- Email: [email protected]
Colocalization of NADPH-diaphorase and acetylcholinesterase in the rat olfactory bulb
Journal of Chemical Neuroanatomy 9 (1995) 207-216
Colocalization of NADPH-diaphorase and acetylcholinesterase rat olfactory bulb C. Crespo, R. Arkvalo, J.G. Brihn,
A. Porteros, LG. Bravo, J. Aijh,
Dpto. Biologia Celular y Patologia. hive&dad
in the
J.R. Alonso*
de Salamanca, E-37007 Salamanea. Spain
Received 23 November 1994; revision received 15 May 1995; accepted 23 July 1995
The colocalixation of NADPH-diaphorase and acetylcholinesterase activities in the rat main and accessory olfactory bulbs has been studied by successive double histochemical staining of the same sections. In the main olfactory bulb, three patterns of glomerular labeling were found: typicalMADPH-diaphorase-positive, typicaliNADPH-diaphorase-negative, and atypical/ NADPH-diaphorase-negative glomeruli. Although both enzymatic activities were present in periglomerular cells and superficial short-axon cells, colocalization of NADPH-diaphorase and acetylcholinesterase was not observed in these neuronal types. By contrast, both enzymes were colocalized in a small subpopulation (less than 3% of NADPH-diaphorase- or acetylcholinesterasepositive cells) of short-axon cells located in the external plexiform layer, internal plexiform layer, granule cell layer, and white matter. In the accessory olfactory bulb, deep short-axon cells were the only neurons where both enzymes were present, and colocahzation of both markers was observed in some of these cells located in the granule cell layer. Keywords: Cholinoceptive; Interneuron;
Nitric oxide synthase; Olfactory system
1. Introduction NADPH-diaphorase (ND) is an oxidative enzyme that can be detected by a colored histoohemical reaction using a water-soluble tetrazolium salt which is transformed into an insoluble blue formazan reaction product (Kiernan, 1990; Alonso et al., 1995a). ND activity has been described in the olfactory system of several species of mammals such as hamster (Davis, 1991), mouse (Kishimoto et al., 1993), hedgehog (Alonso et al., 1995b), and rat (Scott et al., 1987; Davis, 1991; Vincent and Kimura, 1992; Alonso et al., 1993; Porteros et al., 1994), revealing a heterogeneity in the staining pattern among different species. In the rat main olfactory bulb (MOB), ND-staining has been reported in a subpopulation of olfactory fibers, in centrifugal fibers, and in several types of intrinsic neurons. ND-positive cells have been identified as pe.riglomeruIar cells, granule * Corresponding author, Tel.: +34 23 294400 ext. 1854; Fax: +34 23 294549; e-mail: [email protected]. 089l-0618/95/$09.50 0 1995 Ekvier SSDI 0891-0618(95)0082-I
and deep short-axon cells, and external tufted cells (Scott et al., 1987; Alonso et al., 1993), although the presence of ND activity in this latter type has been controversial (Alonso et al., 1993). In the accessory olfactory bulb (AOB), the vomeronasal fibers and glomeruli, and different neuronal types identified as periglomerular cells, granule cells and deep short-axon cells were ND-positive (Porteros et al., 1994). In the rat MOB, the cholinergic system has been studied using histochemical, immunocytochemical, and biochemical techniques for acetylcholinesterase (AChE), choline acetyltransferase (ChAT), and choline@ receptors (Godfrey et al., 1980; Jaffe and Cuello, 1980; Large et al., 1986; Nickel1 and Shipley, 1988; Phelps et al., 1992; Le Jeune and Jourdan, 1991, 1993, 1994). It is currently accepted that the MOB has no intrinsic choline@ elements (Le Jeune and Jourdan, 1991, 1993). Nevertheless, AChE histochemistry demonstrates the existence of a population of AChEpositive neurons in the rat MOB. These AChE-positive cells have been characterized as non-choline@ and
cells, superficial
Science B.V. All rights reserved
208
C. Crespo et al. /Journal of Chemical Neuroanatomy 9 (199.5) 207-216
presumably cholinoceptive neurons (Nickel1 and Shipley, 1988; Le Jeune and Jourdan, 1994). They have been typified as periglomerular cells, superficial and deep short-axon cells, and external tufted cells (Nickel1 and Shipley, 1988; Le Jeune and Jourdan, 1991, 1993, 1994), the same neuronal types that demonstrate ND staining. A deeper understanding of the cholinergic system in the MOB and AOB requires more detailed knowledge of the neurochemical identity of the cholinoceptive cells. Le. Jeune and Jourdan (1994) reported that cholinoceptive neurons in the rat MOB, identified by using AChE histochemistry, are not GABAergic and only a few of them express tyrosine hydroxylase. Their data confirm the previously reported neurochemical heterogeneity of some MOB intemeurons (see Hal&z, 1990, for a review). Both ND and AChE are expressed in similar neuronal populations of both second-order interneurons, superficial and deep short-axon cells, and first-order intemeurons, periglomerular cells. Thus, it is possible that both enzymatic activities are partially or totally colocalized by the same cells or, on the contrary, the AChE-positive and the ND-positive neurons are different neuronal populations. In order to establish the possible expression of ND in AChE-containing cells, the present report analyzes the degree of colocalization of both enzymatic markers in the MOB and AOB, carried out by successive AChE and ND histochemistry in the same sections.
2. Material ad methods Five adult female and five adult male Wistar rats (250-280 g body weight) were deeply anaesthetized with ketamine (Ketolar, 50 mg/ kg body weight, i.m.). After packing the heads with ice, the animals were perfused through the ascending aorta with 100 ml Ringer solution followed by 500 ml fixative solution containing 4% paraformaldehyde and 15% saturated picric acid in 0.1 M phosphate buffer, pH 7.3 (PB). The brains were removed and the olfactory bulbs were dissected out, postfixed 4 h at 4°C in the same fixative solution, and cryoprotected with 30% sucrose (v/v) in PB at 4°C overnight. Ten- and thirty-micrometer coronal and sag&al sections were cut using a cryostat and serially collected in cold (4°C) PB or in cold 0.1 M sodium acetate buffer, pH 6.0 (AB). The histochemical procedures were carried out on serial free-floating sections. A one-in-four series from each animal was processed for the demonstration of ND activity, a second series was used for AChE histochemistry, and a third series was sequentially processed for both enzymatic activities (double-labeled series). The fourth series was used for the specificity controls.
2.1. NADPH-diaphorase histochemistry Series collected in PB were washed in 0.1 M Tris-HCl buffer pH 8.0, and processed for ND histochemistry as described elsewhere (Scherer-Singler et al., 1983; Alonso et al., 1995a). Briefly, sections were incubated for 60-90 min at 37°C in an incubation solution made up of 1 mM reduced @-NADPH (Sigma #N1630), 0.3 mM nitro blue tetrazolium (Sigma #N6876) and 0.08% Triton X-100 in 0.1 M Tris-HCl buffer, pH 8.0. The course of the reaction was controlled under the microscope. When the histochemical reaction was concluded, sections were washed in 0.1 M Tris-HCl buffer, pH 8.0. 2.2. Acetylcholinesterase histochemistry Series collected in AB were processed for the AChE histochemical technique using a modification (Hedreen et al., 1985) of the Koelle method (Geneser-Jensen and Blackstad, 1971). Sections were washed in AB and placed in agitation in a freshly prepared incubation medium at room temperature for 30 min. The incubation medium was made up of 1.7 mM acetylthiocholine iodide, 0.49 mM sodium citrate, 2.9 mM cupric sulfate, 1.25 mM potassium ferricyanide and 0.2 mM ethopropazine, an inhibitor of non-specific cholinesterases, in AB. AChE activity was visualized using two different techniques. In the first method, the sections were rinsed, after incubation, in AB (3 x 10 min), treated for 1 min with 2% ammonium sulfide and washed (3 x 10 min) in 0.1% sodium nitrate. The free-floating sections were exposed to 0.05% silver nitrate for 1 min and washed in 0.1% sodium nitrate (3 x 5 min) and PB (2 x 5 min). In the second procedure, the sections were rinsed in 0.2 M Tris-HCl buffer, pH 7.6 (3 x 10 min) and the reaction was visualized using 0.0125% 3,3 ’ diaminobenzidine and 0.003% hydrogen peroxide in 0.2 M Tris-HCl buffer, pH 7.6 under microscope control. The reaction was stopped by rinsing the sections with 0.2 M Tris-HCl buffer pH 7.6 (3 x 5 min) and distilled water (2 x 5 mm). A few histochemical sections were darkened with 0.5% nickel ammonium sulfate in the DAB solution. 2.3. NADPH-diaphorase-acetylcholinesterase
double la-
beling
Using both described ND- and AChE-histochemical protocols, some sections previously processed for ND were stained for the detection of AChE activity. Tests showed that when the ND histochemistry was carried out in sections previously processed for AChE, no ND reaction was detected. When the double staining was carried out beginning with the ND technique, both enzymatic activities were clearly observed. In order to clarify which reactions could affect the ND detection, some assays were carried out. In these tests, ND histochemistry was carried out in sections previously in-
C. Crespo et al. /Journal
of Chemical Neuroanatomy
