Shear mechanical force induces an increase of intracellular Ca²⁺ in cultured Merkel cells prepared from rat vibrissal hair follicles

Abstract

Merkel cells have been proposed to play a role in mechanical transduction of light touch in mammals. In the present study, Merkel cells were prepared from upper segments of rat vibrissal hair follicles and maintained in culture. Reponses of these cells to shear mechanical forces were examined by Ca²⁺ imaging technique. Shear forces of ≥0.8 dyn/cm² that were delivered to the cells by the application of normal bath solution significantly increased intracellular Ca²⁺ levels ([Ca²⁺]_i) in some of these cells, and up to 30% cells responded to 1.6 dyn/cm² shear force applied for 20 s. Gd³⁺ (100 μM), a compound widely used to inhibit mechanically activated channels, abolished shear force-induced increases of [Ca²⁺]_i in these cells. Reduction of extracellular Ca²⁺ concentration from 2 mM to 0.2 mM also abolished shear force-induced increases of [Ca²⁺]_i in these cells. In addition to shear force, we found that many shear force-responding cells also responded to hypotonic solution. However, the response to hypotonic solution was not abolished by Gd³⁺ (100 μM). We also found that all shear force-responding cells responded to ATP (100 μM) with large increases of [Ca²⁺]_i. The responses to ATP remained in the presence of Gd³⁺. Taken together, our results suggest that Merkel cells in culture are sensitive to shear force stress, osmotic, and chemical stimuli and that shear force-induced increases of [Ca²⁺]_i may be mediated by the activation of mechanically activated channels.

merkel cells were first identified and described by Friedrich Sigmund Merkel in 1875 and were thought to be “touch cells” (Merkel 1875). These cells are found at touch-sensitive areas of glabrous and hairy skin as well as the inner cell layer of the superior enlargement of hair follicles of mammals. Merkel cells are innervated by large myelinated (Aβ) axonal neurites, and their contact points morphologically resemble chemical synapses and are called Merkel cell-neurite complexes (Halata et al. 2003). Merkel cell-neurite complexes are proposed to be functional units for transduction of slowly adapting (SA) mechanical stimuli such as light touch and hair movement (Halata et al. 2003; Iggo and Muir 1969). However, how Merkel cells sense mechanical stimuli remains uncertain.

It has been hypothesized that Merkel cells are sensory transducer cells directly sensing mechanical stimuli, which in turn triggers the release of chemical messengers from Merkel cells onto Aβ fiber neurites to transmit mechanical stimuli into sensory nerve impulses. Consistent with this hypothesis, deletion of Merkel cells from animals by different approaches such as irradiation of quinacrine-loaded Merkel cells or chemical damage of Merkel cells by chloroquine (Ikeda et al. 1994; Senok et al. 1996) or genetic deletion of the transcription factor Atoh1 and Merkel cells (Maricich et al. 2009) resulted in the loss of the characteristic neurophysiological responses to light touch. However, conflicting results were obtained in other studies. For example, in a different study selective elimination of the Merkel cells from rat touch domes by irradiation of quinacrine-loaded Merkel cells showed little change in mechanosensory function of these touch domes (Mills and Diamond 1995). In neurotrophin receptor p75 knockout mice in which Merkel cells were progressively lost, neither the survival nor the mechanical sensitivity of slowly adapting mechanoreceptive Aβ fibers was severely affected (Kinkelin et al. 1999). The transducer function of Merkel cells was also challenged by the extremely fast mechanical responses recorded from Merkel cell-neurite complexes, which suggested that mechanical transducers might be located on the neurites of Merkel cell-neurite complexes (Gottschaldt and Vahle-Hinz 1981). In agreement with this idea, several types of mechanically activated currents have been recorded from primary afferent neurons (Cho et al. 2002, 2006; Drew et al. 2002; Drew and Wood 2007).

