A micropuncture study of Henle s thin loop in Brattleboro

Transcription

1 AMERICAN JOURNAL OF PHYSIOLOGY Vol. 224, No. 1, January Printed in U.S.A. A micropuncture study of Henle s thin loop in Brattleboro rats REX L. JAMISON, JOHN BUERKERT, AND FRANK LACY Departments of Medicine and of Physiology and Biophysics, Washington University School of Medicine and the Jewish Hospital of St. Louis, St. Louis, Missouri 63110; and Division of Nephrology, Department of Medicine, Stanford University, Stanford, California JAMISON, REX L., JOHN BUERKERT, AND FRANK LACY. A micropuncture study of Uenle s thin loop in Brattleboro rats. Am. J. Physiol. 224 (1): Micropuncture was performed on rats with hereditary diabetes insipidus to study changes between water diuresis and vasopressin-induced antidiuresis. U/P osmolality increased from 0.53 t 0.02 SE in water diuresis to 4.23 =t 0.22 in antidiuresis. Measured in the descending limb, juxtamedullary SGFR did not change significantly; TF/P inulin was 3.28 & 0.15 in water diuresis and 5.06 zt 0.47 in antidiuresis (P <.Ol); TF/P osm was 1.96 zt 0.22 and 3.34 =k 0.44 (P <.Ol); TF/PNB, 1.76 & 0.18 and 2.24 zt 0.28 (P >.05); TF/PK, 2.43 =k 0.17 and 4.12 =t 0.75 (P <.05); and the concentration of nonelectrolyte was 95 rt 16 mosm and 358 =k 58 (P <.Ol). Superficial single-nephron glomerular filtration rate and fractional reabsorption by the proximal tubule of the superficial nephron did not change. Assuming that the descending limb was impermeable to sodium, it was calculated that, in antidiuresis, the increase in osmolality of fluid at the end of the descending limb was accounted for in equal measure by water removal and nonelectrolyte solute addition. urinary concentrating mechanism; descending limb of Henle s loop; water diuresis; antidiuresis IT IS NOW GENERALLY ACCEPTED that the thin loop of Henle acts as a countercurrent multiplier in the urinary concentrating mechanism (1, 5, 11, Z), and attention has turned to the molecular movements involved in loop multiplication. One aspect of this problem is the relative contribution of water reabsorption versus transtubular solute addition (secretion) to the increase in solute concentration in descending limb fluid. Urea must be secreted somewhere in the proximal juxtamedullary nephron of Psammomys and hamster, since the quantity of urea in fluid at the bend of Henle s loop exceeds the filtered load of urea (20, 21, 23, 25); the site of urea addition is very likely the descending limb rather than the proximal tubule. The same is probably true for the descending limb of the rat juxtamedullary nephron (24). However, Kokko ( 16) reported a urea reflection coefficient of 0.96 for the rabbit descending limb perfused in vitro, implying osmotic equilibration occurs 96 % by water abstraction and only 4 % by urea entry ( 16). A corresponding conflict regarding sodium exists between the data for the descending limb of Psammomys in viva (25), and of the rat in vitro (24), which strongly suggest the 180 net entry of sodium, and the in vitro studies of the rabbit descending limb, which yielded a reflection coefficient = 0.95 for NaCl (15). Each of the foregoing in vivo studies was performed only upon animals excreting a hypertonic urine, since it is very difficult to induce a sustained water diuresis in a normal rodent prepared for micropuncture. The purpose of the present experiments was to study the flow and composition of fluid in the thin descending limb, first in water diuresis and then in antidiuresis, to ascertain those changes in loop function associated with transformation from water diuresis to antidiuresis, that is, with the production of a hypertonic urine. To achieve this goal, we used rats with hereditary hypothalamic diabetes insipidus (Brattleboro strain (3 1, 32)). In this paper are reported for the first time the concentrations of solutes and the fractional and absolute flows of fluid at the bend of Henle s thin loop in water diuresis. Analysis of the results suggests that water removal and solute entry contribute equally to the concentrating process in the descending limb in the transformation