Potassium transport across renal distal tubules during acid-base disturbances

Transcription

1 AMERCAN JOURNAL OF PHYSOLOGY Vol. 221, No. 4, October Pinted in U.S. A; Potassium transport across renal distal tubules during acid-base disturbances GERHARD MALNC, MARGARDA DE MELLO ARES, AND GERHARD GEBSCH Departments of Physiology, Cornell University Medical College, New York City 121; Yale University School of Medicine, New Haven, Connecticut 651; and University of &To Paul Medical School, Sa o Paula, Basil ~MALNC, GERHARD, MARGARDA DE MELLO ARES, AND GERHARD GEBSCH. Potassium transport across renal distal tubules during acid-base disturbances. Am. J. Physiol. 221(4) : Free-flow samples of tubular fluid (TF) were collected from superficial nephrons of rat kidneys for the analysis of potassium (K), ph, bicarbonate and inulin-w. Rats were subjected to metabolic or respiratory acidosis and alkalosis, to combinations of these acid-base disturbances, and to acute K loading. 1) The distal tubular epithelium is the main site where alterations in the acid-base status effect changes in K transport, 2) Distal tubular K secretion is reduced in acute metabolic and respiratory acidosis despite a sharp increase in plasma K, whereas metabolic and respiratory alkalosis enhance distal K secretion despite a fall in plasma K levels. 3) An increase in distal intratubular ph is associated with increasing potassium secretion. 4) ntracellular acidification, independent of intratubular ph, reduces distal K secretion. 5) K entry into the distal tubule proceeds down an electrochemical potential gradient; and 6) no evidence was obtained of a consistent reciprocal relationship between rates of distal tubular K and hydrogen secretion. distal tubular function; tubular potassium transport; tubular in situ ph; tubular bicarbonate reabsorption T S WELL ESTABLSHED that deviations from the normal acid-base status have a profound effect on renal potassium excretion. Particularly pertinent are observations that indicate the existence of a reciprocal relationship between urinary acidity and renal potassium loss. Supportive experimental evidence may be cited in which an increased urinary potassium excretion was observed during urinary alkalinization brought about either by induction of metabolic (16, 29, 35, 54) or respiratory (3, 17, 59) alkalosis or the administration of acetazolamide (7). Conversely, although not consistently (1, 35, 39, 49) a decrease in urinary potassium excretion can be observed in acute metabolic (65) and respiratory (3, 13, 15, 17) acidosis. nfusion of potassium salts may, on the other hand, lead to decreased bicarbonate reabsorption and an increase in urinary ph (18, 47, 54, 55, but see 6). States of potassium deficiency generally induce metabolic alkalosis and enhance tubular bicarbonate reabsorption (23, 36, 55). These observations have led to the thesis of some competition between the exchange rates of distal cellular potassium and hydrogen ions for tubular sodium ions such that potassium and hydrogen ions compete for a common secretory pathway (4, 6, 7, 47) The present micropuncture study was undertaken to delineate the distal tubular potassium transport system in the rat under a wide variety of acid-base disturbances and to compare the rate of K secretion to the simultaneously measured transport patterns of bicarbonate and hydrogen ions. Thus, some insight was gained into site and extent of tubular potassium secretion under these conditions. n addition, some factors were elucidated which regulate potassium transport during acid-base disturbances. METHODS Male albino or Long-Evans hooded rats weighing g were used. Animals were anesthetized by intraperitoneal injection of nactin (8-1 mg/kg body wt). The experimental conditions of the rats were similar to those outlined in another paper which describes in greater detail the findings related to tubular bicarbonate and hydrogen ion transport (44). The following experimental groups of rats were studied: 1) control rats kept on a diet of Purina laboratory chow (.72 % K,.46 % Na) until the night preceding the experiment and having free access to tap water; 2) rats in which acute respiratory acidosis was induced by breathing a mixture of 15 % COZ-85 % 2; 3) rats in which metabolic acidosis resulted from the dietary pretreatment with an acidifying salt (CaC12) at a concentration of 4% and the substitution of.5 isotonic calcium chloride for drinking water (these animals received an intravenous solution of isotonic ammonium chloride in 5 % mannitol (.1 ml/ min) during the experiment); 4) rats hyperventilated artificially by means of a Harvard or Palmer respirator (rats listed under, 2, and 4) received the same diet and an intravenous infusion of isotonic sodium chloride in 5 % mannitol at a rate of.1 ml/min during the experiment) ; 5) rats in which the infusion of a solution of 5 % sodium bicarbonate at the rate of.1 ml/min led to metabolic alkalosis; 6) rats in which respiratory alkalosis (mechanical hyperventilation) was superirnposed on intravenous bicarbonate loading; 7) rats in which respiratory acidosis (15 % COZ-85 %,) was superirnposed on intravenous bicarbonate loading (the experiments listed under 6 and 7 were done to compare the distal tubular potassium secretory system under conditions of similar intratubular ph but differing intracellular ph (high during hyperventilation, low during breathing 15 % COZ-85 %,)); 8) rats in which the effect of an intravenous potassium chloride load (.1

