| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
* Institut für Pharmakologie, Charité Campus Benjamin Franklin, Berlin, Germany;
Fachbereich Biologie, Chemie, Pharmazie, Freie Universität Berlin, Berlin, Germany;
Institut für Klinische Physiologie, Charité Campus Benjamin Franklin, Berlin, Germany; and
Medizinische Klinik I Gastroenterologie, Infektiologie und Rheumatologie, Charité Campus Benjamin Franklin, Berlin, Germany
1Correspondence: Institut für Pharmakologie, Charité Campus Benjamin Franklin, Thielallee 69–73, 14195 Berlin. E-mail: christian.harteneck{at}charite.de
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Key Words: transepithelial electrical resistance • paracellular pathway • transcellular pathway • calcium homeostasis • BK channel
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
. These proteins assemble into functional complexes shared by the lateral plasma membranes of two adjacent cells and form tight junction, adherens junction, desmosomal, and gap junction domains. In addition, association of transmembrane proteins with cytosolic scaffold proteins like the zonula occludens proteins ZO-1, -2, and -3 link the actin cytoskeleton to integral membrane proteins and recruit signaling molecules (2)
. Tight junctions seal the space between single cells, but also allow for paracellular exchange of ions and other small solutes. They also restrict diffusion of lipids and proteins between apical and basolateral plasma membrane thus maintaining polarization of epithelial cells. A broad range of disease states such as inflammation and cancer are accompanied by a breakdown of the epithelial barrier (3)
.
Expression of TRPV4, a member of the transient receptor potential (TRP) superfamily of nonselective cation channels, has been found in many different epithelial cells. The TRP family is classified into at least three subfamilies: TRPC, TRPM, and TRPV (4
, 5)
. The TRPV channels appear to be integrators of physical and chemical stimuli activated by altered temperature, changes in extracellular osmolarity, and chemical ligands like capsaicin for TRPV1 and 2-aminoethoxydiphenyl borate (2-APB) for TRPV1, TRPV2, and TRPV3 (6
7
8)
. The epithelial calcium channels TRPV5 and TRPV6 are highly calcium selective and play a role in controlling calcium homeostasis in epithelia (9)
. However, agonists activating TRPV5 and TRPV6 have not yet been described.
TRPV4 is activated by extracellular hypoosmolarity, moderate temperature (>25°C), phorbol esters, the endocannabinoid anandamide and its metabolites arachidonic acid, and epoxyeicosatrienoic acid (10
11
12
13)
. High expression of TRPV4 has been found in the kidney, where it localizes to the basolateral side of water-impermeable nephron segments (10
, 14)
. Furthermore, TRPV4 has been described in airway epithelia lining the trachea and the lung alveoli (15
, 16)
, in the plexus choroideus of the circumventricular organ in the brain (11)
, in endothelial cells (17)
, in keratinocytes of the skin (18)
, in sweat glands (15)
, and in cochlear hair cells (11)
.
Here we functionally characterize TRPV4 in epithelial cells of the mouse mammary gland cell line HC11. Cultured on filter membrane supports in the presence of differentiation medium (19)
, the cells form epithelial layers serving as an in vitro model for epithelia. TRPV4 was selectively localized in the basolateral cell membrane. Activation of TRPV4 by a phorbol ester resulted in a fast decrease of transepithelial electrical resistance. This decrease was mediated by an instantaneous transcellular conductance provided by the basolaterally localized TRPV4, which increased intracellular calcium concentration, and thus subsequently activated large conductance calcium-activated potassium (BK) channels. The changes in transcellular conductance were accompanied by a slower increase in paracellular permeability for small molecules, fluorescein or mannitol, resulting from a rearrangement of the tight junction structure. In summary, our data show that TRPV4 regulates transcellular as well as paracellular permeability in HC11 cells.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
-phorbol-12,13-didecanoate (4-PDD), ruthenium red, fluorescein sodium and capsaicin were from Sigma (Steinheim, Germany). 2-Aminophenoxyborate was from Tocris (Avonmouth, UK). The anti-ZO-1, anti-occludin, anti-E-cadherin, and anti-claudin-1, -2, -3, -4, -5, -7, -8, -14, -15, -16 antibodies were from Zytomed (Berlin, Germany). The anti-ZO-2, anti-ZO-3, and anticlaudin-10 antibody (Ab) were purchased from Santa Cruz Biotechnology (Santa Cruz, CA, USA). The anti-tubulin Ab and the anti-BK channel Ab were from Lab Vision (Fremont, CA, USA) and Alomone Labs (Jerusalem, Israel), respectively. The Alexa Fluor 488-linked donkey anti-rabbit IgG conjugate and Alexa Fluor 594-linked donkey anti-mouse IgG conjugate were from Molecular Probes (Leiden, The Netherlands). Transwell membrane supports, clear polyester membrane with a 3.0 µm pore size, were from Costar (Cambridge, MA, USA). Millicell-PCF membrane supports with a 3.0 µm pore size were from Millipore (Carrigtwohill, Cork, Ireland).
