Inflammatory bowel diseases (IBDs) like Crohn's disease and ulcerative colitis affect millions, with current treatments having limitations in terms of side effects, targeting, and administration. Engineered probiotics offer a promising alternative, capable of modulating the gut microbiome, delivering therapeutic payloads, and interacting with the immune system to reduce inflammation. A major challenge is achieving controlled and prolonged retention of these engineered microbes within the gastrointestinal (GI) tract, specifically at sites of inflammation. Many probiotics, including *Saccharomyces boulardii* (*S.b.*), are rapidly cleared from the gut due to colonization resistance. This research aimed to overcome this limitation by engineering *S.b.* to express "synthetic adhesins," cell surface proteins that bind to ECM proteins overexpressed in inflammatory lesions of the GI tract. The hypothesis was that these adhesins would improve *S.b.* retention and thus therapeutic efficacy in IBD.

The gut microbiome's role in IBD pathogenesis is increasingly recognized, yet treatment remains challenging. Current drugs often have significant side effects and lack targeted delivery to inflammatory sites. Engineered microbes offer localized action, reducing systemic side effects and enabling oral administration. However, achieving prolonged gut residence time and targeted delivery remains a key obstacle. Studies have shown that many probiotics experience rapid clearance, necessitating frequent dosing. Inspiration was drawn from gut-resident microbes and pathogens which utilize cell surface proteins (adhesins) to bind to host cells or proteins, establishing spatial niches for survival. This study proposed harnessing this mechanism through the expression of synthetic adhesins on the surface of *S.b. to target it to the inflamed regions of the gut.

The research employed a platform approach involving genetic engineering of *S.b.* to display monomeric streptavidin (mSA) on its cell surface. This mSA acted as a handle for attaching biotinylated antibodies specific to ECM proteins (fibronectin, fibrinogen, collagen IV). The selection of these ECM proteins was based on their overexpression in colonic mucosa of patients with ulcerative colitis, as confirmed by analysis of seven whole-genome transcriptional datasets. Binding assays using biotin-coated plates and flow cytometry were performed to characterize the binding affinities of the engineered *S.b.* strains to ECM proteins. In vitro assays evaluated whether the genetic modifications altered *S.b.*'s probiotic phenotype, including its resistance to GI tract conditions, production of short-chain fatty acids (SCFAs), and ability to stimulate IL-10 production in murine bone marrow-derived dendritic cells (BMDCs) and inhibit IL-8 production in TNFα-stimulated HT-29 intestinal epithelial cells. Acute and chronic dextran sulfate sodium (DSS)-induced colitis models in mice were used to assess the pharmacokinetics (PK) and pharmacodynamics (PD) of the engineered *S.b.* strains. Fecal and colon tissue samples were collected to quantify *S.b.* levels. Clinical biomarkers of disease severity (body weight, colon length, cytokine expression, histological inflammation scores) were measured. Immunofluorescence staining visualized *S.b.* biodistribution in colon tissue.

Genetic engineering successfully enabled *S.b.* to express mSA on its surface, facilitating the attachment of biotinylated antibodies targeting specific ECM proteins. The engineered *S.b.* strains exhibited concentration-dependent binding to the corresponding ECM proteins in vitro. Crucially, the genetic modifications did not impair *S.b.*'s probiotic characteristics. In the acute DSS model, *S.b.* targeted to fibronectin (*S.b.* FN) showed a significantly prolonged gut residence time (at least 72h), with a 100-fold increase in colon concentrations compared to controls. This resulted in markedly improved pharmacodynamic parameters, including increased colon length, reduced pro-inflammatory cytokine (TNFα) expression, increased anti-inflammatory cytokine (IL-10) expression, and lower histological inflammation scores. In the chronic DSS model, *S.b.* targeted to collagen IV (*S.b.* CIV) demonstrated the highest fecal and colon tissue concentrations, and exhibited significant therapeutic efficacy, as evidenced by improved body weight, colon length, cytokine expression profile, and histological scores. Immunofluorescence confirmed colocalization of *S.b.* FN with fibronectin deposits in the inflamed colon.

The results demonstrate that targeting *S.b.* to abundant ECM proteins in the inflamed colon significantly enhances its gut residence time and therapeutic efficacy in both acute and chronic colitis models. The tunable nature of this platform allows for easy exploration of different ECM targets, potentially leading to patient-specific therapies. The prolonged gut residence time achieved through ECM targeting allows for reduced dosing frequency, moving from daily to every three days, improving patient compliance. The study's findings support the potential of this targeted probiotic approach for treating IBDs. However, further research is needed to elucidate the exact mechanisms of *S.b.*'s anti-inflammatory effects and assess its efficacy in other preclinical models and eventually in human clinical trials. The dynamic nature of ECM remodeling highlights the need for adapting the targeting strategy to the specific phase of inflammation.

This study presents a novel platform for targeted delivery of the probiotic *S.b.* to the ECM in the inflamed gut. ECM targeting significantly increases *S.b.* residence time, leading to enhanced therapeutic effects in preclinical colitis models. This approach has the potential to improve treatment of IBDs by enhancing efficacy and patient compliance. Future research should focus on optimizing the targeting strategy, exploring other ECM targets, and conducting human clinical trials.

The study utilized DSS-induced colitis models in mice, which may not perfectly replicate the complexity of human IBD. The mechanisms by which *S.b.* exerts its anti-inflammatory effects remain incompletely understood. Further research is needed to validate these findings in other relevant preclinical models and human clinical trials. The current study focused on a few selected ECM proteins and additional investigation into the expression and dynamics of other relevant ECM components in IBD is warranted.