Available online at www.sciencedirect.com R
Hearing Research 176 (2003) 1^10 www.elsevier.com/locate/heares
Ecto-ATPase activity sites in vestibular tissues: an ultracytochemical study in frog semicircular canals Luciana Gioglio a , Giancarlo Russo b , Mariarosa Polimeni a , Ivo Prigioni b
b;
a Dipartimento di Medicina Sperimentale, Universita' di Pavia, Via Forlanini 8, I-27100 Pavia, Italy Dipartimento di Scienze Fisiologiche e Farmacologiche, Universita' di Pavia, Via Forlanini 6, I-27100 Pavia, Italy
Received 28 March 2002; accepted 9 July 2002
Abstract The present study describes the localization and distribution of putative ecto-nucleoside-triphosphate-diphosphohydrolases in the frog semicircular canals. These enzymes provide the terminating mechanism of adenosine-5P-triphosphate (ATP) signalling. The localization of the ATP hydrolysis was mapped ultracytochemically using a one-step cerium citrate reaction. Electron-dense precipitates, indicating ecto-adenosine-triphosphatase (ecto-ATPase) activity, were found at the outer surface of plasma membranes of crista hair cells and supporting cells of the sensory epithelium, transitional cells and undifferentiated cells of the ampullar wall and dark cells constituting the secretory epithelium. Non-sensory cells of the ampulla usually exhibited reaction deposits at the level of both apical and basolateral membranes coming into contact with the endolymph and the perilymph respectively, while cells constituting the sensory epithelium showed evident differences in relation to their position. Hair cells and supporting cells of the peripheral regions exhibited clear reaction products both at the level of apical and basolateral membranes, while those of the isthmus region showed abundant reactivity only at the level of their apical membranes. Of particular interest was the observation that hair cell stereocilia exhibited an abundant ecto-ATPase activity, thus suggesting a possible colocalization of enzymatic sites with purinergic receptors and mechanotransduction channels. This strategic expression of ecto-ATPase sites could provide a rapid mechanism of ATP removal able to rapidly restore the sensitivity of transduction channels. In conclusion, the widespread distribution of ecto-ATPase sites at the level of sensory and non-sensory cells of the frog semicircular canals suggests that ATP may have a key role in controlling vestibular function. 2 2002 Elsevier Science B.V. All rights reserved. Key words: Ecto-ATPase activity ; Frog; Semicircular canal; Hair cell; Supporting cell; Dark cell; Ultracytochemistry
1. Introduction Considerable evidence has accumulated that extracellular nucleotides, particularly adenosine-5P-triphosphate (ATP), act as signalling molecules in the inner ear (Bobbin and Thompson, 1978; Eybalin, 1993; Kujawa et al., 1994; Housley, 1997). The nucleotide was found in the
* Corresponding author. Tel.: +39 (382) 507605; Fax: +39 (382) 507527. E-mail address:
[email protected] (I. Prigioni). Abbreviations: ATP, adenosine-5P-triphosphate; GTP, guanosine-5P-triphosphate; ecto-ATPase, ecto-adenosine-triphosphatase; E-NTPD-ase, ecto-nucleoside-triphosphate-diphosphohydrolase; Naþ /Kþ -ATPase, sodium/potassium adenosine-triphosphatase; P2X, ligand-gated purinergic; P2Y, metabotropic purinergic
perilymphatic and endolymphatic compartments of the cochlea (Munoz et al., 1995) in which it modulates the sound transduction and neurotransmission (Thorne and Housley, 1996; Housley and Ryan, 1997) as well as the activity of secretory marginal cells of the stria vascularis (Liu et al., 1995; Wangemann, 1995). In the vestibular system, recent studies have suggested a possible regulating role of ATP on neurotransmission (Aubert et al., 1994, 1995) and on the activity of secretory dark cells (Liu et al., 1995; Wangemann, 1995). Finally, evidence was provided about the presence of ligand-gated purinergic (P2X)- and metabotropic purinergic (P2Y)-like receptors in sensory and non-sensory cells of the ampullary epithelium in the frog (Butlen et al., 1998) and rat (Housley et al., 1998). Given that ATP exerts a widespread action on inner
0378-5955 / 02 / $ ^ see front matter 2 2002 Elsevier Science B.V. All rights reserved. PII: S 0 3 7 8 - 5 9 5 5 ( 0 2 ) 0 0 5 8 3 - X
HEARES 3977 22-1-03
2
L. Gioglio et al. / Hearing Research 176 (2003) 1^10
