Harnessing and delivering microbial metabolites as therapeutics via advanced pharmaceutical approaches

Abstract

Microbial metabolites have emerged as key players in the interplay between diet, the gut microbiome, and host health. Two major classes, short-chain fatty acids (SCFAs) and tryptophan (Trp) metabolites, are recognized to regulate inflammatory, immune, and metabolic responses within the host. Given that many human diseases are associated with dysbiosis of the gut microbiome and consequent reductions in microbial metabolite production, the administration of these metabolites represents a direct, multi-targeted treatment. While a multitude of preclinical studies showcase the therapeutic potential of both SCFAs and Trp metabolites, they often rely on high doses and frequent dosing regimens to achieve systemic effects, thereby constraining their clinical applicability. To address these limitations, a variety of pharmaceutical formulations approaches that enable targeted, delayed, and/or sustained microbial metabolite delivery have been developed. These approaches, including enteric encapsulations, esterification to dietary fiber, prodrugs, and nanoformulations, pave the way for the next generation of microbial metabolite-based therapeutics.

In this review, we first provide an overview of the roles of microbial metabolites in maintaining host homeostasis and outline how compromised metabolite production contributes to the pathogenesis of inflammatory, metabolic, autoimmune, allergic, infectious, and cancerous diseases. Additionally, we explore the therapeutic potential of metabolites in these disease contexts. Then, we provide a comprehensive and up-to-date review of the pharmaceutical strategies that have been employed to enhance the therapeutic efficacy of microbial metabolites, with a focus on SCFAs and Trp metabolites.

1. Introduction

Trillions of microbes reside in the human body, rendering humans composite organisms consisting of both human and microbial cells. A dynamic collection of mutualistic, commensal, and pathogenic microorganisms in the human gastrointestinal (GI) tract together contribute the gut microbiome, an additional genome comprising millions of genes. These genes encode for a vast and diverse assortment of proteins that regulate both microbial and host functions. Recent groundbreaking studies in the field of gut microbiome research have helped to illuminate the complexity of the host-gut microbe interface. In these studies, microbial metabolites produced by commensal bacteria during their digestion of dietary components have emerged as key players in the interplay between diet, the gut microbiome, and host health. Two major classes of microbial metabolites, short-chain fatty acids (SCFAs) and tryptophan (Trp) metabolites, are now understood to be crucial regulators of host immune, metabolic, and neuroendocrine processes. In fact, alterations in microbial metabolite production are hypothesized to contribute to the development and progression of a multitude of human diseases.

Altered microbial metabolite production is an inherent consequence of gut dysbiosis, which occurs when the balance between helpful and harmful bacteria is disrupted. In recent years, the emergence of evidence linking gut dysbiosis to a wide range of human diseases has sparked the development of microbiome-based therapeutic approaches. Classic approaches to counteract gut dysbiosis, including probiotics supplementation and fecal microbiota transplantation (FMT), involve administering live bacteria. Although these approaches have shown promise in various disease contexts, their widespread use has been limited by numerous challenges (Suez & Elinav, 2017). For orally administered probiotics, controversy exists as to whether the bacteria can survive their transit through the stomach to reach the GI tract in sufficient amounts. Additionally, the relevance of specific strains in treating specific diseases has not been clearly defined. On the other hand, FMT is often accomplished using intrarectal delivery. Although this ensures that the bacteria are delivered to the distal GI tract, it is not a preferred administration route for most patients. Further, introducing an entire microbial community is associated with risks, as pathogens could also be transferred, producing unpredictable effects in their new host. Other microbiome-based therapeutic approaches include supplementation with prebiotics, which are fermentable substrates designed to promote the expansion of beneficial bacteria, or a combination of probiotics and prebiotics (synbiotics) (L. Du et al., 2022). While the administration of prebiotics can positively modulate the composition of the gut microbiota, it requires large amounts of supplementary material to be consumed.

For of all the aforementioned microbiome-based therapeutic strategies, interindividual variability in gut microbial composition is a significant limiting factor. The existing microbial community influences the efficacy of prebiotics, as they must be microbially degraded to produce their effects, and also influences the ability of live bacteria to effectively colonize the gut. As such, classical applications of microbiome-based therapeutics have resulted in limited and highly variable clinical efficacies. Growing evidence supports that microbial metabolite-based therapeutics can overcome these caveats. As microbial metabolites are largely responsible for the beneficial effects of commensal gut microbes, their administration may provide a more direct method of promoting host homeostasis. Their efficacies are not as dependent on the composition of the gut microbiota, and the potential for unwanted effects resulting from transplanting an entirely new microbial community are avoided. Moreover, while the phylogenetic composition of the human microbiota exhibits a high degree of interindividual variability, microbial metabolic pathways are largely stable among individuals, indicating that administering metabolites may be more universally applicable than targeting phylogeny (Huttenhower et al., 2012). Other noteworthy advantages of microbial metabolites include their relative ease of production, especially compared to administration of live bacteria that are not readily culturable, ability to be administered in specified doses, and suitability for administration by various routes. Lastly, microbial metabolites regulate diverse aspects of host physiology. Given the multifactorial nature of many chronic human diseases, novel therapeutics capable of targeting multiple features of disease pathogenesis, such as microbial metabolites, offer the opportunity to greatly improve clinical outcomes.

Like classical microbiome-based therapeutic approaches, there are challenges associated with administering microbial metabolites. As their beneficial effects are maximal in the distal gut, a major challenge lies in shielding them from degradation and absorption in the upper GI tract after oral administration. Rapid metabolism also limits their efficacy after parenteral administration, resulting high dosage requirements and frequent administrations. Pharmaceutical formulations approaches aiming to provide controlled and targeted microbial metabolite delivery have greatly contributed to the advancement of microbial metabolite-based therapeutics. Drug delivery systems such as enteric coating, esterification to dietary fiber, prodrugs, and nanoformulations have enabled the targeted and/or sustained delivery of microbial metabolites to their sites of action, resulting in improved therapeutic efficacy and reduced side effects. In this review, we first overview the roles of the two major classes of microbial metabolites, SCFAs and Trp metabolites, in maintaining host homeostasis. Then, we describe how gut dysbiosis in various disease contexts associates with dysregulated microbial metabolite production and outline preclinical and clinical studies demonstrating the ability of these metabolites to alleviate disease progression. Finally, we overview the novel pharmaceutical approaches to harness and deliver microbial metabolites that have brought us closer to realizing their therapeutic potential.

2. Roles of microbial metabolites in maintaining host homeostasis

Microbial metabolites are intermediates or end-products resulting from microbial metabolism of exogenous food molecules, endogenous host-derived compounds, or microbiota-derived compounds. These small molecules coordinate communication amongst gut microbes and between gut microbes and their host. The two most extensively studied classes are short-chain fatty acids (SCFAs) and tryptophan (Trp) metabolites, though there are many other products that play important roles in microbiota-host communication.

SCFAs are the end-products of bacterial fermentation of the otherwise indigestible carbohydrates of dietary fibers. The three major SCFAs produced are acetate, propionate, and butyrate. As energy sources, butyrate and propionate are largely utilized by the colonic epithelium and liver, respectively, while acetate reaches the systemic circulation in the highest concentrations (den Besten et al., 2013). SCFAs signal via activation of GPCRs, such as GPR43 (FFAR2), GPR41 (FFAR3) and GPR109A, and via inhibition of histone deacetylase (HDAC). While acetate is more selective for GPR43, butyrate preferentially binds GPR41, and propionate is the most potent agonist of both receptors (Le Poul et al., 2003). Butyrate is the only SCFA that can activate GPR109A (J. K. Tan et al., 2017). For HDAC inhibition, in vitro studies in both epithelial and immune cells have shown that butyrate is the most potent, followed by propionate, then acetate (Waldecker et al., 2008; Zou et al., 2021).

In recent years, Trp metabolites have also emerged as key metabolites produced in the colon as a result of microbial Trp catabolism. These metabolites include indole, tryptamine, indole-3-pyruvic acid (IPyA), indole-3-lactic acid (ILA), indole-3-acrylic acid (IA), indole-3-propionic acid (IPA), indole-3-ethanol (IE), indole-3-aldehyde (IAld), indole-3-acetic acid (IAA), and others, which predominantly act as ligands of the aryl hydrocarbon receptor (AHR) (Roager & Licht, 2018). Microbial catabolism of Trp reduces the amount Trp available for the indoleamine 2,3-dioxygenase (IDO)-1 enzyme expressed in host epithelial and immune cells, which catalyzes the production of kynurenine (Kyn) (Agus et al., 2018). Both SCFAs and Trp metabolites have been found to play crucial roles in regulating the functions of the intestinal epithelium, the immune system, and other processes in diverse tissues throughout the body (Figure 1). In this section, we aim to outline the key regulatory mechanisms by which these microbial metabolites regulate host homeostasis.

Figure 1: Effects of SCFAs and Trp metabolites on the intestinal epithelium, the immune system, and host metabolism.

SCFAs are generated via microbial fermentation of dietary fiber, while indole, tryptamine, and other microbial Trp metabolites are derived from dietary tryptophan. Oxidative metabolism of SCFAs after facilitated diffusion through MCT1 and SMCT1 as well as GPCR signaling contribute to enhanced intestinal barrier functions. Trp metabolites signal via AHR and PXR to modulate the cytoskeleton, upregulate TJP, and enhance IL-10 signaling. Both metabolite classes possess immunomodulatory functions including promotion of anti-inflammatory innate immune cell phenotypes, inducing Tregs and Bregs, and inhibiting pro-inflammatory effector T cells. Stimulation of IgA production by B cells serves to positively modulate microbiome composition against pathogens. SCFA-mediated promotion of energy expenditure in the liver, adipose tissue, and skeletal muscle occurs largely via AMPK-dependent mechanisms. Metabolites that escape metabolism in the gut and liver become available in small quantities in distant tissues, such as the lungs and brain. Graphic created with BioRender.com.

2.1. Effects on the intestinal epithelium

SCFAs

Intestinal epithelial cells (IECs) are the first cells to absorb SCFAs via facilitated diffusion through monocarboxylate transporter (MCT)-1 and sodium-dependent (S)MCT-1, where they fulfill as much as 60–70% of IEC energy requirements (den Besten et al., 2013). In humans, molar fractions of SCFAs in the hepatic portal vein were found to be ~70:20:10 for acetate:propionate:butyrate, compared to ~60:20:20 in the intestinal lumen, indicating preferential metabolism of butyrate by IECs (Cummings et al., 1987). Oxidative metabolism of butyrate and other SCFAs contributes to establishing the hypoxic colonic environment that is ideal for anaerobic bacteria that produce SCFAs via dietary fiber fermentation. Additionally, butyrate was found to modulate peroxisome proliferator-activated receptor γ (PPAR-γ) expression to drive epithelial β-oxidation (Byndloss et al., 2017). Oxygen consumption by IECs also leads to HIF stabilization, which plays a fundamental role in maintaining epithelial barrier integrity (Kelly et al., 2015). Via GPR43 and GPR109A signaling, SCFAs activate a well-characterized molecular mechanism of epithelial barrier maintenance, namely IL-18 upregulation resulting from NLRP3 inflammasome activation (Macia et al., 2015). Additionally, SCFAs are known to promote the assembly of tight junction protein (TJP) complexes and regulate TJP expression (Pérez-Reytor et al., 2021). As we will describe in the next section, fortification of intestinal barrier functions is a crucial mechanism by which SCFAs alleviate the progression of various diseases.

Besides regulating epithelial barrier integrity, SCFAs also modulate secretion of soluble mediators by IECs. Butyrate activates TGF-β production by IECs via HDAC inhibition, which supports immune tolerance by promoting the expansion of anti-inflammatory regulatory T cells (Tregs) (Martin-Gallausiaux et al., 2018). In the context of infections, however, SCFAs acutely stimulate mechanisms that afford protective immunity against pathogens. SCFA-mediated induction of cytokines, chemokines, and antimicrobial peptides (AMPs) in IECs are mediated through GPR41 and GPR43, which activate mitogen/extracellular signal protein kinase (MEK)/extracellular signal regulated kinase 1/2 (ERK) and p38/mitogen activated protein (MAP) kinase pathways (M. H. Kim et al., 2013; Schauber et al., 2003; Y. Zhao et al., 2018).

Trp metabolites

In recent years, growing evidence has revealed the roles of microbially-produced Trp metabolites in maintaining intestinal epithelial homeostasis. The microbial Trp metabolite indole was found to both enhance barrier functions and attenuate inflammation in IECs (Bansal et al., 2010). Indole increased the expression of TJP, which corresponded with increased transepithelial resistance. Since then, other Trp metabolites have also been shown to positively modulate intestinal barrier functions. IE, IPyA, and IAld were shown to modulate the contractility of the actin cytoskeleton via AHR signaling to protect against barrier disruption (Scott et al., 2020). Interestingly, IAld was recently found to reverse the deleterious effects of aging on the intestinal epithelium by increasing proliferation and goblet cell differentiation (Powell et al., 2020). In this study, IAld’s effect on goblet cell differentiation was dependent on upregulation of IL-10, which is largely produced by anti-inflammatory Tregs. Additionally, AHR activation by IA was found to promote both goblet cell differentiation and mucus production (Wlodarska et al., 2017). On the other hand, evidence suggests that IPA promotes epithelial integrity by signaling via the pregnane X receptor (PXR) (Dodd et al., 2017; Venkatesh et al., 2014). IPA-mediated activation of PXR signaling restrains immune responses in the intestinal mesenchyme, thereby reducing inflammation and dampening fibrosis (Flannigan et al., 2023). Trp metabolites may also contribute to establishing anti-inflammatory IEC phenotypes as both IPA and IAld were found to induce the expression of IL-10R1, the receptor for IL-10 (Alexeev et al., 2018).

2.2. Effects on the immune system

SCFAs

Immunological tolerance towards harmless bacterial- and food-derived enteric antigens, as well as self-antigens, is essential for maintaining host homeostasis. Disruption of immune tolerance mechanisms can lead to the development of chronic inflammation, autoimmunity, and allergies. SCFAs support immune tolerance by modulating both adaptive (T cells and B cells) and innate (dendritic cells (DCs) and macrophages) immune cells towards more tolerogenic phenotypes.

Regulatory T cells (Tregs) negatively regulate the activities of effector T cells, such as Th1, Th2 and Th17 cells, and thus are essential for maintaining immune tolerance. SCFA production by gut commensals promotes the expansion of Foxp3⁺ Tregs and their production of the anti-inflammatory cytokine IL-10 (Arpaia et al., 2013; Furusawa et al., 2013; Smith et al., 2013). They accomplish this, in part, by functioning as HDAC inhibitors to epigenetically regulate histone acetylation at the Foxp3 locus. Additionally, SCFA-mediated upregulation of TGF-β, a Treg-polarizing cytokine, contributes to Treg induction. As we will discuss in the next section, SCFAs also inhibit the differentiation of pro-inflammatory effector T cells in a variety of disease contexts. Through these mechanisms, SCFAs drive the Treg/T effector cell balance towards a more tolerogenic profile.

Additionally, tolerogenic T cell profiles are established by SCFAs through their regulation of innate immune cells. SCFAs have consistently been found to promote tolerogenic DC and macrophage phenotypes, rendering them more potent inducers of Tregs and weaker inducers of effector T cells (Corrêa-Oliveira et al., 2016). A key feature of tolerogenic DCs and macrophages is their production of retinoic acid (RA). Evidence supports both GPCR signaling and HDAC inhibition as mechanisms by which SCFAs activate RA production in these cell types (Kaisar et al., 2017; Wu et al., 2017). SCFA treatment also results in decreased co-stimulatory marker expression (Millard et al., 2002) and more anti-inflammatory cytokine secretion profiles (Chang et al., 2014; Nastasi et al., 2017). Notably, butyrate was found to be more potent than propionate at inhibiting pro-inflammatory cytokine secretion by both macrophages and DCs (Ciarlo et al., 2016).

SCFAs also regulate the functions of B cells, the other major adaptive immune cell type. Like Tregs, regulatory B cells (Bregs) are immunosuppressive cells that support immunological tolerance (Rosser & Mauri, 2015). By releasing IL-10 and TGF-β, Bregs contribute to establishing tolerogenic T cell profiles. SCFAs have been shown to stimulate Breg generation, where the mechanism for butyrate and propionate is dependent on their HDAC inhibitory activities (Zou et al., 2021). On the other hand, the conversion of acetate to acetyl-CoA drove Breg differentiation by fueling the TCA cycle and enabling post-translational acetylation of cytoplasmic proteins (Daïen et al., 2021). SCFAs also promote B cell differentiation to antibody-producing plasma cells (PCs). SCFAs stimulated the differentiation of IgA- and IgG-producing PCs by regulating gene expression and by stimulating metabolism to support the heightened energy requirements of antibody production (M. Kim et al., 2016). IgA production by PCs modulates the composition of the microbiota, enhancing commensal colonization while reducing that of pathogens (Kawamoto et al., 2012, 2014; M. Kim et al., 2016; Uchimura et al., 2018). Further regulation of IgA responses is provided by T follicular helper (TFH) and T follicular regulatory (TFR) cells (J. Tan et al., 2016). While TFH cells support B cell differentiation and antibody production, TFR cells check these processes. Imbalanced TFR/TFH cell profiles have been linked with certain autoimmune diseases. As we will describe in the next section, HDAC-inhibiting SCFAs have been shown promote TFR differentiation in preclinical autoimmune disease models, which is associated with inhibited PC differentiation and impaired autoantibody responses (Sanchez et al., 2020; Takahashi et al., 2020).

While the immunosuppressive effects of SCFAs are advantageous in ameliorating chronic inflammation, they are undesirable in the context of infection. Remarkably, SCFAs can also acutely enhance intestinal immunity to promote colonization resistance against pathogens. During an infection when an active immune response is needed, SCFAs induce Th1 and Th17 cell differentiation to promote protective immunity (Park et al., 2015). Additionally, SCFAs induce the differentiation of Tregs even in Th1 and Th17-polarizing conditions, which restrains the development of chronic inflammation. In another mechanism, butyrate and propionate induce neutrophil migration to inflammatory sites, augment their ability to fight infections (Vinolo et al., 2009), while also diminishing their production of pro-inflammatory cytokines, thereby alleviating inflammation (Vinolo et al., 2011). Acetate also enhances the bactericidal effects of both neutrophils and macrophages by inducing the release of reactive oxygen species (ROS) and stimulating phagocytic activity (Machado et al., 2022; Maslowski et al., 2009).

Lastly, SCFAs regulate innate lymphoid cells (ILC), the innate counterparts of T cells that play a crucial role in regulating the functions of intestinal epithelial and immune cells. ILCs provide an early source of IL-22, a crucial cytokine for epithelial homeostasis as it promotes AMP production and enhances barrier functions, before CD4⁺ T cells take over as the dominant source of IL-22 during active infections. SCFAs have been found to promote IL-22 production by both of these cell types, conferring protection against infection (W. Yang et al., 2020). In the absence of an active infection, however, butyrate instead suppresses ILCs. Type 2 ILCs (ILC2s) are a critical source of cytokines that drive Th2 responses, namely IL-13 and IL-5. Butyrate treatment markedly reduced expression of these cytokines in ILC2s, highlighting its potential to treat allergic diseases characterized by hyperactive Th2 responses (Thio et al., 2018). Additionally, evidence suggests that butyrate also suppresses type 3 ILCs (ILC3s) in Peyer’s patches along the intestine, reducing immune responses to harmless enteric antigens (S.-H. Kim et al., 2017).

Trp metabolites

Although the immunomodulatory functions of microbially-produced Trp metabolites are just beginning to be studied in-depth, the past decade has seen an increasing recognition of their roles in host immune regulation. AHR is widely expressed by immune cells and is a key mediator of microbiota-host immune interactions (J. Gao et al., 2018). As mentioned, IL-22 is a crucial protective cytokine expressed by ILCs and T cells. An early study revealed the ability of microbially-produced Trp catabolites to regulate the immune system of their host, demonstrating that IAld produced by Lactobacilli induces IL-22 transcription via AHR (Zelante et al., 2013). This functions to provide resistance against infection while protecting against intestinal inflammation. Since then, other studies have identified other AHR-activating microbial Trp metabolites, such as IAA and IA, that also upregulate IL-22 levels (D. Liu et al., 2023; X. Zhang et al., 2023). Importantly, reduced IL-22 signaling in AHR-deficient mice was associated with deficient ILC3 cells, which increased their susceptibility to enteric infection (Qiu et al., 2013).

Like SCFAs, microbial Trp metabolites have been shown to drive tolerogenic T cell profiles. For example, tryptamine has been shown to induce Foxp3⁺ Tregs (Dopkins et al., 2020). IPyA was found to modulate T cell differentiation, where its induction of Foxp3⁺ Tregs vs. Foxp3⁻ IL-10⁺ T cells and its inhibition of Th17 cells vs. Th1 cells vary in different disease models (Aoki et al., 2018; T. Huang et al., 2023). Additionally, Lactobacillus (L).murinus-derived ILA inhibits Th17 cell polarization, and the presence of L.murinus determines intestinal Th17 frequencies (Wilck et al., 2017). Finally, L.reuteri-derived IAld and ILA stimulate the differentiation of immunoregulatory CD4⁺CD8αα⁺ double-positive intraepithelial T lymphocytes through activation of AHR in CD4⁺ T cells (Cervantes-Barragan et al., 2017).

Regarding innate immune cells, AHR agonists are well-supported to induce tolerogenic DC phenotypes (Quintana et al., 2010; C. Wang et al., 2014). However, the roles of specific microbial Trp metabolites on DC phenotypes remain understudied. So far, one study has demonstrated that IPyA attenuates the ability of DCs in the mesenteric lymph nodes (mLNs) to induce Th1 differentiation (Aoki et al., 2018). On the other hand, microbial Trp metabolites are well-supported to positively modulate macrophage functions. IA was found to suppress TNF-α while inducing IL-10 in cultured macrophages (Wlodarska et al., 2017). Both tryptamine and IAA were also found to inhibit pro-inflammatory cytokine production as well as chemotaxis in macrophages, with IAA exhibiting greater potency (Krishnan et al., 2018). However, like those of SCFAs, the immunomodulatory effects of Trp metabolites are context dependent. In the context of infection, IPA was shown to instead increase the pro-inflammatory and phagocytic capacities of macrophages (Z.-B. Huang et al., 2022). Finally, microbially produced indoles were found to inhibit the myeloperoxidase activity of neutrophils and their production of pro-inflammatory cytokines, mitigating their tissue-damaging effects during acute inflammatory responses (Alexeev et al., 2021).

2.3. Effects on host metabolism

SCFAs

A major role of SCFAs is to regulate host metabolic homeostasis by modulating metabolic processes in the liver, adipose tissue, and skeletal muscle (den Besten et al., 2013). In these tissues, SCFAs activate AMP-activated protein kinase (AMPK), which activates PPAR-γ coactivator (PGC)-1α (Z. Gao et al., 2009; Kondo, Kishi, Fushimi, & Kaga, 2009; Lu et al., 2016). PGC-1α regulates the transcriptional activity of not only PPAR-γ, but also PPAR-α and PPAR-β/δ. This mainly results in upregulation of fatty acid oxidation and thermogenesis, leading to increased energy expenditure. Interestingly, host metabolic homeostasis has been shown to be supported both by SCFA-mediated upregulation of PPAR-α (Kondo, Kishi, Fushimi, & Kaga, 2009; B. Sun et al., 2018) and downregulation of PPAR-γ (den Besten et al., 2015). These mechanisms ultimately stimulate the expression of carnitine palmitoyl transferases (CPT) and mitochondrial uncoupling proteins (UCP) to promote energy expenditure (den Besten et al., 2015; Kondo, Kishi, Fushimi, & Kaga, 2009; Lu et al., 2016). Additionally, SCFAs may play a role in appetite regulation. Colonic-derived acetate in the systemic circulation was found to cross the blood-brain barrier and inactivate AMPK within the CNS (Frost et al., 2014). This led to acetyl-CoA carboxylase (ACC) activation, which induced the expression of appetite-suppressing regulatory neuropeptides. On the other hand, butyrate reduced food intake after oral, but not intravenous, administration, suggesting that it activates a gut-brain axis mechanism (Z. Li et al., 2018). One gut-brain axis mechanism may be SCFA-mediated induction of glucagon-like peptide (GLP)-1 and peptide YY (PYY) expression in the enteroendocrine L cells of the intestinal epithelium, which act within the brain to reduce appetite (Christiansen et al., 2018; Psichas et al., 2015; Tolhurst et al., 2012).

