Minimum Urine Flow Rate During Water Deprivation: Importance of the Nonurea versus Total Osmolality in the Inner Medulla

STEVEN DEMETRI SOROKA,* SIRITHON CHAYARAKS,* SURINDER CHEEMADHADLI,* JEFFREY ADAM MYERS,* STANLEY RUBIN,t HARALD SONNENBERG,t and MITCHELL LEWIS HALPERIN* *Rena/ Division, St. Michael ‘s Hospital, Toronto, Ontario, canada; tDepartment of Internal Medicine, School of Veterinary Medicine, University of Saskatchewan, Saskatoon, Saskatchewan; and Department of Physiology, University of Toronto, Toronto, Ontario, Canada.

Abstract. Antidiuretic hormone leads to an increase in the mosmol/kg H,O). Nevertheless, they had identical urine flow permeability for water and urea in the inner medullary collect- rates (0.5 ± 0.01 and 0.5 ± 0.02 mI/mm, respectively), and ing duct. Hence, urea may not be an “effective” osmole in the their nonurea osmolality also was similar (587 ± 25 and 475 ± inner medulla during maximal renal water conservation. Ac- 14 mosmollkg H,O, respectively) to the water-deprived normal cordingly, the purpose of this study was to evaluate whether subjects. The composition of their urine differed in that the differences in the rate of urea excretion would influence max- principal nonurea osmoles became NH4 and j3-hydroxybu- imum renal water conservation in humans. tyrate rather than Na and Cl. During water deprivation in In water-deprived rats, the concentration of urea and total normal subjects, the ingestion of urea caused a twofold rise in osmolality were somewhat higher in the urine exiting the inner urine flow rate, a fall in the nonurea osmolality, and a rise in medullary collecting duct than in interstitial fluid obtained the rate of excretion of nonurea osmoles. from the entire papillary tip. Nevertheless, the “nonurea” (total The nonurea osmolality of the urine, and presumably the osmolality minus urea in millimolar terms) osmolality was medullary interstitial fluid as well, was inversely related to the virtually identical in both locations. Chronically fasted human urea excretion rate. In chronic fasting, the nature, but not the subjects that were water-deprived for 1 6 h had a lower rate of quantity, of nonurea osmoles changed. The similar minimum urea excretion (71 ± 7 versus 225 ± 14 p.mollmin) and a urine volume was predictable from an analysis based on nonurea somewhat lower urine osmolality (745 ± 53 versus 918 ± 20 osmole considerations. (J Am Soc Nephrol 8: 880-886, 1997)

When water must be conserved, antidiuretic hormone (ADH) be inserted in the IMCD (9). Therefore, it is reasonable to acts on the distal part of the kidney (late distal convoluted suggest that urea in the inner medulla may not have an appre- tubule, cortical collecting duct, and medullary collecting ducts) ciable direct influence on the rate of excretion of water during to increase the permeability for water (reviewed in reference water deprivation if its concentration is similar in the lumen of I ). Because water is likely to achieve diffusion equilibrium in the IMCD and the interstitial fluid of the inner medulla (6-8). this setting, the sum of the solute concentrations in the inner Accordingly, the urine flow rate would be determined by the medullary interstitium and the urine should be equal. The urine number of nonurea osmoles (osmolality - urea concentration flow rate would then be a function of the number of osmoles to in millimolar terms) excreted when ADH acts (10-14) (Equa- be excreted (2-5) (Equation I ). Nevertheless, not all solutes tion 1). Therefore, the question to be addressed is, What is the have the same impact on water distribution. Only particles that influence of variations in the rate of excretion of urea within its achieve a concentration difference across a semipermeable physiologic range on the urine nonurea osmolality and flow membrane obligate water movement here. Urea concentrations rate? are usually similar in the lumen of, and in interstitial fluid surrounding, the inner medullary collecting duct (IMCD) when Volume (L) ADH acts (6-8) because ADH also causes a urea transporter to = Amount of solute (mmol)/Solute concentration (mM)