Merkel cells contain dense-cored vesicles in which a number of transmitter substances such as VIP, substance P, enkephalin, CGRP, and 5-HT have been identified (English et al. 1992; Garcia-Caballero et al. 1989; Tachibana and Nawa 2005; Zaccone 1986; Zaccone et al. 1995). These substances were thought to be releasable and necessary for the characteristic slowly adapting response of SA type I (SAI) mechanoreceptors (Findlater et al. 1987). Agonists and/or antagonists for 5-HT (He et al. 2003) and ionotropic (Fagan and Cahusac 2001) and metabotropic (Cahusac and Senok 2006) glutamate receptors were found to have effects on SAI response, and therefore 5-HT and glutamate were suggested to be either transmitters or modulators of Merkel cell-neurite complexes. Molecular profiling revealed that Merkel cells have neuronal synaptic release machinery including presynaptic molecules, vesicular glutamate transporter 2, voltage-gated Ca²⁺ channels, and K⁺ channels (Haeberle et al. 2004). Merkel cells in the intact skin were specifically labeled by antibodies against voltage-activated Ca²⁺ channels (Ca_V2.1) and voltage- and Ca²⁺-activated K⁺ (BKCa) channels (Piskorowski et al. 2008). Voltage-clamp recordings from acutely dissociated Merkel cells revealed voltage-gated Ca²⁺ channel currents and K⁺ channel currents (Piskorowski et al. 2008; Yamashita et al. 1992). Ca²⁺ entry through voltage-gated Ca²⁺ channels produced small intracellular Ca²⁺ transients, but the signal was amplified severalfold by Ca²⁺-induced Ca²⁺ release (Piskorowski et al. 2008). Voltage-activated K⁺ currents in Merkel cells were carried predominantly by BKCa channels (Piskorowski et al. 2008). These properties suggest that Merkel cells have neuronal-like excitability, although recent studies suggested that they were originated from epidermal lineage (Morrison et al. 2009).

Attempts have been made to determine whether Merkel cells respond directly to different types of mechanical stimulation. A previous study using patch-clamp recording technique failed to show any mechanical response in Merkel cells that were acutely dissociated from rat footpad epidermis (Yamashita et al. 1992). In a sinus hair preparation, mechanical stimulation of Merkel cells with fine glass rods resulted in small transient increases in intracellular Ca²⁺ levels in the group of Merkel cells around the stimulating probes (Chan et al. 1996). In acutely dissociated Merkel cells, swelling of Merkel cells by a hypotonic solution triggered cytoplasmic Ca²⁺ transients. These Ca²⁺ signals appeared to arise from swelling-activated Ca²⁺-permeable ion channels since Ca²⁺ transients were abolished in extracellular Ca²⁺-free solution (Boulais et al. 2009; Haeberle et al. 2008). Using the same preparation, a recent study showed a sustained inward current that was evoked by the application of a high negative pressure through patch-clamp electrodes (Boulais et al. 2009). The study also used the Ca²⁺ imaging technique to show that swelling Merkel cells with hypotonic solution increased intracellular Ca²⁺ levels in Merkel cells. Although both negative pressure and cell swelling could elicit responses in Merkel cells, the mechanotransduction and osmoreception were suggested to be mediated by different mechanisms (Boulais et al. 2009).

Merkel cells have been successfully cultured under different conditions (Fukuda 1996; Haeberle et al. 2008; Shimohira-Yamasaki et al. 2006). Merkel cells usually have a fibroblast-like morphology with some fine processes when they are grown in serum-containing culture medium (Fukuda 1996; Haeberle et al. 2008). When Merkel cells and nerve cells were cultured together, both nerve fibers and cytoplasmic processes of Merkel cells outgrew and cooperatively organized synapselike structures at their contact points (Shimohira-Yamasaki et al. 2006). Cultured Merkel cells may provide an important system to investigate molecular mechanisms by which mechanical stimuli are detected by Merkel cells and subsequently transmitted to sensory nerve fibers. Only recently has a study demonstrated that cultured Merkel cells respond to hypotonic solution with increases of intracellular Ca²⁺ levels (Haeberle et al. 2008). Whether Merkel cells in culture may respond to other types of mechanical stimulation remains to be determined. In the present study, we used cultured Merkel cells isolated from vibrissal hair follicles of rat whisker pads and calcium imaging technique to characterize the sensitivity of these cells to shear force, cell swelling, and ATP.

MATERIALS AND METHODS

Animal care and use conformed to National Institutes of Health guidelines for care and use of experimental animals. Experimental protocols were approved by the University of Cincinnati Institutional Animal Care and Use Committee. Sprague-Dawley rats at the age of 14–21 days were deeply anesthetized with isoflurane and killed by decapitation. Whisker pads were removed and placed in Leibovitz L-15 medium (Cellgro, Manassas, VA), and vibrissal hair follicles were dissected out under a light microscope. Free follicles were truncated to obtain enlargement segments where Merkel cells are located (Fig. 1). The Merkel cell-containing segments were transferred to a tube containing 4 mg of collagenase (Sigma, St. Louis, MO) and 10 mg of dispase (Roche Applied Science, Penzberg, Germany) in 2 ml of MEM (HyClone Lab, Logan, UT) and incubated at 37°C for 1 h on an orbital shaker. Enzyme activity was ceased by washing the follicle segments three times with MEM. Merkel cells were then dissociated from the follicle segments by repeated trituration with a Pasteur pipette. The dissociated cells were plated on coverslips precoated with poly-d-lysine and maintained in a MEM culture medium that also contained 5% heat-inactivated horse serum (JRH Biosciences, Lenexa, KS), uridine/5-fluoro-2′-deoxyuridine (10 μM; Sigma), 8 mg/ml glucose, and 1% vitamin solution (Invitrogen, Eugene, OR). The cultures were maintained at 37°C in a humidified atmosphere of 95% air and 5% CO₂ for 24 h, and then the culture medium was replaced with a pure MEM culture medium. The cultures remained in the incubator at 37°C with 95% air and 5% CO₂. Cells were used within 72 h in culture.

