glomerular territories in the olfactory bulb from the...

Glomerular Territories in the OlfactoryBulb from the Larval Stage of the Sea

Lamprey Petromyzon marinus

ANDREA FRONTINI,1 ALIYA U. ZAIDI,1 HONG HUA,1 TOMASZ P. WOLAK,1

CHARLES A. GREER,2 KARL W. KAFITZ,3 WEIMING LI,4

AND BARBARA S. ZIELINSKI1*1Department of Biological Sciences, University of Windsor,

Windsor, Ontario N9B 3P4, Canada2Sections of Neurosurgery & Neurobiology, Yale University School of Medicine, New

Haven, Connecticut 065203Pysiologisches Institut, Zellulaere Physiologie, Ludwig-Maximilians-Universitaet

Muenchen, 80336 Muenchen, Germany4Department of Fisheries and Wildlife, Michigan State University,

East Lansing, Michigan 48824

ABSTRACTThe goal of this study was to investigate the spatial organization of olfactory glomeruli

and of substances relevant to olfactory sensory neuron activity in the developing agnathan,the sea lamprey Petromyzon marinus. A 45-kD protein immunoreactive to Golf, a cAMP-dependent olfactory G protein, was present in the ciliary fraction of sea lamprey olfactoryepithelium and in olfactory sensory neurons of larval and adult sea lampreys. This resultimplies that Golf expression was present during early vertebrate evolution or evolved inparallel in gnathostome and agnathostome vertebrates. Serial sectioning of the olfactory bulbrevealed a consistent pattern of olfactory glomeruli stained by GS1B4 lectin and by antero-grade labeling with fluorescent dextran. These glomerular territories included the dorsalcluster, dorsal ring, anterior plexus, lateral chain, medial glomeruli, ventral ring, and ventralcluster. The dorsal, anterior, lateral, and ventral glomeruli contained olfactory sensory axonterminals that were Golf-immunoreactive. However, a specific subset, the medial glomeruli,did not display this immunoreactivity. Olfactory glomeruli in the dorsal hemisphere of theolfactory bulb, the dorsal cluster, dorsal ring, anterior plexus, lateral chain, and medialglomeruli, were seen adjacent to 5HT-immunoreactive fibers. However, glomeruli in theventral hemisphere, the ventral ring, and ventral cluster did not display this association. Thepresence of specific glomerular territories and discrete glomerular subsets with substancesrelevant to olfactory sensory neuron activity suggest a spatial organization of informationflow in the lamprey olfactory pathway. J. Comp. Neurol. 465:27–37, 2003.© 2003 Wiley-Liss, Inc.

Indexing terms: Agnatha; olfactory sensory neurons; serotonin; Golf

Grant sponsor: Natural Sciences and Engineering Research Council ofCanada; Grant number: 46383 (B.Z.): Grant number: PG-S1 (H.H.); Grantsponsor: Ontario Council on Graduate Studies (A.Z.); Grant sponsor: theGreat Lakes Fishery Commission (W.L., B.Z.); Grant sponsor: NationalInstitutes of Health; Grant number: DC002210; Grant number: NS10174(C.A.G.); Grant sponsor: the University of Windsor Faculty of GraduateStudies (A.F., A.Z., H.H.).

*Correspondence to: Barbara S. Zielinski, Department of Biological Sci-ences, University of Windsor, Windsor, Ontario N9B 3P4, Canada.E-mail: [email protected]

Received 25 July 2002; Revised 14 February 2003; Accepted 7 April 2003DOI 10.1002/cne.10811Published online the week of August 18, 2003 in Wiley InterScience

(www.interscience.wiley.com).

THE JOURNAL OF COMPARATIVE NEUROLOGY 465:27–37 (2003)

© 2003 WILEY-LISS, INC.

In all vertebrates, from agnathans to primates, axons ofolfactory sensory neurons (OSNs) project from soma in theolfactory epithelium to olfactory bulb glomeruli (Iwahoriet al., 1987; Halasz and Greer, 1993; Klenoff and Greer,1998; Lin and Ngai, 1999). Although functional differ-ences in OSNs that terminate in specific glomerular ter-ritories have been observed, glomerular organizing prin-ciples are not fully understood (Hildebrand and Shepherd,1997; Xu et al., 2000). The arrangement of OSN axonterminals into glomerular units may be affected by trans-duction mechanisms fundamental to OSN function and bymodulatory influences from nonolfactory neuronal fibersadjacent to OSN axons. The cAMP-dependent olfactoryGTP-binding protein linked to olfactory receptors (Jonesand Reed, 1989), Golf, is a constituent of olfactory sensorytransduction and a requisite for olfactory responses inmammals (Belluscio et al., 1998). Because OSN subpopu-lations in amphibians (Mezler et al., 2001) and teleosts(Hansen et al., 2001) express Golf, this G�s subunit mayhave appeared early in vertebrate evolution. AlternateG-proteins are found in vomeronasal sensory neurons (Jiaand Halpern, 1996) and in OSN subpopulations of am-phibians (Mezler et al., 2001) and teleosts (Hansen et al.,2001). If Golf is a vital component for vertebrate OSNfunction, its localization in the olfactory organ of the sealamprey, an ancestral jawless fish, is expected. If trans-duction by alternate G-proteins is also a fundamentalcharacter or if it evolved in parallel (Eisthen, 2002),clearly defined subpopulations of OSNs that do not ex-press Golf should be present in the lamprey.

Previous studies have provided evidence that OSN ter-minals are distributed according to functional parame-ters. In mammals, axons of OSNs expressing putativeodorant receptors within an epithelial zone converge tospatially conserved glomeruli (Mombaerts, 1996). In addi-tion, OSNs containing elements of a cGMP signal trans-duction pathway project to specific glomeruli (Juilfs et al.,1997) and identified odorants stimulate synaptic activityin specific glomeruli (e.g., Wachowiak and Cohen, 2001).In zebrafish (Teleostei, Cypriniformes), olfactory glomer-uli are arranged in a stereotyped configuration (Baier andKorsching, 1994) and glomerular responsiveness to odorstimulation follows spatial patterning (Friedrich and Kor-sching, 1998). In catfish (Teleostei, Siluriformes), twoOSN subpopulations project axons to specific regions ofthe olfactory bulb (Morita and Finger, 1998), and in thelamprey, Lampetra fluviatilis, calretinin-immunoreactiveOSNs extend to particular glomerular locations (Pombalet al., 2002).

