Progression and regression of renal fibrosis is observed in patients with diabetic nephropathy (DN). The underlying pathways, especially those that promote regression of fibrosis, remain poorly understood in part due to the fact that most rodent DN models only mirror the early features of human DN. Another obstacle for optimizing treatment strategies is that albuminuria, the current gold standard biomarker of renal damage in DN, often lacks sensitivity and specificity for identification of those patients with diabetes who are at risk of a rapid decline in renal function. A novel DN model, in which diabetes was induced with streptozotocin in Cyp1a1mRen2 rats and hypertension was generated by inducing renin transgene expression with dietary indole-3-carbinol (I-3-C), mimicked many of the key biochemical, pathological and transcriptomic changes observed in the kidney of patients with DN. Recently, the model was extended to include a ‘reversal phase’ in which glycaemia was tightly controlled and blood pressure normalized for eight weeks after an ‘injury phase’ of 28 weeks. The present study aims to employ this novel rodent model to examine pathways activated in the kidney during and following reversal of hyperglycaemia and hypertension and to identify new biomarkers that might complement albuminuria in assessing risk of renal deterioration in patients with diabetes. Methods Tissue and urinary specimen from the Cyp1a1mRen 2 model of DN were analysed by realtime-PCR, Western-Blot, ELISA and staining techniques including immunohistochemistry, immunofluorescence and zymography. To establish in-situ zymography a model of ureteric obstruction was used. Urinary peptidomic analysis as well as measurement of urinary exosomes and microparticles was performed in the model and in patients with DN utilizing liquid chromatography/tandem mass-spectrometry, nanoparticle tracking analysis (NTA) or flow cytometry. Results Tight control of blood glucose and blood pressure during an 8 week ‘reversal phase’ did not significantly reverse the degree of renal fibrosis accrued during a 28wk ‘injury phase’. However, it did result in a reduction in expression of genes encoding myofibroblast markers and extracellular matrix (ECM) proteins. Genes that were up-regulated during both injury and reversal phases were implicated in adaptive immunity, phagocytosis, lysosomal processing and degradative metalloproteinases (MMPs). Paradoxically MMP activity was massively reduced during both injury and reversal phases. This may be due to an elevated level of tissue inhibitor of metalloproteinase-1 (TIMP-1) protein in both phases. After separating TIMP1 from MMP in renal tissue homogenates from animals of both the injury and reversal phases using gel electrophoresis, MMP activity was restored above that of controls. For biomarker discovery peptidomic analysis was performed on urine from rats at baseline and during the injury and reversal phases of the Cyp1a1mRen2 model of DN and from patients with moderately advanced DN and from normal controls. The use of two different search and analyse tools (Maxquant, Progenesis QI) resulted in the discovery of significantly altered peptides in the urine in rodent and human DN. Further studies focused on peptides derived from those proteins for which the corresponding gene was similarly regulated in the DN model and in human DN. Urinary epidermal growth factor (uEGF) matched these criteria as the reduction of excretion during the injury phase in the DN model was paralleled by reduced EGF protein expression in renal tissue. Key biomarker candidates identified in the first two chapters were measured in urinary specimens of patients from the Edinburgh Type 2 Diabetes study (ET2DS) to test translational utility. MMP7 and other candidates, such as osteopontin or vascular endothelial growth factor (VEGF) were not of value in predicting renal outcomes. Reduced uEGF was significantly associated with increased mortality rate. In a subgroup of 642 study participants who were normoalbuminuric and had a preserved renal function at baseline, a lower uEGF to creatinine ratio was a risk factor for either developing an estimated glomerular filtration rate less than 60 ml/min per 1.73m2, rapid (over 5% per annum) decline in renal function or the combination of both. The latter remained significant after correction for other covariates. Addition of uEGF resulted in a marginal improvement in a model derived from traditional risk factors for predicting rapid decline and the composite end-point. Urinary microparticle (20nm-1000nm) analysis was established in the rodent DN model and translated to patients with DN. Total urinary exosomes (20nm-100nm) or exosomes derived from specific renal cell types including podocytes and tubular cells, increased during the injury phase in the Cyp1a1mRen2 model followed by a decrease after reversal phase. In a pilot study comprising participants with advanced chronic kidney disease, the urinary exosome concentration correlated with renal function. In the ET2DS an increased exosome concentration at baseline indicated a higher risk for renal deterioration during four years follow-up even after correction for baseline eGFR. Urinary microvesicles (100nm-1000nm) concentration increased during the injury phase in the DN model though correlation with renal function in humans was only significant if kidney-specific marker (podocalyxin) positive microvesicles were measured. Conclusion Normalisation of hyperglycaemia and hypertension in the DN model allows the study of genetic and protein regulation during the injury and reversal phases. ECM-production but not ECM-degradation genes are down-regulated during the reversal phase. The lack of reduction in ECM during the reversal phase might be caused by persistently reduced MMP activity due to the presence of TIMP-1. Targeting TIMP might be a treatment strategy to promote reduction of renal fibrosis. For the first time, the analysis of urinary peptidomics was integrated with previous transcriptomic findings in the Cyp1a1mRen2 model and patients with DN for biomarker discovery. The approach was validated using different analysis tools and successfully identified candidate markers which were increased or reduced in DN. Candidates included uEGF, which identified patients with DN who were at risk of a rapid decline of renal function. Though the marker requires further confirmation in other cohorts, it might be especially useful for patients with type 2 diabetes, in whom renal decline is often uncoupled from the development of albuminuria. Finally, the DN model helped to develop the methodology of microparticle analysis. For the first time a potential prognostic value of urinary exosome analysis in patients with diabetes has been demonstrated. Future work will include further optimisation of the methodologies, including labelling of microparticles with multiple antibodies and increasing study participant numbers.
Identifer | oai:union.ndltd.org:bl.uk/oai:ethos.bl.uk:726602 |
Date | January 2017 |
Creators | Betz, Boris Bernhard |
Contributors | Conway, Bryan ; Hughes, Jeremy |
Publisher | University of Edinburgh |
Source Sets | Ethos UK |
Detected Language | English |
Type | Electronic Thesis or Dissertation |
Source | http://hdl.handle.net/1842/25412 |
Page generated in 0.0021 seconds