cubated (30 min) in cupric sulfate, potassium ferricyanide and combination of both reactives. These tests revealed that the cupric sulfate and not the potassium ferricyanide interferes in the ND histochemistry. The ND-, AChE-, and double-stained sections were serially mounted on gelatin-coated slides, dehydrated in ethanol series, cleared with xylene, and coverslipped with Entellan. The intense ND-staining observed in some neuronal types made the detection of double labeling in these cells difficult, since the dark blue formazan obtained could mask the brown reaction product of the AChE histochemistry. For the detection of possible colocalization of both enzymes in these cells, the ND-reaction was stopped in a group of sections when these cells were weakly ND-stained, followed by the AChEhis&hen&al procedure. In some sections, the doublelabeled cells were drawn using a camera lucida, photographed, and the sections treated with dimethylformamide as described (Alonso et al., 1995a). The sections were then re-examined and the cells photographed again (Fig. lg, h). This treatment eliminates the formazan without affecting the AChE reacting product, confirming the presence or absence of double labeling. 2.4. Controls For ND, the following controls were carried out as described (Alonso et al., 1995a): (1) omission of the substrate &NADPH in the incubation media; (2) omission of the chromogen nitro blue tetrazolium; (3) denaturation of the enzyme by heating the tissue at 84°C for 5 min; (4) Overfixation of tissue during 2 weeks in 10% formalin; and (5) substitution of &NADPH by NADP (Sigma #NO505). For the AChE histochemistry, two controls were carried out: (1) omission of the substrate acetylthiocholine iodide; (2) addition of 10” M BW284C51 as inhibitor of AChE activity (Carson and Burd, 1980). For all controls, no residual activity was observed. 2.5. Analysis The number of reactive cells was counted in five animals from the same sex (females), with a 16x planapochromatic objective. In neuronal types that were represented by a high number of positive elements, 10 000 randomly chosen neurons were taken as statistically representative. For neuronal types with less than 40 positive elements per section, all positive neurons were counted. The maximum diameters of the neuronal types were calculated with a digitizer tablet connected to a semiautomatic image analysis system (MOP-Videoplan, Kontron). Only the cells where the nucleus was clearly visible were used for the measurement of cell sizes. In the neuronal types where the dense reaction product obscured the nucleus, only the neurons
9 (1995) 207-216
209
located in the middle of the section and with at least two visible primary dendrites were measured. 3. Results The ND histochemical technique provided a Golgilike staining of positive neurons. No glial cells were labeled. In many ND-stained elements, not only the cell body, but also the dendrites and long portions of the axons were visualized, allowing a good morphologic characterization of these cells (Fig. le). In the AChE histochemistry, by contrast, only the cell bodies and, at most, initial portions of the dendrites of the labeled intrinsic neurons were stained (Fig. lb,d,h). In addition, networks of AChE-positive fibers, presumably centrifugal fibers, were observed in most OB layers (Fig. la). When the reaction was visualized with silver nitrate, a dark fiber staining was observed, whereas cell bodies were weakly labeled. The use of 3,3 ’ diaminobenzidine as chromogen increased the labeling intensity of cell bodies. The staining patterns obtained in double-labeled sections were similar to those observed in the ND- or AChE-single-stained series: the number of ND- or AChEpositive elements, their distribution, and their labeling intensity were not altered in the double-labeled series. In these sections, the blue formazan of the ND histochemistry and the brown reaction product of the normally AChE histochemistry were clearly distinguishable (Fig. 1 a-g). The colocalization of both labels was identified by the presence in cell bodies of both reaction products at the same focusing plane (Figs. lc, f, g) and confirmed by dimethylformamide treatment (Fig. lh). The distribution pattern of both enzymatic activities in the MOB and AOB was similar in males and females. No clear differences were observed in the NDand AChE-stainings nor in the colocalization pattern between animals from either sex. 3.1. Main olfactory bulb (MOB) After both ND and AChE staining, positive cells and/or fibers were found in all layers of the MOB, except in the olfactory nerve layer (ONL) that did not demonstrate AChE reactivity. Both ND and AChE activities showed clear differences across layers (Fig. la). The highest density of ND-positive elements was observed in the granule cell layer (GCL) (Fig. la), where a dense network of positive fibers was stained, and in the deep portion of the glomerular layer (GL), with a high number of ND-positive cells (Fig. la). The NDcontaining neurons were identified as periglomerular cells, granule cells, and superficial and deep short-axon cells (Fig. 2). For AChE, dense plexuses of AChEpositive fibers were found in the deep portion of the GL, in the superficial portion of the external plexiform layer (EPL), and in the internal plexiform layer (IPL) (Fig.
210
C. Crespo et al. / Journal of Chemical Neuroanaromy 9 (1995) 207-216
C. Crespo et al. / Journal of Chemical Neuroanatomy 9 (1995) 207-216
211
COLOC!ALlZATION
AC&El
,.““....
.‘.“.‘..
GL
:
._....,
,.:
;i
‘... :, :.. :.’ .___.. .*:...._..’ ......_..L.....‘........’
EPL
EPLC
.... ..-: ,...:“.. :
:
.:
..: i
i:
:.:
::
: ..
,A. i:
1.:’ .. . .. ‘..j
:
a#
! .i ,.“.. :.:
;.‘.,
:.. : : .: .,j
MCL IPL
EPL
EPL
EPLC
.... :“: .... :“....‘..:...; .. : : : T: : : ; : : : : : ., ..‘., .,..:: ..... :..: .._::.,: :... :. i ; i MCL ..: ..: IPL &e
DSA
GCL
GL
MCL IPL
loC GCL
GCL
A DSA*
WM
DSF
WM
Fig. 2. Schematic drawing showing the acetylcholinesterase-labeled (colocalization) cells in the rat olfactory bulb. DSA: deep short-axon cells; EPL: external plexiform layer; EPLC: external plexiform layer cells: GC: granule cells; GCL: granule cell layer; GL: glomerular layer; IPL: internal plexiform layer; MCL: mitral cell layer; PG: periglomerular cells; SSA: superilicial short-axon cells SSAIETC: superficial sort axon cells and/or external tufted cells; WM: white matter.
la). AChE-labeled cells were identified as periglomerular cells, external tufted cells and superficial and deep short-axon cells (Fig. 2). In the ONL, a subpopulation of ND-positive olfactory fibers was observed. These fibers formed a subpopulation of ND-labeled glomeruli in the GL. These ND-positive glomeruli demonstrated a clear-cut spatial segregation: whereas glomeruli located in the dorsomedial portion of the GL were ND-positive, those of the ventrolateral portion were ND-negative. The AChE histochemistry technique demonstrated a dense network of AChE-positive fibers surrounding all glomemli (Fig. la). The innervation was similar around
ND-positive and ND-negative glomeruli. Nevertheless, in the caudal portion of the MOB a few glomemli showed a remarkably intense and atypical AChEpositive innervation. These atypical glomeruli contained the highest density of AChE-positive fibers in the rat MOB, and they were clearly distinguishable from the typical glomeruli. They showed a clear spatial segregation and were placed in the dorsomedial and ventrolateral portions of the caudalmost sections of the MOB. Although in the dorsomedial portions the atypical glomeruli were close to ND-positive glomeruli, no atypical glomeruli were ND-positive. In the periglomerular region and at the GL/EPL
Fig. 1. AChE-positive (brown) and ND-positive (blue) stainings in the rat olfactory bulb. Scale bar = IO0 Cm (a) and 25 pm (b-h). (a) Overview of a coronal double-labeled section of the rat olfactory bulb showing the different ND- and AChE-staining patterns across layers. Some ND-positive periglomerular cells can be observed (arrows). ONL: olfactory nerve layer; GL: glomerular layer; EPL: external plexiform layer; MCL: mitral cell layer; IPL: internal plexiform layer; GCL: granule cell layer. (b) large AChE-stained cell in the GWEPL boundary (open arrow), and ND-positive periglomerular cells (arrows). (c) Double-labeled cell (open arrow) and AChE-positive fibers (arrows) in the innter portion of the EPL. (d) AChEstained deep short-axon cell (open arrow), and ND-positive granule cell (arrow) in the GCL. (e) ND-labeled deep short-axon cell in the WM. (f) Colocalization of ND and AChE in a deep short-axon cell (open arrow) in the accessory granule cell layer. (g) Colocalization of ND and AChE in a deep short-axon cell of the WM. (h) Same cell as in (g) after treatment with dimethylformamide showing the AChE-labeling.
212
C. Crespo et al. /Journal of Chemical Neuroanatomy 9 (1995) 207-216
Table 1 ND- and/or AChE-positive neuronal types in the rat olfactory bulb Cell type
Staining
Max. diameter pm (mean * S.E.M)
Location
Frequency”
Periglomerular cell
ND AChE ND AChE
8.7 f 0.3 9.2 f 0.4 16.7 ZIZ 0.6 16.2 zt 0.7
GLJEPL GLJEPL GUEPL GUEPL
++++ ++ + ++
ND AChE ND+AChE ND AChE ND+AChE ND
16.2 zt 0.4 16.1 f 0.6 16.0 zt 0.5 17.8 f 0.3 18.2 f 0.2 17.4 f 0.4 7.7 f 0.2
EPL EPL EPL IPL/GCL/WM IPL/GCL/wM IPUGCL/wM MCUIPUGCL
+ + + ++ +++ + ++++
Superficial short-axon cell Superficial short-axon cell and/or external tufted cell External plexiform layer cell
Deep short-axon cell
Granule cell
aFrequency: +: less than 1 cell per section. ++: 1 to 10 cells per section. +++: 10 to 100 cells per section. ++++: more than 100 cells per section.
boundary, populations of periglomerular and superficial short-axon cells were ND or AChE labeled. AChEpositive periglomerular cells were found surrounding the typical and atypical glomeruli. ND-stained periglomerular cells demonstrated one or two dendrites. Bidendritic periglomerular cells innervated one or two adjacent glomeruli. The ND-positive periglomerular cells were homogeneously distributed in the GL, and they surrounded the ND-positive, and the ND-negative typical and atypical glomeruli. Comparing the number of ND-positive and AChE-positive periglomerular cells, ND-positive cells were more abundant (more than 200 cells per 3Oqm section) than AChE-positive cells (less than 10 cells per 30qm section) (Table 1). The two populations of periglomerular cells were different, in that no periglomerular cells colocalized ND and AChE. Close to the periglomerular cells, a few superficial short-axon cells demonstrated ND- or AChE-labeling (Fig. lb). The ratio of ND-positive superficial shortaxon cells to ND-positive periglomerular cells was approximately 1:2000. Of these scarce superficial shortaxon cells, none colocalized both neuronal markers, being therefore two different neuronal populations, as is the case for the periglomerular cells.. In the inner half of the EPL, a few ND- or AChElabeled cells were occasionally observed (Fig. lc). Only one or two proximal dendrites were labeled in these cells, making their classification difficult. We refer to them as EPL cells (Fig. 2). A few of these AChE- or NDpositive EPL cells (less than 3% of both populations) colocalized both enzymes (Figs. lc, 2; Table 1). In the deep layers (IPL, GCL, and white matter (WM)) only one neuronal population, identified as deep short-axon cells, showed both AChE and ND activities (Fig. Id, e). In addition, several granule cells were NDstained (Fig. Id), but all of them were AChEnegative.
