Abstract BACKGROUND AND AIMS Tight junctions (TJ) and transcellular ion channels and transporters define solute transport characteristics across cellular barriers, which is of particular interest in patients on peritoneal dialysis (PD). Little is known about their cell specific expression, and regulation in PD. We studied their expression in human endothelial and mesothelial cell lines and in paediatric peritoneal tissues. METHOD In vitro, polarized primary human peritoneal mesothelial cells (HPMC), immortalized mesothelial cells (MeT-5A), human umbilical vein endothelial cells (HUVEC) and human capillary microvascular endothelial cells (HCMEC) underwent RNA sequencing, and gene enrichment analysis (GSEA, ClueGO/Cluepedia) for functional annotation. Key findings were reconfirmed by western blotting and confocal laser scanning immunofluorescence microscopy. Transepithelial electrical resistance (TER) and permeability fluxes of fluorescent 4-, 10- and 70-kDa dextran were measured in Transwells. Ex vivo, whole transcriptome and proteome data from microdissected omental arterioles were used for targeted pathway analysis in non-CKD children, chronic kidney disease (CKD5) and on PD with low and high glucose degradation product (GDP) content (n = 6/group). Mesothelial and endothelial peritoneal solute transporting proteins were quantified in parietal peritoneum of independent paediatric non-CKD, CKD5 and PD cohorts by digital immunohistochemistry. RESULTS A total of 9853 of 12 760 transcripts were common between all four cell lines. A total of 631 transcripts were MeT-5A, 366 HPMC, 99 HUVEC and 87 HCMEC specific. Next to the tissue origin and transformation status, the transcripts reflected major differences in extracellular matrix, glycocalyx and adhesion organization between HCMEC and HUVEC, and extracellular matrix, migration, growth factor and immune response between HPMC and MeT-5A cells. While total counts of cell junction, transmembrane and endocytosis related transcripts were similar among cell lines, the specific TJ, transmembrane and endocytosis related transcript patterns, differed substantially between endothelial and mesothelial cells. Of the functionally well described sealing TJs, claudin (CLDN)1 was expressed in mesothelial cells, and CLDN5 in endothelial cells. Findings were reconfirmed by western blotting and immunofluorescence staining. Functionally, transepithelial resistance (TER) was 50% lower for HCMEC compared to HPMC, MeT-5A and HUVEC; 4-, 10- and 70-kDa dextran permeability was increased in HCMEC. Ex vivo, human arteriolar pathway analysis demonstrated upregulation of paracellular transport-related pathways in CKD5 versus non-CKD patients on protein level. Compared to CKD5, low GDP PD upregulated and high GDP PD downregulated these pathways on the transcriptome and protein levels. Transcellular transporter pathway regulation was variable. In the parietal peritoneum the endothelial surface area for transport was age dependently 1.5–2-fold higher than the mesothelial surface area and the ratio markedly increased with low GDP PD. Validation in parietal peritoneal tissues, reconfirmed arteriolar sealing TJ regulation. Arteriolar CLDN2, a paracellular pore forming cation and water transporter, correlated with D/PCreatinine (r = 0.58) and D/D0Glucose (r = –0.62), mesothelial pore forming cation transporter CLDN15 with D/PCreatinine (r = 0.57) and D/D0Glucose (–0.66). Transcellular transporters for sodium, glucose and phosphate were hardly affected by PD; phosphate transporter PIT1 abundance correlated with serum phosphate (r = –0.48). CONCLUSION We provide the first comprehensive analysis of the peritoneal paracellular and transcellular determinants of solute transporters and its regulation by CKD and PD. Mesothelial and endothelial cell barrier sealing and transporter abundance differed substantially, and associated with PD membrane function, with functional data suggesting a key role of both the mesothelial and endothelial cell barrier.
Read full abstract