ear tissues, the localization of the enzymes which provide the terminating mechanism of ATP signalling is clearly relevant for inner ear function. In the cochlea (Vlajkovic et al., 1996, 1998), as in other tissues (Zimmermann, 1996), one of the main pathways removing ATP from the extracellular space is represented by ectonucleotidases, which comprise ecto-adenosine-triphosphatase (ecto-ATPase), ecto-apyrase, and 5P-nucleotidase. The nomenclature of these enzyme systems has recently been revised and the former two enzymes are now represented as ecto-nucleoside-triphosphate-diphosphohydrolase (E-NTPDase)-2 and E-NTPDase-1 respectively (Zimmermann, 1999). Ecto-ATPase is generally considered as a plasma membrane-bound enzyme specialized in hydrolyzing nucleotide triphosphates (Plesner, 1995). This enzyme is a glycoprotein represented by a dimer or trimer of homoligomeric units of 66 kDa (Stout and Kirley, 1996), oriented towards the extracellular space. Ecto-ATPase di¡ers from other known ATPases in location and other important characteristics such as (1) activation by either Ca2þ or Mg2þ , (2) ability to use di¡erent nucleoside 5P-triphosphates as a substrate and (3) insensitivity to inhibitors of ATPases with transport function (Lin, 1985; Plesner, 1995; Zimmermann, 1996). The ecto-ATPase and related ectonucleotidase activity was found in rat cochlear tissues using biochemical (Vlajkovic et al., 1996, 1998) and molecular (Vlajkovic et al., 1999) approaches, which, however, do not allow the localization of the enzyme and thus the tissues and cells types involved in ATP degradation. Here, for the ¢rst time, we describe the cellular localization of ectoATPase activity in vestibular tissues. Experiments were performed in frog semicircular canals using a one-step cerium reaction, an ultracytochemical method applied in a variety of other tissues (Robinson and Karnovsky, 1983; Salama et al., 1987; Kortje et al., 1990; Gleeson et al., 1992; Van Noorden and Frederiks, 1993; Kittel et al., 1996; Kittel, 1999). In this method, the phosphate group released by the action of ATPase (in the presence of extracellular cerium ions) precipitates as CePO3 forming granulated reaction deposits which are su⁄ciently electron opaque to be detected with conventional electron microscopy. The present paper provides evidence that in frog semicircular canals, ecto-ATPase activity is widespread at the level of plasma membranes of both sensory and non-sensory cells of ampullary epithelium facing the endolymphatic and the perilymphatic compartments.
2. Materials and methods The experiments were performed on vertical posterior semicircular canals of 10 adult frogs (Rana esculenta) purchased from local suppliers. Frogs were anaesthetized by immersion in 0.1% tricaine methanesulfonate (MS222, Sandoz, Basel, Switzerland) solution. After decapitation, the otic capsule was opened to expose the posterior semicircular canal, and the ampulla was removed according to a procedure described elsewhere (Prigioni et al., 1983). Specimens were ¢xed in 2% paraformaldehyde and 0.5% glutaraldehyde in 0.05 M cacodylate bu¡er (pH 7.2) at 4‡C for 1 h and washed overnight in the same bu¡er. We used low concentrations of aldehydes, which seem to have a negligible in£uence on the enzymatic activity of ecto-ATPase (Thirion et al., 1996). Ecto-ATPase activity was demonstrated by the modi¢ed method of Robinson and Karnovsky (Robinson and Karnovsky, 1983). Brie£y, specimens were incubated in a medium containing 80 mM Tris^maleate bu¡er (pH 7.2), 3 mM ATP as substrate (Sigma, St. Louis, MO, USA), 5 mM CaCl2 , 2 mM CeCl3 , 5 mM levamisole (inhibitor of alkaline phosphatase, Sigma) and 1 mM ouabain (inhibitor of sodium/potassium adenosine-triphosphatase (Naþ /Kþ -ATPase), Sigma) for 1 h at 30‡C. Incubation was followed by rinses in Tris^maleate bu¡er, and samples were post¢xed in 1% OsO4 dissolved in cacodylate bu¡er for 1 h. After rinses in cacodylate bu¡er, specimens were dehydrated in graded ethanol series, processed through propylene oxide and embedded in epoxy resin. Ultrathin sections were cut with a Reichert OM12 ultramicrotome and stained with uranyl acetate. The sections were examined with a Zeiss 109 transmission electron microscope. Semithin sections of control preparations were used for light microscopy. To con¢rm ecto-ATPase activity some experiments were performed by substituting CaCl2 with 5 mM MgCl2 . Another set of experiments was performed by using 3 mM guanosine-5P-triphosphate (GTP) instead of ATP. Inhibitors of F-type (quercetin, 0.1 mM, Sigma) and V-type (N-ethylmaleimide, 1 mM, Sigma) ATPases were tested. Some experiments were also performed in the presence of the inhibitor of the Ca2þ pump ATPase, vanadium oxide (10 mM, Sigma). Vanadium oxide and GTP were used in the presence of 1 mM Ca2þ and 1 mM Mg2þ to avoid spontaneous precipitation. Another set of experiments was performed by using suramin, a non-competitive antagonist