Trp metabolites

Recent evidence suggests that Trp metabolites also play a role in regulating metabolic homeostasis (Qi et al., 2022). Like SCFAs, indole was also found to modulate the secretion of GLP-1 in enteroendocrine L cells (Chimerel et al., 2014). As we will discuss in the next section, administration of various Trp metabolites in preclinical metabolic models has greatly contributed to elucidating the mechanisms by which these metabolites reduce inflammation associated with metabolic dysregulation. However, unlike the mechanisms for SCFAs, those by which Trp metabolites might directly regulate host metabolism in the liver, adipose tissue, or other tissues remain ill-defined. One recent study demonstrated that IDO-1 deletion or inhibition improves insulin sensitivity and regulates lipid metabolism in the liver and adipose tissue (Laurans et al., 2018). These beneficial effects were associated with IL-22 induction, which is known to be induced by microbial Trp metabolites. These findings suggest that increasing Trp availability for gut microbes upon loss of IDO-1 may regulate host metabolism via microbial Trp metabolites. Nonetheless, further research is needed to delineate the precise mechanisms involved.

3. Gut dysbiosis in disease and therapeutic applications of microbial metabolites

Gut dysbiosis, which describes compositional alterations to the gut microbiota, has been linked to myriad inflammatory, metabolic, autoimmune, allergic, and other diseases. The incidence of these multifactorial, chronic conditions has risen alongside lifestyle, environmental, and dietary changes that impact the gut microbiota, including improved hygiene, increased antibiotic use, air pollution, and the transition from Paleolithic nutrition to Westernized diets. Dysbiosis can be described by 1) reduction of α diversity, which describes the species richness within the GI tract, 2) proliferation of typically low-abundant, pathogenic microbes, and/or 3) loss of typically high-abundant, beneficial microbes (Levy et al., 2017). These compositional alterations are inherently accompanied by over or under-abundances of microbial metabolites. Given the challenges associated with classical approaches to reverse dysbiosis, administering microbial metabolites has emerged as an attractive therapeutic strategy. Direct administration of metabolites circumvents the need to understand the totality of effects that may occur after administering live bacteria. Additional advantages include their relative ease of production, stability during long-term storage, and ability to be formulated for various routes of administration.

In this section, we aim to provide an overview of the key findings demonstrating how microbial metabolites and the bacterial species that produce them are altered in various disease states. We further summarize preclinical and clinical studies showing how the in vivo administration of microbial metabolites influences the progression of these diseases, with a focus on SCFAs and Trp metabolites.

An overview of important microbes in the human gut

The composition of a healthy human gut microbiota is 90% comprised of two phyla, Bacteroidetes and Firmicutes, while Verrucomicrobia, Actinobacteria, and Proteobacteria are less represented (Arumugam et al., 2011). Firmicutes contains many of the key families involved in SCFA production (Fusco et al., 2023). Members of Clostridium clusters XIVa and IV from the Clostridiaceae, Ruminococcaceae, and Lachnospiraceae families are the most well-known SCFA producers of this phylum. Decreases in the relative abundance Firmicutes (increased Bacteroidetes-to-Firmicutes ratio) are associated with many chronic disease states. Other important SCFA contributors are Prevotellaceae (in Bacteroidetes phylum) and Bifidobacteriaceae (in Actinobacteria phylum). Many of these same families, such as Clostridiaceae and Bifidobacteriaceae, also produce Trp metabolites (Roager & Licht, 2018). Additionally, certain Lactobacillus species, such as L.reuteri and L.murinus, are especially important for the generation of certain Trp metabolites.

Many bacteria within the Proteobacteria phylum, including those from the Enterobacteriaceae family, are opportunistic pathogens. As these species are facultative anaerobes that can generate ATP aerobically when oxygen is present, Proteobacteria expansion has been suggested to be a signature of epithelial dysfunction as it indicates a disruption in the normally hypoxic environment of the colon (Litvak et al., 2017). Epithelial oxygenation compromises the survival ability of obligate anaerobes such as Bacteroidetes and Firmicutes, resulting in altered microbial metabolite production profiles (Vester-Andersen et al., 2019). Such dysbiosis contributes to chronic disease by negatively impacting epithelial barrier functions, intestinal immune tolerance, and other host processes.

3.1. Inflammatory bowel diseases (IBD)

Inflammatory bowel diseases (IBD) are primarily comprised of two conditions, Crohn’s disease and ulcerative colitis, which are characterized by chronic inflammation and epithelial barrier disruption in the GI tract. IBD can involve inappropriate immune responses to otherwise harmless antigens in the intestinal environment, which can be gut microbes, food components, or even host tissues. Current treatment strategies include anti-inflammatory drugs (e.g. aminosalicylates), immunosuppressants (e.g. corticosteroids), immunomodulators (e.g. methotrexate), and biologics (e.g. anti-TNF antibodies). Additionally, lifestyle and dietary modifications are often recommended. A major challenge is targeting the delivery of therapeutics to the distal GI tract, the region most severely affected by inflammation. Thus, medications can be administered via rectal enema, which is not a preferred administration route for most patients and has therefore been associated with low patient compliance. As such, novel pharmaceutical approaches that enable distal gut-targeting after oral administration and deliver therapeutic payloads capable of modulating multiple facets of immune dysregulation would greatly improve clinical IBD management.

While IBD is driven by a complex interplay between many factors, there is substantial evidence to support a pivotal role for the gut microbiota in the development and progression of IBD. Gut dysbiosis in IBD patients is characterized by significant shifts in the diversity, composition, and metabolite production profile of gut commensals (Golob et al., 2023; Lavelle & Sokol, 2020; Santana et al., 2022). Research has consistently demonstrated that the gut microbiome of IBD patients exhibits reduced α diversity, signifying a reduction in the variety of species present. Compositionally, the IBD-associated microbiota displays increased abundances of Bacteroidetes and Proteobacteria and a decreased abundance of Firmicutes compared to that of control individuals (Litvak et al., 2017; Santana et al., 2022). A key component of dysbiosis in IBD is reduced microbial metabolite production, which we describe here with a focus on SCFA and Trp metabolites. We also outline studies demonstrating the therapeutic potential for administering these metabolites to individuals with IBD.

SCFAs

SCFA-producing bacteria belonging to Clostridium clusters XIVa and IV are commonly reported to be decreased in IBD. In particular, reductions in Faecalibacterium prausnitzii, a key butyrate-producer in this group, are linked to IBD and inversely correlate with disease activity (Duboc et al., 2013; Machiels et al., 2014; Prosberg et al., 2016; Sokol et al., 2009). A reduced genetic capacity for butyrate synthesis by the microbiome is further evidenced by lower butyryl-CoA:acetate-CoA transferase gene ratios in fecal samples from IBD patients (Laserna-Mendieta et al., 2018). In accordance with reduced abundances of SCFA producers, fecal levels of SCFAs in IBD patients are lower than in healthy controls (Bjerrum et al., 2015; Parada Venegas et al., 2019). Additionally, Bifidobacteria, known to produce acetate and lactate, are frequently reported to be reduced in IBD (Prosberg et al., 2016; Sokol et al., 2009).

SCFA administration has been investigated in preclinical IBD models, where most studies report an amelioration of colitis symptoms upon treatment. SCFAs have been shown to enhance epithelial barrier functions, attenuate inflammation, and drive tolerogenic immune cell profiles in the intestinal environment. Signaling through GPCRs, especially GPR109A-mediated IL-18 induction by butyrate, has been shown to mediate the beneficial effects of SCFAs on epithelial barrier functions in colitis models (Macia et al., 2015; Singh et al., 2014). Additionally, SCFAs inhibit NF-κB activation in both epithelial and immune cells, thereby mitigating colitis-induced inflammation (G. Chen et al., 2018; C. Lee et al., 2017). Foxp3⁺ Treg cells are induced by SCFAs and function to moderate immune responses in colitis models, partly through their production of IL-10 (Arpaia et al., 2013; Furusawa et al., 2013; M. Zhang et al., 2016). Butyrate also enhances IL-10 expression in Th1 cells and Bregs, contributing to its anti-inflammatory role in colitis (M. Sun et al., 2018; Zou et al., 2021). Interestingly, butyrate was still protective against colitis in IL-10-knockout mice, suggesting that some of its beneficial effects are independent of IL-10 upregulation (C. Lee et al., 2017). In line with this, butyrate was found to upregulate TGF-β and downregulate IL-6 to drive Treg rather than Th17 cell differentiation, resulting in reduced pro-inflammatory IL-17 levels (M. Zhang et al., 2016). Amelioration of colitis by butyrate has also been linked to its promotion of alternatively-activated, anti-inflammatory macrophage phenotypes (J. Ji et al., 2016). Lastly, butyrate treatment significantly reduced pro-inflammatory cytokine production and migration in neutrophils both from colitic mice and from IBD patients, indicating its potential to inhibit neutrophil-associated immune responses during chronic intestinal inflammation (G. Li et al., 2021).

Other SCFAs, such as propionate and acetate, are also ameliorative in experimental colitis models. Importantly, both are also capable of inducing Foxp3⁺ Treg cells (Arpaia et al., 2013). Acetate administered either orally or systemically (via i.p. injection) attenuated intestinal inflammation in colitis models through GPR43 signaling (Ishiguro et al., 2007; Macia et al., 2015; Maslowski et al., 2009). Acetate’s protective effects on epithelial integrity might be partially mediated by the upregulation of IL-18 production in IECs via activation of the NLRP3 inflammasome (Macia et al., 2015). In a T cell transfer colitis model, administration of propionate or SCFA mixtures effectively inhibited disease progression only in mice with less severe colitis, underlining the importance of considering disease severity when evaluating the efficacy of SCFA treatment (Smith et al., 2013).

As orally administered SCFAs are rapidly absorbed or metabolized before reaching the distal gut, preclinical studies outlined above have relied on high concentrations and prolonged periods of exposure to achieve therapeutic efficacy. As such, clinical studies in IBD patients have historically investigated the efficacy of SCFAs administered directly to the colon via rectal enema (Table 1). While significant improvements were observed in preliminary studies with small IBD cohorts that received enemas containing SCFA mixtures (80 mM acetate, 30 mM propionate, 40 mM butyrate), a larger randomized, double-blind, placebo-controlled trial did not indicate a therapeutic value for the same SCFA enemas (Breuer et al., 1991, 1997; Guillemot et al., 1991; Patz et al., 1996; Vernia et al., 1995). Other studies employing butyrate-only enemas at higher concentrations (80–100 mM) similarly reported inconsistent results (Hamer et al., 2010; Lührs et al., 2002; Scheppach, 1996; Scheppach et al., 1992; Steinhart et al., 1996; Vernia et al., 2003). Given that intrarectal delivery is not a preferred route of administration, more practical approaches for efficient and prolonged SCFA delivery are needed to translate preclinical findings into long-term therapeutic benefits for IBD patients. Novel pharmaceutical approaches for oral SCFA administration and their clinical applications in IBD patients are described in the next section.

Table 1:

Administration of free microbial metabolites in clinical studies

Patients	Metabolite(s)	Admin. route, dose, regimen	Key findings (ClinicalTrials.gov identifier and/or references)
Inflammatory bowel diseases (IBD)
IBD	SCFA mixtures	Enema, 100 mL containing sodium acetate (80 mM), sodium propionate (30 mM), and sodium butyrate (40 mM), twice daily	Improved clinical and histological disease activity scores (Breuer et al., 1991; Vernia et al., 1995; Patz et al., 1996).
			Did not significantly improve clinical and histological activity scores (Breuer et al., 1997; Guillemot et al., 1991).
	Butyrate	Enema, 60–80 mL containing 80–100 mM sodium butyrate, once daily	Improved clinical and histological activity scores (Scheppach et al., 1992; Vernia et al., 2003) and reduced NF-kB activation in macrophages (Lührs et al., 2002).
			Did not significantly improve clinical and histological disease activity scores (Scheppach, 1996; Steinhart et al., 1996) and showed only minor effects on inflammatory and oxidative stress parameters (NCT00696098, Hamer et al., 2010).
Metabolic diseases
Obesity	SCFA mixtures	Enema, 200 mL containing sodium acetate (18–24 mM), sodium propionate (8–14 mM), and sodium butyrate (8–14 mM), twice only	Increased fasting fat oxidation, energy expenditure, and PYY, and decreased lipolysis (Canfora et al., 2017).
	Acetate	Colonic infusion, 100–180 mM sodium acetate, 3d duration	Distal, but not proximal, infusions increased fasting fat oxidation, PYY, and postprandial glucose and insulin levels (NCT01826162, van der Beek et al., 2016).
		p.o., 750–1500 mg/d acetic acid as dietary vinegar	Reduced body weight, visceral fat, and triglyceride levels (Kondo, Kishi, Fushimi, Ugajin, et al., 2009).
Type 2 diabetes (T2D)		p.o., ~1000–1500 mg acetic acid/d as dietary vinegar	Reduced postprandial blood glucose and insulin levels and improved insulin sensitivity in insulin-resistant (non-T2D) subjects (Johnston et al., 2004) and T2D subjects (NCT02309424, Mitrou et al., 2015; Liatis et al., 2010).
			Reduced fasting blood glucose levels in T2D subjects (White & Johnston, 2007; Gheflati et al., 2019).
			Reduced HbA1c levels in T2D subjects (Johnston et al., 2009).
			Did not improve insulin sensitivity or reduce postprandial glucose or insulin levels in T2D subjects (Johnston et al., 2004; van Dijk et al., 2012).
Atherosclerosis	Propionate	p.o. 500 mg calcium propionate in capsules, BID	Reduced total and LDL cholesterol levels (NCT03590496, Haghikia et al., 2022).
Autoimmune diseases
Multiple sclerosis (MS)	Propionate	p.o. 500 mg sodium propionate in capsules, BID	Reduced annual relapse rates, disease progression, and brain atrophy, promoted regulatory T cell profile (Duscha et al., 2020).
			Modulated osteoporosis biomarkers toward increased bone formation and reduced bone resorption (Duscha et al., 2022).
Type 1 diabetes (T1D)	Butyrate	p.o. 4g/d sodium butyrate microgranules in capsules	Did not significantly affect glucose regulation or measures of innate and adaptive immunity (de Groot et al., 2020).
Infectious diseases
Shigellosis	Butyrate	Enema, 80 mM sodium butyrate, BID	Led to early reduction of inflammation and enhanced anti-microbial peptide expression, but did not significantly improve clinical disease scores (NCT00800930, Raqib et al., 2012).
Cancer
Acute myeloid leukemia	Butyrate	i.v. infusion, 500 mg/kg/d sodium butyrate	Lack of clinical efficacy. Low plasma levels due to rapid clearance/short in vivo half-life (Miller et al., 1987)
Metastatic colorectal cancer		i.v. infusion, 2000 mg/kg/d arginine butyrate (MTD)	Lack of clinical efficacy. Dose escalation could not be performed due to toxicities, including fatigue and liver toxicity (Douillard et al., 2000).
Epstein-Barr virus-associated lymphoid malignancies		i.v. infusion, 1000 mg/kg/d arginine butyrate (MTD)	In combination with the anti-viral ganciclovir, was well-tolerated and induced significant anti-tumor responses (NCT00006340, Perrine et al., 2007).

Trp metabolites

Various studies have demonstrated that microbial Trp catabolism is altered in IBD patients. Compared with fecal samples from healthy patients, those from IBD patients typically have reduced levels of Trp metabolites, including indole, IPA, and IAA, while levels of Kyn and its metabolites are increased (Flannigan et al., 2023; Lamas et al., 2016; Nemoto et al., 2012; Nikolaus et al., 2017). Moreover, IBD patients exhibit upregulation of IDO-1, the enzyme responsible for Kyn metabolism (Lamas et al., 2016; Nikolaus et al., 2017; L.-P. Zhao et al., 2022). This suggests that impaired production of Trp metabolites by the dysbiotic IBD microbiome is associated with increased Trp utilization via the host IDO-1-mediated Kyn pathway. In accordance with lower IPA production, expression of PXR and its target genes is suppressed in the intestinal tissues of IBD patients (Flannigan et al., 2023). Notably, both in mice with dextran sodium sulfate (DSS)-induced colitis and in patients with IBD, both serum and fecal IPA concentrations are reduced (Alexeev et al., 2018; Flannigan et al., 2023).

Preclinical studies have demonstrated that administering microbial Trp metabolites can alleviate the symptoms of IBD. Oral administration of IPA to colitic mice attenuated intestinal pro-inflammatory cytokine production (Alexeev et al., 2018), and inhibited neutrophil-mediated tissue damage by inhibiting the MPO enzyme (Alexeev et al., 2021). Modulation of the actin cytoskeleton by IE, IPyA, and IAld protected against intestinal barrier disruption during colitis induction (Scott et al., 2020).

Both IAId and IAA modulated IL-22 production by intestinal immune cells, mitigating colonic inflammation in colitic mice (Zelante et al., 2013; X. Zhang et al., 2023). In a T cell transfer colitis model, IPyA abrogated inflammation by increasing the frequency of IL-10⁺ T cells and decreasing the frequency of Th1 cells (Aoki et al., 2018). Interestingly, IPyA did not affect the frequency of Foxp3⁺ Tregs or IL-17⁺ Th17 cells, which, as described above, are known to be modulated by SCFAs in IBD models. This suggests that AHR-agonizing Trp metabolites and SCFAs may complement one another in dampening IBD-associated intestinal inflammation when co-administered.

As we will outline later, a few of these Trp metabolites have been formulated to enable their distal gut delivery after oral administration using pharmaceutical approaches, which have successfully ameliorated IBD symptoms in preclinical models.

3.2. Metabolic diseases

Metabolic diseases are a group of disorders characterized by alterations in the process of converting food into energy, resulting in disrupted metabolic homeostasis. The major conditions include obesity and type 2 diabetes (T2D), followed by non-alcoholic fatty liver disease and steatohepatitis (NAFLD/NASH) and atherosclerosis. Each of these conditions is associated with chronic, low-grade inflammation, termed “metaflammation”, that can occur in the adipose tissue, liver, and/or blood vessel walls. Additionally, insulin resistance, dyslipidemia, and hepatic steatosis other are common features of metabolic diseases. In the cases of obesity and NAFLD/NASH, there are currently no approved medications available as therapeutics, rendering lifestyle modifications the primary treatment approach. In addition to lifestyle modifications, patients with T2D must continually monitor their blood sugar levels and control them with oral (metformin) or injectable (insulin) medications, while patients with atherosclerosis are prescribed cholesterol-lowering statins and/or blood pressure medications. While these interventions are largely effective in treating one aspect of metabolic dysfunction, novel therapeutics that restore whole-body metabolic homeostasis via multiple complimentary mechanisms are needed.

Gut dysbiosis is now recognized as a significant contributing factor to various metabolic disorders (Dabke et al., 2019). There is a clear correlation between the α diversity of the gut microbiome and metabolic markers, suggesting that individuals with low microbial richness present more pronounced dysmetabolism and insulin resistance (Le Chatelier et al., 2013). According to one theory, compromised intestinal barrier integrity resulting from gut dysbiosis allows gut microbes and harmful microbial byproducts such as lipopolysaccharide (LPS) to permeate into the bloodstream (Musso et al., 2011). Subsequently, these molecules may trigger metabolic dysfunction by contributing to chronic inflammation in metabolically-relevant tissues, such as the liver and adipose tissue. Thus, therapeutics that counteract these processes by strengthening the epithelial barrier and reducing inflammation in both the GI tract and other metabolically-active tissues are strong candidates for treating metabolic diseases. Here, we review evidence describing dysbiosis, altered production of SCFAs and Trp metabolites, and the potential of administering microbial metabolites therapeutically in the context of metabolic diseases.

SCFAs

In contrast to IBD, the proportion of Firmicutes members is often reported to increase in obesity (Pinart et al., 2021; Turnbaugh et al., 2009). This compositional alteration may enable the host to extract more energy from their diet, as SCFA-producing Firmicutes members crucially support host energy production. In accordance with this notion, the gut microbiomes of individuals with metabolic diseases exhibit increased genetic capacities for energy harvest, including increased expression of KEGG orthologues for starch metabolism/SCFA production (Karlsson et al., 2013; Turnbaugh et al., 2006). Additionally, higher concentrations of SCFAs are found in fecal samples from obese patients and genetically obese (ob/ob) mice (Schwiertz et al., 2010; Turnbaugh et al., 2006). Despite these findings, decreased abundances of certain SCFA-producing bacteria within Firmicutes are also commonly observed in obese and T2D microbiomes (Forslund et al., 2015; Karlsson et al., 2013; Le Chatelier et al., 2013; J. Qin et al., 2012). Furthermore, the potential of the gut microbiome to produce butyrate and propionate was notably higher in T2D patients treated with metformin, suggesting that some therapeutic effects of metformin might arise from its modulation of microbial SCFA production (Forslund et al., 2015).

Although the effects of metabolic disease on the abundance of SCFA-producing bacteria remain to be fully elucidated, preclinical studies have repeatedly demonstrated that oral administration of acetate, propionate, butyrate, or their admixture attenuates body weight gain and improves insulin sensitivity after high-fat diet (HFD) feeding (den Besten et al., 2015; Z. Gao et al., 2009; Kondo, Kishi, Fushimi, & Kaga, 2009; Z. Li et al., 2018; H. V. Lin et al., 2012; Lu et al., 2016; Sahuri-Arisoylu et al., 2016). The major findings from these studies will be briefly summarized here. Via AMPK-dependent mechanisms outlined in the previous section, SCFAs elevate energy expenditure in the adipose tissue, liver, and skeletal muscle. As a result of increased thermogenic activity, SCFA administration leads to reduced adipocyte size and whole-body adiposity. Also in the adipose tissue, SCFAs consistently induce the expression of the satiety hormone leptin. Improvements in insulin sensitivity are consistently indicated by reduced fasting insulin levels and improved HOMA-IR values. However, it should be noted that findings from these studies regarding the impact of SCFAs on fasting plasma glucose levels and food intake are inconsistent. However, in genetically diabetic (db/db) mice, butyrate consistently led to decreased fasting blood glucose levels, along with improved liver function and reduced adipose tissue inflammation (X. Wang et al., 2015; Y.-H. Xu et al., 2018; W.-Q. Zhang et al., 2019). Additionally, SCFAs were found to increase Tregs, decrease Th1 and Th17 cells, and decrease serum pro-inflammatory cytokines to reduce “metaflammation” in HFD-fed mice (Mandaliya et al., 2021). Finally, butyrate upregulates TJP expression to ameliorate obesity-associated intestinal barrier disruption (Fang et al., 2019; Y.-H. Xu et al., 2018; D. Zhou et al., 2017).

HFD feeding also causes liver inflammation, which models NAFLD/NASH. In HFD-induced NAFLD/NASH models, administration of butyrate mitigates liver inflammation, lipid accumulation, and hepatic steatosis while reducing circulating ALT and AST levels (Fang et al., 2019; B. Sun et al., 2018; D. Zhou et al., 2017). In HFD-fed ApoE knockout mice, a preclinical model of obesity-associated atherosclerosis, butyrate induced cholesterol efflux in macrophages, resulting in fewer, more stable plaques (Y. Du et al., 2020). Further, butyrate downregulated endothelial expression of the chemokine MCP-1/CCL2, the cell adhesion molecule VCAM-1, and matrix metalloproteinase (MMP)-2 to slow macrophage infiltration, mediated through NF-κB inhibition (Aguilar et al., 2014). In the livers of methionine and choline-deficient diet (MCD)-induced NASH mice, butyrate induced pro-healing macrophage phenotypes, inhibited macrophage infiltration, and induced apoptosis in pro-inflammatory macrophages (Sarkar et al., 2023).

Clinical studies demonstrate that administration of SCFAs benefits patients with metabolic disorders (Table 1). In contrast to IBD, most of these studies administered SCFAs via the oral route, and there is considerably more congruency among the findings. The effects of acetate administration in the form of acetic acid via dietary vinegar consumption have been tested in obese, insulin-resistant (non-T2D), and T2D patients (Gheflati et al., 2019; Johnston et al., 2004, 2009; Kondo, Kishi, Fushimi, Ugajin, et al., 2009; Liatis et al., 2010; Mitrou et al., 2015; van Dijk et al., 2012; White & Johnston, 2007). This has led to reductions in body weight, visceral fat, and serum triglycerides in obese adults and improved insulin sensitivity in insulin-resistant (non-T2D) and T2D subjects. Acetic acid appears to be well-tolerated, has no adverse side effects, and has clinical potential to ameliorate both obesity and T2D, but is rapidly absorbed before reaching the distal GI tract (Valdes et al., 2021). Notably, distal colonic acetate infusions promoted fasting fat oxidation and energy expenditure in overweight/obese men while proximal infusions did not (van der Beek et al., 2016). Further, delivering SCFA mixtures directly to the distal regions of the colon via rectal enema produced similar beneficial effects on fat oxidation in energy expenditure in a similar population (Canfora et al., 2017). These findings suggest that the efficacy of SCFAs to treat metabolic dysfunction could be optimized by targeting their delivery to the distal gut. Additionally, one clinical study investigated the efficacy of propionate to attenuate atherosclerosis (Haghikia et al., 2022). Oral propionate administration reduced cholesterol levels and lesion severity by an immune-mediated mechanism involving Treg induction.

