THE ANATOMICAL RECORD 292:1771–1779 (2009)

Immunolocalization of G-ProteinAlpha Subunits in the Olfactory System

of the Cartilaginous FishScyliorhinus Canicula

SARA FERRANDO,1* CHIARA GAMBARDELLA,1,2 SILVIA RAVERA,3

SERGIO BOTTERO,3 TIZIANA FERRANDO,1 LORENZO GALLUS,1

VALENTINA MANNO,1 AMIR P. SALATI,4 PAOLA RAMOINO,5

AND GRAZIA TAGLIAFIERRO1

1LIBiOM, Department of Biology, University of Genoa, Genoa, Italy2Department of Animal Biology and Marine Ecology, University of Messina, Messina, Italy

3Department of Biology, University of Genoa, Genoa, Italy4Department of Fisheries, Faculty of marine natural resources, Khoramshahr University

of Marine Science and Technology, Khoramshahr, Iran5DipTeRis, University of Genoa, Genoa, Italy

ABSTRACTIn the olfactory and vomeronasal systems of vertebrates, the morphol-

ogy of the receptor neurons, the receptor gene family they express, theG-protein coupled with the receptor (in particular the G-protein alpha sub-unit), and their projection to the olfactory bulb are correlated. Much infor-mation about this complicated system have been collected in differentgroups, but nothing is known about Chondrichthyes. In this work, the pres-ence and distribution of immunoreactivity for different types of G-proteinalpha subunit (Gao, Gaq and Gas/olf) were investigated in the olfactorymucosa and olfactory bulb of the shark Scyliorhinus canicula. Only Gao-like immunoreactivity was detected in the olfactory mucosa and bulb, bothin tissues and homogenates. Its distribution was partially similar to thatfound in other vertebrates: it was localized in the microvillous receptor neu-rons, in numerous axon bundles of the fila olfactoria, in the stratum nervo-sum and in the most of glomeruli in the stratum glomerulosum. Noimmunoreactivity was instead observed in the crypt neurons, the secondtype of olfactory neurons present in cartilaginous fish. The projections ofcrypt neurons to olfactory bulb probably correspond to the few ventrally-located glomeruli which were negative to the antiserum against Gao. Thesedata suggest, in S. canicula, different olfactory neuron types send projec-tions to the olfactory bulb with a segregated distribution, as observed inother vertebrates. Anat Rec, 292:1771–1779, 2009. VVC 2009 Wiley-Liss, Inc.

Keywords: olfactory mucosa; olfactory bulb; Gao; crypt neurons

INTRODUCTION

Two chemosensory systems are present in the nose ofvertebrates: the olfactory and the vomeronasal systems.Their sensory epithelia are differentially located in tetra-pods, while, in fish, the two systems share the same local-ization and thence epithelium (Hansen et al., 2003, 2004).In mammals, the receptor neurons (RNs) in the olfactoryepithelium (OE) express receptors of the olfactory recep-tor (OR) gene family, whereas the RNs in the vomeronasal

Additional Supporting Information may be found in theonline version of this article.

*Correspondence to: Sara Ferrando, LIBiOM, Department ofBiology, University of Genoa, Viale Benedetto XV, 5, I-16132,Genoa, Italy. Fax: þ39 010 3538047. E-mail: sara.ferrando@unige.it

Received 18 January 2009; Accepted 2 July 2009

DOI 10.1002/ar.21003Published online 18 September 2009 in Wiley InterScience (www.interscience.wiley.com).

VVC 2009 WILEY-LISS, INC.

epithelium (VE) express receptors of the vomeronasalreceptors 1 and 2 (V1R and V2R) gene families (Buck andAxel, 1991; Dulac and Axel, 1995; Dryer, 2000; Halpernand Martı́nez-Marcos, 2003). In fish, all the receptor genefamilies are expressed by the RNs located into the same,unique OE (Hansen et al., 2003, 2004).

Each receptor molecule is oligospecific, i.e., specific fora few odorants, and in all vertebrates only one of them isexpressed by each olfactory or vomeronasal RN (Korsch-ing, 2002; Nagao et al., 2002). Moreover, a correlationbetween the morphology of the RN and the family of theexpressed receptor molecule has been demonstrated inmany species (Sato and Suzuki, 2001; Lipschitz andMichel, 2002; Halpern and Martı́nez-Marcos, 2003;Hansen et al., 2003, 2004; Einsthen, 2004; Hansen andZielinski, 2005; Vielma et al., 2008). The receptor mole-cules belong to the gene superfamily GPCRs (G-proteincoupled receptors). Different types of G-protein havebeen shown to have a functional co-expression with dif-ferent receptor molecules; in particular the distributionof different G-protein alpha subunits have been describedin the olfactory/vomeronasal system of vertebrates (Taki-gami et al., 2000; Belanger et al., 2003; Halpern andMartı́nez-Marcos, 2003; Hansen et al., 2003, 2004).