(I) Received January 31, 1996. Accepted December 30. 1996. Correspondence to Dr. Mitchell L. Halperin, Division of Nephrology, St. Michael’s Hospital. 38 Shuter Street, Toronto. Ontario. Canada M5B 1A6. Results to be reported indicate that in water-deprived sub- l()46-6673/0806-0880$03.00/0 jects who were fasted for 2 to 3 wk, the rate of excretion of Journal of the American Society of Nephrology urea was much lower than in water-deprived normal subjects. Copyright 0 1997 by the American Society of Nephrology Nevertheless, their urine flow rates were similar because there Control of Urine Flow Rate 881

was no significant difference in the nonurea osmolality of the small volume of water. Urine samples were collected by spontaneous urine; despite this, the composition of the urine was very voiding every hour, and the values reported represent the collections different in these two groups. When the rate of excretion of in which two adjacent samples had similar high rates of urea excretion (samples in the 2 to 4 h after urea ingestion). urea was increased in water-deprived normal subjects, the urine flow rate rose. In this case, both a fall in the urine nonurea osmolality and a rise in the rate of excretion of Analytic Techniques nonurea osmoles could explain their higher urine output. Urine samples were analyzed for Na, K. Cl, urea, creatinine, NH4 , /3-hydroxybutyrate (f3-HB), and osmolality as described previously Materials and Methods (17). Studies in Rats Sprague-Dawley rats (250 to 325 g body wt) were kept without Calculations food and water overnight; they were anesthetized and prepared for Nonurea Osmolality. The nonurea osmolality of the urine is the IMCD microcatheterization as described previously (15). Interstitial measured urine osmolality minus its urea concentration in millimolar fluid was obtained from the exposed papilla by the method of Lee and terms. The rate of excretion of osmoles or nonurea osmoles was equal Williams (16). The inner (white) medulla was excised, blotted, to the product of the urine flow rate and the total osmolality or weighed, and centrifuged twice under oil at 300 and 1500 X g with nonurea osmolality in the urine. fluid obtained during the second centrifugation used for analysis of Statistical Analysis. Values were presented as the mean ± SEM. urea and osmolality. Such fluid is virtually free ofcontamination with Mean values were compared using unpaired values and a I test. IMCD fluid, the contribution of intracellular fluid is minimal, and the composition changes appropriately with dehydration ( 16). Analysis of Results papillary interstitial fluid and urine from the same kidney is provided Studies Rats in Table I. in Fluid obtained from the terminal IMCD had an osmolality and a urea concentration that were somewhat higher than that Studies in Humans in the corresponding interstitial fluid ( 10 16 ± 137 versus The objectives of this study were to examine the effect of variations in the rate of excretion of urea within the physiologic range in 881 ± 150 mosmollkg H,O, and 439 ± 86 versus 296 ± 34 water-deprived subjects on the urine flow rate and to determine mM, respectively, Table 1 ). Nevertheless, there was no differ- whether changes in the urine flow rate correlated with changes in ence in the nonurea osmolality in these fluids (585 ± 50 versus nonurea osmolality. Volunteers (a 37, ages 17 to 58, 26 male and 575 ± 49 mosmollkg H,O, Table I). 1 1 female) followed their usual diet and took no medications. In- formed consent and ethics approval were obtained for all protocols. Studies in Humans Comparison of Overnight Water Deprivation in Controls and Effect of a Low Rate of Urea Excretion on the Compo- Fasted Subjects. All subjects provided a second-voided urine after sition of the Urine During Water Restriction. The compo- I 2 to I 6 h of water deprivation (collection time close to I 0:00 h am.). To study maximal renal water conservation when there was a low rate sition of the urine in subjects who followed their usual diet and of urea excretion, identical studies were also carried out in four were water deprived for 12 to 16 h is shown in Table 2; these markedly obese subjects who had fasted for 2 to 3 wk in a weight values were compared to urine values in obese, fasted subjects reduction program. who had a much lower rate of excretion of urea (7 1 ± 7 versus Effect of a Urea Load on the Urine Composition After Over- 225 ± 14 p.mol/min, P < 0.01, Table 2). In the control night Water Deprivation. The rationale for this protocol was to subjects, approximately half of the urine osmoles were urea. study the effect of raising the rate of excretion of urea within the The urine flow rate after 16 h of water deprivation was similar physiologic range on the urine flow rate and its composition during in both groups (0.5 ± 0.02 and 0.5 ± 0.01 ml/min). The urine antidiuresis. Six subjects followed their regular diet and did not eat or nonurea osmolality was 475 ± 14 mosmollkg H,O in the drink for 1 2 h. A second-voided morning urine sample was obtained, control subjects, and it was slightly higher in the fasted sub- after which each subject ingested 6 mmol urea/kg body wt with a jects (587 ± 25 mosmol/kg H,O, Table 2). When the rate of nonurea osmole excretion was plotted against the rate of urea

