chemical organization of the macaque monkey olfactory bulb: ii. … · 2017. 9. 6. · monkey...

Chemical Organization of the MacaqueMonkey Olfactory Bulb: II. Calretinin,

Calbindin D-28k, Parvalbumin, andNeurocalcin Immunoreactivity

JOSE R. ALONSO,* JESUS G. BRINON, CARLOS CRESPO, IGNACIO G. BRAVO,

ROSARIO AREVALO, AND JOSE AIJON

Departamento de Biologıa Celular y Patologıa, Facultad de Medicina. Universidad deSalamanca, Campus Miguel de Unamuno, Salamanca E-37007, Spain

ABSTRACTThe distribution and morphologic features of calcium-binding protein- (calbindin D-28k,

calretinin, neurocalcin, and parvalbumin) immunoreactive elements were studied in themacaque monkey olfactory bulb by using specific antibodies and the avidin-biotin-immunoperoxidase method. A characteristic laminar pattern of stained elements was ob-served for each marker. Scarce superficial short-axon cells and superficial stellate cellsdemonstrated calbindin D-28k immunoreactivity in the outer layers, whereas a moderatenumber of calbindin D-28k–immunoreactive granule cells and scarce deep short-axon cellswere observed in the inner layers. Calretinin-staining demonstrated abundant periglomeru-lar cells and granule cells and a scarce number of other interneuronal populations. Mostneurocalcin-immunopositive elements were external and medial tufted cells and periglo-merular cells, although other scarcer interneuronal populations were also immunostained. Afew superficial and deep short-axon cells as well as small interneurons in the externalplexiform layer were the only elements immunoreactive to parvalbumin. The distribution ofthe immunoreactive elements in the olfactory bulb of the macaque monkey showed a highsimilarity to that reported in the human, whereas it demonstrated a different and simplerpattern to what has been reported in the olfactory bulb of macrosmatic animals. It suggestsmore homogeneous calcium-mediated cell responses after stimulation that could be corre-lated to the lower capability to modulate olfactory signals in microsmatic animals. In addi-tion, these results indicate that experimental models in rodents do not provide an accurateestimation of calcium-binding protein-immunoreactive neuronal populations in the primateolfactory system. J. Comp. Neurol. 432:389–407, 2001. © 2001 Wiley-Liss, Inc.

Indexing terms: calcium-binding protein; microsmatic; primate; olfaction

Calbindin D-28k (CB), calretinin (CR), neurocalcin(NC), and parvalbumin (PV) are all members of the EF-hand homologue family, a group of calcium-binding pro-teins (CaBPs), which bind intracellular calcium with dis-sociation constants in the micromolar range (Persechini etal., 1989). These four CaBPs are widely expressed in dif-ferent neuronal populations in the peripheral and centralnervous systems (Taylor, 1974; Baimbridge and Miller,1982; Celio, 1990; Resibois and Rogers, 1992, Hidaka andOkazaki, 1993). They may contribute to the regulation offree calcium concentration through their ability to com-plex it, and thereby promote or restrict the actions of thiscation. They may also facilitate the diffusion of intracel-lular calcium, thus, modulating calcium gradients inside

neurons (Kretsinger et al., 1982; Heizmann and Hunziker,1991).

The basic structural components of the olfactory bulb (OB)are remarkably constant among vertebrates (Allison, 1953).However, there are significant variations in its neuronal

Grant sponsor: Spanish DGESeIC; Grant number: PB99-0068; Grantsponsor: CICyT; Grant number: 1FD97-1325; Grant sponsor: “Junta deCastilla y Leon.”

*Correspondence to: Jose R. Alonso, Dpto. Biologıa Celular y Patologıa,Universidad de Salamanca, Salamanca E-37007, Spain.E-mail: [email protected]

Received 18 May 2000; Revised 10 January 2001; Accepted 10 January2001

THE JOURNAL OF COMPARATIVE NEUROLOGY 432:389–407 (2001)

© 2001 WILEY-LISS, INC.

types, intrinsic circuitry, and target zones between macros-matic and microsmatic animals (Takagi, 1981, 1984, 1986;Mestre et al., 1992). The chemoarchitecture of the OB alsodemonstrates a considerable interspecies variability forsome markers (Takeuchi et al., 1982; Baker, 1986; Matsu-tani et al., 1989; Baker et al., 1991; Davis, 1991; Alonso et al.,1995, 1998), whereas other neurochemically identified pop-ulations are similar in microsmatic and macrosmatic mam-mals (Sanides-Kohlrausch and Wahle, 1991; Bassett et al.,1992). There is an extensive knowledge concerning the neu-ronal circuitry in the OB of macrosmatic animals, whereasless information is available in microsmatic animals, espe-cially in primates.

Antibodies against CR, CB, PV, and NC are valuableneuroanatomic tools in the characterization of specificneuronal and fiber populations in the brain, because cellspositive for each of them are frequently stained in a Golgi-like way, showing an extensive detail of the dendritic andaxonal processes. The distribution and morphologic char-acteristics of PV-, CB-, CR-, and NC-immunoreactive ele-ments have been studied in the OB of the rat by usingradioimmunoassay, immunocytochemistry, and/or in situhybridization (CB: Baimbridge and Miller, 1982; Baim-bridge et al., 1982; Feldman and Christakos, 1983; Garcıa-Segura et al., 1984; Halasz et al., 1985; Sequier et al.,1988, 1990; Celio 1989, 1990; Brinon et al., 1992, 1998a,1999; Alonso et al., 1993; PV: Celio, 1990; Kosaka et al.,1994; CR: Jacobowitz and Winsky, 1991; Resibois andRogers, 1992; NC: Bastianelli et al., 1993; Bastianelli andPochet, 1995; Porteros et al., 1996). In addition, the dis-tribution of CB and NC in the hedgehog OB, an animalwith an extraordinary development of the olfactory struc-tures, revealed important interspecies differences withprevious results in rat, including new neuronal types(Alonso et al., 1995; Bravo, 1997).

In the present study, we describe the distribution pat-tern and morphologic characteristics of the neurons andprocesses immunoreactive for each of these four CaBPs inthe macaque monkey OB. Although the specific functionsof these CaBPs remain to be elucidated, their expressionin particular neuronal types may provide insight into theroles subserved by each neuronal type in the OB circuitry.

MATERIAL AND METHODS

Animals and tissue preparation

Four adult male cynomolgus monkeys (Macaca fascicu-laris), each weighing between 2.7 and 3.8 kg, five adultmale rhesus monkeys (Macaca mulatta), each weighing

between 4.7 and 5.3 kg, and five adult male pig-tail mon-keys (Macaca nemestrina), each weighing between 4.5 and6.5 kg, were used in the present study. The brains of theseanimals were used in other neuroanatomic studies, notrelated with the olfactory system. The OBs were donatedby different groups working in the neuroanatomy of otherparts of the brain (see Acknowledgments section). In allcases, animal manipulations were performed according toNIH guidelines and subjected to authorization from thepertinent ethical committees.

The cynomolgus monkeys were tranquilized withketamine-HCl (10 mg/kg, i.m.) and deeply anesthetizedwith sodium pentobarbital (50 mg/kg, i.p.). After rinsingthe vascular tree with 500 ml of 0.9% saline, the followingfixatives were transcardially perfused: 4% paraformalde-hyde in 0.1 M sodium acetate buffer, pH 6.5 (250 ml/minfor 5 minutes and 100 ml/min for 15 minutes), and 4%paraformaldehyde in 0.1 M sodium borate buffer, pH 9.5(100 ml/min for 30 minutes). The OBs were dissected outand post-fixed in the same fixative for 6 hours at 4°C. Therhesus monkeys and pig-tail monkeys were anesthetizedwith Nembutal and ketamine-HCl (10 mg/kg, i.m.). Thevascular tree was rinsed with 0.8-1 liters of saline, fol-lowed by the perfusion of 4-5 liter of 4% paraformaldehydein 0.1 M phosphate buffer, pH 7.3 (PB). After perfusion,the brains were removed and the OBs were dissected outand stored in the same fixative for 2–3 hours at 4°C. Aftercareful washing, the OBs were cryoprotected in either 30%sucrose or 20% glycerol with 2% dimethylsulfoxide in PBfor 1 day. The OBs were frozen with cooled isopentane(Rosene et al., 1986) and stored at 270°C until furtherprocessing. Each OB was cut in series at 30 mm in thecoronal or sagittal plane by using a freezing, sliding mic-rotome.

The immunohistochemical procedure was carried out onfree-floating sections and involved the use of the avidin-biotin-peroxidase complex method (Hsu et al., 1981). Thesections were preincubated in 0.01% Triton X-100 in PBfor 2 hours at room temperature. Thereafter, the sectionswere incubated with the following primary antibodies: arabbit polyclonal antiserum against human CR, mousemonoclonal antibody against chicken CB, mouse monoclo-nal antibody against carp PV, or rabbit polyclonal anti-serum against bovine NC. The sections were incubated inprimary antibody diluted (PV, 1:1,000; CB, 1:1,000; CR,1:10,000; NC, 1:8,000) in PB for 48 hours at 4°C. Theseantibodies have been fully characterized (CB, clone 300:Celio et al., 1990; PV, clone 235: Celio et al., 1988; CR:Rogers, 1987, 1989; NC: Nakano et al., 1992) and widelyused in different studies of the central nervous system (seeCelio, 1990; Resibois and Rogers, 1992; Rogers and Resi-bois, 1992; Bastianelli et al., 1993). The sections werewashed in PB and incubated with biotinylated anti-mouse(PV and CB) or anti-rabbit (CR and NC) immunogamma-globulin (Vectastain ABC kit, Vector Laboratories, Bur-lingame, CA) diluted 1:250, for 2 hours at 20°C, and thenin Vectastain ABC reagent diluted 1:250, for 2 hours.Tissue-bound peroxidase was visualized by using 0.05%3,39-diaminobenzidine (DAB; Sigma) and 0.003% hydro-gen peroxide in 0.1 M Tris-HCl buffer (pH 7.6) for 5–10minutes, under visual control.

