Pfliigers Arch. 357, 243--252 (1975) 9 by Springer-Verlag 1975
Effects of Indomethacin on Renal Hemodynamics and on Water and Sodium Excretion by the Isolated Dog Kidney J.-L. Vanherweghem, J. Dueobu, and A. D ' H o l l a n d e r Laboratoire de M6decine Exp6riment~le, Fondation M6dieale Reine Elisabeth, Universit6 Libre de Bruxelles, Brussels, Belgium Received February 20, 1975
Summary. Blood-perfused isolated dog kidneys demonstrate steady increases in blood flow and in water and sodium excretion which could be attributed to the accumulation of renal prostaglandins in the perfusing blood. This hypothesis was tested by adding indomethacin, a potent inhibitor of prostaglandins synthesis, to the perfusing blood. Indomethacin completely prevented the vasodilation observed in control kidneys, without affecting glomerular filtration rate. Urine volume was not modified but sodium excretion was enhanced while the steady free water clearance increment observed in the control kidneys was depressed by indomethaein. The sum of sodium and free water clearances which, in the absence of antidiuretie hormone, constitutes an index of the part of the filtered load of solutes which escapes proximal tubular reabsorption, was not modified by indomethacin. Finally, indomethaein partially maintained the osmotic cortico-papillary gradient which was abolished after 2 hrs perfusion in control kidneys. These data suggest that prostaglandins accumulation in the blood is probably the major cause of the vasodilation taking place in isolated blood-perfused kidneys. This vasodilation does not account for decreased proximal reabsorption but partially explains the ADH-resistant diabetes insipidus developing in the isolated kidney. l~oreover, indomethacin inhibits sodium reabsorption in the ascending limb of Henle's loop and increases water transport in the collecting duct. Key words: Isolated Kidney -- Indomethacin -- Pros~aglandins -- Renal Hemodynamics -- Water and Sodium Excretion. I n the course o f perfusion with heparinized whole blood at 37~ isolated dog kidneys develop 3[. vasodilation, 2. decreased tubular sodium reabsorption, 3. increased urine o u t p u t and 4. enhanced free water excretion, w i t h o u t significant changes in glomerular filtration rate [4,18, 23]. (Vanherweghem, J.-L., Dueobu, J., D'Hollander, A., Toussaint, C. : Interactions between furosemide and vasopressin on hemodynamics and on water excretion b y the isolated dog kidney. Unpublished data.) As prostaglandins administration to the whole animal leads to kidney functional changes similar to those observed in the isolated dog kidney [9,13,17, 22], the accumulation of renal prostaglandins in the perfusing blood could explain the changes observed in the isolated kidney in the course of perfusion. I f this hypothesis is correct, the addition o f indo-
244
J.-L. Vanherweghem e$ al.
m e t h a e i n , a p o t e n t i n h i b i t o r o f p r o s t a g l a n d i n s synthesis [1,6,10], t o t h e perfusing b l o o d should p r e v e n t t h e f u n c t i o n a l changes u s u a l l y observed in t h e perfused i s o l a t e d k i d n e y .
Methods After the intravenous administration of 250 mg heparin (Heparin | Leo), bilateral nephrectomy was performed in ten 13--18kg Mongrel dogs, under pentobarbital (30mg/kg, Nembutal| anesthesia. After 5--10 rain ischemia, both kidneys of each animal were individually connected to two separate Nizet pump oxygenators [18], and perfnsed simultaneously. Each pump oxygenator was primed with 450 ml of heparinized arterial blood obtained from a common pentobarbital anesthetized 25--35 kg Mongrel dog. In each case, the donor dog was bled during the 5 rain preceding the start of perfusions on the pump oxygenators; 25 mg promethazin (Phenergan | Specia) and 80 mg creatinine (Merck| were added to each aliquot of priming blood at the start of perfusion. Kidney perfnsions were maintained during 150 vain, at 37~ and at constant mean blood pressure (120 mm Hg). The ureters were catheterized and urine allowed to flow back to the perfnsing blood. From 20 to 150 rain after starting perfusion, six 5--10 min urine collection periods were secured at regular intervals and arterial blood samples were obtained at mid-point of each clearance period. To alleviate depriming of the pump, urine and blood samples were replaced by equivalent volumes of 0.08 M NaC1. Prevention of hypokalemia was secured by 3 separate additions, at regular intervals, of 0.5 mM KCI. Care was taken to add I~aC1 and KC1 aliquots at least 10 min before the next urine collection period. 15 rain prior to the first clearance period, 8 mg crystallized indomethacin (Indocid| Merck Sharp & Dohme) dissolved in 30 ml 0.056 M NaHCO~ were added to one of the paired perfusions (experimental kidney), while 30 ml 0.056 M NaHCO 3 were added to the other perfusion (control kidney), care being taken to alternate left and right organs in successive experiments. During each paired perfnsion the following measurements were simultaneously compared for the experimental and for the control kidneys: 1. renal blood flow (RB~), measured directly at the renal vein with a 25 ml calibrated pipette and a chronometer; 2. renal plasma flow (RPF) derived from RBF and arterial hematocrit (Itet) ; 3. glomerular filtration rate (GI~R) measured by the creatinine clearance; 4. free water clearance (CH2o) derived from urine flow rate (V) and osmolar clearance (Costa); 5. reabsorbed ~Ta/filtered Na ratio (RNa/FNa) derived from Na excretion rate (U~aV), GFR and plasma Na concentration (P~a), without correction for Gibbs-Donnan equilibrium; 6. Na clearance (CN~), calculated from UNaV and P~a; 7. K excretion rate (UxV). In the absence of antidiuretie hormone (ADH), V was considered as an index of the filtered load of water escaping proximal tubular reabsorption and (CNa ~ C~2o) as an index of the filtered load of solute escaping proximal reabsorption [20]. Each experiment was divided in 3 successive chronological periods: A (20--60 rain after initiating kidney perfusion), B (60--120rain) and C (120--150rain).
