Pflugers Arch - Eur J Physiol (2007) 453:787–796 DOI 10.1007/s00424-006-0158-2

CELL AND MOLECULAR PHYSIOLOGY

Aquaporin expression and cell volume regulation in the SV40 immortalized rat submandibular acinar cell line Ann-Kristin Hansen & Hilde Kanli Galtung

Received: 24 May 2006 / Revised: 15 July 2006 / Accepted: 11 August 2006 / Published online: 5 October 2006 # Springer-Verlag 2006

Abstract The amount of aquaporins present and the cellular ability to perform regulatory volume changes are likely to be important for fluid secretions from exocrine glands. In this work these phenomena were studied in an SV40 immortalized rat submandibular acinar cell line. The regulatory cell volume characteristics have not previously been determined in these cells. Cell volume regulation following hyposmotic exposure and aquaporin induction was examined with Coulter counter methodology, radioactive efflux studies, fura-2 fluorescence, and polymerase chain reaction and Western blot techniques. Cell volume regulation was inhibited by the K+ channel antagonists quinine and BaCl2 and the Cl− channel blocker 5-nitro-2(3-phenypropylamino)benzoic acid. A concomitant increase in cellular 3H-taurine release and Ca2+ concentration was also observed. Chelation of both intra- and extracellular Ca2+ with EGTA and the Ca2+ ionophore A23187 did not, however, affect cell volume regulation. Aquaporin 5 (AQP5) mRNA and protein levels were upregulated in hyperosmotic conditions and downregulated upon return to isosmotic solutions, but were reduced by the mitogenactivated ERK-activating kinase (MEK) inhibitor U0126. A 24-h MEK inhibition also diminished hyposmotically induced cell swelling and cell volume regulation. In conclusion, it was determined that regulatory volume changes in this immortalized cell line are due to KCl and taurine efflux. In conditions that increased AQP5 levels, the cells showed a faster cell swelling and a more complete volume recovery following hyposmotic exposure. This response could be overturned by MEK inhibition. A.-K. Hansen : H. Kanli Galtung (*) Institute of Oral Biology, Faculty of Dentistry, University of Oslo, P.O. Box 1052, Blindern, 0316 Oslo, Norway e-mail: [email protected]

Keywords Osmoregulation . KCl efflux . Taurine . Calcium . Aquaporin induction . MAP kinase . Salivary

Introduction Cell volume regulation is essential for cell function and viability. Transporting epithelia such as those in the salivary glands, kidneys and intestines are of special interest in this respect. For example, the salivary glands in human adults normally secrete around 1 l of saliva each day. As a result of solute and water fluxes across the cell membranes during saliva formation, cell volumes will tend to vary. Indeed, diseases that change the volume or the composition of saliva pose a significant health issue for the patient concerned [19]. The amount of aquaporins present in the plasma membrane and the ability of cells to perform regulatory volume changes are likely to be important factors for fluid secretions from exocrine glands, including salivary glands. To study these phenomena in salivary gland acinar cells, a continuous cell line that has not originated from cancerous lesions is most useful. The SV40 immortalized rat submandibular acinar cell line developed by Quissell and colleagues fulfills this criterion [26]. This cell line has maintained physiologic functional responses to adequate hormones and neurotransmitters. In addition, it shows presence of submandibular acinar cell specific secretory proteins, tripartite junctional complexes, cellular polarization, secretory granules and rough endoplasmic reticulum [26]. The regulatory cell volume characteristics, however, had not prior to this study been determined in these cells. Furthermore, due to the importance of AQP5 in saliva production, it was of interest to study (a) how expression of

788

this water channel could be modified by variations in the osmolality of the extracellular medium and (b) how changes in AQP5 expression would alter cell volume regulation. To our knowledge, the occurrence and regulation of aquaporin induction and its effect on cell volume regulation have not been studied in these cells. In addition, the mechanisms that regulate aquaporin expression are not completely elucidated. Thus, in this work, regulatory volume decreases (RVDs) following hyposmotic exposure and aquaporin expression in hyperosmotic solutions in this immortalized cell line were studied. The cells were not stimulated with parasympathetic or sympathetic agonist since the main focus of this initial study was the mechanisms for RVDs and the role of aquaporin expression for this process. Within a few minutes of cell swelling, the volume is in most cells gradually decreased (RVD). This mechanism is generally a result of KCl, amino acid and water efflux [13, 30]. Notably, efflux of the β-amino acid taurine has been found to increase significantly during RVD [10, 11, 15]. The signal for the RVD-associated KCl efflux varies somewhat from cell type to cell type. Specifically, intracellular Ca2+ has been implemented in RVD in a variety of cells [1, 7, 29]. In a recent study it has been found that exposing cells to hyposmolar solutions activates Ca2+ entry and that this influx is necessary for RVD to occur [21]. Aquaporins, or water channels, are responsible for osmotic water flux across cell membranes in many cells. These water channels possess extraordinary water conductivity. For example, aquaporin 1 allows 3×109 water molecules per monomer per second [33, 35, 36]. Aquaporin 5 (AQP5) is found in salivary glands, and during saliva production, the channel facilitates osmotically obligated water transport following ion movement. It is also expected that AQP5 is involved in the cell volume regulation that occurs during ion and water fluxes across the cell membranes [20]. New work points to a close functional relationship between the transient receptor potential vanilloid 4 (TRPV4) and AQP5 during regulatory volume decreases [21]. Furthermore, it is possible to induce expression of the channel by exposing mouse lung cells to hypertonic stress [14]. This upregulation seems to involve the mitogen-activated protein (MAP) kinase pathway [14], but has, however, not yet been studied in salivary gland cells. There is also an indication that there is an abnormal cellular distribution of AQP5 in some types of Sjögren’s disease [3, 27], an ailment associated with decreased saliva secretion. In these patients it seems as though AQP5 is localized primarily to the basal acinar membranes, whereas in normal subjects, AQP5 is found at the apical and AQP3 at the lateral and basal acinar membranes [12, 27].