Additional series stained with 0.25% thionin or withNADPH-diaphorase histochemistry (see Alonso et al.,1998) were used to help in the localization of immunore-active elements. All sections were dehydrated in an in-

Abbreviations

CaBP calcium-binding proteinCB calbindin D-28kCR calretininDAB 3,3’ diaminobenzidineEPL external plexiform layerGCL granule cell layerGL glomerular layerIPL internal plexiform layerMCL mitral cell layerNC neurocalcinOB olfactory bulbONL olfactory nerve layerPB phosphate buffer 0.1 M, pH 7.3PV parvalbuminWM white matter

390 J.R. ALONSO ET AL.

creasing ethanol series, mounted in Entellan (Merck), andstudied under the light microscope.

The specificity of the immunostaining procedure wasassessed by preadsorption of the immune serum with thecorresponding purified protein it was raised against. Falsepositives were left out by omitting the specific antibody inthe first incubation step, and by staining some sectionswith chromogen and hydrogen peroxide, to rule out theputative interference by endogenous peroxidase. Neitherresidual immunoreactivity nor reaction product formationwas found in any of the controls.

Analysis

Sections were analyzed by using Zeiss Axiophot andZeiss III photomicroscopes. Immunoreactive neurons andfibers were traced by using a camera lucida drawing tube.Original images were captured with an Olympus digitalcamera system, transferred to a Power Macintosh com-puter and adjusted with Adobe Photoshop 5.5 software(Adobe Systems, Inc., Mountain View, CA). Only size andgeneral luminosity were modified and the lettering wasadded. The resulting image files were then assembled intomontages by using Canvas 5.0.2 software (Deneba Sys-tems, Miami, FL).

For the quantitative analysis, only the OBs from Ma-caca nemestrina were used because series of these sectionswere simultaneously processed in independent vials. In-dividual labeled cells were drawn with a camera lucida byusing a 1003 oil immersion objective and plotted on adigitizer tablet connected to a semiautomatic image anal-ysis system (MOP-Videoplan, Kontron), and their diame-ters were measured. For the measurement of weakly la-beled neurons, only cells where the nucleus was clearlyvisible were used. In the darkly stained neuronal types,the chosen neurons were those located in the middle of thethickness of the section and showing at least two primarydendrites. For each neuronal type, 300 cells from all fiveanimals were measured, with the exception of those neu-ronal types that were rarely observed (1 in Table 1). Inthis case, all observable cells, always over 40 neurons,were measured.

RESULTS

The OB of the macaque monkey is a laminar structurewhere seven layers can be differentiated. The classifica-tion and nomenclature developed by Ramon y Cajal (1904)

for the OB of macrosmatic mammals can be used, becausethe same layers are present and their boundaries areeasily discernible. According to this classification, frommost external to most internal, the following layers aredistinguished: olfactory nerve layer (ONL), glomerularlayer (GL), external plexiform layer (EPL), mitral celllayer (MCL), internal plexiform layer (IPL), granule celllayer (GCL), and white matter (WM).

General characteristics of theimmunostaining

For each of the antibodies used in this study, the mor-phologic characteristics and distribution of labeled ele-ments in the macaque monkey OB were consistent acrossthe animals studied. Neuronal populations with specificdistributions across the macaque monkey OB layers wereimmunostained for CB, CR, PV, or NC. Similar resultswere obtained in the OB of the three species of monkeysstudied, Macaca fascicularis, Macaca mulatta, and Ma-caca nemestrina. Light microscopic examination of immu-nostained sections of the macaque monkey OB revealedpositive somata, dendritic arborizations, axons, and ax-onic collaterals with a laminated distribution. No immu-noreactivity was found in the controls.

Most positive neurons were stained in a “Golgi-like”way, and the brown DAB reaction product was observed asa dense precipitate that masked the cell nucleus and al-lowed the visualization of cell bodies, dendrites, axon-likeprocesses, and occasionally, dendritic spines. These mor-phologic details usually made easy the characterization ofthe immunoreactive neurons.

The morphologic characteristics and arborization pat-terns of the positive cells were analyzed in coronal andsagittal sections. A high heterogeneity of labeled neuronswas observed for all four CaBPs. Most stained cells couldbe identified as belonging to some of the neuronal typespreviously described in the OB by using Golgi impregna-tion, immunocytochemical, and histochemical techniques.Nevertheless, the morphology and location of other la-beled neuronal subpopulations made difficult their iden-tification with the neuronal types described in the OB ofthe macaque monkey or in other species.

Calbindin D-28k

After CB immunocytochemistry, positive elements wereobserved in all layers of the monkey OB (Fig. 1), except inthe ONL where neither cells nor the olfactory fibers dem-onstrated immunostaining (Fig. 2a). In the GL, a reducednumber of immunolabeled cells were observed surround-ing the glomeruli. These CB-immunopositive cells weremore abundant in the deep zone of the GL, close to theEPL, and at the GL/EPL boundary (Fig. 2b,c). Positiveneurons showed a high heterogeneity in their size andmorphology, and they could be classified into two differentneuronal types. Most CB immunolabeled cells in the GLwere medium in size (12–16 mm in maximum diameter)with round or fusiform cell bodies and intensely immuno-stained. They showed three to five thin aspiny dendritescoursing in the boundary between the GL and the EPL. Inboth coronal and sagittal sections, these dendrites dis-played radial orientation. They ramified in the superficialregion of the EPL and in the periglomerular region, butthey did not enter within the glomeruli. According to thesemorphologic features they were typified as superficial stel-late cells (Fig. 2b).

TABLE 1. Relative Cell Density of Each Immunoreactive Neuronal Type1

Cell CB CR NC PV

Periglomerular cells 2 1111 1 2External tufted cells 2 2 1111 2Inverted monopolar cells 2 11 1 2Superficial short-axon cells 11 2 2 11Superficial stellate cells 1 2 2 2Pyriform cells of the EPL 11 11 2 1Fusiform cells of the EPL 11 11 2 2Middle tufted cells 2 2 11 2Horizontal cells 11 11 1 2Vertical cells of Cajal 11 11 1 2Deep stellate cells 1 1 2 2Deep short-axon cells 2 2 1 1Granule cells 111 1111 11 2

1Relative cell density of each neuronal type demonstrating immunoreactivity to theanalyzed CaBPs in 30-mm thick sections. CB, calbindin D-28k; CR, calretinin; NC,neurocalcin; PV, parvalbumin; EPL, external plexiform layer. 2, no immunoreactivity;1, less than 5 cells per section; 11, 5–25 cells per section; 111, 25–100 cells persection; and 1111, more than 100 cells per section.

391CALCIUM-BINDING PROTEINS IN THE MONKEY OB

Scarce larger neurons (18–22 mm in maximum diame-ter) with heterogeneous morphology and intense immuno-reactivity were observed in the same region. They had

fusiform or ovoid somata with two to five thick dendritesoriented parallel to the OB layers. Dendrites coursed inthe EPL/GL boundary surrounding the inner portion of

Fig. 1. Camera lucida drawing of neuronal elements demonstrat-ing CB immunoreactivity in the macaque monkey OB. Profiles ofolfactory glomeruli and cell bodies of mitral cells are delineated bydashed lines. a: Superficial stellate cell. b: Superficial short-axon cell.

c: Pyriform cell of the EPL. d: Fusiform cell of the EPL. e: Horizontalcell. f–k: Granule cells. l: Vertical cell of Cajal. m,n: Deep stellatecells. For abbreviations, see list. Scale bar 5 100 mm.


the glomeruli and could be followed for long distances.They ramified in several branches restricted to the peri-glomerular region, i.e., no intraglomerular dendrites werefound (Fig. 2c). According to their size, morphology, and

dendritic pattern, these cells were identified as superficialshort-axon cells (Fig. 2c). In a few of these large cells, thedendritic pattern was different. They had primary den-drites coursing parallel to the GL/EPL boundary and an-

Fig. 2. Immunoreactive elements to CB in the macaque monkeyOB. a: Low-power photomicrograph showing the predominance ofCB-immunoreactive elements in the deep layers of the OB. b: Super-ficial stellate cell close to two glomeruli (G). c: Superficial short-axon

cell, extending its dendrites in parallel to the inner side of two glo-meruli (G). d,e: Fusiform cells of the EPL. f,g: Pyriform cells of theEPL (arrows in g). For abbreviations, see list. Scale bars 5 50 mm ina–g.


other dendrite that coursed toward the EPL. This dendritecrossed occasionally the MCL, branching in the superficialportion of the GCL.

In the EPL, only a few cells, scattered throughout theEPL and in the GL/EPL boundary, were CB immunoreac-tive. Two different neuronal populations were distin-guished: the first one included fusiform neurons (11–13mm in maximum diameter) with two primary dendritesdirected in opposite directions and oriented parallel orperpendicular to the OB layers (Fig. 2d,e). They did notextend for long distances, and did not cross the innerboundary of the EPL. Dendrites of the cells placed in theouter third of the layer coursed in the periglomerularregion of the GL but did not enter within the glomeruli.Dendritic spines were occasionally observed.