Table 1. Urea and osmolality in the renal papilla” excretion in the 37 control subjects with and without urea supplements, there was a significant inverse correlation be- Kidney tween these two parameters (Figure 1); the urine flow rate was

Papilla Urine now a function of the rate of excretion on nonurea osmoles (Figure 1). Osmolality mosmollkg H2O 88 1 ± 150 1 0 1 6 ± I 37 The compositions of the urine in the control and fasted Nonurea mosmollkg H,O 585 ± 50 575 ± 49 water-deprived subjects were very different. In the fasted sub- osmolality jects, the urine contained very little Na, Cl, K, and urea; its Urea mM 296 ± 34 439 ± 86 major constituents were the ions NH4 and -HB (Table 2). There was a somewhat higher urine nonurea osmolality, but ‘t For details, see text. The plasma urea concentration was 6.8 ± little difference in the rate of nonurea osmole excretion in these 0.5 mM in these hydropenic rats. Urine was obtained from the terminal inner medullary collecting duct just before excision of the two groups despite the major difference in the rate of excretion papilla. of urea (Table 2). 882 Journal of the American Society of Nephrology

Table 2. Urine composition during maximum antidiuresis in normal and chronically fasted subjectsa

Diet

Normal Chronic Fasting (,z=37) (n4)

Flow rate (ml/min) 0.5 ± 0.02 0.5 ± 0.01 Concentrations in the urine osmolality mosmollkg H,O 917 ± 23 745 ± 53 nonurea osmolality mosmollkg H2O 475 ± 14 587 ± 25 urea mM 442 ± 17 158 ± 29b Na mM 122±9 4±lb K mM 99±9 37±7’’ Cl mM 171±10 46±Sb NH4 mM 24 ± 6 255 ± 12b

f3-HB mM <5 155 ± 12b Urinary excretion rates osmoles p.osmol/min 475 ± 25 352 ± 30 nonurea osmoles p.osmol/min 244 ± 14 281 ± 33 urea p.mol/min 225 ± 14 71 ± 7”

Na p.mol/min 64 ± 6 2 ± 01b

K p.mol/min 48 ± 4 18 ± 4b Cl p.mol/min 88 ± 7 2 1 ± 3b

NH4 p.mol/min 12 ± 4 122 ± 14b

a For details, see text. Values were obtained after 16 h of water deprivation from a second-voided specimen. The fasted subjects received a supplement of 16 to 64 mmol of KCI daily. Effective osmoles are calculated as urine osmolality minus the urine urea concentration. bp < 0.01.

Studies During a Urea Load. Urea was ingested in water- (6,7, 1 8). Accordingly, the nonurea or effective osmolality of deprived subjects on their regular diet. There was no signifi- the medullary interstitium and the number of impermeant par- cant change in the urine osmolality after urea ingestion (884 ± tides in the luminal fluid of the IMCD should dictate what the 40 versus 845 ± 34 mosmollkg H,O, Table 3), but the urine urine flow rate will be when ADH acts (substitute nonurea flow rate virtually doubled (0.6 ± 0. 1 to 1 . 1 ± 0. 1 ml/min). terms for solute in Equation 1). When the components of the urine were examined in nonurea The rationale of the present study was to determine whether osmolality terms. the urine nonurea osmolality declined signif- variations in the rate of excretion of urea within the physiologic icantly from 448 ± 45 to 3 1 1 ± 34 mosmol/kg H2O, and the range would influence the nonurea osmolality of the urine, and nonurea osmole excretion rate rose significantly from 261 ± by inference, of the interstitial fluid in the inner medulla. 57 to 337 ± 60 p.Osmlmin (Table 3). Because both urea and water are reabsorbed from the IMCD at the same total osmolality of interstitial fluid at that horizontal Discussion plane, this will cause an initial fall in the nonurea, but not the As background, building on well-accepted concepts, only total osmolality of this interstitial fluid (Figure 2) (19). Thus, particles restricted to one of two compartments influence water the entry of water along with urea from the IMCD causes the distribution between these compartments in a direct manner. It [Na] and [Cl] to fall in interstitial fluid. This in turn promotes is generally accepted that urea does not contribute to the the diffusion of Na and Cl without water from the water- effective osmolality or tonicity of plasma because it achieves impermeable, thin ascending limb into the interstitium, raising an equal concentration in the intracellular and extracellular the interstitial osmolality. This extra water-free solute addition fluid. Thus, the total osmolality minus the urea concentration in leads to water diffusion out of the thin descending limb of the millimolar terms should reflect the effective osmolality in body loop of Henle, raising is luminal NaCI concentration (20, 21). fluid compartments. Applying this same logic to the IMCD, When this luminal fluid passes the tip of the loop and enters the again one should consider only impermeant particles and not thin ascending limb of the loop of Henle, the higher concen- urea when ADH acts, because in many circumstances there tration of NaCl will augment its diffusion into the interstitial seems to be a high enough permeability for urea in the IMCD fluid (Figure 2). Hence, urea might lead to a decrease (by water to avoid a major concentration difference between the lumen of addition) or increase (by NaC1 addition) in the nonurea osmo- the IMCD and the corresponding interstitial fluid (Table 1) lality of the inner medulla in water-deprived humans. Control of Urine Flow Rate 883