Centering on deep short-axon cells, the number of AChE-positive deep short-axon cells (more than 20 cells per 30-pm section) was higher than the number of NDpositive cells (less than 10 per 30-pm section). There was a spatial segregation of both neuronal subpopulations. Although they were present throughout the IPL, GCL, and WM, most AChE-positive cells were mainly located in the IPL and in the outer half of the GCL, whereas the ND-positive cells were more abundant in the inner portion of the GCL and in the WM. Colocalization of both enzymatic activities was found in a few of them (less than 2% of the AChEpositive neurons and less than 3% of the ND-positive cells). The colocalized neurons were present in the IPL, in the GCL, and in the WM (Fig. lg, h), but most of them were located in the IPL and in the superficial portion of the GCL, the same zones where AChE-positive cells were more abundant. 3.2. Accessory olfactory bulb (AOB) ND-positive fibers and cells were found in all layers of the AOB. All vomeronasal fibers and glomeruli were ND-positive. The ND-stained cells were identified as periglomerular cells, granule cells, and deep short-axon cells. After AChE-histochemistry, only scarce neurons in the IPL and in the GCL, identified as deep short-axon cells, were labeled. Colocalization of ND and AChE was observed in a few (less than 3%) deep short-axon cells located in the accessory granule cell layer (Fig. If). 4. Discussion In this report, a colocalization study of ND and AChE activities was carried out in the rat MOB and AOB. The present data confirm previous observations of the distribution patterns of both neuronal markers in the rat MOB and AOB (Godfrey et al., 1980; Scott et al.,
C. Crespo et al. /Journal
of Chemical Neuroanatomy
1987; Nickel1 and Shipley, 1988; Davis, 1991; Le Jeune and Jourdan, 1991, 1994; Vincent and Kimura, 1992; Alonso et al., 1993; Porteros et al., 1994). The main results in the present report are that although both enzymes coexist in the same neuronal types of the MOB and AOB, only a few cells (less than 3% in all cases) identified as EPL cells and deep short-axon cells colocalized both enzymatic activities, whereas for other neuronal types (i.e., periglomerular cells and superficial short axon cells) no colocalization was observed. The degree of colocalization between ND and AChE in the MOB and the AOB is low in contrast to other regions such as the medial septum and diagonal band of Broca that demonstrate an abundant colocalization of ND and choline@ markers (Kitchener and Diamond, 1993). These regions are two of the main sources of cholinergic afferents to the MOB (De Olmos et al., 1978). Therefore, it can be assumed that a large population of centrifugal fibers is both ND- and AChE-positive. However, in our double histochemistry, double labeled NDpositive and AChE-positive afferents were not clearly identifiable. These data suggest that whereas the doublestaining technique provides an accurate somal staining pattern for both enzymatic markers, it is not reliable for the study of colocalization of both enzymatic activities in extrinsic aBerents. We and others have observed that roughly half of all rat olfactory glomeruli are ND-positive (Scott et al., 1987; Davis et al., 1991; Alonso et al., 1993) whereas only a small population of caudal glomeruli have an atypical, dense AChE-positive innervation (Zheng et al., 1987; Zheng and Jourdan, 1988; Le Jeune and Jourdan, 1994). Our results show that after using both neuronal markers, three glomerular staining patterns may be distinguished: typical/ND-positive, typical/ND-negative and atypical/ND-negative glomeruli. The present data demonstrate that the ND-positive olfactory fibers did not innervate the atypical glomeruli in spite of their spatial proximity in the dorsomedial region of the MOB. This fact confirms the topographically specific heterogeneity of the MOB glomeruli. By contrast, the AOB demonstrated more homogeneous staining, in that all vomeronasal glomentli were ND-positive and AChEnegative. The periglomerular cells represent the first-order intemeurons at the glomerular level. A high neurochemical heterogeneity of periglomerular cells has been previously reported (see Hal&z, 1990, for a review). In agreement with this heterogeneity, we described three different populations of periglomerular cells: NDpositive/AChE-negative, ND-negative/AChE-positive, and ND-negative/AChE-negative. The absence of colocalization of both enzymatic activities in the periglomerular cells add both markers to the neuroactive substances such as calbindin D-28k and ND (Alonso et al., 1993) or calbindin D-28k and tyrosine
9 (1995) 207-216
213
hydroxylase (Hal&z et al., 1985) that do not colocalize in these cells. Le Jeune and Jourdan (1994) described that the AChE-positive periglomerular cells in the rat MOB are non-GABAergic and non-dopaminergic. In addition, our results indicate that the AChE-positive periglomerular cells are also ND-negative. ND-positive periglomerular cells were more abundant (more than 200 ND-positive cells per 30-pm section) than the AChE-positive periglomerular cells (less than 10 AChEstained cells per 30-pm section). These results suggest that ND and AChE label distinct modulatory systems at the glomerular level: AChE was expressed by abundant extrinsic centrifugal afferents and a few intrinsic cells whereas ND was mostly expressed by interneurons, a different subpopulation of periglomerular cells, located in this layer. As periglomerular cells, superficial short-axon cells constitute a biochemically heterogeneous neuronal type (Hal&z, 1990). Le Jeune and Jourdan (1994) have demonstrated that the AChE-containing superficial short-axon cells do not correspond to the few GABAergic cells previously described by Mugnaini et al. (1984). They noticed that the cholinoceptive superficial short-axon cells have similar morphological characteristics and distribution as other populations containing neuropeptide Y (Gall et al., 1986; Scott et al., 1987), somatostatin and ND (Scott et al., 1987). Our data demonstrated, however, that there is no coexpression of both AChE and ND by the superficial shortaxon cells. Therefore, despite those previously mentioned similarities, the AChE-containing and the NDpositive superficial short-axon cells are different neuronal populations. Since ND is a selective marker for nitric oxide synthase-immunoreactive neurons in the olfactory bulb (Kishimoto et al., 1993), these results indicate that ND is a specific marker for noncholinoceptive periglomerular and superficial shortaxon neurons and, reciprocally, AChE histochemistry can be used as a selective marker for neurons that do not express nitric oxide synthase in the periglomerular region of the rat MOB. In the inner half of the EPL, a low number of NDpositive, AChEpositive and colocalized neurons have been observed. The weak labelings for ND- and/or AChE make morphological characterization of these cells difficult. Medium-sized neurons in the EPL constitute a very heterogeneous population where at least live neuronal types can be distinguished (Kosaka et al., 1994). According to their location, size, and shape, these EPL cells may correspond to external tufted cells and/or non-periglomerular short-axon cells. Deep short-axon cells are probably the most heterogeneous neuronal population in the vertebrate olfactory bulb (Hal&z, 1990). Le Jeune and Jourdan (1994) reported that no AChEcontaining deep shortaxon cells were GABAergic and only a limited propor-
214
C. Crespo et al. /Journal of Chemical Neuroanatomy 9 (1995) 207-216
tion of them exhibited tyrosine hydroxylase immunoreactivity. In agreement with this biochemical variability, our data indicate that some deep short-axon cells of the MOB and the AOB express ND, some AChE, and a few cells both. According to the morphology and distribution of the colocalized neurons in the MOB, they belong to the four morphological subtypes defined by Ramon y Cajal (1911) and Schneider and Macrides (1978) using Golgi impregnation. The cells located in the IPL and in the superficial portion of the GCL belong presumably to the Ramon y Cajal and/or horizontal cells, and the cells located in the deep portion of the GCL and in the WM, to the Blanes and/or Golgi types. Nitric oxide has been proposed as a presumptive retrograde transmitter between the postsynaptic and the presynaptic neuron (Bredt and Snyder, 1992). The AChEcontaining neurons in the rat MOB are considered as cholinoceptive (Le Jeune and Jourdan, 1994). The neurotransmitter used by cholinoceptive secondary interneurons in the MOB remains unknown (Le Jeune and Jourdan, 1994). Nevertheless, our data support the hypothesis that it is unlikely to be nitric oxide. Reciprocally, the ND-positive interneurons were principally not cholinoceptive. The cholinergic effects in the MOB could take place at two distinct levels, superficial and deep (Le. Jeune and Jourdan, 1994). Elaagouby et al. (1991) proposed the existence of cholinoceptive superficial interneurons projecting to deep layers and influencing granule cell activity. Several granule cells were ND-positive, but it is presently unknown whether the cholinoceptive cells and the ND-positive granule cells are selectively related among them in the same neuronal circuitries. The localization of ND is coincident with that of nitric oxide synthase in the neurons of the MOB (Kishimoto et al., 1993). On the other hand, the AChEcontaining intrinsic neurons in the MOB are considered as presumably cholinoceptive (Le Jeune and Jourdan, 1994). In both the peripheral and the central nervous systems, there are multiple interactions between the production of nitric oxide and acetylcholine. Nitric oxide may modulate the cholinergic transmission at the pre- and/or post-junctional level in the nerve muscle pathway (Baccari et al., 1994), and appears to exert a marked inhibitory effect on acetylcholine release in different regions such as the trachea and taenia coli (Belvisi et al., 1991; Knudsen and T@trup, 1992). In addition, acetylcholine evokes release of nitric oxide from endothelium and, through muscarinic receptors activation contributes to stimulation of nitric oxide formation in the brain (Wiklund et al., 1993). Nitric oxide activates soluble guanylyl ciclase in target cells (Bredt and Snyder, 1992) and cGMP levels may play an important role in olfactory signalling, including recruitment, adaptation and modulation at the olfactory mucosa and
olfactory bulb levels (Breer and Shepherd, 1993). In this sense, periglomerular cells have been proposed as a NO/cGMP system that modulates intraglomerular and interglomerular synaptic integration of sensory inputs (Breer and Shepherd, 1993). Our data demonsrate that in the rat OB, AChE-positive and ND-positive periglomerular cells are different populations. In the rest of the OB, only a low proportion of AChEcontaining neurons express nitric oxide synthase in the MOB. Pinching and Powell (1972) proposed that periglomerular ceils are the main cellular target of the cholinergic afferents in the glomerular layer but Le Jeune and Jourdan (1991, 1993, 1994) observed after AChE histochemistry and ChAT immunocytochemistry that only a subset of periglomerular cells are cholinoceptive. Our observations are in agreement with Le Jeune and Jourdan (1991, 1993, 1994) since only a low number of periglomerular cells, compared with the ND-positive ones, were AChE-stained. Nevertheless, all periglomerular cells are closely surrounded by a dense network of AChEpositive fibers and nitric oxide, the soluble gas synthesized in the ND-positive neurons, diffuses freely across membranes in a reduced area. Then, despite the practical absence of colocalization, both AChE- and ND-positive systems may interact directly and extensively through synaptic and non-synaptic links. Both choline@ and ND-active systems have been associated with neurodegenerative disorders such as Alzheimer’s disease (Sugaya and McKinney, 1994). NDactive neurons are selectively spared in the hippocampal formation (Hyman et al., 1992) while cholinergic neurons degenerate in the nucleus basalis of Meynert in this neuropathology (Big1 et al., 1990). The olfactory system is severely affected in Alzheimer’s disease (Talamo et al., 1991) but there is little information as to the impact on specific classes of neurons. Thus, further study of the heterogeneity of AChE and ND activity in MOB neurons, as described here, might shed some light on the potentially selective susceptibility of some MOB neurons to degeneration in Alzheimer’s disease and related disorders. Acknowledgements
The authors express their gratitude to Dr. Thomas A. Schoenfeld for his valuable comments and suggestions and Mr. J. Marcos for his photographic assistance. This work was supported by grants from the ‘Junta de Castilla y Leon’, , and the DGICyT (93-065 and PB94-13). References Alonso,J.R., Arhalo, R., Brikn, J.G., Garcia-Ojeda, E., Porteros, A. and Aij6n,J. (1995a) NADPH-diaphorase staining in the central nervous system. Neurosci. Protocols (in press).
