Neal S. Bricker, Division of Biomedical Sciences, University of California, Riverside, California
Edgar Sanclemente, Loma Linda University, Loma Linda, California
Laura Zea, Loma Linda University, Loma Linda, California
Stewart Shankel M.D., Division of Biomedical Sciences, University of California, Riverside, California
Studies were performed on Sprague-Dawley rats to examine the nature and characteristics of the process widely known as "K sequestration." Large loads of KCl (from two to six times the ECF content of K MB2-6) were infused IV over 10 minutes. Over 50% of all loads was transported from the ECF into the ICF by the 10th minute while only 10% was excreted in the urine. No rats died of hyperkalemia. K sequestration is generally believed to be due to Na, K-ATPase activation. However, the K transport (i.e. the rate of transport and the slope of this rate versus time) from the ECF into ICF was unaffected by two potent ATPase inhibitors (digoxin and ouabain). It also was not altered by insulin. The data suggest that the life preserving mechanism of early protection against rapid K entry into the ECF (i.e. The First Responder) is under the aegis of a biologic control system which is independent of the whole body K control system. It is inborn, activated within 90 seconds of K entry into the ECF, on call continuously and independent of Na, K-ATPase and insulin activities.
The Potassium Control System(s)
For many years our laboratory has focused attention on the control of the extracellular fluid volume and its key solutes in health and in advancing renal disease (2,3,4,34,35,36). The present paper deals with the potassium control system. We have challenged rats with a range of K loads varying from moderate to massive in amount, delivered intravenously by constant infusion, in precisely ten minutes. We monitored closely the response to the K loads under numerous experimental conditions.
Both the first responder (see below) and the long-term day-night component of the K control system make major contributions to the protection against abnormalities of potassium distribution and concentration. They also participate in the preservation of the constancy of potassium homeostasis in health and in many disease states. The two systems act in a complimentary fashion. But are they independent components of a double-tiered system? Or are they separate entities with different controls, different modes of operation, and different responses to specific drugs, hormones and messengers?
We have coined the term "First Responder Component" to emphasize the fact that the primary role of this component is to protect the organism against acute K loads, both small and large. Within approximately one minute of the start of each intravascular K infusion (ranging from 2 to 6 times the total ECF K content) the first responder begins to affect the relocation of extracellular fluid K into the intracellular fluid compartment. This process requires that K traverse a double-layered lipid membrane that encases most (or possibly all) living cells and subcellular organelles. This bilipid membrane rejects the transcellular traffic of K except through the K channels, co-transporters, or by the Na, K-ATPase pump.
The long- term component of the K control system is the kidney (17,37). It is the organ that preserves the dictum that "what goes in must come out." The renal excretion of K involves a major contribution of Na, K-ATPase, whereas the data presented in this manuscript suggest that this enzyme is not the primary engine that transfers over half of the 10 minute K loads from the ECF into the ICF.
While the two systems are essential for life, our view is that the First Responder Component is a separate entity designed for the protection of the organism against acute, potentially fatal, hyperkalemia while the major long-term K control system is charged with the maintenance of external K balance and overall K homeostasis.
The present paper addresses the constancy of that portion of total body K (approximately 2%) that resides in the extracellular fluid. Our studies differ from previously reported studies in the following areas. The K loads are intentionally large and the time of infusion very short in an effort to define the characteristics and the limitations of the First Responder. All studies were done in rats while awake, no barbiturates were used, the loads of K given were much larger and given more rapidly over 10 minutes, some rats received 3 consecutive loads of K MB, followed by MB3 and MB4 while other rats were pretreated with either insulin, digoxin or ouabain. Balance studies were done which allowed us to calculate the intracellular uptake of K+ and the cellular polarity.
The first responder component operates virtually independently of the kidneys and it must be "on call" continuously so as to respond to any potentially threatening dose of K within 60 to 90 seconds of its entry into the extracellular fluid. This short-term action is often referred to as "K sequestration" (23,32). We have termed this extremely important, protective process the First Responder Component of the Potassium Control System.
In the resting state the distribution of total body K ions is weighted heavily towards an intracellular location maintained primarily by Na, K-ATPase which pumps 3Na+ out of the cell and 2K+ into the cell. The loss of 3Na+ and the gain of only 2K+ produces a negative charge. An electrical potential gradient across the cell membrane of -80 mV in the cell, making the interior of the cell negative to the ECF (5,16,22,31).
In the resting state, the constancy of the K concentration of the ECF must be preserved in the face of a steeply downhill chemical gradient (cell to ECF). This is largely offset by a negative electrical potential gradient for K (25) and by the movement of K into the cell through the action on their Na, K-ATPase pumps (25). Although the permeability of the skeletal muscle epithelial cells to Na is quite small (approximately 1% of that for K), and there are no specialized Na+ channels (ENaC) in skeletal muscle (16), some Na ions do enter the ICF through non-ENaC Na+ channels and possibly through gap junctions. The current view is that this influx of Na resets the activity of Na, K-ATPase in the cell membrane and that this enhanced activity will increase the rate of pumping of three Na ions (from the cell interior to the ECF) and two potassium ions (from the ECF into the ICF). The Na ions pumped out will equal the Na ions that entered the cell and thereby restore Na balance. The two K ions have a different destiny.
Within 60 to 90 seconds of the start of the intravenous infusion of any of the five acute K loads, a series of potentially life-saving events is initiated. By the end of the 10th minute, over half of all the K infused will have been transferred into the ICF, irrespective of the magnitude of the load and of the impermeability of the cells to K. It is generally believed that Na, K-ATPase is the ultimate vehicle for the transfer of acute K loads into the cells (4,23,32).