Effects of Indomcthacin on Isolated Dog Kidney
245
T h e results were expressed by mean values ~ SEE of two successive clearance periods for each period A, B and C. At the end of the experiment, both kidneys were deeapsulated and weighed. Several 0.6--1.0 g pieces of cortex and of inner medulla were homogenized with 2 ml distilled water using an Ultra Turrax (Jancke and Kunkel). The osmolality of the renal tissue aliquots was measured with a Fiske osmometer. Statistical calculations were conducted according to variance and covariance analysis, each measurement being compared to the paired measurement simultaneously obtained in the control kidney. On the other hand, the chronological evolution of each parameter from period A to periods ]3 and C was also examined. All the statistical tests were expressed by the ratio of variance F according to Fisher and Yates [7]. Degrees of significance were graded with asterisks: * 0.001 < P < 0.01.
** P < 0.001. Results
A. Blood Composition Plasma Na Concentration. Pica remained constant at 150 ~ 1 mEq/1 during the whole perfusion, and did not differ significantly in experimental and in control kidneys. Plasma K Concentration. PK rose slightly (**) from A (3.9 4- 0.1 mEq/ 1) to C (4.9 4- 0.2 mEq/1). The increment did not differ significantly in the experimental and in the control kidneys. Hematocrit. t t c t fell (**) from 39 4- 1 to 36 4- 1 ~ from A through C. This decrement was similar in experimental and in control kidneys. B. Hemodynamies (Fig. 1) In the control kidneys I~BF and g P F rose steadily (**) from A (295 • 24 and 179 q- 16 ml/min/100 g kidney) through C (507 4- 40 and 323-b 38ml/mln/100g kidney). Indomethaein significantly (**) inhibited this spontaneous vasodilation for R B F and R P F remained very stable during the course of the pcrfusion: 231 q- 23 and 128 q- 17 ml/ rain/100 g kidney in period A, 237 q- 15 and 136 4- 11 ml/min/100 g kidney in period C. G F g remained stable from the start to the end of the perfusion (31 4- 2 ml/min/100 g kidney) and it was not affected by indomethacin. F F fell significantly (**) from 18 to 11~ from period A through C in the control kidneys while indomethaein maintained F F at higher and stable values (**): 2i~ at the start and 19~ at the end of the perfusion. C. Urine Output (Fig.2) Urine excretion rate steadily increased (**) from period A (4.54 4- 0.76 ml/min/100 g kidney) through 0 (8.28 q- 1.01 ml/min/100 g kidney) in control kidneys. In indomethacin-treated kidneys, urine excretion in17 PflfigersArch.,Vol.857
246
J.-L. Vanherweghemet al. Renal blood flow ml/min/100g kidney
Gtomerui-ar fittrati0n rate mI./rain/100g kidney
60040
oo: 0
20A
B
C
I
I
i
0
20-50 60-120 120-150 rain
A 20-'6 0
B
C
' ' 60-120 120-150 min
Filtered fraction %
1 A
B
C
1
I
I
20-50 50-120 120-150 rnin
Fig. 1. Effects of indomethacin on hemodynamics. Mean values of the data collected during the three successive periods (A, B and C) of 10 paired perfusions are presented. Open circles represent control kidneys and closed circles indomethacin-treated kidneys
creased (**) from 5.41 • 1.01 through 9.50 =k 1.23 ml/min/100 g kidney but for the same periods this diuretic effect was not significant when compared with control values. V/GFR ratio changes had the same statistical meanings as the changes observed for V. D. N a and K Excretions (Fig. 2)
In comparison with kidneys in situ, RNa/FNa ratio was reduced to 91~ in isolated control kidneys. This ratio was not significantly decreased in the course of the perfusion, from period A to C. Indomethaein further decreased (**) RIqa/FNa ratio to 85~ (period A) and to 780/0 (period C). K excretion rate increased steadily (**) from period A (87 4- 9 FEq/min/100 g kidney) to C (126 q- 12 ~Eq/min/100 g kidney) in control kidneys. Indomethaein increased insignificantly K excretion rate to 97 q- 12 (period A) and 136 =k 16 FEq/100 g kidney (period C).
Effects of Indomethacin on Isolated Dog Kidney Urine volume (V) mt/min/lOOg kidney
247
CNa + C H20 m[/min/100g kidney -
10,0-
10.0-
8.0-
8.0-
6.0-
504.02.0-
4,0220
A
B
C
0
A I
2~ 60 60 120 1~0150 rain
CH20 m[/min/100g kidney
20 60
B I
C /
60 120 120 150 rain
RNa ] FNa
% gO
20 1
I II
/
8O
7O 60
20-60 60420 120-150 rain
20-60 60-120 120-150 min
Fig.2. Effects of indomethacin on water and sodium excretion, l~ean values of the data collected during the three successive periods (A, B and C) of l0 paired perfusions are presented. Open circles represent control kidneys and closed circles indomethacin-treated kidneys
E. Free Water Clearance (Figs. 2 and 3) In control kidneys, CEde steadily increased (**) from 4- 0.21 =t= 0.57 (period A) to 4-4.19 4- 0.50 ml/min/100 g kidney (period C). Indomethaein significantly (**) decreased C~2o but did not modify its progressive rise for CH~o increased (**) from --0.49 4- 0.09 (period A) to 4-2.10 4- 0.25 (period C ml/min/100 g kidney). The changes observed in C ~ o / G F R ratio had the same statistical meanings as those of CH,e. CH2o/V ratio steadily increased (**) from A to C and indomethacin significantly (**) decreased this ratio in periods A, B and C.
