INTRODUCTION

Top Abstract Introduction Materials & Methods Results Discussion References

RENAL LITHIUM ION CLEARANCE is used as an index of the delivery of tubular fluid from the proximal tubules to the loop of Henle in humans and Na⁺-replete rats (15). The technique is based on the assumptions 1) that Li⁺ in the proximal tubule is reabsorbed to the same extent as Na⁺ and water so that the concentration of Li⁺ in the proximal tubular fluid remains constant and 2) that transtubular transport of Li⁺ is negligible in the postproximal segments of the tubule, i.e., the loop of Henle, the distal convoluted tubules, and the collecting ducts (4, 28). Consistent with these assumptions, it has been found repeatedly that the renal clearance of Li⁺ in Na⁺-replete rats is unaffected by administration of amiloride, which inhibits apical membrane Na⁺ entry into the cells of the collecting duct (2). However, during Na⁺ depletion, the clearance of Li⁺ is markedly decreased but can be normalized by amiloride, indicating that, in this situation, a significant proportion of the distal tubular load of Li⁺ may be reabsorbed in the distal nephron (31).

The reason why amiloride-sensitive transport of Li⁺ is quantitatively significant in Na⁺-depleted but not in Na⁺-replete rats is unknown. The limitation in reabsorption of Li⁺ during Na⁺-replete conditions may be either the cellular influx across the apical membrane or extrusion through the basolateral membrane, and, in both cases, the distal tubular load of Na⁺ may be crucial for the reabsorption of Li⁺. If Na⁺ and Li⁺ compete for the same apical transporter, an increased tubular load of Na⁺ would be expected to result in a parallel reduction in Li⁺ transport. Furthermore, Li⁺ extrusion across the basolateral membrane may require a large concentration gradient between the intracellular fluid and the plasma, as the extrusion does not undergo active transport (9). In this situation, hyperpolarization of the apical membrane, which increases apical inward cation transport, may be crucial for elevating the intracellular Li⁺ concentration to levels necessary for substantial transepithelial Li⁺ transport.

In the present study, we have examined, by use of a conscious rat model, the mechanism(s) underlying distal tubular Li⁺ reabsorption by modulation of the Na⁺ reabsorption. The influence of amiloride on the fractional extraction of Li⁺ (FE_Li), an index of distal tubular Li⁺ reabsorption, was investigated in rats on normal and low-Na⁺ intakes, in which distal tubular transport rates of Na⁺ were manipulated by administration of bendroflumethiazide (BFTZ), angiotensin III (ANG III), and aldosterone (Aldo). The effect of an 80-90% reduction in plasma Li⁺ concentration was also investigated. The results are compatible with the hypothesis that, during Na⁺-depleted conditions, the apical membrane in the more distal parts of the amiloride-sensitive region of the collecting ducts becomes hyperpolarized due to a low-Na⁺ influx, and distal Li⁺ reabsorption can occur.

	MATERIALS AND METHODS

Top Abstract Introduction Materials & Methods Results Discussion References

Animals. Specific pathogen-free female Wistar rats (200-250 g) were obtained from the Department of Experimental Medicine, Panum Institute, University of Copenhagen (Copenhagen, Denmark). They were housed in a temperature (22-24°C)- and humidity (40-70%)-controlled room with a 12:12-h light-dark cycle (light on from 6:00 AM to 6:00 PM). The rats were given free access to tap water and pelleted rat diet with a K⁺ content of 220 mmol/kg and Na⁺ contents of either 5 mmol/kg (Na5 diet) or 200 mmol/kg (Na200 diet) for at least 8 days before experimentation. Three days before the clearance study, lithium citrate was added to the food (12 and 7 mmol/kg for Na200 and Na5, respectively) to obtain measurable plasma Li⁺ concentration ([Li⁺]) without influencing renal function (26). In one group of rats, Li⁺ was withdrawn 1 day before the experiment to reduce the plasma [Li⁺] in this group (Na5-low Li).