papillary samples could suggest that there is a limitation for C I diffusion of urea and/or a technical problem. In the latter I 600 I. I context, when the papilla is resected, the tissue sample includes II 1 I I .1 tissue from virtually the entire papilla so one has an average . I value for a span of papillary interstitial fluid. Moreover, if J I water did achieve osmotic equilibrium, the osmolalities in the S I I I terminal IMCD and the papillary tissue should always be I ..‘ 400 I equal. Thus, the finding of a somewhat higher osmolality in the I I I terminal IMCD than in the papillary fluid osmolality suggests

0 I that a sampling problem was present. In this context, the E I I somewhat higher luminal urea concentration could also repre- 0 sent this sampling problem, as well as a somewhat less-than- I - 200 complete equilibration for urea diffusion. It is of interest that there was no difference in the nonurea osmolality in these fluids, a fact that could represent the small variation in nonurea 0 z osmolality throughout the inner medulla (6-8). Overall, the

0 -, nonurea osmolality of urine seems to provide a reasonable 0 200 400 600 800 estimate of the effective osmolality of the interstitial fluid in A Urea excretion rate (sosm/min) the inner medulla during water deprivation. It is also of interest to consider what kinetics of the urea transport system could permit a high luminal (and interstitial) .‘. 600 urea concentration in the inner medulla and not allow the rate of urea reabsorption to be limited. Two possibilities come to mind: first, if the number of and/or maximum velocity of the urea carriers would have to be very large; and second, if the affinity of its carrier for urea was low, this carrier could still

400 II I II have a high transport rate if the concentration of urea rose in the luminal fluid. Indeed, the KT for this transporter was in the

III I I 800-mM range (22, 23). ) I1I I4III The impermeable solutes in the lumen of the IMCD and the x II I 0) I interstitial fluid are not identical. The major nonurea solutes in II Q) interstitial fluid are Na and Cl; other solutes that contribute in 200 I #{149} I I an important way to this nonurea osmolality are K (24) and I’ VI) I I NH4 salts, the latter being more important in chronic meta- 0 1% bolic acidosis (25, 26). In the urine, the nonurea osmoles depend on the physiologic condition. For example, with a low-salt diet, the urine may not contain much Na and Cl (27);