F. C~a 4- CE2o (Fig. 2) (C~a 4- Cja:~o) gradually increased (**) from 3.73 -t- 0.70 (period A) to 7.21 4- 0.90 (period C ml/min/100 g kidney). Indomethaein did not significantly affect (CNa 4- CE2o) : 4.30 4- 0.89 ml/min/100 g kidney in period A and 8.36 4- 1.26 ml/min/100 g kidney in period C. 17"
J.-L. Vanherweghem et al.
248 0/0 CH20/V
50 /
40
/
/
/
/
D
o/ / ! ! / !
30 84
/ ! /
/
20
/
iI
10
-10
~A
20:60
B
C
6;-12o 2c;-lso
Min Fig. 3. Effects of indomethacin on CE~o/Y. Mean values of the data collected during the three successive periods (A, B and C) of 10 paired perfusions are presented. Open circles represent control kidneys and closed circles indomethacin-treated kidneys
Changes in (C~a 4-CH2o)/GFI~ ratio had the same statistical meanings as those of (CNa -k C~2o).
G. Osmotic Cortieo-Papillary Gradient At the end of perfusion, the osmotic cortieo-papiUary gradient was abolished in control kidneys: cortex 303 =E 3 mOsm/kg, deep medulla 313 4- 26 mOsm/kg. These values did not differ significantly. On the other hand, the kidneys which had been submitted to indomethacin retained a significant (*) cortieo-papillary gradient: cortex 299 4- 4 mOsm/kg, deep medulla 398 4- 29 mOsm/kg. Discussion Several hypotheses have been put forward to explain the vasodilation which takes place in the course of the perfusion of the isolated kidney. The linear relationship between the precocity of the vasodilation and the kidney weight [4] suggests either destruction of a vasoconstrictive substance, or synthesis of a vasodilating material b y the kidney.
Effects of Indomethacin on Isolated Dog Kidney
249
Decreased angiotensin concentration, suggested by decreased concentration of renin substrate accompanied by renin concentration increase could explain the vasodilation [3J, but it should be stressed that the vasodilation of the isolated kidney is refractory to angiotensin [19]. Bradykinins elaborated at the oxygenator interfaces might induce vasodilation, as suggested by partial abolition of vasodilation by the addition to the perfusing blood of trypsinic inhibitors extracted from soya bean [181. Prostaglandins secreted by the kidney after isehemia [12] or under bradykinins stimulation [16] could be responsible for the vasodilation observed in the course of the perfusion of the isolated kidney. The inhibition of the vasodilation by indomethacin makes indeed the role of endogenous prostaglandins very plausible [10]. High levels of prostaglandins were recently reported in the perfusing blood of the isolated kidney; renal blood flow was, as in our own experiments, reduced by the addition of indomethacin, and prostaglandins levels were decreased in the perfusing blood [15]. Prostaglandins would also explain the resistance of the vasodilation to angiotensin [2]. Besides, the vasodilation of the isolated kidney is characterized by a medullary blood flow increment which is relatively larger than the cortical blood flow increment [4]. This would readily explain the suppression of the corticopapillary osmotic gradient. Partial maintenance of this gradient by indomethaein is compatible with the inhibition of the intrarenal hemodynamic effect of prostaglandins [11]. In the course of its perfusion the isolated kidney develops a progressive reduction of ~proximal water and _~a reabsorption rate as demonstrated by the steady increase, in the absence of ADH, in V/GFR and in (CNa + CH~o)/GFR. This phenomenon may be explained by several mechanisms: 1. Prostaglandins, through decreased FF, could reduce proximal lqa reabsorption [9,17] in the isolated control kidneys but the maintenance of a decreased proximal Na reabsorption rate in the indomethacin-treated kidneys, with normal I~F, would militate against this hemodynamic explanation. 2. As impairement of proximal Na reabsorption was not improved by a potent prcstaglandins inhibitor such as indomethacin, direct action of prostaglandins on proximal Na transport seems unlikely. 3. Decreased Iqa reabsorption could be due to ischemie injury or to some metabolic defect [23] in the proximal tubular epithelium. It should be stressed that the addition of various substrates to the perfusing blood may alleviate tubular sodium reabsorptive defect in the isolated kidney [233. The increased C~2o in the course of the perfusion of control kidneys is partially explained by enhanced Na reabsorption in the ascending limb of Henle's loop secondary to decreased proximal Na reabsorption
250
J.-L. Vanherweghemet al.
which results in constant RNa/FNa. This compensation of proximal escape of Na by increased Na transport in Kenle's loop would explain the maintenance of RNa/FNa ratio but would not be sufficient to explain enhanced CH~O for CHio/V ratio (Fig.3) increased also in the course of the perfusion. This is probably due to decreased water reabsorption in collecting ducts, secondary to the disappearance of ADH in the per. fusing blood [19] and also to the progressive dissipation of the osmotic cortico-papillary gradient during the perfusion. This explanation is supported by the observation that CH,O increment was greater from period A to B than from period B to C. Indomethacin did not affect proximal Na reabsorption but significantly decreased CK~o as well as CK2o/V ratio. This effect could result from: 1. increased H~O reabsorption in the collecting duct, 2. inhibition of Na reabsorption in the ascending limb of Henle's loop. Enhanced H20 reabsorption is demonstrated by the negative CK~o which was observed in period A in the indomethacin group while it was already positive in the control group during the same period. Such an increase in water transport by indomethaein may be due to inhibition of the dissipation of the osmotic cortieo-papillary gradient or to enhancement of residual ADH, or to suppression of ADH antagonism by prostaglandins [8,17]. In later periods (B and C) of the perfusion, ADH disappearance [5,14] makes the collecting duct impermeable to water. It follows that Ca2o decrement induced by indomethaein observed in these periods is due to inhibition of Na transport in the ascending limb of Henle's loop. This latter phenomenon is responsible for decreased RNa/ FNa ratio in the indomethacin group. Further arguments for the localization of indomethacin action on the ascending limb of Henle's loop arc: 1. the absence of significant change in (C~a ~-Ca2o)/GFI~ and accordingly in proximal Na reabsorption rate; 2. the lack of significant decrease in UKV/GFR which would suggest an action on Na/K exchange at distal site. Moreover, decreased Na reabsorption in the ascending limb of Hcnle's loop would readily explain the incomplete restoration of the osmotic cortico-papillary gradient by indomethaein despite of the effect of this drug on the renal blood flow. Acknowledgments. We are grateful to Dr. G. Toussaint for his suggestions, encouragmentsand criticisms. The technical assistance of MM. R. Vanderstraeten, P. Stroobants and R. Sehmidt is gratefully acknowledged. This work was supported by the Fonds de la Recherche SeientifiqueM6dieMe (contract no 1208) and by the Groupementpour l'Etude, le Traitement et la R6habilitation SoeiMedes InsuffisantsR6naux Chroniques,Brussels. Part of this workwas presented at the Eighth AnnumMeeting of the European Society for ClinicalInvestigation, Rotterdam, April 26, 1974.
