Cardiac function in rats selectively bred for low- and high-capacity running.

Rat genetic models of intrinsic (i.e., untrained) low-capacity runners (LCR) and high-capacity runners (HCR) are being developed by artificial selective breeding for treadmill running. At generation 3, these lines differed in running capacity by 114%. We used generation 3 rats to test the hypotheses that HCR, relative to LCR, have 1) greater isolated cardiac performance and 2) more resistance to myocardial ischemic insult. The LCR ran for 227 +/- 7 m, and the HCR ran 994 +/- 11 m at exhaustion (337% difference, P < 0.001). Isolated heart performance was assessed from cardiac output (CO) generated at constant preload (15 mmHg) and afterload (70 mmHg) using a Langendorff-Neely working heart preparation. CO averaged 33.5 +/- 2.0 ml. min(-1). g(-1) in LCR hearts and 49.9 +/- 1.4 ml. min(-1). g(-1) in HCR hearts (49% difference, P < 0.001). Recovery of CO after 25 min of global ischemia was not different between the lines. These results suggest that 1) increased cardiac performance accounts for part of the difference in running capacity between the lines; and 2) unlike exercise training, genetically determined intrinsic capacity for exercise does not influence the recovery from 25 min of global low-flow cardiac ischemia.

A self-powered constant infusion device for use in unrestrained rats.

We developed a device that delivers fluid through a catheter at a constant rate and can be used in conscious animals to solve a variety of problems. For example, this device can be used for delivering drugs and maintaining intravascular catheter patency. The device provides infusions at low flows (1.0-1.5 ml/day), so that experimental agents may be administered with minimal volume loading of the rat. Arterial and venous catheter patency is maintained by infusion of heparinized saline through indwelling catheters attached to the device. The catheters exit from the rat in the intrascapular area and are routed through a protective spring to the device, which is suspended above the cage. The catheters may be attached to pressure transducers, blood may be sampled, and injections or infusions may be made without disturbing the rat. Because the device is self-contained, it can be suspended by a fluid-free swivel that rotates through 360 degrees, providing minimal restraint. The device has been used successfully to measure arterial and central venous blood pressures in two studies using rats.

Body fluid expansion is not essential for salt-induced hypertension in SS/Jr rats.

To evaluate the importance of volume in the development of hypertension in inbred Dahl salt-sensitive rats (SS/Jr), we measured the changes in blood pressure (BP) that occurred with oral intake of food (salt) and water in rats whose body weight was permitted to increase versus those in which body weight was maintained constant with a servo-control system. We hypothesized that if volume expansion is essential in the development of hypertension, then BP would not increase if body weight was held constant. We found that oral presentation of chow containing 4% salt to SS/Jr rats caused BP to increase 32.2 +/- 2.9 mmHg over 4 days when body weight was controlled at its initial value. Plasma sodium increased from 142.0 to 145.2 meq/l during 4 days of high salt. Neither plasma volume, hematocrit, nor central venous pressure changed significantly on the high-salt diet. In contrast, the inbred Dahl salt-resistant rats (SR/Jr) did not increase their BP during body weight control when given 4% salt. This demonstrates that volume expansion is not an obligatory step in the pressure response to increased salt in SS/Jr rats. Our results obtained with oral presentation of salt, in contrast to intravenous, represent a physiological evaluation of the significance of volume changes in response to dietary salt because no potential regulatory reflexes have been bypassed.

Phenotypic variation in sensorimotor performance among eleven inbred rat strains.

As a first step toward identifying the genes that determine sensorimotor ability (motor coordination) we subjected 11 inbred strains of rats to three different tests for this trait. Rats were tested at 13 wk of age to determine how long they could remain on 1) a rotating cylinder as the velocity of rotation increased every 5 s (1-direction rotation test), 2) a rotating cylinder that reversed direction every 5 s and increased velocity every 10 s (2-direction rotation test), and 3) a platform that was tilted 2 degrees every 5 s from 22 to 47 degrees (tilt test). On all three tests, rats of the PVG strain demonstrated the greatest sensorimotor ability. In contrast, rats of the MNS strain were most often represented among the group of strains that demonstrated the lowest performance on all tests. Considering all three tests, there was a 3- to 13-fold range in sensorimotor performance between the highest and lowest strains. This large divergence between the highest and lowest strains provides a genetic model that can be used to identify intermediate phenotypes and quantitative trait loci that contribute to sensorimotor ability.

Phenotypic variation in strength among eleven inbred strains of rats.