Cell culture
HC11 cells were cultured in medium composed of RPMI 1640 medium containing 10% FCS, 5 µg/ml insulin, 10 ng/ml epidermal growth factor (EGF), 2 mM L-glutamine, 100 U/ml penicillin, and 100 µg/ml streptomycin. The cells were cultured at 37°C in an atmosphere of 5% CO2 and 95% air. Differentiation of the cells was achieved by replacing the growth medium with differentiation medium varying from growth medium in the omission of EGF and addition of 5 µg/ml prolactin and 1 µM dexamethasone. For single cell calcium imaging experiments, cells were seeded on coverslips and used for experiments 1 to 3 days later. HC11 cell were seeded in growth medium at a density of 130,000 and 240,000 cells in 6.5 mm and 12 mm filters, respectively. Medium was replaced by differentiation medium after 2 days to obtain differentiated cell layers. The transepithelial electrical resistance ("TEER", Rt) was measured across cell monolayer using an EVOM voltohmmeter (World Precision Instruments, Sarasota, FL, USA). Cell layers with transepithelial electrical resistances > 0.7 k
cm2 were used for the experiments.
Immunofluorescence, Western blot analysis, and RT-polymerase chain reaction (RT-PCR)
Epithelial cell layers grown on Transwell supports were fixed with methanol/acetic acid (97:3). After neutralization of unspecific binding sites by incubation with 3% BSA in PBS, the specimens were incubated with the primary antibodies at 37°C for 90 min. The staining was visualized by Alexa Fluor-488-linked donkey anti-rabbit-IgG conjugate and/or Alexa Fluor-594-linked donkey anti mouse-IgG conjugate. Images were acquired with a Zeiss confocal laser scanning microscope (Zeiss, Jena, Germany). Lysates and membrane fractions from untreated and 4-PDD- or ruthenium red-treated HC11 cells were subjected to SDS-PAGE on 8% and 14% gels according to Laemmli (20)
. The proteins were transferred to nitrocellulose membranes and the blots were probed with the anti-TRPV4, anti-E-cadherin, anti-occludin or anti-claudin antibodies as described recently (21)
. Total RNA was extracted from HC11 cells using RNeasy kits (Qiagen, Hilden, Germany) according to the manufacturer’s protocol. RNA was transcribed by the RNase H-deficient reverse transcriptase of Moloney murine leukemia virus (Promega, Madison, WI, USA). PCR analysis was performed using TRPV-specific oligonucleotides (see Supplemental Table 1). As a positive control, glyceraldehyde-3-phosphate dehydrogenase was amplified using gene-specific primers.
Measurements of [Ca2+]i
Measurements of the intracellular Ca2+ concentration ([Ca2+]i) in single cells using the Fura-2-AM were performed as described previously (10
, 21)
. At least 20 cells were measured on the same coverslip, representing one independent experiment.
Patch clamp measurements
Membrane currents were recorded in the whole-cell configuration of the patch-clamp technique, using an Axopatch 200 B amplifier (Axon Instruments, Union City, CA, USA), subsequently low-pass filtered at 1 kHz, digitized with a sampling rate of 5 kHz, and analyzed using pCLAMP software (version 7.0; Axon Instruments). The pipette resistance varied between 3 and 5 M. Pipettes were filled with a solution composed of 50 mM CsCH3O3S, 25 mM CsCl, 10 mM 1,2-bis(2-aminophenoxy)ethane-N,N,N',N'-tetraacetic acid (BAPTA), 4.6 mM CaCl2, 4 mM Na2ATP, 2 mM MgCl2, and 10 mM HEPES (pH 7.2 with 
52 mM CsOH). The concentration of free Ca2+ in this solution was calculated to be 100 nM using the software WinMAXC (v.2.05; Chris Patton, Stanford, CA, USA). The bath solution contained 140 mM NaCl, 2 mM CaCl2, 1 mM MgCl2, 10 mM glucose (Glc), and 10 mM HEPES (pH 7.4 with NaOH). Whole-cell currents were elicited by voltage ramps from –100 mV to +100 mV (400 ms duration) applied every 2 s from a holding potential of 0 mV. Pooled data from patch-clamp experiments are expressed as means ± SEM from n cells.