In view of these examples of spatially and functionallydistinct glomerular arrangements, we propose that sub-populations of OSNs expressing the G-protein, Golf, ex-tend axons into spatially distinct glomeruli in the lam-prey. Earlier reports have shown a ring-like arrangementof olfactory glomeruli in Lampetra fluviatilis and L. pla-neri (Schober, 1964), Ichthyomyzon unicuspis (Northcuttand Puzdrowski, 1988), Petromyzon marinus (Tobet et al.,1996), and Lampetra japonica (Iwahori et al., 1997). How-ever, details of the arrangement of the glomerular terri-tories in lampreys are lacking. The olfactory receptor fam-ily is relatively small in lampreys (Berghard and Dryer,1998; Freitag et al., 1999) and, not surprisingly, few odor-ants stimulate olfactory activity (Li and Sorensen, 1995).These findings suggest that the larval lamprey has con-siderably fewer olfactory glomeruli than the 1,800–2,400glomeruli located in the mammalian olfactory bulb (Royet

et al., 1988; Meisami and Sendera, 1993) and that map-ping of glomerular territories is feasible in serially sec-tioned preparations of the lamprey olfactory bulb.

In sea lamprey, nonolfactory serotonergic nerve fibersare present along the primary olfactory pathway and en-ter the olfactory bulb (Zielinski et al., 2000). These fibersmay be associated with particular olfactory sensory inputand may innervate specific glomerular regions. Therefore,the goal of this study was to probe the olfactory bulb of thedeveloping lamprey for spatially conserved glomerularterritories and for the distribution of two substances thatmay be relevant to OSN activity: the protein Golf and thebiogenic amine neurotransmitter serotonin.

In this study we observed six glomerular territories,Golf-immunoreactivity at specific glomerular sites, anddistinct glomerular subsets with serotonergic nonolfactoryfibers. These results imply differences in transductionmechanisms for some OSNs and modulation of specificOSNs by biogenic amines.

MATERIALS AND METHODS

Experimental animals

Year two and three class larval sea lampreys were ob-tained from naturally occurring populations collected fromcreeks in Michigan, then housed at the Lake Huron Bio-logical Station, Millersburg, Michigan, or from OshawaCreek and Bronte Creek, Ontario. All larval lampreyswere maintained at 10°C at the Department of BiologicalSciences, University of Windsor. Approximately 90 larvaewere used for this study (total length 80–130 mm, weight0.6–2.7 g). Adult sea lampreys were trapped or collectedby hand from tributaries to Lakes Huron and Michigan,transported to the U.S. Geological Survey Lake HuronBiological Station, Millersburg, Michigan. These adultswere held in flow-through tanks (1,000 L) with Lake Hu-ron water (7–20°C) before being euthanized for collectionof olfactory organs. Prior to experimentation, all lampreyswere deeply anesthetized with tricaine methane sulfonate(MS-222) and killed by decapitation. All experimental pro-tocols reported in this study were in compliance withguidelines established by the Canadian Council of AnimalCare.

Western immunoblot

The cilia were dissociated from OSNs by calcium shock(Schandar et al., 1998). In brief, sea lamprey olfactory

Fig. 1. Western immunoblot of Golf in a preparation from theolfactory epithelium of an adult sea lamprey. Molecular weight mark-ers are shown on the left. Lane A: 20 �g deciliated lamprey olfactoryepithelium. Lane B: 10 �g deciliated lamprey olfactory epithelium.Lane C: 20 �g ciliated lamprey prep. Lane D: 10 �g ciliated lampreyprep. The Golf antibody concentration was tested at a 1:500 dilution.

28 A. FRONTINI ET AL.

epithelia were dissected out and agitated in a high cal-cium buffer (10 mM CaCl, 20 mM ACES, 0.3 M sucrose, 10�g/ml leupeptid, 76.8 nM aprotinin, 0.7 �M pepstatin,0.83 mM benzamide, 0.23 mM PMSF, 1 mM iodoacet-amide) in an end-over-shaker at 4°C for 20 min. Thesolution was spun for 15 min at 6,000g and the superna-tant was collected. The pellet was resuspended in thesame buffer and spun down again for collection of thesupernatant. The combined supernatant was spun for 15min at 18,000g to isolate the cilia. The ciliary pellet waswashed with TME buffer (10 mM Tris, 3 mM MgCl, 2 mMEGTA, pH 8.2) and resuspended. Aliquots were stored at–80°C until use. The deciliated olfactory mucosa was pre-pared from the pellet of the first centrifugation accordingto DellaCorte et al. (1996). The deciliated tissue was ho-mogenized using a mortar and pestle and centrifuged at30,000g for 90 min at 0°C. The supernatant was removedand stored at –80°C. The protein concentration was de-termined by DCA protein analysis kit (Pierce Biotechnol-ogy, Rockford, IL). Ten and 20 �g protein of cilia anddeciliated mucosa were loaded onto a 10% SDS PAGE geland 5% stacking gel which were subjected to 150 V for 1hour in a standard running buffer. The protein bands inthe gel were transferred to a PVDF membrane. The mem-brane was then blocked with a blocking solution contain-ing 5% nonfat dry milk in buffer, incubated with theprimary antibody (anti-Golf 1:200 or 1:500) in the blockingsolution, washed three times with buffer, and incubatedwith HRP-conjugated secondary antibody (1:5,000) in theblocking solution. Finally, the membrane was washedthree times and incubated in chemiluminescence sub-strate for 10–15 minutes and film was exposed and devel-oped.

Tissue fixation

The larval heads were immersed in Zamboni’s fixativefor 4–20 hours (2% paraformaldehyde, 1.5% picric acid,0.1 M phosphate buffer, pH 7.4) at 4°C, then cryoprotectedby passage through a sucrose gradient (10–20–30% inPB). Horizontal sections (20–30 �m) were cut on a cryo-stat (Microm). Sections were air-dried for at least 1 hour,postfixed in acetone at –20°C for 10 minutes, and rehy-drated with 0.1 M phosphate buffered saline (PBS) (pH7.4) for 10 minutes at room temperature.