As a first step toward the long-range goal of identifying the genes that determine strength, we subjected 11 inbred strains of rats to three tests of muscular strength. The tests consisted of measuring (1) the force exerted by the rat as it was pulled by the base of the tail off a grid on the pan of a top-loading electronic balance (scale test); (2) the length of time the rat hung from a 2.5-mm-diameter U-shaped wire (wire-hanging test); and (3) the length of time the rat hung from a vertically oriented grid consisting of 4-mm-diameter rods (grid-hanging test). Six rats of each gender from each strain were tested at 12 weeks of age, once/day for 5 consecutive days. For the two tests that required use of all four limbs (the scale and grid-hanging tests), one strain performed best (DA). In contrast, on the test that required primarily the use of the front limbs (wire-hanging test), the DA was the lowest performing strain and the F344 rats the best. This differential ranking suggests that the tests selected for variance in the morphological distribution of strength among the strains. There was a 1.5- to 5.2-fold divergence observed between the males of the highest and lowest strains on the scale test and grid hanging tests. This large divergence provides the opportunity to search for intermediate phenotypes and quantitative trait loci that contribute to the different performances of the strains on these strength tests.

Heritability of treadmill running endurance in rats.

Treadmill running was evaluated as a phenotype for selective breeding for high- and low-endurance performance from a starting population of 18 male and 24 female outbred Sprague-Dawley rats. Each rat was exercised to exhaustion once per day for 5 consecutive days. The treadmill was set at a constant 15 degrees slope, and the initial velocity of 10 m/min was increased by 1 m/min every 2 min. The total distance run on the single best day out of the five trials was taken as the measure of endurance performance. The original population (males and females combined, n = 42) ran on average for 396 m. The two lowest-performing pairs and two highest-performing pairs were selectively bred through three successive generations. After three generations of selection, performance of the offspring from the high selected line averaged 659 +/- 36 m (n = 20), whereas low-performance offspring (n = 13) averaged 388 +/- 28 m. The narrow-sense heritability, calculated as the regression of individual offspring performance on midparental value for each family, was 0.39 across the three generations. This implies that 39% of the variation in running endurance performance between the low and high selected lines was determined by heritable factors.

Spectrum of aerobic endurance running performance in eleven inbred strains of rats.

The goal of this study was to identify inbred rat strains that could serve as useful models for exploration of the genetic basis of aerobic endurance performance. Six rats of each gender from 11 different inbred strains were tested for 1) maximal running capacity on a treadmill and 2) isolated cardiac performance. Running performance was estimated from 1) duration of the run, 2) distance run, and 3) vertical work performed. Cardiac output, during constant preload and afterload, was taken as a measure of cardiac performance from an isolated working heart preparation. The COP rats were the lowest performers and the DA rats were the best performers by all estimates of running performance. Across the 11 strains, the distance run correlated positively with isolated cardiac performance (r = 0.87). Estimates of performance were as follows (COP vs. DA strain, respectively): duration of run, 19.9 +/- 1.8 vs. 41.5 +/- 2. 2 min; distance run, 298 +/- 30 vs. 840 +/- 64 m; vertical work, 15 +/- 1.7 vs. 40 +/- 4.4 kg/m. These approximately 2.5-fold differences in running performance between the COP and DA suggest that these strains could serve as models for evaluation of the genetic basis of variance in aerobic performance.

Evaluation of the renin-angiotensin system in a congenic renin Dahl salt-sensitive rat.

When an approximately 30 centiMorgan (cM) region of chromosome 13 containing the renin gene from the Dahl salt-resistant rat (R) was introgressed into the Dahl salt-sensitive rat (S), the resulting congenic rat (designated S.R-Ren) had a systolic blood pressure on a 2% (w/w) salt diet that was 24 mmHg lower than that of its S counterpart. Due to the large size of the transferred segment (over 30 million bp), the question remained as to whether or not the renin gene was the cause of the blood-pressure difference between the strains. We evaluated the role of the renin-angiotensin system in S.R-Ren and S rats fed a 0.05% salt diet by examining differences between strains in (1) expression of renin in three tissue types, (2) the blood-pressure response to blockade of both angiotensin-converting enzyme and angiotensin II receptors, and (3) pressure natriuresis. No differences were found in renin levels in plasma, kidney or adrenal gland between strains. The blood-pressure responses to the angiotensin-converting-enzyme inhibitor captopril and to the angiotensin II-receptor blocker saralasin in conscious S and S.R-Ren rats were similar. Furthermore, renal function, evaluated by a pressure-natriuresis index that took into account both the time and the arterial pressure needed to excrete an acute salt load, did not differ between strains. Our findings therefore fail to demonstrate a role for the renin gene in conferring lower blood pressure in the congenic rat and suggest that there is an unknown arterial-pressure-regulating locus in this 30 cM region of chromosome 13.

