Life-saving renal replacement therapy by peritoneal dialysis (PD) is limited in use and duration by progressive impairment of peritoneal membrane integrity and homeostasis. Preservation of peritoneal membrane integrity during chronic PD remains an urgent but long unmet medical need. PD therapy failure results from peritoneal fibrosis and angiogenesis caused by hypertonic PD fluid (PDF)–induced mesothelial cytotoxicity. However, the pathophysiological mechanisms involved are incompletely understood, limiting identification of therapeutic targets. We report that addition of lithium chloride (LiCl) to PDF is a translatable intervention to counteract PDF-induced mesothelial cell death, peritoneal membrane fibrosis, and angiogenesis. LiCl improved mesothelial cell survival in a dose-dependent manner. Combined transcriptomic and proteomic characterization of icodextrin-based PDF-induced mesothelial cell injury identified αB-crystallin as the mesothelial cell protein most consistently counter-regulated by LiCl. In vitro and in vivo overexpression of αB-crystallin triggered a fibrotic phenotype and PDF-like up-regulation of vascular endothelial growth factor (VEGF), CD31-positive cells, and TGF-β–independent activation of TGF-β–regulated targets. In contrast, αB-crystallin knockdown decreased VEGF expression and early mesothelial-to-mesenchymal transition. LiCl reduced VEGF release and counteracted fibrosis- and angiogenesis-associated processes. αB-crystallin in patient-derived mesothelial cells was specifically up-regulated in response to PDF and increased in peritoneal mesothelial cells from biopsies from pediatric patients undergoing PD, correlating with markers of angiogenesis and fibrosis. LiCl-supplemented PDF promoted morphological preservation of mesothelial cells and the submesothelial zone in a mouse model of chronic PD. Thus, repurposing LiCl as a cytoprotective PDF additive may offer a translatable therapeutic strategy to combat peritoneal membrane deterioration during PD therapy.

No abstract

Background and Aims Renal replacement therapy by peritoneal dialysis (PD) is limited in use and duration by progressive impairment of peritoneal membrane integrity and homeostasis. Preservation of peritoneal membrane integrity during chronic PD remains an urgent but long-unmet medical need. PD therapy failure results from peritoneal fibrosis and angiogenesis caused by hypertonic PD fluid (PDF)-induced mesothelial cytotoxicity. The incompletely defined pathophysiological mechanisms involved confound informed selection of therapeutic targets. Addition of cytoprotective agents to PDF have been shown to counteract pathophysiological mechanisms induced by current PDF. Lithium is a well described inhibitor of glycogen synthase kinase 3β and has recently been shown to also have nephroprotective effects in low doses. Here, we aim to characterize icodextrin-based, PDF-induced cellular injury with a combined omics approach and to investigate the effects of LiCl on the PD-induced observed molecular perturbations. Method To investigate mechanisms of acute cellular damage by PDF we chose an in vitro model of primary omental-derived peritoneal mesothelial cells with direct exposure to icodextrin-based PDF, followed by short-term or extended recovery for detection of short-term and long-term changes in transcriptome, proteome, and cell injury. 0, 2.5 or 10 mM LiCl were added to the PDF. In-vitro findings were validated in peritoneal biopsies (n=41) from pediatric PD and CDK5 patients or healthy controls and peritoneal effluents from adult and pediatric PD patients (n=27) or ascites samples (n=4) as control. For in-vivo experiments, healthy and uremic mice (C57/Bl6, female) were chronically exposed to PD-fluid without or with the addition of 5 mM LiCl via an implanted catheter. In-vivo overexpression of CRYAB was induced by i.p. injection of an adenoviral vector. All animal experiments and use of patient samples were approved by the local ethics committees and performed according to animal protection laws or the Declaration of Helsinki, respectively. Results LiCl significantly improved mesothelial cell survival in a dose-dependent manner. Combined transcriptomic and proteomic characterization of icodextrin-based PDF-induced mesothelial cell injury identified αB-crystallin as the mesothelial cell protein most significantly and consistently counter-regulated by LiCl. In-vitro and in-vivo overexpression of αB-crystallin triggered a fibrotic phenotype and PDF-like upregulation of vascular endothelial growth factor (VEGF), CD31-positive cells, and TGFβ-independent activation of TGFβ-regulated targets. In contrast, αB-crystallin knock-down decreased VEGF expression and early mesothelial-to-mesenchymal transition (MMT). LiCl reduced VEGF release and counteracted fibrosis- and angiogenesis-associated processes. αB-crystallin in patient-derived mesothelial cells was specifically upregulated in response to PDF and increased in peritoneal mesothelial cells from pediatric PD patient biopsies, correlating with markers of angiogenesis and fibrosis. Conclusion The cytoprotective effects of LiCl-supplemented PDF may be explained by counter-regulation of PD-induced angiogenesis via the novel target αB-crystallin. Reduction of mesothelial cell damage, peritoneal fibrosis and VEGF suggests therapeutic potential of this intervention. Repurposing LiCl as a cytoprotective PDF additive may offer a translatable therapeutic strategy to combat peritoneal membrane deterioration during PD therapy. Further study of LiCl-supplemented PDF is merited as a realistic approach to improving treatment longevity and patient outcomes during PD treatment.