A principal purpose of the present studies is to present data that appear to refute the primary role of Na, K-ATPase in the transcellular (ECF→ICF), life sustaining role of the First Responder component. A second goal of the studies is to present data that indicate that the First Responder component and the 24-hour day-night K control system are separate entities.
MATERIALS AND METHODS Definition and Derivation of Terms
[K]0 = plasma potassium values expressed either as µEq/ml or mEq/L.
U KV = K excretion values were determined by measuring the total urine volume gravimetrically (in tared tubes) during each timed period. The urine K concentration was measured on a 40 µl aliquot of the pooled sample and the value (in µEq/ml was multiplied by the sample volume (in ml) to obtain total K excretion.
ΔUKV = UKV for each intra-load period minus the mean preload control (value) for UKV, all in µEq/min.
ΣΔUKV = the sum of ΔUKV (per period) from the onset of the KCl infusion through the last intraload period.
ECF = is equal either to the inulin space for each period or 20% of body weight (in ml).
ECFK = ECF volume x the plasma K concentration.
ΔECFK = equals the cumulative value for ECFK during a given number of intra-load periods. The control value for ECFK is subtracted from each value.
ICFK = 40% of body weight in ml x ICFK concentration in mEq/l
ΔICFK = the load of K infused minus (ΔUKV + ΔECFK).
[K] i = The ICFK concentration for individual periods. (This calculation is developed further in the text).
ICFK mass = Ki x ICFK volume. Any K transferred into the ICF is added to or subtracted from the prior value for ΔICFK.
[K]0/[K]i = This calculation requires that the values, one measured ([K]0], one derived ([K]i), be contemporaneous.
The Calculation of Serial Values for Intracellular K Concentration [K]i
The control value for [K]i is assumed to be 140 mEq/L (29). To obtain [K]i values during the intraload periods also requires a value for intracellular water content. In this example, 40% of body weight is used for this term (31). In a 200 gm rat, the resting K mass thus equals 140 x 80, or 11,200 μEq.
To obtain [K]i during each consecutive infusion period, the value for ΔICFK (in μEq) is added to (or subtracted from) the prior value for ICFK mass and the new value is divided by % of body weight (the value for ICF water). Because the ECF volume (inulin space) remained relatively constant during the 10-minute infusion periods (see Table 1), it is assumed that the ICF volume also remained relatively constant in the normal and remnant rats (18).
We would have preferred to weigh muscle samples and directly measure intracellular Na and K+ and intracellular pH at the conclusion of the IL-10 period, however, all of our studies were continued for another hour. See accompanying paper (Shankel et al. Part II Potassium Control System).
The Experimental Model
Studies were performed on female Sprague-Dawley rats weighing 200-225 gm. Each rat was housed in an individual cage in a vivarium in which temperature and humidity were held
constant and the day-night cycle was reversed (lights on from 9:00 p.m. to 9:00 a.m. and off from 9:00 a.m. to 9:00 p.m.). The 24 hour intake of K+ (as KCl) was 3.0 mEq dissolved in 5.0 ml of demineralized water, and administered by gavage, half at 8:00 a.m. and half at 6:00 to 7:00 p.m. The commercial rat chow contained less than 1 mEq of K per kg, and, along with demineralized water, was allowed ad libitum.
Finally, the total population of test animals was divided into three groups:
a) Rats with two normal kidneys
b) "Remnant rats" in which approximately 70-85% of one kidney was infarcted by ligating second and third order branches of its renal artery and in a separate procedure, the contralateral kidney was removed surgically.
c) Anephric rats in which both kidneys were removed 16 hours before initiating an acute study.
The Standard Protocol
The general design of the protocol was as follows: All animals were fasted overnight and then under a light plane of ether anesthesia, two jugular veins and both femoral arteries were cannulated. The venous catheters were PE 10. One was used to infuse the sustaining infusion of 0.77 mM NaCl at 0.02 ml/minute; the other was used to infuse the KCl loads and any drugs. The femoral catheters were Teflon tipped silastic tubes (internal diameter 0.02 mm); one was heparinized and used to obtain serial blood samples. The other was attached to a transducer to measure online blood pressure. With the exception of the anephric rats, the urinary bladder was cannulated with a PE-50 tube through a puncture wound in the anterior abdominal wall. Four holes were cut in the distal end of the tube to facilitate complete collection of urine samples and a ligature was placed around the urethral orifice.
The time under anesthesia rarely exceeded 30 minutes and most rats were awake within three to five minutes. Each rat then was transferred to a specially constructed bivalved Lucite restrainer. At the time of the transfer, the sustaining solution was started and continued throughout the remainder of the study. An equilibration period of ninety minutes was allowed to permit the rats time to recover from the surgical procedures and the anesthetic.
Blood and urine collections were obtained at the beginning of the sustaining infusion and at 10 minute intervals during the control period. After an MB K load was initiated, blood and urine samples were collected at 1 min (if possible), 2.5, 5.0, 7.5 and 10 minutes.
Remnant Rat Studies
Studies were performed on 22 remnant rats. GFR at the time of study ranged from 0.17 to 1.03 ml per minute. The animals were maintained under the same conditions as the normal rats and were fed 3 mEq of K per 24 hours for at least three days before an acute K loading experiment was performed. The number of rats in each MB (MB2-MB6) subgroup varied from three to seven. Blood and urine collections were as described for the normal rats and the experimental procedure was identical to that employed in the normal rats. The results mirrored closely those described in the normal rats with the exception of the fact that the mean value for the renal excretion of K during the ten-minute K infusion was only 3.85% of the various K loads in contrast to 9.74% in the normal rats.