GS1B4 lectin histochemistry

The staining of larval sea lamprey olfactory glomeruliby Griffonia simplicifolia-1 (GS-1 isolectin B4) was previ-ously described (Tobet et al., 1996; Zielinski et al., 2000).Slides were incubated with biotin-conjugated lectin, G.simplicifolia-1 (GS-1 isolectin B4, Vector, Burlingame, CA;

Fig. 2. Localization of Golf-IR in the olfactory epithelium. A and Bshow a double labeled preparation. A: Acetylated tubulin (AT)-immunolabeling in the larval nasal cavity shows intense labeling ofthe ciliary layer lining the luminal surface and of OSN soma in theolfactory epithelium (horizontal plane). Nonsensory epithelium is lo-cated on the right side of the micrograph. B: The luminal surface ofthe olfactory epithelium is Golf-IR, and the surface of the nonsensoryepithelium is unlabeled (indicated by asterisks). OSN cilia, dendrites,cell bodies, and axons are Golf-IR (arrow). C: Obliquely sectionedOSNs show intense Golf-immunoreactivity in olfactory cilia. D: AnOSN in a metamorphosing sea lamprey is intensely Golf-IR (Z-series ofseven sections, taken at 0.2 �m steps. A, B, and D are the samemagnification. Scale bar in B � 50 �m; 25 �m in C.

29OLFACTORY BULB IN LARVAL SEA LAMPREY

Figure 3


10 �g/ml in 0.1 M PBS, pH 7.5) for 3 hours. These wererinsed in PBS, then incubated in DCS avidin-fluorescein(1:100 in 0.1 M bicarbonate buffer, pH 8.5) for 1 hour,washed, then mounted with Vectashield.

Immunoctyochemistry

Slides with tissue sections were incubated in dilutednormal horse serum for 20 minutes, then in primary an-tiserum raised in rabbits against Golf (1:1,000; Santa CruzBiotechnology, Santa Cruz, CA) or serotonin (5-HT,1:3,000; DiaSorin, Stillwater MN) in 0.1 M PBS, pH 7.4,containing 0.1% Triton X-100 overnight at 4°C. The sec-tions were rinsed in PBS, incubated in Alexa 568 goatantirabbit IgG (Molecular Probes, Eugene, OR; 1:100 inPBS, pH 7.4) for 1 hour and rinsed in PBS. Negativecontrols, with primary antibody omitted from the stainingprocedure, were included with each immunocytochemicalpreparation. Specificity of the 5HT antibody was tested byomitting this primary antibody from the staining protocoland by a preadsorption control experiment. Preadsorptioncontrol preparations for 5-HT were prepared with dilutedantiserum that had been preadsorbed with 10 mM or 100mM 5HT (Sigma, St. Louis, MO) for 24 hours at 4°C.

Application of calcium green dextran

The glomerular units containing OSN projections in theolfactory bulb were anterogradely labeled following theapplication of calcium green dextran into the nasal cavity.Prior to experimentation all lampreys were individuallyanesthetized by immersion in a 0.05% solution of MS222.After the animal was sedated it was wrapped in wet tissuepaper. A solution of 2.5% calcium green 1-dextran (potas-sium salt 3,000 MW anionic, Molecular Probes), 0.2%TritonX-100, and 1 mM NaCl, 0.1 M Na bicarbonate inwater was injected into the nasal cavities using a Hamil-ton syringe with a 23 gauge needle. Following this treat-ment the lamprey recovered and another injection wasapplied the following day. On the fifth day following thisinjection the lamprey was immersed in a solution ofMS222, deeply anesthetized, and killed by decapitation.

The larval heads were immersed into 4% paraformalde-hyde fixative overnight at 4°C, then cryoprotected by pas-sage through a sucrose gradient (10–20–30% in PB).Cryosections (20 �m) were cut in horizontal planes on acryostat (Microm). Coverslips were mounted withVectashield (Vector). The sections were viewed by fluores-cence microscopy or confocal microscopy (BioRad 1024,Hercules, CA).

Morphometry

Larval olfactory bulbs prepared for GS1B4 lectin histo-chemistry, fluorescent dextran, Golf, and 5HT were seri-ally sectioned and stained (n � 25). These sections wereviewed on an Axioskop (Zeiss) and prepared as a 3D movieusing Empix Eclipse software (Mississauga, Canada).This computer program was used to measure the glomer-ular diameters and the total glomerular volume. The num-ber of olfactory glomeruli olfactory bulbs was estimated byapplying the dissector technique (Gundersen et al., 1988)to the 3D reconstruction of the serially sectioned olfactorybulbs.

Production of photomicrographs

The confocal images were acquired in BioRad PIC for-mat and converted to TIFF format with Confocal Assis-tant. The fluorescence microscopy images were acquiredusing Empix Eclipse software and saved in TIFF format.The photomicroscope images were assembled into figuresand labeled with Adobe PhotoShop (Mountain View, CA).

RESULTS

Western blotting of the ciliary fraction from adult lam-prey olfactory epithelium demonstrated the specificity ofthe Golf antibody. The molecular weight of Golf in the ciliaof lamprey OSNs was 45 kDa, with a slightly highermolecular weight component likely from the phosphory-lated form of this G-protein (Fig. 1). The strong Golf-immunoreactive (IR) band in the ciliary fraction is sup-ported through immunolocalization. Ciliary localizationwas seen in double-labeled preparations with acetylatedtubulin and Golf antibodies (Fig. 2A,B). Aceylated tubulinimmunoreactivity indicated the presence of cilia (Fig. 2A).In larvae, the boundary between the olfactory epitheliumand non-Golf-expressing nasal epithelium was sharp andclear (Fig. 2B). High-power views confirmed strong Golf-IRin cilia of OSNs (Fig. 2C). Golf expression in cilia, den-drites, cell bodies, and axons persisted after metamorpho-sis (Fig. 2D), when OSNs were considerably larger thanduring larval development (Vandenbossche et al., 1995,1997). OSN subcellular components located in the olfac-tory epithelium, dendrites, soma, and axons containedGolf-IR (Fig. 2D). The presence of Golf-immunoreactivity inthe OSN axons led us to examine spatial distribution ofGolf-axons in the olfactory bulb.

