BioRender.com/lkm6co8

Hydrogels are naturally porous structures, with pore sizes from micrometers to nanometers depending on polymer type, crosslinker, and concentration [160]. This porosity spatially compartments microbes, facilitating co-culturing and interaction studies without explicit barriers [161, 162].

Pore size can prevent bacterial transfer between compartments while allowing nutrient and small molecule diffusion, governed by hydrogel formulation and bacterial size.. Bacterial motility within hydrogels depends critically on matrix pore size and rheology. For example, agarose median pore size decreases from 400 nm at 0.33% to 200 nm at 1.0% polymer concentration [163]. Polymer concentration also affects rheology, shifting materials between solid-like or liquid-like states, which researchers tune to mimic native parameters [120].

Bacteria exhibit optimal motility in suspended environments with appropriate rheological gradients [164–166], while dense gels (~ 0.7% agarose) may fully inhibit movement [17]. Thus, hydrogel pore sizes influence motility and nutrient diffusion, mirroring the selective barrier function of the GI mucus, allowing controlled exchange of nutrients and signaling molecules while restricting larger entities. Accurate in vitro measurements of pore size remain challenging, as common techniques (e.g., electron microscopy) require lyophilization, which can distort hydrogel morphology. Alternatives like atomic force microscopy allow pore size measurements in hydrated, native-shaped hydrogels [163, 167].

Researchers also exploit extrudability of hydrogels to “print” layered mimicking GI gradients, physical constraints, and spatial hierarchy [168]. Significant work focuses on microbe-laden hydrogels suitable for 3D printing [124, 169–171].

Microfluidic chips integrate hydrogels to generate oxygen gradients through diffusion, with sections exposed to air, enabling simultaneous anaerobic and aerobic conditions. Controlling oxygen levels and gradienttailors systems to diverse experimental setups, enhancing flexibility and precision [172]. Besides air-gel interface diffusion, the liquid phase in hydrogel systems, which is crucial considering fast dehydration at physiological temperatures, may introduce gradients along the hydrogel.

Hydrogel structural adaptations can emulate large-scale GI features. Their moldability allows generation of architectures for studying cell adhesion [126, 133, 173, 174]. Such structures may be coated with physiologically relevant proteins like mucins, allowing bacterial adhesion and interaction with mucus-like environments, thereby providing insight into microbial colonization and behavior [175, 176].

Functionalization strategies summarized in Fig. 3 highlight hydrogel flexibility and their potential to mimic the physical and chemical GI microenvironments (Fig. 2).

Hydrogel applications across microbiology

Hydrogels emerged as a powerful tool in microbiology due to their immobilization capability, spatial structure, diffusion properties, and synthetic and biological polymer options and properties. Despite the large range of applications, hydrogels benefit experimental setups in similar ways, for example through immobilization, which benefits most applications, though other tasks require other specific properties (Table 4).

Table 4.

