Am J Physiol Cell Physiol 306: C230 –C240, 2014. First published November 20, 2013; doi:10.1152/ajpcell.00219.2013.

Functional expression of a Kir2.1-like inwardly rectifying potassium channel in mouse mammary secretory cells Akihiro Kamikawa and Toru Ishikawa Department of Basic Veterinary Medicine, Obihiro University of Agriculture and Veterinary Medicine, Obihiro, Japan Submitted 19 July 2013; accepted in final form 16 November 2013

Kamikawa A, Ishikawa T. Functional expression of a Kir2.1-like inwardly rectifying potassium channel in mouse mammary secretory cells. Am J Physiol Cell Physiol 306: C230 –C240, 2014. First published November 20, 2013; doi:10.1152/ajpcell.00219.2013.—K⫹ channels in mammary secretory (MS) cells are believed to play a role in transcellular electrolyte transport and thus determining ionic composition of the aqueous phase of milk. However, direct evidence for specific K⫹ channel activity in native MS cells is lacking at the single-cell level. Here, we show for the first time that an inwardly rectifying K⫹ (Kir) channel is functionally expressed in fully differentiated MS cells that were freshly isolated from the mammary gland of lactating mice. Using the standard whole cell patch-clamp technique, we found that mouse MS cells consistently displayed a K⫹ current, whose electrophysiological properties are similar to those previously reported for Kir2.x channels, particularly Kir2.1: 1) current-voltage relationship with strong inward rectification, 2) slope conductance approximately proportional to the square root of external K⫹ concentration, 3) voltage- and time-dependent and high-affinity block by external Ba2⫹, and 4) voltage-dependent inhibition by external Cs⫹. Accordingly, RT-PCR analysis revealed the gene expression of Kir2.1, but not Kir2.2, Kir2.3, and Kir2.4, in lactating mouse mammary gland, and immunohistochemical staining showed Kir2.1 protein expression in the secretory cells. Cell-attached patch recordings from MS cells revealed that a 31-pS K⫹ channel with strong inward rectification was likely active at the resting membrane potential. Collectively, the present work demonstrates that a functional Kir2.1-like channel is expressed in lactating mouse MS cells. We propose that the channel might be involved, at least in part, in secretion and/or preservation of ionic components of milk stored into the lumen of these cells. ion channel; Kir2.1; mammary secretory cell; patch-clamp; lactation MAMMARY GLAND is a unique exocrine gland, which develops after birth and exerts its physiological function exclusively during the period from parturition to weaning. In the prepregnant mammary gland of a nulliparous mammal, primary ducts spread with occasional bifurcations in the subcutaneous adipose tissue, and no acinar structures appear. After the onset of pregnancy, epithelia begin to proliferate and differentiate to form small ductal side branches and acini on their endings. During lactation, the secretory epithelia in both side branches and acini enlarge in size, produce milk fat globules inside the cells, and finally occupy the mammary gland instead of adipocytes (2). The mammary secretory (MS) cells secrete milk continuously and slowly into the lumens of glandular acini and/or ducts. Milk contains not only organic nutrients such as sugars, lipids, and proteins, but also electrolytes, including Na⫹, K⫹, and Cl⫺. The primary secretion is stored (and its

Address for reprint requests and other correspondence: A. Kamikawa, Dept. of Basic Veterinary Medicine, Obihiro Univ. of Agriculture and Veterinary Medicine, Inada-cho, Obihiro, Hokkaido 080-8555, Japan (e-mail: akami @obihiro.ac.jp). C230

ionic components may be reabsorbed and thus modified) in the lumens until the sucking juvenile removes it. Although there are some variations among species in the contents of Na⫹, K⫹, and Cl⫺ in the aqueous phase of milk, especially depending upon the lactose content, these ions are believed to be transported across the mammary secretory epithelium at least in part through the transcellular pathway involving the coordinated activities of ion channels and transporters in the basolateral and apical membranes (23, 26, 38). K⫹ channels play a vital role in diverse cellular functions including epithelial fluid and electrolyte transport primarily by regulating the membrane potential (27, 36, 45). Previous studies in the lactating mammary gland, at least in some species, have shown that the basolateral membranes of MS cells not only express Na⫹-K⫹-ATPase (16), but are also permeable to K⫹, suggesting the expression of K⫹ channels (22). Some MS cells also display a loop-diuretic-sensitive, basolateral Na⫹K⫹-2Cl⫺ symport that accumulates Cl⫺ into the cytosol above the predicted electrochemical equilibrium (37). Therefore, K⫹ channels at the basolateral membranes of MS cells would not only mediate a pathway, through which K ions transported into the cells by the Na⫹-K⫹-ATPase and Na⫹-K⫹-2Cl⫺ symporter are recycled, but would also hyperpolarize the membrane potential to provide the driving force for the apical Cl⫺ efflux into the lumen. Likewise, K⫹ channels in the apical membrane would be supportive for apical Cl⫺ secretion and for apical K⫹ secretion and Na⫹ absorption that might occur in mammary epithelial cells (3, 31). In line with these potential roles of K⫹ channels, previous patch-clamp studies have identified a Ca2⫹activated K⫹ current in mouse mammary secretory epithelial cells (of acinar cell origin) in primary culture (6, 8). The K⫹ current was blocked by external charybdotoxin, but not by apamin (6), and at the single-channel level it had a unitary conductance of 40 pS with a high-K⫹ pipette solution and was blocked by internal TEA and Ba2⫹, and its gating was voltage independent (8). Therefore, some of these properties appeared to be similar to those reported for cloned KCa3.1 (5, 13, 15). A KCa3.1-like, Ca2⫹-activated K⫹ channel has been also shown to be present in both the basolateral and apical membranes of an immortalized cell line derived from human mammary epithelial cells (21, 31), which are supposed to have properties of ductal epithelial cell (21). Nevertheless, it remains unknown whether other types of K⫹ channel are expressed in mammary epithelial cells. Furthermore, little information is available particularly about functional K⫹ channel in native MS cells that might be constitutively active and thus support the continuous and slow transport of salts and water in mammary gland. In this work, we used the whole cell patch-clamp technique to show that an inwardly rectifying K⫹ (Kir) conductance, whose electrophysiological properties match those reported for

0363-6143/14 Copyright © 2014 the American Physiological Society

http://www.ajpcell.org

Kir2.1-LIKE CHANNEL IN MOUSE MAMMARY SECRETORY CELL

C231

Kir2.x, especially Kir2.1, is present in fully differentiated secretory cells freshly isolated from mammary glands of lactating mice. We also used RT-PCR, Western blot, and immunohistochemical staining to reveal that Kir2.1 is a molecular candidate that may encode the K⫹ conductance. Moreover, we performed single-channel recordings in the cell-attached patch configuration to show that a 31-pS K⫹ channel possibly underlying the whole cell Kir conductance is likely active at the resting membrane potential of unstimulated, intact MS cells. Finally, we examined the effect of pharmacological inhibition of Kir currents on the cell potentials of MS cells with whole cell current clamp. Our work provides the first evidence at the single-cell level for functional expression of Kir2.1-like K⫹ channels in native MS cells and proposes possible roles of the channel in lactation. MATERIALS AND METHODS

