Renal interstitial cGMP mediates natriuresis by direct tubule mechanism

The objective of this study was to test the hypothesis that renal interstitial (RI) cGMP is natriuretic in vivo. In conscious rats (n=8), urinary sodium excretion (UNaV) was significantly greater on days 3 and 4 of RI infusion of cGMP (1.17+/-0.14 and 1.61+/-0.11 mmol/24 h, respectively) than during vehicle infusion (0.56+/-0.15 and 0.70+/-0.17 mmol/24 h, respectively) (P<0.01). Similarly, UNaV was greater on days 3 and 4 of RI infusion of 8-bromo-cGMP (2.15+/-0.42 and 2.16+/-0.1 mmol/24 h, respectively). Protein kinase G inhibitor Rp-8-pCPT-cGMPS reduced cGMP-induced and 8-bromo-cGMP-induced UNaV to control levels. Acute RI infusion of L-arginine (L-Arg, 40 mg . kg(-1) . min(-1)), but not D-arginine, caused an increase in UNaV from 1.65+/-0.11 to 4.07+/-0.1 mol/30 min (P<0.01). This increase was blocked by RI infusion of N-G-nitro-L-arginine methyl ester (100 ng . kg(-1) . min(-1)) by the phosphodiesterase (PDE II) activator 5,6DMcBIMP (0.01 mol/muL), by PDE II (0.03 U . kg(-1) . min(-1)) itself, or by the soluble guanylyl cyclase inhibitor 1-H-[1,2,4]oxadiazolo-[4,2-alpha ]quinoxalin-1-one (ODQ, 0.12 mg . kg(-1) . min(-1)). The PDE II activator also blocked L-Arg-stimulated cGMP levels. The NO donor S-nitroso-N-acetylpenicillamine (SNAP, 0.12 mu mol . L-1 . kg(-1) . min(-1)) increased UNaV from 1.65+/-0.11 to 2.93+/-0.08 mu mol/30 min (P<0.01), and this response was blocked completely by ODQ. Renal arterial but not RI administration of the heat-stable enterotoxin of Escherichia coli induced natriuresis. RA infusion of cGMP (3 g/min) increased UNaV, renal blood flow (RBF), and glomerular filtration rate (GFR). Renal cortical interstitial cGMP infusion increased UNaV with no effect on total RBF, renal cortical blood flow, or GFR. Similarly, the natriuretic actions of renal interstitial L-Arg or SNAP were not accompanied by any change in RBF or GFR. Medullary cGMP infusion had no effect on UNaV, total RBF, or medullary blood flow. Texas red-labeled cGMP infused via the RI space was distributed exclusively to cortical renal tubular cells. The results demonstrate that RI cGMP inhibits renal tubular sodium absorption via protein kinase G independently of hemodynamic changes. These observations indicate that the cortical interstitial compartment provides a potentially important domain for cell-to-cell signaling within the kidney.