Applications of across microbiology research

Application		Key study	Microbial system	Hydrogel characteristic leveraged						Hydrogel advantage
				Degradability	Preservation	Gradients	Immobilization	3D printing	Spatial hierarchy/organization
Bio-industry	Production	[139]	Escherichia coli, Saccharomyces cerevisiae	✖	✔	✖	✔	✔	✔	- Controlling consortium population dynamics - Sustain metabolic activity for over 1 year of repeated use
		[177]	Escherichia coli	✖	✔	✖	✔	✖	✖	- Controlling consortium population dynamics
	Nitrogen removal	[143, 144, 147, 178]	Undefined comammox/anammox communities	✖	✖	✖	✔	✖	✖	- Controlling consortium population dynamics - Enhanced nitrogen removal with reduced aeration and low temperature
Specialized assays	Spatial organization and communication	[179]	Pseudomonas aeruginosa, Escherichia coli, Bacillus subtilis, Lactococcus lactis	✖	✖	✔	✔	✖	✔	- Steer community spatial hierarchy to study communication
	MIC determination	[180]	Staphylococcus aureus	✖	✖	✔	✔	✖	✖	- High-throughput screening with minimal sample requirement by using diffusion-based microscale culture chambers
	Cellular stiffness	[181]	Escherichia coli, Pseudomonas aeruginosa	✖	✖	✖	✔	✖	✖	- Simpler and more accessible methodology
	Tracking	[182–185]	Bacillus subtilis, Agrobacterium tumefaciens, Escherichia coli, Vibrio alginolyticus	✖	✖	✔	✖	✖	✔	- Controlled mobility and chemotaxis tracking - Physiologically relevant migration
		[186]	Klebsiella pneumoniae Escherichia coli	✖	✖	✖	✔	✖	✖	- Label-free immobilization maintaining normal behavior and clear, long-term, and high-resolution imaging without blur or background noise
Library screening		[132, 187]	Agrobacterium tumefaciens	✔	✖	✖	✔	✖	✖	- High throughput selective release of variant from large pool using UV light
		[116, 188]	Escherichia coli	✖	✖	✖	✔	✖	✖	- Microbes embedded in microparticles (compatible with FACS) to study mutant libraries
Biomedical applications	Probiotic delivery	[189, 190]	Akkermansia muciniphila, Bifidobacterium pseudocatenulatum	✔	✔	✖	✔	✖	✖	- GI region targeted probiotic release
		[191–194]	Akkermansia muciniphila, Bifidobacterium lactis	✖	✔	✖	✔	✖	✖	- Enhanced probiotic aerobic storage
		[142, 192, 194–196]	Lactobacillus rhamnosus, Lactobacillus plantarum	✖	✔	✖	✔	✖	✖	- Enhanced GI tract survival rate
	Prebiotic	[197]	–	✖	✖	✖	✖	✖	✖	Stimulate microbiome composition in vivo
	Wound healing	[125, 198]	Lactobacillus reuteri, photosynthetic bacteria	✖	✖	✖	✔	✔	✖	Improved infected wound healing through living biomaterial treatment

MIC = minimum inhibitory concentration, GI = gastrointestinal

– = not applicable

Implementation of hydrogel microbial co-cultures in bioproduction or nitrogen removal may outperform traditional planktonic co-cultures and enhance fermentation process longevity and efficiency [139, 143, 177, 178], in a similar fashion to what can be achieved through using porous membranes to separate microbial communities for maintaining community longevity in vitro [9, 10]. Related works, such as those from Jeong and Irudayaraj (2023) or Gottshall et al. (2021) focus on hierarchical co-culturing of microbes and leverage hydrogel-enabled spatial configuration, though outside of the scope of bio-industry, utilizing hydrogel-encapsulation as a more fundamental tool to understand microbial intra- and interspecies interactions through nutrient gradients and spatial constraints [144, 179].

More conventional microbiological assays also know some hydrogel-assisted counterparts, such as MIC determination [180] and the determination of cellular stiffness [181] may be performed utilizing hydrogels to assist, enable, or streamline the assay. Assays which can be considered challenging or impossible without the assistance of hydrogel encapsulation are primarily related to those that require tracking. On the one hand, such assays may require controlling mobility or completely immobilizing microbes to facilitate long-term imaging on microbes where it is crucial that microbes do not move away from the tracking focus plane [186]. On the other hand, lower concentration gels may allow the study of microbial motility and movement-dependent characteristics in a controlled fashion [17, 182–185].

Hydrogels may also be employed to screen libraries of microbes (i.e., complex samples, variant libraries) encapsulated in hydrogel matrices, such as beads which could be sorted using FACS. Such as demonstrated in the works of Fattahi et al. (2020) and Van Der Vlies et al. (2019) who can optically identify microbial variants with desirable phenotypes from large immobilized libraries and release them from a photodegradable gel using targeted UV beams [132, 187].

Lastly, hydrogels are also abundantly applied in probiotics, prebiotics, and wound healing, with various matrices characterized to enhance the viability of probiotic species like Akkermansia, Bifidobacterium, and Lactobacillus during storage and GI transit. These researchers follow multiple philosophies, ranging from pH-responsive degrading types for targeted delivery to highly stable and non-degradable matrices that minimize interference from native gut microbes [142, 189].

Hydrogel models of the GI tract: a typology based on microbial localization and application

While these applications span industrial, ecological, and synthetic biology contexts, a particularly rich application domain lies in GI modeling. In this context, hydrogels not only serve as native analogs but also provide spatial structure for microbial localization, barrier co-culture, and functional readouts. These are aspects that may be challenging to incorporate into the wide array of more conventional cultivation strategies and (host-microbe co-culture) gut-on-a-chip models, despite their broad popularity in the field [199–203]. The following section focuses on hydrogel-based gut models, organized by microbial placement and experimental scope.