Animals. Experimental procedures and animal care were carried out in accordance with the guidelines of animal care and use from Obihiro University of Agriculture and Veterinary Medicine and were approved by the university committee for the care and use of laboratory animals. Female and male C57BL/6J mice obtained from Nihon SLC (Shizuoka, Japan) and their offspring were used for the experiment. The mice were housed at 22°C with a 12:12-h light-dark cycle and given food and water ad libitum. Cell preparation for patch clamp. To obtain the lactating mammary epithelial cell, nulliparous female mice (10 –17 wk old) were mated with male mice, and the postpartum mice were allowed to raise litters until the day of the cell preparation. The averaged litter size was 7.5 ⫾ 0.2 (ranging from 3 to 11, n ⫽ 64) in this experiment. Dams were killed by cervical dislocation at days 11–18 of lactation (15.4 ⫾ 0.2, n ⫽ 64) and their abdominal and/or thoracic mammary glands were collected. The mammary glands were minced with scissors in a digestion buffer, which is a nominally divalent cation-free standard bath solution containing (in mM) 145 NaCl, 5 KCl, 10 HEPES, 10 glucose, and 4.6 NaOH at pH 7.4 supplemented with collagenase (type I, 300 U/ml; Wako, Osaka, Japan) and hyaluronidase (100 U/ml; Sigma-Aldrich, St.Louis, MO). The minced tissue in the digestion buffer was gassed with 100% O2 and incubated for 30 min at 37°C in a shaking water bath. After gentle trituration with a pipette, the tissue was incubated once more for 30 min in the fresh digestion buffer. The digested tissue was filtered through 100-␮m nylon mesh, then centrifuged (40 g, 5 min) and washed three times with the divalent cation-free standard bath solution. The filtrate that contained isolated single epithelial cells and small clumps of the cells with a diameter of ⬃25 ␮m (Fig. 1, A and B) was resuspended in the divalent cation-free standard bath solution and stored at 4°C. These cells, which could not be obtained from prepregnant mammary glands (data not shown), were designated as mammary secretory (MS) cells in the present study. Patch-clamp experiment. The cell preparations described above were used for the patch-clamp analyses within 8 h after isolation. The cells were plated out onto a noncoated borosilicate cover glass (Matsunami Glass Industry, Osaka, Japan) in a chamber mounted on an inverted microscope and were allowed to attach to the glass for a few minutes. Attached cells were initially superfused with the standard bath solution (in mM): 145 NaCl, 5 KCl, 10 HEPES, 10 glucose, 1 CaCl2, 1 MgCl2 and 4.6 NaOH at pH 7.4. Current recordings were made with an EPC7 amplifier (HEKA Electronik, Lambrecht, Germany) in the whole cell and cell-attached configurations of the patch-clamp technique (9). Patch-clamp pipettes were pulled from glass capillaries (G-1.5; Narishige, Tokyo, Japan) using a vertical puller (model PP-830; Narishige) so as to have resistances of 5–10 M⍀ when filled with the standard pipette solution (in mM): 110 K-glutamate, 10 KCl, 10 HEPES, 10 EGTA, 10 glucose, 1 MgCl2

Fig. 1. Mammary secretory cells freshly isolated from lactating mice and their membrane currents. A and B: typical cell clusters and single cells isolated from lactating mammary gland. Mammary glands collected from mice at days 11–18 of lactation were enzymatically digested, and the resultant cell suspension contained cell clusters (A) and single cells (B). Single cells were used for subsequent patch-clamp analysis. A patch pipette is also shown (B). White bars indicate 100 ␮m. C and D: representative traces of whole cell currents recorded from a single mouse mammary secretory (MS) cell. The currents were measured with the pipette solution containing 10⫺7 M free Ca2⫹ and the standard bathing solution containing 5 mM K⫹. The membrane potential was held at ⫺48 mV at resting. Whole cell currents were evoked by 800-ms voltage ramps from ⫺108 to ⫹48 mV (C) and by 400-ms voltage steps from ⫺128 to ⫹42 mV at 10-mV intervals (D), as indicated.

(0.51 free Mg2⫹ calculated with WEBMAXC Standard), and 25 KOH at pH 7.4. In some whole cell experiments, the following pipette solutions were also used: an acidic pipette solution containing (in mM): 110 K-glutamate, 10 KCl, 10 MES, 10 EGTA, 10 glucose, 0.54 MgCl2 (0.51 free Mg2⫹), and 22 KOH at pH 6.2; and a Ca2⫹ pipette solution containing 110 K-glutamate, 10 KCl, 10 HEPES, 10 EGTA, 10 glucose, 6.03 CaCl2 (0.1 ␮M free Ca2⫹), 0.71 MgCl2 (0.51 free Mg2⫹), and 37 KOH at pH 7.4. A 30 mM K⫹-containing pipette solution for single-channel recordings was prepared by mixing appropriately the standard pipette solution and a K⫹-free pipette solution [110 Na-glutamate, 10 NaCl, 10 HEPES, 10 EGTA, 10 glucose, 1 MgCl2 (0.51 free Mg2⫹) and 28 NaOH at pH 7.4]. Bath solutions containing 30 and 90 mM K⫹ were prepared by mixing the standard bath solution and a high K⫹ (154 mM)-containing bath solution (150 KCl, 10 HEPES, 10 glucose, 1 CaCl2, 1 MgCl2 and 4.2 KOH at pH 7.4) in the appropriate ratio. For the pH-modified bath solution, MES (for pH 6.6) or Tris (for pH 8.2) was used instead of HEPES. The reference electrode was an Ag-AgCl electrode, which was connected to the bathing solution via an agar bridge filled with the standard bath solution described above. The whole cell and single-channel currents were filtered at 1 kHz and 500 Hz, respectively, with an internal four-pole Bessel filter, sampled at 2 kHz and stored directly on the computer’s hard disk through the PowerLab (AD Instruments, Sydney, Australia). For whole cell current measurements, the amplifier was driven by Scope

AJP-Cell Physiol • doi:10.1152/ajpcell.00219.2013 • www.ajpcell.org

C232

Kir2.1-LIKE CHANNEL IN MOUSE MAMMARY SECRETORY CELL

software (AD Instruments) to allow the delivery of voltage-step or voltage-ramp protocol with concomitant digitization of the current. Membrane potential was held at ⫺40 mV at resting. Current-voltage (I-V) relations were studied using a 400-ms step pulse (commanded from ⫺120 mV to ⫹20 or ⫹50 mV, with 10-mV and 3-s intervals) or a 800-ms ramp pulse (commanded from ⫺100 mV to ⫹50 mV). In experiments, where the effects of extracellular Ba2⫹ were examined, 2.5-s step pulses at ⫺100 mV from the resting potential at 0 mV were commanded. The capacitance transient current was compensated. The cell capacitance was 21.8 ⫾ 0.5 pF (n ⫽ 84). The average series resistance (Rs), which was 18.8 ⫾ 0.6 M⍀ (n ⫽ 84), was not electrically compensated because the high-level (typically more than 50%) electrical compensation caused instability of the monitored current in our experimental condition. In the figures showing the I-V relationships (see Figs. 1– 4), the voltages were not corrected for the voltage decrease across the Rs. Therefore, the conductances of currents in the nanoampere range were seemingly underestimated as a result of this error. However, when reversal potentials and slope conductances were estimated from the slope of I-V relationships (i.e., Fig. 2D), the actual membrane voltages were determined by subtracting the dropped voltage from commanded voltage. For data analysis and/or graphical display, the digital data obtained from whole cell patch-clamp analysis was processed with the spreadsheet software (Microsoft Excel; Microsoft, Redmond, WA) and the data analysis software (Igor Pro; WaveMetrics, Lake Oswego, OR). Single-channel data were acquired with LabChart software (AD Instruments), and the amplitude of unitary current was measured manually. Traces of single-channel currents were plotted in graph with Igor Pro software after being processed with Excel software. Zero-current voltages were measured with current-clamp mode in whole cell configuration, and the voltage data were filtered at 500 Hz and sampled at 1 kHz. Obtained data were processed with Igor Pro and Excel software. The pipette potential was corrected for the liquid junction potential between the pipette solution and the external solution, and between the external solution and the agar bridge, as described elsewhere (29). BaCl2 (dissolved in pure water at 1 M), CsCl (dissolved in pure water at 1 M), and clotrimazole (dissolved in DMSO at 1 mM) were appropriately diluted by bath solution containing 5, 30, 90, or 154 mM K⫹. When TEA-Cl (1 mM) was used, equimolar NaCl (1 mM) was removed from the bath solution. Chemicals employed were of reagent grade. TEA-Cl, clotrimazole, HEPES, MES, Tris, and EGTA were obtained from Sigma-Aldrich. All experiments were performed at room temperature. Bath solution changes were accomplished by gravity feed from reservoirs. The results are reported as means ⫾ SE of independent experiments (n), where n refers to the number of cells patched. Data analysis for Ba2⫹ block. The effect of Ba2⫹ (10⫺3 and 10⫺4, ⫺5 10 , 10⫺6, or 10⫺7 M) was examined on whole cell currents recorded at ⫺108 mV in a 2.5-s step pulse in a 90 mM K⫹ bath solution. The unblocked fractional current (If) was calculated as:

If ⫽ 共Iss,Ba ⫺ Iss,back兲 ⁄ 共Iss,Pre ⫺ Iss,back兲

(1)

where Iss,Pre and Iss,Ba are the steady-state currents before and after the addition of Ba2⫹ (10⫺4–10⫺7 M) to the bath solution, respectively, and Iss,back is the steady-state background current in the presence of 10⫺3 M Ba2⫹. The mean values of If were fitted with the Hill equation: If ⫽ 1 ⁄ 兵1 ⫹ 共关Ba2⫹兴 ⁄ Kd兲h其

(2)

2⫹

where h is the Hill coefficient, [Ba ] is the molar concentration of Ba2⫹ applied, and Kd is the concentration at which one-half of the current is blocked. Analysis of mRNA expression. Abdominal mammary glands were collected from female C57BL/6J mice (10 –18 wk old) at prepregnant and day 15 of lactation. After removal of lymph nodes in mammary glands, the mammary glands were immersed in RNAlater (Life Technologies, Carlsbad, CA) and stored at ⫺30°C until used. From the mammary glands homogenized by vigorous agitation with a zirconia bead (Micro Smash; TOMY, Tokyo, Japan), total RNA was extracted with TRIzol (Life Technologies) and purified with RNA clean-up kit (RNeasy; QIAGEN, Hilden, Germany) according to the manufacturer’s protocols. The cleaned-up RNA (2 ␮g) and 1 ␮g of oligo(dT) primer were treated at 70°C for 5 min and reverse transcribed using 200 units of Moloney murine leukemia virus reverse transcriptase (Promega, Madison, WI), 24 units of RNase inhibitor, and 12.5 nmol of dNTPs in a total volume of 25 ␮l at 42°C for 1 h. After heating at 94°C for 5 min, PCR amplification was performed with 0.625 units of Taq polymerase (Promega), 37.5 nmol MgCl2, 5 nmol of dNTPs, and 25 pmol of sense and antisense primers specific for the respective genes in a total volume of 25 ␮l. The sense and antisense primers used are listed in Table 1. Samples were subjected to 94°C for 2 min followed by 30 cycles of 30 s at 94°C, 30 s at 57°C, and 30 or 60 s at 72°C, and finally incubated at 72°C for 5 min. PCR products were electrophoresed on 2% agarose gels containing ethidium bromide. Immunohistochemical analysis. Abdominal mammary glands collected from female C57BL/6J mice (12–18 wk old) at day 15 of lactation were fixed in 4% paraformaldehyde, embedded in paraffin, and sliced into 5-␮m-thick slices. Using the immunostaining kit (Histofine Simple Stain Mouse MAX-PO [M]; Nichirei, Tokyo, Japan) and anti-Kir2.1 mouse monoclonal antibody (1:200 dilution in PBS; Abcam, Cambridge, UK), Kir2.1 in the sliced section was visualized. In brief, deparaffinized and rehydrated sections were heated in 1 mM EDTA (pH 8.0) solution at 121°C for 10 min for antigen retrieval, and incubated in methanol containing 3% H2O2 for 30 min at room temperature to inactivate the intrinsic peroxidase. According to the manufacturer’s instructions, the sections were sequentially incubated in the blocking solutions, the primary antibody (4° C, overnight), the peroxidase-conjugated secondary antibody, and diaminobenzidine as a substrate. The sections were counterstained with hematoxylin, dehydrated by ethanol, and

Fig. 2. Extracellular K⫹ dependence of Kir currents in MS cells. A and B: representative traces of whole cell currents recorded from MS cells dialyzed with the standard Ca2⫹-free pipette solution. The bathing solutions contained 5 and 90 mM K⫹. The membrane potential was held at ⫺49 mV for 5 mM [K⫹]o (or ⫺48 mV for 90 mM [K⫹]o) at resting. Whole cell currents were evoked by 800-ms voltage ramps from ⫺109 mV (or ⫺108 mV) to ⫹41 mV (or ⫹42 mV) (A) and 400-ms voltage steps from ⫺129 mV (or ⫺128 mV) to ⫹11 mV (or ⫹12 mV) at 10-mV and 3-s intervals (B). C: relationships between steady-state current densities and membrane potentials from MS cells in bathing solutions containing different [K⫹]o. Whole cell currents at 5, 30, 90, and 154 mM [K⫹]o were elicited by the voltage steps. The current density was defined as the steady-state whole cell current (measured at 380 ms of test pulse) normalized for cell capacitance. Each point represents the mean ⫾ SE (n ⫽ 22, 7, 9, and 6 for 5, 30, 90, and 154 mM [K⫹]o, respectively) D: log-log plot of the relative slope conductance (Grel) of the inward current as a function of [K⫹]o. Each point represents the mean ⫾ SE (n ⫽ 7, 9, and 6 for 30, 90, and 154 mM [K⫹]o, respectively) E and F: relationships between the current densities of Ba2⫹-sensitive whole cell currents and the membrane potentials at different [K⫹]o. The whole cell currents were measured with the standard pipette solution. The currents were elicited by voltage ramps as shown in Fig. 1C. The Ba2⫹-sensitive currents in each [K⫹]o were determined by subtracting the currents during application of Ba2⫹ (10⫺4 M) from those before its application in the same cells. Expanded scale figure for the same data is also shown (F). Each point represents the mean ⫾ SE (n ⫽ 5, 5, 6, and 7 for 5, 30, 90, and 154 mM [K⫹]o, respectively) G: semilogarithmic plot of the reversal potentials of the Ba2⫹-sensitive currents as a function of [K⫹]o. The reversal potentials were determined as x-intercepts of linear regressions of Ba2⫹-sensitive inward currents in E. The line indicates the relationship between EK and [K⫹]o predicted from the Nernst equation. AJP-Cell Physiol • doi:10.1152/ajpcell.00219.2013 • www.ajpcell.org

Kir2.1-LIKE CHANNEL IN MOUSE MAMMARY SECRETORY CELL

cleared by xylene. The mounted sections were microscopically observed. Western blot analysis. Abdominal and thoracic mammary glands of lactating C57BL/6J mice (18 –22 wk old, at days 13–17 of lactation) were enzymatically digested according to the above-mentioned protocol. The digested cells were resuspended in 0.25 M sucrose solution containing protease inhibitor cocktail (cOmplete; Roche, Manheim, Germany) and homogenized with Potter-Elvehjem grinder on ice. The homogenate was centrifuged at 9,000 g for 10 min at 4°C and the

C233

supernatant was additionally centrifuged at 100,000 g for 90 min at 4°C. The supernatant was collected as a cytosolic fraction and the precipitant was solubilized in a buffer for membrane protein (CellLyEx MP; TOYO B-Net, Tokyo, Japan) as a microsomal fraction. The protein concentrations of the cytosolic and microsomal fractions were determined by the method of Lowry et al. (25) using BSA as a standard. The proteins in the microsomal fraction were deglycosylated by incubation at 37°C for 20 h with 2.5 mU of glycopeptidase F (Takara Bio, Otsu, Japan). Aliquots of the lysate were separated by

AJP-Cell Physiol • doi:10.1152/ajpcell.00219.2013 • www.ajpcell.org

C234

Kir2.1-LIKE CHANNEL IN MOUSE MAMMARY SECRETORY CELL

Table 1. Primer for RT-PCR Target (Gene Name)

Accession No.