2 ACD-BASE AND RENAL K SECRETON ml/min) on distal and urinary potassium excretion was tested in control; and 9) in bicarbonate-loaded rats (see group 5). The state of mild osmotic diuresis (5 % mannitol in isotonic NaCl) was chosen to minimize changes in distal tubular potassium secretion which could have resulted from unspecific changes in distal tubular fluid reabsorption. A fairly comparable range of inulin U/P ratios was accomplished (Table 1) and makes it unlikely that distal fluid reabsorption varied greatly in the individual groups. This precaution is of importance in view of the experimental evidence demonstrating that changes in distal tubular volume flow rate per se can significantly modify the apparent rate of distal tubular potassium secretion (34). All animals received a priming infusion of 5 pc inulinj4c in.5 ml isotonic sodium chloride solution. The rate of inulin-i4c delivery in the different sustaining infusions was 3-5 &hr. The general methods of preparation, micropuncture, and localization of tubular puncture sites have been described in previous publications (4 1,42). A flank approach to expose the lefi kidney was used consistently. During the surgical preparation the animals received ml of isotonic SaCl intravenously to replace fluid loss. Collection of tubular fluid was begun 3-6 min after the start of the various infusions or after changes in the respi ratory gas mixture. Tubular potassium concen trations were measured by means of ultramicro flame photometry as described previously (4 1) with the addition of digital read-out devices for measuring the integrated signals. Replicate measurements of potassium in fluid samples (l-2 ml) containing potassium chloride in physiological concentrations could be carried out with an accuracy of A4.3 % (SD, n = 21). The preparation of ph-sensitive antimony electrodes for the measurement of intratubular in situ ph and for the measurement of the concentration of bicarbonate in fluid samples has been presented in detail in another paper (66). Blood samples were collected from the carotid artery and the volume withdrawn replaced by an equal amount of isotonic saline. Plasma and urinary potassium concentrations were determined by a Baird flame photometer, the bicarbonate concentration in plasma was measured by means of a Natelson gas microgasometer, and urinary and whole blood ph were determined by a Metrohm glass electrode system. Radioactivity of tubular, plasma, and urine samples was assayed by means of a Nuclear-Chicago or Beckman liquid scintillation spectrometer system. Replicates of appropriate inulin- 14C solutions could be measured with an accuracy of +2.3 % (SD, n = 19). Transepithelial electrical potential differences were measured under the different experimental conditions with glass micrcelectrodes according to methods previously described (4 1, 42). The micropuncture data have generally been evaluated by plotting cation concentration against distal tubular length. Slopes of these relationships were calculated by the method of least squares, and standard tests using the i distribution were used for assessing the significance cf the regression lines and the significance of the differences between slopes. RESULTS This paper will be concerned with the assessment cf the distal tubular potassium transport system during widely 1193 varying patterns of tubular hydrogen ion transport. Most of the-data dealing with the acid-base status of the animals are presented in another paper (44). Table 1 summarizes data on mean glomerular filtration rates (GFR), plasma potassium levels, and fractional excretion rates of potassium. Overall GFR was found moderately decreased during respiratory and metabolic acidosis and during infusion of bicarbonate during mechanical hyperventilation. nspection of the range of inulin U/P ratios indicates that a similar state of moderate osmotic diuresis was achieved in all experimental groups. This makes it unlikely that unspecific effects of larger variations in tubular flow rate per se effect significant changes in potassium excretion. This is of importance since recent cxperimental evidence indicates that, in states of an initially normal or increased urinary potassium excretion, the augmentation of distal tubular flow rate by a variety of maneuvers significantly elevates ( secretion and thus achieves kaliuresis (2). n agreement with observations by others, plasma potassium levels were found to be reduced in respiratory and metabolic alkalosis (3, 21, 23, 35, 54, 59) and increased in situations of acidosis and exogenous potassium loading (3, 13, 15, 21). t is also apparent that a wide range of excretion patterns of K was achieved in the different experimental conditions. Compared to control K data, an increased fractional excretion rate was observed during states of metabolic and respiratory alkalosis. n contrast, the excretion rate of potassium was moderately decreased during acute respiratory acidosis but more dramatically so in metabolic acidosis. Elevating the arterial carbon dioxide tension reduces fractional potassium excretion rate in bicarbonate-loaded rats (compare 5 % sodium bicarbonate with 5% sodium bicarbonate + 15 % CO& w hercas the highest fractional excretion rates were achieved during; - the combination of metabolic and respiratory alkalosis. A summary of micropuncture data obtained in control rats (mean ph = 7.36 s+.12) is given in Table 2 and