Axons from the RNs reach the glomeruli of the olfac-tory bulb (OB), where they synapse with the mitral cells(Korsching, 2002; Nagao et al., 2002). The RNs express-ing the same receptor molecule, project their axons tothe same glomerulus, synapsing with few mitral cells(Takigami et al., 2000; Firestein, 2001). Thus it has beensuggested that, in the olfactory/vomeronasal system of avertebrate, the localization of a RN, its morphology, thereceptor gene family it expresses, the G-protein subtypecoupled with this receptor, and the zone of projection tothe OB are somehow correlated (Halpern and Martı́nez-Marcos, 2003; Hansen et al., 2004).

Studies about the organization of the olfactory systemand the modality of signal transduction have been car-ried out in different species of bony fishes and also injawless fishes (Freitag et al., 1999; Speca et al., 1999;Belanger et al., 2003; Hansen et al., 2003, 2004; Korsch-ing, 2009), but these topics have yet to be explored inChondrichthyes. The olfactory system of Chondrichthyeshas been studied mainly from a morphological point ofview (Ebbesson and Heimer, 1970; Theisen et al., 1986;Franceschini and Ciani, 1993; Takami et al., 1994; Fer-rando et al., 2006a,b, 2007a); its development has beenonly partially described (Fishelson and Baranes, 1997;Ferrando et al., 2007b; Ferrando, 2008).

Chondrichthyes are located, in the phylogenetic tree,before the split between Sarcopterygians (from whichtetrapods arose) and Actinopterygians (such as the morestudied Osteichthyes). For this reason, the study of theolfactory system in this systematic group, can give infor-mation about the ancestral features of this system invertebrates. Moreover, olfaction is generally well devel-oped in sharks and it has a great importance in feeding,in social-sexual behavior, and in predator avoidance(Hueter et al., 2004), thus a better knowledge of thistopic can be of great interest to a greater understandingof shark biology and the improvement of the protectionof numerous endangered species.

To approach the complex olfactory functional organiza-tion and the transduction pathways in Chondrichthyes,in the present work, we began to study the presence and

distribution of different G-protein alpha subunits in theolfactory rosette and bulb of the shark Scyliorhinus cani-cula. The olfactory rosette of the elasmobranch is consti-tuted by numerous lamellae lined up by the OE(Fishelson and Baranes, 1997; Kajura et al., 2005). As inother vertebrates, the axons of the RNs project from theOE to glomeruli in the OB, synapsing with the mitralcells (Franceschini and Ciani, 1993). For this prelimi-nary investigation we have used commercial antiseraagainst the subunit Gas/olf (from rat), Gaq (from mouse)and Gao (from rat), previously tested with good resultsin bony fish (Hansen et al., 2004).

It is known from the literature that the receptor mole-cule and the G-protein specific for this receptor aredetectable not only in the dendritic process of the RNs,as their function suggest, but also along the axons, com-prehending the glomeruli (Hansen et al., 2004). Thusthe detection of immunoreactivity (ir) for differentG-protein alpha subunits in the glomeruli can reflect thedifferent primary projections from the OE, helping toidentify different types of RNs and their primary projec-tion into the OB, highlighting an eventual topographicalorganization in the OB of Chondrichthyes as alreadyobserved in other vertebrates (Zippel et al., 1997;Laberge and Hara, 2001).

MATERIALS AND METHODS

Six adult specimens, four males and two females of S.canicula, with a length ranged between 48 and 53 cm,were collected in the Irish Sea (NE Atlantic) by bottomtrawlers. They were brought on board, anesthetized with0.01% ethyl 3-aminobenzoate methanesulfonate salt(Sigma-Aldrich, St Louis, MO; dilution 1:1,000 in seawater), sacrificed and then dissected to collect the olfac-tory organs.

Histological and Immunohistochemical Analysis

Olfactory organs (olfactory rosettes and bulbs) werefixed in 4% paraformaldehyde in 0.1 M phosphatebuffered solution (PBS, pH 7.4) at 4�C, Paraplast (Bio-Optica, Milano, Italy) embedded and 5-lm thick sec-tioned. Histological observations were performed byhematoxylin–eosin (Bio-Optica, Milano, Italy). Immuno-histochemical reactions were carried out to highlightthree different G-protein alpha subunits using the follow-ing polyclonal antisera from Santa Cruz Biotechnology(CA): anti-Gao (Cat. sc-387), anti-Gaq (Cat. sc-393) andanti-Gas/olf (Cat. Sc-383) raised against mammalian pro-teins. As secondary antisera, Alexa 488 conjugated anti-rabbit (1:800 in PBS, Molecular Probes, Invitrogen,Carlsbad, CA) or the EnVision System–HRP (DAB), anti-rabbit (Dako Cytomation, DK) were used. As only theantisera anti-Gao and anti-Gas/olf gave a positive result,specificity controls were performed by neutralising thesetwo primary antisera with their respective antigens: Gao

blocking peptide (Santa Cruz Biotechnology, Cat. No. sc-387P) and Gas/olf blocking peptide (Santa Cruz Biotech-nology, Cat. No. sc-383P). The highest working dilutionswere 1:100 for the anti-Gas/olf and 1:400 for the anti-Gao.The storage dilution for the antisera was 200 lg/mL andfor blocking peptides was 200 lg/mL. For anti-Gas/olf, 94lL of PBS were mixed with 1 lL of antiserum and with 5lL of specific blocking peptide, obtaining finally 100 lL