The second population was composed of pyriform cells(12–15 mm in maximum diameter) (Fig. 2f,g). These cells

had a primary dendrite with different orientations: di-rected toward either the superficial or the deep portion ofthe EPL or parallel to the OB layers. The primary den-drite ramified in several branches restricted to the EPL.In a few cells, one or two short thinner dendrites arosefrom the opposite side of the cell body. These two neuronalpopulations located in the EPL were named pyriform andfusiform cells of the EPL.

In the MCL, mitral cells did not show CB immunoreac-tivity. In the IPL, a few CB-immunopositive, medium(11–15 mm in maximum diameter) fusiform neurons wereobserved. They had two dendrites arising from oppositepoles of the somata and coursing parallel to the OB layersin coronal sections. They were identified as horizontalcells (Fig. 3a). In a few of them, the proximal axonic regionwas observed coursing in the IPL, parallel to the OBlayers.

Fig. 3. Immunoreactive elements to CB in the inner layers of themacaque monkey OB. a: Horizontal cell (arrow) among positive pro-files of granule cells. b: Immunopositive granule cells in the MCL andin the superficial region of the GCL, extending their apical dendrites

toward the EPL. c: Vertical cell of Cajal (arrow) in the IPL. d: Deepstellate cell (arrow) in the central region of the GCL. For abbrevia-tions, see list. Scale bars 5 50 mm in a–d.


In the GCL, most CB-immunoreactive neurons weresmall (6–8 mm in maximum diameter), round or ovoidcells with a main dendrite extending radially to the EPLand one to three thin dendrites directed deeply, towardthe GCL. Morphologically identical neurons were ob-served in the IPL, MCL, and in the deep region of the EPL(Fig. 3b). Their primary dendrites extended throughoutthe EPL and arborized repeatedly in this layer. Brancheswere covered with large pedunculated spines. They wereidentified as granule cells.

Another large (19–22 mm in maximum diameter) andintensely stained neuronal population was observed in theouter portions of the GCL. Cells belonging to this popula-tion had pyriform cell bodies with a thick primary den-drite directed toward the EPL and a thinner secondaryone, arising from the opposite side of the cell body, di-rected toward the deep portions of the GCL (Fig. 3c).Primary dendrites crossed the IPL and the MCL, reachingthe EPL, whereas secondary dendrites ramified close tothe cell body. The location, morphology, and dendriticpattern of these positive neurons allowed us to identifythem as vertical cells of Cajal.

In the inner region of the GCL, and in the WM, a fewcells with spherical somata shape (10–15 mm in maximumdiameter) demonstrated CB immunoreactivity. Two to fivethin aspiny dendrites arose from the somata, giving thema bipolar or radial aspect in both sagittal and coronalsections. The size and morphologic characteristics weresimilar to the superficial stellate cells; therefore, wenamed them as deep stellate cells (Fig. 3d).

A few immunoreactive fibers were found coursingthrough the WM. They entered from the olfactory pedun-cle into the GCL as bundles located in the dorsal andventral regions of the OB. Because the projecting neurons,mitral, and tufted cells, were CB immunonegative, thesestained fibers were identified as centrifugal fibers.

Calretinin

The highest density of positive elements in the monkeyOB was obtained after CR immunocytochemistry (Fig. 4).The ONL displayed a moderate density of stained fibersalthough some axonal bundles in this layer also remainedunstained (Fig. 5a). Both stained and unstained axonsappeared to be segregated at least when entering theglomeruli. Hence, there were some glomeruli innervatedby CR-stained olfactory axons, and, thus, they displayedstrongly stained neuropil, whereas some other innervatedby unstained olfactory axons showed a lightly stainedneuropil (Fig. 5b). In all other layers, especially in the GLand in the GCL, numerous CR-immunopositive elementswere observed (Fig. 5a).

A large number of CR-immunolabeled neurons was ob-served in the GL. These cells had the typical location andmorphologic features of periglomerular cells. Their smallcell bodies (7–10 mm of maximum diameter) surroundedthe glomeruli by the inner side (Fig. 5a,b). They had asingle dendrite that entered and branched within the glo-merulus, forming a dense dendritic field. In a few labeledperiglomerular cells, two primary dendrites arose fromopposite sides of the cell body, innervating one or twoadjacent glomeruli. This heterogeneity allowed us to dis-tinguish three different populations of periglomerularcells: monodendritic, bidendritic monoglomerular, and bi-dendritic biglomerular cells. CR-immunostaining in theglomeruli was composed by the dendritic arborizations of

the stained periglomerular cells, as well as, in some ofthem, by stained olfactory nerve fibers.

Another CR-immunostained neuronal population waspresent at the GL/EPL boundary. The size (7–10 mm inmaximum diameter) and shape were similar to those ofperiglomerular cells, but their dendritic pattern was dif-ferent. They showed one long smooth dendritic trunk ori-ented toward the EPL, crossing this layer and reachingthe MCL. They were called inverted monopolar cells (Fig.5c). Scattered in the EPL, scarce CR-immunostained neu-rons with similar size (11–14 mm in maximum diameter)and with morphologic features similar to those identifiedafter CB immunostaining as pyriform and fusiform cells ofthe EPL were observed (Fig. 5d).

In the MCL, mitral cells were not stained for CR. Belowthis layer, in the IPL, there was a small population ofhorizontal cells with fusiform shape and sizes rangingfrom 12 to 15 mm. They had two dendrites coursing in theIPL parallel to the OB layering.

A large population of immunolabeled granule cells wereobserved in the GCL as well as through the IPL and theMCL. The highest density of positive cells was found inthe outer portion of the GCL, close to the IPL (Fig. 5e).Their primary dendrites could be followed for long dis-tances through the EPL where they created a dense net-work of spiny immunolabeled dendrites. The secondarydendrites ramified in the GCL, close to the cell body.

Another morphologically different neurons showed CRimmunoreactivity in the GCL. According to their morpho-logic characteristics and location, different neuronal typescould be distinguished. The first group was composed ofpyriform cells, similar to those CB-immunopositive verti-cal cells of Cajal. The cell body gave rise to a thick primarydendrite directed to the EPL. A secondary spiny dendrite,arising from the opposite side of the cell body extendedthrough the deep region of the GCL.

The second population was located in the inner portionsof the GCL and in the WM. They showed variable somaticmorphologies (mostly fusiform or spherical) and differentdendritic patterns (from two to five dendrites with vari-able orientations) similar to the deep stellate cells identi-fied after CB immunostaining. In the WM, most of themhad only one or two stained dendrites with variable ori-entations (Fig. 5f,g).

Neurocalcin

Immunoreactivity against NC in the macaque monkeyOB demonstrated positive elements throughout all layers(Fig. 6), except the ONL which was devoid of immuno-staining (Fig. 7a).

The GL showed the most intense NC immunoreactivity(Fig. 7a). All glomeruli exhibited immunopositive neuropiland displayed stained cells surrounding them, whichcould be ascribed to two neuronal types. The largest pop-ulation was formed by intensely stained cells of round orpyriform shape, with maximum diameter ranging from 9to 15 mm, placed in the inner side of the glomeruli. Theyshowed one smooth and thick dendrite that entered intoone glomerulus where ramified giving rise to a character-istic tuft-like dendritic tree (Fig. 7b). According to thesemorphologic characteristics, they were identified as exter-nal tufted cells.

The second neuronal population positive for NC in theGL showed the characteristics of periglomerular cells.They were small (5–8 mm maximum diameter) sphericalneurons, faintly stained, located between adjacent glomer-


uli and in the internal side of the GL. They had thin spinydendrites ramifying inside the glomeruli.

Most NC-immunoreactive elements in the EPL werelocated close to the border with the GL. Two types of

positive neurons could be discerned in this layer. The firstone was identified as middle tufted cells and was formedby weakly stained neurons with triangular shape andmaximum diameters ranging from 15 to 19 mm, located in

Fig. 4. Camera lucida drawing of neuronal elements demonstrat-ing CR immunoreactivity in the macaque monkey OB. Profiles ofolfactory glomeruli and cell bodies of mitral cells are delineated bydashed lines. a–c: Periglomerular cells. d: Inverted monopolar cell.

e,f: Fusiform cells of the EPL. g,h: Pyriform cells of the EPL.i–m: Granule cells. n: Horizontal cell, o: Vertical cell of Cajal.p,q: Deep stellate cells. For abbreviations, see list. Scale bar 5 100 mm.


Fig. 5. Immunoreactive elements to CR in the macaque monkeyOB. a: Low-power photomicrograph showing the distribution of CR-immunopositive elements. b: CR-immunopositive periglomerularcells surrounding olfactory glomeruli. Note some glomeruli inner-vated by positive fibers (asterisks) and the adjacent ones innervated

by negative olfactory axons. c: Inverted monopolar cells (arrows) inthe border between the GL and the EPL. d: Interneurons in the EPL.e: Abundant positive granule cells in the outer half of the GCL.f,g: Deep stellate cells in the inner half of the GCL. For abbreviations,see list. Scale bars 5 50 mm in a–g.


Fig. 6. Camera lucida drawing of neuronal elements demonstrat-ing NC immunoreactivity in the macaque monkey OB. Profiles ofolfactory glomeruli and cell bodies of mitral cells are delineated bydashed lines. a–c: External tufted cells. d–f: Periglomerular cells.

g: Inverted monopolar cell. h: Middle tufted cells. i: Horizontal cell.j: Vertical cell of Cajal. k: Granule cell. l–n: Deep short-axon cells. Forabbreviations, see list. Scale bar 5 100 mm.

the external third of the EPL (Fig. 7c). They showed athick and smooth dendritic trunk directed to a singleglomerulus and accessory thinner dendrites coursing par-allel to the GL/EPL boundary.

