Structural and immunohistological modifications in olfactory bulb of the staggerer mutant mouse

29

Biology of the Cell 91 (1999129-44 o Elsevier, Paris

Original article

ations Structural and immunohistobgical mu in olfactory butb of the staggerer mutant mouse Zor6 Monnier a, Malika Bahjaoui-Boubaddi a, Jacqueline Bride a, Michelle Bride anb, Francois Math a and Alain Propper a a Laboratoire de Neurosciences, UFR Sciences et Techniques, place Leclerc, 25030 Besancon; bCentre de Microscopic Electronique, Universitk de Franche-Comtk, France

In the present study, we describe the structural and cytological changes observed in stuggerer mutant olfactory bulbs, as compared to normal mice. On the basis of photonic and ultrastructural observations we tried to define the alterations induced by the mutation: ie a reduction of bulb size, a reduction in the volume of three out of the six architectonic layers (glomerular, external and internal plexiform), a reduction of glomeruli size, a loss of half the mitral cells and a slight decrease in juxtaglomerular interneuron number. In stuggerer, an hypertrophy of glial ensheathing cell processes was especially evident at the level of each glomerulus, whereas the density of the astrocyte network was weaker in the granular layer and the nerve layer not apparently impaired. An immunofluorescent labelling study combined with confocal scanning microscopy was performed in order to identify the cellular type and the differentiation degree of the various elements. Antibodies anti-GFAP, a protein present in both ensheathing cells and astrocytes, and anti-OMP, the specific maturation protein of the nerve layer, were used for that purpose. Data confirmed the reality of the gliosis and the persistence of the sensory component in the mutant. All the structural alterations described in staggerer olfactory bulb were in close agreement with the functional troubles previously recorded. Our results are discussed in connection with the present knowledge on embryonal origin, fetal development and adult cellular renewal of the olfactory bulb. 0 Elsevier, Paris

olfactory bulb / electron-confocal microscopy / OMP / GFAP/ immunolabelling

INTRODUCTION Murine neurological mutations are interesting experimental models for olfactory apparatus studies because they offer an opportunity to establish direct correlations between gene, structure and function. Moreover, olfactory cells display, throughout the whole animal life, a continuous renewal which in turn induces a periodic synaptical reorganisation in the olfactory bulb. This renders the olfactory system very sensitive to any functional abnormality, including gene defects, which may affect the central nervous system (Math et ~2, 1997).

*Correspondence

and reprints

Modifications in olfactory bulb of the staggerer mutant mouse

Recently, the genetic mapping of the mouse demonstrated in stuggerev a deletion affecting the RORa gene located on chromosome 9, which encodes a member of the nuclear hormone receptor superfamily (Hamilton et al, 1996). The knockout of this gene, experimentally performed, has confirmed that the deficiencies observed in staggerer mutants are linked to a defect in RORa functioning (Dussault et al, 1998; Steinmayr et nl, 1998). The staggerer mutation is known since a long time (Sidman et al, 1962) for cerebellar neuron degeneration. The gene defect goes along with numerous behavioural alterations: impaired interindividual recognising as well as troubles in mating behaviour (Guastavino, 1982; Baudoin et al, 1991, 1994; Feron and Baudoin, 1992; Guastavino and Larrson, 1992). Most of these behavioural altera-

Monnier et al

30 tions strongly suggest the occurrence of olfactory system defects. Starting from this observation, we have recently demonstrated, that, in staggerer mice, olfactory bulb potentials induced by odours are dramatically modified or even absent in some cases (Math et aI, 1995). The aim of the present cytological study was to establish whether the morphological organisation of olfactory bulb, in a mutant which presents well characterised disturbances of odour recognition we described in our previous functional results, was in correlation with structural modifications. There were until now just a few works devoted to olfactory troubles study in mutant mice or at least dealing with olfactory bulb cytological alterations among which one study on the reeZev mutation (Wyss et al, 1980) describing a reduction in olfactory bulb size and, another one, on Purkinje cell degeneration (PCD) mutant in which an extensive mitral cell loss was pointed out in olfactory bulb (Greer and Sheperd, 1982). In the present work, general morphology of the olfactory system was studied by photonic microscopy, both fine cellular identification and quantification by transmission electron microscopy, and immunolabelling of specific markers by confocal microscopy. On the basis of descriptions previously given for other mammals (Price and Powell, 1970a, b, c; Pinching and Powell, 1971a, b, c; Farbman, 1992; MC Lean and Shipley, 1992) we identified the laminar structure and the various cellular types in the mouse olfactory bulb. In order to discover putative structural and cytological alterations of the olfactory bulb, we precisely compared, point by point, sfuggever mutant to normal mouse. Since a neuronal specific marker in the olfactory bulb is still lacking, we chose as a maturation sign the olfactory marker protein (OMP) (Margolis, 1972) and a glial fibrillary marker (GFAP), because neuronal loss frequently goes along with glial cell hyperactivity. Moreover, olfactory bulb is characterised by the fact that both astrocytes and ensheathing cells contain the same protein in their cytoskeletal intermediary filaments (Barber and Lindsay, 1982).

MATERIALS AND METHODS Animals Adult (2-3 months) male mice from the C57/BL56 strain (Iffa Credo) were used as controls. They were bred in OUT lab under 12 h light-12 h dark and fed ad libitum. with a mouse diet (UAR, Villemoisson-sur-Orge, France). Staggerer mice were obtained from animals reared in the Physiology and Behavior Biology Laboratory (Nancy, France), all originating from professor Baudoin’s lab (University Paris XIII). The staggerermutation may be

Modifications in olfactory bulb of the staggerer mutant mouse

Biology of the Cell 91 (1999) 2944

identified in young animals as soon as 14 days after birth by evident troubles in locomotion (Guastavino, 1982). Mutant animals were given the same diet, but in the form of a flour mixed with water to produce a large ball put on the ground of the cages. This kind of feeding allows mutant mice to overcome their locomotory handicap and therefore to develop almost normally. Before killing, the mice were anaesthetised by intraperitoneal injection of 0.3 M chloral hydrate (Merck) in NaCl solution at a dose of 0.8 mL/lOO g. For each experimental series, 10 mutants and 10 normal mice were used. After appropriate fixation, all olfactory bulbs were embedded so as to always present an identical antero-posterior orientation. Serial transverse sections were then performed, in all experimental cases.

