Activation of P2Y₁ and P2Y₂ receptors induces chloride secretion via calcium-activated chloride channels in kidney inner medullary collecting duct cells

Abstract

Dysregulation of urinary sodium chloride (NaCl) excretion can result in extracellular fluid (ECF) volume expansion and hypertension. Recent studies demonstrated that urinary nucleotide excretion increases in mice ingesting a high-salt diet and that these increases in extracellular nucleotides can signal through P2Y₂ receptors in the kidney collecting duct to inhibit epithelial Na⁺ channels (ENaC). However, under conditions of ECF volume expansion brought about by high-dietary salt intake, ENaC activity should already be suppressed. We hypothesized that alternative pathways exist by which extracellular nucleotides control renal NaCl excretion. We used an inner medullary collecting duct (mIMCD-K2) cell line in an Ussing chamber system as a model to study additional ion transport pathways that are regulated by extracellular nucleotides. When ENaC was inhibited, the addition of adenosine triphosphate (ATP) to the basal side of cell sheets activated both P2Y₁ and P2Y₂ receptors, inducing a transient increase in short-circuit current (I_sc); addition of ATP to the apical side activated only P2Y₂ receptors, inducing first a transient and then a sustained increase in I_sc. The ATP-induced increases in I_sc were blocked by pretreatment with a phospholipase C (PLC) inhibitor, a calcium (Ca²⁺) chelator, or Ca²⁺-activated Cl⁻ channel (CACC) inhibitors, suggesting that ATP signals through both PLC and intracellular Ca²⁺ to activate CACC. We propose that P2Y₁ and P2Y₂ receptors operate in tandem in IMCD cells to provide an adaptive mechanism for enhancing urinary NaCl excretion in the setting of high-dietary NaCl intake.

defects in the regulation of urinary sodium chloride (NaCl) excretion can lead to extracellular fluid (ECF) volume expansion and high blood pressure. The collecting duct (CD) of the kidney plays a critical role in fine-tuning NaCl balance and is subject to extensive hormonal regulation. For example, under conditions of ECF volume depletion, the mineralocorticoid hormone aldosterone can increase ECF volume by activating the epithelial Na⁺ channel (ENaC), which is the principal ion channel responsible for Na⁺ absorption in the kidney CD. While the regulation of ENaC-mediated Na⁺ transport is thought to be a major determinant of ECF volume and blood pressure, Cl⁻ transport in the kidney CD may also contribute to the regulation of NaCl balance by a variety of pathways whose hormonal regulatory mechanisms are not well-established.

Extracellular nucleotides such as adenosine triphosphate (ATP) are becoming recognized as autocrine or paracrine hormones that can be released by renal epithelial cells in response to various environmental cues (9, 33, 36). Extracellular ATP activates plasma membrane-localized purinergic receptors belonging to one of two classes, the P2X ion channels and the P2Y guanine nucleotide-binding protein (G protein)-coupled receptors, either of which regulate a diverse set of physiological processes (9). Recent studies suggested that urinary nucleotides and the P2Y₂ receptor system contribute to the regulation of NaCl homeostasis and blood pressure. High-NaCl feeding in mice stimulates an increase in urinary excretion of nucleotides, including UTP, ATP, and their hydrolytic products, which coincides with decreased ENaC activity in the cortical CD (26). ATP in the tubular fluid of the cortical CD activates P2Y₂ receptors, which decrease the open probability and net activity of ENaC (25, 30). Studies of P2Y₂ knockout mice further confirm that P2Y₂ receptors participate in maintaining salt balance and blood pressure. P2Y₂ knockout mice exhibit renal NaCl retention and hypertension (25, 30); moreover, these mice develop a greater level of salt-sensitive hypertension than do wild-type mice when exogenous mineralocorticoids are infused to maintain ENaC activity at high levels (26).

However, when wild-type mice (21) or humans (17, 23, 31) ingest a high-NaCl diet, ECF volume expands, leading to low levels of serum aldosterone and, as a consequence, diminished ENaC activity in the kidney CD. Thus, the observed increase in urinary nucleotide excretion in mice fed a high-NaCl diet suggests that ATP-mediated downregulation of ENaC may not be important when aldosterone levels are suppressed. We hypothesized that alternative pathways exist by which ATP induces urinary NaCl excretion in the inner medullary collecting duct (IMCD), a nephron segment that is known to secrete Cl⁻ under conditions of high-NaCl intake (14, 35).

The purpose of our study was to characterize the specific P2Y receptor signaling pathways that control NaCl transport in the IMCD under conditions where ENaC activity is diminished. We used the mIMCD-K2 cell line as a model system to study the direct effects of nucleotides on NaCl transport across IMCD epithelia. The mIMCD-K2 cell line was originally derived from the initial segment of the IMCD and retains characteristic features of the IMCD, including the signaling and ion transport machinery that support ATP-stimulated Cl⁻ secretion (22). Although previous studies demonstrated a role for ATP in either increasing intracellular calcium (Ca²⁺) concentration (4, 41) or stimulating Cl⁻ secretion in IMCD cells (5, 22), the identity of the specific P2Y receptor subtypes involved in the integrated response to extracellular ATP has not been determined. Here, we demonstrate that ATP activates and signals through P2Y₁ and P2Y₂ receptors on the basal side, and P2Y₂ receptors on the apical side of mIMCD-K2 cell sheets, to increase Cl⁻ secretion through calcium-activated Cl⁻ channels (CACC). ATP-stimulated transcellular Cl⁻ secretion generates an electromotive force that also increases paracellular Na⁺ secretion, leading to net NaCl secretion. We propose that these two purinergic signaling pathways could represent another mechanism by which urinary nucleotides enhance urinary NaCl excretion under conditions of high-dietary NaCl intake.