C Fig. 1. Light micrographs of transverse sections of peripheral, intermediate, and central regions of the crista ampullaris. In the peripheral region (A), the sensory epithelium (E) is in contact with a layer of undi¡erentiated cells (U) which constitute the ampullary wall. In the intermediate (B) and central (C) regions, the layer of transitional cells (T) is clearly evident below the sensory epithelium (E). This layer is in contact with that of dark cells (D) which extends toward the canal duct. Bar = 20 Wm.
HEARES 3977 22-1-03
L. Gioglio et al. / Hearing Research 176 (2003) 1^10
HEARES 3977 22-1-03
3
4
L. Gioglio et al. / Hearing Research 176 (2003) 1^10
Fig. 2. Electron micrographs of ecto-ATPase histochemistry in hair cells and supporting cells of the central, peripheral and intermediate regions of the crista ampullaris. In the central region (A), an intense reaction is clearly evident on stereocilia, microvilli and apical membrane of cylindrical hair cells (C). The apical membrane of supporting cells (S) also shows a marked reactivity. Note the lack of staining at the level of basolateral membranes of sensory and supporting cells. In the peripheral region (B), the ecto-ATPase activity, although less abundant, is present on stereocilia, microvilli and apical membranes of club-shaped hair cells (L). These cells also show reaction products at the level of basolateral membranes. Similarly supporting cells (S) of this region show enzymatic activity on apical and basolateral membranes. In the intermediate region (C), the enzymatic activity is present only on the apical pole of pear-shaped hair cells (P) and supporting cells (S). Bar = 1 Wm.
of ecto-ATPases (Plesner, 1995) (100 WM, Sigma). Finally, exclusion of ATP or Ca2þ and Mg2þ from the reaction medium was used as an additional negative control.
The care and the use of animals in this study were in accordance with the guidelines of Italian ‘Ministero della Sanita'’ (authorization: 9194/A; 1-08-94 and 68/97-A; 23-10-97).
HEARES 3977 22-1-03
L. Gioglio et al. / Hearing Research 176 (2003) 1^10
5
3. Results The presence of ecto-ATPase activity was studied in sensory and non-sensory areas of frog ampullary epithelium. We investigated the localization and the distribution of the enzyme activity at the level of the peripheral (Fig. 1A), intermediate (Fig. 1B) and central (Fig. 1C) epithelial regions. The sensory epithelium consists of hair cells surrounded by supporting cells and basal cells located close to the basement membrane (Gioglio et al., 1995). Three types of hair cells characterize the three crista regions : club-shaped cells in the peripheral region, cylindrical cells in the central region and pearshaped cells in the intermediate region. Supporting cells have a £ask-like shape in all regions and, as for hair cells, vary in length in relation to their location in the crista. The basal cells are small and ovoid in shape with a large nucleus and show no evident morphological di¡erences in all three regions of the crista. The nonsensory epithelium is formed by a monocellular layer of three types of cells : the transitional, dark and undi¡erentiated cells (Oudar et al., 1988). The transitional cells constitute the lateral wall of the crista and are usually cuboid in shape. These cells are in contact with the layer of dark cells which extends toward the canal duct. The dark cells are usually cylindrical in shape, show scanty cytoplasm and are characterized by numerous basolateral membrane infoldings which extend up to half of the cell height. The undi¡erentiated cells are very £attened and elongated, and form a layer that constitutes the remainder of the ampulla. Morphological evidence for ecto-ATPase activity in the ampullary epithelium was represented by grained electron opaque precipitates. We found that the staining was rather widespread both in sensory and nonsensory epithelial areas, and that the reaction deposits were invariably present at the outer surface of plasma membranes of di¡erent cell types. No cytoplasmatic organelles