Primer Pair

Product Size (Amplified Region)

Kir2.1 (Kcnj2)

NM_008425.4

157 bp

Kir2.2 (Kcnj12)

NM_010603.5

Kir2.3 (Kcnj4)

NM_008427.4

Kir2.4 (Kcnj14)

NM_145963.2

␤-Actin (Actb)

NM_007393.3

5=-CAGTGTCTTGGGAATTCTCAC-3= 5=-ACCTTAGTAACTCAGCTGAC-3= 5=-TGCATCATTGACTCCTTCATGAT-3= 5=-TCATCAATCTCGTGCAAGATGGT-3= 5=-TTGTCCAGTCCATTGTG-3= 5=-GCTGTCCTCGTCGATTTCATG-3= 5=-TAAGTGACCTGTTCACCACATG-3= 5=-CAATCTCATGCACAATGGTAATG-3= 5=-CAGCTTCTTTGCAGCTCCTT-3= 5=-TCACCCACATAGGAGTCCTT-3=

SDS-PAGE and transferred on polyvinylidene difluoride (PVDF) membranes (Millipore, Bedford, MA). The membranes were incubated first in a blocking buffer [20 mM Tris/HCl (pH 7.4), 150 mM NaCl containing 0.1% Tween 20 and 5% (wt/vol) skimmed milk], then in the buffer containing the anti-Kir2.1 antibody (1:4,000 dilution in blocking buffer) at 4°C overnight. The bound antibody was detected with horseradish peroxidase (HRP)-linked anti-mouse immunoglobulin (1:4,000 dilution in blocking buffer; GE Healthcare, Little Chalfont, UK) and an enhanced chemiluminescence HRP Substrate (Wako) on X-ray film. RESULTS

Expression of strongly inwardly rectifying K⫹ currents in mouse MS cells. We used the standard whole cell patch-clamp technique to identify functional K⫹ channels in acutely dissociated, single mouse mammary secretory (MS) cells (Fig. 1, A and B). Figure 1, C and D, shows whole cell currents recorded from a representative single MS cell in the standard bath solution. When dialyzed with a pipette solution having 10⫺7 M free Ca2⫹ concentration ([Ca2⫹]), which lies in the range of basal free [Ca2⫹] level reported for mammary acinar cells (40), the cell exhibited membrane currents with slight inward rectification (Fig. 1, C and D). The current appeared to be composed of at least two different components, because whole cell dialysis with a pipette solution containing ⬍1 nM free [Ca2⫹] abolished a linear component at potentials more positive than about ⫺80 mV with little effect on an inwardly rectifying one at potentials more negative than ⫺80 mV (Fig. 2, A and B). Under these experimental conditions, the inwardly rectifying whole cell current was observed in all cells tested (n ⫽ 84), albeit with some variations in current density. When extracellular K⫹ concentration ([K⫹]o) was raised from 5 to 90 mM, the inward current increased and its reversal potential shifted toward the positive direction (Fig. 2, A and B). Figure 2C summarizes steady-state current-voltage (I-V) relationships of whole cell currents in bath solutions with different [K⫹]o and shows that inward current increased as [K⫹]o was elevated. The slope conductances (G) of the inward currents at 5, 30, 90, and 154 mM [K⫹]o, which were determined from the linear section of the I-V relationships, were 9.3 ⫾ 1.0, 15.7 ⫾ 6.2, 42.2 ⫾ 9.3, and 107.1 ⫾ 30.7 nS, respectively (n ⫽ 22, 7, 9, and 6, respectively). When examining the relationship between the values of G and [K⫹]o, G at 30, 90 or 154 mM [K⫹]o was normalized to that at 5 mM [K⫹]o measured in the same cells to minimize the error induced by variations of the current densities among each cells. The normalized slope conductances (Grel) at 30, 90, and 154 mM [K⫹]o were 2.04 ⫾ 0.21, 4.51 ⫾ 0.49, and 9.71 ⫾ 0.92, respectively (Fig. 2D). The plot

320 bp 347 bp 615 bp 219 bp

of the logarithm of Grel as a function of the logarithm of [K⫹]o revealed a rectilinear relation with a slope of 0.63, which was roughly proportional to the square root of [K⫹]o. The reversal potentials of the inward current estimated by extrapolation from the data lying in the range in which its I-V relation was linear were ⫺44.4 ⫾ 2.2 mV (5 mM [K⫹]o), ⫺20.0 ⫾ 3.9 mV (30 mM [K⫹]o), ⫺5.7 ⫾ 1.2 mV (90 mM [K⫹]o), and ⫹6.2 ⫾ 2.0 mV (154 mM [K⫹]o). Although these values, especially at 5 and 30 mM [K⫹]o, were deviated to some extent from equilibrium potentials for K⫹ (EK) (⫺86.5 mV, ⫺40.5 mV, ⫺12.3 mV, and ⫹1.5 mV at 5, 30, 90, and 154 mM [K⫹]o, respectively) predicted from the Nernst equation, such a deviation may be explained assuming that the membrane had a significant leak conductance and/or other ionic conductance (e.g., Na⫹ or Cl⫺). In our preliminary experiments, replacement of Na⫹ by impermeable monovalent cation (NMDG⫹) in the bath solution containing 5 mM K⫹ shifted the reversal potentials of the whole cell currents to the hyperpolarized direction as much as 15 mV, suggesting the presence of Na⫹ channels and/or nonselective cation channels in MS cells. Therefore, to ensure that the inward current was solely carried by K⫹ ions, we took advantage that it was highly sensitive to external Ba2⫹ (see Fig. 3, A–C), and subsequently estimated the reversal potentials of the Ba2⫹ (10⫺4 M)-sensitive currents at different [K⫹]o. (Fig. 2, E and F). The Ba2⫹-sensitive K⫹ current displayed a strong inward rectification, namely a large inward current at potentials negative to EK and a small outward current at potentials positive to EK (Fig. 2, E and F). The G of the Ba2⫹-sensitive inward current (GBa) at 5, 30, 90, and 154 mM [K⫹]o were 4.3 ⫾ 0.8, 26.1 ⫾ 14.2, 37.2 ⫾ 15.4, and 55.9 ⫾ 17.2 nS, respectively (n ⫽ 5, 5, 6, and 7, respectively). The relationship between GBa and [K⫹]o (GBa ⬀ [K⫹]o0.69) was similar to the relationship between Grel and [K⫹]o. As shown in Fig. 2G, the reversal potentials of the Ba2⫹-sensitive current were ⫺79.4 mV (5 mM [K⫹]o), ⫺41.0 mV (30 mM [K⫹]o), ⫺16.2 mV (90 mM [K⫹]o), and ⫺4.4 mV (154 mM [K⫹]o), the values being indeed in good agreement with the expected EK values. Collectively, these results indicate the functional expression of a strongly inwardly rectifying K⫹ channel (Kir) in mouse MS cells. Block by external Ba2⫹ and Cs⫹ of native Kir currents in mouse MS cells. As described above, biophysical properties of native Kir currents in mouse MS cells were similar to those of the current cloned and native Kir currents, especially mediated by Kir2 subfamily members (Kir2.x) (7, 14, 19). Since it is well known that Kir2.x channels are inhibited by extracellular Ba2⫹ in a manner that is dose-, voltage-, and time-dependent

AJP-Cell Physiol • doi:10.1152/ajpcell.00219.2013 • www.ajpcell.org

Kir2.1-LIKE CHANNEL IN MOUSE MAMMARY SECRETORY CELL

C235

Fig. 3. Effects of external Ba2⫹, Cs⫹, and TEA on Kir currents in MS cells. A: reversible inhibition of the Kir current by external Ba2⫹ in a representative MS cell. The whole cell currents were recorded with the standard pipette solution and the 90 mM [K⫹]o bath solution with or without 10⫺4 M Ba2⫹. Voltage-ramp protocol used was identical to that shown in Fig. 1C. B: dose- and time-dependent Ba2⫹ block of Kir currents in MS cells. The whole cell currents were recorded with the pipette and bath solutions as described in A in the absence (Pre) or presence of Ba2⫹ (10⫺3 M–10⫺7 M). The whole cell currents were elicited by 2.5-s voltage step to ⫺108 mV from the holding potential of ⫺8 mV. The currents were normalized by Iss,Pre (measured at 2.25 s of voltage-step, in the absence of Ba2⫹). C: unblocked fractional currents plotted against external Ba2⫹ concentrations. The unblocked fractional currents were determined with Eq. 1. The line is a fit to the Hill equation (see Eq. 2). Each point represents the mean ⫾ SE (n ⫽ 4, 3, 4, and 3 for 10⫺4, 10⫺5, 10⫺6, and 10⫺7 M Ba2⫹, respectively) D and E: effects of external Cs⫹ (10⫺3 M) and TEA (10⫺3 M) on the Kir currents in representative MS cells. The experimental protocol and condition for current recording were identical to those in A.

