Expression of neuron-specific markers by the vomeronasal neuroepithelium in six species of primates.код для вставкиСкачать
THE ANATOMICAL RECORD PART A 281A:1190 –1200 (2004) Expression of Neuron-Speciﬁc Markers by the Vomeronasal Neuroepithelium in Six Species of Primates JOHN C. DENNIS,1* TIMOTHY D. SMITH,2,3 KUNWAR P. BHATNAGAR,4 CHRISTOPHER J. BONAR,5 ANNE M. BURROWS,3,6 1 AND EDWARD E. MORRISON 1 Department of Anatomy, Physiology, and Pharmacology, College of Veterinary Medicine, Auburn University, Auburn, Alabama 2 School of Physical Therapy, Slippery Rock University, Slippery Rock, Pennsylvania 3 Department of Anthropology, University of Pittsburgh, Pittsburgh, Pennsylvania 4 Department of Anatomical Sciences and Neurobiology, University of Louisville School of Medicine, Louisville, Kentucky 5 Cleveland Metroparks Zoo, Cleveland, Ohio 6 Department of Physical Therapy, Duquesne University, Pittsburgh, Pennsylvania ABSTRACT Vomeronasal organ (VNO) morphology varies markedly across primate taxa. Old World monkeys display no postnatal VNO. Humans and at least some apes retain a vestigial VNO during postnatal life, whereas the strepsirrhines and New World Monkeys present a morphologically well-deﬁned VNO that, in many species, is presumed to function as an olfactory organ. Available microanatomical and behavioral studies suggest that VNO function in these species does not precisely duplicate that described in other mammalian taxa. The questions of which species retain a functional VNO and what functions they serve require inquiry along diverse lines but, to be functional, the vomeronasal epithelium must be neuronal and olfactory. We used immunohistochemistry to establish these criteria in six primate species. We compared the expression of two neuronal markers, neuron-speciﬁc ␤-tubulin (BT) and protein gene product 9.5, and olfactory marker protein (OMP), a marker of mature olfactory sensory neurons, in parafﬁn-embedded VNO sections from two strepsirrhine and four haplorhine species, all of which retain morphologically well-deﬁned VNOs during postnatal life. The infant Eulemur mongoz, adult Otolemur crassicaudatus, neonatal Leontopithicus rosalia, and adult Callithrix jacchus express all three proteins in their well-deﬁned vomeronasal neuroepithelia. The infant Tarsius syrichta showed some BT and OMP immunoreactivity. We establish that two strepsirrhine species and at least some New World haplorhines have mature sensory neurons in the VNO. In contrast, at all ages examined, Saguinus geoffroyi VNO expresses these markers in only a few cells. © 2004 Wiley-Liss, Inc. Key words: immunohistochemistry; olfaction; olfactory marker protein; primate; ␤-tubulin; PGP 9.5; vomeronasal The tetrapod vomeronasal organ (VNO) is bilaterally symmetric and lies along the ventrorostral aspect of the nasal septum. Sensory neurons in the vomeronasal neuroepithelium (VNNE) of snakes, rodents, and opossums detect chemical signals that evoke behavioral and/or physiological changes regarding prey identiﬁcation, social status, and reproductive state (Wysocki, 1979; Halpern, 1987; Takami, 2002; Halpern and Martı́nez-Marcos, 2003). The rodent VNO is the best studied in terms of gene expression during development, its morphological organization, its secondary and higher-order connections, and its effects on the organism’s behavior and physiology. For some time, the rodent VNO was thought to epitomize the © 2004 WILEY-LISS, INC. mammalian vomeronasal system but VNO function as exempliﬁed in rodents is not precisely replicated in other mammalian taxa, namely, ferrets (Weiler et al., 1999; *Correspondence to: John C. Dennis, Department of Anatomy, Physiology, and Pharmacology, College of Veterinary Medicine, Auburn University, Auburn, AL 36849. Fax: 334-844-4542. E-mail: firstname.lastname@example.org Received 20 May 2004; Accepted 1 July 2004 DOI 10.1002/ar.a.20124 Published online 7 October 2004 in Wiley InterScience (www.interscience.wiley.com). 