METHODS

Cell culture.

The mIMCD-K2 cell line was kindly provided by Dr. Bruce Stanton (Dartmouth Medial School, Hanover, NH). Cells of passages 38–46 were expanded and plated on polycarbonate Snapwell inserts (Corning Costar), as described previously (28). Transepithelial voltage (V_te) and resistance (R_te) were measured with an EVOM “chopstick” voltmeter (World Precision Instruments, Sarasota, FL). Cells were in defined medium until R_te reached resistance values between 800 and 1,200 Ω·cm² within 10–14 days.

Ussing chamber measurements.

Cell sheets were mounted between the Lucite half chambers of the Ussing chamber apparatus (Physiological Instruments, San Diego, CA) for electrophysiological studies, as described previously (27, 28). Cell sheets were bathed in Krebs-Henseleit solution (in mM: 140 NaCl, 25 NaHCO₃, 5 KCl, 5 glucose, 2 CaCl₂, and 1 MgCl₂) and gassed with a mixture of 95% O₂-5% CO₂. V_te across cell sheets was clamped to 0 mV, and a set voltage pulse of 1 mV was applied across cell sheets for 200 ms every 20 s. The short-circuit current (I_sc) and R_te across cell sheets were continuously recorded using Acquire and Analyze Software (Physiological Instruments).

Ion substitution experiments.

To establish the presence of Na⁺ secretion across cell sheets, Na⁺ substitution experiments were also performed, in which NaCl was replaced with N-methyl-d-glucamine chloride (NMDG-Cl) in the Krebs-Henseleit solution of the apical bath, while the concentrations of all other ions were kept constant. In these experiments, the basal bath solution contained 140 mM NaCl and 25 mM NaHCO₃, whereas the apical bath solution contained 25 mM NaHCO₃ and 140 mM NMDG-Cl, which created a Na⁺ concentration gradient in which the basal Na⁺ concentration was 6.6-fold higher than that in the apical bath. All I_sc measurements were made after correcting for junction potentials created by the asymmetric bath solutions.

Pharmacological agents.

Cell sheets were exposed to a series of pharmacological agents once I_sc and R_te stabilized in the Ussing chamber. The apical side of all cell sheets was treated with 10⁻⁵ M amiloride (Sigma, St. Louis, MO), an ENaC inhibitor. ATP (10⁻⁵ M) was then added sequentially to the apical and basal sides of cell sheets. Some cell sheets were exposed to the P2Y₁ receptor agonist MRS2365 (10⁻⁶ M) or the P2Y₂ receptor agonist thio-UTP (10⁻⁶ M). Other cell sheets were pretreated with the P2Y₁ receptor antagonist MRS2500 (10⁻⁶ M) before addition of P2Y₁ and P2Y₂ receptor agonists. To rule out involvement of adenosine receptors in ATP-induced I_sc (I_sc^ATP), the A2b receptor antagonist PSB603 (10⁻⁷ M) and the broad spectrum adenosine receptor antagonist DPCPX (10⁻⁴ M) were added to both sides of cell sheets, as described previously (28). All agonists and antagonists were obtained from Tocris Bioscience (Ellisville, MO).

To identify the classes of ion channels or transporters that respond to P2Y receptor activation, several small molecule inhibitors were used. CFTR inhibitor-172 (10⁻⁵ M, Sigma) was used to block cystic fibrosis transmembrane conductance regulator (CFTR) (20). Flufenamic acid (FFA; 10⁻³ M, Sigma) and niflumic acid (2 × 10⁻⁴ M, Sigma) were used to block CACC (13, 40) in mIMCD-K2 cells. FFA induced a small change in baseline I_sc, which was attributed to a change in the pH of the bath solution. Diphenylamine-2-carboxylate (DPAC; 10⁻⁴ M, Sigma) was used to inhibit voltage-gated Cl⁻ channels, and 4,4′-diisothiocyano-2, 2-stilbenedisulfonate (DIDS; 10⁻⁵ M, Sigma) was used as a broad spectrum inhibitor of Cl⁻ channels and Cl⁻/HCO₃⁻ exchangers. Bumetanide (2 × 10⁻⁴ M, Sigma) was used to inhibit Na⁺-K⁺-2Cl⁻ cotransporters, and acetazolamide (2 × 10⁻⁴ M, Sigma) was used to inhibit carbonic anhydrase. Barium chloride (BaCl₂; 5 × 10⁻³ M, Sigma) was used as a broad spectrum K⁺ channel inhibitor.

To determine whether I_sc^ATP was dependent on phospholipase C (PLC) activity, cell sheets were pretreated for 10 min with U73122 (10⁻⁶ M, Tocris Bioscience). The involvement of intracellular Ca²⁺ in I_sc^ATP responses was analyzed by pretreating cell sheets for 15 min with the Ca²⁺ chelator BAPTA-AM (5 × 10⁻⁵ M). To determine whether I_sc^ATP depended on protein kinase A (PKA) activity, cell sheets were pretreated for 30 min with H-89 (10⁻⁵ M, Sigma). The involvement of protein kinase C (PKC) in I_sc^ATP responses was analyzed with GF109203X (10⁻⁶ M, Tocris Bioscience) and Go6983 (2 × 10⁻⁵ M, Tocris Bioscience).

