Renal Physiology

Inhibition of the renal betaine transporter by calcium ions

Stephen A. Kempson, Jason M. Edwards, Michael Sturek


Chronic upregulation of the renal betaine/GABA transporter (BGT1) by hypertonic stress has been well documented, but it is not known whether BGT1 can be regulated acutely after insertion in the basolateral plasma membrane. Related transporters, such as the rat brain GABA transporter, can be rapidly removed from the plasma membrane through activation of G protein-coupled receptors. The goal of the present study was to determine whether acute changes in extracellular and/or intracellular Ca2+ will regulate BGT1 transport activity at the plasma membrane level in Madin-Darby canine kidney cells subjected to 24-h hypertonic stress. After brief pretreatment with a Ca2+-free solution, the addition of extracellular Ca2+ in the transport assay produced dose-dependent inhibition of Na+-GABA cotransport. Maximum inhibition was 49% at 2 mM Ca2+ (P < 0.05). Fura 2 imaging confirmed that addition of 2 mM Ca2+ produced a transient increase in intracellular Ca2+ that preceded transport inhibition. Acute inhibition of Na+-GABA cotransport was reproduced by addition of thapsigargin (5 μM) and ionomycin (10 μM). Amino acid transport system A, assayed as a control, was not inhibited. Brief treatment with phorbol esters reproduced the specific inhibition of Na+-GABA cotransport, and the inhibition was blocked by staurosporine. Surface biotinylation confirmed that the response to phorbol esters was accompanied by loss of BGT1 protein from the plasma membrane, and immunohistochemistry showed a shift to an intracellular distribution. We conclude that BGT1 can be inhibited acutely by extracellular Ca2+ through a mechanism involving BGT1 protein internalization, and protein kinase C may play a role.

  • system A
  • Madin-Darby canine kidney cells
  • GABA
  • confocal microscopy
  • biotinylation

during regulation of cell volume in response to osmotic stress, a variety of mechanisms are employed to alter cell solute content so that the cell will gain or lose water (37, 59). Long-term adaptation to chronic hypertonic stress involves accumulation of organic osmolytes that do not perturb normal cell function (7, 10). Betaine is an important osmolyte in the kidney, and the renal betaine/GABA transporter (BGT1) protects the cells of nephron segments in the hypertonic milieu of the inner medulla by mediating cell uptake and accumulation of betaine (22, 65). Prolonged hypertonicity in this region of the kidney is required for the normal operation of the urinary concentrating mechanism. Betaine also may act as a chaperone to stabilize protein structure (52) as part of the cellular response to the complications of osmotic stress (15, 34, 68).

BGT1 transport activity is low in Madin-Darby canine kidney (MDCK) cells in isotonic medium but is strongly induced in response to prolonged hypertonic stress (57, 64). An increase in intracellular ionic strength, following a volume decrease, leads to an increase in abundance and activity of the tonicity-responsive enhancer binding protein transcription factor (45), which migrates to the nucleus and activates BGT1 gene transcription (43, 61). Hypertonic upregulation of BGT1 transport requires more than 18–24 h for a maximal response in cultured kidney cells (31, 57). This delay is likely due to the need for de novo synthesis and intracellular trafficking rather than plasma membrane insertion (33).

BGT1 belongs to the large SLC6 family of transporters for neurotransmitters, amino acids, and osmolytes (13). Recent evidence suggests that some of these transporters, such as the rat brain GABA transporter (GAT1), can be rapidly retrieved from the plasma membrane through activation of G protein-coupled receptors (5). The goal of the present study was to determine whether acute changes in Ca2+, a possible intracellular signal, will regulate BGT1 transport activity at the plasma membrane level.


Cell culture.

