Effects of Saline and Mannitol on Renin and Distal Tubule Na in Rats

Transcription

1 786 Effects of Saline and Mannitol on Renin and Distal Tubule Na in Rats PAUL C. CHURCHILL, MONIQUE C. CHURCHILL, AND FRANKLIN D. MCDONALD SUMMARY The purpose of these experiments was to investigate the relationship between renin secretion (the product of renal plasma flow and renal venous minus arterial plasma renin) and distal tubular Na* concentration and load (the product of Na* concentration and tubular fluid flow rate) in pentobarbital-anesthetized rats. Induction of saline diuresis decreased renin secretion (from 274 ± 82 to 56 ± 18 ng/hr per min per kg, n = 6 rats, P < 0.03) and increased Na* load (from 494 ± 40 to 1179 ± 181 peq min, n 13 tubules, P < 0.004). Induction of mannitol diuresis also decreased renin secretion (from 226 ± 62 to 21 ± 12 ng/hr per min per kg, n - 6 rats, P < 0.03) and increased Na + load (from 479 ± 68 to 969 ± 147 peq/min, n» 13 tubules, P < 0.004). Na 4 concentration was increased by induction of saline diuresis (from 59 ± 2 to 79 ± 5 meq/liter, P < 0.004) and decreased by induction of mannitol diuresis (from 66 ± 5 to 49 ± 4 meq/liter, P < 0.004). Na* concentration, Na* load, and renin secretion rate were stable over time in a third group of rats (controls). The acute changes elicited by induction of osmotic diuresis were not consistent with a relation between Na* concentration and renin secretion, but these changes were consistent with a reciprocal relation between Na* load and renin secretion. Circ Res 45: , 1979 RENIN is secreted by the juxtaglomerular apparatus in mammals. The juxtaglomerular apparatus consists of both vascular (juxtaglomerular cells) and tubular (macula densa cells) components, and it is well established that events at either site can affect the rate of renin secretion (Davis and Freeman, 1976; Vander, 1967). Thus, it has been shown that renin secretion is related reciprocally to the renal perfusion pressure, independently of changes in the activity of the renal nerves, of changes in plasma or perfusate composition, and of changes in tubular fluid flow or composition (Blaine et al., 1970; Fray, 1976). Based on such observations, the hypothesis has been advanced that secretion is stimulated by decreased stretch of the afferent arteriolar juxtaglomerular cells, and vice versa (the baroreceptor, or the renal vascular receptor, mechanism of control of renin secretion). On the other hand, it also has been shown that renin secretion is related reciprocally to dietary Na + and /or Na + excretion on a chronic basis (Davis and Freeman, 1976; Vander, 1967; Brubacher and Vander, 1968; Reeves and Sommers, 1965), and the administration of several osmotic diuretics acutely increases Na + excretion and suppresses renin secretion in the absence of changes in renal hemodynamics that are consistent with baroreceptor control (Nash et al., 1968; Vander and Miller, 1964). Based on such observations, the From the Departments of Physiology and Internal Medicine, Wayne State University School of Medicine and Hutzel Hospital, Detroit, Michigan. This research was supported by grants from the National Science Foundation (GB and PCM ), the Kidney Foundation of Michigan, and the Skillman Foundation of Detroit. Address for reprints: Dr Paul C. Churchill, 5263 Scott Hall, Department of Physiology, Wayne State University School of Medicine, 540 East Canfield, Detroit, Michigan Received April 2, 1979; accepted for publication August 28,1979. hypothesis has been advanced that macula densa cells can mediate changes in secretion in response to changes in tubular fluid flow and/or composition (the macula densa mechanism of control). Further evidence for the existence of a tubular mechanism of control of renin secretion comes from the observations that furosemide and ethacrynic acid have little (Corsini et al., 1975) or no (Freeman et al., 1974) effect on renin secretion from nonfiltering kidneys, but greatly stimulate secretion from filtering kidneys (Corsini et al., 1975; Freeman et al., 1974; Bailie et al., 1973; Blaine and Zimmerman, 1978; Cooke et al., 1970; Vander and Carlson, 1969) even after pharmacological blockade of the baroreceptor mechanism (Freeman et al., 1974; Blaine, 1977). Despite the consensus concerning both the existence of macula densa control of renin secretion and the general nature of the controlling