(11), we next examined the effect of Ba2⫹ on the Kir current in MS cells bathed in a solution containing 90 mM [K⫹]o (Fig. 3, A–C). Figure 3A demonstrates that extracellular Ba2⫹ (10⫺4 M) strongly and reversibly inhibited the inward current. The Ba2⫹ inhibition was evidently dose- and time-dependent (Fig. 3B). The plots of the unblocked fractional current (If) (see Eq. 1) as a function of [Ba2⫹]o were fitted with the Hill equation (see Eq. 2) (Fig. 3C). The Kd value and the Hill coefficient (h) were 5.1 ␮M and 1.4, respectively, when the voltage was commanded at ⫺100 mV. Moreover, the fractional Ba2⫹ (10⫺5 M)-sensitive currents measured at command voltages of ⫺100 mV and ⫺50 mV in the same cells were 0.72 ⫾ 0.06 and 0.56 ⫾ 0.08, respectively, suggesting that Ba2⫹ block of the current was voltage dependent. The voltage dependence of the Ba2⫹ inhibition could not be investigated in detail because the currents often showed rundown. Other pharmacological fingerprints of Kir2.x channels include the time-independent and voltage-dependent block by external Cs⫹ (11, 18, 19). Kir currents in MS cells were indeed inhibited by external Cs⫹ (10⫺3 M) in a time-independent (data not shown) and apparently voltage-dependent manner (Fig. 3D). However, Kir currents in MS cells were not affected by external application of TEA (10⫺3 M) or clotrimazole (10⫺6 M), an inhibitor of a large-conductance, Ca2⫹- and voltage-

dependent K⫹ (Maxi-K) channel, or an intermediate conductance, Ca2⫹-dependent K⫹ channel (IK1/SK4), respectively (n ⫽ 4 each) (Fig. 3E and data not shown). Effects of intracellular and extracellular pH on Kir current in mouse MS cells. Among the Kir2.x channels, Kir2.3 and Kir2.4 are known to be distinguished from the other members in that they are highly sensitive to external pH (pHo) (4, 12). Therefore, we next examined the pHo sensitivity of the Kir current. The whole cell currents at ⫺100 mV at pHo 6.6 or 8.2 were normalized to the currents measured at pHo 7.4 in the same cells. The normalized currents at pHo 6.6 and 8.2 were 0.99 ⫾ 0.01 and 1.09 ⫾ 0.01, respectively (Fig. 4A). It has been also reported that Kir2.3 is inhibited by intracellular acidification (35). However, the whole cell current density of MS cells dialyzed with an acidic pipette solution (pHi 6.2) was similar to that with the control solution (pHi 7.4) (Fig. 4B). The Ba2⫹-sensitive current densities at pHi 7.4 and pHi 6.2 were also comparable (data not shown). Kir currents in MS cells were thus insensitive to both pHo and pHi in the range tested. Molecular expression of Kir2.1 in mammary gland of lactating mice. To elucidate the molecular basis underlying the Kir current in MS cells, the mRNA expressions of Kir2.x in the mammary gland of lactating mice were examined with RTPCR. The expression of Kir2.1 mRNA was intensely detected

AJP-Cell Physiol • doi:10.1152/ajpcell.00219.2013 • www.ajpcell.org

C236

Kir2.1-LIKE CHANNEL IN MOUSE MAMMARY SECRETORY CELL

Fig. 4. Effects of extracellular and intracellular pH on Kir current in MS cells. A: inward currents in MS cells in different extracellular pH solutions. The whole cell currents were recorded with the standard pipette solution from MS cells and elicited by voltage ramps as shown in Fig. 1C. Inward currents at ⫺100 mV with pH 6.6 and 8.2 bath solutions (90 mM K⫹) were normalized to the current measured with pH 7.4 solution. Data are means ⫾ SE (n ⫽ 4, each). B: effect of intracellular acidification (pH 6.2) on Kir currents in MS cells. The whole cell currents were measured with an acidic (pH 6.2) pipette solution and the bath solution containing 90 mM K⫹. Experimental protocol for current recording was identical to that in Fig. 2C. The data with a pipette solution at pH 7.4 are from the experiments in Fig. 2C and shown for comparison. Each point represents the mean ⫾ SE (n ⫽ 4 and 9 for pHi 6.2 and 7.4, respectively)

in lactating mammary glands at day 15 of lactation, but not in prepregnant mammary glands (Fig. 5, A and B). The transcripts of Kir2.2, Kir2.3, and Kir2.4 were not detected in either lactating or prepregnant mammary glands (Fig. 5A and data not shown). We next investigated the localization of Kir2.1 protein in lactating mammary glands with a specific primary antibody. Immunohistochemical analysis showed that positive staining with the Kir2.1 antibody was detected in mammary secretory cells (Fig. 6, A and B). The labeling was diffusively distributed in the adjacent region of apical and basolateral membrane. The punctate immunoreactivity was also found in the cytoplasmic region (Fig. 6B, inset). In Western blot analysis, strong antiKir2.1 antibody-reactive bands were observed in the micro-

somal lysate of lactating mammary glands, and the detected sizes were larger than the size predicted from its amino acid sequence (48 kDa) (Fig. 6C). The treatment of the microsomal lysate with N-glycosidase F shifted the band to the predicted size (Fig. 6D). These data indicate that Kir2.1 is expressed abundantly in microsomal fraction of the mammary secretory cells in a N-glycosylated form. Single-channel recording of Kir current in mouse MS cells. To monitor Kir.2.1-like channel activity in intact (i.e., nondialyzed) MS cells, single-channel recording was performed in the cell-attached configuration. The pipette contained the standard

Fig. 5. mRNA expression of Kir channels in mammary glands. A: mRNA expressions of Kir2.x channels in mammary glands of lactating mice. Total RNA was extracted from the abdominal mammary glands of mice on day 15 of lactation. mRNA expressions of Kir2.1, Kir2.2, Kir2.3, Kir2.4, and ␤-actin were examined using RT-PCR in the presence (⫹) or absence (⫺) of a reverse transcriptase. Amplifications obtained from 3 different mice are shown. Positive amplifications were obtained by using total RNA isolated from kidney (for Kir2.2) and brain (for Kir2.3, 2.4). B: expression of Kir2.1 in prepregnant and lactating mammary gland. Total RNA was extracted from the abdominal mammary glands of nulliparous prepregnant mice and lactating mice (day 15 of lactation). Data obtained by RT-PCR from 3 different mice at each stage are shown.

Fig. 6. Expression and localization of Kir2.1 in lactating mammary glands. A and B: immunohistochemical analysis for Kir2.1. The sections of mammary glands from lactating mice were immunostained for Kir2.1 in the presence (B) or absence (A) of the primary antibody. Representative data from three different experiments are shown. Scale bars indicate 50 ␮m. Arrows indicate the punctate staining pattern. C and D: Western blot analysis for Kir2.1. The proteins (100 ␮g) in the cytosolic and microsomal fractions of lactating mammary glands were immunoblotted using the anti-Kir2.1 antibody (C). The microsomal proteins (100 ␮g) incubated with or without N-glycosidase F (NGase F) and the nonincubated microsomal proteins (as a control) were immunoblotted (D). Representative data from three different experiments are shown. The asterisk denotes a deglycosylated form of Kir2.1.