TABLE 1. Summary of glomeruhr fihation r&es, blasma 1. botassium 1 concentrations, cud fractionul excretion rates of potassium - Control 157c co2 Metabolic acidosis Hypervent Group 57c NaHC3 57c NaHC3 iv + Hyperven t 5yc NaHCOa iv + 15% CO2.15 M KC1 5yo NaHC3 iv +.15 M KC1 ml/min - kg 9.12 *o. 55 (3) 6.98 ho.5 (26) 5.31 *o.43 (22) 8.51 ho.47 (32) 8.38 zto.33 (34) 5.86 rto.43 (22) 7.42 zto.28 (22) 8.47 ho.27 (2) 8.41 zto.35 (28) U/P nulin 47.3 rt6.45 (3) 28.1 zt3.86 (26) 19.5 *.13 (22) 31.o h5.18 (32) 27.9 h3.11 (34) 24.2 U.99 (23) 17.2 *.35 (22) 34.2 h2.78 (2) 17.5 ko.96 (28) Plasma i,,meq/liter 4.16 *o..5 (3) 6.41 zto.19 (26) 5.86 *o.45 (22) 3.46 zko.3 (32) 3.4 *o.7 (34) 2.99 zko.9 (23) 4.79 zto.4 (18) 6.18 zto. 15 (2) 5.76 *.22 (28) Values are means & SE. Numbers in parentheses numbers of observations. U/P.36 zko.21 (28).2:) zk -17 (28).87 zko.12 (21).53 *o.24 (31).4:) zto.13 (32).83 *o.ogo (21).38 zto.16 (22).68 ~.25 (18).72 zto. 24 (29) indicate

3 1194 MALNC, DE MELLO ARES, AND GEBSCH TABLE 2. Potassium and indin concentrations of tubular jluid, plasma, and urine, and corresponding GFR ualues from rals on a control diet - Rat o LOG 5% P or D 4P 2 P 4 P, /SOP 3 D 2 D 25 D 26 D 2 D 8 D 7 D 76 D 9D 79 D 16 D 3 D 35 D 5 D 68 D 25 D D 39 D 83 D 1D 74 D 7 D 73 D ml/kg - min , nulin u/p o Potassium, TF meq/liter u/p Lot = location, P = proximal, D = distal, P = plasma, TF = tubular fluid. Fig. 1. A few proximal potassium ratios confirm our previous data in which values slightly but significantly below unity were regularly observed (41, 42). Across the early part of the distal tubule transepithelial concentration ratios below unity were consistently observed. The lowest localized value (.27) corresponds to a concentration of 1.17 meq/liter. Confirming work in a number of laboratories (12, 38, 45) distal ratios show a progressive increase as a function of tubular length with peak values approaching 4. at the late distal tubular level. On the right side of Fig. 1, potassium/inulin ratios are plotted as a function of tubular length. t is apparent that, at the early distal puncture sites, some 9-95 % of the filtered potassium has been reabsorbed. Potassium/inulin ratios show a significant increase from early to late distal tubular sites as demonstrated by a significantly positive slope (P <.1, correlation coefficient r =.76). Calculations indicate that at least some 86 % of the potassium in the final urine is derived from distal tubular secretion. Obviously this estimate represents a minimum value in view of the inaccessibility of the earliest part of the distal tubule for puncture. A comparison of ureteral urine potassium/inulin U/P and corresponding late distal ratios shows that the collecting-duct epithelium contributes at best in a very small way to the excreted potassium moiety. This indi- rectly derived conclusion is supported also by direct punctures of papillary collecting ducts (14) as well as by previous microcatheterization studies of single collecting ducts in another rodent species (32, 33). n general the distribution of potassium and potassium/inulin ratios is similar to previous data in Pmx. o/o Distal e U _ A _ m loo- Kh Control K [W/P] [] 4 O 8 o % Distal u FG. 1. Summary of potassium and potassium-to-inulin concentration ratios as function of distal tubular length from animals under control conditions. Regression line was obtained by least-square method. Area between broken lines represents &l SE (C.. Bliss, The Statistics of Bioassay. New York: Academic, 1952, p. 496). a 6b

4 ACD-BASE AND RENAL K SECRETON animals under control conditions (41). The only difference concerns the absence of significant net reabsorption of potassium along the collecting ducts in the present control group at the higher rate of fractional fluid excretion (mean U/P inulin ratio: 42), compared to the frequently observed significant net reabsorption of potassium along the terminal nephron segment in the earlier control group during the state of antidiuresis (mean U/P inulin ratio: 58). n Table 3, data from rats breathing a mixture of 15% C2-85% 2 are shown. Respiratory acidosis manifested itself by a low mean blood ph (7. =t.13) and an elevated level of arterial PCO~ ( 11.5 AZ 6.5 mm Hg). The graphic summary of potassium and potassium/ inulin ratios along the nephron is presented in Fig. 2. t is apparent that proximal and early distal tubular concentration ratios are similar to control values. This indicates the absence of important modifications of fractional potassium reabsorption along the proximal convolution and the loop of Henle. n contrast, late distal values show differences as compared to control conditions. n respiratory acidosis, late distal ( > 5 % distal tubular length) potassium/inulin ratios averaged.18 =t.3 whereas the corresponding value in control animals was.3 &.29. This difference is statistically significant (.1 > P >.1) and permits the conclusion that distal tubular potassium secretion, although still present (P <.1, =.74), is depressed during acute respiratory acidosis. Metabolic acidosis even more depresses distal tubular potassium secretion. Blood ph averaged 7.2 =t.16 TABLE 3. Potassium and i&in concentrations of tubular fluid, plasma, and urine, and corresponding GFR values from rats breathing 1.5% COZ-85% Rat Lot, % P or D 45 P : 3 : 65 P 4 65 P 7 6OP v P 42 D / 5 D j 33 D 1 29 D 3 ; 8 D i 39 D : 2 D /ND 29 D 4 (61D 25 D 9D 2 D 5 15 D 16 D 6 D t 7D 8D 1 ml/k - min LO