stimulus, there is controversy concerning the directionality of the response to a change in the stimulus. That is, it has been proposed that secretion is inhibited (Davis and Freeman, 1976; Vander, 1967; Nash et al., 1968; Vander and Miller, 1964; Corsini et al., 1975; Freeman et al., 1974; Bailie et al., 1973; Blaine and Zimmerman, 1978; Vander and Carlson, 1969; Blaine, 1977; Kirchner et al., 1978) or, conversely, that secretion is stimulated (Cooke et al., 1970; Birbari, 1972; Ploth et al., 1977; Schnermann et al., 1970, 1976; Thurau et al., 1967), by increases in the same parameter (Na + and/or Cl~ concentration, delivered load, or reabsorptive flux). This controversy is attributable, at least in part, to the impossibility of measuring both the controlling stimulus and the response simultaneously. Tubular fluid composition and flow can be measured slightly downstream from the macula densa segment in rats,

2 RENIN AND DISTAL TUBULE Na/Churchilletal. 787 but only one experiment has been reported in which this was done simultaneously with renin secretion measurements. It was found that chronic changes in dietary Na + intake induced changes consistent with a reciprocal relationship between renin secretion rate and Na + load (Churchill et al., 1978). In the experiments described below, both renin secretion rate and early distal tubular fluid flow and Na + concentration were measured in anesthetized rats, before and after the induction of diuresis with mannitol and with saline. The purpose was to determine whether, on an acute basis, renin secretion and Na + load are related reciprocally. Methods Adult Sprague-Dawley rats were obtained from Spartan Research Animals and kept in individual cages in a constant temperature room. They were provided with tap water and Purina Laboratory Chow (0.45% Na by weight) until the experimental day. On the experimental day, each rat was weighed (body weight averaged 252 ± 12 g, mean ± SE, n = 18 rats) and anesthetized (Na pentobarbital, 45 mg/ kg body weight, administered iv). Supplemental doses of pentobarbital were given intravenously as required during the surgery and experimental procedures. The rat was placed on a heated operating table (body temperature was maintained at C), and previously described methods were used to expose and catheterize a jugular vein, a femoral artery, both ureters, and the left renal vein (Churchill et al., 1973). To minimize any obstruction to renal blood flow, the tip diameter of the renal venous catheter was reduced by heating and stretching the polyethylene tubing (PE 50) before insertion. After catheterization, the left kidney was prepared for micropuncture (Churchill et al., 1978). After surgical preparation, an intravenous priming injection of inulin and />-aminohippuric acid (PAH) was given. Plasma inulin and PAH concentrations were maintained by continuous intravenous infusion of both, dissolved in 150 mm NaCl, at a rate of approximately 0.1 ml/min per kg body weight (Churchill et al., 1978). An interval of minutes was allowed for stabilization. During this time period, several cortical distal tubules were identified and marked with nigrosine (Churchill et al., 1978). Then, a 45-minute clearance period was begun. Urine was collected in preweighed test tubes, and arterial and renal venous blood samples were collected in ice-cold syringes at the clearance midpoint. Blood sample volumes were approximately 0.4 ml. During the clearance period, several distal tubules were punctured as close as possible to the beginning of the distal tubule. Oil-filled pipettes, of 6-10 jim tip diameter, were used. Oil blocks were injected, and timed fluid samples were collected. At the end of the first clearance period, during which all the rats were treated identically, three protocols were followed. In one group (n = 6), saline diuresis was induced by increasing the rate of the intravenous infusion to approximately 1 ml/min per kg. In the second group (n = 6), mannitol diuresis was induced by changing the infusate to 10% mannitol, which also contained 10 mm NaCl and 5 mm KC1, and increasing the delivery rate to 1 ml/min per kg. The inulin and PAH concentrations were reduced by a factor of 10 in both cases. In the third group (controls, n = 6), the composition and the rate of administration of the infusion remained constant. Thirty to 45 minutes later, the second 45- minute clearance period