Anephric Rat Studies
Six normal rats were selected for the anephric studies. They were fed and maintained under the same conditions as the normal and remnant rats. Sixteen hours before initiating the acute K loading studies (MB-2), both kidneys were removed surgically. The procedure was performed under light ether anesthesia and was associated with minimal blood loss (estimated to be 0.5 ml or less).
All of the rats in the normal, remnant and anephric groups survived. If they had had no First Responder Component, the plasma K concentrations at the end of the 10 minute infusions in the normal rats receiving MB 4, 5 or 6 loads would have been at least 20, 24, and 28 mEq/L, respectively
Inulin Space (Table 1)
In a number of rats, ECF volume was measured sequentially, as the inulin space. The method used in the normal and remnant rats was as follows:
The control value for ECF volume was assumed to be 20% of body weight. A dose of 0.4 μCi of 14C or 3H inulin was infused in one ml of 0.77 mM NaCl at the time of initiating the sustaining solution. In the normal rats, at the conclusion of the 90-minute equilibration period, the rates of infusion and excretion of radioinulin were equal. In addition the volume of the sustaining infusion and the rate of urine flow were closely comparable.
Inulin space (Table 1) was calculated as follows: At the end of the last control period, the volume of inulin space in the ECF was calculated as the CPM/ml of plasma x 20% of body weight in ml. Inulin space, during each succeeding period was calculated as the previous inulin content plus the amount of inulin infused during each period, minus the amount of inulin excreted in the urine during the same period divided by the contemporaneous plasma value for 14C or 3H in CPM/min/ml.
In the anephric rats, the initial priming dose of radioinulin was infused, but no further inulin was administered. In these animals, the values for inulin space were approximately 10% higher than in the normal and remnant rats (see Table 1). We attribute this to the fact that for 16 hours post-nephrectomy, the anephric rats had free access to water; they were infused with 1 ml of .77 mM NaCl to deliver the inulin; and 3.2 ml of .20 mMol NaCl during the 10 minute K loading phase. Blood loss during the three control periods and four intraload periods totaled 1.4 ml.
In representative rats from all three groups, no inulin was found in the pleural or peritoneal cavities or in the feces; and the fecal K content was the same at the completion of the study as in the pre-load periods.
The Five K Loads
As noted earlier, the individual amounts of K (as KCl) were based on the calculated total quantity of ECFK in the basal state. The latter is equal to the product of the value for K (in mEq/l) times the volume of the ECF in ml (equal either to the inulin space [Table 1] or to 20% of body weight). This value was termed the Basal K Content.
The experiments were performed using one of five multiples of the basal K content and were termed "Multiple of Basal or MB." The K loads employed were MB-2; MB-3; MB-4; MB-5; and MB-6.
All of these studies were approved by an AUCU board at UCLA and previously similar studies on rats using these same techniques were approved at Washington University and Albert Einstein College of Medicine in New York.
The Distribution of the Infused K Loads (MB-2 through MB-6) in 27 Normal Rats (Table 3)
All K loads were dissolved in 3.2 ml of 20 mM NaCl and infused at a rate of 320 µl/minute throughout the 10 minutes. All K loads were delivered in the same volume and at the same rate.
In Table 2A, the % of the five K loads excreted in the urine during the 10-minute K infusions ranged from 7.29 (MB-6) to 13.4 (MB-2). The % of the load excreted diminished with increasing loads of K. The mean value for all 27 rats was 9.74 (± 2.07).
Thus the kidney's contribution to the very early protection against potentially dangerous elevations of plasma K concentration is minimal. Indeed, in the remnant rats, the urinary value for the % of the five loads excreted was 3.8 and in the anephric rats the value was zero.
In Table 2B the % of the five loads retained in the ECF at the conclusion of the 10-minute infusions ranged from 33.5 to 40.7. There was an increase from MB-2 through MB-5; but the value dropped in MB-6. For the group, the mean value was 36.8%.
In Table 2C, the % of the respective loads calculated to be in the ICF at the conclusion of the 10-minute K infusions ranged from 50.5 to 61.8 with a mean of 54%. There was no correlation between the load, expressed either as MB or as μEq, and the % calculated (in the same units) to be in the ICF.
In summary, in examining all three parameters, the % of loads MB 2-6 excreted fell as the load increased; the % retained in the ECF increased in loads MB 2 through 5; but with the MB-6 load, the percent retained in the ECF fell. The % of the loads entering the ICF averaged 54.0% at the end of the infusions.
K loads vs. Time: Kinetics
A. Δ[K]0 vs. Time: A Linear Function
Fig. 1 depicts 10 plots of Δ[K]0 vs. time in minutes from 2.5 minutes through the tenth minute of the intra-infusion periods in both normal and remnant rats. Regression analysis reveals a highly linear pattern for each of the ten plots (R2 = 0.99 to 1.0). Moreover, the slopes of the functions increase progressively from the MB-2 load through the MB-6 load. Comparison of the plots of the normal with remnant rats shows good agreement at each load. Thus, the loss of over 70% of the nephron population did not sever the relationship between [K]0 and time when both normal and remnant rats received the same K loads.
B. ΔECFK vs. Time
The constancy of ECF volume (inulin space, Table 1) in normal rats receiving KCl loads ranging from MB-2 through MB-6, coupled with the linearity of [K]0 vs. time, at each of the loads, requires that the product of ECF volume times [K]0, at any point in time fit on a linear regression line. Therefore the slopes of this family of regressions increased as the loads rose.