:Effects of Indomethacin on Isolated Dog Kidney
251
References 1. Aiken, J . W . , Vane, J. R.: Blockade of angiotensin induced prostaglandin release from dog kidney by indomethacin. Pharmacologist 13, 293 (1971) 2. Aiken, J . W . , Vane, J . R . : Intrarenal prostaglandin release attenuates the renal vasoconstrictor activity of angiofensin. J. Pharmacol. exp. Ther. 184, 678--687 (1973) 3. Berkowitz, H. D., Miller, L. D., Itskovitz, H. D. : Renal function and the renin angiotensin system in the isolated perfused kidney. Amer. J. Physiol. 218, 928--934 (1967) 4. Berkowifz, H. D., Miller, L. D., Itskovitz, H. D., Bovee, K. C.: Renal function in the isolated perfused kidney. Surg. Gynec. Obstet. 127, 1257--1266 (1968) 5. Czaczkes, J. W., Kleeman, C. R. : The effects of various states of hydratafion and the plasma concentration on the turnover of antidiuretic hormone in mammals. J. clin. Invest. 48, 1649--1658 (1964) 6. Davis, A. tI., Herren, E. W. : Output of prostaglandins from the rabbit kidney, its increase on renal nerve stimulation and its inhibition by indomethacin. Brit. J. Pharmacol. 46, 658--675 (1972) 7. Fischer, R. Ao, Yates, F.: Statistical tables for biological, agricultural and medical research. 3rd Edit. Edinburgh: Oliver and Boyd Ltd. 1948 8. Grantham, J . J . , Orloff, J. : The effect of prostaglandins E 1 on the permeability response of the isolated collecting tubule to vasopressin, adenosine 3',5'-rochephosphate and theophylline. J. clin. Invest. 47, 1154--1161 (1968) 9. Gross, J. B., Bartter, F. C. : Effects of prostaglandins E 1, A1 and F2cr on renal handling of salt and water. Amer. J. Physiol. 225, 218--224 (1973) 10. Herbaczynska-Cedro, K., Vane, J. R.: Contribution of intrarenal generation of prostaglandin to auforegulation of renal blood flow in the dog. Circular. Res. 88, 428--436 (1973) 11. Itskovitz, H.D., Stemper, J., Pacholczyk, D., McGiff, J. C.: Renal prostaglandins: determinants of intrarenal distribution of blood flow. Clin. Sci. Mol. Med. 45 (Suppl. I), 321--324 (1973) 12. Jaffe, B. IV[., Parker, C.W., Marshall, G. R., Needleman, P.: Renal concentrations of prostaglandin E in acute and chronic renal ischemia. Biochem. biophys. Res. Commun. 49, 799--805 (1972) 13. Johnston, H. H., tterzog, J. P,, Lauler, D. P.: Effect of prostaglandin E1 on renal hemodynamics, sodium and water excretion. Amer. J. Physiol. 213, 939--946 (1967) 14. Lauson, H. : ~etabolism of antidiuretic hormones. Amer. J. Med. 42, 713--744 (1967) 15. Lonigro, A. J., Itskovitz, H. D., Crowshaw, K., McGiff, J. C.: Dependency of RBF on prostaglandin synthesis in the dog. Circular. Res. 32, 712--717 (1973) 16. McGiff, J. C., Terragno, N. A., Malik, K. U., Lonigro, A. J. : Release of prosfaglandin E like substance from canine kidney by bradykinin. Circular. Res. 81, 36--43 (1972) 17. Martinez-Maldonado, M., Tsaparas, N., Eknoyan, G., Suki, W . N . : Renal actions of prostaglandins comparison with acetylcholine and volume expansion. Amer. J. Physiol. 222, 1147--1152 (1972) 18. Nizet, A., Cuyl3ers, Y., Deetjen, P., Kramer, K.: Functional capacity of the isolated perfused dog kidney. Pflfigers~Arch. ges. Physiol. 296, 179--195 (1967) 19. Nizet, A. : Physiologic du rein en survie extracorporelle. J. TYrol. N6phrol. 77, 909--927 (197i)
252
J.-L. Vanherweghem et al.