Fig. 1.

Quinacrine staining and cytokeratin 20 (CK20) immunoreactivity (ir) of cultured Merkel cells dissociated from vibrissal hair follicles of rat whisker pads. A, left: an intact vibrissal hair follicle dissected out from a rat whisker pad and viewed under a light microscope. Right: schematic diagram illustrates the structures of a vibrissal hair follicle. Highlighted region in both panels is the root sheath of upper hair follicle segment where Merkel cells are present at a high density. B: cultured cells labeled with quinacrine. Left: viewed under a light microscope. Right: viewed under a fluorescent microscope. C: cells after immunostaining with an antibody against CK20. Left: viewed under a light microscope. Right: viewed under a fluorescent microscope. In both B and C cells were in culture for 48 h before staining. Scale bars, 50 μm in A and 10 μm in B and C.

To stain Merkel cells with quinacrine, a coverslip with cultured cells was mounted on a perfusion chamber. Cells were incubated 1 h with 10 nM quinacrine in normal bath solution at room temperature. Normal bath solution contained (in mM) 150 NaCl, 5 KCl, 2 MgCl₂, 2 CaCl₂, 10 glucose, and 10 HEPES, pH 7.3 adjusted with NaOH and osmolarity 320 mosM adjusted with sucrose. Cells were then perfused with the normal bath solution at 1 ml/min. Quinacrine-stained cells were examined under an inverted fluorescent microscope with an FITC filter set (Olympus IX70 microscope, Lake Success, NY). To immunostain cells against cytokeratin 20 (CK20), cells in culture dishes were fixed with a fixation solution that contained 4% paraformaldehyde and 0.4% Triton X-100 overnight at 4°C. Cells were rinsed five times with distilled water and then incubated with 2% Triton X-100 for 1.5 h at room temperature. Cells were further rinsed three times with a phosphate-buffered saline (PBS) solution that contained 1% goat serum and 2% Triton X-100 (1% GS-PBS-T). Cells were then exposed to a 1:30 goat serum GS-PBS-T solution for 1 h at room temperature and incubated with a monoclonal anti-mouse CK20 antibody overnight at 4°C. After being rinsed with 1% GS-PBS-T three times, the cells were exposed to 1:100 goat anti-mouse IgG conjugated with Alexa 594 (Invitrogen) for 3 h at room temperature, washed three times with 1% GS-PBS-T, and mounted on coverglasses. CK20 immunoreactivity (CK20-ir) was examined under a ×10 objective with a Texas red filter set. Fluorescent images for quinacrine staining and CK20-ir were taken by a Peltier-cooled charge-coupled device (CCD) camera (PentaMAX-III System, Roper Scientific, Trenton, NJ) under a ×10 objective with 100-ms exposure time, and positively stained cells were counted off-line.

For calcium imaging experiments, the calcium indicator fluo-3 was loaded into Merkel cells on coverslips by incubation of cells with 5 μM fluo-3 AM (Invitrogen) in normal bath solution at 37°C for 1 h. After dye loading, a coverslip was mounted on a perfusion chamber (width of 20 mm and height of 2 mm) and the chamber was then placed on the stage of an inverted Olympus IX70 microscope. Cells on the coverslip were continuously perfused with normal bath solution flowing at 0.5 ml/min. At this perfusion flow rate, the shear force at the center of the chamber was 0.0024 dyn/cm², estimated based on the equation S = 3μQ/wh², where μ is fluid viscosity of the bath solution (0.0077 P), Q is the flow rate (ml/s), w is the width of the perfusion chamber, and h is the height of the chamber (Jacobs et al. 1995). Fluo-3 was excited at 450 nm with a mercury lamp; fluorescence emission was collected at 550 nm, and the wave lengths of excitation and emission were achieved by a fluorescence filter set. Fluo-3 fluorescence in the cells was detected with a Peltier-cooled CCD camera (PentaMAX-III System) under a ×10 objective. Images were acquired at one frame every 5 s, 100-ms exposure time per frame, with MetaFluor Imaging System software (Molecular Devices, Sunnyvale, CA). To test shear force, normal bath solution was applied to cells at the center of the chamber through a glass tube (1-mm ID) positioned 1.0 mm away from cells. At flow rates of 0.3, 0.6, 1.2, and 2.4 ml/min, the shear stress force was 0.4, 0.8, 1.6, and 3.2 dyn/cm², estimated based on the equation S = 4μQ/πr³, where μ is fluid viscosity (0.0077 P), Q is the flow rate (ml/s), and r is the internal radius of the tube (Jacobs et al. 1995; Olesen et al. 1988). Unless otherwise indicated, the time interval was at least 15 min between each test. In some experiments, Gd³⁺ (100 μM) was preapplied through the bath and then coapplied during the application of shear force. To test the hypotonic solution effect, a modified bath solution with 224 mosM was applied through the bath. To test the effect of ATP, ATP at a concentration of 100 μM was applied to cells through the above glass tube in the presence of Gd³⁺. All experiments were carried out at room temperature of ∼24°C.