Reverse transcriptase-PCR.

mIMCD-K2 cells were grown to resistance on permeable supports, and total RNA was harvested using an RNeasy Mini Kit (Qiagen, Valencia, CA) according to the manufacturer's instructions. Mouse brain and kidney total RNA were obtained from Zyagen (San Diego, CA) to serve as positive controls. Reverse transcriptase (RT) reactions were performed according to the manufacturer's instructions (New England BioLabs, Ipswich, MA). Thermal cycling parameters were the following: incubation at 98°C for 30 s followed by 35 cycles at 98°C for 10 s, 50°C for 10 s, 72°C for 50 s. PCR primers were designed and used for detecting gene amplification of bestrophin-1 and TMEM16A (11, 24). Specificity of each set of primers was confirmed by BLAST search against GenBank. PCR products were resolved using a 1% agarose gel-dissolved Tris-Acetate EDTA buffer and visualized with ethidium bromide.

Western blot analysis.

Seventy micrograms of protein from mIMCD-K2 lysates were resolved by 10% SDS-PAGE and transferred to PVDF membranes (Bio-Rad, Hercules, CA). The membranes were immunostained with rabbit anti-human P2Y₁ receptor antibody (1.5 μg/ml, Sigma) or rabbit anti-rat P2Y₂ receptor antibody (4 μg/ml, Alomone Labs, Jerusalem, Israel) for detection of P2Y₁ and P2Y₂ receptor-immunoreactive protein, respectively. Membranes were incubated with horseradish peroxidase-conjugated secondary antibody (Amersham Biosciences, Piscataway, NH) and processed as described previously (28).

The specificity of the anti-P2Y₁ receptor antibody was previously tested against whole cell lysates from the P2Y₁ receptor knockout mouse (34). The specificity of the anti-P2Y₂ receptor antibody was tested by preabsorbing the antibody for 1 h with its corresponding antigen peptide at a 1:1 ratio. The membranes were then incubated with preabsorbed antigen-antibody mixture (4 μg/ml).

Statistics.

Statistical analyses for comparisons between different treatment groups of mIMCD-K2 cells were performed using paired or unpaired two-tailed Student t-tests. Differences were considered to be significant at P < 0.05.

RESULTS

ATP induces anion secretion through CACC.

Figure 1, A and B, shows that addition of ATP (10⁻⁵ M) to the basal side of mIMCD-K2 cell sheets induced a transient increase in I_sc (9.8 ± 5.4 μA/cm²) without a sustained increase in I_sc, while addition of ATP (10⁻⁵ M) to the apical side caused a sharp, transient increase in I_sc (25.8 ± 1.4 μA/cm²), and within 2 min, this transient current transitioned to a sustained I_sc (1.98 ± 0.15 μA/cm²). Addition of ATP to either side of cell sheets also stimulated a parallel decrease in R_te, with a time course that mirrored the ATP-induced increase I_sc (Fig. 1, C and D). Basal addition of ATP induced a transient decrease in R_te (−180.23 ± 4.97 Ω·cm²), while apical addition of ATP induced a larger transient decrease in R_te (−325.8 ± 29.1 Ω·cm²).

Fig. 1.

Sidedness of ATP-induced short-circuit current (I_sc) in inner medullary collecting duct (mIMCD)-K2 cells. A: representative trace of I_sc; Am indicates 10⁻⁵ M amiloride added to the apical bath; ATP(b) indicates 10⁻⁵ M ATP added to the basal side; ATP(a) indicates 10⁻⁵ M ATP added to the apical side. B: average I_sc values in response to ATP addition (ΔI_sc^ATP) to the apical (open bars) or the basal (closed bars) sides. I_sc^ATP are categorized as transient or subsequent sustained responses. Values are represented as means ± SE (n = 42 filters). *, #Values significantly different from those induced by ATP added to the opposite side. C: representative trace of transepithelial resistance (R_te); Am indicates 10⁻⁵ M amiloride added to the apical bath; ATP(b) indicates 10⁻⁵ M ATP added to the basal side; ATP(a) indicates 10⁻⁵ M ATP added to the apical side. D: average changes in R_te values in response to ATP addition (ΔRteATP) to the apical (open bars) or the basal (closed bars) sides. ΔRteATP are categorized as transient or subsequent sustained responses. Values are represented as means ± SE (n = 48 filters). *Value significantly different from that induced by ATP added to the opposite side.

Since amiloride was added to the apical bath solution to inhibit ENaC-mediated Na⁺ absorption, the observed changes in I_sc^ATP likely involved a change in electrogenic anion secretion. To identify the ion channels involved in ATP-induced anion secretion, we pretreated cell sheets with a series of small molecular inhibitors. CFTR has been implicated as an ion channel that mediates apical Cl⁻ secretion in mIMCD-K2 cells (18, 28, 38); however, apical addition of CFTR inhibitor-172, a selective inhibitor of CFTR, did not influence I_sc^ATP (Fig. 2A). Similarly, apical addition of DPAC (10⁻⁴ M, Sigma), an inhibitor of voltage-gated Cl⁻ channels, did not influence I_sc^ATP (not shown). In contrast, addition of FFA (10⁻³ M), an inhibitor of CACC (13, 40), to the basal side of cell sheets significantly reduced the transient I_sc^ATP elicited by basal addition of ATP, and it diminished, but did not completely block, the transient and sustained I_sc^ATP elicited by apical addition of ATP (Fig. 2, B-D). Addition of FFA to the apical side of cell sheets completely blocked I_sc^ATP responses elicited by ATP addition to either side of cell sheets (Fig. 2, B-D). Similar results were observed with addition of niflumic acid (2 × 10⁻⁴ M), another inhibitor of CACC, to either side of cell sheets before ATP addition (Fig. 3).

