SEARCH

SEARCH BY CITATION

Keywords:

  • crypt neuron;
  • olfactory epithelium;
  • OMP;
  • skate;
  • development

Abstract

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. LITERATURE CITED

Crypt neurons are olfactory receptor cells located in the olfactory epithelium of fishes. They exhibit a peculiar and well-recognizable morphology, although their odorant specificity is still unknown. Data on their appearance during development are few and far between. This study set out to identify the time at which crypt neurons appeared in the skate, Raja clavata, using histological and immunohistochemical methods. For this purpose, embryos and juveniles at different stages of development, from 13 weeks after laying (11 weeks before hatching) to 24 weeks after hatching, were examined. The crypt neurons were identified on a morphological basis. An anti–α-tubulin antibody and two lectins (wheat germ agglutinin and peanut agglutinin) were used to highlight morphological details. The olfactory marker protein was detected by immunohistochemistry, because this protein is a marker of neuronal maturity in vertebrates. The crypt neurons could be detected by their morphology at 15 weeks after laying and became strongly olfactory marker protein immunoreactive 22 weeks after laying. Although involvement of crypt neurons in reproductive behavior has been inferred in various studies on bony fishes, their early presence in skate embryos and juveniles may suggest that they are not exclusively involved in sexual behavior. Anat Rec, 290:1268-1272, 2007. © 2007 Wiley-Liss, Inc.

Crypt neurons (CNs) are olfactory receptor cells located in the olfactory epithelium (OE) of some vertebrates (Hansen et al., 1997; Hansen and Finger, 2000; Schmachtenberg, 2006). Their presence has been assessed in actinopterigians, where they represent the third olfactory neuron type, while they are absent in both sarcopterigians and tetrapods (Hansen and Finger, 2000). Recently, CNs have been described in Chondrichthyes (Ferrando et al., 2006a; Bottaro et al., submitted). CNs are characterized by a typical morphology, clearly distinguished from that of common olfactory receptor neurons (ORNs). They exhibit an oval shape and an apical invagination where short cilia protrude. Although their presence is uncommon, they are located in the upper third of the OE and scattered along the olfactory lamellae (Hansen et al., 1997; Hansen et al., 1999, 2003, 2004; Hansen and Finger, 2000; Catania et al., 2003; Germanà et al., 2004; Ferrando et al., 2006a). Their presence and distribution in fishes seem to vary from specimen to specimen and from season to season, suggesting a certain variability and feedback control of the expression of the CN population (Belanger et al., 2003; Hamdani and Døving, 2006).

Studies into their projection to the olfactory bulb (OB) have been conducted to understand the function of CNs. In Ictalurus punctatus and Carassius carassius, neural tracer injection into the OB showed that CNs project to specific ventral sites of the bulb (Hansen et al., 2003; Hamdani and Døving, 2006) where a population of secondary sex pheromone-specific neurons are located (C. carassius, Lastein et al., 2006). These observations suggest that CNs are involved in reproductive behavior (Hamdani and Døving, 2006). On the other hand, a recent study into the electrophysiological properties of CNs suggests these neurons respond to amino acids, generally thought to mediate feeding behavior (Schmachtenberg, 2006). However, the receptor gene family expressed in this cell type remains unknown (Hansen et al., 2004) and its role can only be suggested.

Data concerning the point at which CNs appear during development is scarce in the literature. This cell type has been observed in the juvenile stage of Acipenser sp. and Verasper moseri (Zeiske et al., 2003; Yamamoto et al., 2004). Zeiske and colleagues (2003), however, could not rule out the presence of CNs in the embryonic or larval stages of Acipenser sp. due to the small number and range of the TEM sections used. No data have been provided on the appearance of CNs in Chondrichthyes.