Considerable effort has gone into standardizing hydrogel for diverse modeling applications, with a strong focus on mimicking mucus-like environments due to natural similarities with hydrogels [21, 78, 204, 205]. Nonetheless, they pose considerable ability to mimic non-mucosal contexts such as soil, where they successfully replicate native-like conditions [129, 134, 135, 206, 207]. In the GI tract, hydrogels are used to model interactions between microbes, mucus, luminal contents, and host tissues. We propose a typology of gut hydrogel systems based on two axes: (1) microbial localization, interfaced with the gel versus encapsulated within it, and (2) functional scope, microbial-only versus host–microbe co-culture. This yields four major types of hydrogel-based models: Type IA–IIB (Fig. 4).

Fig. 4.

Classification of hydrogel-based gut models for microbial integration and application. Experimental models are categorized into four types: hydrogel-interfaced systems (Type I), hydrogel-encapsulated systems (Type II), microbial culture systems (Type A), and co-culture systems with human cell lines (Type B). The schematic highlights key features of each setup and application focus

Created in BioRender. Sieders, M. (2025) https://BioRender.com/in47uk6

Type IA: Interfaced, microbial-only

Type IA models feature microbes cultured in a liquid medium interfaced with a hydrogel. These systems have been primarily used to simulate the mucus layer, with microbes seeded onto or near the surface, with early adaptations using this platform as an assay to observe, quantify, and study mucus adhesion in vitro using various mono-cultures and infected complex undefined communities [208–211]. These models are critical for studying colonization, adhesion, and spatial organization at the luminal–mucosal interface, often under defined gradients or flow conditions. Such systems allow complex sampling, for instance, analyzing both the liquid suspension, surface-attached, or hydrogel-embedded fractions to assess microbial behavior across compartments [212, 213]. The utility of Type IA systems extends primarily to revealing microbial spatial organization features of complex communities and is utilized to investigate the environment. These models provide a highly controlled platform for observing microbial behavior, including attachment, biofilm formation, and other interactions with mucus layers, which are critical to understanding luminal–mucosal dynamics [175, 176].

Type IB: interfaced, host-microbe

Type IB models introduce additional complexity by enabling co-culturing of human cell lines with microbes at hydrogel interfaces. Here, epithelial cells are cultured adjacent to or on top of a hydrogel, with microbes added apically. These systems allow real-time interrogation of microbial adhesion, epithelial response, or barrier integrity in co-culture and are widely used in intestinal infection and probiotic interaction studies. Models featuring villi-like structures provide a substrate for human epithelial cell growth while simultaneously enabling microbial adherence and interaction with the cells, mimicking the intestinal lumen [126, 173, 174]. Similarly, Swaminathan et al. (2021) utilized a hydrogel model to study microbial attachment to intestinal epithelial layers, shedding light on microbe-host dynamics in the GI tract [133].

Type IIA: encapsulated, microbial-only

Type IIA models involve encapsulating microbes directly within the hydrogel matrix. These models involve embedding microbes within hydrogels to simulate the gut lumen’s spatial and physicochemical constraints [120, 122]. By simulating the physical and chemical constraints of the mucosal layer within the gut, the Type IIA models demonstrated by Sardelli et al. (2022, 2024) and Pajoumshariati et al. (2018) provide an effective tool for studying microbial growth and interaction under conditions that resemble those in vivo, as microbes are forced to inhabit localized biofilm colonies [120, 122, 123]. These types of models enable researchers to observe how nutrient gradients and diffusion dynamics within the hydrogel influence microbial growth patterns and competition, phenomena that are difficult to replicate in conventional culture systems [123, 134]. Additionally, encapsulating microbes within the hydrogel matrix allows for detailed studies of microbial metabolic exchanges and quorum sensing in spatially constrained environments, offering insights into how biofilm communities may coordinate their activities [214].

Type IIB: encapsulated, host-microbe

Type IIB models integrate hydrogel-encapsulated microbes with co-cultured human cell lines, significantly broadening their applications as an extension of Type IIA systems. They are used to model biofilm formation, oxygen gradients, or host responses in structured environments, and increasingly include immune components to capture more physiologically relevant interactions. Notable examples are the GI models developed by Zheng et al. (2023) and Huang et al. (2022), which mimic the mucosal layer and incorporate spatial and chemical gradients to mimic the heterogeneity of the GI tract. This model facilitates the study of interactions between human cells and microbes under region-specific conditions guided by hydrogel interfaces [121, 168].

