Brain Research, 547 (1991) 239-248 © 1991 Elsevier Science Publishers B.V. 0006-8993/91/$03.50 A-DONIS 000689939116542M
239
BRES 16542
Subsets of olfactory and vomeronasal sensory epithelial cells and axons revealed by monoclonal antibodies to carbohydrate antigens Gerald A. Schwarting 1'2 and James E. Crandall 1'2 Departments of 1Biochemistry and 2Developmental Neurobiology, E. K. Shriver Center, Waltham, MA 02254 (U.S.A.) (Accepted 13 November 1990) Key words: Olfactory; Vomeronasal; Glycolipid; Glycoprotein; Monoclonal antibody
Cell surface glycoconjugates are believed to play an important role in cell-cell interactions during development of CNS pathways. In order to identify developmentally regulated glycoconjugates in the nervous system, monoclonal antibodies were raised and selected for reactivity with carbohydrate antigens. Three monoclonal antibodies were identified, each of which reacts with a defined carbohydrate epitope and reveals a unique pattern of immunoreactivity within the olfactory sensory epithelia, vomeronasal and olfactory nerves and their terminal regions in rats. Antibody CCI reacts with a globoside-like glycolipid which contains a terminal N-acetylgalactosamine residue. CCl-immunoreactivity is present in just the vomeronasal organ, vomeronasal nerve and in the rostral half of the accessory olfactory bulb. Antibody CC2 reacts with a complex glycolipid which contains a branched chain oligosaccharide terminating with a-galactose and ct-fucose. CC2-immunoreactivity is seen throughout the vomeronasal organ, in dorsomedial regions of the olfactory sensory epithelia, in the vomeronasal and olfactory nerves, the accessory olfactory bulb and dorsomedial glomeruli of the main olfactory bulb. Antibody 1B2 reacts with lacto-N-glycosyl ceramides. IB2-immunoreactivity is highest at the luminal surfaces of receptor cells throughout the vomeronasal organ and in portions of the olfactory sensory epithelia. 1B2 is also expressed on the surface of a subset of receptor cell bodies, their dendrites and the proximal region of their axons in dorsomedial regions of the main olfactory epithelium.
INTRODUCTION The olfactory system of vertebrates displays several unique features which m a k e it very useful for studies of neural d e v e l o p m e n t and regeneration. T h r o u g h o u t the life of vertebrates, r e p l a c e m e n t of olfactory r e c e p t o r neurons occurs. Thus olfactory function is maintained as proliferation, neurite outgrowth, synapse formation, and cell d e a t h is taking place. H o w e v e r , the topographical organization of olfactory r e c e p t o r neuron projections to the olfactory bulb is p o o r l y understood, and there is not a great deal of information available on the molecular mechanisms which guide growing axons of olfactory sensory neurons to their targets. Recently, however, m o n o c l o n a l antibodies have been described which may be useful in characterizing molecularly distinct compartments within the olfactory system. Mori and co-workers described an antigen (R4B12) which defined subsets of olfactory nerve fibers in rabbits 16, and Schwob and G o t t l i e b have similarly described an antigen (RB-8) which is spatially segregated in the primary olfactory projection in rats 24. Both of these antigens are expressed on axons in the ventrolateral olfactory epithelium and in ~heir terminal portions in the ventrolateral olfactory bulb.
The RB-8 antigen has b e e n characterized as a 125-kDa cell surface glycoprotein 25. A l t h o u g h the function of this protein is not known, it is a c a n d i d a t e for involvement in axonal guidance. Several recent reports of c a r b o h y d r a t e expression in the olfactory system indicate that glycoconjugates are also t e m p o r a l l y and spatially regulated and thus may participate in guidance mechanisms. Lectins have been used to show that specific c a r b o h y d r a t e antigens are spatially regulated in the olfactory system. In Xenopus, soybean agglutinin l a b e l e d dorsal regions of the olfactory nerve and bulb m o r e intensely than ventral regions 13. Ulex europus agglutinin I was shown to bind exclusively to p r i m a r y olfactory and v o m e r o n a s a l neurons in the rat 5. M o n o c l o n a l antibodies to c a r b o h y d r a t e antigens have also defined d e v e l o p m e n t a l l y regulated and spatially restricted glycoconjugates in the rat olfactory system 3. In this study we r e p o r t that 3 unique c a r b o h y d r a t e moieties, as characterized by m o n o c l o n a l antibodies, are preferentially expressed in subsets of olfactory sensory neurons. O n e a n t i b o d y identifies subsets of the olfactory sensory epithelium and their axons within restricted portions of the olfactory bulb. A second a n t i b o d y stains vomeronasal sensory cells and their axons into the
Correspondence: G.A. Schwarting, Department of Biochemistry, E.K. Shriver Center, 200 Trapelo Road, Waltham, MA 02254, U.S.A.
240 accessory o l f a c t o r y bulb. A third a n t i b o d y stains the luminal surface of o l f a c t o r y sensory n e u r o n s in the v o m e r o n a s a i o r g a n as well as a subset of sensory cells in the m a i n o l f a c t o r y e p i t h e l i u m . P r e l i m i n a r y results h a v e b e e n r e p o r t e d p r e v i o u s l y 2~.
MATERIALS AND METHODS
Glycosphingolipid extraction Olfactory bulbs together with caudal regions of the olfactory epithelium were dissected from 60-day-old rats. Rostral regions of the olfactory epithelium and vomeronasal organ were not included in these dissections. The glycolipid antigens described below occur in very small amounts, thus requiring large amounts of tissue in order to visualize antibody-reactive species. This was best accomplished using these older animals. The tissue was homogenized in a small amount of water, extracted twice in 20 vol. chloroform:methanol (2:1) and the extracts were filtered and dried by rotary evaporation as previously described 26. Extracts were desalted by C-18 reversed-phase chromatography, and neutral and acidic glycosphingofipids (GSLs) were separated by diethyl aminoethyl (DEAE) Sephadex A-25 (Pharmacia, Piseataway, NJ) ion exchange chromatography. Acidic GSLs were desalted; neutral GSLs were treated with 0.6 N sodium hydroxine in methanol for 1 h, neutralized with 0.6 N HCi in methanol and desalted. After removal of solvent by evaporation under nitrogen, the GSLs were dissolved in chloroform:methanol (1:1).
Thin layer chromatogram-immunostaining GSL immunoehromatography was performed essentially as described by Yamamoto et al. 2s. GSI.s were separated by thin layer chromatography (TLC) on aluminum-backed silica gel 60 highperformance TLC (HPTLC) plates (E. Merck, Applied Analytical Industries, Wilmington, NC). Plates were developed in chloroform: methanol:water (60:35:8). The standard lane was cut from the plate and visualized with orcinol (0.2%). Plates were then coated with 0.05% polyisobutyl methacrylate (Polyscience, Warrington, PA) in hexane, dried thoroughly, and soaked in 0.05 M PBS, pH 7.4, containing 1% bovine serum albumin (BSA) for 1 h, The plates were incubated with antibody (undiluted hybridoma supernatant) for 2 h at 4 °C, then with horseradish peroxidase (HRP-conjugated goat anti-mouse IgG/IgM) (Boehringer Mannheim, Indianapolis, IN) at a 1:400 dilution, for 1 h at room temperature. Plates were rinsed with PBS, and developed with 33 mM 4-chloro-l-naphthol (Sigma, St. Louis, MO) in 0.02 M Tri-HCI buffer, 0.5 M NaCI, pH 7.5, containing 20% methanol and 0.02% H20 2.
Protein blotting Adult olfactory bulbs were homogenized in 0.05 M Tris pH 7.4 buffer containing 1% Triton X-100, 1% aprotinin, and 0.1% leupeptin. The homogenate was centrifuged at 100,000 g for 1 h. 20 ~g of protein was boiled in sample buffer (1.5% Triton X-100, 20% glycerol, 10% fl-mereaptoethanol, 4% sodium dodecylsulfate in 0.1 M Tris, pH 6.8), and run on a 12% polyacrylamide gel. It was transblotted to 0.45/~m nitrocellulose (Bio-Rad) exposed to primary antibody (hybridoma supernatant diluted 1:1) overnight at 4 °C, followed by peroxidase conjugated goat anti-mouse IgM for 2 h at room temperature. Immunoreactive proteins were visualized using 4-chloronaphthol as described above for TLC-immunostaining. For N-glycosidase F treatment, 20/ag of protein was boiled in sample buffer, diluted in phosphate buffer, pH 8.6 with 1.0% NP-40 and incubated with 2 units of enzyme overnight at 37 °C.
products were screened by an antibody-binding assay, ant] i~v immunochromatography against PC12 glycolipids. The production and use of antibodies CCI ~md CC2 have been previously described in the study of PC12 cells:~. The carbohydrate specificity of monoclonal antibody 1B2 was studied by Young et at. :'~ and was obtained from The American Type Culture Collection (Rockville, MD). Each is a mouse IgM, reacting with specific carbohydrate epitopes described in more detail below.