1191 EXPRESSION OF NEURON-SPECIFIC MARKERS TABLE 1. Species, age, sex, source, and procedures used on primates* Specimen number SG4 SG5 MM105 MM101 SG6 LR1 LR3 CJ21 CJ16 P96 EM1 P2855 P2766 P2868 Species Age Sex Source Saguinus geoffroyi Saguinus geoffroyi Saguinus geoffroyi Saguinus geoffroyi Saguinus geoffroyi Leontopithecus rosalia Leontopithecus rosalia Callithrix jacchus Callithrix jacchus Tarsius syrichta Eulemur mongoz Otolemur crassicaudatus Otolemur crassicaudatus Otolemur crassicaudatus n (0) n (0) 1 month 2 month 2.75 years n (3) 4 months 21 years 8 years n (0) n (0) 9 years 12 years 8 years F F M F F F F M M ? ? F F M CZ CZ CZ CZ CZ CZ CZ DM DM DPC DPC DM DM DM Procedures BT, BT, BT, BT, BT, BT, BT, BT, BT, BT, BT, BT, BT, BT, OMP PGP, PGP, PGP, PGP, PGP, PGP, PGP, PGP, OMP PGP, PGP, PGP, PGP, OMP OMP OMP OMP OMP OMP OMP OMP OMP OMP OMP OMP *n, neonatal (days postpartum); CZ, Cleveland Metroparks Zoo; DM, Duke University Medical Center; DPC, Duke University Primate Center. Kelliher et al., 2001), swine (Dorries et al., 1997), sheep (Cohen-Tannoudji et al., 1989), and, in particular, primates (Barrett et al., 1993; Aujard, 1997). The phylogenetic distribution among primates is confusing in that the VNOs of several New World monkeys and the Old World lemurs and their kin retain morphological traits that support a presumption of functionality (Loo and Kanagasuntheram, 1972; Hunter et al., 1984; Taniguchi et al., 1992; Mendoza et al., 1994; Evans and Schilling, 1995; Smith et al., 2003). On the other hand, the Old World monkeys and the great apes, including humans, are not thought to possess a functional VNO. Although chimps and humans retain anatomically identiﬁable VNOs, these appear, by biochemical and morphological criteria, to be vestigial in postnatal life (Meredith, 2001; Smith et al., 2001, 2002). Ethical considerations mandate various indirect approaches to describe the extent and quality of VNO function in those primates that are presumed to retain functional VNOs. One avenue of inquiry has detailed differences in lectin binding between species and between age groups within particular species (Takami, 2002; Halpern and Martı́nez-Marcos, 2003). While those studies contribute to our understanding of the development and microanatomical organization of the vomeronasal epithelium (VNE) as a tissue, they do not directly identify the differentiated state of VNE cell populations as olfactory and therefore cannot address the question of function of the primate VNO as an olfactory organ. We undertook to compare directly the extent to which the VNOs of two strepsirrhine and four haplorhine species contain cells that are both neuronal and olfactory. To that end, antisera raised against the neuronal markers neuron-speciﬁc ␤-tubulin (BT) and protein gene product 9.5 (PGP) and the marker of olfactory identity, olfactory marker protein (OMP), were applied to tissue sections from the six species. BT is expressed by neurons throughout the rodent nervous system (Burgoyne et al., 1988). In particular, BT is expressed in the embryonic, neonatal, and adult rodent main olfactory epithelium (MOE) (Lee and Pixley, 1994; Roskams et al., 1998) as well as in the postnatal rodent VNNE (Hofer et al., 2000; Witt et al., 2002) and the adult canine VNNE (Dennis et al., 2003). PGP is a ubiquitin hydrolase ﬁrst isolated from brain (Jackson and Thompson, 1981; Wilkinson et al., 1989) and is a marker of neurons and neuroendocrine cells generally (Thompson et al., 1983). Speciﬁcally, PGP is expressed in rodent (Johnson et al., 1994) and canine (Dennis et al., 2003) VNNE as well as rodent, canine, and marmoset accessory olfactory bulb (Taniguchi et al., 1993; Johnson et al., 1994; Nakajima et al., 1998a, 1998b, 2003). OMP is a phylogenetically highly conserved cytoplasmic protein expressed by mature olfactory chemosensory neurons of the MOE and VNO (Margolis, 1972, 1980; Farbman and Margolis, 1980). We show here that, in the age groups sampled, all six primate species examined express both neuronal markers and OMP. In one species, Saguinus geoffroyi, the pattern of marker expression by VNNE cells differs signiﬁcantly from the other ﬁve species. MATERIALS AND METHODS The histological sample included four Leontopithecus rosalia (three neonates, one juvenile), six Saguinus geoffroyi (three neonates, two infants, and one 2.75-year-old adult), two adult Callithrix jacchus, one neonatal Tarsius syrichta, two adult Otolemur crassicaudatus, and one infant Eulemur mongoz (Table 1). All specimens were sectioned similarly (Smith et al., 2003). Brieﬂy, tissues were dissected free of the nasal chambers after ﬁxation in 10% buffered neutral formalin (Fisher Scientiﬁc, Pittsburgh, PA), decalciﬁed using a formic acid-sodium citrate solution, dehydrated in a graded series of ethanols, and embedded in parafﬁn. Blocks were sectioned serially at 10 –12 m and every ﬁfth section was stained alternately with Gomori trichrome or hematoxylin-eosin procedures. Intervening sections were saved for immunohistochemistry. Immunohistochemistry Mounted tissue sections were deparafﬁnized in Hemo-D (Scientiﬁc Safety Products) and hydrated to distilled water (dH2O). To abolish endogenous peroxidase-like activity, the sections were incubated in absolute methanol made to 0.9% hydrogen peroxide (H2O2) for 20 min at 1192 DENNIS ET AL. Fig. 