Fig. 2.

Effects of cystic fibrosis transmembrane conductance regulator (CFTR) inhibitor-172 and flufenamic acid (FFA) on ATP-induced I_sc in mIMCD-K2 cells. A: representative trace of I_sc in response to sequential addition of amiloride (Am; apical addition at 10⁻⁵ M), CFTR inhibitor-172 (CI; apical addition at 10⁻⁵ M), and basal or apical addition of ATP (10⁻⁵ M). Pretreatment with CFTR inhibitor-172 did not have a significant effect on ATP-induced I_sc. B: superimposed representative traces of I_sc responses to basal and apical addition of 10⁻⁵ M ATP following treatment with 10⁻³ M FFA added to either the basal (dashed trace) or the apical (solid trace) bath. Am, 10⁻⁵ M amiloride (apical addition); FFA(b), 10⁻³ M FFA added to the basal bath; FFA(a), 10⁻³ M FFA added to the apical bath; ATP(b), 10⁻⁵ M ATP added to the basal bath; ATP(a), 10⁻⁵ M ATP added to the apical bath. C: average I_sc values in response to basal addition of ATP (ΔI_sc^ATP) to control cells (open bars), cells pretreated with FFA added to the apical side (closed bars), and cells pretreated with FFA added to the basal side (dashed bars). Values are represented as means ± SE (n = 11 filters). D: average I_sc values in response to apical addition of ATP to control cells (open bars), cells pretreated with FFA added to the apical side (closed bars), and cells pretreated with FFA added to the basal side (dashed bars). Values are represented as means ± SE (n = 14 filters). *, **, #Values significantly different from control values.

Fig. 3.

Effect of niflumic acid (NFA) on ATP-induced I_sc in mIMCD-K2 cells. A: superimposed representative traces of I_sc responses to basal and apical addition of 10⁻⁵ M ATP following treatment with control (solid trace) or with NFA (dashed traces) added to cell sheets. Am, apical addition at 10⁻⁵ M amiloride; N, 2 × 10⁻⁴ M NFA (added to both sides); ATP(b) and ATP(a) indicate the addition of ATP added to basal and apical sides, respectively. B: average I_sc values in response to basal addition of ATP (I_sc^ATP) in control cells (open bars) or in cells pretreated with NFA (closed bars). I_sc responses are categorized as transient or subsequent sustained responses. Values are means ± SE (n = 12 filters). C: average I_sc values in response to apical addition of ATP in control cells (open bars) or in cells pretreated with NFA (closed bars). Values are means ± SE (n = 12 filters). *Values significantly different from those in control cell sheets.

To begin the process of determining the molecular identities of CACC expressed in mIMCD-K2 cells, we performed RT-PCR to evaluate mRNA expression of bestrophin-1 and TMEM16A, which have been implicated as candidates responsible for CACC currents (19). We used a PCR-based approach because we anticipated that expression of these genes in mIMCD-K2 cells would be low. Mouse brain and kidney cDNA were run in parallel as positive controls. We detected both bestrophin-1 and TMEM16A mRNA expression in mouse brain, kidney, and mIMCD-K2 cells (Fig. 4), indicating that mIMCD-K2 cells express two gene products that may mediate CACC activity.

Fig. 4.

Expression of bestrophin-1 and TMEM16A mRNA in mIMCD-K2 cells. Photograph of gel showing PCR amplification products in mouse brain and kidney tissue (positive controls) and mIMCD-K2 cells (IMCD) after reverse transcription of mouse bestrophin-1 and TMEM16A mRNA. Samples containing reverse transcriptase (RT; +) or negative controls lacking RT (−) are included for each tissue/primer combination. BP 500 indicates a marker of 500 base pairs.

To evaluate the ion transporters and channels that are required for basolateral Cl⁻ uptake during the ATP response, we pretreated mIMCD-K2 cells with bumetanide (Na⁺-K⁺-2Cl⁻ cotransporter inhibitor), DIDS (Cl⁻/HCO₃⁻ exchanger and broad spectrum Cl⁻ channel inhibitor), acetazolamide (carbonic anyhydrase inhibitor), and BaCl₂ (broad spectrum K⁺ channel inhibitor) before ATP addition. Basal application of bumetanide, DIDS, and acetazolamide did not inhibit Cl⁻ secretion (not shown), indicating that bumetanide-sensitive Na⁺-K⁺-2Cl⁻ cotransporters and Cl⁻/HCO₃⁻ exchangers are not likely to be involved in ATP-induced basolateral Cl⁻ uptake in mIMCD-K2 cells. However, basal application of BaCl₂ (5 × 10⁻³ M) significantly inhibited I_sc^ATP elicited by either apical or basal addition of ATP (Fig. 5), suggesting that a basolateral K⁺ channel is responsible for recycling K⁺ to the extracellular compartment to promote Cl⁻ secretion and uptake during the ATP response.

Fig. 5.