1772 FERRANDO ET AL.

of solution with 0.2 lg of antiserum and 1 lg of blockingpeptide (five-fold by weight as suggested in the datasheet of the product). For anti-Gao, 197 lL of PBS weremixed with 0.5 lL of antiserum and with 2.5 lL of spe-cific blocking peptide, obtaining finally 200 lL of solutionwith 0.1 lg of antiserum and 0.5 lg of blocking peptide(five-fold by weight as suggested in the data sheet of theproduct). The mixtures of the antisera with their block-ing peptides were left to react for 24 hr at 4�C before use.Alternatively, negative controls were performed by omis-sion of the primary antisera. Sections were examined bya BX60 Olympus light and epi-fluorescence microscopeand visualised through an Olympus CCD Color-ViewIICamera (Olympus, Tokyo, Japan) with analySIS software(Soft Imaging System GmbH, Germany). Some sectionswere also visualized by a TCS SP2 Leica confocal laserscanning microscope. Measurements on the micrographswere performed using the ImageJ 1.41 public domainsoftware (Rasband, 2006). Schemes of the organization ofthe OB at different level of sectioning have been per-formed scanning the slides at high resolution to obtain-ing the silhouette of the histological sections. Thedrawing has been made using Photoshop CS and the or-ganization of the tissues has been drawn using micro-graphs and direct microscopy observation as references.

Electrophoresis Separation and WesternBlot Analysis

Olfactory organs were frozen at �80�C for westernblot analysis. All operations to obtain S. canicula masshomogenate were carried out at 4�C. Samples werehomogenated by a Politron system in 1 mL of homoge-nizing medium containing 10 mM Tris HCl pH 7.4 and aprotease inhibitors cocktail (Ravera et al., 2007). Thehomogenate was stored at �80�C and the protein con-centration was determined by Bradford methods (Brad-ford, 1976).

Denaturing electrophoresis was performed using aLaemmli protocol (Laemmli, 1970) with minor modifica-tions. In these experiments 50 lg of homogenate wasloaded in the gel. Electrophoresis was carried out usinga Mini Protean III (BioRad, Hercules) apparatus (60 �80 � 1.5 mm3), in which both faces of the gel sandwichwere immersed in the buffer. Separating gel was a gradi-ent from 10% to 14% (w/v) of polyacrylamide and 0.1%SDS, pH 8.8. Stacking gel contained 4.5% w/v polyacryl-amide and 0.1% SDS. Samples were incubated with 8%SDS w/v in 125 mM Tris-HCl (pH 6.8), and 1.25% v/vDTT, for 15 min, in order to improve its denaturation.Then, samples were boiled for 5 min and a second solu-tion, contained 40% w/v sucrose and 0.008% w/v Bromo-phenol blue, was added. Run was performed at 4�C, at50 mA for each gel, for 120–150 min with running buffer[0.05 M Tris (pH 8.0); 0.4 M glycine; 1.8 mM EDTA (eth-ylenediaminetetraacetic acid), and 0.1% SDS]. ProteinMolecular Weight (MW) markers were purchased fromFermentas (Fermentas).

After the run, the gel was stained with a Blue Silverstain (Candiano et al., 2004) or the proteins were trans-ferred onto nitrocellulose (NC) membranes (HybondECL; GE Healthcare, Sweden) in a mini transblot device(Bio-Rad Laboratories, Hercules) at 400 mA for 2 hr.Sheets were blocked over night at 4�C, in PBS plus 5%BSA. After washing with PBS, NC was incubated and

stirred with the antibodies against Gao, Gaq, and Gas/olf

(Santa Cruz Biotechnology, CA) diluted 1:200 in PBSplus 3% BSA, for 1 hr at 20�C. After extensive washingwith PBS plus 0.15% Tween, binding of Ab was revealedby the ECL detection system (Roche, Germany) using ananti-rabbit IgG, as secondary antibody (Amersham Phar-macia Biothech, Sweden) diluted 1:7,500 in PBS plus 1%BSA. Blots were then autoradiographed onto HyperfilmECL (Amersham Pharmacia Biothech, Sweden) andfilms acquired.

Blast Method

After immunohistochemistry, specificity controls, andwestern blot analysis, only the antiserum against Gao

could be considered to give positive results in S. caniculasamples. As the peptide used to obtain this antiserumwas the 105–124 aa of Gao from Rattus norvegicus(Hansen et al., 2004), we performed a BLASTP search(Basic Local Alignment Search Tool) of the aminoacidicsequence KMVCDVVSRMEDTEPFSAEL against nrdatabase. As the sequences of G-protein alpha subunitswe investigated are unknown in Chondrichthyes, thissearch was made to evaluate the possibility, for the anti-serum, to recognize a G-protein type different from itstarget when used in different vertebrates.