Glomerular territories

The detailed glomerular organization in the lampreyolfactory bulb has not been documented. Therefore, wemapped the olfactory bulbar neuropil innervated by OSNsbefore charting Golf-IR glomeruli. Examination of the ol-factory bulb from serial sections in the horizontal plane,stained with GS1B4 lectin (Fig. 3), and by anterogradelabeling with fluorescent dextran, revealed a consistentpattern of glomerular organization. All glomeruli thatwere identified were labeled by both techniques. The spa-

Fig. 3. The organization of olfactory glomeruli from a 125-mmlarval sea lamprey. The horizontal sections were labeled with GS1B4lectin and visualized by confocal microscopy. The arrows in A showthe orientation of the images. R is rostral, M is medial. The seriesshows sections taken at intervals of �100 �m; however, sections withprominent medial glomeruli are shown at 50-�m intervals. A: 100 �m.Dorsal cluster. A dorsal group of six glomeruli is seen at the anteriorand lateral edges of the olfactory bulb. B: 200 �m. Dorsal ring.Glomeruli are arranged in a circular configuration. The olfactorynerve layer is prominent at the anterior/lateral edge. C: 300 �m.Anterior plexus and lateral chain. The anterior plexus (AP) and lat-eral chain (LC) are present at the dorsal edge of the junction of theolfactory nerve with the olfactory bulb. D: 350 �m and E: 400 �m.Anterior plexus, lateral chain and medial glomeruli. Three medialglomeruli seen at a depth of 350 �m: anterior radial (rA), radialposterior (RP), and the glomerulus adjacent to the anterior commis-sure (AC). At a depth of 400 �m the medial glomerulus adjacent to theanterior commissure is present and individually labeled fibers areseen in the neuropil lateral to this glomerulus. F: 500 �m and G: 550�m. Ventral ring and medial glomeruli. Ventral to the olfactory nervethe anterior plexus and the anterior medial glomeruli form a circulararrangement. A medial glomerulus extends toward the anterior com-missure (500 �m) and a posterior medial glomerulus (PM) is promi-nent (550 �m). H: 600 �m. Ventral ring. I: 700 �m. Ventral cluster.An aggregate of glomeruli is located at the ventral edge of the olfac-tory bulb. Scale bar � 100 �m in A.


Figure 4

tial distribution of the glomerular groupings was consis-tent in the larvae examined (80–130 mm long). The thick-ness of the olfactory bulb was �500 �m in the smallerlampreys. Spatial analysis is presented in a series ofevenly spaced sections through the olfactory bulb of larvaewith a length of 12–13 cm, where the thickness of theolfactory bulb was 725–800 �m. The olfactory nerve en-tered the olfactory bulb at a depth of 300–425 �m. Thediameter of olfactory glomeruli ranged from 45–90 �m.Tallies of glomerular modules in serially sectioned larvalolfactory bulbs indicated that the total number of glomer-uli ranged from 41–65 (n � 10). In addition to formingolfactory glomeruli, OSN fibers were seen entering theolfactory bulb from the olfactory nerve and proceedingmedially beyond the olfactory bulb to the ventral dien-cephalon, as previously reported (Tobet et al., 1996).

We consistently observed the following glomerular ter-ritories in larval sea lampreys.

Dorsal cluster, 40–100 �m. A cluster of 6–10 olfac-tory glomeruli was apparent close to the dorsal edge of theolfactory bulb (Fig. 3A).

Dorsal ring, 100–200 �m. Approximately 7–10closely spaced glomerular modules were arranged in acircular pattern along the medial, anterior, and lateraledges and the olfactory nerve layer was prominent at theanterior and lateral edge (Fig. 3B).

Anterior plexus, 225–475 �m. The anterior plexusconsisted of a cohesive mesh of OSN axon terminals (Fig.3C–E). Intense GS1B4 labeling at the medial edge indi-cated the location of olfactory nerve fibers entering theolfactory bulb.

Lateral chain, 225–475 �m. A space about 50 �mwide separated the anterior plexus from the lateral chain,a relatively narrow group of glomerular modules extend-

ing along the lateral edge of the olfactory bulb (Fig. 3C–E).The anterior plexus and lateral chain were previouslyobserved in the zebrafish olfactory bulb (Baier and Kor-sching, 1994).

Medial glomeruli, 350–575 �m. At bulbar depths of350–425 �m, medial glomeruli were positioned posteriorto the junction of the olfactory nerve and olfactory bulb(Fig. 3D,E). At 350 �m, a small anterior-radial glomeruluswas positioned posterior to the medial edge of the anteriorplexus; a second medial glomerulus, the posterior-radialglomerulus, was positioned radially (Fig. 3D), and a thirdglomerulus faced the anterior commissure (Fig. 3D,E).Medial glomeruli were located ventral to the olfactorynerve (450–600 �m): a posterior glomerulus and a glomer-ulus adjacent to the anterior commissure (Fig. 3F,G).

Ventral ring, 500–675 �m. A circular arrangement of10–14 glomerular units were present in sections in theolfactory bulb taken ventral to the olfactory nerve atdepths of 500–600 �m (Fig. 3F–H). At 500 �m, GS1B4-positive fibers were seen extending from anterior glomer-uli of this ring into the granular layer.

Ventral cluster, 700–750 �m. The bottom portion ofthe olfactory bulb contained a cluster of 7–10 olfactoryglomeruli (Fig. 3I).

Glomerular Golf-immunolabeling. Glomerular terri-tories in the dorsal, anterior, lateral, and ventral regionsof the olfactory bulb contained Golf-IR. However, the me-dial glomeruli were not Golf-IR (Fig. 4). The Golf-immunolabeling was also intense in the olfactory nerveand the olfactory nerve layer of the olfactory bulb.

Dorsal cluster and dorsal ring. The glomerularunits in this dorsal region of the olfactory bulb containedGolf-immunoreactivity (Fig. 4A).

Anterior plexus and lateral chain. These glomerulicontained Golf-immunoreactivity (Fig. 4B).

Medial glomeruli. Golf-immunoreactivity was absentfrom the region occupied by medial glomeruli. In prepara-tions double-labeled with Golf immunocytochemistry andanterogradely applied fluorescent dextran, medial glomer-uli were devoid of Golf-immunoreactivity (Fig. 4C–E).These included the posterior radial glomerulus, the glo-merulus facing the anterior commissure, the posteriormedial glomerulus, and individual fibers extending intothe neuropil. The anterograde labeling shows directly thatOSN axons extend to the medial glomeruli of the olfactorybulb, and that these axons do not stain by Golf-immunocytochemistry. The anterograde labeling was lim-ited to the OSN axons that took up the dextran when itwas applied to the nasal cavity in vivo, and labeled OSNaxons were less populous than with GS1B4 lectin labeling(Fig. 4E). The lectin, which reacts with carbohydrate res-idues on extracellular axonal surface, stained more ro-bustly than the dextran, which was confined to the cyto-plasm within the narrow intracellular axonal space.