In summary, promising preclinical findings regarding the therapeutic potential of SCFAs have generally translated well into short-term clinical applications in patients with metabolic disease. However, high doses are still required and it remains unknown whether long-term benefits can be achieved. As we will discuss, novel oral formulation approaches for delivering SCFAs to the distal gut represent a promising therapeutic strategy for treating metabolic diseases.

Trp metabolites

In both plasma and fecal samples from individuals with obesity and T2D, levels of microbial Trp metabolites are consistently lower than in those from healthy individuals. These include tryptamine, indole, IAA, IPA, and ILA, some of which are negatively correlated with BMI (Cussotto et al., 2020; Jennis et al., 2018; Ma et al., 2020; Natividad et al., 2018; Virtue et al., 2019). In line with these observations, the AHR activity of fecal samples from obese individuals is compromised, which is recapitulated in fecal samples from HFD-fed mice (Krishnan et al., 2018; Natividad et al., 2018; Virtue et al., 2019). Similarly to IBD, obesity is associated with higher Kyn levels and increased IDO-1 activities (Cussotto et al., 2020; Laurans et al., 2018; Natividad et al., 2018). In the context of NAFLD, there is also evidence of the shift in Trp metabolism from microbially-mediated indole production to host-mediated Kyn production, as NAFLD individuals also present increased IDO-1 activities (J. Chen et al., 2022; Q. Zhou et al., 2021).

IPA is a particularly promising candidate for treatment of metabolic diseases. Higher levels of IPA correlate with a reduced risk of T2D, improved insulin secretion, and lower inflammation (de Mello et al., 2017; Tuomainen et al., 2018). Studies have shown that IPA administration can reduce fasting plasma glucose levels in db/db mice and healthy rats (Abildgaard et al., 2018; Z. Liu et al., 2020). Moreover, IPA alleviated HFD-induced fatty liver, hepatic steatosis and liver inflammation (Z.-H. Zhao et al., 2019). This study provided evidence that IPA restructures the gut microbiome of HFD-fed mice, reducing its capacity for nutrient and energy metabolism. Additionally, IPA mitigated the intestinal barrier disruption induced by HFD feeding (Jennis et al., 2018). In contrast, one study found that IPA administration could not improve cardiometabolic outcomes in mice fed a Western diet, highlighting that its effects may be diet- or model-dependent (D. M. Lee et al., 2020).

Indole has been shown to alleviate liver inflammation in HFD-induced NAFLD/NASH models and in ob/ob mice (Beaumont et al., 2018; Knudsen et al., 2021; Ma et al., 2020). Interestingly, indole was found suppress pro-inflammatory macrophage activation by stimulating the expression of the glycolysis activator PFKFB3 (Ma et al., 2020). This resulted in alleviation of HFD-induced hepatic steatosis and inflammation. In the MCD-induced model of NASH, however, indole was found to alleviate symptoms by decreasing pro-inflammatory responses in hepatocytes, but not macrophages, suggesting its mechanism of action may also be diet- or model-dependent (B. Zhu et al., 2022).

Similarly, contradictory findings have been reported regarding the mechanism by which IAA ameliorates inflammation in HFD-fed mice. One study found that oral administration of IAA promoted anti-inflammatory macrophage phenotypes in HFD-fed mice, but did not significantly affect hepatocyte phenotypes (Y. Wang et al., 2021). On the other hand, another found that IAA promoted anti-inflammatory phenotypes in both macrophages and hepatocytes (Krishnan et al., 2018). In addition to the oral route, administration of IAA via i.p. injection was also shown to ameliorate HFD-induced oxidative and inflammatory liver damage and improve insulin resistance (Y. Ji et al., 2019). Finally, IA produced by P.distasonis attenuated metabolic dysfunction in T2D rats by repairing the intestinal barrier, which prevented transepithelial LPS leakage (D. Liu et al., 2023). IA activated the AHR pathway to increase IL-22 levels, which promoted barrier functions by increasing epithelial TJP expression.

More research is surely needed to define the precise mechanisms of action of Trp metabolites in regulating metabolic dysfunction and to determine whether these preclinical findings can be translated to humans. However, important progress has been made by employing novel pharmaceutical formulations of Trp metabolites in preclinical metabolic disease models.

3.3. Autoimmune diseases

The incidence of autoimmune diseases, which now impact nearly 5% of the world’s population, has risen dramatically in recent years (Fugger et al., 2020). These conditions occur when the immune system erroneously targets the body’s own cells and tissues, which is known as self-tolerance loss. As a result, autoimmunity is characterized by persistent inflammation and tissue damage. Although they are categorized as primarily T cell- or B cell-mediated, the pathogenesis of all autoimmune diseases involves both T and B cells as well as DCs and macrophages. Current disease-modifying therapies include biologics, such as inflammatory cytokine inhibitors, and broad-spectrum immunosuppressants. As both of these interventions produce only modest improvements and are associated with significant side effects, such as increased risk of infection, more effective therapies that re-establish self-tolerance rather than suppress natural immune functions are needed.

The emergence of autoimmune disease can be attributed to a mix of genetic and environmental factors. Given the crucial role of the gut microbiota in regulating the immune system, it is no surprise that gut dysbiosis is now recognized as a significant environmental factor that influences the onset and progression of autoimmunity (Miyauchi et al., 2023). Here, we review literature regarding gut dysbiosis and the therapeutic potential for microbial metabolite-based interventions in the context of three major autoimmune diseases: multiple sclerosis (MS), rheumatoid arthritis (RA), and type 1 diabetes (T1D).

3.3.1. Multiple sclerosis (MS)

Multiple sclerosis (MS) is a chronic neurodegenerative disease primarily caused by autoreactive Th1 and Th17 cell-mediated inflammatory reactions against the myelin sheathes surrounding the nerves of the central nervous system (CNS) (Goverman, 2009). After being activated in the periphery, these cells infiltrate the CNS by crossing the blood-brain and blood-spinal cord barriers. Recent research has highlighted the potential to treat MS by targeting the gut, where epithelial barrier disruption and an imbalance of regulatory vs. effector T cells are thought to contribute to heightened inflammation and self-tolerance loss (Ghezzi et al., 2021). In this respect, gut dysbiosis has been identified in a variety of MS patient cohorts (Ordoñez-Rodriguez et al., 2023). While α diversity is generally unchanged, there are marked alterations in specific bacterial constituents, including those that produce SCFAs and Trp metabolites. For example, a common feature of MS-associated is a reduction in Prevotella species, which can produce both SCFAs and Trp metabolites (Shahi et al., 2017). In fact, Prevotella abundance is inversely related intestinal Th17 cell frequencies, which positively correlate with disease activity in MS (Cosorich et al., 2017). This serves as just one example that illustrates how alterations to the composition of gut microbiota may contribute to CNS autoimmunity via the gut-brain axis. As we will discuss, abundant evidence supports the use of SCFAs and Trp metabolites as therapeutics for MS.

SCFAs

MS patients consistently exhibit reductions in SCFA-producing bacteria, particularly butyrate-producers, including members of Clostridium clusters XIVa and IV (Miyake et al., 2015; Ordoñez-Rodriguez et al., 2023). Notably, a machine-learning algorithm identified butyrate-producing bacteria as key discriminators between MS patients and controls (Levi et al., 2021). Accordingly, reduced fecal SCFA concentrations are frequently reported in MS patients (Duscha et al., 2020; Takewaki et al., 2020; Zeng et al., 2019).

SCFA administration has yielded ameliorative effects on the progression of experimental autoimmune encephalomyelitis (EAE), the preclinical model of MS. Oral administration of butyrate significantly reduced clinical EAE scores, demyelination, and CNS inflammation (Calvo-Barreiro et al., 2021). In another study, it was found to increase Tregs and downregulate production of the Th1 cytokine IFN-γ in EAE mice (Mizuno et al., 2017). In a cuprizone-induced demyelination model, oral butyrate treatment suppressed demyelination and enhanced remyelination (T. Chen et al., 2019). Interestingly, the less well-known SCFA valerate (pentatoate) was also found to ameliorate EAE severity. Valerate treatment reduced the number of effector T cells that infiltrated the CNS while increasing both Tregs and Bregs (Luu et al., 2019).

Administration of acetate and propionate has also been found to ameliorate the progression and severity of preclinical MS. These SCFAs were found to modulate immune responses in EAE mice: acetate downregulated pro-inflammatory TNF-α production (Fragas et al., 2022), whereas propionate increased intestinally-derived Tregs and upregulated anti-inflammatory IL-10 production (Haghikia et al., 2015). Interestingly, the latter study also revealed that the administration of long-chain fatty acids suppresses SCFA production, which exacerbates disease progression by expanding pathogenic Th1 and Th17 cells. In another study, propionate administration prevented the aggravation of EAE progression induced by HFD feeding (Haase et al., 2021). Propionate inhibited Th17-mediated inflammatory reactions in the gut and spleen by enhancing the suppressive capacity of Tregs. Interestingly, obese MS patients were found to have increased Th17 cells and decreased Tregs compared to non-obese patients, which was associated with reduced fecal propionate levels. As such, propionate may have the potential to counteract the enhanced inflammation driven by obesity in MS patients by restoring the intestinal Treg/Th17 cell balance.

While these preclinical findings demonstrated the benefits of SCFA administration in preventative settings, two recent clinical studies showcased the therapeutic potential of propionate supplementation in MS patients (Table 1). Following a 14d treatment, propionate upregulated the Treg-cell inducing genetic capacity of the gut microbiome, enhanced Treg cell numbers and functions, and decreased Th1 and Th17 cells (Duscha et al., 2020). Chronic supplementation led to reduced annual relapse rates, disability stabilization, and reduced brain atrophy. Propionate also increased levels of osteocalcin, a circulating biomarker of bone formation, in MS patients (Duscha et al., 2022). Osteocalcin levels were positively and negatively correlated with Treg and Th17 cell numbers, respectively. These clinical studies highlight the potential for propionate to modulate the progression of MS and one of its most common co-morbidities, osteoporosis.

Trp metabolites

Microbial Trp metabolism is now recognized to be regulated differently in the MS disease context (Gaetani et al., 2020). The gut microbiome in MS patients was found to be deficient in bacteria that produce ILA, causing reduced serum levels of ILA and its downstream metabolite IPA (Levi et al., 2021). Global metabolomics in pediatric MS cases revealed that higher levels of ILA, IPA, and IAA associated with either lower risk or better outcomes, while higher relative amounts of Kyn were linked to worse outcomes (Nourbakhsh et al., 2018). As such, Trp metabolism may be directed toward the Kyn pathway at the expense of the indole pathway in MS, like in IBD and metabolic diseases. One of the most significantly reduced genera of commensal bacteria in MS patients is Parabacteroides, with a notable reduction in the IA-producing P.distasonis (Shahi et al., 2017).

A crucial role for microbially-produced AHR ligands in ameliorating CNS autoimmunity has been elegantly demonstrated by Rothhammer et al (Rothhammer et al., 2016, 2018). They report that Trp metabolites such as indole, IPA, and IAld contributed by the gut microbiota activate AHR signaling in brain-resident astrocytes and microglia, leading to reductions in CNS inflammation and EAE progression. Tryptamine was also found to attenuate EAE by decreasing the number of infiltrating CD4⁺ T effector cells in the CNS and increasing Foxp3⁺ Tregs (Dopkins et al., 2020). Supplementation of Trp-metabolite producing L.reuteri has been reported to ameliorate EAE by reducing Th1/Th17 cells and their associated cytokines (B. He et al., 2019; Johanson et al., 2020; Wilck et al., 2017). However, another study reported that this species exacerbates EAE and enhances IL-17 production (Montgomery et al., 2022), where the microbial Trp metabolites produced by L.reuteri were implicated. As such, further research is needed to determine the precise mechanisms of Trp metabolite-mediated regulation of CNS autoimmunity.

3.3.2. Rheumatoid arthritis (RA)

Rheumatoid arthritis (RA) is another primarily Th1 and Th17 cell-driven autoimmune disease, where the synovial joints are targeted by inflammatory reactions. In addition to autoreactive T cells, autoreactive B cells contribute to disease progression by synthesizing autoantibodies against antigens within the joints. Bregs, the B cell counterparts of Tregs, are particularly important for suppressing excessive immune responses in RA as they inhibit both autoantibody production and effector T cell activation. As in MS, self-tolerance loss has been linked to dysregulation of the intestinal immune system in RA. For example, in the collagen-induced arthritis (CIA) preclinical model of RA, clinical symptoms are preceded by upregulation of pro-inflammatory cytokines, including the Th17 cell-derived cytokine IL-17, and the subsequent activation of autoreactive T cells in the intestinal tissue (M. Li & Wang, 2021). Interactions like these exemplify the presence of the gut-joint axis, raising the possibility of targeting the gut to treat RA.

Gut dysbiosis plays a critical role in RA pathogenesis. In contrast to MS, Prevotella abundance, particularly of P.copri, is increased in RA (Zaiss et al., 2021). The increased abundance of P.copri has been linked to the activation of autoreactive intestinal Th17 cells and enhanced IL-17 production (Maeda et al., 2016). However, administration of P.histicola, another Prevotella species, suppressed arthritis development in preclinical studies (Maeda & Takeda, 2017), indicating the complexity of Prevotella-mediated immune regulation in the context of RA. Adding to the complexity, conflicting reports exist as to whether the major SCFA-producing bacterial families are increased or decreased in RA (L. Lin et al., 2023). Nevertheless, there is convincing evidence to support that RA patients have lower fecal SCFA levels, which correlate positively with Breg frequencies (Rosser et al., 2020; Yao et al., 2022). Moreover, He et al. used a quasi-paired cohort strategy to demonstrate that RA patients present a deficiency of butyrate producers and an enrichment of butyrate consumers, where net butyrate yield exhibited a negative correlation to disease activity (J. He et al., 2022).

Immunomodulation by SCFAs after oral administration has been shown to ameliorate disease progression in preclinical RA models. Administration of acetate, propionate, and butyrate mixtures prior to CIA induction reduced bone erosion by promoting Breg differentiation via GPR43 (Yao et al., 2022). Both butyrate and propionate have also been shown to modulate gene expression in osteoclasts (bone-resident macrophages) to protect against inflammatory bone loss in CIA mice (D. S. Kim et al., 2018; Lucas et al., 2018). Administration of these metabolites promoted Treg expansion, along with increased IL-10 production, decreased Th17 cells, and decreased TFH cells (Bai et al., 2021; J. He et al., 2022; Hui et al., 2019). The therapeutic effects of butyrate may be dependent on IL-10 upregulation, as IL-10-knockout mice were not protected by butyrate (D. S. Kim et al., 2018). Interestingly, butyrate was also found to increase the levels of the Trp metabolite 5-hydroxyindole-3-acetic acid (5-HIAA), which promoted IL-10 expression in Breg cells via AHR (Rosser et al., 2020; Yu et al., 2021). In another mechanism, butyrate suppressed autoantibody production by stimulating TFR cells (Takahashi et al., 2020).

Preventative administration of acetate is similarly protective against arthritis development. Acetate directly induces the differentiation of Breg cells, and adoptive transfer of acetate-treated B cells leads to reduced disease severity in CIA mice (Daïen et al., 2021). Findings in the K/BxN serum-induced arthritis (SIA) model of RA are less consistent, with one study reporting amelioration by SCFAs (Lucas et al., 2018), and another reporting no effect (Mizuno et al., 2017). As this model represents the effector phase of inflammation rather than the initiation phase of autoimmunity, these findings highlight that SCFA efficacies may be dependent on disease status in RA.

A few studies have also investigated the therapeutic potential of Trp metabolites in preclinical RA. IPyA was found to modulate Treg/Th17 cell balance in CIA mice, leading to disease amelioration (T. Huang et al., 2023). Two other Trp metabolites, IAld and IAA, were found to differentially modulate processes associated with autoimmune arthritis (Langan et al., 2021). While IAld inhibited pro-inflammatory cytokine expression and exhibited both pro-angiogenic and pro-osteoclastogenic activities, IAA exhibited anti-angiogenic activity. As recent evidence suggests that Trp metabolism may be dysregulated in RA (H. Xu et al., 2022; Yu et al., 2021), further investigations into the therapeutic potential of Trp metabolites in the context of RA are warranted.

3.3.3. Type I diabetes (T1D)

In type 1 diabetes (T1D), the insulin-producing β-cells in the pancreas are attacked by T-cell mediated autoimmune reactions, causing chronically elevated blood glucose levels. As in the previously described autoimmune conditions, targeting the gut is an evolving approach toward better clinical T1D management. This has come through recognition of the intimate interactions between the GI tract and the pancreas. For example, disruption of barrier functions and consequent bacterial translocation from the gut is posited to contribute immune cell activation and self-tolerance loss in the pancreatic lymph nodes (Menezes-Silva & Fonseca, 2019). Gut microbes may be involved as early initiators of T1D pathogenesis, as demonstrated by increasing evidence from both preclinical and clinical studies (Zheng et al., 2018).

Gut dysbiosis in T1D is consistently characterized by an increased Bacteroidetes-to-Firmicutes ratio (Gülden et al., 2015). The decline in Firmicutes can largely be attributed to the loss of butyrate-producing genera, including Clostridium clusters XIVa and IV (de Goffau et al., 2013; de Groot et al., 2017; J. Huang et al., 2020; Yuan et al., 2022). Like IBD fecal samples, T1D fecal samples present significantly decreased butyryl-CoA:acetate-CoA transferase gene ratios, reflecting a reduction in the gut microbiome’s genetic capacity for butyrate synthesis (de Groot et al., 2017). In accordance, significant reductions in butyrate concentrations are found T1D fecal samples, along with reductions in acetate and propionate concentrations (de Groot et al., 2017; J. Huang et al., 2020; Yuan et al., 2022). Lower abundances of lactate-producing Bifidobacteria are have also been reported (de Goffau et al., 2013).

A few studies have examined the effects of microbial metabolites in T1D murine models and patients. In streptozotocin-induced T1D mice, oral butyrate administration upregulated preproinsulin-coding genes and downregulated genes involved in inflammation and immune responses in the nucleotide-binding oligomerization domain (NOD)-like receptor and toll-like receptor (TLR) signaling pathways (Yuan et al., 2022). Butyrate has also shown efficacy in the non-obese diabetic (NOD) mouse model of T1D. When treatment occurred after the onset of hyperglycemia, oral butyrate restored immunological tolerance by inducing Tregs capable of migrating from the gut-associated lymphoid tissues to the pancreatic lymph nodes (Jacob et al., 2020). Interestingly, production of cathelicidin-related antimicrobial peptide (CRAMP) by pancreatic β-cells, which induces regulatory rather than inflammatory immune cell phenotypes, was found to be controlled by SCFAs (J. Sun et al., 2015). Administration of butyrate via i.p. injection to NOD mice (which showed reduced fecal and systemic levels of butyrate) increased pancreatic CRAMP production. In another study, acetate was also found to be effective in treating pancreatic islet inflammation in NOD mice (J. Huang et al., 2020). In this study, acetate increased the α diversity of the gut microbiota, promoted Treg induction, and reduced gut bacteria–induced IgA responses. While SCFAs normally stimulate intestinal IgA responses to modulate the composition of the microbiota against pathogens, IgA responses are hyperactivated in T1D. As such, this study serves as another example of the context-dependency of SCFA-mediated immunomodulation.

One clinical study investigated the efficacy of oral sodium butyrate in patients with T1D (de Groot et al., 2020) (Table 1). Even at the high dose of 4 g/d, this treatment did not produce changes in the innate immune phenotype or blood markers of pancreatic islet autoimmunity in T1D patients after 1 month of treatment. In this study, it is unlikely that butyrate concentrations were increased in the distal GI tract enough to reach the pancreas in sufficient concentrations after oral administration, even with the high dose, given its rapid degradation and metabolism. Additionally, as pancreatic β-cells are irreversibly lost in T1D, it is possible that this short-term butyrate administration was not sufficient to significantly restore insulin homeostasis.

3.4. Allergic diseases

Allergies represent another manifestation of inappropriate immune responses. Rather than the loss of self-tolerance as seen with autoimmune disease, allergic diseases are characterized by the loss of tolerance to exogenous antigens in foods or the environment that are otherwise harmless in most individuals. While Th1 and Th17 cells are the effector T cells primarily involved in many of the aforementioned chronic diseases, Th2 cells are the primary T cell effectors in allergic diseases. Activation of Th2 cells in response to an antigen causes them to release IL-4, IL-5, and IL-13, which stimulate B cells to produce IgE antibodies against that antigen in a process known as sensitization. These antibodies bind to mast cells and basophils, which reside in the lungs, GI tract, and/or skin, through FcεRI receptors. When these IgE antibodies encounter antigens again upon re-exposure or “challenge”, those cells then release histamine and other inflammatory mediators, causing allergic responses known as a type I hypersensitivity reactions.

The prevalence of allergic diseases has surged in the past few decades (Gutowska-Ślesik et al., 2023). This increase has been particularly noticeable in developed countries and cannot be simply explained by genetic factors, suggesting that changes in lifestyle and environmental factors may be contributing to this trend. Microbial dysbiosis has been linked to various allergic diseases, including asthma, allergic rhinitis, and food allergies (Pascal et al., 2018). Here, we aim to summarize how allergy-associated dysbiosis contributes to altered levels of SCFAs and Trp metabolites, and how their administration has the potential to ameliorate allergic diseases.

3.4.1. Allergic airways disorders

Allergic airways disorders (AAD) are primarily comprised of asthma and allergic rhinitis. Dysbiosis of the microbial communities, both in the lungs and the gut, has been linked to these conditions (Hufnagl et al., 2020). In the gut microbiomes of infants who later develop asthma and in those of asthmatic children, decreased abundances of SCFA-producing genera including Lachnospiraceae, Roseburia, and Faecalibacterium, have been reported (Chiu et al., 2019; Stiemsma et al., 2016; Stokholm et al., 2018). Lower fecal butyrate levels are observed in children that develop AAD and correlate with serum IgE levels, while children that do not develop AAD have higher fecal butyrate levels (Chiu et al., 2019; Roduit et al., 2019). Moreover, the microbiome of infants who later developed allergic sensitization displayed a reduced genetic capacity for butyrate production (Cait et al., 2019). Acetate and propionate levels have also been reported to be reduced in fecal samples from asthmatic patients (Ivashkin et al., 2019).

SCFAs produced by the gut microbiota are known to act via gut-lung axis mechanisms to modulate lung homeostasis (Dang & Marsland, 2019). Exemplifying this concept, oral administration of SCFAs has been shown to resolve airway inflammation in mouse AAD models. Administration of all three SCFAs, either individually or in combination, reduces airway hyperresponsiveness and immune cell infiltration into the lungs of AAD mice (M.-T. Huang et al., 2023; Roduit et al., 2019). Interestingly, while SCFAs induced neither Tregs nor immunosuppressive myeloid cells in the lungs of non-asthmatic mice, SCFA mixtures increased the levels of these anti-inflammatory cell populations in mice rendered asthmatic by ovalbumin (OVA) sensitization (M.-T. Huang et al., 2023). In another study, butyrate inhibited type 2 innate lymphoid cell (ILC2)-driven airway hyperresponsiveness and eosinophilic inflammation after administration by oral or intranasal routes (Thio et al., 2018). Additionally, enhanced allergic inflammation in vancomycin-treated mice, which was associated with impaired butyrate production, was counteracted by butyrate supplementation (Cait et al., 2018). A variety of studies in preclinical AAD models implicate the HDAC inhibitory activity of butyrate in its phenotypic regulation of immune cells, including ILC2s, Tregs, and eosinophils (Cait et al., 2018; Islam et al., 2022; Roduit et al., 2019; Theiler et al., 2019). HDAC inhibition may also be the mechanism by which acetate suppresses AAD pathology, as acetate treatment increased Treg cell numbers and functions, associated with increased acetylation at the Foxp3 promoter (Thorburn et al., 2015).

SCFA-mediated effects on macrophages and DCs also mediate their protective effects against AAD. Propionate was shown to act in a GPR41-dependent manner to enhance generation of macrophage and DC precursors with anti-inflammatory phenotypes, leading to reduced Th2 effector cell polarization (Trompette et al., 2014). Furthermore, both propionate and butyrate regulated macrophage polarization to attenuate airway inflammation in AAD mice, potentially through a GPR43- and/or HDAC-dependent mechanism (C. Huang et al., 2022).

Further studies are needed to investigate the effects of Trp metabolites in the context of AAD, as the administration of an indole-producing probiotic species demonstrated therapeutic effects in AAD mice (L. Li et al., 2020). As we will illustrate in the next section, targeting the delivery of AHR-agonizing Trp metabolites to either the distal gut or the lungs using novel pharmaceutical formulations approaches has demonstrated the potential to improve their efficacy in promoting lung homeostasis.