RESULTSHistology and Immunohistochemistry

Olfactory mucosa. The olfactory mucosa is made upby the OE, lining the olfactory lamellae and the connectivetissue, forming the lamina propria, where the axons of theRNs run toward the OB. The structure of the OE of

Fig. 1. Olfactory mucosa of S. canicula—Light microscope–Hema-toxylin–eosin. In the epithelium are well recognizable the microvillousORNs (N; with roundish nuclei in the middle part of the epithelium),the ciliated sustaining cells (SC; with oval nuclei localized in the upperpart of the epithelium), the basal cells (BC; with little nuclei at thebase of the epithelium) and the large, clear ionocytes (IC).

G-PROTEINS IN THE OLFACTORY SYSTEM OF A SHARK 1773

S. canicula has been already described (Ferrando et al.,2006a,b, 2007a). Briefly, the RNs in the OE can be dividedin microvillous receptor neurons (ORNs) and crypt neu-rons (CNs) according to their morphology. ORNs arehighly numerous and show a round clear nucleus, with aprominent nucleolus; they are generally localized in themiddle zone of the OE. CNs are characterized by a crypt atthe top, and a typical bean-shaped nucleus; their cellbodies are localized at the upper third of the OE, they arequite rare and scattered along the olfactory lamellae.Other cell types can be recognized: ciliated sustainingcells, basal cells, ion exchanging cells, and mucous cells(Fig. 1).Gao-like immunoreactivity. An intense Gao -like

immunostaining was observed in all the ORNs, espe-cially at their dendritic apex, and in the numerous axonbundles of the fila olfactoria located in the lamina prop-ria of the olfactory lamellae (Fig. 2a). Preincubation ofthe antiserum with its antigen prevents the immuno-staining (Fig. 2b). The CNs were never immunoreactive(Fig. 2c). A strong ir was also observed in the big nervebundles running towards the OB (Fig. 2d). Tangentialsections of the OE, showed the immunoreactive dendrites

of the ORNs, well distributed among the sustaining cells(Fig. 3a). Observations by confocal microscope (Fig. 3b)allows to localize the ir on the ORN cell membrane andin a narrow region around the nucleus (Fig. 3c).Gas/olf -like immunoreactivity. Some scattered ORNs

showed a very weak immunoreactive cytoplasm forGas/olf. Preincubation of the antiserum with its antigendid not abolish the immunostaining and so the irobserved should be considered aspecific (Fig. 4a,b).Gaq-like immunoreactivity. No Gaq-like ir has been

detected in the olfactory mucosa.The distribution of the immunoreactivities in the ol-

factory mucosa and the results of the negative controlsare shown in Table 1.

Olfactory bulb. The OB of cartilaginous fish isa paired oval structure linked to the forebrain by apeduncle. The OBs of the collected sharks were mea-sured in toto and after sectioning. The measurementsand histological observations allowed to build schematicreconstruction of the shape, size, and histological organi-zation of S. canicula OB (Fig. 5). The bulb organization

Fig. 2. Olfactory mucosa of S. canicula—Light microscope—Immu-nohistochemical detection of Gao (immunoperoxidase counterstainedwith hematoxylin). (a) Only the ORNs show ir, weaker in the neuronalcell body and stronger in the dendritic (apical) portion. In the laminapropria the axons travelling toward the OB are immunoreactive (arrow-

heads), (b) Preincubation of the antiserum with its antigen prevent theimmunostaining, (c) Scattered along the OE is possible finding CNs(arrow), completely devoid of immunoreactivity. (d) As the nerve fibersin the lamina propria (arrowhead), also the big nerve bundles betweenthe olfactory rosette and the OB were immunoreactive.

1774 FERRANDO ET AL.

is characterized by a thick layer of primary olfactoryfibers, making up the stratum nervosum that was themost external layer in the medio-ventral part of the OB,and by numerous roundish glomeruli constituted thestratum glomerulosum. The stratum mitrale described inother vertebrates, was not easily visible because themitral cells were intermingled with the glomeruli, sur-rounding them. The stratum granulosum was the mostinternal one, and in this part it can be found the mostlateral part of the lateral ventricle (Figs. 5, 6a).Gao-like immunoreactivity. About all the axon bun-

dles of the stratum nervosum showed a strong ir (Fig. 6b),that can be abolished both by preincubating the primaryantiserum with its antigen or by omission of the primaryantiserum. In the stratum glomerulosum most the glo-meruli showed an intense ir (Fig. 6c), but few of them,located in the ventralmost part of the anterior and middlebulb, showed a weak brownish color, that was not abol-ished by preincubation. This aspecific ir is probably due tothe fact that the commercial antiserum used can weaklyreact in an aspecific way in the shark tissues (Figs. 5c,6d). No ir was present in the mitral cells, and, in the stra-tum granulosum only a slight ir, not removed by preincu-bation of the antiserum, was observed (Fig. 6c,d). Theomission of the antiserum removed completely the ir.

Thus, we can describe two kind of brown labelling inthe histological sections of OB treated with the antise-rum against Gao, one was strong and abolished by thepreincubation of the antiserum with its antigen, theother was very weak and not abolished by the preincu-bation. We will discuss the strong immunolabeling asGao-like ir, while the weak one has been considered anaspecific staining.Gas/olf-like immunoreactivity. No ir was observed for

Gas/olf.Gaq-like immunoreactivity. No ir was observed for

Gaq.The distribution of the immunoreactivities in the OB

and the results of the negative controls are shown inTable 2.