AJP-Cell Physiol • doi:10.1152/ajpcell.00219.2013 • www.ajpcell.org

Kir2.1-LIKE CHANNEL IN MOUSE MAMMARY SECRETORY CELL

C237

Fig. 7. Single-channel Kir current in MS cells. A: representative traces of a Kir-like single-channel current in MS cells. Single-channel currents were recorded with the standard pipette and bath solutions at pipette potentials (Vp) from ⫺69 to ⫹91 mV in the cell-attached configuration. Bars on the right indicate the current level of the closed channels. B: single-channel current-voltage relationships recorded from MS cells in cell-attached mode with the standard bath solution and the pipette solutions containing 30 or 145 mM K⫹. Channel openings were not clearly observed at potentials (⫺Vp) more positive than ⫹9 mV with 145 mM [K⫹]pipette and ⫺31 mV with 30 mM [K⫹]pipette. Each point represents the mean ⫾ SE (n ⫽ 4 and 17 for 30 and 145 mM [K⫹]pipette, respectively). C: relationships between the unitary currents and pipette potentials of Kir channels at different [K⫹]bath. Single-channel currents were recorded in the cell-attached configuration with the standard pipette solution from MS cells bathed in solutions containing 5 or 154 mM K⫹. Channel openings were not clearly observed at potentials (⫺Vp) more positive than ⫹9 mV and ⫺31 mV at 5 and 154 mM [K⫹]bath, respectively. Each point represents the mean ⫾ SE (n ⫽ 17 and 4 for 5 mM and 154 mM [K⫹]bath, respectively).

potassium glutamate-rich solution (145 mM K⫹) and the bath was the standard NaCl-rich solution (5 mM K⫹). Under these conditions, we observed at least two types of channel currents. One of these was a channel characterized by fast flickering between the open and closed states with an intermediate single-channel conductance (3 of 25 patches), which was not studied in the present study in detail. The other was a 31-pS K⫹ channel current with an incidence of 28% (7 of 25 patches). Examples of the currents are shown in Fig. 7A. The channel showed inward rectification over a range of potentials (⫺Vp) from ⫺91 to 69 mV, as revealed by an increased unitary current amplitude with membrane hyperpolarization and the lack of unitary currents at potentials more positive than 29 mV (Fig. 7, A and B). It should be noted that channel openings could be observed at ⫹9 mV (⫺Vp), indicating that the channel was active at around resting membrane potential (RMP). The average unitary conductance calculated from the I-V relationships (Fig. 7B) was 31 ⫾ 1 pS (n ⫽ 17), which is similar to that of Kir2.1 (24, 33). Additionally, when K⫹ concentration in the pipette was lowered from 145 to 30 mM by replacing with equimolar Na⫹, the reversal potentials of the currents as estimated by extrapolation shifted toward the negative direction (Fig. 7B). This finding suggests that the unitary current was mediated by a K⫹-selective channel. We also found that when [K⫹]o (i.e., [K⫹]bath) was increased from 5 to 154 mM, unitary current amplitudes decreased at potentials (⫺Vp) more negative than ⫺51 mV and the estimated reversal potentials of the currents shifted toward the negative direction without changing the slope conductance (Fig. 7C). Based on the assumption that intracellular and extracellular (i.e., pipette solution) ionic concentrations across the patch were constant, the results strongly suggest that membrane K⫹ conductance plays a role in setting RMP of intact MS cells.

The possible role of Kir2.1-like channels in the maintenance of the resting membrane potential. To further elucidate the role of Kir current on the maintenance of RMP of native MS cells, the effects of Ba2⫹ on zero-current voltages were assessed in current-clamp experiments using a pipette solution containing 10⫺7 M free Ca2⫹. This free Ca2⫹ concentration is equivalent to that of the basal level reported for freshly isolated mammary acinar cells (40). We first confirmed that the Kir current was strongly and moderately inhibited by 10⫺4 and 10⫺5 M external Ba2⫹, respectively, even in this experimental condition (data not shown). As shown in Fig. 8, addition of Ba2⫹ (10⫺4 M) to the bath solution induced depolarization of zero-current voltages by 9.7 ⫾ 2.1 mV (n ⫽ 6). We also found that a lower concentration of Ba2⫹ (10⫺5 M) depolarized by 3.8 ⫾ 0.9 mV (n ⫽ 4). These results indicate that cell potentials of MS cells

Fig. 8. Effects of external Ba2⫹ on zero-current voltages of MS cells. A: a continuous current-clamp recording of zero-current voltage of a MS cell. Inside and outside of the cell were perfused with the Ca2⫹ (10⫺7 M)containing pipette solution and the standard bath solution with or without Ba2⫹ (10⫺4 M), respectively. B: responses of zero-current voltages of six MS cells to external Ba2⫹ (10⫺4 M). The data represent the averaged voltages during last 2 s in each treatment.

AJP-Cell Physiol • doi:10.1152/ajpcell.00219.2013 • www.ajpcell.org

C238

Kir2.1-LIKE CHANNEL IN MOUSE MAMMARY SECRETORY CELL

were highly sensitive to external Ba2⫹. Although the pipette solution could not fully replicate the intracellular ionic composition of intact MS cells, these results may also suggest that Kir2.1-like channel activity contributes, at least in part, to the maintenance of the RMP. DISCUSSION

Potassium channels are critical for a variety of cellular functions including transepithelial fluid and electrolyte transport in exocrine glands (27, 36, 45). However, direct evidence for functional expression of a K⫹ channel in native MS cells has been lacking at the single-cell level. In the present study, we show that Kir2.1-like channel activity exists in fully differentiated MS cells freshly isolated from lactating mouse mammary gland. To the best of our knowledge, this is the first report of a specific K⫹ channel functionally expressed in native MS cells. Using the standard whole cell patch-clamp technique, we found that mouse MS cells exhibit an inwardly rectifying K⫹ (Kir) current. Our data also provide compelling evidence that the native Kir current has similar characteristics to those reported for heterologously expressed Kir2.x channels, especially Kir2.1. Firstly, I-V relationship of the native whole cell Kir current showed strong inward rectification (Fig. 2, C and E), and a slope conductance of the inward current was approximately proportional to the square root of external K⫹ concentration ([K⫹]o) (Fig. 2D) (11). Second, in addition to being inhibited voltage dependently by external Cs⫹, the native Kir current was inhibited by external Ba2⫹ in a manner that is concentration-, voltage- and timedependent, as shown for Kir2.x channels (11). Particularly, a Kd value (5.1 ␮M) for Ba2⫹ block estimated for the native current is quantitatively similar to those reported for cloned Kir2.1 (Kd ⫽ 2.7– 8 ␮M) (1, 32, 34, 42, 44) and Kir2.2 (Kd ⫽ 0.5– 6 ␮M) (32, 34, 44). Considering that the sensitivity of the native Kir current to external Ba2⫹ were underestimated due to the uncompensated voltage error derived from series resistance, the native Kir current appears to be more sensitive to Ba2⫹ than Kir2.3 (Kd ⬎ 10 ␮M) (34, 46) and Kir2.4 (Kd ⬎ 65 ␮M) (42, 44). From this viewpoint, the fractional Cs⫹ (1 mM)-sensitive currents that are estimated by the ratio of the currents in the presence of the blocker to those in its absence at different membrane potentials (Fig. 3D) seem to be comparable to that determined under similar experimental conditions for Kir2.1 (43), which has ⬃100-fold lower sensitivity to block by external Cs⫹ than Kir2.2 and Kir2.3 (28, 41). Third, singlechannel recordings with a high K⫹ pipette solution from MS cells in the cell-attached patch configuration revealed a K⫹ channel current with strong inward rectification and with an average unitary conductance of 31 pS, which is within the range reported for Kir2.1 (33) and Kir2.2 (32) but out of the range reported for Kir2.3 and Kir2.4 (28, 44). Finally, consistent with these functional similarities between the native and Kir2.x currents, particularly Kir2.1, RT-PCR analyses revealed that transcripts of Kir2.1, but not the other Kir2.x channels, were expressed in lactating mammary gland. The protein expression of Kir2.1 in mouse MS cells was also confirmed by Western blot and immunohistochemical analyses. Therefore, at this stage, it is practical to