MDCK cells (CCL-34, American Type Culture Collection, Rockville, MD) were used between passages 15 and 40 and were grown as monolayers in a 1:1 mixture of DMEM-Ham's F-12 containing 10% bovine calf serum, 15 mM HEPES, 25 mM NaHCO3 (pH 7.4), penicillin (100 IU/ml), and streptomycin (100 μg/ml), as in previous studies (6, 33). Cultures were maintained in an atmosphere of 5% CO2 in air. Cells were grown on glass coverslips for fura 2 fluorescence studies and immunohistochemistry and in 24-well plates for transport measurements. For direct measurement of basolateral transport in some experiments, the cells were grown on collagen-coated Millicell-CM 12-mm filters (Millipore, Bedford, MA) by seeding at confluent density (35–40 × 104 cells/filter). The cells were used 4 days later when tight junctions and cell polarity are established (9, 58). Hypertonic stress was induced using normal growth medium containing NaCl added to achieve a final osmolality of 500 mosmol/kgH2O. Low Ca2+ growth medium was S-MEM modified for suspension culture (GIBCO/Invitrogen) containing 10% bovine calf serum. The serum was previously dialyzed for 24 h against two changes of PBS containing 0.2 mM EGTA. Total Ca2+ content of the complete medium was 0.05 ± 0.01 compared with 1.60 ± 0.06 mM (means ± SD, n = 3) in DMEM/F12 medium, as determined by direct measurement (14). To eliminate any other differences in medium composition for these initial experiments, normal growth medium was made by addition of 1.55 mM CaCl2 to low-Ca2+ growth medium. Cells were washed five times in PBS to remove extracellular Ca2+ before transfer to low-Ca2+ medium for 30 min. Standard DMEM/F12 medium was used for cell culture for the subsequent studies on the effects of Ca2+ addition.

Transport measurements.

The osmolality of all solutions used for transport was matched to the osmolality of the growth medium by addition of sucrose where necessary. BGT1 transports GABA, for which it has a higher affinity than betaine (1, 42). Therefore, the transport function of endogenous BGT1 in cell monolayers was determined as cell uptake of [3H]GABA during a 10- to 30-min incubation, as described previously in detail (31). Briefly, [3H]GABA uptake was determined in both medium containing Na+ and medium in which Na+ was replaced by N-methyl-d-glucamine-HCl. The difference is the Na+-dependent component that represents transport specifically via BGT1. Standard Na+ medium consisted of (in mM) 137 NaCl, 5.4 KCl, 2.0 CaCl2, 1.2 MgSO4, and 10 HEPES, pH 7.4. Low-Ca2+ uptake medium was similar except that CaCl2 was replaced by MgCl2 to keep divalent cations constant, and 0.05 mM EGTA was added as a Ca2+ chelator. This solution was used solely to determine uptake in cells pretreated with low-Ca2+ growth medium. In complementary experiments on the effects of adding back Ca2+ on BGT1 transport, the cells were first washed five times in Ca2+-free solution [(in mM) 137 NaCl, 5.4 KCl, 1.2 MgSO4, 0.05 EGTA, and 10 HEPES, pH 7.4] before being preincubated for 30 min at 37°C in the same solution to which CaCl2 was added at the concentration to be tested. An identical solution, containing CaCl2 and radioactive solute, was used for the transport assay. The small dissociation constant (1 × 10−7 M) of the Ca2+-EGTA complex (56) ensures there will be negligible free EGTA when Ca2+ is added back to the Ca2+-free solution. The presence of Mg2+ will produce a negligible increase in free Ca2+ because it binds to EGTA with a much lower affinity than Ca2+ (30).

Basolateral transport measurements were performed on filter-grown cells by the method described previously (64). Less than 1% of the [3H]GABA added to the basolateral compartment was recovered in the apical compartment, consistent with an intact confluent monolayer (64). Identical protocols were used for to determine cell uptake of α-[14C]methylaminoisobutyric acid (MeAIB), a specific substrate for amino acid transport system A in the basolateral membrane (12, 28, 38). All tested compounds (ionophores, phorbols, etc.) were used from stock solutions in DMSO that were diluted 1:1,000 after addition to experimental solutions. When cells were pretreated with one of these compounds, the same compound was included during the transport assay. An equivalent volume of DMSO was added to the controls.

Intracellular Ca2+ measurements.