Western Blotting Analysis

To confirm the presence of Gao, Gaq, and Gaolf inS. canicula, WB analysis (Western Blotting analysis)

Fig. 3. Immunohistochemical detection of Gao (immunofluores-cence) in the olfactory mucosa of S. canicula. (a) Epifluorescence micro-scope—Tangential section of the OE. The dendritic process of theORNs present Gao-like ir and are well visible intermingled with the sus-taining cells and with some mucous cells (M). (b) Confocal micro-scope—The immunofluorescence confirm the immunoperoxidasedistribution, in the epithelium and also in the lamina propria. The strogerir is visible in the dendritic process (arrows) of ORNs and in the axons(arrowheads). (c) Confocal microscope—The optical sectioning of anORN allows to observe the Gao-like ir especially localized along the cellmembrane (arrows) and in a thin layer around the nucleus (arrowhead).

Fig. 4. Immunohistochemical detection of Gas/olf (immunofluores-cence) in the olfactory mucosa of S. canicula. Epifluorescence micro-scope (a) Some ORNs show a weak immunoreactive cytoplasm(arrow). A fluorescent labeling is also present in the apical part of theepithelium (arrowhead) (b) Preincubation of the antiserum with its anti-gen did not abolish the immunostaining in the ORNs (arrows) nor inthe apical zone of the epithelium (arrowhead).

G-PROTEINS IN THE OLFACTORY SYSTEM OF A SHARK 1775

was performed and results was reportes in Fig. 7. PanelA showed the protein pattern stained with Blue Silver(lane 1, molecular weight markers, and lane 2, theS. canicula homogenate). In Panel B, a WB analysis,using an Ab against Gao was reported; the signal band,present in lane 2, was around 42 kDa. At the otherhand, the signals of Gaq or Gas/olf were absent in S. cani-cula olfactory organs (Supporting Information Fig. 1).

BLAST Search

The sequence KMVCDVVSRMEDTEPFSAEL (repre-senting the antigen used to obtain the antiserum againstGao) gave, on BLASTP, a significant alignment only

with G-protein alpha subunits of type o. The similarityranged from 20/20 to 16/20 in different vertebrates(mammals, birds, amphibians and bony fishes) andinvertebrates (e.g., Gao of Aplysia californica, Lymnaeastagnali and, Apis mellifera), reaching 15/20 in Drosoph-ila melanogaster. No Gaq, or Gas/olf showed a significantalignment with the query sequence.

DISCUSSION

Although the sequence of Gao, Gas/olf, and Gaq inChondrichthyes is not known, the data obtained usingthese commercial antisera can be considered reliable:the antiserum against Gao has been raised against a

TABLE 1. Overview of the ir in the olfactory mucosa of S. canicula

ORNs CNs Fibers in the lamina propria

Gao Immunohistochemistry þþþ � þþþPrimary antiserum neutralized by its antigen � � �Primary antiserum omitted � � �

Gas/olf Immunohistochemistry þ� � þ�Primary antiserum neutralized by its antigen þ� � þ�Primary antiserum omitted � � �

Gaq Immunohistochemistry � � �Primary antiserum neutralized by its antigen // // //Primary antiserum omitted � � �

þ�, weak ir; þþþ, strong ir; �, no ir; //, not performed.

Fig. 5. Schemes of the OB of S. canicula. (a) Dorsal view of theforebrain (left half). The OB is connected to rest of the brain by apeduncule. (b) Anterior view of the forebrain. The peduncule connectthe OB to the brain in the upper part. (c) Virtual sectioning of the OB.The different level of sectioning are reported in millimeters and should

be reported to the ruler of reference in Fig. 5b. The stratum glomerulo-sum is localized between the stratum nervosum and the stratum granu-losum in the ventral part of the OB, while in the dorsal part the stratumnervosum is pratically absent. The glomeruli negative to the Gao antise-rum are in the ventral anterior and middle zone of the OB (stars).

1776 FERRANDO ET AL.

peptide representing a very conserved region of this spe-cific type of G-protein alpha subunit, as verified by thesearch performed with BLASTP. Moreover, the proteindetected by the antiserum against Gao had a molecular

weight consistent with the data in the literature (Han-sen et al., 2004), while no signals were obtained in theWestern blot analysis using antisera against Gaq andGas/olf. Similar data were obtained in histological

Fig. 6. OB of S. canicula—Light microscope, (a) Hematoxylin–eosin. From the edge to the center it is possible to observe differentlayers in the OB: the stratum nervosum (SN), the stratum glomerulo-sum (SGl) and the stratum granulosum (SG) (b, c, d) Immunohisto-chemical detection of Gao (immunoperoxidase counterstained withhematoxylin) in the OB of S. canicula. (b) The stratum nervosum

shows a strong immunoreactivity (c) The major part of the glomeruli(G) are immunoreactive. The tissue around the glomeruli can be con-sidered the mitral layer, and show a weak, aspecific immunoreactivity.The mitral cells (MCs) are completely negative. (d) Some glomeruli inthe ventral part of the OB show a very weak aspecific staining.