conclude that inwardly rectifying K⫹ currents in mouse MS cells may be mediated by the channels composed of Kir2.1, but not the other Kir2.x, although the available data do not allow us to distinguish whether the native Kir currents are solely mediated by a homo-oligomer of Kir2.1 or not. Further studies are required to address this issue specifically. In the present study, the native Kir currents in MS cells were insensitive to both extracellular and intracellular pH in the range tested (pHo ⫽ 6.6 – 8.2, pHi ⫽ 6.2–7.4, Fig. 4, A and B). The findings appear to be consistent with the characteristics reported for Kir2.1 (4, 35) and thus may strengthen the conclusion that the native currents are mediated by Kir2.1-like channels. However, Kir2.1 is also shown to be inhibited by internal protons with a Ki value of pH 6.2 in the absence of intracellular Mg2⫹ and polyamine that block the channel by direct voltage-dependent pore block and by reduction of single-channel conductance (39). It would be thus interesting to examine if the native currents become sensitive to intracellular protons under such an experimental condition. What are the physiological roles of the Kir channel in native MS cells? It has been well known that despite its strong inward rectification, small outward currents through Kir2.1 contribute to the control of the resting membrane potential (RMP) in a wide variety of cell types (11) including exocrine gland acinar cells (10, 17). Some of our data may allow us to speculate that Kir2.1-like channels could be involved, at least in part, in the generation and maintenance of RMP also in MS cells. The Kir2.1-like channels in MS cells also conducted small but significant outward currents (Fig. 2F). An inwardly rectifying 31 pS K⫹ channel could be active at RMP (Fig. 7, A and B). In addition, RMP might be maintained partly by membrane K⫹ permeability (Fig. 7C). Furthermore, external Ba2⫹ (10⫺4 and 10⫺5 M) that inhibited the native Kir current shifted the zero-current voltage to the depolarizing direction. Thus, by maintaining RMP, Kir2.1-like channels could contribute to slow and continuous secretion of electrolytes of the aqueous phase of milk, a feature of mammary gland secretion. It has been reported that K⫹ influx through Kir channels plays a role in keeping the extracellular K⫹ homeostasis in astrocytes (30) and retinal pigment cell (20). Likewise, if Kir2.1-like channels are expressed in the apical membrane of MS cells, their strong inward rectifying property could be suitable for absorption and thus modification of K⫹ content stored in the lumen of MS cells, although this scenario requires the hyperpolarization of the apical membrane beyond the potassium equilibrium potential (EK) that is determined by the transmembrane concentration gradient of K⫹. In cell-attached patch-clamp experiments, we detected Kir2.1-like channel activity at a relatively low rate (28%), even though whole cell Kir2.1-like currents were observed in all MS cells tested. The apparent discrepancy might imply that functional Kir2.1 channels are localized to the distinctive region of the apical or basolateral membrane. We, however, could not observe such a specific localization of Kir2.1 to the plasma membrane in the present immunohistochemical analysis. Further detailed studies on localization of Kir2.1 will provide an important clue to understand its physiological role in MS cells.

AJP-Cell Physiol • doi:10.1152/ajpcell.00219.2013 • www.ajpcell.org

Kir2.1-LIKE CHANNEL IN MOUSE MAMMARY SECRETORY CELL

Finally, our data do not preclude the possibility that other types of ion channels are also functional in intact MS cells under physiological conditions. In fact, when dialyzed with a pipette solution containing 10⫺7 M free Ca2⫹, MS cells exhibited other membrane currents, which were not characterized in detail in this work, as well as the Kir currents (Fig. 1, C and D). In any case, further investigation will be needed to elucidate the physiological impacts of the native Kir2.1-like channel in intact MS cells. In conclusion, we have demonstrated that functional Kir2.1like channels are expressed in secretory cells freshly isolated from the mammary gland of lactating mouse. Because the ionic contents in the aqueous phase of milk are varied among mammalian species, different types of K⫹ channel in MS cells might play a role in producing species-specific milk. Conversely, particular types of K⫹ channel might be generally expressed and participate in milk production in mammalian MS cells. Therefore, it would be interesting to investigate whether Kir2.1-like currents are specific for native MS cells of mice or they are commonly active in MS cells of other species. This study is a first step toward better understanding the mechanisms of lactation. Characterizing functional ion channels and their cross-talk in native MS cells will be helpful to know how MS cells secrete and preserve ionic components of milk stored into the lumen. GRANTS This work was supported by Japan Society for the Promotion of Science KAKENHI Grant No. 22880003. DISCLOSURES No conflicts of interest, financial or otherwise, are declared by the author(s). AUTHOR CONTRIBUTIONS Author contributions: A.K. and T.I. conception and design of research; A.K. performed experiments; A.K. analyzed data; A.K. and T.I. interpreted results of experiments; A.K. prepared figures; A.K. drafted manuscript; A.K. and T.I. edited and revised manuscript; A.K. and T.I. approved final version of manuscript. REFERENCES 1. Alagem N, Dvir M, Reuveny E. Mechanism of Ba2⫹ block of a mouse inwardly rectifying K⫹ channel: differential contribution by two discrete residues. J Physiol 534: 381–393, 2001. 2. Anderson SM, Rudolph MC, McManaman JL, Neville MC. Key stages in mammary gland development. Secretory activation in the mammary gland: it’s not just about milk protein synthesis! Breast Cancer Res 9: 204, 2007. 3. Blaug S, Rymer J, Jalickee S, Miller SS. P2 purinoceptors regulate calcium-activated chloride and fluid transport in 31EG4 mammary epithelia. Am J Physiol Cell Physiol 284: C897–C909, 2003. 4. Coulter KL, Périer F, Radeke CM, Vandenberg CA. Identification and molecular localization of a pH-sensing domain for the inward rectifier potassium channel HIR. Neuron 15: 1157–1168, 1995. 5. Dunn PM. The action of blocking agents applied to the inner face of Ca2⫹-activated K⫹ channels from human erythrocytes. J Membr Biol 165: 133–143, 1998. 6. Enomoto K, Furuya K, Maeno T, Edwards C, Oka T. Oscillating activity of a calcium-activated K⫹ channel in normal and cancerous mammary cells in culture. J Membr Biol 119: 133–139, 1991. 7. Fang Y, Schram G, Romanenko VG, Shi C, Conti L, Vandenberg CA, Davies PF, Nattel S, Levitan I. Functional expression of Kir2.x in human aortic endothelial cells: the dominant role of Kir22. Am J Physiol Cell Physiol 289: C1134 –C1144, 2005. 8. Furuya K, Enomoto K, Furuya S, Yamagishi S, Edwards C, Oka T. Single calcium-activated potassium channel in cultured mammary epithelial cells. Pflügers Arch 414: 118 –124, 1989.