C. ΔICFK vs. Time
Fig. 2 depicts the regression of ΔICFK on time (2.5; 5.0; 7.5; and 10.0 minutes) in normal rats receiving 10 minute K infusions ranging from MB-2 through MB-6. All of the functions are linear and again, the slopes increase from 20.5 to 69.7 with the increments in the MB value from MB-2 to MB-6.
The resting values for [K]O ranged from 3.1 to 4.6 mEq/L (mean 4.12 ± .53 mEq/L) and there was no relationship between GFR and the resting [K]0 (see Figure 1).
As noted above, in all five remnant groups during the K loading, the values for [K]O increased in a linear fashion when plotted against time. At the end of the ten-minute infusions, [K]O values ranged from 6.84 to 14.3 mEq/L. The relationship between [K]0 and ΔICFK was also linear over the range of K loads (312 to 943 μEq) and the values for ΔICFK at 10 minutes, expressed as a percentage of the absolute loads, ranged from 45.5 to 62.5 with a mean value of 53.8 (± 5.76 S.D.).
In summary, the reduction in GFR in this group of rats ranged from moderate to severe. However, on the same K intake per 24 hr as in the normal rats, the control values for [K]O were not affected by the nephron loss. Indeed, the only difference between the two groups was the lower percentage of K excreted in the urine by the remnant rats following the K load.
In Table 3, detailed results are presented for six anephric rats, each receiving an MB-2 load of KCl. The resting values for [K]O ranged from 4.26 to 5.34 mEq/L and in 4 of the 6 rats, the values for [K]O were 5.2 or greater. The mean value was 4.97 mEq/L.
The same anephric rats exhibited an unanticipated finding in the plot of ΔICFK vs. time. In all six studies the regression lines were linear with R2 values of 1.0. But the one and 2.5 minute values for ΔICFK were uniformly negative. The mean value for the six one-minute values was -55.2 (± S.D. 9.65 μEq) and the mean value at 2.5 min was -31.9 (+ S.D. 14.4) μEq. Thus in these animals there was apparently net K efflux (ICF U+02192 ECF) during the first 2.5 minutes of the KCl infusions. Yet all negative values fell directly on the respective regression lines.
Based on the foregoing findings in the anephric rats, the possibility was examined that the earliest values for ICFk, in the normal and remnant rats might also be negative. In recalculating approximately 200 experiments we found negative values for ΔICFK in 20 studies at 2.5 minutes, in all 15 studies where 1 minute samples were obtained and in all eight studies where 30 second samples were obtained.
Several modifications were made in the basic protocol in an effort to examine certain properties of the First Responder.
Doubling the Time of the Intraload Period of KCl Infusion
Two consecutive MB-3 loads were administered and the intraload period was doubled (from 10 minutes to 20 minutes). All of the other components of the protocol (e.g. blood and urine collections, etc.) were kept the same.
The relationship between [K]0 and time in minutes remained linear throughout and there was no evidence of saturation kinetics or splay. The formula for the prolonged infusion was: Y = 0.28 x 6.21 (R2 = .995). The value for [K]O at 10 minutes was 9.01 mEq/L and at 20 minutes, 11.8 mEq/L. Hence, the linear pattern observed during a 10 minute infusion was replicated when the dose and time of the infusion was doubled.
The Effects of a Major Discontinuity in K Delivery
At the conclusion of the equilibration period an MB-6KCl infusion was initiated. However, rather than continuing the infusion for 10 minutes, the pump was turned off precisely at 60 seconds. The pump was left off for the next 60 seconds and the on-off sequence was repeated through the 10th minute. Blood samples were obtained at the end of each 60 second interval. The "pump-on" and "pump-off" data were analyzed separately. For the pump-on analysis, there were five blood samples obtained at one, three, five, seven and nine minutes. The same number of points were available for the pump-off analysis (at two, four, six, eight and 10 minutes).
Both sets of data fit different linear regression lines. The R2 value for each was 0.99. Thus, there was no discontinuity in either function, but the "reset" button must have been activated early within each 60 second interval. The equations were:
Pump-on: Y = 0.52 x + 7.36
Pump-off: Y = 0.33 x + 5.55
To test the boundaries of protection against K intoxication provided by the First Responder Component, three consecutive loads of K were administered intravenously to individual rats.
Two sequences were employed. In a group designated as "upload", the individual loads were MB-2, MB-3 and MB-4. In the second group, designated as "download", the three loads were MB-4, MB-3 and MB-2. Each individual upload was 10 minutes in duration and the intervals between loads was 10 minutes. Each download was given over 10 minutes and the interval between loads was 10 minutes.
To have a standard of reference for the multiple load studies, single load experiments were performed on 24 rats in which the values for ΔICFk, were plotted against time in minutes. The results for 2 rats in each group are summarized in Table 4.
In comparing the upload with the download data, the response to the maximum loads (MB-4) is of special interest. The K loads were identical. However, for the upload, MB-4 is the third consecutive challenge load having been preceded by MB-2 and MB-3 loads. On the contrary, the download MB-4 value represents the first load, without prior exposure to MB-2 or MB-3.