0 , . , . , . K and NH4 salts (Table 2) could constitute the excreted 0 0.4 0.8 1.2 1.6 nonurea osmoles, depending on the need to excrete either of B Urine flow rate (mi/mm) these major cations. Nevertheless, from an osmotic point of view, there would be little difference in urine volume during Figure 1. Effect of the rate of excretion of urea on the nonurea water deprivation if the excreted nonurea osmoles were Na and osmolality in water-deprived subjects. Each dot represents a single Cl or NH4 and -HB, providing that their excretion rates subject on their normal diet. The nonurea osmolality is the urine were equal (Table 2). osmolality minus the urine urea in millimolar terms. The top portion of the figure indicates that there is a significant influence of the rate The data presented in Figure 1, Table 2, and Table 3 can be of excretion of urea on the nonurea osmolality. (r = 0.275, P < summarized as follows. The minimum urine flow rate and the 0.01). The bottom portion of the figure indicates that the urine flow nonurea osmolality of the urine were not different when the rate was directly proportional to the nonurea osmole excretion rate rate of excretion of urea was at its usual or a lower rate (Table (r2 0.691, P < 0.01). For details, see Materials and Methods. 2). Rather, all that changed was the urine osmolality and urea concentration, factors that do not directly regulate the urine flow rate. When the rate of excretion of urea was increased to Central to the interpretation of our data is the fact that urea >500 p.mol/min in water-deprived subjects, there was a small is a permeable solute when ADH acts (reviewed in references rise in the concentration of urea in the urine, a fall in its 2 and 3). Data to support this interpretation would be similar nonurea osmolality (and presumably the inner medullary in- concentrations of urea in the luminal fluid exiting the IMCD terstitial fluid nonurea osmolality), and a rise in the nonurea and papillary interstitial fluid (Table 1 ) (6-8). The failure to osmole excretion rate (Table 3). If urea is not an effective find true equilibration for urea between the urine and the osmole in the papillary tip, why did the urine flow rate rise 884 Journal of the American Society of Nephrology

Table 3. Effect of a urea load on the urine flow rate during antidiuresis

Urea Load

No Yes

Plasma urea mM 5.4 ± 0.6 9.9 ± 0.4” Urine flow rate ml/min 0.6 ± 0.1 1.1 ± 0.1” Concentrations osmolality mosmol/kg H,O 888 ± 40 845 ± 34 nonurea osmolality mosmol/kg H,O 448 ± 45 3 1 1 ± 34

urea mM 439 ± 49 534 ± 21b

Na mM 131 ± 32 102 ± 24b K mM 99±16 77±15 Cl mM 190 ± 17 142 ± 17” Excretion rates

osmoles p.osmol/min 5 1 8 ± 91 902 ± 107” nonurea osmoles p.osmol/min 261 ± 57 337 ± 60”

urea p.mol/min 257 ± 52 565 ± 57b creatinine p.mol/min 11.2± 1.1 11.0± 1.2 Na p.mol/min 85 ± 31 1 15 ± 33” K p.mol/min 52 ± 8 75 ± 8h Cl p.mollmin 115 ± 26 155 ± 29”

a Six subjects consumed urea at I 0:00 h. A timed urine sample was obtained before and 150 to 250 mm later. Results are presented as the mean ± SEM for six subjects on their usual diet. h p < 0.01 for paired observations.

OIL AlL AVR IMCD

Requires NaCI H20 high [Urea]

NaCI ‘I”

‘4 I ,__ --

.- : H20 I Critical H 20 step I I Urea I ,.___.1

#{174}= Actions of ADH

Figure 2. Events in the inner medulla: role of urea. The nonurea osmolality of interstitial fluid in the inner medulla is influenced by urea. This osmolality rises because NaCl without water is reabsorbed from the ascending thin limb (ATL) of the loop of Henle (point I). The interstitial INaClI falls due to dilution when more electrolyte-free water is reabsorbed from the inner medullary collecting duct (IMCD). One can classify this electrolyte-free water reabsorption in two categories: solute-free water (point 2) and water reabsorbed along with urea from the terminal IMCD. This latter fluid has the same total osmolality as the corresponding interstitial fluid in the inner medulla (point 3), but is still electrolyte-free. When more urea is delivered to the IMCD. the concentration of nonurea osmoles in its luminal fluid must be lower (same total osmolality). This should augment the reabsorption of both solute-free water (point 2) and a solution of urea that is isosmolal to fluid in the medullary interstitium (point 3). The net effect is a lower nonurea osmolality of interstitial fluid (Table 3). AVR, ascending vasa recta.