to antidiuresis. METHODS Three groups of rats of either sex with hereditary hypothalamic diabetes insipidus (Brattleboro strain (3 1, 32)), weighing g, were studied by the micropuncture technique. The first group (15 rats) was used to examine simultaneously collecting duct and thin-loop function. The collecting duct data and loop osmolalities from this group of rats were the subject of a previous paper (13). After anesthesia with intraperitoneal Inactin, 80 mg kg-1 body wt, a tracheostomy was performed and polyethylene catheters were inserted into both jugular veins and into the left femoral artery. The left kidney was exposed through a midline abdominal incision, the left ureter was excised to uncover the tip of the papilla, and the kidney was placed gently in a glass cup and bathed with mineral oil at 38 C. The papilla was illuminated by a fiberoptic light guide. The cut end of the left ureter was ligated and a catheter was inserted in the bladder to collect urine from the right kidney. The animal received a prime and maintenance solution of inulin calculated to achieve a plasma concentration of 80 mg/ 100 ml. In the first period, Tyrode solution, diluted 1: 1 with distilled water, was infused at a rate between 0.08 and 0.1 ml

2 FUNCTION OF HENLE S THIN LOOP min-l to insure a brisk water diuresis. After at least 1 hr to permit equilibration, a loop was punctured at or near the hairpin turn and the tubule fluid collected, usually at less than the intratubular flow rate. At the end of the first (water diuresis) period, the infusate was changed to undiluted Tyrode solution containing inulin and antidiuretic hormone (Pitressin, Parke, Davis & Company, Detroit), 2.2 mu ml-l. The infusion rate was set to equal exactly half the initial measured rate. Thus these animals received antidiuretic hormone at a rate ranging I from 88 to 110 $J min- l. After 1 hr a loop of Henle different from that in water diuresis was punctured and a sample of fluid obtained. Samples of blood (65 ~1) were obtained from the femoral catheter approximately every 30 min during puncture. The blood pressure was measured through the same catheter. One or more urine samples were collected from the right kidney during each period. The second group ( 10 rats) was used to study exclusively thin-loop function under conditions identical to those for group I. At least two loops were punctured in water diuresis and in antidiuresis. A drop of Kel-F oil, at least 5 tubule diameters in length, was injected into the loop and the tubule fluid was collected as the drop was held steady or as it moved slowly downstream, but not out of sight. The collection was timed, the volume of fluid measured, and the single-nephron glomerular filtration rate (SGFR) was calculated (14). Superficial nephron function was studied in a third group of eight rats under conditions identical to those for the first two groups. In four rats the recollection micropuncture technique (4) was used to compare fractional reabsorption of water by the superficial proximal tubule in water diuresis with that in antidiuresis. The end-accessible portion was identified by the injection of 0.03 ml of 5 % lissamine green in isotonic saline. No more than 0.75 ml of the dye solution was injected per rat. The fluid was collected at a rate deliberately less than the intratubular flow rate. In the other four rats the SGFR of the superficial nephron was determined by a technique reported previously (14). Four to six proximal tubules, selected at random, and one distal tubule, identified by the intravenous injection of lissamine green, were punctured during water diuresis; in antidiuresis four to six other proximal tubules were punctured at random while the same distal tubule was repunctured. The SGFR determined in the distal puncture was used as a check on the proximal SGFR ( 14) : there was no significant difference between mean proximal SGFR and mean distal SGFR. The manufacture and treatment of the pipettes and the analytic methods have been described previously (9). Nonelectrolyte solute was defined as equal to the total osmolality minus (Na+ X K+ X 2). (The