was begun. Arterial and renal venous bloods were resampled, and tubular fluid samples were re-collected from the same sites during this period. In most of the rats, femoral arterial blood pressure was monitored with a pressure transducer and polygraph. After the second clearance period, the rats were killed. Punctured tubules were filled with a nigrosine solution, and the kidneys were removed and macerated in preparation for microdissection (Churchill et al., 1978). Total length of a punctured tubule was estimated as the distance between the point where the ascending limb touched the glomerulus and the junction with another distal tubule. Micropuncture site was expressed as percent of total length. Volumes and Na + concentrations of tubular fluid samples were measured with a calibrated constant bore capillary and an Aminco Helium Glow photometer, respectively. Na + load was calculated as the product of tubular fluid flow rate and Na + concentration. Inulin and PAH concentrations were determined by colorimetric methods and Na + by flame photometry with internal Li + standardization (Churchill et al., 1978). Inulin clearance was used as the index of glomerular filtration rate (GFR). Renal plasma flow (RPF) was calculated according to the Fick principle (Pitts, 1974), using PAH and/ or inulin concentrations in urine and arterial and renal venous plasma. The method for determination of plasma renin differed slightly from that originally described (Churchill et al., 1973) but has been described and validated in a more recent publication from this laboratory (Churchill et al., 1978). Briefly, plasma samples were mixed with rat renin substrate (containing inhibitors of angiotensinases and converting enzyme) and incubated at 37 C. Samples from nondiuretic rats were incubated for 60 minutes; samples from diuretic rats, in which renin was expected to be lower, were incubated for 120 minutes. It was shown previously that angiotensin I production rate was linear for up to 120 minutes (Churchill et al., 1978). Angiotensin I was measured by radioimmunoassay, and plasma renin was expressed in units of ng/hr per ml (nanograms of angiotensin I per hour of incubation with substrate per milliliter of plasma sample). Renin secretion rate was calculated as the difference between renal venous and arterial plasma renin, multiplied by the renal plasma flow.

3 788 CIRCULATION RESEARCH VOL. 45, No. 6, DECEMBER 1979 The modifications of the originally described (Churchill et al., 1973) renin method are as follows. (1) Plasma samples were not subjected to the acidification and neutralization procedure of the original method. This modification increased the apparent plasma renin by approximately 5-fold. Plasma from an anesthetized rat was separated into two aliquots. One was kept at 4 C and acidified for 30 minutes (1 part of 1.2 N HC1 added to 20 parts of plasma) then neutralized (1 part 0.6 N NaOH to 20 parts plasma) as previously described (Churchill et al., 1973). The other aliquot was kept at 4 C for 30 minutes. Plasma renin in both was measured by incubating 20-fA samples with 200 fil of substrate for 2 hours. Plasma renin averaged 21 ± 1 (n = 20 trials) and 98 ± 5 (n = 20) in acidified and nonacidified aliquots, respectively. (2) Rat renin substrate was not semipurified before use, as previously described (Churchill et al., 1973), but consisted of a mixture of one part 24-hour nephrectomized rat plasma and four parts of sodium phosphate buffer, ph 6.25 (100 mm phosphate; 2.4 mg/ml of sodium EDTA; 0.2 mg/ml of neomycin sulfate). Before use, 10 id each of 6.6 g/100 ml of 8-hydroxyquinoline sulfate and 1.8 g/100 ml of 2,3-dimercaprol (in peanut oil containing benzyl benzoate, 3 g/100 ml) were added per ml. The concentration of substrate in this mixture was 1158 ± 25 ng of angiotensin I per ml found by exhaustive reaction of substrate with purified hog renin (Churchill et al., 1978). Carvalho et al. (1975) used a similar method for renin determination but diluted the nephrectomized rat plasma until a substrate concentration of 500 ng/ ml was obtained, which is below the apparent K m they found for the reaction (625 ng/ml). (3) Previously, plasma was incubated with substrate for 4 hours and the reaction was stopped by boiling the mixture (Churchill et al., 1973). Currently, plasma is incubated for 2 hours at most, and the reaction is stopped by cooling to 4 