Hepatorenal reflex in the rat.

The possible existence of a hepatorenal reflex was evaluated in male Sprague-Dawley rats. Sodium excretion was measured in two groups of six rats each, during the first 4 h following acute ingestion of a known amount of high salt chow (2.0-2.5 mequiv. NaCl). Hourly excretion rates for sodium before surgery were compared with results following 7 days of recovery from either hepatic denervation (n = 6) or sham denervation (n = 6). Before denervation, hourly sodium excretion between the groups was not different. Following surgery for hepatic denervation, sodium excretion was 91% lower than presurgery values for the 1st h (p < 0.02) and 44% lower in the 2nd h (p < 0.04). Sham denervation caused no significant change in sodium excretion when compared with presurgery results. A test for completeness of denervation showed that norepinephrine concentration in liver tissue taken from denervated rats was 5.1 +/- 8 ng/g and that taken from sham rats was 22.8 +/- 1 ng/g (p < 0.001). These data demonstrate that the liver is essential for the normal postprandial excretion of sodium following ingestion of a high salt meal in rats.

Contribution of cytosolic and membrane-bound 5'-nucleotidases to cardiac adenosine production.

The purpose of this study was to evaluate the relative contributions of AMP-specific cytosolic 5'-nucleotidase and ecto-5'-nucleotidase to cardiac adenosine production and its regulation by ADP and Mg2+. 5'-Nucleotidase activity was measured spectrophotometrically in the total homogenate, the 150,000-g supernatant fraction (cytosolic 5'-nucleotidase), and the membrane pellet fraction (ecto-5'-nucleotidase) of dog left ventricles. Increasing [MgCl2] over a range from 0 to 6 mmol/l increased 5'-nucleotidase activity in both the supernatant and pellet; only cytosolic 5'-nucleotidase exhibited an absolute requirement for Mg2+. ADP, (20-480 mumol/l) activated supernatant and inhibited membrane-bound 5'-nucleotidase activity. At 80 mumol/l ADP, 5 mmol/l MgCl2, 100 mumol/l AMP, and pH 7.3, the average 5'-nucleotidase activities of the supernatant vs. pellet were 74% of total and 26% of total, respectively. Total adenosine production in unfractionated samples of ventricular homogenates decreased an average of 73% by specific inhibition of cytosolic 5'-nucleotidase, using antibodies against the cytosolic enzyme, and 46% by specific inhibition of ecto-5'-nucleotidase with alpha, beta-methylene adenosine 5'-diphosphate (AOPCP). These findings support the hypotheses that 1) both cytosolic and ecto-5'-nucleotidase contribute to cardiac adenosine production in dog heart homogenates; 2) AMP-specific cytosolic 5'-nucleotidase activity exceeds ecto-5'-nucleotidase activity at physiological concentrations of ADP, AMP, and Mg2+; and 3) Mg2+ is an important regulator of cardiac adenosine production via activation of both ecto- and AMP-specific cytosolic 5'-nucleotidases.

Spontaneous changes in arterial blood pressure and renal interstitial hydrostatic pressure in conscious rats.

1. Previous work has demonstrated a positive relationship between experimentally induced changes in arterial pressure (AP) and renal interstitial hydrostatic pressure (RIHP). The purpose of the present study was to test the hypothesis that RIHP is positively correlated with the normal changes in AP that occur spontaneously in conscious rats. 2. Rats were chronically instrumented for the recording of AP (via an aortic catheter) and RIHP. RIHP was measured by implanting a Millar microtransducer, whose tip had been encapsulated in a 35 microns pore polyethylene matrix (5 mm long, 2 mm o.d.), approximately 5 mm below the renal cortical surface. 3. A total of 56 h of simultaneous analog recording of AP and RIHP was obtained from ten rats. Each 1 h segment was digitized and evaluated at frequencies of 1, 0.1, 0.02 and 0.01 Hz. 4. In forty-nine out of fifty-six of these 1 h recordings taken at 1 Hz, there were significant positive linear correlations between AP and RIHP (mean r = 0.32) with a mean slope of 0.11 mmHg RIHP/1 mmHg AP. Low-pass filtering to 0.01 Hz significantly increased the r value to 0.48. 5. These results demonstrate that spontaneous changes in AP and RIHP are positively correlated. The spontaneous coupling of AP and RIHP may be of importance in the regulation of salt and water excretion by the pressure diuresis mechanism.