Transmission

electron

microscopy

(TEM)

Animals were given an intracardiac perfusion with a mixture of 4% glutaraldehyde, 0.2 M Na cacodylate buffer to which 0.03% CaCl, was added. The final pH was adjusted at 7.2 and osmotic pressure at 400 mosM. After perfusion, olfactory bulbs were removed and post-fixed for 1 h with 2% buffered osmium tetroxide. After alcoholic dehydration, tissues were embedded in ERL 4206 (Spiirr, 1969). Semi-thin and ultra-thin sections were cut on a Reichert ultra slicer. The semi-thin sections were observed and photographed with a Zeiss photonic microscope and ultra-thin sections with a Jeoll200 EX TEM.

lmmunofluorescence

and confocal microscopy

Some sample series were fixed in 4% buffered paraformaldehyde (PFA) for 3 h and others in Bouin’s fixative. All immunofluorescence techniques were carried out on thick slices (100 p) cut with a vibratome. Each section was stored overnight at 4 “C in PBS to be later assayed for GFAP or OMP. Rabbit polyclonal anti-GFAP (Sigma) was used at l/80 dilution in phosphate buffer saline (PBS), containing 0.1% Triton X-100, 1% bovine serum albumin (BSA) and 10% sheep serum. A goat polyclonal antibody against olfactory marker protein (OMP), a gift from Dr FL Margolis (Nutley, USA) (Margolis, 1972), was used at l/200 dilution in the same physiological solution, containing Triton BSA and 10% normal rabbit serum. The immunostaining sequence included the following steps: i) 10 min incubation in PBS containing 1% glycin; ii) washing three times (5 min each) in PBS, 0.4% Triton X100, 1% BSA, pH 7.6; iii) preincubation in a blocking solution for 1 h at room temperature (PBS, 0.1% Triton X-100, 1% BSA,lO% normal goat serum for GFAP, or PBS, 0.1% Triton X-100, 5% BSA, 10% normal rabbit serum for OMP); iv) incubation with primary antibodies overnight at room temperature; v) washing three times (5 min each) in 0.2% Triton X-100, 1% BSA in PBS; vi) incubation for 1 h at room temperature in goat anti-rabbit FITC labelled immunoglobulins (Sigma) at a l/80 dilution in PBS containing 1% BSA, 0.1% Triton X-100 for GFAP, or with rabbit anti-sheep FITC labelled immunoglobulins (Institut Pasteur) at a l/90 dilution in the same physiological solution; and vii) washing three times (5 min each) in PBS. Sections were spread on glass slides and then mounted in Vectashield (Biosys).

Monnier et al

31

Biology of the Cell 91 (1999) 29-44

Controls were .performed by omitting the first antibody. The slides were observed in a Leica confocal laser scanning microscope. After acquisition of the series of optical sections in the Z dimension, the final pictures were obtained by linear projection of the required number of successive horizontal sections (maximal projection) or presented as stereo-pairs.

Morphometric

cell counts

Mitral cells were counted using a light microscope at x 25 magnification, on both semi-thin and paraffin sections for a 300~pm length along the mitral cell layer. The interval between two successive observed paraffin sections was higher than 30 ,um. Sixty counts were carried out for both normal mice and sfuggever mutants.

data at the glomerular

observed sections was higher than 20 pm. Given numbers correspond to a 25000 pm3 volume.

Results are given with the standard deviation and the statistic significance of all variations was determined by

Glial cells were counted directly under the microscope on serial transverse paraffin sections and the number of cells entered in the NBC1 picture logiaal which was used to determine the section area. The counted area was registered by a camera and the pictures normalised (contrast, brightness) then digitalised to 256 grey levels. 90 counts were carried out for both normal and stuggerer mice, the interval betweeen two consecutive paraffin sections being higher than 15 pm.

Morphometric

olfactory bulbs. The interval between two successive

Statistics

studies

Glial cell counts in nervous layer

Mitral

Granular cell counts Granular cells were counted under photonic microscopy at x 4.0 magnification, on various sections of adult male

level

To compare the juxtaglomerular peripheral cell number and the glomerular size in normal and stuggerer mutant male mice, an automatic analysis on a Biocom 500 analyser was performed. For each glomerulus, the dimensions were measured at the place where the diameter reached its maximum size. Countings were repeated on 20 glomeruli for 10 animals in each group. The three components of glomerular neuropils: sensory terminal axons, dendritic processes arising from mitral, tufted and ,iuxtaglomerular cells, as well as glial cell processes, were easily recognised under TEM, due to the fact that electron transparence increased from nerve layer (dark11 to glial processes (pale). We used these characters to perform an automatic morphometrical analysis, with the help of the Biocom 500 analyser and Mima and Mjmal logicials (1991). Moreover, we

designed a specific automatic program suitable to our problem in which the used pictures were the TEM negative films. Each registered image was normalised (contrast increased, brightness) and digitalized to 256 grey levels. The adequate thresholds were established after numerous assays, and checked on 10 cases for each animal group for the quantitative analysis was based on automatic contours, and area estimations. The good correspondence between these automatically determined areas and the three desired components was checked on th.e original photographs, and allowed us to validate the method. The procedure has been carried out on 30 photographs (X 5000) randomly chosen in each experimental group (staggerer and control mice). The mean percentage of the three components was calcu-

lated on the samearea (138pm*). Modifications in olfa,ctory bulb of the staggerer mutant mouse

the Anova Fisher test (variance

analysis)

(Scherrer,

1984).

RESULTS Laminar structures of mice olfactory bulbs: comparison between mutants and controls (table I) All mice bulb slice observations showed the six well described olfactory bulb layers (Farbman, 1992): nerve, glomerular, external plexiform, mitral, internal plexiform and granular (fig lE, F). In the

mutant, three layers appeared thinner than in normal animals: the glomerular, and the two plexiform zones (fig ZE). As a result, a slight decrease in the bulb volume in mutant was often perceptible. Thus our quantitative and comparative studies were founded on an evaluation of the cell number for the same units of area or of volume in mutant and control

mice.

To discriminate

as closely

as possible

putative differences in cell density between the two groups, we focussed our numeration on the main olfactory bulb functional levels, ie the glomeruli, the mitral and the granular cells. Histologically, it was quite impossible to distinguish small neurons nuclei from those of glial cells. Therefore, glial cells were only counted in the nerve layer where they represent the unique cell type. Results obtained for controls (73.28 f 16.78 cells / 25000 ,um3) and mutants (75.03 + 16.42 cells / 25000 prn3) indicated that in the staggerevnerve layer, the number of glial cells, which are mostly ensheathing ones, were similar when compared to control. An estimation of the difference, at the glomerular layer level, was given by morphometric studies of each glomerular unit (fig lA, B). Between mutants and normal mice, measurements of the glomeruli surface revealed a 15% difference for the mean area. In the mutant glomerular area, this decrease was significant (P < 0.05) (table II). This difference was enhanced by a juxtaglomerular cell number (counted on 10 sections) lower than in the normal mouse (330.3 + 25.5 against 370.7 f 35.0), due to the fact that in mutants large intercellular spaces were present between these surrounding cells (fig 1A). Intercellular space increasing is a rather general feature characteristic Monnier et al

32

Biology of the Cell 91 (1999) 29-44

Table I. Comparative table emphasising cytological and morphometrical differences disclosed between the wild-type and staggerer olfactory bulbs.