The second group of NC-immunopositive elements in theEPL was less frequent than the former. They were roundneurons with maximum diameters ranging from 10 to 12 mmand strongly stained. Their somata were located in the outerregion of the EPL, and displayed one smooth dendrite ori-ented toward the deeper layers, reaching the MCL where itdiscretely ramified. These neurons were identified as in-

verted monopolar cells (Fig. 7d), because they were similar tothose cells observed after CR immunostaining.

Another group of NC-immunoreactive elements wasfound in the innermost third of the EPL, and occasionallyin the MCL and IPL. They demonstrated round or fusi-form shapes, with maximum diameters ranging from 6.5to 9.5 mm. A main dendrite was visible, at least in itsproximal region, coursing toward the EPL. Thinner sec-ondary dendrites arose from the basal portion of the somaand ramified close to it. According to this description,these neurons were identified as granule cells (Fig. 7e).

Fig. 7. Immunoreactive elements to NC in the macaque monkeyOB. a: Low-power photomicrograph showing the distribution of NC-immunopositive elements. b: NC-immunopositive external tuftedcells with their dendrites arborizing inside the glomeruli. c: NC-immunopositive middle tufted cell, extending its principal dendrite(arrow) toward one glomerulus and an axon-like process (arrowhead)

toward the EPL. d: Inverted monopolar cell (arrow) in the borderbetween the GL and the EPL, extending its dendrite (arrowhead)toward the deep region of the EPL. e: Displaced granule cell (arrow)located in the MCL, extending its apical dendrite (arrowheads) to-ward the superficial region of the EPL. For abbreviations, see list.Scale bars 5 50 mm in a–e.


In the MCL and the IPL, scarce positive elements wereobserved. Mitral cells were immunonegative to NC, leav-ing their unstained cell bodies distinguishable over thebackground staining. Two populations of oriented neuronscould be described as immunoreactive. The first one wasformed by round or fusiform cells with maximum diame-ters ranging from 9 to 13 mm. They showed two mainvaricose dendrites arising from opposite poles of the somaand coursing parallel to the OB lamination. These positiveneurons were identified as horizontal cells (Fig. 8a).

The second group of positive elements in this layerdisplayed round cell bodies with maximum diameters be-tween 8 and 12 mm. Two dendrites arose from oppositepoles of the soma and coursed radially to the OB layering.The apical dendrites reached occasionally the internalregion of the EPL without branching, although they weremainly restricted to the MCL border. Their basal pro-cesses were short and ramified close to the soma. Theseneurons were classified as vertical cells of Cajal (Fig. 8a).

In the GCL and in the WM, the neuropil staining wasmore intense than in the superficial layers. In fact, theIPL was clearly delimited by the immunonegative somataof the mitral cells and the more intensely stained neuropilof the GCL, where rows of negative granule cells wereclearly distinguishable. Strongly immunopositive ele-ments in these layers were rare and mainly placed in thedeep portions of the GCL and throughout the WM. Theywere typified as deep short-axon cells, showing variablemorphologies and dendritic branching patterns (Fig. 8b,c).They were mainly fusiform- or round-shaped, with maxi-mum diameters between 12 and 16 mm. Commonly, onlythe cell body and the varicose initial segments of thedendrites arising from opposite poles of the somata werestained.

Parvalbumin

PV immunostaining demonstrated the lowest number ofimmunoreactive elements from all four CaBPs studied.Only a few cells, mainly located in the GL and in the EPL,were PV immunolabeled (Fig. 9). Neither the olfactoryfibers nor centrifugal fibers showed PV immunoreactivity.

In the GL and in the superficial portion of the EPL,some PV-immunoreactive neurons were distinguished.

These cells were large (19–24 mm in maximum diameter),darkly stained, and showed a heterogeneous morphology,with round, oval, or fusiform cell bodies. Most of themwere located in the inner border of the GL or in theGL/EPL boundary, with two to five thick dendrites cours-ing in the periglomerular region, but no intraglomerulardendrites were observed. In coronal and sagittal sections,the dendritic trees were observed extending parallel to theOB layers. Occasionally, they had a long varicose dendritedirected toward the inner layers, reaching the IPL and theGCL. These large neurons were identified, according totheir size and morphology, as superficial short-axon cells(Fig. 10a–c).

The second PV-immunoreactive neuronal population ofthe EPL was formed by small cells (8–10 mm in maximumdiameter) with similar features to those immunoreactivefor other CaBPs and called “pyriform cells of the EPL.”They had a primary dendrite with variable orientationsand some other thin prolongations arising from the oppo-site side of the somata (Fig. 10c,d).

Mitral cells and all other elements in the MCL were PVimmunonegative. In the internal layers (IPL, GCL, andWM), only a few neurons demonstrated PV immunostain-ing. Their somatic morphology and dendritic branchingpattern was heterogeneous. They had usually large roundor ovoid cell bodies (17–21 mm in maximum diameter), andshowed weakly stained dendrites, so only their proximalregions could be observed. According to their size andlocation, they were identified as deep short-axon cells (Fig.10e,f).

DISCUSSION

This study provides a description of the distribution andmorphologic characteristics of CB-, CR-, PV-, and NC-immunoreactive neurons and fibers in the macaque mon-key OB. The main results are that antibodies against eachCaBP demonstrated a characteristic laminated patternwith specific immunostained neuronal types and thatthese labelings are different in primates from previousreports in macrosmatic mammals. There are constantpoints of convergence and divergence with the descrip-tions of distribution and frequency of appearance of ele-

Fig. 8. Immunoreactive elements to NC in the deep layers of the macaque monkey OB. a: Immu-nopositive vertical cell of Cajal and horizontal cell in the MCL. b,c: Deep short-axon cells with differentsomatic shapes in the inner half of the GCL. For abbreviations, see list. Scale bars 5 25 mm in a–c.


ments positive to these CaBPs, resulting that none of thefour proteins studied displays a similar expression patternin the macaque monkey and in macrosmatic mammals(rodents and insectivores).

Identification of immunoreactive elements

Among the different neuronal types found in the macaquemonkey OB after immunohistochemistry for the four inves-

Fig. 9. Camera lucida drawing of neuronal elements demonstrating PV-immunoreactivity in themacaque monkey OB. Profiles of olfactory glomeruli and cell bodies of mitral cells are delineated bydashed lines. a,b: Superficial short-axon cells. c,d: Pyriform cells of the EPL. e,f: Deep short-axon cells.For abbreviations, see list. Scale bar 5 100 mm.


Fig. 10. Elements immunoreactive to PV in the macaque monkeyOB. a,b: PV-immunoreactive superficial short-axon cells, extendingtheir dendrites along the inner side of some glomeruli (G). c: PV-immunostained elements in the superficial EPL. Superficial short-axon cell (arrow) and pyriform cell of the EPL (arrowhead). d: Pyri-

form cell of the EPL in the deep half of this layer. e: Group ofPV-immunoreactive deep short-axon cells in the GCL. f: Multipolardeep short-axon cell with the characteristic multipolar shape in theinner region of the GCL. For abbreviations, see list. Scale bars 5 50mm in a–f.


tigated CaBPs, some of them, including periglomerular cells,tufted cells, superficial short-axon cells, and granule cellswere easily identified on the basis of previous descriptions(Schneider and Macrides, 1978; Halasz, 1990). However, themorphologic features of other immunostained cells were notconsistent with any of the classic neuronal types described inthe OB. Nevertheless, there is a progressive increase in thenumber and typology of short-axon cells not previously de-scribed by means of silver-impregnation methods, detectedas new histochemical and cytochemical tools become avail-able and new species are studied.

The superficial and deep stellate cells have been describedin the macaque monkey OB after NADPH-diaphorase histo-chemistry (Alonso et al., 1998). The main differential fea-tures of superficial stellate cells with respect to the superfi-cial short-axon cells is their smaller cell body size and theabsence of dendritic spines (Alonso et al., 1998). The deepstellate cells were described as displaying round, ovoid, orpyriform medium cell bodies with several thin dendritesarising from different regions of the somata, giving them astellate aspect. The features and location of both cell typesare consistent with those found in our material.

Other interneuronal types observed in the EPL of themacaque monkey were identified as inverted monopolarcells, pyriform cells of the EPL, and fusiform cells of the EPLon the basis of their similar features and location to identi-cally named neuronal types previously described in thehedgehog OB after either NADPH-diaphorase staining orCB immunostaining (Alonso et al., 1995). According to thesedescriptions, inverted monopolar cells were located at theGL-EPL border, with a small cell body, spherical in shape,and with a characteristic single unbranched dendrite with astraight course toward the deep part of the EPL (Alonso etal., 1995). All these features were found in the invertedmonopolar cells of the monkey with the exception that theywere CB negative but CR and NC immunopositive.

Fusiform cells of the EPL are characterized by a smallfusiform cell body with two dendrites arising from oppo-site poles and oriented radially toward the external andinternal boundaries of the EPL (Alonso et al., 1995). Neu-rons identified as fusiform cells of the EPL in the monkeyOB displayed similar features, although, occasionally,their dendrites were oriented parallel to the OB layering.

Concerning the pyriform cells of the EPL, they had alsobeen described in the hedgehog OB as located in the mid-dle region of the EPL, with pyriform cell bodies and anevident dendritic trunk arborizing close to the cell body(Alonso et al., 1995). Neurons identified as pyriform cellsof the EPL in the macaque monkey OB closely resembledthose of the hedgehog, although they often show addi-tional thinner dendrites.

