- Email: [email protected]
Divalent cation transport in isolated tubules
Kidney International, Vol. 22 (1982), pp. 492—497
Divalent cation transport in isolated tubules ROLAND C.K. No, ROBERT A. PERAINO, and WAD! N. SuKI Department of Medicine, Baylor College of Medicine and The Methodist Hospital, Houston, Texas
Since Burg introduced the technique of microperfusion of precipitation of calcium carbonate or binding of calcium and isolated renal tubules in 1966 [11, a number of papers has been loss of calcium to the oil-water interface [15]. Proximal convoluted tubule. Approximately 60% of the filpublished on sodium and water transport in various segments of the nephron. However, it was not until 1977 that transport of tered calcium is reabsorbed in the proximal convoluted tubule
divalent ions was described using this technique [21. Subse- [161. The bulk of calcium reabsorption in this segment is quently, 12 additional papers were published dealing with generally thought to be passive primarily because of(1) parallel calcium transport in various segments of the nephron [3—14].
handling of sodium and calcium such that the tubular fluid-to-
This review will deal primarily with the contents of these plasma (TF/P) sodium: (TF/UF) calcium concentration ratio is nearly unity under nondiuretic conditions [161, and (2) in situ microperfusion experiments demonstrating high permeability to Calcium transport calcium (an unidirectional efflux nearly two and one-half times Methodologic pitfalls. In studying calcium transport, the larger than the net flux) [171. However, two lines of evidence investigator must be cognizant of certain characteristics pecu- point to an element of active reabsorption: (I) The tubular fluidliar to calcium which may lead to artifactual errors. The first is to-ultrafiltrable calcium concentration ratio (TF/UF)ca falls to its propensity to bind to glass. Errors related to bound calcium less than unity in the presence of a mannitol diuresis 116]; and will lead to inaccuracies in the direction of greater transport. (2) in the presence of an equilibrium solution in the lumen, and Coating the perfusing, collecting, and constant volume pipettes therefore, absent convective forces, more calcium is transportwith a layer of silicon using Prosil-28 (PCR, Gainesville, Flori- ed than can be accounted for by the electrochemical gradient da), Sigmacote (Sigma Chemical Co., St. Louis, Missouri), [18]. Furthermore, ouabain, or replacement of sodium with Siliclad (Clay Adams, New York, New York), or Dow Corning choline or lithium on the peritubular side, inhibits the active 7 compound (Dow Corning, Midland, Michigan) will minimize component of calcium transport suggesting the presence of a this error. After coating the inner surface with a 6% vol/vol sodium-calcium antiport at the basilar surface of the tubular solution of SC-87 in chloroform, the pipettes are then flushed cells. Preliminary findings by Bomsztyk and Wright [19, 20] with toluene and air-dried. An additional precaution is to expell corroborate in vivo the presence of net calcium absorption in the sample from the constant volume pipette with 10 mM EGTA the absence of water transport, albeit in the presence of a to ensure complete removal of calcium after each sampling. slightly positive transepithelial voltage. The mechanism of calcium transport in the proximal convoWhen using glass scintillation vials, there may be a progressive decrease in 45Ca with time unless glacial acetic acid is also luted tubule has been studied only recently in vitro. Friedman et added to the scintillation fluid to solubilize calcium precipitates a! [21] studied the effect of lowering the peritubular sodium concentration on water absorption (J) in the proximal convo[5]. A second characteristic of calcium is its tendency to precipi- luted tubule and found an inhibitory effect which they attributed tate as calcium carbonate in alkaline solutions. This is generally to increased cytosolic calcium. In that same paper net calcium noticeable as whitish particulate matter which on bubbling with flux, as measured in a group of six tubules by helium glow 95% 02/5% CO2 redissolves. Use of solutions with none or low photometry, was 6.7 pEq/mm mi Isosmotic replacement of bicarbonate concentrations minimizes this problem. When solu- sodium in the bath by lithium or choline lowered this net tions with higher bicarbonate concentrations are used, the calcium absorption by 60%. This magnitude of net calcium flux solutions must be bubbled with carbon dioxide-containing gas compares favorably with in vivo findings of Murayama, Morel, mixtures and every precaution taken to avoid loss of carbon and LeGrimellec [17] and Quamme and Dirks [22] but is lower dioxide. Additionally, all oils coming in contact with solutions than those of Ullrich et at [23] and Frick et al [24]. Our laboratory has also examined the mechanism of calcium must be pre-bubbled with 100% CO2 to lessen the loss of carbon dioxide which, if not prevented, can lead to alkalinization and transport in the superficial proximal convoluted tubule in preliminary fashion [25]. In the absence of water transport or any electrochemical gradient, unidirectional efflux was 9.3 pEq/ mm mm and was not different from the unidirectional backflux —7.1 pEq/mm mm. Net calcium flux under these same condiReceived for publication June 15, 1982 tions was 0.43 pEq/mm mm, a value not different from zero; the calculated permeability (bath-to-lumen) was 4.5 x i0 cm! 0085—2538/82/0022—0492 $01.20 sec. These results suggest that calcium transport in this segment © 1982 by the International Society of Nephrology papers.
492
Divalent cation transport in isolated tubules
net calcium transport. Lowering the temperature from 37 to
Table 1. Calcium flux in cortical thick ascending limb JCa
NETa
Reference
lb pEq/inm mm
bI
2.00 12.58 3.90
0.99 3.50 2.48 0.42 0.39 160h
Rocha, Magaldi. and Kokko [21 Imai [4, 10] Shareghi and Stoner [3] Bourdeau and Burg [51 Suki et al [81
Shareghi and Agus [141
5.64 to 7.40 1.58
1.06 to 2.20 1.48 1.30" 0.82
1.48
250b
493
Net values were obtained independent of lb and bI values. Medullary thick ascending limb.
of the nephron is due primarily to passive forces, and there is little active transport, if any. Bomsztyk and Wright [20] also made similar preliminary observations in in vivo micropuncture studies. In the absence of any water transport, net calcium flux was determined by the electrochemical gradient (absorption with a positive gradient and secretion with a negative gradient). In addition, these investigators noted a residual component of absorption in the absence of any electrochemical gradient, suggesting a small component of active transport. Finally, an unconfirmed report suggested that 25-hydroxycholecalciferol may enhance unidirectional calcium efliux from 15 pEq/mm mm to 26 pEq/mm mm [26]. Proximal straight tubule (pars recta). Roughly 20% of the filtered calcium is reabsorbed in the loop of Henle [161. By the end of the superficial proximal convoluted tubule, the tubular fluid-to-plasma (TF/P) sodium:(TF/UF) calcium concentration ratio is approximately unity, suggesting parallel handling of sodium and calcium by this segment. However, micropuncture data at the tip of the 1oop of Henle indicates a dissociation in the handling of these two ions: concentration ratio of (TF/UF)Na/Ca = 1.5 was found in Psammomys [27] and in rats [281. Both
25°C, however, did inhibit calcium transport indicating a dependence on an energy-producing metabolic process. Almeida, Kudo and Rocha [29] in a preliminary communica-
tion similarly found net calcium transport of 7.3 to 9.4 pJEq/ mm mm in the pars recta, and neither ouabain nor use of a sodium-free perfusate altered this net calcium flux. Thin descending limb of Henle. Rocha, Magaldi, and Kokko [2] studied the thin descending limb of Henle and found a low apparent permeability to calcium (Pib = 0.76 x l0— cm/sec; Pbi = 0.92 X l0— cm/sec) with insignificant net transport even when an osmotic gradient (bath lumen) was imposed. Findings from our laboratory [61 confirm the low apparent permeability (Fib = 0.64 x l0— cm/sec) and insignificant transport in this segment.
Thin ascending limb of Henle. In similar fashion to the thin descending limb of Henle, apparent permeability to calcium in the thin ascending limb is low (Fib = 1.09 >< l0 cm/see; Pb! = 1.67
x l0 cm/sec) and is one to two orders of magnitude
lower than the permeability to chloride [2]. These findings by
Rocha and his colleagues do not suggest a significant net calcium transport in this nephron segment. Thick ascending limb of Henle. A number of papers dealing with calcium transport in the thick ascending limb of Henle have been published [2—5, 7—10, 141. These reports disagree on
whether transport of calcium is normally passive or active. However, it seems clear that hormones such as parathyroid hormone and calcitonin—both mediated by cyclic AMP—play an important role in the regulation of transport of calcium in this segment.
The magnitude of net calcium flux in this tubule segment
ranges from 0.8 to 7.4 pEq/mm mm [2—5, 7—10, 14] (Table 1).
