Among ion channels with a mechanosensitive role are some members of the transient-receptor potential (TRP) family, and of the degenerin/epithelial Na+ channel (DEG/ENaC) superfamily, especially those belonging to the acid-sensing ion channel (ASIC) family (see Lumpkin and Caterina, 2007; Tsunozaki and Bautista, 2009). Thus, the ion channels involved in mechanosensation are expected to be expressed in the mechanosensory neurons and/or in the mechanoreceptors where the mechanotransduction occurs (see for a review Del Valle et al., 2012). In particular, β-ENaC and γ-ENaC subunits, but not α-ENaC subunit, have been detected in the axons innervating Pacinian corpuscles, Meissner corpuscles, Merkel cells, and small lamellate corpuscles of the rat and mouse (Drummond et al., 2000; Montaño et al., 2009). ASIC ion channels are expressed in large and intermediate mechanosensory neurons of dorsal root ganglia (McIlwrath et al., 2005; Kawamata et al., 2006) as well as in different kinds of mechanoreceptors including Meissner corpuscles, Pacinian corpuscles, Ruffini's corpuscles, and hair-associated sensory corpuscles (Waldmann et al., 1997; García-Añoveros et al., 2001; Price et al., 2001; Duggan et al., 2002; Montaño et al., 2009; Calavia et al., 2010; Rahman et al., 2011;). On the other hand, mechanosensory neurons also express TRP vanilloid 2 (TRPV2), TRPV4, and TRP ankyrin 1 (TRPA1) (Liedtke et al., 2000; Caterina and Julius, 2001; Tamura et al., 2005; Kwan et al., 2009), and Meissner corpuscles, Merkel cells, penicillate nerve endings, and intraepidermal terminal display TRPV4 and TRPA1 immunoreactivity (Liedtke et al., 2000; Suzuki et al., 2003; Kwan et al., 2009).
All the data referred above apply to mammals and as far as is known, no information exists about the presence of mechanoproteins in avian sensory corpuscles. Therefore, we decided to investigate its presence in Herbst corpuscles. Herbst corpuscles are the avian sensory corpuscles equivalent to the mammalian Pacinian corpuscles. They consist of the peripheral dendritic zone of a sensory axon, specialized Schwann-related periaxonic cells forming the inner core, and fibroblasts arranged as external capsule. Between the inner core and the capsule there is a boundary space containing a more or less dense network of microfibrils and fibroblasts (Halata and Munger, 1980; Malinovsky and Pac, 1980; Halata and Grim, 1993; Zelena et al., 1997). Immunohistochemical studies have also demonstrated that the corpuscular constituents of Herbst corpuscles have a protein composition that matched that of Pacinian corpuscles (Germanà et al., 1995). On the other hand, Herbst corpuscles typically express high levels of calcium-binding proteins (Del Valle et al., 1995; Chouchkov et al., 2002).
In the present study, we have used immunohistochemistry to investigate whether or not Herbst corpuscles localize the putative mechanoproteins TRPV4, ASIC2, and ENaC subunits, which have been previously detected in mammalian mechanoreceptors. The study was aimed to investigate the molecular basis of the mechanosensation and to elucidate whether or not they are similar to those in mammals.
MATERIAL AND METHODS
Materials and Treatment of the Tissues
Samples of glabrous skin taken from the rictus of ten adult pigeon (Columba livia, c. Morini) were used throughout this study. Animals were sacrificed by decapitation after deep chloral hydrate anesthesia. The rictus was removed and placed in Bouin's fixative for 24 hr, then dehydrated and processed for routine embedding in paraffin. The pieces were cut into 10 μm thick sections perpendicular to the skin surface and mounted on gelatin-coated microscope slides. The structure of the tissue was assessed using H&E staining. Experimental procedures were carried out according to the guidelines of the European Community on welfare of research animals (directive 86/609/EEC).
Deparaffinized and rehydrated sections were processed for detection of ENaC subunits, TRPV4, and ASIC2 using the EnVision antibody complex detection kit (Dako, Copenhagen, Denmark), following supplier's instructions. Briefly, non-specific binding was blocked with 10% bovine serum albumin for 20 min, and the endogenous peroxidase activity was blocked with 3% hydrogen peroxide. Sections were then incubated overnight at 4°C with the primary antibodies. The anti-ENaC subunit antibodies were rabbit polyclonal antibodies against specific epitopes of α-ENaC (131–225 aminoacids; H95, catalogue sc-21012, Santa Cruz Biotechnology, Santa Cruz, CA), β-ENaC (271-480 aminoacids; H190, catalogue sc-21013, Santa Cruz Biotechnology), and γ-ENaC (411-520 aminoacids; H110, catalogue sc-21014, Santa Cruz Biotechnology), all of them diluted to 1:100. The anti-ASIC2 antibody was a rabbit polyclonal antibody that binds the extracellular domain of mouse ASIC2 (Lifespan Biosciences, Seattle, WA; catalogue LS-C93915), and it was used diluted 1:200. The anti-TRPV4 antibody was a rabbit polyclonal antibody against a synthetic peptide derived from the cytoplasmic N-terminus conserved in mouse, human, and rat TRPV4 conjugated to immunogenic carrier protein (Abcam plc, Cambridge, UK; catalogue ab63003), used diluted 1:100. After incubation with the primary antibody, sections were incubated with the anti-rabbit EnVision system-labeled polymer (Dako-Cytomation) for 30 min, washed in buffer solution, and treated with peroxidase blocking buffer. Finally, the slides were washed with buffer solution and the immunoreaction was visualized with diaminobenzidine as a chromogen, washed, dehydrated, and mounted with Entellan® (Merk, Dramstadt, Germany). To ascertain structural details the sections were counterstained with Mayer's hematoxylin.
Sections were also processed for simultaneous detection of ASIC2 or TRPV4 and neuron specific enolase (NSE), and ASIC2 or TRPV4 and S100 protein. NSE was used to label the axon, and S100 protein was used to label the inner-core (Germanà et al., 1995). The antibodies against NSE (clone BBS/NC/VI-H14; Glöstrup, Denmark) and against S100 (clone 4C4.9; Thermo Scientific, Freemont, CA) were raised in mouse. Non-specific binding was reduced by incubation for 30 min with a solution of 10% bovine serum albumin in tris buffer solution (TBS). The sections were then incubated overnight at 4°C in a humid chamber with a 1:1 mixture of anti-ASIC2 and anti-NSE antibodies (diluted 1:200 and 1:100, respectively, in the blocking solution), anti-TRPV4 and anti-NSE antibodies (both diluted 1:100 in the blocking solution), anti-ASIC2, and anti-S100 protein antibodies (diluted 1:200 and 1:500, respectively, in the blocking solution), and anti-TRPV4 and anti-S100 protein antibodies (both diluted 1:100 in the blocking solution). After rinsing with TBS, the sections were incubated for 1 h with Alexa fluor 488-conjugated goat anti-rabbit IgG (Serotec, Oxford, UK), diluted 1:1,000 in TBS containing 5% mouse serum (Serotec), then rinsed again and incubated for another hour with CyTM3-conjugated donkey anti-mouse antibody (Jackson-ImmunoResearch, Baltimore, MD) diluted 1:50 in TBS. Both steps were performed at room temperature in a dark humid chamber. Finally, to ascertain structural details sections were counterstained with 4′,6-diamidino-2-phenylindole (DAPI) (10 ng/mL). Sections were then washed, dehydrated, and mounted with Entellan®. Triple staining was detected using a Leica DMR-XA automatic fluorescence microscope coupled with a Leica Confocal Software, version 2.5 (Leica Microsystems, Heidelberg GmbH, Germany) and the captured images were processed using the software Image J version 1.43 g Master Biophotonics Facility, MacMaster University, Ontario, (www.macbiophotonics.ca).
For control purposes, representative sections were processed in the same way as described above using non-immune rabbit or mouse sera instead of the primary antibodies or omitting the primary antibodies in the incubation. Furthermore, when available, additional controls were carried out using specifically preabsorbed antisera. Under these conditions no positive immunostaining was observed (data not shown).
In the skin of the pigeon rictus numerous Herbst corpuscles were observed, showing a typical morphology and structure. Occasionally, they were found isolated immersed in the dermis, often associated with feather follicles (Fig. 1a). As a rule, immunoreactivity for ENaC subunits was never found in Herbst corpuscles but positive β-ENaC immunostaining was detected in small non-capsulated nerve endings associated with feathers (Fig. 1b,c).
ASIC2 and TRPV4 Immunoreactivity in Cutaneous Herbst Corpuscles
Simple immunohistochemistry demonstrated the occurrence of specific ASIC2 and TRPV4 immunoreactivity in Herbst corpuscles (Fig. 2). An intense ASIC2 immunostaining was always detected in the central axon of Herbst corpuscles, whereas the cells forming the inner core and the capsule were unreactive or weakly reactive (Fig. 2a,b). The localization of TRV4 immunostaining was widespread and specific immunostaining was observed in both the central axons and the inner core cells (Fig. 2c–e). Apparently all Herbst corpuscles identified displayed immunoreactivity for both assessed mechanoproteins.
ASIC2 is Localized in Axons Supplying Herbst Corpuscles
Since the precise localization of ASIC2 within the Herbst corpuscles cannot be ensured with light immunohistochemistry, double immunofluorescence was carried out with antibodies against ASIC2 and NSE, and against ASIC2 and S100 protein. ASIC2 antibody conjugated with Alexa fluor 488 (green fluorescence), and anti-NSE antibody conjugated with CyTM3 (red fluorescence), showed almost exact co-localization (Fig. 3a–c), and this was confirmed by the analysis of cytofluorograms (data not shown). Therefore, ASIC2 in cutaneous Herbst corpuscles from pigeon is axonic. Conversely co-localization of ASIC2 and S100 protein was never observed (data not shown), thus excluding the expression of ASIC2 in cells forming the inner core.
TRPV4 in Localized in the Axon and Periaxonic Cells
Intense immunofluorescence for TRPV4 was found in the central axon, but also a faint, diffuse and specific immunofluorescence was observed in the inner core cells (Fig. 4a,d). The results of the double immunolabeling study were conclusive and demonstrated that TRPV4 is axonic since it co-localizes with NSE (Fig. 4b,c,e,f). Nevertheless, although TRPV4 was detected in the inner core cells, and S100 protein as well (data not shown) no specific co-localization of both antigens was clearly observed (Fig. 4g,h).
The vertebrate skin contains specialized sensory organs known collectively as mechanoreceptors that are responsible for the detection of different types of mechanosensation such as touch, vibration, or pressure. Some kinds of mechanoreceptors are common for all terrestrial species whereas others are specific for birds or mammals (Malinovsky and Pac, 1982; Malinovsky, 1996). The Herbst corpuscles of birds are the equivalent of the Pacinian corpuscles in mammals, and likely they are regarded as rapidly adapting low-threshold mechanoreceptors, that detect vibration and movement of the feathers (Höster, 1990). In recent years evidence has emerged that at the basis of the process of touch there is the activation of ion channels in response to mechanical stimuli, which is necessary for the conversion of a mechanical stimulus into an electrical signal (Lumpkin and Caterina, 2007; Lumpkin et al., 2010). Thus, different members of the DEG/ENaC, TRP channels, and the two-pore domain potassium (K2P) channel super-families are currently being considered as candidate mechanotransducing channels (see for a review Lumpkin and Caterina, 2007; Arnadóttir and Chalfie, 2010; Lumpkin et al., 2010; Del Valle et al., 2012).
Some of those putative mechanoproteins have been detected in Pacinian corpuscles (see Del Valle et al., 2012). In particular, human cutaneous Pacinian corpuscles express ASIC1 in the central axon and ASIC2 in the inner core cells (Calavia et al., 2010), whereas the murine tarsal Pacinian corpuscles express ASIC2 in both the inner core cell and the central axon (Montaño et al., 2009). Murine Pacinian corpuscles also express the β and γ ENaC subunits (Drummond et al., 2000; Montaño et al., 2009). Regarding the other protein investigated here, that is TRPV4, it has been never detected in Pacinian corpuscles, while it was observed in the axon of murine Meissner-like corpuscles (Suzuki et al., 2003). Therefore, since no data are available for avian sensory corpuscles, the results presented here are completely new and original.
We failed to demonstrate occurrence of ENaC subunits in Herbst corpuscles as reported in mammals (Drummond et al., 2000; Montaño et al., 2009) whereas β-ENaC was detected in sensory endings associated with the feathers as observed by Hitomi et al. (2009) in Ruffini's corpuscles. On the other hand, as for mice (Montaño et al., 2009), and monkeys (Cabo et al., 2012), the central axon of Herbst corpuscles display ASIC2 immunoreactivity. These and present results clearly differ from those of Calavia et al. (2010), who showed ASIC2 in the inner core cells of human Pacinian corpuscles. These findings clearly suggest species-specific differences in the expression of ASIC2, but the functional significance, if any, remains to be clarified.
Regarding TRPV4 its presence in mammalian Pacinian corpuscles has never been reported. In the pigeon Herbst corpuscles, however, TRPV4 immunoreactivity was clearly detected in the axon, co-localized with NSE. Its presence in the inner core cannot be ensured since a co-localization with S100 protein was not obvious. However, in both light and laser confocal microscopy, specific TRPV4 immunoreactivity was observed in the Schwann-related cells of the inner core. The low affinity and cross-reactivity of the antibodies used may account for these technical inconsistencies.
The present results provide new data about the occurrence of putative mechanoproteins in encapsulated mechanoreceptors, and are the first evidence of the presence of those proteins in avian sensory corpuscles. Experiments are in progress in our laboratory to localize new putative mechanoproteins in these organs.
In spite of the accumulating data demonstrating the presence of mechanoproteins in both mechanoreceptors and mechanosensory neurons (see Del Valle et al., 2012), the role of these molecules in the cutaneous sensation remains controversial based on the results obtained with mechanoproteins deficient mice (Price et al., 2000, 2001; Drew et al., 2004). Therefore more studies are necessary to elucidate the role of these molecules in mechanosensation.
Authors thank Dra. M.A. Guervós for excellent assistance and help in laser confocal microscopy