C. (4) As before, the angiotensin I generated during the incubation is measured by radioimmunoassay, but antibody to angiotensin I produced in our laboratory, rather than from a commercial kit, is used. The paired and the unpaired <-tests were used to evaluate statistical significance of observed changes within groups and differences between groups, respectively. Results All rats received the same intravenous infusion at the same rate during the first clearance period. As can be seen in Table 1, there were no significant differences (P > 0.05, unpaired t-test) between the control and either the saline or the mannitol group with respect to arterial blood pressure, plasma Na + and K + concentrations, RPF, GFR, urine flow, or Na + excretion during the first clearance period. As a group, the control rats appeared to be in a steady state; these rats continued to receive the same infusion at the same rate during the second clearance period, and there were no significant changes (P > 0.05, paired f-test) in any of the variables in Table 1. It can be calculated from the intravenous infusion rate that this group received from 1 to 2% of body weight in fluid during the course of the experiments. In contrast, both mannitol and saline groups received from 8 to 10% of body weight in fluids, and in both these groups, urine flow and Na + excretion were increased and plasma K + concentration was decreased, during the second clearance period. Plasma Na + was decreased significantly in the mannitol group but was unaffected in the saline group. Although renal plasma flow tended to increase in both groups during the second period, the change was significant only in the mannitol group. TABLE 1 Arterial Blood Pressure, Plasma Na* and K*, and Renal Function during Two Clearance Periods Arterial blood pressure (mm Hg) Plasma Na + (meq/liter) Plasma K + (meq/liter) Renal plasma flow (ml/min per kg) Glomerular filtration (ml/min per kg) Urine flow (jil/min per kg) Na + excretion OiEq/min per kg) 127 ±9 137 ±2 4.1 ± ± ± ±4 0.8 ± 0.3 Saline 124 ±5 138 ± ± 0.1* 15 ± ± ± 26* 26 ±6* 124 ± ±2 3.7 ± ±2 3.3 ± ±4 1.1 ±0.5 Control 122 ±3 142 ± n ±2 2.8 ± ±2 1.5 ± ±4 140 ±2 4.4 ± ± ± ±3 0.8 ± 0.2 Mannitol 116 ± ± 1* 4.1 ± 0.1* 13 ± If 3.3 ± ± 22* 10 ± 2* During the first clearance, all rats received 150 mw NaCl at approximately 0.1 ml/min per kg; the control group continued to receive 150 mm NaCl at the same rate during the second clearance, whereas diuresis was induced in the saline and manmto) groups before the second clearance period. Renal functiona] data are factored by body weight and are for the left kidney only. Results are expressed as mean ± SEM; n 6, except n 4 for blood pressure in the saline group. ' P < 0.01 and fp < 0.05, comparing first vs. second clearance period (paired l-test).

4 RENIN AND DISTAL TUBULE Na/Churchill et al. 789 TABLE 2 Tubular Fluid Na + Concentration and Load, Arterial Plasma Renin and Renin Secretion Rate during Two Clearance Periods Micropuncture site (% total length) Na + concentration (meq/liter) Na + load (peq/min) Arterial renin (ng/hr per ml) Renin secretion rate (ng/hr per min per kg) 18 ±2 59 ±2 494 ± ± ±82 Saline 79 ±5* 1179 ± 181* 36 ±6* 56 ± 18f 22 ± 1 61 ±3 447 ± ± ± 57 Control 61 ±6 494 ± ± ± ±2 66±5 479 ± ± ±62 Mannitol 49 ±4* 969 ± 147* 27 ± 5* 21 ± 12f During the first clearance, all rats received 150 mil NaCl at approximately 0.1 ml/min per kg; the control group continued to receive 150 mm NaCl at the same rate during the second clearance, whereas diuresis was induced in the saline and mannitol groups before the second clearance period. Renin secretion rate is for the left kidney only. Results are expressed as mean ± SEM. There were six rata in each group, n 6 for rerun data, and n 13, 12, and 13 for tubular fluid data for the saline, control, and mannitol groups, respectively. " P < and fp < 0.03, comparing first vs. second measurement (paired /-test). Neither blood pressure nor glomerular filtration rate was significantly affected (P > 0.05, paired t- test) by the induction of saline or mannitol diuresis. Renin and micropuncture data are summarized in Table 2. Neither the saline nor the mannitol group differed significantly (P > 0.05, unpaired t- test) from the control group with respect to any of the variables measured. On the average, distal tubular fluid was sampled from the same location, and Na + concentrations and