Animal preparation. Animals were surgically prepared for clearance studies 2 wk before experimentation as described previously (21). In summary, animals were anesthetized with halothane-N₂O, and Tygon catheters were inserted into the abdominal aorta and the inferior caval vein via femoral vessels. A suprapubic bladder catheter was also implanted. After instrumentation, animals were housed individually. During the last week before clearance studies, rats were adapted to the restraining cage used for these experiments by training them for two periods of ~90 min each.

Clearance technique. On the day of the experiment, the rats were transferred to restraining cages, and the catheters were opened and flushed with 150 mM glucose containing 20 U/ml of heparin. Throughout the experiments, the animals received an infusion of this solution at a rate of 0.25 ml/h to keep the arterial catheter open. In addition, each animal received throughout the experiment an infusion of 150 mmol/l glucose (bolus 0.5 ml, sustained 0.5 ml/h) containing [³H]inulin (batch no. 119-121; bolus 5 µCi, sustained 5 µCi/h; Amersham), LiCl (bolus 3.3 µmol, sustained 3.3 µmol/h), and 1.0 ml/h of 150 mM glucose as vehicle for drugs. Each experiment comprised a 15-min inulin-Li⁺ bolus period, a 105-min equilibration period, three 30-min preamiloride periods, and three 30-min periods during which amiloride was added to the ongoing intravenous infusion.

The total rate of infusion of 150 mM glucose was kept constant at 1.75 ml/h in all experiments. In addition, during administration of BFTZ or amiloride, urinary fluid losses exceeding 1.75 ml/h were replaced by infusion of a 150 mmol/l NaCl solution by a computer-driven servo-controlled system. This system, which includes a personal computer, an electronic balance (type MC 1; Sartorius, Goettingen, Germany), and an infusion pump (Perfusor Secura; Braun Melsungen, Melsungen, Germany), monitored urine flow rate gravimetrically at 5-min intervals and adjusted the rate of the pump accordingly (5).

Blood samples of 250 µl were collected from the arterial catheter at the end of the equilibration period, after the preamiloride periods, and at the end of the experiment. All blood samples were replaced immediately with heparinized donor blood. At the end of the experiment, all catheters were sealed with 50% glucose, 500 units of heparin, and 10,000 units of streptokinase per milliliter, and the rats were returned to the animal unit. When different experiments were performed on the same animal, they were separated by at least 1 wk, allowing full recovery between experiments.

Drugs were dissolved in 150 mM glucose and were given as an intravenous priming dose followed by a maintenance infusion at a rate of 0.5 ml/h. The priming doses of BFTZ (83 µg), ANG III (1.5 µg), and amiloride (0.17 mg) were given over a period of 5 min, whereas Aldo (600 pmol) was given over a period of 15 min together with the inulin bolus. The corresponding maintenance infusions provided 250 µg/h, 1.5 µg/h, 0.5 mg/h, and 600 pmol/h, respectively.

Analyses. Urine volume was determined gravimetrically. Na⁺ concentration ([Na⁺]), K⁺ concentration ([K⁺]), and [Li⁺] in plasma and urine were determined by atomic absorption spectrophotometry, using a Perkin-Elmer model 2380 atomic absorption spectrophotometer. In the Na5-low Li group, plasma and urine [Li⁺] were determined by flameless atomic absorption spectrophotometry, using a Perkin-Elmer electrothermal atomic absorption spectrophotometer (16, 24). [³H]inulin in plasma and urine was determined by liquid scintillation counting using a Packard Tri-Carb liquid scintillation analyzer (model 2250CA). The urinary excretion rate of amiloride was determined in eight of the Na5 and eight of the Na200 rats given vehicle infusion. Urinary amiloride concentrations were determined using a high-performance liquid chromatography system with fluorescent analysis. Blood pressure and heart rate were determined continuously using Baxter Uniflow pressure transducers (Bentley Laboratories) connected to Hugo Sachs pressure and heart couplers. Signals were displayed on a Watanabe Instruments WR 3101 Line recorder Mark VII (Watanabe Instruments).