TABLE 2. Overview of the ir in the OB of S. canicula

Fibers in thestratumnervosum

Mitralcells

Mitralzone

Few glomeruliin the

ventral OB

All theother

glomeruliStratum

granulosum

Gao Immunohistochemistry þþþ � þ� þ� þþþ þ�Primary antiserum neutralized

by its antigen� � þ� þ� þ� þ�

Primary antiserum omitted � � � � � �Gas/olf Immunohistochemistry � � � � � �

Primary antiserum neutralizedby its antigen

// // // // // //

Primary antiserum omitted � � � � � �Gaq Immunohistochemistry � � � � � �

Primary antiserum neutralizedby its antigen

// // // // // //

Primary antiserum omitted � � � � � �þ�, weak ir; þþþ, strong ir; �, no ir; //, not performed.

G-PROTEINS IN THE OLFACTORY SYSTEM OF A SHARK 1777

sections where only the antiserum against Gao showed aspecific ir. Thus, in this study, we demonstrate for thefirst time the presence of Gao-like ir in the olfactory sys-tem of the shark S. canicula and the absence of ir forthe subunits Gas/olf and Gaq.

In the literature the subunit Gao has been shown tohave different localizations according to the taxa consid-ered but, generally, it is expressed in all or most of themicrovillous RNs expressing either V1R or V2R receptorgene family. In mammals, Gao is expressed by the V2R-VRN populations of the VE, projecting into the posteriorpart of the accessory OB (Berghard and Buck, 1996;Halpern and Martı́nez-Marcos, 2003) and the same hasbeen found for Xenopus laevis (Hagino-yamagishi et al.,2004). In teleost (where the vomeronasal epithelium islacking) the immunolocalization of Gao in the olfactoryepithelium can vary according to the species; in Neogo-bius melanostomus Gao-like ir is localized in the micro-villous RNs (while the ciliated ones express Gas/olf) andin CNs (Belanger et al., 2003); in goldfish Gao-like ir ishighlighted in most of the microvillous V2R–RNs, theCNs (that express also the Gaq) and in a substantialpart of the olfactory nerve (Hansen et al., 2004). In con-trast to other teleosts, in channel catfish, Gao isexpressed in the CNs only, while the microvillous RNsexpress Gaq and the ciliated ones express Gas/olf.

The olfactory epithelium of Chondrichthyes seems tobe more primitive than that found in other vertebrates,showing no ciliated RNs and just two types of RN: themicrovillous ORNs and the CNs (Theisen et al., 1986).However, in S. canicula, as in mammals, amphibians,and some teleost fishes, the Gao subunit is typicallyimmunolocalized in neurons characterized by a microvil-lous dendritic zone. In contrast to those found in teleost,the CNs found in S. canicula did not show Gao-like ir.CNs are RNs present in both cartilaginous and bonyfishes (Hansen and Finger, 2000; Ferrando et al., 2006a)but their role in olfaction, their odorant specificity, andthe family of receptors they express, are still unknown.Despite an involvement in reproductive behaviour whichhas been suggested for this kind of RNs (Hamdani andDøving, 2006; Hamdani et al., 2008), studies on theirelectrophysiological properties in teleost suggest their

response to amino acids, generally thought to mediatefeeding behaviour (Schmachtenberg, 2006; Vielma et al.,2008). In Chondrichthyes, their involvement in repro-duction has been ruled out by their early appearanceduring the development in Raja clavata (Ferrando et al.,2007b). Our data suggest that the G-protein alpha subu-nit type expressed by microvillous ORNs in S. caniculadiffers from that expressed in the CNs, highlightingtheir different function. In Sarcopterygians the linkamong microvillous RNs, Gao and the V2R receptor fam-ily seems to be quite common (Halpern and Martinez-Marcos, 2003; Hagino-Yamagishi et al., 2004) while inActinopterygians the situation seems to be more variable(Hansen et al., 2003; 2004). Our data suggest that inChondrichthyes, as in most Sarcopterygians and someActinopterygians, the microvillous RNs present the Gao

as an expressed G-protein alpha subunit.The presence of 32 genes of the V2R receptor family

has been highlighted in sharks (Grus and Zhang, 2009).The next step should be to investigate the expression ofthese receptors in the microvillous RNs of S. canicula.

The distribution of G-protein alpha subunits in theOB, and in particular in the stratum glomerulosum,reflects the topographical order of the afferent fibers(Hansen et al., 2003) and hence it would be reasonableto expect the Gao-negative CNs to project into glomerulithat were also Gao-negative.

In the OB of S. canicula, almost all the glomeruli areGao-like immunoreactive and probably represent thezones of projection of the microvillous ORNs. Few nega-tive glomeruli have been observed in the ventral part ofthe OB. Thus, we have observed that, in the OE ofS. canicula, the only RNs negative to the immunodetec-tion of Gao seem to be the CNs and, again, the only glo-meruli negative to the immunodetection of Gao are inthe ventral (anterior and medial) part of the OB. Theseobservations suggest that in S. canicula the projectionsof CNs to the ventral OB, as already demonstrated inthe teleost Ictalurus punctatus and Carassius carassiusthrough neural tracer injection into the OB (Hansenet al., 2003; Hamdani and Døving, 2006). These datasuggest that the segregated distribution of the projec-tions of the two olfactory receptor neuron types to theOB, and indicate a topographical organization of the OBalso in Chondrichthyes.