Values from five laboratories are in the range of 1.0 pEq/ mm mm, or roughly net flux one order of magnitude lower than the proximal tubule. Values from Imai's laboratory [4, 101 have been significantly higher than those reported by Rocha, Magaldi, and Kokko [2], Shareghi and Stoner [3], Bourdeau and Burg [5], Suki et al [81, and Shareghi and Agus [14], and may be in part related to the use of different solutions and experimental conditions.
findings are in contrast to Lassiter's finding in hamsters (TF/ UF)NaICa = 0.5) [16], and suggest either nephron heterogeneity or calcium reabsorption in excess of sodium in the pars recta or Application of the Ussing or the Nernst equation [301 by descending thin limb of Henle. several investigators has shown flux ratios that exceed those Since studies by Rocha, Magaldi and Kokko [2] and from our laboratory [6] showed insignificant net calcium transport and low permeability to calcium in the thin descending limb, the proximal straight tubule was the likely site for this transport of calcium. Studies from our laboratory [6] confirm the proximal straight tubule as a major site of calcium transport. In flux studies of proximal straight tubules with 45Ca in the bath and perfusate, JNETCa was 11.2 pEq/mm mm; JNETCa calculated from the difference of JlbCI (lumen-to-bath, efflux) and Jbi (bath-to-lumen, influx) studies, was similar—l3.0
expected for simple passive diffusion [2, 4, 5, 8], while Shareghi
and Stoner [3] report a concentration ratio consistent with simple passive diffusion. To explain the high ratios, some investigators have invoked an active transport mechanism [2, 4, 8]. In the presence of a positive transepithelial voltage, Rocha, Magaldi, and Kokko [21 found a calcium flux ratio that exceeded the value predicted by the Ussing equation, suggesting either active transport or a membrane interactive process. Imai [4], on
the other hand, found calcium flux ratios that were high with both positive transepithelial voltage and zero voltage, suggest-
pEq/mm mm [6]. Since net transport occurred against a ing active calcium transport. He further characterized this concentration (collected:perfused concentration ratio = 0.88) and electrical gradient (PD —1.2 mY), transport in this segment was thought to be active. As perfusion rate was increased, net calcium transport also increased without evidence of saturability. When ouabain (l0 M) was added to the bath, transepithelial voltage and water absorption decreased to near zero, but there was no change in
transport as being saturable when the perfusate calcium concentration was increased. He examined metabolic inhibitors and found that with the addition of l0 M NaCN to the bath, calcium efflux decreased in parallel with a decrease in transepithelial voltage [4]. In contrast, iodoacetamide and ouabain did not affect calcium efflux despite a decrease in voltage. Ruthenium red, a presumed selective inhibitor of Ca-ATPase, reduced
494
!Vg et
transepithelial voltage but had no effect on calcium efflux. Finally, the effect of l0 M furosemide in the perfusate was examined. In contrast to the subsequent findings of Suki et al [8], Imai demonstrated a decrease in both voltage and calcium efflux in the cortical thick ascending limb. Bourdeau and Burg [5], on the other hand, showed a dependence of calcium transport on transepithelial voltage, with net absorption in the presence of lumen-positive voltage and net secretion with lumen-negative voltage, the latter artificially created by the use of furosemide and substitution of sodium nitrate for sodium chloride in the perfusate. Application of the Ussing flux ratio equation under these circumstances showed greater transport than could be accounted for by the voltage; there was no net transport with zero voltage. These findings are incompatible with simple passive transport but are consistent with single file pore diffusion as described by Hodgkin and Keynes [311. Because of clearance and micropuncture studies linking bicarbonaturia to enhanced reabsorption of calcium in the distal nephron [32, 331, Bourdeau and Burg [5] also exam-
al
calcium transport by l0 M 8-BrcAMP in both groups of tubules [91.
The response of the thick ascending limb of Henle to hormones has also been studied by other investigators [3, 7, 10, 14]. Parathyroid hormone (PTH) at a concentration of 0.1 to 1.0 stimulated net calcium absorption in the cortical thick ascending limb [7, 8, 10, 14]. This finding is consistent with the presence of PTH-sensitive adenylate cyclase activity in this segment [34]. Shareghi and Stoner [31 did not find an effect of PTH but used concentrations of PTH which were below the
threshold concentration as determined from the studies of Chabardes et al [34]. In a subsequent report, Shareghi and Agus [14] did find enhancement of calcium transport with 1.0 U/mI PTH in the bath. Imai [10], however, found that PTH enhanced
calcium efflux only if the perfusate contained 3 mM P04,
whereas no such augmentation was seen if the perfusate contained 1 mM P04. That PTH works in this segment via the adenylate cyclase system has been well substantiated by studies demonstrating that cyclic AMP analogues similarly enhance net ined the effect of 25 m bicarbonate in the perfusate. No calcium transport [7, 9, 10, 14]. change in net calcium absorption or transepithelial voltage was The mechanism whereby parathyroid hormone stimulates observed suggesting a site distal to the thick ascending limb calcium transport in this segment is of considerable interest. where bicarbonate augments calcium absorption. Bourdeau and Burg [5] had earlier shown voltage-dependent Voltage dependence of calcium transport has also been basal transport consistent with single file pore diffusion. Howreported recently by Shareghi and Agus [14]. These investiga- ever, in their subsequent studies [7], PTH at a concentration of tors confirmed enhanced net calcium absorption in the presence 1.0 U/ml increased lumen-to-bath flux without any effect on of a positive transepithelial voltage, secretion in the presence of backflux despite the absence of any change in transepithelial negative voltage, and no net transport with zero voltage. In voltage, This would appear to be incompatible with passive addition, doubling calcium concentration in the perfusate in- transport. When a "chemical voltage clamp" was applied and creased the net transport rate, while a doubling of the bath the luminal voltage was reduced to near zero, neither parathyconcentration of calcium reduced net transport: Altering mag- roid hormone nor 8-BrcAMP had any effect on calcium transnesium concentration either in the perfusate or bath did not port. In the presence of a lumen-negative voltage, however, parathyroid hormone was able to augment lumen-to-bath flux affect the net calcium flux. Our laboratory [81 has also examined the role of voltage and decrease backflux. This latter finding is incompatible with a changes in the thick ascending limb, and we find evidence for simple increase in passive permeability induced by the hormone axial heterogeneity. Medullary thick ascending limbs of Henle since one would have expected greater backflux and decreased were segregated from cortical thick limbs, and when furosemide lumen-to-bath flux in the presence of a lumen-negative voltage. In similar fashion, Shareghi and Agus [14] also found voltagewas placed intraluminally, transepithelial voltage decreased in both groups of tubules. However, calcium transport was re- independent PTH enhancement of net calcium transport but did duced only in the medullary thick ascending limb, consistent not examine the effect of PTH under conditions of zero or with passive movement of calcium in this segment. When the negative transepithelial voltage. Such findings again seem influx ratios were calculated for both groups of tubules, the ratio compatible with purely passive transport but would be consoin the cortical thick ascending limb of Henle exceeded that nant with the findings of Suki et al [81 in which a voltagepredicted for simple passive diffusion, in contrast to the ratio in independent component of active transport was enhanced with the medullary tubules which was consistent with simple passive PTH. Currently there is no satisfactory single explanation for these findings. However, it is possible that parathyroid hordiffusion. Further evidence of axial heterogeneity in medullary and mone uncovers or enhances an active transport mechanism of cortical thick limbs was advanced by the differing effects of low basal activity, or alternatively, it induces rectification of the PTH and calcitonin in these two groups of tubules [8, 9]. membrane barrier to calcium transport, allowing greater calciParathyroid hormone increased net calcium transport in the um efflux and lesser backflux [7]. Such a situation may also exist in the medullary thick ascending limb in which calcium cortical thick ascending limb from 1.38 to 2.30 pEq/mm mm but had no effect in medullary tubules [81. In marked contrast, transport appears to be passive according to the Ussing flux calcitonin augmented net calcium transport from 1.30 to 1.83 ratio equation [81, yet calcitonin or 8-BrcAMP is able to pEq/mm mm only in the medullary thick ascending limb with augment net transport without a change in transepithelial voltno change noted in cortical tubules [9]. These findings are age [9]. Distal convoluted tubule. Beyond the loop of Henle, an consistent with the presence of PTH- and calcitonin-sensitive adenylate cyclase activities in the cortical and medullary tu- estimated 10% of the filtered calcium is reabsorbed [16]. This bules, respectively [34, 35]. That the effects of both hormones portion of the nephron includes the distal convoluted tubule, on calcium transport are mediated via the adenylate cyclase which is an extremely complex segment in addition to being a system was substantiated further by the finding of enhanced net difficult segment to dissect. If defined as that portion of the
Divalent ration transport in isolated tubules
495
In a recent study Shareghi and Agus [13] seemingly confirmed the absence of any significant net calcium transport in this segment and a lack of response to PTH. Bourdeau and Hellstrom-Stein [12] also reported similar findings. They measured a very small net transport rate of 0.060 pEq/mm mm and found that calcium permeability was less than 2% that of the cortical thick ascending limb. Transport was voltage-dependent, as demonstrated under conditions of extreme luminal electronegativity induced by chronic treatment of rabbits with DOCA and following reduction of the voltage with amiloride. Imbert [381; however, the results would be compatible with the There was no evidence of active transport. presence of a portion of the connecting tubule, which has been In marked contrast to the above reports, Imai [10] and Holt demonstrated to have a PTH-sensitive adenylate cyclase [10]. and Lechene [111 report significant calcium transport in the The effect of the diuretics furosemide and amiloride was also cortical collecting tubule. Imai [10] found a net calcium efflux of studied in these tubules, and no effect on net calcium transport 12.8 pEq/mm mm, which is flux one to two orders of magninephron extending from the macula densa to the branching with
a collecting tubule, the segment will contain a portion of the thick ascending limb, the connecting tubule, as well as the true distal convoluted tubule [36, 37]. Shareghi and Stoner [31 studied net calcium transport in this segment, which may have included a portion of the connecting tubule. The control net calcium absorption was 0.42 pEq/mm mm and increased to 3.42 pEq!mm mm in response to PTH in the bath. This finding conflicts with the absence of PTH-sensitive adenylate cyclase activity in this segment as reported by Morel, Chabardes, and
was found [31.
tude greater than that obtained by other investigators [3, 11—13].