Calculations. Values for renal clearance (C) and fractional excretion (FE) were calculated using standard formulas where U is the concentration in the urine, V is the urine flow rate, and P is the concentration in plasma. [³H]inulin clearance (C_in) was used as a marker for the glomerular filtration rate (GFR).

The urinary amiloride concentrations were 346 ± 95 (means ± SE) and 170 ± 18 (means ± SE) µmol/l, and the excretion rates were 7.3 ± 0.8 and 6.5 ± 0.8 nmol · min¹ · 100 g body wt¹ in the Na5-vehicle and the Na200-vehicle groups, respectively. From these values, intratubular amiloride concentrations were estimated to range from 5 to 10 µmol/l in the proximal tubules and higher in the cortical collecting ducts. These concentrations are well below those reported to block the Na⁺-H⁺ exchange mechanism in the proximal tubules [inhibitory constant (K_i) >50-100 µmol/l] but above the concentrations required to block movement of Na⁺ through the conductive Na⁺ channel (K_i <1 µmol/l; see Refs. 2 and 11).

Assuming that amiloride completely blocked the conductive Na⁺ channels in the collecting ducts and that Na⁺ is not reabsorbed by other mechanisms in this segment or further downstream (22), the reabsorption (R) of Na⁺ expressed as a fraction (f) of the load delivered to the reabsorptive site can be calculated as where V₁×U_Na1 and V₂×U_Na2 represent the absolute excretion of Na⁺ before and after administration of amiloride, respectively. To account for changes in GFR and plasma [Na⁺] between the two observation periods, these variables should be factored, resulting in the following equation expressing the relative (r) amiloride-sensitive Na⁺ reabsorption which is equal to the equation Similarly, the relative amiloride-sensitive Li⁺ reabsorption can be expressed as The expressions fR_Na-Amil and rR_Na-Amil are similar, as no differences between GFR and P_Na are seen before or after administration of amiloride. In contrast, as P_Li showed time-dependent changes in some series, a meaningful estimate of the relative reabsorption of Li⁺ in the amiloride-sensitive segment can only be accomplished through determination of rR_Li-Amil. Furthermore, Li⁺ and Na⁺ transport in the amiloride-sensitive segment can be directly compared using the expressions rR_Li-Amil and rR_Na-Amil. Because the Na⁺ load on which rR_Na-Amil is based (FE_Na2) may be underestimated because of Na⁺ reabsorption in the distal nephron by mechanisms not being amiloride sensitive, the true rR_Na-Amil may be less than the calculated value. However, this inaccuracy is not important for the data presentation and interpretation performed.

Data presentation and statistics. Renal clearance variables are presented as mean values of the last two preamiloride periods and the last two periods during amiloride infusion. Effects of amiloride were evaluated statistically by comparing the two average values using Student's paired t-test. Differences were considered statistically significant at the 0.05 level. Values presented are means ± SE.

	RESULTS

Top Abstract Introduction Materials & Methods Results Discussion References

Heart rate, mean arterial blood pressure, GFR, and plasma electrolytes before and during amiloride are given in Table 1. Between series, the blood pressure values were similar except for the series given ANG III. Administration of ANG III increased blood pressure by ~15%. In most experiments, GFR was not influenced by the dietary Na⁺ intake and the various pharmacological treatments, but, in the ANG III series, it was reduced by ~20% compared with the other appropriate control series. Plasma [Li⁺] averaged 125-170 µmol/l, except in the Na5-low Li group in which plasma [Li⁺] was 10-20% of this level. There were no significant differences in plasma [Na⁺] or [K⁺] among the groups.

The renal handling of Na⁺ and Li⁺ is reported in terms of fractional excretion before and during amiloride administration (Table 2). The difference between the FE_Li before and during amiloride administration (delta FE_Li) may be taken as an index of the Li⁺ reabsorption in the distal tubular segments, predominantly in the collecting ducts (Table 3). With regard to the general pattern of changes in FE_Li, it is noticeable that 1) conditions during which distal tubular reabsorption of Na⁺ is intensified, resulting in low fractional excretion rates of Na⁺, were associated with amiloride-reversible reductions in the FE_Li; and 2) FE_Li after amiloride always was close to 35% irrespective of large changes in Na⁺ intake or plasma [Li⁺] or of administration of Aldo, ANG III, or thiazide.