Nerve layer Glomerular layer

Mitral layer Granular layer (granular cell density (25 000 pm3 + SD))

Glial cell density (25 000 pm3 + SD) OMP labelling GFAP labelling Juxtaglomerular cell number Diameter of glomeruli (pm + SD) Surface of glomeruli (pm2 + SD) OMP labelling GFAP labelling mitral cell number (= 300 pm1 Deeper zone

Mutant

Wild-type

75.03 + 16.42 +-I-++ ++++* 330.3 f 25.5’ 103.3 f 13.2* 840.6 + 22.2* ++++ ++++ 21.8 + 6.0* 105 + 8.4*

73.28 * 16.78 ++++

782 11 ++*

68 + 7.9* ++++*

Upper zone Astrocyte GFAP labelling

370.7+:35.0* 111.6 f 14.8*

994.6 zt 26.2* ++++ * 41.3+:7.6* 65 f 8.9*

* Significant difference.

Table II. Glomerular diameters and areas. Glomeruli

(10 mice) Normal (n = 10) Mutant (n = 10)

Diameter (ym + SD1

Area (pm2 + SD)

111.6 * 14.8 103.3 * 13.2

840.6 f 22.2

994.6 2 26.2

of staggerer olfactory bulb structure and is most prominent in the external plexiform layer (fig 1E). It was at the level of mitral cells, the main output neurons, that the differences between mutants and controls were the most obvious. The mitral cell layer, characterised by a quite continuous line of tightly arranged soma (fig lD), appeared rather discontinuous in mutants (fig 1C). Mitral cell counting revealed in mutants a density lower than in the same area of control animals (table III). This estimation demonstrated the deficiency in mitral cells which occurred in mutants, nearly one half of the deutoneuronal population. This difference was highly significant (P < 0.05). The mitral cell paucity allowed the different deeper associative neurons to extend in the free

large interspaces between mitral cells. For that reason, the internal plexiform layer was less distinct in the mutants (fig 1E). The granular cell layer of mutant mice seemed roughly similar to that of controls. To check that point, measurements of the cellular density of these interneurons were carried out. As it was difficult, in this counting, to discriminate short axon cells from granular ones, the total population of the layer was taken into account. Two zones have been considered: firstly the upper granular zone which contains the mature neurons and still increases in size throughout life (Hinds, 1968); secondly the deeper zone, including the subependymal layer from which probably differentiate new granule cells (Kaplan ef al, 1985) (table IV). Values recorded in the mature granular zone, were quite similar in mutants and controls, the dif-

Table Ill. Mitral cell number (= 300 ,uml. Normal (n = 60) Mutant In = 60)

41.3 -c 7.6 21.8 + 6.0

Fig 1. Micrographs and semi-thin sections of olfactory bulbs. A. In staggerer mouse, the size of glomeruli is smaller, the ) neuropil structure less dense and the periglomerular crown is less regular. B. Control mouse glomerulus: the neuropil displays a denser structure and the peripheral juxtaglomerular cells are arranged in a regular crown. C. Mitral cells are less numerous and their arrangement is discontinuous in staggerer, D. The line of mitral cells appears continuous in normal mice (arrow). E. The laminar structure in staggerer mutant mice. F. The laminary structure in control mice. nl, nerve layer; gll, glomerular layer; epl, external plexiform layer; M, mitral cell; ipl, internal plexiform layer; Grl, granular cell; j, crown of peripherical juxtaglomerular cells; n, glomerular neuropil. Modifications in olfactory bulb of the staggerer mutant mouse

Monnier et al

Modifications in olfactory bulb of the staggerer mutant mouse

Monnier et al

34

Biology of the Cell 91 (1999) 2944

Table IV. Cell number for 25 000 ruma.

Normal (n = 301 Mutant (n = 30)

size. However, in these latter, the full differentiation of the nerve layer was not impaired.

Deeper zone with ependymal line

Upper granular zone

65 f 8.9

68 f 7.9

105 f 8.4

785

11

ference being not significant (P < 0.95). The mean number increase in sfaggerer was probably due to a slightly reduced nuclear size (< 10%). According to the role played in granular cell renewal by some subependymal cells, we have measured the density of presumed cycling cells in both groups, The results revealed an important and significant (P < 0.05%) difference between mutants (105) and controls (65). The general characteristics of mutant olfactory bulb tissues were a rather less dense structure (fig lA, E). The paler appearing regions seemed to correspond both to glial cell process development and to the presence of enlarged intercellular spaces. The other remarkable feature of the staggerer olfactory bulb was the very irregular distribution of cellular deficiencies: some zones displaying a normal appearance were randomly located close to very injured others.

Immunological

features

Glial fibrillaty associated protein (GFAP) Thirty sections for each group were stained with anti-GFAP antibodies. In mutants, GFAP labelling was more intense at the glomerular level in spite of the fact that ensheathing glial cells were mainly located at the olfactory bulb periphery. There was good evidence that glial cell extensions were much more developed in mutants than in controls (fig 2F, E) in glomeruli of which astrocytes were preferentially labelled (fig 2E). Astrocytes appeared far lessnumerous in mutant olfactory bulbs. This change was very clear in the granular layer where the maximal projections of 40 sections, observed in confocal scanning microscopy, revealed an astrocyte network less dense in mutants than in control mice (fig 2C, D).

Ultrastructural characteristics olfactory bulb

of mutant

The various cellular types present in the sfaggevev bulb were identified by comparison with those previously described in other mammals. The given illustrations refer mostly to mutants. Whenever some variations between sfaggerer and normal mice appeared of interest to notice, their description was comparatively presented.

Olfactory marker protein (OMP)

Olfactory nerve layer

Alternate slices were labelled with anti-OMP while, in control slices, this first antibody was omitted. The obtained results remained constant from one sample to another, while in controls labelling was lacking (data not shown). The olfactory nerve, outside and inside olfactory glomeruli, showed an intense fluorescence and no clear labelling difference existed between mutant and control groups (fig 2B, A). The stereo pairs, obtained from thick sections, provided information on the location of this OMP in the whole glomerular sphere, and confirmed that mutant glomeruli displayed a smaller

The thickness of the nerve layer varied from one part of the bulb to another, both in mutant and control mice as reported in other mammals. It consisted in intermingled bundles of unmyelinated axons of the smallest diameter (about 0.3 pm) (fig 3A, B), which extended from the sensory cells present in the nasal mucosa. The only cellular elements were glial cells ensheathing axons bundles and some astrocytes extending between the nerve bundles. The ensheathing cell processes were more abundant in mutants (fig 3B) as a sign of glial cell process’ general development.