Relative fluorescence intensity (ΔF/F₀) was used as response to stimuli, and cells with ΔF/F₀ values of ≥0.15 (i.e., equal to or above 15% of increase in fluorescence intensity) were designated as responsive cells. Sample sizes are numbers of cells in each experimental group. Percentages of responding cells were calculated in each dish, and sample sizes are numbers of dishes in each experimental group. Unless otherwise specified, data are presented as means ± SE. Analysis of variance (ANOVA, 1 way) was used for statistical analyses of data sets of multiple groups, followed by Student-Newman-Keuls post hoc test. Student's t-test was used to evaluate the significance of changes in mean values between two groups. Statistical significance was considered at the level of P < 0.05.

RESULTS

Intact vibrissal hair follicles dissected out from rat whisker pads have the following structures when viewed under a light microscope: hair shaft, root sheath, ring sinus, attached nerves, and capsule (Fig. 1A). Merkel cells have been found to be present at high density in root sheath of upper hair follicle segment (Fukuda 1996). Enzyme treatment of the upper hair follicle segments and subsequent mechanical dissociation yielded high-density Merkel cells in culture dishes as evidenced by staining with quinacrine (Fig. 1B) or CK20-ir (Fig. 1C), two often-used markers for Merkel cell identification (Fukuda et al. 2003). The sizes of these cells were ∼10–20 μm. Quinacrine-labeled cells and CK20-ir-positive cells were 69.8 ± 4.2% (n = 14 dishes, total 233/335 cells, 3 dishes) and 60.8 ± 4.4% (n = 19 dishes, total 203/340 cells, 4 dishes), respectively. Merkel cells were round in shape when immediately plated in dishes (not shown), and they extended some processes in culture overnight (Fig. 1, B and C).

Figure 2A is a schematic diagram that shows the experimental setup for the application of shear force onto cells. In these experiments, cells were continuously perfused with normal bath solution (perfusion bath) at a constant flow rate of 0.5 ml/min. At this flow rate in our recording chamber, the perfusion bath only produced 0.0024 dyn/cm² shear force on the cells at the center of the recording chamber, and no cell responded to this baseline shear force (Fig. 2B, top left). Accordingly, only those cells at the center field of the recording chamber were tested, in order to keep baseline shear force constant. To test cell mechanical sensitivity, a shear force bath (same as normal bath solution) was applied from a tube whose opening was located near the center of the recording chamber (Fig. 2A). Applications of shear forces could elicit a rapid increase of intracellular Ca²⁺ levels in some cells, as indicated by the increase of fluo-3 fluorescence intensity (Fig. 2B, bottom).

Fig. 2.

Experimental setup for delivering shear force and measuring cell responses. A: schematic diagram illustrates the experimental setup for delivering shear force to cells. Shear forces were generated by delivering a shear force bath at different flow rates from a glass tube positioned very close to the center of the recording chamber. The shear force bath had chemical composition identical to the perfusion bath in control experiments. B: an example shows Ca²⁺ imaging obtained under baseline conditions (top left) and after the application of 1.6 dyn/cm² shear force (bottom left). Cells in the recording chamber were continuously perfused with perfusion bath flowing at 0.5 ml/min (baseline), which yielded negligible shear force on the cells at the center of the recording chamber. Top right: cells viewed under light microscope (×10 objective). Scale bars, 50 μm in each panel. Bottom right: 2 traces show the increases of intracellular Ca²⁺ levels in the 2 cells indicated by 2 arrows in the images. ΔF/F₀, relative fluorescence intensity.

We examined cell responses to different magnitudes of shear forces. In this experiment, shear forces were applied to cells for 20 s with a time interval of 20 min between each application (Fig. 3, A–D). At the shear force of 0.4 dyn/cm², no cell was found to have any response (not shown). When shear force increased to 0.8 dyn/cm², 3.2 ± 0.9% cells (n = 13 dishes, 64/1,997 cells) showed responses, and the peak responses (ΔF/F₀) were 0.38 ± 0.05 (n = 64 cells). A further increase of shear force to 1.6 dyn/cm² resulted in significant increase in the percentage of responding cells to 29.9 ± 3.6% (n = 13 dishes, 574/1,997 cells; P < 0.05) and also increases in peak responses (ΔF/F₀) to 0.55 ± 0.02 (n = 574 cells, 13 dishes; P < 0.05). There was no further increase in the cell numbers and peak responses when shear force increased to 3.2 dyn/cm² (not shown). While shear force of 1.6 dyn/cm² elicited responses in the cells that morphologically resemble Merkel cells, it had no effect on the cells that morphologically resemble keratinocytes (Fig. 3, E–G). Of 26 keratinocytes tested with a shear force of 1.6 dyn/cm², none responded to the shear force. This is in sharp contrast to Merkel cells (n = 26; Fig. 3G).

Fig. 3.

Increases of intracellular Ca²⁺ levels following the application of shear force. A: images from left to right show a cultured cell viewed under light microscope (1st panel), baseline fluo-3 fluorescence (2nd panel), and increases of fluo-3 fluorescence intensity following the application of 0.8 dyn/cm² (3rd panel) and 1.6 dyn/cm² (4th panel) shear forces. Scale bars, 10 μm. B: 2 traces show the responses of the cell shown in A to 0.8 dyn/cm² and 1.6 dyn/cm² shear forces. The horizontal bar above the traces indicates the duration (20 s) of shear force application. C: summary of % of responding cells to 0.8 dyn/cm² (n = 13 dishes, total 64/1,997 cells) and 1.6 dyn/cm² (n = 13 dishes, total 574/1,997 cells) shear forces. D: summary of peak responses to 0.8 dyn/cm² (n = 64 cells) and 1.6 dyn/cm² (n = 574 cells) shear forces. E: images from left to right are bright field (1st panel) and fluo-3 fluorescence images at baseline (2nd panel) and after the application of 1.6 dyn/cm² (3rd panel) shear force. Scale bars, 20 μm. In the image field there were several Merkel cells, and 3 of them responded to 1.6 dyn/cm²⁺ shear force (indicated by red, orange, and green arrows). There was a keratinocyte (a big cell indicated by a blue arrow) that had no response to the shear force. F: time courses of fluo-3 fluorescence intensity in the cells indicated by arrows. The horizontal bar above the traces indicates shear force application. G: peak fluo-3 fluorescence intensity in keratinocytes (n = 26) and Merkel cells (n = 26). Data represent means ± SE. *P < 0.05.

We next examined effects of shear force duration on cell responses. At the same shear force of 1.6 dyn/cm², the peak responses (ΔF/F₀) were 0.22 ± 0.02 (n = 43 cells, 5 dishes), 0.30 ± 0.03 (n = 81, 5 dishes), 0.36 ± 0.02 (n = 94, 5 dishes), and 0.38 ± 0.02 (n = 145, 5 dishes) when shear force stimulation duration was 2, 5, 10, and 20 s, respectively (Fig. 4, A and B). Responses evoked by 1.6 dyn/cm² shear force at stimulation durations of 10 and 20 s were significantly larger than the responses evoked by 2-s stimulation (P < 0.05).

Fig. 4.

Responses to shear force at different stimulation durations and to repeated shear force stimulation. A: sample traces show responses of a cell to 1.6 dyn/cm² shear force at durations of 2, 5, 10, and 20 s. The horizontal bar above each trace indicates the duration of shear force application. B: summary of peak responses of cells to 1.6 dyn/cm² shear force at durations of 2, 5, 10, and 20 s. C: sample traces show responses of a cell to repeated shear force stimulation. The intensity of each stimulus was 1.6 dyn/cm² and the duration 20 s. The time intervals between each stimulus are shown in D. D: bar graph shows summary of peak responses of cells to repeated shear force stimulation at different intervals. Data represent means ± SE. *P < 0.05.

Shear force-evoked increases of intracellular Ca²⁺ levels showed some degree of adaptation to consecutive stimulation. As shown in Fig. 4, C and D, the peak responses to the first stimulation were 0.47 ± 0.02 (n = 543 cells, 7 dishes) and were significantly reduced to 0.19 ± 0.01 (n = 543, 7 dishes; P < 0.05) in the subsequent stimulation at the time interval of 1 min. As the time intervals between shear force stimulation increased to 3, 5, 10, and 20 min, responses were 0.17 ± 0.05 (n = 543, 7 dishes), 0.18 ± 0.01 (n = 543, 7 dishes), 0.27 ± 0.01 (n = 543, 7 dishes), and 0.39 ± 0.02 (n = 543, 7 dishes), respectively (Fig. 4D). Peak responses at the intervals of 10 and 20 min showed significant recovery from the adaptation (P < 0.05).

Increases of intracellular Ca²⁺ level responding to shear force stimulation could be due to Ca²⁺ entry from the extracellular side or Ca²⁺ release from intracellular Ca²⁺ stores. To test these possibilities, we performed experiments in either normal bath solution that contained 2 mM Ca²⁺ or a low-Ca²⁺ bath solution that contained 0.2 mM Ca²⁺ and examined shear force effect under these two conditions (Fig. 5). Cell responses (ΔF/F₀) to 1.6 dyn/cm² shear force were 0.98 ± 0.05 (n = 157 cells, 6 dishes) in normal bath solution and significantly reduced to 0.05 ± 0.002 (n = 157, 6 dishes, P < 0.05) in the low-Ca²⁺ bath solution (Fig. 5). After return to normal Ca²⁺ bath solution, responses to the shear force recovered to 0.60 ± 0.04 (n = 157 cells, 6 dishes).

Fig. 5.

Extracellular Ca²⁺ dependence of shear force-induced responses. A: sample images show responses of a cell (indicated by arrow) to shear force in normal bath solution with 2 mM Ca²⁺ (2nd and 4th panels) and in a bath solution with low Ca²⁺ (0.2 mM, 3rd panel). The images in the 1st and 5th panels are baseline and bright field, respectively. Scale bars, 10 μm. B: traces show the responses of the cell indicated in A. C: bar graph shows summary of peak responses under each condition described in A (n = 157 cells). In all experiments, 1.6 dyn/cm² shear force was applied for 20 s and the time interval between each test was 15 min. Data represent means ± SE. *P < 0.05.

Shear force-evoked Ca²⁺ entry may be due to the activation of mechanically activated channels expressed on the cells of our Merkel cell preparation. To test this possibility, we examined whether shear force-induced responses could be blocked by Gd³⁺, a widely used blocker for mechanically activated channels (Boulais et al. 2009). As shown in Fig. 6, peak responses to 1.6 dyn/cm² shear force were 0.66 ± 0.04 (n = 219, 5 dishes) in the absence of Gd³⁺ and were significantly reduced to 0.03 ± 0.002 in the presence of 100 μM Gd³⁺ (n = 219, 5 dishes; P < 0.05). Thus, in cells in which intracellular Ca²⁺ levels were increased by shear force, the effect was almost completely abolished in the presence of 100 μM Gd³⁺. The blockage by Gd³⁺ was reversed (peak response: 0.75 ± 0.06, n = 219, 5 dishes) after 20-min wash in baseline normal bath solution (Fig. 6).

Fig. 6.

Inhibition of cell responses to shear force by Gd³⁺. A: from left to right, a bright field image (1st panel) and fluorescent images taken under baseline conditions (2nd panel) and after the application of 1.6 dyn/cm² shear force with normal bath solution (3rd panel), the application of 1.6 dyn/cm² shear force with bath solution that also contained Gd³⁺ (100 μM, 4th panel), and the application of 1.6 dyn/cm² shear force after washing out of Gd³⁺ (5th panel). Scale bars, 10 μm. B: traces show the responses of the cell illustrated in A under different conditions. The duration of shear force application is indicated by a bar above the traces. C: bar graph shows the summary of peak responses and inhibition by Gd³⁺ (n = 219 cells). In each experiment, 1.6 dyn/cm² shear force was applied for 20 s and the time interval between each test was 15 min. Data represent means ± SE. *P < 0.05.

Many cells that responded to shear force also responded to hypotonic solution, and the effect of hypotonic solution remained in the presence of 100 μM Gd³⁺ (Fig. 7). In this set of experiments, the percentage of cells that responded to 1.6 dyn/cm² shear force was 18.5 ± 2.8% (n = 7 dishes, 196/1,143 cells) and the percentage responding to hypotonic solution was 19.0 ± 6.0% (n = 7 dishes, 189/1,143 cells; Fig. 7C, left). Of all 196 cells that responded to the shear force, 59 (30%) also responded to hypotonic solution that contained 100 μM Gd³⁺. The peak responses to the shear force were 0.75 ± 0.03 (n = 196, 7 dishes), significantly higher than the responses induced by hypotonic solution (0.46 ± 0.03, n = 189, 7 dishes, P < 0.05; Fig. 7C, right).

Fig. 7.

Responses induced by hypotonic solution. A: images show Fluo-3 fluorescence intensity in cells before (1st panel) and after the application of 1.6 dyn/cm² shear force (2nd panel) or the application of a hypotonic solution (3rd panel). The hypotonic solution was applied in the presence of 100 μM Gd³⁺. A cell that responded to both shear force and hypotonic solution is indicated by arrows in 2nd and 3rd panels. Scale bars, 20 μm. B: traces are the responses of the cell indicated in A to 1.6 dyn/cm² shear force (left) and the hypotonic solution (right). The horizontal bar above each trace indicates the stimulation duration. C, left: summary of the percentage of the cells that responded to shear force (n = 7 dishes, total 196/1,143 cells) and hypotonic solution (n = 7 dishes, 189/1,143 cells). Right: peak responses to the shear force (n = 196 cells) and the hypotonic solution (n = 189). In each experiment, the hypotonic solution was applied in the presence of 100 μM Gd³⁺. Data represent means ± SE. *P < 0.05.

The cells that responded to shear force also responded to 100 μM ATP (Fig. 8). However, more cells responded to ATP than to shear force. For a shear force of 1.6 dyn/cm², 24.8 ± 1.2% (n = 9 dishes, 588/2,407 cells) cells responded (Fig. 8C). The percentage of cells that responded to 100 μM ATP was 86.7 ± 2.5% (n = 9 dishes, 2,096/2,407 cells) in the absence of Gd³⁺ and 84.3 ± 2.2% (n = 9 dishes, 2,044/2,407 cells) in the presence of Gd³⁺, significantly higher than the percentage of cells responding to shear force (P < 0.05). Peak responses to 100 μM ATP in the absence and presence of 100 μM Gd³⁺ were 1.26 ± 0.06 (n = 205) and 1.04 ± 0.06 (n = 205) respectively, significantly higher than shear force-induced responses (0.83 ± 0.04, n = 205, P < 0.05; Fig. 8D).

Fig. 8.

Responses induced by ATP. A: images from left to right show responses of a cell (indicated by arrow) to 1.6 dyn/cm² shear force in the absence (2nd panel) and presence (3rd panel) of 100 μM Gd³⁺ and to 100 μM ATP in the absence (4th panel) and presence (5th panel) of 100 μM Gd³⁺. Scale bars, 10 μm. B: traces show the time course of the responses of the cell indicated in A. The horizontal bar over each trace indicates the application of shear force or ATP. C: summary of the percentage of responding cells in the experiments illustrated in A (n = 9 dishes in each group, total 2,407 cells). D: summary of the peak responses in the experiments illustrated in A (n = 205 cells in each group). Data represent means ± SE. *P < 0.05.

DISCUSSION

This is the first study to use shear force to demonstrate mechanical sensitivity of Merkel cells grown in culture. In this study, we isolated cells from the upper segment of rat vibrissal hair follicles. Our preparation yielded ∼70% cells that were stained with quinacrine, a chemical marker often used for identifying Merkel cells (Fukuda et al. 2003). There were ∼61% of cells in our preparation that were immunoreactive to CK20, a specific immunochemical marker for Merkel cells (Fukuda et al. 2003). The percentage of Merkel cells in our preparation as based on the staining with quinacrine and CK20-ir were lower than that of a previous study using the same cell preparation with quinacrine staining (Fukuda 1996). In that study, >90% of cells were labeled by quinacrine and considered to be Merkel cells (Fukuda 1996). The discrepancy between the previous study and ours could be because we excluded weak-staining cells as positive cells and therefore may underestimate the percentage of Merkel cells in our Merkel cell preparation. Although most cells (61–70%) in our preparation were defined as Merkel cells by the staining methods, shear force-sensitive cells were only up to 30%. This indicated that not all cultured Merkel cells were sensitive to shear force under our experimental condition.

Previous studies have demonstrated mechanical sensitivity of noncultured Merkel cells with several other stimulation methods including probing Merkel cell membrane with fine glass rods (Chan et al. 1996), swelling Merkel cells with hypotonic solution (Boulais et al. 2009; Haeberle et al. 2008), and application of negative pressures into cells with patch-clamp electrodes (Boulais et al. 2009). None of these approaches is a physiological stimulus for Merkel cells in vivo. Shear force is also not a physiological stimulus for Merkel cells. However, using shear force as a method to mechanically stimulate Merkel cells may have certain advantages over other approaches when experiments are performed with cultured Merkel cells. For example, probing cell membranes with fine glass rods may potentially damage cell membranes. This may be particularly a problem for cultured Merkel cells because they are small and flat in culture. On the other hand, this is not a concern when shear force is used to stress cell membranes. The application of negative pressure into cells through patch-clamp electrodes may change intracellular environment due to intracellular dialysis. This is also not a concern when shear force is used to apply mechanical force onto Merkel cells. Although both cell swelling and shear force can cause mechanical stress on cell membranes, the effect of hypotonic solution is slow-on and slow-off. On the other hand, membrane stress induced by shear force has a short delay in both on-time and off-time of mechanical force, thereby probably better mimicking natural stimuli experienced by Merkel cells in vivo.

