P2 purinoceptors regulate calcium-activated chloride and fluid transport in 31EG4 mammary epithelia



It has been reported that secretory mammary epithelial cells (MEC) release ATP, UTP, and UDP upon mechanical stimulation. Here we examined the physiological changes caused by ATP/UTP in nontransformed, clonal mouse mammary epithelia (31EG4 cells). In control conditions, transepithelial potential (apical side negative) and resistance were −4.4 ± 1.3 mV (mean ± SD, n = 12) and 517.7 ± 39.4 Ω · cm2, respectively. The apical membrane potential was −43.9 ± 1.7 mV, and the ratio of apical to basolateral membrane resistance (RA/RB) was 3.5 ± 0.2. Addition of ATP or UTP to the apical or basolateral membranes caused large voltage and resistance changes with an EC50 of ∼24 μM (apical) and ∼30 μM (basal). Apical ATP/UTP (100 μM) depolarized apical membrane potential by 17.6 ± 0.8 mV (n = 7) and decreasedRA/RB by a factor of ≈3. The addition of adenosine to either side (100 μM) had no effect on any of these parameters. The ATP/UTP responses were partially inhibited by DIDS and suramin and mediated by a transient increase in free intracellular Ca2+ concentration (427 ± 206 nM; 15–25 μM ATP, apical; n = 6). This Ca2+ increase was blocked by cyclopiazonic acid, by BAPTA, or by xestospongin C. 31EG4 MEC monolayers also secreted or absorbed fluid in the resting state, and ATP or UTP increased fluid secretion by 5.6 ± 3 μl · cm−2 · h−1(n = 10). Pharmacology experiments indicate that 31EG4 epithelia contain P2Y2 purinoceptors on the apical and basolateral membranes, which upon activation stimulate apical Ca2+-dependent Cl channels and cause fluid secretion across the monolayer. This suggests that extracellular nucleotides could play a fundamental role in mammary gland paracrine signaling and the regulation of milk composition in vivo.

extracellular atp or utp and other purine nucleotides mediate multiple cellular responses through specific membrane-bound purinoreceptors in diverse tissues. Purinoceptor stimulation affects antiproliferative and contractile properties in vascular tissue, endothelial PGI2release, protein phosphorylation in numerous cell types, bile acid secretion in the gastrointestinal tract, and an increase in mucociliary clearance and mucin release in the lung (7, 18, 22, 37, 45,50). In various epithelia, metabotropic P2 purinoceptors have been shown to stimulate second messenger-induced activation of Cl (and K) channels (6, 39, 43, 46, 53, 55).

In mammary epithelia, ATP, UTP, and UDP are released in the extracellular space after mechanical stimulation (16). Subsequent ATP-dependent contraction of adjacent myoepithelial cells can stimulate further nucleotide release (40) and perhaps lead to increased fluid secretion from mammary cells. Because ATP, UTP, and other P2 agonists are released from cells, their presence in human cyst fluid and milk secretions is highly likely. Indeed, ATP has been found in bovine milk (14, 44), and purinoceptors have been observed in MCF-7 human tumor breast cells (17), suggesting that paracrine regulation may play a vital role in fluid homeostasis in the breast. In addition to regulating the composition of milk, purinoceptors may play a role in the accumulation of abnormal breast fluid common in premenopausal women (5, 36). This cyst fluid consists of a myriad of compounds, including hormones, growth factors, glycoprotein, bile acids, proteins, and glucose.

Previously, we showed that fluid secretion across intact monolayers of confluent 31EG4 mammary epithelia is in part mediated by apical membrane Na [epithelial Na channel (ENaC)] and Cl [cystic fibrosis transmembrane conductance regulator (CFTR)] channels and, possibly, Ba2+-sensitive K channels (3). We now provide evidence for P2 purinoceptors at both the apical and basolateral membranes of 31EG4 cells, using electrophysiological, fluid transport, and intracellular Ca2+ measurements. The activation of these receptors by either ATP or UTP significantly increased intracellular Ca2+ levels ([Ca2+]in), apical membrane conductance, and steady-state fluid secretion across the epithelium. These data suggest that in vivo mammary gland function is determined by paracrine, autocrine, and hormonal activation of metabotropic P2 purinoceptors.


Cell culture.

31EG4 cells were cultured in DMEM/F-12 media containing 5% FBS, 5 μg/ml insulin, and 5 μg/ml gentamicin sulfate. Upon reaching confluence, cells were plated on Transwell filters (Costar) at a density of 105 cells/well. When the cells became confluent monolayers on the filters, 1 μM dexamethasone in 2% FBS was added to stop growth and induce differentiation, including formation of tight junctions (58) and polarization of ion transport proteins to the apical and basolateral membranes (49). Transepithelial resistance (RT) was recorded using an EVOM (Epithelial Voltammeter; World Precision Instruments, New Haven, CT).


Total RNA was extracted from 3lEG4 cells using Qiagen RNeasy per the manufacturer's instructions. First-strand cDNA synthesis was carried out with 0.5 pg of total RNA, 20 pM oligo(dT) primers, 0.5 mM dNTP mix, and 200 units of Moloney murine leukemia virus RT in 20 ml (final volume) of 50 mM Tris · HCl (pH 8.3), 75 mM KCl, and 3 mM MgCl2. The RNA and oligo(dT) were annealed by mixing two items, heating to 70°C for 2 min, and then cooling on ice followed by addition of the remaining reaction components. The mixture was incubated at 42°C for 1 h for first-strand synthesis and then heated to 94°C for 4 min to stop the reaction. The mixture was then diluted 1:5 with sterile distilled water. A control reaction containing no RT was included for each tissue RT reaction to ensure that no genomic DNA was being amplified (data not shown).

PCR was carried out using oligonucleotide primers (sense 5′-GAC ACC ATC AAT GGC ACC TGG GAT-3′; anti-sense 5′-TGA TGC AGG TGA GGA AGA GGA TGC-3′) designed to cover a 355-bp region of P2Y2. The primers (0.5 mM final concentration) were added to a 1-ml aliquot of the first-strand synthesis mixture with the following: 0.2 mM dNTPs (each), 1.25 units of Pfu turbo polymerase, PCR buffer with Mg (Boehringer Mannheim), and water to bring to a total volume of 50 μl. The reaction was overlaid with two drops of mineral oil (Sigma), and the mixture was then incubated in a thermal cycler (Stratagene) with the following amplification profile: 1 cycle at 94°C for 4 min, 39 cycles at 94°C for 1 min, 55°C for 1 min, 72°C for 1 min, and 1 cycle of 72°C for 10 min. The PCR product was run on a 1.5% agarose gel and stained with ethidium bromide.