lmmunohistochernistry Rats (Sprague-Dawley) were deeply anesthetized with ether, and perfused through the left ventricle with 0.9% saline, then with 2% paraformaldehyde in 0.1 M PB (pH 7.4). Alternative fixatives included 2% paraformaldehyde with 2% glutaraldehyde, as well as 2% paraformaldehyde with 0.2% glutaraldehyde. All fixatives gave similar results. The olfactory epithelium and olfactory bulbs were removed and kept in 15% sucrose/2% paraformaldehyde overnight at 4 °C, then in 15% sucrose for 24 h at 4 °C. For animals that were less than 3 days old, the snout was dissected from the overlying skin and kept intact with the olfactory nerves and bulbs. Serial coronal or sagittal brain sections (12-16#m thick) were cut in a cryostat and thaw-mounted on gelatin-coated slides. Immunohistochemistry was performed on tissue sections as described by Yamamoto et al. 2s. Sections were exposed to the primary antibody overnight at 4 °C. After a 30-min incubation with 0.05 M PBS, pH 7.4, containing 1% normal goat serum (NGS), sections were incubated with goat anti-mouse IgM (Boehringer Mannheim, diluted 1:100 in 1% NGS in PBS) for 2 h at room temperature, tmmunoreactivity was visualized with DAB in PBS containing (I,004% H202. Some sections were stained using fluorescence techniques, with fluorescein-eonjugated secondary antibodies (Boehringer-Mannheim). This was especially useful for visualizing immunoreactive elements in the main olfactory epithelium and vomeronasal organ. Avidinbiotin (Vector Labs) conjugated antibody techniques were also tested. However, this method did not improve the quality of immunohistochemistry, probably due to the IgM primary antibodies used in this study. Sections reacted with peroxidase or avidin-biotin conjugated secondary antibodies were dehydrated and coverslipped, or eounterstained with Cresyl violet, dehydrated and coverslipped~ Sections reacted with fluorescence-conjugated secondary antibodies were coverslipped with PBS-glycerol. Control sections were incubated with NGS, fetal calf serum or an IgM antibody which does not react with postnatal olfactory tissue.
RESULTS
Biochemistry of glycolipid antigens O l f a c t o r y bulbs a n d a s s o c i a t e d c a u d a l r e g i o n s o f the o l f a c t o r y e p i t h e l i u m w e r e e x t r a c t e d for e i t h e r p r o t e i n s o r lipids, and the s e p a r a t e fractions w e r e r e a c t e d w i t h the a n t i b o d i e s C C 1 , C C 2 and 1B2. E a c h of the a n t i b o d i e s reacts with a u n i q u e glycotipid as d e m o n s t r a t e d using thin layer c h r o m a t o g r a m - i m m u n o s t a i n i n g t e c h n i q u e s ( F i g . 1). CC1 and 1B2 r e a c t with glycolipids which m i g r a t e o n thin layer c h r o m a t o g r a m s
n e a r g l o b o s i d e , a n d C C 2 reacts
with a m a j o r c o m p l e x glycolipid which is s e e n m i g r a t i n g n e a r the origin in a d d i t i o n to t w o m i n o r f a s t e r - m i g r a t i n g glycolipids (Fig. 1). T h e r e are n o d e t e c t a b l e i m m u n o r e -
Production of monoclonal antibodies
active glycolipids p r e s e n t in w h o l e b r a i n e x t r a c t s after the
Balb/c mice were immunized intraperitoneaUy at 2-week intervals with a cell suspension from the rat pheochromocytoma cell line, PC12. Mice were sacrificed 4 days after the final immunization, and spleen cells were fused with the NS-1 myeloma cell line. Cells were plated in 96-well microtiter dishes (Corning, Corning, NY). Fusion
o l f a c t o r y tissue has b e e n r e m o v e d . T h i s i n d i c a t e s that t h e s e a n t i g e n s are e x p r e s s e d at m u c h h i g h e r c o n c e n t r a tions in the o l f a c t o r y system t h a n in o t h e r C N S regions. T a k e n t o g e t h e r with i m m u n o c y t o c h e m i c a l e v i d e n c e ( d a t a
241
CTH
GLOBO
AGM1
116
--84 --58
O
B
O
1B2
B
o
CC1
B
S
CC2
Fig. l. Olfactory glycolipid antigens. Thin layer chromatography and immunostaining of the neutral glycolipid fraction obtained from olfactory tissue (O) compared to non-olfactory brain (B) tissue from adult rats. Standards (S) are at the right. CTH, globotriaosylceramide; GLOBO, globotetraosylceramide; AGM1, gangliotetraosylceramide.
--48
not shown) that little or no CC1, CC2 or 1B2 immunoreactivity is seen in the remainder of the CNS, these data suggest that the CC1, CC2 and 1B2 glycolipids and glycoproteins may be olfactory specific antigens within the CNS of rats. Although none of these glycoconjugates has been completely characterized from olfactory tissue, studies using exoglycosidases to remove the terminal sugar residue of the glycolipids suggest that CC1 reacts with a globoside-like glycolipid which terminates in iI-N-acetyl-galactosamine, and CC2 reacts with a complex branched chain glycoconjugate which contains terminal ,t-galactose and a-fucose 23. 1B2 has been shown previously to react with glycolipids which terminate with fl-galactose 29. Immunoblotting analysis of proteins obtained from adult olfactory bulbs indicated that there were no detectable CC1 proteins and weak 1B2 reactive proteins were detectable only following neuraminidase treatment. However, CC2 antibodies recognized a prominent protein at 38 k D a and a less prominent protein at 55 k D a (Fig. 2). However, reactivity was not abolished by
--36
"FABLE I
Antibody reactivity in olfactory structures VNO, vomeronasal organ; MOE, main olfactory epithelium; AOB, accessory olfactory bulb; OB, olfactory bulb.
Antibody
CC1 CC2 1B2
Structure VNO
MOE
AOB
OB
+ + +
+ +
+ + -
+ +
m26
Fig. 2. Immunoblot analysis of the cell lysate from rat olfactory bulb using the CC2 monoclonal antibody. An intense 38-kDa band and a minor 55-kDa band were specifically detected.
treatment with N-glycosidase F (Genzyme), suggesting that CC2 antibodies may not recognize a typical N-linked glycoprotein.
lmmunohistochemistry of glycolipid antigens The sensory neurons in the vomeronasal epithelia send their axons in the vomeronasal nerves to terminate in glomeruli of the accessory olfactory bulb. Similarly, the sensory neurons in the main olfactory epithelia project their axons in the olfactory nerves to terminate in the glomeruli of the main olfactory bulb. Results are organized below by each sensory epithelium and target structure in the CNS, and are outlined in Table I. Vomeronasal epithelium (VNO) The vomeronasal organ (Fig. 3A) consists of the receptor cell epithelial layer at the right and a respiratory epithelium (arrowheads), separated by a luminal space (asterisk). CC2 immunoreactivity is observed throughout the VNO. (Fig. 3B). Dendrites, axons and cell bodies of
242
Fig. 3. Vomeronasal organ. A: phase-contrast micrograph of a coronal section of the rat VNO. Calibration bar = 50/~m. Asterisk marks the lumen, arrowheads mark a portion of the respiratory epithelium. B: CC2 immunoreactivity in the VNO. The cell bodies and surfaces of many receptor cells show immunoreactivity. The respiratory epithelium is also stained. C: 1B2 immunoreactivity is highest at the luminal surface of all receptor cells but is also detected on numerous fiber-like structures in the epithelium. The luminal surface of the respiratory epithelium is also stained. D: CC1 immunoreactivity is present on cell surfaces of many receptor cells but not on the respiratory cpithelium
most olfactory receptor cells are stained while basal cells and supporting cells are not. Respiratory epithelial cells are also CC2 positive. The basement membranes of the vomeronasal and respiratory epithelia are especially heavily stained. 1B2 immunoreactivity is observed mainly on the luminal surface (asterisk in Fig. 3A) of olfactory receptor cells in the V N O , and weakly on processes of receptor cells (Fig. 3C). No 1B2 immunoreactivity could be seen on axons or vomeronasal nerve bundles. Respiratory epithelial cells are strongly reactive at the luminal surface. CC1 immunoreactivity is observed on dendrites, cell somata, and axons (Fig. 3D) in a pattern similar to that of CC2 immunoreactivity. Nearly all olfactory receptor cells in the V N O appear to express the CC1 antigen. However, unlike CC2 immunoreactivity, respiratory epithelial cells are CC1 negative.
Main olfactory epithelium CC2 immunoreactivity is seen in a subset of olfactory receptor cells (Fig. 4B). In CC2-positive regions, most cells are positive, and most cellular elements are positive within those regions including dendrites, cell bodies and axons (Fig. 5A). In other regions of the epithelium, no CC2 immunoreactivity is seen. Approximately onequarter to one-third of the olfactory epithelium is CC2 positive, mainly in dorsomedial regions. 1B2 immunoreactivity is observed on a subset of olfactory receptor cells (Fig. 4C). As in the VNO, immunoreactivity is concentrated at the luminal surface of receptor cells. 1B2 is also expressed mainly in dorsomediai regions of the olfactory epithelium and there is considerable overlap with CC2-positive areas. However, 1B2 and CC2 recognize separate antigens, each occurring in a distinct distribution within the nasal cavity. Using fluorescence immunocytochemistry (Fig. 5B), a
243 s u b s e t of r e c e p t o r cells is visible. T h e surfaces of cell
stain with 1B2 a n t i b o d i e s . I m m u n o r e a c t i v i t y disappears
b o d i e s , d e n d r i t e s , d e n d r i t i c k n o b s , a n d p r o x i m a l axons
as the axons p r o j e c t t h r o u g h t h e basal l a m i n a .