ACKNOWLEDGMENTS

The authors are grateful to Prof. I. McCarthy (Schoolof Ocean Sciences, Bangor University) and Dott. M.Vacchi (ISPRA, Rome, Italy) for collecting the specimens.They also thank Prof. Alberto Diaspro and Dott. PaoloBianchini (L.A.M.B.S., Department of Physics, Univer-sity of Genoa) for having allowed the use of the confocalmicroscope and greatly improved the quality of theobservations with their experience. They also thankProf. R. Galliano Court, University of Wisconsin, Madi-son, for the linguistic revision.

LITERATURE CITED

Belanger RM, Smith CM, Corkum LD, Zielinski BS. 2003. Mor-phology and histochemistry of the peripheral olfactory organ inthe round goby, Neogobius melanostomus (Teleostei: Gobiidae).J Morphol 257:62–71.

Fig. 7. Western blot detection of Gao in homogenates of olfactoryorgans of S.canicula. (a) The protein pattern as stained with Blue Silver,the loaded samples were: molecular weight markers (lane 1), olfactoryorgans homogenate (lane 2). (b) The chemiluminescence signal of Gao

had an apparent molecular mass of about 42 � 4 kDa (mean � SD).

1778 FERRANDO ET AL.

Berghard A, Buck B. 1996. Sensory transduction in vomeronasal neu-rons: evidence for Gao, Gai2, and adenylyl cyclase II as major com-ponents of a pheromone signalling cascade. J Neurosci 16:909–918.

Bradford MM. 1976. A rapid and sensitive method for the quantita-tion of microgram quantities of proteins utilizing the principles ofprotein-dye binding. Anal Biochem 72:248–254.

Buck L, Axel R. 1991. A novel multigene family may encode odorantreceptors: a molecular basis for odor recognition. Cell 65:175–187.

Candiano G, Bruschi M, Musante L, Santucci L, Ghiggeri GM, Car-nemolla B, Orecchia P, Zardi L, Righetti PG. 2004. Blue silver: avery sensitive colloidal Coomassie G-250 staining for proteomeanalysis. Electrophoresis 25:1327–1333.

Dulac C, Axel R. 1995. A novel family of genes encoding putativepheromone receptors in mammals. Cell 83:195–206.

Dryer L. 2000. Evolution of odorant receptors. BioEssays 22:803–810.Ebbesson SO, Heimer L. 1970. Projection of the olfactory tract fibers

in the nurse shark (Ginglymostoma cirratum). Brain Res 17:47–55.Eisthen HL. 2004. The goldfish knows: olfactory receptor cell mor-

phology predicts receptor gene expression. J Comp Neurol 477:341–346.

Ferrando S, Bottaro M, Gallus L, Girosi L, Vacchi M, TagliafierroG. 2006a. Observations of crypt neuron-like cells in the olfactoryepithelium of a cartilaginous fish. Neurosci Lett 403:280–282.

Ferrando S, Bottaro M, Gallus L, Girosi L, Vacchi M, TagliafierroG. 2006b. Naþ/KþATPase immunoreactivity in the olfactory epi-thelium of small-spotted catshark Scyliorhinus canicula (L.): pos-sible presence of ion exchanging cells? J Fish Biol 69 (Suppl B):278–282.

Ferrando S, Bottaro M, Gallus L, Girosi L, Vacchi M, TagliafierroG. 2007a. First detection of olfactory marker protein (OMP)immunoreactivity in the olfactory epithelium of a cartilaginousfish. Neurosci Lett 413:173–176.

Ferrando S, Bottaro M, Pedemonte F, De Lorenzo S, Gallus L,Tagliafierro G. 2007b. Appearance of crypt neurons in the olfac-tory epithelium of the skate Raja clavata during development.Anat Rec 290:1268–1272.

Ferrando S. 2008. Ionocytes in the olfactory epithelium of develop-ing Raja clavata. Ital J Zool (Modena) 75:233–236.

Firestein S. 2001. How the olfactory system makes sense of scent.Nature 413:211–218.

Fishelson L, Baranes A. 1997. Ontogenesis and cytomorphology ofthe nasal olfactory organs in the Oman shark, Iago omanensis (Tri-akidae), in the Gulf of Aqaba, Red Sea. Anat Rec 249:409–421.

Franceschini V, Ciani F. 1993. Lectin binding to olfactory system ina shark, Scyliorhinus canicula. Folia Histochem Cytobiol 31:133.90–137.90.

Freitag J, Beck A, Ludwig G, von Buchholtz L, Breer H. 1999. Onthe origin of the olfactory receptor family: receptors genes of thejawless fish. Gene 226:165–174.

Grus WE, Zhang J. 2009. Origin of the genetic components of thevomeronasal system in the common ancestor of all extant verte-brates. Mol Biol Evol 26:407–419.