However, his findings that neither PTH nor dibutyryl cAMP enhance calcium absorption is in accord with the finding by Shareghi and Stoner [3] and Shareghi and Agus [13]. Thus, the distal site of PTH enhancement of calcium absorption appears to be proximal to the light segment of the cortical collecting 10.3 to 17.1 pEq/mm mm, which increased to 27.4 pEq/ tubule and is probably localized in the connecting segment as mm mm with the addition of PTH to the bath; calcium influx outlined by Imai [36]. was not altered. Dibutyryl cAMP l0- M in the bath mimicked Holt and Lechene [11] also noted significant net calcium the effect of PTH and augmented net calcium transport, con- efilux in the cortical collecting tubule on the order of 2.0 pEq/ firming the presence of PTH-sensitive adenylate cyclase activi- mm mm, which is comparable to the values obtained for ty in this segment of the nephron. However, it should be noted transport in the thick ascending limb of Henle although still that while the changes in net calcium flux appear to be significantly less than the value obtained by Imai [10]. These qualitatively consistent, the magnitude of calcium flux obtained investigators found an inhibitory effect of vasopressin on net Connecting tubule. The connecting tubule has been characterized recently hormonally by Imai [361. His study includes the granular portions of the distal convoluted tubule (formerly DCTg) and cortical collecting tubule (formerly CCTg) [36, 37]. Imai 1101 found in this segment a basal calcium efflux rate of
in Imai's laboratory in the distal nephron is one order of absorption of calcium which was mimicked by pro staglandin E2 magnitude greater than those of other laboratory findings [2, 3, 5, 7—9, 11]. Whether this is due to technical or experimental differences is unclear.
and reversed by meclofenamate, an inhibitor of prostaglandin synthesis. Since net calcium transport occurred in the absence of fluid
Three tubules consisting of the granular portions of the transport and in the presence of a lumen-negative voltage cortical collecting tubule were studied by Shareghi and Stoner (—12.1 mV) with equal concentrations of calcium in bath and [3]. A basal net absorptive rate for calcium of 0.74 pEq/ perfusate, transport appears to be active. This finding conflicts mm mm was stimulated to 2.16 pEq/mm mm in the presence with the report by Bourdeau and Hellstrom-Stein [121 of the of PTH. No firm conclusions were drawn because of the small absence of active transport and only a very small voltagenumber of tubules, but these findings are qualitatively consist- dependent net calcium flux. Explanations offered by Holt and Lechene [11] for the disparate findings include their use of ent with Imai's study [10]. It is becoming increasingly clear that the connecting tubule is immature rabbits and differences in length of the equilibration probably the distal site of enhanced calcium absorption respon- period (at least 180 mm in their study) prior to sampling. sive to PTH that other investigators had suggested from in vivo However, the reversibility of the inhibition of calcium transport studies [39—4 1], and may be responsible for a major fraction of mitigates against artifact of calcium binding to glass or complexthe filtered calcium reabsorbed in the terminal portion of the ing with various anions. Further studies will be needed to resolve the issue of magnitude of calcium flux and mechanism nephron. Collecting tubule. A number of investigators have examined of transport in this segment. Summary. From in vitro studies, the majority of calcium is calcium transport in the cortical collecting tubule [3, 10—13], and there is marked disagreement as to the magnitude of reabsorbed proximally in the convoluted and straight tubules. calcium transport. Shareghi and Stoner [3] first reported studies The absorptive rate ranges from 6.7 to 13 pEq/mm mm. of calcium transport in the light segment (CCT1) and found Recent preliminary studies suggest that transport in the convoinsignificant net transport—0.20 pEq!mm mm; the addition of luted segment is primarily passive although there may be a PTH or dibutyryl cAMP to the bath did not alter the magnitude small component of active transport. Transport in the proximal of transport. To obviate any possible effects of phosphate in the straight tubule may be active. There is no net transport in the perfusate, bicarbonate was substituted, and change in net thin descending or ascending limbs. The thick ascending limbs calcium absorption was not observed. This is an interesting transport calcium on the order of 1.0 pEq/mm mm. Transport finding in that clearance and micropuncture studies have linked in the cortical, and probably the medullary, thick limb is bicarbonaturia with enhanced calcium reabsorption [32, 33], voltage-dependent. The mechanism of transport may be via and an in vitro study by Bourdeau and Burg [5] had already single file pore diffusion. PTH uncovers active transport in the excluded the thick ascending limb as the site of enhanced cortical limb, whereas calcitonin uncovers active transport in the medullary limb. Calcium absorption in the DCT and concalcium transport.
496
Ng
ci
a!
effect of one cation on the other. Whether this is due simply to thick ascending limb; PTH stimulates transport in both seg- increased backflux along a concentration gradient or to some ments. The magnitude of calcium transport in the cortical other mechanism is unknown. Thus, magnesium and calcium collecting tubule (CCT1) is unclear and ranges from insignificant luminal transport appear to be independent of one another in to as great as that in the thick ascending limb. PTH has no effect this segment in contrast to clearance [45] and micropuncture fleeting tubule (DCTg + CCTg) is probably less than that of the
in this segment, and the mechanism of transport remains [46] studies suggesting a competitive transport mechanism, and unclear.
Magnesium transport Although magnesium is often considered with calcium, it has been until recently an infrequently studied divalent cation. The
reason for this lies with the difficulty in obtaining and using isotopic 28Mg which has a half-life of 21.3 hr. Thus micropunc-
ture studies and isolated tubule studies have been limited to centers with electron microprobe capabilities.
Unlike virtually any other ion or solute, magnesium is relatively impermeant in the proximal convoluted tubule, and only 20 to 30% of the filtered magnesium is reabsorbed in this segment of the nephron. The majority of the magnesium—SO to 60%—is reab sorbed in the loop of Henle with an important role ascribed to the thick ascending limb [42, 431. The distal convoluted tubule may reabsorb 2 to 5%, while the remainder of the nephron may reabsorb 1 to 3% of the filtered magnesium [431. Two publications [13, 14] and one preliminary report [441 deal
there appears to be no competitive effect of these two ions at the basolateral membrane [21, 43]. Finally, in the cortical collecting tubule, Shareghi and Agus [13] could not demonstrate any significant net magnesium transport or any effect of PTH to increase this transport. Summary, Permeability to magnesium is low in the proximal straight, and probably convoluted, tubule. The thin descending and ascending limbs have not been studied. The cortical thick ascending limb transports magnesium in a voltage-dependent way similar to calcium. PTH and dibutyryl cAMP uncover active transport in the cortical thick ascending limb, which has a basal absorptive rate of 0.8 to 0.9 pEq/mm mm. The distal convoluted tubule and connecting tubule have not been studied. The cortical collecting tubule has no significant net transport, and there is no effect of PTH to enhance transport. Acknowledgments This work was supported by National Institutes of Health grants AM00770, AM-00424, and AM-2 1394. Drs. Ng and Peraino are recipients of
with magnesium transport in isolated tubules. In the proximal straight tubule Quamme [44] examined the effect of increasing Clinical Investigator Awards. The secretarial assistance of Ms. L. magnesium concentration in the perfusate of superficial and Brazil is gratefully acknowledged. juxtamedullary pars recta. In superficial tubules, raising the Reprint requests to Dr. W. N. Suki, Renal Section, The Methodist perfusate magnesium concentration from 0.7 to 3.1 m elevated Hospital, MS F505, 6565 Fannin, Houston, Texas 77030, USA the net absorptive rate from 0.44 to 0.88 pEq/mm mm. Similarly, injuxtamedullary tubules, the net absorptive rate rose from References
0.92 to 1.20 pEq/mm mm. Elevating the bath magnesium concentration from 0.5 to 1.85 mr'i did not result in secretion of
magnesium in either superficial or juxtamedullary tubules. Thus, in this segment of the nephron, permeability to magnesium is low as compared with sodium and calcium, and there is no active or passive secretion. Magnesium transport appears to be dependent on luminal concentration and is probably passive in view of the collected-to-perfused magnesium concentration ratios which were close to 1.1. In the cortical thick ascending limb Shareghi and Agus [141
found a net reabsorptive rate of 0.64 pEq!mm mm with a lumen-positive transepithelial voltage of +8.8 mY. When the voltage was increased to + 13.6 mV with hypotonic perfusate,
the net reabsorptive rate increased to 0.90 pEq/mm mm. Similarly, when the voltage was changed to —10.4 mY with furosemide and an isosmolar, hyponatric bath, net secretion of —0.90 pEq/mm mm was demonstrated. The correlation between voltage (x) and net transport (v) was y = 0.055x — 0.042, r = 0.69, P < 0.001. Thus, magnesium transport in this segment appears to be voltage-dependent in a fashion similar to calcium. In analogous fashion with calcium, PTH and dibutyryl cAMP enhanced magnesium transport in the cortical thick ascending
limb from 0.94 to 1.78 pEq/mm mm [14]. Doubling luminal magnesium concentration to 1.4 mivi increased magnesium absorption but had no effect on calcium absorption. When luminal calcium concentration was doubled to 2.5 mM, net calcium transport was increased, but magnesium absorption was not affected. When the bath concentration of magnesium or
calcium was doubled, net outward transport of either magnesium or calcium, respectively, was reduced, but there was no
1. BURG M, GRANTHAM J, ABRAMOW M, ORLOFF J: Preparation and
study of fragments of single rabbit nephrons. Am J Physiol 210:1293—1298, 1966
2. ROCHA AS, MAGALDI JB, KoKK0 JP: Calcium and phosphate transport in isolated segments of rabbit Henle's loop. J C/in Invest 59:975—983, 1977
3. SHAREGHI GR, STONER LC: Calcium transport across segments of the rabbit distal nephron in vitro. Am J Physiol 235:F367—F375, 1978 1MM M: Calcium
transport across the rabbit thick ascending limb of Henle's loop perfused in vitro. Pflugers Arch 374:255—263, 1978 5. BOURDEAU JE, BURG MB: Voltage dependence of calcium trans4.