Accessory olfactory bulb (AOB) CC1 i m m u n o r e a c t i v i t y w i t h i n the A O B is d i s t r i b u t e d in the v o m e r o n a s a l n e r v e layer, the g l o m e r u l a r layer a n d o u t e r p o r t i o n s of the e x t e r n a l p l e x i f o r m layer (Fig. 6 A , B). A n t i g e n expression is also restricted in the rostroc a u d a l axis to the rostral half of the A O B . This is particularly e v i d e n t in the sagittal p l a n e of section (Fig.
7A). C C 2 i m m u n o r e a c t i v i t y w i t h i n the A O B is d i s t r i b u t e d similarly b u t n o t identically to CC1 i m m u n o r e a c t i v i t y (Fig. 6C, D). A biphasic g r a d i e n t of reactivity can be seen in which s t r o n g e r reactivity is e v i d e n t in the v o m e r o n a s a l n e r v e a n d g l o m e r u l a r layer b u t w e a k reac-
|
< P / f
/I
A
."Qp
0
J
C Fig. 4. Olfactory epithelium. A: Nissl-stained section of the olfactory epithelium• Arrows delimit the extent of the olfactory cpithelium. Regions in boxes A and B are shown at higher power in Fig. 5. Turbinates I, lid and Ilv are labeled. B: CC2 immunoreactivity is restricted to dorsomedial regions of the olfactory cpithelium (arrows)• Bundles of the vomeronasal nerve are also stained (arrowheads). C: 1B2 staining also is limited to dorsomedial regions of the olfactory epithelium. Staining of 1B2-positive receptor cell bodies is not visible at this low power with peroxidaseconjugated techniques. Calibration bar = 500 #m for A-C.
Fig. 5. Olfactory epithelium. A: CC2 immunofluorescence staining of a dorsomedial region (see box A in Fig. 4) of olfactory epithelium. Immunoreactivity is detected on most receptor cell bodies, their dendrites and axon bundles (arrow)• Calibration bar = 50 pm for A and B. B:IB2 immunofluorescence staining of a dorsomedial region (see box B in Fig. 4) of P2 olfactory epithelium. Immunoreactivity is present on a subset of receptor cell bodies, their dendrites, and their dendritic knobs. 1B2 is expressed on the proximal region of axons, as they project along the basal lamina but is not detected in olfactory nerve bundles (arrow).
244
Fig. 6. Accessory olfactory bulb. (A, B) CC1 immunoreactivity in coronal sections through the rat accessory olfactor~ bulb. A: darkfield illumination of the section in B shows clearly the extent of the CCI immunostaining to include the VNL and G[.. B: section counterstained with Cresyl violet shows the layers of the AOB. VNN, vomeronasal nerve; VNL, vomeronasal nerve layer; GL, glomerular layer; EPL, external plexiform layer. C, D: CC2 immunoreactivity in an adjacent section. Nissl-counterstained section reveals the layers as in B. C: darkfield illumination of the section in D shows CC2 immunoreactivity extending beyond the VNL and GL into the EPL. Calibration bar = 100 !~m for A-D.
tivity is evident in the external plexiform layer. CC2 immunoreactivity appears to extend d e e p e r into the external plexiform layer than CC1 immunoreactivity. As seen in sagittal sections, CC2 immunoreactivity is present t h r o u g h o u t the A O B and is not restricted in its expression in a rostrocaudal gradient, as is seen with CC1 a n t i b o d y (Fig. 7B).
Olfactory bulb (OB) CC2 immunoreactivity is seen in the olfactory nerve layer, and is c o n c e n t r a t e d in d o r s o m e d i a l areas of the olfactory bulb (Fig. 8). O n l y glomeruli in d o r s o m e d i a l regions are filled with CC2 immunoreactivity. In lateral and ventral regions of the olfactory bulb little CC2 i m m u n o r e a c t i v i t y is o b s e r v e d in the olfactory nerve layer
and no staining could be d e t e c t e d in glomeruli in these regions. We have also d e t e c t e d weak 1B2 staining of selected glomeruli in the olfactory bulb (data not shown). This staining would a p p e a r to be associated with terminal portions of some olfactory r e c e p t o r axons, even though 1B2 immunoreactivity is not d e t e c t a b l e on m o r e proximal portions of the axon that travel from the olfactory epithelium to the olfactory bulb. DISCUSSION T h r e e monoclonal antibodies reacting with glycolipid and glycoprotein antigens are described. E a c h defines a selective group of cells found uniquely within the olfac-
245 tory system of rats (summarized in Fig. 9). This diagram depicts the location of the CC1, CC2 and 1B2 antigens and the structure of their terminal carbohydrate epitopes. lmmunocytochemical studies in progress revealed few differences from embryonic day 15 to adult (data not shown) suggesting that the regulation of expression of the
glycolipids described here is dissimilar to the majority of stage-specific embryonic carbohydrate antigens which have been described in the developing nervous system22. However, a detailed analysis of the development profile of these antigens has not yet been completed and subtle differences in their expression during development may
lit 4
!!A m
Fig. 7. Accessory olfactory bulb. A: CC1 immunoreactivity in sagittal sections through the rat accessory olfactory bulb. Staining is present in rostral (solid arrow) but not caudal (open arrow) portions of the AOB. B: CC2 immunoreactivity in an adjacent section shows staining throughout the rostrocaudal extent of the AOB. Straight arrows show staining in the dorsal and rostral portions of the main olfactory bulb. C: Nissl-counterstained section. The layers of the main olfactory bulb and the AOB can be seen. Calibration bar = 100 jtm for A - C .
Fig. 8. Olfactory bulb. CC2 immunoreactivity in the main olfactory bulb of the rat. A: at low power in a coronal section dorsal and medial regions of the olfactory bulb (arrowheads) are stained as well as the vomeronasal nerve (small arrows). B and C: darkfield (B) and brightfield (C) views of CC2 immunoreactivity in a Nissl-counterstained section. Only glomeruli in this dorsomedial region of the olfactory bulb (between the arrows) contain CC2 immunoreactivity. Calibration bars = 100 ~m for A and 50/~m for B and C.
246 A CC1
~- GALNAC...
B
CC2
C 1B2
:i~~
~-GAL...
OE
Fig. 9. Schematic summary of the localization of the glycolipid antigens (left) and their carbohydrate specificity (risht)/for CCI (A), CC2 (B) and 1B2 (C) antibodies. AOB, accessory olfactory bulb; FUC, fucose; GAL, galactose; GALNAC, n-acetylglucosamine; OB, olfactory bulb; OE, olfactory epithelium; VNO, vomeronasal organ.
help to reveal their functional significance. The biochemical specificity and the anatomical localization of the 3 monoclonal antibodies presented here are similar, yet different, from antibodies and lectins which have been previously described. CC1 antibody recognizes an N-acetyl-galactosamine terminal glycolipid, which could be similar to antigens recognized by the lectin soybean agglutinin (SBA). In Xenopus, the ventral portion of the olfactory nerves and bulb contained more SBA reactivity than the dorsal portions 13. In rats SBA has been shown to bind subsets of olfactory bulb glomeruli 2°. Although SBA and CC1 antibodies both recognize fl-N-acetyl-galactosamine, their reactivity in the olfactory system appears to differ considerably. Likewise, antibodies to CC2 and to blood group B antigens both recognize similar branched chain oligosaccharides containing a-galactosyl and a-fucosyl structures. However, CC2 and blood group B immunoreactivities differ considerably in the olfactory epithelium 3. Blood group B antigens are expressed on only a few neuroreceptor cells up to postnatal day 3, after which most main olfactory and vomeronasal receptor cells become positive. CC2, however, is continuously expressed on a subset of the olfactory epithelium from embryonic day 14
or earlier. The anatomical localization of CC2 is also similar to 2B8 antigens, a group of glycoproteins in the rat olfactory system located predominantly in dorsomedial regions of the main olfactory epithelium and bulb j'~ However, CC2 antibodies detect proteins with different molecular weights than the previously described, 2B8 reactive glycoproteins. This suggests that the two antibodies are distinct. 1B2 immunoreactivity is similar to an anti-carbohydrate antibody (4C9) that reacts with fucosyl-poly Nacetyllactosamine described by Mori 15. 4C9 also reacts strongly with structures at the luminal surface of the vomeronasal neuroepithelium and respiratory epithelium. However, 4C9 also stains axons and their projection sites in the olfactory bulb, whereas, with the current protocols, we are not able to detect 1B2 immunoreactivity in axons. At P14 we have been able to detect 1B2 immunoreactivity in subsets of glomeruli in the OB. Mori et al. ~4 previously demonstrated, using monoclohal antibodies reacting with restricted regions of the rabbit AOB, that there is no correlation of the location of receptor somata in the vomeronasal organ with their termination site in the AOB. This finding is confirmed here with an additional monoclonal antibody in the rat. CCl-positive cell bodies are distributed throughout the vomeronasal organ while CCl-positive axons terminate only in the rostral half of the AOB. This pattern of sensory neuron to target structure resembles the organization of cutaneous afferent fiber projections to the spinal cord in which subtypes intermix with fibers of different subtypes in the nerve bundles and dorsal roots ~'ls. However, different subtypes of fibers terminate in separate laminae of the dorsal horn of the spinal cord. Segregation of dorsal root ganglion neurons on the basis of their characteristic patterns of carbohydrate cell surface antigens has been studied in detail by Jessell and co-workers 9. They showed in sensory neurons that lacto-series and globo-series carbohydrate antigens were expressed on non-overlapping populations of dorsal root ganglia neurons and in their terminals in the dorsal horn of the spinal cord s . Lactose binding proteins were detected in the same cells which expressed the lactoseries carbohydrates, but were also released from cultured D R G neurons. Thus, the binding proteins may be laid down in the extracellular space and therefore play a key role in the guidance of axons between D R G neurons and their target cells 19. It is possible that the carbohydrate antigens described here, which define subsets of olfactory receptor cells, play a role in the guidance of axons from the olfactory receptor neurons to their targets in the MOB and AOB, just as the lacto-series carbohydrates are involved in guidance of D R G fibers to their targets in the spinal cord,
247 In addition, it is conceivable that these chemical cues are
between dorsal
continuously needed along the pathway leading from the
epithelium. This suggests that RB-8 may be involved in
olfactory receptor cells to the bulb, due to the fact that the t u r n o v e r of receptor ceils is ongoing throughout the life of the animal4'11'17.