Hagino-Yamagishi K, Moriya K, Kubo H, Wakanayashi Y, Isobe N,Saito S, Ichikawa M, Yazaki K. 2004. Expression of vomeronasalreceptor genes in Xenopus laevis. J Comp Neurol 472:246–256.

Halpern M, Martı́nez-Marcos A. 2003. Structure and function of thevomeronasal system: an update. Prog Neurobiol 70:245–318.

Hamdani EH, Døving KB. 2006. Specific projection of the sensorycrypt cells in the olfactory system in crucian carp, Carassius car-assius. Chem Senses 31:63–67.

Hamdani EH, Lastein S, Gregersen F, Døving KB. 2008. Seasonalvariations in olfactory sensory neurons—fish sensitivity to sexpheromones explained? Chem Senses 33:119–123.

Hansen A, Anderson KT, Finger TE. 2004. Differential distributionof olfactory receptor neurons in goldfish: structural and molecularcorrelates. J Comp Neurol 477:347–359.

Hansen A, Finger TE. 2000. Phyletic distribution of crypt-typeolfactory receptor neurons in fishes. Brain Behav Evol 55:100–110.

Hansen A, Rolen SH, Anderson K, Morita Y, Caprio J, Finger TE.2003. Correlation between olfactory receptor cell type and func-tion in the channel catfish. J Neurosci 23:9328–9339.

Hansen A, Zielinski BS. 2005. Development, lamellar arrangement,sensory neuron cell types and transduction components.J Neurocytol 34:183–208.

Hueter RE, Mann DA, Maruska KP, Sisneros JA, Demski LS. 2004.Sensory biology of Elasmobranchs. In: Carrier JC, Musick JA,Heithaus MR, editors. Biology of sharks and their relatives. CRCPress, Boca Raton, FL. p 325–361.

Kajura SM, Forni JB, Summers AP. 2005. Olfactory morphology ofcarcharhinid and sphyrnid sharks: does the cephalofoil confer asensory advantage? J Morphol 264:253–263.

Korsching S. 2002. Olfactory maps and odor images. Curr OpinNeurobiol 12:387–392.

Korsching S. 2009. The molecular evolution of teleost olfactoryreceptor gene families. In: Meyerhof W, Korsching S, editors.Chemosensory systems in mammals, fishes and insects. Series:Results Probl Cell Differ. Vol. 47. Springer, Berlin/Heidelberg.p. 1–19.

Laberge F, Hara TJ. 2001. Neurobiology of fish olfaction: a review.Brain Res Rev 36:46–59.

Laemmli UK. 1970. Cleavage of structural proteins during theassembly of the head of bacteriophage T4. Nature 227:680–685.

Lipschitz DL, Michel WC. 2002. Amino acid odorants stimulatemicrovillar sensory neurons. Chem Senses 27:277–286.

Nagao H, Yamaguchi M, Takahash Y, Mori K. 2002. Grouping andrepresentation of odorant receptors in domains of the olfactorybulb sensory map. Microsc Res Tech 58:168–175.

Rasband WS. 2006. ImageJ. US National Institutes of Health,Bethesda, Maryland, USA. Available at: http://rsb.info.nih.gov/ij/.

Ravera S, Calzia D, Panfoli I, Pepe IM, Morelli A. 2007. Simulta-neous detection of molecular weight and activity of adenylatekinase after electrophoretic separation. Electrophoresis 28:291–300.

Sato K, Suzuki N. 2001. Whole-cell response characteristics of cili-ated and microvillous olfactory receptor neurons to amino acids,pheromone candidates and urine in rainbow trout. Chem Senses26:1145–1156.

Schmachtenberg O. 2006. Histological and electrophysiological prop-erties of crypt cells from the olfactory epithelium of the marineteleost Trachurus simmetricus. J Comp Neurol 495:113–121.

Speca DJ, Lin DM, Sorensen PW, Isacoff EY, Ngai J, Dittman AH.1999. Functional identification of a goldfish odorant receptor.Neuron 23:487–498.

Takami S, Luer CA, Graziadei PPC. 1994. Microscopic structure ofthe olfactory organ of the clearnose skate, Raja eglanteria. AnatEmbryol 190:211–230.

Takigami S, Mori Y, Ichikawa M. 2000. Projection pattern of vomer-onasal neurons to the accessory olfactory bulb in goats. ChemSenses 25:387–393.

Theisen B, Zeiske E, Breucker H. 1986. Functional morphology ofthe olfactory organs in the spiny dogfish (Squalus acanthias L.)and the small-spotted catshark (Scyliorhinus canicula (L.)). ActaZool 67:73–86.

Vielma A, Ardiles A, Delgado L, Schmachtenberg O. 2008. The elu-sive crypt olfactory receptor neuron: evidence for its stimulationby amino acids and cAMP pathway agonist. J Exp Biol 211:2417–2422.

Zippel HP, Hansen A, Caprio J. 1997. Renewing olfactory receptorneurons in goldfish do not require contact with the olfactory bulbto develop normal chemical responsiveness. J Comp Physiol A181:425–437.

G-PROTEINS IN THE OLFACTORY SYSTEM OF A SHARK 1779