port in the thick ascending limb of Henle's loop. Am J Physiol 236:F357—F364, 1979
6. RoUSE D, NG RCK, SuKI WN: Calcium transport in the pars recta and thin descending limb of Henle of the rabbit, perfused in vitro. J Clin Invest 65:37—42, 1980 7. BOURDEAU JE, BURG MB:
Effect of PTH on calcium transport across the cortical thick ascending limb of Henle's loop. Am Physiol 239:F121—F126, 1980
J
8.
SUKI WN, ROUSED, NG RCK, KOKKO JP: Calcium transport in the
thick ascending limb of Henle. Heterogeneity of function in the medullary and cortical segments. J C/in Invest 66:1004—1009, 1980 9. SUM WN, ROUSE D: Hormonal regulation of calcium transport in thick ascending limb renal tubules. Am J Physiol 24l:F17l—F174, 1981
10. IMAI M: Effects of parathyroid hormone and N6,02-dibutyryl cyclic
AMP on Ca transport across the rabbit distal nephron segments perfused in vitro. P//tigers Arch 390:145—151, 1981 11. HOLT WF, LECHENE C: ADH-PGE2 interactions in cortical collect-
ing tubule. II.
Inhibition of
Ca and P reabsorption. Am J Physiol
241:F461—F467, 1981 12.
BOURDEAU JE, HELLSTROM-STEIN Ri: Voltage-dependent calcium
movement across the cortical collecting duct. Am J Physiol 242:F285—F292, 1982 13. SHAREGIII
GR, AGus ZS: Phosphate transport in the light segment
Divalent cation transport in isolated tubules
of the rabbit cortical collecting tubule. Am
J Physiol 242:F379—
F384, 1982 14. SHAREGHI GR, AGUS ZS: Magnesium transport in the cortical thick
ascending limb of Henle's loop of the rabbit. J
Clin Invest 69:759—
497
isolated perfused pars recta of proximal tubule, Abst Proc VII Int Congr Nephrol, Montreal, 1978, p. E16 30. UssiNo HH: The distinction by means of tracers between active transport and diffusion. The transfer of iodide across the isolated
and other electrolytes from microvolume samples. Am J Physiol
frog skin. Acta Physiol Scand 19:43—56, 1949 31. HODGKIN AL, KEYNES RD:The potassium permeability of a giant nerve fibre. J Physiol (Land) 128:61—88, 1955
242:F202—F206, 1982 16. LAS5ITER WE, GOTTSCHALK CW, MYLLE M: Micropuncture study
32. PE1WN0 RA, SUKI WN: Urine HC03 augments renal CA2 absorption independent of systemic acid-base changes. Am J
769,
1982
15. MUHLERT M, JULITA M, QUAMME G: Disappearance of calcium
of renal tubular reabsorption of calcium in normal rodents. Am J Physiol 204:771—775, 1963
17. MURAYAMA Y, MOREL F, LEGRIMELLEC C: Phosphate, calcium,
Physiol 238:F394—F398, 1980 33. SUTTON RAL, WONG NLM, DIRK JH: Effects of metabolic acidosis and alkalosis on sodium and calcium transport in the dog kidney. Kidney mt 15:520—533, 1979
and magnesium transfer in proximal tubules and 1oops of Henle, as measured by single nephron microperfusion experiments in the rat. Pflugers Arch 333:1—16, 1972 18. ULLRICH KJ, RUMRICH G, KLoss S: Active CA2 reabsorption in the proximal tubule of the rat kidney. Dependence on sodium and buffer transport. Pflugers Arch 364:223—228, 1976 19. BOMSZTYK K, WRIGHT FS: Effects of transepithelial fluid flux on transepithelial voltage and transport of calcium, sodium, chloride
nephron. Pflugers Arch 354:229—239, 1975 35. CHABARDES D, IMBERT-TEBOUL M, MONTEGUT M, CLIQUE A,
and potassium by renal proximal tubule (abstract). Kidney Int
36. IMAI M: The connecting tubule: A functional subdivision of the
21:269, 1982 20. BOMSZTYK K, WRIGHT ES: Contribution of diffusion to transport of calcium by renal proximal tubule (abstract). Clin Res 30:443A, 1982 21. FRIEDMAN PA, FIGUEIREDO JF, MAACK T, WINDHAGER EE:
rabbit distal nephron segments. Kidney Int 15:346—356, 1979 37. JACOBSON HR: Functional segmentation of the mammalian nephron. Am J Physiol 241:F203—F218, 1981 38. MOREL F, CHABARDES D, IMBERT M: Functional segmentation of the rabbit distal tubule by microdetermination of hormone-dependent adenylate cyclase activity. Kidney mt 9:264—277, 1976 39. SUTTON RAL, WONG NLM, DIRKS JH: Effects of parathyroid hormone on sodium and calcium transport in the dog nephron. Clin
Sodium-calcium interactions in the renal proximal convoluted tubule of the rabbit. Am J Physiol 240:F558—F568, 1981 22. QUAMME GA, DIRKS JH: Intraluminal and contraluminal magnesium on magnesium and calcium transfer in the rat nephron. Am J Physiol 238:Fl87—Fl98, 1980 23. ULLRICH KJ, CAPA550 G, RUMRICH G, KLOSS S: Effect of parathy-
roid hormone on the active Ca2-reabsorption in the proximal convolution of the rat kidney (abstract). Pfldgers Arch 359:Rl 18, 1975
24. FRICK A, RUMRICH G, ULLRICH KJ, LA5SITER WE: Microperfu-
sion study of calcium transport in the proximal tubule of the rat kidney. Pfiugers Arch 286:109—117, 1965 25. NG RCK, Suiu WN: Calcium transport in the superficial proximal convoluted tubule of the rabbit (abstract). Clin Res 30:460A, 1982 26. BRENNAN JP, K0KKO JP: Effect of 25-hydroxycholecalciferol on the transport of Ca and Na in the proximal tubule (abstract). Clin Res 21:678, 1973 27. DE ROUFFIGNAC C, MOREL F, Moss N, ROINEL N: Micropuncture
study of water and electrolyte movements along the ioop of Henle in Psammomys with special reference to magnesium, calcium and phosphorus. Pflügers Arch 344:309—326, 1973
28. JAMISON RL, FREY NR, LACY FB: Calcium reabsorption in the thin loop of Henle. Am J Physiol 227:745—751, 1974
29. ALMEIDA AL KUDO LH, ROCHA AS: Calcium transport in
34. CHABARDES D, IMBERT M, MONTEGUT M, MOREL F: PTH sensi-
tive adenyl cyclase activity in different segments of the rabbit MOREL F: Distribution of calcitonin-sensitive adenylate cyclase activity along the rabbit kidney tubule. Proc NatI Acad Sci USA 73:3608—3612, 1976
Sci 51:345—351, 1976
40. AGUS ZS, CHIU PJS, GOLDBERG M: Regulation of urinary calcium excretion in the rat. Am J Physiol 232:F545—F549, 1977 41. HARRIS CA, BURNATOWSKA MA, SEELY JF, SUTTON RAL,
QUAMME GA, DIRKS JH: Effects of parathyroid hormone on
electrolyte transport in the hamster nephron. Am J Physiol 236:F342—F348, 1979 42. BRUNETTE MG, VIGNEAULT N, CARRIERE S: Micropuncture study
of renal magnesium transport in magnesium-loaded rats. Am J Physiol 229:1695—1701, 1975 43. QUAMME GA, DIRKS JH: Magnesium transport in the nephron. Am J Physiol 239:F393—F401, 1980 44. QUAMME GA: Magnesium transport in isolated proximal straight
tubules of the rabbit (abstract). Kidney mt 21:139, 1982
45. MA55RY SG, COBURN JW, CHAPMAN LW, KLEEMAN CR: Effect of
NaC1 infusion on urinary Ca2 and Mg2 during reduction in their filtered loads. Am J Physiol 213:1218—1224, 1967 46. LE GRIMELLEC C, ROINEL N, MOREL F: Simultaneous Mg, Ca, P,
K, Na and Cl analysis in rat tubular fluid. III. During acute Ca plasma loading. Pflugers Arch 346:171—188, 1974
Divalent cation transport in isolated tubules ROLAND C.K. No, ROBERT A. PERAINO, and WAD! N. SuKI Department of Medicine, Baylor College of Medicine and The Methodist Hospital, Houston, Texas
Since Burg introduced the technique of microperfusion of precipitation of calcium carbonate or binding of calcium and isolated renal tubules in 1966 [11, a number of papers has been loss of calcium to the oil-water interface [15]. Proximal convoluted tubule. Approximately 60% of the filpublished on sodium and water transport in various segments of the nephron. However, it was not until 1977 that transport of tered calcium is reabsorbed in the proximal convoluted tubule
divalent ions was described using this technique [21. Subse- [161. The bulk of calcium reabsorption in this segment is quently, 12 additional papers were published dealing with generally thought to be passive primarily because of(1) parallel calcium transport in various segments of the nephron [3—14].