loads were similar in the three groups, as were average arterial reruns and renin secretion rates. Also, with respect to the variables in Table 2, the control rats appeared to be in a steady state; there were no significant changes in Na + concentration or load or in arterial renin or renin secretion rate {P > 0.05, paired t- test). In both groups of diuretic rats, arterial plasma renin and renin secretion rate were significantly reduced; saline and mannitol groups were not significantly different from each other with respect to the second period values of arterial renin and renin secretion rate (P > 0.05, unpaired t-test). Na + load was significantly increased in both groups, and second period values were not significantly different from each other (P > 0.05, unpaired t-test). Distal tubule Na + concentration was significantly increased during saline diuresis and significantly decreased during mannitol diuresis. Thus, increased Na + load and decreased renin secretion rate were simultaneous events in both diuretic groups. Arterial plasma renin was correlated directly with renin secretion rate; the correlation coefficient, calculated by linear regression analysis using both first and second period values for each rat, was somewhat low (r = 0.7) but highly significant (P < 0.001). In all, 18 rats were studied before and during saline diuresis and 14 rats before and during mannitol diuresis. The same protocols were followed, but data obtained from the 12 additional "saline" rats and the eight additional "mannitol" rats were not included in the averages presented in Tables 1 and 2 because at least one of the critical measurements, usually of renin secretion rate, was missing. However, the Na + concentration and load data from all diuretic rats are presented in Figures 1 and 2. These are standard re-collection plots, showing the effects of saline and mannitol on early distal tubule Na + concentration and load, respectively. It can be seen that all rats responded as did those of Table 2, in that Na + load of nearly every tubule was increased during either type of diuresis and in that Saline A Mannitol Collection [No] meq/l FIGURE 1 The effects of saline and mannitol diureses on distal tubule Na + concentration (meq/liter). During the initial collection period, all rats were receiving 150 mm NaCl at approximately 0.1 ml/min per kg (nondiuretic). Diuresis was induced with saline or mannitol prior to re-collection from the same tubular sites.

5 790 CIRCULATION RESEARCH VOL. 45, No. 6, DECEMBER OOOr 8 2OO0 ction f Eq/mi a> a or x ^ A ' A A Saline A Manmtol Collection No Load peq/min FIGURE 2 The effects of saline and mannitol diureses on distal tubule Na* load, the product of tubular fluid Na + concentration and flow rate (peq/min). During the initial collection period, all rats were receiving 150 m\i NaCl at approximately 0.1 ml/min per kg (nondiuretic). Diuresis was induced with saline or mannitol prior to re-collection from the same tubular sites. Na + concentration almost invariably increased during saline diuresis but decreased during mannitol diuresis. Discussion In a study we reported earlier (Churchill et al., 1973), rats were anesthetized and infused with 150 RIM NaCl at approximately 0.2 ml/min per kg; arterial plasma renin and renin secretion rate averaged 11 ± 4 ng/hr per ml and 37 ± 32 ng/hr per min per kg, respectively. In a more recent study (Churchill et al., 1978), rats were anesthetized, infused with 150 mm NaCl at approximately 0.1 ml/ min per kg, and averages of arterial plasma renin and renin secretion rate were 148 ± 16 ng/hr per ml and 269 ± 48 ng/hr per min per kg, respectively. In both studies, the rats were fed the same diet before the experiments. The large differences in arterial plasma renin and in renin secretion rate between these two studies can be explained, at least in part, by differences in the experimental conditions and in the renin methods. As shown in this and in the previous study (Churchill, 1973), increasing the rate of fluid administration suppresses renin secretion, and in the previous study, the rate was 2-fold higher than in either the more recent (Churchill, 1978) or the present studies. Several modifications were made in the renin methods between the earlier, the more recent, and the present studies. Perhaps the most significant modification was the omission of the acidification-neutralization