Intestinal Barrier Dysfunction in Chronic Kidney Disease

Abstract

Chronic kidney disease (CKD) is increasingly understood not as an organ-confined pathology but as a manifestation of a bidirectional gut-kidney axis. Intestinal dysbiosis shifts bacterial metabolism toward proteolytic fermentation, generating protein-bound uremic toxins such as indoxyl sulfate that erode tight-junction integrity. The resulting "leaky gut" permits translocation of lipopolysaccharide and other microbial products into the systemic circulation, sustaining low-grade inflammation that accelerates renal and cardiovascular decline. Given that this cycle is governed by communication between microbial signals and the host epithelium, this thesis centers on the aryl hydrocarbon receptor (AhR), a ligand-activated transcription factor coupling microbial and dietary cues to epithelial homeostasis, as a candidate mechanistic node within the axis. It further examines whether the human milk oligosaccharide 2'-fucosyllactose (2'-FL), previously shown to support epithelial integrity, can preserve barrier function through AhR-dependent or AhR-independent pathways. Three experimental systems were employed, structured to progress from reductionist to increasingly clinically representative models. At the cellular level, CRISPR-Cas9-generated AhR-knockout Caco-2 monolayers exhibited lower transepithelial electrical resistance over a 21-day differentiation protocol than wild-type cells, implicating AhR in epithelial maturation. Within this same model, 2'-FL conferred barrier protection irrespective of AhR status, indicating a parallel pathway independent of canonical receptor activation. Extending this to a whole-organism setting, a 5th/6th-nephrectomy rat model of CKD showed that 2'-FL modulated immune parameters, although circulating uremic metabolites known to engage AhR did not produce the receptor hyperactivation anticipated from the in vitro data. The final, ex vivo model approximated clinical physiology by exposing differentiated Caco-2 cultures to fecal water from CKD patients and matched healthy controls. Here the contrast was most pronounced: only control-derived suspensions, enriched in short-chain fatty acids, bile acid salts, and tryptophan catabolites, induced AhR and its downstream target CYP1A1, whereas CKD-derived fecal water failed to elicit this response and instead impaired barrier integrity, coinciding with a distinct metabolomic signature marked by elevated hippuric acid and reduced p-cresyl glucuronide. Considered collectively, these findings converge on a consistent pattern: uremic toxin accumulation does not straightforwardly hyperactivate AhR, as a purely ligand-driven model would predict. Rather, receptor output in CKD appears blunted or qualitatively altered, consistent with partial agonism, competitive antagonism, or ligand competition among circulating solutes. This altered signaling helps explain why barrier compromise was neither uniform nor fully attributable to AhR status; reductions in transepithelial resistance were not accompanied by proportional increases in macromolecular flux, suggesting a barrier that becomes selectively rather than globally permeable. Against this backdrop, the consistent protective effect of 2'-FL across all three models, despite the disrupted AhR signaling observed within them, points to convergent mechanisms only partially dependent on the receptor, and positions HMO-based intervention as a plausible adjunct to conventional CKD management across multiple disease stages. Taken together, these results support conceptualizing the gut-kidney axis as a genuinely multi-factorial system spanning epithelial, microbial, and immune compartments, one whose translation into clinical practice will require longitudinal and stratified models capable of capturing this complexity.