use of two for the osmotic coefficient of K+ instead of its precise value, l , introduces an error of no more than 6 mosm-and usually less than 3 mosm- into the calculations, which is within the limits of the analytic error.) Standard statistical methods were used (27). RESULTS Whole kidney. The mean arterial blood pressure and function of the right kidney are summarized in Table 1. The TABLE I. Comparison of kidney weights, blood pressure, and clearance data among groups of rats Groupa : Number of rats: Kidney weight, mg: GFR, ~1 min-l c V, ~1 min-l d I and II AZ 8.7 WD AD WD AD III 371 rl Al.8 h2.0 dz2.9 ~~ *21 302*19k 435zt23 350*303' ztl.3 *0. lk AZ.6 *0.4k u;k,,e zto.02 &O.ZZk zto.04 rt0.24.k (l/(ww) x loof o zto.6 4x0. lk ho.6 &O. lk (U/P(ml,In>) x lo@ dxo.3 ho.2 zko.2 zto.2 (U/PtNa/In)) X looh dzo.2 *o.oj zto.1 zto.2 Values are means * SE. a For definition of groups I, II, III, See METHODS. bwd = water diuresis; AD = antidiuresis. c Glomerular filtration rate. d Urinary flow rate. e Urineto-plasma osmolality. f Fractional excretion of water in percent. g Fractional excretion of solute in percent. h Fractional excretion of sodium in percent. i Value in AD compared to that in WD. j P <.05. k P <.OOl. mean values for group I did not differ significantly from those for group 11 (except that fractional sodium excretion was significantly less in antidiuresis in group II, but not in group 1); the data were therefore combined. The kidney weights of the rats used for group III were slightly greater, so the data are presented separately. There was a striking decrease in urinary flow rate and fractional excretion of water and a marked increase in the urinary osmolality in antidiuresis. The mean arterial blood pressure and the fractional excretion of total solute remained unchanged. The glomerular filtration rate declined 15 %. Micro/wzcture. In Table 2 is presented a summary of the micropuncture data obtained in groups I and II. There was accordance for each parameter in the two conditions between groups I and II except that the total osmolality, sodium, and nonelectrolyte concentrations were slightly but not significantly higher in group II. The experiments in group I were focused upon collecting tubule function (13); only one loop was punctured during each period. Technical problems prevented the successful determination of loop fluid potassium in 6 of the 15 animals; calculation of nonelectrolyte solute was thus limited to the other 9. The loop collections in the group 11 animals were considered superior to those in group I because there was more time to do them, two or more were done per period, and the operator had the advantage of the previous experience gained in the group I experiments. For these reasons we elected to present the loop data in group I separately from group 11 and analyze the latter in greater detail (see below). However, we found no significant differences in the results whether the two 181

3 182 R. L. JAMISON, J. BUERKERT, AND F. LACY groups were analyzed separately or together, except that the difference between loop TF/PN& in antidiuresis and that in water diuresis achieved statistical significance in the combined group. The fraction of filtered water remaining unreabsorbed at the end of the descending limb was approximately 30 % in water diuresis. The osmolality of the loop fluid was persistently hypertonic even though, as shown in a previous study (13), the collecting duct fluid was significantly hypotonic. In antidiuresis TF/P osmolality of loop fluid was more than 50 % higher than in water diuresis (Table 2). The fraction of filtered water remaining unreabsorbed at the end of the descending limb declined to approximately 20 % with a commensurate decline in tubular fluid flow rate. Juxtamedullary SGFR determinations again displayed the scatter noted in previous SGFR measurements (10, 14) There was a decrease in the mean value from 62.1 =t 7.01 nl min-l to 50.9 zt 4.72 nl min-l g kidney wt-l (or, in nl min-l, from 20.4 A 2.5 to 16.9 zt 1.4, respectively), but the decline was not statistically significant. (See Table 3 for individual experiments.) In