Ventral ring and ventral cluster. Glomeruli in thisregion were Golf-IR (Fig. 4F,G).

Glomerular subsets with differing Golf expression per-sisted in the adult stage of the sea lamprey. Medial glo-meruli maintained an absence of Golf-IR compared to theremaining glomerular units (Fig. 4H–I).

Spatial relationship between 5HT-IR fibersand olfactory glomeruli

The dorsal region of the olfactory bulb contained morerobust 5HT-IR than the ventral region (Fig. 5). Double-

Fig. 4. Golf-IR in olfactory glomeruli of the larval and adult lamprey.Scale bar in A � 100 �m. The depth of each section is indicated. A, B, F,and G are single-labeled preparations (Golf-immunocytochemistry). Cand D are double-labeled (Golf-immunocytochemistry and fluorescentdextran). E is double-labeled (Golf-immunocytochemistry and GS1B4 lec-tin). A: 200 �m. Dorsal ring. B: 300 �m. Anterior plexus and lateralchain. C: 350 �m and D: 400 �m. Anterior plexus, lateral chain, andmedial glomeruli. In C the medial edge of the olfactory bulb is located onthe right edge of the micrograph. In D the center of the micrograph showsthe medial regions of adjacent olfactory bulbs. The anterior and lateralglomeruli show double-labeling for Golf and anterogradely applied fluo-rescent dextran. These labels do not show complete colocalization, as thegreen (dextran) is within the cytoplasm of the OSN axons and the red Golfis localized on the membranous surface. OSN axons in the medial glo-merulus facing the anterior commissure (AC) labeled with the dextran,but not with Golf immunocytochemistry. E: 500 �m. Ventral ring andmedial glomeruli (Z series of eight sections taken at 1-�m intervals). Theglomeruli of the ventral ring show double-labeling for Golf and for GS1B4lectin. Both stains are localized on the cell membrane. The medial glo-meruli, including posterior medial glomeruli (PM), are not Golf-IR. F: 600�m. Ventral ring. G: 700 �m. Ventral cluster. H to J show a horizontalview of the olfactory bulb from an adult sea lamprey, double-labeled,GS1B4 lectin histochemistry, and Golf immunocytochemistry. H and I arefluorescence micrographs shown at the same magnification. Scale bar inH � 500 �m. H: GS1B4 lectin stains olfactory glomeruli in the anterior,lateral, and medial regions. I: Anterior and lateral olfactory glomeruliare Golf-IR. The area enclosed by a green box is shown at high powerin J. J: A high-power confocal micrograph of the medial olfactory glo-merular region. Green is GS1B4 lectin histochemistry, red is Golf-immunolabeling, and yellow is colocalized Golf-immunolabeling andGS1B4 lectin staining. The red and green channels were imaged inseries, to avoid fluorescein bleed-through on the red channel. Scale bar �25 �m.


labeling experiments from eight larvae showed a constantpattern of 5HT innervation of glomerular territories.

Dorsal cluster and dorsal ring. Serotonergic fibersentered the glomerular units and were located betweenthe closely spaced glomeruli (Fig. 5A).

Anterior plexus and lateral chain. Serotonergic fi-bers were present in the space between the anteriorplexus and the lateral chain and surrounding the periph-ery of the anterior plexus and the lateral chain (Fig. 6B).As previously observed, 5HT-IR fibers extended along the

Fig. 5. Double-labeled preparations of the olfactory bulb from thelarval lamprey. GS1B4 lectin is green and 5HT immunocytochemistryis red. Yellow indicates regions of overlap between the axons of OSNsand 5HT-IR fibers. Glomerular regions with 5HT-IR fibers adjacent toOSN axons are circled in white. A, B, and C are the same magnifica-tion. The distance from the dorsal surface of the olfactory bulb of eachsection is indicated for each micrograph. A: 100 �m. Dorsal cluster.Five regions with close apposition between OSN and 5HT-IR fibersare present. The neuropil proximal to the glomeruli is populated by

5HT-IR fibers. B: 300 �m. Anterior plexus and lateral chain. 5HT-IRfibers are located in glomerular units adjacent to the olfactory nerveand in the space between the anterior plexus and lateral chain (ar-row). C: 400 �m and D: 420 �m. Medial glomeruli. High-power viewsof medial glomerular neuropil show seven regions with 5HT-IR fibersadjacent to OSN axons. E: 500 �m. Ventral ring and medial glomer-ulus. In the ventral region 5HT-IR fibers are weakly stained andconfined to the granular cell layer of the olfactory bulb. Scale bars �100 �m in A; 50 �m in C.


olfactory nerve into the olfactory bulb (Zielinski et al.,2000). In the lateral portion of this region, 5HT-IR fibersextended into the olfactory bulb neuropil through the an-terior plexus.

Medial glomeruli. Serotonergic fibers extended intothe medial posterior glomerulus from the medial region ofthe olfactory nerve (Fig. 5C,D). Two medial glomeruli, theradiomedial glomeruli and the glomerulus adjacent to theanterior commissure, were devoid of 5HT-IR.

Ventral ring and ventral cluster. Glomerular mod-ules in this region of the olfactory bulb did not contain5HT-IR fibers (Fig. 5E).

In summary, 5HT-IR fibers and OSN axons were inclose proximity in the dorsal cluster and the medial pos-terior glomerulus, and 5HT-immunoreactive fibers wereleast populous in the ventral olfactory bulb.

DISCUSSION

Examination of the distribution of substances that maybe relevant to OSN function in the sea lamprey revealedspatially conserved glomerular associations. The olfactoryreceptor linked protein Golf was expressed by sea lampreyOSNs projecting to dorsal, anterior, lateral, and ventralglomerular subsets. Innervation by nonolfactory 5HT-IRfibers was concentrated in the dorsal cluster and the me-dial posterior glomerulus. The principle differences be-tween the glomerular groups is summarized in Table 1and in Figure 6. Although these lampreys were from nat-urally occurring populations that represented diversegene pools, there was remarkable consistency in thesemorphological, biochemical, and spatial characteristics.