nulin u/p P TF l Potassium, meq/liter - - u/p Ol oo sa % co, K [TFh] % Distal O m loo K /,, [TFh ] FG. 2. Summary of potassium and potassium-to-inulin concentration ratios as function of distal tubular length from animals breathing 15% CO~-85/~ 2. and the bicarbonate level was depressed to a mean of 9.55 A.36 meq/liter. Pertinent results are shown in Table 4 and Fig. 3. Potassium ratios remain low along the distal tubule and net addition of potassium along this part of the nephron is conspicuously absent. Notably, the progression of potassium/inulin ratios fails to increase ( =.6). nspection of the right part of Fig. 3 also shows consistently low late-distal potassium/ inulin ratios. The mean value of.13 s.23 is the lowest one of the various groups studied. Since some of the corresponding ratios in the final urine are still lower (mean values:.87 =t O.OlZ), some net reabsorption of

5 1196 MALNC, DE MELLO ARES, A4ND GEBSCH TABLE 4. Potassium and inulin concentrations of tubular fluid, plasma, and urine, and corresponding GFR values from rats in metabolic acidosis f 6 46D 38 D 8 7 D 8 D 9 ' 7 D 35 D :9OD a 8 D / 62 D 1 / 3 D j 43 D i 6 D j 2 D f D ;loo D / 7 D ml/kg. min 6P P P P P P 2.28 nulin Potassium, meq/li ter - u/p P TF potassium along the collecting duct may have taken place in some animals. n Table 5 and Fig. 4, data from rats undergoing mechanical hyperventilation are shown. Blood ph averaged 7.64 &.7 and mean arterial Pcoz was depressed to a level of 17.1 =t.42 mm Hg. The similarity of the origin of the regression lines relating the fraction of potassium present at the earliest accessible part of the distal tubule in control and hyperventilated rats indicates that potassium transfer during hyperventilation was not importantly modified along the proximal convoluted tubule and Henle s loop. n contrast, distal tubular potassium secretion was stimulated.,4 comparison with control distal tubular data shows: 1) a regression line of distal potassium/inulin ratios with a significant positive slope (P <.1, r =.91), 2) this regression line is steeper than the corresponding slope in control conditions, 3) late-distal potassium/ inulin ratios (.52 AZ.43) are also significantly higher than control values (P <.1). The effects of acute metabolic alkalosis induced by the intravenous infusion of sodium bicarbonate solutions are summarized in Table 6 and Fig. 5. Mean blood ph was 7.58 Z&.9 and the arterial bicarbonate level was elevated to meq/liter. Distal tubular potassium secretion is markedly stimulated. This is apparent from the steeper than normal progression of the slope relating fractional secretion of potassium to tubular length (P <.1, r =.87), and the significantly elevated late-distal potassium/inulin ratios (.79 =t.87). One aspect of the potassium transport pattern deserves mention. t concerns the consistent observation of net reabsorption of some secreted potassium at a nephron site beyond the distal tubular epithelium as demonstrated by the reduction of potassium/inulin ratios in the final urine. Provided that functional differences between superficial and deep nephrons do not play an important role, this observation is consistent with the notion that in metabolic alkalosis induced by bicarbonate loading the collecting-duct epithelium reduces the amount of potassium below that present at the late-distal tubular level. The combination of metabolic and respiratory alkalosis (mean blood ph: 7.84 &.13, mean plasma bicarbonate 34.8 & 1.3 meq/liter), achieved by intravenous sodium c % Distal U P 4 m lm Mo tobolic Acidosis K tlf/p] ws 8 % Distal U 2 4 a 8 roe [] -W ,,,-- ~ - ---a -Co He-- Q) --a- -9 t FG. 3. Summary of potassium and potassium-to-inulin concentration ratios as function of distal tubular length from animals in metabolic acidosis. t# 8 a oep

6 l ACD-BASE AND RENAL K SECRETON TABLE 5. Potassium and inuh concentrations of tubular fluid, plasma, and urine, and corresponding GFR values from rats undergoing mechanical hyperventilation - -- Rat, / Lot. cr, l PorD 5 i 7 /6OP : i 1 8 / ; 1 / / 86 D 7D 3 D 2 D 2 2 D 82 D 15D ; 27 D 38 D 5 / 7 D i 3 D! 87 D 35 D 6 :lood 1 D j 64 D 1 D 1 D 76 D ml/kg - min n.in u/p o o Potassium, meq/liter _ u/p P TF o A l bicarbonate loading during mechanical 1 hyperventilation, leads to one of the highest late-distal potassium/inulin ratios of the present series of experiments (.88 Z.24). Table 7 and Fig. 6 summarize the pertinent data. Similar to the other experimental conditions in which potassium secretion was present, the regression line of distal potassium/inulin ratios had a significant positive slope (P <.1, r =.95). This regression line is significantly steeper than that found under control conditions (P <.1). Again, some reduction of the distally secreted potassium moiety by reabsorption along the collecting ducts is apparent. The effect of a combination of metabolic alkalosis and breathing 15 % COZ-85 % 2 on distal tubular potassium transport is shown in Table 8 and Figure 7. Mean blood ph averaged 7.26 &.22, plasma bicarbonate 59.7 rfr 3.5 meq/liter. Fractional urinary potassium excretion was markedly depressed compared to that during acute metabolic alkalosis at normal levels of arterial PCO~. Late-distal potassium/inulin values (.51 &.6) are lower than those observed in animals given intravenous sodium bicarbonate alone, but still significantly elevated compared to control ratios at comparable nephron sites (P <.1). The regression line