group 111 only the superficial nephron was studied. In four rats the mean TF/P inulin in fluid from the end-accessible proximal tubule was 2.28 & 0.08 in water diuresis and TABLE 2. Micropuncture of Henle s loop Grou#: na: I 15 II 10 TF/PI,c V, nl min-ld SGFR, Tmmm TF/PN, TF/PK Nonelectrolyte, nl min-le Conditionb O/(TF/Rn) x lo@ ~TF/hosm/~n)) mosmf X looh CJWP(N~/I~)) X 100 WD AD WD AD 3.50 It =t 0.7gk m 3.28 & zt rt O.lln 1.51 It * It zt & * zt & & zt * AZ III rt zt zt zt St * s+: zt 0.47km 3.42 rt & & zt zt A zt * 5.0 (TF/P(K/I,)) x 1ooj 83 zt zk zt 4 98 zk 17 - Values are means & SE. B Number of rats. bwd = water diuresis; AD = antidiuresis. c Tubule fluid-to-plasma inulin ratio. d V, tubule fluid flow. e Single-nephron glomerular filtration rate. f Nonelectrolyte solute = total osmolality minus (NaS X K+ X 2), where Na+ and KS- are the concentrations of sodium and potassium, respectively, of the loop fluid. g Frac= tion of filtered water remaining in percent. h Fraction of filtered solute remaining in percent. i Fraction of filtered sodium rem maining in percent. j Fraction of filtered potassium remaining in percent. k Value in AD compared to that in WD. l P <.05, In P <.Ol. n P <.OOl. TABLE 3. Group 11 rats: micropuncture of Henle s loop, individual experiments - Exp No, WDe ADe TJV Rn Vb SGFRC TW osm TF/PN~ TWPK. WD AD WD AD AD WD AD WD AD WD AD WD AD I Ol , Mean values are presented. a Number of loops punctured. b Tubule fluid flow rate in nl min? c Single-nephron glomeruiar filtration rate in nl min-l. d Nonelectrolyte solute = total osmolality minus (Na+ X K+ X Z), where Na+ and K+ are the, concentrations of sodium and potassium, respectively, of the loop fluid (in mosm). WD = water diuresis; AD = antidiuresis,

4 FUNCTION OF HENLEY3 THIN LOOP 2.31 & 0.08 in antidiuresis (n = 17). In four other rats the mean superficial SGFR was 29.7 =t 2.3 nl min-l g kidney wt-l (or in absolute units 11.2 & 0.7 nl min-l) (n = 23) in water diuresis and 35.4 XI= 1.6 nl min-l g kidney wt-l (or 13.6 ZIZ 1.3 nl min-l) in antidiuresis (P >.l) (n = 21). DISCUSSION, It is apparent that the animals sustained a brisk water diuresis in period I and a marked antidiuresis in period 2. Although the experiment was designed to insure a constant solute delivery in the two states, the actual experiments fell somewhat short of this objective. The GFR declined 15 % in antidiuresis, for which we have no ready explanation. However, for reasons given in a previous paper (13), it is very unlikely that the decline in GFR by itself could account for the marked changes in urinary osmolality and fractional excretion of water which occurred. Fractional sodium excretion fell from 0.5 % in water diuresis to 0.1 % in antidiuresis (P <.05) in groups I and 11 but not in group 111. That a change in sodium delivered to the descending limb was a major factor in the observed changes in loop fluid seems unlikely for two reasons. First, analyzed separately, the decrease in fractional sodium excretion was significant in group II but not in group 1, yet the changes in loop fluid were similar in both groups Secondly, we have previously demonstrated in antidiuretic rats of comparable size that to effect changes in loop fluid composition similar to those reported here by changing urinary sodium excretion required a profound natriuresis (fractional sodium excretion = 6 % ( 14)). There was no significant change between water diuresis and antidiuresis in either SGFR of the superficial nephron or in fractional reabsorption by the superficial proximal tubule, as noted by others (2, 26). The principal changes occurred in the juxtamedullary nephron. The mean juxtamedullary SGFR was 62.1 & 7.0 nl min-l g kidney wt-l in water diuresis, and 50.9 Ifi 4.7 nl min-l g kidney wt-l in antidiuresis, but the decrease was not statistically significant. These values are similar to those reported previously in the nondiuretic rat (8, 10, 14, 29). In contrast, an increase in the juxtamedullary SGFR was reported by Davis and Schnermann using Hanssen s technique (3). The present work shows that a significant increase in juxtamedullary