- Email: [email protected]
Developmental changes of fucosylated glycoconjugates in rabbit dorsal root ganglia
74
Neuroscienee Research, 15 (1992) 74-80 ~'~ 1992 Elsevier Science Publishers Ireland, Ltd. All rights reserved 0168-0102/92/$05.00
NEURES 00568
Developmental changes of fucosylated glycoconjugates in rabbit dorsal root ganglia S. K u s u n o k i a, K. Inoue a, M. Iwamori
b, y .
Nagai c, T. M a n n e n a and I. Kanazawa a
Department of" Neurology and h Biochemistry, Faculty of Medicine, University of Tokyo, ' Tokyo Metropolitan Institute of Medical Science, Tokyo 113, Japan (Received 11 May 1992; accepted 6 July 1992)
Key words: Ganglioside; Lectin; Glycoprotein; Peripheral nerve; Immunohistochemistry
Summary Developmental changes of the fucosylated glycoconjugates in the dorsal root ganglia ( D R G ) of the rabbit were investigated histochemically using anti-fucosyl GM1 antibody and Ulex europaeus agglutinin 1 (UEA-1) lectin. Neither anti-fucosyl GM1 antibody nor UEA-1 lectin bound to the neural tubes or to the neural crest on embryonic day 14 (El4). Anti-fucosyl GM1 antibody binds diffusely to the D R G of E25. Large neurons unreactive with anti-fucosyl GM1 antibody appeared at 1 month and increased within 6 months after birth. Schwann cells immunoreactive with anti-fucosyl GM1 antibody came to be limited to the satellite cells surrounding the positive neurons. No staining with UEA-1 lectin was observed in the D R G of E25. Some small neurons became reactive with UEA-1 lectin within 1 month and remained to be so at 6 months after birth. Schwann cells including satellite cells were unreactive with this lectin. Since fucosyl GM1 was detected in the lipid fraction of DRGs from 1-month-old and 6-month-old rabbits, fucosyl GM1 itself should be the antigen molecule recognized by the anti-fucosyl GM1 antibody. Further study is necessary to elucidate the association between these developmental changes of the fucosylated glycoconjugates in D R G and their possible functional roles.
Introduction We have previously shown that two fucose-binding probes, anti-fucosyl GM1 antibody and the lectin Ulex europaeus agglutinin 1 (UEA-1), recognize the same subpopulation of neurons in the adult human dorsal root ganglia ( D R G ) (Kusunoki et al., 1989, 1991). In addition, anti-fucosyl GM1 antibody recognizes the
Correspondence to: Susumu Kusunoki, MD, Department of Neurology, Institute of Brain Research, Faculty of Medicine, University of Tokyo 7-3-1 Hongo, Bunkyo-ku, Tokyo 113, Japan. Tel.: 03-38155411, ext. 8763. Abbreviations: UEA-I, Ulex europaeus agglutinin 1; DRG, dorsal root ganglia.
satellite cells surrounding the positive neurons (Kusunoki et al., 1989), whereas UEA-1 lectin did not bind to any satellite cells but to unmyelinated axons in the peripheral nervous system (Nakagawa et al., 1986) including the biopsied sural nerve (Kusunoki et al., 1991). Since neurons in the human sympathetic ganglia, which have unmyelinated axons, were not recognized by either of these two probes (Kusunoki et al., 1991), UEA-1 lectin can be used to discriminate unmyelinated axons in the somatosensory peripheral nervous system from those in the sympathetic nervous system. Considering that cell surface glycoconjugates have been shown to be involved in cell-cell interaction and cell adhesion (Kunemund et al., 1988; Lowe et al., 1990), these fucosylated glycoconjugates might have
75 some functional role in the development and the organization of DRG. Developmental studies may be of help to elucidate the function of such antigens in DRG. It has been reported that UEA-1 lectin binds to some neurons in the rabbit D R G (Mori, 1986). In this paper, therefore, we studied rabbit D R G immunohistochemically by using anti-fucosyl GM1 antibody, and found that it immunostained some neurons and the satellite cells surrounding the stained neurons in adult rabbit DRG. Then the developmental changes of the cellular expression of these fucosylated glycoconjugates recognized by the anti-fucosyl GM1 antibody and UEA-1 lectin were investigated. We also found that the antigen molecule recognized by the anti-fucosyl GM1 antibody is fucosyl GM1 itself by using a thin-layer chromatography (TLC) immunostaining procedure.