For the two MB-4 upload studies (interval 10 minutes), the ICF slope averaged 33.6; for the two MB-4 download (given as the first load), the ICF slope averaged 32.2. The mean value for the ICF slope of the nine MB-4 reference studies was 36.4. For MB-3 up vs. down, the ICF slopes were 24.1 and 24.0. The ICF slopes for the reference group averaged 28.7. Studies using a 5 minute interval were performed in the upload group only. The values were MB-2, 17.9; MB-3, 29.3; and MB-4, 33.3. The reference values were 20.6, 28.7 and 36. So whether the first load was MB-4 or MB-2 the ICF slopes were almost identical to the control slopes for that load.
The effect of Na, K-ATPase Inhibition on Extrarenal Transport of K: Digoxin (24, 33)
Contact between the intracellular fluid K and extracellular fluid K is made possible through a series of K specific channels which penetrate the cell membrane, or through cotransport pathways (e.g. NaK2C1, KCC) and by the all important four-sided (tetrametric) protein with a central water-filled pore, Na, K-ATPase, "pump" (5,23).
To examine the role of Na, K-ATPase in the extrarenal transport of K, a series of studies was performed on rats in which large doses of digoxin or ouabain were given by constant infusion to poison the pump. The drug being studied was added to the sustaining solution and the mixture was infused for the remainder of the study. After 40 to 60 minutes exposure to digoxin, either an MB-2 or MB-3 K load was administered by constant infusion through a venous catheter over a period of precisely 10 minutes. This experiment is described below and the data are depicted graphically in Fig. 3.
A 200 gm normal rat was prepared in the manner described in Methods. During two control periods (20 minutes each), there was no net flux of K. At -60 minutes, digoxin (1 mg/hr) was added to the sustaining solution. The value for ΔICFK dropped dramatically from the resting value of zero to -180 μEq during the following 60 minutes. At this time (time 0 on the graph), an MB-3 load of KCl was infused through an access separate from that used for the sustaining solution. Despite the continuing infusion of digoxin, the K load induced a striking change in the direction of K flux. Within 2.5 minutes (and probably earlier), the direction of the K flux reversed sharply and ΔICFK rose from -180 μEq to +84 μEq, reaching a total of 264 μEq in 10 minutes. A plot of ΔICFK vs. time fit a linear line (R2 = .99). The slope of the function was 29 (in normal rats receiving an MB-3 load without digoxin, the slope averaged 28; n = 16).
The Effects of Digoxin on Extrarenal K Transport Followed by an MB-3 K Load in a Remnant Rat
The identical study shown in Fig. 3 was performed on a series of remnant rats subjected to 60 minutes exposure to digoxin (one mg/hr) followed by a 10-minute MB-3 K infusion. The results replicated in precise detail the preceding studies on normal rats.
Anephric Rats Treated with Digoxin: The Effects of an MB-2 KCl Load
The anephric rats were studied 16 hours after bilateral nephrectomy. After the control and recovery periods, each rat received an infusion of digoxin (1 mg/hr). In a representative study, ΔICFK was 0 during the control periods; fell to -83.5 μEq during the 60 minute exposure to digoxin; and then during the 10-minute infusion of K, ΔICFK rose to +78.5 μEq. The total increase in the ΔICFK from the onset of the K infusion to the completion of the MB-2 load was 162 μEq.
[K]0 rose from a resting level of 5.2 mEq/L to 6.78 during the 1 hour exposure to digoxin and 13.3 mEq/L at the completion of the 10-minute MB-2 K infusion. The increasing values for [K]O during the K infusion when plotted against time were linear. Despite the striking elevation of the value for [K]O prior to the K load, the First Responder Component was not activated.
The Effects of the Inhibition of Na, K-ATPase: Ouabain (7,24)
A 225 gm rat was prepared in the standard manner following which two 20-minute control periods were obtained. The coordinates were ΔICF vs. time (Fig. 4).
During the control periods, there was no net flow of K. At 40 minutes, ouabain (1 mg/hr) was added to the sustaining solution, which was K free. During the 40 minutes of ouabain administration, the value for ΔICFK dropped by -79 μEq. An MB-3 load of KCl was then infused intravenously in 10 minutes. The administration of ouabain was continued at the same rate through the remainder of the study. The K load again had a striking effect. During the first 2.5 minutes, there was a marked loss of ICF-K; the efflux of K then reversed and the consecutive values for ΔICFK rose from -115 μEq to + 107 μEq (total 222 μEq). The plot of time in minutes vs. ΔICFK during the 10 minute K infusion was linear (Fig. 4).
The same experiment described above produced results virtually superimposable in 200¬225 gram remnant rats.
Anephric Rats Treated with Ouabain: The Effects of an MB-2 KCI Load
Following the control and recovery periods, each rat received an infusion of ouabain (0.5 mg/hr) administered with the sustaining solution. In the first 30 minutes of exposure to ouabain, [K]0 rose from 5.53 to 8.13 mEq/L. During the 10 minute KCl load (MB-2) K0, rose to 13.3 mEq/L. The resting value for ΔICFK was zero. Following the addition of ouabain, ΔICFK fell by -133 μEq. During the 10 minute KCl infusion, ΔICFK rose to + 79.2 μEq/L. The total amount of K transferred during the 10 minute MB-2 K infusion was 212.2 μEq, an amount that is 32.6% greater than the normal ECFK content.
It is an old and well documented finding that the administration of insulin will produce a fall in ECF [K]0. A number of routes for the transfer of ECFK to ICFK have been considered but the predominant opinion is that the insulin acts by increasing the rate of Na, K-ATPase mediated K influx.
A series of studies was performed to examine the effects of K flux produced by insulin alone, insulin followed by an MB-3 K load, and insulin plus ouabain followed by an MB-3 load.