SGFR need not occur in antidiuresis, and thus is not essential to the mechanism by which the renal medulla becomes more hypertonic in antidiuresis. The papilla was hypertonic in water diuresis, as evidenced by the osmolality of the loop fluid at the hairpin turn, which is in osmotic equilibrium with that of the medullary interstitium (12). The concentrations of all constituents of loop fluid increased in antidiuresis, in particular that of nonelectrolyte solute. The concentrating process in the descending limb was first assumed to be exclusively water efflux (7) and then exclusively solute entry (19) in early models. Gottschalk and co-workers (6, 20) reported the mean TF/P inulin = 1 l-l 2 at the bend of the thin loop in nondiuretic hamsters, suggesting water removal as the principal if not exclusive means by which loop fluid constituents are concentrated. Since the TF/P inulin ratio at the end of the proximal tubule of the iuxtamedullarv nephron is unknown, however, it is difficult to assess the relative contributions of water removal and solute addition to the increased solute concentration in fluid in the descending limb. To get around this difficulty, de Rouffignac and Morel (25) plotted TF/P solute and TF/P sodium as functions of TF/P inulin in their studies of the long descending limbs of Psammomys, and interpreted the results as demonstrating that net sodium chloride entry accounted for 85 % of the increase in total osmolality in loop fluid. Critical to that interpretation was the estimation by extrapolation of a TF/P inulin = 4 at the beginning of the descending limb. Kokko ( 15) studied the descending limb of the rabbit nephron removed by dissection from the midmedullary region and perfused in vitro. From the results of adding sodium chloride to the bathing medium, he calculated a reflection coefficient for NaCl of Recently ( 16) he reported a similarly high reflection coefficient for urea, These findings demonstrate that the constituents of fluid in the isolated, perfused descending limb are concentrated almost entirely by water extraction. Based on these results Kokko and Rector ( 17) have proposed a new model for countercurrent multiplication in which Henle s thin limbs operate as purely passive equilibrating segments. However, the extent to which these in vitro observations hold for the descending limb in the inner medulla in vivo is uncertain. The composition of the inner medullary interstitium may be different from that of the bathing media employed in the in vitro experiments. Furthermore, the descending limbs in the inner medulla are supplied only by juxtamedullary nephrons and Kriz et al. ( 18) have reported significant morphological differences between the epithelium of the descending limb of the juxtamedullary nephron and that of the superficial descending limb in the rat. It should be noted that the concentrating process need not be exclusively water removal or solute entry. Recently Stephenson (28) has constructed a model in which Henle s ascending and descending limbs and collecting ducts exchange with a central vascular core of vasa recta. Osmolality of the core is increased by urea reabsorbed from collecting ducts and salt reabsorbed from ascending limbs. Concentration in descending limbs can occur by osmotic extraction of water, by solute entry, or by a combination of the two processes. The relatively low TF/P sodium ratios in the loop fluid in the present study, particularly in antidiuresis, suggested that transtubular entry of sodium into the descending limb was minimal. Assuming, as a first approximation, that the descending limb is impermeable to sodium and that the sodium concentration of fluid at the beginning of the descending limb is the same as that of systemic plasma, permits the computation of the TF/P inulin ratios at the beginning of the descending limb? The calculation was performed for the group 11 data and yielded a TF/P inulin of 2.01 zt 0.19 in water diuresis and 2.43 rfi 0.22 in antidiuresis. Although these ratios for the TF/P inulin at the end of the proximal tubule of the juxtamedullary nephron seem low, particularly for water diuresis, when compared to the TF/P inulin ratio of 2.2 in the superficial proximal tubule at a puncture site l The TF/P inulin at the beginning of the descending limb was calculated by dividing the TF/P inulin of each loop collection by the TF/PN~ in the same specimen. 183