Materials and methods
Fresh tissue was taken from animals killed by an intravenous overdose of pentobarbital. D R G and spinal cord were obtained from female rabbits, 1 month old (lmo) and 6 months old (6mo). Cross-sections of the fetal rabbits, embryonic day 14 (El4) and day 25 (E25), were also obtained. The sections were cut in a cryostat, mounted on bovine serum albumin (BSA)-coated slides, and fixed in 4% formalin in phosphate-buffered saline (PBS) for 5 min. After preincubation with 10% goat serum in PBS, the sections were incubated overnight at 4°C with the rabbit anti-fucosyl GM1 antiserum (Iwamori et al., 1983) (diluted 1:100 with 10% goat serum in PBS) or with biotinylated UEA-1 lectin (Vector Laboratories, Burlingame, CA, USA, diluted 1 : 100, 20 /xg lectin/ml). The anti-fucosyl GM1 antiserum used in this study showed its activity up to a dilution of 1 : 5120 in enzyme-linked immunosorbent assay (ELISA). The slides were then washed 3 times with PBS for 10 min and incubated with a peroxidase-conjugated goat anti-rabbit IgM antibody (Cappel, West Chester, PA, USA, diluted 1:100) or avidin-biotinperoxidase (Vector) at room temperature for 2 h. The slides were then washed 3 times with PBS for 10 min and stained with a diaminobenzidine solution (0.5 m g / m l 3,3'-diaminobenzidine tetrahydrochloride, 0.01% H 2 0 2 in PBS). For controls, only the dilution buffer (10% goat serum in PBS) was applied to the sections instead of the anti-fucosyl GM1 antibody or biotinylated UEA-1 lectin, followed by the procedure as described above. Each immunostained section was counterstained with hematoxylin to identify the ceils and tissues.
The anti-fucosyl GM1 anti serum was preabsorbed with fucosyl GM1 by incubating the serum in Linbro ELISA plate (Flow Laboratories, McLean, VA, USA) coated with fucosyl GM1 (1 /xg/well) at 4°C overnight. The sections were incubated with the absorbed antiserum followed by the same immunohistochemical procedure as described above. To remove lipids from the sections, frozen sections were immersed in c h l o r o f o r m / m e t h a n o l (1 : 1, v / v ) for 20 min and air-dried. Then the reactions of the sections with the anti-fucosyl GM1 antibody or UEA-1 lectin were investigated as described above. DRGs (100 mg wet weight) were obtained from rabbits ( l m o and 6mo), homogenized in 1 ml of chlorof o r m / m e t h a n o l (1: 1) and incubated at 40°C for 30 min. The extracted lipid solution was evaporated with a rotary evaporator, dissolved in c h l o r o f o r m / m e t h a n o l (9: 1) and applied to an iatrobeads column (3 ml), which was eluted with 9 ml each of c h l o r o f o r m / methanol (9: 1), c h l o r o f o r m / m e t h a n o l (8:2) and chlor o f o r m / m e t h a n o l (1:2), successively. The final fraction, eluted by c h l o r o f o r m / m e t h a n o l (1 : 2), was evaporated to dryness with a nitrogen evaporator, and dissolved in 300 Ixl c h l o r o f o r m / m e t h a n o l (1 : 1). Six txl of each lipid fraction of D R G (corresponding to 2 mg wet weight) was applied to the TLC plate (Macherey-Nagel, Germany), and developed with chlor o f o r m / m e t h a n o l / 0 . 2 % CaCI 2 (50:45: 10, v/v). The plate was then immunostained with anti-fucosyl GM1 antiserum or UEA-1 lectin as described previously (Kusunoki et al., 1987). In summary, the plate was dipped in a 0.4% polyisobutylmethacrylate/hexane solution for 1 min and dried for 30 min. The anti-fucosyl GM1 antiserum, diluted 1 : 100 with 1% BSA in PBS, or UEA-1 lectin, diluted 1:100 with PBS, was then applied to the plate and incubated at room temperature for 1.5 h. The plate was washed with PBS and incubated with peroxidase-conjugated anti-rabbit IgM antibody, diluted 1:200 with 1% BSA in PBS, or avidin-biotin=peroxidase at room temperature for 1.5 h. The plate was then washed with PBS, and the diaminobenzidine solution as described above was applied to reveal the immunoreactive bands. The immunostained lanes were compared with the pattern visualized with the orcinol reagent on the same plate, in order to identify the immunostained bands. Results
On E14, neither the anti-fucosyl GM1 antibody nor UEA-1 lectin bound to the neural tube or the neural crest.