directing axons from the ventral olfactory epithelium to ventral regions of the olfactory bulb. Similarly, CC2 may
The relationship of CC1 and CC2 antigens to previously described olfactory axon surface antigens is interesting. I m a m u r a et al. 12 described a monoclonal antibody (R4B12) which stained a subset of vomeronasal nerve fibers that t e r m i n a t e d in the rostrolateral portion of the A O B . A second antibody (R2D5) identified a complim e n t a r y subgroup of fibers which terminated in the caudomedial glomeruli of the A O B 14'15. W h e t h e r either of these antigens is structurally related to CC1 oligosaccharides which, like R4B12, are expressed preferentially in the rostral half of the A O B , is not known. A n additional antigen (RB-8) has been described 24, which
and ventral zones of the
olfactory
direct dorsomedially situated olfactory axons to the dorsomedial glomeruli of the olfactory bulb. Likewise, CC1 antigen may function to direct vomeronasal axons to the rostral portion of the A O B . Studies underway which utilize primary cultures of vomeronasal and olfactory epithelial n e u r o n s 7 will permit us to study the role played by CC1, CC2 and 1B2 antigens in the organization of axon projections in the olfactory system. Co-cultures of olfactory epithelial n e u r o n s and olfactory bulb explants are already proving valuable for analyzing the interaction between carbohydrate antigens on axon surfaces and structures in the extracellular matrix or on the surfaces of adjacent cells.
complements CC2 expression in the main olfactory epithelium and in the glomeruli of the olfactory bulb. RB-8 is expressed in ventrolaterai regions of the olfactory epithelium and of the olfactory bulb, whereas CC2 is expressed in the dorsomedial regions of the olfactory epithelium and olfactory bulb. RB-8 has been identified as a 125-kDa glycoprotein and its expression is regulated REFERENCES 1 Allen, W.K. and Akeson, R., Identification of a cell surface glycoprotein family of olfactory receptor neurons with a monoclonal antibody, J. Neurosci., 5 (1985) 284-296. 2 Allen, W.K. and Akeson, R., Identification of an olfactory receptor neuron subclass: cellular and molecular analysis during development, Dev. Biol., 109 (1985) 393-401. 3 Astic, L., Le Pendu, J., Mollicone, R., Saucier, D. and Oriol, R., Cellular expression of H and B antigens in the rat olfactory system during development, J. Comp. Neurol., 289 (1989) 386-394. 4 Barber, P.C. and Raisman, G., Cell division in the vomeronasal organ of the adult mouse, Brain Research, 141 (1978) 57-66. 5 Barber, P.C., Ulex europus agglutinin I binds exclusively to primary olfactory neurons in the rat nervous system, Neuroscience, 30 (1989) 1-9. 6 Brown, A.G., Organization in the Spinal Cord, Springer, New York, 1981. 7 Calof, A.L. and Chikaraishi, D.M., Analysis of neurogenesis in a mammalian neuroepithelium: proliferation and differentiation of an olfactory neuron precursor in vitro, Neuron, 3 (1989) 115-127. 8 Dodd, J. and Jessell, T.M., Lactoseries carbohydrates specify subsets of dorsal root ganglion neurons projecting to the superficial horn of rat spinal cord., J. Neurosci., 5 (1986) 3278-3294. 9 Dodd, J., Solter, D. and Jessell, T.M., Monoclonal antibodies against carbohydrate differentiation antigens identify subsets of primary sensory neurons, Nature, 314 (1984) 469-472. l0 Fujita, S.C., Mori, K., Imamura, K. and Obata, K., Subclasses of olfactory receptor cells and their segregated central projections demonstrated by a monoclonal antibody, Brain Research, 326 (1985) 192-196. 11 Graziadei, P.P.C. and Monti Graziadei, G.A., Neurogenesis and neuron regeneration in the olfactory system of mammals. I. Morphological aspects of differentiation and structural organi-
Acknowledgements. We thank Drs. T. Schoenfeld, S. Tobet and M. Yamamoto for their helpful discussion and critical reading of the manuscript. We gratefully appreciate the technical assistance of Gail Deutsch, Devin Gattey, Darlene Butler, and Ms. Denise Brescia for her expert assistance in the preparation of the manuscript. Portions of this work were supported by NIH Grants HD05515 and NS25580 to G.A.S. and NS24386 to J.E.C.
zation of the olfactory sensory neurons, J. Neurocytol., 8 (1978) 1-8. 12 Imamura, K., Mori, K., Fujita, S.C. and Obata, K., Immunochemical identification of subgroups of vomeronasal nerve fibers and their segregated terminations in the accessory olfactory bulb, Brain Research, 328 (1985) 362-366. 13 Key, B. and Giorgi, P.P., Selective binding of soybean agglutinin to the olfactory system of Xenopus, Neuroscience, 18 (1986) 507-515. 14 Mori, K., Fujita, S.C., Imamura, K. and Obata, K., Immunohistochemical study of subclasses of olfactory nerve fibers and their projections to the olfactory bulb in the rabbit, J. Cornp. Neurol., 242 (1985) 214-229. 15 Mori, K., Monoclonal antibodies (2C5 and 4C9) against lactoseries carbohydrates identify subsets of olfactory and vomeronasal receptor cells and their axons in the rabbit, Brain Research, 408 (1987) 215-221. 16 Mori, K., Imamura, K., Fujita, S.C. and Obata, K., Projections of two subclasses of vomeronasal nerve fibers to the accessory olfactory bulb in the rabbit, Neuroscience, 20 (1987) 259-278. 17 Moulton, D.G., Celebi, G. and Fink, R.P., Olfaction in mammals - - two aspects: proliferation of cells in the olfactory epithelium and sensitivity to odours. In G.E.W. Wolstenholme and J. Knight, (Eds.), Ciba Foundation Symposium on Taste and Smell in Vertebrates, Churchill, London, 1970, pp. 227-250. 18 Perl, E.R., Characterization of nociceptors and their activation of neurons in the superficial dorsal horn: first steps for the sensation of pain, Adv. Pain Res. Ther., 6 (1983) 23-51. 19 Regan, L.J., Dodd, J., Barondes, S.H. and Jessell, T.M., Selective expression of endogenous lactose-binding lectins and lactoseries glycoconjugates in subsets of rat sensory neurons, Proc. Natl. Acad. Sci. U.S.A., 83 (1986) 2248-2252. 20 Riggott, M.J. and Scott, J.W., Lectin labeling of rat olfactory bulb glomeruli, Soc. Neurosci. Abst., 15 (1989) 926. 21 Schwarting, G.A. and Crandall, J.E., Monoclonal antibodies to carbohydrate antigens identify subsets of olfactory and vomeronasal axons in the rat, Soc. Neurosci. Abst., 15 (1989) 590.
248 22 Schwarting, G.A. and Yamamoto, M., Expression of glycoconjugates during development of the vertebrate nervous system, BioEssays, 9 (1988) 19-23. 23 Schwarting, G.A., Tischler, A.S. and Donahue, S.R., Fucosylation of glycolipids in PC12 cells is dependent on the sequence of nerve growth factor treatment and adenylate cyclase activation, Develop. Neurosci., 12 (1989) 159-171. 24 Sehwob, J.E. and Gottlieb, D.I., The primary olfactory projection has two chemically distinct zones, J. Neurosci., 6 (1986) 3393-3404. 25 Schwob, J.E. and Gottlieb, D.I., Purification and characterization of an antigen that is spatially segregated in the primary olfactory projection, J. Neurosci., 8 (1987) 3470-3480. 26 Suchy, S.F., Yamamoto, M., Barbero, L. and Schwarting, G.A., A monoclonal antibody WCC4 recognizes a developmentally
regulated ganglioside containing ~z-galactose and a-tucose present in the rat nervous system, Brain Research, 440 (1988) 25-34. 27 Towbin, H., Staehelin, T. and Gordon, J., Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets: procedure and some applications. Proc. Natl. Acad. Sci. U.S.A., 76 (1979) 4350-4354. 28 Yamamoto, M., Boyer, A. and Schwarting, G.A., Fucosccontaining glycolipids are stage- and region-specific antigens in developing embryonic brain of rodents, Pr~c. Natl. Acad. Sci. U.S.A., 82 (1985) 3045-3049. 29 Young Jr., W.W., Portoukalien, J. and Hakomori, S.-t., Two monoclonal anticarbohydrate antibodies directed to glycosphingolipids with a Lacto-N-glycosyl Type II chain, J: Biol. Chem., 256 (1981) 10967-10977.