Classically described interneurons of the deep layers com-prise several morphologically different cell types. Neuronsidentified in the monkey OB as horizontal cells and verticalcells of Cajal were consistent with descriptions of Schneiderand Macrides (1978) for these cell types in the hamster OBas well as for those reported in the OB of other mammaliangroups such as rodents and insectivora (Halasz, 1990; Alonsoet al., 1995). More problematic, however, was the exact iden-tification of another population of immunostained cells lo-cated in the GCL and the WM. Their general morphologyand location resembled those of deep short-axon cells de-scribed after Golgi impregnation in the hamster OB (Schnei-der and Macrides, 1978). However, it is unclear whetherthey correspond to one or more of the classes of deep short-axon cells, because these interneurons comprise a heteroge-

neous population, differing in their size, morphology, posi-tion, connections, and chemical content (Halasz, 1990). Thelarge immunopositive cells in the GCL may correspond toBlanes cells or Golgi cells. The main distinctive feature ofeach neuronal type is the presence or absence of dendriticspines (Schneider and Macrides, 1978), a morphologic char-acteristic that was not always distinguishable in our immu-nostained material. Therefore, it was difficult to ascribe theobserved deep-short axon cells to one of the classes includedin this neuronal type.

Interspecies differences in theimmunostaining

The olfactory nerve layer. We only observed immu-nopositive olfactory fibers for CR, whereas they were nega-tive for either CB, NC, or PV. In the rat, similar results havebeen previously reported for CB (Garcıa-Segura et al., 1984),PV (Kosaka et al., 1994), and NC (Brinon et al., 1998a),although in the case of CR, no fiber staining was described inthe ONL (Jacobowitz and Winsky, 1991; Resibois and Rog-ers, 1992). In addition, there are some descriptions of thepresence of a small number of CB-positive olfactory fasciclesin the rat (Celio, 1990) and human OB (Ohm et al., 1991),and, despite olfactory fibers have been reported to be CRimmunonegative in rat (Jacobowitz and Winsky, 1991; Resi-bois and Rogers, 1992), a similar pattern of CR staining tothat found in the primate OB was reported in the hedgehog(Bravo, 1997). Because, at least in rodents, CB, CR, and NCare expressed by a wide number of olfactory receptors (Bas-tianelli et al., 1995), the lack of immunoreactivity in thedistal olfactory fibers may indicate the differential intracel-lular distribution and role of these CaBPs. Nevertheless,although there are abundant data demonstrating the heter-ogeneity of the olfactory fibers (Akeson, 1988), the location,characteristics, and functional significance of these stainedfibers remain unknown.

The output neurons. CaBPs are commonly expressedin subpopulations of local circuit neurons in the brain(Celio, 1990), and so it is in the OB. The four proteinsstudied in the macaque monkey OB are not expressed inmitral cells, the true output neurons of this structure(Ramon y Cajal, 1904; Allison, 1953; Macrides et al.,1985). This fact is in partial agreement with the distribu-tion pattern for these CaBPs in other species. Mitral cellsin the rat OB express neither CB (Celio, 1990; Brinon etal., 1992) nor PV (Celio, 1990; Kosaka et al., 1994) nor NC(Brinon et al., 1998a). In the hedgehog, which shows adramatic increase in complexity of olfactory related areas,mitral cells are immunonegative for CB (Alonso et al.,1995) and for NC (Bravo, 1997). By contrast, a smallpopulation of mitral cells expresses CR in the rat OB,although this labeling seems to be weak and relativelyvariable (Jacobowitz and Winsky, 1991; Resibois and Rog-ers, 1992; Wouterlood and Hartig, 1995; Brinon et al.,1997). Therefore, no expression of any of the four proteinsis detected in mitral cells in our material, as it occurs, withthe exception of CR, in macrosmatic mammals.

Tufted cells are a heterogeneous neuronal population thatincludes true projecting neurons, neurons presumably in-volved in interbulbar connections, and local circuit neurons(Schoenfeld et al., 1985). In macrosmatic mammals, thesegroups of tufted cells can be discerned by means of theirposition, dendritic tree pattern, and input and output con-nections (Orona et al., 1984; Macrides et al., 1985; Schoen-feld et al., 1985). Tufted cells in the rat OB have been de-scribed as immunonegative for PV (Kosaka et al., 1994), CR


(Wouterlood and Hartig, 1995) and, despite some contro-versy, for CB (Garcıa-Segura et al., 1984; Celio, 1990; Brinonet al., 1992). In the case of NC, immunopositive external,medial, and internal tufted cells have been reported in therat and in the hedgehog OB (Bravo, 1997; Brinon et al.,1998a). In the macaque OB, the location and morphologicfeatures of NC-positive tufted cells are consistent with thoseof external and middle tufted cells.

The periglomerular cells. Subsets of periglomerularcells are positive in our material for CR and NC, whereasthey are negative for PV and CB. By contrast, periglo-merular cells are positive in the rat OB for all of the fourstudied proteins. Periglomerular cells are positive for CBin human and in rat, where 26% of these neurons expressthis protein (Halasz et al., 1985; Ohm et al., 1991; Brinonet al., 1992). PV is also present in an important subgroupof periglomerular cells in the rat OB (Kosaka et al., 1994).Almost 50% of periglomerular cells in rat contain CR(Jacobowitz and Winsky, 1991; Resibois and Rogers,1992). Finally, an important group of periglomerular cellsexpresses NC in the rat OB (Crespo et al., 1997; Brinon etal., 1998a), and in the hedgehog OB (Bravo, 1997).

Periglomerular cells constitute a neuronal populationrather homogeneous according to morphologic criteria, byusing the Golgi technique and conventional electron mi-croscopy (Pinching and Powell, 1971). However, studies ontheir biochemistry have revealed a high chemical hetero-geneity, making it possible to ascertain different subsetsof periglomerular cells according to the presence, accumu-lation, and/or expression of CaBPs, taurine, tyrosine hy-droxylase, cholecystokinin, methionine-enkephalin, soma-tostatin, gamma-aminobutyric acid, chlorine, or NADPH-diaphorase activity (Mugnaini et al., 1984; Halasz et al.,1985; Seroogy et al., 1985, 1989; Gall et al., 1987; Sakai etal., 1987; Halasz, 1990; Vincent and Kimura, 1992; Alonsoet al., 1993, 1995; Siklos et al., 1995; Kosaka et al., 1995,1997, 1998). Moreover, there are several different degreesof overlapping between these subgroups of periglomerularcells. In the rat OB, there is no colocalization of CR andCB, or NADPH-diaphorase and CB in the same periglo-merular cell (Rogers, 1992; Rogers and Resibois, 1992;Alonso et al., 1993); coexpression of CB and NC is veryrare; approximately 20% of PV-immunopositive periglo-merular cells colocalize with 2.6% of NC-immunopositiveperiglomerular cells (although percentages vary acrossrostrocaudal and ventrolateral axes); and approximately40% of NC-positive periglomerular cells coincide with 15%of CR-positive periglomerular cells (Brinon et al., 1999).According to the expression patterns of CaBPs in themacaque OB, only colocalization of CR and NC is expect-able in the periglomerular cells. These proteins exhibit thehighest percentage of coexpression in the periglomerularcells of the rat OB. This high content of calcium-chelatorsin periglomerular cells could correlate with the fact thatthese neurons are targets of excitatory impulses comingfrom olfactory fibers, output neurons, and short-axon cells.The presence of these molecules could enable the neuronto bear the sustained elevation of cytosolic calcium con-centration and to modulate its output, shortening thepostsynaptic potentials, in addition to other putative col-lateral actions. Differential expression of CaBPs in peri-glomerular cells might also account, on the other hand, forvariability in their behavior, which would generate diver-sity in the modulation of the primary response of outputneurons to the outcoming odorant signals.

The short-axon cells. Superficial short-axon cells arepositive after CB and PV immunodetection, and preferen-tially located in the outer third of the EPL of the macaqueOB. This is in agreement with previous reports about thepresence of superficial short-axon cells expressing CB in therat OB (Seroogy et al., 1989; Celio, 1990) and in the hedge-hog OB (Alonso et al., 1995), about the presence of superficialshort-axon cells expressing PV in the rat OB (Kosaka et al.,1994; Toida et al., 1996) and the human OB (Ohm et al.,1990), and about the absence of superficial short-axon cellsexpressing NC in the rat and hedgehog OB (Bravo, 1997;Brinon et al., 1998a). However, a subset of superficial short-axon cells containing CR is present in the rat OB (Resiboisand Rogers, 1992), but absent in the macaque OB.

In the EPL, immunopositive elements display charac-teristic morphologies, some of them not observed in ro-dents and/or insectivores. Thus, neurons identified as su-perficial stellate cells are immunopositive to CB. Thisneuronal type has not been described in macrosmatic spe-cies, although it is also observed after NADPH-diaphorasestaining in the macaque OB (Alonso et al., 1998). Invertedmonopolar neurons, described as CB positive in the hedge-hog OB (Alonso et al., 1995), are immunopositive for CRand NC in the macaque monkey EPL. Other elements inthis layer have been described as fusiform cells of theEPL, positive to CB and CR, and pyriform cells of the EPL,positive to CB, CR, and PV. These data account again fora wide morphologic and biochemical variability of the in-terneuronal types in the EPL of the primate OB.