76 The anti-fucosyl GM1 antibody diffusely immunostained D R G from the E25 rabbit (Fig. 1). However, the dorsal horn of the spinal cord of E25 was not recognized. On the other hand, there was no staining with UEA-1 lectin, either in D R G or in the spinal cord, at this stage. In lmo, the anti-fucosyl GM1 antibody immunostained the D R G neurons except for a few large ones (Fig. 2a). The staining was most intense at the cell surface of the neurons. The satellite cells and the interstitial tissue surrounding the stained neurons, some axons, and the dorsal horn of the spinal cord were also immunostained. UEA-1 lectin stained some small neurons, some axons (Fig. 2b) and the dorsal horn of the spinal cord. The number of the neurons stained by UEA-1 lectin was less than that by the anti-fucosyl GM1 antibody. The satellite ceils and the interstitial tissue were not recognized by UEA-1 lectin. In 6mo, the anti-fucosyl GM1 antibody immunostained some small to intermediate neurons and the satellite cells surrounding the positive neurons (Fig. 3a and b). The number of neurons unstained with the anti-fucosyl GM1 antibody increased as compared with lmo. The neurons immunostained with the anti-fucosyl GM1 antiserum appeared to be aggregated. Some axons and the dorsal horn of the spinal cord were also immunostained. UEA-1 lectin recognized some small neurons, some axons (Fig. 3c), and the dorsal horn of the spinal cord, but no satellite cells. No vascular endothelium of the rabbit was stained by UEA-1 lectin.
In the control sections immunostained only with the peroxidase-conjugated second antibody or with avidinbiotin-peroxidase, no immunostaining was observed in any stage of the development. In the sections immunostained by the antifucosyl GM1 antiserum preincubated in the fucosyl GMl-coated wells, the immunostaining was greatly reduced (Fig. 2c). On the other hand, no such absorption of immunostaining was observed with the anti-fucosyl GM1 antiserum preincubated in the uncoated wells. In the sections treated with chloroform/methanol (1:1), the staining by the anti-fucosyl GM1 antibody was completely eliminated, whereas the staining with UEA-1 lectin diminished but remained. TLC-immunostaining with the anti-fucosyl GM1 antibody revealed that an immunoreactive band with the same mobility as purified fucosyl GM1 existed in the lane of each lipid fraction from lmo- and 6mo-rabbit DRGs (Fig. 4). The stronger immunoreactivity was observed in the lane of lmo. UEA-1 lectin did not bind to this band or any other band in the lane of the lipid fraction of the lmo-rabbit DRG. UEA-1 lectin did not bind to purified fucosyl GM1 either.
Discussion
Specific carbohydrate epitopes in a subpopulation of D R G neurons have been reported in several animal species and in humans (Dodd et al., 1984; Omlin et al.,
Fig. 1. E25. D R G was diffusely immunostained by the anti-fucosyl GMI antibody (arrow). Nerve bundles from the DRG were also immunostained (arrowhead). SC: spinal cord; B: vertebral body; L: vertebral lamina. Bar - 0.1 mm.
77 1984; Dodd and Jessell, 1985; Mori, 1986; Nakagawa et al., 1986; Streit et al., 1986; Plenderleith et al., 1988; Chou et al., 1989; Kusunoki et al., 1989, 1991; Scott et al., 1990). The rat is the best studied species among them. However, the developmental changes of these glycoconjugates have not yet been well characterized. We found that the anti-fucosyl GM1 antibody, as well as UEA-1 lectin (Mori, 1986), recognizes some D R G neurons in adult rabbit, and then investigated their expression during development from E l 4 to 6mo. The developmental changes in the present histochemical study are summarized in the Table 1. The anti-fucosyl GM1 antibody does not bind to the neural tube or to the neural crest at El4, but binds diffusely to the E25-rabbit DRG, which is the stage just before birth. After birth, large neurons unreactive
iii¸?i!¸¸i~
with anti-fucosyl GM1 antibody began to develop and increase. In the meantime, the interstitial cells immunoreactive with anti-fucosyl GM1 antibody became limited to the satellite cells surrounding the positive neurons. There was no staining with UEA-1 lectin in the neural tube, in the neural crest or in the D R G before birth. Some neurons became reactive with UEA-1 lectin after birth, but the interstitial ceils, Schwann cells including satellite ceils, were unreactive with this lectin during the span of the present study. The fact that the staining by the anti-fucosyl GM1 antibody was eliminated by treatment with chlorof o r m / m e t h a n o l (1 : 1) suggested that the antigen molecule is a glycolipid. The anti-fucosyl GM1 antibody detected a band in the lane of each lipid fraction from
i ~ii
Fig. 2. 1 - m o n t h - o l d . a: t h e a n t i - f u c o s y l G M 1 a n t i b o d y i m m u n o s t a i n e d t h e D R G n e u r o n s e x c e p t f o r a f e w l a r g e o n e s ( a r r o w ) . B a r = 0.1 m m . b: U E A - 1 lectin b o u n d to s o m e s m a l l n e u r o n s in D R G . S o m e n e r v e fibers ( a r r o w h e a d s ) w e r e a l s o s t a i n e d . B a r = 0.1 m m . c: t h e i m m u n o s t a i n i n g w i t h t h e a n t i - f u c o s y l G M 1 a n t i s e r u m w a s a b s o r b e d by p r e i n c u b a t i o n w i t h fucosyl G M 1 . B a r = 0.1 m m .