5 184 R. L. JAMISON, J. BUERKERT, AND F. LACY approximately two-thirds of its total length beyond the glomerulus, they cannot be discarded simply on that basis. There is no a priori reason to suppose fractional reabsorption by the proximal tubule of the two kinds of nephrons is the same, since the filtration rate of the juxtamedullary nephron is higher than that of the superficial nephron. The higher TF/P inulin in the juxtamedullary proximal tubule in antidiuresis than in water diuresis has at least two possible explanations. It may truly reflect a higher fractional reabsorption of water in the juxtamedullary proximal tubule during an tidiuresis. This does not necessarily indicate an effect of antidiuretic hormone (ADH) on the juxtamedullary proximal tubule : those factors (unknown) that contributed to the lower whole-kidney GFR may have played a role. Secondly, there may actually be sodium reabsorption by the descending limb in antidiuresis (which would lead to an overestimation of the TF/P inulin at the beginning of the descending limb). Marsh (2 1) has shown that the urea concentration of vasa recta plasma is significantly higher than that of fluid from the hairpin turn of the hamster. Since the contents of vasa recta and hairpin turn are in osmotic equilibrium ( 12), this implies that the sodium concentration of fluid in the descending limb is significantly higher than that of the vasa recta plasma and medullary interstitium- There being no electrical restraints (33), a concentration gradient for sodium due to the osmotic extraction of water from the descending limb by the high concentration of urea in the medullary interstitium in antidiuresis could drive sodium passively out of the descending limb (28). The calculated TF/P inulin at the beginning of the descending limb and the actual TF/P inulin at the end of the descending limb for each loop collection were used to calculate the contributions of water reabsorption and solute secretion to the final nonelectrolyte solute concentration, TF/P potassium, and TF/P osmolality in fluid at the end of the descending limb in group II. Results are presented in Table 4.2 Mean increment in TF/P,,, was 2.34 in antidiuresis ( = 688 mosm since mean P,,, = 294 mosm). Of this, 362 mosm was due to water removal along the descending limb and nearly all of the rest, 326 mosm, was due to transtubular entry of nonelectrolyte solute (which is 80 % urea in the papilla of vasopressin-treated Brattleboro rats (30, 3 1)). In short, water abstraction and solute addition contributed in equal amounts to the concentration process along the descending limb. Of course, the descending limb of the rat may not be impermeable to sodium (24). In that case some of the solute entry ascribed to nonelectrolyte solute may be instead accounted for by sodium, particularly in water diuresis. But if there is significant osmotic water extraction from the descending limb secondary to urea in the medullary interstitium, the concentration of sodium chloride would be higher in the luminal fluid than in the adjacent interstitium since salt and urea are the principal solutes in the medullary interstitium (2 1,30). Thus sodium would have to be entering the descending limb against a concentration gradient. On the other hand, an interpretation that solute entry contributes little to the increase in loop fluid solute concen- 2 Calculations for the combined group II and III data yielded similar results. TABLE 4. Relative contributions of water removal and solute addition to nonelectrolyte, potassium, and total solute concentrations of fluid at end of Henle s descending limb, assuming sodium impermeability of descending limb Water Antidiuresis Difference Condition diuresis Beginning Of Owing Owing Descen- to Water to Solute ding Reabsorption Addition Limb Nonelectrolyte solute, mosm Potassium, 11 & zt