Spontaneous pressure-flow patterns in the kidney of conscious rats.

We have developed a model that permits a quantitative analysis of the contribution of different mechanisms to the spontaneously occurring pressure-flow patterns of a vasculature. In this study we evaluated the spontaneous relationship between arterial pressure (P) and renal blood flow (F) in resting conscious rats during control conditions, autonomic ganglionic blockade (hexamethonium), and nonselective alpha-adrenoreceptor blockade (phentolamine). In a total of 250 trials in 29 rats, we measured the average P and F for each cardiac cycle over 13-min periods (approximately 4,000 cardiac cycles/trial). The P and F values for each cardiac cycle were expressed as percentage change from each 13-min average (beat-to-beat changes). The slope and angle of each consecutive beat-to-beat P-F change were calculated and collated into one of eight zones representing the physiological mechanisms responsible for the concurrent spontaneous changes in P and F. Our results reveal that, in the absence of any chemical or mechanical intervention (control), the renal circulation demonstrated a baroreflex-like P-F pattern approximately 38% of the time. An autoregulatory-like P-F pattern occurred, at the most, 35% of the time. Autonomic ganglionic blockade significantly (P < 0.05) decreased the baroreflex-like pattern and increased the presence of P-F patterns indicative of autoregulation. alpha-Adrenoreceptor blockade resulted in a P-F pattern that was qualitatively similar to that produced by hexamethonium, but with considerably more variability. These results indicate that, in the resting conscious undisturbed state, the autonomic nervous system exerts a tonic influence on the renal circulation that facilitates arterial pressure regulation via a baroreflex-like pattern.

Dynamic, short-term coupling between changes in arterial pressure and urine flow.

Pressure diuresis refers to the direct effect of arterial pressure (AP) on the rate of urine flow (UF). On the basis of computer modeling, pressure diuresis has been viewed as a long-term mechanism that acts to set the level of the blood volume and, thus, the steady-state AP. There are no systematic studies, however, on the rapidity with which changes in AP induce changes in UF in vivo. Therefore, we measured the delay between induced changes in AP and the subsequent change in UF. Nine anesthetized rats were instrumented with arterial, venous, and ureteral catheters. AP and UF were measured every 2 s, while acute changes in AP were induced by 1) occlusion of the aorta above or below the renal vessels; 2) brief tail pinch; or 3) intravenous administration of acetylcholine (1 microgram), phenylephrine (1 microgram), or angiotensin II (0.1 microgram). The rapidity of the urinary response to induced changes in AP was determined by calculating the delay between a significant change in AP (+/- 2 SD from baseline) and a significant change in UF. The delay averaged 6.0 +/- 0.5 s for all conditions. Also, examining the relationship between the magnitude of the induced changes in AP and the magnitude of the responses in UF revealed an exponential influence of AP on UF. That is, there were proportionately larger changes in UF compared with AP (< or = 10 times greater magnitude) in response to the experimental interventions.(ABSTRACT TRUNCATED AT 250 WORDS)

Gravimetric method for the dynamic measurement of urine flow.

The rate of urine formation is a primary index of renal function, but no techniques are currently available to accurately measure low rates of urine flow on a continuous basis, such as are normally found in rats. We developed a gravimetric method for the dynamic measurement of urine flow in anesthetized rats. Catheters were inserted directly into the ureters close to the renal pelves, and a siphon was created to collect all of the urine formed as rapidly as it was produced. Urine flow was determined by measuring the weight of the urine using a direct-reading analytical balance interfaced to a computer. Basal urine flow was measured at 2-sec intervals for 30 to 60 min. The dynamic response of urine flow to a rapid decrease in arterial pressure produced by a bolus intravenous injection of acetylcholine (0.5 micrograms) was also measured. Intrinsic drift, evaporative losses, and the responsiveness of the system to several fixed pump flows in the low physiologic range were evaluated in vitro. The gravimetric method described was able to continuously measure basal urine flows that averaged 37.3 +/- 12.4 microliters/min. Error due to drift and evaporation was negligible, totaling less than 1% of the measured urine flow. Acetylcholine-induced declines in arterial pressure were followed within 8 sec by a decline in urine flow. These data demonstrate that this new gravimetric method provides a simple, inexpensive, dynamic measurement of urine flow in the microliter/min range.