Pflugers Arch - Eur J Physiol (2007) 453:787–796

In this report it was found that cell volume regulation in an immortalized rat salivary acinar cell line is due to KCl and taurine efflux. This efflux appears to be Ca2+independent. Growth of the cells in hyperosmotic conditions leads to an induction of AQP5. Following such an induction period, the cells showed a faster swelling and a more complete regulatory volume decrease upon hyposmotic shock. Both the AQP5 induction and the positive effects on cell volume regulation could be decreased by inhibition of the MAP kinase system.

Materials and methods Cell culture Experiments were performed on rat submandibular acinar cells in culture (SM 10, passage 12). The immortalized cell line was a gift from Dr. D. Quissell, University of Colorado, USA. Cells were grown in Primaria Falcon flasks (Falcon/Becton Dickenson, NJ, USA) at 37°C in 5% CO2 under standard cell culture conditions. Doubling times were around 24 h. The cell culture solution consisted of Ham’s F12/Dulbecco’s modified Eagle medium (1:1; Gibco BRL) containing 15 mM HEPES, 14 mM NaHCO3, 2.5 mM glutamine and 16 mM D-glucose to which was added 2% fetal bovine serum (Gibco BRL), 5 μg/ml insulin (Gibco BRL), 80 ng/ml epidermal growth factor (Collaborative Biomedical Products), 0.1 μM retinoic acid, 1.1 μM hydrocortisone, 5 μg/ml transferrin (Gibco BRL), 2 nM T3 (3,3′,5-triiodo-L-thyroxine), trace element mix [BioFluids, Rockville, MD, USA; containing (in milligrams per liter) manganese 0.02, silicate 14.21, molybdate 0.124, vanadate 0.058, nickel 0.02, tin salts 0.011 and selenium 0.519], 2.5 mM glutamine and 50 μg/ml gentamicin sulfate (unless otherwise noted, all chemicals were from Sigma). Cells were dissociated with trypsin EDTA for 10 min at 37°C, after which fresh culture medium was added to the flasks, the cells centrifuged for 5 min at 1,000 rpm, the supernatant discarded and the cells resuspended in culture medium. All cells were kept for at least 20 min in this solution before experiments were commenced. Isosmotic growth conditions Cell volume regulation (Coulter counter technique) Cells were dissociated by trypsination, isolated by centrifugation and resuspended in Ham’s F12/Dulbecco’s cell culture medium as described above (310 mOsm; pH 7.4; 37°C). The cells were kept in this solution for up to 1 h and 20 min. An aliquot of this cell suspension was then placed in culture medium to which the pharmacologic inhibitors/

Pflugers Arch - Eur J Physiol (2007) 453:787–796

agents to be tested had been added (see below). The cell volumes of cells with or without test chemicals were measured with a Coulter counter ZB Channelyzer every minute for 3 min (isosmotic exposure, 310 mOsm). At this time a new aliquot of cells pre-exposed to isosmotic solution with or without test chemicals was added to a hyposmotic solution (100 mOsm; cell culture medium diluted with water) with or without the same test chemicals. Time 0 was taken as the first volume reading in the hyposmotic solution. There is, however, a time lag of about 15 s from when the cells are suctioned into the Coulter counter apparatus until a read-out can be made. Cell volumes were measured every 30 s for the first 5 min in hyposmotic solutions, then every 3 min for the next 10 min. For each experiment a parallel control experiment was performed. The cells were kept in a 37°C water bath for the duration of the experiments. The pH of the solutions the cells were suspended in remained constant at 7.4. To determine if the ionic strength of the solutions could influence the cell volume readings, experiments were also performed with Coulter size standard latex beads (diameter 10 μm, Beckman Coulter Inc., Fullerton, CA, USA) added to the isosmotic and hyposmotic solutions. The pharmacologic inhibitors used were as follows: K+ channel inhibitors–quinine (0.5 mM, n =4) and BaCl2 (1, 5, 10 and 20 mM, n= 4 for all); Cl− channel inhibitors— IAA-94 (R(+)-[(6,7-dichloro-2-cyclopentyl-2,3-dihydro2-methyl-1-oxo-1H-inden-5-yl)-oxy]acetic acid) (0.1 mM, n=5 and 1 mM, n=4) and NPPB (5-nitro-2-(3-phenypropylamino)benzoic acid) (0.1 mM, n=4 and 0.8 mM, n=4); K+/Cl− symporter inhibitor—furosemide (0.1 mM, n=4; 1 mM, n=5 and 10 mM, n=3). Finally, the Ca2+ chelator EGTA (1 mM) alone or EGTA and the Ca2+ ionophore A23187 (10 μM) in combination were used to test the cells’ requirement for extra- and intracellular Ca2+ during cell volume regulation (n=4 for both sets). In these experiments the Ca2+ concentration in the isosmotic and hyposmotic solutions was reduced from 2 to 0.1 mM prior to addition of A23187 or EGTA (S-MEM Eagle Spinner Culture Ca2+-free medium to which 0.1 mM CaCl2 was added). The cells were kept for 3 min in the isosmotic solution with EGTA or EGTA and A23187 and for 15 min in the hyposmotic solution, still in the presence of EGTA or EGTA/A23187. Taurine efflux during cell volume regulation ( 3H-taurine measurements) Cells grown on plastic cell culture dishes were incubated for 24 h with 1 μCi 3H-taurine (Amersham Chemical Company). The cells were then washed for 10 min in Ham’s F12/Dulbecco’s cell culture medium. The dishes were placed in a flow-through chamber that allowed