C239

9. Hamill OP, Marty A, Neher E, Sakmann B, Sigworth FJ. Improved patch-clamp techniques for high-resolution current recording from cells and cell-free membrane patches. Pflügers Arch 391: 85–100, 1981. 10. Hayashi M, Komazaki S, Ishikawa T. An inwardly rectifying K⫹ channel in bovine parotid acinar cells: possible involvement of Kir2.1. J Physiol 547: 255–269, 2003. 11. Hibino H, Inanobe A, Furutani K, Murakami S, Findlay I, Kurachi Y. Inwardly rectifying potassium channels: their structure, function, and physiological roles. Physiol Rev 90: 291–366, 2010. 12. Hughes BA, Kumar G, Yuan Y, Swaminathan A, Yan D, Sharma A, Plumley L, Yang-Feng TL, Swaroop A. Cloning and functional expression of human retinal Kir2.4, a pH-sensitive inwardly rectifying K⫹ channel. Am J Physiol Cell Physiol 279: C771–C784, 2000. 13. Ishii TM, Silvia C, Hirschberg B, Bond CT, Adelman JP, Maylie J. A human intermediate conductance calcium-activated potassium channel. Proc Natl Acad Sci USA 94: 11651–11656, 1997. 14. Jang SS, Park J, Hur SW, Hong YH, Hur J, Chae JH, Kim SK, Kim J, Kim HS, Kim SJ. Endothelial progenitor cells functionally express inward rectifier potassium channels. Am J Physiol Cell Physiol 301: C150 –C161, 2011. 15. Jensen BS, Strobaek D, Christophersen P, Jorgensen TD, Hansen C, Silahtaroglu A, Olesen SP, Ahring PK. Characterization of the cloned human intermediate-conductance Ca2⫹-activated K⫹ channel. Am J Physiol Cell Physiol 275: C848 –C856, 1998. 16. Johnson MP, Wooding FB. Adenosine triphosphatase distribution in mammary tissue. Histochem J 10: 171–183, 1978. 17. Kim SJ, Kerst G, Schreiber R, Pavenstädt H, Greger R, Hug MJ, Bleich M. Inwardly rectifying K⫹ channels in the basolateral membrane of rat pancreatic acini. Pflügers Arch 441: 331–340, 2000. 18. Kubo Y, Adelman JP, Clapham DE, Jan LY, Karschin A, Kurachi Y, Lazdunski M, Nichols CG, Seino S, Vandenberg CA. International Union of Pharmacology. LIV. Nomenclature and molecular relationships of inwardly rectifying potassium channels. Pharmacol Rev 57: 509 –526, 2005. 19. Kubo Y, Baldwin TJ, Jan YN, Jan LY. Primary structure and functional expression of a mouse inward rectifier potassium channel. Nature 362: 127–133, 1993. 20. Kusaka S, Inanobe A, Fujita A, Makino Y, Tanemoto M, Matsushita K, Tano Y, Kurachi Y. Functional Kir7.1 channels localized at the root of apical processes in rat retinal pigment epithelium. J Physiol 531: 27–36, 2001. 21. Lee SY, Palmer ML, Maniak PJ, Jang SH, Ryu PD, O’Grady SM. P2Y receptor regulation of sodium transport in human mammary epithelial cells. Am J Physiol Cell Physiol 293: C1472–C1480, 2007. 22. Linzell JL, Peaker M. Intracellular concentrations of sodium, potassium and chloride in the lactating mammary gland and their relation to the secretory mechanism. J Physiol 216: 683–700, 1971. 23. Linzell JL, Peaker M. Mechanism of milk secretion. Physiol Rev 51: 564 –597, 1971. 24. Lopatin AN, Nichols CG. Inward rectifiers in the heart: an update on IK1. J Mol Cell Cardiol 33: 625–638, 2001. 25. Lowry OH, Rosebrough NJ, Farr AL, Randall RJ. Protein measurement with the Folin phenol reagent. J Biol Chem 193: 265–275, 1951. 26. McManaman JL, Neville MC. Mammary physiology and milk secretion. Adv Drug Deliv Rev 55: 629 –641, 2003. 27. Melvin JE, Yule D, Shuttleworth T, Begenisich T. Regulation of fluid and electrolyte secretion in salivary gland acinar cells. Annu Rev Physiol 67: 445–469, 2005. 28. Morishige K, Takahashi N, Jahangir A, Yamada M, Koyama H, Zanelli JS, Kurachi Y. Molecular cloning and functional expression of a novel brain-specific inward rectifier potassium channel. FEBS Lett 346: 251–256, 1994. 29. Neher E. Correction for liquid junction potentials in patch clamp experiments. In: Methods in Enzymology, edited by Rudy B. San Diego, CA: Academic, 1992, vol 207, p 123–131. 30. Olsen ML, Sontheimer H. Functional implications for Kir4.1 channels in glial biology: from K⫹ buffering to cell differentiation. J Neurochem 107: 589 –601, 2008. 31. Palmer ML, Peitzman ER, Maniak PJ, Sieck GC, Prakash YS, O’Grady SM. KCa3.1 channels facilitate K⫹ secretion or Na⫹ absorption depending on apical or basolateral P2Y receptor stimulation. J Physiol 589: 3483–3494, 2011. 32. Panama BK, McLerie M, Lopatin AN. Functional consequences of Kir2.1/ Kir22 subunit heteromerization. Pflügers Arch 460: 839 –849, 2010.

AJP-Cell Physiol • doi:10.1152/ajpcell.00219.2013 • www.ajpcell.org

C240

Kir2.1-LIKE CHANNEL IN MOUSE MAMMARY SECRETORY CELL

33. Picones A, Keung E, Timpe LC. Unitary conductance variation in Kir2.1 and in cardiac inward rectifier potassium channels. Biophys J 81: 2035– 2049, 2001. 34. Preisig-Müller R, Schlichthörl G, Goerge T, Heinen S, Brüggemann A, Rajan S, Derst C, Veh RW, Daut J. Heteromerization of Kir2x potassium channels contributes to the phenotype of Andersen’s syndrome. Proc Natl Acad Sci USA 99: 7774 –7779, 2002. 35. Qu Z, Zhu G, Yang Z, Cui N, Li Y, Chanchevalap S, Sulaiman S, Haynie H, Jiang C. Identification of a critical motif responsible for gating of Kir2.3 channel by intracellular protons. J Biol Chem 274: 13783– 13789, 1999. 36. Schubert ML. Gastric secretion. Curr Opin Gastroenterol 26: 598 –603, 2010. 37. Shennan DB, McNeillie SA. Efflux of chloride from lactating rat mammary tissue slices. Comp Biochem Physiol A Comp Physiol 95: 367–371, 1990. 38. Shennan DB, Peaker M. Transport of milk constituents by the mammary gland. Physiol Rev 80: 925–951, 2000. 39. Shieh RC, John SA, Lee JK, Weiss JN. Inward rectification of the IRK1 channel expressed in Xenopus oocytes: effects of intracellular pH reveal an intrinsic gating mechanism. J Physiol 494: 363–376, 1996.

40. Sudlow AW, Burgoyne RD. A hypo-osmotically induced increase in intracellular Ca2⫹ in lactating mouse mammary epithelial cells involving Ca2⫹ influx. Pflügers Arch 433: 609 –616, 1997. 41. Takahashi N, Morishige K, Jahangir A, Yamada M, Findlay I, Koyama H, Kurachi Y. Molecular cloning and functional expression of cDNA encoding a second class of inward rectifier potassium channels in the mouse brain. J Biol Chem 269: 23274 –23279, 1994. 42. Tennant BP, Cui Y, Tinker A, Clapp LH. Functional expression of inward rectifier potassium channels in cultured human pulmonary smooth muscle cells: evidence for a major role of Kir2.4 subunits. J Membr Biol 213: 19 –29, 2006. 43. Thompson GA, Leyland ML, Ashmole I, Sutcliffe MJ, Stanfield PR. Residues beyond the selectivity filter of the K⫹ channel Kir2.1 regulate permeation and block by external Rb⫹ and Cs⫹. J Physiol 526: 231–240, 2000. 44. Töpert C, Döring F, Wischmeyer E, Karschin C, Brockhaus J, Ballanyi K, Derst C, Karschin A. Kir2.4: a novel K⫹ inward rectifier channel associated with motoneurons of cranial nerve nuclei. J Neurosci 18: 4096 –4105, 1998. 45. Warth R. Potassium channels in epithelial transport. Pflügers Arch 446: 505–513, 2003. 46. Welling PA. Primary structure and functional expression of a cortical collecting duct Kir channel. Am J Physiol Renal Physiol 273: F825–F836, 1997.

AJP-Cell Physiol • doi:10.1152/ajpcell.00219.2013 • www.ajpcell.org