Bulbar Golf-IR and OSN subtypes

The molecular weight of the lamprey Golf, 45 kDa, isidentical to that of Golf that regulates adenlyl cyclaseactivity in mammals (Jones and Reed, 1989) and teleosts(Abogadie et al., 1995; Dellacorte et al., 1996). The ciliarylocalization of lamprey Golf, shown both through Westernimmunoblotting and immunocytochemically, supports theinvolvement of Golf in olfactory sensory transduction.Therefore, this G-protein may have been present in ver-tebrate ancestors over 400 million years ago, or it mayhave evolved in parallel during agnathan evolution. Golf-labeling was prominent in the glomerular units of thedorsal ring, anterior plexus, lateral chain, and ventralcluster. This is somewhat in contrast with prior reports inthe rodent where Golf was not detected in the olfactorybulb, although other molecules, including olfactorymarker protein (Danciger et al., 1989) and mRNA for the

olfactory marker protein (Wensley et al., 1995) and odorreceptor are present in OSN axons (Mombaerts, 1996).This difference may underscore a species-specific pheno-type but nevertheless emphasizes that subsets of lampreyOSNs may use Golf transduction cascades. Detection inthe axons may reflect further on the importance of Golf innot only odor transduction, but also modulation of growthcone behavior through cascades that involve Golf and cy-clic nucleotide-gated channels, as suggested by Kafitz etal. (2000). The absence of Golf-IR in the lamprey medialglomeruli points to a differing signal transduction mech-anism for these compared to the other glomerular group-ings. In various teleosts the medial olfactory path is asso-ciated with bile acid and steroid perception (e.g.,Thommesen, 1978; Hara and Zhang, 1998), and in thezebrafish, medial glomeruli respond to chemostimulationby these compounds (Friedrich and Korsching, 1998). TheOSN medial glomeruli in the lamprey may also constitutea subtype of OSN specialist expression. It is not surprisingto find a subpopulation of lamprey OSNs without Golf-IR.The expression of various G-proteins in OSN subtypesappears to be a principle of the vertebrate olfactory system(e.g., Shinohara et al., 1992; Berghard and Buck, 1996; Jiaand Halpern, 1996; Wekesa and Anholt, 1999; Hansen etal., 2001; Mezler et al., 2001). The medial location of theglomeruli that do not express Golf implies that the OSNsprojecting to these glomeruli use an alternate G-proteinduring olfactory sensory transduction. In tetrapods, che-moreceptive neurons of the vomerosal system contain theG-proteins Gi�2 and G0� (Halpern et al., 1995; Berghardand Buck, 1996; Jia and Halpern, 1996), and in the mam-

TABLE 1. Summary of the Distribution of Relevant Molecules inGlomerular-Territories in the Olfactory Bulb from 12–13 cm

Long Larval Sea Lampreys

Glomerularterritory

Depth(�m)

Golf-IRglomeruli 5HT-IR fibers

Dorsal cluster 40–100 Labeled Within glomeruliDorsal ring 100–200 Labeled Within glomeruliAnterior plexus 225–475 Labeled Medial edge of anterior plexus*Lateral chain 225–475 Labeled Outer edge of the lateral chain*Medial glomeruli 350–575 Unlabeled Through medial glomeruliVentral ring 500–675 Labeled AbsentVentral cluster 700–775 Labeled Absent

The glomerular territories were labeled by GS1B4 lectin and by anterograde applicationof fluorescent dextran.*The region between anterior plexus and lateral chain contains 5HT-IR fibers.

Fig. 6. A schematic summary of the position of glomerular terri-tories with Golf-IR and 5HT-IR labeling in the olfactory bulb fromlarval lampreys 12–13 cm long. The diagram represents an anteriorview of the olfactory bulb. The depth of the olfactory bulb is shown onthe left (�m). The glomerular territories are shown as rectangles andthe regions with 5HT-IR fibers are shown as cylinders. Glomeruli withGolf-IR are dark gray; the medial glomeruli, which are not Golf-IR, areshown by a patterned rectangle.


malian olfactory system, “necklace olfactory glomeruli”adjacent to the accessory olfactory bulb contain cGMP-stimulated phosphodiesterase and guanylyl cyclase-D(Juilfs et al., 1997) and heterogenous immunoreactivityfor neural proteins (Ring et al., 1997). Therefore, the use ofG-protein subtypes may have preceded the gnathostomeradiation, or have evolved in parallel in gnathostome andagnathostome vertebrates.

Nonolfactory fibers adjacent to dorsal andmedial glomeruli

In the lamprey olfactory bulb 5HT-IR fibers were lo-cated between glomerular groupings, particularly in lat-eral and medial locations of the dorsal hemisphere. Inmammals, periglomerular serotonergic fibers originatefrom raphe nuclei (McLean and Shipley, 1987; Philpot etal., 1994). However, in the lamprey the nonolfactory5HT-IR fibers from the primary olfactory pathway enterthe olfactory bulb (Zielinski et al., 2000). The presentstudy extends previous results by investigating the spatialrelationship between these fibers and olfactory glomeruli.The most prominent glomerular sites with 5HT innerva-tion were the dorsal cluster and the medio-posterior glo-merulus. This anterior dorsal/posterior medial ventraldistribution is reminiscent of the dorsomedial pathway ofthe terminal nerve in the African lungfish (Von Bartheldand Mayer, 1988; Schober et al., 1994) and may indicatethat this 5HT-immunoreactive pathway is a derivative ofthe nasal terminal nerve.