BACKGROUND AND AIMS Infectious complications and peritoneal inflammation occurring in a significant proportion of PD patients lead to progressive peritoneal membrane injury and account for most peritoneal dialysis (PD) technique failures. Reduced peritoneal immune-competence, caused by the continuous exposure to PD fluids, has been described as a therapy-related pathomechanism. We therefore hypothesized a relationship between dialysate interleukin 6 (IL-6) concentrations and peritoneal host defense. We established an ex vivo stimulation assay to test host defense mechanisms in only 9 mL of PD effluent. The aim of this study was to analyse basal inflammation and immune-competence in the PD population at routine conditions to evaluate the assay as surrogate parameter of immune competence and linking it to PD vintage and clinical outcome parameters. METHOD We prospectively analysed 195 serial routine peritoneal equilibration tests (PETs) of 123 stable PD patients treated exclusively with low-GDP, multi-chamber PD fluids during the glucose dwells and compared the data to routine follow-up clinical data. The cohort represents 76% of all eligible PD patients treated between April 2013 and September 2020 at the local Department of Nephrology. The study was approved by the local ethics committee and was conducted in accordance with the Declaration of Helsinki. All participants underwent a standardized 4-h PET with 3.86% glucose PDF as routine follow-up to test peritoneal membrane transport function, with an additional ex vivo cell stimulation protocol. PD effluent samples were obtained at the 1- and 4-h PET timepoints and immediately processed. Effluent samples were collected directly from the drainage bags into standard 9 mL additive-free sample tubes. For ex vivo stimulation, 100 ng/mL toll-like receptor (TLR) 4 agonist LPS and TLR2 agonist Pam3Cys were added to the effluent in the 9 mL collection tubes in duplicates and incubated at 37°C for 24 h. Unstimulated samples kept in parallel were used as controls. IL-6 concentrations were measured with ELISA in the supernatants. RESULTS Patients were stratified into two cohorts (incident and prevalent) at 120 days of continuous PD therapy. The cohorts were statistically indistinguishable except for stratification-specific variables. During a total follow-up period of 2499 patient-months, 55 patients underwent repeated PETs. Ex vivo stimulation of peritoneal cells significantly increased the IL-6 release 20-fold compared to unstimulated controls and resulted in a dwell-time dependent increase, with a significant lower cytokine released at the 1 h PET timepoint. To assess local inflammation, IL-6 concentrations in PD effluent of the 4 h PET were analysed. PD effluent IL-6 concentrations from the entire study cohort or from incident and prevalent patient subcohorts correlated neither with patient characteristics, RRF, disease background (including prior peritonitis) nor with time on PD. Effluent IL-6 concentrations correlated significantly with markers of systemic inflammation, serum CRP and albumin, with PSTR and with peritoneal protein loss. Interestingly, effluent IL-6 concentrations also had predictive value for risk of a subsequent peritonitis episode during follow-up. Effluent IL-6 levels were, in contrast, negatively correlated with ex vivo–stimulated IL-6 release from PD effluent cells, most remarkably in PD patients with history of prior peritonitis. CONCLUSION This longitudinal study provides the first direct evidence in PD patients of a correlation between peritoneal inflammation and impaired peritoneal immune cell function, likely driving infectious complications. Our study results suggest a mechanistic link between peritoneal inflammation and host defense that may foster innovative therapeutic approaches to improve clinical outcomes of chronic PD.

Used hemodialysis membranes (HD-M) are a valuable reservoir of biological information. Proteins bind to HD-M, but whether this process depends on the type of membrane or patient factors or selectively affects specific protein classes has not been adequately elucidated. State-of-the-art proteomics techniques are capable of identifying and quantifying this therapy-specific subproteome to enable the analysis of disease- or membrane-induced pathophysiologies. We demonstrate the feasibility of the deep proteomic characterization of the extracorporeal proteome adsorbed to HD-M. A shotgun proteomics approach using nano-flow liquid chromatography coupled to mass-spectrometry identified 1648 unique proteins eluted by a chaotropic buffer from the HD-M of eight patients. In total, 995 proteins were present in all eluates; a more stringent approach showed that a core proteome of 310 proteins could be identified independently in all samples. Stability of the dialyzer proteome was demonstrated by a >90% re-identification rate on longitudinal samples of a single patient. The core proteome showed an overrepresentation of pathways of hemostasis and the immune system, and showed differences in membrane materials (polysulfone vs. helixone). This study demonstrates that optimized conditions combined with high-performance proteomics enable the in-depth exploration of the subproteome bound to HD-M, yielding a stable core proteome that can be exploited to study patient-specific factors and improve hemodialysis therapy.