C. Crespo et al. /Journal
of Chemical Neuroanatomy 9 (1995) 207-216
Alonso, J.R., Amvalo, R., Garcia-Gjeda, E., Porteros, A., Brition, J.G. and Aijon, J. (1995b) NADPHdiaphorase active and calbindin D28k-immunoreactive neurons and fibers in the olfactory bulb of the hedgehog (Erinacew europaeus). J. Comp. Neurol. 35 1,307-327. Alonso, J.R., ArivaIo, R., Porteros, A., B&on, J.G., Lara, J. and Aijbn, J. (1993) Calbindin D28K and NADPH-diaphorase activity are localized in different populations of periglomerular cells in the rat olfactory bulb. J. Chem. Neuroanat. 6, 1-6. Baccari, M.C., Calamai, F. and Staderini, G. (1994) Modulation of chohnergic transmission by nitric oxide, VIP and ATP in the gastric muscle. Neuroreport 5, 905-908. Eklvisi, M.G., Stretton, D. and Barnes, P.J. (1991) Nitric oxide as an endogenous modulator of cholinergic neurotransmission in guinea pig airways. Eur. J. Pharmacol. 198, 219-221. Bigi, V., Arendt, T. and Biesold, D. (1990) The nucleus basalis of Meynert during aging and in dementing disorders. In Brain cholinergic systems (eds Steriade, M. and Biesold, D.) pp. 364-386. Oxford University Press, Oxford. Bredt, D.S. and Snyder, S.H. (1992) Nitric oxide, a novel neuronal messenger. Neuron 8, 3- 11. Breer, H. and Shepherd, G.M. (1993) Implications of the NO/&IMP system for olfaction. Trends Neurosci. 16, 5-9. Carson, K.A. and Burd, G.D. (1980) Localization of acetylcholinesterase in the main and accessory olfactory bulbs of the mouse by light and electron microscopic histochemistry. J. Comp. Neurol. 191, 353-371. Davis, B.J. (1991) NADPH-diaphorase activity in the olfactory system of the hamster and rat. J. Comp. Neurol. 314, 493-511. De Olmos, J., Hardy, H. and Heimer, L. (1978) The afferent connections of the main and accessory olfactory bulb formations in the rat: an experimental HRP-study. J. Comp. Neurol. 181, 213-244. Elaagouby, A., Ravel, N. and Gervais, R. (1991) Cholinergic modulation of excitability in the rat olfactory bulb: effect of local application of cholinergic agent on evoked field potentials. Neuroscience 45, 653-662. Gall, C., Seroogy, K.B. and Brecha, N. (1986) Distribution of VIPand NPY-like immunoreactivities in rat main olfactory bulb. Brain Res. 374, 389-394.
Geneser-Jensen, F.A. and Blackstad, T.W. (1971) Distribution of acetyl cholinesterase in the hippocampal region of the guinea pig. 1. Entorhinal area, parasubicuhnn, and presubiculum. Z. Zellforsch. Mikrosk. Anat. 114, 460-48 1. Godfrey, D.A., Ross, C.D., Hermann, A.D. and Matschinsky, F.M. (1980) Distribution and derivation of cholinergic elements in the rat olfactory bulb. Neuroscience 5, 273-292. Hal&z, N. (1990) The vertebrate olfactory system. Chemical neuroanatomy, function and development. Akademiai Kiado, Budapest. Hal&z, N., Hokfelt, T., Norman, A.W. and Goldstein, M. (1985) Tyrosine hydroxylase and 28k-vitamin D-dependent calcium binding protein are located in different subpopulations of periglomerular cells of the rat olfactory bulb. Neurosci. L&t. 61, 103-107. Hedreen, J.C., Bacon, S.J. and Price, D.L. (1985) A modified histochemical technique to visualize acetylcholinesterase-containing axons. J. Histochem. Cyfochem. 33, 134-140. Hyman, B.T., Manloff, K., Wenninger, J.J., Dawson, T.M., Bredt, D.S. and Snyder, S.H. (1992) Relative sparing of nitric oxide synthase-containing neurons in the hippocampal formation in Alxheimer’s disease. Ann. Neural. 32, 818-820. Jaffe, E.H. and Cuello, AC. (1980) The distribution of catecholamines, glutamate decarboxylase and choline acetyltransferase in layers of the rat olfactory bulb. Brain Res. 186, 232-237. Kieman, J.A. (1990) Histological and hisrochemical methodr. Theory and practice. Pergamon Press, Oxford. Kishimoto, J., Keveme, E.B., Hardwick, J. and Emson, P.C. (1993) Localization of nitric oxide synthase in the mouse olfactory and
215
vomeronasal system: a h&chemical, immunological and in situ hybridization study. Eur. J. Neurosci. 5, 1684-1694. Kitchener, P.D. and Diamond, J. (1993) Distribution and colocahxation of choline acetyhransferase immunoreactivity and NADPH diaphorase reactivity in neurons within the medial septum and diagonal band of Broca in the rat basal forebrain. J. Comp. Neural. 335, l-15. Knudsen, M.A. and Tdtrup, A. (1972) A possible role of the Larginine-nitric oxide pathway in the modulation of cholinergic transmission in the guinea-pig taenia coli. Br. J. Pharmacol. 107, 837-841.
Kosaka, K., Heizmann, C.W. and Kosaka, T. (1994) Calcium-binding protein parvalbumin-immunoreactive neurons in the rat olfactory bulb. Exp. Brain Res. 99, 191-204. Large, T.H., Lambert, M.P., Gremillion, M.A. and Klein, W.L. (1986) Parallel postnatal development of choline acetyltransferase activity and muscarinic acetylcholine receptors in the rat olfactory bulb. J. Neurochem. 46, 671-680. Le Jeune, H. and Jourdan,
F. (1991) Postnatal development of cholinergic markers in the rat olfactory bulb: a histochemical and immunocytochemical study. J. Comp. Neural. 314, 383-395. Le Jeune, H. and Jourdan, F. (1993) Choline&z innervation of olfactory glomeruli in the rat: an ultrastructural immunocytochemical study. J. Comp. Neural. 336, 279-292. Le Jeune, H. and Jourdan, F. (1994) Acetylcholinesterase-containing intrinsic neurons in the rat main olfactory bulb: cytological and neurochemical features. Eur. J. Neurosci. 6, 1432-1444. Mugnaini, E., Wouterlood, F.G., Dahl, A.L. and Gertel, W.H. (1984) Immunocytochemical identification of GABAergic neurons in the main olfactory bulb of the rat. Arch. Ital. BioI. 122, 83- I1 3. Nickell, W.T. and Shipley, M.T. (1988) Two anatomically specific classes of candidate cholinoceptive neurons in the rat olfactory bulb. J. Neurosci. 8, 4482-4491. Phelps, P.E., Houser, C.R. and Vaughn, J.E. (1992) Small choline@ neurons within field of choline@ axons characterize olfactoryrelated regions of rat telencephalon. Neuroscience 48, 121-136. Pinching, A.J. and Powell, T.P.S. (1972) The termnation of centrifugal libres in the glomerular layer of the olfactory bulb. J. Cell Sci. 10, 621-635. Porteros, A., Alonso, J.R., Amvalo, R., Garcia-Ojeda, E., Crespo, C. and Aijon, J. (1994) Histochemical locahxation of NADPHdiaphorase in the rat accessory olfactory bulb. Chem. Senses 23, 1083-1093. Ramon y Cajal, S. (1911) Histologie du Systeme Nerveux de I’Homme et des Vertebres. Maloine, Paris. Scott, J.W., McDonald, J.K. and Pemberton, J.L. (1987) Short axon cells of the rat olfactory bulb display NADPH-diaphorase activity, neuropeptide Y-like inununoreactivity, and somatostatin-like immunoreactivity. J. Comp. Neurol. 260, 378-391. Scherer-Singler, U., Vincent, S.R., Kimura, H. and McGeer, E.G. (1983) Demonstration of a unique population of neurons with NADPH-diaphorase histochemistry. J. Neurosci. Methoa!s 9, 229-234.
Schneider, S.P. and Macrides, F. (1978) Laminar distribution of intemeurons in the main olfactory bulb of the adult hamster. Brain Res. Bull. 3, 73-82.
Sugaya, K. and McKinney, M. (1994) Nitric oxide synthase gene expression in cholinergic neurons in the rat brain examined by comin situ hybridization bined immunocytochemistry and histochemistry. Brain Res. 23, 111-125. Talamo, B.R., Feng, W.H., Perez-Cruet, M., Adelman, L., Kosik, K., Lee, V.M. Y., Cork, L.C. and Kauer, J.S. (1991) Pathological changes in olfactory neurons in Alzheimer’s disease. Ann. N. Y. Acad. Sci. 640, 1-7. Vincent, S.R. and Kimura, H. (1992) Histochemical mapping of nitric oxide synthase in the rat brain. Neuroscience 46, 755-784. Wiklund, C.U., Wiklund, N.P. and Gustafsson, L.E. (1993) Modula-
216
C. Crespo et al. /Journal of Chemical Neuroanatomy 9 (1995) 207-216
tion of neuroeffector transmission by endogenous nitric oxide: a role for acetylcholine receptor-activated nitric oxide formation, as indicated by measurements of nitric oxide/nitrite release. Eur. J. Pharmacol. 240, 235-242.
Zheng, L.M. and Jourdan, F. (1988) Atypical olfactory glomeruh con-
tam original olfactory axon terminals: an ultrastructural horscradish peroxidase study in the rat. Neuroscience 26, 367-378. Zheng, L.M., Ravel, N. and Jourdan, F. (1987) Topography of centrifugal acetylcholinesterase-positive fibres in the olfactory bulb of the rat: evidence for original projections in atypical glomeruh. Neuroscience 23, 1083-1093.