Neuronal populations identified as short-axon cells in theinner layers of the macaque OB are positive for all the fourproteins studied. Some of them can be identified as belong-ing to different subsets of oriented neurons. Thus, horizontalcells, located in the IPL, are immunopositive for CB, CR,and/or NC; and vertical cells of Cajal in the IPL and the GCLcontain CB and/or NC in our material. In the rat and thehedgehog OB, previous results are comparable to ours in thecase of CB (Celio, 1990; Brinon et al., 1992; Alonso et al.,1995) and NC (Bravo, 1997; Brinon et al., 1998a). Neverthe-less, no CR-immunopositive horizontal cells have been ob-served in the rat OB, whereas they are present in the ma-caque OB. By contrast, horizontal and vertical cells of Cajalhave been found as PV immunopositive in the rat OB (Ko-saka et al., 1994), but they were immunonegative in themacaque OB. In the deep layers, short-axon cells were im-munopositive for the four proteins studied. In the case of CB,they can be identified as similar to those described asNADPH-diaphorase positive in the macaque OB, andtermed deep stellate cells (Alonso et al., 1998). They couldalso somehow relate to stellate multipolar neurons located inthe inner part of the GCL or in the WM, immunopositive forCB in the hedgehog OB (Alonso et al., 1995). No cell withthese characteristics is positive after CB immunodetection inthe rat or human OB (Celio, 1990; Ohm et al., 1991; Brinonet al., 1992; Alonso et al., 1995). By contrast, there areneither giant nor large cells positive for CB in the macaqueOB, as is the case in the rat and human OB, respectively(Celio, 1990; Ohm et al., 1991; Brinon et al., 1992). Becausethese studies were carried out by using similar immuno-staining procedures and the same antibodies, we considerthat these differences reflect true interspecies differences,rather than variations due to distinct methodologic ap-proaches.

The granule cells. Granule cells have been describedas positive for NC in the hedgehog OB (Bravo, 1997), andfor CR, PV, and NC in the rat OB (Jacobowitz and Winsky,


1991; Rogers and Resibois, 1992; Kosaka et al., 1994;Brinon et al., 1998a). The putative expression of CB ingranule cells is controversial. Garcıa-Segura et al. (1984)and Rogers and Resibois (1992) in the rat OB, and Ohm etal. (1991) in the human OB, describe elements similar togranule cells as positive for this CaBP, whereas these datado not agree with those of other authors (Baimbridge andMiller, 1982; Celio, 1990; Brinon et al., 1992; Alonso et al.,1995) who did not find CB immunoreactivity in the gran-ule cells. In the monkey OB, subsets of granule cells can beidentified as CB, CR, and NC immunopositive between theinnermost portion of the EPL and the outermost IPL. ForCB and CR, this immunoreactivity is also presentthroughout the IPL and the external half of the GCL.

The heterogeneous intracellular distribution of CaBPsin neurons is a rule rather than a exception. An exampleis the uneven distribution of PV in Purkinje cells (Kosakaet al., 1993) or CR in mitral cells (Wouterlood and Hartig,1995). The distributions of these proteins appear to becorrelated with the location of different sets of channels indistinct cellular compartments that might require partic-ular types of CaBPs for the maintenance of cell metabo-lism. Although DAB labeling does not allow a clear sub-cellular localization, immunostaining of distal dendritesand spines was observed in granule cells after either CB orCR immunohistochemistry, whereas the NC immunoreac-tion product was restricted to the cell body and the prox-imal region of their dendritic shafts, being the distal partsas well as the dendritic spines unobservable, suggesting adifferent intracellular distribution of these markers. Inother brain regions, the buffer type CaBPs such as CBhave been proposed to be involved in the compartmenta-tion of calcium within spines where N-methyl-D-aspartatereceptors associated to calcium channels exist (Pickel andHeras, 1996). In the case of NC, it has a myristoyl consen-sus sequence, which may constitute a calcium-myristoylswitch, which allows NC to bind cell membranes in acalcium-dependent manner (Zozulya and Stryer, 1992).This suggests that anchoring to cell membranes may befunctionally relevant and that this protein develops itsrole locally, associated to specific cell membrane domains(Sanada et al., 1996). Although we have no direct evi-dence, given the large number of immunostained granulecells with CB, CR, and NC, the possible coexpression ofsome of them in the same cell cannot be ruled out. In thiscase, the differential expression of CaBPs would revealsubcellular compartments within granule cells with par-ticular calcium-handling requirements. Nevertheless, ifthe expression of these CaBPs occurs in segregated sets ofgranule cells, as was described in the rodent OB (Rogersand Resibois, 1992; Brinon et al., 1999), they could iden-tify different populations of granule cells on the basis oftheir synaptic relationships with particular groups of neu-rons in the EPL or with different centrifugal inputs.

The heterogeneity for granule cell differential expres-sion of CaBPs contrasts with the fact that GABA is themain neurotransmitter for most granule cells (Ribak etal., 1977; Kosaka et al., 1985, 1987; Gall et al., 1987).Thus, being complex, the pool of neuroactive substancesknown to be present in granule cells is more restrictedthan in other OB cell types. This diversity might be trans-lated to a differential modulation of the dendrodendriticsynapses between granule cells and mitral cells. It hasbeen proposed that granule cells would not need actionpotentials to inhibit mitral cells (Shepherd, 1981). Thisinhibitory response would then affect exclusively the

mitral/tufted cell that has excited the granule cell (feed-back inhibition) and not to their neighbors (lateral inhibi-tion) (Scott et al., 1993). The presence of a CaBP at one ofthese dendrodendritic synapses would then affect the abil-ity of the granule cell to develop autoinhibition of themitral/tufted cell, preventing this way the dominance ofthe most active output neuron in every region. Therefore,differential expression of CaBPs in granule cells may sug-gest not only biochemical divergences between differentsubsets of this neuronal type but different physiologicresponses of different granule cell subsets.

Functional implications

CaBPs modulate the actions of calcium by complexing it(Heizmann and Hunziker, 1991; Ikura, 1996). The seques-tering of calcium causes depletion on its availability, thusdown-regulating the mechanisms triggered by increasesin its concentration. Buffering CaBPs such as CB, PV, andCR would shorten calcium oscillations, preserving the cellfrom damage due to a sustained rise in this cation concen-tration (Baimbridge et al., 1992; Andressen et al., 1993)and/or affect the amplitude of calcium entrance and itstiming. This shifts the generation of action potentialsmodulating the neuronal response (Keinanen et al., 1990;Trombley and Westbrook, 1990; Martin et al., 1993). Inaddition, CaBPs can also control the behavior of cells by adifferential interaction with some effectors in the presenceof threshold calcium concentration (Ikura, 1996), so theyare able to modify the cell dynamics by affecting secondmessengers systems and/or ion channels (Dizhoor et al.,1991, 1993; Zozulya and Stryer, 1992). In the case of NC,its putative action on phosphorylation of ionic channelscould regulate the response and firing pattern of the neu-ron expressing it (Zozulya and Stryer, 1992; Jonas andKackmarek, 1996; Moss and Smart, 1996).

In the OB, particular CaBPs are expressed by definitesubsets of neurons in a constant and conservative manner,although there are not only interspecies differences butalso differences between cells not discernible by means ofmorphology or presence of other neuroactive substances.This clearly points to a segregation of roles subserved foreach CaBP into the cell machinery, leading to variation incell responses modulating the olfactory signal. Hence, amore complex pattern regarding the expression of theseCaBPs could indicate a higher variability in the calcium-mediated response to the olfactory stimulus with morepossibilities to modulate it. Our results demonstrate thatthe distribution pattern of the analyzed CaBPs in themonkey OB is qualitatively and quantitatively simplerthan that reported for the OB of macrosmatic animals,with the only exception of CR, which shows a similardistribution pattern. This lower presence of biochemicallydistinct neuronal subtypes is specially evident in the caseof periglomerular cells. In addition, there is a reducednumber of neurons expressing PV in the monkey OB,whereas in macrosmatic mammals most interneurons inthe EPL express it. The existence of a limited number ofbiochemically different neuronal populations in the mon-key OB compared with rodents and insectivores wouldindicate more homogeneous calcium-mediated responsesof neurons to olfactory stimuli. This could be correlatedwith a more restricted capability to modulate the olfactorysignals and to the notable differences in the developmentand physiologic role that the olfactory system has betweenmicrosmatic and macrosmatic animals.


Available data on the expression of CB and PV in thehuman OB (Ohm et al., 1990, 1991) demonstrate a highlysimilar distribution pattern of elements containing thesemarkers with those described in the present study, butclearly different to those reported from macrosmatic mam-mals. Other markers such as NADPH-diaphorase also re-vealed a high degree of similarity in the distribution andmorphologic features of the positive elements in human(Brinon et al., 1998b) and the macaque monkey OB (Alonsoet al., 1998) and clear differences to those found in the OB ofmacrosmatic mammals. These data and our results indicatethat experimental models based on rodents do not provide anaccurate estimation of the organization of different neuronalpopulations in the primate olfactory system. Therefore, thepresent study provides baseline information about the orga-nization of CaBPs-containing neurons in the primate OBthat can be useful in the studies of human diseases affectingthis sensory system.

ACKNOWLEDGMENTS

The authors thank Dr. C. Cavada, Dr. E. Rausell, andDr. D.G. Amaral for kindly providing some of the macaquemonkey OBs used in this study.

LITERATURE CITEDAkeson RA. 1988. Primary olfactory neuron subclasses. In: Margolis FL,

Getchell TV, editors. Molecular neurobiology of the olfactory system.New York: Plenum Press. p 297–318.

Allison AC. 1953. The morphology of the olfactory system in the verte-brates. Biol Rev 28:195–244.