20. Seldin, D. W., Eknoyan, G., Suki, W. N., Rector, F. C., Jr.: Localization of diuretic action from pattern of water and electrolytes excretion. Ann. N. Y. Acad. Sci. 189, 328--343 (1966) 21. Reference deleted. 22. Vander, A. J.: Direct effects of prostaglandin on renal function and renin release in anesthetized dog. Amer. J. Physiol. 214, 218--221 (1968) 23. Waugh, W. H., Kubo, T. : Development of an isolated perfused dog kidney with improved function, Amer. J. Physiol. 217, 277--290 (1969) J.-L. Vanherweghem D6partement de N~phrologie HSpital Universit. Brugmann Place Van Gehuchten, 4 B-1020 Brussels Belgium
Effects of Indomethacin on Renal Hemodynamics and on Water and Sodium Excretion by the Isolated Dog Kidney J.-L. Vanherweghem, J. Dueobu, and A. D ' H o l l a n d e r Laboratoire de M6decine Exp6riment~le, Fondation M6dieale Reine Elisabeth, Universit6 Libre de Bruxelles, Brussels, Belgium Received February 20, 1975
Summary. Blood-perfused isolated dog kidneys demonstrate steady increases in blood flow and in water and sodium excretion which could be attributed to the accumulation of renal prostaglandins in the perfusing blood. This hypothesis was tested by adding indomethacin, a potent inhibitor of prostaglandins synthesis, to the perfusing blood. Indomethacin completely prevented the vasodilation observed in control kidneys, without affecting glomerular filtration rate. Urine volume was not modified but sodium excretion was enhanced while the steady free water clearance increment observed in the control kidneys was depressed by indomethaein. The sum of sodium and free water clearances which, in the absence of antidiuretie hormone, constitutes an index of the part of the filtered load of solutes which escapes proximal tubular reabsorption, was not modified by indomethacin. Finally, indomethaein partially maintained the osmotic cortico-papillary gradient which was abolished after 2 hrs perfusion in control kidneys. These data suggest that prostaglandins accumulation in the blood is probably the major cause of the vasodilation taking place in isolated blood-perfused kidneys. This vasodilation does not account for decreased proximal reabsorption but partially explains the ADH-resistant diabetes insipidus developing in the isolated kidney. l~oreover, indomethacin inhibits sodium reabsorption in the ascending limb of Henle's loop and increases water transport in the collecting duct. Key words: Isolated Kidney -- Indomethacin -- Pros~aglandins -- Renal Hemodynamics -- Water and Sodium Excretion. I n the course o f perfusion with heparinized whole blood at 37~ isolated dog kidneys develop 3[. vasodilation, 2. decreased tubular sodium reabsorption, 3. increased urine o u t p u t and 4. enhanced free water excretion, w i t h o u t significant changes in glomerular filtration rate [4,18, 23]. (Vanherweghem, J.-L., Dueobu, J., D'Hollander, A., Toussaint, C. : Interactions between furosemide and vasopressin on hemodynamics and on water excretion b y the isolated dog kidney. Unpublished data.) As prostaglandins administration to the whole animal leads to kidney functional changes similar to those observed in the isolated dog kidney [9,13,17, 22], the accumulation of renal prostaglandins in the perfusing blood could explain the changes observed in the isolated kidney in the course of perfusion. I f this hypothesis is correct, the addition o f indo-
244
J.-L. Vanherweghem e$ al.
m e t h a e i n , a p o t e n t i n h i b i t o r o f p r o s t a g l a n d i n s synthesis [1,6,10], t o t h e perfusing b l o o d should p r e v e n t t h e f u n c t i o n a l changes u s u a l l y observed in t h e perfused i s o l a t e d k i d n e y .
Methods After the intravenous administration of 250 mg heparin (Heparin | Leo), bilateral nephrectomy was performed in ten 13--18kg Mongrel dogs, under pentobarbital (30mg/kg, Nembutal| anesthesia. After 5--10 rain ischemia, both kidneys of each animal were individually connected to two separate Nizet pump oxygenators [18], and perfnsed simultaneously. Each pump oxygenator was primed with 450 ml of heparinized arterial blood obtained from a common pentobarbital anesthetized 25--35 kg Mongrel dog. In each case, the donor dog was bled during the 5 rain preceding the start of perfusions on the pump oxygenators; 25 mg promethazin (Phenergan | Specia) and 80 mg creatinine (Merck| were added to each aliquot of priming blood at the start of perfusion. Kidney perfnsions were maintained during 150 vain, at 37~ and at constant mean blood pressure (120 mm Hg). The ureters were catheterized and urine allowed to flow back to the perfnsing blood. From 20 to 150 rain after starting perfusion, six 5--10 min urine collection periods were secured at regular intervals and arterial blood samples were obtained at mid-point of each clearance period. To alleviate depriming of the pump, urine and blood samples were replaced by equivalent volumes of 0.08 M NaC1. Prevention of hypokalemia was secured by 3 separate additions, at regular intervals, of 0.5 mM KCI. Care was taken to add I~aC1 and KC1 aliquots at least 10 min before the next urine collection period. 15 rain prior to the first clearance period, 8 mg crystallized indomethacin (Indocid| Merck Sharp & Dohme) dissolved in 30 ml 0.056 M NaHCO~ were added to one of the paired perfusions (experimental kidney), while 30 ml 0.056 M NaHCO 3 were added to the other perfusion (control kidney), care being taken to alternate left and right organs in successive experiments. During each paired perfnsion the following measurements were simultaneously compared for the experimental and for the control kidneys: 1. renal blood flow (RB~), measured directly at the renal vein with a 25 ml calibrated pipette and a chronometer; 2. renal plasma flow (RPF) derived from RBF and arterial hematocrit (Itet) ; 3. glomerular filtration rate (GI~R) measured by the creatinine clearance; 4. free water clearance (CH2o) derived from urine flow rate (V) and osmolar clearance (Costa); 5. reabsorbed ~Ta/filtered Na ratio (RNa/FNa) derived from Na excretion rate (U~aV), GFR and plasma Na concentration (P~a), without correction for Gibbs-Donnan equilibrium; 6. Na clearance (CN~), calculated from UNaV and P~a; 7. K excretion rate (UxV). In the absence of antidiuretie hormone (ADH), V was considered as an index of the filtered load of water escaping proximal tubular reabsorption and (CNa ~ C~2o) as an index of the filtered load of solute escaping proximal reabsorption [20]. Each experiment was divided in 3 successive chronological periods: A (20--60 rain after initiating kidney perfusion), B (60--120rain) and C (120--150rain).