78 the l m o a n d 6too rabbit D R G s . T h e s e b a n d s have the same mobility as purified fucosyl GM1. Thus, the antigen molecule recognized by the anti-fucosyl G M 1 antibody should be fucosyl GM1 itself. T h e stronger imm u n o r e a c t i v i t y of the b a n d of fucosyl G M I in the lane of l m o may reflect the higher p e r c e n t a g e of fucosyl G M l - p o s i t i v e n e u r o n s in l m o D R G . Because the staining with U E A - 1 lectin d i m i n i s h e d but r e m a i n e d in the D R G section t r e a t e d with chlorof o r m / m e t h a n o l (1 : 1), a n d t h e r e was n o b a n d reactive with U E A - 1 lectin in the lipid fraction from D R G , the
a n t i g e n molecule recognized by U E A - 1 lectin may not be in the glycolipid fraction, but is probably some glycoprotein. F u r t h e r study is necessary to identify the functional roles of fucosylated glycoconjugates in primary sensory n e u r o n s . Some n e u r o n s begin to c h a n g e from fucosyl G M l - p o s i t i v e to negative a r o u n d l m o . Some n e u r o n s c h a n g e from U E A - l - n e g a t i v e to positive in the p e r i n a tal period. T h e s e changes might be associated with the differentiation of i m m a t u r e n e u r o n s to m a t u r e ones conveying a specific modality of sensation, either
Fig. 3. B-month-old.a: the anti-fucosyl GM1 antibody immunostained some small to intermediate neurons and some nerve fibers. Representative immunostained neurons and nerve fibers are indicated by arrows and arrowheads, respectively. Bar = 0.1 ram. b: the satellite cells (arrowheads) surrounding the fucosyl GMl-positive neuron (N) were also stained by the anti-fucosyl GM1 antibody. Bar = 0.05 mm. c: UEA-1 lectin bound to some small neurons and nerve fibers, but no satellite cells. Representative immunostained neurons and nerve fibers are indicated by arrows and arrowheads, respectively. Bar = 0.1 mm.
79
GM1
(-,,
fGM1
1
2
8B g GTlh 4
3
5
6
7
Fig. 4. Detection of fucosyl GM1 in the lipid fraction from lmo and 6mo rabbit DRGs by thin-layer chromatography (TLC) immunostaining procedure. Lanes 1 and 6: purified fucosyl GM1 (0.6/xg), lanes 2 and 4: lipid fraction from lmo-rabbit DRG (corresponding to 2 mg wet weight); lanes 3 and 5: lipid fraction from 6mo rabbit DRG (2 mg wet weight); lane 7: standard glycolipids (0.7/xg each). Lanes 1-3 were immunostained by the anti-fucosyl GM1 antibody, and lanes 4-7 were stained with orcinol. The arrow shows an immunostained band of fucosyl GMI, the arrowhead shows weak cross-reaction with GM1.
myelinated or not. Fucosyl GM1 is negative in the spinal cord of E25 but becomes positive in the dorsal horn of the spinal cord of lmo. Thus, it is possible that they are involved in the organization of the projection of dorsal root afferent fibers to the dorsal horn of the spinal cord. It is possible that such glycoconjugates may interact with complementary lectins on neighboring neurons or interstitial tissues. The immunostaining with the anti-fucosyl GM1 antibody appeared to be most intense at the cell surface. Fucosyl GM1 antigen was also expressed in the satellite ceils and the interstitial tissue surrounding the fucosyl GMl-positive neurons. Fucosyl GMl-positive neurons in DRG appeared to be aggregated. From
TABLE 1 DEVELOPMENTAL CHANGES OF THE STAINING PATTERN BY ANTI-FUCOSYL GM1 ANTIBODY AND UEA-1 LECTIN IN RABBIT DRG
antifucosyl GM1 antibody UEA-I lectin
E25
1month
6month
neuron Schwann cells
+ + + +
+ + +
+ +
q- -[- -1- q-
-t- -b -1-
_1_ a
neuron Schwann cells
-
+ + -
+ + -
+ + + +, all cells recognized; + + +, almost all cells recognized; + + , about half the cells recognized; +, only a few cells recognized; - , none of the cells recognized. a Only the satellite cells surrounding the fucosyl GMl-positive neurons were recognized.
these facts, it is possible that fucosyl GM1 is directly involved in cell adhesion in DRG. Since the staining pattern of the neurons and the satellite ceils in the rabbit DRG resembles that in the human DRG (Kusunoki et al., 1989, 1991), further analysis on the physiological roles of the fucosylated glycoconjugates in the rabbit DRG should provide us with useful information for understanding those in the human DRG.