Basal addition of BaCl₂ to mIMCD-K2 cells inhibits ATP-induced I_sc. A: superimposed representative traces of I_sc responses to apical and basal addition of 10⁻⁵ M ATP following treatment with control (solid trace) or 5 × 10⁻³ M BaCl₂ (dashed trace). Am, 10⁻⁵ M amiloride (apical addition); Ba²⁺, 5 × 10⁻³ M BaCl₂ (basal addition); ATP(a) and ATP(b) indicate 10⁻⁵ M ATP added to the apical and basal baths, respectively. B: average I_sc values in response to apical addition of ATP (ΔI_sc^ATP) in control cells (open bars) or in cells pretreated with BaCl₂ to the basal side (closed bars). I_sc responses are categorized as transient or subsequent sustained responses. Values are means ± SE (n = 12 filters). C: average I_sc values in response to basal addition of ATP in control filters (open bars) or in filters pretreated with BaCl₂ to the basal side (closed bars). Values are means ± SE (n = 12 filters). *Values significantly different from those in control cell sheets.

ATP does not directly influence paracellular Na⁺ transport.

For transcellular Cl⁻ secretion to persist, there must also be a parallel pathway for Na⁺ secretion to offset the electrochemical gradient generated by Cl⁻ secretion. Under conditions where ENaC is downregulated, ATP-induced Cl⁻ secretion may drive paracellular Na⁺ secretion. To establish the presence of paracellular Na⁺ transport, we performed Na⁺ substitution experiments, in which some of the Na⁺ was replaced with NMDG⁺ in the apical bath solution so that the Na⁺ concentration in the basal bath was 6.6-fold greater than that in the apical bath. Figure 6 shows that, in the presence of amiloride, I_sc was significantly lower in mIMCD-K2 cells bathed in asymmetric solutions (I_sc 0.51 ± 0.23 μA/cm² for control apical bath solution compared with I_sc −2.39 ± 0.23 μA/cm² for NMDG-containing apical bath solution), suggesting that Na⁺ can be secreted through a paracellular pathway to decrease net I_sc. We next repeated these Na⁺ substitution experiments in the presence of amiloride and FFA, to block transcellular NaCl transport, and then stimulated mIMCD-K2 cells with apical or basal ATP. Under these conditions, we observed no significant changes in I_sc or R_te in response to apical (ΔI_sc = 0.26 ± 0.16 μA/cm²; ΔR_te = −0.21 ± 20 Ω·cm²) or basal (ΔI_sc = −1.06 ± 0.17 μA/cm²; ΔR_te = −38 ± 31 Ω·cm²) addition of ATP, indicating that ATP does not directly influence paracellular Na⁺ transport.

Fig. 6.

Effect of substitution of Na⁺ with NMDG⁺ in the apical bath solution on basal I_sc in mIMCD-K2 cells. NaCl was replaced with NMDG-Cl in the apical bath solution so that a basal-to-apical Na⁺ concentration gradient of 6.6:1 was established. The concentrations of all other ions in the apical and basal bath solutions were kept constant. I_sc measurements were made after accounting for tight junction potentials created by the asymmetric bath solutions and after addition of 10⁻⁵ M amiloride to the apical bath. Average I_sc values in cell sheets in control apical bath solution (Reg KHS) and in bath solution with Na⁺ substitution (NMDG). Values are represented as means ± SE (n = 12 filters). *Value significantly different from control value.

P2Y₁ and P2Y₂ receptors are expressed in mIMCD-K2 cells.

Although mIMCD-K2 cells are known to express P2Y₁ and P2Y₂ receptor mRNA (22), P2Y₁ and P2Y₂ receptor protein expression has not been demonstrated. Western blotting of mIMCD-K2 cell lysates with a previously characterized P2Y₁ antibody (34) detected an immunoreactive band of 42 kDa, which matched the predicted size of mouse P2Y₁ receptor protein (Fig. 7A). Western blotting of mIMCD-K2 cell lysates with an anti-P2Y₂ receptor antibody showed an immunoreactive band of ∼36 kDa, which is slightly lower than the predicted molecular weight of 42 kDa. Anti-P2Y₂ receptor antibody preincubated with blocking peptide, however, did not recognize this 36-kDa band, suggesting that this band represents P2Y₂-immunoreactive protein (Fig. 7B).

Fig. 7.

P2Y₁ and P2Y₂ receptor protein expression by Western immunoblot in mIMCD-K2 cells. A: representative immunoblot showing whole cell lysates from 3 different passages of mIMCD-K2 cells probed with an anti-P2Y₁ receptor (P2Y₁R) antibody. Blots were also probed with an anti-α-tubulin antibody to assess protein loading. B: representative immunoblot showing whole cell lysates from 2 different passages incubated with either P2Y₂ receptor (P2Y₂R) antibody or P2Y₂ receptor antibody preadsorbed with immunizing peptide.

P2Y₁ and P2Y₂ receptor activation stimulates Cl⁻ secretion.

Since mIMCD-K2 cells express P2Y₁ and P2Y₂ receptors, we reasoned that ATP would stimulate Cl⁻ secretion through activation of these receptors. To verify P2Y₁ and P2Y₂ receptor activation is responsible for I_sc^ATP, we used specific agonists, MRS2365 and thio-UTP, to activate P2Y₁ and P2Y₂ receptors, respectively. Basal addition of MRS2365 (10⁻⁶ M) to cell sheets caused a transient increase in I_sc (4.8 ± 0.77 μA/cm²) without a subsequent sustained change in I_sc (Fig. 8). Apical addition of MRS2365 induced a small, transient increase in I_sc (0.7 ± 0.2 μA/cm²; Fig. 8).