relating potassium/inulin 1197 ratios to distal tubular length is significantly positive (P <.1, r =.71). The last two experimental groups concern rats in which the effectiveness of a comparable intravenous potassium chloride load (9 peq/hr) on the distal tubular potassium transport system was assessed in control conditions and during acute metabolic alkalosis. Relevant data are summarized in Tables 9 and 1, and Figs. 8 and 9. As expected, and confirming previous observations (41, 42), acutely potassium-loaded animals respond with increased distal potassium secretion. This is apparent from w 2 la Prox. % Dis to. U 8. & la s i b( O.! % HypWvfbtilation K 8 [lf/p] K /n [] e-e Q -e-s---e o o O % Distal J..,,, elo loo ---e FG. 4. Summary of potassium and potassium-to-inulin concentration ratios as function of distal tubular length from animals undergoing mechanical hyperventilation.

7 l 1198 TABLE 6. Potassium and inukn concentrations of tubular fluid, plasma, and urine, and corresponding GFR values, from rats receiving 5 % sodium bicarbonate intravenously nulin Potassium (meq/l) Rat -. ml/kg min 4P P P P P P P P D ; 64D 72 D 9D 1 D 8 D 9D 3 D 3 D 5 D 3 D 28 D 68 D 33 D 4 44 D 74 D 1 D 23 D 5 38 D 68 D 2 D 33 D 96D 7 46 D, 35 D / 15D u/p P TF l Ol u/p the significantly elevated late-distal potassium/inulin, ratios of.42 &.6 in K-loaded animals as compared to a value of.3 &.29 in control animals (P <.1, r =.85). Administering potassium chloride to sodium bicarbonate-loaded animals results in very much elevated late-distal potassium/inulin ratios (.9 rt.95). Calculations indicate that net secretion along the distal tubule, expressed in absolute amounts, was dramatically augmented during metabolic alkalosis. Corresponding values are : 16.4 peq/min kg in rats receiving KC1 at normal bicarbonate levels, and 47.2 peq/minkg in rats made acutely alkalotic by sodium bicarbonate-loading (see also Fig. 1 and Table 12). Thus, it appears that alkalosis sensitizes the distal tubular secretory mechanism to potassium loading. Again, a curious and as yet unexplained feature of these experiments is the different behavior of the collecting-duct system in the two experimental conditions. A significant reduction of the amount of potassium present at the late-distal tubular level is apparent in bicarbonate-loaded animals receiving potassium chloride, but is absent in potassium-loaded control animals. n the latter, the urinary potassium/inulin ratios exceed latedistal values, implyin, e continued net secretion along the MALNC, DE MELLO ARES, AND GEBSCH % Distal U a> loo 5 O/o No HGO, O/o Distal U la3 [lf/p] FG. 5. Summary of potassium and potassium-to-inulin concentration ratios as function of distal tubular length from animals receiving 5y sodium bicarbonate intravenously. collecting ducts, whereas corresponding ratios are lower in the urine of animals made alkalotic. This confirms the observation made in rats undergoing sodium bicarbonate diuresis. n an effort to examine whether net secretion of potassium in the present experimental conditions occurs against or down an electrochemical potential gradient, the electrical potential difference across the distal tubular epithelium was evaluated with respect to the observed mean transtubular concentration differences of potassium ions. Since secretion of potassium is a function of the second half of the distal tubule, late-distal tubular concentration ratios and mean electrical potential differences were compared.l l Wright (7) has recently shown that the transepithelial electrical potential difference across the early distal tubule is lower than that across the late distal tubule. The range of distal transepithelial potential differences in the present study is quite similar to late distal values in the study of Wright. This indicates that most of our measurements were obtained from the second half of the distal tubule. a A

8 ACD-BASE AND RENAL K SECRETON 1199 TABLE 7. Potassium and inulin concentrations of plasma, and urine, and corresponding GFR values from rats receiving 5 5% sodium bicarbonate and being mechanically hvberventilated Rat LOG % P or D 57 P 55 P 73 D 72 D 81 D 83 D 77 D 83 D 36 D 71 D 27 D 85 D 21 D 26 D 18 D 9 D 89 D 18 D ml/kg emin nulin Potassium, meq/liter u/p P TF lO Using the Nernst equation, a limiting value of distal transtubular concentration ratios of potassium for the corresponding transepithelial electrical potential difference may be obtained. The latter value is the highest possible concentration that could be achieved if potassium ions were distributed solely according to the electrical driving force. nspection of Table 11 indicates that, under all experimental conditions, the observed potassium ratios were lower than the calculated ones. A small reduction of transepithelial electrical potential differences is apparent in acidotic animals. According to the comparison between calculated and observed transepithelial concentration differences, it is apparent that net entry of potassium ions under all conditions studied proceeds down an electrochemical potential gradient. potassium during acute acid-base disturbances are shown in Figs. 1 and 11. Figure 1 summarizes the estimated absolute distal tubular secretion rates, and Fig. 11 gives a summary of the relationship between fractional distal tubular potassium secretion and the mean plasma potassium level during control conditions and during various acid-base alterations. From inspection