Bathing solutions and materials.

The control Ringer solution for all measurements (intracellular recording, Ca2+ measurement, and fluid transport) contained (in mM) 113.5 NaCl, 5 KCl, 26 NaHCO3, 1.8 CaCl2, 0.8 MgSO4, 1.0 NaH2PO4, 5.5 glucose, and 5 taurine, pH 7.4. The following drugs were obtained from Sigma Chemical (St. Louis, MO): ATP, UTP, ADP, UDP, benzoylbenzoyl-ATP (BzBz-ATP), adenosine 3′,5′-diphosphate (A3P5P),N6-methyl-2′-deoxyadenosine-3′,5′-bisphosphate (MRS2179), ADPβS, 2-methylthio-ATP (2-MeSATP), α,β-methylene ATP (α,β-MeATP), suramin, and DIDS. H2DIDS and BAPTA-AM were obtained from Molecular Probes (Eugene, OR), and xestospongin C was obtained from Calbiochem (La Jolla, CA).

Microelectrode electrophysiology.

Transwells with confluent 31EG4 monolayers andRT >300 Ω · cm2 were used for electrophysiology measurements, as previously described (3). Monolayers on filters were mounted on a nylon mesh support and clamped in a modified Ussing chamber, apical side up. The perfusion of Ringer to each side of the tissue was controlled separately. The apical perfusion rate was ∼26 chamber vol/min; the basal perfusion rate was ∼79 chamber vol/min. Ringer solution changes were delivered to the chamber from a short distance, which delayed arrival by ∼30 s. Calomel electrodes in series with Ringer solution were used to measure the transepithelial potential (TEP), and the signals from intracellular microelectrodes were referenced to either the apical or basal bath to measure the membrane potentials,VA and VB, where TEP = VBVA(30, 38). Conventional microelectrodes, with resistances of 80–250 MΩ, were made from fiber-filled borosilicate glass tubing with 0.5 mm inner diameter and 1 mm outer diameter (Sutter Instrument, Novato, CA) and were back-filled with 150 mM KCl.

The ratio of the apical (RA) to basolateral (RB) membrane resistance (a value) and RT were obtained by passing 4 or 8 μA current pulses (peak to peak) across the tissue and measuring the resultant changes in TEP, VA, andVB. Current pulses were bipolar, with a period of 3 s applied at various time intervals.RT is the resulting change in TEP divided by the appropriate current, and a is the absolute value of the ratio of voltage change in VA divided by the change in VB (a = ΔVAVB). The current-induced voltage deflections were digitally subtracted from the records for clarity.

Calculating membrane and shunt resistances and equivalent electromotive forces.

The electrical properties of the 31EG4 monolayer can be modeled as a simple equivalent circuit (3) where the apical and basolateral membranes are electrically coupled by the paracellular shunt pathway and possibly by edge damage around the circumference of the mounted tissue. The apical and basolateral membranes of the monolayer are each represented as an equivalent electromotive force (EMF), EA or EB, in series with a resistor, RA orRB, respectively (30, 38). The paracellular pathway is represented as a shunt resistor,Rs, which is a parallel combination of the paracellular resistances between neighboring cells and the resistance of the mechanical seal around the tissue. In general, a shunt current (IS) will flow around the circuit becauseEAEB.IS is given by

IS=(EAEB)/(RA+RB+RS)Equation 1
Because the TEP is basolateral side positive,IS flows through shunt resistance (RS) and then inward across the apical membrane. This circulating current hyperpolarizes the recorded apical membrane potential (VA) relative toEA and depolarizes the recorded basolateral (VB) membrane potential relative toEB. The observed steady-state membrane potentials, VA and VB, are related to the respective EMFs by
RT and the ratio of apical to basolateral membrane resistance (a) are expressed in terms of the membrane and RS as follows
RT=[(RA+RB)RS]/(RA+RB+RS)anda=RA/RBEquation 3
The apical and basolateral membrane voltages are electrically coupled via Rs so that a voltage change at one membrane will be shunted to the opposite membrane. For example, a membrane voltage change, ΔVA, originating at the apical membrane will produce a smaller change with the same time course at the basolateral membrane, ΔVB(30).

We could calculate RA,RB, and RS by usingEqs. 2, 3, and 4 before and after the addition of a K channel blocker (Ba2+) to the apical bath (responses in Ba2+ denoted with *). This calculation assumes that Ba2+ affected only the apical membrane, whereas RS and RBremained constant (3, 30)

RS=[RTRT*(a*a)]/[RT(1+a*)RT*(1+a)]Equation 4


[Ca2+]in levels were monitored with the fluorometric ratioing dye fura 2-AM (Molecular Probes) in a modified Ussing chamber. In this chamber, the tissue is mounted apical side down. The chamber and recording setup have been described previously (3, 31, 35, 41). In these experiments, the perfusion of Ringer solution to each side of the tissue was controlled separately. The apical perfusion rate was ∼20 chamber vol/min and the basal perfusion rate was ∼5 chamber vol/min. Ringer solutions were delivered over a short distance, delaying the arrival by ∼30 s.

Cell monolayers were loaded with dye by bathing them in Ringer solution containing 8.2–12.5 μM fura 2-AM (dissolved in DMSO + 10% pluronic acid) for 2 h (8% CO2) at room temperature. In addition, 1 mM probenicid was included in the apical bath and in all subsequent apically perfused Ringer solutions to inhibit dye extrusion, presumably mediated by an organic anion transporter, as in other epithelia (31). Photic excitation was achieved using a Xenon light source filtered at λ = 340 and 380 nm every 0.5 s; the emission fluorescence was measured at 510 nm with a photomultiplier tube (Thorn; EMI). The ratio of the fluorescence intensities (R) at 340/380 nm was determined every second. The technique and computer software for data acquisition have been described previously (2). Calibration of [Ca2+]in was performed at the end of each experiment by first perfusing both membranes with a zero-Ca2+ Ringer solution containing 10 mM EGTA, which chelates any residual free Ca2+, and 10 μM ionomycin, a Ca2+ ionophore that facilitates the equilibration of [Ca2+]in and extracellular Ca2+concentration. After this zero Ca2+ calibration, the tissue was exposed to a saturating (1.8 mM) concentration of Ca2+. [Ca2+]in was determined according to the following equation: [Ca2+]in =K([R − Rmin]/[Rmax − R]), where K =Kd(Fmin/Fmax).Kd is the dissociation constant for fura 2-AM, and Fmin and Fmax are the fluorescence intensities at 380 nm in the presence and absence, respectively, of saturating Ca2+. With the use of solutions of known [Ca2+]in and 6 μg/ml digitonin or 20 μM ionomycin, the value for K was determined to be ∼140 nM.