Fig 2. lmmunofluorescentstudiesby confocal laser scanningmicroscopy on thick sections(100 pm). A. lmmunodetection ) of OMP in control mice nerve and glomerular layers. Stereo-imagereconstituted from 26 focal planes in 30 pm Z dimension. B. OMP immunodetectionin staggerer nerve and glomerularlayers. The fluorescenceis as intenseas in control animals, but the glomerular size looks smaller.(25 focal planes, Z, 30 pm). C. GFAPimmunodetectionin the control mice granular layer pointing out the density of the astrocyte network. Image reconstituted by linear maximal projection of 25 successivehorizontal sectionsin 40;um Z dimension.D. GFAPimmunodetectionin the staggerer granular layer reveal the less denseastrocyte network (maximalprojection with the sameconditionsas in C). E. GFAPimmunodetectionin control mice nerve and glomerularlayers demonstratingglial processesand especiallyglomerularastrocytes. Stereo-image reconstituted from 26 focal planesin 30;um Z dimension.F. GFAPimmunodetectionin the staggerer nerve and glomerular layers. lmmunostainingis more intensein glomerularneuropiland underlinesthe glial processes(stereo-imageobtained in the sameconditionsas in E). nl, nerve layer; gll, glomerularlayer. Modifications in olfactory bulb of the staggerer mutant mouse

Monnier et al

Modifications in olfactory bulb of the staggerer mutant mouse

Monnier et al

Modifications in olfactory bulb of the staggerer mutant mouse

Monnier et al

Biology of the Cell ‘31 (1999) 29-44

37

Table V. Area ratio for same glomerular components.

Olfactory axon component Dendritic component Glial component

Glomerular

Non-mutant (% area)

Mutant (9/oarea)

33.3 * 9.1 45.2 f 9.4 19.3 f 7.9

35.5 + 6.8 34.5 f 13.3 32.2 + 11.4

layer

The most important variations between mutants and controls were observed at the level of glomeruli, main functi.onal relay between sensory cells and mitral deutoneurons (fig 3C, D). The core of glomeruli just contained a network of nerve endings to which participate sensory terminal axons, dendritic processes from mitral, tufted and juxtaglomerular cells, as well as glial cell processes (fig 3C, D). In the control bulb, some balance existed between these three elements (fig 3C). Contrariwise, in many areas of mutant glomeruli, the usual triad was absent, giving the glomerular core an heterogeneous structure very different from that of control (fig 3D). Glial cell hypertrophy was accompanied by an intracellular organelle paucity which rendered them very transparent to electrons (fig 3D). The morphometric data confirmed these ultrastructural observations (table V). These significant results (P < 0.05 ) confirmed that, in the mutant, olfactive receptors preserved a normal structure while the deuto- and interneuronal relays were markedly reduced (> 20%), a decrease which was compensated by the glial hypertrophy. If the previous quantitative estimations of the juxtaglomerular cells (periglomerular, short-axon, and external tufted cells) have demonstrated a decrease in the peripheral cell population, in the mutant the ultrastructural observations did not reveal dramatical modifications in the neuron cell types. Arranged in a crown shape,. the three interneuron types of the mutant glomerul:i displayed all characteristics usually described in previously studied mammals. The periglomerular cells were the smallest (about 7 ,um in diameter), rounded in cross section with a reduced perikaryon (fig 4A).

4

The external tufted cells appeared more fusiform and their average size was about 12 pm; their larger cytoplasmic compartment contained numerous cell organelles, including rough endoplasmic reticulum, mitochondria and a prominent nucleolus (fig 4B). The third type of peripheral glomerular neurons was represented by the superficial short-axon cells displaying a diameter of about lo-12 pm, a usually spherical nucleus with homogeneous chromatin and sometimes indentations of the nuclear envelope (fig 4B).

External

plexiform

layer

This layer of synaptic integrations, involved in olfactory information processing, displayed in mutants the same organisation than in control mice. It was especially characterised by an important fibrous network comprising the lateral (secondary) dendrites of mitral and tufted cells which could be recognised by their relatively large size, the regular arrangement of their microtubules and the abundance of mitochondria in all neuronal processes (fig 4D). In mutant, the weaker density of the secondary dendrites was evident even at the ultrastructural level. The cellular bodies observed in this layer were those of different types of tufted cells, originally described from their position (external, middle, internal). These latter, referred to as displaced mitral cells, displayed similar features: abundant euchromatin, prominent nuclear pores, RER and mitochondria (fig 5A). The astrocyte network was also extensive in this zone. These glial cells could be identified by their morphology, by the presence of intermediary filament fascia labelled by an anti-GFAP antibody and often by their pinocytotic vesicles (fig 4D). The intermediary (GFAP) filaments were running in the processes extended between the fibrous network (fig 5A) which contacted capillary endothelial cells, and some mitral cell dendrites (fig 4D). However, more than anywhere else in the olfactory bulb, observations of the mutant plexiform layer ultrastructure revealed a structural impairment. This layer appeared very heterogeneous: quite normal areas (fig 4D) were located close to others fitted with degenerating cells and large empty intercellular spaces. Numerous astrocytes surrounded by a

Fig 3. Ultrastructural comparison between control and mutant olfactory bulbs. A. The nerve layer of control mice contains many unmyelinated axons. The ensheathing cells appear fusiform. B. The nerve layer of staggerer mice displays similar features, but glial processes appear more developed. C. Control mice glomerular neuropil; dark appearance of the nerve component, transparency of the glial processes to electrons and rather reduced density of mitral primary dendrites. D. The staggerer glomerular neuropil exhibits a very different aspect: glial processes appear enlarged and paler. E, ensheathing glial cell. On, olfactory nerve: G, glial processus; Dl, mitral primary dendrite. Modifications in olfactory bulb of the staggerer mutant mouse

Monnier et al

38

Modifications in olfactory bulb of the staggerer mutant mouse

Biology of the Cell

91 (1999) 29-44

Monnier et al

Biology of the Cell !31 (1999) 29-44

recognisable basement membrane contained a number of lysosomes, a sign of their phagocytotic activity (fig 4C).

Mitral

cell layer

Even if the mitral cell layer was discontinuous in the mutant olfactory bulb, each mitral cell was easily recognisable in TEM (fig 4E). It comprised the largest of the output neurons of the bulb, reaching approximately 20-30 pm in diameter. A large nucleus, with abundant euchromatin occupied the center of the cell body. The nuclear activity of this cellular type was assessed by a prominent nucleolus with both fibrillar and granular components, and conspicuou,’ = nuclear pores (fig 4E). The particularly well developed cytoplasmic compartment contained numerous mitochondria, and was filled by RER whose abundant cisternae surrounded the nucleus. The ax’ons of mitral cells present in the deeper zone were myelinated. In the neighbouring, displaced mitral or tufted cells could only be recognized by their slightly smaller size, for their ultrastructural features were similar to those of regular mitral cells (fig 5A).