of Fig. 11 it is apparent that the stimulation of distal tubular potassium secretion is clearly not solely related to the plasma potassium level. t is apparent that intravenous administration of isosmotic potassium chloride, during control and alkalotic conditions, leads to both marked hyperkalemia and a significant stimulation of distal tubular potassium secretion (see also Fig. 9). This observation confirms the results of clearance studies by Toussaint and Vereerstraeten (65). These authors demonstrated that in experiments in which the plasma potassium levels progressively increased during exogenous potassium loading, the magnitude of urinary O a.3 POX. % Distal U - b b t loo 5 % No HCO, + Hypwent ilation B [jb,p] ~ O b % - Qistal, Q loo DSCUSSON The present series of experiments lends strong support to the view that the distal tubular epithelium is the nephron site where the most important acid-base dependent changes in tubular potassium transport occur. This is evident from inspection of Figs. l-8, which shows that the fraction of filtered potassium present at the earliest distal tubular puncture site is small and varies but little despite widely differing secretion patterns of potassium along the distal tubule. The secretory contribution of the distal tubule shows large differences depending on the metabolic situation. n general, potassium secretion is enhanced during alkalosis and depressed in acidosis. A summary of the absolute secretion rates during the different experimental conditions is given in Fig. 1. The widely differing distal tubular transport patterns of [] Q) FG. 6. Summary of potassium and potassium-to-inulin concentration ratios as function of distal tubular length from animals receiving 57 sodium bicarbonate intravenously and being mechanically hyperventilated. /-

9 l min me ---B-Y MALNC, DE MELLO ARES, AND GEBSCH TABLE 8. Potassium and inulin concentrations of plasma, and urine, and corresponding GFR values from rats receiving 5 5% sodium bicarbonate and breathing 15% COS-85% 2 Rat LOG % P or D 55 P 4 P 55 P 5 P 45 P D 6 D 79 D 21 D 25 D 7 D 45 D 8 D 84 D 91 D 51 D 6 D ml/k nulin u/p P TF Potassium, meq/li ter z u/p potassium excretion was directly related to the plasma potassium level and the duration of the potassium infusion. The latter observation is consistent with the interpretation that continued uptake of potassium by the distal tubular epithelium during the development of a positive potassium balance constitutes the initial event triggering enhanced potassium secretion. n agreement with Toussaint and Vereerstraeten (65), we observed that metabolic alkalosis sensitizes the kidney to potassium loading. nspection of Fig. 11 also indicates that marked changes in the plasma potassium level resulted from the various experimental procedures in the absence of exogenous potassium loading. Distal tubular secretion rate and plasma potassium levels may even be inversely related. Thus, respiratory and metabolic acidosis both led to a significant rise in the plasma potassium level whereas both respiratory and metabolic alkalosis induced a fall in plasma potassium. At the same time, acidosis generally depressed distal potassium secretion and alkalosis significantly stimulated the secretory mechanism. Changes in PCO~, superimposed on metabolic alkalosis, have effects similar to those observed under control conditions. Although starting from a different base level of secretory activity, breathing 15 % CO2 suppresses distal potassium secretion despite a rise in the plasma potassium level. Further alkalinization by mechanical hyperventilation, on the other hand, enhances distal secretory rate beyond that achieved by 5 % sodium bicarbonate administration alone, despite a further drop in the plasma potassium level. Thus, in experiments involving the induction of acute respiratory and metabolic acidosis and alkalosis, increased distal tubular potassium secretion is generally associated with low plasma potassium levels, whereas the opposite holds for states of reduced potassium secretion in acidosis. Several lines of evidence indicate that potassium transfer into body cells increases with a rise in arterial ph (3, 8, 2 1, 57, 58, 63, 64, 65). According to such evidence, the drop in plasma potassium level during metabolic and respiratory alkalosis, also noted in, the present experiments, reflects significant cell penetration of potassium, whereas the shayp rise in plasma potassium during acidosis is due to the release of potassium from tissues upon acidification of the body fluids. f these results of ph-dependent shifts of potassium between extra- and intracellular body fluid compartments are extrapolated to distal tubular epithelium it is apparent that the observed modifications of distal secretory rate may be related to changes in distal tubular potassium concentration. A rise in distal cell potassium by increased peritubular uptake in alkalosis, and the reverse event in acidosis, could be responsible for stimulating secretion in alkalosis and depressing it during acute acidosis. According to this view, cellular changes in distal tubular potassium content or concentration would be fundamen- O J Rox. % Distal U - 6 t m 4b cp 5 % No HCO, + s,% CO, K [. K/, [] --mm % Distal U.. 4 l FG. 7. Summary of potassium and potassium-to-inulin concentration ratios as function of distal tubular length from animals receiving 5% sodium bicarbonate intravenousy and breathing 15% CO2-85% 2.