Characterization of epithelial barrier function
HC11 grown on filter inserts (MillicellTM-PCF, Millipore, Bedford, MA, USA) were mounted in Ussing chambers (22)
. Water-jacketed gas lifts were filled with 7 ml circulating bath solution containing (mM): NaCl 113.6, KCl 5.4, CaCl2 1.2, MgCl2 1.2, Na2HPO4 2.4, NaH2PO4 0.6, NaHCO3 21, Glc 10, equilibrated with 5% CO2 in O2. Under these conditions, short circuit current (ISC, µA/cm2 alternate µAcm–2) and transepithelial resistance (Rt, ·cm2) were determined. Paracellular flux was determined by application of fluorescein to the apical compartment of the Ussing chamber (250 µg/ml final concentration). At different time points, aliquots (100 µl) were taken from the basolateral compartment. Fluorescein concentration was determined in a fluorometer (excitation 485 nm, emission 535 nm; Mithras LB940, Berthold Technologies, Bad Wildbad, Germany).
Mannitol flux
Measurement of unidirectional tracer flux from the apical to the basolateral side and vice versa was performed under short circuit conditions in the presence of 10 mM mannitol on both sides and the addition of 25 kBq/ml of [3H]-mannitol (Biotrend, Cologne, Germany) to the donor side. Samples (100 µl and 1 ml) were taken from the donor and the receiving side, respectively, to determine starting activities. Upon application of 4-PDD, 1 ml samples were taken from the receiving side every hour and replaced with fresh Ringer’s solution. Four milliliters of Ultima Gold high flash point liquid scintillation cocktail (Packard Bioscience, Groningen, The Netherlands) was added. All 5 ml samples were subsequently analyzed with a Tri-Carb 2100TR Liquid Scintillation counter (Packard, Meriden, CT, USA).
Impedance measurements
Impedance analysis was performed as described previously to distinguish between changes in the resistance of the paracellular and transcellular pathways (23)
This method is based on a model describing the apical and the basolateral membrane of the confluent cell layer each as a resistor and a capacitor in parallel (RA, CA, and RB, CB, respectively). The paracellular pathway is represented by a resistor (RSh) that shunts these two RC circuits and the filter support by a resistor in series (RS) (24)
. After application of alternating current (35 µA cm–2, frequency range 1.3 Hz to 65 kHz), changes in tissue voltage were detected by phase-sensitive amplifiers (402 frequency response analyzer, Beran Instruments, Glen Allen, VA, USA; 1286 electrochemical interface; Solartron Schlumberger, Atlanta, GA, USA). Complex impedance (Zreal, Zimaginary) values were calculated and plotted in a Nyquist diagram. As long as apical and basolateral membranes do not clearly differ (RA 
RB and CA CB), this plot yields a semicircle. For RA 
RB the curve is deformed (minimum shifted to smaller or higher values). To facilitate discrimination between changes in RA or RB on the one hand and RSh on the other, values of the Nyquist diagram were normalized with respect to total resistance (RT, measured at the lowest frequencies). Initial values of RSh were estimated from experiments during which impedance spectra and fluorescein fluxes were obtained before and after chelating extracellular Ca2+ with EGTA. This caused tight junctions to open and increased fluorescein flux by a factor of 38 ± 3 (n=5). This factor was taken to also reflect the decrease in RSh. From these experiments, initial values of RSh were calculated and found to depend linearly on RT. This relationship was used to estimate RSh in all further experiments. RA and RB were subsequently estimated from experiments during which 50, 100, and 150 U/ml nystatin were applied to the basolateral side of the epithelium, thus decreasing RB.
In the present study, a moderate decrease in RSh would be expected to decrease the absolute value of RT without changing the shape of the normalized curve. In contrast, a decrease in RB would not only cause a decrease in RT but also shift the minimum of Zimaginary toward lower or higher values of Zreal. This shift can be reversed by a subsequent decrease in RShif it is sufficiently large to short circuit the transcellular pathway.
An appendix containing the detailed description of the impedance analysis is available from D. G. upon request (dorothee.guenzel@charite.de).