1. Antineuron-speciﬁc ␤-tubulin immunoreactivity. A: Otolemur crassicaudatus, adult. BT immunoreactivity is restricted to the VNNE, fascicles of the VNN, and nerves associated with vasculature. VL, vomeronasal lumen. Scale bar ⫽ 100 m. B: Otolemur crassicaudatus, adult. BT immunoreactivity in an individual different than that shown in A is expressed differentially by sensory neuron somata (arrows) and dendrites (arrowheads) in the VNNE. BT immunoreactivity also appears in the axon fascicles (Ax) of the VNN. Scale bar ⫽ 20 m. C: Eulemur mongoz, neonate. BT immunoreactivity is most prominent in the medial VNNE and VNN fascicles in the lamina propria. Some immunoreactivity is present in the NSE (arrowheads). Scale bar ⫽ 200 m. D: Tarsius syrichta, neonate. BT immunoreactivity is present but the signal is light in this preparation. Clusters of immunoreactive cells (arrowheads) occur throughout the VNNE. Axon fascicles are labeled with intensity similar to that expressed by the BT⫹ VNNE cell somata. Scale bar ⫽ 50 m. room temperature (23.5–25°C). Subsequently, the tissues were washed in dH2O, then in 10 mM phosphate-buffered saline (PBS; 2.7 mM KCl, 137 mM NaCl; Sigma). Tissues to be analyzed by epiﬂuoresence microscopy were not incubated in H2O2/methanol but transferred from dH2O directly to PBS. All tissues were incubated 20 min in the appropriate blocking solution [5% normal serum (Sigma) of the species in which the secondary antibody was made and 2.5% BSA (Sigma) in PBS], then washed brieﬂy in PBS. Primary antibody appropriately diluted in blocking solution was applied and the tissue sections were left overnight at room temperature. For double-label BT/PGP assays, anti-BT (Covance) and anti-PGP (Chemicon) were applied as a cocktail. For OMP/BT double-label assays, sections were incubated overnight in anti-OMP (gift of Dr. Frank Margolis) and then in anti-BT for 1 hr at room temperature. Sections to be analyzed with bright ﬁeld optics were treated with biotinylated secondary antibodies (Vector) diluted 1:200, then with ABC Elite reagent (Vector), reacted with diaminobenzidine (Vector), dehydrated, and mounted with VectaMount (Vector). Sections to be analyzed using epiﬂuoresence microscopy were incubated 1 hr with appropriate Alexa-conjugated secondaries (Molecular Probes) diluted at 1:500, mounted with VectaShield (Vector), and sealed with clear nail polish. Tissues were examined with a Nikon Eclipse E600 microscope equipped with epiﬂuoresence. Images were made with an RT Slider digital camera (Diagnostic Instruments) using EXPRESSION OF NEURON-SPECIFIC MARKERS 1193 Fig. 2. Antineuron-speciﬁc ␤-tubulin immunoreactivity. A: Saguinus geoffroyi, neonate. BT signal is present in a small number of isolated cells in the thin VNNE (arrowheads). Scale bar ⫽ 50 m. B: Saguinus geoffroyi, adult. The VNNE is thicker than the neonatal epithelium and more bipolar BT⫹ cells (arrowhead) are present. The small scattered BT⫹ axon fascicles reﬂect the small number of BT⫹ cells in the VNNE. Scale bar ⫽ 50 m. C: Leontopithecus rosalia, neonate. The VNNE is several cells deep and contains numerous BT⫹ cells occurring in clusters (ar- rowhead). Heavily labeled axon fascicles occur immediately below the VNNE. Scale bar ⫽ 50 m. D: Leontopithecus rosalia, juvenile. After 4 months of development, the VNNE contains many BT⫹ cells, a few of which are intensely labeled (arrow). Label in the axon fascicles is not uniformly heavy (arrowhead). Scale bar ⫽ 50 m. E: Callithrix jacchus, adult. The vomeronasal lumina are deﬁned by sensory epithelia. The BT⫹ cell layers are two or three cells thick in the basal compartment. NS, nasal septum. Scale bar ⫽ 200 m. Spot Advanced software and processed with Photoshop 7.0 (Adobe). is not uniform in that not all cell bodies and dendrites are BT⫹ (Fig. 1B). The BT expression pattern manifest in the VNNE is reﬂected in the nonuniform labeling within and among the VNN axon fascicles. The VNNE and VNN axon bundles of neonatal E. mongoz are intensely BT⫹ (Fig. 1C). In contrast, fewer VNNE cell bodies of T. syrichta are BT⫹ (Fig. 1D). The signal is much less intense in both RESULTS Antineuron-Speciﬁc ␤-Tubulin The VNNE and associated axon fascicles of adult O. crassicaudatus are BT⫹ (Fig. 1A). The immunoreactivity 1194 DENNIS ET AL. Fig. 3. Antineuron-speciﬁc ␤-tubulin and antiprotein gene product 9.5 immunoreactivity. A: Eulemur mongoz, neonate. The VNNE cell somata label heavily with PGP (green) in this preparation. Some cell bodies express immunoreactivity to both antibodies (yellow/orange) and a few are BT⫹ (red)/PGP⫺. Most dendrites are BT⫹/PGP⫹ or BT⫹ only. A few dendrites are BT⫺/PGP⫹. The NSE is