We showed that shear force-induced increases of intracellular Ca²⁺ levels depended on Ca²⁺ entry from the extracellular side. This could be due to a direct entry of Ca²⁺ through a mechanically activated channel. Other mechanisms including voltage-gated Ca²⁺ channels and intracellular stores might also contribute to the increases of intracellular Ca²⁺ levels following shear force stimulation. Voltage-gated Ca²⁺ channels have been found to be expressed in Merkel cells, and depolarization of Merkel cell membranes by high concentrations of potassium chloride could result in the increases of intracellular Ca²⁺ levels due to the activation of voltage-gated Ca²⁺ channels (Haeberle et al. 2008; Piskorowski et al. 2008; Yamashita et al. 1992). The initial Ca²⁺ signals induced by mechanical stimulation could be further amplified and prolonged through calcium-induced calcium release in Merkel cells (Haeberle et al. 2008; Piskorowski et al. 2008). There has been evidence indicating the presence of caffeine- and ryanodine-sensitive calcium release stores in Merkel cells isolated from rat vibrissal hair follicles (Senok and Baumann 1997). Release of Ca²⁺ from intracellular Ca²⁺ stores may also be the mechanism by which ATP increased intracellular Ca²⁺ levels in Merkel cells shown in the present study. This suggestion is based on a previous study showing that ATP-induced responses in Merkel cells were mediated by P_2Y receptors (Chan et al. 1996).

A previous study suggested that mechanical responses induced by intracellular negative pressure and hypotonic solution may be mediated by two different types of transducer molecules (Boulais et al. 2009). This suggestion was based on the observation that Gd³⁺, a widely used blocker of mechanically activated channels/receptors, abolished mechanical responses induced by intracellular negative pressure but not by hypotonic solution (Boulais et al. 2009). Consistent with this previous work, we found that Gd³⁺ abolished shear force-induced responses but not hypotonic solution-induced response. The differential effects of Gd³⁺ in our study raised a possibility that responses to shear force and hypotonic solution have different underlying mechanisms. Alternatively, both shear force and hypotonic solution may act on the same stretch-activated channel but extracellular Gd³⁺ may only block the channel when stretch force is applied extracellularly by shear force. The clear answers to these questions will need highly specific blockers of mechanically activated ion channels, but such blockers are currently not available.

Molecular identities of mechanically activated channels/receptors remain unclear in Merkel cells and other somatosensory tissues in mammals. Other types of cells such as endothelial cells have been shown to respond to shear stress, and the responses in these cells could be blocked by Gd³⁺ (Lansman et al. 1987; Naruse and Sokabe 1993). Shear force-induced responses in these cells were thought to be mediated by stretch-activated ion channels whose molecular identities remain poorly defined (Sachs 2010). It is possible that the ion channels mediating shear force responses in Merkel cells are the same stretch-activated ion channels that mediate shear force responses in other cell types. Alternatively, the shear force-induced response in Merkel cells was mediated by different mechanically activated ion channels. Transcripts of several TRP channels have been identified in Merkel cells, including PKD1, PKD2, and TRPC1, and they were considered to be good candidates of mechanically activated channels in Merkel cells (Haeberle et al. 2008). A previous study indicated that TRPC1 and TRPC6 channels cooperate with TRPV4 channels to mediate mechanical hyperalgesia and primary afferent nociceptor sensitization (Alessandri-Haber et al. 2009). More recently, two new molecules, Piezo1 and Piezo2, have been described to play a broad role in mechanotransduction in different forms of life, and Piezo2 is suggested to be involved in rapidly adapting mechanically activated currents in somatosensory neurons (Coste et al. 2010). Of interest, Piezo1 but not Piezo2 was found to be expressed in a high amount in mouse skin (Coste et al. 2010), making Piezo1 one of the candidates for mechanical transducers in Merkel cells. The establishment of shear force-induced mechanical sensitivity in cultured Merkel cells will allow the use of a cell-based knockdown strategy to screen candidates of mechanically activated channels/receptors in Merkel cells. It may also be useful in studying potential transmitters released by Merkel cells in responding to mechanical stimulation.

DISCLOSURES

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