Microbial physiology and behavior in hydrogel models

Hydrogel-based culture systems offer a powerful alternative to traditional suspension cultures by better mimicking the structured, spatially complex in vivo microenvironments, as shown in the previous section. Here, we review how hydrogel cultivation influences microbial metabolism, growth patterns, community composition, and interactions with host or matrix components. This section highlights key differences between hydrogel-based and conventional liquid culture systems.

Metabolic profiles

Microbes cultured in hydrogels often exhibit distinct metabolic activity compared to those grown in suspension. For example, hydrogel-encapsulated yeast and bacterial systems have shown enhanced cellulose degradation and biomass production [129]. Similarly, Ruminiclostridium cellulolyticum produced different fermentation profiles depending on the growth system; ethanol predominated in suspension, while acetate and other products dominated under encapsulated conditions [134]. These shifts may arise from several factors, including a more physiologically relevant microenvironment and a higher achievable cell densities in the hydrogel matrix. The latter could cause a bulk increase in metabolic output without necessarily enhancing cell-specific metabolic activity. Detangling these effects would require direct measurements of per cell-specific metabolic rates, which remain technically challenging.

Hydrogel properties can also drive specific metabolic adaptations. For example, oxygen-limited diffusion in F127-bis-urethane methacrylate hydrogels promoted a shift from respiratory to fermentative metabolism in Saccharomyces cerevisiae, increasing ethanol yields by 3.7% [138]. These findings highlight the importance of accounting for matrix physiochemical properties when interpreting metabolic readouts in hydrogel-based cultures.

Co-culture of Escherichia coli with HepG2 cells in a 3D-hydrogel system led to the production of indole-3-propionic acid, a metabolite absent in 2D cultures. This, in turn, stimulated IL-22 release from HepG2 cells, illustrating how hydrogel environments can reveal context-dependent metabolic signaling relevant to host immunity [215].

Community dynamics and composition

Microbial consortia cultured using hydrogels often exhibit altered dynamics and compositions compared to suspension cultures. For example, Type IA systems with mucin-coated hydrogel surfaces supported the formation of biofilms and the emergence of strain-specific spatial patterns and attachment-mediated genomic invertrons in defined (123-species, taken from a previous study from Chen et al. (2022)) and undefined communities [175, 176, 216, 217]. These systems sustained higher species richness than liquid cultures and increased in mucin-binding genes, echoing luminal-mucosal dynamics observed in the human gut. Spatial organization patterns can extend to the strain level, with specific taxa preferentially colonizing and self-organizing in distinct micro-niches over time[143, 144, 176]. Calvigioni et al. (2023b) show that Bacillus cereus, when added to GI-derived consortia in a type IA model, revealed mucus concentration-dependent adhesion and virulence gene expression, underscoring the regulatory role of host-like interfaces [175, 218].

Type IA models so far have classified localization in broad categories (e.g., supernatant vs. mucus-bound), but future research should employ high-resolution imaging at the micron scale to reveal finer spatial structures at liquid-gel interfaces. These investigations could uncover how local gradients, viscosity, or binding affinities shape emergent localized community patterns.

Growth patterns and biofilm formation

Hydrogel matrices, particularly Type IB systems, promote biofilm-like growth characterized by dense, spherical aggregates embedded in the gel. This has been observed across species, including Staphylococcus aureus [115, 127], Pseudomonas aeruginosa [127, 130], Saccharomyces cerevisiae [136, 139], and Escherichia coli [116, 130, 139], among others [130, 132]. Aggregate morphology is influenced by hydrogel properties, such as pore size and deformability, which impact nutrient access and mechanical cues. Growth rates typically differ from those in liquid media due to diffusion limitations and altered cellular states [116, 118, 130, 219]. For example, reduced nutrient penetration can impair microbial proliferation in certain hydrogels [135]. Yet this confinement may better mimic natural environments, where microbes often occupy spatial niches separated by only a few hundred microns. Hydrogel systems also support stable multi-species co-cultures, in part by reducing competitive exclusion. Composition varies with hydrogel formulation and initial inoculum [168, 178]. Metabolic cross-feeding may further regulate population balance. In lower-density gels, satellite colony formation has been observed, potentially facilitated by extracellular motility structures [220].