Immunogold localization of adenosine 5'-monophosphate-specific cytosolic 5'-nucleotidase in dog heart.

Adenosine has a major regulatory function in the heart and many tissues. Our previous work showed that a cytosolic (not a membrane, as previously hypothesized) 5'-nucleotidase from dog heart has the kinetic properties consistent with it being the enzyme responsible for adenosine formation from adenosine 5'-monophosphate (AMP) in response to hypoxia or ischemia. In the present study, we evaluated the spatial distribution of AMP-specific cytosolic 5'-nucleotidase in dog heart using electron microscopic immunogold localization. Polyclonal antibodies raised against purified cytosolic 5'-nucleotidase recognized the 43-kd subunit of the enzyme on Western blots of both purified enzyme and the soluble fraction of dog heart homogenates but did not react with proteins extracted from the membrane fraction. Purified cytosolic 5'-nucleotidase and 5'-nucleotidase activity present in the soluble fraction of heart homogenates were inhibited by anti-cytosolic 5'-nucleotidase, but the membrane fraction was not. The monospecific antibodies against the cytosolic 5'-nucleotidase were used for electron microscopic immunogold localization of cytosolic 5'-nucleotidase in dog heart tissue sections. Cytosolic 5'-nucleotidase was found in the cytoplasm of red blood cells, cardiac myocytes, and endothelium; the plasma membrane and interstitium were devoid of gold label. These results are the first to document the presence cytosolic 5'-nucleotidase in specific cell types in the heart and demonstrate the potential for these cell types to produce adenosine via cytosolic 5'-nucleotidase.

Purification and regulation of an AMP-specific cytosolic 5'-nucleotidase from dog heart.

The major enzyme responsible for adenosine production during myocardial hypoxia or ischemia is 5'-nucleotidase. We purified an AMP-specific 5'-nucleotidase to homogeneity from the 150,000-g supernatant of dog heart homogenate using phosphocellulose, DEAE-cellulose, and ADP-agarose affinity chromatography. Sodium dodecyl sulfate-poly-acrylamide gel electrophoresis of the purified enzyme yielded a single protein band of 43 kDa. The molecular mass of the holoenzyme, determined by gel filtration and sucrose density-gradient centrifugation, was approximately 166 kDa, suggesting a tetrameric structure. Dog heart cytosolic 5'-nucleotidase was active at physiological pH (6.8-7.8) and demonstrated a preference for AMP over IMP as substrate. The enzyme exhibited sigmoidal saturation kinetics, with half-maximal activity at 2.6 mM AMP in the absence of ADP. ADP (0-250 microM) activated cytosolic 5'-nucleotidase by increasing maximal velocity and affinity for AMP. The enzyme was inhibited by 4 mM ATP, but 5'-nucleotidase activity increased as [ATP] was reduced. Mg2+ was required for activity, with maximal activation at approximately 3.5 mM free Mg2+. These data suggest that the regulation of AMP-specific cytosolic 5'-nucleotidase by adenine nucleotides and free Mg2+ may be important in the production of adenosine during conditions promoting ATP hydrolysis, such as myocardial hypoxia or ischemia.

Spontaneous pressure-flow relationships in renal circulation of conscious dogs.

Renal pressure-flow (P-F) relationships are usually evaluated by measuring effects of mechanically induced changes in renal arterial pressure (AP) on renal blood flow (RBF). We devised a method allowing evaluation of renal P-F relationships during normal changes in AP occurring spontaneously in a conscious animal rather than during artificially induced changes in AP. In 18 trials in 6 dogs standing at rest, we measured average AP and RBF for each cardiac cycle over periods of approximately 35 min (approximately 3,100 cardiac cycles/trial). AP and RBF values for each cardiac cycle were expressed as percent change (%delta) from the 35-min average (beat-to-beat changes). Slope and angle of each consecutive beat-to-beat P-F change were calculated and collated into one of eight zones representing the possible physiological mechanisms responsible for concurrent, spontaneous changes in RBF and AP. In a predominance of the cardiac cycles (approximately 43%), the spontaneous AP-RBF relationship was consistent with being mediated by arterial baroreflexes (i.e., increases in AP were accompanied by proportionately greater increases in RBF during 44.4% of cardiac cycles in which AP increased, and decreases in AP were accompanied by proportionately greater decreases in RBF during 41.4% of cardiac cycles in which AP decreased). Blockade of autonomic ganglionic transmission with hexamethonium markedly attenuated this pattern. Our results indicate that renal circulation participates in moment-to-moment control of AP via a predominant baroreflex-like pattern.