Fluid transport.

The rate of transepithelial fluid flow (JV) was measured using the capacitance probe technique, as previously described (25, 29). A monolayer of cells on a filter (0.5 cm2 exposed area) was mounted between two water-impermeable Kel F half-chambers. JV was determined using a sensitive oscillator circuit (Acumeasure 1000; Mechanical Technology, Lantham, NY) connected to two probes, one on either side of the tissue, that measure the capacitance between the probe tips and the fluid meniscuses connected to each half-chamber. Net fluid movement across the epithelium, from the apical to basolateral surface, or vice versa, is recorded by the changes in probe output voltage. This technique has a resolution of ∼1 nl/min. Ports in the bottom of separate half-chambers allow for the perfusion of solutions with different chemical compositions to either side of the tissue. Voltage-sensing and current-passing agar bridges built in each half-chamber permit continuous monitoring of TEP and RT, the latter calculated from the TEP voltage deflections in response to transepithelial current pulses of 4–10 μA.

All data are presented as means ± SD, unless otherwise specified. Student's unpaired t-test was used to compare groups, andP < 0.05 is considered statistically significant.



RT-PCR was performed on 31EG4 cells grown to confluence in the presence of dexamethasone to increase differentiation of the cells. The PCR experiment summarized in Fig. 1 shows that mRNA for the P2Y2 purinoceptor is present in 31EG4 cells.

Fig. 1.

Fig. 1.PCR product for the P2Y2 purinoceptor (355 bp). The amplified product was run on a 1.5% agarose gel and stained with ethidium bromide. The 100-bp ladder is on left(marker).

ATP- and/or UTP-induced electrical responses.

Figure 2 shows typical intracellular microelectrode recordings from 31EG4 mammary epithelial cells in the absence and presence of ATP in the apical (A) or basal (B) baths. In Fig. 2A, the resting, unstimulated levels of VA and VB were −37.5 and −33.5 mV, respectively, with a TEP of 4 mV, whereasRT andRA/RB were 385 Ω · cm2 and 2.6. The addition of 100 μM ATP to the apical bath transiently increased TEP by 3 mV, since VA depolarized at a faster rate thanVB; concurrently, RTdecreased by 70 Ω · cm2 andRA/RB decreased from ∼2.6 to 0.2 and then slowly returned to baseline. This return was accelerated by the removal of ATP from the apical bath. Figure2B shows that a similar set of responses was obtained with the addition of 100 μM ATP to the basal bath. Basal addition of ATP caused a rapid increase in TEP of 2.8 mV becauseVA depolarized faster thanVB; concomitantly, there was an 80 Ω · cm2 drop inRT, andRA/RB decreased from 2.75 to 0.75. The magnitude and direction of the resistance changes provide strong evidence that ATP (or UTP) increased apical membrane conductance (42).

Fig. 2.

Fig. 2.A: effects of apical ATP (100 μM) on membrane voltages and resistances in 31EG4 cells. VA andVB, apical and basolateral membrane potentials, respectively; RA/RB, ratio of apical to basolateral membrane resistance; TEP (=VBVA), transepithelial potential; RT, total tissue resistance (materials and methods).B: effects of basal ATP (100 μM) on membrane voltages and resistances in 31EG4 cells.

Figure 3, A and B, shows that apical or basal bath UTP (20 μM) produced a similar set of electrical responses. The summary data for a set of ATP/UTP experiments at 50 μM are presented in Table 1 (the responses to ATP or UTP are indistinguishable and have been combined). Table 1 indicates that apical or basal addition of secretagogue elicited similar responses. Addition of ATP/UTP caused membrane voltage and resistance changes that are consistent with a conductance increase of an apical membrane channel whose equilibrium potential is depolarized with respect to VA.

Fig. 3.

Fig. 3.A: effects of apical UTP (20 μM) on membrane voltages and resistances in 31EG4 cells. B: effects of basal UTP (20 μM) on membrane voltages and resistances in 31EG4 cells.

Table 1. Summary of electrophysiological data from control and ATP- or UTP (50 μM)-treated tissues

TEP, mVRT,Ω · cm2VA, mVVB, mVRA/RB
Untreated−4.4 ± 1.3517.7 ± 39.4−43.9 ± 1.7−48.2 ± 1.63.5 ± 0.2
ApicalATP or UTP−6.5 ± 1.3409.2 ± 19.6−26.9 ± 1.1−33.9 ± 1.21.8 ± 0.3
BasalATP or UTP−7.6 ± 1.7385.5 ± 16.4−23.7 ± 1.4−31.2 ± 1.31.4 ± 0.2

Values are means ± SD. TEP, transepithelial potential;RT, transepithelial resistance;VA, apical membrane voltage;VB, basal membrane voltage;RA/RB, ratio of apical to basal membrane resistance; untreated, mean values of membrane voltage and resistance in control Ringer (n = 12 experiments); apical, membrane voltages and resistances after addition of ATP or UTP (50 μM) to the apical bath (n = 7); basal, membrane voltages and resistances after addition of ATP or UTP to the basal bath (n = 5). The difference between control and apical or basal ATP (or UTP)-induced electrical responses is statistically significant for all parameters (P < 0.001).

Effects of apical or basal DIDS or suramin on the ATP- and/or UTP-induced changes in membrane voltage and resistance.

It has been shown that DIDS and suramin can block P2Y receptors (8, 23, 41). Figure4A shows the UTP-induced changes in membrane voltage and resistance first in the absence and then in the presence of basal DIDS. After addition of 50 μM UTP to the basal bath, VA depolarized more thanVB (TEP increased by 1.5 mV),RT decreased by 75 Ω · cm2, andRA/RB decreased from ∼3.0 to 1.0. After 16 min in control Ringer, DIDS (500 μM) was added to the basal bath where it had little effect onVA, VB,RA/RB,RT, or TEP. Subsequent addition of 50 μM basolateral UTP had little effect on membrane voltage and produced no significant resistance changes. DIDS blockade of the UTP responses was equally effective in three other experiments (ΔTEP = −0.2 ± 0.1 mV, ΔRT= −1.7 ± 1.5).

Fig. 4.

Fig. 4.A: basal UTP (50 μM)-induced changes in membrane voltage and resistance blocked by basal DIDS. B: apical ATP (100 μM)-induced changes in membrane voltage and resistance blocked by apical DIDS. C, top: sequential UTP (50 and 25 μM) basal control responses.Bottom: sequential ATP (100 μM) apical and basal control responses.