handling of sodium and calcium such that the tubular fluid-to-
This review will deal primarily with the contents of these plasma (TF/P) sodium: (TF/UF) calcium concentration ratio is nearly unity under nondiuretic conditions [161, and (2) in situ microperfusion experiments demonstrating high permeability to Calcium transport calcium (an unidirectional efflux nearly two and one-half times Methodologic pitfalls. In studying calcium transport, the larger than the net flux) [171. However, two lines of evidence investigator must be cognizant of certain characteristics pecu- point to an element of active reabsorption: (I) The tubular fluidliar to calcium which may lead to artifactual errors. The first is to-ultrafiltrable calcium concentration ratio (TF/UF)ca falls to its propensity to bind to glass. Errors related to bound calcium less than unity in the presence of a mannitol diuresis 116]; and will lead to inaccuracies in the direction of greater transport. (2) in the presence of an equilibrium solution in the lumen, and Coating the perfusing, collecting, and constant volume pipettes therefore, absent convective forces, more calcium is transportwith a layer of silicon using Prosil-28 (PCR, Gainesville, Flori- ed than can be accounted for by the electrochemical gradient da), Sigmacote (Sigma Chemical Co., St. Louis, Missouri), [18]. Furthermore, ouabain, or replacement of sodium with Siliclad (Clay Adams, New York, New York), or Dow Corning choline or lithium on the peritubular side, inhibits the active 7 compound (Dow Corning, Midland, Michigan) will minimize component of calcium transport suggesting the presence of a this error. After coating the inner surface with a 6% vol/vol sodium-calcium antiport at the basilar surface of the tubular solution of SC-87 in chloroform, the pipettes are then flushed cells. Preliminary findings by Bomsztyk and Wright [19, 20] with toluene and air-dried. An additional precaution is to expell corroborate in vivo the presence of net calcium absorption in the sample from the constant volume pipette with 10 mM EGTA the absence of water transport, albeit in the presence of a to ensure complete removal of calcium after each sampling. slightly positive transepithelial voltage. The mechanism of calcium transport in the proximal convoWhen using glass scintillation vials, there may be a progressive decrease in 45Ca with time unless glacial acetic acid is also luted tubule has been studied only recently in vitro. Friedman et added to the scintillation fluid to solubilize calcium precipitates a! [21] studied the effect of lowering the peritubular sodium concentration on water absorption (J) in the proximal convo[5]. A second characteristic of calcium is its tendency to precipi- luted tubule and found an inhibitory effect which they attributed tate as calcium carbonate in alkaline solutions. This is generally to increased cytosolic calcium. In that same paper net calcium noticeable as whitish particulate matter which on bubbling with flux, as measured in a group of six tubules by helium glow 95% 02/5% CO2 redissolves. Use of solutions with none or low photometry, was 6.7 pEq/mm mi Isosmotic replacement of bicarbonate concentrations minimizes this problem. When solu- sodium in the bath by lithium or choline lowered this net tions with higher bicarbonate concentrations are used, the calcium absorption by 60%. This magnitude of net calcium flux solutions must be bubbled with carbon dioxide-containing gas compares favorably with in vivo findings of Murayama, Morel, mixtures and every precaution taken to avoid loss of carbon and LeGrimellec [17] and Quamme and Dirks [22] but is lower dioxide. Additionally, all oils coming in contact with solutions than those of Ullrich et at [23] and Frick et al [24]. Our laboratory has also examined the mechanism of calcium must be pre-bubbled with 100% CO2 to lessen the loss of carbon dioxide which, if not prevented, can lead to alkalinization and transport in the superficial proximal convoluted tubule in preliminary fashion [25]. In the absence of water transport or any electrochemical gradient, unidirectional efflux was 9.3 pEq/ mm mm and was not different from the unidirectional backflux —7.1 pEq/mm mm. Net calcium flux under these same condiReceived for publication June 15, 1982 tions was 0.43 pEq/mm mm, a value not different from zero; the calculated permeability (bath-to-lumen) was 4.5 x i0 cm! 0085—2538/82/0022—0492 $01.20 sec. These results suggest that calcium transport in this segment © 1982 by the International Society of Nephrology papers.
492
Divalent cation transport in isolated tubules
net calcium transport. Lowering the temperature from 37 to
Table 1. Calcium flux in cortical thick ascending limb JCa
NETa
Reference
lb pEq/inm mm
bI
2.00 12.58 3.90
0.99 3.50 2.48 0.42 0.39 160h
Rocha, Magaldi. and Kokko [21 Imai [4, 10] Shareghi and Stoner [3] Bourdeau and Burg [51 Suki et al [81
Shareghi and Agus [141
5.64 to 7.40 1.58
1.06 to 2.20 1.48 1.30" 0.82
1.48
250b
493
Net values were obtained independent of lb and bI values. Medullary thick ascending limb.
of the nephron is due primarily to passive forces, and there is little active transport, if any. Bomsztyk and Wright [20] also made similar preliminary observations in in vivo micropuncture studies. In the absence of any water transport, net calcium flux was determined by the electrochemical gradient (absorption with a positive gradient and secretion with a negative gradient). In addition, these investigators noted a residual component of absorption in the absence of any electrochemical gradient, suggesting a small component of active transport. Finally, an unconfirmed report suggested that 25-hydroxycholecalciferol may enhance unidirectional calcium efliux from 15 pEq/mm mm to 26 pEq/mm mm [26]. Proximal straight tubule (pars recta). Roughly 20% of the filtered calcium is reabsorbed in the loop of Henle [161. By the end of the superficial proximal convoluted tubule, the tubular fluid-to-plasma (TF/P) sodium:(TF/UF) calcium concentration ratio is approximately unity, suggesting parallel handling of sodium and calcium by this segment. However, micropuncture data at the tip of the 1oop of Henle indicates a dissociation in the handling of these two ions: concentration ratio of (TF/UF)Na/Ca = 1.5 was found in Psammomys [27] and in rats [281. Both
25°C, however, did inhibit calcium transport indicating a dependence on an energy-producing metabolic process. Almeida, Kudo and Rocha [29] in a preliminary communica-
tion similarly found net calcium transport of 7.3 to 9.4 pJEq/ mm mm in the pars recta, and neither ouabain nor use of a sodium-free perfusate altered this net calcium flux. Thin descending limb of Henle. Rocha, Magaldi, and Kokko [2] studied the thin descending limb of Henle and found a low apparent permeability to calcium (Pib = 0.76 x l0— cm/sec; Pbi = 0.92 X l0— cm/sec) with insignificant net transport even when an osmotic gradient (bath lumen) was imposed. Findings from our laboratory [61 confirm the low apparent permeability (Fib = 0.64 x l0— cm/sec) and insignificant transport in this segment.
Thin ascending limb of Henle. In similar fashion to the thin descending limb of Henle, apparent permeability to calcium in the thin ascending limb is low (Fib = 1.09 >< l0 cm/see; Pb! = 1.67
x l0 cm/sec) and is one to two orders of magnitude
lower than the permeability to chloride [2]. These findings by
Rocha and his colleagues do not suggest a significant net calcium transport in this nephron segment. Thick ascending limb of Henle. A number of papers dealing with calcium transport in the thick ascending limb of Henle have been published [2—5, 7—10, 141. These reports disagree on
whether transport of calcium is normally passive or active. However, it seems clear that hormones such as parathyroid hormone and calcitonin—both mediated by cyclic AMP—play an important role in the regulation of transport of calcium in this segment.
The magnitude of net calcium flux in this tubule segment
ranges from 0.8 to 7.4 pEq/mm mm [2—5, 7—10, 14] (Table 1).