Colocalization of NADPH-diaphorase and acetylcholinesterase rat olfactory bulb C. Crespo, R. Arkvalo, J.G. Brihn,
A. Porteros, LG. Bravo, J. Aijh,
Dpto. Biologia Celular y Patologia. hive&dad
in the
J.R. Alonso*
de Salamanca, E-37007 Salamanea. Spain
Received 23 November 1994; revision received 15 May 1995; accepted 23 July 1995
The colocalixation of NADPH-diaphorase and acetylcholinesterase activities in the rat main and accessory olfactory bulbs has been studied by successive double histochemical staining of the same sections. In the main olfactory bulb, three patterns of glomerular labeling were found: typicalMADPH-diaphorase-positive, typicaliNADPH-diaphorase-negative, and atypical/ NADPH-diaphorase-negative glomeruli. Although both enzymatic activities were present in periglomerular cells and superficial short-axon cells, colocalization of NADPH-diaphorase and acetylcholinesterase was not observed in these neuronal types. By contrast, both enzymes were colocalized in a small subpopulation (less than 3% of NADPH-diaphorase- or acetylcholinesterasepositive cells) of short-axon cells located in the external plexiform layer, internal plexiform layer, granule cell layer, and white matter. In the accessory olfactory bulb, deep short-axon cells were the only neurons where both enzymes were present, and colocahzation of both markers was observed in some of these cells located in the granule cell layer. Keywords: Cholinoceptive; Interneuron;
Nitric oxide synthase; Olfactory system
1. Introduction NADPH-diaphorase (ND) is an oxidative enzyme that can be detected by a colored histoohemical reaction using a water-soluble tetrazolium salt which is transformed into an insoluble blue formazan reaction product (Kiernan, 1990; Alonso et al., 1995a). ND activity has been described in the olfactory system of several species of mammals such as hamster (Davis, 1991), mouse (Kishimoto et al., 1993), hedgehog (Alonso et al., 1995b), and rat (Scott et al., 1987; Davis, 1991; Vincent and Kimura, 1992; Alonso et al., 1993; Porteros et al., 1994), revealing a heterogeneity in the staining pattern among different species. In the rat main olfactory bulb (MOB), ND-staining has been reported in a subpopulation of olfactory fibers, in centrifugal fibers, and in several types of intrinsic neurons. ND-positive cells have been identified as pe.riglomeruIar cells, granule * Corresponding author, Tel.: +34 23 294400 ext. 1854; Fax: +34 23 294549; e-mail: [email protected]. 089l-0618/95/$09.50 0 1995 Ekvier SSDI 0891-0618(95)0082-I
and deep short-axon cells, and external tufted cells (Scott et al., 1987; Alonso et al., 1993), although the presence of ND activity in this latter type has been controversial (Alonso et al., 1993). In the accessory olfactory bulb (AOB), the vomeronasal fibers and glomeruli, and different neuronal types identified as periglomerular cells, granule cells and deep short-axon cells were ND-positive (Porteros et al., 1994). In the rat MOB, the cholinergic system has been studied using histochemical, immunocytochemical, and biochemical techniques for acetylcholinesterase (AChE), choline acetyltransferase (ChAT), and choline@ receptors (Godfrey et al., 1980; Jaffe and Cuello, 1980; Large et al., 1986; Nickel1 and Shipley, 1988; Phelps et al., 1992; Le Jeune and Jourdan, 1991, 1993, 1994). It is currently accepted that the MOB has no intrinsic choline@ elements (Le Jeune and Jourdan, 1991, 1993). Nevertheless, AChE histochemistry demonstrates the existence of a population of AChEpositive neurons in the rat MOB. These AChE-positive cells have been characterized as non-choline@ and
cells, superficial
Science B.V. All rights reserved
208
C. Crespo et al. /Journal of Chemical Neuroanatomy 9 (199.5) 207-216
presumably cholinoceptive neurons (Nickel1 and Shipley, 1988; Le Jeune and Jourdan, 1994). They have been typified as periglomerular cells, superficial and deep short-axon cells, and external tufted cells (Nickel1 and Shipley, 1988; Le Jeune and Jourdan, 1991, 1993, 1994), the same neuronal types that demonstrate ND staining. A deeper understanding of the cholinergic system in the MOB and AOB requires more detailed knowledge of the neurochemical identity of the cholinoceptive cells. Le. Jeune and Jourdan (1994) reported that cholinoceptive neurons in the rat MOB, identified by using AChE histochemistry, are not GABAergic and only a few of them express tyrosine hydroxylase. Their data confirm the previously reported neurochemical heterogeneity of some MOB intemeurons (see Hal&z, 1990, for a review). Both ND and AChE are expressed in similar neuronal populations of both second-order interneurons, superficial and deep short-axon cells, and first-order intemeurons, periglomerular cells. Thus, it is possible that both enzymatic activities are partially or totally colocalized by the same cells or, on the contrary, the AChE-positive and the ND-positive neurons are different neuronal populations. In order to establish the possible expression of ND in AChE-containing cells, the present report analyzes the degree of colocalization of both enzymatic markers in the MOB and AOB, carried out by successive AChE and ND histochemistry in the same sections.
2. Material ad methods Five adult female and five adult male Wistar rats (250-280 g body weight) were deeply anaesthetized with ketamine (Ketolar, 50 mg/ kg body weight, i.m.). After packing the heads with ice, the animals were perfused through the ascending aorta with 100 ml Ringer solution followed by 500 ml fixative solution containing 4% paraformaldehyde and 15% saturated picric acid in 0.1 M phosphate buffer, pH 7.3 (PB). The brains were removed and the olfactory bulbs were dissected out, postfixed 4 h at 4°C in the same fixative solution, and cryoprotected with 30% sucrose (v/v) in PB at 4°C overnight. Ten- and thirty-micrometer coronal and sag&al sections were cut using a cryostat and serially collected in cold (4°C) PB or in cold 0.1 M sodium acetate buffer, pH 6.0 (AB). The histochemical procedures were carried out on serial free-floating sections. A one-in-four series from each animal was processed for the demonstration of ND activity, a second series was used for AChE histochemistry, and a third series was sequentially processed for both enzymatic activities (double-labeled series). The fourth series was used for the specificity controls.
2.1. NADPH-diaphorase histochemistry Series collected in PB were washed in 0.1 M Tris-HCl buffer pH 8.0, and processed for ND histochemistry as described elsewhere (Scherer-Singler et al., 1983; Alonso et al., 1995a). Briefly, sections were incubated for 60-90 min at 37°C in an incubation solution made up of 1 mM reduced @-NADPH (Sigma #N1630), 0.3 mM nitro blue tetrazolium (Sigma #N6876) and 0.08% Triton X-100 in 0.1 M Tris-HCl buffer, pH 8.0. The course of the reaction was controlled under the microscope. When the histochemical reaction was concluded, sections were washed in 0.1 M Tris-HCl buffer, pH 8.0. 2.2. Acetylcholinesterase histochemistry Series collected in AB were processed for the AChE histochemical technique using a modification (Hedreen et al., 1985) of the Koelle method (Geneser-Jensen and Blackstad, 1971). Sections were washed in AB and placed in agitation in a freshly prepared incubation medium at room temperature for 30 min. The incubation medium was made up of 1.7 mM acetylthiocholine iodide, 0.49 mM sodium citrate, 2.9 mM cupric sulfate, 1.25 mM potassium ferricyanide and 0.2 mM ethopropazine, an inhibitor of non-specific cholinesterases, in AB. AChE activity was visualized using two different techniques. In the first method, the sections were rinsed, after incubation, in AB (3 x 10 min), treated for 1 min with 2% ammonium sulfide and washed (3 x 10 min) in 0.1% sodium nitrate. The free-floating sections were exposed to 0.05% silver nitrate for 1 min and washed in 0.1% sodium nitrate (3 x 5 min) and PB (2 x 5 min). In the second procedure, the sections were rinsed in 0.2 M Tris-HCl buffer, pH 7.6 (3 x 10 min) and the reaction was visualized using 0.0125% 3,3 ’ diaminobenzidine and 0.003% hydrogen peroxide in 0.2 M Tris-HCl buffer, pH 7.6 under microscope control. The reaction was stopped by rinsing the sections with 0.2 M Tris-HCl buffer pH 7.6 (3 x 5 min) and distilled water (2 x 5 mm). A few histochemical sections were darkened with 0.5% nickel ammonium sulfate in the DAB solution. 2.3. NADPH-diaphorase-acetylcholinesterase
double la-
beling
Using both described ND- and AChE-histochemical protocols, some sections previously processed for ND were stained for the detection of AChE activity. Tests showed that when the ND histochemistry was carried out in sections previously processed for AChE, no ND reaction was detected. When the double staining was carried out beginning with the ND technique, both enzymatic activities were clearly observed. In order to clarify which reactions could affect the ND detection, some assays were carried out. In these tests, ND histochemistry was carried out in sections previously in-
C. Crespo et al. /Journal
of Chemical Neuroanatomy
cubated (30 min) in cupric sulfate, potassium ferricyanide and combination of both reactives. These tests revealed that the cupric sulfate and not the potassium ferricyanide interferes in the ND histochemistry. The ND-, AChE-, and double-stained sections were serially mounted on gelatin-coated slides, dehydrated in ethanol series, cleared with xylene, and coverslipped with Entellan. The intense ND-staining observed in some neuronal types made the detection of double labeling in these cells difficult, since the dark blue formazan obtained could mask the brown reaction product of the AChE histochemistry. For the detection of possible colocalization of both enzymes in these cells, the ND-reaction was stopped in a group of sections when these cells were weakly ND-stained, followed by the AChEhis&hen&al procedure. In some sections, the doublelabeled cells were drawn using a camera lucida, photographed, and the sections treated with dimethylformamide as described (Alonso et al., 1995a). The sections were then re-examined and the cells photographed again (Fig. lg, h). This treatment eliminates the formazan without affecting the AChE reacting product, confirming the presence or absence of double labeling. 2.4. Controls For ND, the following controls were carried out as described (Alonso et al., 1995a): (1) omission of the substrate &NADPH in the incubation media; (2) omission of the chromogen nitro blue tetrazolium; (3) denaturation of the enzyme by heating the tissue at 84°C for 5 min; (4) Overfixation of tissue during 2 weeks in 10% formalin; and (5) substitution of &NADPH by NADP (Sigma #NO505). For the AChE histochemistry, two controls were carried out: (1) omission of the substrate acetylthiocholine iodide; (2) addition of 10” M BW284C51 as inhibitor of AChE activity (Carson and Burd, 1980). For all controls, no residual activity was observed. 2.5. Analysis The number of reactive cells was counted in five animals from the same sex (females), with a 16x planapochromatic objective. In neuronal types that were represented by a high number of positive elements, 10 000 randomly chosen neurons were taken as statistically representative. For neuronal types with less than 40 positive elements per section, all positive neurons were counted. The maximum diameters of the neuronal types were calculated with a digitizer tablet connected to a semiautomatic image analysis system (MOP-Videoplan, Kontron). Only the cells where the nucleus was clearly visible were used for the measurement of cell sizes. In the neuronal types where the dense reaction product obscured the nucleus, only the neurons
9 (1995) 207-216
209
located in the middle of the section and with at least two visible primary dendrites were measured. 3. Results The ND histochemical technique provided a Golgilike staining of positive neurons. No glial cells were labeled. In many ND-stained elements, not only the cell body, but also the dendrites and long portions of the axons were visualized, allowing a good morphologic characterization of these cells (Fig. le). In the AChE histochemistry, by contrast, only the cell bodies and, at most, initial portions of the dendrites of the labeled intrinsic neurons were stained (Fig. lb,d,h). In addition, networks of AChE-positive fibers, presumably centrifugal fibers, were observed in most OB layers (Fig. la). When the reaction was visualized with silver nitrate, a dark fiber staining was observed, whereas cell bodies were weakly labeled. The use of 3,3 ’ diaminobenzidine as chromogen increased the labeling intensity of cell bodies. The staining patterns obtained in double-labeled sections were similar to those observed in the ND- or AChE-single-stained series: the number of ND- or AChEpositive elements, their distribution, and their labeling intensity were not altered in the double-labeled series. In these sections, the blue formazan of the ND histochemistry and the brown reaction product of the normally AChE histochemistry were clearly distinguishable (Fig. 1 a-g). The colocalization of both labels was identified by the presence in cell bodies of both reaction products at the same focusing plane (Figs. lc, f, g) and confirmed by dimethylformamide treatment (Fig. lh). The distribution pattern of both enzymatic activities in the MOB and AOB was similar in males and females. No clear differences were observed in the NDand AChE-stainings nor in the colocalization pattern between animals from either sex. 3.1. Main olfactory bulb (MOB) After both ND and AChE staining, positive cells and/or fibers were found in all layers of the MOB, except in the olfactory nerve layer (ONL) that did not demonstrate AChE reactivity. Both ND and AChE activities showed clear differences across layers (Fig. la). The highest density of ND-positive elements was observed in the granule cell layer (GCL) (Fig. la), where a dense network of positive fibers was stained, and in the deep portion of the glomerular layer (GL), with a high number of ND-positive cells (Fig. la). The NDcontaining neurons were identified as periglomerular cells, granule cells, and superficial and deep short-axon cells (Fig. 2). For AChE, dense plexuses of AChEpositive fibers were found in the deep portion of the GL, in the superficial portion of the external plexiform layer (EPL), and in the internal plexiform layer (IPL) (Fig.