Granular

cells

In mutants, the whole granular cell population was arranged in sevleral layers or clusters (fig 1E) of anaxonal interneurons exhibiting smaller cell bodies (< 10 pm). Their nuclear features varied along a reduced range which could depend on the sectioning angle, and/or on the presence of short-axon cells, a fact difficult to unequivocally distinguish (fig 4F). Two subtypes of granular cells (Struble and Walters, 1582) were described in the rat while, according to another nomenclature (Blanes, Golgi) (MC Lean and Shipley, 1992) short-axon cells were sometimes considered as also including the granular cell population. However, in the stuggerer bulbs, the nuclear size of granular cells seemed slightly smaller than in control mice (approximately 6%); small clusters of heterochromatin were dispersed in the nucleoplasm and somewhat concentrated just beneath the nuclear envelope, while the nucleolus was highly condensed. In the mutant, the granular

4

39

layer exhibited the most frequently modified cells with pycnotic nuclei, ruptures of membrane, more or less developed vacuolizations leading to an almost complete disappearance of their cytoplasmic compartment (fig 4F). All these events gave the mutant olfactory bulb structure a loose and highly representative cytoarchitectony. These alterations were not supposed to be due to fixation defects or to artifacts for, in control bulbs, treated by the same procedures, similar features were occasionally seen but only in a reduced number of cases.

Synapses In staggerer, the specific arrangement of glomerular neuropil and its deficit in primary mitral dendrites made difficult to find axo-dendritic fundamental synapses, between proto and deutoneurons. On the contrary, dendro-dendritic asymmetrical synapses, either alone or reciprocal (fig 5B, C) were the most abundant. So, in the plexiform layer, presynaptic vesicles of a rather spherical type were seen on the mitral dendrites (fig 5C), while presynaptic vesicles of rather flattened and irregular type were observed on interneuron dendrites (granule cell for example) (fig 5D). The contacts between accessory dendrites of mitral cells and spines of granular cells displayed a fuzzy appearance particularly developed at the level of the postsynaptic membrane as was the case for an asymmetrical synapse (fig 5D). Some symmetrical synapses and punctia adherentia might also be observed (fig 58). The organisation of synapses in the glomerular zone and in gemmules of periglomerular cells was similar to previous descriptions.

DISCUSSION Cytological features of the staggerer olfactive bulb

mutant

Our cytological observations throw some light on the stnggerer mutant olfactory bulb organisation at the cellular level, a prerequisite for further studies devoted to the functional analysis of this model. They also contribute to a better knowledge of adult

Fig 4. Ultrastructure features of main cellular types in staggerer olfactory bulbs. A. Periglomerular cells: the smallest interneurons. B. Two types of neurons can be observed in the discontinuous cell crown. The short axon cells display an homogeneous nucleus. The external tufted cells, bigger than the two other cell types. C. An astrocyte containing a number of lysosomes, a sign of their phagocytotic activity. D. Some areas in the external plexiform layer appear unmodified. An astrocyte makes contact with a secondary dendrite from a mitral cell. E. A metabolically active mitral cell: nucleus with abundant euchromatin, prominent nucleolus, conspicuous nuclear pores, dispersed RER cisternae, numerous mitochondria. F. The granular cell population is not modified, but some degeneration signs often appear: pycnotic nucleus (short arrow), cytoplasmic compartement disappeared (large arrow). T, tufted ceil; Sac, short-axon cell; P, periglomerular cell; C, capillary; L, lysosomes; N, nucleus; Nu, nucleolus; np, nuclear pores; RER, rough endoplasmic reticulum; mi, mitochondria; D2, mitral secondary dendrite; Gr, granular cell; A, astrocyte; n, glomerular neuropil. Modifications in olfactory bulb ot the staggerer mutant mouse

Monnier et al

40

Modifications in olfactory bulb of the staggerer mutant mouse

Biology of the Cell 91 (1999) 29-44

Monnier et al

Biology of the

Cell ‘91 (1999) 29-44

mutant and normal mice olfactory bulb structure. In fact, previous works, from which we started the present study, were essentially devoted to other mammals (mostly rat: Farbman, 1992; MC Lean and Shipley, 1992) or to mouse fetal development (Doucette, 1989), mitral cell differentiation (Hinds, 1972), synaptogenesis (Hinds and Hinds, 1976; MarinPadilla and Amieva, 1989) or to glial cells (Doucette, 1984, 1989,1993). In the mutant olfactory bulb, the most striking modifications we observed are summarized in table I and were the following: i) a decrease in the number of neurons, and periglomerular interneurons, especially pronounced for mitral cells (-45%) first functional relay in the olfactory bulb; and ii) changes in the granular zone character&d by an abundance of glial cell process extension mostly from ensheathing cells and an important reduction of the astrocyte network. As a consequence, this decrease in cell density leads to a prominent reduction in the glomerular and in both external and internal plexiform layers together with a decreased bulb size but without any structural impairment in the general organisation of the bulb. However, a set of deficiencies linked to this mutation is well. known and their appearance follows a precise cihronology. .For our own study, we choose to discri.minate mutation effects on young adults male olfactory bulb as compared to controls reared in the same conditions. A somewhat identical reduction in the number of the main neurons associated to a diminished organ size but without disturbance in the topographical organisation of the bulb, was already described for other mutants like reeler (Wyss ef al, 1980) or PCD (Greer and Sheperd, 1982) as well as for cerebellum and olivocerebellar projections in sfaggerer (Blatt and Eisenmann, 1989). And it is interesting to note that ageing equally induces a volume decrease and a reduction in mitral cell number, for example in rats (Hinds and MC Nelly, 1977) and in the human species (Bhatnagar ef aI, 1987). Neuron number reduction gives the mutant olfactory bulb tissue, as a whole, a looser and modified architectony which is striking on ultrastructural observations; tog;ether with the presence of degenerating cells, they explain the observed decrease in size. Our quantitative data indicated an average reduction of 15% for glomeruli size. Of course, variations of the glomeruli size take place along the rostro-caudal axis of mouse olfactory bulb (Royet et al,

4

41 1988), but these authors specified that the profile density (size x number) remained homogenous in the whole bulb. In the sfaggerer, the profile density was smaller than in normal mice, the whole thickness of the glomerular layer being reduced. Also to be noticed, is the reduction of the mitral dendritic trees in the external plexiform layer, which gives the sfaggerer the originality to display at the same time a decrease in deutoneurons as well as in primary and basal dendrites together with a reduction in the periglomerular cell number. These periglomerular cells are lacking in fishes, while in the course of mitral cell evolution the primary dendrites decrease when basal ones progressively emerge (Andres, 1970; Dryer and Graziadei, 1994). The fact that, in mutants, more numerous cells present in the deeper ependymal zone yield, in the upper granular zone, a number of cells identical to that of controls, may be explained by a higher percentage of cell degeneration during cell differentiation and migration from one layer to the other. Cell death occurring at a higher rate in mutants will stimulate cell division in the ependymal cell layer resulting in a higher cell density at this level.