In many epithelia, DIDS has been shown to specifically block Ca2+-activated Cl channels at 0.5 mM (47). Figure 4B compares the electrical responses obtained by the addition of 100 μM ATP to the apical (or basal) bath, first in the absence (control) and then in the presence of apical DIDS. After the addition of apical ATP, the tissue was returned to control Ringer for 8 min, and then 500 μM DIDS was added to the apical bath to block Ca2+-activated Cl channels and P2Y receptors. Again, DIDS itself produced no noticeable changes in voltage or resistance. In the presence of apical DIDS, ATP was added to the basal bath and produced electrical responses much smaller than control (by a factor of >10). Subsequent addition of ATP to the apical bath was also without appreciable effect, because both apical membrane Cl channels and P2 purinoceptors were blocked. The control experiment in Fig.4C shows that, in the absence of DIDS, repeated application of ATP or UTP produces typical changes in TEP andRT (and VA,VB, andRA/RB; data not shown).

Figure 5A compares the electrical responses obtained by the addition of 100 μM ATP to the basal bath, first in the absence and then in the presence of basal suramin. In Fig. 5A, basal addition of ATP produced typical control responses (Table 1). After the ATP control response, 100 μM suramin was added to the basal bath. In its presence, ATP (100 μM) had no effect on membrane voltage or resistance. This result was also obtained in three other tissues. Figure 5B (same tissue) shows that the subsequent addition of suramin to the apical bath completely blocked the apical ATP responses (n = 4).

Fig. 5.

Fig. 5.A: basal ATP-induced changes in membrane voltage and resistance are blocked by basal suramin. B: apical ATP-induced changes in membrane voltage and resistance are blocked by apical suramin. Continuation of recording in A.

ATP- and/or UTP-regulated [Ca2+]in.

The electrical effects of ATP/UTP and the abolition of those effects by suramin or DIDS suggest the presence of metabotropic (G protein-coupled) purinoceptors at both the apical and basolateral membranes of the 31EG4 mammary cells. Activation of metabotropic receptors is normally followed by an inositol trisphosphate (IP3)-mediated increase in [Ca2+]in (32). Figure6A, top, shows the change in [Ca2+]in after multiple additions of 25 μM ATP to the apical bath, whereas Fig. 6B shows the [Ca2+]in changes after multiple additions of 100 μM ATP to the basal bath. In six experiments, we found no significant difference between ATP- and UTP-induced rises in [Ca2+]in (15–25 μM). However, apical addition of agonist (ATP or UTP) produced a significantly larger increase in [Ca2+]in than basal addition (seediscussion). Apical ATP/UTP (15–25 μM) increased [Ca2+]in by 427 ± 206 nM from an initial value of 29 ± 12 nM (mean ± SD, n = 6), whereas basal addition (15–25 μM) increased [Ca2+]in by 158 ± 61 nM from an initial value of 45 ± 12 nM (mean ± SD,n = 3). Basal addition of 100 μM ATP increased [Ca2+]in by 268 ± 96 nM from a baseline of 35 ± 7 nM (n = 2).

Fig. 6.

Fig. 6.A: apical ATP-induced elevation of intracellular Ca2+ concentration ([Ca2+]in) of 31EG4 mammary monolayer (25 μM ATP). B: basal ATP-induced elevation of [Ca2+]in (100 μM).

Figure 7A shows that basal H2DIDS, a nonfluorescent DIDS analog, blocked the UTP-induced increase in Ca2+, presumably by blocking the P2 receptor (41). The control response shows that 25 μM UTP, added to the basal bath, caused a 6.5-fold increase in free [Ca2+]in, from 25 to 130 nM. After the addition of H2DIDS (500 μM) to the basal bath, basal UTP had no effect on cell Ca2+. In other experiments, suramin produced similar results (data not shown; n = 3). Figure 7B compares the ATP-induced rise in cell Ca2+ in the absence and presence of apical H2DIDS (100 μM). Addition of 25 μM ATP to the apical bath caused a 600 nM increase in cell Ca2+. In the presence of H2DIDS, the ATP-induced rise in [Ca2+]in was practically abolished.

Fig. 7.

Fig. 7.A: effects of basal DIDS (500 μM) on UTP-induced increase in [Ca2+]in (25 μM UTP). B: effects of apical DIDS (100 μM) on ATP-induced increases in [Ca2+]in (25 μM ATP).

If the ATP- and/or UTP-induced changes in cell Ca2+ are the result of IP3-induced release from the ER stores, then the purinoceptor-linked increase in [Ca2+]in should be inhibited by depletion of these stores. Cyclopiazonic acid (CPA), a specific inhibitor of endoplasmic reticulum (ER) Ca2+-ATPase, was added to the apical bath to block Ca2+ uptake without affecting Ca2+ leakage out of the stores (9, 33). Figure8A shows that apical ATP (25 μM) almost doubled the fura 2 ratio (from 2 to 3.8). After ATP was washed out, 5 μM CPA was added to the apical bath and caused a transient increase in cell Ca2+ followed by a decrease to a sustained, elevated plateau. In the presence of CPA, the ATP-induced rise in cell Ca2+ was abolished. The electrical responses (Fig. 8B) followed the same time course as the Ca2+ changes; ATP addition increased the TEP (1 mV) and concomitantly decreased RT (∼25 Ω · cm2). These electrical responses were also abolished in the presence of CPA.

Fig. 8.

Fig. 8.Blockade of ATP-induced changes in [Ca2+]in (given as fura 2 ratio, F340/380), TEP, and RT by treatment with cyclopiazonic acid (CPA; 5 μM) (25 μM ATP). A: [Ca2+]in; B: TEP,RT.

Figure 9 shows that the addition of 25 μM ATP to the apical bath increased [Ca2+]in from 30 to 350 nM. The subsequent addition to the basal bath produced a much smaller response from 27 to 50 nM, as expected for a basal vs. apical and second vs. first pulse. The addition of 2.5 μM CPA caused a small, slow increase in steady-state [Ca2+]in compared with 5 μM CPA (Fig. 8). In the presence of CPA, apical ATP only increased [Ca2+]in from ∼50 to 75 nM, whereas basal ATP had no effect on [Ca2+]in. These experiments strongly suggest that the UTP- and/or ATP-induced increases in [Ca2+]in originate from the ER stores and mediate the membrane voltage and resistance changes (TEP andRT). If the ATP- and/or UTP-induced increases in cell Ca2+ originate from the ER stores, then these increases should also be blocked by BAPTA-AM, a cell-permeable Ca2+ chelator (52), and xestospongin C, an IP3 receptor blocker (21). Figure 10,A and B, shows that BAPTA-AM (n = 3 tissues) and xestospongin C (n = 1) almost completely block the apical ATP-induced rise in [Ca2+]in.