210
C. Crespo et al. / Journal of Chemical Neuroanaromy 9 (1995) 207-216
C. Crespo et al. / Journal of Chemical Neuroanatomy 9 (1995) 207-216
211
COLOC!ALlZATION
AC&El
,.““....
.‘.“.‘..
GL
:
._....,
,.:
;i
‘... :, :.. :.’ .___.. .*:...._..’ ......_..L.....‘........’
EPL
EPLC
.... ..-: ,...:“.. :
:
.:
..: i
i:
:.:
::
: ..
,A. i:
1.:’ .. . .. ‘..j
:
a#
! .i ,.“.. :.:
;.‘.,
:.. : : .: .,j
MCL IPL
EPL
EPL
EPLC
.... :“: .... :“....‘..:...; .. : : : T: : : ; : : : : : ., ..‘., .,..:: ..... :..: .._::.,: :... :. i ; i MCL ..: ..: IPL &e
DSA
GCL
GL
MCL IPL
loC GCL
GCL
A DSA*
WM
DSF
WM
Fig. 2. Schematic drawing showing the acetylcholinesterase-labeled (colocalization) cells in the rat olfactory bulb. DSA: deep short-axon cells; EPL: external plexiform layer; EPLC: external plexiform layer cells: GC: granule cells; GCL: granule cell layer; GL: glomerular layer; IPL: internal plexiform layer; MCL: mitral cell layer; PG: periglomerular cells; SSA: superilicial short-axon cells SSAIETC: superficial sort axon cells and/or external tufted cells; WM: white matter.
la). AChE-labeled cells were identified as periglomerular cells, external tufted cells and superficial and deep short-axon cells (Fig. 2). In the ONL, a subpopulation of ND-positive olfactory fibers was observed. These fibers formed a subpopulation of ND-labeled glomeruli in the GL. These ND-positive glomeruli demonstrated a clear-cut spatial segregation: whereas glomeruli located in the dorsomedial portion of the GL were ND-positive, those of the ventrolateral portion were ND-negative. The AChE histochemistry technique demonstrated a dense network of AChE-positive fibers surrounding all glomemli (Fig. la). The innervation was similar around
ND-positive and ND-negative glomeruli. Nevertheless, in the caudal portion of the MOB a few glomemli showed a remarkably intense and atypical AChEpositive innervation. These atypical glomeruli contained the highest density of AChE-positive fibers in the rat MOB, and they were clearly distinguishable from the typical glomeruli. They showed a clear spatial segregation and were placed in the dorsomedial and ventrolateral portions of the caudalmost sections of the MOB. Although in the dorsomedial portions the atypical glomeruli were close to ND-positive glomeruli, no atypical glomeruli were ND-positive. In the periglomerular region and at the GL/EPL
Fig. 1. AChE-positive (brown) and ND-positive (blue) stainings in the rat olfactory bulb. Scale bar = IO0 Cm (a) and 25 pm (b-h). (a) Overview of a coronal double-labeled section of the rat olfactory bulb showing the different ND- and AChE-staining patterns across layers. Some ND-positive periglomerular cells can be observed (arrows). ONL: olfactory nerve layer; GL: glomerular layer; EPL: external plexiform layer; MCL: mitral cell layer; IPL: internal plexiform layer; GCL: granule cell layer. (b) large AChE-stained cell in the GWEPL boundary (open arrow), and ND-positive periglomerular cells (arrows). (c) Double-labeled cell (open arrow) and AChE-positive fibers (arrows) in the innter portion of the EPL. (d) AChEstained deep short-axon cell (open arrow), and ND-positive granule cell (arrow) in the GCL. (e) ND-labeled deep short-axon cell in the WM. (f) Colocalization of ND and AChE in a deep short-axon cell (open arrow) in the accessory granule cell layer. (g) Colocalization of ND and AChE in a deep short-axon cell of the WM. (h) Same cell as in (g) after treatment with dimethylformamide showing the AChE-labeling.
212
C. Crespo et al. /Journal of Chemical Neuroanatomy 9 (1995) 207-216
Table 1 ND- and/or AChE-positive neuronal types in the rat olfactory bulb Cell type
Staining
Max. diameter pm (mean * S.E.M)
Location
Frequency”
Periglomerular cell
ND AChE ND AChE
8.7 f 0.3 9.2 f 0.4 16.7 ZIZ 0.6 16.2 zt 0.7
GLJEPL GLJEPL GUEPL GUEPL
++++ ++ + ++
ND AChE ND+AChE ND AChE ND+AChE ND
16.2 zt 0.4 16.1 f 0.6 16.0 zt 0.5 17.8 f 0.3 18.2 f 0.2 17.4 f 0.4 7.7 f 0.2
EPL EPL EPL IPL/GCL/WM IPL/GCL/wM IPUGCL/wM MCUIPUGCL
+ + + ++ +++ + ++++
Superficial short-axon cell Superficial short-axon cell and/or external tufted cell External plexiform layer cell
Deep short-axon cell
Granule cell
aFrequency: +: less than 1 cell per section. ++: 1 to 10 cells per section. +++: 10 to 100 cells per section. ++++: more than 100 cells per section.
boundary, populations of periglomerular and superficial short-axon cells were ND or AChE labeled. AChEpositive periglomerular cells were found surrounding the typical and atypical glomeruli. ND-stained periglomerular cells demonstrated one or two dendrites. Bidendritic periglomerular cells innervated one or two adjacent glomeruli. The ND-positive periglomerular cells were homogeneously distributed in the GL, and they surrounded the ND-positive, and the ND-negative typical and atypical glomeruli. Comparing the number of ND-positive and AChE-positive periglomerular cells, ND-positive cells were more abundant (more than 200 cells per 3Oqm section) than AChE-positive cells (less than 10 cells per 30qm section) (Table 1). The two populations of periglomerular cells were different, in that no periglomerular cells colocalized ND and AChE. Close to the periglomerular cells, a few superficial short-axon cells demonstrated ND- or AChE-labeling (Fig. lb). The ratio of ND-positive superficial shortaxon cells to ND-positive periglomerular cells was approximately 1:2000. Of these scarce superficial shortaxon cells, none colocalized both neuronal markers, being therefore two different neuronal populations, as is the case for the periglomerular cells.. In the inner half of the EPL, a few ND- or AChElabeled cells were occasionally observed (Fig. lc). Only one or two proximal dendrites were labeled in these cells, making their classification difficult. We refer to them as EPL cells (Fig. 2). A few of these AChE- or NDpositive EPL cells (less than 3% of both populations) colocalized both enzymes (Figs. lc, 2; Table 1). In the deep layers (IPL, GCL, and white matter (WM)) only one neuronal population, identified as deep short-axon cells, showed both AChE and ND activities (Fig. Id, e). In addition, several granule cells were NDstained (Fig. Id), but all of them were AChEnegative.