Development mutant

of glial cell processes in the

As it may be expected, the lack of neurons is partly compensated by the development of glial cell processesemanating from ensheathing cells, the number of cell bodies in mutants being very close to that of controls. This is particularly clear at the glomerular level as demonstrated by our morphometrical ultrastructural and GFAP immunolabelling studies. In olfactory bulb, GFAP is present both in astrocytes and ensheathing cells which differ from Schwann cells (Barber and Lindsay, 1982; RamonCueto and Valverde, 1995) while other proteins (vimentin, laminin, connexin 43 ) characterise only the ensheathing cells (Miragall et al, 1992; RamonCueto and Valverde, 1995).

Is the lack of alterations in the staggerer nervous layer linked to its developmental

olfactory origin?

Our immunohistochemical observations using OMP as a differentiation marker (Farbman and Margolis, 1980; Monti-Graziadei ef al, 1980) demonstrate that the nervous layer is well preserved in mutants. Both

Fig 5. Synapsesin staggerer olfactory bulbs.A. An internaltufted cell whose cellularintegrity is well preserved. An astrocyte cell process c:ontainingintermediary filament bundlesis runningbetweenthe fiber network. B. Asymmetrical, reciprocal t-4 and punctia adherentia(4 synapsesin the fiber network of the external plexiform layer. C. Dendrodendritic synapses in the external plexiform layer. Some secondary dendrites of mitral cells may be observed. D. Dendro-dendritic synapsesin the gr’snularlayer: asymmetrical(4 synapsebetween secondarydendrites of mitral cell and spinesof granular cells. Ap, astrocyte process; S, synapse. Modifications in olfactory bulb of the staggerer mutant mouse

Monnier et al

42

ensheathing cells, which are specific to olfactory bulb, and the sensory receptive cells derived from the embryonic nasal placode, belong to the same lineage than Schwann cells, whereas astrocytes arise from another progenitor (Doucette, 1989, 1993; Chuah and Au, 1991; Pixley, 1992). Astrocytes and neurons originate from the central nervous system and their settlement takes place by successive waves and migrations bulb (Chalmers et al, 1996).

Heterogeneity of the staggerer olfactory bulb structure. A comparison with genetic cerebellar alterations The staggerer mutation could be due to a genomic deletion affecting the common coding region of RORa four isoforms generated by differential splicing (Matysiak-Scholze and Nehls, 1997). Mutants therefore exhibit a defect in neurons and astrocyte number in their olfactory bulb, the inactivation of the RORa gene having apparently the same consequences there than in cerebellum. During rodent brain postnatal development, RORa could be involved, by means of these four different isoforms encoded, in the regulation of specific types of neuron maturation, especially in the olfactory bulb (Sashihara et al 1996). Two of these isoforms, RORal and a4, appear to be required for Purkinje cell development, a fact which plainly explains cerebellar impairment in staggerer mutants (Matysiak-Scholze and Nehls, 1997). This mutation is also associated to an hyperexpression of IL-1 -A expression leading to an increased production of this cytokine and to macrophage hyperexcitability (Kopmels et al, 1991; Bakalian ef aI, 1992; LemaigreDubreuil et al, 1996). We regularly noticed in the mutant bulb that the deficiency, instead of being homogeneous and uniform, affects more heavily some particular zones, giving this part of the brain a mosaic pattern. This is to bring together with reports establishing that in sfaggerer cerebellar Purkinje cells, various proteins exhibit either a similar expression pattern (ie integrin pl subunit: Murase and Hayashi, 1996) or regional variations (ie calbindin and II’3 type 1 receptor mRNA: Nakagawa et al, 1997). In the mouse, a regional heterogeneous expression of the RORa gene was particularly reported for thalamus, cerebellum and olfactory bulb (Matsui et al, 1995). These authors suggest that the encoded protein acts as a transcription factor not only in neuronal lineage differentiation, but also in mature brain, a very suitable hypothesis for the mutant olfactory bulb.

Cell adhesion molecules may be implied in staggerer olfactive bulb impairment The role played by the N-CAM adhesion molecules, and especially by their polysialylated counterparts (Rousselot et al, 1995), during neuron migration in Modifications in olfactory bulb of the staggerer mutant mouse

Biology of the Cell 91 (1999) 2944

the olfactory bulb which occurs in the fetal life (Hinds, 1968) and in adult neurogenesis originating from the ventricular subependymal zone (Gobetto et aZ, 1995), is well known. Deletion of the N-CAM-180 exon in transgenic mice induces in their olfactive bulb a reduction in granule cells number and an accumulation of their precursors in the subependyma1 zone (Tomasiewicz et al, 1993). This may be explained by the loss of the polysialylated form of this N-CAM which is necessary for cell migration. The fact that in staggerer cerebellum maturation deficiencies could imply a lack of cadherins (BahjaouiBouhaddi et al, 1997) justifies a look at an identical absence of these molecules in the olfactory bulb. Results obtained by cell numbering in the granular layer sustain our opinion that mitotic abilities are still present in the subependymal zone cells and that impairments in the cell migration pathway may be as responsible as neuron degeneration for the defective granular neuron’s homing.

Relationship between the olfactive bulb structure and electrophysiological data In fact, olfactory electrocorticograms of mutants yielded bursts of potentials far less frequently than in normal mice but of a longer duration. Evoked potentials induced by odours display a longer latency and a longer duration while in most cases, the late phase and late oscillatory potentials are lacking. These variations may surely be correlated with a glomerular structural deficiency, a decrease in post-synaptic elements, and the absence of a late phase in the evoked potential may be directly related to degeneration and lack of mitral cells. Additionally, it would be of interest to complete these studies by a look upon the malleable synapses with the 5’ nucleotidase enzyme as a marker (Schoen and Kreutzberg, 1995), to quantify their extension but it was not our purpose in this first approach. Our ultrastructural investigations establish that, in mutants, synapses belonging either to the spherical synaptic vesicles type (type I, stimulatory, according to Gray, 1959) do persist for mitral cells, or to the flattened vesicle type (type II, inhibitory, according to Gray, 1959) for interneurons. Their characteristics correspond to functional records or to physiological features. These synapses preserve in the mutant olfactory bulb functional capacities since the dendritic spines are considered as multifunctional integrative units underlying odour learning and memory (Shepherd, 1996).

