This is the first direct demonstration, to our knowledge, that renal cells possess EPO-Rs through which EPO stimulates mitogenesis. This suggests currently unrecognized cytokine functions for EPO in the kidney, which may prove beneficial in the repair of an injured kidney while being potentially detrimental in renal malignancies.

An interferon-'y, tumor necrosis factor, and interleukin-1-inducible, high-output pathway synthesizing nitric oxide (NO) from L-arginine was recently identified in rodents. High-dose interleukin-2 (IL-2) therapy is known to induce the same cytokines in patients with advanced cancer. Therefore, we examined renal cell carcinoma (RCC; n = 5) and malignant melanoma (MM; n = 7) patients for evidence of cytokine-inducible NO synthesis. Activity of this pathway was evaluated by measuring serum and urine nitrate (the stable degradation product of NO) during IL-2 therapy. IL-2 administration caused a striking increase in NO generation as reflected by serum nitrate levels (10-and 8-fold increase IP < 0.001, P < 0.0031 for RCC and MM patients, respectively) and 24-h urinary nitrate excretion (6.5-and 9-fold increase [both P < 0.0011 for RCC and MM patients, respectively). IL-2-induced renal dysfunction made only a minor contribution to increased serum nitrate levels. Metabolic tracer studies using L-[guanidino-5N2Jarginine demonstrated that the increased nitrate production was derived from a terminal guanidino nitrogen atom of L-arginine. Our results showing increased endogenous nitrate synthesis in patients receiving IL-2 demonstrate for the first time that a cytokine-inducible, highoutput L-arginine/NO pathway exists in humans. (J. Clin. Invest. 1992. 89:867-877.) Key words: acute renal failure * malignant melanoma * nitrate * nitrite * renal cell carcinoma

These data are the first to demonstrate, to our knowledge, that human RCCs express EPO-R message and protein and that receptor activation stimulates their proliferation in vitro. If these mitogenic effects of EPO are also operative in patients with RCC, endogenous EPO or its administration for the treatment of anemia could potentially hasten proliferation of renocellular malignancies.

It has been proposed that adenosine, derived from ATP and released into the renal interstitium, mediates a reduction in renal function in ischemic acute renal failure. Because no direct measurements of interstitial adenosine are available, we evaluated an in vivo microdialysis technique to assess the levels of adenosine and its metabolites in the cortex of the normal rat kidney (n = 6). Microdialysis probe implantation did not alter cortical renal blood flow, glomerular filtration rate, or fractional sodium excretion. The interstitial concentration of adenosine was 199 +/- 53 nM, and relative concentrations of inosine, hypoxanthine, xanthine, and uric acid were 99 +/- 47, 182 +/- 29, and 183 +/- 70 nM and 1.8 +/- 0.4 microM, respectively. Infusion of ATP-MgCl2 (n = 5) resulted in a significant increase in the dialysate levels of adenosine (67 +/- 11 to 378 +/- 97 nM), inosine (230 +/- 102 to 803 +/- 219 nM), and uric acid (3.5 +/- 1.3 to 6.9 +/- 1.7 microM). In conclusion, this study demonstrates that the microdialysis technique is suited to monitor metabolically important substances in the renal interstitium.

The purpose of this study was to examine proximal and distal tubular function in rats with nonoliguric, myohemoglobinuric acute renal failure (ARF). ARF was induced with glycerol (50%, 10 ml/kg of body wt, i.m.), and renal function was studied 24 hours after glycerol or saline (controls) injection. Glycerol injection caused a 50 to 90% depression in GFR and a significant rise in blood urea nitrogen concentration. Animals with ARF exhibited glycosuria with normal blood sugar levels and a striking depression in tubular glucose reabsorption per milliliter of GFR. The capacity to reabsorb (mEq/liter GFR) was intact at normal blood bicarbonate levels, but was markedly depressed when blood bicarbonate was raised. The tubular maximum for para-aminohippurate (PAH) secretion and the renal extraction fraction of PAH were strikingly depressed in rats with ARF. Distal acidification as assessed by the urine-to-blood gradient of PCO2 (UB PCO2) was normal both during maximal alkalinization of the urine with bicarbonate (urine pH, approximately 7.8) or during neural phosphate infusion (urine pH, approximately 7.0). Net acid excretion per milliliter GFR and minimal urine pH (less than 5.5) following 3 days of ammonium chloride ingestion was similar in control and ARF animals. Potassium excretion was intact in maximal urinary osmolality were significantly altered in animals with ARF. Cortical and outer medullary Na-K-ATPase specific activities were significantly depressed in ARF rats. This occurred as a consequence of enzyme loss and not secondary to alterations in enzyme kinetics of absolute tubular sodium reabsorption. Light and electron microscopy showed diffuse proximal tubular damage, whereas glomeruli and distal tubules were intact. These data demonstrate that glycerol injection produces a diffuse proximal tubular transport defect associated with histologic and enzymatic alterations.