predominantly negative for either antibody. Scale bar ⫽ 50 m. B: Saguinus geoffroyi, neonate. The VNO at this level is almost entirely BT⫺. A small axon bundle above the basement membrane is BT⫹/PGP⫹ (arrowhead). Most PGP⫹ cell somata occur in the apical epithelial compartment. Scale bar ⫽ 50 m. C: Saguinus geoffroyi, 2 months. The VNNE is relatively thick compared to that of the neonate. Labeled cell somata are BT⫹ (red) or PGP⫹ (green) as well as BT⫹/PGP⫹. Double-labeled axon bundles are present in the VNNE (arrow) and a few are present below the VNNE (arrowhead). Scale bar ⫽ 50 m. D: Saguinus geoffroyi, 2 months. A cluster of labeled cells in the VNNE of the same individual as B shows variable signal intensity from cell to cell. Some cell somata are BT⫹ (red) or PGP⫹ (green; arrows). A few show degrees of double labeling (arrowhead). Scale bar ⫽ 20 m. E: Leontopithicus rosalia, 4 months. Many cells in the VNNE are either BT⫹ or PGP⫹ or both. Cytoplasm of double-labeled cells varies from yellow through red-orange. Dendrites are BT⫹/PGP⫹ or BT⫹/PGP⫺ (arrowhead). Small axon fascicles (arrow) are BT⫹. Scale bar ⫽ 20 m. EXPRESSION OF NEURON-SPECIFIC MARKERS Fig. 4. Olfactory marker protein immunoreactivity. A: Eulemur mongoz, neonate. OMP⫹ cells vary in label intensity. Axon fascicles are more lightly labeled but are above background. The NSE is OMP⫺. Scale bar ⫽ 100 m. B: Tarsius syrichta, neonate. A few sensory cells are OMP⫹ but just above background (arrowheads). C, vomeronasal cartilage. Scale bar ⫽ 100 m. C: Otolemur crassicaudatus, adult. The VNNE and axon fascicles are OMP⫹. The NSE is OMP⫺. Scale bar ⫽ 100 m. D: Callithrix jacchus, adult. Cell bodies, dendrites, and axon fascicles are 1195 strongly OMP⫹. Scale bar ⫽ 50 m. E: Saguinus geoffroyi, 1 month. The VNNE is largely OMP⫺ but does contain a small number of scattered OMP⫹ cells that label above background. Axon fascicles are above background. Scale bar ⫽ 100 m. F: Leontopithicus rosalia, juvenile. The VNNE is OMP⫹ and, in this section, roughly equals the length of the NSE. OMP⫹ cells are present among the axon fascicles below the basement membrane (arrowhead). Scale bar ⫽ 100 m. 1196 DENNIS ET AL. somata and axon fascicles and the VNE displays some segregation of labeled cells to the medial side (Fig. 1D). The VNO of neonatal S. geoffroyi is almost completely BT⫺ with few BT⫹ cell bodies and small BT⫹ axon bundles (Fig. 2A). As individuals develop, BT expression increases somewhat. An adult S. geoffroyi VNO contains more BT⫹ cells compared to the neonate but a small number relative to the entire VNO epithelial volume (Fig. 2B). Small axon fascicles are BT⫹ and are present both in the VNE and in the lamina propria. Cell bodies of neonatal L. rosalia VNO are BT⫹ but not pervasive (Fig. 2C). Axon fascicles of the VNN are BT⫹. After 4 months of development, most of the L. rosalia VNNE is BT⫹ (Fig. 2D). In adult C. jacchus VNO, BT⫹ cells occur around the entire circumference of the VNO (Fig. 2E). These BT⫹ cells are interrupted by regions of BT⫺ cells, many of which are cytokeratin⫹ and are glandular duct cells (data not shown). Axon bundles in the lamina propria are BT⫹. Antineuron-Speciﬁc ␤-Tubulin and Protein Gene Product 9.5 Double Labeling A VNO section from the same E. mongoz individual shown in Figure 1C was double-labeled with anti-BT and anti-PGP antibodies (Fig. 3A). In this preparation, most of the sensory cell somata are BT⫺/PGP⫹, although a small number are clearly double-labeled. Double-labeled dendrites are more numerous than double-labeled cell bodies and BT⫹/PGP⫺ dendrites are more prevalent than BT⫹/ PGP⫺ cell bodies. In the relatively thin neonatal S. geoffroyi VNO, PGP⫹ cells are present in the apical compartment but BT expression by cell bodies is largely absent (Fig. 3B). After 1 month of postnatal development, the VNNE has thickened and the number of BT⫹ cells has increased compared to the neonate (Fig. 3C). Cells expressing immunoreactivity to either or both antibodies are present largely in the basal compartment, although occasional cell clusters span the thickness of the epithelium (Fig. 3D). Double-labeled axon fascicles are present both above the basement membrane and below it (Fig. 3C and D). In contrast, the 4-month L. rosalia VNNE (Fig. 2D) contains many BT⫹/PGP⫹ cells across the thickness of the epithelium, although cells negative for either antibody are present in the middle region of the epithelium (Fig. 3E). The PGP signal intensity varies among cell nuclei and the relative expression of BT and PGP in the cytoplasm varies among cell somata and some dendrites as inferred from the range of color from yellow