Acknowledgments This work was supported in part by a Grant-in-Aid for scientific research from the Ministry of Education, Science and Culture of Japan (02259203, 02770445), and a grant (61-A) from the National Center of Neurology and Psychiatry (NCNP) of the Ministry of Health and Welfare of Japan. The authors wish to thank Ms. M. Sagishima for technical assistance.
References Chou, D.K.H., Dodd, J., Jessell, T.M., Costello, C.E. and Jungalwala, F.B. (1989) Identification of a-galactose (a-fucose)-asialoGM1 glycolipid expressed by subsets of rat dorsal root ganglion neurons, J. Biol. Chem., 264: 3409-3415. Dodd, J. and Jessell, T.M. (1985) Lactoseries carbohydrates specify subsets of dorsal root ganglion neurons projecting to the superficial dorsal horn of rat spinal cord, J. Neurosci., 5: 3278-3294. Dodd, J., Solter, D. and Jessel, T.M. (1984) Monoclonal antibodies against carbohydrate differentiation antigens identify subsets of primary sensory neurons, Nature, 311: 469-472. Iwamori, M., Shimomura, J., Tsuyuhara, S., Mogi, M., Ishizaki, S. and Nagai, Y. (1983) Differential reactivities of fucosyl GM1 and GM1 gangliosides on rat erythrocyte membrane revealed by analysis with anti-fucosyl GM1 and GM1 antisera, J. Biochem., 94: 1-10. Kunemund, V., Jungalwala, F.B., Fischer, G., Chou, D.K.H., Keilhauer, G. and Schachner, M. (1988) The L2/HNK-1 carbohydrate of neural cell adhesion molecules is involved in cell interactions, J. Cell Biol., 106: 213-223. Kusunoki, S., Craft, J.E., Roach, B., Hardin, J.A. and Yu, R.K. (1987) A human IgM M-protein in a patient with unknown bleeding disorder binds to /3-galactosyl and /3-glucosyl residues, Arch. Biochem. Biophys., 255: 226-232. Kusunoki, S., Inoue, K., Iwamori, M., Nagai, Y. and Mannen, T. (1989) Discrimination of human dorsal root ganglion cells by anti-fucosyl GM1 antibody, Brain Res., 494: 391-395. Kusunoki, S., Inoue, K., lwamori, M., Nagai, Y. and Mannen, T. (1991) Fucosylated glycoconjugates in human dorsal root ganglion cells with unmyelinated axons, Neurosci. Lett., 126: 159-162. Lowe, J.B., Stoolman, L.M., Nair, R.P., Larsen, R.D., Berhend, T.L. and Marks, R.M. (1990) ELAM-l-dependent cell adhesion to vascular endothelium determined by a transfected human fucosyltransferase cDNA, Cell, 63: 475-484.
80 Mori, K. (1986) Lectin Ulex europaeus agglutinin I specifically labels a subset of primary afferent fibers which project selectively to the superficial dorsal horn of the spinal cord, Brain Res., 365: 404408. Nakagawa, F., Schulte, B.A. and Spicer, S.S. (1986) Lectin cytochemical evaluation of somatosensory neurons and their peripheral and central processes in rat and man, Cell Tissue Res., 245: 579-589. Omlin, F.X., Matthieu, J.M., Philippe, E., Roch, J.M. and Droz, B. (1984) Expression of myelin-associated glycoprotein by small neurons of the dorsal root ganglion in chickens, Science, 227:1359 1360.
Plenderleith, M.B., Cameron, A.A., Key, B. and Snow, P.J. (1988) Soybean agglutinin binds to a subpopulation of primary sensory neurons in the cat, Neurosci. Lett., 86: 257-262. Scott, S.A., Patel, N. and Levine, J.M. (1990) Lectin binding identifies a subpopulation of neurons in chick dorsal root ganglia, J. Neurosci., 10: 336-345. Streit, W.J., Schulte, B.A., Balentine, J.D. and Spicer, S.S. (1986) Evidence for glycoconjugate in nociceptive primary sensory neurons and its origin from the Golgi complex, Brain Research, 377: 1-17.
Neuroscienee Research, 15 (1992) 74-80 ~'~ 1992 Elsevier Science Publishers Ireland, Ltd. All rights reserved 0168-0102/92/$05.00
NEURES 00568
Developmental changes of fucosylated glycoconjugates in rabbit dorsal root ganglia S. K u s u n o k i a, K. Inoue a, M. Iwamori
b, y .
