Abstract
Background
Attenuating progressive fibrosis is fundamental to preventing chronic kidney disease and, ultimately, any accompanying end-stage kidney disease. Lithium treatment is associated with nephrogenic diabetes insipidus, changes to cellular morphology of the collecting duct and dilation of tubules. In the long term, lithium treatment causes irreversible chronic interstitial fibrosis. Nephrogenic diabetes insipidus is reversible when co-administration with amiloride, a sodium epithelial blocker (ENaC). In an established model of long-term lithium-induced interstitial fibrosis, delayed co-administration with amiloride reduced the progression of fibrosis. However, amiloride did not modify the cellular changes or prevent the dilation of tubules. How amiloride reduces fibrosis is unclear and was investigated here using animal models treated for a short and long term with lithium and amiloride.
Hypothesis and aims
Here we hypothesised that amiloride, when combined with lithium, modifies key cellular pathways enabling a reduction in fibrosis. Therefore, this thesis aimed to elucidate the critical signalling pathways these two drugs modulate together, thereby providing a better understanding of how lithium induces and how amiloride attenuates chronic interstitial fibrosis.
Methods
Amiloride was administered to a lithium-induced interstitial fibrosis rat model for a long-term (5 months) of a six-month lithium treatment period. Amiloride was also administered to rats for shorter periods of 14 and 28 days and co-administered with lithium from day zero. Following treatment, kidneys were excised, and RNA was extracted from the kidney cortical tissue. RNA was sequenced to identified gene expression differences, and microRNA expression obtained using a commercial microRNA probe set specific to rats. Pathways of interest were determined from genes that were differentially expressed with lithium alone, and lithium with amiloride treatment compared to control treatment (FDR<0.05, log2fold change of more or less than 0.5). Ontologies overrepresented amongst the differential genes were determined using various online tools. Genes of interest from the candidate pathways were verified using immunohistochemistry and immunofluorescence on kidney tissues on an automated staining platform. Notch1 was investigated using RNAscope in situ hybridisation on kidney tissue using a manual staining method. Stained slides were scanned, and images were captured digitally using Aperio Imagescope for brightfield imaging and Aperio Versa for fluorescent imaging and analysed using Fiji ImageJ. Furthermore, some of the genes were cross-validated using in vitro cell culture. Stained slides were scanned, and images were captured digitally using Aperio Imagescope for brightfield imaging and analysed using Fiji ImageJ. All statistics were performed using ordinary one-way ANOVA with multiple comparisons on PRISM (version 9.0.0). A p-value of less than 0.05 was considered significant.
Results
Pathways identified by RNA sequencing of kidney tissue were related to ‘promoting inflammation’ for lithium and ‘reducing inflammation’ for amiloride, while many microRNAs were increased by amiloride, including miR34a. In situ validation of inflammatory pathways, found amiloride profoundly reduced inflammatory pathways induced by lithium, including NF-B (2.2 fold decrease, p<0.01), activated Akt (3.5 fold decrease, p<0.0001), and increased anti-inflammatory pathways, including increased p53 (1.5 fold increase, p<0.05) compared to lithium. Proteins of interest were localised in the damaged and dilated tubular epithelial cells near the areas of fibrosis. Amiloride also reduced Notch1 positive cells (4 fold decrease p<0.05) in the interstitium. Notch1 positive cells were also positive for PDGFr and were likely pericytes. These pericytes were sparsely co-localised with SMA positive cells, which were likely myofibroblasts. There was also decreased CD3 positive T cells in the interstitium with lithium and amiloride compared to lithium alone.
In cell culture, reduced activated Akt (10 fold reduction, p<0.0001) and increased p53 (4 fold, p<0001) was found with lithium and amiloride compared to lithium alone, replicating the in vivo findings. Lithium treated cells had increased PDGF (4 fold, p<0.0001) compared to lithium amiloride, suggesting the ligand for recruiting PDGFr pericytes was induced with lithium
Conclusions
Using a combination of in vivo and in vitro models, we have expanded our understanding of the key regulatory pathways that are modified by lithium leading to chronic interstitial fibrosis, and how amiloride counteracts these regulatory pathways to reduce the extent of interstitial fibrosis. This thesis found that amiloride attenuates a multitude of pro-inflammatory pathways early and late following exposure suggesting a new indication for an old drug.