Values from five laboratories are in the range of 1.0 pEq/ mm mm, or roughly net flux one order of magnitude lower than the proximal tubule. Values from Imai's laboratory [4, 101 have been significantly higher than those reported by Rocha, Magaldi, and Kokko [2], Shareghi and Stoner [3], Bourdeau and Burg [5], Suki et al [81, and Shareghi and Agus [14], and may be in part related to the use of different solutions and experimental conditions.
findings are in contrast to Lassiter's finding in hamsters (TF/ UF)NaICa = 0.5) [16], and suggest either nephron heterogeneity or calcium reabsorption in excess of sodium in the pars recta or Application of the Ussing or the Nernst equation [301 by descending thin limb of Henle. several investigators has shown flux ratios that exceed those Since studies by Rocha, Magaldi and Kokko [2] and from our laboratory [6] showed insignificant net calcium transport and low permeability to calcium in the thin descending limb, the proximal straight tubule was the likely site for this transport of calcium. Studies from our laboratory [6] confirm the proximal straight tubule as a major site of calcium transport. In flux studies of proximal straight tubules with 45Ca in the bath and perfusate, JNETCa was 11.2 pEq/mm mm; JNETCa calculated from the difference of JlbCI (lumen-to-bath, efflux) and Jbi (bath-to-lumen, influx) studies, was similar—l3.0
expected for simple passive diffusion [2, 4, 5, 8], while Shareghi
and Stoner [3] report a concentration ratio consistent with simple passive diffusion. To explain the high ratios, some investigators have invoked an active transport mechanism [2, 4, 8]. In the presence of a positive transepithelial voltage, Rocha, Magaldi, and Kokko [21 found a calcium flux ratio that exceeded the value predicted by the Ussing equation, suggesting either active transport or a membrane interactive process. Imai [4], on
the other hand, found calcium flux ratios that were high with both positive transepithelial voltage and zero voltage, suggest-
pEq/mm mm [6]. Since net transport occurred against a ing active calcium transport. He further characterized this concentration (collected:perfused concentration ratio = 0.88) and electrical gradient (PD —1.2 mY), transport in this segment was thought to be active. As perfusion rate was increased, net calcium transport also increased without evidence of saturability. When ouabain (l0 M) was added to the bath, transepithelial voltage and water absorption decreased to near zero, but there was no change in
transport as being saturable when the perfusate calcium concentration was increased. He examined metabolic inhibitors and found that with the addition of l0 M NaCN to the bath, calcium efflux decreased in parallel with a decrease in transepithelial voltage [4]. In contrast, iodoacetamide and ouabain did not affect calcium efflux despite a decrease in voltage. Ruthenium red, a presumed selective inhibitor of Ca-ATPase, reduced
494
!Vg et
transepithelial voltage but had no effect on calcium efflux. Finally, the effect of l0 M furosemide in the perfusate was examined. In contrast to the subsequent findings of Suki et al [8], Imai demonstrated a decrease in both voltage and calcium efflux in the cortical thick ascending limb. Bourdeau and Burg [5], on the other hand, showed a dependence of calcium transport on transepithelial voltage, with net absorption in the presence of lumen-positive voltage and net secretion with lumen-negative voltage, the latter artificially created by the use of furosemide and substitution of sodium nitrate for sodium chloride in the perfusate. Application of the Ussing flux ratio equation under these circumstances showed greater transport than could be accounted for by the voltage; there was no net transport with zero voltage. These findings are incompatible with simple passive transport but are consistent with single file pore diffusion as described by Hodgkin and Keynes [311. Because of clearance and micropuncture studies linking bicarbonaturia to enhanced reabsorption of calcium in the distal nephron [32, 331, Bourdeau and Burg [5] also exam-
al
calcium transport by l0 M 8-BrcAMP in both groups of tubules [91.
The response of the thick ascending limb of Henle to hormones has also been studied by other investigators [3, 7, 10, 14]. Parathyroid hormone (PTH) at a concentration of 0.1 to 1.0 stimulated net calcium absorption in the cortical thick ascending limb [7, 8, 10, 14]. This finding is consistent with the presence of PTH-sensitive adenylate cyclase activity in this segment [34]. Shareghi and Stoner [31 did not find an effect of PTH but used concentrations of PTH which were below the
threshold concentration as determined from the studies of Chabardes et al [34]. In a subsequent report, Shareghi and Agus [14] did find enhancement of calcium transport with 1.0 U/mI PTH in the bath. Imai [10], however, found that PTH enhanced
calcium efflux only if the perfusate contained 3 mM P04,
whereas no such augmentation was seen if the perfusate contained 1 mM P04. That PTH works in this segment via the adenylate cyclase system has been well substantiated by studies demonstrating that cyclic AMP analogues similarly enhance net ined the effect of 25 m bicarbonate in the perfusate. No calcium transport [7, 9, 10, 14]. change in net calcium absorption or transepithelial voltage was The mechanism whereby parathyroid hormone stimulates observed suggesting a site distal to the thick ascending limb calcium transport in this segment is of considerable interest. where bicarbonate augments calcium absorption. Bourdeau and Burg [5] had earlier shown voltage-dependent Voltage dependence of calcium transport has also been basal transport consistent with single file pore diffusion. Howreported recently by Shareghi and Agus [14]. These investiga- ever, in their subsequent studies [7], PTH at a concentration of tors confirmed enhanced net calcium absorption in the presence 1.0 U/ml increased lumen-to-bath flux without any effect on of a positive transepithelial voltage, secretion in the presence of backflux despite the absence of any change in transepithelial negative voltage, and no net transport with zero voltage. In voltage, This would appear to be incompatible with passive addition, doubling calcium concentration in the perfusate in- transport. When a "chemical voltage clamp" was applied and creased the net transport rate, while a doubling of the bath the luminal voltage was reduced to near zero, neither parathyconcentration of calcium reduced net transport: Altering mag- roid hormone nor 8-BrcAMP had any effect on calcium transnesium concentration either in the perfusate or bath did not port. In the presence of a lumen-negative voltage, however, parathyroid hormone was able to augment lumen-to-bath flux affect the net calcium flux. Our laboratory [81 has also examined the role of voltage and decrease backflux. This latter finding is incompatible with a changes in the thick ascending limb, and we find evidence for simple increase in passive permeability induced by the hormone axial heterogeneity. Medullary thick ascending limbs of Henle since one would have expected greater backflux and decreased were segregated from cortical thick limbs, and when furosemide lumen-to-bath flux in the presence of a lumen-negative voltage. In similar fashion, Shareghi and Agus [14] also found voltagewas placed intraluminally, transepithelial voltage decreased in both groups of tubules. However, calcium transport was re- independent PTH enhancement of net calcium transport but did duced only in the medullary thick ascending limb, consistent not examine the effect of PTH under conditions of zero or with passive movement of calcium in this segment. When the negative transepithelial voltage. Such findings again seem influx ratios were calculated for both groups of tubules, the ratio compatible with purely passive transport but would be consoin the cortical thick ascending limb of Henle exceeded that nant with the findings of Suki et al [81 in which a voltagepredicted for simple passive diffusion, in contrast to the ratio in independent component of active transport was enhanced with the medullary tubules which was consistent with simple passive PTH. Currently there is no satisfactory single explanation for these findings. However, it is possible that parathyroid hordiffusion. Further evidence of axial heterogeneity in medullary and mone uncovers or enhances an active transport mechanism of cortical thick limbs was advanced by the differing effects of low basal activity, or alternatively, it induces rectification of the PTH and calcitonin in these two groups of tubules [8, 9]. membrane barrier to calcium transport, allowing greater calciParathyroid hormone increased net calcium transport in the um efflux and lesser backflux [7]. Such a situation may also exist in the medullary thick ascending limb in which calcium cortical thick ascending limb from 1.38 to 2.30 pEq/mm mm but had no effect in medullary tubules [81. In marked contrast, transport appears to be passive according to the Ussing flux calcitonin augmented net calcium transport from 1.30 to 1.83 ratio equation [81, yet calcitonin or 8-BrcAMP is able to pEq/mm mm only in the medullary thick ascending limb with augment net transport without a change in transepithelial voltno change noted in cortical tubules [9]. These findings are age [9]. Distal convoluted tubule. Beyond the loop of Henle, an consistent with the presence of PTH- and calcitonin-sensitive adenylate cyclase activities in the cortical and medullary tu- estimated 10% of the filtered calcium is reabsorbed [16]. This bules, respectively [34, 35]. That the effects of both hormones portion of the nephron includes the distal convoluted tubule, on calcium transport are mediated via the adenylate cyclase which is an extremely complex segment in addition to being a system was substantiated further by the finding of enhanced net difficult segment to dissect. If defined as that portion of the
Divalent ration transport in isolated tubules
495
In a recent study Shareghi and Agus [13] seemingly confirmed the absence of any significant net calcium transport in this segment and a lack of response to PTH. Bourdeau and Hellstrom-Stein [12] also reported similar findings. They measured a very small net transport rate of 0.060 pEq/mm mm and found that calcium permeability was less than 2% that of the cortical thick ascending limb. Transport was voltage-dependent, as demonstrated under conditions of extreme luminal electronegativity induced by chronic treatment of rabbits with DOCA and following reduction of the voltage with amiloride. Imbert [381; however, the results would be compatible with the There was no evidence of active transport. presence of a portion of the connecting tubule, which has been In marked contrast to the above reports, Imai [10] and Holt demonstrated to have a PTH-sensitive adenylate cyclase [10]. and Lechene [111 report significant calcium transport in the The effect of the diuretics furosemide and amiloride was also cortical collecting tubule. Imai [10] found a net calcium efflux of studied in these tubules, and no effect on net calcium transport 12.8 pEq/mm mm, which is flux one to two orders of magninephron extending from the macula densa to the branching with
a collecting tubule, the segment will contain a portion of the thick ascending limb, the connecting tubule, as well as the true distal convoluted tubule [36, 37]. Shareghi and Stoner [31 studied net calcium transport in this segment, which may have included a portion of the connecting tubule. The control net calcium absorption was 0.42 pEq/mm mm and increased to 3.42 pEq!mm mm in response to PTH in the bath. This finding conflicts with the absence of PTH-sensitive adenylate cyclase activity in this segment as reported by Morel, Chabardes, and
was found [31.
tude greater than that obtained by other investigators [3, 11—13].