Insulin Alone followed by an MB-3 K Load
A representative experiment is shown in Fig. 5 for a normal rat receiving 2 μU/hr of insulin. The insulin was added to the sustaining solution at the completion of the control periods and continued through the remainder of the study. During the first 40 minutes of exposure to insulin, the plasma potassium concentration decreased from the control value of 43 to 3.56 mEq/L and ΔICFK rose by 87.8 μEq. With the insulin infusion continuing, an MB-3 load of KCl was infused during the following 10 minutes. KO rose only to 7.79 mEq/L and an additional 172 μEq was transferred from the ECF into the ICF.
The Effects of Insulin Plus Ouabain Followed by an MB-3 K Load
In Fig. 6, an experiment is shown in a normal rat in which the coordinates are ΔICFk and Time. During the control periods, the cells were "at rest" in an equilibrium state. Hence there was no net K flux. At 40 minutes. insulin (2 μU/hr) and ouabain (1 mg/hr) were added to the sustaining solution and continued through the end of the study. The result was an influx of K totaling 110 μEq in the next 40 minutes. Thus, the addition of a potent ATPase inhibitor to the insulin did not appear to impair the K influx produced by insulin alone.
In Fig. 6, after 40 minutes of infusion of insulin and ouabain, the infusion of an MB-3 load resulted in an abrupt increase in the net K influx. The amount of K transferred during the 10 minute period was an additional 110 μEq. The plot of time vs. ΔICFK was linear. The slope was 25.9 and the value for R2 = 1.00. Thus the rate of K influx following an MB-3 K load was 172 uEq for insulin alone (Fig. 5) and only 110 μEq for insulin plus ouabain (Fig. 6).
Blood Pressure and Glomerular Filtration Rate (GFR)
As noted in the Methods section, an arterial catheter was attached to a transducer for the online measurement of mean arterial blood pressure. Despite an anticipated drop in blood pressure, particularly at the MB-4 and above K loads, there was no statistically significant change in B.P. comparing resting, and intraload values.
The mean GFR of 23 normal rats was 2.1 ml/min prior to study. At 10 min IL it was 2.73, at 10 min PL it was 1.95 ml/min, and 1.67 ml/min at PL 60.
SUMMARY AND CONCLUSIONS
Our studies were undertaken to examine the distribution of massive intravenous loads of K given over a 10- minute period under various conditions. The loads were varied from 2 MB to 6 MB and were given to normal, remnant and anephric rats. Rats on digoxin, ouabain, insulin, and ouabain plus insulin. Multiple loads were given in sequence 3 loads (2,3,4 MB). Changes were made every 5-10 minutes within the IL infusion period. Blood and urine samples were taken at 2.5, 5, 7.5 and 10 minutes and in some studies at 1.0 or 1.25 minute. This allowed us to observe the acute distribution of K in an intact animal with a rise in serum K from 4 mEq/L to 8¬13 mEq/L over a 10-minute period. Previous studies used much smaller loads, often given over longer periods of time with rises in serum K of up to 1.5 mEq/L in dogs and rats (12,13,20,28).
A constant feature in all studies under all conditions was a liner rise in serum K from 2.5 to 10 minutes IL (R2 0.99). This linearity persisted even after blockade of Na, K-ATPase with either digoxin or ouabain (Figs. 3 and 4). The rate of rise in serum K varied with the load and with the potassium balance of the rats. Much more rapid rise in K depleted rats and a slower rise in K loaded rats, due to a high urine K in K loaded rats and very little urinary K in K depleted rats.
Uremia did not appear to influence the rate of sequestration. At IL-10 normal rats 54%, and remnant rats 57%. ΔICFK 17 hours after total nephrectomy had only 26% of the infused K sequestered at IL-10. Their baseline serum K+ prior to infusion was 4.97 mEq/L and reached a level of 10.78 mEq/L at IL-10, very similar to our remnant rats with a PK of 4.14 mEq/L at start of infusion and 9.08 mEq/L at IL-10. From these studies it would appear that uremia per se has no effect on the initial rate of sequestration but the absence of the kidneys may influence the rate of sequestration during the IL period.
The rise in serum K in the first 1.25 minutes averaged 2.3 mEq/L in 22 normal rats on MB2-8 loads. The amount of K administered to these same 22 rats in the first 1.25 minutes was 54.4 μEq. A total of 92 μEq was added to the ECF. Therefore 37.6 μEq was added to the ECF from some other source, presumably from the ICF (negative sequestration). At some point between 0.5 and 1.5 minutes the negative sequestration reversed and became positive in a linear fashion, as noted above. The mechanism producing the initial negative sequestration is not clear from our studies as it was present in all studies mentioned above. However, the magnitude of the negative sequestration may determine the slope of the subsequent rate of sequestration (see Fig. 2).
It is possible that the rapid rise in serum K, with large loads of IV K, would cause the polarity in the cell to become less negative, allowing greater loss of K from the cell (see the Nernst modification of the Goldman equation. One of the major driving forces for Na, K¬ATPase is the alfa2 activity, present in the T-tubules, and an increase in the K concentration in the T-tubule. However, the steady depolarization of the cell as serum K rises would inactivate excitation-contraction coupling before there would be any effect on Na, K-ATPase activity (14). The slope of the rate of sequestration involved can be reset within seconds as shown by varying the loads every 5-10 minutes over the infusion period (see Table 4). The slope can also be changed by adding drugs that increase the rate of sequestration (i.e. insulin) (Fig. 5).
Insulin is known to increase Na, K-ATPase (1) activity and produce increased K sequestration in the cell. Figure 5 demonstrates the effect of insulin in our rats with a linear increase in K sequestration during the 40-minute insulin infusion.