Changes in intracellular Ca2+ concentration ([Ca]i) were measured with the Ca2+-sensitive probe fura 2, which was loaded into the cells as the acetoxymethyl ester (Invitrogen/Molecular Probes) (29, 54). After 24 h in hypertonic growth medium, MDCK cells on coverglasses were loaded with 2.5 μM fura 2-AM in hypertonic solution containing sucrose and (in mM) 118 NaCl, 5 KCl, 2 CaCl2, 1 MgCl2, 25 NaHCO3, 10 glucose, and 20 HEPES (pH 7.4) at 37°C. After 30 min, the cells were rinsed in fresh hypertonic solution and placed in the perfusion chamber on the stage of an epifluorescence microscope. Cells were studied at room temperature using the InCa Ca2+ imaging system (Intracellular Imaging, Cincinnati, OH) described previously (24, 26). Fura 2 was excited by 340- and 380-nm light, and emitted fluorescence at 510 nm was collected by a CCD camera attached to a computer for data acquisition by the InCa ratiometric fluorescence program. The initial perfusion was for 60 s with the Ca2+-free solution to obtain a baseline before perfusion of the test solution for 3–6 min. For each experiment the maximum fluorescence signal of at least 20 cells was averaged for analysis.

Cell fractionation and immunostaining.

MDCK cells collected from four flasks (75 cm2) were disrupted by nitrogen cavitation, and a cell membrane fraction was recovered by differential centrifugation. This fraction was used to test possible direct effects of Ca2+ on BGT1 by measurement of Na+-GABA cotransport in the presence and absence of Ca2+ using the rapid filtration procedure. All methods have been described previously in detail (4, 31). Cells on glass coverslips were fixed in methanol and processed for antibody staining and confocal microscopy as described previously (6, 33). Primary antibodies to BGT1 and E-cadherin were used at dilutions of 1:100 and detected with goat anti-rabbit IgG conjugated to FITC (Jackson ImmunoResearch, West Grove, PA) diluted 1:100. Cells were washed and counterstained for 10 min in propidium iodide (2 μg/ml) to visualize nuclei (36). For visualization of ZO1 and occludin, cells were fixed in 4% paraformaldehyde, permeabilized in 0.2% Triton X-100, and primary antibodies were used at 1:100 dilution, as above.

Surface biotinylation, gel electrophoresis, and Western blotting procedures were performed as described elsewhere in detail (6, 33). Data are expressed as means ± SD of at least three experiments. In each transport experiment, the mean value (n = 1) was derived from triplicate measurements. Where appropriate, different groups were compared by Student's t-test or by analysis of variance and Tukey's test for multiple comparisons using Instat v.3.06 software (GraphPad, San Diego, CA). A probability of P < 0.05 was considered statistically significant.


In MDCK cells grown on a plastic substrate, omission of Ca2+ from the uptake solution (low Ca2+) produced almost a doubling of BGT1 transport activity in both isotonic control cells and in cells previously adapted to hypertonic growth medium (500 mosmol/kgH2O) for 24 h (Fig. 1). Only the Na+-dependent transport was increased in both groups, and there was no change in GABA uptake determined when Na+ in the uptake solution was replaced by N-methyl-d-glucamine. This suggests that the stimulation of GABA uptake was not due simply to a change in membrane permeability. The effect of Ca2+ omission was relatively specific because Na+-dependent transport of MeAIB, a substrate for amino acid transport system A, remained unchanged when determined in the presence and absence of extracellular Ca2+. Na+-MeAIB cotransport, determined after 6-h hypertonic stress, was 1,265 ± 268 in normal uptake medium compared with 1,334 ± 241 pmol·mg−1·20 min−1 (means ± SD, n = 3) in low-Ca2+ solution. System A in MDCK cells, like BGT1, is activated by hypertonic stress and resides primarily in the basolateral plasma membrane (28, 31, 38). When basolateral transport was determined directly, using filter-grown polarized MDCK cells (64) adapted to the same hypertonic stress, the effect of acute Ca2+ omission was similar. Na+-GABA cotransport determined in low-Ca2+ uptake solution was increased to 2,606 ± 390 compared with 921 ± 234 pmol·mg−1·20 min−1 (means ± SD) in normal uptake solution (P < 0.02, n = 3). In contrast, Na+/MeAIB transport was not significantly different (not shown). Because the specific effect of low-Ca2+ uptake medium was independent of establishment of cell polarity, plastic-grown cells were used for subsequent experiments.

Fig. 1.