3.4.2. Food allergies

The gut microbial alterations observed in AAD are largely similar to those observed in patients with food allergy. Enrichment of Firmicutes, specifically of the Clostridia class, was found in children whose milk allergy resolved compared to those whose allergy persisted (Bunyavanich et al., 2016). Supporting these findings, a study comparing twins concordant or discordant for food allergy revealed an enrichment of SCFA-producing members of the Clostridia class, including Lachnospiraceae and Ruminococcaceae family members, in healthy twins compared to those with food allergies (Bao et al., 2021). Supporting the role of Clostridia in preventing food allergies, colonization of gnotobiotic mice with Clostridia species blocked food allergen sensitization (Stefka et al., 2014). These bacteria likely functioned to promote intestinal barrier functions, as upregulation of IL-22 production by intestinal ILCs and T cells was associated with reduced transfer of the orally administered dietary antigen to the systemic circulation. Moreover, germ-free (GF) mice colonized with feces from healthy infants, but not cow’s milk allergic infants, were protected against allergic sensitization (Feehley et al., 2019). This protective effect was specifically associated with the Lachnospiraceae family member Anaerostipes caccae, which produces both butyrate and the Trp metabolite ILA.

Butyrate was found to enhance desensitization of basophils and mast cells induced by oral immunotherapy in a model of cow’s milk allergy, leading to a reduction of acute allergic skin responses and mast cell degranulation upon challenge (Vonk et al., 2019). In a mouse model of peanut allergy, reduced anaphylaxis clinical scores were observed when mice were treated with acetate and butyrate prior to sensitization (J. Tan et al., 2016). Protection correlated with induction of CD103⁺ DCs and Treg cells, which are both crucial for maintaining intestinal immune tolerance, in the mesenteric LNs (mLNs). As highlighted in the next section, oral SCFA formulations capable of modulating immune cell populations in the mLNs represent a next-generation approach towards promoting oral tolerance in the context of food allergy.

3.5. Infectious diseases

Microbial metabolites not only help restrain inappropriate immune activation in chronic diseases to promote immune tolerance, but they also acutely enhance immunity in infectious diseases to accelerate pathogen clearance. Thus, it is no surprise that compromised microbial metabolite production in chronic diseases has been linked to increased risk of infections. Conversely, enteric infections can seed the development of dysbiosis by disrupting both microbial and host homeostasis. This can both contribute to chronic disease pathogenesis and increase the risk of secondary infections. Here, we highlight the potential of SCFAs and Trp metabolites to interrupt this detrimental feedback loop between gut dysbiosis and infectious diseases.

SCFAs

SCFA production is a well-recognized mechanism by which commensal gut microbes limit the colonization of pathogenic bacteria in the GI tract. In mice, lower microbial SCFA production is associated with increased susceptibility to enteric infection with Citrobacter (C).rodentium, the murine model of enterohemorrhagic E. coli (EHEC) infection (Osbelt et al., 2020). Interestingly, mouse models have also shown that pulmonary viral infections contribute to gut dysbiosis and reduced SCFA production (Sencio et al., 2020). Completing the negative feedback loop, this reduction in SCFA production was found to contribute to the development of secondary bacterial infections, both in the lungs (Sencio et al., 2020) and the GI tract (Sencio et al., 2021). These studies point to SCFAs as key players in the commensal microbe-host-pathogen interactions that act to prevent infectious diseases.

As outlined in the first section of this review, SCFAs acutely activate antimicrobial defense mechanisms in both IECs and immune cells to promote pathogen clearance during infection. Moreover, they activate anti-inflammatory pathways to simultaneously prevent the development of chronic inflammation. For example, butyrate treatment via enema or oral administration simultaneously induces anti-microbial peptides (AMPs) and barrier repair processes in the IECs of enterically-infected mice, leading to reduced disease scores (Jiminez et al., 2017; Raqib et al., 2006). In patients with enteric infections, butyrate enemas recapitulated the preclinical findings of stimulated AMPs and reduced inflammation (Raqib et al., 2012) (Table 1). However, this treatment did not yield significant improvements in clinical recovery or disease scores.

Importantly, SCFA-mediated activation of immunity is context-dependent, occurring only when active immune responses are needed. Park et al. demonstrated that orally administered SCFAs significantly increase the frequencies of Th1 and Th17 cells in mice during C.rodentium infection, but do not alter these populations in healthy, uninfected mice (Park et al., 2015). This has also been demonstrated in the context of enteric fungal infections. SCFAs were found to stimulate Th17 cells and IL-17 production to promote fungal clearance (Bhaskaran et al., 2018). Butyrate-mediated induction of IL-22 production by CD4⁺ T cells and ILCs is another mechanism by which it contributes to pathogen resistance (W. Yang et al., 2020). At the same time, SCFAs induce Tregs to promote the resolution of inflammation. Additionally, SCFA-mediated downregulation pro-inflammatory cytokines in innate immune cells, such as macrophages and DCs, limits their ability to perpetuate chronic inflammation after infection (Ciarlo et al., 2016).

Oral acetate administration has been shown to induce neutrophil chemotaxis toward sites of infection (Maslowski et al., 2009) and upregulate their production of ROS via GPR43 (Schlatterer et al., 2021). Through these mechanisms, acetate improves their capacity to eliminate pathogenic bacteria from the bloodstream, conferring protection against severe sepsis (Schlatterer et al., 2021). Additionally, acetate has been shown to boost immunity against pulmonary infections via the gut-lung axis. Both oral and intranasal acetate administrations upregulated IFN-β-mediated antiviral type-I interferon (IFN1) responses in pulmonary epithelial cells via GPR43, leading to reduced viral loads in respiratory syncytial virus-infected mice (Antunes et al., 2019, 2022, 2023). Notably, acetate’s upregulation of retinoic acid-inducible gene (RIG)-I, a key mediator of IFN1 responses, was necessary for its protective effects (Antunes et al., 2022). Oral acetate supplementation also lowered the severity of secondary bacterial infections in the lungs of mice infected with respiratory influenza A virus (Sencio et al., 2020), suggesting that acetate might also engage antibacterial signaling pathways in pulmonary epithelial and/or immune cells.

Recently, orally administered SCFAs were found to protect mice against pulmonary SARS-CoV-2 infection (Brown et al., 2022). In this study, SCFAs reduced viral loads both in the lungs and the GI tract. In addition to enhancing adaptive immunity via GPCRs, SCFAs downregulated ACE2, the receptor that the virus exploits to enter host cells. With these findings, SCFAs have emerged as potential alternatives to broad-spectrum immunosuppressant corticosteroids (e.g. dexamethasone), which can cause serious side effects with long-term use. Notably, an intranasal formulation developed to harness the antiviral properties of acetate was found to improve acetate’s control of SARS-CoV2 infection, which we will discuss in the next section. Additionally, chronic viral infections, such as HIV and hepatitis, are associated with a reduced abundance of SCFA-producing bacteria (Dillon et al., 2017; Milosevic et al., 2021), and recent studies have identified this same trend in COVID-19 patients (Galperine et al., 2023; Sajdel-Sulkowska, 2021). The ability of SCFAs to promote immunity while attenuating hyperinflammation supports that they could be used to control the cytokine storm associated with viral infection without compromising global immune responses. Collectively, these findings pave the way for pioneering SCFA-based therapeutics for infectious diseases.

Trp metabolites

In addition to their roles in microbiota-host communication, indoles have long been recognized as quorum-sensing molecules, utilized by bacteria to coordinate communication amongst their community (J.-H. Lee & Lee, 2010). Notably, indole is well-known for its antimicrobial properties. Interestingly, Clostridium (C).difficile infection was found to stimulate indole production by commensal microbes, which was hypothesized to be a mechanism by which C.difficile compromises the survival of protective bacteria to enhance its own colonization ability (Darkoh et al., 2019). However, evidence suggests that indole production by gut commensals actually decreases the ability of enteric pathogens, such as C.rodentium and EHEC, to colonize the gut (Kumar & Sperandio, 2019). Moreover, in human studies, high fecal indole levels were predictive of successful treatment of recurrent C.difficile infection (Hibbard et al., 2019), and resistance against parasitic Cryptosporidium infection (Chappell et al., 2016).

Starting with their direct effects on pathogen growth, a recent study determined that D-tryptophan (D-Trp) administration protects mice against C.rodentium infection (Seki et al., 2022). IA produced by microbially-mediated metabolism of D-Trp was found to be responsible for the inhibition of C.rodentium growth, as mice treated with IA were similarly protected. In another recent study, oral administration of indole reduced the severity of parasitic infection in mice both by directly inhibiting the growth of the parasite and by modulating host mitochondrial functions (Funkhouser-Jones et al., 2023).

Administration of Trp metabolites also regulates host immunity to promote resistance against colonization by pathogens. Oral administration of indole, IEt, IPyA, or IAld reduced bacterial load and ameliorated colonic inflammation in mice infected with C.rodentium (Scott et al., 2023). These metabolites were found to activate the dopamine receptor D2 on IECs, which caused the downregulation of an actin regulatory protein involved in pathogen attachment to the gut epithelium. Thus, this study revealed both a novel receptor for microbially-produced Trp metabolites and a novel mechanism by which they promote colonization resistance. In a murine model of sepsis, oral IPA administration protected against mortality and alleviated bacterial burden in the serum, peritoneal lavage fluid, and spleen (Z.-B. Huang et al., 2022). Mechanistically, IPA enhanced the phagocytic capacity of peritoneal macrophages via AHR signaling, as the IPA-induced increase in survival rates was ablated upon macrophage depletion. Lastly, IAld administration provided resistance against fungal pathogens while inhibiting inflammation via AHR-mediated upregulation of IL-22 in IECs (Zelante et al., 2013). As discussed in the next section, both oral and inhalable pharmaceutical formulations of IAld have been developed to target its delivery to the GI tract and lungs, respectively, which have yielded promising findings in the context of fungal infections.

3.6. Cancer

With research dating back nearly three decades, butyrate’s role in the prevention and treatment of cancer is one of the earliest explorations of its therapeutic potential. As the role of butyrate and the other SCFAs in the context of cancer has been reviewed extensively in other literature (O’Keefe, 2016), we will briefly touch upon the critical discoveries from this large area of research. This section is primarily included as important background for the subsequent section, as this work greatly fueled the quest for novel pharmaceutical formulations that enhance butyrate’s anti-cancer attributes through targeted- and/or sustained-delivery.

In the context of colorectal cancer, a preventative role for SCFAs produced by microbial fermentation of dietary fiber has long been well-understood (D’Argenio & Mazzacca, 1999). Interestingly, while butyrate stimulates proliferation of normal colonic epithelial cells, it inhibits the proliferation of colon cancer cells (Scheppach et al., 1995). The anti-tumor properties of butyrate include its promotion of apoptosis, downregulation of angiogenesis, and induction of cell differentiation (Williams et al., 2003). As histone acetylation was associated with these effects, HDAC inhibition was highlighted as the major mechanism at play. However, selective effects on GPCR activation and Wnt signaling have also been hypothesized to explain the disparate effects of butyrate in normal vs. colon cancer cells (Bordonaro et al., 2008; Velázquez et al., 1996). Additionally, butyrate may negatively affect the glycolysis-driven metabolism of cancer cells, known as the Warburg effect, to exert its growth-inhibiting effects (Donohoe et al., 2012; Geng et al., 2021).

Butyrate was found to similarly inhibit the growth of a variety of other cancer cell lines in vitro. However, its effects were dependent on prolonged exposure to a minimum concentration in the mM range (Newmark & Young, 1995). Due to rapid metabolism and clearance, many vivo studies did not support a therapeutic effect for butyrate in a phenomenon termed “the butyrate paradox” (Lupton, 2004). Thus, clinical applications of butyrate in cancer patients required high doses and administration via continuous i.v. infusions (Table 1) (Douillard et al., 2000; Miller et al., 1987). Even at doses of 0.5–2 g/kg/d, these treatments demonstrated a lack of clinical efficacy, largely due to butyrate’s short circulating half-life. Further, some studies were associated with significant side effects, such as serious liver toxicity (Douillard et al., 2000). However, in patients with Epstein-Barr virus-associated lymphoid malignancies, butyrate administered via i.v. infusion at doses up to 1 g/kg/d showed efficacy in sensitizing lymphoma cells to apoptosis induced by ganciclovir by inducing expression of the viral enzyme target thymidine kinase via HDAC inhibition (Perrine et al., 2007). Additionally, supplementing anti-cancer treatment by co-administering lower doses of butyrate emerged as an attractive strategy to not only improve efficacy, but also mitigate the side effects of anti-cancer drugs by virtue of butyrate’s anti-inflammatory properties (Al-Qadami et al., 2022).

As described in the next section of this review, a variety of novel pharmaceutical formulations to enhance delivery of butyrate and the other SCFAs to tumor cells have greatly contributed to realizing their therapeutic potential in cancer treatment. As increasing evidence continues to support gut dysbiosis associated with reduced SCFA production as a major contributing factor to colorectal and other cancers (Compare & Nardone, 2011; Kumar et al., 2023), oral formulations approaches that target SCFA delivery to the distal gut or prolong its circulating half-life will undoubtedly be the focus of future research aiming to develop SCFA-based anti-cancer therapeutics.

4. Pharmaceutical approaches towards improving the efficacy of microbial metabolite-based therapeutics

The therapeutic potential of microbial metabolites in such a multitude of disease contexts arises from their ability to simultaneously regulate many aspects of host physiology. As such, microbial metabolite-based therapeutics offer an opportunity to escape the dogma of “one drug, one target, one disease”, which has dominated the industrial development of disease-modifying therapies with limited success, owing to the complex etiology of multifactorial diseases (Puccetti et al., 2023). There are abundant advantages of microbial metabolites, including their defined chemical structures, relative ease of production, long shelf-lives, and amenability to formulation into various dosage forms, which are suitable for various administration routes. However, clinical translation of microbial metabolites into effective therapeutics presents significant challenges.

Given their poor stability in the acidic gastric fluids and tendency toward premature absorption in the proximal GI tract, a major challenge is to target the delivery of microbial metabolites to the distal gut after oral administration, the most preferred administration route for most patients. During their transit through the GI tract, microbial metabolites are subject to the effects of pH changes and metabolic degradation by either host cells or resident gut microbes. As a result, the preclinical studies outlined previously relied on administration of high oral doses, which often were not feasibly translated into clinical applications. In addition to limited or no efficacy in some patient populations, microbial metabolites administered in high doses, especially when delivered via i.v. injection, have led to worrying side effects, which has greatly limited their widespread adoption in the clinic. In the case of butyrate, its foul odor/taste is associated with low patient compliance, and has caused termination of clinical trials as it makes blinding impossible (e.g. NCT03010865).

Pharmaceutical approaches that enable targeted and/or sustained delivery while masking unpleasant organoleptic properties have played a pivotal role in unlocking the full potential of microbial metabolite-based therapeutics. These innovative drug delivery systems, which include enteric encapsulations, esterification to dietary fibers, prodrugs, nanoformulations, and other targeted delivery approaches, can stabilize metabolites during their transit through the body to ensure they reach their intended site of action. They also facilitate more desirable release kinetics, thereby extending therapeutic windows of efficacy. Because most metabolites have a free carboxyl or hydroxyl group, they can be conjugated to various delivery vehicles via a simple esterification reaction. Some approaches, such as prodrugs, aim to delay metabolite release until the compound becomes systemically bioavailable, enabling it to reach more distant tissues. There are also approaches that formulate metabolites for other administration routes, such as injectable nanoformulations and inhalable dry powders. Here, we aim to provide a comprehensive and up-to-date review of the pharmaceutical approaches that have been employed for SCFA and Trp metabolite-based therapeutics (Figure 2). These approaches will undoubtedly bring us closer to translating the promising findings observed in preclinical studies into the clinic.

Figure 2: Pharmaceutical approaches to deliver microbial metabolites as therapeutics.

Microbial metabolites have been formulated into delivery systems designed for oral, parenteral, topical, inhalable, and intranasal administration. Orally administered formulations include enteric encapsulations, esterification to dietary fiber, polymeric micelles, solid lipid NPs, prodrugs, and polymeric NPs. Prodrugs and polymeric NPs, as well as liposomes, have also been developed for administration via injection. Topically administered polymers, lung-targeted liposomes, and inhalable dry powders comprise formulations developed for administration routes other than oral and parenteral. Graphic created with BioRender.com.

4.1. Enteric encapsulation

One approach to enhance the stability of small molecule drugs during their transit through the GI tract is to use gastro-resistant materials in oral formulations. In this approach, gastro-resistant polymer-, lipid-, or carbohydrate-based materials are used to coat solid dosage forms, such as capsules and tablets, or to entrap an active compound within a gastro-resistant matrix, termed microencapsulation. These materials provide a moisture barrier to slow water penetration, which protects the active ingredient from metabolism within the gut. Further, they can impart delayed-, modified-, and/or controlled-release characteristics by virtue of their selective disintegration at specific pH values or upon exposure to intestinal enzymes (Figure 3).

Figure 3: Enterically-encapsulated microbial metabolites.

Gastro-resistant polymeric enteric coatings dissolve upon pH increase, causing burst release of metabolites. Microencapsulation matrices, namely polymeric, lipid-based, and dietary fiber-containing, release metabolites upon pH-dependent dissolution, enzymatic hydrolysis, and bacterial fermentation, respectively. Graphic created with BioRender.com.

As we describe here, enteric encapsulation approaches have been utilized to deliver both SCFAs and Trp metabolites to the distal GI tract. While these approaches have largely been tested in IBD patients, there are also examples of their application in patients with metabolic diseases (Table 2). As microbial metabolites are normally produced in the distal gut, targeting their delivery to this site allows researchers to mimic their production by a eubiotic microbiome. Enteric coatings and microencapsulations also mask unwanted organoleptic properties, which has been especially useful in clinical studies by facilitating patient compliance.

Table 2:

Administration of enterically-encapsulated and dietary fiber-esterified microbial metabolites in clinical studies

Metabolite dosage form	Study population	Dose, metabolite	Key findings (ClinicalTrials.gov identifier and/or references)
Enterically-coated tablets and capsules
Eudragit® S-coated tablets	IBD patients	2.4–4 g/d sodium butyrate	Improved clinical and histological disease activity scores (Vernia et al., 2000). Reduced inflammation in the terminal ileum and cecum, but not in the colon (Di Sabatino et al., 2005).
Eudragit® S-coated capsules	Healthy subjects	200 mg sodium acetate, 170 mg sodium propionate, or 495 mg sodium butyrate	Enteric coating enabled SCFA release slightly before inulin, likely in the terminal ileum and/or cecum (NCT01757379, Boets et al., 2017).
Eudragit® FS 30 D-coated capsules		~ 5 g acetate, ~0.5 g propionate, and ~2.3 g butyrate	Distal gut delivery of SCFA mixtures increased serum SCFA levels and attenuated the cortisol response to acute psychosocial stress (NCT03688854, Dalile et al., 2020).
Shellac-coated tablets		1 g (50–200 mg sodium butyrate)	Enteric coating enabled butyrate release in the terminal ileum/cecum (Roda et al., 2007).
Sodium alginate-coated capsules	Type 1 diabetes (T1D) patients	3.6 g/d sodium butyrate	No significant effects on intestinal inflammation or kidney function (diabetic nephropathy) (NCT04073927, Tougaard et al., 2022).
ZACOL-NMX® tablets	IBD patients	~1 g/d calcium butyrate and ~1.5 g/d inulin	Targeted butyrate delivery to all regions of the colon, significantly reduced disease activity and improved symptoms (Assisi & GISDI Study Group, 2008).
Microencapsulations
Hydrogenated plant triglycerides (C14:C20)	IBD patients	1000 mg/d Debutir® (300 mg/d sodium butyrate)	Debutir® decreased the frequency of certain IBD symptoms and improved patients’ quality of life (Banasiewicz et al. 2013; Krokowicz et al., 2014).
			Intesta administration reduced self-reported disease severity (Lewandowski et al., 2022).
		1200–1800 mg/d Butyrose® Lsc® Microcaps (BLM) (1000–1500 mg/d sodium butyrate)	Increased SCFA-producing bacteria at 1800 mg/d for 2 mo., with no effect on clinical disease activity scores (NCT04879914, Facchin et al., 2020). Significantly improved IBD symptoms and reduced inflammation at 1200 mg/d for 12 mo (Vernero et al. 2020).
	Children/adolescents with IBD	1000 mg/d Debutir® (300 mg/d sodium butyrate)	Debutir® did not show efficacy in reducing disease activity (NCT05456763, Pietrzak et al., 2022).
	Children with pediatric obesity	20 mg/kg/d, max. 800 mg/kg/d BLM (~17 mg/kg/d sodium butyrate).	Reduced BMI, waist circumferences, and inflammation while improving insulin sensitivity (NCT04620057; Coppola et al., 2022).
Medium-chain triglycerides (MCT) (C6:C12)	Type 2 diabetes (T2D) patients	p.o., 600 mg/d (550 mg (sodium butyrate)	Reduced blood pressure, but did not significantly decrease postprandial blood glucose or HOMA-IR (Khosravi et al., 2022; Roshanravan et al., 2017). Reduced expression of inflammatory and oxidative stress markers (Roshanravan et al., 2020).
Dietary fiber esterifications
High amylose maize starch (HAMS)	Healthy subjects	20–40g/d butyrylated HAMS (~3.5–7g/d butyrate)	80% of ingested esterified butyrate is released after HAMSB consumption; 60% is targeted to the colon (J. M. Clarke et al., 2011).
	Subjects at risk for colorectal cancer	40 g/d butyrylated HAMS (~7 g/d butyrate)	Prevented development of high red meat diet-induced rectal O6-methyl-2-deoxyguanosine adducts in rectal tissue, which are associated with increased risk of colorectal cancer (Le Leu et al., 2015).
	Type 1 diabetes (T1D) patients	40 g/d aceylated and butyrylated HAMS	Positively modulated the gut microbiome composition and the promoted regulatory immune cell phenotypes (Bell et al., 2022).
Inulin	Healthy subjects	10 g/d inulin propionate ester (~2.5 g/d propionate)	Improved pancreatic β-cell function and insulin secretion (Pingitore et al., 2017) and attenuated reward-based eating behavior (Byrne et al., 2016) without changes in PYY or GLP-1 levels (both: NCT00750438).
			Did not significantly increase weight loss relative to inulin only control, but reduced appetite and increased GLP-1 levels, which were not observed with inulin only control (NCT03322514, Khatib et al., 2018).
	Overweight adults		Acute supplementation increased PYY and GLP-1 levels. Long-term supplementation inhibited weight gain and preserved insulin sensitivity (NCT00750438, Chambers et al., 2015)
			Enhanced fat oxidation when combined with moderate intensity exercise without changes in PYY and GLP-1 (NCT04016350, Malkova et al., 2020).
	Non-diabetic overweight and obese adults	20 g/d inulin propionate ester (~5 g/d propionate)	Both IPE and inulin only control similarly improved insulin resistance. Significantly reduced proinflammatory interleukin-8 levels, which was not observed with inulin only control (Chambers et al., 2019).
	NAFLD patients		Prevented significant increase in intrahepatocellular lipid content (Chambers et al., 2019).

4.1.1. Enterically-coated tablets and capsules

Targeted delivery of SCFAs to the distal GI tract after oral administration has been achieved by coating tablets or capsules with gastro-resistant, pH-dependent polymers such as Eudragit^® polymethacrylate copolymers (Evonik Industries AG, Darmstadt, Germany). Delayed-release Eudragit^® copolymers include Eudragit^® L, a 1:1 copolymer of methacrylic acid (MAA) and methyl methacrylate (MMA) that dissolves above pH 5.5–6 in the small intestinal region, and Eudragit^® S, a 1:2 copolymer of MAA and MMA that dissolves above pH 7 in the terminal ileum and cecal regions.

In IBD patients receiving treatment with mesalazine, co-administration 2.4 g/d sodium butyrate tablets coated with Eudragit S yielded a significantly greater improvement in clinical and disease activity scores compared to mesalazine treatment alone (Vernia et al., 2000). Encouraged by the improved compliance and absence of adverse side effects, a later study tested a higher dose of 4 g/d of these tablets (Di Sabatino et al., 2005). The compliance remained excellent, and the high doses were found to be safe and well-tolerated. Endoscopical and histological assessments showed significant improvements after treatment in the terminal ileum and cecum, but not in the ascending colon, suggesting that butyrate was primarily released in the ileo-cecal region. More recently, hard gelatin capsules loaded with ¹³C-labelled acetate, propionate, or butyrate and coated with Eudragit S and diethyl phthalate (as plasticizer) were used to investigate the systemic bioavailability and metabolism of SCFAs in humans (Boets et al., 2017). To prevent early dissolution of the coating resulting from exposure to butyrate, the capsules were filled with 9% citric acid to maintain the pH of the contents below 6.5. Co-administering ¹⁴C-labelled inulin, which is microbially degraded and absorbed in the cecum/proximal colon, allowed the authors to determine whether the SCFAs were released before or after reaching this region. The results indicated that the SCFAs were again released in the ileo-cecal region. In a subsequent study, administration of SCFA mixtures in capsules coated with Eudragit FS 30 D (the aqueous dispersion formulation of Eudragit S) and PlasACRYL^™ T20 (Evonik Industries AG) (a glidant and plasticizer premix) attenuated cortisol responses to psychosocial stress and fear tests in healthy men after just one week of treatment (Dalile et al., 2020). Notably, increases in circulating SCFA levels were significantly associated with decreases in cortisol response, indicating that the formulation enabled the SCFAs to become systemically bioavailable.