Centering on deep short-axon cells, the number of AChE-positive deep short-axon cells (more than 20 cells per 30-pm section) was higher than the number of NDpositive cells (less than 10 per 30-pm section). There was a spatial segregation of both neuronal subpopulations. Although they were present throughout the IPL, GCL, and WM, most AChE-positive cells were mainly located in the IPL and in the outer half of the GCL, whereas the ND-positive cells were more abundant in the inner portion of the GCL and in the WM. Colocalization of both enzymatic activities was found in a few of them (less than 2% of the AChEpositive neurons and less than 3% of the ND-positive cells). The colocalized neurons were present in the IPL, in the GCL, and in the WM (Fig. lg, h), but most of them were located in the IPL and in the superficial portion of the GCL, the same zones where AChE-positive cells were more abundant. 3.2. Accessory olfactory bulb (AOB) ND-positive fibers and cells were found in all layers of the AOB. All vomeronasal fibers and glomeruli were ND-positive. The ND-stained cells were identified as periglomerular cells, granule cells, and deep short-axon cells. After AChE-histochemistry, only scarce neurons in the IPL and in the GCL, identified as deep short-axon cells, were labeled. Colocalization of ND and AChE was observed in a few (less than 3%) deep short-axon cells located in the accessory granule cell layer (Fig. If). 4. Discussion In this report, a colocalization study of ND and AChE activities was carried out in the rat MOB and AOB. The present data confirm previous observations of the distribution patterns of both neuronal markers in the rat MOB and AOB (Godfrey et al., 1980; Scott et al.,
C. Crespo et al. /Journal
of Chemical Neuroanatomy
1987; Nickel1 and Shipley, 1988; Davis, 1991; Le Jeune and Jourdan, 1991, 1994; Vincent and Kimura, 1992; Alonso et al., 1993; Porteros et al., 1994). The main results in the present report are that although both enzymes coexist in the same neuronal types of the MOB and AOB, only a few cells (less than 3% in all cases) identified as EPL cells and deep short-axon cells colocalized both enzymatic activities, whereas for other neuronal types (i.e., periglomerular cells and superficial short axon cells) no colocalization was observed. The degree of colocalization between ND and AChE in the MOB and the AOB is low in contrast to other regions such as the medial septum and diagonal band of Broca that demonstrate an abundant colocalization of ND and choline@ markers (Kitchener and Diamond, 1993). These regions are two of the main sources of cholinergic afferents to the MOB (De Olmos et al., 1978). Therefore, it can be assumed that a large population of centrifugal fibers is both ND- and AChE-positive. However, in our double histochemistry, double labeled NDpositive and AChE-positive afferents were not clearly identifiable. These data suggest that whereas the doublestaining technique provides an accurate somal staining pattern for both enzymatic markers, it is not reliable for the study of colocalization of both enzymatic activities in extrinsic aBerents. We and others have observed that roughly half of all rat olfactory glomeruli are ND-positive (Scott et al., 1987; Davis et al., 1991; Alonso et al., 1993) whereas only a small population of caudal glomeruli have an atypical, dense AChE-positive innervation (Zheng et al., 1987; Zheng and Jourdan, 1988; Le Jeune and Jourdan, 1994). Our results show that after using both neuronal markers, three glomerular staining patterns may be distinguished: typical/ND-positive, typical/ND-negative and atypical/ND-negative glomeruli. The present data demonstrate that the ND-positive olfactory fibers did not innervate the atypical glomeruli in spite of their spatial proximity in the dorsomedial region of the MOB. This fact confirms the topographically specific heterogeneity of the MOB glomeruli. By contrast, the AOB demonstrated more homogeneous staining, in that all vomeronasal glomentli were ND-positive and AChEnegative. The periglomerular cells represent the first-order intemeurons at the glomerular level. A high neurochemical heterogeneity of periglomerular cells has been previously reported (see Hal&z, 1990, for a review). In agreement with this heterogeneity, we described three different populations of periglomerular cells: NDpositive/AChE-negative, ND-negative/AChE-positive, and ND-negative/AChE-negative. The absence of colocalization of both enzymatic activities in the periglomerular cells add both markers to the neuroactive substances such as calbindin D-28k and ND (Alonso et al., 1993) or calbindin D-28k and tyrosine
9 (1995) 207-216
213
hydroxylase (Hal&z et al., 1985) that do not colocalize in these cells. Le Jeune and Jourdan (1994) described that the AChE-positive periglomerular cells in the rat MOB are non-GABAergic and non-dopaminergic. In addition, our results indicate that the AChE-positive periglomerular cells are also ND-negative. ND-positive periglomerular cells were more abundant (more than 200 ND-positive cells per 30-pm section) than the AChE-positive periglomerular cells (less than 10 AChEstained cells per 30-pm section). These results suggest that ND and AChE label distinct modulatory systems at the glomerular level: AChE was expressed by abundant extrinsic centrifugal afferents and a few intrinsic cells whereas ND was mostly expressed by interneurons, a different subpopulation of periglomerular cells, located in this layer. As periglomerular cells, superficial short-axon cells constitute a biochemically heterogeneous neuronal type (Hal&z, 1990). Le Jeune and Jourdan (1994) have demonstrated that the AChE-containing superficial short-axon cells do not correspond to the few GABAergic cells previously described by Mugnaini et al. (1984). They noticed that the cholinoceptive superficial short-axon cells have similar morphological characteristics and distribution as other populations containing neuropeptide Y (Gall et al., 1986; Scott et al., 1987), somatostatin and ND (Scott et al., 1987). Our data demonstrated, however, that there is no coexpression of both AChE and ND by the superficial shortaxon cells. Therefore, despite those previously mentioned similarities, the AChE-containing and the NDpositive superficial short-axon cells are different neuronal populations. Since ND is a selective marker for nitric oxide synthase-immunoreactive neurons in the olfactory bulb (Kishimoto et al., 1993), these results indicate that ND is a specific marker for noncholinoceptive periglomerular and superficial shortaxon neurons and, reciprocally, AChE histochemistry can be used as a selective marker for neurons that do not express nitric oxide synthase in the periglomerular region of the rat MOB. In the inner half of the EPL, a low number of NDpositive, AChEpositive and colocalized neurons have been observed. The weak labelings for ND- and/or AChE make morphological characterization of these cells difficult. Medium-sized neurons in the EPL constitute a very heterogeneous population where at least live neuronal types can be distinguished (Kosaka et al., 1994). According to their location, size, and shape, these EPL cells may correspond to external tufted cells and/or non-periglomerular short-axon cells. Deep short-axon cells are probably the most heterogeneous neuronal population in the vertebrate olfactory bulb (Hal&z, 1990). Le Jeune and Jourdan (1994) reported that no AChEcontaining deep shortaxon cells were GABAergic and only a limited propor-
214
C. Crespo et al. /Journal of Chemical Neuroanatomy 9 (1995) 207-216
tion of them exhibited tyrosine hydroxylase immunoreactivity. In agreement with this biochemical variability, our data indicate that some deep short-axon cells of the MOB and the AOB express ND, some AChE, and a few cells both. According to the morphology and distribution of the colocalized neurons in the MOB, they belong to the four morphological subtypes defined by Ramon y Cajal (1911) and Schneider and Macrides (1978) using Golgi impregnation. The cells located in the IPL and in the superficial portion of the GCL belong presumably to the Ramon y Cajal and/or horizontal cells, and the cells located in the deep portion of the GCL and in the WM, to the Blanes and/or Golgi types. Nitric oxide has been proposed as a presumptive retrograde transmitter between the postsynaptic and the presynaptic neuron (Bredt and Snyder, 1992). The AChEcontaining neurons in the rat MOB are considered as cholinoceptive (Le Jeune and Jourdan, 1994). The neurotransmitter used by cholinoceptive secondary interneurons in the MOB remains unknown (Le Jeune and Jourdan, 1994). Nevertheless, our data support the hypothesis that it is unlikely to be nitric oxide. Reciprocally, the ND-positive interneurons were principally not cholinoceptive. The cholinergic effects in the MOB could take place at two distinct levels, superficial and deep (Le. Jeune and Jourdan, 1994). Elaagouby et al. (1991) proposed the existence of cholinoceptive superficial interneurons projecting to deep layers and influencing granule cell activity. Several granule cells were ND-positive, but it is presently unknown whether the cholinoceptive cells and the ND-positive granule cells are selectively related among them in the same neuronal circuitries. The localization of ND is coincident with that of nitric oxide synthase in the neurons of the MOB (Kishimoto et al., 1993). On the other hand, the AChEcontaining intrinsic neurons in the MOB are considered as presumably cholinoceptive (Le Jeune and Jourdan, 1994). In both the peripheral and the central nervous systems, there are multiple interactions between the production of nitric oxide and acetylcholine. Nitric oxide may modulate the cholinergic transmission at the pre- and/or post-junctional level in the nerve muscle pathway (Baccari et al., 1994), and appears to exert a marked inhibitory effect on acetylcholine release in different regions such as the trachea and taenia coli (Belvisi et al., 1991; Knudsen and T@trup, 1992). In addition, acetylcholine evokes release of nitric oxide from endothelium and, through muscarinic receptors activation contributes to stimulation of nitric oxide formation in the brain (Wiklund et al., 1993). Nitric oxide activates soluble guanylyl ciclase in target cells (Bredt and Snyder, 1992) and cGMP levels may play an important role in olfactory signalling, including recruitment, adaptation and modulation at the olfactory mucosa and
olfactory bulb levels (Breer and Shepherd, 1993). In this sense, periglomerular cells have been proposed as a NO/cGMP system that modulates intraglomerular and interglomerular synaptic integration of sensory inputs (Breer and Shepherd, 1993). Our data demonsrate that in the rat OB, AChE-positive and ND-positive periglomerular cells are different populations. In the rest of the OB, only a low proportion of AChEcontaining neurons express nitric oxide synthase in the MOB. Pinching and Powell (1972) proposed that periglomerular ceils are the main cellular target of the cholinergic afferents in the glomerular layer but Le Jeune and Jourdan (1991, 1993, 1994) observed after AChE histochemistry and ChAT immunocytochemistry that only a subset of periglomerular cells are cholinoceptive. Our observations are in agreement with Le Jeune and Jourdan (1991, 1993, 1994) since only a low number of periglomerular cells, compared with the ND-positive ones, were AChE-stained. Nevertheless, all periglomerular cells are closely surrounded by a dense network of AChEpositive fibers and nitric oxide, the soluble gas synthesized in the ND-positive neurons, diffuses freely across membranes in a reduced area. Then, despite the practical absence of colocalization, both AChE- and ND-positive systems may interact directly and extensively through synaptic and non-synaptic links. Both choline@ and ND-active systems have been associated with neurodegenerative disorders such as Alzheimer’s disease (Sugaya and McKinney, 1994). NDactive neurons are selectively spared in the hippocampal formation (Hyman et al., 1992) while cholinergic neurons degenerate in the nucleus basalis of Meynert in this neuropathology (Big1 et al., 1990). The olfactory system is severely affected in Alzheimer’s disease (Talamo et al., 1991) but there is little information as to the impact on specific classes of neurons. Thus, further study of the heterogeneity of AChE and ND activity in MOB neurons, as described here, might shed some light on the potentially selective susceptibility of some MOB neurons to degeneration in Alzheimer’s disease and related disorders. Acknowledgements
The authors express their gratitude to Dr. Thomas A. Schoenfeld for his valuable comments and suggestions and Mr. J. Marcos for his photographic assistance. This work was supported by grants from the ‘Junta de Castilla y Leon’, , and the DGICyT (93-065 and PB94-13). References Alonso,J.R., Arhalo, R., Brikn, J.G., Garcia-Ojeda, E., Porteros, A. and Aij6n,J. (1995a) NADPH-diaphorase staining in the central nervous system. Neurosci. Protocols (in press).