The staggerer mutation: a good model for olfactory studies Finally, the major interest of the olfactory system is that its functional organisation is still to be underMonnier et al

Biology of the Cell 91 (1999) 29-44

stood (Shipley and Ennis, 1996). These authors remind that many critical gaps are still existing in our knowledge of this integrative structure. So, the major aim of our work was to give the functional alterations previously recorded in staggerer olfactory bulb (Math et al, 1995) a structural explanation and to deduce how the olfactory system may function This relationship between structure and function confirmed the idea that the staggerer mutant studied in comparison with normal mouse, offers a very good natural model to solve thoroughly the enigma of how olfactory messages are learnt and kept into memory and how this may be compatible with the plasticity of this unique organ. Some changes associated with this mutation have also been reported in ageing (Zanjani et al, 1992) or Alzheimer’s disease (Rezek, 1987). The utilisation of mutant mice may complete other frequently used models like in vivo experimental anosmia induced by ZnSO, treatment. Recent experiments have allowed us to note that staggerer mice, in spite of a sensory alteration by ZnSO,, remained able to localize odours, and that regeneration of epithelial cells occurs there faster than in normal mice (Bensoula et al, 1996). This result is in close agreement with our cytological observations that the nerve layer is not impaired in mutants. Our laboratory is currently running further these investigations w’ith the aid of an in vitro experimental model: olfactory bulb organ and slice cultures.

ACKNOWLEDGEMENTS are grateful to Dr L Astic for her slides on stuggerer mouse olfactory bulbs which were used for cell counting. We also thank Professor D Fellmann who permitted us to use his vibratome equipment. We

REFERENCES Andres KH (1970) Anatomy and ultrastructure of the olfactory bulb in fish, amphibia, reptiles, birds and mammals. ln: Taste and Smell in vertebrates. A Ciba foundation symposium (Wolstenholme GEW, Knight J. eds) Churchill, London, pp 177-194 Bahjaoui-Bouhaddi M, Padilla F, Nicolet M, Cifuentes-Diaz C, Fellman D and Mege RM (1997) Localized deposition of Mcadherin in the g;lomeruli of the granular layer during the post-natal development of mouse cerebellum. ] Comp Neurol378,180-195 Bakalian A, Kopmels B, Messer A, Fradelizi D, Delhaye-Bouchaud N, Wollman E and Manani J (1992) Peripheral macrophage abnormalities in mutant mice with spinocerebellar degeneration. REB ImmunoZl43,129-139 Barber PC and Lindsay RM (1982) Schwann cells of the olfactory nerves contain glial fibrillary acidic protein and resemble astrocytes. Neuroscience 7,3077-3090 Baudoin C, Feron C and Magnusson MS (1991) Male-female interactions in skggerer and non-mutant mice: impairment to react to novelty as a possible explanation of stuggerer male social behavior. Behav Process 24,49-58

Modifications in olfactory bulb of the staggerer mutant mouse

43

Baudoin C, Feron C and Deiss V (1994) Olfactory deficit induced by the murine staggerer mutation: Behavioural data. Adv Biosci 93,413419 Bensoula AN, Math F, Guastavino JM, Cousin X and Davrainville JL (1996) Recuperation fonctionnelle de l’olfaction consecutive a une anosmie experimentale chez la souris normale et mutante staggerer. Sci Tech Anim lab 21,4147 Bhatnagar KP, Kennedy RC, Baron G and Greenberg RA (1987) Number of mitral cells and the bulb volume in the aging human olfactory bulb: a quantitative morphological study. Anat Ret 218,73-87 Blatt GJ and Eisenman LM (1989) Regional and topographic organization of the olivocerebellar projection in homozygous Stuggerer (Sg/sg) mutant mice: an anterograde and retrograde tracing study. Neuroscience 30,703-715 Chalmers GR, Peterson DA and Gage FH (1996) Sprouting adult CNS cholinergic axons express nile and associate with astrocytic surfaces expressing neural cell adhesion molecules. 1 Comp Neural 371,287-299 Chuah Ml and Au C (1991) Olfactory Schwann cells are derived from precursor cells in the olfactory epithelium. ] Neurosci Res 29,172-180 Doucette JR (1984) The glial cells in the nerve fiber layer of the rat olfactory bulb. Anat Ret 210,385-391 Doucette JR (1989) Development of the nerve fiber layer in the olfactory bulb of mouse embryos. I Comp Neural 285, 514-527 Doucette JR (1993) Glial cells in the nerve fiber layer of the main olfactory bulb of embryonic and adult mammals. Microsc Res Tech 24,113-130 Dryer L and Graziadei PPC (1994) Mitral cell dendrites a comparative approach. Anat Embryo1 189,91-106 Dussault 1, Fawcett D, Matthyssen A, Bader ]A and Giguere V (1998) Orphan nuclear receptor ROR alpha-deficient mouse display the cerebellar defects of Staggerer. Mech Dev 70, 147-153 Farbman AI (1992) Olfactory bulb. In: Cell biology olfaction (Barlow PW, Bray D, Green PB and Slack JMW, eds) Cambridge University Press, New-York Cambridge, pp 132-163 Farbman AI and Margolis FL (1980) Olfactory marker protein during ontogeny: immunohistochemical localization. Dev Biol74,205-215 Feron C and Baudoin C (1992) Reactions of Staggerer and nonmutant male mice to female urine and vaginal secretions odors. Behav Process 27,165-170 Gobetto A, Aimar I’, Bonfanti L, Ghidella S, Lossi L and Merighi A (1995) Cell proliferation in the post-natal and adult mammalian central nervous system. Inf J Anat Embryo1 100, 167-175 Gray EG (1959) Axo-somatic and axo-dendritic synapses of the cerebral cortex and electron microscopic study. ] Anaf 93, 420433 Greer CA and Shepherd GM (1982) Mitral cell degeneration and sensory function in the neurological mutant mouse Purkinje cell degeneration (PCD). Brain Res 235,156-161 Guastavino JM (1982) Sexual experience and successful mating in sfaggerer mutant mice. Behav Process 7, 183-188 Guastavino JM and Larrson K (1992) The staggerer gene curtails the reproductive life span of females. Behav Genetics 22, 101-112 Hamilton BA, Frankel WN, Kerrebrock AW, Hawking TL, Fitzhugh W, Kusumi K, Russell LB, Mueller KL, Van Berkel V, Birren BW, Kruglyak L and Lander ES (1996) Disruption of the nuclear hormone receptors ROR alpha in Staggerer mice. Nature 379,736739 Hinds JW (1968) Autoradiographic study of histogenesis in the mouse olfactory bulb. II Cell proliferation and migration. ] Camp Neurot 134,305-322 Hinds JW (1972 ) Early neuron differentiation in the mouse olfactory bulb. II. Electron microscopy. 1 Comp Neural 146, 253-276 Hinds JW and Hinds PL (1976) Synapse formation in the mouse olfactory bulb. II. Morphogenesis. I Comp Neurol 169, 41-62

Monnier et al

Biology of the Cell 91 (1999) 2944

Hinds JW and MC Nelly NA (1977) Aging of the rat olfactory bulb: growth and atrophy of constituent layers and changes in size and number of mitral cells. ] Comp Neural 171, 345-368