Acute stimulation of Na+-dependent GABA transport in Madin-Darby canine kidney (MDCK) cells by low-Ca2+ medium. The Na+-independent component, determined when Na+ in the uptake medium was replaced by N-methyl-d-glucamine (MG), remained unchanged. The cells, grown in 24-well plates, were pretreated for 30 min in low-Ca2+ growth medium (Low Ca), and transport was determined by measuring GABA uptake at 10 min in Ca2+-free solution as described. Controls (Norm Ca) were maintained in normal growth and uptake media throughout. Cells were previously adapted for 24 h to either hypertonic (HYP) or normal isotonic (ISO) growth medium. Values are means ± SD from 3 experiments.

A complementary study was undertaken in which extracellular Ca2+ was added back during measurement of GABA uptake in cells previously washed in essentially Ca2+-free solution, one containing 0.05 mM EGTA and no added Ca2+. The results revealed a dose-dependent inhibition of Na+-dependent GABA uptake by extracellular Ca2+ in the concentration range 0.1–10.0 mM. At 2.0 mM Ca2+, the inhibition was 49%, which was almost maximum (Fig. 2). Ca2+ addition did not significantly change system A transport, except at 5 mM, where it tended to increase transport activity (Fig. 3).

Fig. 2.

Dose-dependent inhibition of Na+-GABA cotransport by extracellular Ca2+. Following adaptation to hypertonic growth medium for 24 h, the cells were pretreated 30 min in hypertonic Ca2+-free solution ± added Ca2+ before measurement of Na+-GABA cotransport in the same solution (10-min uptakes) as described. Values are means ± SD from 5 experiments. *Significantly different (P < 0.05) from transport at 0 Ca2+, which was 2,706 ± 1,057 pmol·mg−1·10 min−1.

Fig. 3.

Acute increases in extracellular Ca2+ did not inhibit Na+-MeAIB cotransport (system A). Cells were previously adapted to hypertonic growth medium for either 6 (Na+-MeAIB) or 24 h (Na+-GABA) (31) before determination of the effects of extracellular Ca2+, as described in the legend to Fig. 2. Values are means ± SD from 3 experiments. *Significantly different (P < 0.05) compared with transport at 0 Ca2+, which was 2,396 ± 1,157 pmol·mg−1·10 min−1 (Na+-GABA) and 1,807 ± 366 pmol·mg−1·20 min−1 (Na+-MeAIB).

The possibility that Ca2+ acts as a direct inhibitor of Na+-GABA cotransport was tested using an isolated membrane fraction containing plasma membranes (4, 31). Membranes were preincubated for 5 min with 2 mM Ca2+ before measurement of transport, also in the presence of 2 mM Ca2+. GABA transport (as 1-min uptakes) in Na+ medium was 34 ± 11 pmol·mg−1·min−1 in control membranes, not significantly different compared with 37 ± 10 pmol·mg−1·min−1 (means ± SD, n = 3) in membranes treated with Ca2+. Na+-independent GABA transport, determined when Na+ in the uptake solution was replaced by K+, was 4 ± 4 pmol·mg−1·min−1 in controls and also was not altered by Ca2+. These findings rule out a direct effect of Ca2+ on BGT1 transport function.

Under isotonic conditions, as expected (11, 44), application of 2.0 mM extracellular Ca2+ to MDCK cells preperfused in Ca2+-free solution produced a rapid and transient increase in the fura-2 fluorescence ratio, indicating an increase in intracellular free Ca2+. The increase reached a peak within 100 s after Ca2+ addition (Fig. 4). Similar results were obtained in cells adapted to hypertonic medium for 24 h using hypertonic perfusates. Examination of the time course of the acute action of extracellular Ca2+ at the same concentration on Na+-GABA cotransport showed that the onset of inhibition was not significant until 30 min after Ca2+ addition (Fig. 5). Thus the increase in intracellular free Ca2+ precedes the inhibition of BGT1 transport activity.

Fig. 4.

Addition of 2 mM extracellular Ca2+ to MDCK cells produces a prompt increase in fura 2 fluorescence ratio, consistent with increased intracellular Ca2+. Cells were grown on coverglasses, loaded with fura 2-AM, and perfused with Ca2+-free solution at the start of the experiment. Isotonic solutions were used throughout. Values are means of 24 cells imaged simultaneously.