239
BRES 16542
Subsets of olfactory and vomeronasal sensory epithelial cells and axons revealed by monoclonal antibodies to carbohydrate antigens Gerald A. Schwarting 1'2 and James E. Crandall 1'2 Departments of 1Biochemistry and 2Developmental Neurobiology, E. K. Shriver Center, Waltham, MA 02254 (U.S.A.) (Accepted 13 November 1990) Key words: Olfactory; Vomeronasal; Glycolipid; Glycoprotein; Monoclonal antibody
Cell surface glycoconjugates are believed to play an important role in cell-cell interactions during development of CNS pathways. In order to identify developmentally regulated glycoconjugates in the nervous system, monoclonal antibodies were raised and selected for reactivity with carbohydrate antigens. Three monoclonal antibodies were identified, each of which reacts with a defined carbohydrate epitope and reveals a unique pattern of immunoreactivity within the olfactory sensory epithelia, vomeronasal and olfactory nerves and their terminal regions in rats. Antibody CCI reacts with a globoside-like glycolipid which contains a terminal N-acetylgalactosamine residue. CCl-immunoreactivity is present in just the vomeronasal organ, vomeronasal nerve and in the rostral half of the accessory olfactory bulb. Antibody CC2 reacts with a complex glycolipid which contains a branched chain oligosaccharide terminating with a-galactose and ct-fucose. CC2-immunoreactivity is seen throughout the vomeronasal organ, in dorsomedial regions of the olfactory sensory epithelia, in the vomeronasal and olfactory nerves, the accessory olfactory bulb and dorsomedial glomeruli of the main olfactory bulb. Antibody 1B2 reacts with lacto-N-glycosyl ceramides. IB2-immunoreactivity is highest at the luminal surfaces of receptor cells throughout the vomeronasal organ and in portions of the olfactory sensory epithelia. 1B2 is also expressed on the surface of a subset of receptor cell bodies, their dendrites and the proximal region of their axons in dorsomedial regions of the main olfactory epithelium.
INTRODUCTION The olfactory system of vertebrates displays several unique features which m a k e it very useful for studies of neural d e v e l o p m e n t and regeneration. T h r o u g h o u t the life of vertebrates, r e p l a c e m e n t of olfactory r e c e p t o r neurons occurs. Thus olfactory function is maintained as proliferation, neurite outgrowth, synapse formation, and cell d e a t h is taking place. H o w e v e r , the topographical organization of olfactory r e c e p t o r neuron projections to the olfactory bulb is p o o r l y understood, and there is not a great deal of information available on the molecular mechanisms which guide growing axons of olfactory sensory neurons to their targets. Recently, however, m o n o c l o n a l antibodies have been described which may be useful in characterizing molecularly distinct compartments within the olfactory system. Mori and co-workers described an antigen (R4B12) which defined subsets of olfactory nerve fibers in rabbits 16, and Schwob and G o t t l i e b have similarly described an antigen (RB-8) which is spatially segregated in the primary olfactory projection in rats 24. Both of these antigens are expressed on axons in the ventrolateral olfactory epithelium and in ~heir terminal portions in the ventrolateral olfactory bulb.
The RB-8 antigen has b e e n characterized as a 125-kDa cell surface glycoprotein 25. A l t h o u g h the function of this protein is not known, it is a c a n d i d a t e for involvement in axonal guidance. Several recent reports of c a r b o h y d r a t e expression in the olfactory system indicate that glycoconjugates are also t e m p o r a l l y and spatially regulated and thus may participate in guidance mechanisms. Lectins have been used to show that specific c a r b o h y d r a t e antigens are spatially regulated in the olfactory system. In Xenopus, soybean agglutinin l a b e l e d dorsal regions of the olfactory nerve and bulb m o r e intensely than ventral regions 13. Ulex europus agglutinin I was shown to bind exclusively to p r i m a r y olfactory and v o m e r o n a s a l neurons in the rat 5. M o n o c l o n a l antibodies to c a r b o h y d r a t e antigens have also defined d e v e l o p m e n t a l l y regulated and spatially restricted glycoconjugates in the rat olfactory system 3. In this study we r e p o r t that 3 unique c a r b o h y d r a t e moieties, as characterized by m o n o c l o n a l antibodies, are preferentially expressed in subsets of olfactory sensory neurons. O n e a n t i b o d y identifies subsets of the olfactory sensory epithelium and their axons within restricted portions of the olfactory bulb. A second a n t i b o d y stains vomeronasal sensory cells and their axons into the
Correspondence: G.A. Schwarting, Department of Biochemistry, E.K. Shriver Center, 200 Trapelo Road, Waltham, MA 02254, U.S.A.
240 accessory o l f a c t o r y bulb. A third a n t i b o d y stains the luminal surface of o l f a c t o r y sensory n e u r o n s in the v o m e r o n a s a i o r g a n as well as a subset of sensory cells in the m a i n o l f a c t o r y e p i t h e l i u m . P r e l i m i n a r y results h a v e b e e n r e p o r t e d p r e v i o u s l y 2~.
MATERIALS AND METHODS
Glycosphingolipid extraction Olfactory bulbs together with caudal regions of the olfactory epithelium were dissected from 60-day-old rats. Rostral regions of the olfactory epithelium and vomeronasal organ were not included in these dissections. The glycolipid antigens described below occur in very small amounts, thus requiring large amounts of tissue in order to visualize antibody-reactive species. This was best accomplished using these older animals. The tissue was homogenized in a small amount of water, extracted twice in 20 vol. chloroform:methanol (2:1) and the extracts were filtered and dried by rotary evaporation as previously described 26. Extracts were desalted by C-18 reversed-phase chromatography, and neutral and acidic glycosphingofipids (GSLs) were separated by diethyl aminoethyl (DEAE) Sephadex A-25 (Pharmacia, Piseataway, NJ) ion exchange chromatography. Acidic GSLs were desalted; neutral GSLs were treated with 0.6 N sodium hydroxine in methanol for 1 h, neutralized with 0.6 N HCi in methanol and desalted. After removal of solvent by evaporation under nitrogen, the GSLs were dissolved in chloroform:methanol (1:1).
Thin layer chromatogram-immunostaining GSL immunoehromatography was performed essentially as described by Yamamoto et al. 2s. GSI.s were separated by thin layer chromatography (TLC) on aluminum-backed silica gel 60 highperformance TLC (HPTLC) plates (E. Merck, Applied Analytical Industries, Wilmington, NC). Plates were developed in chloroform: methanol:water (60:35:8). The standard lane was cut from the plate and visualized with orcinol (0.2%). Plates were then coated with 0.05% polyisobutyl methacrylate (Polyscience, Warrington, PA) in hexane, dried thoroughly, and soaked in 0.05 M PBS, pH 7.4, containing 1% bovine serum albumin (BSA) for 1 h, The plates were incubated with antibody (undiluted hybridoma supernatant) for 2 h at 4 °C, then with horseradish peroxidase (HRP-conjugated goat anti-mouse IgG/IgM) (Boehringer Mannheim, Indianapolis, IN) at a 1:400 dilution, for 1 h at room temperature. Plates were rinsed with PBS, and developed with 33 mM 4-chloro-l-naphthol (Sigma, St. Louis, MO) in 0.02 M Tri-HCI buffer, 0.5 M NaCI, pH 7.5, containing 20% methanol and 0.02% H20 2.
Protein blotting Adult olfactory bulbs were homogenized in 0.05 M Tris pH 7.4 buffer containing 1% Triton X-100, 1% aprotinin, and 0.1% leupeptin. The homogenate was centrifuged at 100,000 g for 1 h. 20 ~g of protein was boiled in sample buffer (1.5% Triton X-100, 20% glycerol, 10% fl-mereaptoethanol, 4% sodium dodecylsulfate in 0.1 M Tris, pH 6.8), and run on a 12% polyacrylamide gel. It was transblotted to 0.45/~m nitrocellulose (Bio-Rad) exposed to primary antibody (hybridoma supernatant diluted 1:1) overnight at 4 °C, followed by peroxidase conjugated goat anti-mouse IgM for 2 h at room temperature. Immunoreactive proteins were visualized using 4-chloronaphthol as described above for TLC-immunostaining. For N-glycosidase F treatment, 20/ag of protein was boiled in sample buffer, diluted in phosphate buffer, pH 8.6 with 1.0% NP-40 and incubated with 2 units of enzyme overnight at 37 °C.
products were screened by an antibody-binding assay, ant] i~v immunochromatography against PC12 glycolipids. The production and use of antibodies CCI ~md CC2 have been previously described in the study of PC12 cells:~. The carbohydrate specificity of monoclonal antibody 1B2 was studied by Young et at. :'~ and was obtained from The American Type Culture Collection (Rockville, MD). Each is a mouse IgM, reacting with specific carbohydrate epitopes described in more detail below.