to dark orange and red. Olfactory Marker Protein Neonatal E. mongoz VNNE is clearly deﬁned by a population of OMP⫹ cells and some dendrites (Fig. 4A). OMP⫹ axon fascicles of variable size occur in the lamina propria. The axon bundles do not label strongly but the label is above background. Neonatal T. syrichta VNO contains a few cells that label above background in this preparation (Fig. 4B). Compared to neonatal E. mongoz, the VNNE of adult O. crassicaudatus contains a prominent population of OMP⫹ cells and more strongly OMP⫹ axon fascicles (Fig. 4C). Adult C. jacchus shows a heavily OMP⫹ VNNE (Fig. 4D). The smaller lightly OMP⫹ axon fascicles of the adult C. jacchus contrast with relatively larger and more heavily OMP⫹ axon fascicles in the lamina propria of O. crassicaudatus (Fig. 4C). One-month-old S. geoffroyi contains a small number of OMP⫹ cells relegated in the TABLE 2. Summary of immunohistochemical results* Species Saguinus geoffroyi Saguinus geoffroyi Saguinus geoffroyi Saguinus geoffroyi Leontopithecus rosalia Leontopithecus rosalia Callithrix jacchus Tarsius syrichta Eulemur mongoz Otolemur crassicaudatus Age BT PGP OMP n 1 month 2 month Adult n 4 month Adult n n Adult ⫹ ⫹ ⫹ ⫹ ⫹ ⫹⫹ ⫹⫹ ⫹ ⫹⫹ ⫹⫹ ⫹⫹ ⫹⫹ ⫹⫹ ⫹⫹ ⫹⫹ ⫹⫹ ⫹⫹ ND ⫹⫹ ⫹⫹ ⫺ ⫹ ⫹ ⫹ ⫺ ⫹⫹ ⫹⫹ ⫹ ⫹⫹ ⫹⫹ *n, neonate; ⫹⫹, many reactive cells; ⫹, few reactive cells; ⫺, no reactive cells; ND, data not available. main to the basal compartment (Fig. 4E). In the VNOs of the two 1-month-old individuals that were studied, lightly labeled axon fascicles are present but not numerous (not shown). Four-month-old L. rosalia VNNE contains a strongly OMP⫹ cell population in the basal compartment and, apically, OMP⫹ dendrites (Fig. 4F). Axon fascicles in the lamina propria are less strongly OMP⫹. In this section, OMP⫹ cells observed among the axon bundles are below the basement membrane. Antineuron-Speciﬁc ␤-Tubulin Olfactory Marker Protein Double Labeling The adult O. crassicaudatus VNNE contains cell bodies positive for either or both BT and OMP immunoreactivity and the nonsensory epithelium (NSE) negative for either antibody (Fig. 5A). The axon fascicles in the lamina propria immediately below the basement membrane are largely BT⫹/OMP⫺. Adult C. jacchus contains double-labeled cell somata and largely BT⫹ dendrites (Fig. 5B). A section of C. jacchus main olfactory epithelium doublelabeled for BT and OMP is shown for comparison with VNO labeling in all species represented (Fig. 5C). The VNE of neonatal L. rosalia is OMP⫺ and largely, but not completely, BT⫺ (Fig. 5D). Scattered BT⫹ cell bodies and dendrites are present and BT⫹ axon bundles are present in the lamina propria. The 4-month-old L. rosalia VNNE is BT⫹/OMP⫹ and is segregated from an NSE that is OMP⫺ (Fig. 5E). Intensely BT⫹ axon fascicles of variable size occur in the lamina propria. At higher magniﬁcation, immunoreactive cells in the VNNE are labeled with either antibody and, in a few cases, with both antibodies (Fig. 5F). The axon fascicles are BT⫹/OMP⫺. In contrast to neonatal and 4-month-old L. rosalia, 1-month-old S. geoffroyi contains BT⫹ cells and axon bundles in the VNNE (Fig. 5G). The OMP⫹ signal is slight compared with other species (Fig. 5A, B, E, and F), but the signal is above background. Adult S. geoffroyi VNO contains few cells immunopositive for either antibody but many small BT⫹ axon fascicles are located in the lamina propria immediately below the basement membrane (Fig. 5H). The few OMP⫹ cells are bipolar and resemble OMP⫹ cells of E. mongoz, O. crassicaudatus, and L. rosalia VNNE. The remaining OMP signal is diffuse and resembles that observed in the 1-month-old individual shown in Figure 5G. Immunohistochemical observations are summarized in Table 2. DISCUSSION Data from this study demonstrate that the VNE of all species examined express immunoreactivity to the neu- Figure 5. 1198 DENNIS ET AL. Fig. 5. Olfactory marker protein and neuron-speciﬁc ␤-tubulin double-label immunoreactivity. A: Otolemur crassicaudatus, adult. The VNNE is strongly OMP⫹ (green) and BT⫹ (red) but the NSE is not immunoreactive. The axon fascicles below the VNNE are predominantly BT-labeled. Scale bar ⫽ 100 m. B: Callithrix jacchus, adult. The cytoplasm in a cluster of VNNE sensory cells (arrowhead) is double-labeled (yellow-orange) but the dendrites are predominantly BT⫹. Scale bar ⫽ 20 m. C: Callithrix jacchus, adult MOE. OMP (green)/BT (red) labeling in the main olfactory epithelium for comparison with immunoreactivity in the VNO. Most cell somata are OMP⫹/BT⫺, but some are doublelabeled. Dendrites are double-labeled or OMP⫺/BT⫹. Axon bundles vary in the relative amount of OMP/BT label. N, nasal space. Scale