In this study, the OE of embryos and juveniles of the elasmobranch Raja clavata were examined using a histologicaland immunohistochemical approach. The presence of cilia and mucus in the apical crypt was highlighted by tubulin immunohistochemistry and lectin histochemistry, as it was for the shark Scyliorhinus canicula (Ferrando et al., 2006a). The probable mature state of the CNs was assessed by olfactory marker protein (OMP) immunoreactivity. It is known that the OMP is only expressed in the mature olfactory sensory neurons of many vertebrates (Monti Graziadei et al., 1977; Farbman and Margolis, 1980; Rama Krishna et al., 1992; Çelik et al., 2002). In mammals, OMP immunoreactivity has only been demonstrated in sensory neurons after their synaptogenesis in the olfactory bulb (Farbman and Margolis, 1980). Nevertheless, OMP is not present in all vertebrates, although it was recently immunohistochemically detected in elasmobranches (Ferrando et al., 2007; Bottaro et al., manuscript submitted for publication).

MATERIALS AND METHODS

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. LITERATURE CITED

Twenty-one R. clavata eggs, laid in captivity in tanks at Acquario di Genova (Costa Edutainment S.p.A.), were kept at 15°C in the aquaria of the Department of Biology (University of Genoa, Italy). Incubation, from egg-laying to hatching, lasted 24 weeks. Three embryos were killed 13, 15, 19, and 22 weeks after egg laying (wl). Three juveniles were killed 2, 8, and 24 weeks after hatching (wh). The animals were anesthetized with 0.01% ethyl 3-aminobenzoate methanesulfonate salt (Sigma-Aldrich, St. Louis, MO; dilution 1:1,000 in seawater), killed and then dissected to collect the olfactory organs (OOs). OOs were fixed in 4% paraformaldehyde in a 0.1 M phosphate buffered solution (PBS, pH 7.4) at 4°C, Paraplast (Bio-Optica, Italy) embedded and 5-μm-thick sectioned. Each OO was divided into two halves along the rostrocaudal axis; as only the first 25 sections were considered for each half OO, only the central 250 μm were examined for each OO. Histological observations were performed by hematoxylin–eosin (Bio-Optica, Italy). Two Alexa 488-conjugated lectins (Molecular Probes, The Netherlands) were used to label different sugar residues: peanut agglutinin (PNA; source: Arachis hypogea, binding specificity: β-Galactose, working dilution: 20 μg/ml) and wheat germ agglutinin (WGA; source: Triticum vulgaris, binding specificity: N-acetylglucosamine and N-acetylneuraminic acid, working dilution: 10 μg/ml). Specificity controls were performed by pre-adsorbing lectins with their specific sugar residue. For ciliary microtubules, a mouse monoclonal anti–α-tubulin (1:100, Sigma) and an fluorescein isothiocyanate-conjugated anti-mouse secondary antiserum (1:400 in PBS, DAKO, Denmark) were used. For OMP, a goat polyclonal anti-OMP antiserum (1:800 in PBS, Wako-Chemicals, Richmond, VA) and an Alexa 488-conjugated anti-goat secondary antiserum (1:800 in PBS, Molecular Probes, The Netherlands) were used.

Negative controls were performed by omitting the primary antibody or antiserum. The sections were examined using a BX60 Olympus microscope (light and epifluorescence microscope) and visualized through an Olympus CCD Color-ViewII Camera (Olympus, Japan).

RESULTS

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. LITERATURE CITED

Morphology

At 13 wl, well-developed olfactory lamellae can be detected. The OE is approximately 17 μm thick, with ill-defined cellular types. The nuclei are located in two different layers, and their shape ranges from round to oval, regardless of their position (Fig. 1a). At 15 wl, the OE is approximately 20 μm thick. Two types of nuclei can be seen: dark-stained oval nuclei and pale round nuclei with prominent nucleolus (Fig. 1b). CNs, in their typical morphology, can be seen in the upper part of the epithelium (Fig. 1c), while they are rare and scattered along the lamellae. At 19 wl, the OE is approximately 35 μm thick. At this stage, the cellular organization approaches to the mature adult condition. The supporting cells are characterized by an oval nucleus, generally situated in the upper third of the cell. Numerous large, lightly stained cells can be seen in the hematoxylin– eosin stained olfactory epithelium, running from the base to the apex of the epithelium and exhibiting a flask-like shape. Due to its morphology and its characteristic Na+/K+ ATPase immunoreactivity (ir), a role in ion exchange has been suggested for this cell type (Ferrando et al., 2006b). ORNs present pale round nuclei in the central zone of the epithelium (Fig. 1d). CNs exhibit a typical chestnut-like shape and a clear nucleus with a prominent nucleolus in the upper zone of the epithelium (Fig. 1e). From 22 wl to 24 wh (Fig. 1f), the epithelium thickens to approximately 70 μm, while its morphology remains unchanged. CNs are always few and far between, and it is not possible to appreciate any differences in density and/or localization among different specimens.