of any of the cells showed evidence of cerium precipitates. When ecto-ATPase activity labelling was examined in detail, some di¡erences were seen among cells constituting the sensory epithelium in relation to their location within the crista. An intense labelling was found on the plasma membranes of cylindrical cells of the central region of the crista particularly along and around the stereocilia, on microvilli, and on the apical membrane lining the cuticular plate (Fig. 2A). The club-shaped hair cells of the peripheral regions (Fig. 2B) and pearshaped hair cells of the intermediate regions (Fig. 2C) showed less abundant ecto-ATPase activity at the level of their apical poles. In these cells, cerium precipitates on stereocilia and microvilli, and on the outer surface of apical membrane were usually lighter. Greater staining di¡erences were seen at the level of the basolateral
Fig. 3. Electron micrograph of ecto-ATPase labelling in crista basal cells. These cells are characterized by poor reaction products; cerium precipitates are indeed clearly evident only under the tight junction-like (arrows) between cells. Bar = 0.2 Wm.
membranes of hair cells. Ecto-ATPase activity was present in club-shaped hair cells of the peripheral regions (Fig. 2B), whereas the basolateral membranes of cylindrical (Fig. 2A) and pear (Fig. 2C) hair cells usually showed no clear evidence of cerium precipitates. The supporting cells of the crista exhibited a pattern of ecto-ATPase distribution which largely matched that seen in their associated hair cells in the three regions. In fact, abundant reaction products were observed on the outer surface of the apical membrane of cells located in the central region of the crista (Fig. 2A), while cerium precipitates were less abundant on apical membranes of cells of the peripheral (Fig. 2B) and intermediate (Fig. 2C) regions of the crista. Basolateral membranes of supporting cells showed clear reaction products only in the peripheral regions (Fig. 2A). The ovoid basal cells showed poor ecto-ATPase activity with the exception of the lateral membranes located under the tight junction-like connecting the cells to each other (Fig. 3). No evident di¡erences in the reaction products were observed in relation to the localization of basal cells along the crista. Some di¡erences in ecto-ATPase activity were also seen among cells constituting the non-sensory epithelium of the ampulla. Transitional cells showed abundant
HEARES 3977 22-1-03
6
L. Gioglio et al. / Hearing Research 176 (2003) 1^10
Fig. 4. Electron micrographs of ecto-ATPase labelling in transitional and undi¡erentiated cells of the non-sensory epithelium of the ampulla. In transitional cells (A), an intense reaction is present on the apical membrane and on the numerous microvilli, while less intense enzymatic activity is present on the basolateral membrane. In undi¡erentiated cells (B), clear ecto-ATPase activity is present on apical and basolateral membranes. Bar = 1 Wm.
precipitates at the level of the apical membrane and on the short microvilli, while the basolateral membranes usually exhibited scattered cerium precipitates (Fig. 4A). The £attened undi¡erentiated cells exhibited clear reaction products both at the level of apical and basolateral membranes (Fig. 4B). The dark cells also displayed an evident ecto-ATPase activity. Abundant cerium precipitates were seen on the apical membranes having few microvilli and on the basolateral membranes characterized by large infoldings (Fig. 5). To con¢rm that the reaction products we observed in di¡erent cell types of the crista ampullaris are due to ecto-ATPase activity, some experiments were repeated using Mg2þ (instead of Ca2þ ) as enzyme activator. We found an intense staining both in sensory and non-sensory cells of the crista (Fig. 6A). Similar ecto-ATPase labelling was observed when GTP substituted for ATP, or when inhibitors of Ca2þ pumps (vanadium oxide) and of F-type (quercetin) and V-type (N-ethylmaleimide) ATPases were tested. The only substance able to prevent the staining of plasma membranes in di¡erent cell types was suramin (Fig. 6B), a non-competitive antagonist of ecto-ATPases (Plesner, 1995). Similarly, no reaction products were seen when the substrate ATP or divalent cations Ca2þ or Mg2þ were omitted
from the medium. All these observations are indicative of the presence of ecto-ATPase activity in sensory and non-sensory cells of ampullary epithelium.