Fig. 5.

Time course of acute inhibition of Na+-GABA cotransport by 2 mM extracellular Ca2+. Cells were previously adapted to hypertonic growth medium for 24 h. Values are means ± SD from 4 experiments. *Significantly different (P < 0.05) compared with uptake in controls (Ca2+-free medium) at the same time point.

When intracellular free Ca2+ was increased pharmacologically by use of the Ca2+-ATPase inhibitor thapsigargin or Ca2+ ionophores, there was significant inhibition of Na+-GABA cotransport (Fig. 6A) but no inhibition of Na+-MeAIB cotransport (Fig. 6B). Direct measurement of intracellular free Ca2+ confirmed that prompt increases occurred in response to both thapsigargin and ionomycin (Fig. 7). The increase due to thapsigargin was comparable to that produced when cells were perfused with hypertonic solution containing 2 mM Ca2+. In contrast to previous reports (46, 47), arginine vasopressin did not change intracellular free Ca2+ in MDCK cells (Fig. 7). However, consistent with the unchanged intracellular Ca2+, there was no effect of vasopressin on Na+-GABA cotransport (Fig. 6A).

Fig. 6.

A: inhibition of Na+-GABA cotransport by thapsigargin (TG), ionomycin (IONO), and A-23187 (A23) but not vasopressin (VP) in cells adapted to hypertonic growth medium for 24 h. B: same treatment produced no inhibition of Na+-MeAIB cotransport. The tested compounds were added during a 30-min pretreatment and also during the 30-min uptake period. Normal hypertonic solutions containing Ca2+ were used throughout. Values are means ± SD from 3 experiments. *Significantly different (P < 0.05) from control group (C).

Fig. 7.

Application of 5 μM TG or 10 μM IONO produces prompt and transient increases in intracellular Ca2+ in MDCK cells, but 10−7 M arginine vasopressin (AVP) did not. Extracellular Ca2+ was 2 mM. Peak responses to TG and IONO were significantly different (P < 0.0001, paired t-test, n = 20–26 cells) compared with baseline.

To investigate the possible intracellular steps subsequent to the increase in intracellular Ca2+, we next tested the effects of protein kinase C activators. The cell-permeable diacylglycerol analog, 1,2-dioctanoyl-sn-glycerol (DOG) produced dose-dependent inhibition of Na+-GABA cotransport over the concentration range 0.1–10.0 μM. Inhibition was ∼70% at 10 μM (Fig. 8). In contrast, Na+-MeAIB cotransport was much less sensitive to DOG. There was 25% inhibition at 1 μM DOG, but higher concentrations of DOG did not increase the inhibition further (Fig. 8). The β-phorbol ester PMA was a more potent inhibitor than DOG. Na+-GABA cotransport was inhibited by 53% at 50 nM (Fig. 9). A biologically inactive α-analog of PMA, 4α-phorbol 12,13-didecanoate (53, 55), had no effect at the same concentration. Staurosporine (10 μM), a broad spectrum protein kinase C inhibitor, had no significant effect alone but completely blocked the action of PMA (Fig. 9).

Fig. 8.

Treatment with 1,2-dioctanoyl-sn-glycerol (DOG) produced dose-dependent inhibition of Na+-GABA cotransport (renal betaine-GABA cotransporter; BGT1) but not Na+-MeAIB cotransport (system A). After 24-h hypertonic stress, MDCK cells were preincubated with DOG for 30 min before measurement of transport (20-min uptake) in the presence of DOG. Values are means ± SD from 3 experiments.

Fig. 9.

Inhibitory action of DOG was reproduced by 50 nM PMA. The inactive analog 4α-phorbol 12,13-didecanoate (PDD; 50 nM) had no effect. The action of PMA was blocked by 10 μM staurosporine (STAU). Other details are as in Fig. 8. Values are means ± SD from 6 experiments. *P < 0.001 compared with control group (C).