Freeze fracture electron microscopy
Freeze fracture analysis was carried out as described earlier (25)
. Briefly, epithelial layers were fixed with phosphate-buffered 2% glutaraldehyde. After incubation in 10% and 30% glycerol, tissues were frozen in Freon 22 cooled with liquid nitrogen. Subsequently, tissues were fractured at –100°C and shadowed with platinum and carbon in a Denton DV-502 vacuum evaporator (Cherry Hill, NJ, USA). Replicas were bleached in sodium hypochlorite, picked up on grids (Ted Pella Inc., Tustin, CA, USA), and examined in a Zeiss electron microscopy (EM) 902 electron microscope (Carl Zeiss AG, Oberkochen, Germany). Morphological analysis was performed on at least 10 TJ regions per condition in which both an apical and a contra-apical strand of the meshwork could be clearly demarcated.
Statistical analysis
Measured values are presented as means ± SEM. Differences between means were analyzed by the Student’s t test.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
. Upon removal of EGF and addition of prolactin and dexamethasone to the medium (differentiation medium), the cells differentiate and form epithelial layers of high transepithelial electrical resistance (19)
. We analyzed the expression of TRPV cation channels in undifferentiated HC11 cells by RT-PCR (Fig. 1
A). The semiquantitative analysis revealed low expression of TRPV1 and TRPV2, moderate expression of TRPV3, TRPV5, and TRPV6, and high expression of TRPV4. Analysis of differentiated HC11 cells showed comparable results with the exception of TRPV5, which was not detected (data not shown).
|
Expression and localization of TRPV4 in HC11 cells was studied by immunofluorescence experiments staining HC11 cells with an anti-TRPV4 Ab (Fig. 1B-D
, Supplemental Fig. 1). In HC11 cells cultured in growth medium, most of the TRPV4 protein was pooled in intracellular compartments in single (Fig. 1B
) and clustered (Fig. 1C
) cells, whereas in confluent HC11 cells the staining was prominent in the plasma membrane (Fig. 1D
). HC11 cell layers cultured for at least 3 days in the presence of differentiating medium were used to clarify the localization of TRPV4. To this end, we performed immunofluorescence experiments using antibodies directed against occludin, E-cadherin, and TRPV4 (Fig. 1E
). Occludin and E-cadherin were found to be differentially distributed in plasma membranes. Occludin was localized in a small distinct region in the apical portion of the lateral membrane, defining the junctional region. In contrast, anti-E-cadherin and the anti-TRPV4 antibodies stained regions of the lateral plasma membrane more distant from the junctional region (see Fig. 1E
). The pattern of TRPV4 staining, which excluded an apical distribution, was quite similar to that of E-cadherin. Therefore, we performed double labeling experiments using the polyclonal anti-TRPV4 rabbit Ab and the monoclonal mouse antibodies raised against all cadherin proteins (Fig. 1F
) and E-cadherin (Fig. 1G
). The experiments showed that the anti-TRPV4 Ab and both anti-cadherin antibodies stained similar regions of the plasma membrane, indicating a basolateral localization of TRPV4 in HC11 cells. In summary, the mouse mammary gland line HC11 expresses TRPV4 basolaterally localized in differentiated cell layers.
Functional characterization of TRPV4 in single HC11 cells
We performed calcium imaging experiments on single undifferentiated HC11 cells using different stimuli of the calcium-permeable channels TRPV1, TRPV2, TRPV3, and TRPV4. The TRPV1 activator capsaicin (10 µM) did not change the intracellular calcium concentration [Ca2+]i (Fig. 2
A). Treatment with 2-APB (1 mM), an activator of TRPV1, TRPV2, and TRPV3, also had no effect on [Ca2+]i (see Fig. 2A
). In contrast, different stimuli resulting in an activation of TRPV4 were able to increase [Ca2+]i in HC11 cells. Figure 2A
summarizes the effects of hypotonic stimulation, arachidonic acid, and 4-PDD. The elevation of [Ca2+]i due to hypotonic stimulation was largely suppressed by addition of the TRPV channel blocker ruthenium red (see Fig. 2A
) and was fully abolished by chelating extracellular Ca2+ with EGTA (Fig. 2A, B
), indicating Ca2+ entry via the plasma membrane. The increase in [Ca2+]i evoked by application of arachidonic acid (Fig. 2C
) was more transient than during hypotonic treatment, but mean values for changes in [Ca2+]i were similar for either stimulus (see Fig. 2A
). Application of 10 µM 4-PDD resulted in a large increase in [Ca2+]i, which could also be blocked by addition of ruthenium red (Fig. 2A, D
) or EGTA (see Fig. 2A
). Responses to 4-PDD showed a clear concentration dependence, with maximal effects at 10 µM (see Fig. 2A
). At 5 µM or below, 4-PDD-induced increases in [Ca2+]i that were even smaller than those in response to hypotonicity or arachidonic acid. The experiments provide evidence for the functional expression of TRPV4 in mammary epithelial cells. Due to its high specificity for TRPV4 and its large effects at 10 µM in HC11 cells, 4-PDD is a valuable pharmacological tool for investigation of TRPV4 in this cell line.