Fig. 9.

Fig. 9.CPA (2.5 μM) blocks apical or basal ATP-induced changes in [Ca2+]in (100 μM ATP).

Fig. 10.

Fig. 10.A: block of apical ATP-induced changes in [Ca2+]in by treatment with BAPTA-AM (50 μM) and 25 μM ATP. B: blockade of apical ATP-induced changes in [Ca2+]in by treatment with xestospongin C (10 μM) (25 μM ATP).

Purinoceptor receptor subtype.

ATP hydrolysis products such as adenosine may have produced the ATP-induced changes observed in 31EG4 cells. However, the addition of 50 or 100 μM adenosine to either the apical or basal bath produced no change in membrane voltage or resistance (n = 8; data not shown). By measuring agonist-induced changes in TEP, we examined the potency of several different concentrations of ATP or UTP at each membrane (Fig. 11). We also tested the efficacy of several other P2Y purinoceptor agonists (ADP, UDP, 2-MeSATP, and α,β-MeATP) on each side of the epithelium. The selected agonists had the following rank in potency at all concentrations: ATP = UTP; ATP > > > > > 2-MeSATP; and ADP, UDP, or α,β-MeATP were without effect, as expected for the P2Y2 receptor subtype. We fit our ΔTEP/concentration data using first-order kinetics to estimate EC50, but full saturation was not achieved at the highest concentration tested. The approximate EC50 for ATP and UTP is ∼24 ± 4 μM (± SE) for the apical membrane and ∼30 ± 7 μM for the basolateral membrane.

Fig. 11.

Fig. 11.Dose-response curves; effects of ATP, UTP, ADP, UDP, α,β-methylene-ATP (α,β-MeATP), and 2-methylthio-ATP (2-MeSATP) on TEP. A: apically applied agonists.B: basolaterally applied agonists. The rank order of affinity to the putative purinoceptor receptor (apical or basolateral) is as follows: ATP = UTP, > > > > 2-MeSATP and not ADP, UDP, or α,β-MeATP.

The data summarized in Fig. 11 show that the addition of ADP to the apical or basal baths produced practically no TEP orRT (data not shown) responses, suggesting that P2Y1 purinoceptors do not significantly contribute to the ATP/UTP responses. This notion was tested further using ADPβS, a P2Y1 purinoceptor agonist (43). Figure12A shows that, compared with the same concentration of ATP (25 μM), apical ADPβS produces a 16-fold smaller increase in cell Ca2+ (n = 3). In addition, apical ADPβS produced no change inJV. Figure 12B shows that basal application of ADPβS produces no change in cell Ca2+(n = 2). In addition, the rise in [Ca2+]in induced by 25 μM apical ATP (Fig.6A) was completely unaffected by the presence of P2Y1 blockers [MRS2179 −10 μM; A3P5P −100 μM (43)] (not shown). The ATP- and/or UTP-induced changes in cell physiology could also have been induced by activation of P2X receptors also located at the apical or basolateral membranes. Figure13 shows that 25 μM apical BzBz-ATP, a P2X agonist (43), caused a small increase in cell Ca2+ relative to the increase caused by an equal concentration of ATP (n = 2). At higher concentrations (50 μM; n = 2), the magnitude of the BzBz-ATP-induced increases in [Ca2+]in were comparable to those of ATP but produced no change in JV. On the basal side of the cell, 100 μM BzBz-ATP elicited no change in cell Ca2+ (data not shown).

Fig. 12.

Fig. 12.A: effect of apical ADPβS (25 μM) compared with ATP (25 μM) on [Ca2+]in.B: basal ATP-induced elevation of [Ca2+]in (100 μM) compared with ADPβS (100 μM).

Fig. 13.

Fig. 13.Apical benzoylbenzoyl-ATP (BzBz-ATP)-induced elevation of [Ca2+]in of 31EG4 mammary monolayer (25 μM ATP, 25 μM BzBz-ATP).

Agonist-induced changes on fluid transport.

JV, TEP, and RT were first measured under control conditions and then after the addition of ATP/UTP to the apical or basal baths. Figure14 shows that apical or basal ATP induced fluid secretion, increased TEP, and decreasedRT. Figure 15summarizes the results from 10 cultures that exhibited baseline secretion or absorption that ranged from −5.0 to 2.0 μl · cm−2 · h−1. In 10 cultures, the apical or basal ATP-induced increase in secretion ranged from −2.0 to −11.0 μl · cm−2 · h−1; for apical addition, the mean increase in secretion was 6.2 ± 2.4 μl · cm−2 · h−1(mean ± SD; n = 6; P < 0.001), and for basal addition the mean increase was 5.6 ± 3.1 μl · cm−2 · h−1(n = 4, P < 0.02).

Fig. 14.

Fig. 14.A: apical ATP-induced alteration in steady-state fluid absorption. Top: rate of transepithelial fluid flow (JV) in μl · cm−2 · h−1; positive values of JV indicate net fluid absorption and negative values fluid secretion. Arrows indicate that the probes have been moved away from the fluid surface during solution composition change. In these time periods, JVwas arbitrarily set to 0. Bottom: continuous trace of TEP and RT. B: same as Aexcept that ATP was added to the basal bath (50 μM ATP).

Fig. 15.

Fig. 15.Summary of fluid transport experiments with apical (n = 6) or basal (n = 4) addition of ATP (50 μM). The ATP-induced increase in fluid secretion ranged from 2.0 to 10 μl · cm−2 · h−1.

Figure 16 shows that the ATP-induced changes in TEP, RT, andJV were reduced by 50–75% in the presence of apical DIDS (500 μM). In two other monolayers, inhibition in all three parameters was even greater (data not shown). These experiments show that activation of apical or basolateral P2 purinoceptors activates fluid secretion across 31EG4 monolayers and that the apical membrane mechanisms(s) is DIDS sensitive.

Fig. 16.

Fig. 16.Apical UTP-induced increase in fluid secretion partially blocked by apical DIDS (50 μM UTP). Arrows (and horizontal dashed line) indicate that the probes have been moved away from the fluid surface during solution composition changes. In these time periods,JV was arbitrarily set to 0.


Purinoceptor-mediated ion and fluid transport in 31EG4 epithelial cells.