Pretreatment of MDCK cells for 15 min with either 200 μM IBMX or 50 μM forskolin, followed by a 20-min uptake measurement in the presence of either at the same concentration, led to inhibition of Na+-GABA cotransport by only 13 and 11% (n = 3), respectively, compared with controls. These effects of IBMX and forskolin were not statistically significant and were not additive when cells were treated with both. Increasing concentrations to 400 μM IBMX and 100 μM forskolin produced no additional inhibition. These concentrations are within the range shown previously to be effective in regulating cAMP-mediated pathways and control of membrane transport in MDCK cells (40, 63, 67) and frog oocytes (55).

The inhibition of several different membrane transport proteins in response to protein kinase C activation was reported to involve retrieval of the transport proteins from the plasma membrane. Specific examples include the renal type IIa Na+-phosphate cotransporter (18), the renal Na+-independent organic ion transporter (60), the neuronal Na+-glutamate cotransporter (55), and the brain GAT1 (5). A combination of immunohistochemistry and surface biotinylation was used to determine the intracellular location of BGT1 protein in MDCK cells after 40-min treatment with PMA followed by methanol fixation (33). In control cells adapted to hypertonic growth medium, the location of BGT1 protein, as expected, was primarily in the plasma membrane (Fig. 10, top row). However, after PMA treatment there was a marked shift of BGT1 into the cytoplasm. The fluorescent signal was punctate, consistent with entrapment in intracellular vesicular structures such as endosomes. Staurosporine alone had no effect on BGT1 distribution (not shown). However, when staurosporine was added at the same time as PMA more BGT1 was retained in the plasma membrane (Fig. 10, top row). In contrast, the plasma membrane distribution of E-cadherin remained unchanged by similar treatment with PMA in both the absence and presence of staurosporine (Fig. 10, bottom row). PMA treatment also had no effect on the distribution of ZO-1 and occludin, which remained exclusively in the plasma membrane (not shown).

Fig. 10.

Internalization of BGT1 protein was induced by PMA and partially blocked by STAU. After 24-h adaptation to hypertonic medium (Hyp), the cells were treated for 40 min with vehicle (control), 50 nM PMA, or 50 nM PMA plus 10 μM STAU. After methanol fixation, the cells were processed for staining with antibodies to BGT1 (top row, green), E-cadherin (CAD; bottom row, green), and propidium iodide (red). Distribution of CAD remained unchanged throughout.

Surface biotinylation confirmed the loss of BGT1 protein from the plasma membrane during 40-min treatment with PMA (Fig. 11). BGT1 abundance was estimated by densitometry and expressed relative to surface E-cadherin abundance, which did not change (Fig. 10), and was used as a loading control (33). The BGT1/cadherin ratio decreased from 1.95 in controls to 0.25 in PMA-treated cells.

Fig. 11.

Surface biotinylation of intact cells confirmed loss of plasma membrane BGT1 following 40-min treatment with PMA. Other details are as in Fig. 10. Surface abundance of CAD remained unchanged.

Finally, to test whether the inhibitory action of extracellular Ca2+ on Na+-GABA cotransport (Fig. 2) might be mediated in part via protein kinase C activation, cells were pretreated with staurosporine (10 μM) in Ca2+-free medium before addition of extracellular Ca2+ (2 mM). The inhibitory action of Ca2+ was significantly reduced by staurosporine (Fig. 12). In contrast, similar use of other inhibitors of protein kinase C (calphostin C, chelerythrine chloride, and Ro-31–8220) failed to block the action of extracellular Ca2+ (Fig. 12).

Fig. 12.

Inhibitory action of extracellular Ca2+ (2 mM) on Na+-GABA cotransport was blocked by pretreatment with STAU (10 μM) but not by calphostin C (CAL; 1 μM), chelerythrine chloride (CHE; 1 μM), or Ro-31–8220 (Ro; 1 μM). Values are means ± SD from 4–6 experiments. *Significantly different (P < 0.01) compared with both control (no additions) and STAU-pretreated groups.