Spatial pattern of OSN projections

The pattern of glomerular organization in the larvallamprey was similar yet considerably reduced from thepattern formed by 18 glomerular groups in the olfactorybulb of the teleost Danio rerio (Baier and Korsching,1994). In both, glomeruli were clustered in dorsal andventral regions, the anterior plexus stretched from thedorsal to the ventral olfactory bulb, and the lateral groupappeared as a chain of glomeruli. The overall simplicity ofthe lamprey’s glomerular arrangement can be seen fromthe ring of glomeruli encircling the dorsal and ventralregions. Images from previous studies of the adult silverlamprey Ichthyomyzon unicuspis, (Northcutt and Puzd-rowski, 1988) and larval Petromyzon marinus (Tobet etal., 1996) have shown this ring-like pattern, as well as themedially located glomeruli. The four medial glomeruli (ra-dial anterior, radial posterior, adjacent to the anteriorcommissure, and posterior medial) that we observed in thelarval lamprey are fewer than the several medial glomer-uli located in the olfactory bulb of the zebrafish (Baier andKorsching, 1994). The scarcity of glomeruli in the lam-prey’s dorsomedial region of olfactory bulb may be due tothe fact that this neuropil is occupied by contralateralsecondary olfactory projections (Northcutt and Puzd-rowski, 1988). The larval olfactory glomeruli were smallerthan the 140 �m (average) diameter reported for glomer-uli in adult Lampetra japonica (Iwahori et al., 1987), andwithin the range observed in adult zebrafish, (25 and 140�m; Baier and Korsching, 1994) and rats (50–100 �m;Pinching and Powell, 1971). It is not surprising that theglomerular size is larger in adult lampreys than in larvae,as the olfactory epithelial surface area and the diameter ofthe olfactory nerve fascicles also increase during meta-morphosis (Vandenbossche et al., 1997).

In conclusion, there are spatially conserved glomerularterritories and substances that may be relevant to OSN

activity which are distributed in a consistent manner.Serotonin-IR fibers are adjacent to specific glomerularmodules and the G-protein Golf is expressed by lampreyOSNs; however, a subpopulation of OSNs do not expressGolf. These data are the first to show that expression ofGolf, the GTP-binding protein linked to olfactory receptors,is present at the base of gnathostome radiation.

ACKNOWLEDGMENTS

The authors thank the staff at Department of Fisheriesand Oceans, Sault Ste Marie, Ontario, Canada; HammondBay Biological Station, Millersburg, MI, and the U.S. Fishand Wildlife Service, Marquette Biological Station, Mar-quette, MI. for animal collection; and assistance from Dr.S.S. Yun (Michigan State University, Dept. of Fisheriesand Wildlife), A. Belanger, C. Dakhil, and D.W. Dingler(University of Windsor, Ontario, Canada).

LITERATURE CITED

Abogadie FC, Bruch RC, Farbman AI. 1995. G-protein subunits expressedin catfish olfactory receptor neurons. Chem Senses 20:199–206.

Baier H, Korsching S. 1994. Olfactory glomeruli in the zebrafish form aninvariant pattern and are identifiable across animals. J Neurosci 14:219–230.

Belluscio L, Gold GH, Nemes A, Axel R. 1998. Mice deficient in G(olf) areanosmic. Neuron 20:69–81.

Berghard A, Buck LB. 1996. Sensory transduction in vomeronasal neu-rons: evidence for G alpha o, G alpha i2, and adenylyl cyclase II asmajor components of a pheromone signaling cascade. J Neurosci 16:909–918.

Berghard A, Dryer L. 1998. A novel family of ancient vertebrate odorantreceptors. J Neurobiol 37:383–392.

Chiba A, Sohn YC, Honma Y. 1996. Immunohistochemical and ultrastruc-tural characterization of the terminal nerve ganglion cells of the Ayu,Plecoglossus altivelis (Salmoniformes, Teleostei). Anat Rec 246:549–556.

Danciger E, Mettling C, Vidal M, Morris R, Margolis F. 1989. Olfactorymarker protein gene: its structure and olfactory neuron-specific expres-sion in transgenic mice. Proc Natl Acad Sci U S A 86:8565–8569.

DellaCorte C, Restrepo D, Menco BP, Andreini I, Kalinoski DL. 1996. Galpha 9/G alpha 11: immunolocalization in the olfactory epithelium ofthe rat (Rattus rattus) and the channel catfish (Ictalurus punctatus).Neuroscience 74:261–273.

Eisthen HL. 2002. Why are olfactory systems of different animals sosimilar? Br Behav Evol 59:273–293.

Freitag J, Beck A, Ludwig G, von Buchholtz, Breer H. 1999. On the originof the olfactory receptor family: receptor genes of the jawless fish(Lampetra fluviatilis). Gene 226:165–174.

Friedrich RW, Korsching SI. 1998. Chemotopic, combinatorial, and non-combinatorial odorant representations in the olfactory bulb revealedusing a voltage-sensitive axon tracer. J Neurosci 18:9977–9988.

Greer CA. 1991. Structural organization of the olfactory system. In: Getch-ell TV, editor. Smell and taste in health and disease. New York: RavenPress. p 65–81.

Gundersen HJG, Bagger P, Bendtesen TF, Evans SM, Korbo L, MarcussenN, Moller A, Nielsen K, Nyengaard JR, Pakkenberg B, Sorense FB,Vesterby A, West MJ. 1988. The new stereological tools: dissector,fractionator, nucleator and point sampled intercepts and their use indiagnosis and diagnosis. APMIS 96:857–881.

Halasz N, Greer CA. 1993. Terminal arborizations of olfactory nerve fibersin the glomeruli of the olfactory bulb. J Comp Neurol 337:307–316.

Halpern M, Shapiro LS, Jia C. 1995. Differential localization of G proteinsin the opossum vomeronasal system. Brain Res 17:157–161.

Hansen A, Anderson KT, Finger TE. 2001. Immunohistochemical andultrastructural identification of G-proteins in the olfactory epitheliumof catfish. Chemical Senses 26:1128.

Hara TJ, Zhang C. 1998. Topographic bulbar projections and dual neuralpathways of the primary olfactory neurons in salmonid fishes. Neuro-science 82:301–313.

Hildebrand JG, Shepherd GM. 1997. Mechanisms of olfactory discrimina-


tion: converging evidence for common principles across phyla. AnnuRev Neurosci 20:595–631.

Iwahori N, Kiyota E, Nakamura KA. 1987. Golgi study on the olfactorybulb in the lamprey, Lampetra japonica. J Neurosci Res 5:126–139.

Jia C, Halpern M. 1996. Subclasses of vomeronasal receptor neurons:differential expression of G proteins (Gi alpha 2 and G(o alpha)) andsegregated projections to the accessory olfactory bulb. Brain Res 719:117–128.

Jones DT, Reed RR. 1989. Golf: an olfactory neuron specific-G proteininvolved in odorant signal transduction. Science 244:790–795.

Juilfs DM, Fulle HJ, Zhao AZ, Houslay MD, Garbers DL, Beavo JA. 1997.A subset of olfactory neurons that selectively express cGMP-stimulatedphosphodiesterase (PDE2) and guanylyl cyclase-D define a unique ol-factory signal transduction pathway. Proc Natl Acad Sci U S A 94:3388–3395.