Addition of micromolar amounts of ATP/UTP to either the apical or basal baths of mammary epithelia caused an increase in free [Ca2+]in followed by large voltage and resistance changes at the apical membrane and fluid secretion across the epithelium. These agonist-induced changes are best explained in terms of purinoceptors located on the apical or basolateral membranes. Metabotropic purinoceptors increase [Ca2+]inby phospholipase Cβ activation and IP3 formation (43, 57). In the present study, prerelease of ER Ca2+ stores using CPA, or treatment with BAPTA or xestospongin C greatly reduced the ATP- and/or UTP-evoked Ca2+ and electrical responses (Figs. 8-10), providing a link between the activation of plasma membrane purinoceptors, the ER-mediated increase in [Ca2+]in, and plasma membrane voltage and resistance changes. The intracellular data (Table1 and Figs. 2 and 3) strongly suggest that ATP/UTP increases apical membrane conductance and activates DIDS-inhibitable, Ca2+-activated Cl channels in the apical membrane. Basal DIDS (Fig. 4A) or suramin blocked the electrical responses produced by basal UTP, indicating that changes in apical membrane voltage and resistance can be mediated by activation of P2 purinoceptors on the basolateral membrane.

Figure 1 shows that 31EG4 mammary epithelial cells contain the message for P2Y2 receptors. These receptors are equally sensitive to UTP and ATP but not sensitive to 2-MeSATP, ADP, UDP, and α,β-MeATP and are suramin and DIDS inhibitable (4, 24, 32,43). The ATP/UTP-sensitive receptors that we have examined fit all of these characteristics but not those of other P2 purinoceptors, as confirmed in Figs. 2-5, 7, and 11. The purinoceptors have an approximate EC50 of ≈24 μM for both UTP and ATP at the apical membrane and ≈30 μM at the basolateral membrane. Because the apical and basolateral membrane electrophysiological responses to ATP, UTP, 2-MeSATP, ADP, UDP, and α,β-MeATP are practically identical, it appears that the receptor subtype is the same on both membranes. The basal bath agonist-induced Ca2+ responses are smaller than those produced by apical addition of secretagogue, probably because of the relatively slow perfusion rate into the basal bath of the fluorescence chamber (materials and methods).

Previously, we showed that the 31EG4 mammary epithelial cells express CFTR and ENaC in the apical membrane and that both channels help determine apical membrane resting potential and net fluid transport across the monolayer (3). CFTR regulation by protein kinase C has been reported in various cell lines (10, 28,56) and may play a role in the movement of Cl in mammary epithelia. Addition of ATP to the basal bath did not evoke any electrical changes when preceded by the addition of DIDS to the apical bath to block the Ca2+-activated Cl channel (and apical purinoceptors; Fig. 4B). This result suggests that activation of basolateral metabotropic purinoceptors does not cause observable changes in apical membrane CFTR conductance.

The localization of these transport proteins and the separate pathways that mediate fluid absorption and secretion are shown in the model in Fig. 17. We have preliminary evidence for the presence of Ba2+-sensitive apical membrane K channels, as shown in other epithelia (1). Thus the apical membrane resting potential is determined by a combination of Cl, Na, and probably K channels. In addition, there is evidence for Na/H exchangers and Na-K-2Cl cotransporters on the basolateral membrane of mammary epithelia (48, 49). On the basis of present data, fluid absorption is most likely driven by active NaCl absorption (Na through the cell, Cl through the paracellular pathway driven by the TEP), and fluid secretion is driven by active (K + Na) Cl transport (KCl through the cell and Na through the paracellular pathway driven by the TEP). Net fluid transport is determined by the balance between these two pathways. For example, fluid secretion can be induced either by activating apical CFTR in the secretory pathway or by inhibiting apical ENaC in the absorption pathway (3). Fluid secretion is also induced (Fig. 14) by activating apical membrane Ca2+-dependent Cl channels via activation of P2Y2-purinoceptors at either the apical or basolateral membranes.

Fig. 17.

Fig. 17.Model of ion and fluid transport in 31EG4 mammary epithelia. Shown are apical and basolateral membrane mechanisms and second messengers involved in the purinoceptor-mediated alterations in cell Ca2+, membrane potential, and resistance. Also included are transport proteins identified in previous studies (see text). ENaC, epithelial Na+ channel; CFTR, cystic fibrosis transmembrane conductance regulator.

Mammary gland physiology and function.

The present data provide the first demonstration of functional purinoceptors on both membranes of the mammary epithelium. The presence of functional P2Y2 receptors in the apical and basolateral membranes of 31EG4 cells is consistent with previous studies showing the presence of purinoceptors in mammary cell lines and primary cultures (15, 17, 19, 20, 40). The activation of these receptors likely occurs in vivo. ATP has been found in bovine milk (14, 44), and ATP, UTP, and UDP are released from human mammary epithelia upon mechanical stimulation. In vivo, this mechanical stimulation is provided by myoepithelial cell contraction probably induced by ATP, perhaps released from acinar/duct cells, and oxytocin, released from the posterior pituitary (40). Activation of fluid secretion across the 31EG4 monolayer suggests that extracellular nucleotides could play a fundamental role in mammary gland paracrine signaling and the regulation of milk composition.

In addition, P2 purinoceptors may be involved in abnormal mammary cell growth (11). Several studies have shown the existence of metabotropic P2 purinoceptors on tumor cell lines (11, 13,26). It has also been observed that, at physiological concentrations, extracellular ATP can cause mitogenic activity in a variety of cell types (27) and can act as a comitogen in concert with other growth factors to enhance cellular proliferation in cell culture models (54). In MCF-7 breast cancer cells, ATP- and UTP-stimulated P2 purinoceptors were found to increase [Ca2+]in and induce cell proliferation (possibly by a K current-dependent mechanism; see Refs. 11and 34). Because breast cyst fluid often consists of several growth factors, it is possible that the abnormal accumulation of ATP in cyst fluid could cause cell proliferation over time.

ATP- and/or UTP-induced fluid secretion may also play a role in fibrocystic disease of the breast, a frequent mammary pathology in premenopausal women that is characterized by the abnormal accumulation of breast cystic fluid (5, 36). The ATP- and/or UTP-induced increases in fluid secretion (2.0–10 μl · cm−2 · h−1) reported here would be equivalent to the addition of ∼0.2–1.0 ml fluid/day for a large cyst (51). These cysts are characterized by two different electrolyte disturbances. Type I cysts (lined by epithelia of apocrine morphology) are characterized by relatively low concentrations of Na (∼41 meq/l) and Cl (∼15 meq/l) levels and high concentrations of K (∼101 meq/l). In contrast, type II or transudative cysts (lined with flattened epithelium) contain moderately high Na (∼140 meq/l) and Cl levels (∼90 meq/l) and relatively low K levels (∼9 meq/l; see Refs. 12 and 36).