bar ⫽ 20 m. D: Leontopithecus rosalia, neonate. The VNNE contains a few BT⫹ (red) cells and BT⫹ axon bundles are located in the lamina propria near the epithelium. No OMP⫹ labeling is present. Scale bar ⫽ 100 m. E: Leontopithecus rosalia, 4 months. OMP⫹ and BT⫹ cells occur in the basal compartment throughout most of the VNNE and strongly BT⫹ axon bundles occur in the lamina propria. The NSE contains some BT signal. Scale bar ⫽ 100 m. F: Leontopithecus rosalia, 4 months. A detail of the VNNE from the individual shown in D demonstrates a range of labeling in the cell bodies and dendrites. Some cell bodies are OMP⫹ (green) only or BT⫹ (red) only and several are OMP⫹/BT⫹ (yellow-orange). Axon bundles are predominantly BT⫹. Scale bar ⫽ 40 m. G: Saguinus geoffroyi, 1 month. Some cells (arrows) express OMP immunoreactivity and three or four cells (arrowhead) are BT⫹ in the VNNE. Small axon fascicles below the VNNE and NSE are BT⫹. Scale bar ⫽ 20 m. H: Saguinus geoffroyi, adult. OMP (green) and BT (red) labeling is sparse in the VNNE but a few bipolar OMP⫹ cells are present (arrowhead). BT⫹ axon bundles are present below the basement membrane. Scale bar ⫽ 40 m. ronal marker BT and the speciﬁc olfactory marker OMP at some point during postnatal life. The VNE of the two strepsirrhine (E. mongoz and O. crassicaudatus) and three haplorhine (C. jacchus, L. rosalia, S. geoffroyi) species examined also expressed PGP immunoreactiv- ity. We presume PGP expression in the neonatal tarsier VNE, Tarsius syrichta, but tissue was limited and we did not conﬁrm PGP expression by assay. Of these species, S. geoffroyi showed the least expression, by relative numbers of labeled cells in the VNE, of the three EXPRESSION OF NEURON-SPECIFIC MARKERS marker proteins and that expression was greatest in the 1-month age group but, by adulthood, very few VNNE cells were positive for any of the markers. Among the neonatal individuals examined, a large number of cells in E. mongoz VNE expressed OMP, indicating that at birth, mature sensory neurons are present. The tarsier was intermediate between E. mongoz and the haplorhine species in that a few cells were labeled with BT and OMP antibodies at levels above background. Neonatal L. rosalia expressed BT but not OMP immunoreactivity and neonatal S. geoffroyi expressed immunoreactivity to neither antibody but did express PGP immunoreactivity. In preliminary assays of neonatal C. jacchus, no reactivity to BT or OMP antibodies was observed (data not shown). These observations suggest that, at nativity, the VNE is mostly populated by nonneuronal cells and/or neuronal cells that have not differentiated to the stage in which this set of markers is expressed. Therefore, by immunohistochemistry, and accepting the supposition that OMP expression is required in the VNO for olfactory signaling, only E. mongoz could possess a functional VNO at birth among the species considered here. As the postnatal individual develops, expression of all three markers increases. By 4 months, L. rosalia VNE is strongly OMP⫹/BT⫹ and, in its expression pattern of immunoreactivity, resembles the adult state demonstrated for O. crassicaudatus and C. jacchus (compare Fig. 5E and F with A and B). In the 1- and 2-month-old S. geoffroyi individuals examined, marker expression was greater compared to that seen in the neonate but was not comparable to the level seen at 4 months in L. rosalia. The VNE in both species are about equal thickness and labeled cells occur in small clusters and are rarely observed in layers even two cells deep. Of the labeled cells, BT and PGP signal is most prominent. In S. geoffroyi, OMP signal is above background but weak compared to the OMP signal in L. rosalia. Additionally, the VNE does not appear well organized in as much as axon fascicles occur in the epithelium, although that judgement is based on a limited number of available specimens and may be sampling artifact. By adulthood, BT and OMP signal is present but in only a few cells in Saguinus, and more BT⫹ cells are observed than OMP⫹ cells. These labeled cells occur in loose clusters, as in the 1-month-old, suggesting a clonal genesis, but the dearth of OMP⫹ cells suggests that many of the BT⫹ cells are not in the OMP-expressing lineage or these cells die before they, or their daughter cohorts, terminally differentiate and begin expressing OMP. These possibilities suggest that BT and OMP expression, rather than PGP expression, are better indicators of capacity for VNO function vis-à-vis the VNE. First, not all cells are PGP⫹, begging the question of the PGP⫺ cells’ identity. Second, although PGP is a neuronal marker, it is expressed in the