The inhibitory effect of increased extracellular Ca2+ was specific for BGT1 (Fig. 3) and may be mediated by the increase in intracellular Ca2+ because this event occurred before any change in BGT1 transport (Figs. 4 and 5). The increase in intracellular Ca2+ appears to be a key step because when it was increased pharmacologically by thapsigargin and ionophores (Fig. 7), the specific inhibition of BGT1 was reproduced (Fig. 6). One consequence of the increase in intracellular Ca2+ may be the activation of signaling pathways involving protein kinase C because phorbol esters produced dose-dependent inhibition of BGT1 but not system A (Fig. 8). The mechanism cannot be attributed to a generalized disruption of the Na+ gradient because the Na+ gradient-dependent transport of MeAIB by system A was not changed by extracellular Ca2+ or PMA. It is more likely that PMA reduced the plasma membrane abundance of BGT1 protein (Fig. 11) by endocytic internalization (Fig. 10). The actions of PMA on BGT1 transport and internalization were both blocked, at least in part, by staurosporine, which supports a role for protein kinase C. PMA was also shown to deplete ERK activity in MDCK cells (35), but this effect required treatment for 20 h with 150 nM PMA. This suggests that the rapid action of 50 nM PMA on BGT1 is unlikely to be mediated by inhibition of ERK. Use of IBMX and forskolin to activate protein kinase A produced no significant change in BGT1 transport, suggesting that BGT1 could be regulated primarily by protein kinase C-mediated signal pathways. Whether a decrease in extracellular Ca2+ stimulates rapid membrane insertion of BGT was not studied, but such a process could help explain the activation of BGT1 transport observed during acute exposure to low Ca2+ medium (Fig. 1).

The potential role for protein kinase C in the intracellular mechanism for rapid internalization of BGT1 should be regarded with caution. First, it should be noted that not all actions of phorbol esters are mediated by protein kinase C. For example, the stimulation of neurotransmitter secretion by phorbol esters may be due to a direct action of the phorbol ester on Munc13, a SNARE protein required for priming the process of vesicle fusion with the plasma membrane (53). Second, the action of staurosporine to block the inhibitory action of extracellular Ca2+ on BGT1 transport was not reproduced by other protein kinase C inhibitors (Fig. 12). The lack of absolute specificity of most of the available kinase C inhibitors (see Refs. 24a, 39, and 66, for example) further hampers accurate interpretation of these apparently conflicting data. Additional studies using complementary techniques, such as siRNA or use of dominant negative isoforms of specific kinases, will be required to resolve the intracellular mechanisms suggested by these initial pharmacological studies with activators and inhibitors of kinase C.

The effects of PMA and DOG on BGT1 activity in MDCK cells confirm and extend the results of previous reports (41, 48). The same authors also reported that protein kinase A activation led to inhibition of BGT1 transport, a finding that was not reproduced in the present study and additional work will be required to resolve this discrepancy. The rapid response to PMA and DOG is in marked contrast to the relatively slow (at least 24 h) downregulation of BGT1 during release from hypertonic stress (4, 32, 57) and suggests that different regulatory steps are involved. Furthermore, the rapid downregulation was observed in cells adapted to hypertonic medium, indicating that it can override the chronic upregulation by hypertonicity. It is also intriguing that BGT1 internalization can still occur in these cells because it is well known that a hypertonic medium inhibits endocytosis in various cell types, including MDCK (19), probably by disrupting the formation of clathrin-coated vesicles (23, 25). Whatever the specific mechanism, this suggests that a similar process could occur within the hypertonic medulla in vivo. These findings are consistent with other reports of the rapid regulation of plasma membrane transporters by PMA and DOG [for example, brain GAT1 (5) and EAAC1 (55) and renal type IIa Na+-phosphate cotransporter (18), fNaDC-3 (21), and hOAT1 (60)]. Because BGT1 has several consensus sequences for phosphorylation (48, 65), the direct phosphorylation of BGT1 protein may inhibit the transport function, increase the normal turnover rate at the plasma membrane, specifically target the protein for endocytic retrieval, or some combination of these possibilities. However, based on findings from other membrane transport systems, the mechanism may be more complex. For example, site-directed mutagenesis of putative phosphorylation sites in fNaDC-3 and hOAT1 failed to block downregulation by PMA (21, 60). Similarly, activation of Ca2+-sensitive K+ channels by PMA was not affected by mutation of phosphorylation sites (62). It was concluded that the phosphorylation step could occur at some unknown, accessible site on the transporter protein, or it could involve a specific regulatory protein involved in plasma membrane insertion/retrieval of the transporter. In the case of BGT1, it has been suggested that the observed phosphorylation of BGT1 induced by a phorbol ester could lead to disruption of the interaction of BGT1 with Lin7, a PDZ-domain-containing protein that normally prevents internalization of BGT1 (41). The authors of this report, performed with MDCK cells overexpressing BGT1, also demonstrated that acute treatment with a phorbol ester caused BGT1 internalization under isotonic conditions and that this action was blocked in cells treated acutely with hypertonic medium to block endocytosis (41). Perhaps during chronic hypertonicity, as used in the present study, the cells are able to internalize BGT1 because they can eventually overcome the constraints on endocytosis. This could be part of the adaptive mechanism essential for cell survival during hypertonic stress.