Coating with the natural polymer Shellac (also dissolves above pH 7) also enabled butyrate release primarily in the terminal ileum of human subjects (Roda et al., 2007). In this formulation, internal pre-coating with hydroxypropyl methylcellulose (HPMC) was essential to preventing shellac exposure to butyrate, which would cause its early dissolution. In one unsuccessful example, administration of sodium butyrate in sodium alginate (SA)-coated capsules had no significant effects on intestinal inflammation or kidney function parameters in patients with T1D (Tougaard et al., 2022). While SA, a mucopolysaccharide derived from seaweed, can protect against acid degradation by forming a gelatinous mass in the stomach, it likely is not able delay butyrate release enough to target butyrate delivery to the distal GI tract.

Whereas most of these enteric coating approaches targeted delivery to the ileo-cecal region, one formulation successfully achieved colon-targeted butyrate delivery using Multimatrix^® (MMX^®) technology (Cosmo Pharmaceuticals, Ireland). MMX^® tablets are made by dispersing a lipophilic/amphiphilic (e.g. stearic acid/lecithin) matrix into a hydrophilic matrix (e.g. hydroxypropylcellulose) (Villa et al., 2015). The tablets are then enclosed in a gastro-resistant coating, which is generally comprised of Eudragit^® L and S copolymers in ~1:1 ratio. The coating delays the release until the terminal ileum is reached. Then, the hydrophilic excipients drive the tablet to swell into a viscous gel mass, slowing the release of the embedded compound. Finally, the lipophilic excipients slow water penetration into the tablet core. These properties enable MMX^® technology to function as a delayed- and extended-release carrier to enable colon-targeted drug delivery, which was demonstrated using radiolabeling for the MMX^® formulations of the pharmaceutical IBD drugs mesalamine and budesonide in clinical pharmacokinetic studies (Nardelli et al., 2017). ZACOL-NMX^® is a nutraceutical version of MMX^® technology that delivers a combination of calcium butyrate (0.3 mg/tablet) and inulin, a prebiotic dietary fiber described in detail below (see section 4.2.2) (0.5 mg/tablet) to the colon. In IBD patients receiving treatment with mesalazine, co-administration of three ZACOL-NMX^® tablets/d significantly reduced disease activity and improved symptoms (Assisi & GISDI Study Group, 2008). Since this study, however, ZACOL-NMX^® has not undergone further clinical evaluation.

Recently, a noteworthy preclinical study examining the efficacy of enterically-encapsulated sodium butyrate in treating Fusobacterium (F).nucleatum-mediated chemoresistance was published (L. Chen et al., 2023). F.nucleatum bacteria have been found to faciliate the formation of a tumor-promoting microenvironment, leading to the development of colorectal cancer chemotherapy resistance. Interestingly, butyrate was found to inhibit F.nucleatum proliferation, adhesion, and colonization. In this study, sodium butyrate tablets containing chitosan (CS) (2% w/w) and sodium alginate (SA) (2% w/w) were prepared and coated with Eudragit S100 (8% w/v). The CS and SA were selected to give adhesive and self-gelling properties to the tablet powder. Levels of indocyanine green, a fluorescent dye encapsulated in the tablet to investigate the release properties, peaked in the mouse intestinal tract at ~6h after administration. In an orthotopic colorectal cancer mouse model, these enterically-coated butyrate tablets successfully improved the therapeutic efficacy of the chemotherapy drug oxaliplatin (OXA) in reducing the growth of F.nucleatum-infected tumors. This was not achieved when free sodium butyrate was administered with free OXA in solution rather than in the tablet form, signifying the importance of the enteric coating.

In summary, enteric-coated dosage forms offer advantages such as increased compliance due to their odor- and taste-masking abilities, as well as delayed, pH- or enzyme-dependent release kinetics. However, high doses (often in the range of g/d) are typically required in humans, necessitating a large number of capsules and an excessive amount of polymers to be administered. Additionally, delivery is generally targeted to the ileo-cecal region, with minimal exposure in the colon.

4.1.2. Microencapsulation

Microencapsulation has been employed as an approach to deliver microbial metabolites to the distal gut. The microencapsulation process entraps a core active ingredient within a gastro-resistant matrix, which can be comprised of lipid, protein, carbohydrate, or polymeric materials (Choudhury et al., 2021). As such, microencapsulated compounds are protected against the harsh environment of the stomach and can pass with food components into the small and large intestines. Like enteric coating, microencapsulation also aids in masking unpleasant organoleptic properties.

Lipid matrices

Sodium butyrate has been microencapsulated into lipid matrices comprised of hydrogenated plant-derived triglycerides, which are generally 14–20 carbons in length (Facchin et al., 2020; Krokowicz et al., 2014; Lewandowski et al., 2022; Roshanravan et al., 2020). These formulations are commercially available, marketed as Debutir^® (Polfa Łódź), Butyrose^® Lsc^® Microcaps (Sila Srl), Intesta (Bioton), and a butyrate gut health supplement (Body Bio), which have all been tested in clinical studies. The lipid matrix protects against the acidic stomach environment and is degraded upon exposure to intestinal lipases.

Debutir^®, at a dose of 1000 mg/d (two 500 mg capsules daily, each containing 150 mg sodium butyrate), reduced the clinical severity of IBS and diverticulitis in adult patients without causing side effects (Banasiewicz et al., 2013; Krokowicz et al., 2014). However, the same efficacy was not achieved in children/adolescents with IBD when Debutir^® was administered at the same dose (Pietrzak et al., 2022). Similarly, administration of 1000 mg/d of Intesta (two 500 mg capsules daily, each containing 150 mg sodium butyrate), which also contains HPMC as a matrix constituent, yielded substantial reductions in self-reported disease severities in IBS patients (Lewandowski et al., 2022). Butyrose^® Lsc^® Microcaps (BLM), given at a dose of 1800 mg/d (three 600 mg capsules daily, each containing 500 mg sodium butyrate), were found to alter the gut microbiota in IBD patients by increasing SCFA-producing bacteria (Facchin et al., 2020). However, no significant effects on clinical activity were observed in this study, potentially due to the small sample size and short treatment duration of 2 months. In support of this notion, BLM therapy at the lower dose of 1200 mg/d (2 capsules) for a longer duration of 12 months was significantly more effective than a placebo in maintaining clinical remission, improving residual symptoms, and reducing inflammatory markers in IBD patients concurrently receiving standard therapy with mesalamine (Vernero et al., 2020).

Microencapsulated sodium butyrate has also been tested in patients with obesity and T2D. Children with pediatric obesity who received BLM at a dose of 20 mg/kg/d (to a maximum of 800 mg/d) exhibited significantly higher rates of BMI reduction, decreased waist circumferences, improved insulin sensitivities, and diminished inflammation compared to those who received placebo (Coppola et al., 2022). In T2D patients, sodium butyrate microencapsulated in a matrix comprised of medium-chain triglycerides (MCT) and HPMC significantly reduced in blood pressure at a dose of 600 mg/d (~550 mg sodium butyrate) (Khosravi et al., 2022; Roshanravan et al., 2017). Additionally, this treatment led to reductions in inflammation and oxidative stress biomarkers (Roshanravan et al., 2020). Although these capsules did not significantly improve glycemic control on their own, co-administration with the prebiotic fiber inulin significantly reduced fasting blood sugar (Khosravi et al., 2022; Roshanravan et al., 2017). Notably, a protocol for a clinical trial evaluating the effects of this treatment on PGC-1α, PPARα, and UCP-1 regulation in obese individuals has been recently published (Amiri et al., 2023). While the effects of butyrate on these parameters in preclinical species have been elucidated, this study marks the first evaluation in humans.

In summary, while microencapsulated sodium butyrate formulations have generally demonstrated efficacy in IBD patients, their effects in patients with metabolic diseases are mixed. As each application remained reliant on high oral doses, further development is needed to lower the dosage requirements.

The Trp metabolite indole has also been microencapsulated in a lipid matrix comprised of MCT (Shimada et al., 2013). These microcapsules were originally developed to protect probiotic Bifidobacteria from the acidic gastric environment, thereby enabling their activity in the intestine (Taki et al., 2005). In mice, the indole-containing microcapsules were found to dissolve in the distal small intestine after oral administration (Shimada et al., 2013). Evidently, the released indole traversed to the colon, as the mice showed increased expression of TJP in colonic, but not small intestinal, IECs. Germ-free (GF) mice, which had reduced fecal indole levels and decreased expression of TJP, showed enhanced resistance to colitis after receiving indole-containing microcapsules (15 mg/d (0.369 mg indole/d) for 2 wks). While over 90% of GF mice died upon colitis induction, only 15% died when treated with the indole-containing microcapsules. Additionally, this treatment was also protective against colitis in specific pathogen-free (SPF) mice with normal fecal indole levels. These results support future clinical applications testing the effects of microencapsulated indole in patients with IBD.

Polymer matrices

The Trp metabolite IAld has been entrapped into a polymer-based microcapsule formulation (IAld-MP), and well-characterized in preclinical studies. The formulation includes Eudragit^® S, Eudragit^® L, and ethyl cellulose as encapsulating materials (Puccetti et al., 2018). The most acid-resistant formulation was obtained by spray-drying a 2:1 ratio of Eudragit^® S:Eudragit^® L in an ethanol solvent system. The pharmacokinetics demonstrated that IAld was detected at high levels in the mouse intestine within 1 h after oral administration, corresponding to an intestinal half-life of ~3 h (Puccetti, Pariano, Borghi, et al., 2021). C_max values, while high in the intestine, were very low in all other tissues, suggesting that IAld is extensively metabolized. Pharmacodynamically, IAld-MP upregulated AhR-responsive genes significantly for as long as 6 h after administration, which, given the short half-life of IAld in the intestine, was likely mediated by downstream AhR agonists derived from IAld.

At a dose of 18 mg/kg, IAld-MP showed efficacy in ameliorating both intestinal barrier disruption and metabolic dysfunction in HFD-fed mice (Puccetti, Pariano, Borghi, et al., 2021). In another study, this treatment mitigated liver inflammation and fibrosis in a mouse model of primary sclerosing cholangitis via IL-22 induction, which strengthened the intestinal barrier to prevent the transepithelial passage of gut microbes (D’Onofrio et al., 2021). In this model, IAld-MP demonstrated an ability to positively influence the composition of the gut microbiome as it expanded the abundance of the IAld-producing species L.reuteri.

IAld-MP has also shown promise in mitigating infections in both the lungs and the GI tract. In mice with cystic fibrosis (CF) whose intestines were infected with Candida (C).albicans fungus, IAld-MP greatly reduced inflammation in the GI tract and the liver, where IAld-MP prevented fungal dissemination across the gut epithelium (Puccetti, Pariano, Renga, et al., 2021). When fungal infection instead occurred in the lungs of CF mice, orally administered IAld-MP also ameliorated pulmonary inflammation. Interestingly, delivering IAld to the gut in IAld-MP positively modulated the composition of the lung microbiome in CF mice. In another study, IAld-MP protected mice from immune checkpoint (CTLA-4) inhibitor-induced gastrointestinal toxicity (Renga et al., 2022). Importantly, IAld-MP preserved CTLA-4 inhibitor efficacy, demonstrating its potential to be effectively limit side effects when co-administered with immune checkpoint inhibitors.

Recently, another Trp metabolite, indole-3-propionic acid (IPA), was microencapsulated within a matrix comprised of sodium alginate (SA) and prebiotic resistant starch. These microcapsules (IPA@MC) were further coated with chitosan using a microfluidic electrospray platform to minimize IPA leakage and increase drug loading efficiency (K. Yang et al., 2022). The chitosan coating also resulted in smaller, denser microcapsules due to electrostatic interactions between the negatively-charged chitosan amino groups and the positively-charged SA carboxyl groups. The IPA@MC allowed for slower IPA release in the acidic SGF and a more rapid release in SIF, demonstrating its ability to protect IPA during its transit through the stomach. Additionally, their strong adhesion ability due to their wrinkled surface texture enabled longer retention times in the GI tract. IPA@MC was significantly more protective against colitis than either IPA or MC alone, demonstrating the synergistic effects accomplished by this dual prebiotic-postbiotic formulation. Importantly, IPA@MC increased the abundance of SCFA-producing bacteria, likely owing to the fermentable nature of the prebiotic microcapsule components.

Lastly, sodium butyrate has also been formulated into polymer-based microcapsules for systemic, rather than oral, delivery. Poly(lactic-co-glycolic acid) (PLGA) and poly(N-isopropylacrylamide) (PNIPAM) copolymers were formulated into microspheres loaded with sodium butyrate to form sodium butyrate (NaB)-loaded PLGA-PNIPAM (PP-NaB) microparticles (Cheng et al., 2016). PLGA is a biocompatible copolymer that degrades to non-toxic, metabolizable lactic and glycolic acids. Despite the high stability of PLGA-based particles, they tend to display poor drug encapsulation efficiencies and rapid drug release. To address this limitation, the amphiphilic polymer PNIPAM was incorporated into the formulation. PNIPAM copolymers possess thermosensitive properties in that they are liquid below 30°C, but morph into a gel-like solid state above 30°C. As such, PNIAM-containing microspheres are suitable for administration via injection. Additionally, PNIPAM coating slows particle degradation, enabling both prolonged retention in the body and extended drug release. In a mouse model of acute myocardial infarction (AMI), i.v. injection of PP-NaB directly into the ischemic areas of the heart ameliorated cardiac dysfunction, inflammation, and fibrosis in AMI mice to a greater extent than NaB alone. Mechanistically, butyrate inhibited ROS generation and promoted autophagy in cardiomyocytes while promoting angiogenesis in endothelial cells. As such, PP-NaB is a promising drug delivery system for the controlled release of butyrate after targeted injection.

4.2. Esterification to dietary fiber

Esterification of microbial metabolites to dietary fibers such as high amylose maize starch (HAMS) and inulin represents another pharmaceutical approach to achieve colon-targeted delivery, whereby large amounts of the metabolite can be released upon bacterial fermentation of the fiber. Esterification eliminates the unpleasant organoleptic properties of SCFAs and greatly reduces the amounts of fiber that need to be consumed to raise colonic concentrations via dietary fiber supplementation. Further, in contrast to consumption of ordinary dietary fibers, administration of acetylated, propionylated, and butyrylated fibers can selectively elevate the concentration of those specific SCFAs in the colon. In comparison to enteric coating, esterification to dietary fiber results in a more extended-release profile and is targeted to the colon, where the majority of fermentative bacteria reside. This approach also has the added advantage of supplying the prebiotic fiber itself as a fermentable substrate for commensal bacteria, which can contribute to expanding the abundance of beneficial microbes (Figure 4).

Figure 4: Dietary fiber-esterified microbial metabolites.

Metabolites esterified to high amylose maize starch or inulin are released upon enzymatic (host or bacterial) hydrolysis. Fermentation of the fiber backbone produces additional SCFAs. Graphic created with BioRender.com.

4.2.1. High amylose maize starch (HAMS)

The first study of SCFAs esterified to HAMS demonstrated that consumption of acetylated, propionylated, and butyrylated HAMS (HAMSA, HAMSP, and HAMSB, respectively) selectively raises the concentration of the corresponding SCFA in the distal GI tract in rats (Annison et al., 2003). In humans, it was found that roughly 80% of ingested esterified butyrate is released after HAMSB consumption, where 60% is targeted to the colon (J. M. Clarke et al., 2011). HAMSB administration has shown promising effects in reducing colonocyte DNA damage in both rats and humans, suggesting its potential to reduce the risk of colorectal cancer (J. M. Clarke et al., 2012; Conlon et al., 2012; Le Leu et al., 2015; Toden et al., 2014). A recently published protocol describes what will be the first clinical evaluation of HAMSB’s protective effects in individuals with a genetic form of colon cancer (J. Clarke et al., 2023).

HAMSB supplementation has shown promise in both preclinical IBD and metabolic disease models. A diet containing 15% HAMSB ameliorated T cell transfer-induced colitis in mice, associated with increased IL-10-producing Treg cells in the colon (Furusawa et al., 2013). HAMSB also reduced both colonic and hepatic inflammation in a dextran sodium sulfate (DSS)-induced colitis model, indicating a positive modulation of the gut-liver axis (L. Li et al., 2021). In genetically-diabetic (db/db) mice, HAMSB supplementation reduced fasting blood glucose and hepatic lipid accumulation, increased insulin sensitivity, and ameliorated inflammation in both liver and adipose tissues (Pedersen et al., 2023).

Intake of HAMSA, but not HAMSB, ameliorated the clinical severity of C.rodentium infection in mice (Yap et al., 2021). HAMSB administration did not result in increased fecal butyrate concentrations in this model, suggesting that enteric infection may affect the composition or function of commensal bacteria such that their ability to release butyrate from the fiber is reduced. Protection against infection by HAMSA was associated with increased IL-22-producing IELs, which are the first line of immunity against enteric infection, and increased IgA-producing B cells. Additionally, antimicrobial peptides and markers of epithelial repair in IECs were upregulated by HAMSA.

Both HAMSA and HAMSB provided a high degree of protection against diabetes in NOD mice, a preclinical T1D model (Mariño et al., 2017), where each operated through distinct mechanisms. While acetate hindered the ability of pathogenic B cells to expand populations of autoreactive T cells, butyrate boosted the number and function of Treg cells. As a result, their combined supplementation largely prevented T cell-mediated pancreatic β-cell destruction. Based on these findings, HAMS esterified to both acetate and butyrate (HAMSAB, Ingredion Incorporated, Bridgewater, NJ, USA) was then tested in patients with T1D at a dose of 40 g/day and found to recapitulate the preclinical findings in that circulating B and T cells were modulated towards a more regulatory phenotype (Bell et al., 2022). Another clinical trial examining the effects of HAMSAB in children and adolescents with recently diagnosed T1D was recently completed (NCT04114357), with no results yet posted.

Many of these studies report that supplementation with SCFA-conjugated HAMS induces an expansion of bacteria that can utilize the starch. For example, P.distasonis is reported to be increased by HAMSA, HAMSB, and HAMSAB (Bell et al., 2022; J. M. Clarke et al., 2011; Le Leu et al., 2015; West et al., 2013; Yap et al., 2021). Additionally, HAMSAB was found to induce a switch in the metabolism of gut microbes, where SCFAs are primarily produced via glycolysis rather than fermentation post-administration (Bell et al., 2022). Further studies are needed to explore the long-term effects of HAMS intervention on the gut microbiota.

A recent study synthesized IAA esterified to HAMS (HAMSIAA) to target IAA delivery to the colon (Song et al., 2023). Whereas administering IAA via other routes, including intragastric (i.g.), i.p., and oral, only increased colonic IAA concentrations by 10–30x relative to baseline, administration of HAMSIAA resulted in a 200-fold increase in colonic IAA concentrations after just one week. HAMSIAA was more effective in ameliorating colitis-induced intestinal pathology than HAMS or IAA administered alone, indicating that synergistic effects can be achieved with HAMSIAA administration. As HAMS can be fermented by gut microbes to produce SCFAs, esterification of Trp metabolites to dietary fibers represents a valuable approach for co-delivering SCFAs and Trp metabolites to the distal GI tract.

Notably, HAMSIAA with a degree of esterification (DE) ~0.3 exhibited the highest efficiency of colon-targeted IAA in mice. At DE>0.3, the steric hindrance caused by closely packed IAA side chains likely prevented bacterial esterase enzymes from accessing the ester bonds, resulting in reduced delivery efficiency. The HAMSB applied in the aforementioned studies had a similar DE of ~0.25, and the colonic fermentation efficiency also decreases with increases in DE (Nielsen et al., 2018). As butyrate is much smaller than IAA, it is possible that this can be attributed to the decline in solubility resulting from replacing polar hydroxyl groups with more nonpolar butyryl groups rather than to increased steric hindrance.

4.2.2. Inulin

Inulin is a fructan-type dietary fiber that is widely accessible from natural sources, such as chicory root. Inulin on its own or in novel formulations, such as the recently developed orally administered inulin gels, is known to positively modulate diverse host processes, with applications ranging from metabolic disease to anti-cancer applications (Han et al., 2021; Y.-Q. Qin et al., 2023). While acetate and butyrate are the most common SCFAs esterified to HAMS, studies on SCFAs esterified to inulin have primarily focused on inulin propionate ester (IPE).

The first study demonstrated that administration of 10 g/d IPE reduced energy intake after acute administration in overweight adults, associated with increased plasma levels of the anorectic hormones GLP-1 and PYY (Chambers et al., 2015). With long-term supplementation, IPE inhibited weight gain, reduced hepatic lipid content, and prevented deterioration in glycemic responses. IPE-27, which contains 27% propionate with a degree of esterification of ~0.75, produced the maximum amount of propionate when incubated with microbes derived from human fecal samples. Similarly to HAMSB, the efficiency of propionate release declined as the degree of propionate loading increased, likely due to the decline in solubility resulting from the addition of more nonpolar propionate groups (Polyviou et al., 2016). This may explain why IPE-27 was more effective than IPE-54 (54% propionate) at acutely lowering energy intake.

In another clinical study, long-term IPE supplementation improved pancreatic β-cell functions, where propionate was found to act directly on β-cells to stimulate insulin secretion (Pingitore et al., 2017). These findings suggest that the ability of IPE to release propionate in the colon allows it to become systemically available in high enough concentrations to reach distant tissues such as the pancreas and the brain. At a higher dose of 20 g/d, IPE administration inhibited increases in hepatocellular lipid content in NAFLD patients (Chambers, Byrne, Rugyendo, et al., 2019). IPE has been incorporated into food products (a bread roll and a fruit smoothie) without changing their palatabilities, providing a more attractive way to deliver IPE on a population scale (Byrne et al., 2019). Another group found that IPE improved metabolic health in a mouse model of obesity-induced fatty liver disease, further underscoring its utility in treating metabolic diseases (X. Zhu et al., 2020). Lastly, both inulin acetate ester and inulin butyrate ester, in addition to IPE, were effective in preventing diet-induced obesity in mice (Drew et al., 2018). It is important to note that some conflicting findings have been reported after IPE administration. In contrast to previous findings (Chambers et al., 2015), IPE administration was found to attenuate reward-based eating behavior (Byrne et al., 2016) and enhance fat oxidation (Malkova et al., 2020) independently of changes in GLP-1 and PYY. On the other hand, another study found that IPE-mediated appetite reduction was associated with increased GLP-1 levels (Khatib et al., 2018). Finally, it has been reported that IPE is not significantly more effective than inulin alone at inhibiting weight gain (Khatib et al., 2018) or improving insulin resistance (Chambers, Byrne, Morrison, et al., 2019).

4.3. Prodrugs

Prodrugs have emerged as a promising pharmaceutical approach to improve the stability, bioavailability, and duration of action of small molecule drugs (Chien et al., 2023; Dahan et al., 2014). Prodrugs are designed to be metabolized into the active drug through enzymatic conversion. When formulated for oral administration, small molecule prodrugs must resist premature degradation in the stomach and small intestine in order to reach tissues in the distal GI tract. Prodrugs can also be designed to target specific tissues or cells outside of the GI tract, where they must resist metabolism for even longer. Through prodrug approaches, the efficacy and specificity of small molecule drugs are enhanced while minimizing side effects associated with off-target delivery.

Microbial metabolite prodrug formulations are currently only comprised of SCFA prodrugs, which are, in turn, predominantly comprised of butyrate prodrugs. Early clinical studies employing butyrate for its HDAC-inhibitory, anti-cancer properties resulted in largely unimpressive outcomes owing to its low potency, short circulating half-life, and inability to be administered orally due to rapid degradation and absorption. This necessitated administration via continuous i.v. infusion, which was sometimes associated with side effects due to high systemic/liver exposures (Table 1). Thus, research began to focus on the development of butyrate prodrugs, designed with the aims of prolonging butyrate’s half-life and enabling oral administration. As our understanding of the immunomodulatory effects of SCFAs have expanded, investigations have extended into the application of butyrate prodrugs for other therapeutic uses, and some have formulated other SCFAs into prodrugs (Table 3). Generally, the SCFA molecule is chemically conjugated to a promoiety through esterification. This eliminates their unpleasant organoleptic properties such that they can be orally administered, provides protection against degradation in the acidic gastric fluids, and, in some cases, improves their pharmacokinetic properties.