MOE and the VNO throughout life but the expression pattern within a given cell’s life span is unknown. Finally, PGP is involved with ubiquitin and is therefore part of the cellular metabolism. Its expression pattern may be episodic or it may undergo regular changes, for example, conformational, that make it invisible to the antibody used in any particular study. On the other hand, the BT expression pattern in the VNO is likely to conform to the pattern of olfactory differentiation observed in the MOE, wherein cell cohorts in the olfactory neuron lineage express BT before they express OMP (Lee and Pixley, 1994). As individual cells begin to express 1199 OMP, BT expression in the cell body subsides but remains high in the dendrites. BT expression suggests that neurogenesis and the passage of precursors through the olfactory developmental program to terminal differentiation marked by OMP expression continues in some adults considered here. Expression of growth-associated protein 43, a marker of new neurons that are extending processes, in adult O. crassicaudatus VNO (data not shown) lends credence to the supposition of continuous birth and maturation of mature vomeronasal cells in the VNE. OMP expression conﬁrms that mature sensory neurons are present in the VNE of the several species considered. That expression does not afﬁrm the inevitability of vomeronasal function, nor does it address the extent to which the vomeronasal system may function in these species. Compared to rodent VNO, the histology of the primate VNO is more similar to that of goats (Takigami et al., 2000), sheep (Cohen-Tannoudji et al., 1989), swine (Dorries et al., 1997), and ferrets (Weiler, et al., 1999). In the last three of these species, the VNO is not involved in all of the functions mediated by the rodent VNO (CohenTannoudji et al., 1989; Dorries et al., 1997; Weiler, et al., 1999; Kelliher et al., 2001). Likewise, several behaviors that in rodents are mediated by the VNO are not solely VNO-mediated in primates (Barrett et al., 1993; Aujard, 1997; Kraus et al., 1999). The histological similarities to sheep, swine, and ferret VNO together with the behavioral observations argue that primate VNO function is likely to be restricted compared to rodents and more similar to that of the former species. The extent to which VNO function resembles, or differs from, that in the several nonrodent species mentioned above not withstanding, we establish here that the VNE of infant E. mongoz, adult O. crassicaudatus, adult C. jacchus, and 4-month-old L. rosalia express mature olfactory neurons. By that criterion, VNE function as a chemosensory vomeronasal epithelium may be possible in those species. ACKNOWLEDGMENTS The authors are grateful to Dr. Frank Margolis for the generous gift of the anti-OMP antisera used in this study. They also thank J.H. Kinzinger and K.L. Shimp for sectioning some of the specimens used in the study and A.B. Taylor and C.J. Vinyard for providing nasal tissues from adult marmosets. Supported in part by Federal Aviation Administration grant 01-6-022 and U.S. Army Robert Morris Acquisition Center grant N66001-1099-0072 to (E.E.M.). This is Duke Primate Center publication number 789. LITERATURE CITED Aujard F. 1997. Effect of vomeronasal organ removal on male sociosexual responses to female in a prosimian primate (Microcebus murinus). Physiol Behav 62:1003–1008. Barrett J, Abbott DH, George LM. 1993. Sensory cues and the suppression of reporduction in subordinate female marmoset monkeys, Callithrix jacchus. J Repro Fert 97:301–310. Burgoyne RD, Cmabray-Deakin MA, Lewis SA, Arkar S, Cowan NJ. 1988. Differential distribution of ␤-tubulin isotypes in cerebellum. EMBO J 7:2311–2319. Cohen-Tannoudji J, Lavenet C, Locatelli A, Tillet Y, Signoret JP. 1989. Non-involvement of the accessory olfactory system in the LH response of anoestrous ewes to male odour. J Reprod Fert 86:135– 144. 1200 DENNIS ET AL. Dennis JC, Allgier JG, Desouza LS, Eward WC, Morrison EE. 2003. Immunohistochemistry of the canine vomeronasal organ. J Anat 203:329 –338. Dorries KM, Adkins-Regan E, Halpern BP. 1997. Sensitivity and behavioral responses to the pheromone androstenone are not mediated by the vomeronasal organ in domestic pigs. Brain Behav Evol 49:53– 62. Evans C, Schilling A. 1995. The accessory (vomeronasal) chemoreceptor system in some prosimians. In: Alterman L, editor. Creatures of the dark: the nocturnal prosimians. New York: Plenum. p 393– 411. Farbman AI, Margolis FL. 1980. Olfactory marker protein during ontongeny: immunohistochemical localization. Dev Biol 74:205– 215. Halpern M. 1987. The organization and function of the vomeronasal