The response to changes in extracellular Ca2+ could be mediated by the calcium-sensing receptor, a G protein-coupled receptor that is present in several tissues including kidney (20, 50, 51) and MDCK cells (2, 3). Numerous agonists activate intracellular signal pathways through this receptor, which has been linked to regulation of secretion, proliferation, differentiation, and ion channel activity in various cell types (27). This is consistent with a role in short-term events mediated by second messengers. An elevation in extracellular Ca2+ produces intracellular Ca2+ signals through interactions between the receptor and phospholipase C (27). This could explain the prompt increase in intracellular Ca2+ in response to the addition of 2 mM extracellular Ca2+ (Fig. 4).

The physiological relevance of acute downregulation of BGT1 by Ca2+, and possibly protein kinase C, must remain speculative. During normal function, it could provide a mechanism for rapid fine-tuning of BGT1 transport during fluctuations in medullary hypertonicity. For example, high urinary Ca2+ levels during hypercalcemia are likely detected by the calcium-sensing receptor in the collecting duct, which then blocks water reabsorption by inhibiting cAMP production (16) and interfering with water channel trafficking (49, 51). In this way the reabsorption of mineral ions and water is coordinated to avoid kidney stone formation (8, 16). When water reabsorption is reduced, the hypertonic milieu in the medulla may dissipate, especially if NaCl reabsorption is also reduced (17, 49, 51). Under these conditions, continued betaine uptake would be unnecessary and downregulation could be initiated by extracellular Ca2+. The calcium-sensing receptor also is strongly stimulated by aminoglycosides (27), which could lead to internalization of BGT1 in the medullary sections of the nephron. This would impair the hypertonic adaptation and may contribute to the nephrotoxic effects of these antibiotics.


This work has been published in part in abstract form (J Am Soc Nephrol 16: 571A, 2005) and supported by a grant-in-aid from the American Heart Association Midwest Affiliate (S. A. Kempson) and National Heart, Lung, and Blood Institute Grant HL-062552 (M. Sturek).


We thank Dr. Gerhard Burckhardt (Georg-August University, Gottingen, Germany) for helpful suggestions and Kimberly Wilson for technical assistance.


  • The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.


  1. 1.
  2. 2.
  3. 3.
  4. 4.
  5. 5.
  6. 6.
  7. 7.
  8. 8.
  9. 9.
  10. 10.
  11. 11.
  12. 12.
  13. 13.
  14. 14.
  15. 15.
  16. 16.
  17. 17.
  18. 18.
  19. 19.
  20. 20.
  21. 21.
  22. 22.
  23. 23.
  24. 24.
  25. 24a.
  26. 25.
  27. 26.
  28. 27.
  29. 28.
  30. 29.
  31. 30.
  32. 31.
  33. 32.
  34. 33.
  35. 34.
  36. 35.
  37. 36.
  38. 37.
  39. 38.
  40. 39.
  41. 40.
  42. 41.
  43. 42.
  44. 43.
  45. 44.
  46. 45.
  47. 46.
  48. 47.
  49. 48.
  50. 49.
  51. 50.
  52. 51.
  53. 52.
  54. 53.
  55. 54.
  56. 55.
  57. 56.
  58. 57.
  59. 58.
  60. 59.
  61. 60.
  62. 61.
  63. 62.
  64. 63.
  65. 64.
  66. 65.
  67. 66.
  68. 67.
  69. 68.
View Abstract