However, his findings that neither PTH nor dibutyryl cAMP enhance calcium absorption is in accord with the finding by Shareghi and Stoner [3] and Shareghi and Agus [13]. Thus, the distal site of PTH enhancement of calcium absorption appears to be proximal to the light segment of the cortical collecting 10.3 to 17.1 pEq/mm mm, which increased to 27.4 pEq/ tubule and is probably localized in the connecting segment as mm mm with the addition of PTH to the bath; calcium influx outlined by Imai [36]. was not altered. Dibutyryl cAMP l0- M in the bath mimicked Holt and Lechene [11] also noted significant net calcium the effect of PTH and augmented net calcium transport, con- efilux in the cortical collecting tubule on the order of 2.0 pEq/ firming the presence of PTH-sensitive adenylate cyclase activi- mm mm, which is comparable to the values obtained for ty in this segment of the nephron. However, it should be noted transport in the thick ascending limb of Henle although still that while the changes in net calcium flux appear to be significantly less than the value obtained by Imai [10]. These qualitatively consistent, the magnitude of calcium flux obtained investigators found an inhibitory effect of vasopressin on net Connecting tubule. The connecting tubule has been characterized recently hormonally by Imai [361. His study includes the granular portions of the distal convoluted tubule (formerly DCTg) and cortical collecting tubule (formerly CCTg) [36, 37]. Imai 1101 found in this segment a basal calcium efflux rate of
in Imai's laboratory in the distal nephron is one order of absorption of calcium which was mimicked by pro staglandin E2 magnitude greater than those of other laboratory findings [2, 3, 5, 7—9, 11]. Whether this is due to technical or experimental differences is unclear.
and reversed by meclofenamate, an inhibitor of prostaglandin synthesis. Since net calcium transport occurred in the absence of fluid
Three tubules consisting of the granular portions of the transport and in the presence of a lumen-negative voltage cortical collecting tubule were studied by Shareghi and Stoner (—12.1 mV) with equal concentrations of calcium in bath and [3]. A basal net absorptive rate for calcium of 0.74 pEq/ perfusate, transport appears to be active. This finding conflicts mm mm was stimulated to 2.16 pEq/mm mm in the presence with the report by Bourdeau and Hellstrom-Stein [121 of the of PTH. No firm conclusions were drawn because of the small absence of active transport and only a very small voltagenumber of tubules, but these findings are qualitatively consist- dependent net calcium flux. Explanations offered by Holt and Lechene [11] for the disparate findings include their use of ent with Imai's study [10]. It is becoming increasingly clear that the connecting tubule is immature rabbits and differences in length of the equilibration probably the distal site of enhanced calcium absorption respon- period (at least 180 mm in their study) prior to sampling. sive to PTH that other investigators had suggested from in vivo However, the reversibility of the inhibition of calcium transport studies [39—4 1], and may be responsible for a major fraction of mitigates against artifact of calcium binding to glass or complexthe filtered calcium reabsorbed in the terminal portion of the ing with various anions. Further studies will be needed to resolve the issue of magnitude of calcium flux and mechanism nephron. Collecting tubule. A number of investigators have examined of transport in this segment. Summary. From in vitro studies, the majority of calcium is calcium transport in the cortical collecting tubule [3, 10—13], and there is marked disagreement as to the magnitude of reabsorbed proximally in the convoluted and straight tubules. calcium transport. Shareghi and Stoner [3] first reported studies The absorptive rate ranges from 6.7 to 13 pEq/mm mm. of calcium transport in the light segment (CCT1) and found Recent preliminary studies suggest that transport in the convoinsignificant net transport—0.20 pEq!mm mm; the addition of luted segment is primarily passive although there may be a PTH or dibutyryl cAMP to the bath did not alter the magnitude small component of active transport. Transport in the proximal of transport. To obviate any possible effects of phosphate in the straight tubule may be active. There is no net transport in the perfusate, bicarbonate was substituted, and change in net thin descending or ascending limbs. The thick ascending limbs calcium absorption was not observed. This is an interesting transport calcium on the order of 1.0 pEq/mm mm. Transport finding in that clearance and micropuncture studies have linked in the cortical, and probably the medullary, thick limb is bicarbonaturia with enhanced calcium reabsorption [32, 33], voltage-dependent. The mechanism of transport may be via and an in vitro study by Bourdeau and Burg [5] had already single file pore diffusion. PTH uncovers active transport in the excluded the thick ascending limb as the site of enhanced cortical limb, whereas calcitonin uncovers active transport in the medullary limb. Calcium absorption in the DCT and concalcium transport.
496
Ng
ci
a!
effect of one cation on the other. Whether this is due simply to thick ascending limb; PTH stimulates transport in both seg- increased backflux along a concentration gradient or to some ments. The magnitude of calcium transport in the cortical other mechanism is unknown. Thus, magnesium and calcium collecting tubule (CCT1) is unclear and ranges from insignificant luminal transport appear to be independent of one another in to as great as that in the thick ascending limb. PTH has no effect this segment in contrast to clearance [45] and micropuncture fleeting tubule (DCTg + CCTg) is probably less than that of the
in this segment, and the mechanism of transport remains [46] studies suggesting a competitive transport mechanism, and unclear.
Magnesium transport Although magnesium is often considered with calcium, it has been until recently an infrequently studied divalent cation. The
reason for this lies with the difficulty in obtaining and using isotopic 28Mg which has a half-life of 21.3 hr. Thus micropunc-
ture studies and isolated tubule studies have been limited to centers with electron microprobe capabilities.
Unlike virtually any other ion or solute, magnesium is relatively impermeant in the proximal convoluted tubule, and only 20 to 30% of the filtered magnesium is reabsorbed in this segment of the nephron. The majority of the magnesium—SO to 60%—is reab sorbed in the loop of Henle with an important role ascribed to the thick ascending limb [42, 431. The distal convoluted tubule may reabsorb 2 to 5%, while the remainder of the nephron may reabsorb 1 to 3% of the filtered magnesium [431. Two publications [13, 14] and one preliminary report [441 deal
there appears to be no competitive effect of these two ions at the basolateral membrane [21, 43]. Finally, in the cortical collecting tubule, Shareghi and Agus [13] could not demonstrate any significant net magnesium transport or any effect of PTH to increase this transport. Summary, Permeability to magnesium is low in the proximal straight, and probably convoluted, tubule. The thin descending and ascending limbs have not been studied. The cortical thick ascending limb transports magnesium in a voltage-dependent way similar to calcium. PTH and dibutyryl cAMP uncover active transport in the cortical thick ascending limb, which has a basal absorptive rate of 0.8 to 0.9 pEq/mm mm. The distal convoluted tubule and connecting tubule have not been studied. The cortical collecting tubule has no significant net transport, and there is no effect of PTH to enhance transport. Acknowledgments This work was supported by National Institutes of Health grants AM00770, AM-00424, and AM-2 1394. Drs. Ng and Peraino are recipients of
with magnesium transport in isolated tubules. In the proximal straight tubule Quamme [44] examined the effect of increasing Clinical Investigator Awards. The secretarial assistance of Ms. L. magnesium concentration in the perfusate of superficial and Brazil is gratefully acknowledged. juxtamedullary pars recta. In superficial tubules, raising the Reprint requests to Dr. W. N. Suki, Renal Section, The Methodist perfusate magnesium concentration from 0.7 to 3.1 m elevated Hospital, MS F505, 6565 Fannin, Houston, Texas 77030, USA the net absorptive rate from 0.44 to 0.88 pEq/mm mm. Similarly, injuxtamedullary tubules, the net absorptive rate rose from References
0.92 to 1.20 pEq/mm mm. Elevating the bath magnesium concentration from 0.5 to 1.85 mr'i did not result in secretion of
magnesium in either superficial or juxtamedullary tubules. Thus, in this segment of the nephron, permeability to magnesium is low as compared with sodium and calcium, and there is no active or passive secretion. Magnesium transport appears to be dependent on luminal concentration and is probably passive in view of the collected-to-perfused magnesium concentration ratios which were close to 1.1. In the cortical thick ascending limb Shareghi and Agus [141
found a net reabsorptive rate of 0.64 pEq!mm mm with a lumen-positive transepithelial voltage of +8.8 mY. When the voltage was increased to + 13.6 mV with hypotonic perfusate,
the net reabsorptive rate increased to 0.90 pEq/mm mm. Similarly, when the voltage was changed to —10.4 mY with furosemide and an isosmolar, hyponatric bath, net secretion of —0.90 pEq/mm mm was demonstrated. The correlation between voltage (x) and net transport (v) was y = 0.055x — 0.042, r = 0.69, P < 0.001. Thus, magnesium transport in this segment appears to be voltage-dependent in a fashion similar to calcium. In analogous fashion with calcium, PTH and dibutyryl cAMP enhanced magnesium transport in the cortical thick ascending
limb from 0.94 to 1.78 pEq/mm mm [14]. Doubling luminal magnesium concentration to 1.4 mivi increased magnesium absorption but had no effect on calcium absorption. When luminal calcium concentration was doubled to 2.5 mM, net calcium transport was increased, but magnesium absorption was not affected. When the bath concentration of magnesium or
calcium was doubled, net outward transport of either magnesium or calcium, respectively, was reduced, but there was no
1. BURG M, GRANTHAM J, ABRAMOW M, ORLOFF J: Preparation and
study of fragments of single rabbit nephrons. Am J Physiol 210:1293—1298, 1966
2. ROCHA AS, MAGALDI JB, KoKK0 JP: Calcium and phosphate transport in isolated segments of rabbit Henle's loop. J C/in Invest 59:975—983, 1977
3. SHAREGHI GR, STONER LC: Calcium transport across segments of the rabbit distal nephron in vitro. Am J Physiol 235:F367—F375, 1978 1MM M: Calcium
transport across the rabbit thick ascending limb of Henle's loop perfused in vitro. Pflugers Arch 374:255—263, 1978 5. BOURDEAU JE, BURG MB: Voltage dependence of calcium trans4.