Kaplan MS, dynamics tory bulb. Kopmels B, Wollman stimulated Lurcher 345-353

MC Nelly NA and Hinds JW (1985) Population of adult formed granule neurons of the rat olfacJ Camp Neurol239,117-125 Mariani J, Taupin V, Delhaye-Bouchaud N and EE (1991) Differential IL-6 mRNA expression by peripheral macrophages of Staggever and cerebellar mutant mice. Eur Cytokine Netw 2,

Lemaigre-Dubreuil Y, Brugg B, Chianale C, Delhaye-Bouchaud N and Mariani J (1996) Overexpression of interleukin 1 beta converting enzyme mRNA in Stqgerer cerebellum. Neuroreport 7,1777-1780 MC Lean JH and Shipley MT (1992) Neuroanatomical substrates of olfaction. II?: Science of olfaction (Serby JM and Chobor KL, eds) Springer Verlag, New-York, pp 148-171 Margolis FL (1972) A brain protein unique to the olfactory bulb. Proc Nut1 Acad Sci USA 69,1221-1224 Marin-Padilla M and Amieva MR (1989) Early neurogenesis of the mouse olfactory nerve: Golgi and electron microscopic studies. J Comp Neural 288,339-352 Math F, Baudoin C, Feron C, Vallat F, Guillermoz I’, Astic L and Bride M (1995) Modifications de l’electrocorticogramme du bulbe olfactif et des potentiels evoques transglomerulaires chez la souris mutante cerebelleuse Staggerer. CR Acad Sci 318,843-849 Math F, Greer CA and Sheperd GM (1997) Determinants of olfactory system organization: view from neurological mutations in rodents and genomic maps. Trends Neurosci, in press Matsui T, Sashihara S, Oh Y and Waxman SG (1995) An orphan nuclear receptor, mRORalpha, and its spatial expression in adult mouse brain. Bruin Res Mel Brain Res 2, 217-226 Matysiak-Scholze U, Nehls M (1997) The structural integrity of ROR alpha isoforms is mutated in Sfaggerer mice: cerebellar coex@ession of ROR alpha 4. Gene&s 43,78-84 Miraeall F. Hwane TK. Traub 0, Hertzbere EL and Dermietzel Ry1992) Expr&ion of connexins in the developing olfactory system of the mouse. J Comp Neural 325,359-378 Monti-Graziadei GA, Stanley RS and Graziadei PPC (1980). The olfactory marker protein in the olfactory system of the mouse during development. Neuroscience 5,1239-1252 Murase S and Hayashi Y (1996) Expression pattern of integrin beta 1 subunit in Purkinje cells of rat and cerebellar mutant mice. J Comp Neural 375,225-237 Nagakawa S, Watanabe M and Inoue Y (1997) Prominent expression of nuclear hormone receptor ROR alpha in Purkinje cells from early development. Neurosci Res 28,177-184 Pinching AJ and Powell TPS (1971a) The neuron types of the glomerular layer of the olfactory bulb. J Cell Sci 9,305-345 Pinching AJ and Powell TPS (1971b) The neuropil of the glomeruli of the olfactory bulb. J Cell Sci 9,347-377 Pinching AJ and Powell TPS (1971~) The neuropil of the periglomerular region of the olfactory bulb. J Cell Sci 9,379409 Pixley SK (1992) The olfactory nerve contains two populations of glia, identified both it1 viva and i?z vitro. Glia 5,269-284

Modifications in olfactory bulb of the staggerer mutant mouse

Price JL and Powell TPS (1970a) The synaptology of the granule cells of the olfactory bulb. J Cell Sci 7, 125-155 Price JL and Powell TPS (1970b) An electron microscopic study of the termination of the afferent fibres to the olfactory bulb from the cerebral hemisphere. J Cell Sci 7,157-187 Price JL and Powell TPS (1970~) The mitral and short axon cells of the olfactory bulb. J Cell Sci 7,631-651 Ramon-Cueto A and Valverde F (1995) Olfactory bulb ensheathing glia: a unique cell type with axonal growthpromoting properties. G&r 14,163-I73 Rezek DL (1987) Olfactorv deficits as a neuroloeical sien of dementia of the Alzheimer type. Arch Neural 44:1030-TO32 Rousselot I’, Lois C and Alvarez-Buylla A (1995) Embryonic (PSA)N CAM reveals chains of migrating neuroblasts between the lateral ventricle and the olfactory bulb of adult mice. J Comp Neural 351,51-61 Royet JP, Souchier C, Jourdan F and Ploye H (1988) Morphometric study of the glomerular population in the mouse olfactory bulb: numerical density and size distribution along the rostrocaudal axis. J Comp Neural 270,559-568 Sashihara S, Felts PA, Waxman SG and Matusi T (1996) Orphan nuclear receptor ROR alpha gene: isoform-specific spatiotemporal expression during postnatal development of brain. Bruin Res Mol Bruin Res 42,109-117 Scherrer B (1984) Biostnfistique. Gaetan Morin (eds) 850 p Schoen SW and Kreutzberg GW (1995) Evidence that 5’ nucleotidase is associated with malleable synapses: an enzyme cytochemical investigation of the olfactory bulb of adult rats. Neuroscience 65,37-50 Shepherd GM (1996) The dendritic spine: a multifunctional integrative unit. J Neurophysiol75,2197-2210 Shipley MT and Ennis M (1996) Functional organization of the olfactory system. J Neurobiol30, 123-176 Sidman R, Lane P and Dickie M (1962) Staggerer: a new mutation in the mouse affecting the cerebellum. Science 137, 610-612 Sptirr AR (1969) A low viscosity epoxy resin embedding medium for electron microscoov. I Ultrastruct Res 26,3143 Steinmayr M, Andre E, Conqudt’F; Rondi-Reig L, DelhayeBouchaud N, Auclair N, Daniel H, Crepe1 F, Mariani J, Sotelo C and Becker-Andre M (1998) Staggerer phenotype in retinoid-related orphan receptor alpha-deficient mice. Proc Nat1 Acad Sci USA 95,3960-3965 Strubble RG and Walters Cl’ (1982) Light microscopic differentiation of two populations of rat olfactory bulb granule cells. Brain Res 236, 237-251 Tomasiewicz H, Ono K, Yee D, Thompson C, Goridis C, Rutishauser U and Magnuson T(1993) Genetic deletion of a neural cell adhesion molecule variant (N-CAM-180) produces distinct defects in the central nervous system. Neuron 11,1163-1174 Wyss JM, Stanfield BB and Cowan WM (1980) Structural abnormalities in the olfactory bulb of the Reeler mouse. Bruin Res 188,566-571 Zanjani H, Mariani J, Delaye-Bouchaud H and Herrup K (1992) Neuronal cell loss in heterozygous sfaggerer mutant mice: a model for genetic contribution to the aging process. Brain Res 67,153-160 Received

13 April

1998; accepted

25 January

1999

Monnier et a/