system. Ann Rev Neurosci 10:325–362. Halpern M, Martı́nez-Marcos A. 2003. Structure and function of the vomeronasal system: an update. Prog Neurobiol 70:245–318. Hofer D, Shin D-W, Drenckhahn D. 2000. Identiﬁcation of cytoskeletal markers for the different microvilli and cell types of the rat vomeronasal sensory epithelium. J Neurocytol 29:147–156. Hunter AJ, Fleming D, Dixson AF. 1984. The structure of the vomeronasal organ and nasopalatine ducts in Aotus trivirgatus and some other primate species. J Anat 138:217–225. Jackson P, Thompson RJ. 1981. The demonstration of new human brain speciﬁc proteins by high-resolution two-dimensional polyacrylamide gel electrophoresis. J Neurol Sci 49:429 – 438. Johnson EW, Eller PM, Jafek BW. 1994. Protein gene product 9.5 in the developing and mature rat vomeronasal organ. Dev Brain Res 78:254 –259. Kelliher KR, Baum MJ, Meredith M. 2001. The ferret’s vomeronasal organ and accessory olfactory bulb: effect of hormone manipulation in adult males and females. Anat Rec 263:280 –288. Kraus C, Heistermann M, Kappeler PM. 1999. Physiological suppression of sexual function of subordinate males: a subtle form of intrasexual competition among male sifakas (Propithecus verreauxi)? Physiol Behav 66:855– 861. Lee VM, Pixley SK. 1994. Age and differentiation-related differences in neuron-speciﬁc tubulin immunostaining of olfactory sensory neurons. Dev Brain Res 83:209 –215. Loo SK, Kanagasuntheram R. 1972. The vomeronasal organ in tree shrew and slow loris. J Anat 112:165–172. Margolis FL. 1972. A brain protein unique to the olfactory bulb. Proc Natl Acad Sci USA 69:1221–1224. Margolis FL. 1980. A maker protein for the olfactory chemoreceptor neuron. In: Bradshaw RA, Schneider D, editors. Proteins of the nervous system. New York: Raven. p 59 – 84. Mendoza AS, Kuderling I, Kuhn HJ, Kuhnel W. 1994. The vomeronasal organ of the new world monkey Saguinus fuscicollis (Callitrichidae): a light and transmission electron microscopic study. Ann Anat 176:217–222. Meredith M. 2001. Human vomeronasal organ function: a critical review of best and worst cases. Chem Senses 26:433– 445. Nakajima T, Murabayashi C, Ogawa K, Taniguchi K. 1998a. Immunoreactivity of protein gene product 9.5 (PGP 9.5) in the developing hamster olfactory bulb. Anat Rec 250:238 –244. Nakajima T, Sakaue M, Kato M, Saito S, Ogawa K, Taniguchi K. 1998b. Immunohistochemical study on the accessory olfactory bulb of the dog. Anat Rec 252:393– 402. Nakajima T, Tanioka Y, Taniguchi K. 2003. Distribution of protein gene product 9.5-immunopositive and NADPH-diaphorase-positive neurons in the common marmoset (Callithrix jacchus) accessory olfactory bulb. J Vet Med Sci 65:1307–1311. Roskams AJI, Cai X, Ronnett GV. 1998. Expression of neuron-speciﬁc beta-III tubulin during olfactory neurogenesis in the embryonic and adult rat. Neuroscience 83:191–200. Smith TD, Siegel MI, Bonar CJ, Bhatnagar KP, Mooney MP, Burrows AM, Smith MA, Maico LM. 2001. The existence of the vomeronasal organ in postnatal chimpanzees and evidence for its homology with that of humans. J Anat 198:77– 82. Smith TD, Siegel MI, Bhatnagar KP, Shimp KL, Kinzinger JH, Bonar CJ, Burrows AM, Mooney MP, Siegel MI. 2002. Histological deﬁnition of the vomeronasal organ in humans and chimpanzees, with a comparison to other primates. Anat Rec 267:166 –176. Smith TD, Siegel MI, Bhatnagar KP. 2003. Observations on the vomeronasal organ of prenatal Tarsius bancanus borneanus with implications for ancestral morphology. J Anat 203:473– 481. Takami S. 2002. Recent progress in the neurobiology of the vomeronasal organ. Microsc Res Tech 58:228 –250. Takigami S, Mori Y, Ichikawa M. 2000. Projection pattern of vomeronasal neurons to the accessory olfactory bulb in goats. Chem Senses 25:387–393. Taniguchi K, Matsusaki Y, Ogawa K, Saito TR. 1992. Fine structure of the vomeronasal organ in the common marmoset (Callithrix jacchus). Folia Primatol 59:169 –176. Thompson RJ, Doran JF, Jackson P, Dhillon AP, Rode J. 1983. PGP 9.5: a new marker for vertebrate neurons and neuroendocrine cells. Brain Res 278:224 –228. Weiler E, Apfelbach R, Farbman AI. 1999. The vomeronasal organ of the male ferret. Chem Senses 24:127–136. Wilkinson KD, Lee K, Deshpande S, Duerksen-Hughes P, Boss JM, Pohl J. 1989. The neuron-speciﬁc protein PGP 9.5 is a ubiquitin carboxyl-terminal hydrolase. Science 246:670 – 673. Witt M, Georgiewa B, Knecht M, Hummel T. 2002. On the chemosensory nature of the vomeronasal epithelium in adult humans. Histochem Cell Biol 117:493–509. Wysocki CJ. 1979. Neurobehavioral evidence for the involvement of the vomeronasal system in mammalian reproduction. Neurosci Biobehav Rev 3:301–341.