10 ACD-BASE AND RENAL K SECRETON TABLE 9. Potassium and inuh n concentrations of plasma, and urine, and corresponding GFR values from rats receiving.15 M KCZ intravenously - Rat LOG % P or D ml/kg -min nulin Potass. 1 urn (meq/liter) u/p P TF H/n U/P D D D D D D D D D D D D D D D D D D D *TABLE 1. Potassium and inulin concentrations of tubular fluid, plasma, and urine, and corresponding GFR values from rats receiving.15 M KC and 5% sodium bicarbonate Rat LOG % P or D ml/k l min P P P P D D D D D D D D D D D 11.o 8 D D D D D D 6.74 nulin Potassium, meq/liter u/p P TF l oo Ol tally mvolved and responsible for the 1. aajustment of the potassium secretory rate to changes in the acid-base status of the organism. 121 The observation that an acute exogenous potassium load is more effective in increasing distal potassium secretion at an alkaline blood ph (see Figs. 9 and 1) than at normal blood ph is also best interpreted by an increased rate of uptake of potassium ions by distal tubule cells in alkalosis. Supportive evidence for this view is the increased cell penetration of potassium as well as the enhanced effectiveness of a potassium load in promoting kaliuresis which was observed by Toussaint and Vereerstraeten in C *o o,os,. S M KCL K [lf/p] % Distal U % Distol U FG. 8. Summary of potassium and potassium-to-inulin concentration ratios as function of distal tubular length from animals receiving.15 M KC1 intravenously.

11 , Prox.., % Distal 2, 5 % No Hco, t.15 M KCL K kf/p K /n [] loo CP o 8 % Distal U 4 * loo / O 4 - & ?@ FG. 9. Summary of potassium and potassium-to-inulin concentration ratios as function of distal tubular length from animals receiving.15 M KC.l and 5% sodium bicarbonate intravenously. MALNC, DE MELLO ARES, AND GEBSCH absorption along the nephron that distal volume flow rate is ultimately augmented ( 1, 39, 67). Similarly, increased excretion of chloride in ammonium chloride-induced acidosis (56), or enhanced ketoacid excretion during diabetic acidosis (35) would all be expected to result in increased fluid delivery into the distal tubule. An increased delivery of tubular fluid to the site of potassium secretion has recently been shown to result in stimulation of distal tubular potassium secretion (2) and a similar mechanism may be responsible for the observed kaliuresis in chronic acidotic conditions. n the present series of experiments, both distal intratubular ph and the progression of bicarbonate concentrations along the distal tubule were measured under the different experimental conditions. Although discussed in detail in another paper (44), some pertinent aspects of the relationship between distal in situ ph and distal tubular bicarbonate reabsorption, on the one hand, and net potassium transport across the distal tubular epithelium, on the other, will be considered. Figure 12 illustrates the relationship between intratubular in situ ph and distal secretory activity. t is apparent that an increase in intratubular ph is generally associated with stimulation of distal tubular potassium secretion. However, it is also apparent that other factors than intratubular ph can markedly affect distal tubular potassium secretion. This follows from the observation that auite different secretion rates may obtain at al- PEq /min. kg metabolic alkalosis of similar degree (65). These observations are also consistent with the view that potassium uptake into secretory distal tubule cells is ph dependent. t should be noted that the clear-cut relationship which we and others have observed between acute acid-base alterations and both plasma potassium levels and urinary potassium excretion is not regularly seen in acid-base disturbances of longer duration. For instance, normal plasma potassium levels and a slightly elevated excretion rate of potassium have been noted during chronic respiratory acidosis ( 1, 39, 49), and the situation is similarly complicated during chronic metabolic acidosis in which kaliuresis may be observed (35, 56). t is likely that a number of secondary, time-dependent responses are responsible for the differences in potassium excretion between acute and chronic acidosis. Thus, diminished tubular chloride reabsorption in chronic respiratory acidosis develops with a time lag and may result in such redistribution of fluid re- FG. 1. Summary of absolute rates of distal tubular potassium secretion under different experimental conditions. Rate of distal tubular potassium secretion was calculated according to equation: 1 (K/inulin x) - (K/in&n x) K, = 1% 2% distal distal l GFR*Px where K, is amount of K secreted in peq*kgl min-l; GFR and Px represent mean values of individual experimental groups, and K/inulin values are taken from the individual slopes of Figs. l-9. Rate of distal tubular bicarbonate reabsorption was similarly calculated from values taken from individual slopes relating transepithelial inulin concentration ratios to distal tubular length. Mean distal tubular bicarbonate concentrations were used since no progressive concentration change of this ion was observed in the different experimental conditions (44, 66).