Nagai c, T. M a n n e n a and I. Kanazawa a
Department of" Neurology and h Biochemistry, Faculty of Medicine, University of Tokyo, ' Tokyo Metropolitan Institute of Medical Science, Tokyo 113, Japan (Received 11 May 1992; accepted 6 July 1992)
Key words: Ganglioside; Lectin; Glycoprotein; Peripheral nerve; Immunohistochemistry
Summary Developmental changes of the fucosylated glycoconjugates in the dorsal root ganglia ( D R G ) of the rabbit were investigated histochemically using anti-fucosyl GM1 antibody and Ulex europaeus agglutinin 1 (UEA-1) lectin. Neither anti-fucosyl GM1 antibody nor UEA-1 lectin bound to the neural tubes or to the neural crest on embryonic day 14 (El4). Anti-fucosyl GM1 antibody binds diffusely to the D R G of E25. Large neurons unreactive with anti-fucosyl GM1 antibody appeared at 1 month and increased within 6 months after birth. Schwann cells immunoreactive with anti-fucosyl GM1 antibody came to be limited to the satellite cells surrounding the positive neurons. No staining with UEA-1 lectin was observed in the D R G of E25. Some small neurons became reactive with UEA-1 lectin within 1 month and remained to be so at 6 months after birth. Schwann cells including satellite cells were unreactive with this lectin. Since fucosyl GM1 was detected in the lipid fraction of DRGs from 1-month-old and 6-month-old rabbits, fucosyl GM1 itself should be the antigen molecule recognized by the anti-fucosyl GM1 antibody. Further study is necessary to elucidate the association between these developmental changes of the fucosylated glycoconjugates in D R G and their possible functional roles.
Introduction We have previously shown that two fucose-binding probes, anti-fucosyl GM1 antibody and the lectin Ulex europaeus agglutinin 1 (UEA-1), recognize the same subpopulation of neurons in the adult human dorsal root ganglia ( D R G ) (Kusunoki et al., 1989, 1991). In addition, anti-fucosyl GM1 antibody recognizes the
Correspondence to: Susumu Kusunoki, MD, Department of Neurology, Institute of Brain Research, Faculty of Medicine, University of Tokyo 7-3-1 Hongo, Bunkyo-ku, Tokyo 113, Japan. Tel.: 03-38155411, ext. 8763. Abbreviations: UEA-I, Ulex europaeus agglutinin 1; DRG, dorsal root ganglia.
satellite cells surrounding the positive neurons (Kusunoki et al., 1989), whereas UEA-1 lectin did not bind to any satellite cells but to unmyelinated axons in the peripheral nervous system (Nakagawa et al., 1986) including the biopsied sural nerve (Kusunoki et al., 1991). Since neurons in the human sympathetic ganglia, which have unmyelinated axons, were not recognized by either of these two probes (Kusunoki et al., 1991), UEA-1 lectin can be used to discriminate unmyelinated axons in the somatosensory peripheral nervous system from those in the sympathetic nervous system. Considering that cell surface glycoconjugates have been shown to be involved in cell-cell interaction and cell adhesion (Kunemund et al., 1988; Lowe et al., 1990), these fucosylated glycoconjugates might have
75 some functional role in the development and the organization of DRG. Developmental studies may be of help to elucidate the function of such antigens in DRG. It has been reported that UEA-1 lectin binds to some neurons in the rabbit D R G (Mori, 1986). In this paper, therefore, we studied rabbit D R G immunohistochemically by using anti-fucosyl GM1 antibody, and found that it immunostained some neurons and the satellite cells surrounding the stained neurons in adult rabbit DRG. Then the developmental changes of the cellular expression of these fucosylated glycoconjugates recognized by the anti-fucosyl GM1 antibody and UEA-1 lectin were investigated. We also found that the antigen molecule recognized by the anti-fucosyl GM1 antibody is fucosyl GM1 itself by using a thin-layer chromatography (TLC) immunostaining procedure.
Materials and methods
Fresh tissue was taken from animals killed by an intravenous overdose of pentobarbital. D R G and spinal cord were obtained from female rabbits, 1 month old (lmo) and 6 months old (6mo). Cross-sections of the fetal rabbits, embryonic day 14 (El4) and day 25 (E25), were also obtained. The sections were cut in a cryostat, mounted on bovine serum albumin (BSA)-coated slides, and fixed in 4% formalin in phosphate-buffered saline (PBS) for 5 min. After preincubation with 10% goat serum in PBS, the sections were incubated overnight at 4°C with the rabbit anti-fucosyl GM1 antiserum (Iwamori et al., 1983) (diluted 1:100 with 10% goat serum in PBS) or with biotinylated UEA-1 lectin (Vector Laboratories, Burlingame, CA, USA, diluted 1 : 100, 20 /xg lectin/ml). The anti-fucosyl GM1 antiserum used in this study showed its activity up to a dilution of 1 : 5120 in enzyme-linked immunosorbent assay (ELISA). The slides were then washed 3 times with PBS for 10 min and incubated with a peroxidase-conjugated goat anti-rabbit IgM antibody (Cappel, West Chester, PA, USA, diluted 1:100) or avidin-biotinperoxidase (Vector) at room temperature for 2 h. The slides were then washed 3 times with PBS for 10 min and stained with a diaminobenzidine solution (0.5 m g / m l 3,3'-diaminobenzidine tetrahydrochloride, 0.01% H 2 0 2 in PBS). For controls, only the dilution buffer (10% goat serum in PBS) was applied to the sections instead of the anti-fucosyl GM1 antibody or biotinylated UEA-1 lectin, followed by the procedure as described above. Each immunostained section was counterstained with hematoxylin to identify the ceils and tissues.
