Mechanism by which superoxide potentiates tubuloglomerular feedback

The macula densa detects changes in NaCl concentration in tubular fluid and transmits a feedback signal, known as tubuloglomerular feedback (TGF), which helps to control glomerular afferent arteriole resistance. We and other investigators have reported that synthesis of NO in the macula densa inhibits TGF. NO can be scavenged by superoxide (O(-)(2)) to form peroxynitrite, effectively reducing the bioavailability of NO; there is growing evidence that O(-)(2) regulates vascular tone in the kidney. We hypothesized that O(-)(2) produced in the macula densa enhances TGF and this effect acts only in an autocrine manner within the cells of the macula densa. Afferent arterioles and attached macula densas from Sprague-Dawley rats were simultaneously microperfused in vitro and TGF response examined before and after perfusing the tubular lumen, bath, or vascular lumen with a superoxide scavenger. The macula densa was perfused with solutions containing either 5 mmol/L Na(+) and 3 mmol/L Cl(-) (low NaCl) or 80 mmol/L Na(+) and 77 mmol/L Cl(-) (high NaCl) while keeping pressure in the afferent arteriole constant at 60 mm Hg. When 10(-4) M Tempol, a stable membrane-permeant superoxide dismutase (SOD) mimetic, was added to the tubular lumen, it inhibited TGF by 56% (before Tempol: TGF, 3.2 +/- 0.3 microm; after Tempol: TGF, 1.4 +/- 0.2 microm; n=6; P<0.05, control versus Tempol). Adding Tempol to the bath inhibited TGF by 48% (before Tempol: TGF, 2.5 +/- 0.25 microm; after Tempol: TGF, 1.3 +/- 0.18 microm; n=6; P<0.05). However, adding Tempol to the vessel lumen did not change TGF response significantly (before Tempol: TGF, 2.7 +/- 0.37 microm; after Tempol: TGF, 3.2 +/- 0.25 microm; n=7; P=0.25). When 300 U/mL of the enzyme SOD, which is not membrane-permeant, was added to either the tubular lumen or bath, it had no effect on TGF response. Finally, to determine whether the effect of O(-)(2) in the macula densa is mediated by its scavenging of NO, 7-nitroindazole (7-NI) was added to the macula densa to inhibit neuronal nitric oxide synthase (nNOS). In the presence of 7-NI, Tempol had no effect (7-NI only: TGF, 3.0 +/- 0.4 microm; 7-NI plus Tempol: TGF, 2.8 +/- 0.5 microm; n=6; P=0.343). Our findings suggest that (1) reducing O(-)(2) increases the bioavailability of NO, which inhibits TGF, (2) both O(-)(2) and NO act within the macula densa, and (3) O(-)(2) appears to have no effect on its own.