step, which increased apparent renin activity by approximately 5- fold. All rats of the present study were infused with 150 mm NaCl at approximately 0.1 ml/min per kg during the first clearance period, and averages of arterial plasma renin and renin secretion rate during this period were comparable to those we reported most recently (Churchill, 1978). It is well known that hemorrhagic hypotension can be a potent stimulus for renin secretion (Davis and Freeman, 1976; Vander, 1967). Apparently, however, secretion was not stimulated by withdrawal of blood samples in the present study; in none of the three groups was secretion during the second period higher than during the first. The fact that renin secretion rate and arterial plasma renin were directly related suggests that the measurements were made during steady state conditions, in both the control and the diuretic rats. With respect to arterial plasma renin and renin secretion rate, the saline and the mannitol groups were not significantly different from each other prior to, or during, diuresis. The induction of diuresis suppressed plasma renin and secretion rate equally in both groups. These observations extend our previous observation (Churchill, 1973) that induction of either mannitol or saline diuresis suppresses renin secretion and plasma renin in rats fed a diet low in Na +. Renin secretion and/or plasma renin can be suppressed acutely in other species also, by a variety of osmotic diuretics including saline and mannitol (Davis and Freeman, 1976; Vander, 1967; Nash et al., 1968; Vander and Miller, 1964). Induction of osmotic diuresis could suppress renin secretion by a variety of mechanisms, both extra- and intrarenal. In experiments such as ours, volume expansion could alter plasma concentration of a humoral factor and/or reflexly suppress the activity of the renal nerves, either or both of which could in turn suppress renin secretion. The experiments of Humphreys et al. (1975) argue against a role for circulating factors; volume expansion of donor anesthetized dogs increased Na + excretion and decreased renin secretion of their intact in situ kidneys but failed to alter either Na + excretion or renin secretion of isolated kidneys perfused with blood from the donors. A role for the renal nerves might be suggested on the basis that only the innervated kidney responded in their experiments. Our experiments provide no direct evidence for or against such extrarenal mechanisms. However, given the facts that both mannitol- and saline-infused rats received fluids at identical rates, but their urine flows differed by a factor of four on the average, it seems unlikely that equivalent volume expansion existed in both groups; nevertheless, arterial plasma renin and renin secretion rate were suppressed to an equivalent degree in both groups. Moreover, the experiments of Nash et al. (1968) suggest that osmotic diuresis per se, independently of volume expansion, suppresses renin secretion; unilateral intrarenal arterial infusion of saline induced unilateral increases in Na + excretion and

6 RENIN AND DISTAL TUBULE Nn/Churchill et al. 791 decreases in renin secretion. Thus, an intrarenal mechanism could mediate, at least in part, the suppression of renin secretion when osmotic diuresis and volume expansion coexist. A tubular, rather than a vascular, intrarenal mechanism is suggested by the observations that intrarenal arterial infusion of saline suppresses renin secretion in filtering (Nash et al., 1968) but not in nonfiltering (Shade et al., 1972) dog kidneys. Indeed, it has been argued that, if the vascular controlling mechanism were predominant during osmotic diuresis, renin secretion should increase rather than decrease (Vander, 1967). This follows from the premises that renal interstitial pressure is increased during diuresis, that transmural pressure is decreased and that decreased transmural pressure (or decreased stretch, a function of transmural pressure) acts on the juxtaglomerular apparatus to stimulate renin secretion (Vander, 1967). Again our experiments provide no direct evidence for or against mediation of the suppressed renin secretion by an intrarenal baroreceptor mechanism. At any rate, blood pressure did not increase and overall renal hemodynamics did not change in both groups of diuretic rats, but renin secretion did. Vander (1967) and Vander and Miller (1964) proposed that