FE_Li was markedly reduced by a low-Na⁺ diet (Na5-vehicle group), and this reduction was completely reversed by amiloride. The difference amounted to 16% of the filtered load equal to ~40% of the estimated distal delivery of Li⁺ (Table 3). A practically identical reversal of FE_Li was observed after amiloride in the Na5-low Li group in which plasma [Li⁺] was reduced by 80-90%. Here, amiloride increased FE_Li with 14% of the filtered load, indicating that the amiloride-reversible reduction in FE_Li induced by the low-Na⁺ diet was independent of [Li⁺] in plasma. Administration of BFTZ completely reversed the decrease of FE_Li otherwise observed in rats on a low-Na⁺ diet (Na5-BFTZ group vs. the Na5-vehicle and Na200-vehicle groups), and subsequent administration of amiloride did not affect FE_Li of the Na5-BFTZ group at all. Similar to the effects of dietary Na⁺ restriction, pretreatment with Aldo led to a reduction of FE_Li, which also was fully reversed by subsequent administration of amiloride. A less pronounced effect was seen after the administration of ANG III. All conditions during which fractional Na⁺ excretion was reduced (low-Na⁺ intake, Aldo or ANG III pretreatment) were associated with a decrease in FE_Li, which was reversible by amiloride, whereas conditions during which fractional excretion of Na⁺ was normal or high (normal Na⁺ diet, thiazide + normal Na⁺ diet or thiazide + low-Na⁺ diet) revealed FE_Li values close to 35% and completely amiloride insensitive.

The amiloride-sensitive reabsorption of Na⁺ and Li⁺ expressed as a fraction of the distal tubular load (rR_Na-Amil and rR_Li-Amil, respectively) showed almost the same pattern as delta FE_Na and delta FE_Li (Table 3). In general, rR_Na-Amil was ~95% in the groups with low-Na⁺ excretion (Na5-vehicle, Na5-low Li, and Na200-Aldo), suggesting an almost complete distal tubular Na⁺ reabsorption. In the groups with a high Na⁺ excretion (Na200-vehicle, Na200-BFTZ, and Na5-BFTZ), rR_Na-Amil ranged from 35 to 80% and was thereby somewhat lower compared with rats exhibiting Na⁺ retention. rR_Li-Amil was ~35% in groups with a low-Na⁺ excretion, but, in contrast to rR_Na-Amil, it completely vanished in groups with a high Na⁺ excretion. These data emphasize that rR_Na-Amil and rR_Li-Amil did not change proportionally between conditions with a high and a low urinary Na⁺ excretion.

To gain further insight into the mechanism of Li⁺ transport in the distal nephron, the relationships between Li⁺ and Na⁺ handling were investigated in more detail, and K⁺ was included in the analysis. Because the amiloride-sensitive Na⁺ channel is probably the only quantitatively important mechanism for Na⁺ reabsorption in the cortical collecting ducts and further downstream (22), the fractional excretion of Na⁺ after amiloride administration can be used as an estimate of the fractional delivery of Na⁺ to the amiloride-sensitive nephron site. It is evident that BFTZ administration was associated with an increase in the fractional distal tubular Na⁺ delivery (Table 2). This increase in Na⁺ delivery led to a decrease of the amiloride-sensitive Li⁺ reabsorption, and, overall, there was an inverse relationship between delta FE_Li and the fractional distal delivery of Na⁺ (Fig. 1A). Similarly, there was an inverse correlation between delta FE_Li and the fractional excretion of Na⁺ in the urine before amiloride (Fig. 1B).

View larger version (11K): [in this window] [in a new window]   Fig. 1.   A: relationship between amiloride-sensitive reabsorption of Li+ (Delta FELi) and the fractional delivery of Na+ to the site of reabsorption [(FENa)Amiloride]. B: relationship between Delta FELi and the fractional excretion of Na+ [(FENa)Control]. Data are expressed as means ± SE. , 5 mmol/kg Na+ (Na5-Vehicle); , Na5-bendroflumethiazide (BFTZ); , Na5-low Li; , 200 mmol/kg Na+ (Na200-Vehicle); , Na200-BFTZ; , Na200-ANG III; , Na200-aldosterone.