789

sampling of the fluid that had passed over the cells. The fluid was collected at 1-min intervals. Cells were exposed to the cell culture medium (310 mOsm) for 5 min before being introduced to the hyposmotic solution described above (100 mOsm) for 15 min (n=7). The radioactivity in the efflux samples was measured with standard isotope detection using Triton X-100. Cell proteins were precipitated with 5% TCA and the radioactivity of the supernatant measured. Finally, the rate coefficient for 3H-taurine efflux (1/min) was calculated according to standard procedure [10]. Intracellular Ca2+ variations during RVD (fura-2-AM measurements) The salivary gland cells were grown in glass-bottomed chambers following the same procedure outlined above (Lab-Tek chambered cover glass, Nalge Nunc Intl., Rochester, NY, USA). Cells were loaded with the fluorescent Ca2+-binding dye fura-2-AM (Molecular Probes, Eugene, OR, USA) in an isosmotic Ringer’s solution (1 μM fura-2-AM for 30 min at 37°C). The Ringer’s solution contained the following (in millimolars): 130 NaCl, 5.4 KCl, 1.8 CaCl2, 0.8 MgSO4, 1 NaH2PO4, 10 glucose, 20 HEPES and 2 glutamine (290 mOsm, pH 7.4) The cells were washed twice with dye-free isosmotic Ringer’s solution, and the chamber with cells was placed on the temperature-regulated stage of an inverted Zeiss microscope fitted for ratiometric fluorescence analysis (37°C). After a few minutes the chamber solution was switched from isosmotic to hyperosmotic (100 mOsm, NaCl content reduced), and fura-2 fluorescence was recorded at regular intervals for 6 min (n=7). The dye was excited at 340 and 380 nm, and the emission was recorded at 510 nm. As a measure of the intracellular Ca2+ concentration, the 340/380 (nanometers) emission ratio was calculated using the MetaFluor 6.2 software (Universal Imaging Corporation, PA, USA). Hyperosmotic growth conditions Western blot analysis of AQP5 expression Cells were grown in a 500-mOsm culture medium for 24 h and then in the normal isosmotic culture medium for 24 h (310 mOsm; medium described above) with or without the mitogen-activated ERK-activating kinase (MEK) inhibitor U0126 [1,4-diamino-2,3-dicyano-1,4-bis(o-aminophenylmercapto) butadiene; 50 μM; inhibits activation of MAP kinase]. The hyperosmotic medium was constructed by adding sucrose to the normal isosmotic cell culture medium. Cells were harvested at 0, 4, 8, 12 and 24 h in the hyperosmotic medium, and at 3, 6 and 24 h after the

790

isosmotic medium was reintroduced. At time 0 h, the cells were at 60–70% confluence, and at the end of the experimental series, the cells were at 100% confluence. The cells’ doubling times were about 24 h. The collected cells were lysed at 4°C in a 10% NP-40 solution. Lysates were centrifuged for 15 min at 15,000 rpm to remove insoluble material. Total protein concentrations were determined by the BioRad protein assay using bovine serum albumin as the standard (Bio-Rad, CA, USA). A total of 50 μg of protein was loaded per well in a 10% linear gradient sodium dodecyl sulfate (SDS)-polyacrylamide gel. After gel electrophoresis, proteins were transferred to a nitrocellulose membrane from BioRad. The membrane was treated with 5% non-fat dry milk powder in TBST buffer at 4°C overnight to block non-specific binding sites. The blot was then incubated with the affinity purified anti-AQP5 antibody (Alpha Diagnostic International, San Antonio, TX, USA) diluted 1:1,000 in 1% TBST at room temperature for 2 h. The blot was then washed in TBST followed by incubation with AP-conjugated rabbit anti-rat IgG (Jackson ImmunoResearch, Cambridgeshire, UK) diluted 1:5,000 in 1% TBST for 1 h. After washing with TBST, the immunoblot was developed using the ECF kit from Amersham. Blots were visualized with a STORM scanner from Amersham Biosciences. Relative band intensities were determined using the associated software Image Quant. Reverse transcriptase polymerase chain reaction analysis of AQP5 expression Cells were exposed to hyperosmotic growth conditions with and without MEK inhibitor as described for the Western blot procedure above. For reverse transcriptase polymerase chain reaction (RT-PCR), total RNA from the cells were extracted by columns (RNeasy, Qiagen, Valencia, CA, USA) according to the manufacturer’s instructions. PCR was carried out using a OneStep RT-PCR kit from Qiagen. The kit contains optimized components that allow both reverse transcription and PCR amplification to take place in a ‘one-step’ reaction. Primer sequences (Invitrogen, Oslo, Norway) used were as follows: AQP5 (730 bp) sense, 5′-C AAG GCG GTG TTC GCA GAG TTC C-3′, and antisense, 5′-C CTC TCG ATG ATC TTC CCA GTC C-3′; ribosomal protein L-27 (520 bp) sense, 5′-CCG ACC GCC CTT ACA GC-3′, and antisense, 5′-CCT GGC CTT GCG TTT CA-3′. Ribosomal protein L-27 was used as a housekeeping gene. Reaction mixtures consisted of 25 μl final volume containing 0.25 μg total RNA, 5× QIAGEN OneStep RT-PCR Buffer (contains 12.5 mM MgCl2), 10 mM of each dNTP, 0.6 μM of each primer and 2 μl QIAGEN OneStep RT-PCR Enzyme Mix

Pflugers Arch - Eur J Physiol (2007) 453:787–796

(contains Omniscript Reverse Transcriptase, Sensiscript Reverse Transcriptase and HotStarTaq DNA Polymerase). PCR was performed in an Eppendorf thermocycler. Cycle parameters were as follows: reverse transcriptase reaction at 50°C for 30 min, initial activation of HotStarTaq DNA Polymerase at 90°C for 15 min, followed by 30 cycles of 94°C for 1 min (denature), 55°C for 1 min (anneal) and 72°C for 1 min (extend). PCR was completed for a final extension at 72°C for 10 min. PCR products were separated by electrophoresis on a 3% agarose gel. The gel was stained with ethidium bromide and scanned in a ProXPRESS Proteomic Imaging System from Perkin-Elmer. The signal was then analyzed using the associated software Totallab. Cell volume regulation following inhibition of aquaporin expression The cells’ regulatory volume response following a period of growth in hyperosmotic solutions was studied by first keeping the cells in the 500-mOsm culture medium described above for 24 h. The cells were then harvested, and their cell volumes in isosmotic (310 mOsm, 3 min) and then hyposmotic (100 mOsm, 15 min) culture medium were measured as described above. In an attempt to compromise AQP5 expression both during isosmotic and hyperosmotic growth conditions, the MAP kinase pathway was inhibited by addition of the MEK antagonist U0126 (50 μM, 24 h). Following a 24-h incubation period, cell volume measurements in isosmotic and hyposmotic solutions were performed as above (n=4 for all).