osmotic diuretics suppress renin secretion by increasing the delivery of Na + to the macula densa segment. A quantitative test of the relationship between macula densa Na + load and renin secretion is precluded by the inaccessibility of the macula densa segment to micropuncture. However, the distal tubules of rats can be punctured downstream from the macula densa, and even if variable amounts of Na + are reabsorbed between the macula densa and the puncture site, it is clear that Na + load at the macula densa must approximate (be equal to, or exceed slightly) the load measured at the puncture site. We found that Na + load increased during either mannitol or saline diuresis. The increased load could be explained in two ways. Perhaps macula densa Na + load was not increased, and the increased delivery downstream was due entirely to reduced reabsorption in the intervening nephron segment. In view of the facts that, on the average, the puncture site was very close to the macula densa segment and that measured load was approximately doubled during diuresis, this explanation seems unlikely, even if reabsorption in the intervening segment is reduced during diuresis. The second explanation is the more likely, that macula densa Na + load was increased during diuresis. If this was the case, our observations are consistent with the proposal that Na + load and renin secretion are related reciprocally and that increases in load suppress secretion, as proposed by Vander (1967). Although consistent, our observations do not establish a cause-and-effect relationship; suppressed secretion and increased load might be independent events. On the other hand, our observations argue strongly against the proposal that Na + load and renin secretion are directly, rather than reciprocally, related that increases in Na + load stimulate renin secretion, as proposed by Thurau et al. (1967). Moreover, since Na + concentration was affected in opposite directions by saline and mannitol but renin secretion was inhibited by both, either a direct or a reciprocal causal relation between renin secretion and Na + concentration per se seems improbable. Vander (1967) suggested that renin secretion might be responsive to macula densa intracellular Na + concentration rather than to the load of Na + delivered to the macula densa segment. He reasoned that increases in delivered load induce increases in reabsorptive flux, which entail increases in intracellular Na + concentration. Our findings provide no evidence for or against this possibility; to date, neither reabsorptive flux across macula densa cells nor their intracellular Na + concentration has been measured. Recently, by analogy with the transporting characteristics of the ascending limb of Henle's loop, it has been suggested that macula densa cells actively transport Cl~ rather than Na +, and attention has been directed toward the possibility that renin secretion responds to Cl" rather than to Na + (Kirchner et al., 1978; Schnermann et al., 1976). Mason et al. (1978) have suggested that Na + and Cl" concentrations are virtually identical in the macula densa segment, in which case the suppression of renin secretion in our experiments could be attributed to increased Cl~ load rather than to increased Na + load. Regardless of which ion is transported actively by macula densa cells, renin secretion probably is related to a normal concomitant of reabsorptive flux, such as intracellular Na +, as suggested by Vander (1967), rather than to flux per se. This is suggested by the observations that furosemide, ethacrynic acid, and ouabain all inhibit NaCl reabsorptive flux in the ascending limb (Burg and Green, 1973; Burg et al., 1973); however, ouabain inhibits (Churchill and McDonald, 1974), whereas furosemide and ethacrynic acid stimulate (Corsini et al., 1975; Freeman et al., 1974; Bailie et al., 1973; Blaine and Zimmerman, 1978; Cooke et al., 1970; Vander and Carlson, 1969; Blaine, 1977), renin secretion. Intracellular Na + seems a likely candidate; undoubtedly intracellular Na + increases after inhibition of peritubular Na,K-ATPase activity by ouabain (Nechay 1974) and, most probably, it decreases after furosemide or ethacrynic acid attaches to luminal membrane receptors (Burg and Green, 1973; Burg et al., 1973) and blocks Na + and Cl~ influx from tubular fluid. However, further speculation along these lines is unwarranted until information becomes available concerning the transporting characteristics and intracellular fluid composition of macula densa cells. References Bailie MD, Davis LE, Loutzenhiser R (1973) Intrarenal secretion of renin in the dog: Effect of furosemide. Am J Physiol 224:

7 792 CIRCULATION RESEARCH VOL. 45, No. 6, DECEMBER 1979 Birbari A (1972) Intrarenal factors in the control of renin secretion. Pfluegers Arch 337: Blaine EH, Davis JO, Witty RT (1970) Renin release after hemorrhage and after suprarenal aortic constriction in dogs without sodium delivery to the macula densa. Circ Res 27: Blaine EH (1977) Renin secretion after papaverine and furosemide in conscious sheep. Proc Soc Exp Biol Med 154: Blaine EH, Zimmerman MB (1978) Renal function and renin secretion after administration of ouabain and ouabain plus furosemide in conscious sheep. Circ Res 43: Brubacher ES, Vander AJ (1968) Sodium deprivation and renin secretion in unanesthetized dogs. Am J Physiol 214: Burg M, Green N (1973) Effect of ethacrynic acid on the thick ascending limb of Henle's loop. Kidney Int 4: Burg M, Stoner L, Cardinal J, Green N (1973) Furosemide effect on isolated perfused tubules. Am J Physiol 225: Carvalho JS, Shapiro R, Hopper P, Page LB (1975) Methods for serial study of renin-angiotensin system in the unanesthetized rat. Am J Physiol 228: Churchill PC, Churchill MC, McDonald FD (1973) Renin release in anesthetized rats. Kidney Int 4: Churchill PC, McDonald FD (1974) Effect of ouabain on renin secretion in anesthetized dogs. J Physiol (Lond) 242: Churchill, PC, Churchill, MC, McDonald FD (1978) Renin secretion and distal tubule Na* in rats. Am J Physiol 235: F611- F616 Cooke CR, Brown TC, Zacherle BJ, Walker WG (1970) The effect of altered sodium concentration in the distal nephron segments on renin release. J Clin Invest 49: Corsini WA, Hook JB, Bailie MD (1975) Control of renin secretion in the dog: Effects of furosemide on the vascular and macula densa receptors. Circ Res 37: Davis JO, Freeman RH (1976) Mechanisms regulating renin release. Physiol Rev 56: 1-56 Fray JCS (1976) Stretch receptor model for renin release with evidence from perfused rat kidney. Am J Physiol 231: Freeman RH, David JO, Gotshall RW, Johnson JA, Spielman WS (1974) The signal perceived by the macula densa during changes in renin release. Circ Res 35: Humphreys MH, Reid IA, Ufferman RC, Lieberman RA, Earley LE (1975) The relationship between sodium excretion and renin secretion by the perfused kidney. Proc Soc Exp Biol Med 150: Kirchner KA, Kotchen TA, Galla JH, Luke RG (1978) Importance of chloride for acute inhibition of renin by sodium chloride. Am J Physiol 235: F444-F450 Mason J, Takabatake T, Olbricht C, Thurau K (1978) The early phase of experimental acute renal failure. III. Tubuloglomerular feedback. Pfluegers Arch 373: Nash FD, Rostorfer HH, Bailie MD, Wathen RL, Schneider EG (1968) Renin release. Relation to renal sodium load and dissociation from hemodynarruc changes. Circ Res 22: Nechay BR (1974) Relationship between inhibition of renal Na* + K + -ATPase and natriuresis. Ann NY Acad Sci 242: Pitts RF (1974) Physiology of the Kidney and Body Fluids, 3rd ed, Chicago, Year Book Medical Publishers, pp Ploth DW, Schnermann J, Dahlheim H, Hennle M, Schmidmeier E (1977) Autoregulation and tubuloglomerular feedback in normotensive and hypertensive rats. Kidney Int 12: Reeves, G, Sommers SC (1965) Sensitivity of the renal macula densa to urinary sodium. Proc Soc Exp Biol Med 120: Schnermann J, Wright FS, Davis JM, v Stackelberg W, Grill G (1970) Regulation of superficial nephron filtration rate by tubuloglomerular feedback. Pfluegers Arch 318: Schnermann J, Ploth DW, Hermle M (1976) Activation of tubuloglomerular feedback by chloride transport. Pfluegers Arch 362: Shade RE, Davis JO, Johnson JA, Witty RT (1972) Effects of renal arterial infusion of sodium and potassium on renin secretion in the dog. Circ Res 31: Thurau K, Schnermann J, Nagel W, Horster M, Wahl M (1967) Composition of tubular fluid in the macula densa segment as a factor regulating the function of the juxtaglomerular apparatus. Circ Res 20/21 (suppl 2): Vander AJ, Miller R (1964) Control of renin secretion in the dog. Am J Physiol 207: Vander AJ (1967) Control of renin release. Physiol Rev 47: Vander AJ, Carlson J (1969) Mechanism of the effects of furosemide on renin secretion in anesthetized dogs. Circ Res 25: