Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2009 Oct 14.
Published in final edited form as: Chemosens Percept. 2008 Jun 1;1(2):110–118. doi: 10.1007/s12078-008-9012-6

Merkel Cells in Somatosensation

Henry Haeberle 1,2, Ellen A Lumpkin 2
PMCID: PMC2761664  NIHMSID: NIHMS92065  PMID: 19834574

Abstract

Merkel cells are rare epidermal cells whose function in the skin is still debated. These cells localize to highly touch-sensitive areas of vertebrate epithelia, including palatine ridges, touch domes and finger tips. In most cases, Merkel cells complex with somatosensory afferents to form slowly adapting touch receptors; it is unclear, however, whether mechanosensory transduction occurs in the Merkel cell, the somatosensory afferent or both. Classic anatomical results suggests that Merkel cells are sensory cells that transduce mechanical stimuli and then communicate with sensory afferents via neurotransmission. This model is supported by recent molecular, immunohistochemical and physiological studies of Merkel cells in vitro and in intact tissues. For example, Merkel cells express essential components of presynaptic machinery, including molecules required for release of the excitatory neurotransmitter glutamate. Moreover, Merkel cells in vitro and in vivo are activated by mechanical stimuli, including hypotonic-induced cell swelling. Although these findings support the hypothesis that Merkel cells are sensory receptor cells, a definitive demonstration that Merkel cells are necessary and sufficient to transduce touch awaits future studies.

Keywords: epidermis, mechanotransduction, somatosensory, touch

Somatosensory transduction in the oral cavity

Along with the senses of taste and olfaction, the somatic senses of touch, thermoreception and pain (nociception) contribute to our ingestion and enjoyment of food. In the oral cavity, somatosensory neurons of the trigeminal, glossopharyngeal and vagal nerves encode information about the chemical and physical qualities of foodstuffs (Simon et al., 2006). For example, polymodal nociceptors innervating the tongue and palate are essential for detecting the pungent tastants found in chili peppers and wasabi (Caterina et al., 2000; Bautista et al., 2006; Kwan et al., 2006), as well as cooling agents such as menthol (Daniels and McKemy, 2007). In addition to these chemical irritants, polymodal nociceptors are activated by noxious temperatures, harsh mechanical stimuli and inflammatory mediators.

The oral epithelium is also innervated by mechanosensitive neurons that monitor the positions of the tongue and foodstuffs during chewing. In the palate, at least four subtypes of somatosensory receptors have been classified as touch receptors based on morphological similarities to touch-sensitive neurons that innervate the skin. One receptor subtype, the Merkel cell-neurite complex, is particularly abundant in palatine ridges (Tachibana et al., 1997; Halata et al., 1999; Nunzi et al., 2004). Although the role of Merkel cell-neurite complexes in innocuous touch is well established, the function of the Merkel cell within the complex is still debated. Here, we present an historical overview of the discovery of somatosensory mechanoreceptors and review progress on the role of Merkel cells in somatosensation. This manuscript is intended to complement excellent reviews of somatosensation in the oral cavity (Simon et al., 2006) and Merkel cells across vertebrate species (Tachibana and Nawa, 2002; Halata et al., 2003).

Somatosensory modalities

The mechanisms by which animals sense touch, temperature and pain have been the subject of scientific inquiry since at least the third century BCE, when Aristotle included touch in his five cardinal senses. Anatomical experiments in the nineteenth century revealed a remarkable assortment of sensory structures within the dermis and underlying tissues of the skin. Concurrently psychophysical experiments discovered the presence of sensory spots on the skin that were highly sensitive to specific stimulus modalities in humans (Hamann, 1995). This spatial localization of structure and function led von Frey and others to postulate the specificity theory of somesthesis – cutaneous senses are subserved by neurons that are specialized to respond to only one sensory modality (Boring, 1942).

The sensory modalities of somesthesis are innocuous touch, cool, warm and nociception. Which sensory modalities are mediated by the various somatosensory structures was the subject of much speculation: most sensory structures are not visible in living humans, and the inability to record the activity of sensory neurons in animals prevented definite attribution of individual sensory modalities to specific neuronal populations. Consequently, the strict labeled-line doctrine of somesthesis was controversial.

Physiological analysis of somatosensation awaited the development of single unit recordings of primary afferent fibers. In 1926, Adrian and Zottermann developed the first live, semi-intact nerve recordings of primary sensory afferents and their associated cutaneous targets (Adrian and Zotterman, 1926; Zotterman, 1939). They discovered individual afferents specifically activated by light touch, supporting von Frey's specificity theory. Subsequent work revealed the presence of specific thermosensitive and nociceptive afferents (Zotterman, 1939; Dodt and Zotterman, 1952; Hensel and Boman, 1960).

Given that different sensory modalities are mediated by labeled lines, one possibility is that the intensity of a given stimulus is encoded by firing rates of individual neurons. Mountcastle and colleagues tested this theory by recording from sensory afferents while applying quantitative indentations of the skin (Werner and Mountcastle, 1965). They described mechanically sensitive fibers responding to indentation of mechanosensitive touch domes with the following power function: R = K • Sn where R is the mean response number, K is a constant, S is the stimulus intensity measured as net skin indentation and the exponent n is a constant. Subsequent research has described a number of power functions that fit the stimulus-response relationship, depending on the type of mechanosensitive afferent measured and the response period monitored (Knibestöl, 1975). Mountcastle extended his work by recording similar mechanically induced stimulus-response functions in the spinal cord and the primary somatosensory areas of the brain, indicating that stimulus intensity is primarily encoded by the sensory afferent terminals and that transformations in central areas of the nervous system are linear (Mountcastle, 1967; Harrington and Merzenich, 1970). More recent research in the human hand revealed that different mechanoreceptors have varied stimulus-response functions despite a linear relationship between stimulus intensity and percept strength (Knibestöl and Vallbo, 1980). These data indicate, contrary to Mountcastle's hypothesis, a non-linear relationship between afferent firing and percept strength, indicating the presence of diverse non-linear transforms in higher brain regions.

Response profiles of mechanosensory afferents

Additional studies of somatosensory afferents in various vertebrates revealed five consistent stimulus-response profiles, which are characterized by their adaptive properties (Hamann, 1995). Rapidly adapting afferents fire bursts of action potentials to skin movement and vibration, but are insensitive to sustained skin deformation (Fig. 1). Slowly adapting afferents, by contrast, can continue to respond to sustained skin deformation for more than 30 min (Iggo and Muir, 1969). Both rapidly adapting and slowly adapting response profiles are further subdivided into two categories: type I and type II. Type I responses have receptive fields with small, sharp borders whereas type II responses have large, poorly defined receptive fields.

Figure 1. Representative response profiles of RA, SAI and SAII afferents to touch.

Figure 1

Open in a new tab

Extracellular recordings were made from individual somatosensory afferents using a semi-intact preparation consisting of mouse hairy skin innervated by the saphenous nerve. Plots of instantaneous firing frequency versus time are shown for an RA afferent (conduction velocity = 7 m·s-1), SAI afferent (conduction velocity = 12 m·s-1) and SAII afferent (conduction velocity = 11 m·s-1). Arrows indicate the onset and offset of mechanical stimuli. The prolonged response of the RA fiber reflects vibration in the stimulus probe. Afferents were classified based on conduction velocity, adaptation properties and variance of firing frequency.

Rapidly adapting type I (RAI) afferents respond to bending of hairs, gentle rubbing of skin, and vibration (Brown and Iggo, 1967 2630; Petit and Burgess, 1968; Iggo and Findlater, 1984). In the hand, RAI afferents are the most sensitive mechanoreceptors, responding to surface features as small as 2 μm. Rapidly adapting type II responses (RAII), which have been definitively correlated with Pacinian corpuscles (Iggo and Findlater, 1984; Bell et al., 1994), respond most robustly to high frequency vibration (Hunt and McIntyre, 1960). Slowly adapting type I (SAI) afferents respond specifically to indentation of skin containing Merkel cell-neurite complexes (Iggo and Muir, 1969), and have the highest spatial resolution of the afferent response profiles. Slowly adapting type II (SAII) afferents, many of which are spontaneously active, respond more effectively to skin stretch than to indentation (Brown and Iggo, 1967; Leem et al., 1993). In addition, many respond preferentially to specific orientations (Hamann, 1995). Finally, C-mechanoreceptors are rapidly adapting C-fibers that respond to both skin indentation and bending of hairs (Nordin, 1990).

Definitive correlation of afferent responses and sensory perception was not possible until the advent of primary afferent microneurography in awake humans. In this experimental paradigm, extremely thin electrodes record electrical impulses from neurons innervating the skin. Mechanical stimuli such as a pen point are applied to the hand, and single afferent responses associated with a tactile unit are identified. The electrode is then used to stimulate the single afferent fiber, and the subject reports the quality of sensation generated. Most of these studies have been performed in the hand, which has a high density of mechanosensitive afferents (Vallbo et al., 1984; Johnson and Hsiao, 1992). Such studies, combined with traditional electrophysiological recordings in primates, indicate that rapidly adapting afferents encode skin motion. These afferents are integral to transmitting information related to low frequency vibration and the sensation of slip between skin and a grasped object (Johnson et al., 2000). Microstimulation of RA afferents in humans has been described as a “wobbling” or “fluttering feeling” with very short electrical trains described as a “tap” (Vallbo et al., 1984). RAII or PC afferents, which respond best to vibratory stimuli, are described, not surprisingly, by human subjects as a “vibration” sensation. Since SAII responses are more sensitive to forces tangential to the skin than indentation, they likely encode skin stretch. Microstimlation of SAII afferents has been described as percepts ranging from sustained lateral pulling of the skin to flutter (Vallbo et al., 1984). By contrast, SAI afferents encode texture and small feature discrimination, and are necessary for fine tactile discrimination tasks, such as detecting Braille-like patterns (Johnson et al., 2000). They are extremely sensitive to local deformations: a single Braille-like dot elicits 20 times more firing than a smooth surface (Phillips and Johnson, 1981). Microstimulation of single SAI afferents induced humans to feel a faint and uniform pressure, much like a “soft painting brush held tangentially against the skin” (Vallbo et al., 1984).

Merkel cell-neurite complexes underlie SAI responses

In 1969, Iggo and Muir identified highly touch-sensitive spots in hairy skin, called touch domes, as the source of a slowly adapting response (Iggo and Muir, 1969). These structures are visible in depilated rodents and cats, and Iggo and Muir were able to characterize the response profile. They noted that SAI fibers are completely silent in the absence of mechanical stimulation of the touch dome. Sustained touch-dome indentation elicits a burst of action potentials followed by a slowly adapting response (Fig. 1). The sustained element of the response is characterized by an irregular firing pattern, with interspike intervals ranging from <20 to 220 ms. This feature, in particular, differentiates SAI responses from SAII responses, which have a highly regular sustained firing pattern. Iggo and Muir noted that the apex of the touch dome was highly sensitive to indentation, but firing rates dropped precipitously with indentation of the surrounding tissue. Measurement of conduction velocity by electrical stimulation of touch domes indicated that they are innervated by rapidly conducting Aβ afferents.

Each touch dome generally is innervated by a single Aβ afferent that branches repeatedly, ultimately contacting a cluster of ∼10–100 epidermal Merkel cells (Fig. 2A; Munger, 1965; Iggo and Muir, 1969). The afferent has large terminals that contact the side of the Merkel cell nearest the basal lamina. This anatomy is reminiscent of other sensory receptor cells, such as inner-ear hair cells. Merkel cells are located in the basal layer of the epidermis, just above the basal lamina. In touch domes, the keratinocytes that overlie Merkel cells are columnar and form more layers, contributing to its dome-like appearance. The dermal portion of the dome contains collagen bundles woven in a tight mesh. The dermis beneath the dome is well vascularized, suggesting high metabolic activity by cells in the touch dome.

Figure 2. Structure of Merkel cell-neurite complexes.

Figure 2

Open in a new tab

(A) A confocal micrograph shows a cluster of Merkel cell-neurite complexes in a mouse touch dome. Skin was dissected from a transgenic Math1/nGFP mouse. In these transgenic mice, Math1 enhancer sequences drive expression of nuclear-localized GFP (green) specifically in Merkel cells in the skin (Lumpkin et al., 2003). Within a touch tome, clusters of Merkel cells are innervated by branches (arrowheads) of a single SAI afferent (red), which was labeled in vivo by subcutaneous injection of the styryl dye FM5-95 (Meyers et al., 2003). (B) Schematic of a Merkel cell (green) and its somaotsensory afferent (red) in a touch dome (adapted from Iggo and Muir 1969). Merkel cells are connected to overlying keratinocytes by desmosomes and actin-filled microvilli. Large, dense core vesicles are cluster below the Merkel cell's characteristically lobulated nucleus. These vesicles are closely apposed to the afferent terminal and often cluster around synaptic densities. Merkel cells are contacted exclusively by myelinated axons.

Recent experiments by Woodbury and Koerber have confirmed Iggo and Muir's original correlative work indicating that Merkel cell-neurite complexes mediate SAI response (Woodbury and Koerber, 2007). In a technical tour de force, they developed a mouse semi-intact preparation to perform electrophysiological recordings from the somata of somatosensory neurons with their cutaneous receptive fields intact. By filling electrodes with a tracer, these authors labeled the afferent projections of recorded neurons. Afferents with SAI-response profiles innervated Merkel-cell clusters in touch domes.

Ultrastructure of Merkel cells

Merkel cell-neurite complexes are found only in highly touch-sensitive areas of skin, such as touch domes, finger tips, whisker follicles and palatine ridges (Fig. 3). Based on this localization and the ultrastructure of Merkel cell-neurite complexes, Merkel cells were first hypothesized to be sensory receptor cells that transduce innocuous touch.

Figure 3. Merkel cells (green) cluster in touch domes in hairy skin (A; arrowheads) and palatine ridges in oral epithelia (B; arrows).

Figure 3

Open in a new tab

Projections of confocal z-series show Merkel cells in intact tissue from Math1/nGFP transgenic mice.

If this model is correct, Merkel cells must fulfill two criteria: 1) they must be intrinsically mechanically sensitive and 2) they must signal the associated afferent through synaptic transmission. Merkel-cell ultrastructure provides some evidence for both functions (Fig. 2B). Merkel cells have thin microvilli, 1–2 μm in length, which project into overlying keratinocytes (Brown and Iggo, 1967; Iggo and Muir, 1969). These processes are reminiscent of mechanosensitive steriocilia in hair cells and are postulated to detect deformations due to indentation of epithelial surfaces (Iggo and Muir, 1969). Additionally, Merkel cells and keratinocytes are linked by desmosomes, suggesting stiff structural attachments (Munger, 1965; Iggo and Muir, 1969).

Ultrastructural and molecular evidence also supports the notion of synaptic transmission between Merkel cells and the afferent terminal. There is a close association between Merkel cells and enlarged terminals of somatosensory afferents (Iggo and Muir, 1969). The contact between a Merkel cell and an afferent terminal is marked by a synapse-like granular cleft and a postsynaptic-like thickening of the afferent terminal's membrane (Iggo and Muir, 1969; Hartschuh and Weihe, 1980). In Merkel cells, the presynaptic active-zone protein Piccolo localizes to these contact sites, which confirms that they are presynaptic active zones (Haeberle et al., 2004). Along with Piccolo, Merkel cells express a number of molecules that are essential for evoked synaptic vesicle release, including SNARE-complex proteins, synaptotagmins and voltage-activated Ca2+ channels (Haeberle et al., 2004).

Afferent terminals contain an abundance of mitochondria and have small clear-core vesicles (Munger, 1965). In one report, a possible reciprocal synapse was observed with an asymmetrical electron density facing the Merkel cell, accompanied by a cluster of small clear-core vesicles in the neuron (Mihara et al., 1979)

The presence of synapses suggests that Merkel cells are sensory cells that function in the skin much as hair cells and photoreceptors function in hearing and vision; however, Merkel cells do not contain the small clear-core vesicles characteristic of these sensory synapses and fast neurotransmission in general. Instead, Merkel cells contain large, dense-core granules that cluster between their nuclei and the afferent terminal (Mihara et al., 1979; Hartschuh and Weihe, 1980). Large dense-core vesicles are generally thought to release neuropeptides or hormones. Consistent with this notion, immunohistochemical studies demonstrate that Merkel cells produce neuropeptides (Halata et al., 2003), including vasoactive intestinal peptide (Hartschuh et al., 1983) and cholecystokinin octapeptide (Haeberle et al., 2004). Merkel cells also express vesicular monoamine transporter 1, which is an isoform predominantly found in neuroendocrine cells (Weihe et al., 1998).

Dense-core vesicles may also contain the excitatory neurotransmitter glutamate (Morimoto et al., 2003). Thus, Merkel cells may employ large dense-core vesicles for mechanically triggered, excitatory transmission. The recent discovery that Merkel cells express vesicular glutamate transporters further supports the presence of glutamatergic synaptic transmission (Hitchcock et al., 2004; Nunzi et al., 2004). Moreover, whole-cell recordings and live-cell imaging of dissociated Merkel cells revealed voltage-activated ion channels and depolarization-evoked Ca2+ transients (Yamashita et al., 1992; Haeberle et al., 2004; Piskorowski et al., 2008), indicating that Merkel cells are excitable cells. Although these studies collectively suggest that Merkel cells are presynaptic cells, a direct demonstration that depolarization elicits neurotransmitter release from Merkel cells is still lacking.

Do Merkel cells contribute to mechanotransduction?

These histological and molecular findings have fueled a long-standing controversy over whether Merkel cells, somatosensory afferents or both transduce touch in Merkel cell-neurite complexes. Iggo and Muir (1969) noted that SAI responses have an irregular firing pattern as opposed to the regular firing rate of other slowly-adapting, mammalian mechanoreceptors. This irregular firing pattern could be generated by chemical synaptic transmission between Merkel cells and their afferent terminals, if quanta are released in response to mechanical stimulation with a Poisson-like process (Horch et al., 1974). Tellingly, synaptic transmission has not been implicated in other, regularly firing, cutaneous mechanoreceptors. On the other hand, such firing patterns could be generated by direct mechanical stimulation of the branched afferent terminal itself. Indeed, Gottschaldt and Vahle-Hinz observed SAI mechanoelectric response latencies of ∼0.2 ms, which they interpreted as being too fast for chemical synaptic transmission (Smith and Creech, 1967; Gottschaldt and Vahle-Hinz, 1981).

Because the location of Merkel cells in the epidermis prevents direct electrophysiological recordings with microelectrodes in intact epithelia, several groups have attempted pharmacological manipulations of semi-intact skin-nerve preparations to determine if synaptic transmission is required for normal SAI responses. Pacitti and Findlater reported that voltage-activated Ca2+ channel inhibitors impair SAI responses in skin-nerve preparations (Pacitti and Findlater, 1988). Additionally, kynurenate, a broad spectrum glutamate-receptor antagonist, reduced the responsiveness of Merkel cell-neurite complexes to touch, indicating that glutamatergic neurotransmission plays a role in touch reception (Fagan and Cahusac, 2001). Furthermore, the peripheral afferents of some somatosensory neurons contain ionotropic glutamate receptors (Coggeshall and Carlton, 1998) and Merkel cells express metabotropic glutamate receptors (Tachibana et al., 2003). The NMDA-type glutamate receptor antagonist MK-801 inhibited the static phase of the SAI response, but an AMPA-type glutamate receptor antagonist had no effect (Cahusac et al., 2005). SAI responses in rat sinus-hair follicles were also reduced by antagonists of two serotonin receptors, 5-HT2 and 5-HT3 (He et al., 2003). Surprisingly, these Merkel cell-neurite complexes show immunoreactivity against 5-HT1 receptors, but not 5-HT2 and 5-HT3 (Tachibana et al., 2005). Other antagonists of neuromodulators have been reported to have no effect on SAI responses (Gottschaldt and Vahle-Hinz, 1982).

Studies asking whether Merkel cells are required for the SAI response have produced conflicting results. Eliminating Merkel cells by photoablation greatly diminished SAI responses in one study (Mills and Diamond, 1995) but not in another (Ikeda et al., 1994). The contradictory outcomes can be explained either by incomplete photoablation of Merkel cells or by collateral damage of the somatosensory afferent (Senok et al., 1996). Another study examined slowly adapting responses in mice lacking the low affinity neurotrophin receptor, p75, which have dramatically reduced numbers of touch domes containing Merkel cells (Kinkelin et al., 1999). The authors noted that there were no differences in slowly adapting responses in skin nerve preparations in these mice compared to control animals. Based on these phenotypes the authors conclude that Merkel cells are not required for slowly adapting responses; however, there are a few caveats that confound this interpretation. First, this report did not distinguish between SAI and other slowly adapting fiber types. Second, maximal firing rates of slowly adapting responses were at least an order of magnitude lower (∼5 s-1) than those characteristic of SAI afferents. Finally, touch domes that were detected had only slightly reduced numbers of Merkel cells.

Several groups have attempted to determine whether Merkel cells directly respond to mechanical stimulation. Baumann and colleagues have used a semi-intact sinus hair preparation to investigate Merkel cells in situ (Chan et al., 1996). They incubated the entire sinus-hair follicle in fluorescent, Ca2+ indicators and treated the preparation with mechanical stimuli and solutions containing depolarizing concentrations of K+. Intriguingly, these authors observed that direct touch and hypotonic solutions evoked Ca2+ transients in presumptive Merkel cells. Because neuronal contacts are preserved in this preparation, these studies cannot distinguish whether mechanical stimuli directly activate Merkel cells, or whether mechanical stimuli activate afferent terminals, which then signal to Merkel cells through reciprocal synapses (Mihara et al., 1979).

To test the hypothesis that Merkel cells are directly activated by mechanical stimuli, several groups have studied dissociated Merkel cells. Tazaki and Suzuki (1998) observed hypotonic-induced [Ca2+] increases in a few dissociated Merkel cells labeled with the fluorescent marker quinacrine. Though promising, the difficulty of dissociating and identifying Merkel cells likely prevented analysis of additional cells. To circumvent these challenges, Lumpkin and colleagues have developed methods to dissociate, purify and culture Merkel cells (Haeberle et al., 2004) by employing transgenic mice that specifically express green fluorescent protein (GFP) in Merkel cells in the skin (Lumpkin et al., 2003). Recent studies confirm that Merkel cells are directly activated by hypotonic-evoked cell swelling and indicate that this stimulus is transduced by Ca2+-permeable ion channels (Haeberle et al., 2008). Other groups have failed to demonstrate mechanically evoked responses in dissociated Merkel cells, which may reflect the loss of Merkel-cell processes in those experimental conditions (Nurse and Cooper, 1988; Yamashita et al., 1992). By contrast, in serum-free culture conditions, Merkel cells display actin-filled processes (Fig. 4) as well as fine, cytoplasmic tendrils (Haeberle et al., 2008).

Figure 4. Morphology of dissociated Merkel cells.

Figure 4

Open in a new tab

Merkel cells were dissociated from the hairy skin of Math1/nGFP transgenic mice, purified by fluorescence-activated cell sorting based on GFP fluorescence (green), then cultured for two days in serum-free keratinocyte media. Actin-filled, dendritic processes were visualized by labeling Merkel cells with rhodamine-phalloidin (red). Details of the isolation and culture methods are given in Haeberle et al. (2004 and 2008).

Merkel cells in the oral epithelium

Merkel cells are abundant in oral epithelia of numerous species, including salamanders, rodents and primates (Tachibana et al., 1997; Halata and Baumann, 1999; Toyoshima et al., 1999). In most species, they are found at highest concentration in palatine ridges (Fig. 3B; Tachibana et al., 1997; Halata et al., 1999; Nunzi et al., 2004). SA responses characteristic of Merkel cell-neurite complexes have been recorded from the goat palate and human oral mucosa (Halata et al., 1999; Trulsson and Johansson, 2002). Nonetheless, several groups have suggested that Merkel cells in the oral cavity may participate in functions other than touch reception. For example, they have been postulated to function in taste-bud morphogenesis in salamanders (Toyoshima et al., 1999) and in wound healing in mammals (Righi et al., 2006).

Morphological evidence suggests that distinct subpopulations of Merkel cells reside in the palate (Tachibana et al., 1997). About half of Merkel cells in the rodent palate are morphologically similar to those in touch domes (Fig. 2 and Tachibana et al., 1997). As described above, these roughly spherical cells are characterized by microvilli (Toyoshima et al., 1998), synaptic contacts with sensory afferents and expression of vesicular glutamate transporters (Nunzi et al., 2004). In the goat palate, only innervated, spheroid Merkel cells have been observed (Halata et al., 1999).

In rodent palates, the remaining Merkel cells, termed dendritic Merkel cells, lack obvious innervation (Tachibana et al., 1997). Instead, they have long cytoplasmic processes containing dense-core vesicles, reminiscent of neuronal projections. Intriguingly, dendritic Merkel cells have recently been described in rat touch domes (Nakafusa et al., 2006). Moreover, Merkel cells dissociated from touch domes adopt this morphology when cultured in the absence of sensory neurons (Fig. 4). Because they apparently lack innervation, dendritic Merkel cells have been proposed to be a functionally distinct subpopulation that releases paracrine signals to influence the development or repair of oral epithelia (Tachibana et al., 1997). Alternatively, dendritic Merkel cells may reflect a developmentally immature state (Kim and Holbrook, 1995). One argument against the latter interpretation is that dendritic Merkel cells are highly abundant in the palates of adult rodents, whereas Merkel cells in the oral cavity develop in the first few postnatal weeks (Tachibana et al., 2000).

Conclusions and open questions

Since their discovery in 1875, Merkel cells have been proposed to function as mechanosensory cells based on their morphology and location (Merkel, 1875; reviewed in Tachibana and Nawa, 2002; Halata et al., 2003). Consistent with this model, recent molecular and physiological studies demonstrate that Merkel cells are neuronal-like cells that are poised to play an active role in touch reception. Despite this progress, key questions remain open. First, is mechanotransduction in Merkel cells necessary for SAI responses? Second, is direct touch sufficient to depolarize Merkel cells and evoke synaptic vesicle release? Third, does neurotransmitter release from Merkel cells send excitatory or modulatory signals to sensory neurons? Alternatively, Merkel cells may release autocrine signals or paracrine signals to surrounding keratinocytes. Finally, do dendritic Merkel cells reflect the development of Merkel cells in adult animals and, if so, what progenitor cell type begets new Merkel cells in mature tissues? This issue is particularly important in the oral epithelium, which is an environment prone to trauma and high mechanical stress (Ishizaki et al., 2006).

Acknowledgments

We thank Ms. Jennifer LaRey for providing valuable comments on the manuscript. Electrophysiological recordings were performed by E.A.L. in the laboratory of Dr. Jonathon Howard at the University of Washington. The authors were supported by grants from the Gillson Longenbaugh Foundation (to E.A.L.), the National Institutes of Health (NIAMS grant AR051219 to E.A.L.), and the National Science Foundation (graduate fellowship to H.H.). The original publication is available at springerlink.com: http://www.springerlink.com/content/n240t84631716wm5/?p=f055da8df911443bb78ee5b67afd4e4b&pi=4

Abbreviations

PC

Pacinian corpuscle

RA

rapidly adapting

SAI

slowly adapting type I

SAII

slowly adapting type II

SNARE

soluble N-ethylmaleimide sensitive factor attachment receptor

5-HT

5-hydroxytryptamine/serotonin

References

  1. Adrian ED, Zotterman Y. The impulses produced by sensory nerve endings: Part 3. Impulses set up by Touch and Pressure. J Physiol. 1926;61:465–483. doi: 10.1113/jphysiol.1926.sp002308. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Bautista DM, Jordt SE, Nikai T, Tsuruda PR, Read AJ, Poblete J, Yamoah EN, Basbaum AI, Julius D. TRPA1 mediates the inflammatory actions of environmental irritants and proalgesic agents. Cell. 2006;124:1269–1282. doi: 10.1016/j.cell.2006.02.023. [DOI] [PubMed] [Google Scholar]
  3. Bell J, Bolanowski S, Holmes MH. The structure and function of Pacinian corpuscles: a review. Prog Neurobiol. 1994;42:79–128. doi: 10.1016/0301-0082(94)90022-1. [DOI] [PubMed] [Google Scholar]
  4. Boring EG. Sensation and perception in the history of experimental psychology. New York: Appleton Century Crofts Inc.; 1942. [Google Scholar]
  5. Brown AG, Iggo A. A quantitative study of cutaneous receptors and afferent fibres in the cat and rabbit. J Physiol. 1967;193:707–733. doi: 10.1113/jphysiol.1967.sp008390. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Cahusac PM, Senok SS, Hitchcock IS, Genever PG, Baumann KI. Are unconventional NMDA receptors involved in slowly adapting type I mechanoreceptor responses? Neuroscience. 2005;133:763–773. doi: 10.1016/j.neuroscience.2005.03.018. [DOI] [PubMed] [Google Scholar]
  7. Caterina MJ, Leffler A, Malmberg AB, Martin WJ, Trafton J, Petersen-Zeitz KR, Koltzenburg M, Basbaum AI, Julius D. Impaired nociception and pain sensation in mice lacking the capsaicin receptor. Science. 2000;288:306–313. doi: 10.1126/science.288.5464.306. [DOI] [PubMed] [Google Scholar]
  8. Chan E, Yung WH, Baumann KI. Cytoplasmic Ca2+ concentrations in intact Merkel cells of an isolated, functioning rat sinus hair preparation. Exp Brain Res. 1996;108:357–366. doi: 10.1007/BF00227259. [DOI] [PubMed] [Google Scholar]
  9. Coggeshall RE, Carlton SM. Ultrastructural analysis of NMDA, AMPA, and kainate receptors on unmyelinated and myelinated axons in the periphery. J Comp Neurol. 1998;391:78–86. doi: 10.1002/(sici)1096-9861(19980202)391:1<78::aid-cne7>3.3.co;2-8. [DOI] [PubMed] [Google Scholar]
  10. Daniels R, McKemy D. Mice left out in the cold: commentary on the phenotype of TRM8-nulls. Mol Pain. 2007;3:23. doi: 10.1186/1744-8069-3-23. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Dodt E, Zotterman Y. Mode of action of warm receptors. Acta Physiol Scand. 1952;26:345–357. doi: 10.1111/j.1748-1716.1952.tb00916.x. [DOI] [PubMed] [Google Scholar]
  12. Fagan BM, Cahusac PM. Evidence for glutamate receptor mediated transmission at mechanoreceptors in the skin. Neuroreport. 2001;12:341–347. doi: 10.1097/00001756-200102120-00032. [DOI] [PubMed] [Google Scholar]
  13. Gottschaldt KM, Vahle-Hinz C. Merkel cell receptors: structure and transducer function. Science. 1981;214:183–186. doi: 10.1126/science.7280690. [DOI] [PubMed] [Google Scholar]
  14. Gottschaldt KM, Vahle-Hinz C. Evidence against transmitter function of met-enkephalin and chemosynaptic impulse generation in “Merkel cell” mechanoreceptors. Exp Brain Res. 1982;45:459–463. doi: 10.1007/BF01208608. [DOI] [PubMed] [Google Scholar]
  15. Haeberle H, Bryan LA, Vadakkan TJ, Dickinson ME, Lumpkin EA. Swelling-activated Ca2+ channels trigger Ca2+ signals in Merkel cells. PLoS ONE. 2008 doi: 10.1371/journal.pone.0001750. In press. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Haeberle H, Fujiwara M, Chuang J, Medina MM, Panditrao M, Bechstedt S, Howard J, Lumpkin EA. Molecular profiling reveals synaptic release machinery in Merkel cells. Proc Natl Acad Sci U S A. 2004;101:14503–14508. doi: 10.1073/pnas.0406308101. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Halata Z, Baumann KI. Sensory nerve endings in the hard palate and papilla incisiva of the rhesus monkey. Anat Embryol (Berl) 1999;199:427–437. doi: 10.1007/s004290050241. [DOI] [PubMed] [Google Scholar]
  18. Halata Z, Grim M, Bauman KI. Friedrich Sigmund Merkel and his “Merkel cell”, morphology, development, and physiology: Review and new results. Anat Rec. 2003;271A:225–239. doi: 10.1002/ar.a.10029. [DOI] [PubMed] [Google Scholar]
  19. Halata Z, Cooper BY, Baumann KI, Schwegmann C, Friedman RM. Sensory nerve endings in the hard palate and papilla incisiva of the goat. Exp Brain Res. 1999;129:218–228. doi: 10.1007/s002210050892. [DOI] [PubMed] [Google Scholar]
  20. Hamann W. Mammalian cutaneous mechanoreceptors. Prog Biophys Mol Biol. 1995;64:81–104. doi: 10.1016/0079-6107(95)00011-9. [DOI] [PubMed] [Google Scholar]
  21. Harrington T, Merzenich MM. Neural coding in the sense of touch: human sensations of skin indentation compared with the responses of slowly adapting mechanoreceptive afferents innervating the hairy skin of monkeys. Exp Brain Res. 1970;10:251–264. doi: 10.1007/BF00235049. [DOI] [PubMed] [Google Scholar]
  22. Hartschuh W, Weihe E. Fine structural analysis of the synaptic junction of Merkel cell-axon-complexes. J Invest Dermatol. 1980;75:159–165. doi: 10.1111/1523-1747.ep12522555. [DOI] [PubMed] [Google Scholar]
  23. Hartschuh W, Weihe E, Yanaihara N, Reinecke M. Immunohistochemical localization of vasoactive intestinal polypeptide (VIP) in Merkel cells of various mammals: evidence for a neuromodulator function of the Merkel cell. J Invest Dermatol. 1983;81:361–364. doi: 10.1111/1523-1747.ep12519966. [DOI] [PubMed] [Google Scholar]
  24. He L, Tuckett RP, English KB. 5-HT2 and 3 receptor antagonists suppress the response of rat type I slowly adapting mechanoreceptor: an in vitro study. Brain Res. 2003;969:230–236. doi: 10.1016/s0006-8993(03)02335-7. [DOI] [PubMed] [Google Scholar]
  25. Hensel H, Boman KK. Afferent impulses in cutaneous sensory nerves in human subjects. J Neurophysiol. 1960;23:564–578. doi: 10.1152/jn.1960.23.5.564. [DOI] [PubMed] [Google Scholar]
  26. Hitchcock IS, Genever PG, Cahusac PM. Essential components for a glutamatergic synapse between Merkel cell and nerve terminal in rats. Neurosci Lett. 2004;362:196–199. doi: 10.1016/j.neulet.2004.02.071. [DOI] [PubMed] [Google Scholar]
  27. Horch KW, Whitehorn D, Burgess PR. Impulse generation in type I cutaneous mechanoreceptors. J Neurophysiol. 1974;37:267–281. doi: 10.1152/jn.1974.37.2.267. [DOI] [PubMed] [Google Scholar]
  28. Hunt CC, McIntyre AK. An analysis of fibre diameter and receptor characteristics of myelinated cutaneous afferent fibres in cat. J Physiol. 1960;153:99–112. doi: 10.1113/jphysiol.1960.sp006521. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Iggo A, Muir AR. The structure and function of a slowly adapting touch corpuscle in hairy skin. J Physiol. 1969;200:763–796. doi: 10.1113/jphysiol.1969.sp008721. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Iggo A, Findlater GS. A review of Merkel cell mechanisms. In: Hamann W, Iggo A, editors. Sensory receptor mechanisms. Singapore: World Scientific Publ. Co.; 1984. pp. 117–131. [Google Scholar]
  31. Ikeda I, Yamashita Y, Ono T, Ogawa H. Selective phototoxic destruction of rat Merkel cells abolishes responses of slowly adapting type I mechanoreceptor units. J Physiol. 1994;479(Pt 2):247–256. doi: 10.1113/jphysiol.1994.sp020292. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Ishizaki K, Sakurai K, Tazaki M, Inoue T. Response of Merkel cells in the palatal rugae to the continuous mechanical stimulation by palatal plate. Somatosens Mot Res. 2006;23:63–72. doi: 10.1080/08990220600741069. [DOI] [PubMed] [Google Scholar]
  33. Johnson KO, Hsiao SS. Neural mechanisms of tactual form and texture perception. Annu Rev Neurosci. 1992;15:227–250. doi: 10.1146/annurev.ne.15.030192.001303. [DOI] [PubMed] [Google Scholar]
  34. Johnson KO, Yoshioka T, Vega-Bermudez F. Tactile functions of mechanoreceptive afferents innervating the hand. J Clin Neurophysiol. 2000;17:539–558. doi: 10.1097/00004691-200011000-00002. [DOI] [PubMed] [Google Scholar]
  35. Kim DK, Holbrook KA. The appearance, density, and distribution of Merkel cells in human embryonic and fetal skin: their relation to sweat gland and hair follicle development. J Invest Dermatol. 1995;104:411–416. doi: 10.1111/1523-1747.ep12665903. [DOI] [PubMed] [Google Scholar]
  36. Kinkelin I, Stucky CL, Koltzenburg M. Postnatal loss of Merkel cells, but not of slowly adapting mechanoreceptors in mice lacking the neurotrophin receptor p75. Eur J Neurosci. 1999;11:3963–3969. doi: 10.1046/j.1460-9568.1999.00822.x. [DOI] [PubMed] [Google Scholar]
  37. Knibestöl M. Stimulus-response functions of slowly adapting mechanoreceptors in the human glabrous skin area. J Physiol. 1975;245:63–80. doi: 10.1113/jphysiol.1975.sp010835. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Knibestöl M, Vallbo AB. Intensity of sensation related to activity of slowly adapting mechanoreceptive units in the human hand. J Physiol. 1980;300:251–267. doi: 10.1113/jphysiol.1980.sp013160. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Kwan KY, Allchorne AJ, Vollrath MA, Christensen AP, Zhang DS, Woolf CJ, Corey DP. TRPA1 contributes to cold, mechanical, and chemical nociception but is not essential for hair-cell transduction. Neuron. 2006;50:277–289. doi: 10.1016/j.neuron.2006.03.042. [DOI] [PubMed] [Google Scholar]
  40. Leem JW, Willis WD, Weller SC, Chung JM. Differential activation and classification of cutaneous afferents in the rat. J Neurophysiol. 1993;70:2411–2424. doi: 10.1152/jn.1993.70.6.2411. [DOI] [PubMed] [Google Scholar]
  41. Lumpkin EA, Collisson T, Parab P, Omer-Abdalla A, Haeberle H, Chen P, Doetzlhofer A, White P, Groves A, Segil N, Johnson JE. Math1-driven GFP expression in the developing nervous system of transgenic mice. Gene Expr Patterns. 2003;3:389–395. doi: 10.1016/s1567-133x(03)00089-9. [DOI] [PubMed] [Google Scholar]
  42. Merkel F. Tastzellen und Tastkörperchen bei den Haustehieren und beim Menschen. Arch Mikrosk Anat. 1875;11:636–652. [Google Scholar]
  43. Meyers JR, MacDonald RB, Duggan A, Lenzi D, Standaert DG, Corwin JT, Corey DP. Lighting up the senses: FM1-43 loading of sensory cells through nonselective ion channels. J Neurosci. 2003;23:4054–4065. doi: 10.1523/JNEUROSCI.23-10-04054.2003. [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Mihara M, Hashimoto K, Ueda K, Kumakiri M. The specialized junctions between Merkel cell and neurite: an electron microscopic study. J Invest Dermatol. 1979;73:325–334. doi: 10.1111/1523-1747.ep12550322. [DOI] [PubMed] [Google Scholar]
  45. Mills LR, Diamond J. Merkel cells are not the mechanosensory transducers in the touch dome of the rat. J Neurocytol. 1995;24:117–134. doi: 10.1007/BF01181555. [DOI] [PubMed] [Google Scholar]
  46. Morimoto R, Hayashi M, Yatsushiro S, Otsuka M, Yamamoto A, Moriyama Y. Co-expression of vesicular glutamate transporters (VGLUT1 and VGLUT2) and their association with synaptic-like microvesicles in rat pinealocytes. J Neurochem. 2003;84:382–391. doi: 10.1046/j.1471-4159.2003.01532.x. [DOI] [PubMed] [Google Scholar]
  47. Mountcastle VB. The problem of sensing and the neural coding of sensory events. In: Quarton GC, Melnechuk T, Schmitt FO, editors. The Neurosciences. New York: Rockfeller University Press; 1967. pp. 393–408. [Google Scholar]
  48. Munger BL. The intraepidermal innervation of the snout skin of the opossum. A light and electron microscope study, with observations on the nature of Merkel's Tastzellen. J Cell Biol. 1965;26:79–97. doi: 10.1083/jcb.26.1.79. [DOI] [PMC free article] [PubMed] [Google Scholar]
  49. Nakafusa J, Narisawa Y, Shinogi T, Taira K, Tanaka T, Inoue T, Misago N. Changes in the number of Merkel cells with the hair cycle in hair discs on rat back skin. Br J Dermatol. 2006;155:883–889. doi: 10.1111/j.1365-2133.2006.07441.x. [DOI] [PubMed] [Google Scholar]
  50. Nordin M. Low-threshold mechanoreceptive and nociceptive units with unmyelinated (C) fibres in the human supraorbital nerve. J Physiol. 1990;426:229–240. doi: 10.1113/jphysiol.1990.sp018135. [DOI] [PMC free article] [PubMed] [Google Scholar]