4. Discussion In the present study, we provide the ¢rst evidence for the localization of ecto-ATPase activity in inner ear vestibular tissues. The speci¢city of the cytochemical method of Robinson and Karnovsky (1983) to mark ecto-ATPases has been emphasized by recent works (Kortje et al., 1990; Gleeson et al., 1992; Kittel et al., 1996; Kittel, 1999). In support to the speci¢city of this method we found that cerium precipitates were localized at the outer surface of the cell plasma membranes, a localization that is not expected for Ca2þ -transporting ATPases (Kortje et al., 1990; Cardy and Firth, 1993). We also found similar staining of the cell plasma membranes in the presence of mM concentrations of Ca2þ or Mg2þ or both ions. The equal activation by these cations of ATP hydrolysis represents a common feature of ecto-ATPases in a variety of tissues (Plesner, 1995; Zimmermann, 1996). It is well known that Ca2þ pumps strictly require ATP as a substrate and, at variance with
HEARES 3977 22-1-03
L. Gioglio et al. / Hearing Research 176 (2003) 1^10
7
Fig. 5. Electron micrograph of ecto-ATPase activity in a dark cell. An intense reactivity characterizes the apical and basolateral membranes of the cell (A) which are shown at high magni¢cation in B and C respectively. Bar = 1 Wm.
Fig. 6. Electron micrographs of hair cells and supporting cells of the crista central region in di¡erent incubation conditions. An intense ectoATPase activity (A) is present in hair cells (C) and supporting cells (S) using a medium containing Mg2þ as enzyme activator. No detectable reaction products (B) were seen in hair cells (C) and supporting cells (S) in the presence of the ecto-ATPase antagonist, suramin. Bar = 1 Wm.
HEARES 3977 22-1-03
8
L. Gioglio et al. / Hearing Research 176 (2003) 1^10
ecto-ATPases, are unable to cleave other triphosphates (Cardy and Firth, 1993). The fact that we observed comparable staining when GTP was the substrate (Plesner, 1995) further supports the speci¢city of our reaction and bolsters our identi¢cation of ecto-ATPase activity sites in semicircular canals. This view is further supported by the observation that the staining was prevented in the presence of non-competitive ecto-ATPase antagonists such as suramin (Plesner, 1995; Zimmermann, 1996). Further, putative ecto-ATPase labelling was not a¡ected by inhibitors of Ca2þ -transporting ATPase (vanadate), Naþ /Kþ -ATPase (ouabain) and other intracellular ATPases. The presence of ectonucleotidase activity has been demonstrated in the cochlea by using biochemical methods. Evidence was provided that E-NTPDase activity coupled with that of ecto-5P-nucleotidase provides the terminating mechanism of purinergic signalling, leading to adenosine production in both endolymphatic and perilymphatic compartments (Vlajkovic et al., 1996, 1998). One of the main functions of ecto-ATPases is to terminate the action of ATP on purinergic receptors through ATP hydrolysis (Plesner, 1995; Zimmermann, 1996). The presence of purinergic receptors has been demonstrated in cochlear tissues facing both the endoand the perilymphatic compartments by using both imaging and immunocytochemical methods (Mockett et al., 1994; Housley and Ryan, 1997). However, much less is known about the vestibular system in which putative purinergic receptors are thought to be localized at the level of dark cell apical membranes (Liu et al., 1995; Wangemann, 1995) and hair cell basolateral membranes (Rennie and Ashmore, 1993; Aubert et al., 1995). Our data demonstrating a widespread expression of ecto-ATPases at the level of both apical and basolateral membranes of sensory and non-sensory cells of the crista ampullaris suggest a possible parallel distribution of purinergic receptors. Therefore, it is likely that ATP exerts a paracrine action in the vestibular system, as in the cochlea (Housley, 1997), providing a widespread in£uence over the activity of sensory and non-sensory cells both at the level of the endolymphatic and perilymphatic compartments. Concerning hair cells, it is particularly interesting that the present study provides the ¢rst evidence that ecto-ATPase activity is present at the stereocilia level, thus suggesting a possible colocalization of enzymatic sites with purinergic receptors and mechanotransduction channels (Mockett et al., 1994; Housley, 1997; Housley and Ryan, 1997). The strategic expression of ecto-ATPase sites at the stereocilia level could provide a rapid mechanism of ATP removal which would enable rapid restoration of sensitivity of mechanotransduction after a transient ATP release into the endolymph. However, on the basis of our data this mechanism would be