Table 3:

SCFA prodrugs

SCFA prodrugs tested in vivo
Structurea	Name(s)	Species/disease model (route of admin., dose, dosage form)	Key findings (ClinicalTrials.gov identifier and/or references)
	Tributyrin (TB)	Healthy mice (p.o.)	Improved PK characteristics compared to i.v. bolus sodium butyrate. Similar plasma butyrate AUC to p.o. sodium butyrate, suggesting complete hydrolysis in GI tract (Egorin et al., 1999).
		Preclinical LPS-, HFD- and ethanol-induced liver injury models (p.o. 1–2 g/kg/d or wk)	Transiently elevates butyrate conc. in hepatic portal vein, reduces liver inflammation (Donde et al., 2020; Miyoshi et al., 2011, 2015; Sato et al., 2020; Singhal et al., 2021; Vinolo et al., 2012).
		Clinical trials in patients with solid tumors (p.o. 400–600 mg/kg/d, hard gelatin capsules)	Low/variable plasma butyrate exposure, limited efficacy, side effects ranging from grade 1 to grade 3 (NCT00002677, Conley et al., 1998, Edelman et al., 2003).
		Healthy rats (p.o, cholesterol-based lipid emulsion)	Increased plasma butyrate AUC compared to p.o. free TB, enhanced absorption and increased p.o. bioavailability (Su et al., 2006).
	Pivaloyloxymethyl butyrate (AN-9, Pivanex)	Clinical trials in patients with solid tumors (i.v. infusion, ≤90 mg/kg/d, Intralipid® 20% emulsion)	Modest outcomes including disease stabilization in some patients, no dose-limiting toxicity (Patnaik et al., 2002; Reid et al., 2004).
		Spinal muscular atrophy mouse model (p.o., 400 mg/kg/d, Intralipid® 20% emulsion)	Increased survival and growth rate (Edwards & Butchbach, 2016).
	Butyroyloxymethyl diethylphosphate (AN-7)	Tumor-bearing mice (p.o., 50 mg/kg 3x/wk)	High oral bioavailability, inhibited tumor growth and increased survival without adverse events (Rephaeli et al., 2005, 2006). More potent than AN-9 (Tarasenko et al., 2008).
		Ischemic cardiac injury model (p.o.,25 mg/kg in mice, 15 mg/kg in rats)	Selectively affects cardiomyocytes to confer improved cardiac recovery from ischemia and reperfusion performed ex-vivo (Kessler-Icekson et al., 2012).
	3-O-[butanoyl-1,2-O-isopropylidene-α-D-glucofuranose] (monoacetone glucose 3-butyrate, MAG-3But)	Healthy rabbits (i.v.)	Greater mean residence times than i.v. bolus arginine butyrate, limited toxicity (Pouillart, Ronco, et al., 1992).
	6-O-[butanoyl…] (MAG-6But)	Tumor-bearing mice (i.p., 3×10−6 M ~ 0.468 ug/mL per mouse)	Extended survival times compared to i.p. arginine butyrate (Pouillart et al., 1991).
		Mice infected with encephalomyocarditis virus (i.p., 3×10−6 M ~ 0.468 ug/mL per mouse)	More effectively protected mice against infection than i.p. arginine butyrate (Pouillart, Cerutti, et al., 1992, p. 11).
	N-(1-carbamoyl-2-phenyl-ethyl) butyramide (FBA)	Healthy mice (p.o)	Similar serum butyrate PK compared to p.o. sodium butyrate, suggesting complete butyrate release in GI tract (Russo et al., 2021).
		HFD-fed mice and rats (p.o., 127.5 mg/kg/d)	Similar or improved efficacy compared to sodium butyrate, depending on the outcome measure (Mattace Raso et al., 2013; Mollica et al., 2017).
		Mice with DSS-induced colitis (p.o., 42.5 mg/kg/d) and mice with antibiotic-induced intestinal and liver injury (p.o., 637.5 mg/kg/d)	Improved intestinal barrier functions, reduced intestinal and liver inflammation, modulated microbiome (Simeoli et al., 2017; Lama et al., 2019).
	SCFA-Azithromycin esters (PAE and BAE)	Healthy mice (i.v. and p.o.)	High distribution into intestine, liver, and other tissues. Stable during their transit through the stomach, hydrolyzed in GI tract and liver. Some pAe metabolites detected in brain (Straß et al., 2021).
	2-seryl butyrate (SerBut)	Healthy mice (p.o.)	Increased butyrate levels in plasma and various tissues, notably the spinal cord and brain, compared to p.o. sodium butyrate (Cao, Budina, Raczy, et al., 2023).
Structures below	pMan-But and pMan-PhBut	T2D (db/db) excisional wound mouse model (topical, 1% hyaluronic acid + pMan-But or - PhBut (10% butyrate w/w))	pMan-But, but not pMan-PhBut, induced pro-regenerative shifts in cytokines and chemokines in the wound microenvironment and accelerated wound healing (Lauterbach et al., 2023).
SCFA prodrugs tested only in vitro
Retinoyloxymethyl butyrate (RN1)	2,3-dibutyroil-1-O-octadecyl glycerol (D-SCAKG)		Tyrosol-SCFA esters
		CIA mouse model of rheumatoid arthritis (p.o., 25 mg/mouse once daily) and EAE mouse model of multiple sclerosis (p.o., 100 mM SerBut in DW + 24 mg SerBut/mouse post-EAE induction once or twice daily)	Ameliorated disease progression, promoted anti-inflammatory immune cell phenotypes. Did not interfere with global immune response to vaccination (Cao, Budina, Raczy, et al., 2023).

^a:chemical structures generated using ChemDraw^® (PerkinElmer).

Tributyrin

Tributyrin (TB) is an oil-form butyrate prodrug consisting of three butyrate moieties esterified to a glycerol backbone, which are released upon hydrolysis. A preclinical pharmacokinetic study in mice demonstrated that oral TB administration at a dose of 7.8 g/kg could achieve therapeutically relevant peak plasma butyrate concentrations of ~1 mM, supporting its potential to be a more convenient clinical alternative to i.v. sodium butyrate (Egorin et al., 1999). Additionally, butyrate levels remained in therapeutic range for 2–3 times longer than i.v. bolus dose of sodium buyrate, suggesting the potential to achieve a longer duration of action. However, compared to oral sodium butyrate, oral TB administration did not result in enhanced butyrate exposure (AUC) (Egorin et al., 1999), and was later found to be completely hydrolyzed within 20 min of exposure to pancreatic lipase (Martín et al., 2011).

In a Phase I trial in patients with solid tumors, TB-filled gelatin capsules administered at a dose of 400 mg/kg/d failed to maintain sustained plasma butyrate concentrations above 0.45 mM (NCT00002677) (Conley et al., 1998). Similarly, a later clinical study aiming to achieve continuous exposure through three times daily dosing of 200 mg/kg reported even lower mean peak plasma butyrate concentrations of ~0.1 mM, which exhibited significant variability between patients and were associated with low/variable AUC values (NCT00002677), (Edelman et al., 2003). Plasma concentrations of TB were undetectable in most patients, further supporting that it is rapid hydrolyzed in the GI tract. It should also be noted that this dosage regimen required the consumption of 28 TB capsules three times daily, for a total of 96 capsules per day. Although many preclinical studies have demonstrated the efficacy of oral TB to reduce liver inflammation in various liver injury models, they have also relied on high doses of 1–2 g/kg, administered multiple times per day or week, to achieve therapeutic effects (Donde et al., 2020; Miyoshi et al., 2011, 2015; Sato et al., 2020; Singhal et al., 2021; Vinolo et al., 2012).

To address the issue of instability, TB has been formulated into both oral and injectable oil-in-water (O/W) emulsion dosage forms where it comprises the inner oil phase (Kang et al., 2011; Su et al., 2006; Su & Ho, 2004). Lipid emulsions are stable drug delivery systems that can enhance stability within the GI tract, enabling improved oral bioavailabilities and increased systemic exposures. A notable example is the successful development of a TB (10% w/v) nanoemulsion containing cholesterol, cholesterol oleate, and Lipoid E80 (phospholipids with ~80% phosphatidylcholine) (Su & Ho, 2004). Oral administration of this TB emulsion in rats increased plasma butyrate C_max by ~15-fold and decreased T_max by ~3-fold relative to free TB, indicating enhanced intestinal absorption (Su et al., 2006). In accordance, butyrate’s bioavailability was greatly increased, causing a ~5-fold increase in systemic exposure (AUC). Interestingly, this cholesterol-based formulation also enables tumor-targeted butyrate delivery by binding to the low-density lipoprotein receptor, which is upregulated on tumor cells to support their high rates of membrane synthesis (Su & Ho, 2004).

An additional advantage of TB emulsions is their high solubilization capacity, which allows them to function as effective carriers for poorly soluble drugs. Thus, TB emulsions are well-suited for combination anti-cancer therapeutic approaches as they enhance the delivery of poorly soluble anti-cancer drugs while providing their own anti-cancer effects. For example, the cholesterol-based TB nanoemulsion was utilized to carry all-trans-retinoic acid (atRA) (Su et al., 2008). Other TB emulsions have been also formulated to carry atRA and other poorly soluble anti-cancer drugs (Carvalho et al., 2017; Kang et al., 2012; Salata & Lopes, 2022). In summary, TB emulsion dosage forms can enhance systemic exposure to butyrate after oral administration and can overcome the solubility limitations of other anti-cancer drugs, leading to marked improvements in pharmacokinetic and pharmacodynamic properties.

Retinoyloxymethyl butyrate (RN1)

In addition to the atRA-tributyrin emulsion described above, retinoyloxymethyl butyrate (RN1) represents another example of prodrug designed to synergize the anti-cancer effects of butyrate and atRA. In a pilot study, retinoyloxymethyl butyrate (RN1) induced differentiation in cancer cells more potently than atRA or butyric acid (the conjugate acid of butyrate) alone by 40-fold and 9000-fold, respectively (A. Nudelman & Rephaeli, 2000). Interestingly, the effect of applying 0.5 μM atRA and 50 μM butyric acid together as a combined mixture resulted in lower anti-cancer effects than those obtained with 0.5 μM RN1. This was surprising given the low potency of butyrate (whose IC₅₀ is generally in mM range for anti-cancer effects). It was suggested that the lipophilicity imparted by atRA facilitated greater cellular uptake and intracellular release of butyrate. Moreover, in atRA-resistant cancer cells, RN1 reduced the IC50 for proliferation inhibition by 7-fold compared to that of atRA. A later study found that butyrate causes a DNA conformation that facilitates RA-induced gene transcription, thus providing the mechanism by which RN1 can overcome atRA resistance in certain cancer cell lines (Mann et al., 2003). Despite the ability of RN1 to co-deliver atRA and butyrate to the same cells, thus reducing the required doses of each drug, RN1 was never tested in preclinical or clinical studies.

Pivaloyloxymethyl butyrate (AN-9, Pivanex) and butyroyloxymethyl diethylphosphate (AN-7)

Pivaloyloxymethyl butyrate (AN-9, Pivanex, Titan Pharmaceuticals) is a prodrug which releases butyrate, along with formaldehyde and pivalic acid, upon hydrolysis. It was found to induce the differentiation and inhibit the proliferation of cancerous cells at concentrations around 10-fold lower and at a rate and over 100 times faster compared to butyric acid alone (Aviram et al., 1994; Rephaeli et al., 1991). Similarly to RN1, the higher potency of AN-9 was attributed to its higher lipophilicity, facilitating greater cellular uptake and enhanced intracellular release of butyrate (Zimra et al., 2000). AN-9 was then formulated as an emulsion in Intralipid^® 20% (20% soybean oil, 1.2% egg yolk phospholipids, 2.25% glycerin, and water) for i.v. infusion for clinical trials (Patnaik et al., 2002). At doses up to 90 mg/kg/d, AN-9 demonstrated modest outcomes, with up to 3% of patients exhibiting partial responses and 30% exhibiting disease stabilization (Reid et al., 2004). This likely halted further development of AN-9 as a cancer therapeutic. Interestingly, the AN-9 emulsion was recently found to be protective in a mouse model of spinal muscular atrophy, a childhood-onset degenerative disease caused by selective loss of motor neurons in the spinal cord, at an oral dose of 400 mg/kg/d (Edwards & Butchbach, 2016). Further research is needed to determine whether its therapeutic effects in this model were due to sufficient concentrations of butyrate becoming available in the CNS after oral AN-9 administration or were mediated by its activation of a gut-CNS axis mechanism.

Butyroyloxymethyl diethylphosphate (AN-7) is another butyrate prodrug that is structurally similar to AN-9, but releases butyrate along with phosphoric acid rather than pivalic acid upon hydrolysis. Unlike AN-9, this compound was initially developed using oral administration. Evidence supports that AN-7 has high oral bioavailability, as equivalent anti-metastatic activities in mice were achieved after oral and i.p. administrations (Rephaeli et al., 2006). AN-7 was found to be more potent than AN-9, as evidenced by its more significant inhibition of tumor growth and reduction of mean vessel density in a human colon carcinoma xenograft model, and superior efficacy in reducing lung lesions in a metastatic breast carcinoma model (Tarasenko et al., 2008). The enhanced efficacy of AN-7 may be due to its solubility in both aqueous and lipophilic solutions, whereas AN-9 is highly lipophilic and water-insoluble. Moreover, AN-7 demonstrated less toxicity to normal cells and greater selectivity for cancerous cells than AN-9 in vitro (Rephaeli et al., 2006). In mice inoculated with human prostate cancer cells, orally administered AN-7 (50 mg/kg, three times a week) inhibited tumor growth and increased survival without causing adverse effects (Rephaeli et al., 2005). However, AN-7 was never tested in human trials. Beyond its oncological applications, AN-7 has been investigated for its cardioprotective effects, which are mainly mediated by the HDAC-inhibitory effects of the released butyrate. A single oral dose was shown to confer protection against ischemic injury in cardiomyocytes in a mouse model of ischemic cardiac injury (Kessler-Icekson et al., 2012). Here, AN-7 upregulated heme oxygenase-1 expression in cardiomyocytes while leaving cardiofibroblasts unaffected. This selective effect was also evidenced in vitro (V. Nudelman et al., 2020).

Butyrate esters derived from cyclic monosaccharides

Butyrate esters derived from cyclic monosaccharides demonstrated promising pharmacokinetic profiles in preclinical studies, supporting their use as anti-cancer therapeutics. These prodrugs were developed for systemic administration with the aim of providing an alternative to continuous i.v. infusions of butyrate. In studies with rabbits, several esters were found to exhibit rapid tissue distribution post-i.v. administration and exhibited ~ 100 to 300 times greater mean residence times than arginine butyrate, an arginine salt of butyric acid with similar pharmacokinetic properties to sodium butyrate (Daniel et al., 1989; Pouillart, Ronco, et al., 1992). Two esters, namely 3- and 6-O-butanoyl-1,2-O-isopropylidene-α-D-glucofuranose (monoacetone glucose 3- and 6-butyrate, or MAG-3But and MAG-6But, respectively) were chosen for further evaluation given their promising pharmacokinetic characteristics, limited toxicities, and exceptional aqueous solubilities.

Cancer cells treated with MAG-3But or MAG-6But exhibited decreased tumorigenicity relative to those treated with arginine butyrate upon their inoculation into mice. Their amphipathic nature likely contributed to their superior performance by facilitating their transfer through biological membranes. In agreement with their pharmacokinetics, i.v. MAG-3But and MAG-6But administration extended the duration of action of butyrate to promote the maintenance of differentiated tumor cell phenotypes in vivo. Administration of these butyric esters to tumor-bearing mice either every 3 days (total of 6 treatments) or every 6 days (total of 3 treatments) significantly extended mean survival times compared to arginine butyrate. Importantly, this improved efficacy was not apparent when the treatments occurred every 2 days (total of 9 treatments). This indicates that the butyric ester prodrugs only outperformed arginine butyrate when longer periods of exposure were needed, such as when the number of doses was reduced. In addition to exhibiting improved anti-tumor properties in mice, MAG-3B protected mice against encephalomyocarditis virus infection twice as effectively as arginine butyrate (Pouillart, Cerutti, et al., 1992). The antiviral activity was shown not to be virucidal, but was speculated to involve the stimulation of cell-mediated immunity.

N-(1-carbamoyl-2-phenyl-ethyl) butyramide (FBA)

N-(1-carbamoyl-2-phenyl-ethyl) butyramide (FBA) is a butyrate prodrug with improved palatability developed for oral administration. In HFD-fed rats, FBA showed comparable efficacy and potency to sodium butyrate in ameliorating most outcome measures of hepatic steatosis and insulin resistance (Mattace Raso et al., 2013). However, FBA showed higher potency in reducing liver inflammation via the specific pathways of suppressing TLR signaling and inhibiting NF-κB activation. In a subsequent study, FBA was more effective than sodium butyrate in inhibiting weight gain and lipid accumulation in HFD-fed mice, with hepatic mitochondria identified as the primary target for these benefits (Mollica et al., 2017).

FBA administration has also emerged as a valuable strategy for alleviating intestinal injury. The efficacy of FBA was examined in colitic mice with DSS-induced colitis (Simeoli et al., 2017). While histopathological and disease activity scores were similar in mice treated with FBA or sodium butyrate, FBA was outperformed sodium butyrate in restoring TJP expression and increasing colon lengths. This study also confirmed the earlier observation that FBA leads to greater NF-κB inhibition than butyrate in the preventative setting. Additionally, FBA reduced inflammation in both the intestine and liver following antibiotic-induced injuries (Lama et al., 2019). Lastly, FBA has demonstrated protective effects against SARS-CoV-2 infection in IECs in vitro by reducing NF-κB activation and regulating various genes involved in antiviral pathways (Paparo et al., 2022).

The authors speculated that the stronger effects of FBA may be due to its potential to be transported into IECs by other members of the SLC family in addition to MCT1, facilitating improved absorption from the GI lumen. However, a recent study found that oral FBA administration results in similar serum butyrate concentrations and pharmacokinetic profiles to those achieved with oral sodium butyrate. This demonstrates that FBA releases butyrate within the GI tract and, therefore, there are no differences in absorption kinetics (Russo et al., 2021). Indeed, FBA underwent almost complete degradation during in vitro simulated oro-gastro-duodenal digestion, with the residual concentration falling below 10% of the initial FBA levels.

2,3-dibutyroil-1-O-octadecyl glycerol (D-SCAKG)

2,3-dibutyroil-1-O-octadecyl glycerol (D-SCAKG) is an alkylglycerol butyrate prodrug. D-SCAKG demonstrated slower hydrolysis during in vitro intestinal digestion than tributyrin (TB), a triacylglycerol butyrate prodrug described earlier (Martín et al., 2011). Complete hydrolysis took 20 min for TB as opposed to 60 min for D-SCAKG. Further, while butyrate was the only final product of TB hydrolysis, D-SCAKG mainly yielded 2-butyroil-1-O-octadecyl glycerol (M-SCAKG), a stable form of esterified butyrate that could potentially serve as an extended-release source enabling prolonged butyrate release in the GI lumen relative to TB. Indeed, the M-SCAKG level progressively decreased after 10 min of digestion and remained at approximately 40% of its initial level after 4h, indicating the molecule’s significant resistance to pancreatic lipases. It remains to be determined whether M-SCAKG can cross the gut epithelium to become systemically available. As such, in vivo investigations are warranted to determine whether D-SCAKG can increase the circulating half-life of butyrate via production of M-SCAKG.

SCFA-Azithromycin esters

The macrolide drug Azithromycin was adapted to enable delivery of SCFAs to the lysosomal compartment of immune and epithelial cells (Straß et al., 2021). Azithromycin is a derivative of erythromycin with more extensive tissue distribution, a longer elimination half-life, and better stability in acidic environments. Its two amine groups become positively charged at lower pH, trapping Azithromycin in acidic cellular compartments like the lysosome and resulting in high tissue-to-serum ratios. Esterification of the hydroxyl groups of Azithromycin in positions 11, 2’, and 4” with butyric anhydride and propionic anhydride produced butyrate- and propionate-Azithromycin ester prodrugs (BAE and PAE, respectively), capable of releasing three SCFA equivalents. Pharmacokinetic analysis in mice demonstrated that the systemic concentrations were similar after oral and i.v. administrations despite that the oral doses were 5x greater. Additionally, their whole blood concentrations rapidly decreased after both i.v. and oral administration, suggesting rapid tissue distribution. While the esters were stable during their transit through the stomach, they distributed extensively into the intestinal and liver tissues, as indicated by the abundance of downstream metabolites in those tissues after oral administration.

BAE was more hydrophobic, less soluble, less distributed, and less labile than PAE. Accordingly, PAE exhibited higher uptake, metabolic diversity, and systemic distribution than BAE. Interestingly, although macrolides are generally not considered to cross the blood-brain barrier (BBB) due to their size and polar surface area, metabolites of PAE were detected in the brain following both i.v. and oral administration. The increased hydrophobicity due to the propionate groups may have rendered PAE and its metabolites more amenable to non-specific BBB transcytosis routes, resulting in a promising potential strategy for gut-to-brain delivery of propionate. As Azithromycin has antibacterial properties, further research to enable more widespread use should focus on employing more practical, non-antibacterial macrolide carriers.

Tyrosol-SCFA esters (Tyr-SEs)

Tyrosol-SCFA esters (Tyr-SEs) have recently been developed as extended-release carriers that can deliver SCFAs to the distal GI tract. These phenolipids are produced by reacting the polyphenol and fatty acid using Candida antarctica lipase as a catalyst. The first studies of tyrosol-acyl esters incorporating long-chain fatty acids revealed their stability in the stomach and their ability to reach the cecum and colon, where they were hydrolyzed by Lactobacillus bacteria (X. Wang, Chen, Qiu, et al., 2022; X. Wang, Chen, Wang, et al., 2022). Importantly, however, the majority of hydrolysis occurred in the small intestine, where the formulation displayed sustained-release behavior. The most recent studies indicate that Ty-SEs incorporating acetate, propionate, or butyrate as the SCFA also display sustained-release behavior in the GI fluids (X. Wang, Wang, Cai, et al., 2023; X. Wang, Wang, Hu, et al., 2023). The ability of Tyr-SEs to cross the intestinal epithelium was tested using a rat everted gut sac model, which flips the mucosal surface of the intestinal tract outwards to enable the ex vivo study of uptake kinetics. It was shown that all three Tyr-SEs can cross the mucosal epithelium to become systemically bioavailable. Tyr-SEs also exhibited sustained-release behavior in rat plasma, where carboxylesterases in the blood mediated their hydrolysis. In summary, Tyr-SCFA esters may delay SCFA release until they reach the distal GI tract and improve oral SCFA bioavailabilities by virtue of their membrane-crossing abilities.

2-seryl butyrate (SerBut)

A recent innovation to improve the oral bioavailability of butyrate was accomplished by simply esterifying butyrate to serine (2-seryl butyrate, SerBut) (Cao, Budina, Raczy, et al., 2023). This molecule utilizes amino acid transporters to traverse the gut barrier, thereby becoming systemically available. Additionally, it is capable of crossing blood-CNS barriers. Pharmacokinetic analysis after oral administration demonstrated that SerBut outperformed sodium butyrate in increasing plasma butyrate concentrations and significantly increased butyrate concentrations in diverse tissues, including the liver, mesenteric LNs, spleen, lungs, spinal cord, and brain. SerBut was less toxic to bone marrow-derived dendritic cells at high concentrations (~2 mM) than butyrate, but retained its biological activities in suppressing co-stimulatory marker expression. Remarkably, SerBut administration greatly ameliorated disease progression in two preclinical autoimmune disease models. In the collagen antibody-induced arthritis (CAIA) model of rheumatoid arthritis (RA), SerBut reduced immune cell infiltration and prevented collagen loss in the joints, associated with a systemic increase in Tregs, decreased Th17 cells, increased Breg cells, and increased immunoregulatory macrophages. In the experimental autoimmune encephalomyelitis (EAE) model of multiple sclerosis (MS), SerBut decreased immune cell infiltration into the spinal cord, associated with increased inhibitory markers on CD4⁺ T cells, increased Tregs, and decreased activation markers on myeloid cells. SerBut also targeted myelin oligodendrocyte glycoprotein (MOG)-specific pathogenic cells that contribute to EAE progression, including MOG⁺ Th17 cells. Importantly, SerBut did not interfere with the global immune response to vaccination, demonstrating that it is a safe and effective mechanism to modulate immune cell populations and reduce inflammatory responses in the context of autoimmune disease.