Effects of Indomcthacin on Isolated Dog Kidney
245
T h e results were expressed by mean values ~ SEE of two successive clearance periods for each period A, B and C. At the end of the experiment, both kidneys were deeapsulated and weighed. Several 0.6--1.0 g pieces of cortex and of inner medulla were homogenized with 2 ml distilled water using an Ultra Turrax (Jancke and Kunkel). The osmolality of the renal tissue aliquots was measured with a Fiske osmometer. Statistical calculations were conducted according to variance and covariance analysis, each measurement being compared to the paired measurement simultaneously obtained in the control kidney. On the other hand, the chronological evolution of each parameter from period A to periods ]3 and C was also examined. All the statistical tests were expressed by the ratio of variance F according to Fisher and Yates [7]. Degrees of significance were graded with asterisks: * 0.001 < P < 0.01.
** P < 0.001. Results
A. Blood Composition Plasma Na Concentration. Pica remained constant at 150 ~ 1 mEq/1 during the whole perfusion, and did not differ significantly in experimental and in control kidneys. Plasma K Concentration. PK rose slightly (**) from A (3.9 4- 0.1 mEq/ 1) to C (4.9 4- 0.2 mEq/1). The increment did not differ significantly in the experimental and in the control kidneys. Hematocrit. t t c t fell (**) from 39 4- 1 to 36 4- 1 ~ from A through C. This decrement was similar in experimental and in control kidneys. B. Hemodynamies (Fig. 1) In the control kidneys I~BF and g P F rose steadily (**) from A (295 • 24 and 179 q- 16 ml/min/100 g kidney) through C (507 4- 40 and 323-b 38ml/mln/100g kidney). Indomethaein significantly (**) inhibited this spontaneous vasodilation for R B F and R P F remained very stable during the course of the pcrfusion: 231 q- 23 and 128 q- 17 ml/ rain/100 g kidney in period A, 237 q- 15 and 136 4- 11 ml/min/100 g kidney in period C. G F g remained stable from the start to the end of the perfusion (31 4- 2 ml/min/100 g kidney) and it was not affected by indomethacin. F F fell significantly (**) from 18 to 11~ from period A through C in the control kidneys while indomethaein maintained F F at higher and stable values (**): 2i~ at the start and 19~ at the end of the perfusion. C. Urine Output (Fig.2) Urine excretion rate steadily increased (**) from period A (4.54 4- 0.76 ml/min/100 g kidney) through 0 (8.28 q- 1.01 ml/min/100 g kidney) in control kidneys. In indomethacin-treated kidneys, urine excretion in17 PflfigersArch.,Vol.857
246
J.-L. Vanherweghemet al. Renal blood flow ml/min/100g kidney
Gtomerui-ar fittrati0n rate mI./rain/100g kidney
60040
oo: 0
20A
B
C
I
I
i
0
20-50 60-120 120-150 rain
A 20-'6 0
B
C
' ' 60-120 120-150 min
Filtered fraction %
1 A
B
C
1
I
I
20-50 50-120 120-150 rnin
Fig. 1. Effects of indomethacin on hemodynamics. Mean values of the data collected during the three successive periods (A, B and C) of 10 paired perfusions are presented. Open circles represent control kidneys and closed circles indomethacin-treated kidneys
creased (**) from 5.41 • 1.01 through 9.50 =k 1.23 ml/min/100 g kidney but for the same periods this diuretic effect was not significant when compared with control values. V/GFR ratio changes had the same statistical meanings as the changes observed for V. D. N a and K Excretions (Fig. 2)
In comparison with kidneys in situ, RNa/FNa ratio was reduced to 91~ in isolated control kidneys. This ratio was not significantly decreased in the course of the perfusion, from period A to C. Indomethaein further decreased (**) RIqa/FNa ratio to 85~ (period A) and to 780/0 (period C). K excretion rate increased steadily (**) from period A (87 4- 9 FEq/min/100 g kidney) to C (126 q- 12 ~Eq/min/100 g kidney) in control kidneys. Indomethaein increased insignificantly K excretion rate to 97 q- 12 (period A) and 136 =k 16 FEq/100 g kidney (period C).
Effects of Indomethacin on Isolated Dog Kidney Urine volume (V) mt/min/lOOg kidney
247
CNa + C H20 m[/min/100g kidney -
10,0-
10.0-
8.0-
8.0-
6.0-
504.02.0-
4,0220
A
B
C
0
A I
2~ 60 60 120 1~0150 rain
CH20 m[/min/100g kidney
20 60
B I
C /
60 120 120 150 rain
RNa ] FNa
% gO
20 1
I II
/
8O
7O 60
20-60 60420 120-150 rain
20-60 60-120 120-150 min
Fig.2. Effects of indomethacin on water and sodium excretion, l~ean values of the data collected during the three successive periods (A, B and C) of l0 paired perfusions are presented. Open circles represent control kidneys and closed circles indomethacin-treated kidneys
E. Free Water Clearance (Figs. 2 and 3) In control kidneys, CEde steadily increased (**) from 4- 0.21 =t= 0.57 (period A) to 4-4.19 4- 0.50 ml/min/100 g kidney (period C). Indomethaein significantly (**) decreased C~2o but did not modify its progressive rise for CH~o increased (**) from --0.49 4- 0.09 (period A) to 4-2.10 4- 0.25 (period C ml/min/100 g kidney). The changes observed in C ~ o / G F R ratio had the same statistical meanings as those of CH,e. CH2o/V ratio steadily increased (**) from A to C and indomethacin significantly (**) decreased this ratio in periods A, B and C.