lmmunohistochernistry Rats (Sprague-Dawley) were deeply anesthetized with ether, and perfused through the left ventricle with 0.9% saline, then with 2% paraformaldehyde in 0.1 M PB (pH 7.4). Alternative fixatives included 2% paraformaldehyde with 2% glutaraldehyde, as well as 2% paraformaldehyde with 0.2% glutaraldehyde. All fixatives gave similar results. The olfactory epithelium and olfactory bulbs were removed and kept in 15% sucrose/2% paraformaldehyde overnight at 4 °C, then in 15% sucrose for 24 h at 4 °C. For animals that were less than 3 days old, the snout was dissected from the overlying skin and kept intact with the olfactory nerves and bulbs. Serial coronal or sagittal brain sections (12-16#m thick) were cut in a cryostat and thaw-mounted on gelatin-coated slides. Immunohistochemistry was performed on tissue sections as described by Yamamoto et al. 2s. Sections were exposed to the primary antibody overnight at 4 °C. After a 30-min incubation with 0.05 M PBS, pH 7.4, containing 1% normal goat serum (NGS), sections were incubated with goat anti-mouse IgM (Boehringer Mannheim, diluted 1:100 in 1% NGS in PBS) for 2 h at room temperature, tmmunoreactivity was visualized with DAB in PBS containing (I,004% H202. Some sections were stained using fluorescence techniques, with fluorescein-eonjugated secondary antibodies (Boehringer-Mannheim). This was especially useful for visualizing immunoreactive elements in the main olfactory epithelium and vomeronasal organ. Avidinbiotin (Vector Labs) conjugated antibody techniques were also tested. However, this method did not improve the quality of immunohistochemistry, probably due to the IgM primary antibodies used in this study. Sections reacted with peroxidase or avidin-biotin conjugated secondary antibodies were dehydrated and coverslipped, or eounterstained with Cresyl violet, dehydrated and coverslipped~ Sections reacted with fluorescence-conjugated secondary antibodies were coverslipped with PBS-glycerol. Control sections were incubated with NGS, fetal calf serum or an IgM antibody which does not react with postnatal olfactory tissue.
RESULTS
Biochemistry of glycolipid antigens O l f a c t o r y bulbs a n d a s s o c i a t e d c a u d a l r e g i o n s o f the o l f a c t o r y e p i t h e l i u m w e r e e x t r a c t e d for e i t h e r p r o t e i n s o r lipids, and the s e p a r a t e fractions w e r e r e a c t e d w i t h the a n t i b o d i e s C C 1 , C C 2 and 1B2. E a c h of the a n t i b o d i e s reacts with a u n i q u e glycotipid as d e m o n s t r a t e d using thin layer c h r o m a t o g r a m - i m m u n o s t a i n i n g t e c h n i q u e s ( F i g . 1). CC1 and 1B2 r e a c t with glycolipids which m i g r a t e o n thin layer c h r o m a t o g r a m s
n e a r g l o b o s i d e , a n d C C 2 reacts
with a m a j o r c o m p l e x glycolipid which is s e e n m i g r a t i n g n e a r the origin in a d d i t i o n to t w o m i n o r f a s t e r - m i g r a t i n g glycolipids (Fig. 1). T h e r e are n o d e t e c t a b l e i m m u n o r e -
Production of monoclonal antibodies
active glycolipids p r e s e n t in w h o l e b r a i n e x t r a c t s after the
Balb/c mice were immunized intraperitoneaUy at 2-week intervals with a cell suspension from the rat pheochromocytoma cell line, PC12. Mice were sacrificed 4 days after the final immunization, and spleen cells were fused with the NS-1 myeloma cell line. Cells were plated in 96-well microtiter dishes (Corning, Corning, NY). Fusion
o l f a c t o r y tissue has b e e n r e m o v e d . T h i s i n d i c a t e s that t h e s e a n t i g e n s are e x p r e s s e d at m u c h h i g h e r c o n c e n t r a tions in the o l f a c t o r y system t h a n in o t h e r C N S regions. T a k e n t o g e t h e r with i m m u n o c y t o c h e m i c a l e v i d e n c e ( d a t a
241
CTH
GLOBO
AGM1
116
--84 --58
O
B
O
1B2
B
o
CC1
B
S
CC2
Fig. l. Olfactory glycolipid antigens. Thin layer chromatography and immunostaining of the neutral glycolipid fraction obtained from olfactory tissue (O) compared to non-olfactory brain (B) tissue from adult rats. Standards (S) are at the right. CTH, globotriaosylceramide; GLOBO, globotetraosylceramide; AGM1, gangliotetraosylceramide.
--48
not shown) that little or no CC1, CC2 or 1B2 immunoreactivity is seen in the remainder of the CNS, these data suggest that the CC1, CC2 and 1B2 glycolipids and glycoproteins may be olfactory specific antigens within the CNS of rats. Although none of these glycoconjugates has been completely characterized from olfactory tissue, studies using exoglycosidases to remove the terminal sugar residue of the glycolipids suggest that CC1 reacts with a globoside-like glycolipid which terminates in iI-N-acetyl-galactosamine, and CC2 reacts with a complex branched chain glycoconjugate which contains terminal ,t-galactose and a-fucose 23. 1B2 has been shown previously to react with glycolipids which terminate with fl-galactose 29. Immunoblotting analysis of proteins obtained from adult olfactory bulbs indicated that there were no detectable CC1 proteins and weak 1B2 reactive proteins were detectable only following neuraminidase treatment. However, CC2 antibodies recognized a prominent protein at 38 k D a and a less prominent protein at 55 k D a (Fig. 2). However, reactivity was not abolished by
--36
"FABLE I
Antibody reactivity in olfactory structures VNO, vomeronasal organ; MOE, main olfactory epithelium; AOB, accessory olfactory bulb; OB, olfactory bulb.
Antibody
CC1 CC2 1B2
Structure VNO
MOE
AOB
OB
+ + +
+ +
+ + -
+ +
m26
Fig. 2. Immunoblot analysis of the cell lysate from rat olfactory bulb using the CC2 monoclonal antibody. An intense 38-kDa band and a minor 55-kDa band were specifically detected.
treatment with N-glycosidase F (Genzyme), suggesting that CC2 antibodies may not recognize a typical N-linked glycoprotein.
lmmunohistochemistry of glycolipid antigens The sensory neurons in the vomeronasal epithelia send their axons in the vomeronasal nerves to terminate in glomeruli of the accessory olfactory bulb. Similarly, the sensory neurons in the main olfactory epithelia project their axons in the olfactory nerves to terminate in the glomeruli of the main olfactory bulb. Results are organized below by each sensory epithelium and target structure in the CNS, and are outlined in Table I. Vomeronasal epithelium (VNO) The vomeronasal organ (Fig. 3A) consists of the receptor cell epithelial layer at the right and a respiratory epithelium (arrowheads), separated by a luminal space (asterisk). CC2 immunoreactivity is observed throughout the VNO. (Fig. 3B). Dendrites, axons and cell bodies of
242
Fig. 3. Vomeronasal organ. A: phase-contrast micrograph of a coronal section of the rat VNO. Calibration bar = 50/~m. Asterisk marks the lumen, arrowheads mark a portion of the respiratory epithelium. B: CC2 immunoreactivity in the VNO. The cell bodies and surfaces of many receptor cells show immunoreactivity. The respiratory epithelium is also stained. C: 1B2 immunoreactivity is highest at the luminal surface of all receptor cells but is also detected on numerous fiber-like structures in the epithelium. The luminal surface of the respiratory epithelium is also stained. D: CC1 immunoreactivity is present on cell surfaces of many receptor cells but not on the respiratory cpithelium
most olfactory receptor cells are stained while basal cells and supporting cells are not. Respiratory epithelial cells are also CC2 positive. The basement membranes of the vomeronasal and respiratory epithelia are especially heavily stained. 1B2 immunoreactivity is observed mainly on the luminal surface (asterisk in Fig. 3A) of olfactory receptor cells in the V N O , and weakly on processes of receptor cells (Fig. 3C). No 1B2 immunoreactivity could be seen on axons or vomeronasal nerve bundles. Respiratory epithelial cells are strongly reactive at the luminal surface. CC1 immunoreactivity is observed on dendrites, cell somata, and axons (Fig. 3D) in a pattern similar to that of CC2 immunoreactivity. Nearly all olfactory receptor cells in the V N O appear to express the CC1 antigen. However, unlike CC2 immunoreactivity, respiratory epithelial cells are CC1 negative.
Main olfactory epithelium CC2 immunoreactivity is seen in a subset of olfactory receptor cells (Fig. 4B). In CC2-positive regions, most cells are positive, and most cellular elements are positive within those regions including dendrites, cell bodies and axons (Fig. 5A). In other regions of the epithelium, no CC2 immunoreactivity is seen. Approximately onequarter to one-third of the olfactory epithelium is CC2 positive, mainly in dorsomedial regions. 1B2 immunoreactivity is observed on a subset of olfactory receptor cells (Fig. 4C). As in the VNO, immunoreactivity is concentrated at the luminal surface of receptor cells. 1B2 is also expressed mainly in dorsomediai regions of the olfactory epithelium and there is considerable overlap with CC2-positive areas. However, 1B2 and CC2 recognize separate antigens, each occurring in a distinct distribution within the nasal cavity. Using fluorescence immunocytochemistry (Fig. 5B), a
243 s u b s e t of r e c e p t o r cells is visible. T h e surfaces of cell
stain with 1B2 a n t i b o d i e s . I m m u n o r e a c t i v i t y disappears
b o d i e s , d e n d r i t e s , d e n d r i t i c k n o b s , a n d p r o x i m a l axons
as the axons p r o j e c t t h r o u g h t h e basal l a m i n a .
