Background: Tophaceous deposits of uric acid (UA) are an occasional finding in diagnostic kidney biopsies. Whether these UA crystal granulomas are a cause or a bystander of chronic kidney disease (CKD) is subject of debate. We hypothesized that renal UA crystal granulomas would associate with more severe tubular atrophy and interstitial fibrosis compared to patients with hyperuricemia or gout, and speculated on their functional contribution to CKD progression. Moreover, UA crystal granulomas are mainly comprised of infiltrating immune cells including pro-inflammatory M1-macrophages. We thought to target a phenotype switch from M1- to anti-inflammatory M2-macrophages by activating adenosine receptor signaling with adenosine in a mouse model of CKD with UA crystal granuloma formation.
Methods: We screened 81,200 diagnostic kidney biopsies for the presence of UA crystal granuloma to first determine their prevalence and performed a case-control study to compare biopsies with and without granulomas for further morphological abnormalities. Six-week old Alb-creERT2/Glut9lox/lox (ki/ki) or Alb creERT2/Glut9lox/lox without active cre (+/+, control) mice were injected with tamoxifen. Both groups of mice were either fed a high-fat diet with inosine or normal chow diet with inosine for 32 days. We assessed UA crystal deposition, the extent of kidney damage, the glomerular filtration rate (GFR) on day 0, 14 and before sacrifice on day 32. Adenosine treatment was started after renal fibrosis had established on day 14. Tissue and serum analysis included MALDI-FTICR MS imaging, RT-PCR, (immune)-histology, fluorescence microscopy, flow cytometry, and colorimetric assays.
Results: 84 out of 81,200 diagnostic kidney biopsies showed UA crystal granulomas, which revealed significantly more global glomerulosclerosis, moderate to severe interstitial fibrosis, tubular atrophy, arteriosclerosis, and podocyte foot process effacement compared to control biopsies. The ki/ki mice on chow diet with inosine developed hyperuricemia without renal impairment, whereas the ki/ki mice on high-fat diet with inosine developed hyperuricemia and CKD. Control +/+ mice on both diets did neither develop hyperuricemia nor kidney damage. Indeed, urate nephropathy caused a significant decline in GFR compared to HU or +/+ control mice. MALDI-FTICR MS imaging confirmed UA deposits that were associated with morphological signs of tubular atrophy, casts and interstitial fibrosis as well as an infiltration of macrophages. Histological analysis revealed that the formation of UA granulomas occurs after renal fibrosis had established. Interestingly, adenosine therapy significantly reduced the number of renal UA crystal granulomas due to less M1- but more M2-macrophages, a process that attenuated CKD progression (GFR: 111 vs. 61µl/min/100g KG; p<0.05).
Conclusions: Renal UA crystal granulomas are found at a frequency of 1 out of 1,000 diagnostic kidney biopsies and are associated with more kidney fibrosis and tubular atrophy. Our in vivo attempts to test for causation revealed that UA crystal granulomas form not before interstitial fibrosis had established. M1-macropages are essential for UA crystal granuloma formation, and interfering with a switch from M1- into M2- macrophage polarization prevents progressive renal fibrosis. Taken together, renal UA crystal granulomas develop secondary to renal fibrosis but contribute to accelerated kidney atrophy and dysfunction.