Microbe-gel interactions

Microbial adhesion to hydrogels depends on surface chemistry, pore architecture, and bioactive additives. Functionalized surfaces, such as mucin-coated gels, enhance adhesion of mucosa-associated species and influence downstream community assembly [106, 175, 221, 222]. Bacterial adhesion can also be modulated via interactions between microbial surface proteins and hydrogel functional groups like carboxyl or hydroxyl moieties [223].

Conversely, some hydrogels are designed to resist microbial attachment. Chitosan-containing matrices, for example, possess antimicrobial properties that inhibit adhesion and growth [224, 225]. Selective tuning of hydrogel components thus offers a means to either promote or suppress colonization.

Notably, microbial activity can also remodel hydrogel properties. Certain bacteria alter the stiffness or integrity of the gel matrix during colonization [118], especially in mucus models where microbes degrade mucins and modify viscoelasticity [77]. These reciprocal interactions suggest that hydrogel systems can capture feedback loops between microbial physiology and the biophysical environment.

Conclusions and future directions

Hydrogels have transformed in vitro modeling of the GI tract, particularly in replicating structured, gel-like niches such as the mucosal environment. However, their utility extends well beyond mucus analogs. Hydrogel systems can also model less structured regions of the gut, including luminal spaces, by enabling spatial control, nutrient gradients, and microbial aggregation. Despite significant progress, key challenges remain. Accurately mimicking the native glycosylation patterns and rheological properties of human mucus, especially in disease states, is crucial, as microbial adhesion and metabolism are highly sensitive to these features. Inter-individual variability further complicates the task of designing representative models. The highly heterogeneous nature of mucus, comprising global radial and axial gradients but also localized segments with altered physiology and microarchitecture, underscores the need for a more critical assessment of claims regarding hydrogel physiological and structural similarity to native mucus [120, 226, 227]. This will require creative solutions and invites opportunities to connect with conventional GI tract models, such as leveraging cell-line or organoid produced mucus layers for microbial cultivation [228]. Moving forward, personalized and disease-specific hydrogel systems are needed to reflect the diverse conditions of the human gut.

To capture these complex dynamics, next-generation models must incorporate environmental factors such as pH, oxygen gradients, and host-derived signals. These models will be essential to study how microbial behavior shifts in disease, including alterations in metabolism, spatial structure, and biofilm formation.

In parallel, we must deepen our understanding of how microbes interact with hydrogel matrices, both at the surface and within the gel. This includes examining how different 3D environments influence microbial gene expression, protein production, and community organization. Of particular importance are fine-scale host-microbe interactions (~ 100 µm) at the epithelial–mucus interface. While often inferred from meta-omics, these remain poorly validated in situ. Hydrogel platforms offer a tractable way to experimentally resolve such interactions at high spatial resolution.

Emerging tools such as 3D printing, multiplexed imaging, and live in situ microscopy are poised to further advance this field. When integrated into hydrogel models, these technologies will enable time-resolved studies of microbial colonization, metabolite exchange, and biofilm formation, processes central to both health and disease. To support this, a broader library of biocompatible, tunable, and microbe-friendly gelling polymers will be essential. These should accommodate a wide range of microbial species and allow controlled incorporation of disease-relevant features like altered pH, glycan patterns, or viscosity.

In conclusion, hydrogel-based GI models represent a crucial step toward recreating the spatial and biochemical complexity of the human gut. Such controlled, scalable, and physiologically relevant systems models will offer new opportunities to dissect microbial ecology and function. Ultimately, they may open the door to novel interventions that harness or reshape microbial communities for therapeutic benefit.

Authors’ contributions

M.S. wrote the main manuscript text and prepared Tables and Figs. 1, 2, 3, and 4. P.C. provided critical comments on the manuscript, and S.E.A. edited the manuscript content. M.S., P.C., and S.E.A. revised the manuscript.

Funding

Nothing to declare.

Data availability

No datasets were generated or analysed during the current study.

Declarations

Ethics approval and consent to participate

Nothing to declare.

Consent for publication

All authors consented with.

Competing interests

The authors declare no competing interests.