|
To further verify the involvement of TRPV4 channels in 4-PDD-induced increases in [Ca2+]i, we analyzed ion currents in single, undifferentiated HC11 cells. In 10 of 21 cells tested, extracellular application of 4-PDD (10 µM) induced current responses, characterized by a relative long latency of 60–120 s for full activation (Fig. 2E
). The mean current-voltage relationship obtained from whole-cell currents in the presence of 4-PDD displayed a slight outward rectification (see Fig. 2E
) and a reversal potential of –13 ± 10 mV (n=8). The mean current densities were –9.4 ± 3.4 pA/pF and +22.8 ± 5.6 pA/pF (n=8) at –100 mV and +100 mV, respectively. Before stimulation with 4-PDD, current densities were –0.8 ± 0.2 pA/pF and +3.2 ± 0.4 pA/pF (n=8). Analysis of the 4-PDD-sensitive current component by subtracting untreated currents from currents during 4-PDD treatment revealed a reversal potential of +5 ± 3 mV (n=8). Furthermore, we tested the ability of ruthenium red to inhibit currents in HC11 cells. Figure 2F
shows the reversible inhibition of 4-PDD-induced currents by 10 µM ruthenium red. Inward currents were almost completely blocked (92±3%, n=3), whereas outward currents were only partially inhibited (55±9%, n=3). Together, the delayed activation by 4-PDD, the slight outward rectification, the positive reversal potential close to 0 mV and the voltage-dependent block by ruthenium red are typical for currents through the nonselective cation channel TRPV4 (12
, 26)
. The data show that HC11 cells express a cation channel with the electrophysiological and pharmacological properties characteristic for TRPV4.
TRPV4-mediated changes of transepithelial resistance
HC11 cells cultured on membrane filters differentiate and form epithelial cell layers that were monitored by measuring the transepithelial electrical resistance ("TEER", Rt). Two days after seeding the medium was exchanged from growth to differentiation medium, causing Rt to increase to values of at least 0.7 k cm2. To study possible modulation of epithelial permeability by the TRPV4 activator 4-PDD, we performed time-resolved Rt measurements in Ussing chambers. Surprisingly, the application of the TRPV4 activator 4-PDD (10 µM) resulted in a rapid (within 5 min) and sustained decrease in Rt (Fig. 3
A). The rate of onset depended on the site of activation with a fast onset after basolateral application of 4-PDD (data not shown). The decrease in Rt was accompanied by an initial transient current (mean duration 3.0±0.5 min, n=11) with a maximum of 1.5 ± 0.3 µA/cm2 (n=11), followed by a sustained current of –2.4 ± 0.3 µA/cm2 (n=11, negative current is equivalent to negative charge moving from the basolateral to the apical side). Neither the decrease in Rt nor the currents were observed in epithelia treated with the combination of 4-PDD (10 µM) and the TRPV4 inhibitor ruthenium red (10 µM) (Fig. 3B
) or in untreated HC11 cell layers (Fig. 3C
). Application of ruthenium red alone also had no effect on Rt or current (data not shown).
|
To test whether the decrease in Rt is mediated by cytotoxic effects or by modulation of epithelial function, we performed trypan blue staining of cell layers. The number of blue cells were indistinguishable between the different experimental settings (control vs. 4-PDD vs. 4-PDD + RR). Furthermore, the final Rt values of the filters in culture were above values of the empty membrane filters even after 24 h in the presence of 4-PDD. These data and the immunofluorescence images support the idea that 4-PDD changes epithelial resistance via TRPV4 activation.
An involvement of protein kinase C activation in the 4-PDD-mediated Rt decrease was excluded by coapplication of 4-PDD and one of the protein kinase C inhibitors BIM I, calphostin C, or Gö 6976 (1 µM each). The compounds were unable to inhibit the 4-PDD-induced Rt decrease (data not shown).