F. C~a 4- CE2o (Fig. 2) (C~a 4- Cja:~o) gradually increased (**) from 3.73 -t- 0.70 (period A) to 7.21 4- 0.90 (period C ml/min/100 g kidney). Indomethaein did not significantly affect (CNa 4- CE2o) : 4.30 4- 0.89 ml/min/100 g kidney in period A and 8.36 4- 1.26 ml/min/100 g kidney in period C. 17"
J.-L. Vanherweghem et al.
248 0/0 CH20/V
50 /
40
/
/
/
/
D
o/ / ! ! / !
30 84
/ ! /
/
20
/
iI
10
-10
~A
20:60
B
C
6;-12o 2c;-lso
Min Fig. 3. Effects of indomethacin on CE~o/Y. Mean values of the data collected during the three successive periods (A, B and C) of 10 paired perfusions are presented. Open circles represent control kidneys and closed circles indomethacin-treated kidneys
Changes in (C~a 4-CH2o)/GFI~ ratio had the same statistical meanings as those of (CNa -k C~2o).
G. Osmotic Cortieo-Papillary Gradient At the end of perfusion, the osmotic cortieo-papiUary gradient was abolished in control kidneys: cortex 303 =E 3 mOsm/kg, deep medulla 313 4- 26 mOsm/kg. These values did not differ significantly. On the other hand, the kidneys which had been submitted to indomethacin retained a significant (*) cortieo-papillary gradient: cortex 299 4- 4 mOsm/kg, deep medulla 398 4- 29 mOsm/kg. Discussion Several hypotheses have been put forward to explain the vasodilation which takes place in the course of the perfusion of the isolated kidney. The linear relationship between the precocity of the vasodilation and the kidney weight [4] suggests either destruction of a vasoconstrictive substance, or synthesis of a vasodilating material b y the kidney.
Effects of Indomethacin on Isolated Dog Kidney
249
Decreased angiotensin concentration, suggested by decreased concentration of renin substrate accompanied by renin concentration increase could explain the vasodilation [3J, but it should be stressed that the vasodilation of the isolated kidney is refractory to angiotensin [19]. Bradykinins elaborated at the oxygenator interfaces might induce vasodilation, as suggested by partial abolition of vasodilation by the addition to the perfusing blood of trypsinic inhibitors extracted from soya bean [181. Prostaglandins secreted by the kidney after isehemia [12] or under bradykinins stimulation [16] could be responsible for the vasodilation observed in the course of the perfusion of the isolated kidney. The inhibition of the vasodilation by indomethacin makes indeed the role of endogenous prostaglandins very plausible [10]. High levels of prostaglandins were recently reported in the perfusing blood of the isolated kidney; renal blood flow was, as in our own experiments, reduced by the addition of indomethacin, and prostaglandins levels were decreased in the perfusing blood [15]. Prostaglandins would also explain the resistance of the vasodilation to angiotensin [2]. Besides, the vasodilation of the isolated kidney is characterized by a medullary blood flow increment which is relatively larger than the cortical blood flow increment [4]. This would readily explain the suppression of the corticopapillary osmotic gradient. Partial maintenance of this gradient by indomethaein is compatible with the inhibition of the intrarenal hemodynamic effect of prostaglandins [11]. In the course of its perfusion the isolated kidney develops a progressive reduction of ~proximal water and _~a reabsorption rate as demonstrated by the steady increase, in the absence of ADH, in V/GFR and in (CNa + CH~o)/GFR. This phenomenon may be explained by several mechanisms: 1. Prostaglandins, through decreased FF, could reduce proximal lqa reabsorption [9,17] in the isolated control kidneys but the maintenance of a decreased proximal Na reabsorption rate in the indomethacin-treated kidneys, with normal I~F, would militate against this hemodynamic explanation. 2. As impairement of proximal Na reabsorption was not improved by a potent prcstaglandins inhibitor such as indomethacin, direct action of prostaglandins on proximal Na transport seems unlikely. 3. Decreased Iqa reabsorption could be due to ischemie injury or to some metabolic defect [23] in the proximal tubular epithelium. It should be stressed that the addition of various substrates to the perfusing blood may alleviate tubular sodium reabsorptive defect in the isolated kidney [233. The increased C~2o in the course of the perfusion of control kidneys is partially explained by enhanced Na reabsorption in the ascending limb of Henle's loop secondary to decreased proximal Na reabsorption
250
J.-L. Vanherweghemet al.
which results in constant RNa/FNa. This compensation of proximal escape of Na by increased Na transport in Kenle's loop would explain the maintenance of RNa/FNa ratio but would not be sufficient to explain enhanced CH~O for CHio/V ratio (Fig.3) increased also in the course of the perfusion. This is probably due to decreased water reabsorption in collecting ducts, secondary to the disappearance of ADH in the per. fusing blood [19] and also to the progressive dissipation of the osmotic cortico-papillary gradient during the perfusion. This explanation is supported by the observation that CH,O increment was greater from period A to B than from period B to C. Indomethacin did not affect proximal Na reabsorption but significantly decreased CK~o as well as CK2o/V ratio. This effect could result from: 1. increased H~O reabsorption in the collecting duct, 2. inhibition of Na reabsorption in the ascending limb of Henle's loop. Enhanced H20 reabsorption is demonstrated by the negative CK~o which was observed in period A in the indomethacin group while it was already positive in the control group during the same period. Such an increase in water transport by indomethaein may be due to inhibition of the dissipation of the osmotic cortieo-papillary gradient or to enhancement of residual ADH, or to suppression of ADH antagonism by prostaglandins [8,17]. In later periods (B and C) of the perfusion, ADH disappearance [5,14] makes the collecting duct impermeable to water. It follows that Ca2o decrement induced by indomethaein observed in these periods is due to inhibition of Na transport in the ascending limb of Henle's loop. This latter phenomenon is responsible for decreased RNa/ FNa ratio in the indomethacin group. Further arguments for the localization of indomethacin action on the ascending limb of Henle's loop arc: 1. the absence of significant change in (C~a ~-Ca2o)/GFI~ and accordingly in proximal Na reabsorption rate; 2. the lack of significant decrease in UKV/GFR which would suggest an action on Na/K exchange at distal site. Moreover, decreased Na reabsorption in the ascending limb of Hcnle's loop would readily explain the incomplete restoration of the osmotic cortico-papillary gradient by indomethaein despite of the effect of this drug on the renal blood flow. Acknowledgments. We are grateful to Dr. G. Toussaint for his suggestions, encouragmentsand criticisms. The technical assistance of MM. R. Vanderstraeten, P. Stroobants and R. Sehmidt is gratefully acknowledged. This work was supported by the Fonds de la Recherche SeientifiqueM6dieMe (contract no 1208) and by the Groupementpour l'Etude, le Traitement et la R6habilitation SoeiMedes InsuffisantsR6naux Chroniques,Brussels. Part of this workwas presented at the Eighth AnnumMeeting of the European Society for ClinicalInvestigation, Rotterdam, April 26, 1974.