Results Isosmotic growth conditions Cell volume regulation The isosmotic cell volumes remained relatively constant in the control experiments (representative in Fig. 1a), with an average baseline cell volume of 1,340±20 fl (n=38). The cell volumes in the figures are expressed as cell volume changes (delta volumes). Upon hyposmotic stimulation, the control cell volumes rapidly increased to reach their peak within 1-2 min following solution change (Fig. 1a). The average volume at this time for all the control experiments was 1,610±30 fl (n=38), which amounts to a 20±1% swelling (P <0.05). Following this swelling, the cell volumes gradually decreased towards their baseline values. At the end of the experiment (t=15 min), the average cell volume for all the control experiments was 1,360±20 fl

Pflugers Arch - Eur J Physiol (2007) 453:787–796 310

400

100 mOsm *

Delta volume (fL)

300 200

*

100 0

Quinine 0.5 mM (n=4)

Control (n=5)

-100 -3

0

3

6

9

12

15

Time (min)

b

310

100 mOsm

600

Delta volume (fL)

500 BaCl2 (n=4 each):

400

* 1 mM * 10 mM * 20 mM

300 200

* 5 mM

100 Control (n=16)

0 -100 -3

0

3

6

9

12

cell swelling was higher at 1 mM BaCl2 than that for the other Ba2+ concentrations. Furthermore, both inhibitors induced a 1- to 3-min delay in the attainment of peak volumes. This delay was higher for 1 mM BaCl2 than for the other barium concentrations. Of the two Cl− channel inhibitors used, only NPPB affected cell volume regulation (Fig. 2a). NPPB inhibited volume regulation by 70±10% at 0.8 mM (P<0.05, no effect at 0.1 mM). The other inhibitor, IAA, did not change RVD at the two concentrations tested (0.1 and 1 mM; Fig. 2a, only 0.1 mM is shown). The K+/Cl− symporter inhibitor furosemide did not change the cell shrinkage response at low concentrations (0.1 and 1 mM), but exhibited a strong inhibition at 10 mM (60±3% inhibition; P<0.05; Fig. 2b). Furosemide not only inhibited RVD; it also induced a larger peak

a

310

100 mOsm

600

15

*

500

0.8 mM (n=4)

300

NPPB 0.1 mM (n=4)

200

IAA 0.1 mM Control (n=4) (n=5)

0 -100 -3

0

3

6

9

12

15

Time (min)

b

400

310

100 mOsm

Delta volume (fL)

300

(n=38). This value corresponds to a 15±1% shrinkage (P<0.05). There was some variability in the amount of cell swelling between groups of experiments. The reason for this variability could be due to natural changes within the cell line, small variations in trypsination time due to some fluctuations in confluency, or minor variations in cell culture medium. The important point is, however, that for every experiment with inhibitors, a control experiment was performed. Thus, every experiment has its own control. Exposure to the K+ channel blocker quinine inhibited volume regulation by 50±7% (0.5 mM, P<0.05; Fig. 1a). Similarly, the K+ channel inhibitor BaCl2 induced a significant reduction in cell volume decreases (Fig. 1b; RVD inhibition in per cent: 85±20, 50±7, 55±5 and 50±5 at 1, 5, 10 and 20 mM, respectively; P<0.05). The level of

* NPPB

400

100

Time (min)

Fig. 1 Effect of quinine and barium treatment. Time (minutes) is on the x-axis; delta volume (femtoliters) is expressed on the y-axis. Number of experiments is depicted by n. a The K+ blocker quinine was added to both iso- and hyposmotic solutions (0.5 mM). *Value significantly different from control (P<0.05). b Iso- and hyposmotic exposure to the K+ channel inhibitor BaCl2 was added to both iso- and hyposmotic solutions (1, 5, 10 and 20 mM BaCl2). *Final volume value significantly different from control (P<0.05)

800 700

Delta volume (fL)

a

791

*

*

200

* Furosemide 10 mM (n=4)

100

Control (n=5) Furosemide 1 mM (n=5)

0

Furosemide 0.1 mM (n=5)

-100 -3

0

3

6 9 Time (min)

12

15

Fig. 2 Effect of NPPB, indanyloxyacetic acid 94 (IAA-94) and furosemide on RVD. Time (minutes) is on the x-axis; delta volume (femtoliters) is expressed on the y-axis. Number of experiments is depicted by n. a The Cl− channel antagonists NPPB (0.1 and 0.8 mM) and IAA-94 (0.1 mM) were added to both iso- and hyposmotic solutions. *Maximum and final cell volume significantly different from control (P<0.05). b The K+/Cl− symporter inhibitor furosemide was added to both iso- and hyposmotic solutions (0.1, 1 and 10 mM). *Value significantly different from control (P<0.05)

Pflugers Arch - Eur J Physiol (2007) 453:787–796 310

0,30

100 mOsm

250

*

0,25 Rate coefficient (min-1)

swelling than normal (75±30% more swelling than in controls; Fig. 2b). Chelation of extracellular Ca2+ or both extra- and intracellular Ca2+ had no effect on regulatory cell volume decrease (Fig. 3a). Nonetheless, in some, but not all, cells there was a doubling of the intracellular Ca2+ concentration upon hyposmotic stimulation at 37°C (P<0.05; Fig. 3b). The Ca2+ concentration did not normalize throughout the 350-s observation period.