The rapid alteration of overall transepithelial resistance (Rt) was analyzed by impedance measurements in order to distinguish between transcellular and paracellular components (Fig. 3D-I
). In measurements carried out under control conditions (untreated and in the presence of 4-PDD plus RR), transepithelial resistance measured under DC conditions slowly decreased over time (Fig. 3B, C
). Accordingly, the amplitudes of the two components of the complex impedance, Zreal and Zimaginary, also decreased (Fig. 3E, F
). Despite this decrease, the impedance curves retained their general shape, as seen when normalized (Fig. 3H, I
).
The addition of 4-PDD to the bath solution on the apical and basolateral side (or on the basolateral side alone) of the cell layer caused an instantaneous, pronounced drop in transepithelial resistance (Fig. 3A
). Flux data obtained under these conditions (see below) indicate that fluorescein or mannitol permeability constantly increases after the application of 4-PDD (
Fig. 5
A, B). This increase in flux was assumed to reflect the decrease in paracellular resistance (RSh). Based on this assumption, detailed impedance analysis showed that the initial drop in Rt was due almost exclusively to a decrease in apical (RA) and basolateral (RB) resistance (Fig. 5C
). In Fig. 3G
this is reflected by the change in the shape of the normalized curve (dotted line) obtained 5 min after the application of 4-PDD compared with the curve obtained before the addition of 4-PDD (black line). The deviation in shape implies that different conductive transport mechanisms in the apical (RA) and basolateral (RB) membrane are activated. Over a period of 5 h, epithelial resistance continued to decrease. As shown in Fig. 5C
, this decrease is due to a decrease in both paracellular and transcellular resistances. Thus, after several hours (Fig. 3G
, curve 4) the impedance curve adapted to the shape prior to 4-PDD application (black and gray lines). This observation can be explained by a bypass of the transcellular pathways due to the increase in paracellular ion permeability. The effects of 4-PDD on transepithelial resistance and on the shape of the impedance curves were completely blocked by an earlier application of ruthenium red (Fig. 3E, H
). In summary, these data show that in HC11 cells the application of 4-PDD and activation of TRPV4 influences both transcellular and paracellular pathways.
|
|
4-PDD–induced transcellular conductance across HC11 layers
To characterize the components of the transcellular ion conductance, we tested various blocking substances for the identification of other channels or transporters contributing to changes in transcellular pathways induced by 4-PDD. A candidate for an essential role in transepithelial transport could be the large conductance Ca2+-activated K+ (BK) channel (27)
. Indeed, Ba2+ as well as tetraethylammonium were able to inhibit the 4-PDD-induced Rt decrease arguing for the involvement of BK channels (data not shown). To verify the participation of this channel type in the transcellular pathway, we used the specific BK channel blocker paxilline. In calcium imaging experiments, calcium entry stimulated by 4-PDD was not altered by paxilline (data not shown), whereas paxilline (5 µM) resulted in the inhibition of a 4-PDD-induced Rt decrease (Fig. 4
A, B). The expression of BK channels in undifferentiated and differentiated HC11 cells was further studied by Western blot analyses using an anti-BK channel Ab. Figure 4C
clearly shows the expression of a protein of 110 kDa, which is in good agreement with other published data reporting the expression of apically localized BK channels in epithelial cells (28)
. The detection reaction was specific, as coincubation with the immunogenic peptide was able to suppress detection. The data argue for an up-regulation of BK channels during differentiation (Fig. 4C
). In summary, these data show that TRPV4 and BK channels mediate the transcellular pathways initiated by the activation of TRPV4.
TRPV4-induced changes in the paracellular pathway
The paracellular permeability of 4-PDD-treated cell layer was further characterized by measuring transepithelial flux of fluorescein, a molecule of 332 Da. In cell layers incubated with 4-PDD, the decrease in Rt was paralleled by a dramatic increase in fluorescein flux within the first hour of incubation (Fig. 5A
). In contrast, the flux of fluorescein was small in all other untreated and treated cell layers (Fig. 5A
). Fluorescein as fluorescent compound allowing highly sensitive measurements of paracellular transport mechanism has been described to be transported in some cell layers by transcellular processes. To confirm that TRPV4 activation actually affected the paracellular permeability of HC11 cell layers, we measured transepithelial flux of [3H]-mannitol (Fig. 5B
). Independent of the application compartment, increases in [3H]-mannitol were measured in the opposite compartments (see Fig. 5B
). After 6 h, the fluorescein and mannitol experiments resulted in comparable increases in flux of 10-fold. Based on these flux data, impedance analysis allowed discrimination of the components of the transepithelial resistance. Figure 5C
shows that 4-PDD primarily affected the transcellular pathway (RA and RB).