thumbnail image

Figure 1. Morphology of crypt cells during prehatch and posthatch development. Sections of olfactory lamellae of R. clavata, hematoxylin eosin. a: At 13 weeks after egg laying (wl). Although the olfactory lamellae are well developed, it is difficult to recognize different cell types. Mitosis is not uncommon (arrows). b: At 15 wl, some nuclei in the epithelial thickness present the typical appearance of olfactory receptor neurons (asterisks). c: At 15 wl, crypt neurons (CNs), surrounded by supporting cells (arrows), can be seen in the upper zone of the epithelium (arrowhead). d: At 19 wl, large clear cells, probably involved in ion exchange, can be clearly seen scattered along the lamellae (arrow); the nuclei of olfactory receptor neurons are characterized by a round shape and prominent nucleolus (asterisks). e: CNs, surrounded by supporting cells (arrows), are located in the upper third of the epithelium (arrowhead). f: At 24 weeks after hatching (wh), CNs, surrounded by supporting cells (arrows), are scattered in the olfactory epithelium (arrowhead).

Download figure to PowerPoint

Lectin Histochemistry

Although CNs exhibit WGA labeling in their crypt zone in all the stages considered (Fig. 2a), PNA labeling does not appear in any of these stages.

thumbnail image

Figure 2. Crypt cells during prehatch and posthatch development: morphological details and olfactory marker protein (OMP) immunoreactivity (ir). Sections of olfactory lamellae of R. clavata. a: Alexa 488-conjugated wheat germ agglutinin (WGA) lectin. At 15 weeks after egg laying (wl), WGA labeling is located in the crypt of crypt neurons (CNs; arrow). b: α-Tubulin immunofluorescence. At 19 wl, the ir is localized in the crypt of CNs. c,d: OMP immunofluorescence. c: At 19 wl, the ir appears at this stage, barely visible in the ORNs, along the epithelium. The fluorescence in the CN (arrow) is localized in the cytoplasm, delineating the crypt as a darker zone (asterisk). d: At 22 wl, the ir is variably present in the ORNs and clearly visible in the cytoplasm of CN (arrow), while it is not present in the crypt zone (asterisks).

Download figure to PowerPoint

Immunofluorescence

The CNs observed at 15 wl do not present α-tubulin immunoreactivity. Immunolabeling in all the CNs observed, in fact, appears for the first time at 19 wl (Fig. 2b).

A faint OMP ir appears in both CNs (Fig. 2c) and ORNs, at 19 wl. Not all the olfactory neurons show the same ir and fluorescence varies from cell to cell. At 22 wl, the ir exhibits the characteristics seen in adult conditions: marked fluorescence in the cytoplasm of all the CNs and a variable ir in the different ORNs can be seen (Fig. 2d). This ir pattern appears in all the subsequent stages. The results are shown in Table 1.

Table 1. Time of Appearance of Crypt Neurons and Their Histochemical and Immunohistochemical Features in Developing R. clavata
StageOE thickness (μm)Detection of CNsWGA labellingα-Tubulin irOMP ir
  1. OE, olfactory epithelium; WGA, wheat germ agglutinin; ir, immunoreactivity; OMP, olfactory marker protein; wl, weeks after egg laying; wh, weeks after hatching.