Taurine efflux (n=7)

0,20

* 0,10

50 0

0,05 Delta volume (n=5)

0,00

-50 -100

-3

H-taurine efflux increased when the cells were exposed to hyposmotic media (Fig. 4). Taurine efflux started to increase 1–2 min after solution change and reached a

310

0

3

6

9

12

15

Time (min)

3

600

150 100

0,15

Taurine efflux

a

200 Delta volume (fL)

792

Fig. 4 Taurine efflux during cell volume regulation. Taurine efflux is superimposed on cell volume changes during an isosmotic (300 mOsm) to hyposmotic (100 mOsm, t=0) solution change. Time (minutes) is on the x-axis; rate coefficient of 3H-taurine efflux (per minute) and delta volume change (femtoliters) are on the y-axis. *Value significantly different from value at t=0 (P<0.05). §Value significantly different from maximum efflux value (P<0.05)

100 mOsm

Delta volume (fL)

500 400 300 EGTA+ A23187

200

Control

100

EGTA Control

0

n=4 for all

-100 -3

0

3

6

9

12

15

Time (min)

b

peak after 9 min (400 ± 8 times increase in rate coefficient; P<0.05). Within the next 6 min there was a steady decline in the efflux rate coefficient (45±10% decrease, P < 0.05). Thus, the onset of taurine efflux coincides with that of cell swelling, whereas taurine efflux starts to decline about 10 min after cell volume regulation has begun (Fig. 4; volume measurements are from Coulter counter experiments). Hyperosmotic growth conditions with or without MEK antagonist AQP5 expression

28

[Ca2+]i-concentration (µM)

26 24 22

*

20 18 16

n=7

14 12 10 8 -60 -30 0

30 60 90 120 150 180 210 240 270 300 330 360 390

Time (sec)

Fig. 3 Effect of extra- and intracellular Ca2+ on RVD. a Chelation of extracellular Ca2+ or both extra- and intracellular Ca2+ (addition of 1 mM EGTA and 0.1 mM Ca2+ or 1 mM EGTA and 10 μM A23187 to both iso- and hyposmotic solutions). b Intracellular Ca2+ concentration during RVD as measured with fura-2-AM. The extracellular solution was switched from isosmotic 310 mOsm to hyposmotic 100 mOsm at t=0. *Value significantly different from that at t=0 (P<0.05)

Aquaporin 5 mRNA and protein were found in this salivary acinar cell line (Fig. 5a–c). Incubating the cells in 500-mOsm hyperosmotic medium for 24 h increased the amount of AQP5 mRNA and AQP5 protein 3.5and 5-fold, respectively (P<0.05). After 24 h in the reintroduced 310-mOsm isosmotic growth medium, AQP5 mRNA and protein levels returned to near normal. Furthermore, both AQP5 mRNA and protein levels were decreased when cells were incubated for 24 h with the MEK inhibitor U0126 in isosmotic or hyperosmotic solutions (inhibits the MAP kinase pathway; Fig. 6a–c). Addition of U0126 to isosmotic solutions decreased both AQP5 mRNA and protein levels by about 20%, whereas addition of U0126 to the hyperosmotic growth medium reduced the expected increase in mRNA and protein level by 2 and 2.5 times, respectively (P<0.05 for all).

Pflugers Arch - Eur J Physiol (2007) 453:787–796

793

a

a

450

*

% AQP5 protein or mRNA change

*

500

* AQP5 protein AQP5 mRNA

400

* *

300

*

*

*

*

200

*

*

*

100

* *

*

400 350 300 250

*

200

*

150 100

n=5 for all

*

*

50

Isosmotic growth (310 mOsm)

b

Protein

Hyper+U0126

48

Hyperosmotic

Hyperosmotic growth (500 mOsm)

30

Iso+U0126

12 24 27 Time (hours)

Isosmotic

8

Hyper+U0126

4

Isosmotic

0

Iso+U0126

0

0

Hyperosmotic

AQP5 protein or mRNA (% of control)

600

mRNA

b

c

c

Fig. 5 Aquaporin 5 mRNA and protein levels during growth in hyperosmotic solutions. a Cells were grown for 24 h in a hyperosmotic growth medium. At this point, the cell culture medium was returned to the normal isosmotic solution. *Value significantly different from control at t=0 (P<0.05). b Representative result from the RT-PCR experiments indicating AQP5 mRNA levels in hyperosmotic and reintroduced isosmotic solutions. c Representative result from the Western blot experiments showing AQP5 protein levels in hyperosmotic and reintroduced isosmotic solutions

Cell volume regulation Upon acute hyposmotic solution switch, both cells grown for 24 h in the hyperosmotic medium (500 mOsm) and cells grown in isosmotic conditions (310 mOsm) displayed a normal cell swelling and subsequent regulatory volume decrease pattern (Fig. 7). Cells grown in hyperosmotic medium, however, both swelled then shrank more than the control cells (65% difference; P<0.05). Cell swelling in the hyposmotic solution was significantly decreased by adding the MEK inhibitor U0126 for

Fig. 6 Inhibition of the MAP kinase pathway on AQP5 mRNA and protein levels in isosmotic and hyperosmotic solutions. a Cells were grown for 24 h in an isosmotic (310 mOsm) or hyperosmotic (500 mOsm) growth medium to which the MEK inhibitor U0126 was added (50 μM). *Value significantly different from isosmotic control (P<0.05). b Representative result from the RT-PCR experiments indicating AQP5 mRNA levels during MEK inhibition in isosmotic and hyperosmotic solutions (addition of 50 μM U0126). c Representative result from the Western blot experiments showing AQP5 protein levels in the presence of the MEK inhibitor U0126 (50 μM) added to isosmotic (310 mOsm) and hyperosmotic (100 mOsm) solutions. Iso isosmotic, hyper hyperosmotic

24 h to the hyperosmotic growth medium (25% decrease; P<0.05). Furthermore, U0126 significantly decreased cell volume recovery both in cells grown in isosmotic and hyperosmotic solutions (66% decrease; P<0.05).