Type I cysts could result from the activation of the Na absorption pathway, which consists of apical membrane ENaC and basolateral membrane Na/K pumps. Upregulation of Na-K-ATPase activity could drive K secretion into the lumen, perhaps mediated by an ATP-induced increase in Ca2+-activated apical membrane K conductance. Organic anion or bicarbonate transport may provide the accompanying anions needed to mediate fluid secretion. In type II cysts, abnormal levels of extracellular ATP could stimulate Ca2+-activated Cl channels and help drive fluid into the luminal space. The counterion for this fluid secretion could be provided by Ca2+-activated apical membrane K channels or cation movement through the paracellular path, driven by the ATP-induced increase in TEP (Fig. 2). Therefore, blockade of apical membrane K channels may help prevent type I cysts, whereas purinoceptor blockade may be therapeutically advantageous against type II cyst formation.


  • *S. Blaug and J. Rymer contributed equally to this work.


  • Address for reprint requests and other correspondence: S. S. Miller, National Eye Institute, National Institutes of Health, Bldg. 31,6A20, 31 Center Dr., Bethesda, MD 20892-2510 (E-mail:).

  • 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.

  • First published November 27, 2002;10.1152/ajpcell.00238.2002


  • 1 Bialek S, Miller SS.K+ and Cl transport mechanisms in bovine pigment epithelium that could modulate subretinal space volume and composition.J Physiol4751994401417
  • 2 Bialek S, Quong J, Yu K, Miller SS.Nonsteroidal anti-inflammatory drugs alter chloride and fluid transport in bovine retinal pigment epithelium.Am J Physiol Cell Physiol2701996C1175C1189
  • 3 Blaug S, Hybiske K, Cohn J, Firestone GL, Machen TE, Miller SS.ENaC- and CFTR-dependent ion and fluid transport in mammary epithelia.Am J Physiol Cell Physiol2812001C633C648
  • 4 Boarder MR, Hourani SM.The regulation of vascular function by P2 receptors: multiple sites and multiple receptors.Trends Pharmacol Sci19199899107
  • 5 Bodian CA.Benign breast diseases, carcinoma in situ, and breast cancer risk.Epidemiol Rev151993177187
  • 6 Bouyer P, Paulais M, Cougnon M, Hulin P, Anagnostopoulos T, Planelles G.Extracellular ATP raises cytosolic calcium and activates basolateral chloride conductance in Necturus proximal tubule.J Physiol5101998535548
  • 7 Brake AJ, Julius D.Signaling by extracellular nucleotides.Annu Rev Cell Dev Biol121996519541
  • 8 Christoffersen BC, Hug MJ, Novak I.Different purinergic receptors lead to intracellular calcium increases in pancreatic ducts.Pflügers Arch43619983339
  • 9 Darby PJ, Kwan CY, Daniel EE.Use of calcium pump inhibitors in the study of calcium regulation in smooth muscle.Biol Signals21993293304
  • 10 Dechecchi MC, Rolfini R, Tamanini A, Gamberi C, Berton G, Cabrini G.Effect of modulation of protein kinase C on the cAMP-dependent chloride conductance in T84 cells.FEBS Lett31119922528
  • 11 Dixon JM, McDonald C, Elton RA, Miller WR.Breast cancer risk with cyst type in cystic disease of the breast. Larger study found no association between cyst type and breast cancer.Br Med J3151997545546
  • 12 Dixon JM, McDonald C, Elton RA, Miller WR.Risk of breast cancer in women with palpable breast cysts: a prospective study. Edinburgh Breast Group.Lancet353199917421745
  • 13 Dubyak GR.Signal transduction by P2-purinergic receptors for extracellular ATP.Am J Respir Cell Mol Biol41991295300
  • 14 Emanuelson U, Olsson T, Mattila T, Astrom G, Holmberg O.Effects of parity and stage of lactation on adenosine triphosphate, somatic cell count and antitrypsin content in cows' milk.J Dairy Res5519884955
  • 15 Enomoto K, Furuya K, Moore RC, Yamagishi S, Oka T, Maeno T.Expression cloning and signal transduction pathway of P2U receptor in mammary tumor cells.Biol Signals51996921
  • 16 Enomoto K, Furuya K, Yamagishi S, Oka T, Maeno T.The increase in the intracellular Ca2+ concentration induced by mechanical stimulation is propagated via release of pyrophosphorylated nucleotides in mammary epithelial cells.Pflügers Arch4271994533542
  • 17 Flezar M, Heisler S.P2-purinergic receptors in human breast tumor cells coupling of intracellular calcium signaling to anion secretion.J Pharmacol Exp Ther265199314991510
  • 18 Fulford GR, Blake JR.Mucociliary transport in the lung.J Theor Biol1211986381402
  • 19 Furuya K, Enomoto K, Yamagishi S.Spontaneous calcium oscillations and mechanically and chemically induced calcium responses in mammary epithelial cells.Pflügers Arch4221993295304
  • 20 Furuya K, Enomoto K, Nakano H, Yamagishi S.Purinergic and mechanical interactions between myo- and secretory epithelial cells in mammary gland.Jpn J Physiol Suppl11997S62
  • 21 Gafni J, Munsch JA, Lam TH, Catlin MC, Costa LG, Molinski TF, Pessah IN.Xestospongins: potent membrane permeable blockers of the inositol 1,4,5-trisphosphate receptor.Neuron191997723733
  • 22 Gobran LI, Xu ZX, Lu Z, Rooney SA.P2u purinoceptor stimulation of surfactant secretion coupled to phosphatidylcholine hydrolysis in type II cells.Am J Physiol Lung Cell Mol Physiol2671994L625L633
  • 23 Grobben B, Claes P, Van Kolen K, Roymans D, Fransen P, Sys SU, Slegers H.Agonists of the P2Y(AC)-receptor activate MAP kinase by a ras-independent pathway in rat C6 glioma.J Neurochem78200113251338
  • 24 Harden TK, Lazarowski ER.Release of ATP and UTP from astrocytoma cells.Prog Brain Res1201999135143
  • 25 Hughes BA, Miller SS, Machen TE.Effects of cyclic AMP on fluid absorption and ion transport across frog retinal pigment epithelium. Measurements in the open-circuit state.J Gen Physiol831984875899