The increased distal tubular delivery of Na⁺ during BFTZ was associated with increased Na⁺ reabsorption in the distal tubular segments, particularly in the Na5 group, as evidenced by the higher delta FE_Na during amiloride infusion in animals treated with BFTZ compared with vehicle infusion (Table 3). In addition, BFTZ administration increased amiloride-sensitive K⁺ secretion from 16 to 24% and from 10 to 38% in the Na200 and Na5 groups, respectively (Table 3). When evaluating all groups, a strong inverse correlation was observed between amiloride-sensitive Li⁺ and K⁺ transport (Fig. 2). Thus changes in the potential difference across the apical membrane, as reflected in the driving force for K⁺ efflux and K⁺ excretion, seemed related to changes in amiloride-reversible distal tubular Li⁺ reabsorption.

View larger version (10K): [in this window] [in a new window]   Fig. 2.   Relationship between Delta FELi and the amiloride-sensitive secretion of K+ (Delta FEK). Symbols are as in Fig. 1.

	DISCUSSION

Top Abstract Introduction Materials & Methods Results Discussion References

Effect of amiloride on Li⁺ reabsorption in the distal nephron. Amiloride blocks conductive Na⁺ channels in the collecting duct and has often been used as a tool to recognize distal nephron Li⁺ reabsorption (13, 27, 31). Micropuncture studies have shown that, in the doses employed in these studies, as well as in the present study, absolute and fractional proximal reabsorption of Li⁺, Na⁺, and water is not influenced, nor is the reabsorption of Li⁺ and Na⁺ in the loop of Henle. The increase of fractional urinary Li⁺ excretion observed in Na⁺-depleted rats is due to inhibition of Li⁺ reabsorption in the collecting ducts (7, 29, 32). An increased urine flow rate during amiloride administration, as observed in the present study, has consistently been found and is due to decreased water reabsorption in the collecting ducts (7, 29).

Effect of BFTZ on Li⁺ reabsorption in distal nephron. The results of the present study confirm previous observations that dietary Na⁺ depletion in rats leads to reabsorption of Li⁺ via an amiloride-sensitive transport mechanism (27, 31). This amiloride-sensitive reabsorption of Li⁺ is located between the last accessible part of the distal nephron and the end of the collecting ducts (32). In the present study, we show, for the first time, that Li⁺ transport at these distal nephron sites is abolished by administration of the thiazide diuretic BFTZ. Thiazide diuretics inhibit Na⁺-Cl cotransport in the distal convoluted tubules, just proximal to the amiloride-sensitive region (1, 6). BFTZ does not interfere with the tubuloglomerular feedback because of its lacking inhibition of the carbonic anhydrase in the proximal tubules (3) and does not inhibit Li⁺ reabsorption in Na⁺-repleted rats (12). Furthermore, BFTZ does not impair the amiloride-sensitive Na⁺ transport as seen from the similar natriuretic responses to amiloride in the Na200-vehicle and the Na200-BFTZ series (Table 2).

Mechanism of Li⁺ transport in distal nephron. The mechanism of Li⁺ reabsorption in distal nephron segments is not yet known with certainty, but, since it is blocked by amiloride, Li⁺ movement via the amiloride-sensitive Na⁺ channel is presumably involved (2). This channel is located in the apical cell membrane of collecting duct principal cells, and it displays a slightly higher permeability for Li⁺ than for Na⁺ (20). The number of open Na⁺ channels is increased by Na⁺ depletion (18) or chronic Aldo administration (19), but, even in Na⁺-replete animals, a significant number of channels are active, as also suggested by the increase of FE_Na in response to amiloride observed in previous studies and in the present study. It is therefore surprising that Li⁺ is only reabsorbed in the distal nephron in Na⁺-depleted rats and not in Na⁺-replete rats.