The anti-fucosyl GM1 anti serum was preabsorbed with fucosyl GM1 by incubating the serum in Linbro ELISA plate (Flow Laboratories, McLean, VA, USA) coated with fucosyl GM1 (1 /xg/well) at 4°C overnight. The sections were incubated with the absorbed antiserum followed by the same immunohistochemical procedure as described above. To remove lipids from the sections, frozen sections were immersed in c h l o r o f o r m / m e t h a n o l (1 : 1, v / v ) for 20 min and air-dried. Then the reactions of the sections with the anti-fucosyl GM1 antibody or UEA-1 lectin were investigated as described above. DRGs (100 mg wet weight) were obtained from rabbits ( l m o and 6mo), homogenized in 1 ml of chlorof o r m / m e t h a n o l (1: 1) and incubated at 40°C for 30 min. The extracted lipid solution was evaporated with a rotary evaporator, dissolved in c h l o r o f o r m / m e t h a n o l (9: 1) and applied to an iatrobeads column (3 ml), which was eluted with 9 ml each of c h l o r o f o r m / methanol (9: 1), c h l o r o f o r m / m e t h a n o l (8:2) and chlor o f o r m / m e t h a n o l (1:2), successively. The final fraction, eluted by c h l o r o f o r m / m e t h a n o l (1 : 2), was evaporated to dryness with a nitrogen evaporator, and dissolved in 300 Ixl c h l o r o f o r m / m e t h a n o l (1 : 1). Six txl of each lipid fraction of D R G (corresponding to 2 mg wet weight) was applied to the TLC plate (Macherey-Nagel, Germany), and developed with chlor o f o r m / m e t h a n o l / 0 . 2 % CaCI 2 (50:45: 10, v/v). The plate was then immunostained with anti-fucosyl GM1 antiserum or UEA-1 lectin as described previously (Kusunoki et al., 1987). In summary, the plate was dipped in a 0.4% polyisobutylmethacrylate/hexane solution for 1 min and dried for 30 min. The anti-fucosyl GM1 antiserum, diluted 1 : 100 with 1% BSA in PBS, or UEA-1 lectin, diluted 1:100 with PBS, was then applied to the plate and incubated at room temperature for 1.5 h. The plate was washed with PBS and incubated with peroxidase-conjugated anti-rabbit IgM antibody, diluted 1:200 with 1% BSA in PBS, or avidin-biotin=peroxidase at room temperature for 1.5 h. The plate was then washed with PBS, and the diaminobenzidine solution as described above was applied to reveal the immunoreactive bands. The immunostained lanes were compared with the pattern visualized with the orcinol reagent on the same plate, in order to identify the immunostained bands. Results
On E14, neither the anti-fucosyl GM1 antibody nor UEA-1 lectin bound to the neural tube or the neural crest.
76 The anti-fucosyl GM1 antibody diffusely immunostained D R G from the E25 rabbit (Fig. 1). However, the dorsal horn of the spinal cord of E25 was not recognized. On the other hand, there was no staining with UEA-1 lectin, either in D R G or in the spinal cord, at this stage. In lmo, the anti-fucosyl GM1 antibody immunostained the D R G neurons except for a few large ones (Fig. 2a). The staining was most intense at the cell surface of the neurons. The satellite cells and the interstitial tissue surrounding the stained neurons, some axons, and the dorsal horn of the spinal cord were also immunostained. UEA-1 lectin stained some small neurons, some axons (Fig. 2b) and the dorsal horn of the spinal cord. The number of the neurons stained by UEA-1 lectin was less than that by the anti-fucosyl GM1 antibody. The satellite ceils and the interstitial tissue were not recognized by UEA-1 lectin. In 6mo, the anti-fucosyl GM1 antibody immunostained some small to intermediate neurons and the satellite cells surrounding the positive neurons (Fig. 3a and b). The number of neurons unstained with the anti-fucosyl GM1 antibody increased as compared with lmo. The neurons immunostained with the anti-fucosyl GM1 antiserum appeared to be aggregated. Some axons and the dorsal horn of the spinal cord were also immunostained. UEA-1 lectin recognized some small neurons, some axons (Fig. 3c), and the dorsal horn of the spinal cord, but no satellite cells. No vascular endothelium of the rabbit was stained by UEA-1 lectin.