more e⁄cient at the level of sensory cells of the central region of the crista. It may be that these cells have more purinergic receptor sites than hair cells of other crista regions. We also found the presence of ecto-ATPase activity at the level of hair cell basolateral membranes. This agrees with studies performed in isolated frog semicircular canal preparations suggesting the presence of P2Y-subtype receptors involved in the modulation of the a¡erent transmitter release (Aubert et al., 1994, 1995). Therefore, the ecto-ATPase activity at the level of basolateral membranes in canal hair cells may be related to the function of perilymphatic ATP (Mockett et al., 1994). However, our data indicate that an evident ecto-ATPase activity appears peculiar of hair cells of the peripheral regions. This observation further supports the concept that hair cells located in di¡erent regions of the crista have a di¡erent role in processing the natural stimulus (Masetto et al., 1994). Concerning supporting cells, our data indicate an abundant ecto-ATPase activity at the level of their apical membranes adjacent to the apical hair cell labelling. This contrasts with the situation in the cochlea where high levels of PX2 receptor immunoreactivity have been demonstrated throughout the Deiters and Hensen cells (Jarlebark et al., 2000). It appears that the apical ATPgated channels in Hensen cells are probably involved in the regulation of ionic and £uid homeostasis of the scala media endolymph via activation of the Ca2þ -activated Cl channels (Sugasawa et al., 1996). It is possible that this mechanism also operates in the vestibular system and ecto-ATPase provides the mechanism for termination of ATP action. Concerning basolateral membranes of supporting cells, our data indicate that ectoATPase activity is restricted to the peripheral regions of the crista, thus suggesting a possible di¡erential distribution of P2 receptor sites. At present, the expression of P2 receptors at the level of basolateral membranes has been demonstrated only in Deiters’ cells of the cochlea in which the activation of P2X and P2Y receptors are responsible for ATP-induced increase in intracellular calcium concentration (Dulon et al., 1993). The presence of ecto-ATPase in apical and basolateral membranes of dark cells could have relevant functional implications. These cells are responsible for the secretion of the endolymph and for its anomalous ionic composition characterized by a high Kþ concentration (Wangemann, 1995). In these cells, purinergic receptors of P2U and P2Y types mediate the transepithelial Kþ transport (Liu et al., 1995). In addition, dark cells may represent, in analogy with marginal cells of the stria vascularis of the cochlea (White et al., 1995; Housley, 1997), a potential source of ATP present in the endolymph. Therefore, the presence of ecto-ATPase on the apical membranes of dark cells in conjunction with their possible production of ATP may provide an auto-
HEARES 3977 22-1-03
L. Gioglio et al. / Hearing Research 176 (2003) 1^10
crine homeostatic control of the endolymphatic ATP content (Liu et al., 1995; Wangemann, 1995). The presence of ecto-ATPase activity on transitional and undi¡erentiated cells is not clear, since at present there is no evidence of the expression of purinergic receptor sites by these cells. It is known that these cells appear not to be involved in ionic transport and their main function is to maintain the endolymph composition forming a barrier between endolymph and perilymph (Oudar et al., 1988). It is possible that ecto-ATPases in these cells contribute together with other cells of the crista in regulating the ATP content of the endolymphatic and perilymphatic £uids both in normal and in pathological conditions. It is known that ATP levels can double in the cochlear £uids during hypoxia (Munoz et al., 1995). Therefore, the widespread distribution of ecto-ATPases we found in the semicircular canals may also have, in addition to speci¢c roles, an important protective function against ATP-induced damage. In conclusion, the present study provides evidence for the localization of ecto-ATPases in vestibular tissues lining both the perilymphatic and endolymphatic compartments. The widespread distribution of ecto-ATPases on sensory and non-sensory epithelial cells lining the perilymphatic and endolymphatic surfaces suggests that ATP could act both as a humoral factor involved in the regulation of hair cell transduction and electrochemical homeostasis as well as a neuromodulator of the a¡erent transmission, similar to its role in the cochlea (Housley, 1997).