What is striking about this rat is the marked increase in the rate of K sequestration following an MB-3 load of KCl over a 10-minute period. In Fig. 6 the same study was done on a rat receiving ouabain 3 mEq/hr along with insulin. This dose of ouabain in our 8 rats routinely produced marked negative sequestration in the range of 170 μEq in 1 hour, clearly having a marked effect on Na, K-ATPase (Figs. 3 and 4). In spite of this potent blockade of Na, K-ATPase, insulin was able to overcome the effect of ouabain and produce a normal rate of sequestration. Once again the addition of an MB-3 load of KCl produced a marked increase in the rate of sequestration, far above that produced by insulin with or without ouabain.
Na, K-ATPase is primarily stimulated by increase in intracellular sodium and to a lesser degree by an increase in extracellular K (9,14). During muscle activity K is lost from the cell and sodium moves into the cell, stimulating the Na, K-ATPase activity. However in our model where K is infused in massive loads over a short period, following an initial (30-91 second loss from the cell), the concentration of K in the cell increases above normal but because of the high ECF K, according to the Nernst equation, there is marked depolarization of the cell which should decrease movement of Na into the cell. Our rats were in a resting state where 80% of Na, KATPase pool has been shown to either be inactive or unavailable for immediate activation (19). Indeed intracellular sodium concentration appears to be the major activator of Na. K-ATAPase (8). In contrast, acute stimulation of Na, K-ATPase by insulin, catecholamines, etc., increases basal Na, K-ATPase by only 100% (8), which seems inadequate to account for the massive K sequestration in our rats.
It is difficult to compare our studies with studies done on rats where muscle activity is the major stimulator of Na, K-ATPase activity. In these studies, muscle activity produces loss of K+ from the cell and increased intracellular Na which in turn stimulates marked increases in Na, K-ATPase (30). In our studies, the infusion of K intravenously produces marked increase in K sequestration and a depolarization of the cell which should block Na+ from entering the cell and a decrease in Na, K-ATPase.
In summary these studies demonstrate that the protection of the ECF K is essential to life. Under normal conditions Na, K-ATPase seems to be the major protector of the ECF K by increasing sequestration of K into the cell. When massive loads of K are administered intravenously over a very short period of 10 minutes, there is initial negative sequestration for 30-90 seconds followed by massive positive sequestration over the remaining 10 minutes of the IV infusion of KCl (54% of the load) with loads up to 8 times the total ECF-K. The degree of the initial negative sequestration may be related to the subsequent positive rate of sequestration (see Fig. 1). It is also possible that part of the positive sequestration is due to Na, K-ATPase. Obviously, we were unable to totally block Na, K-ATPase with ouabain or digoxin or the animal would die. However, from our studies, we believe that some other mechanism is also operative in this process for the following reasons. The rate, linearity, and degree of sequestration are much greater than would be expected due to activation of the Na, K-ATPase pump by increasing levels of serum K. At physiological concentrations of serum K (4 mm), the Na, K-ATPase pump are to a large extent, saturated by K (9). The high level of blockade of Na, K-ATPase by both ouabain and digoxin was promptly overcome by a KCl load and sequestered in a normal fashion. A KCl load given following insulin, a potent stimulator of Na, K-ATPase, had a marked increased rate of K sequestration above and beyond that seen with insulin alone. We have termed this acute K protective mechanism "the first responder component."
Author, Article and Disclosure Information
Conflict of Interest: Authors report no financial relationships or interests to disclose.
Authorship: Authors are entirely responsible for the content of this manuscript.
Requests for reprint should be addressed to Southern California Journal of Medicine at email@example.com
1. Benziane B, Chibalin, AV. Frontiers: Skeletal muscle sodium pump regulation: a
translocation paradigm. Am J Physiol Endocrinol Metab 295: E553-E558, 2008.
2. Bricker NS. Sodium homeostasis in chronic renal disease. Kidney Int. 21: 866-897, 1982.
3. Bricker NS, Shapiro MS, Levin NM et al. Physiology and pathology of electrolyte
metabolism in chronic renal disease. In: The Kidney, Physiology and Pathophysiology, edited by Seldin GW, Giebisch G. New York, NY: Raven Press, 1985, 1837-1862.
4. Cain CD, Schroeder FC, Shankel SW, Mitchnick M, Schmertzler M, Bricker NS.
Identificaton of xanthurenic acid 8-0-beta-D-glucoside and xanthurenic acid 8-0-sulfate as human natriuretic hormones. Proc Nat Acad Sci 104: 17873-17818, 2007.
5. Cairns SP, Flatman JA, Clausen T. Relation between extracellular [K+], membrane
potential and contraction in rat soleus muscle modulation by the Na+-K+ pump. Pflugers Arch Eur J Physiol 430:909-915, 1995.
6. Clausen T, Kohn PG. The effect of insulin on the transport of sodium and potassium in
rat soleus muscle. J Physiol 265: 19-42, 1977.
7. Clausen T, Hansen, 0. Active Na-K transport and the rate of ouabain binding. The
effect of insulin and other stimuli on skeletal muscle and adipocytes. J. Phsyiol 270: 415-430, 1977.
8. Clausen T. Na÷-K+ pump regulation and skeletal muscle contractility. Physiol Rev 83:
9. Clausen T. Review. Role of Na , ICATPase pump and transmembrane Na+, K+
distribution in muscle function. Acta Physiol 192: 339-349, 2008.