Fig. 8.

Effects of the P2Y₁ receptor agonist MRS2365 and the P2Y₂ receptor agonist thio-UTP on I_sc in mIMCD-K2 cells. A: representative trace of I_sc in response to sequential addition of the following: Am, 10⁻⁵ M amiloride (apical addition); MRS(b) and UTP(b), basal addition of MRS2365 (10⁻⁶ M) and thio-UTP (10⁻⁶ M), respectively; MRS(a) and UTP(a), apical addition of MRS2365 and thio-UTP, respectively. B: average I_sc values after addition of MRS2365 (ΔI_sc) to the apical side (open bars) or the basal side (closed bars) of cell sheets. I_sc responses are categorized as transient or subsequent sustained responses. Values are represented as means ± SE (n = 14 filters). C: average I_sc values after addition of thio-UTP to the apical side (open bars) or the basal side (closed bars). Values are means ± SE (n = 14 filters). *, #Values that are significantly different from those induced by the same agonist added to the opposite side.

Basal addition of the specific P2Y₂ receptor agonist thio-UTP (10⁻⁶ M) to cell sheets induced a small, transient increase in I_sc (2.7 ± 0.8 μA/cm²), which is not apparent in the representative trace in Fig. 8A. Similar to the effect of apical addition of ATP on I_sc, apical addition of thio-UTP induced a large, transient increase in I_sc (14.24 ± 1.6 μA/cm²) that transitioned to a sustained I_sc (1.05 ± 0.12 μA/cm²; Fig. 8).

To further confirm the involvement of P2Y₁ receptors in I_sc^ATP, we incubated cell sheets with the P2Y₁ receptor antagonist MRS2500 before the addition of MRS2365. Pretreatment of both sides of cell sheets with MRS2500 (10⁻⁶ M) completely blocked the stimulatory effect of basal addition of P2Y₁ receptor agonist MRS2365 on I_sc (Fig. 9), confirming that P2Y₁ receptor activation by MRS2365 at the basal side of cell sheets enhances I_sc. In contrast, pretreatment of cell sheets with MRS2500 did not influence the stimulatory effects of basal or apical addition of P2Y₂ receptor agonist thio-UTP on I_sc (Fig. 9). Pretreatment with MRS2500 also did not have any effect on the I_sc response to basal ATP addition (Fig. 9C), indicating that P2Y₂-mediated stimulation of I_sc^ATP can occur despite P2Y₁ receptor blockade.

Fig. 9.

Effects of the P2Y₁ receptor antagonist MRS2500 on MRS2365-, thio-UTP-, or ATP-induced I_sc in mIMCD-K2 cells. A: superimposed representative traces of I_sc responses to the addition of the P2Y₁ agonist MRS2365 or the P2Y₂ agonist thio-UTP following treatment with control (solid trace) or with 10⁻⁶ M P2Y₁ receptor antagonist MRS2500 (dashed trace) added to both sides of cell sheets. Am, 10⁻⁵ M amiloride; MRS2500, 10⁻⁶ M MRS2500; MRS(a) and MRS(b) indicate addition of 10⁻⁶ M P2Y₁ agonist MRS2365 to the apical and basal sides, respectively. UTP(b) and UTP(a) indicate the addition of 10⁻⁶ M P2Y₂ agonist thio-UTP to the basal and apical sides, respectively. B: average I_sc values in the transient response to basal addition of either MRS2365 (ΔI_sc) or thio-UTP in control cells (open bars) or in cells pretreated with MRS2500 (closed bars). Values are means ± SE (n = 9 filters). *Value significantly different from control value. C: representative trace of I_sc showing that pretreatment of both sides of cell sheets with the P2Y₁ antagonist MRS2500 (10⁻⁶ M) blocked the stimulatory effects of the P2Y₁ agonist MRS2365 (10⁻⁶ M) added to the basal side of cell sheets [MRS(b)]. MRS(a) indicates MRS2365 (10⁻⁶ M) added to the apical side. Pretreatment of MRS2500 did not affect the stimulatory effects of ATP on I_sc when added to either the basal [ATP(b)] or apical [ATP(a)] sides of cell sheets (n = 6 filters).

ATP signals through phospholipase C and intracellular calcium.

P2Y receptors are linked to G proteins that activate PLC enzyme activity (8, 9), which can lead to activation of downstream pathways involving PKC and intracellular Ca²⁺. To further characterize the signaling pathways involved in I_sc^ATP, we treated mIMCD-K2 cells with the PLC inhibitor U73122 before the addition of ATP. Pretreatment of cell sheets with U73122 (10⁻⁶ M) completely abolished the I_sc^ATP response elicited from either side of cell sheets (Fig. 10), confirming that ATP action on ion transport in mIMCD-K2 cells involves stimulation of PLC activity. We also used PKC inhibitors GF109203X (10⁻⁵ M) and Go6983 (2 × 10⁻⁵ M) to evaluate whether PKC is involved in I_sc^ATP, but these inhibitors did not alter I_sc^ATP responses (not shown). To confirm whether I_sc^ATP involves Ca²⁺ signaling pathways and CACC-mediated Cl⁻ secretion, we used the Ca²⁺ chelator BAPTA-AM to decrease the intracellular Ca²⁺ concentration in mIMCD-K2 cells. Pretreatment of cell sheets with BAPTA-AM (5 × 10⁻⁵ M) significantly diminished I_sc^ATP responses elicited by either apical (transient response 6.48 ± 1.10, and sustained response 0.57 ± 0.13 μA/cm²) or basal (no response) ATP addition to cell sheets (Fig. 11).