Acknowledgements This work was supported by the Ministero della Ricerca Scienti¢ca e Tecnologica (MURST-Co¢n-2000), Rome, Italy. The authors thank Dr. G. Housley and Dr. S.M. Vlajkovic for their critical reading of the manuscript.
References Aubert, A., Norris, C.H., Guth, P.S., 1994. In£uence of ATP and ATP agonists on the physiology of the isolated semicircular canal of the frog. Neuroscience 62, 963^974. Aubert, A., Norris, C.H., Guth, P.S., 1995. Indirect evidence for the presence and physiological role of endogenous extracellular ATP in the semicircular canal. Neuroscience 64, 1153^1160. Bobbin, R.P., Thompson, M.H., 1978. E¡ects of putative transmitters on a¡erent cochlear transmission. Ann. Otol. 87, 185^190. Butlen, D., Bernard, C., Ferrary, E., 1998. Pharmacological characterization of ATP receptors in ampulla from frog semicircular canal. Am. J. Physiol. 275, 253^261. Cardy, J.D., Firth, J.A., 1993. Adenosine triphosphate-lead histochemical reactions in ependymal epithelia of murine brains do
9
not represent calcium transport adenosine triphosphate. J. Histochem. Cytochem. 44, 103^111. Dulon, D., Moataz, R., Mollard, P., 1993. Characterization of Ca2þ signals generated by extracellular nucleotides in supporting cells of the organ of Corti. Cell Calcium 14, 245^254. Eybalin, M., 1993. Neurotransmitters and neuromodulators of the mammalian cochlea. Physiol. Rev. 73, 309^373. Gioglio, L., Congiu, T., Quacci, D., Prigioni, I., 1995. Morphological features of di¡erent regions in frog crista ampullaris (Rana esculenta). Arch. Histol. Cytol. 58, 1^16. Gleeson, R.A., Trapido-Rosenthal, H.G., McDowell, L.M., Aldich, H.C., Carr, W.E.S., 1992. Ecto-ATPase/phosphatase activity in the olfactory sensilla of the spiny lobster, Panulirus-argus. Localization and characterization. Cell Tissue Res. 269, 439^445. Housley, G.D., 1997. Extracellular nucleotide signaling in the inner ear. Mol. Neurobiol. 16, 21^48. Housley, G.D., Ryan, A.F., 1997. Cholinergic and purinergic neurohumoral signalling in the inner ear: A molecular physiological analysis. Audiol. Neuro-Otol. 2, 92^110. Housley, G.D., Luo, L., Ryan, A.F., 1998. Localization of mRNA encoding the P2X2 receptor subunit of the adenosine 5P-triphosphate-gated ion channel in the adult and developing rat inner ear by in situ hybridization. J. Comp. Neurol. 393, 403^ 414. Jarlebark, L.E., Housley, G.D., Thorne, P.R., 2000. Immunohistochemical localization of adenosine 5P-triphosphate-gated ion channel P2X(2) receptor subunits in adult and developing rat cochlea. J. Comp. Neurol. 421, 289^301. Kittel, A., 1999. Lipopolysaccharide treatment modi¢es pH- and cation-dependent ecto-ATPase activity of endothelial cells. J. Histochem. Cytochem. 47, 393^399. Kittel, A., Siklo's, L., Thuro'czy, G., 1996. Qualitative enzyme histochemistry and microanalysis reveals changes in ultrastructural distribution of calcium and calcium-activated ATPases after microwave irradiation of the medial habenula. Acta Neuropathol. 92, 362^368. Kortje, K.H., Freihofer, D., Rahmann, H., 1990. Cytochemical localization of high-a⁄nity Ca2þ -ATPase activity in synaptic terminals. J. Histochem. Cytochem. 38, 895^900. Kujawa, S.G., Erostegui, C., Fallon, M., Crist, J., Bobbin, R.P., 1994. E¡ects of Adenosine 5P-triphosphate and related agents on cochlear function. Hear. Res. 76, 87^100. Lin, S.H., 1985. The rat liver plasma membrane high a⁄nity (Ca2þ Mg2þ )-ATPase is not a calcium pump. J. Biol. Chem. 264, 14403^ 14407. Liu, J., Kozakura, K., Marcus, D.C., 1995. Evidence for purinergic receptors in vestibular dark cell and strial marginal cell epithelia of the gerbil. Audit. Neurosci. 