794

Pflugers Arch - Eur J Physiol (2007) 453:787–796 350

310

300

Delta volume (fL)

100 mOsm

**

250

$

200 Hyper growth

150

* +U0126 Control * +U0126

100 50

Hyper growth § Control

0 -50 -3

0

3

6

9

12

15

Time (min)

Fig. 7 Effect of inhibition of the MAP kinase pathway on cell swelling and cell volume regulation under isosmotic and hyperosmotic growth conditions. Time (minutes) is on the x-axis; delta volume (femtoliters) is expressed on the y-axis. Number of experiments is depicted by n. Cells were grown for 24 h with or without 50 μM of the MEK inhibitor U0126 in an isosmotic (310 mOsm) or hyperosmotic (500 mOsm) solution before acute exposure to a 100 mOsm hyposmotic solution. *Value significantly different from corresponding control cells without U0126 (P<0.05). §Per cent RVD from max swelling significantly different from controls grown in isosmotic solutions.*$Value significantly different from that in hyperosmotic growth without U0126 (P<0.05). ¤*Per cent RVD from max swelling significantly different from corresponding controls without U0126 (P<0.05). Control cells grown for 24 h in isosmotic solutions, control+U0126 cells grown for 24 h in isosmotic solutions with 50 μM U0126, hyper growth cells grown for 24 h in hyperosmotic solutions, hyper growth+U0126 cells grown for 24 h in hyperosmotic solutions with 50 μM U0126

Discussion It seems that water channels (aquaporins) could be involved either as the causal link to xerostomia (dry mouth) [27] or as part of a possible cure [4] (see the review in [3]). In much the same way as aquaporins are integral to saliva formation, so must cell volume regulation be. Cell volume regulation and osmotic properties of the immortalized cell line used in this study have, to our knowledge, not been characterized. The cell line originates from rat submandibular acinar glands and was developed by Dr. D. Quissell and associates at the University of Colorado, USA [26]. Many of the salivary acinar differential features and cell functions have been retained in these cells [26]. As ions and water move from plasma, through the salivary acinar cells and out into the luminal fluid, the cell volumes must be tightly controlled to avoid significant changes in cell size during fluid secretion. Thus, it is expected that these epithelial cells have fairly well developed regulatory cell volume mechanisms. Indeed, when the acinar cells in this study were exposed to hyposmotic solutions, cell volumes increased rapidly due to water influx. This swelling lasted only 1–2 min and was followed by a cell decrease phase in which cell volumes

were reduced to near baseline values (within 10 min). These results are in agreement with findings in isolated, non-cultured salivary acinar cells [9] and human salivary gland cells [21]. There have been reports, however, of lack of RVD in acinar cells in intact glands [28]. The regulatory cell volume decrease observed in this report seems to be dependent on KCl efflux since RVD could significantly be reduced by inhibitors of both K+ (BaCl2 and quinine) and Cl− (NPPB) channels. Similar KCl efflux dependence has been observed in non-cultured isolated rat salivary acinar cells [9] and cells derived from human salivary parotid ductal tissue [23]. Furthermore, an NPPB-sensitive Cl− current has been documented in noncultured rat salivary acinar tissue [22]. In addition, hyposmotic stimulation of salivary acinar cells from both Clcn2 knockout and wild-type mice activates a large outwardly rectifying chloride current [25]. The Cl− channel blocker IAA, known to inhibit Cl− channels in mammalian skeletal muscle and epithelial cells, had no effect on RVD in the present study, possibly indicating a species or celltype variation in sensitivity. The concentration of the K+/Cl− symporter inhibitor furosemide required to induce RVD inhibition was much higher than that necessary to inhibit the symporter in other tissues [18]. Thus, it is not unlikely that the response at this high concentration is due to a less specific inhibition of other transporters. Chloride channels directly or indirectly furosemidesensitive have been reported [5, 16, 31]. It cannot be ruled out, however, that this immortalized cell line has lost some of the cells’ original properties so that they now only respond to higher levels of RVD inhibitors. Nonetheless, rat salivary acinar cells have been found to express high levels of the NaKCl2 cotransporter and at least five distinct Cl− channels [37]. All inhibitors that affected RVD also induced a larger than normal cell swelling. This response is most likely due to lack of volume regulation that is normally activated while the cells are still swelling. The inhibitors also brought about a delayed peak swelling. Such a delay is not unexpected when the cellular RVD machinery and swelling counteraction are not functioning properly. The β-amino acid taurine also seems to play a role in cell volume regulation in these cells. The onset and duration of the taurine efflux corresponded well with the volume response. Similar results have been found in other cells [11, 15] and are most likely indications of how cells use non-ionic compounds in combating osmotic variations in their external milieus. Salivary secretion is under the control of the autonomic nervous system. It has been observed that cholinergic stimulation of isolated rat salivary gland acinar cells results in a pronounced cell volume decrease [8, 24]. This reduction appears to be due to Ca2+-stimulated Cl− loss