when urea was ingested (Table 3)? Urea could increase the the result of events in the outer medulla where urea is an urine flow rate by leading to a decrease in the nonurea osmo- impermeable solute (3) (Figure 2). At any given horizontal lality in the inner medullary interstitial fluid. Urea does this by plane in the outer medulla, because the total osmolality of increasing the delivery of electrolyte-free water to the IMCD, luminal and interstitial fluids are equal (ADH action), a higher Control of Urine Flow Rate 885 luminal urea concentration will result in a lower luminal elec- 24-h period. Had urea been produced rapidly, it would have trolyte concentration (same total osmolality). been excreted in a matter of hours and obligated a large urine flow rate (Table 3). Physiologic Implications of the Analysis of Water What Would Happen to Water Conservation if NH4 Conservation in Nonurea Terms and j3-HB Were Permeable Solutes in the Inner Medulla?. Four aspects of water conservation will be considered in this A deficit of water would cause death long before a caloric section. The discussion of these points will illustrate the effects problem in subjects deprived of food and water (33, 34). of urea on the number of effective osmoles excreted and on Nevertheless, having a small urine volume is advantageous for their concentration in the fluid in the inner medullary intersti- survival because this minimizes the possibility of renal calculi, tium. hyperkalemia due to failure to excrete K from tissue catabo- Effect of Urea on Water Conservation if Urea Was Not a lism, and a very large retention of urea (an osmotic diuresis Permeable Solute in the IMCD. If urea was not permeable upon refeeding). To ensure this minimum urine output, one in the IMCD, it, like all nonpermeable solutes, would obligate must either have a low interstitial nonurea osmolality (a prob- the excretion of water, because for any given effective urine lem for water conservation if impermeable solutes needed to be osmolality, the urine volume is directly proportional to the excreted) and/or a guaranteed rate of excretion of nonurea number of effective osmoles excreted (Equation I , Figure 1 ). If osmoles. In this latter regard, NH4 and -HB function as impermeable in the IMCD, there would be little addition of nonurea osmoles in the IMCD, and their excretion is a neces- urea into the interstitial fluid in the inner medulla. Moreover, sary accompaniment ofchronic fasting (35). Accordingly, hay- water will still be reabsorbed from inner medullary nephron segments, because water is permeable and the osmolality in the ing these impermeant nonurea urinary osmoles permits a small interstitial fluid rises progressively from the cortex to the (and essential) physiologic urine volume during chronic, total papillary tip (6, 7). This reabsorbed water must exit from the caloric deprivation. Had NH4 and -HB been permeable inner medulla via the ascending vasa recta, making its flow rate solutes in the IMCD, they, like urea, would not obligate the greater than in the descending vasa recta. Even with perfect excretion of water by a luminal effect. Nevertheless, they equilibrium between the descending and ascending vasa recta would have increased the delivery of water from the cortical (equal urea concentrations in both limbs), there will be net collecting duct to the IMCD, where, when reabsorbed, they removal of urea from the inner medulla via the vasa recta; the would have led to a lower effective osmolality in the interstitial higher the concentration of urea in interstitial fluid, and thereby fluid (Figure 2). Thus, there would be a very low urine volume in both limbs of the vasa recta, the greater the quantity of urea and a very low interstitial effective osmolality. removed from the inner medulla. To summarize, the combina- Minimizing the Urine Volume Overnight. To have the tion of more effective osmoles being excreted (urea plus nonu- minimum urine volume overnight, there must be a high inner- rea osmoles) and a lower interstitial total osmolality (due to a medullary interstitial nonurea osmolality and a low rate of low interstitial urea concentration) should lead to a much excretion of nonurea osmoles (29, 36). The former is brought higher minimum urine volume (Equation 1). Hence, to the about by ADH actions together with a rate of excretion of urea extent that urea fails to achieve equal concentrations in the that is not high, as discussed above. The second component lumen of the IMCD and the interstitial fluid, urea will com- contributing to the low urine flow rate is the low rate of promise the ability to minimize water excretion during water excretion of Na and Cl at this time, the usual diurnal variation deprivation. in excretion of these electrolytes (36). Avoiding a High Rate of Urea Excretion After a Protein- Rich Meal. To have the most efficient water conservation during water deprivation, one must avoid a urea excretion rate Conclusion that exceeds 500 p.mol/min (Table 3, Figure 1 ). Urea is the Because urea is permeable in the IMCD under the influence major nitrogenous, water-soluble end product of protein me- of ADH (3, 37, 38), one should examine the urine nonurea tabolism (28). Because most subjects eat the majority of their osmolality and the rate of excretion of nonurea osmoles to protein in a relatively small proportion of the 24-h period, one understand the factors influencing the minimum urine volume might anticipate that the rate of urea production and excretion when ADH acts (Equation I ). The data in Figure 1 and Table would be much higher after that meal. Nevertheless, when a 2 are consistent with the hypothesis that low versus typical usual mixed meal is ingested, there is little diurnal variation in the rate of excretion of urea (29), a point implicit in the concentrations of urea in the urine do not have a major influ- traditional use of the blood urea nitrogen to reflect the GFR. ence on its minimum volume or nonurea osmolality; the data in Moreover, because hepatocytes extract the majority of amino Table 3 suggest that a somewhat larger delivery of urea will acids delivered via the portal vein in a single pass (30-32) and cause a rise in urine flow rate associated with a decrease in its because their conversion to glucose or CO2 must occur slowly nonurea osmolality and a rise in nonurea osmole excretion rate. due to hepatic energy considerations (28), Cheema-Dhadli and Hence, urea seems to permit its own concentration to rise in the Halperin (29) speculated that postprandial protein synthesis interstitial fluid and the final urine without influencing the occurs rapidly in the liver, whereas hydrolysis of these proteins nonurea osmolality in the inner medulla in humans unless its occurs more slowly and at a relatively constant rate over the excretion rate is >500 p.mol/min. 886 Journal of the American Society of Nephrology