  51. Nunzi MG, Pisarek A, Mugnaini E. Merkel cells, corpuscular nerve endings and free nerve endings in the mouse palatine mucosa express three subtypes of vesicular glutamate transporters. J Neurocytol. 2004;33:359–376. doi: 10.1023/B:NEUR.0000044196.45602.92. [DOI] [PubMed] [Google Scholar]
  52. Nurse C, Cooper E. Electrophysiological studies on Merkel cells isolated from rat vibrissal mechanoreceptors. In: Hnik P, Soukup T, Vejsada R, Zelena J, editors. Mechanoreceptors – Develpment, structure and function. New York: Plenum Press; 1988. pp. 189–194. [Google Scholar]
  53. Pacitti EG, Findlater GS. Calcium channel blockers and Merkel cells. Prog Brain Res. 1988;74:37–42. doi: 10.1016/s0079-6123(08)62995-7. [DOI] [PubMed] [Google Scholar]
  54. Petit D, Burgess PR. Dorsal column projection of receptors in cat hairy skin supplied by myelinated fibers. J Neurophysiol. 1968;31:849–855. doi: 10.1152/jn.1968.31.6.849. [DOI] [PubMed] [Google Scholar]
  55. Phillips JR, Johnson KO. Tactile spatial resolution. II. Neural representation of Bars, edges, and gratings in monkey primary afferents. J Neurophysiol. 1981;46:1192–1203. doi: 10.1152/jn.1981.46.6.1192. [DOI] [PubMed] [Google Scholar]
  56. Piskorowski RA, Haeberle H, Panditrao M, Lumpkin EA. Voltage-activated ion channels shape calcium signaling in Merkel cells. Pflugers Archiv Eur J Physiol. 2008 doi: 10.1007/s00424-008-0496-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
  57. Righi A, Betts C, Marchetti C, Marucci G, Montebugnoli L, Prati C, WEusebi L, Muzzi L, Ragazzini T, Foschini M. Merkel cells in the oral mucosa. Int J Surg Pathol. 2006;14:206–211. doi: 10.1177/1066896906290053. [DOI] [PubMed] [Google Scholar]
  58. Senok SS, Baumann KI, Halata Z. Selective phototoxic destruction of quinacrine-loaded Merkel cells is neither selective nor complete. Exp Brain Res. 1996;110:325–334. doi: 10.1007/BF00229133. [DOI] [PubMed] [Google Scholar]
  59. Simon S, de Araujo I, Gutierrez R, Nicholelis M. The neural mechanisms of gustation: a distributed processing code. Nat Rev Neurosci. 2006;7:890–901. doi: 10.1038/nrn2006. [DOI] [PubMed] [Google Scholar]
  60. Smith KR, Creech BJ. Effects of pharmacological agents on the physiological responses of hair discs. Exp Neurol. 1967;19:477–482. doi: 10.1016/0014-4886(67)90167-7. [DOI] [PubMed] [Google Scholar]
  61. Tachibana T, Nawa T. Recent progress in studies on Merkel cell biology. Anat Sci Int. 2002;77:26–33. doi: 10.1046/j.0022-7722.2002.00008.x. [DOI] [PubMed] [Google Scholar]
  62. Tachibana T, Fujiwara N, Nawa T. Postnatal differentiation of Merkel cells in the rat palatine mucosa, with special reference to the timing of peripheral nerve development and the potency of cell mitosis. Anat Embryol (Berl) 2000;202:359–367. doi: 10.1007/s004290000124. [DOI] [PubMed] [Google Scholar]
  63. Tachibana T, Endoh M, Kumakami R, Nawa T. Immunohistochemical expressions of mGluR5, P2Y2 receptor, PLC-beta1, and IP3R-I and -II in Merkel cells in rat sinus hair follicles. Histochem Cell Biol. 2003;120:13–21. doi: 10.1007/s00418-003-0540-5. [DOI] [PubMed] [Google Scholar]
  64. Tachibana T, Endoh M, Fujiwara N, Nawa T. Receptors and transporter for serotonin in Merkel cell-nerve endings in the rat sinus hair follicle. An immunohistochemical study. Arch Histol Cytol. 2005;68:19–28. doi: 10.1679/aohc.68.19. [DOI] [PubMed] [Google Scholar]
  65. Tachibana T, Yamamoto H, Takahashi N, Kamegai T, Shibanai S, Iseki H, Nawa T. Polymorphism of Merkel cells in the rodent palatine mucosa: immunohistochemical and ultrastructural studies. Arch Histol Cytol. 1997;60:379–389. doi: 10.1679/aohc.60.379. [DOI] [PubMed] [Google Scholar]
  66. Tazaki M, Suzuki T. Calcium inflow of hamster Merkel cells in response to hyposmotic stimulation indicate a stretch activated ion channel. Neurosci Lett. 1998;243:69–72. doi: 10.1016/s0304-3940(98)00066-4. [DOI] [PubMed] [Google Scholar]
  67. Toyoshima K, Seta Y, Takeda S, Harada H. Identification of Merkel cells by an antibody to villin. J Histochem Cytochem. 1998;46:1329–1334. doi: 10.1177/002215549804601113. [DOI] [PubMed] [Google Scholar]
  68. Toyoshima K, Seta Y, Toyono T, Takeda S. Merkel cells are responsible for the initiation of taste organ morphogenesis in the frog. J Comp Neurol. 1999;406:129–140. doi: 10.1002/(sici)1096-9861(19990329)406:1<129::aid-cne8>3.0.co;2-p. [DOI] [PubMed] [Google Scholar]
  69. Trulsson M, Johansson RS. Orofacial mechanoreceptors in humans: encoding characteristics and responses during natural orofacial behaviors. Behav Brain Res. 2002;135:27–33. doi: 10.1016/s0166-4328(02)00151-1. [DOI] [PubMed] [Google Scholar]
  70. Vallbo AB, Olsson KA, Westberg KG, Clark FJ. Microstimulation of single tactile afferents from the human hand. Sensory attributes related to unit type and properties of receptive fields. Brain. 1984;107(Pt 3):727–749. doi: 10.1093/brain/107.3.727. [DOI] [PubMed] [Google Scholar]
  71. Weihe E, Hartschuh W, Schafer MK, Romeo H, Eiden LE. Cutaneous Merkel cells of the rat contain both dynorphin A and vesicular monoamine transporter type 1 (VMAT1) immunoreactivity. Can J Physiol Pharmacol. 1998;76:334–339. [PubMed] [Google Scholar]
  72. Werner G, Mountcastle VB. Neural Activity in Mechanoreceptive Cutaneous Afferents: Stimulus-Response Relations, Weber Functions, and Information Transmission. J Neurophysiol. 1965;28:359–397. doi: 10.1152/jn.1965.28.2.359. [DOI] [PubMed] [Google Scholar]
  73. Woodbury CJ, Koerber HR. Central and peripheral anatomy of slowly adapting type I low-threshold mechanoreceptors innervating trunk skin of neonatal mice. J Comp Neurol. 2007;505:547–561. doi: 10.1002/cne.21517. [DOI] [PubMed] [Google Scholar]
  74. Yamashita Y, Akaike N, Wakamori M, Ikeda I, Ogawa H. Voltage-dependent currents in isolated single Merkel cells of rats. J Physiol. 1992;450:143–162. doi: 10.1113/jphysiol.1992.sp019120. [DOI] [PMC free article] [PubMed] [Google Scholar]
  75. Zotterman Y. Touch, pain and tickling: an electro-physiological investigation on cutaneous sensory nerves. J Physiol. 1939;95:1–28. doi: 10.1113/jphysiol.1939.sp003707. [DOI] [PMC free article] [PubMed] [Google Scholar]

RESOURCES