Evaluation of spontaneous baroreflex sensitivity in conscious dogs.

1. We evaluated a method of measuring cardiac baroreflex sensitivity (BRS) derived from spontaneous changes in systolic pressure (SP). SP was measured from the ECG signal in seven conscious, resting dogs. 2. Beat-to-beat changes in PI (dPI) were positively correlated with beat-to-beat changes in SP (dSP) in all dogs tested, suggesting spontaneous baroreflex function. The slope of the regression of dPI on dSP was used as an index of spontaneous BRS. 3. The spontaneous BRS was abolished by hexamethonium, atropine and bilateral carotid sinus denervation. Low dose atropine sulphate produced a paradoxical increase in spontaneous BRS, which has been observed in other studies. The spontaneous BRS was positively correlated with the average pulse interval in resting dogs. 4. Random modulation of heart rate after vagotomy failed to reproduce the strong positive correlation between dSP and dPI; this demonstrated that the correlation was not the result of mechanical coupling between heart rate and arterial blood pressure. 5. The BRS was measured pharmacologically in six dogs using a bolus injection of a vasoconstrictor. The pharmacological BRS was positively correlated with the spontaneous BRS measured after the bolus injection. 6. Finally, the spontaneous BRS was negatively correlated with the average arterial pressure in resting dogs. We conclude that the spontaneous BRS is a useful quantitative indicator of baroreflex function in conscious resting dogs.

Autoregulatory capacity in renal and mesenteric vasculatures of mineralocorticoid hypertensive dogs.

Experiments were performed in seven conscious dogs to quantitate the renal and mesenteric autoregulatory capacity in deoxycorticosterone acetate (DOCA)-salt hypertensive dogs. Each dog was chronically instrumented for the measurement of aortic blood pressure and blood flows through the superior mesenteric and left renal arteries (Doppler measurements); a right nephrectomy was also performed. To induce hypertension, animals were administered 5 mg/kg DOCA each day and given 1% saline as a drinking solution. After arterial pressure had increased by 25% (8-14 days of DOCA), a large-bore catheter was positioned retrograde in the left common carotid artery for subsequent connection to a gravity reservoir system for acute control of arterial pressure. Autoregulatory capacity was evaluated by controlling arterial pressure via the gravity reservoir at hypertensive and then normotensive values for 15-min periods. In neurally intact animals the renal, but not the mesenteric, circulation displayed autoregulatory capacity; the closed-loop gain of renal flow control averaged 0.92 +/- 0.06. Inhibition of autonomic ganglionic transmission (hexamethonium) increased the gain of flow regulation in the mesenteric circulation, but mesenteric autoregulation was not manifest. Ganglionic blockade, inhibition of prostaglandin formation, or blockade of histamine or adenosine receptors did not significantly influence renal autoregulatory capacity. These data demonstrate that the kidney has a high capacity to autoregulate in DOCA hypertension and suggest that autoregulatory-mediated renal vasoconstriction contributes to the increased total peripheral resistance.

Pressure diuresis and autonomic function in conscious dogs.

Pressure diuresis is thought to be a major long-term regulator of arterial blood pressure (AP). Previously, pressure diuresis has been characterized using pharmacological or surgical blockade of other mechanisms known to affect renal function. This study evaluated pressure diuresis in conscious dogs with minimal experimental interference. Dogs were chronically instrumented under pentobarbital anesthesia with aortic and urinary bladder catheters. AP was increased by 10% in resting dogs by exposure to increased light and sound intensity (arousal) for 90 min. During arousal, urine flow (UV) and Na+ excretion (UNa+ V) correlated with AP (UV vs. AP, r = 0.12, P less than 0.05; UNa+ V vs. AP, r = 0.19, P less than 0.005; 17 trials in 7 dogs). Arousal did not affect the plasma concentration of atrial natriuretic factor, suggesting that this hormone did not contribute to the correlations between UV or UNa+ V and AP. Because arousal may induce an autonomically mediated antidiuresis, studies were repeated during autonomic ganglionic blockade with hexamethonium. During autonomic blockade, the correlations between UV or UNa+ V and AP were increased (UV vs. AP, r = 0.72; UNa+ V vs. AP, r = 0.72, P less than 0.001; 6 trials in 4 dogs). We conclude that the effect of pressure diuresis on UV and UNa+ V can be detected in the intact animal, during normal operation of all the mechanisms that control renal function. Furthermore, when autonomic reflexes are blocked, the pressure-diuresis mechanism is a major determinant of UV and UNa+ V.