12 ACD-BASE AND RENAL K SECRETON 123 TABLE 11. Mean distal transepithehal potential dz~erences, expected ratios and mean observed late-distal potassium ratios Control 15% co2 Hypervent 5% NaHCOs Metabolic.15 M KC1 Group Distal PD, mv acidosis Expected -5.2 =t.94 (152) rt 1.17 (53) rt 1.78 (32) * 2.4 (33) =t 1.93 (44) & 1.8 (3) 5. PD values are means =t SE. Numbers of observations are given in parentheses. A K/N NaHC3 Hyperventilatio NaHCO, + a h Hyperventilation \ \ \ 3 $ji ---e P+ NaHC3 + KC \ \ \ \ \ \ \ \ \ \ PLASMA K bx Metabolic y Acidosis 6 7 O!fGFd FG. 11. Summary of fractional distal tubular potassium secretion as function of plasma potassium level, during control conditions, various acid-base disturbances, and during exogenous potassium loading in control and alkalotic conditions. Fractional secretion rates were calculated as difference of K/inulin x ratios at 1% and 27 distal tubular length, respectively. most identical intratubular hydrogen ion concentrations. Thus, at very similar intratubular ph values, an elevation of the Pcoa level (compare points Na HCOS and Na HCOS + 15 % CO2, or control and control + 15 % CO,), or the induction of metabolic acidosis, depresses distal tubular potassium secretion. t is reasonable to infer that with the fall in arterial blood ph in acidotic situations the intracellular ph of distal tubule cells was also more acid when animals were exposed to high Pco:! levels or given ammonium chloride. n particular, since CO2 is freely diffusible, a significant rise in tubule cell PCO~ and hydrogen ion concentration would be expected. These experiments thus provide strong evidence that relative intracellular acidification, independent of the intratubular ph level, can effectively suppress distal tubular potassium secretion. The observation of quite dissimilar values of distal tubular potassium secretion at very similar intratubular hydrogen ion concentrations is taken to indicate that tubular fluid ph is not the most important and most direct determinant of the distal tubular potassium transport system. As shall be discussed subsequently in more detail, the distal tubular fluid ph is a reflection of the relationship between the rate of hydrogen ion secretion, presumably related to tubular cell ph, and distal tubular buffer (bicarbonate) load. Accordingly, distal tubular ph is not only related to 2.21 distal tubular cell ph, which is the most likely factor re sponsible for regulating distal tubular potassium secretion A large amount of experimental and clinical data has been 3.55 interpreted to suggest the competition of H and K ions for a common, presumably carrier-mediated transport mechanism (4, 6, 7, 47). Much of the evidence in support of this thesis is based on the frequently observed, apparently reciprocal, relationship between urinary acidity and potassium excretion. Fundamental to this thesis is the assumption that urinary acidification (low ph) is paralleled by an increase, and urinary alkalinization (high ph) by a decrease in the rate of hydrogen ion secretion at the site of potassium secretion, i.e., at the distal tubular level. To gain insight into the possible relationship between distal tubular hydrogen and potassium secretion, absolute rates of bicarbonate reabsorption and of potassium secretion were measured and compared during various acid-base disturbances. The assumption is made that, inasmuch as bicarbonate ions constitute, besides phosphate and ammonia, a major fraction of the distal tubular buffer load, distal tubular bicarbonate reabsorption reflects rate of hydrogen ion secretion, or at least a very important fraction of this transport process. This view is supported by the presence of a significant distal tubular disequilibrium ph which persists even during marked alkalinization of distal tubular fluid (25, 53, 66). This finding demonstrates continuation of active hydrogen ion secretion even during metabolic alkalosis (25, 53,-66). Table 12 provides a summary of pertinent data on distal tubular potassium and bicarbonate transport. t is obvious that there is not a consistent reciprocal relationship be- A K /N T-F/p F.3 - * Control.1 - m Control + 5 %C2 Meto bolic Acidosis a NaHC3 Hyperventilation m NaHC3 + S % CO DSTAL ph NaHC3+ Hyperventilation FG. 12. Relationship between intratubular ph and fractional distal tubular potassium secretion (ph data from refs. 44 and 66).