C. Crespo et al. /Journal
of Chemical Neuroanatomy 9 (1995) 207-216
Alonso, J.R., Amvalo, R., Garcia-Gjeda, E., Porteros, A., Brition, J.G. and Aijon, J. (1995b) NADPHdiaphorase active and calbindin D28k-immunoreactive neurons and fibers in the olfactory bulb of the hedgehog (Erinacew europaeus). J. Comp. Neurol. 35 1,307-327. Alonso, J.R., ArivaIo, R., Porteros, A., B&on, J.G., Lara, J. and Aijbn, J. (1993) Calbindin D28K and NADPH-diaphorase activity are localized in different populations of periglomerular cells in the rat olfactory bulb. J. Chem. Neuroanat. 6, 1-6. Baccari, M.C., Calamai, F. and Staderini, G. (1994) Modulation of chohnergic transmission by nitric oxide, VIP and ATP in the gastric muscle. Neuroreport 5, 905-908. Eklvisi, M.G., Stretton, D. and Barnes, P.J. (1991) Nitric oxide as an endogenous modulator of cholinergic neurotransmission in guinea pig airways. Eur. J. Pharmacol. 198, 219-221. Bigi, V., Arendt, T. and Biesold, D. (1990) The nucleus basalis of Meynert during aging and in dementing disorders. In Brain cholinergic systems (eds Steriade, M. and Biesold, D.) pp. 364-386. Oxford University Press, Oxford. Bredt, D.S. and Snyder, S.H. (1992) Nitric oxide, a novel neuronal messenger. Neuron 8, 3- 11. Breer, H. and Shepherd, G.M. (1993) Implications of the NO/&IMP system for olfaction. Trends Neurosci. 16, 5-9. Carson, K.A. and Burd, G.D. (1980) Localization of acetylcholinesterase in the main and accessory olfactory bulbs of the mouse by light and electron microscopic histochemistry. J. Comp. Neurol. 191, 353-371. Davis, B.J. (1991) NADPH-diaphorase activity in the olfactory system of the hamster and rat. J. Comp. Neurol. 314, 493-511. De Olmos, J., Hardy, H. and Heimer, L. (1978) The afferent connections of the main and accessory olfactory bulb formations in the rat: an experimental HRP-study. J. Comp. Neurol. 181, 213-244. Elaagouby, A., Ravel, N. and Gervais, R. (1991) Cholinergic modulation of excitability in the rat olfactory bulb: effect of local application of cholinergic agent on evoked field potentials. Neuroscience 45, 653-662. Gall, C., Seroogy, K.B. and Brecha, N. (1986) Distribution of VIPand NPY-like immunoreactivities in rat main olfactory bulb. Brain Res. 374, 389-394.
Geneser-Jensen, F.A. and Blackstad, T.W. (1971) Distribution of acetyl cholinesterase in the hippocampal region of the guinea pig. 1. Entorhinal area, parasubicuhnn, and presubiculum. Z. Zellforsch. Mikrosk. Anat. 114, 460-48 1. Godfrey, D.A., Ross, C.D., Hermann, A.D. and Matschinsky, F.M. (1980) Distribution and derivation of cholinergic elements in the rat olfactory bulb. Neuroscience 5, 273-292. Hal&z, N. (1990) The vertebrate olfactory system. Chemical neuroanatomy, function and development. Akademiai Kiado, Budapest. Hal&z, N., Hokfelt, T., Norman, A.W. and Goldstein, M. (1985) Tyrosine hydroxylase and 28k-vitamin D-dependent calcium binding protein are located in different subpopulations of periglomerular cells of the rat olfactory bulb. Neurosci. L&t. 61, 103-107. Hedreen, J.C., Bacon, S.J. and Price, D.L. (1985) A modified histochemical technique to visualize acetylcholinesterase-containing axons. J. Histochem. Cyfochem. 33, 134-140. Hyman, B.T., Manloff, K., Wenninger, J.J., Dawson, T.M., Bredt, D.S. and Snyder, S.H. (1992) Relative sparing of nitric oxide synthase-containing neurons in the hippocampal formation in Alxheimer’s disease. Ann. Neural. 32, 818-820. Jaffe, E.H. and Cuello, AC. (1980) The distribution of catecholamines, glutamate decarboxylase and choline acetyltransferase in layers of the rat olfactory bulb. Brain Res. 186, 232-237. Kieman, J.A. (1990) Histological and hisrochemical methodr. Theory and practice. Pergamon Press, Oxford. Kishimoto, J., Keveme, E.B., Hardwick, J. and Emson, P.C. (1993) Localization of nitric oxide synthase in the mouse olfactory and
215
vomeronasal system: a h&chemical, immunological and in situ hybridization study. Eur. J. Neurosci. 5, 1684-1694. Kitchener, P.D. and Diamond, J. (1993) Distribution and colocahxation of choline acetyhransferase immunoreactivity and NADPH diaphorase reactivity in neurons within the medial septum and diagonal band of Broca in the rat basal forebrain. J. Comp. Neural. 335, l-15. Knudsen, M.A. and Tdtrup, A. (1972) A possible role of the Larginine-nitric oxide pathway in the modulation of cholinergic transmission in the guinea-pig taenia coli. Br. J. Pharmacol. 107, 837-841.
Kosaka, K., Heizmann, C.W. and Kosaka, T. (1994) Calcium-binding protein parvalbumin-immunoreactive neurons in the rat olfactory bulb. Exp. Brain Res. 99, 191-204. Large, T.H., Lambert, M.P., Gremillion, M.A. and Klein, W.L. (1986) Parallel postnatal development of choline acetyltransferase activity and muscarinic acetylcholine receptors in the rat olfactory bulb. J. Neurochem. 46, 671-680. Le Jeune, H. and Jourdan,
F. (1991) Postnatal development of cholinergic markers in the rat olfactory bulb: a histochemical and immunocytochemical study. J. Comp. Neural. 314, 383-395. Le Jeune, H. and Jourdan, F. (1993) Choline&z innervation of olfactory glomeruli in the rat: an ultrastructural immunocytochemical study. J. Comp. Neural. 336, 279-292. Le Jeune, H. and Jourdan, F. (1994) Acetylcholinesterase-containing intrinsic neurons in the rat main olfactory bulb: cytological and neurochemical features. Eur. J. Neurosci. 6, 1432-1444. Mugnaini, E., Wouterlood, F.G., Dahl, A.L. and Gertel, W.H. (1984) Immunocytochemical identification of GABAergic neurons in the main olfactory bulb of the rat. Arch. Ital. BioI. 122, 83- I1 3. Nickell, W.T. and Shipley, M.T. (1988) Two anatomically specific classes of candidate cholinoceptive neurons in the rat olfactory bulb. J. Neurosci. 8, 4482-4491. Phelps, P.E., Houser, C.R. and Vaughn, J.E. (1992) Small choline@ neurons within field of choline@ axons characterize olfactoryrelated regions of rat telencephalon. Neuroscience 48, 121-136. Pinching, A.J. and Powell, T.P.S. (1972) The termnation of centrifugal libres in the glomerular layer of the olfactory bulb. J. Cell Sci. 10, 621-635. Porteros, A., Alonso, J.R., Amvalo, R., Garcia-Ojeda, E., Crespo, C. and Aijon, J. (1994) Histochemical locahxation of NADPHdiaphorase in the rat accessory olfactory bulb. Chem. Senses 23, 1083-1093. Ramon y Cajal, S. (1911) Histologie du Systeme Nerveux de I’Homme et des Vertebres. Maloine, Paris. Scott, J.W., McDonald, J.K. and Pemberton, J.L. (1987) Short axon cells of the rat olfactory bulb display NADPH-diaphorase activity, neuropeptide Y-like inununoreactivity, and somatostatin-like immunoreactivity. J. Comp. Neurol. 260, 378-391. Scherer-Singler, U., Vincent, S.R., Kimura, H. and McGeer, E.G. (1983) Demonstration of a unique population of neurons with NADPH-diaphorase histochemistry. J. Neurosci. Methoa!s 9, 229-234.
Schneider, S.P. and Macrides, F. (1978) Laminar distribution of intemeurons in the main olfactory bulb of the adult hamster. Brain Res. Bull. 3, 73-82.
Sugaya, K. and McKinney, M. (1994) Nitric oxide synthase gene expression in cholinergic neurons in the rat brain examined by comin situ hybridization bined immunocytochemistry and histochemistry. Brain Res. 23, 111-125. Talamo, B.R., Feng, W.H., Perez-Cruet, M., Adelman, L., Kosik, K., Lee, V.M. Y., Cork, L.C. and Kauer, J.S. (1991) Pathological changes in olfactory neurons in Alzheimer’s disease. Ann. N. Y. Acad. Sci. 640, 1-7. Vincent, S.R. and Kimura, H. (1992) Histochemical mapping of nitric oxide synthase in the rat brain. Neuroscience 46, 755-784. Wiklund, C.U., Wiklund, N.P. and Gustafsson, L.E. (1993) Modula-
216
C. Crespo et al. /Journal of Chemical Neuroanatomy 9 (1995) 207-216
tion of neuroeffector transmission by endogenous nitric oxide: a role for acetylcholine receptor-activated nitric oxide formation, as indicated by measurements of nitric oxide/nitrite release. Eur. J. Pharmacol. 240, 235-242.
Zheng, L.M. and Jourdan, F. (1988) Atypical olfactory glomeruh con-
tam original olfactory axon terminals: an ultrastructural horscradish peroxidase study in the rat. Neuroscience 26, 367-378. Zheng, L.M., Ravel, N. and Jourdan, F. (1987) Topography of centrifugal acetylcholinesterase-positive fibres in the olfactory bulb of the rat: evidence for original projections in atypical glomeruh. Neuroscience 23, 1083-1093.