Inhibitors paxilline and ruthenium red were both effective in inhibiting flux and transepithelial electrical resistance within the initial time period (Table 1
). After 24 h, ruthenium red partially restored the 4-PDD-induced decrease in transepithelial resistance, whereas paxilline failed to block the decrease in transepithelial permeability completely (see Table 1
).
|
To test whether structural changes of the tight junction (TJ) have caused the alterations in paracellular permeability, we performed a freeze fracture analysis to investigate the TJ architecture. In accordance with their high resistance, untreated HC11 cells comprised a complex network of 10 horizontally oriented TJ strands, which were mostly continuous and showed only infrequent small strand breaks (Fig. 5D
). In sharp contrast, HC11 cells treated with 4-PDD exhibited a pattern of completely disrupted TJs, with discontinuous strands (Fig. 5D
, inset) and frequent large strand breaks. In accordance with the functional data, a combined treatment of ruthenium red and 4-PDD resulted in a TJ pattern indistinguishable from untreated HC11 cells. The data show that upon TRPV4 activation, the epithelial layers lost their sealing properties and became permeable for small compounds like fluorescein and mannitol. The increase in fluorescein and mannitol permeability was paralleled, and thus presumably caused, by a change in tight junction structure.
Molecular basis of tight junction alterations
To investigate the molecular basis of tight junction alterations after 4-PDD treatment, we studied the expression of ZO-1, occludin, E-cadherin, and claudin proteins in membrane fractions of HC11 cells (Supplemental Fig. 2). The total content of each protein was analyzed by Western blot from lysates prepared from differentiated control, 4-PDD-, ruthenium red-, or 4-PDD- and ruthenium red-treated HC11 cells (Fig. 6
A). The protein quantity loaded onto each lane was standardized and controlled by a monoclonal anti-tubulin Ab (see Fig. 6A
). Results on expression of tight junction proteins in HC11 cells can be roughly grouped into three patterns. First, claudin-2, -10, -14, and -18 were not detectable in HC11 cells at all. Second, ZO-1, ZO-2, occludin, claudin-15 and -16 were constantly expressed. Third, claudin-1, -3, -4, -5, -7, and -8 were dramatically reduced in 4-PDD-treated HC11 cells compared with controls. All of the latter proteins are barrier-forming claudins. The most striking effect was seen for claudin-4. Therefore, we studied the expression of claudin-4 by immunofluorescence in epithelial cell layers grown on membrane supports (Fig. 6B
). The staining of claudin-4 was indistinguishable in untreated or 4-PDD- and ruthenium red-treated cell layers, whereas signals were reduced in 4-PDD-treated HC11 cells. These data indicate a 4-PDD-induced degradation or down-regulation of claudin proteins with sealing properties, thus mediating the increase in paracellular permeability for medium-sized solutes on TRPV4 activation.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
.
TRPV4 is known to be activated by stimuli such as low extracellular osmolarity, arachidonic acid, or by application of the agonist 4-PDD (10
11
12
13)
. In the present study, all of these stimuli induced increases in [Ca2+]i in HC11 cells. These increases could be demonstrated to be due to calcium entry from the basolateral side of the cell layer, as they were blocked by the TRPV4 inhibitor ruthenium red and were absent if extracellular Ca2+ was buffered by EGTA. As depicted in Fig. 7
, the increase in [Ca2+]i was followed by two independent chains of events. First, the [Ca2+]i increase caused the activation of the Ca2+-sensitive BK channels located in the apical membrane of the HC11 cells, resulting in an instantaneous dramatic increase in transcellular ion conductance. Second, over a period of several hours, TRPV4 activation caused an increase in paracellular permeability for small solutes (a further decrease in transepithelial resistance, increase in fluorescein and mannitol flux). This increase in paracellular permeability could be attributed to the striking increase in the number of TJ strand breaks as seen in freeze fracture analysis and the accompanying down-regulation of claudin expression.
|
Application of paxilline initially blocked the rapid 4-PDD-induced decrease in transepithelial resistance, but was ineffective in long-term (24 h) incubation experiments. In contrast, both the transcellular and paracellular components were affected by the TRPV4 inhibitor ruthenium red.