Kafitz KW, Lenders-Zufall T, Zufall F, Greer CA. 2000. Cyclic GMP evokedcalcium transients in olfactory receptor cell growth cones. Neuroreport11:677–681.

Klenoff JR, Greer CA. 1998. Postnatal development of olfactory receptorcell axonal arbors. J Comp Neurol 390:256–267.

Li W, Sorensen PW. 1995. The olfactory system of migratory adult sealamprey (Petromyzon marinus) is specifically and acutely sensitive tounique bile acids released by conspecific larvae. J Gen Physiol 105:569–587.

Lin DM, Ngai J. 1999. Development of the vertebrate main olfactorysystem. Curr Opin Neurobiol 9:74–78.

McLean JH, Shipley MT. 1987. Serotonergic afferents to the rat olfactorybulb. I. Origins and laminar specificity of serotonergic inputs in theadult rat. J Neurosci 7:3016–3028.

Meisami E, Sendera TJ. 1993. Morphometry of rat olfactory bulbs stainedfor cytochrome oxidase reveals that the entire population of glomeruliforms early in the neonatal period. Brain Res Dev Brain Res 71:253–257.

Mezler M, Fleisher J, Conzelmann S, Korchi A, Widmayer P, Breer H,Boekhoff I. 2001. Identification of a nonmammalian G(olf) subtype:functional role in olfactory signaling of airborne odorants in Xenopuslaevis. J Comp Neurology 439:400–410.

Mombaerts P. 1996. Visualizing an olfactory sensory map. Cell 87:675–686.

Mombaerts P. 1999. Seven-transmembrane proteins as odorant and che-mosensory receptors. Science 286:707–711.

Morita Y, Finger TE. 1998. Differential projections of ciliated and microvil-lous olfactory receptor cells in the catfish, Ictalurus punctatus. J CompNeurol 398:539–550.

Northcutt RG, Puzdrowski RL. 1988. Projections of the olfactory bulb andnervus terminalis in the silver lamprey. Brain Behav Evol 32:96–107.

Philpot BD, Jazaeri AA, Brunjes PC. 1994. The development of serotoner-gic projections to the olfactory bulb of Monodelphis domestica (the greyshort tailed opossum). Brain Res Dev Brain Res 77:265–270.

Pinching AJ, Powell TP. 1971. The neuropil of the glomeruli of the olfactorybulb. J Cell Sci 9:347–377.

Pombal MA, de Arriba MC, Sampedro C, Alvarez R, Megias M. 2000.Immunocytochemical localization of calretinin in the olfactory systemof the adult lamprey, Lampetra fluviatilis. Brain Res Bull. 57:281–283.

Ring G, Mezza RC, Schwob JE. 1997. Immunohistochemical identificationof discrete subsets of rat olfactory neurons and the glomeruli that theyinnervate. J Comp Neurol 388:415–434.

Royet JP, Souchier C, Jourdan F, Playe H. 1988. Morphometric study of theglomerular population in the mouse olfactory bulb: numerical densityand size distribution along the rostrocaudal axis. J Comp Neurol 270:559–568.

Schandar M, Laugwitz KL, Boekhoff I, Kroner C, Gudermann T, Shultz G,Breer H. 1998. Odorants selectively activate distinct G protein sub-types in olfactory cilia. J Biol Chem 273:16669–16677.

Schober W. 1964. Vergleichend-anatomische untersuchungen am gerhirnder larven un adulten tiere von Lampetra fluviatilis (Linne, 1758) undLampetra planeri (Bloch 1784). J Hirnforschung 7:107–209.

Schober A, Mayer DL, von Bartheld CS. 1994. Central projections of thenervus terminalis and the nervus praeopticus in the lungfish brainrevealed by nitric oxide synthase. J Comp Neurol 349:1–19.

Shinohara H, Kato K, Asano T. 1992. Differential localization of G pro-teins, Gi and Go, in the olfactory epithelium and the main olfactorybulb of the rat. Acta Anat (Basel) 144:167–171.

Thommesen G. 1978. The spatial distribution of odour induced potentialsin the olfactory bulb of char and trout (Salmonidae). Acta PhysiolScand 102:205–217.

Tobet SA, Chickering TW, Sower SA. 1996. Relationship of gonadotropin-releasing hormone (GnRH) neurons to the olfactory system in develop-ing lamprey (Petromyzon marinus). J Comp Neurol 376:97–111.

VanDenbossche J, Seelye JG, Zielinski BS. 1995. The morphology of theolfactory epithelium in the larval, juvenile and upstream migrantstages of the sea lamprey, Petromyzon marinus. Brain Behav Evol45:19–24.

VanDenbossche J, Youson H, Pohlman D, Wong E, Zielinski BS. 1997.Metamorphosis of the olfactory organ of the sea lamprey (Petromyzonmarinus L.): morphological changes and morphometric analysis. J Mor-phol 231:41–52.

von Bartheld CS, Mayer DL. 1988. Central projections of the nervusterminalis in lampreys, lungfishes, and bichirs. Brain Behav Evol32:151–159.

Wachowiak M, Cohen LB. 2001. Representation of odorants by receptorneuron input to the mouse olfactory bulb. Neuron 32:723–735.

Wekesa KS, Anholt H. 1999. Differential expression of G proteins in themouse olfactory system. Brain Res 837:117–126.

Wensley CH, Stone DM, Baker H, Kauer JS, Margolis FL, Chikaraishi DM.1995. Olfactory marker protein mRNA is found in axons of olfactoryreceptor neurons. J Neurosci 15:4827–4837.

Wright DE, Demski LS. 1996. Organization of GnRH and FMRF-amidesystems in two primitive bony fishes (Order Polypteriformes). BrainBehav Evol 47:267–278.

Xu F, Greer CA, Shepherd GM. 2000. Odor maps in the olfactory bulb.J Comp Neurol 422:489–495.

Zhang X, Firestein S. 2002. The olfactory receptor gene superfamily of themouse. Nat Neurosci 5:124–133.

Zielinski BS, Moretti N, Hua HN, Zaidi AU, Bisaillon AD. 2000. Seroto-nergic nerve fibers in the primary olfactory pathway of the larval sealamprey, Petromyzon marinus. J Comp Neurol 420:324–334.


glomerular territories in the olfactory bulb from the...

Documents