zt 3.6 >.05 TF/P, Water diuresis 1.0,,;~;.19,.6%&12~.4;H&.l7 An tidiuresis 1.0 Il.24 zt P7 68 rfi 16 <, zk 56 < =fi 59 <,005 End of Dey;zp 95 zt 16 <, zt 58 <.OOl 264 2t zt zt 0.75 <.05 <.OOl Difference 0.48 zt zt <.05 >.I <.05 Water Antidiuresis Difference diuresis Total solute, TF/P,,, 1.o 0.76 zt 0.19o.20 zt zt 0.22 <.025 <.OOl zt zt zt 0.44 <.OOl trations found in the present study has serious obstacles to overcome. If the solute concentrations in fluid at the hairpin turn are the result of water removal alone, the TF/P inulin at the beginning of the descending limb can be calculated by dividing the TF/P inulin found at the end of the descending limb by each respective TF/P solute ratio, since the TF/P solute ratio at the beginning of the descending limb (end of the proximal tubule) should be one or very close to one (25). In group II in water diuresis, the calculated TF/P inulin at the beginning of the descending limb would have to be 1.79 to explain the subsequent increase in total osmolality in fluid at the hairpin turn and 2.0 1, 1.41, and approximately 0.85 to account for the increase in sodium, potassium, and nonelectrolyte, respectively. In antidiuresis the respective TF/P inulin ratios necessary at the beginning of the descending limb to account for the solute concentra-

6 FUNCTION OF HENLE S THIN LOOP 185 tions in the fluid at the bend are, respectively, for total These studies were supported by National Institutes of Health osmolality, 2.43 for sodium, 1.46 for potassium, and 0.27 for Research Grant 2RO 1 HE , and by Jewish Hospital Restricted Fund for Micropuncture Studies no nonelectrolyte. In brief, the TF/P inulin required at the J. Buerkert was a Fellow of the Kidney Foundation of Metropolitan beginning of the descending limb differs markedly for each St. Louis. His current address is Armed Forces Radiobiology Research solute, and net fluid secretion by the nephron has to be Institute, Defense Atomic Support Agency, Bethesda, Md invoked to account for the high concentration of nonelec- R. L. Jamison is a recipient of Research Career Development Award 5 K04 HE42685 and is a Markle Scholar in Academic Medicine. His trolyte. present address, to which reprint requests should be sent, is Division In our view, therefore, the results suggest strongly that of Nephrology, Department of Medicine, Stanford University School water removal and solute entry each contribute signifi- of Medicine, Stanford, Calif cantly and approximately in equal measure to the total Abstracts of portions of this work have been published. (Jamison, solute concentration of loop fluid in the antidiuretic Brattle- R. L., J. Buerkert, and F. B. Lacy. Micropuncture of the renal papilla of rats with hereditary diabetes insipidus. J. Clin. Invest. 50 : 499, boro rat. Jarnison, R. L., J. Buerkert, and F. B. Lacy. Micropuncture of Henle s loop in rats with hereditary diabetes insipidus. Ann. il4eeting, Am. Sot. Nephrol.,.%h, Washington, D. C., 1971, p. 35.) We are greatly indebted to Daniel Marcus, Betty Henton, and Norman Frey for their able technical assistance. We thank Drs. Robert Berliner and John Stephenson for helpful criticism and suggestions. Received for publication 18 May REFERENCES 1. BERLINER, R. W., AND C. M. BENNETT. Concentration of urine in morphology of descending limbs of short and long loops of Henle the mammalian kidney. Am. J. Med. 42 : , in the rat kidney. In: International Symposium on Renal Handling of 2. DAVIS, B. B., F. G. KNOX, AND R. W. BERLINER. The effect of Sodium, Brestenberg, Switzerland, edited by F. Spinelli and H. Wirz. vasopressin on proximal tubule sodium reabsorption in the dog. Base1 : Karger, Am. J. Physiol. 212: , KUHN, W., AND A. RAMEL. Aktiver Salztransport als mijglicher 3. DAVIS, J. M., AND J. SCHNERMANN. The effect of antidiuretic hor- (und wahrscheinlicher) Einzelefl ekt bei dur Harnkonzentrierung mone on the distribution of nephron filtration rates in rats with in der Niere. Helv. Chim. Acta 42: , hereditary diabetes insipidus. Arch. Ges. Physiol. 