Acknowledgments 18. Schmidt-Nielsen B: Function of the pelvis. In: Urinary Concen- We are extremely grateful to Drs. K. S. Kamel, M. Gowrishankar, trating Mechanisms. 2omparative Physiology, edited by Kinne and M. Oh for critique and helpful discussions. We also thank Ching RKH, Basel, Switzerland, Karger, 1990, pp 103-140 Bun-Chen, Stella Tang, and Ellie Singer for expert technical assis- 19. Oh M, Halperin ML: The mechanism of urine concentration in tance and Jolly Mangat for secretarial assistance. the inner medulla. Nephron 75: 383-394, 1997 20. Stephenson JL: Central core model of the renal counterfiow References system. Kidney mi 2: 85-94, 1972 2 1 . Kokko JP, Rector FCJ: Countercurrent multiplication system 1. Robertson GL: Regulation of vasopressin secretion. In: Clinical Disturbances of Water Metabolism, edited by Seldin DW, Gie- without active transport in the inner medulla. Kidney 1w’ 2: bisch OH, New York, Raven Press, 1993, pp 99-1 18 214-223, 1972 2. Jamison RL, Roy DR. Layton HE: Countercurrent mechanisms 22. Chou C-L, Sands JM, Nonoguchi H, Knepper MA: Urea gradi- and its regulation. In: Clinical Disturbances of Water Metabo- em-associated fluid absorption with Surea I in rat terminal lisin, edited by Seldin DW, Giebisch OH, New York, Raven collecting duct. Am J Phvsiol 258: Fl 173-Fl 180, 1990 Press, 1993, pp 1 19-156 23. Chou C-L, Sands JM, Nonoguchi H, Knepper MA: Concentra- 3. Knepper MA, Rector FCJ: Urinary concentration and dilution. tion dependence of urea and thiourea transport in rat inner In: Brenner and Rector’s, The Kidney, 5th Ed., edited by Brenner medullary collecting duct. Am J Physiol 258: F486-F494, 1990 BM, Philadelphia, W. B. Saunders, 1996, 532-570 24. Jamison RL: Potassium recycling. Kidney hit 31: 695-703, 1987 4. Bankir L: Urea and the kidney. In: Brenner and Rector’s, The 25. DuBose TD Jr: Ammonium transport in the kidney: New phys- Kidney, 5th Ed., edited by Brenner BM, Philadelphia, W. B. iological concepts and their clinical implications. J Am Soc Saunders, 1996, 571-606 Nephrol 1: 1193-1203, 1991 5. Teitelbaum I, Berl T, Kleeman CR: The physiology of the renal 26. Buerkert J, Martin D, Trigg D: Ammonium handling by super- concentrating and diluting mechanisms. In: Clinical Disorders of ficial and juxtamedullary nephrons in the rat. J C/in Invest 70: Fluid and Electrolyte Disorders, 4th Ed., edited by Maxwell MH, 1-12, 1982 Kleeman CR, Narins RG, New York, McGraw Hill, 1987, 79-103 27. Schrier RW: Body fluid volume regulation in health and disease: 6. Schmidt-Nielsen B: Urea excretion in mammals. Phvsiol Rev 38: A unifying hypothesis. Ann Intern Med I 13: 155-159, 1990 139-168, 1958 28. Jungas RL, Halperin ML, Brosnan iT: Lessons learnt from a 7. Ullrich KJ, Rumrich 0, Schmidt-Nielsen B: Urea transport in the quantitative analysis of amino acid oxidation and related glu- collecting duct of rats on normal and low protein diet. Pflugers coneogenesis in man. Physiol Rev 72: 419-448, 1992 Arch 295: 147-156, 1967 29. Cheema-Dhadli S. Halperin ML: Diurnal excretion of nitrogen 8. Levitin H, Goodman A, Pigeon 0, Epstein FH: Composition of and sulphur after meals containing protein: Indications for post- the renal medulla during