Based on genome analysis, a total of 24 claudin proteins are known in humans, mice, and rats (29)
. The composition of expressed claudin proteins forming the TJ strands is associated with the sealing properties of the epithelia. The change in expression pattern allows the cell to adapt permeability on physiological conditions. For example, expression of claudin-4 or claudin-8 increases sealing properties of the epithelia (30
, 31)
, whereas expression of claudin-2 results in a decrease due to cation channel-forming characteristics of claudin-2 (23
, 32)
. We investigated the expression of 12 different claudin proteins by Western blot analysis and identified 8 claudins in HC11 cells, namely, claudin-1, -3, -4, -5, -7, -8, -15, and -16. The absence of the pore-forming claudin-2 in HC11 as well as the presence and 4-PDD-induced down-regulation of claudin-4 and -8 explains the high transepithelial resistance in untreated HC11 layers and low values in 4-PDD-treated layers.
Physiological TRPV4 activators such as arachidonic acid and 5',6'-epoxyeicosatrienoic acid are known to be involved in a variety of signal transduction pathways, such as the relaxation of coronary arteries via potassium channel activation (33
, 34)
. In the present study of HC11 cells, we used the phorbol ester 4-PDD as a more selective TRPV4 activator. The effect of 4-PDD is independent of protein kinase C isoforms as the 4-stereo isomers of phorbol esters are unable to activate these proteins. Furthermore, neither inhibitor of protein kinase C (BIM I, calphostin C, or Gö 6976) affected the 4-PDD-induced, TRPV4-mediated Rt decrease. For specific activation of TRPV4, 4-PDD appears to be most appropriate, because other TRP channels—e.g., TRPV2 (35)
and TRPM3 (21)
—have been described to be activated by extracellular hypotonicity but to be insensitive to phorbol esters.
The data show for the first time that TRPV4, a hypotonicity-activated channel, contributes to regulation of epithelial permeability by a rapidly induced transcellular ion pathway as well as by delayed modulation of the paracellular pathway. These findings obtained from a mouse mammary gland epithelial cell line, as an in vitro model for epithelial cell layers expressing TRPV4, may be helpful in understanding the contribution of TRPV4 in the control of the tonicity of body fluids separated by the epithelia. In vitro models of epithelia expressing TRPV4 like the epithelia of the thick ascending limb of Henle’s loop (TALH) and the distal convoluted tubules (DCT) (14)
, as well as in vitro models of endothelial cells (17)
and keratinocytes (18)
, will allow to answer the question whether or not the regulation of barrier permeability by TRPV4 is a general phenomenon.
Furthermore, our findings may contribute to an understanding of increased permeability in inflammation and cancer. During inflammation processes, increased levels of arachidonic acid activating TRPV4 may result in increased epithelial permeability and subsequent exudation and edema formation described in the pathophysiology of mastitis (36)
. New and known inhibitors of TRPV4 will help us to understand the role of TRPV4 in this process, and TRPV4 may provide a new target in the therapy of inflammatory processes.
In summary, our data clarify the cellular role of TRPV4 in epithelial cells by demonstrating that TRPV4 activation results in modulation of epithelial permeability by alterations of both transcellular and paracellular pathways.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
Received for publication January 16, 2006. Accepted for publication April 17, 2006.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  R. N. Willette, W. Bao, S. Nerurkar, T.-l. Yue, C. P. Doe, G. Stankus, G. H. Turner, H. Ju, H. Thomas, C. E. Fishman, et al. Systemic Activation of the Transient Receptor Potential Vanilloid Subtype 4 Channel Causes Endothelial Failure and Circulatory Collapse: Part 2 J. Pharmacol. Exp. Ther., August 1, 2008; 326(2): 443 - 452. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 V. K. Sidhaye, K. S. Schweitzer, M. J. Caterina, L. Shimoda, and L. S. King Shear stress regulates aquaporin-5 and airway epithelial barrier function PNAS, March 4, 2008; 105(9): 3345 - 3350. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. A. Gradilone, A. I. Masyuk, P. L. Splinter, J. M. Banales, B. Q. Huang, P. S. Tietz, T. V. Masyuk, and N. F. LaRusso Cholangiocyte cilia express TRPV4 and detect changes in luminal tonicity inducing bicarbonate secretion PNAS, November 27, 2007; 104(48): 19138 - 19143. [Abstract] [Full Text] [PDF]
|
|
| ||||||||