port in the thick ascending limb of Henle's loop. Am J Physiol 236:F357—F364, 1979
6. RoUSE D, NG RCK, SuKI WN: Calcium transport in the pars recta and thin descending limb of Henle of the rabbit, perfused in vitro. J Clin Invest 65:37—42, 1980 7. BOURDEAU JE, BURG MB:
Effect of PTH on calcium transport across the cortical thick ascending limb of Henle's loop. Am Physiol 239:F121—F126, 1980
J
8.
SUKI WN, ROUSED, NG RCK, KOKKO JP: Calcium transport in the
thick ascending limb of Henle. Heterogeneity of function in the medullary and cortical segments. J C/in Invest 66:1004—1009, 1980 9. SUM WN, ROUSE D: Hormonal regulation of calcium transport in thick ascending limb renal tubules. Am J Physiol 24l:F17l—F174, 1981
10. IMAI M: Effects of parathyroid hormone and N6,02-dibutyryl cyclic
AMP on Ca transport across the rabbit distal nephron segments perfused in vitro. P//tigers Arch 390:145—151, 1981 11. HOLT WF, LECHENE C: ADH-PGE2 interactions in cortical collect-
ing tubule. II.
Inhibition of
Ca and P reabsorption. Am J Physiol
241:F461—F467, 1981 12.
BOURDEAU JE, HELLSTROM-STEIN Ri: Voltage-dependent calcium
movement across the cortical collecting duct. Am J Physiol 242:F285—F292, 1982 13. SHAREGIII
GR, AGus ZS: Phosphate transport in the light segment
Divalent cation transport in isolated tubules
of the rabbit cortical collecting tubule. Am
J Physiol 242:F379—
F384, 1982 14. SHAREGHI GR, AGUS ZS: Magnesium transport in the cortical thick
ascending limb of Henle's loop of the rabbit. J
Clin Invest 69:759—
497
isolated perfused pars recta of proximal tubule, Abst Proc VII Int Congr Nephrol, Montreal, 1978, p. E16 30. UssiNo HH: The distinction by means of tracers between active transport and diffusion. The transfer of iodide across the isolated
and other electrolytes from microvolume samples. Am J Physiol
frog skin. Acta Physiol Scand 19:43—56, 1949 31. HODGKIN AL, KEYNES RD:The potassium permeability of a giant nerve fibre. J Physiol (Land) 128:61—88, 1955
242:F202—F206, 1982 16. LAS5ITER WE, GOTTSCHALK CW, MYLLE M: Micropuncture study
32. PE1WN0 RA, SUKI WN: Urine HC03 augments renal CA2 absorption independent of systemic acid-base changes. Am J
769,
1982
15. MUHLERT M, JULITA M, QUAMME G: Disappearance of calcium
of renal tubular reabsorption of calcium in normal rodents. Am J Physiol 204:771—775, 1963
17. MURAYAMA Y, MOREL F, LEGRIMELLEC C: Phosphate, calcium,
Physiol 238:F394—F398, 1980 33. SUTTON RAL, WONG NLM, DIRK JH: Effects of metabolic acidosis and alkalosis on sodium and calcium transport in the dog kidney. Kidney mt 15:520—533, 1979
and magnesium transfer in proximal tubules and 1oops of Henle, as measured by single nephron microperfusion experiments in the rat. Pflugers Arch 333:1—16, 1972 18. ULLRICH KJ, RUMRICH G, KLoss S: Active CA2 reabsorption in the proximal tubule of the rat kidney. Dependence on sodium and buffer transport. Pflugers Arch 364:223—228, 1976 19. BOMSZTYK K, WRIGHT FS: Effects of transepithelial fluid flux on transepithelial voltage and transport of calcium, sodium, chloride
nephron. Pflugers Arch 354:229—239, 1975 35. CHABARDES D, IMBERT-TEBOUL M, MONTEGUT M, CLIQUE A,
and potassium by renal proximal tubule (abstract). Kidney Int
36. IMAI M: The connecting tubule: A functional subdivision of the
21:269, 1982 20. BOMSZTYK K, WRIGHT ES: Contribution of diffusion to transport of calcium by renal proximal tubule (abstract). Clin Res 30:443A, 1982 21. FRIEDMAN PA, FIGUEIREDO JF, MAACK T, WINDHAGER EE:
rabbit distal nephron segments. Kidney Int 15:346—356, 1979 37. JACOBSON HR: Functional segmentation of the mammalian nephron. Am J Physiol 241:F203—F218, 1981 38. MOREL F, CHABARDES D, IMBERT M: Functional segmentation of the rabbit distal tubule by microdetermination of hormone-dependent adenylate cyclase activity. Kidney mt 9:264—277, 1976 39. SUTTON RAL, WONG NLM, DIRKS JH: Effects of parathyroid hormone on sodium and calcium transport in the dog nephron. Clin
Sodium-calcium interactions in the renal proximal convoluted tubule of the rabbit. Am J Physiol 240:F558—F568, 1981 22. QUAMME GA, DIRKS JH: Intraluminal and contraluminal magnesium on magnesium and calcium transfer in the rat nephron. Am J Physiol 238:Fl87—Fl98, 1980 23. ULLRICH KJ, CAPA550 G, RUMRICH G, KLOSS S: Effect of parathy-
roid hormone on the active Ca2-reabsorption in the proximal convolution of the rat kidney (abstract). Pfldgers Arch 359:Rl 18, 1975
24. FRICK A, RUMRICH G, ULLRICH KJ, LA5SITER WE: Microperfu-
sion study of calcium transport in the proximal tubule of the rat kidney. Pfiugers Arch 286:109—117, 1965 25. NG RCK, Suiu WN: Calcium transport in the superficial proximal convoluted tubule of the rabbit (abstract). Clin Res 30:460A, 1982 26. BRENNAN JP, K0KKO JP: Effect of 25-hydroxycholecalciferol on the transport of Ca and Na in the proximal tubule (abstract). Clin Res 21:678, 1973 27. DE ROUFFIGNAC C, MOREL F, Moss N, ROINEL N: Micropuncture
study of water and electrolyte movements along the ioop of Henle in Psammomys with special reference to magnesium, calcium and phosphorus. Pflügers Arch 344:309—326, 1973
28. JAMISON RL, FREY NR, LACY FB: Calcium reabsorption in the thin loop of Henle. Am J Physiol 227:745—751, 1974
29. ALMEIDA AL KUDO LH, ROCHA AS: Calcium transport in
34. CHABARDES D, IMBERT M, MONTEGUT M, MOREL F: PTH sensi-
tive adenyl cyclase activity in different segments of the rabbit MOREL F: Distribution of calcitonin-sensitive adenylate cyclase activity along the rabbit kidney tubule. Proc NatI Acad Sci USA 73:3608—3612, 1976
Sci 51:345—351, 1976
40. AGUS ZS, CHIU PJS, GOLDBERG M: Regulation of urinary calcium excretion in the rat. Am J Physiol 232:F545—F549, 1977 41. HARRIS CA, BURNATOWSKA MA, SEELY JF, SUTTON RAL,
QUAMME GA, DIRKS JH: Effects of parathyroid hormone on
electrolyte transport in the hamster nephron. Am J Physiol 236:F342—F348, 1979 42. BRUNETTE MG, VIGNEAULT N, CARRIERE S: Micropuncture study
of renal magnesium transport in magnesium-loaded rats. Am J Physiol 229:1695—1701, 1975 43. QUAMME GA, DIRKS JH: Magnesium transport in the nephron. Am J Physiol 239:F393—F401, 1980 44. QUAMME GA: Magnesium transport in isolated proximal straight
tubules of the rabbit (abstract). Kidney mt 21:139, 1982
45. MA55RY SG, COBURN JW, CHAPMAN LW, KLEEMAN CR: Effect of
NaC1 infusion on urinary Ca2 and Mg2 during reduction in their filtered loads. Am J Physiol 213:1218—1224, 1967 46. LE GRIMELLEC C, ROINEL N, MOREL F: Simultaneous Mg, Ca, P,
K, Na and Cl analysis in rat tubular fluid. III. During acute Ca plasma loading. Pflugers Arch 346:171—188, 1974