Pflugers Arch - Eur J Physiol (2007) 453:787–796

[6, 8, 34]. Thus, a similar signal for regulatory volume decrease might also be present in the salivary acinar cells used in this study and as described in many cell types and recently presented in human salivary gland cells [21]. An attempt at reduction in both intra- and extracellular Ca2+ to zero had no effect on cell volume regulation in the present work. Some cells, nonetheless, experienced a sustained increase in the intracellular free Ca2+ concentration following hyposmotic exposure. In about 30% of the cells there was no detectable Ca2+ increase. The reason for this irregularity is not clear. The cells were picked at random, but were nonetheless chosen from a population of attached, actively dividing cells. It could be that the cells’ mitotic stage plays a role in the cellular capacity to mobilize calcium from extra- or intracellular stores. Indeed, cell volumes vary during cell division, and the cells’ osmoregulatory properties could also change in this process. Furthermore, it could be that osmoregulatory properties are not the same in dissociated cells compared with those that are attached. In general, however, the need for Ca2+ influx from the extracellular medium or Ca2+ release from intracellular stores during cell volume regulation varies greatly between cell types and species. Some cells are strictly or partially dependent on an increased intracellular Ca2+ concentration for proper volume regulation [29], whereas many more cell types have no such need [18]. In yet other cells, there is a Ca2+ increase, but it is not necessary for cell volume control to occur [1, 2]. The reason for these differences could be possible variations in the type K+ channels participating in RVD between cell types and species. During KCl and taurine efflux, water follows by osmosis and the cells shrink. This water efflux would most likely occur through water channels. The presence of the water channel AQP5 could be documented in these cells. These aquaporins have been found in a variety of exocrine epithelia, including salivary glands, sweat glands and lacrimal glands. The aquaporins are necessary for proper fluid movement to occur, as demonstrated by the presence of cytoplasmic—but not apical membrane—AQP5 in patients with Sjögren’s syndrome [27]. Furthermore, it appears as though AQP5 is upregulated in mouse lung epithelial cells adapted to hyperosmotic media [14] and submandibular gland tissue samples from hypertonic rats [14]. A similar result was found in the salivary acinar cells in this study, where both AQP5 mRNA and protein were increased by several fold. This increase was dramatically reduced when the MEK inhibitor U0126 was added to the cellular growth medium (inhibits the MAP kinase signalling pathway). The MAP kinase pathway also appears to be important for the aquaporin upregulation during osmotic stress in renal [17, 32] and lung cells [14].

795

In the present study it was found that cells grown for 24 h in hyperosmotic medium both swelled and recovered their volume more than control cells upon an acute hyposmotic shock. Such a response seems inevitable following cellular accumulation of solutes during hyperosmotic growth conditions. When placed in a diluted environment, these cells would experience a higher osmotic water influx. It is not unlikely that such cells also would have a more strongly activated cell volume regulatory signal and should, therefore, also shrink more than control cells. Both ‘hyperosmotic’ cells and control cells shrink to the same basal volume during the 15-min observation period. The cell swelling and volume decrease associated with volume regulation were markedly reduced following a 24-h growth period in the presence of the MEK kinase inhibitor U0126. Such a response would be expected seen in the light of the results above, where inhibition of MEK reduces both AQP5 mRNA and protein levels. Reducing the number of water channels in the cellular membranes should have a negative effect on the cells’ volume regulatory responses. In conclusion, we have characterized osmoregulatory properties of immortalized rat salivary acinar cells, which we hope will give the basis for further work on cell volume regulation in this promising cell line. We found that cell shrinkage following osmotically induced swelling appears to be the result of K+ and Cl− efflux through separate conductive pathways. Taurine efflux may also take part in this process. Neither extra- nor intracellular Ca2+ appears to be necessary for volume regulation to occur. AQP5 is required for proper cell volume regulation to take place, and AQP5 expression is regulated by the extracellular osmolality and the MAP kinase pathway. This study indicates that proper saliva formation depends on functional cell volume regulatory responses and the presence of AQP5. Acknowledgements We are greatly in debt to Dr. David Quissell, Department of Craniofacial Biology, University of Colorado, for providing us with the acinar cell line and to Dr. Peter Agre, John Hopkins University School of Medicine, USA, for supplying the initial anti-rat AQP5 sera. We would also like to thank Dr. Kjell Fugelli, Department of Molecular Biosciences, University of Oslo, Norway, for assistance with the taurine efflux experiments. Excellent imaging service was provided by Hege Avsnes Dale, M.S., at the national technology platform Molecular Imaging Center at the University of Bergen, Norway, and supported by the functional genomics program (FUGE) in the Research Council of Norway.

References 1. Beck JS, Breton S, Laprade R, Giebisch G (1991) Volume regulation and intracellular calcium in the rabbit proximal convoluted tubule. Am J Physiol 260:F861–F867