13 wl17
15 wl20+++
19 wl35++++++
22 wl40+++++++
hatch
2 wh40+++++++
8 wh55+++++++
24 wh70+++++++

DISCUSSION

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. LITERATURE CITED

Our results suggest that the histodifferentiation of OE begins at 15 wl, when morphologically defined CNs can be seen. At this stage, they already exhibit WGA-labeled mucus in the crypt. However, they do not seem to have cilia as they are totally negative to tubulin antiserum. The other olfactory neurons are not clearly detectable at the morphological level. At 19 wl, CNs exhibit cilia in the mucus-filled crypts, as indicated by tubulin ir in the crypt zone. The presence of a faint OMP ir in the CN cytoplasm suggest that these cells could be considered as mature, even though a close link between OMP expression and synaptogenesis in the olfactory bulb has only been demonstrated in mammals (Farbman and Margolis, 1980). At 19 wl, the other olfactory neurons are morphologically recognizable in the OE, as is variable OMP ir. At 22 wl, the CNs develop their definitive structure and strong OMP ir can be seen. This aspect, together with WGA labeling and tubulin ir in the crypt and OMP ir in the cytoplasm, was seen in all subsequent stages of R. clavata until 24 wh, as well as in other adult elasmobranches (Ferrando et al., 2006a).

The peculiar presence of PNA lectin labeling in the crypt of CNs could be seen in S. canicula. This feature was interpreted as a specialization of the mucus composition within the crypt, due to a particular odorant specificity for this neuron type (Ferrando et al., 2006a). The absence of PNA lectin labeling in the CNs of R. clavata, as shown in this work, and of Raja brachyura (data not shown) demonstrates that mucus composition is not the same in all elasmobranches.

The OMP ir in R. clavata has a similar pattern to that observed in S. canicula, where CNs are strongly positive while ORNs are variably immunoreactive, probably due to different maturation stages. However, the possible presence of a family of OMP genes in elasmobranches, as seen in amphibians (Rössler et al., 1998), should not be ruled out. In fact, the different affinity of the antiserum (raised against mammalian OMP) used could explain the difference in OMP ir seen in the ORNs of R.clavata and S. canicula. The OMP ir pattern seen in R. brachyura, where the same antiserum only immunostained the CNs, support the hypothesis of more than one OMP in elasmobranches (Bottaro et al., manuscript submitted for publication).

Our data suggest that, in R. clavata embryos, CNs are morphologically detectable before the other receptor neurons (15 wl). However, they only show their typical morphology (with cilia in the crypt) at 19 wl, when the morphology of ORNs can also be clearly seen. At this stage, both CNs and ORNs display OMP ir, thus demonstrating their probable maturation.

In all vertebrates, the olfactory neurons begin to mature before birth and continue during postnatal development (Farbman and Margolis, 1980; Hansen and Zeiske, 1993; Valverde et al., 1993; Fishelson and Baranes, 1997). In some fishes, the OE is involved in imprinting processes, during development and in the very early stages of life (Werner and Lannoo, 1994; Arvedlund et al., 2000; Harden et al., 2006). No data on imprinting in elasmobranches are present in literature. However, it is widely known that the egg case of cartilaginous fishes has respiratory canals that become unplugged during incubation, allowing the seawater to flow inside the case. We observed that these canals open in R. clavata at approximately 10 wl. As their name suggests, the respiratory canals provide oxygenation but they also allow different substances to come into contact with the embryo. It is therefore possible that a functional OE could transmit information from the surrounding environment during the final weeks of incubation. Together with the maturation steps of the OE before birth, we have identified, for the first time, the presence of mature CNs in R. clavata before hatching and in the very early stages of life. In conclusion, the early presence of CNs could suggest they are involved in other functions in addition to reproduction.

Acknowledgements

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. LITERATURE CITED

The authors thank Acquario di Genova (Costa Edutainment S.p.A.) and Dr. Daniele Zanzi for the kind gift of the R. clavata eggs. Thanks also to Dr. Raffaele Ferrari for his invaluable assistance.

LITERATURE CITED

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. LITERATURE CITED