Fig. 10.

Phospholipase C (PLC) inhibitor U73122 inhibits ATP-induced I_sc in mIMCD-K2 cells. A: superimposed representative traces of I_sc responses to apical and basal addition of 10⁻⁵ M ATP following treatment with control (solid trace) or 10⁻⁶ M U73122 (dashed trace). Am, 10⁻⁵ M amiloride (apical addition); U, U73122; ATP(a) and ATP(b) indicate 10⁻⁵ M ATP added to the apical and basal baths, respectively. B: average I_sc values in response to apical addition of ATP (10⁻⁵ M) in control cells (open bars) or in cells pretreated with U73122 on both sides (closed bars). I_sc responses are categorized as transient or subsequent sustained responses. Values are means ± SE (n = 14 filters). C: average I_sc values in response to basal addition of ATP (10⁻⁵ M) in control filters (open bars) or in filters pretreated with U73122 (closed bars). Values are means ± SE (n = 14 filters). *, #Values significantly different from those in control cell sheets.

Fig. 11.

BAPTA-AM inhibits ATP-induced I_sc in mIMCD-K2 cells. A: superimposed representative traces of I_sc responses to apical and basal addition of 10⁻⁵ M ATP following treatment with control (solid trace) or BAPTA-AM (dashed trace). Am, 10⁻⁵ M amiloride (apical addition); BAPTA, 5 × 10⁻⁵ M BAPTA-AM (added to both sides); ATP(a) and ATP(b) indicate 10⁻⁵ M ATP added to the apical and basal baths, respectively. B: average I_sc values in response to apical addition of ATP (ΔI_sc^ATP) in control filters (open bars) or in filters pretreated with BAPTA-AM (closed bars). I_sc responses are categorized as transient or subsequent sustained responses. Values are means ± SE (n = 12 filters). C: average I_sc values in response to basal addition of ATP in control filters (open bars) or in filters pretreated with BAPTA-AM (closed bars). Values are means ± SE (n = 12 filters). *, #Values significantly different from those in control cell sheets.

We recently showed that adenosine stimulates Cl⁻ secretion though CFTR in mIMCD-K2 cells by activating apical adenosine A2b receptors and signaling through cAMP/PKA (28). Since the magnitude of the sustained I_sc^ATP was similar to that of adenosine-stimulated I_sc (I_sc^adenosine), we used a series of small molecule compounds to determine whether I_sc^ATP is distinct from Iscadenosine. Addition of PSB603 (adenosine A2b receptor inhibitor), DPCPX (broad spectrum adenosine receptor inhibitor), and H-89 (PKA inhibitor) had no effects on the transient or sustained I_sc^ATP elicited by ATP addition to cell sheets (not shown), indicating that adenosine receptor signaling pathways are not involved in the sustained I_sc^ATP response.

DISCUSSION

In this study, we used the mIMCD-K2 cell line to examine direct effects of extracellular nucleotides on NaCl transport across IMCD epithelia. Our data show that ATP stimulates CACC in mIMCD-K2 cells by activating both P2Y₁ and P2Y₂ receptor pathways, which are localized on different sides of polarized epithelia and signal through PLC and intracellular Ca²⁺. Our experiments further demonstrate that ATP stimulates Cl⁻ secretion in the presence of amiloride, indicating that ATP plays an important role in regulating ion transport under conditions of minimal ENaC activity. While ATP stimulates Cl⁻ secretion under these conditions, ATP does not directly stimulate paracellular Na⁺ secretion, but rather it secondarily drives paracellular Na⁺ transport by inducing transcellular Cl⁻ secretion.

The molecular identity of the proteins responsible for CACC currents in mIMCD-K2 cells (4, 5), as well as in other cells and tissues (15), is currently unresolved. However, emerging studies implicated the bestrophin and TMEM16 family of proteins as candidates that recapitulate many of the properties of CACC currents (3, 6, 32, 42). Using a PCR-based approach, we detected bestrophin-1 and TMEM16A mRNA in mIMCD-K2 cells. It is not presently clear, however, whether bestrophin-1, TMEM16A, or both proteins might be responsible for CACC activity in these cells. The expression studies presented here should provide guidance for future studies regarding the molecular identity of the proteins responsible for ATP-regulated CACC currents in mIMCD-K2 cells.

McCoy et al. (22) originally demonstrated that extracellular nucleotides stimulate Cl⁻ secretion in mIMCD-K2 cells and they suggested, based on mRNA expression studies, that P2Y₁ and P2Y₂ receptors regulate Cl⁻ secretion. Using a panel of purinergic receptor agonists and antagonists, we confirmed this hypothesis and showed further that P2Y₁ and P2Y₂ receptors regulate Cl⁻ secretion through activation of CACC. Although P2Y receptor activation stimulated CACC-mediated Cl⁻ secretion from either side of IMCD epithelia, the I_sc response profiles differed depending on which side was stimulated. Apical addition of ATP or UTP, but not the P2Y₁ agonist MRS2365, caused transient increases in I_sc and subsequent sustained increases in I_sc, while basal addition of ATP, UTP, or MRS2365 induced isolated, transient increases in I_sc. The mechanisms responsible for separate I_sc response profiles that vary according to sidedness (apical vs. basal) rather than receptor subtype (P2Y₁ vs. P2Y₂) remain to be determined.