796 2. Best L, Sheader EA, Brown PD (1996) A volume-activated anion conductance in insulin-secreting cells. Pflügers Arch 431:363–370 3. Delporte C, Steinfeld S (2006) Distribution and roles of aquaporins in salivary glands. Biochim Biophys Acta DOI 10.1016/j.bbamem.2006.01.022 4. Delporte C, Redman R, Baum BJ (1997) Relationship between the cellular distribution of the alpha(v)beta3/5 integrins and adenoviral infection in salivary glands. Lab Invest 77:167–173 5. Evans MG, Marty A, Tan YP, Trautmann A (2003) Blockage of Ca-activated Cl conductance by furosemide in rat lacrimal glands. Pflugers Arch 406:65–68 6. Foskett JK (1990) [Ca2+]i modulation of Cl− content controls cell volume in single salivary acinar cells during fluid secretion. Am J Physiol 259:998–1004 7. Foskett JK (1994) The role of calcium in the control of volume regulation transport pathways. In: Strange K (ed) Cellular and molecular physiology of cell volume regulation. CRC, Boca Raton, pp 259–277 8. Foskett JK, Melvin JE (1989) Activation of salivary secretion: coupling of cell volume and [Ca2+]i in single cells. Science 244:1582–1585 9. Foskett JK, Wong MM, Sue-A-Quan G, Robertson MA (1994) Isosmotic modulation of cell volume and intracellular ion activities during stimulation of single exocrine cells. J Exp Zool 268:104—110 10. Fugelli K, Thoroed SM (1986) Taurine transport associated with cell-volume regulation in flounder erythrocytes under anisosmotic conditions. J Physiol Lond 374:245–261 11. Fugelli K, Kanli H, Terreros DA (1995) Taurine efflux is a cell volume regulatory process in proximal renal tubules from the teleost Carassius auratus. Acta Physiol Scand 155:223–232 12. Gresz V, Kwon TH, Hurley PT, Varga G, Zelles T, Nielsen S, Case RM, Steward MC (2001) Identification and localization of aquaporin water channels in human salivary glands. Am J Physiol Gastrointest Liver Physiol 281:G247–G254 13. Hallows KR, Knauf PA (1994) Principles of cell volume regulation. In: Strange K (ed) Cellular and molecular physiology of cell volume regulation. CRC, Boca Raton, pp 3–29 14. Hoffert JD, Leitch V, Agre P, King LS (24-3-2000) Hypertonic induction of aquaporin-5 expression through an ERK-dependent pathway. J Biol Chem 275:9070–9077 15. Hoffmann EK, Hendil KB (1976) The role of amino acids and taurine in isosmotic intracellular regulation in Ehrlich ascites mouse tumour cells. J Comp Physiol 108:279–286 16. Ishikawa T, Cook D (1993) A Ca2+-activated Cl− current in sheep parotid secretory cells. J Membr Biol 135:261–271 17. Itoh T, Yamauchi A, Miyai A, Yokoyama K, Kamada T, Ueda N, Fujiwara Y (1994) Mitogen-activated protein-kinase and its activator are regulated by hypertonic stress in Madin-Darby canine kidney-cells. J Clin Invest 93:2387–2392 18. Kanli H, Norderhus E (1998) Cell volume regulation in proximal renal tubules from trout (Salmo trutta). J Exp Biol 201:1405–1419 19. Kaplan MD, Baum BJ (1993) The functions of saliva. Dysphagia 8:225–229 20. Krane CM, Melvin JE, Nguyen HV, Richardson L, Towne JE, Doetschman T, Menon AG (2001) Salivary acinar cells from aquaporin 5-deficient mice have decreased membrane water permeability and altered cell volume regulation. J Biol Chem 276:23413–23420

Pflugers Arch - Eur J Physiol (2007) 453:787–796 21. Liu X, Bandyopadhyay B, Nakamoto T, Singh B, Liedtke W, Melvin JE, Ambudkar I (2006) A role for AQP5 in activation of TRPV4 by hypotonicity: concerted involvement of AQP5 and TRPV4 in regulation of cell volume recovery. J Biol Chem 281:15485–15495 22. Majid A, Brown PD, Best L, Park K (2001) Expression of volume-sensitive Cl channels and ClC-3 in acinar cells isolated from the rat lacrimal gland and in submandibular salivary gland. J Physiol 534:409–421 23. Moran A, Turner RJ (1993) Secretagogue-induced RVD in HSY cells is due to K+ channels activated by Ca2+ and protein kinase C. Am J Physiol 265:1405–1411 24. Nakahari T, Murakami M, Sasaki Y, Kataoka T, Imai Y, Shiba Y, Kanno Y (1991) Dose effects of acetylcholine on the cell volume of rat mandibular salivary acini. Jpn J Physiol 41:153–168 25. Nehrke K, Arreola J, Nguyen H-V, Pilato J, Richardson L, Okunade G, Baggs R, Shull GE, Melvin JE (2002) Loss of hyperpolarization-activated Cl− current in salivary acinar cells from Clcn2 knockout mice. J Biol Chem 277:23604–23611 26. Quissell DO, Barzen KA, Gruenert DC, Redman RS, Camden JM, Turner JT (1997) Development and characterization of SV40 immortalized rat submandibular acinar cell lines. In Vitro Cell Dev Biol 33:164–173 27. Steinfeld S, Cogan E, King LS, Agre P, Kiss R, Delporte C (2001) Abnormal distribution of aquaporin-5 water channel protein in salivary glands from Sjogren’s syndrome patients. Lab Invest 81:143–148 28. Steward MC, Seo Y, Murakami M, Seo JT, Larcombe-McDouall JB, Case RM (1998) Regulation of cell volume and diffusion of intracellular water in salivary acinar cells. Eur J Morphol 36:103– 106 29. Terreros D, Kanli H (1992) Role of intracellular calcium in renal proximal tubule cell volume regulation. Am J Physiol 263:1086– 1092 30. Terreros D, Kanli H, Coombs J (1990) Possible role of basolateral cell membrane in proximal renal tubule osmoregulation. Am J Physiol 258:1022–1033 31. Uchida S, Sasaki S, Nitta K, Uchida K, Horita S, Nihei H, Marumo F (1995) Localization and functional characterization of rat kidney-specific chloride channel. J Clin Invest 95:104–113 32. Umenishi F, Schrier RW (2003) Hypertonicity-induced aquaporin1 (AQP1) expression is mediated by the activation of MAPK pathways and hypertonicity-responsive element in the AQP1 gene. J Biol Chem 278:15765–15770 33. Walz T, Smith BL, Zeidel ML, Engel A, Agre P (1994) Biologically active two-dimensional crystals of aquaporin CHIP. J Biol Chem 269:1583–1586 34. Wong MM, Foskett JK (1991) Oscillations of cytosolic sodium during calcium oscillations in exocrine acinar cells. Science 254:1014–1016 35. Zeidel ML, Ambudkar SV, Smith BL, Agre P (1992) Reconstitution of functional water channels in liposomes containing purified red cell CHIP28 protein. Biochemistry 31:7436–7440 36. Zeidel ML, Nielsen S, Smith BL, Ambudkar SV, Maunsbach AB, Agre P (1994) Ultrastructure, pharmacologic inhibition, and transport selectivity of aquaporin channel-forming integral protein in proteoliposomes. Biochemistry 33:1606–1615 37. Zeng W, Lee MG, Muallem S (1997) Membrane-specific regulation of Cl− channels by purinergic receptors in rat submandibular gland acinar and duct cells. J Biol Chem 272:32956–32965