330: , 20. LASSITER, W. E., C. W. GOTTSCHALK, AND M. MYLLE. Micro puncture study of urea transport in rat renal medulla. Am. J. 4. DIRKS, J. H., W. J. CIRKSENA, AND R. W. BERLINER. The effect Physiol : , of saline infusion on sodium reabsorption by the proximal tubule 21. MARSII, D. J. Solute and water flows in thin limbs of Henle s loop of the dog. J. Clin. best. 44 : 1160-l 170, in the hamster kidney. Am. J. Physiol. 218: , GOTTSCHALK, C. W. Osmotic concentration and dilution of the 22. MARSH, D. J. Osmotic concentration and dilution of the urine. urine. Am. J. Med. 36: In: The Kidney Morphology, Biochemistry, Physiology, edited by C. 6. GOTTSCHALK, C. W., W. E. LASSITER, M. MYLLE, K. J. ULLRICH, Rouiller and A. F. Muller. New York : Academic, 1971, vol. III, p. B. SCHMIDT-NIELSEN, R. O DELL, AND G. PEHLING. Micropunc ture study of composition of loop of Henle fluid in desert rodents. 23. MOREL, F., AND C. DE ROUFFIGNAC. Micropuncture study of urea Am. J. Physiol. 204: , medullary recycling in desert rodents. In: Urea and the Kidney, 7. HARGITAY, G., AND W. KUHN. Das Multiplikationsprinzip als edited by B. Schmidt-Nielsen. 1970, p Grundlage der Harnkonzentrierung in der Niere. 2. Elektrochem. 24. MORGAN, T., AND R. W. BERLINER. Permeability of the loop of 55 : , Henle, vasa recta, and collecting duct to water, urea, and sodium. 8. HORSTER, M., AND K. THURAU. Micropuncture studies on the Am. J. Physiol. 215: , filtration rate of single superficial and juxtamedullary glomeruli 25. ROUFFIGNAC, C. DE, AND F. MOREL. Micropuncture study of in the rat kidney. Arch. Ges. Physiol. 301: , water, electrolytes, and urea movements along the loops of Henle 9. JAMISON, R. L. Micropuncture study of segments of thin loop of in Psammomys. J. Clin. Inuest. 48: , Henle in the rat. Am. J. Physiol. 215 : , SCHNERMANN, J., H. VALTIN, K. THURAU, W. NAGEL, M. HOR- 10. JAMISON, R. L. Micropuncture study of superficial and juxta- STER, H. FISCHBACK, M. WAHL, AND G. LIEBAU. Micropuncture medullary nephrons in the rat. Am. J. Physiol. 2 18: 46-55, studies on the influence of antidiuretic hormone on tubular fluid 11. JAMISON, R. L. The function of the thin loops of Henle in the reabsorption in rats with hereditary hypothalamic diabetes urinary concentrating mechanism. Proc. Intern. Congr. Nephrol., insipidus. Arch. Ges. Physiol. 306: 103-l 18, th, Washington, D.C., 1970, vol. 1, p. 170-l SNEDECOR, G. W. Statistical Methods. (5th ed.). Ames: Iowa State 12. JAMISON, R. L., C.,M. BENNETT, AND R. W. BERLINER. Counter- Univ. Press, current multiplication by the thin loops of Henle. Am. J. Physiol. 28. STEPHENSON, JOHN L. Concentration of urine in a central core 212: , model of the renal counterflow system. Kidney Intern. 2: 85-94, 13. JAMISON, R. L., J. BUERKERT, AND F. LACY. A micropuncture study of collecting tubule function in rats with hereditary diabetes 29. STUMPE, K. O., L. H. D. LOWITZ, AND B. OCHWADT. Function of insipidus. J. Clin. Invest. 50 : , juxtamedullary nephrons in normotensive and chronically hyper- 14. JAMISON, R. L., AND F. B. LACY. Effect of saline infusion on super- tensive rats. Arch. Ges. Physiol. 313 : 43-52, ficial and juxtamedullary nephrons in the rat. Am. J. Physiol. 221: , KOKKO, J. P. Sodium chloride and water transport in the descending limb of Henle. J. Clin. InueJt. 49 : 1838-l 846, KOKKO, J. P. Urea transport in the proximal convoluted tubule and the descending limb of Henle. J. Clin. Invest. 5 1: , KOKKO, J. P., AND F. C. RECTOR. Countercurrent multiplication system without active transport in inner medulla. Kidney Intern. 2: , KRIZ, W., J. SCHERMANN, AND H. J. DIETRICH. Differences in the 30. VALTIN, H. Sequestration of urea and nonurea solutes in renal tissues of rats with hereditary hypothalamic diabetes insipidus: effect of vasopressin and dehydration on the countercurrent mechanism. J. Clin. Invest. 45 : , VALTIN, H. Hereditary hypothalamic diabetes insipidus in rats (Brattleboro strain). A useful experimental model. Am. J. Med. 42: , VALTIN, H., AND H. A. SCHROEDER. Familial diabetes insipidus in rats (Brattleboro strain). Am. J. Physiol. 206 : , WINDHAGER, E. E. Electrophysiological study of renal papilla of golden hamsters. Am. J. Physiol. 206 : , 1964.