We also found that ATP-induced basolateral Cl⁻ uptake was not sensitive to bumetanide, which has also been described in mIMCD-K2 cells as well as in primary cultures of the rat IMCD cells (16, 18). Additional studies will be required to more fully characterize the molecular mechanisms controlling basolateral Cl⁻ uptake in mIMCD-K2 cells. Interestingly, the mechanisms involved in ATP-induced basolateral Cl⁻ uptake appear to be distinct from those involved in cAMP or vasopressin-induced Cl⁻ uptake via bumetanide-sensitive Na⁺-K⁺-2Cl⁻ cotransporters and Cl⁻/HCO₃⁻ exchangers (18). The mechanisms involved in ATP-induced apical Cl⁻ exit via CACC are also different from those involved in vasopressin-induced Cl⁻ exit via CFTR (18, 38). Collectively, these findings suggest that a variety of hormones can regulate signal transduction and Cl⁻ transport pathways in a context-specific manner in mIMCD-K2 cells.

To our knowledge, this is the first demonstration of P2Y₁ receptor-mediated stimulation of Cl⁻ secretion in a kidney model system. Previous studies described P2Y₁ mRNA expression in mIMCD-K2 cells (22) and implicated P2Y₁ receptors in Ca²⁺ signaling pathways in the outer medullary CD (7). In the few studies that delineated P2Y₁ expression in the kidney, P2Y₁ receptor expression has been localized only to basolateral surfaces (1). This finding is congruent with our data in mIMCD-K2 cells that activation of basal P2Y₁ receptors stimulates Cl⁻ secretion. Bailey et al. (1, 2) demonstrated that P2Y₁ and P2Y₂ receptors are present together in cells from microdissected rat CD and suggested that both receptors share a common pool of intracellular Ca²⁺. Our findings similarly suggest that both P2Y₁ and P2Y₂ receptor pathways signal through PLC to increase intracellular Ca²⁺ concentration in IMCD cells, thus activating CACC and increasing Cl⁻ secretion. Given that ATP and ADP content in rat kidney tissue slices have been observed to increase under conditions of high-dietary salt intake (12), the presence of P2Y₁ receptors on the basolateral surface of IMCD cells may constitute a potential mechanism by which ATP and by-products such as ADP stimulate Cl⁻ secretion as an adaptive response to enhance urine NaCl excretion.

Although P2Y₂ receptors are known to mediate the stimulatory effects of extracellular ATP on CACC-mediated Cl⁻ secretion in cortical CD cells (10), the physiological role of Cl⁻ secretion in the cortical CD is not clear. In fact, the only nephron segment that has been implicated as an in vivo site for Cl⁻ secretion by the kidney is the IMCD. Wallace et al. (39) microperfused isolated rat IMCD segments and demonstrated fluid secretion that was coupled to active NaCl secretion, with the highest rate of fluid secretion in the initial segment of the IMCD, the same segment from which mIMCD-K2 cells are derived. Sonnenberg (35, 39) used retrograde microcatheterization to show that rat IMCD has the capacity to secrete up to 10% of the filtered load of NaCl when ECF volume is expanded. While the capacity for NaCl secretion by the IMCD may appear limited, changes in net tubular excretion in the IMCD as low as 0.055% of the filtered NaCl load can be sufficient to cause significant changes in ECF volume over time (14). Thus, in states of NaCl excess, Cl⁻ secretion pathways in IMCD, including P2Y receptor pathways identified here, could play an important role in the daily regulation of urinary NaCl excretion.

Recently, we demonstrated that extracellular adenosine stimulates Cl⁻ secretion through CFTR in mIMCD-K2 cells by activating apical A2b receptors and signaling through the cAMP/PKA pathway (28). Adenosine is derived from ATP through the action of ecto-nucleotidases, which are abundantly expressed at the apical membrane surface of kidney epithelial cells (37). The action of both extracellular nucleosides and nucleotides, signaling through separate but parallel Cl⁻ channel pathways in the IMCD, may represent a membrane signaling network that can fine-tune the regulation of urinary NaCl excretion.

In conclusion, the integrated control by the kidney of NaCl retaining and excreting systems determines NaCl balance, ECF volume, and blood pressure. Under conditions of high-dietary salt intake in mice, urinary nucleotide concentration increases; we propose that extracellular nucleotides may activate both P2Y₁ and P2Y₂ receptor pathways in IMCD cells, stimulating both sustained and repeated bursts of Cl⁻ secretion through CACC channels as an alternative adaptive mechanism for enhancing urinary NaCl excretion. The activation of these signaling pathways would become particularly important under states of ECF volume expansion, when levels of serum aldosterone are low and ENaC activity is suppressed. Defects in this pathway could contribute to the pathogenesis of salt-sensitive hypertension because inappropriate control of renal salt handling could result in expansion of ECF volume and elevated blood pressure.

GRANTS

This work was supported by grants from the National Institutes of Health (K08-DK-073487 to A. C. Pao and T32-DK07357-26A1 to P. P. Kathpalia), Amgen (2009 Amgen Nephrology Junior Faculty Research Support Program to A. C. Pao), Satellite Healthcare (2008 Norman S. Coplon Extramural Grant to A. C. Pao), the Forest Research Institute (2009 Emerging Leaders in Hypertension Program to P. P. Kathpalia), and Stanford University (Dean's Fellowship Award to S. V. Thomas).

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the author(s).