  • 26 Insel PA, Ostrom RS, Zambon AC, Hughes RJ, Balboa MA, Shehnaz D, Gregorian C, Torres B, Firestein BL, Xing M, Post SR.P2Y receptors of MDCK cells: epithelial cell regulation by extracellular nucleotides.Clin Exp Pharmacol Physiol282001351354
  • 27 Ishikawa S, Kawasumi M, Kusaka I, Komatsu N, Iwao N, Saito T.Extracellular ATP promotes cellular growth of glomerular mesangial cells mediated via phospholipase C.Biochem Biophys Res Commun2021994234240
  • 28 Jia Y, Mathews CJ, Hanrahan JW.Phosphorylation by protein kinase C is required for acute activation of cystic fibrosis transmembrane conductance regulator by protein kinase A.J Biol Chem272199749784984
  • 29 Jiang C, Finkbeiner WE, Widdicombe JH, McCray PB, Miller SS.Altered fluid transport across airway epithelium in cystic fibrosis.Science2621993424427
  • 30 Joseph DP, Miller SS.Apical and basal membrane ion transport mechanisms in bovine retinal pigment epithelium.J Physiol4351991439463
  • 31 Kenyon E, Yu K, La Cour M, Miller SS.Lactate transport mechanisms at apical and basolateral membranes of bovine retinal pigment epithelium.Am J Physiol Cell Physiol2671994C1561C1573
  • 32 King BF, Townsend-Nicholson A, Burnstock G.Metabotropic receptors for ATP and UTP: exploring the correspondence between native and recombinant nucleotide receptors.Trends Pharmacol Sci191998506514
  • 33 Kirischuk S, Scherer J, Kettenmann H, Verkhratsky A.Activation of P2-purinoreceptors triggered Ca2+ release from InsP3-sensitive internal stores in mammalian oligodendrocytes.J Physiol48319954157
  • 34 Klimatcheva E, Wonderlin WF.An ATP-sensitive K(+) current that regulates progression through early G1 phase of the cell cycle in MCF-7 human breast cancer cells.J Membr Biol17119993546
  • 35 Lin H, Kenyon E, Miller SS.Na-dependent pHi regulatory mechanisms in native human retinal pigment epithelium.Invest Ophthalmol Vis Sci33199235283538
  • 36 Malatesta M, Mannello F, Sebastiani M, Cardinali A, Marcheggiani F, Reno F, Gazzanelli G.Ultrastructural characterization and biochemical profile of human gross cystic breast disease.Breast Cancer Res Treat481998211219
  • 37 Mason SJ, Paradiso AM, Boucher RC.Regulation of transepithelial ion transport and intracellular calcium by extracellular ATP in human normal and cystic fibrosis airway epithelium.Br J Pharmacol103199116491656
  • 38 Miller SS, Steinberg RH.Passive ionic properties of frog retinal pigment epithelium.J Membr Biol361977337372
  • 39 Muraki K, Imaizumi Y, Watanabe M.Effects of UTP on membrane current and potential in rat aortic myocytes.Eur J Pharmacol3601998239247
  • 40 Nakano H, Furuya K, Furuya S, Yamagishi S.Involvement of P2-purinergic receptors in intracellular Ca2+ responses and the contraction of mammary myoepithelial cells.Pflügers Arch435199718
  • 41 Peterson W, Meggyesy C, Yu K, Miller SS.Extracellular ATP activates calcium signaling, ion and fluid transport in retinal pigment epithelium.J Neurosci17199723242337
  • 42 Quinn RH, Quong JN, Miller SS.Adrenergic receptor activated ion transport in human fetal retinal pigment epithelium.Invest Ophthalmol Vis Sci422001255264
  • 43 Ralevic V, Burnstock G.Receptors for purines and pyrimidines.Pharmacol Rev501998413492
  • 44 Richardson T, McGann TC, Kearney RD.Levels and location of adenosine 5′-triphosphate in bovine milk.J Dairy Res4719809196
  • 45 Sabater JR, Mao YM, Shaffer C, James MK, O'Riordan TG, Abraham WM.Aerosolization of P2Y(2)-receptor agonists enhances mucociliary clearance in sheep.J Appl Physiol87199921912196
  • 46 Schlenker T, Romac JM, Sharara AI, Roman RM, Kim SJ, LaRusso N, Liddle RA, Fitz JG.Regulation of biliary secretion through apical purinergic receptors in cultured rat cholangiocytes.Am J Physiol Gastrointest Liver Physiol2731997G1108G1117
  • 47 Schultz BD, Singh AK, Devor DC, Bridges RJ.Pharmacology of CFTR chloride channel activity.Physiol Rev Suppl791999109144
  • 48 Shennan DB, Peaker M.Transport of milk constituents by the mammary gland.Physiol Rev802000925951
  • 49 Sjaastad MD, Zettl KS, Parry G, Firestone GL, Machen TE.Hormonal regulation of the polarized function and distribution of Na/H exchange and Na/HCO3 cotransport in cultured mammary epithelial cells.J Cell Biol1221993589600
  • 50 Stutts MJ, Fitz JG, Paradiso AM, Boucher RC.Multiple modes of regulation of airway epithelial chloride secretion by extracellular ATP.Am J Physiol Cell Physiol2671994C1442C1451
  • 51 Takei H, Iino Y, Horiguchi J, Maemura M, Yokoe T, Koibuchi Y, Oyama T, Ohwada S, Morishita Y.Natural history of fibroadenomas based on the correlation between size and patient age.Japn J Clin Oncol291999810
  • 52 Taylor CW, Broad LM.Pharmacological analysis of intracellular Ca2+ signalling: problems and pitfalls.Trends Pharmacol Sci191998370375
  • 53 Viana F, de Smedt H, Droogmans G, Nilius B.Calcium signalling through nucleotide receptor P2Y2 in cultured human vascular endothelium.Cell Calcium241998117127
  • 54 Wang DJ, Huang NN, Heppel LA.Extracellular ATP and ADP stimulate proliferation of porcine aortic smooth muscle cells.J Cell Physiol1531992221233
  • 55 Watt WC, Lazarowski ER, Boucher RC.Cystic fibrosis transmembrane regulator-independent release of ATP. Its implications for the regulation of P2Y2 receptors in airway epithelia.J Biol Chem27319981405314058
  • 56 Winpenny JP, McAlroy HL, Gray MA, Argent BE.Protein kinase C regulates the magnitude and stability of CFTR currents in pancreatic duct cells.Am J Physiol Cell Physiol2681995C823C828
  • 57 Yerxa BR, Johnson FL.P2Y2 receptor agonists: structure, activity, and therapeutic utility.Drugs241999759769
  • 58 Zettl KS, Sjaastad MD, Riskin PM, Parry G, Machen TE, Firestone GL.Glucocorticoid-induced formation of tight junctions in mouse mammary epithelial cells in vitro.Proc Natl Acad Sci USA89199290699073

защита днища

медицинские препараты для повышения потенции у мужчин