10. Clausen T, Everts ME, Kjeldsen K. Quantification of the maximum capacity for active
sodium-potassium transport in skeletal muscle. J Physio1388: 163-181, 1987.
11. Clausen T, Flatman JA. Effect of insulin and epinephrine on Na+-K+-ATPase and
glucose transport in soleus muscle. Am J Physiol 252: E493, 1987.
12. DeFronzo RA, Lee R, Jones A, Bia M. Effect of insulinopenia and adrenal hormone
deficiency on acute potassium tolerance. Kidney Int 7: 586-594, 1980.
13. DeFronzo RA, Sherwin RS, Dillingham M, Hendler R, Tamborlane WV, Felig D.
Influence of basal insulin and glucagon secretion on potassium and sodium metabolism. J Clin Invest 61: 472-479, 1978.
14. DiFranco M, Hakimjavadi H, Lingrel JB, Heiny JA. Na, K-ATPase a2 activity in
mammalian skeletal muscle T-tubules is acutely stimulated by extracellular K. J Gen Physiol 146: 281-294, 2015.
15. Doyle DA, Cabral JM, Pfuetzner RA, Kuo A, Gulbis JM, Cohen SL, Chait BT, MacKinnon R. The Structure of the potassium channel: Molecular Basis of K+ conduction and selectivity. Science 280:69-77, 1998.
16. Garty H, Palmer LG. Epithelial sodium channels: function, structure, and regulation.
Physiol Rev 77: 359-396, 1977.
17. Giebish G, Kropf R, Wagner, C. Renal and extrarenal regulation of K. Kidney Int 72:
18. Graham JA, Lawson MI, Linton AL. Muscle biopsy water and electrolyte contents in
chronic renal failure. Clin Sci 38: 583-591, 1970.
19. Giilden, M, Seibert H, Voss JU. In vitro toxicity screening using cultured rat skeletal
muscle cells. II. Agents affecting excitable membranes. J Toxicol In Vitro 8:197-208, 1994.
20. Hiatt N, Chapman, LW, Davidson MB, Sheinkopf JA, Miller A. Inhibition of
transmembrane K transfer in ureter-ligated dogs infused with KCl. Am J Physiol 231: 1660-1664, 1976.
21. Juel C. Oxidative stress (glutathionylation) of Na, K-ATPase activity in rat skeletal
muscle. PLOS One 9: e, 10514, 2014.
22. Lindinger M, Heigenhauser GJ. The roles of ion fluxes in skeletal muscle fatigue. Can
J Physiol Pharmacol 69: 246-253, 1991.
23. McDonough AA, Thompson CB, Young JH. Skeletal muscle regulates extracellular
potassium. Am J Physiol Renal Physiol 282: F967-F974, 2002.
24. Molier B, Vaag A, Johansen B. Ouabain, Inhibition of the Na-K pump. J Clin
Pharmacol 129: 93, 1990.
25. Moore RD, Rabovsky JL. Mechanism of insulin action on resting membrane potential
of frog skeletal muscle. Am J Physiol Cell Physiol 236: C254, 1979.
26. Moore RD. Effects of insulin upon ion transport. Biochim. Biophys. Acta 737: 1-49,
27. Palmer LG, Sackin H. Electrophysiological analysis of transepithelial transport. In:
The Kidney, Physiology and Pathophysiology, edited by Seldin DW, Giebisch E. New York, NY, Raven Press, pp. 361-405, 1992.
28. Pettit G, Vick RL, Swandler AM. Plasma K and insulin: changes during KCl infusion
in normal and nephrectomized dogs. Am J Physiol 228: 107-109, 1975.
29. Pichard C, Hoshino E, Allard JP, Charlton MP, Atwood HL, Jeejeebhoy KN.
Intracellular potassium and membrane potential in rat muscles during malnutrition and subsequent refeeding. Am J Clin Nutr 54: 489-498, 1991.
30. Pirkmajer S, Chibalin AV. Na, K-ATPase regulation in skeletal muscle. Am J Physiol
Endocrinol Metab 311: EL1-EL31, 2016.
31. Relman AS, Gorham GU, Levinsky NG. The relation between external potassium
concentration and the electrolyte content of isolated rat muscle in the steady state. J Clin Invest 40: 386-393, 1961.
32. Rosa RM, Epstein FH. Extrarenal potassium metabolism. In: The Kidney: Physiology
and Pathophysiology, edited by Seldin DW, Giebisch E. Lippincott, Williams and Wilkins, Philadelphia, 1552-1573, 2000.
33. Schoner W, Scheiner-Bobis G. Endogenous cardiac glycosides: hormones using the
sodium pump as signal transducer. Semin Nephrol 25: 343-351, f2005.
34. Schultze RG, Taggart DS, Shapiro H, Pennell JP, Caglar S, Bricker NS. On the
adaptation in potassium excretion associated with nephron reduction in the dog. J Clin Invest 59: 1061-1068, 1971.
35. Shankel SW, Robson AM, Bricker NS. On the mechanism of the splay in the glucose
titration curve in advanced experimental renal disease in the rat. J Clin Invest 46: 164¬172, 1967.
36. Slatopolsky E, Robson AM, Elkan I, Bricker NS. Control of phosphate excretion in
uremic man. J Clin Invest 47: 1865-1874, 1968.
37. Wright PS, Giebisch G. Regulation of potassium excretion. In: The Kidney: Physiology
and Pathophysiology, vol. 2, New York, Raven Press,1223-1249, 1985.
38. Wright SH. Generation of resting membrane potential. Advan Physiol Edu 28: 138-142, 2004.