:Effects of Indomethacin on Isolated Dog Kidney
251
References 1. Aiken, J . W . , Vane, J. R.: Blockade of angiotensin induced prostaglandin release from dog kidney by indomethacin. Pharmacologist 13, 293 (1971) 2. Aiken, J . W . , Vane, J . R . : Intrarenal prostaglandin release attenuates the renal vasoconstrictor activity of angiofensin. J. Pharmacol. exp. Ther. 184, 678--687 (1973) 3. Berkowitz, H. D., Miller, L. D., Itskovitz, H. D. : Renal function and the renin angiotensin system in the isolated perfused kidney. Amer. J. Physiol. 218, 928--934 (1967) 4. Berkowifz, H. D., Miller, L. D., Itskovitz, H. D., Bovee, K. C.: Renal function in the isolated perfused kidney. Surg. Gynec. Obstet. 127, 1257--1266 (1968) 5. Czaczkes, J. W., Kleeman, C. R. : The effects of various states of hydratafion and the plasma concentration on the turnover of antidiuretic hormone in mammals. J. clin. Invest. 48, 1649--1658 (1964) 6. Davis, A. tI., Herren, E. W. : Output of prostaglandins from the rabbit kidney, its increase on renal nerve stimulation and its inhibition by indomethacin. Brit. J. Pharmacol. 46, 658--675 (1972) 7. Fischer, R. Ao, Yates, F.: Statistical tables for biological, agricultural and medical research. 3rd Edit. Edinburgh: Oliver and Boyd Ltd. 1948 8. Grantham, J . J . , Orloff, J. : The effect of prostaglandins E 1 on the permeability response of the isolated collecting tubule to vasopressin, adenosine 3',5'-rochephosphate and theophylline. J. clin. Invest. 47, 1154--1161 (1968) 9. Gross, J. B., Bartter, F. C. : Effects of prostaglandins E 1, A1 and F2cr on renal handling of salt and water. Amer. J. Physiol. 225, 218--224 (1973) 10. Herbaczynska-Cedro, K., Vane, J. R.: Contribution of intrarenal generation of prostaglandin to auforegulation of renal blood flow in the dog. Circular. Res. 88, 428--436 (1973) 11. Itskovitz, H.D., Stemper, J., Pacholczyk, D., McGiff, J. C.: Renal prostaglandins: determinants of intrarenal distribution of blood flow. Clin. Sci. Mol. Med. 45 (Suppl. I), 321--324 (1973) 12. Jaffe, B. IV[., Parker, C.W., Marshall, G. R., Needleman, P.: Renal concentrations of prostaglandin E in acute and chronic renal ischemia. Biochem. biophys. Res. Commun. 49, 799--805 (1972) 13. Johnston, H. H., tterzog, J. P,, Lauler, D. P.: Effect of prostaglandin E1 on renal hemodynamics, sodium and water excretion. Amer. J. Physiol. 213, 939--946 (1967) 14. Lauson, H. : ~etabolism of antidiuretic hormones. Amer. J. Med. 42, 713--744 (1967) 15. Lonigro, A. J., Itskovitz, H. D., Crowshaw, K., McGiff, J. C.: Dependency of RBF on prostaglandin synthesis in the dog. Circular. Res. 32, 712--717 (1973) 16. McGiff, J. C., Terragno, N. A., Malik, K. U., Lonigro, A. J. : Release of prosfaglandin E like substance from canine kidney by bradykinin. Circular. Res. 81, 36--43 (1972) 17. Martinez-Maldonado, M., Tsaparas, N., Eknoyan, G., Suki, W . N . : Renal actions of prostaglandins comparison with acetylcholine and volume expansion. Amer. J. Physiol. 222, 1147--1152 (1972) 18. Nizet, A., Cuyl3ers, Y., Deetjen, P., Kramer, K.: Functional capacity of the isolated perfused dog kidney. Pflfigers~Arch. ges. Physiol. 296, 179--195 (1967) 19. Nizet, A. : Physiologic du rein en survie extracorporelle. J. TYrol. N6phrol. 77, 909--927 (197i)
252
J.-L. Vanherweghem et al.
20. Seldin, D. W., Eknoyan, G., Suki, W. N., Rector, F. C., Jr.: Localization of diuretic action from pattern of water and electrolytes excretion. Ann. N. Y. Acad. Sci. 189, 328--343 (1966) 21. Reference deleted. 22. Vander, A. J.: Direct effects of prostaglandin on renal function and renin release in anesthetized dog. Amer. J. Physiol. 214, 218--221 (1968) 23. Waugh, W. H., Kubo, T. : Development of an isolated perfused dog kidney with improved function, Amer. J. Physiol. 217, 277--290 (1969) J.-L. Vanherweghem D6partement de N~phrologie HSpital Universit. Brugmann Place Van Gehuchten, 4 B-1020 Brussels Belgium