Alonso JR, Arevalo R, Porteros A, Brinon JG, Lara J, Aijon J. 1993.Calbindin D-28K and NADPH-diaphorase activity are localized in dif-ferent populations of periglomerular cells in the rat olfactory bulb.J Chem Neuroanat 6:1–6.

Alonso JR, Arevalo R, Garcıa-Ojeda E, Porteros A, Brinon JG, Aijon J.1995. NADPH-diaphorase active and calbindin D-28k-immunoreactiveneurons and fibers in the olfactory bulb of the hedgehog (Erinaceuseuropaeus). J Comp Neurol 351:307–327.

Alonso JR, Porteros A, Crespo C, Arevalo R, Brinon JG, Weruaga E, Aijon J.1998. Chemical anatomy of the macaque monkey olfactory bulb: NADPH-diaphorase/nitric oxide synthase activity. J Comp Neurol 402:419–434.

Andressen C, Blumcke I, Celio MR. 1993. Calcium-binding proteins: selec-tive markers of nerve cells. Cell Tissue Res 271:181–208.

Baimbridge KG, Miller JJ. 1982. Immunohistochemical localization ofcalcium-binding protein in the cerebellum, hippocampal formation andolfactory bulb of the rat. Brain Res 245:223–229.

Baimbridge KG, Miller JJ, Parkes CO. 1982. Calcium-binding proteindistribution in the rat brain. Brain Res 239:519–525.

Baimbridge KG, Celio MR, Rogers JH. 1992. Calcium-binding proteins inthe nervous system. Trends Neurosci 15:303–308.

Baker H. 1986. Species differences in the distribution of substance P andtyrosine hydroxylase immunoreactivity in the olfactory bulb. J CompNeurol 252:206–226.

Baker H, Abate C, Szabo A, Joh TH. 1991. Species-specific distribution ofaromatic L-amino acid decarboxylase in the rodent adrenal gland,cerebellum, and olfactory bulb. J Comp Neurol 305:119–129.

Bassett JL, Shipley MT, Foote SL. 1992. Localization of corticotropin-releasing factor-like immunoreactivity in monkey olfactory bulb andsecondary olfactory areas. J Comp Neurol 316:348–362.

Bastianelli E, Pochet R. 1995. Calmodulin, calbindin-D-28k, calretinin andneurocalcin in rat olfactory bulb during postnatal development. DevBrain Res 87:224–227.

Bastianelli E, Okazaki K, Hidaka H, Pochet R. 1993. Neurocalcin immu-noreactivity in rat olfactory bulb. Neurosci Lett 161:165–168.

Bastianelli E, Polans AS, Hidaka H, Pochet R. 1995. Differential distributionof six calcium-binding proteins in the rat olfactory epithelium duringpostnatal development and adulthood. J Comp Neurol 354:395–409.

Bravo IG. 1997. Distribucion de neurocalcina en el bulbo olfatorio demamıferos. Thesis of Licenciature, University of Salamanca.

Brinon JG, Alonso JR, Arevalo R, Garcıa-Ojeda E, Lara J, Aijon J. 1992.Calbindin D-28k-positive neurons in the rat olfactory bulb. Cell TissueRes 269:289–297.

Brinon JG, Alonso JR, Garcıa-Ojeda E, Crespo C, Arevalo R, Aijon J. 1997.Calretinin- and parvalbumin-immunoreactive neurons in the rat mainolfactory bulb do not express NADPH-diaphorase activity. J ChemNeuroanat 13:253–264.

Brinon JG, Arevalo R, Crespo C, Bravo IG, Okazaki K, Hidaka H, Aijon J,Alonso JR. 1998a. Neurocalcin immunoreactivity in the rat main olfac-tory bulb. Brain Res 795:204–214.

Brinon JG, Crespo C, Weruaga E, Alonso J, Sobreviela T, Aijon J, AlonsoJR. 1998b. NADPH-diaphorase/nitric oxide synthase-positive elementsin the human olfactory bulb. Neuroreport 9:3141–3146.

Brinon JG, Martınez-Guijarro FJ, Bravo IG, Arevalo R, Crespo C, OkazakiK, Hidaka H, Aijon J, Alonso JR. 1999. Coexpression of neurocalcinwith other calcium-binding proteins in the rat main olfactory bulb.J Comp Neurol 407:404–414.

Celio MR. 1989. Calcium binding proteins in the brain. Arch Ital AnatEmbryol 94:227–236.

Celio MR. 1990. Calbindin D-28k and parvalbumin in the rat nervoussystem. Neuroscience 35:375–475.

Celio MR, Baier W, Scharer L, de Viragh PA, Gerday C. 1988. Monoclonalantibodies directed against the calcium binding protein parvalbumin.Cell Calcium 9:81–86.

Celio MR, Baier W, Scharer L, Gregersen HJ, de Viragh PA, Norman AW.1990. Monoclonal antibodies directed against the calcium-binding pro-tein, calbindin D-28k. Cell Calcium 11:599–602.

Crespo C, Alonso JR, Brinon JG, Weruaga E, Porteros A, Arevalo R, AijonJ. 1997. Calcium-binding proteins in the periglomerular region of typ-ical and atypical olfactory glomeruli. Brain Res 745:293–302.

Davis BJ. 1991. NADPH-diaphorase activity in the olfactory system of thehamster and rat. J Comp Neurol 314:493–511.

Dizhoor AM, Ray S, Kumar S, Niemi G, Spencer M, Brolley D, Walsh KA,Philipov PP, Hurley JB, Stryer L. 1991. Recoverin; a calcium sensitiveactivator of retinal rod guanylate cyclase. Science 251:915–918.

Dizhoor AM, Chen C-K, Olshevskaya E, Sinelnikova VV, Phillipov P,Hurley JB. 1993. Role of the acylated amino terminus of recoverin inCa21-dependent membrane interaction. Science 259:829–832.

Feldman SC, Christakos S. 1983. Vitamin D-dependent calcium-bindingprotein in the rat brain: biochemical and immunocytochemical charac-terization. Endocrinology 112:290–302.

Gall CM, Hendry SHC, Seroogy KB, Jones EG, Haycock JW. 1987. Evi-dence for coexistence of GABA and dopamine in neurons of the ratolfactory bulb. J Comp Neurol 266:307–318.

Garcıa-Segura LM, Baetens D, Roth J, Norman AW, Orci L. 1984. Immu-nohistochemical mapping of calcium-binding protein immunoreactivityin the rat central nervous system. Brain Res 296:75–86.

Halasz N. 1990. The vertebrate olfactory system. Chemical neuroanatomy,function and development. Budapest: Akademiai Kiado.

Halasz N, Hokfelt T, Norman AW, Goldstein M. 1985. Tyrosine hydroxy-lase and 28k-vitamin D-dependent calcium binding protein are locatedin different subpopulations of periglomerular cells of the rat olfactorybulb. Neurosci Lett 61:103–107.

Heizmann CW, Hunziker W. 1991. Intracellular calcium-binding proteins:more sites than insights. Trends Biochem Sci 16:98–103.

Hidaka H, Okazaki K. 1993. Neurocalcin family: a novel calcium-binding proteinabundant in bovine central nervous system. Neurosci Res 16:73–77.

Hsu SM, Raine L, Fanger H. 1981. The use of avidin biotin peroxidasecomplex (ABC) in immunoperoxidase techniques. A comparison be-tween ABC and unlabeled antibody (peroxidase) procedures. J Histo-chem Cytochem 29:577–590.

Ikura M. 1996. Calcium binding and conformational response in EF-handproteins. Trends Biochem Sci 21:14–17.

Jacobowitz DM, Winsky L. 1991. Immunocytochemical localization of cal-retinin in the forebrain of the rat. J Comp Neurol 304:198–218.

Jonas EA, Kackmarek LK. 1996. Regulation of potassium channels byprotein kinases. Curr Opin Neurobiol 6:318–323.

Keinanen K, Wisden W, Sommer B, Werner P, Herb A, Verdoorn TA,Sakmann B, Seeburg PH. 1990. A family of AMPA-selective glutamatereceptors. Science 249:556–560.

Kosaka K, Heizmann CW, Kosaka T. 1994. Calcium-binding proteinparvalbumin-immunoreactive neurons in the rat olfactory bulb: 1. Distri-bution and structural features in adult rat. Exp Brain Res 99:191–204.

Kosaka K, Aika Y, Toida K, Heizmann CW, Hunziker W, Jacobowitz DM,Nagatsu I, Streit P, Visser TJ, Kosaka T. 1995. Chemically defined


neuron groups and their subpopulations in the glomerular layer of therat main olfactory bulb. Neurosci Res 23:73–88.

Kosaka K, Toida K, Margolis FL, Kosaka T. 1997. Chemically definedneuron groups and their subpopulations in the glomerular layer of therat main olfactory bulb: II. Prominent differences in the intraglomeru-lar dendritic arborization and their relationship to olfactory nerveterminals. Neuroscience 76:775–786.

Kosaka K, Toida K, Aika Y, Kosaka T. 1998. How simple is the organiza-tion of the olfactory glomerulus? The heterogeneity of so-called periglo-merular cells. Neurosci Res 30:101–110.

Kosaka T, Hataguchi Y, Hama K, Nagatsu I, Wu J-Y. 1985. Coexistence ofimmunoreactivities for glutamate decarboxylase and tyrosine hydrox-ylase in some neurons of the periglomerular region of the rat mainolfactory bulb: possible coexistence of gamma-aminobutyric acid(GABA) and dopamine. Brain Res 343:166–171.