Accessory olfactory bulb (AOB) CC1 i m m u n o r e a c t i v i t y w i t h i n the A O B is d i s t r i b u t e d in the v o m e r o n a s a l n e r v e layer, the g l o m e r u l a r layer a n d o u t e r p o r t i o n s of the e x t e r n a l p l e x i f o r m layer (Fig. 6 A , B). A n t i g e n expression is also restricted in the rostroc a u d a l axis to the rostral half of the A O B . This is particularly e v i d e n t in the sagittal p l a n e of section (Fig.
7A). C C 2 i m m u n o r e a c t i v i t y w i t h i n the A O B is d i s t r i b u t e d similarly b u t n o t identically to CC1 i m m u n o r e a c t i v i t y (Fig. 6C, D). A biphasic g r a d i e n t of reactivity can be seen in which s t r o n g e r reactivity is e v i d e n t in the v o m e r o n a s a l n e r v e a n d g l o m e r u l a r layer b u t w e a k reac-
|
< P / f
/I
A
."Qp
0
J
C Fig. 4. Olfactory epithelium. A: Nissl-stained section of the olfactory epithelium• Arrows delimit the extent of the olfactory cpithelium. Regions in boxes A and B are shown at higher power in Fig. 5. Turbinates I, lid and Ilv are labeled. B: CC2 immunoreactivity is restricted to dorsomedial regions of the olfactory cpithelium (arrows)• Bundles of the vomeronasal nerve are also stained (arrowheads). C: 1B2 staining also is limited to dorsomedial regions of the olfactory epithelium. Staining of 1B2-positive receptor cell bodies is not visible at this low power with peroxidaseconjugated techniques. Calibration bar = 500 #m for A-C.
Fig. 5. Olfactory epithelium. A: CC2 immunofluorescence staining of a dorsomedial region (see box A in Fig. 4) of olfactory epithelium. Immunoreactivity is detected on most receptor cell bodies, their dendrites and axon bundles (arrow)• Calibration bar = 50 pm for A and B. B:IB2 immunofluorescence staining of a dorsomedial region (see box B in Fig. 4) of P2 olfactory epithelium. Immunoreactivity is present on a subset of receptor cell bodies, their dendrites, and their dendritic knobs. 1B2 is expressed on the proximal region of axons, as they project along the basal lamina but is not detected in olfactory nerve bundles (arrow).
244
Fig. 6. Accessory olfactory bulb. (A, B) CC1 immunoreactivity in coronal sections through the rat accessory olfactor~ bulb. A: darkfield illumination of the section in B shows clearly the extent of the CCI immunostaining to include the VNL and G[.. B: section counterstained with Cresyl violet shows the layers of the AOB. VNN, vomeronasal nerve; VNL, vomeronasal nerve layer; GL, glomerular layer; EPL, external plexiform layer. C, D: CC2 immunoreactivity in an adjacent section. Nissl-counterstained section reveals the layers as in B. C: darkfield illumination of the section in D shows CC2 immunoreactivity extending beyond the VNL and GL into the EPL. Calibration bar = 100 !~m for A-D.
tivity is evident in the external plexiform layer. CC2 immunoreactivity appears to extend d e e p e r into the external plexiform layer than CC1 immunoreactivity. As seen in sagittal sections, CC2 immunoreactivity is present t h r o u g h o u t the A O B and is not restricted in its expression in a rostrocaudal gradient, as is seen with CC1 a n t i b o d y (Fig. 7B).
Olfactory bulb (OB) CC2 immunoreactivity is seen in the olfactory nerve layer, and is c o n c e n t r a t e d in d o r s o m e d i a l areas of the olfactory bulb (Fig. 8). O n l y glomeruli in d o r s o m e d i a l regions are filled with CC2 immunoreactivity. In lateral and ventral regions of the olfactory bulb little CC2 i m m u n o r e a c t i v i t y is o b s e r v e d in the olfactory nerve layer
and no staining could be d e t e c t e d in glomeruli in these regions. We have also d e t e c t e d weak 1B2 staining of selected glomeruli in the olfactory bulb (data not shown). This staining would a p p e a r to be associated with terminal portions of some olfactory r e c e p t o r axons, even though 1B2 immunoreactivity is not d e t e c t a b l e on m o r e proximal portions of the axon that travel from the olfactory epithelium to the olfactory bulb. DISCUSSION T h r e e monoclonal antibodies reacting with glycolipid and glycoprotein antigens are described. E a c h defines a selective group of cells found uniquely within the olfac-
245 tory system of rats (summarized in Fig. 9). This diagram depicts the location of the CC1, CC2 and 1B2 antigens and the structure of their terminal carbohydrate epitopes. lmmunocytochemical studies in progress revealed few differences from embryonic day 15 to adult (data not shown) suggesting that the regulation of expression of the
glycolipids described here is dissimilar to the majority of stage-specific embryonic carbohydrate antigens which have been described in the developing nervous system22. However, a detailed analysis of the development profile of these antigens has not yet been completed and subtle differences in their expression during development may
lit 4
!!A m
Fig. 7. Accessory olfactory bulb. A: CC1 immunoreactivity in sagittal sections through the rat accessory olfactory bulb. Staining is present in rostral (solid arrow) but not caudal (open arrow) portions of the AOB. B: CC2 immunoreactivity in an adjacent section shows staining throughout the rostrocaudal extent of the AOB. Straight arrows show staining in the dorsal and rostral portions of the main olfactory bulb. C: Nissl-counterstained section. The layers of the main olfactory bulb and the AOB can be seen. Calibration bar = 100 jtm for A - C .
Fig. 8. Olfactory bulb. CC2 immunoreactivity in the main olfactory bulb of the rat. A: at low power in a coronal section dorsal and medial regions of the olfactory bulb (arrowheads) are stained as well as the vomeronasal nerve (small arrows). B and C: darkfield (B) and brightfield (C) views of CC2 immunoreactivity in a Nissl-counterstained section. Only glomeruli in this dorsomedial region of the olfactory bulb (between the arrows) contain CC2 immunoreactivity. Calibration bars = 100 ~m for A and 50/~m for B and C.
246 A CC1
~- GALNAC...
B
CC2
C 1B2
:i~~
~-GAL...
OE
Fig. 9. Schematic summary of the localization of the glycolipid antigens (left) and their carbohydrate specificity (risht)/for CCI (A), CC2 (B) and 1B2 (C) antibodies. AOB, accessory olfactory bulb; FUC, fucose; GAL, galactose; GALNAC, n-acetylglucosamine; OB, olfactory bulb; OE, olfactory epithelium; VNO, vomeronasal organ.
help to reveal their functional significance. The biochemical specificity and the anatomical localization of the 3 monoclonal antibodies presented here are similar, yet different, from antibodies and lectins which have been previously described. CC1 antibody recognizes an N-acetyl-galactosamine terminal glycolipid, which could be similar to antigens recognized by the lectin soybean agglutinin (SBA). In Xenopus, the ventral portion of the olfactory nerves and bulb contained more SBA reactivity than the dorsal portions 13. In rats SBA has been shown to bind subsets of olfactory bulb glomeruli 2°. Although SBA and CC1 antibodies both recognize fl-N-acetyl-galactosamine, their reactivity in the olfactory system appears to differ considerably. Likewise, antibodies to CC2 and to blood group B antigens both recognize similar branched chain oligosaccharides containing a-galactosyl and a-fucosyl structures. However, CC2 and blood group B immunoreactivities differ considerably in the olfactory epithelium 3. Blood group B antigens are expressed on only a few neuroreceptor cells up to postnatal day 3, after which most main olfactory and vomeronasal receptor cells become positive. CC2, however, is continuously expressed on a subset of the olfactory epithelium from embryonic day 14
or earlier. The anatomical localization of CC2 is also similar to 2B8 antigens, a group of glycoproteins in the rat olfactory system located predominantly in dorsomedial regions of the main olfactory epithelium and bulb j'~ However, CC2 antibodies detect proteins with different molecular weights than the previously described, 2B8 reactive glycoproteins. This suggests that the two antibodies are distinct. 1B2 immunoreactivity is similar to an anti-carbohydrate antibody (4C9) that reacts with fucosyl-poly Nacetyllactosamine described by Mori 15. 4C9 also reacts strongly with structures at the luminal surface of the vomeronasal neuroepithelium and respiratory epithelium. However, 4C9 also stains axons and their projection sites in the olfactory bulb, whereas, with the current protocols, we are not able to detect 1B2 immunoreactivity in axons. At P14 we have been able to detect 1B2 immunoreactivity in subsets of glomeruli in the OB. Mori et al. ~4 previously demonstrated, using monoclohal antibodies reacting with restricted regions of the rabbit AOB, that there is no correlation of the location of receptor somata in the vomeronasal organ with their termination site in the AOB. This finding is confirmed here with an additional monoclonal antibody in the rat. CCl-positive cell bodies are distributed throughout the vomeronasal organ while CCl-positive axons terminate only in the rostral half of the AOB. This pattern of sensory neuron to target structure resembles the organization of cutaneous afferent fiber projections to the spinal cord in which subtypes intermix with fibers of different subtypes in the nerve bundles and dorsal roots ~'ls. However, different subtypes of fibers terminate in separate laminae of the dorsal horn of the spinal cord. Segregation of dorsal root ganglion neurons on the basis of their characteristic patterns of carbohydrate cell surface antigens has been studied in detail by Jessell and co-workers 9. They showed in sensory neurons that lacto-series and globo-series carbohydrate antigens were expressed on non-overlapping populations of dorsal root ganglia neurons and in their terminals in the dorsal horn of the spinal cord s . Lactose binding proteins were detected in the same cells which expressed the lactoseries carbohydrates, but were also released from cultured D R G neurons. Thus, the binding proteins may be laid down in the extracellular space and therefore play a key role in the guidance of axons between D R G neurons and their target cells 19. It is possible that the carbohydrate antigens described here, which define subsets of olfactory receptor cells, play a role in the guidance of axons from the olfactory receptor neurons to their targets in the MOB and AOB, just as the lacto-series carbohydrates are involved in guidance of D R G fibers to their targets in the spinal cord,
247 In addition, it is conceivable that these chemical cues are
between dorsal
continuously needed along the pathway leading from the
epithelium. This suggests that RB-8 may be involved in
olfactory receptor cells to the bulb, due to the fact that the t u r n o v e r of receptor ceils is ongoing throughout the life of the animal4'11'17.