1, 331^340. Masetto, S., Russo, G., Prigioni, I., 1994. Di¡erential expression of potassium currents by hair cells in thin slices of frog crista ampullaris. J. Neurophysiol. 72, 443^455. Mockett, B.G., Housley, G.D., Thorne, P.R., 1994. Fluorescence imaging of extracellular purinergic receptor sites and putative ecto-ATPase sites on isolated cochlear hair cells. J. Neurosci. 14, 6992^7007. Munoz, D.J.B., Thorne, P.R., Housley, G.D., Billet, T.E., 1995. Adenosine 5P-triphosphate (ATP) concentrations in the endolymph and perilymph of the guinea-pig cochlea. Hear. Res. 90, 119^125. Oudar, O., Ferrary, E., Feldmann, G., 1988. Ultrastructural study of semicircular canal cells of the frog Rana esculenta. Anat. Rec. 220, 328^334. Plesner, L., 1995. Ecto-ATPase: identities and functions. Int. Rev. Cytol. 158, 141^214. Prigioni, I., Valli, P., Casella, C., 1983. Peripheral organization of the
HEARES 3977 22-1-03
10
L. Gioglio et al. / Hearing Research 176 (2003) 1^10
vestibular e¡erent system in the frog: an electrophysiological study. Brain Res. 269, 83^90. Rennie, K.J., Ashmore, J.F., 1993. E¡ects of extracellular ATP on hair cell isolated from the guinea-pig semicircular canals. Neurosci. Lett. 160, 185^189. Robinson, J.M., Karnovsky, M.J., 1983. Ultrastructural localization of several phosphatases with cerium. J. Histochem. Cytochem. 31, 1197^1208. Salama, A.H., Zaki, A.E.E., Eisenmann, D.R., 1987. Cytochemical localization of Ca2þ -Mg2þ adenosine triphosphatase in rat incisor ameloblasts during enamel secretion and maturation. J. Histochem. Cytochem. 35, 471^482. Stout, J.G., Kirley, T.L., 1996. Control of cell membrane ecto-ATPase by olimerization state. Intermolecular cross-linking modulates ATPase activity. Biochemistry 35, 8289^8298. Sugasawa, M., Erostegui, C., Blanchet, C., Dulon, D., 1996. ATP activates a cation conductance and Ca2þ -dependent Cl3 conductance in Hansen cells of guinea pig cochlea. Am. J. Physiol. 271, C1817^1827. Thirion, S., Troadec, J., Nicaise, G., 1996. Cytochemical localization of ecto-ATPases in rat neurohypophysis. J. Histochem. Cytochem. 44, 103^111. Thorne, P.R., Housley, G.D., 1996. Purinergic signalling in sensory system. Semin. Neurosci. 8, 233^246. Van Noorden, C.J.F., Frederiks, W.M., 1993. Cerium methods for
light and electron microscopical histochemistry. J. Microsc. 171, 3^16. Vlajkovic, S.M., Thorne, P.R., Munoz, D.J.B., Housley, G.D., 1996. Ectonucleotidase activity in the perilymphatic compartment of the guinea pig cochlea. Hear. Res. 99, 31^37. Vlajkovic, S.M., Thorne, P.R., Housley, G.D., Munoz, D.J.B., Kendric, I.S., 1998. Ecto-nucleotidases terminate purinergic signalling in the cochlear endolymphatic compartment. NeuroReport 9, 1559^1565. Vlajkovic, S.M., Housley, G.D., Greenwood, D., Thorne, P.R., 1999. Evidence for alternative splicing of ecto-ATPase associated with termination of purinergic transmission. Mol. Brain Res. 73, 85^92. Wangemann, P., 1995. Comparison of ion transport mechanisms between vestibular dark cells and strial marginal cells. Hear. Res. 90, 149^157. White, P.N., Thorne, P.R., Housley, G.D., Mockett, B., Billett, T.E., Burnstock, G., 1995. Quinacrine staining of marginal cells in the stria vascularis of the guinea-pig cochlea: a possible source of extracellular ATP? Hear. Res. 90, 97^105. Zimmermann, H., 1996. Biochemistry, localization and functional roles of ecto-nucleotidases in the nervous system. Prog. Neurobiol. 49, 589^618. Zimmermann, H., 1999. Two novel families of ectonucleotidases: molecular structures, catalytic properties and search for function. Trends Pharmacol. Sci. 20, 231^236.
HEARES 3977 22-1-03