Kosaka T, Kosaka K, Heizmann CW, Nagatsu I, Wu J-Y, Yanaihara N, HamaK. 1987. An aspect of the organization of the GABAergic system in the ratolfactory bulb: laminar distribution of immunohistochemically definedsubpopulations of GABAergic neurons. Brain Res 411:373–378.

Kosaka T, Kosaka K, Nakayama T, Hunziker W, Heizmann CW. 1993.Axons and axon terminals of cerebellar Purkinje cells and basket cellshave higher levels of parvalbumin immunoreactivity than somata anddendrites: quantitative analysis by immunogold labeling. Exp BrainRes 93:483–491.

Kretsinger RH, Mann JE, Simmons JG. 1982. Model of facilitated diffusionof calcium by the intestinal binding protein. In: Norman AW, SchaeferK, Herrath DV, Grigoleit H-G, editors. Vitamin D: chemical, biochem-ical and clinical endocrinology of calcium metabolism. Berlin: DeGruyter. p 233–248.

Macrides F, Schoenfeld TA, Marchand JE, Clancy AN. 1985. Evidence formorphologically, neurochemically and functionally heterogeneous classesof mitral and tufted cells in the olfactory bulb. Chem Senses 10:175–202.

Martin LJ, Blackstone CD, Levey A, Huganir RL, Price DL. 1993. AMPAglutamate receptor subunits are differentially distributed in rat brain.Neuroscience 53:327–358.

Matsutani S, Senba E, Tohyama M. 1989. Distribution of neuropeptide-likeimmunoreactivities in the guinea pig olfactory bulb. J Comp Neurol280:577–586.

Mestre N, Petter A, Bons N. 1992. Systematisation of the olfactory bulbefferent projections in a lemurian primate: Microcebus murinus. JHirnforsch 33:173–184.

Moss SJ, Smart TG. 1996. Modulation of amino acid-gated ion channels byprotein phosphorylation. Int Rev Neurobiol 39:1–52.

Mugnaini E, Wouterlood FG, Dahl AL, Oertel WH. 1984. Immunocyto-chemical identification of GABAergic neurons in the main olfactorybulb of the rat. Arch Ital Biol 122:83–113.

Nakano A, Terasawa M, Watanabe M, Usuda N, Morita T, Hidaka H. 1992.Neurocalcin, a novel calcium binding protein with three EF-hand do-mains, expressed in retinal amacrine cells and ganglion cells. BiochemBiophys Res Commun 186:1207–1211.

Ohm TG, Muller H, Ulfig N, Braak E. 1990. Glutamic-acid-decarboxylase-and parvalbumin-like-immunoreactive structures in the olfactory bulbof the human adult. J Comp Neurol 291:1–8.

Ohm TG, Muller H, Braak E. 1991. Calbindin-D-28K-like immunoreactivestructures in the olfactory bulb and anterior olfactory nucleus of thehuman adult: distribution and cell typology-partial complementaritywith parvalbumin. Neuroscience 42:823–840.

Orona E, Rainer EC, Scott JW. 1984. Dendritic and axonal organization ofmitral and tufted cells in the rat olfactory bulb. J Comp Neurol 226:346–356.

Persechini A, Moncrief ND, Kretsinger RH. 1989. The EF-hand family ofcalcium-modulated proteins. Trends Neurosci 12:462–467.

Pickel V, Heras A. 1996. Ultrastructural localization of calbindin-D28k andGABA in the matrix compartment of the rat caudate-putamen nuclei.Neuroscience 71:167–178.

Pinching AJ, Powell TPS. 1971. The neuron types of the glomerular layerof the olfactory bulb. J Cell Sci 9:305–345.

Porteros A, Brinon JG, Crespo C, Okazaki K, Hidaka H, Aijon J, Alonso JR.1996. Neurocalcin immunoreactivity in the rat accessory olfactory bulb.Brain Res 729:82–89.

Ramon y Cajal S. 1904. Textura del Sistema Nervioso del hombre y de losvertebrados. Madrid: Imprenta de Nicolas Moya.

Resibois A, Rogers JH. 1992. Calretinin in rat brain: an immunohistochem-ical study. Neuroscience 46:529–538.

Ribak CE, Vaughn JE, Saito K, Barber R, Roberts E. 1977. Glutamate decar-boxylase localization in neurons of the olfactory bulb. Brain Res 126:1–18.

Rogers JH. 1987. Calretinin: a gene for a novel calcium-binding proteinexpressed principally in neurons. J Cell Biol 105:1343–1353.

Rogers JH. 1989. Two calcium-binding proteins mark many chick sensoryneurons. Neuroscience 31:697–709.

Rogers JH. 1992. Immunohistochemical markers in rat cortex: co-localization of calretinin and calbindin-D28k with neuropeptides andGABA. Brain Res 587:147–157.

Rogers JH, Resibois A. 1992. Calretinin and calbindin-D28K in rat brain:patterns of partial co-localization. Neuroscience 51:843–865.

Rosene DL, Roy NJ, Davis BJ. 1986. A cryoprotection method that facili-tates cutting frozen sections of whole monkey brains for histologicaland histochemical processing without freezing artifact. J HistochemCytochem 34:1301–1315.

Sakai M, Yoshida M, Karasawa N, Teramura-Ito M, Nagatsu I. 1987.Coexistence of taurine-like and tyrosine hydroxylase-like immunoreac-tivities in some periglomerular cells of the rat main olfactory bulb. BiogAmines 4:457–464.

Sanada K, Shimizu F, Kameyama K, Haga K, Fukada Y. 1996. Calcium-bound recoverin targets rhodopsin kinase to membranes to inhibitrhodopsin phosphorylation. FEBS Lett 384:227–230.

Sanides-Kohlrausch C, Wahle P. 1991. Distribution and morphology ofsubstance P-immunoreactive structures in the olfactory bulb and olfac-tory peduncle of the common marmoset (Callithrix jacchus), a primatespecies. Neurosci Lett 131:117–120.

Schneider SP, Macrides F. 1978. Laminar distributions of interneurons inthe main olfactory bulb of the adult hamster. Brain Res Bull 3:73–82.

Schoenfeld TA, Marchand JE, Macrides F. 1985. Topographic organizationof tufted cells axonal projections in the hamster main olfactory bulb: anintrabulbar associational system. J Comp Neurol 235:503–518.

Scott JW, Wellis DP, Riggott MJ, Buonviso N. 1993. Functional organiza-tion of the main olfactory bulb. Microsc Res Tech 24:142–156.

Sequier JM, Hunzinker W, Richards G. 1988. Localization of calbindinD-28k mRNA in rat tissues by in situ hybridization. Neurosci Lett86:155–160.

Sequier JM, Hunziker W, Andressen C, Celio MR. 1990. Calbindin D-28kprotein and mRNA localization in the rat brain. Eur J Neurosci2:1118–1126.

Seroogy KB, Brecha N, Gall C. 1985. Distribution of cholecystokinin-likeimmunoreactivity in the rat main olfactory bulb. J Comp Neurol 239:373–383.

Seroogy KB, Hokfelt T, Buchan A, Brown JC, Terenius L, Norman AW,Goldstein M. 1989. Somatostatin like immunoreactivity in rat mainolfactory bulb: extent of coexistence with neuropeptide Y-, tyrosinehydroxylase- and vitamin D-dependent calcium binding protein-likeimmunoreactivities. Brain Res 496:389–396.

Shepherd GM. 1981. Synaptic and impulse loci in olfactory bulb dendriticcircuits. In: Roberts A, Bush BMH, editors. Neurones without im-pulses. Cambridge: Cambridge University Press. p 255–267.

Siklos L, Rickmann M, Joo F, Freeman WJ, Wolff JR. 1995. Chloride is prefer-entially accumulated in a subpopulation of dendrites and periglomerular cellsof the main olfactory bulb in adult rats. Neuroscience 1:165–172.

Takagi SF. 1981. Multiple olfactory pathways in mammals: a review.Chem Senses 6:329–333.

Takagi SF. 1984. The olfactory nervous system of the Old World monkey.Jpn J Physiol 34:561–573.

Takagi SF. 1986. Studies on the olfactory nervous system of the Old Worldmonkey. Prog Neurobiol 27:195–250.

Takeuchi Y, Kimura H, Sano Y. 1982. Immunohistochemical demonstra-tion of serotonin nerve fibers in the olfactory bulb of the rat, cat andmonkey. Histochemistry 75:461–471.

Taylor AN. 1974. Chick brain calcium-binding protein: comparison withintestinal vitamin-D induced calcium-binding protein. Arch BiochemBiophys 161:100–108.

Toida K, Kosaka K, Heizmann CW, Kosaka T. 1996. Electron microscopicserial-sectioning/reconstruction study of parvalbumin-containing neu-rons in the external plexiform layer of the rat olfactory bulb. Neuro-science 72:449–466.

Trombley PQ, Westbrook GL. 1990. Excitatory synaptic transmission incultures of rat olfactory bulb. J Neurophysiol 64:598–606.

Vincent SR, Kimura H. 1992. Histochemical mapping of nitric oxide syn-thase in the rat brain. Neuroscience 46:755–784.

Wouterlood FG, Hartig W. 1995. Calretinin-immunoreactivity in mitralcells of the rat olfactory bulb. Brain Res 682:93–100.

Zozulya S, Stryer L. 1992. Calcium-myristoil protein switch. Proc NatlAcad Sci USA 89:11569–11573.


chemical organization of the macaque monkey olfactory bulb: ii. … · 2017. 9. 6. · monkey...

Documents