directing axons from the ventral olfactory epithelium to ventral regions of the olfactory bulb. Similarly, CC2 may
The relationship of CC1 and CC2 antigens to previously described olfactory axon surface antigens is interesting. I m a m u r a et al. 12 described a monoclonal antibody (R4B12) which stained a subset of vomeronasal nerve fibers that t e r m i n a t e d in the rostrolateral portion of the A O B . A second antibody (R2D5) identified a complim e n t a r y subgroup of fibers which terminated in the caudomedial glomeruli of the A O B 14'15. W h e t h e r either of these antigens is structurally related to CC1 oligosaccharides which, like R4B12, are expressed preferentially in the rostral half of the A O B , is not known. A n additional antigen (RB-8) has been described 24, which
and ventral zones of the
olfactory
direct dorsomedially situated olfactory axons to the dorsomedial glomeruli of the olfactory bulb. Likewise, CC1 antigen may function to direct vomeronasal axons to the rostral portion of the A O B . Studies underway which utilize primary cultures of vomeronasal and olfactory epithelial n e u r o n s 7 will permit us to study the role played by CC1, CC2 and 1B2 antigens in the organization of axon projections in the olfactory system. Co-cultures of olfactory epithelial n e u r o n s and olfactory bulb explants are already proving valuable for analyzing the interaction between carbohydrate antigens on axon surfaces and structures in the extracellular matrix or on the surfaces of adjacent cells.
complements CC2 expression in the main olfactory epithelium and in the glomeruli of the olfactory bulb. RB-8 is expressed in ventrolaterai regions of the olfactory epithelium and of the olfactory bulb, whereas CC2 is expressed in the dorsomedial regions of the olfactory epithelium and olfactory bulb. RB-8 has been identified as a 125-kDa glycoprotein and its expression is regulated REFERENCES 1 Allen, W.K. and Akeson, R., Identification of a cell surface glycoprotein family of olfactory receptor neurons with a monoclonal antibody, J. Neurosci., 5 (1985) 284-296. 2 Allen, W.K. and Akeson, R., Identification of an olfactory receptor neuron subclass: cellular and molecular analysis during development, Dev. Biol., 109 (1985) 393-401. 3 Astic, L., Le Pendu, J., Mollicone, R., Saucier, D. and Oriol, R., Cellular expression of H and B antigens in the rat olfactory system during development, J. Comp. Neurol., 289 (1989) 386-394. 4 Barber, P.C. and Raisman, G., Cell division in the vomeronasal organ of the adult mouse, Brain Research, 141 (1978) 57-66. 5 Barber, P.C., Ulex europus agglutinin I binds exclusively to primary olfactory neurons in the rat nervous system, Neuroscience, 30 (1989) 1-9. 6 Brown, A.G., Organization in the Spinal Cord, Springer, New York, 1981. 7 Calof, A.L. and Chikaraishi, D.M., Analysis of neurogenesis in a mammalian neuroepithelium: proliferation and differentiation of an olfactory neuron precursor in vitro, Neuron, 3 (1989) 115-127. 8 Dodd, J. and Jessell, T.M., Lactoseries carbohydrates specify subsets of dorsal root ganglion neurons projecting to the superficial horn of rat spinal cord., J. Neurosci., 5 (1986) 3278-3294. 9 Dodd, J., Solter, D. and Jessell, T.M., Monoclonal antibodies against carbohydrate differentiation antigens identify subsets of primary sensory neurons, Nature, 314 (1984) 469-472. l0 Fujita, S.C., Mori, K., Imamura, K. and Obata, K., Subclasses of olfactory receptor cells and their segregated central projections demonstrated by a monoclonal antibody, Brain Research, 326 (1985) 192-196. 11 Graziadei, P.P.C. and Monti Graziadei, G.A., Neurogenesis and neuron regeneration in the olfactory system of mammals. I. Morphological aspects of differentiation and structural organi-
Acknowledgements. We thank Drs. T. Schoenfeld, S. Tobet and M. Yamamoto for their helpful discussion and critical reading of the manuscript. We gratefully appreciate the technical assistance of Gail Deutsch, Devin Gattey, Darlene Butler, and Ms. Denise Brescia for her expert assistance in the preparation of the manuscript. Portions of this work were supported by NIH Grants HD05515 and NS25580 to G.A.S. and NS24386 to J.E.C.
zation of the olfactory sensory neurons, J. Neurocytol., 8 (1978) 1-8. 12 Imamura, K., Mori, K., Fujita, S.C. and Obata, K., Immunochemical identification of subgroups of vomeronasal nerve fibers and their segregated terminations in the accessory olfactory bulb, Brain Research, 328 (1985) 362-366. 13 Key, B. and Giorgi, P.P., Selective binding of soybean agglutinin to the olfactory system of Xenopus, Neuroscience, 18 (1986) 507-515. 14 Mori, K., Fujita, S.C., Imamura, K. and Obata, K., Immunohistochemical study of subclasses of olfactory nerve fibers and their projections to the olfactory bulb in the rabbit, J. Cornp. Neurol., 242 (1985) 214-229. 15 Mori, K., Monoclonal antibodies (2C5 and 4C9) against lactoseries carbohydrates identify subsets of olfactory and vomeronasal receptor cells and their axons in the rabbit, Brain Research, 408 (1987) 215-221. 16 Mori, K., Imamura, K., Fujita, S.C. and Obata, K., Projections of two subclasses of vomeronasal nerve fibers to the accessory olfactory bulb in the rabbit, Neuroscience, 20 (1987) 259-278. 17 Moulton, D.G., Celebi, G. and Fink, R.P., Olfaction in mammals - - two aspects: proliferation of cells in the olfactory epithelium and sensitivity to odours. In G.E.W. Wolstenholme and J. Knight, (Eds.), Ciba Foundation Symposium on Taste and Smell in Vertebrates, Churchill, London, 1970, pp. 227-250. 18 Perl, E.R., Characterization of nociceptors and their activation of neurons in the superficial dorsal horn: first steps for the sensation of pain, Adv. Pain Res. Ther., 6 (1983) 23-51. 19 Regan, L.J., Dodd, J., Barondes, S.H. and Jessell, T.M., Selective expression of endogenous lactose-binding lectins and lactoseries glycoconjugates in subsets of rat sensory neurons, Proc. Natl. Acad. Sci. U.S.A., 83 (1986) 2248-2252. 20 Riggott, M.J. and Scott, J.W., Lectin labeling of rat olfactory bulb glomeruli, Soc. Neurosci. Abst., 15 (1989) 926. 21 Schwarting, G.A. and Crandall, J.E., Monoclonal antibodies to carbohydrate antigens identify subsets of olfactory and vomeronasal axons in the rat, Soc. Neurosci. Abst., 15 (1989) 590.
248 22 Schwarting, G.A. and Yamamoto, M., Expression of glycoconjugates during development of the vertebrate nervous system, BioEssays, 9 (1988) 19-23. 23 Schwarting, G.A., Tischler, A.S. and Donahue, S.R., Fucosylation of glycolipids in PC12 cells is dependent on the sequence of nerve growth factor treatment and adenylate cyclase activation, Develop. Neurosci., 12 (1989) 159-171. 24 Sehwob, J.E. and Gottlieb, D.I., The primary olfactory projection has two chemically distinct zones, J. Neurosci., 6 (1986) 3393-3404. 25 Schwob, J.E. and Gottlieb, D.I., Purification and characterization of an antigen that is spatially segregated in the primary olfactory projection, J. Neurosci., 8 (1987) 3470-3480. 26 Suchy, S.F., Yamamoto, M., Barbero, L. and Schwarting, G.A., A monoclonal antibody WCC4 recognizes a developmentally
regulated ganglioside containing ~z-galactose and a-tucose present in the rat nervous system, Brain Research, 440 (1988) 25-34. 27 Towbin, H., Staehelin, T. and Gordon, J., Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets: procedure and some applications. Proc. Natl. Acad. Sci. U.S.A., 76 (1979) 4350-4354. 28 Yamamoto, M., Boyer, A. and Schwarting, G.A., Fucosccontaining glycolipids are stage- and region-specific antigens in developing embryonic brain of rodents, Pr~c. Natl. Acad. Sci. U.S.A., 82 (1985) 3045-3049. 29 Young Jr., W.W., Portoukalien, J. and Hakomori, S.-t., Two monoclonal anticarbohydrate antibodies directed to glycosphingolipids with a Lacto-N-glycosyl Type II chain, J: Biol. Chem., 256 (1981) 10967-10977.