Mannose-decorated, butyrate-conjugated co-polymer prodrugs (pMan-But and pMan-PhBut)

Most recently, topically-administered, mannose-decorated, butyrate-conjugated co-polymer prodrugs were developed to promote wound healing (Lauterbach et al., 2023). The co-polymers consist of two functionalized methacrylamide monomers: one with mannose side chains and the other with either aliphatic hydroxyl butyric esters (pMan-But) or phenyl butyric esters (pMan-PhBut) attached. The mannose groups serve both to increase solubility and to target butyrate delivery to immune cells expressing the mannose receptor for ligand scavenging. Additionally, the high molecular weight of the polymers increases viscosity to promote hydration in the wound. While both pMan-But and pMan-PhBut displayed negligible butyrate release in PBS, the two constructs displayed distinct release kinetics in complete media with fetal bovine serum, which displays esterase-like activity. pMan-PhBut released butyrate faster than pMan-But, with half-lives of 8.0 and 56.9 h, respectively. While this temporal difference allows for tunable butyrate delivery by administering a combination of the two compounds, it was found that the slower release of butyrate by pMan-But is more optimal for promoting wound healing as it does not blunt the important spike in inflammation that facilitates clearance of debris from the wound. In the type 2 diabetic (db/db) excisional wound mouse model, pMan-But, but not pMan-PhBut, was found to accelerate wound healing, by rebalancing the profile of secreted cytokines and chemokines in the wound microenvironment to favor regeneration. These results demonstrate that topical application of pMan-But has the potential to treat non-healing cutaneous wounds in patients with type 2 diabetes by enabling controlled release of butyrate in the wound microenvironment. As noted by the authors, this novel platform could be adapted to deliver other molecules to the wound microenvironment in a controlled and sustained manner.

4.4. Nanoformulations

An emerging pharmaceutical approach toward prolonging drug exposure profiles is to utilize nanoformulations. These formulations can function as targeted-delivery systems, where they preferentially localize the drug at target sites, and/or controlled-release carriers, where they extend the window of efficacy by maintaining drug levels within therapeutic range. Applications of these therapeutics have ranged from anti-cancer, to vaccination improvement, to delivery of biologics, where they have largely been formulated for administration via injection (i.v., intramuscular, or subcutaneous). Formulating these therapeutics into nano-scale carriers avoids the need for frequent, high-dose injections by prolonging the release of the active ingredient, which also mitigates the intensity of fluctuations in systemic concentrations. Additionally, nanoformulations also offer an approach to enhance stability within the harsh environment of the stomach. This affords the ability to target specific locations within the GI tract, increase bioavailabilities, and enable controlled- and/or prolonged-release of the active ingredient, thus greatly enhancing drug efficacies after oral administration.

Nanoformulations have shown great promise, particularly in the field of anti-cancer therapeutics. Nanocarriers can be engineered to target specific cancer cells or respond to unique biochemical signals within the tumor, enabling the targeted delivery of the therapeutic to the tumor microenvironment. Additionally, nanoparticles can be selectively accumulated into tumors due to their leakier blood vessels and impaired lymphatic drainage, which is known as the enhanced permeability and retention (EPR) effect. Nanomedicine enables the combination of multiple therapeutic modalities, such as chemotherapy and immunotherapy, leading to immensely improved anti-cancer efficacies. Besides cancer, immunomodulatory nanocarriers hold great promise for treating other diseases, such as chronic inflammatory, autoimmune, allergic, and infectious diseases. Generally, biocompatible polymer- or lipid-based nanocarriers are used to provide controlled-release, whereas targeted-release can be achieved by modulating the size and/or surface charge or by incorporating a targeting ligand on the surface.

Size is one of most important properties as it influences 1) release properties, as larger nanoparticles with lower surface area-to-volume ratios tend to display slower release kinetics, and 2) interactions with cell membranes, thus dictating kinetics of cellular uptake, systemic absorption, and tissue distribution (Dianzani et al., 2017). Smaller nanocarriers (<50 nm) are used to enable extended circulating half-lives because recognition by the reticuloendothelial and mononuclear phagocyte systems (RES/MPS) is greatly reduced. Surface modification with polyethylene glycol (PEG) is another strategy to mask recognition by RES/MPS cells while increasing the hydrodynamic radius to facilitate slower release. On the other hand, larger nanocarriers (>50 nm) allow for size-dependent accumulation in lymphoid tissues as they are recognized by these systems (Jiang et al., 2017; McCright et al., 2022). This strategy is often employed in the context of vaccination and autoimmune diseases, where targeting therapeutic delivery to sites of adaptive immune activation, namely the lymph nodes, is desirable. Moreover, immune cell recognition offers an approach to selectively target delivery to inflamed tissues, which are highly infiltrated by activated phagocytes.

Here, we will discuss the utilization of nanoformulations to deliver microbial metabolites, specifically SCFAs (Table 4). We will examine various delivery routes, including oral (Figure 5), parenteral, and intranasal administrations, and how they have been employed to modulate host processes in various therapeutic contexts.

Table 4:

SCFA nanoformulations

Abbreviated name	Formulation description	Physiochemical properties	Tested model (admin. route)	Key findings (references)
	Fabrication method
Oral formulations
Chol-But and Dx-Chol-But SLN	Cholesteryl-butyrate solid lipid nanoparticles (Chol-But SLN) and those loaded with dexamethasone	Chol-But SLN Size: 70–200 nm PDI: 0.2–0.65 ζ: −28 / −29 mV	Cancer cells in vitro	More potent anti-proliferative and pro-apoptotic effects than sodium butyrate Pellizzaro et al., 1999; Salomone et al., 2000; Serpe, Laurora, et al., 2004; Ugazio et al., 2001).
	Warm microemulsion dispersed in cold water		Polymorphonuclear neutrophils in vitro	More potent than sodium butyrate in inducing anti-inflammatory effects (Dianzani et al., 2006).
			Peripheral blood mononuclear cells from IBD patients	Reduced pro-inflammatory cytokine secretion more potently than sodium butyrate (Serpe et al., 2010).
		Dx-Chol-But SLN Size (avg): 72.9 nm PDI: 0.28	Mice with DSS-induced colitis (p.o.)	Reduced disease activity, indicating protection against degradation in the stomach (Dianzani et al., 2017).
PV-But NPs	Polyvinyl butyrate (PV-But) polymeric nanoparticles (NPs), coated with Pluronic F-127 (PEG-PPG-PEG)	Size: 120–205 nm PDI: 0.16–0.26 ζ: (−0.4)–(−0.6) mV	Healthy mice (p.o.)	NPs ~120 nm were retained in the GI tract for 3 h (Mu et al., 2021).
			Mice with DSS-induced colitis (p.o.)	NPs ~120 nm suppressed colon length shortening and clinical scores (Mu et al., 2021).
	Single emulsion		Mice with DSS-induced colitis (p.o.)	More effective amelioration of colitis with larger NPs (~200 nm) than smaller NPs (~120 nm), potentially due to improved accumulation in inflammatory lesions (J. Li et al., 2021).
PEG-b-PV(SCFA) ester micelles	Amphiphilic block copolymers comprised of PEG block and polyvinyl (PV) block esterified to either propionate (Pr) or butyrate (Bu) self-assembled into micelles	PEG-b-PV(Pr) micelles Size: 37.04 ± 0.90 nm PDI: 0.0851 ± 0.0313 ζ: −2.32 ± 0.04 mV	Healthy mice (p.o.)	Retained for 48h in GI tract, sustained-release of SCFA (t1/2~360 min for Bu release) (Shashni et al., 2021).
			Mouse model of T2D (db/db) (p.o.)	PEG-b-PV(Bu) ester micelles improved glucose tolerance and decreased pancreatic pathology, no effect of butyrate (administered as butyric acid) alone (Shashni et al., 2021).
	Polymer dissolution in organic solvent (DMF) followed by dialysis in pure water	PEG-b-PV(Bu) micelles Size: 40.31 ± 1.80 PDI: 0.0585 ± 0.0147 ζ: −2.71 ± 0.18 mV	Mouse model of NASH and liver fibrosis (p.o.)	PEG-b-PV(Bu) ester micelles led to elevated butyrate in the liver for 72h, reduced hepatic lipogenesis and fibrosis, no effect of butyrate alone (Shashni et al., 2023).
			Mouse model of melanoma tumor metastasis (p.o.)	Both PEG-b-PV(Pr) and (Bu) micelles exhibited greater anti-metastatic effects than their corresponding free SCFA alone, where PEG-b-PV(Pr) showed a superior effect (Shashni & Nagasaki, 2023).
NtL- and Neg-ButM	Amphiphilic block copolymers comprised of HPMA (NtL) or MAA (Neg) block and MA block esterified to butyrate (But) self-assembled into micelles	NtL-ButM Size: 44.7 ± 0.8 nm PDI<0.1 ζ: −0.34 ± 0.5 mV	Heatlhy mice (p.o.)	NtL-ButM and Neg-ButM released butyrate predominantly in the ileum and cecum, respectively. Neg-ButM raised cecal butyrate conc. between t=4h and t=8h (R. Wang, Cao, Bashir, et al., 2022).
	Co-solvent evaporation (NtL-ButM); base titration (Neg-ButM)	Neg-ButM Size: 39.9 ± 1.6 nm PDI<0.1 ζ: −31.5 ± 2.3 mV	Antibiotic-, DSS-, and T-cell transfer-induced colitis and peanut allergy mouse models (p.o.)	NtL-ButM and Neg-ButM (1:1 mixture) improved barrier integrity and reduced disease severity more effectively than sodium butyrate. Increased the abundance of butyrate-producing bacteria in peanut allergic mice (R. Wang, Cao, Bashir, et al., 2022).
PSBA and PSBA@MAG	Amphiphilic block copolymers comprised of PEG block and polymethacrylate block conjugated to butyrate via disulfide linkage self-assembled into micelles (PSBA) and PSBA loaded with magnolol (MAG)	PSBA Size: 125.67 ± 1.68 nm PDI < 0.1 ζ: −23.9 ± 0.17 mV	Healthy mice and mice with DSS-induced colitis (p.o.)	PBSA@MAG displayed prolonged retention in the distal gut in DSS-colitis mice (~24 h) relative to healthy mice (~12 h). PBSA@MAG alleviated colitis relative to free MAG. Unloaded PBSA improved epithelial integrity and inflammation (Fan et al., 2023).
	Nanoprecipitation	PSBA@MAG Size: 171.67 ± 5.33 nm PDI <0.1 ζ: −22.3 ± 0.9 mV
Injectable formulations
PEG-CS-Au NC-NaB	PEG-coated composite nanoparticles (NPs) containing gold (Au) nanoclusters on chitosan (CS) template and sodium butyrate (NaB)	Size: 99.6 ± 18.3 nm (TEM); 147.62 ± 10.21 nm (DLS) ζ: +8.97 ± 5.8 mV NaB EE: 65%	Tumor-bearing mice (i.p.)	Reduced tumor volume and increased survival, effects not seen with NaB alone (Goswami et al., 2018).
	Ionotropic gelation
CS-PCL-NaB	Chitosan (CS)-coated Poly(ε-caprolactone) (PCL) nanoparticles loaded with sodium butyrate (NaB)	Size: 311.1 ± 3.1 nm PDI: 0.208 ± 0.007 ζ: +56.3 ± 2.6 mV NaB EE: 92.3%	(Healthy rats, intravitreal injection)	NPs preserved the integrity of the retinal epithelial cells due to controlled release while free NaB caused disruption and inflammation (Reis et al., 2022).
	Double emulsion
BS@PEAL-GPC3	Sodium butyrate and sorafenib (BS)-loaded mPEG-PLGA-PLL nanoparticles (PEAL) functionalized with glypican-3 (GPC3) antibodies	Size: 191.09 ± 2.86 nm ζ: +32.79 ± 0.37 mV	Tumor-bearing mice (i.v.)	Inhibited tumor progression more strongly than NPs loaded with sodium butyrate or sorafenib alone, selectively accumulated in tumors, reduced liver damage compared with sorafenib only NPs (Che et al., 2023).
	Double emulsion
LITA and LITA-CAN	Acetate encapsulated into liposomes comprised of DSPC, CDAN, cholesterol, and 1% or 5% DSPE-PEG2000 for LITA and LITA-CAN, respectively	LITA Size: 102.3 ± 7.5 nm	HFD-fed mice (i.p.)	Reduced whole-body and hepatic fat accumulation, improved liver function and promoted adipose tissue “browning” (Sahuri-Arisoylu et al., 2016).
	Thin film hydration	LITA-CAN Size: 108.2 ± 3.3 nm	Tumor-bearing mice (i.p.)	Reduced tumor growth and HDAC expression (Brody et al., 2017).
NaB-Lip	Sodium butyrate encapsulated into liposomes comprised of DPPC, cholesterol, and DSPE-PEG5000	Size: 80 nm NaB EE: 32.83%	F.nucleatum-infected tumor-bearing mice (i.v.)	NaB-Lip preferentially accumulated in tumors and reduced F.nucleatum bacterial loads, leading to sensitization to the chemotherapy drug oxaliplatin (L. Chen et al., 2023).
	Thin film hydration
Neg-ButM	See above in “NtL- and Neg-ButM” under “Oral formulations”.		CAIA mouse model of rheumatoid arthritis (s.c.)	Increased Tregs and suppressed myeloid cell activation to alleviate autoimmune response (Cao, Budina, Wang, et al., 2023).
Intranasal formulations
Ac-ALA	Acetate (Ac) encapsulated in α-linolenic acid (ALA) liposomes	Size: 141.6 nm PDI: 0.136	Mouse model of SARS-CoV-2 infection (intranasal)	Selectively distributed to the lungs, elevating acetate concentrations for ~8h. Reduced lung viral loads and pro-inflammatory cytokine levels to ameliorate lung pathology (McGill et al., 2023).
	Thin film-hydration

Figure 5: Microbial metabolite nanoformulations developed for oral administration.

Cholesteryl butyrate solid lipid nanoparticles and polyvinyl-butyrate nanoparticles are larger in size, enabling them to be recognized and taken up by immune cells. Combined with enhanced epithelial barrier leakiness, this facilitates accumulation in sites of inflammation. Block copolymer micelles are comprised of a hydrophilic block and a hydrophobic, metabolite-conjugated, block. In the distal gut, at concentrations below the critical micelle concentration, demicellization followed by enzymatic hydrolysis occurs. Graphic created with BioRender.com.

4.4.1. Oral formulations

Cholesteryl butyrate solid lipid nanoparticles (Chol-But SLN)

Chol-But SLN are equimolarly composed of cholesterol and butyrate, originally designed to prolong the delivery of butyrate for anti-cancer applications. These nanoparticles are prepared by first creating a warm oil-in-water microemulsion containing Chol-But (the ester of cholesterol and butyrate) as the lipid matrix, Epikuron 200^® (95% soy phosphatidylcholine (PC) as a surfactant, taurocholate as a cosurfactant, and water (Brioschi et al., 2008). Dispersing this microemulsion in cold aqueous medium solidifies the lipid matrix, producing Chol-But SLNs. In vitro, Chol-But SLNs were more potent than sodium butyrate in terms of their anti-proliferative and pro-apoptotic effects on cancerous cells (Pellizzaro et al., 1999; Salomone et al., 2000; Serpe, Laurora, et al., 2004; Ugazio et al., 2001). Additionally, Chol-But SLN were found to inhibit the adhesion and migration of cultured cancer cells, highlighting their potential as anti-metastatic agents (Minelli et al., 2012). Interestingly, formulations with a higher concentration of PC than CholBut exhibited the strongest inhibition of cancer cell growth, likely due to PC’s role in enhancing nanoparticle internalization and hydrolysis (Pellizzaro et al., 1999; Ugazio et al., 2001). In combination with free paclitaxel or free doxorubicin, Chol-But SLN achieved synergistic cytotoxic effects in colorectal cancer cells while the combination with sodium butyrate did not (Serpe, Catalano, et al., 2004).

In addition to their anti-tumor effects, the anti-inflammatory effects of Chol-But SLN has been investigated. In neutrophils, Chol-But SLN inhibited adhesion to vascular endothelial cells, superoxide anion production, and myeloperoxidase release to a greater extent than did sodium butyrate (Dianzani et al., 2006). Additionally, treatment of peripheral blood mononuclear cells (PBMCs) derived from IBD patients with these nanoparticles resulted in significantly reduced pro-inflammatory cytokine secretion at significantly lower concentrations than sodium butyrate (Serpe et al., 2010). The efficient internalization of Chol-But SLN likely enables intracellular mechanisms of butyrate action such as HDAC inhibition, contributing to their greater anti-inflammatory effects.

More recently, a SLN formulation co-loaded with dexamethasone and butyrate (DxCb-SLN) was developed and administered to mice with DSS-induced colitis via oral gavage (Dianzani et al., 2017). After just 3d of treatment with this combination nanoformulation (0.1 mg/kg BW Dx and 4 mg/kg Cb), plasma concentrations of TNF-α and IL-1β significantly decreased, along with body weight recovery and a reduction in disease activity scores. Notably, each drug administered alone produced no significant effects. As noted by the authors, further pharmacokinetic and pharmacologic characterizations comparing the simultaneous administration of the two free drugs versus DxCb-SLN are needed to determine the precise mechanisms that led to the improved efficacy of DxCb-SLN.

In summary, Chol-But SLNs offer the following advantages: 1) protection against the acidic gastric environment, 2) controlled release, and therefore prolonged drug exposure, and 3) targeted delivery to inflamed tissues as a result of preferential uptake by immune cells.

Polyvinyl butyrate nanoparticles (PV-But NPs)

Polyvinyl butyrate nanoparticles (PV-But NPs) were developed to deliver butyrate to the lower intestine in a controlled manner after oral administration (J. Li et al., 2021; Mu et al., 2021). The PV-But polymers were formulated into nanoparticles and coated with Pluronic F-127 as a stabilizer, which is a triblock copolymer comprised of polyethylene glycol (PEG) and polypropylene glycol (PPG). The neutral charge imparted by this stabilizer may enable PV-But NPs to penetrate the mucus layer of the intestine.

PV-But NPs were found to be resistant to hydrolysis by pancreatic lipases, likely due to the limited access of the enzymes to the hydrophobic PV-But chains. The PV-But NPs were retained in the distal GI tract for ~3h after administration to mice by oral gavage, and ameliorated DSS-induced colitis in mice more effectively than sodium butyrate at the same dose. As this was accomplished without raising the concentration of butyrate in the intestinal lumen, it was suggested that the NPs resistance to hydrolysis allows for slow butyrate release in the GI tract, providing a steady supply of sufficient butyrate to produce anti-inflammatory effects.

Owing to their size, PV-But NPs were endocytosed by macrophages, which are abundant in inflamed colonic tissues. Hydrolytic intracellular enzymes can then release the butyrate from the polymer, allowing it to function as an HDAC inhibitor to suppress pro-inflammatory NF-κB signaling. A later study found that larger PV-But NPs (~200 nm) were more effective than smaller ones (~100 nm) in ameliorating colitis in mice, and that all-trans retinoic acid (atRA) can act synergistically with butyrate to improve colitis when incorporated into the NPs (J. Li et al., 2021).

Polyethylene glycol-b-polyvinyl-SCFA ester micelles (PEG-b-PV(SCFA) ester micelles)

Polyethylene glycol (PEG)-b-polyvinyl (PV)-SCFA esters are amphiphilic block copolymers composed of a hydrophilic PEG block and a hydrophobic PV-SCFA ester block, where either propionate or butyrate are esterified to create PEG-b-PV(Pr) and PEG-b-PV(Bu), respectively (Shashni et al., 2021). These copolymers self-assemble into stable nanoparticulate micelles with the hydrophobic, SCFA-conjugated block protected in the core of the micelle and the hydrophilic PEG block as the corona shell. The PEG portion of the polymer functions not only to sterically inhibit interactions with other cells, rendering the NPs non-immunogenic, but also to impart mucoadhesive properties, prolonging their retention in the GI tract (~48 h). Further, the ultra-small size of the nanoparticles (20–50 nm) also contributed to their non-immunogenicity by allowing them to evade detection by the RES/MPS. When administered orally in mice, the nanoparticles stayed in the GI tract while gradually releasing SCFAs, which were internalized into the bloodstream and detected in peripheral organs. The half-life of butyrate release from PEG-b-PV(Bu) was ~6 hrs, about 12 times longer than that of free butyrate (~30 min).

In this first report, the therapeutic efficacies of PEG-b-PV(Pr) and PEG-b-PV(Bu) were evaluated in the db/db mouse model of T2D. When administered through ad libitum drinking (60 mM), only PEG-b-PV(Bu) nanoparticles effectively improved glucose homeostasis relative to control drinking water, while no significant effects were observed for PEG-b-PV(Pr), butyrate, or propionate at the same concentrations. Notably, the mice consumed less drinking water when it contained butyrate or propionate, likely owing to their poor palatability, which may have affected the results. In contrast, consumption of drinking water containing PEG-b-PV(Pr) or PEG-b-PV(Bu) was comparable to that of control water, indicating their improved palatabilities. PEG-b-PV(Bu) was also tested in a choline-deficient, L-amino acid-defined high-fat diet-induced mouse model of NASH and liver fibrosis (Shashni et al., 2023). Whereas administration of 100 mM butyrate in the drinking water failed to prevent the pathogenesis of fatty liver disease, the same concentration of PEG-b-PV(Bu) nanoparticles significantly reduced hepatic lipogenesis and fibrosis. In another application of cancer treatment, surprisingly, PEG-b-PV(Pr) showed greater anti-cancer and anti-metastatic effects in a mouse model of melanoma metastasis than PEG-b-PV(Bu) (Shashni & Nagasaki, 2023). There, it was hypothesized that the lower metabolism of propionate within intestinal epithelial cells enabled more of it to enter the circulation and become available systemically, which is where these effects are primarily mediated. In summary, PEG-b-PV(Pr) and (Bu) showed prolonged retention in the GI tract and extended SCFA release, which yielded promising findings in a variety of disease models.

Neutral and negatively-charged butyrate-conjugated methacrylamide block copolymer micelles (NtL- and Neg-ButM)

Another example of self-assembling, SCFA-containing micelles is the recently developed butyrate-conjugated, methacrylamide-based block copolymer micelles (R. Wang, Cao, Bashir, et al., 2022). These micelles were formed with two co-polymers: one with neural charge and one with negative charge. The neutral charged polymer is a block copolymer comprised of a hydrophilic N-(2-hydroxypropyl) methacrylamide (HPMA) block and a hydrophobic N-(2-butanoyloxyethyl) methacrylamide (BMA) block, thus carrying butyrate esterified to the backbone sidechain of the hydrophobic polymer. Both of these formulations released most of their butyrate in simulated intestinal fluid in the presence of pancreatin with negligible release in simulated gastric fluid, demonstrating their resistance to acid degradation. When orally administered to mice, NtL-ButM predominantly released butyrate in the ileum, while Neg-ButM did so predominantly in the cecum, allowing a significant portion of the distal GI tract to be targeted by administering the two together.

A 1:1 combination of NtL-ButM and Neg-ButM was found to improve barrier integrity both in antibiotic-treated mice and in mice with DSS-induced colitis, and significantly reduce disease severity in a T-cell-transfer model of colitis. Additionally, this treatment was found to ameliorate the anaphylactic response to peanut challenge in peanut-allergic mice. Notably, NtL- and Neg-ButM were more effective than free sodium butyrate at fortifying epithelial barrier functions, and free sodium butyrate did not effectively reduce the allergic response. These micelles were also able to modulate the microbiome to increase the abundance of butyrate-producing genera such as Clostridium cluster XIVa. Thus, targeting butyrate delivery to the distal GI with butyrate-conjugated micelles represents a powerful approach toward reversing dysbiosis while ameliorating immunopathology, which has the potential to be applied in numerous other disease contexts.

Magnolol-loaded polyethylene glycol-b-polymethacrylate-butyrate nanoparticles (PBSA@MAG)

Recently, colonic pH and redox dual-responsive, butyrate-conjugated micelles (PSBA) were developed to target butyrate delivery to the inflamed colonic epithelium in IBD (Fan et al., 2023). Butyrate was conjugated to an amphiphilic polyethylene glycol (PEG)-b-polymethacrylate backbone via a reducible disulfide bond, which enables redox-responsivity due to the low redox potential of the colon lumen and the high level of glutathione in macrophages infiltrating inflamed tissues. Additionally, a sulfhydryl-containing polymer is released upon cleavage of the disulfide bond, which was found to adhere to sites of epithelial inflammation. Lastly, colon-targeted delivery was further enabled by the mucus-penetrating and pH-responsive nature of the PEG-b-polymethacrylate backbone. These micelles were used to encapsulate magnolol (PSBA@MAG), a lignan-type plant-derived polyphenol used in traditional Chinese medicine for its anti-oxidative and -inflammatory properties. By modulating the butyrate content of the polymer, the pH and redox-responsive release behavior of the micelles were optimized to circumvent premature inactivation of both butyrate and magnolol.