water diuresis. J C/in Invest 41: 1145- prandial synthesis of proteins. (‘au J Physiol Pharinacol 7 1: 1151, 1962 120-129, 1993 9. Nielsen 5, Terris J, Smith CP, Heddiger MA, Ecelbarger CA, 30. Feigin RD. Klainer AS, Beisal WR: Circadian periodicity of Knepper MA: Cellular and subcellular localization of the vaso- blood amino-acids in adult men. Nature 215: 512-514, 1967 pressin-regulated urea transporter in the kidney. Proc Nail Acad 3 1 . Frame EG: The levels of individual free amino acids in the Sci USA 93: 5495-5500, 1996 plasma of normal man at various intervals after a high-protein 10. Gamble JL, McKhann CF. Butler AM, Tuthill E: An economy of meal. J C/in Invest 37: 17 10- 1723, 1958 water in renal function referable to urea. Am J Physiol 109: 32. Rerat A, Simoes-Nunes C, Mendy F, Vaissade P. Vaugelade P: 139-154, 1934 Splanchnic fluxes of amino acids after duodenal infusion of I 1. Berliner RW, Levinsky NO, Davidson DO, Eden M: Dilution carbohydrate solutions containing free amino acids or oligopep- and concentration of the urine and the action of antidiuretic tides in the non-anaesthetized pig. Br J Nutr 68: 1 1 1-138, 1992 hormone. Am J Med 27: 730-744, 1958 33. McCance RA: The excretion of urea, salts and water during periods 12. Halperin ML, Gowrishankar M, Mallie JP, Sonnenberg H, Oh M: of hydropaenia in man. J Physiol (Land) 104: 196-209, 1945 Urea recycling, an aid to the excretion of potassium during 34. Owen OE, Morgan AP, Kemp HG, Sullivan JM, Herrera MG, antidiuresis. Nephron 72: 507-5 1 1, I 996 Cahill GFJ: Brain metabolism during fasting. J C/in Invest 46: 13. Rabinowitz L: Accumulation of organic nonelectrolytes in the 1589-1595, 1967 renal medulla: Mechanism and consequences. Kidney hit 17: 323-332, 1980 35. Kamel KS, Stinebaugh BJ, Schloeder FX, Halperin ML: Kidney 14. Vander AJ: Urine concentration: The medullary countercurrent in starvation. In: The Kidney, Physiology and Pathophysiology, system. In: Renal Physiology, 4th Ed., edited by Vander AJ, New 2nd Ed., edited by Seldin DW, Giebisch 0, New York, Raven York, McGraw-Hill International Editions, 1994, pp 99-1 1 1 Press, 1992, pp 3457-3470 15. Sonnenberg H, Wilson DR: Urea handling by the medullary 36. Moore-Ede MC, Herd JA: Renal electrolyte circadian rhythms: collecting duct of the rat kidney during hydropenia and urea Independence from feeding and activity patterns. Am J Phvsiol infusion. PJlUgersArch 390: 131-137, 1981 232: F128-F135, 1977 16. Lee J, Williams P0: Changes of sodium and urea concentrations 37. You 0, Smith CP, Kanai Y, Lee WS, Stelzner M, Hediger MA: in the renal papillary interstitial fluid on dehydration of rats. Expression cloning and characterization of the vasopressin-reg- J Physiol (Lond) 218: 195-204, 1971 ulated urea transporter. Nature 365: 844-847, 1993 17. Halperin ML, Vinay P, Gougoux A, Pichette C, Jungas RL: 38. Smith CP, Lee W-S, Martial 5, Knepper MA, You 0, Sands JM, Regulation of the maximum rate of renal ammoniagenesis in the Hediger MA: Cloning and regulation of expression of the rat dog with chronic metabolic acidosis: A quantitative analysis. kidney urea transporter (rUT2). J C/in Invest 96: 1556-1563, Aiii J Physiol 248: F607-F615, 1985 1995