In vitro studies in simulated gastric, intestinal, and colonic fluids demonstrated the extended-release profile of PSBA@MAG, with minimal release at gastric pH. Studies performed in 10 mM glutathione (GSH) to simulate the low redox potential and macrophage-mediated disulfide reduction confirmed the redox-responsivity of the micelles. Encapsulation into PSBA@MAG ameliorated the macrophage and intestinal epithelial cell cytotoxicity caused by high MAG concentrations. Moreover, fluorescence analysis demonstrated that PSBA@MAG were efficiently taken up by both of these cell types in a time-dependent fashion, suggesting their ability to accumulate in inflamed intestinal tissue. Indeed, orally administered PSBA@MAG showed prolonged retention in the colons of mice with DSS-induced colitis (~24h) relative to healthy mice, enabling significant mitigation of colitis symptoms. Notably, unloaded PSBA failed to ameliorate colitis, likely owing to the relatively low butyrate dose (~46 mg/kg). However, PSBA improved histological indicators of epithelial barrier integrity and inflammation relative to micelles formulated from non-butyrate-conjugated polymers. Future studies using this platform would benefit from testing higher concentrations of unloaded PSBA.

4.4.2. Injectable formulations

PEG-coated chitosan-based Au nanoclusters-sodium butyrate composite nanoparticles (PEG-CS-Au NC-NaB-NPs)

Luminescent chitosan-based gold nanoclusters (CS-Au NC) and sodium butyrate (NaB) were formulated into composite nanoparticles coated with polyethylene glycol (PEG) to form PEG-CS-Au NC-NaB-NPs (Goswami et al., 2018). To synthesize these, Au NCs were embedded into a chitosan (CS) template. The inclusion of the negatively-charged NaB during synthesis led to the formation of the composite chitosan-based nanoparticles, with 65% encapsulation efficiency, where it served as an ion-gelating agent by interacting electrostatically with the positively-charged CS-Au NCs. Lastly, PEGylation on the NPs surface increased the size and colloidal stability of the nanoparticles while reducing their surface charge density. PEG-CS-Au NC-NaB-NPs were designed to prolong butyrate’s half-life for the purpose of improving its anti-cancer efficacy. Metallic nanoparticles (gold, silver, copper, and silica) have been associated with high solubility and stability, effective tumor targeting, and controlled release characteristics. Surface modification with PEG can further improve these parameters, as well as prolong the circulating half-life of nanoparticles by preventing recognition by RES/MPS.

In vitro, the luminescent properties of the NPs enabled visualization of their uptake into cancer cells, where they primarily localized to the lysosomal compartment. The NPs were internalized via endocytosis, which was found to be predominantly caveolae-mediated. Then, they were trafficked to the lysosomal compartment, likely due to the CS template, which contains N-acetylglucosamine residues that bind to N-linked glycans in the lysosomal membrane. Butyrate’s IC₅₀ value for cancer cell cytotoxicity was slightly lower when administered in the form of PEG-CS-Au NC-NaB-NPs than as free NaB. In vivo, i.p. administration of PEG-CS-Au NC-NaB NPs to tumor-bearing mice decreased tumor volume and increased survival, which were not observed in mice administered NaB alone at an equivalent dose. Thus, PEG-CS-Au NC-NaB-NPs enhanced butyrate’s anti-cancer effects, likely primarily due to PEG-mediated extension of butyrate’s in vivo half-life.

Chitosan-coated sodium butyrate-loaded PCL nanoparticles (CS-PCL-NaB NPs)

Poly(ε-caprolactone) (PCL)-based nanoparticles loaded with sodium butyrate and coated with chitosan (CS-PCL-NaB NPs) were formulated for intraocular administration and used to treat age-related macular degeneration (AMD) (Reis et al., 2022). This study explored whether the anti-angiogenic properties of butyrate, which are useful in cancer applications, can be harnessed to inhibit ocular neovascularization in AMD. The nanoparticle formulation was pursued with the aim of maintaining therapeutic butyrate levels in the intravitreal cavity for prolonged periods of time, thus reducing the frequency of administrations, minimizing side effects, and improving patient compliance. PCL was selected for its high molecular weight, which allows for a slower hydrolysis rate and therefore a more controlled release, while the CS enabled both controlled-release and enhanced emulsion stability. The positive surface charge also facilitates electrostatic interactions with negatively-charged cell membranes to promote cellular internalization.

The optimized formulation encapsulated NaB with a high efficiency of 92.3%. In PBS buffer, CS-PCL-NaB controlled NaB release, which displayed zero-order kinetics and reached nearly 100% over 35d. This ideal kinetic profile was attributed to the chemical stability of the PCL polymer and the swelling of the CS, which both provided barriers to NaB diffusion. Whereas free NaB disrupted the structure of the retina after injection into the vitreous cavity of the rat’s eye, CS-PCL-NaB did not, demonstrating their ability to eliminate the toxic side effects induced by burst exposure to NaB by gradually releasing it over time. Additionally, CS-PCL-NaB displayed similar antiangiogenic activity to the widely used AMD therapeutic bevacizumab (Avastin), revealing it to be a viable therapeutic alternative to inhibit neovascularization while reducing side effects and increasing compliance in AMD patients.

mPEG-PLGA-PLL-GPC3-butyrate-sorafenib nanoparticles (BS@PEAL-GPC3 NPs)

mPEG-PLGA-PLL-GPC3-butyrate-sorafenib nanoparticles (BS@PEAL-GPC3 NPs) are anti-GPC3 antibody-functionalized nanoparticles co-loaded with sodium butyrate and the anti-cancer drug sorafenib, originally developed to target the delivery of this therapeutic combination to hepatocellular carcinoma (HCC) cells (Che et al., 2023). The functional units of the co-polymer include: 1) a poly(ethylene glycol) monomethyl ether (mPEG) molecule, which minimizes recognition by the RES/MPS to prolong nanoparticle circulation, 2) a poly(lactic-co-glycolic acid) (PLGA) molecule, a widely used drug carrier that enhances drug loading and encapsulation efficiency, and 3) poly-L-lysine (PLL), a polymerized form of the amino acid Lysine, which contains amino groups in its side chain that can be modified with targeting moieties (P. Liu et al., 2012). In this study, an antibody against glypican-3 (GPC3), a coreceptor for Wnt that promotes Wnt/βcatenin signaling, was selected as the targeting moiety to enable efficient and specific binding to HCC cells since these cells overexpress GPC3 relative to normal liver tissue. Emulsifying these copolymers with sodium butyrate and sorafenib using a double emulsion method yielded BS@PEAL-GPC3 nanoparticles with an average particle size of slightly less than 200 nm. In addition to their size, PEGylation of the NPs promoted tumor accumulation due to the EPR effect.

In a previous study, the non-functionalized nanoparticle formulation was found to have slow and sustained drug release properties, characterized by an initial burst release (~23%) within 1d, followed by gradual release over 16d (~90% cumulative) (P. Liu et al., 2012). These properties proved to be highly beneficial for enhancing the therapeutic efficacy of BS@PEAL-GPC3. In a mouse liver cancer model, 5 total i.v. injections of BS@PEAL-GPC3 (equivalent to 500 mg/kg sodium butyrate and 20 mg/kg sorafenib) markedly suppressed HCC progression, showing stronger inhibition than nanoparticles loaded with the same doses of sodium butyrate or sorafenib alone (Che et al., 2023). Mechanistically, butyrate was supported to act through enhancing T-cell-mediated immunity, likely mediated through HDAC inhibition. Further, the sorafenib-only nanoparticles caused elevated serum levels of the liver damage markers ALT and AST while the combination nanoparticles did not, suggesting that the anti-inflammatory properties of butyrate alleviated the side effects of sorafenib. Lastly, the GPC3 targeting moiety enabled selective accumulation in HCC cells while minimizing accumulation in other organs. In summary, BS@PEAL-GPC3 nanoparticles effectively synergized the anti-cancer effects of butyrate and sorafenib and enabled their targeted delivery to HCC cells, which enhanced the anti-cancer effects while minimizing systemic side effects.

Liposome-encapsulated acetate (LITA) nanoparticles

One group has investigated the efficacy of acetate encapsulated into liposome-based nanoparticles (LITA) and delivered via i.p. injection (Sahuri-Arisoylu et al., 2016). This strategy was employed to prevent acetate from crossing the blood-brain barrier, allowing isolation of its peripherally-mediated effects. The nanoparticles were prepared via thin film hydration using N¹-cholesteryloxycarbonyl-3,7-diazanonane-1,9-diamine (CDAN), 1,2-distearoyl-sn-glycero-3-phosphocholine (DSPC), cholesterol, and (ω-methoxy-polyethylene glycol 2000)-N-carboxy-1,2-distearoyl-sn-glycero-3-phosphoethanolamine (PEG²⁰⁰⁰-DSPE) (32:32:35:1 molar ratio) as lipids. It was found that LITA treatment decreased lipid accumulation, improved hepatic function, and increased mitochondrial efficiency in the liver. In the adipose tissue, LITA treatment increased thermogenic capacity, resulting in reduced adiposity. Notably, these effects were accomplished by only 52.9 μg of acetate delivered in LITA NPs three times per week.

This formulation was modified to deliver acetate to tumor tissues (LITA-CAN) (Brody et al., 2017). The molarity percentage of DSPE-PEG²⁰⁰⁰ in the liposome membrane was increased from 1% to 5% in order to prolong circulation, which enabled accumulation of LITA-CAN in tumor tissues. The combination of their small size (<200 nm) and PEGylation promoted tumor targeting via the EPR effect. LITA-CAN treatment significantly reduced tumor sizes relative to control NPs in an HT-29 colorectal cancer xenograft tumor mouse model. Similar to butyrate’s inhibitory effects on tumor growth, the antitumor effects of acetate may be mediated by HDAC inhibition, as liver HDAC mRNA expression was significantly downregulated by LITA-CAN.

Liposome-encapsulated sodium butyrate (NaB-Lip) nanoparticles

In addition to the enterically-coated sodium butyrate tablets described previously (see section 4.1.1), Chen et al. also synthesized sodium butyrate-encapsulating liposomes (NaB-Lip) and tested their ability to mitigate F.nucleatum-induced chemotherapy resistance after i.v. injection (L. Chen et al., 2023). The liposomes were prepared by hydrating thin films containing 1,2-dihexadecanoyl-sn-glycero-3-phosphocholine (DPPC), cholesterol, DSPE-PEG⁵⁰⁰⁰, and sodium butyrate (5:4:0.5:1 molar ratio). In vitro release studies revealed that NaB-Lip showed a sustained-release profile, where ~50% of sodium butyrate was released over 24 h. NaB-Lip were shown to accumulate in F.nucleatum-infected subcutaneous colorectal tumors after i.v. administration in mice. While treatment with the chemotherapy drug oxaliplatin (OXA) or sodium butyrate alone showed minimal therapeutic effects, treatment with NaB-Lip+OXA significantly reduced tumor growth and improved survival, indicating their ability to overcome F.nucleatum-mediated chemoresistance. Analysis of the F.nucleatum content in the tumors after treatment confirmed the ability of butyrate to effectively eliminate the bacteria. Importantly, no other major organs were adversely affected by the treatment, indicating the safety of NaB-Lip.

Negatively-charged butyrate-conjugated methacrylamide block copolymer micelles (Neg-ButM)

The long-acting efficacy of neutral and negatively-charged butyrate-conjugated methacrylamide block copolymer micelles (NtL-ButM and Neg-ButM, described previously in 1. Oral formulations) administered via subcutaneous (s.c.) injection was recently investigated in a preclinical model of autoimmune arthritis (Cao, Budina, Wang, et al., 2023). Neg-ButM demonstrated superior lymph node-targeting abilities and longer retention than NtL-ButM. In fact, Neg-ButM were retained in systemic tissues including spleen and liver for longer than a month after s.c. administration. The in vitro effects of Neg-ButM to increase Tregs and suppress myeloid cell activation were recapitulated in vivo, leading to a significant amelioration of arthritis disease progression in collagen antibody-induced arthritis (CAIA) mice. Efficacy was achieved regardless of whether the injections directly targeted the site of inflammation (hocks) or occurred at more distant sites (abdominal). Moreover, amelioration was accomplished using only two subcutaneous injections of Neg-ButM. These results yield strong evidence supporting the use of Neg-ButM to target butyrate delivery to the lymphatic system. Given the important roles of the lymphatic system in fluid/inflammatory cell clearance and immune tolerance, this platform has the potential to be employed in numerous other disease contexts.

4.4.3. Pulmonary-targeted formulations

Acetate-encapsulating α-linolenic acid liposomes (Ac-ALA)

To harness the antiviral properties of acetate, acetate-encapsulating α-linolenic acid liposomes (Ac-ALA) were recently developed (McGill et al., 2023). In addition to short-chain fatty acids, unsaturated long-chain fatty acids such as ALA have also been shown to have antiviral properties, and ALA supplementation has been associated with improved recovery in COVID-19 patients. In molecular docking studies, ALA was found to bind surface proteins on viruses such as SARS-CoV-2 and respiratory syncytial virus, thus inhibiting their ability to fuse with host cells. Thus, ALA used to formulate liposomes encapsulating acetate with the aim of synergizing the fusion-inhibiting effects of ALA with the replication-inhibiting effects of acetate. In vitro, co-administration of acetate and ALA synergistically reduced viral plaque formulation and pro-inflammatory cytokine expression in virus-infected lung cells. In agreement with other studies outlined previously, IFN-β signaling was induced upon GPR43 activation by acetate.

Phospholipids, ALA, and cholesterol were formulated into liposomes using the thin-film hydration method, where the film was hydrated with acetate in PBS to produce Ac-ALA liposomes. In vitro, the release of the payload from the liposomes occurred upon the addition of inactivated virus and was dependent on the presence of ALA in the formulation. This indicated that ALA binding to viral surface proteins triggers particle degradation and selective acetate delivery to sites of viral infection. In vivo, fluorescently-labelled Ac-ALA liposomes administered intranasally to mice infected with a murine-adapted SARS-CoV-2 virus (MA10) distributed to only the lungs. Additionally, acetate levels remained elevated in the lungs for 8h post-administration, with no increase detected in the serum.

Finally, the in vivo efficacy of intranasally delivered Ac-ALA was demonstrated in MA10-infected mice. In the lungs of mice that received Ac-ALA liposomes, viral loads were significantly lower than those in mice that were treated with either acetate alone or ALA liposomes without acetate. In agreement with the in vitro studies, pro-inflammatory cytokines were reduced and IFN-β signaling was activated. This was associated with reduced epithelial damage and immune cell infiltration into the lungs. Thus, Ac-ALA liposomes are a promising strategy to control SARS-CoV-2 infection and ameliorate the associated cytokine storm.

IAld-MP inhalable dry powder

In addition to its formulation into microcapsules for enteric delivery (IAld-MP, described previously, see section 4.1.2), IAId has been formulated into an inhalable dry powder to treat conditions associated with pulmonary inflammation (Puccetti, Gomes Dos Reis, Pariano, et al., 2021). The spray-dried formulations contain mannitol with or without D-Leucine, both excipients chosen for their affordability and ability to enhance the dispersibility and flowability of dry powders. The ratio of 2:1 mannitol to IAld (w/w) was first evaluated for its efficacy in a mouse model of pulmonary fungal infection. The IAld-MP inhalable dry powder exhibited dual therapeutic benefits by both reducing infection and alleviating inflammation in the lungs.

This formulation was also tested in mice with cystic fibrosis whose lungs were infected with a fungal pathogen (Puccetti, Pariano, Renga, et al., 2021). Preliminary experiments revealed that maximum inhibition of fungal growth was achieved when aerosolized IAld was delivered intranasally at a dose of 18 mg/kg, where further dose escalations did not yield further inhibition. This dose was effective in diminishing fungal colonization and inflammation, whether administered before or after exposure to the fungal pathogen. However, an endogenous AHR ligand was more effective than intranasal IAld. Therefore, the authors hypothesized that more targeted delivery of IAld via dry powder insufflation rather than intranasal administration could improve the efficacy. Remarkably, a reduced dose of 4.5 mg/kg via dry powder insufflation was more effective than other administration routes, including intranasal or oral (via IAld-MP). Notably, there were no signs of AhR activation in distant organs, indicating successful pulmonary-targeted IAld delivery.

5. Discussion and Future Directions

Over the past few decades, the expansion of the human microbiome research field has uncovered the crucial role of microbial metabolites in maintaining host homeostasis and preventing the development of chronic disease. There are copious advantages of microbial metabolites over traditional microbiome-based therapeutic approaches, including their relative ease of production, formulation, and administration. A major impetus behind the development of metabolite-based therapies is their ability to simultaneously target many different host processes. This stands in stark contrast to single-target drugs with high affinity, selectivity, and potency. With the growing understanding of the limitations of single-target drugs, microbial metabolite are becoming increasingly valued as a versatile therapeutics with the potential to treat a broad spectrum of human diseases (Puccetti et al., 2023). Nonetheless, challenges such as low stability, rapid metabolism and consequent high dosage requirements have greatly hindered their translation into clinical applications.

To address these limitations, various pharmaceutical approaches have been pursued to enhance the efficacy of microbial metabolites. Enteric encapsulations, such as coating with Eudragit^® co-polymers and microencapsulation within triglyceride matrices, have been widely used in clinical applications to deliver butyrate to the distal gut. These strategies appear to largely target the terminal ileum/cecum, which has led to significant improvements in IBD patients. However, studies employing coated or microencapsulated forms of SCFAs have still relied on high doses (in the range of g/d) to achieve significant effects. In these studies, the large number of pills required likely outweighs the odor/taste-masking benefits of these formulations in terms of improving patient compliance. Although the delayed-release profiles achieved by pH-dependent polymers have been useful in targeting SCFA delivery to the distal gut, enteric encapsulation approaches could be further improved by the use of gastro-resistant materials that enable more extended-release profiles, which would likely reduce both the required doses and the administration frequencies. Moreover, approaches to improve the encapsulation efficiency of metabolites into microencapsulated formulations would also be instrumental in lowering required doses.

Esterification to dietary fiber is another approach that has been utilized in clinical studies to achieve colon-targeted delivery of SCFAs. This approach considerably reduces the amount of dietary fiber that needs to be consumed per day in order to yield therapeutically-relevant increases in colonic SCFA concentrations. Moreover, patient compliance is facilitated not only because esterification eliminates unpleasant organoleptic properties, but also because the dried fibers can simply be dissolved in water for easy consumption by the patient. However, as this platform aims to harness the metabolic capacity of the gut microbiota to achieve colon-targeted delivery, it is important to consider that disease-associated dysbiosis may influence its efficacy. In some diseases, the residing microbial community may not be able to efficiently release the metabolite by effectively degrading the delivery vehicle. Indeed, alterations in gut microbial composition have been shown to impact the efficacy of dietary fiber-esterified SCFAs in a disease-specific manner (Yap et al., 2021). More generally, other factors that can be altered in disease, such as changes to GI pH and transit time, should also be accounted for when developing oral formulations for colon-targeted microbial metabolite delivery.

When translating preclinical findings on microbial metabolite efficacies to clinical settings, several crucial questions must be addressed. These questions are particularly salient for platforms such as nanoparticles, which have been solely evaluated in rodent models thus far. Firstly, it is important to understand that key differences in rodent vs. human GI physiology may impact the translatability of novel microbial metabolite delivery systems. These differences include: 1) higher pH of the rodent stomach compared to that of humans, 2) faster intestinal transit time in rodents, matching their much higher metabolic rate, 3) presence of a distinct microbial fermentation chamber, the cecum, in rodents, whereas it is much less pronounced in humans, and 4) larger large intestine relative to body weight in rodents (Hugenholtz & de Vos, 2018). In addition to questions surrounding the translatability of preclinical doses, dosage forms, and administration frequencies, it is presently unclear whether a single microbial metabolite is sufficient to treat a multifaceted human disease. In this respect, co-delivery therapeutics combining multiple microbial metabolites could improve their overall efficacies (Nagler, 2020). Approaches to co-deliver prebiotics and microbial metabolites are promising due to their potential to simultaneously modulate the microbiome (by providing a digestible substrate) and host pathways (via direct effects of the metabolite). Interestingly, it has also been shown that delivering butyrate alone to the distal gut can increase the abundance of butyrate-producing bacteria (R. Wang, Cao, Bashir, et al., 2022), implying that prebiotics might not always be indispensable to modulate the microbiome towards enhanced SCFA production. Therefore, a deeper exploration is warranted to understand how delivering microbial metabolites either alone or in combination influences microbiome structure and for how long such changes can be maintained.

Besides that of SCFAs and Trp metabolites, innovative pharmaceutical approaches could be employed to maximize the therapeutic potential of other microbial metabolites. A noteworthy example is the microbially-produced secondary bile acids (SBAs), namely deoxycholic acid (DCA) and lithocholic acid (LCA). Gut dysbiosis has been linked to decreased colonic SBA production in various disease contexts (Agus et al., 2021; Thomas et al., 2022). Further, SBA supplementation via intrarectal delivery has led to improvements in disease severity (Sinha et al., 2020), and similar benefits can be achieved after oral administration of high SBA doses (Bhargava et al., 2020; Van den Bossche et al., 2017). However, these high doses may cause side effects, such as liver injury (Woolbright et al., 2014). Thus, like for SCFAs and Trp metabolites, the translatability of SBAs as therapeutics could be improved by their formulation into colon-targeting oral formulations that enable reduced doses and extended-release profiles. Similarly, lactate is another microbially-produced metabolite with the potential to ameliorate a variety of chronic diseases, especially when delivered to the colon. Lactate produced by exogenously administered probiotics has been shown to stimulate the growth of commensal gut bacteria, leading to increased SCFA production (Ghyselinck et al., 2020). In addition, lactate plays complementary, but distinct, roles along with SCFAs and Trp metabolites in maintaining host homeostasis. As such, formulating lactate into distal gut delivery platforms, either alone or in combination with other microbial metabolites, may yield further improvements in disease outcomes.

In summary, pharmaceutical strategies such as enteric encapsulations, esterification to dietary fiber, prodrugs, and nanoformulations have enabled significant strides towards translating the disease-modifying effects of microbial metabolites observed in preclinical studies into clinical applications. However, armed with the knowledge acquired through decades of developing more targeted, stable, and long-acting pharmaceutical formulations, the scientific community is well-positioned to develop the next generation of microbial metabolite-based therapeutics with superior translatability.

List of Abbreviations:

AHR

aryl hydrocarbon receptor

AAD

allergic airways disorders

ALA

α-linolenic acid

AMD

age-related macular degeneration

AMPK

AMP-activated protein kinase

AMPs

antimicrobial peptides

Au

gold

B(or P)AE

butyrate (or propionate) Azithromycin ester

BLM

Butyrose^® Lsc^® Microcaps

Bregs

regulatory B cells

CIA

collagen-induced arthritis

C_max

maximum concentration

CS

chitosan

DCs

dendritic cells

D-( or M-)SCAKG

2,3-(or 2-)dibutyroil-1-O-octadecyl glycerol

DSS

dextran sodium sulfate

EAE

experimental autoimmune encephalomyelitis

EPR

enhanced permeability and retention

FBA

N-(1-carbamoyl-2-phenyl-ethyl butyramide)

GF

germ-free

GI

gastrointestinal

GLP-1

glucagon-like peptide-1

GPCR/GPR

G-protein-coupled receptor

HAMS

high amylose maize starch

HDAC

histone deacetylase

HFD

high-fat diet

HIF

hypoxia-inducible factor

IA

indole-3-acrylic acid

IAA

indole-3-acetic acid

IAld

indole-3-aldehyde

IDO-1

indoleamine 2,3-dioxygenase

IE

indole-3-ethanol

IFN

interferon

ILA

indole-3-lactic acid

IPA

indole-3-propionic acid

IPyA

indole-3-pyruvic acid

Ig

immunoglobulin

ILC

innate lymphoid cells

IPE

inulin propionate ester

i.p.

intraperitoneal

i.v.

intravenous

LITA

liposome encapsulated acetate

LNs

lymph nodes

MAG-3-(or 6-)But

monoacetone glucose 3-(or 6-)butyrate

MC

microcapsules

MS

multiple sclerosis

NAFLD/SH

non-alcoholic fatty liver disease/steatohepatitis

NOD

non-obese diabetic

NP

nanoparticles

NtL (or Neg)-ButM

neutral (or negative) butyrate micelles

PCL

poly(ε-caprolactone)

PC

phosphatidylcholine

PEG

poly(ethylene glycol)

PLGA

poly(lactic-co-glycolic acid)

PNIPAM

poly(N-isopropylacrylamide)

PPAR

peroxisome proliferator-activated receptor

PV

poly(vinyl)

PXR

pregnane X receptor

ROS

reactive oxygen species

SARS-CoV-2

severe acute respiratory syndrome coronavirus 2

SCFAs

short-chain fatty acids

SerBut

2-seryl butyrate

(S)MCT-1

(sodium-dependent) monocarboxylate transporter-1

TB

tributyrin

T1D

type 1 diabetes

T2D

type 2 diabetes

TFR

T follicular regulatory

TFH

T follicular helper

Th

T helper

TJP

tight junction protein

TNF

tumor necrosis factor

Tregs

regulatory T cells

Trp

tryptophan

TGF

transforming growth factor