In the control sections immunostained only with the peroxidase-conjugated second antibody or with avidinbiotin-peroxidase, no immunostaining was observed in any stage of the development. In the sections immunostained by the antifucosyl GM1 antiserum preincubated in the fucosyl GMl-coated wells, the immunostaining was greatly reduced (Fig. 2c). On the other hand, no such absorption of immunostaining was observed with the anti-fucosyl GM1 antiserum preincubated in the uncoated wells. In the sections treated with chloroform/methanol (1:1), the staining by the anti-fucosyl GM1 antibody was completely eliminated, whereas the staining with UEA-1 lectin diminished but remained. TLC-immunostaining with the anti-fucosyl GM1 antibody revealed that an immunoreactive band with the same mobility as purified fucosyl GM1 existed in the lane of each lipid fraction from lmo- and 6mo-rabbit DRGs (Fig. 4). The stronger immunoreactivity was observed in the lane of lmo. UEA-1 lectin did not bind to this band or any other band in the lane of the lipid fraction of the lmo-rabbit DRG. UEA-1 lectin did not bind to purified fucosyl GM1 either.
Discussion
Specific carbohydrate epitopes in a subpopulation of D R G neurons have been reported in several animal species and in humans (Dodd et al., 1984; Omlin et al.,
Fig. 1. E25. D R G was diffusely immunostained by the anti-fucosyl GMI antibody (arrow). Nerve bundles from the DRG were also immunostained (arrowhead). SC: spinal cord; B: vertebral body; L: vertebral lamina. Bar - 0.1 mm.
77 1984; Dodd and Jessell, 1985; Mori, 1986; Nakagawa et al., 1986; Streit et al., 1986; Plenderleith et al., 1988; Chou et al., 1989; Kusunoki et al., 1989, 1991; Scott et al., 1990). The rat is the best studied species among them. However, the developmental changes of these glycoconjugates have not yet been well characterized. We found that the anti-fucosyl GM1 antibody, as well as UEA-1 lectin (Mori, 1986), recognizes some D R G neurons in adult rabbit, and then investigated their expression during development from E l 4 to 6mo. The developmental changes in the present histochemical study are summarized in the Table 1. The anti-fucosyl GM1 antibody does not bind to the neural tube or to the neural crest at El4, but binds diffusely to the E25-rabbit DRG, which is the stage just before birth. After birth, large neurons unreactive
iii¸?i!¸¸i~
with anti-fucosyl GM1 antibody began to develop and increase. In the meantime, the interstitial cells immunoreactive with anti-fucosyl GM1 antibody became limited to the satellite cells surrounding the positive neurons. There was no staining with UEA-1 lectin in the neural tube, in the neural crest or in the D R G before birth. Some neurons became reactive with UEA-1 lectin after birth, but the interstitial ceils, Schwann cells including satellite ceils, were unreactive with this lectin during the span of the present study. The fact that the staining by the anti-fucosyl GM1 antibody was eliminated by treatment with chlorof o r m / m e t h a n o l (1 : 1) suggested that the antigen molecule is a glycolipid. The anti-fucosyl GM1 antibody detected a band in the lane of each lipid fraction from
i ~ii
Fig. 2. 1 - m o n t h - o l d . a: t h e a n t i - f u c o s y l G M 1 a n t i b o d y i m m u n o s t a i n e d t h e D R G n e u r o n s e x c e p t f o r a f e w l a r g e o n e s ( a r r o w ) . B a r = 0.1 m m . b: U E A - 1 lectin b o u n d to s o m e s m a l l n e u r o n s in D R G . S o m e n e r v e fibers ( a r r o w h e a d s ) w e r e a l s o s t a i n e d . B a r = 0.1 m m . c: t h e i m m u n o s t a i n i n g w i t h t h e a n t i - f u c o s y l G M 1 a n t i s e r u m w a s a b s o r b e d by p r e i n c u b a t i o n w i t h fucosyl G M 1 . B a r = 0.1 m m .
78 the l m o a n d 6too rabbit D R G s . T h e s e b a n d s have the same mobility as purified fucosyl GM1. Thus, the antigen molecule recognized by the anti-fucosyl G M 1 antibody should be fucosyl GM1 itself. T h e stronger imm u n o r e a c t i v i t y of the b a n d of fucosyl G M I in the lane of l m o may reflect the higher p e r c e n t a g e of fucosyl G M l - p o s i t i v e n e u r o n s in l m o D R G . Because the staining with U E A - 1 lectin d i m i n i s h e d but r e m a i n e d in the D R G section t r e a t e d with chlorof o r m / m e t h a n o l (1 : 1), a n d t h e r e was n o b a n d reactive with U E A - 1 lectin in the lipid fraction from D R G , the
a n t i g e n molecule recognized by U E A - 1 lectin may not be in the glycolipid fraction, but is probably some glycoprotein. F u r t h e r study is necessary to identify the functional roles of fucosylated glycoconjugates in primary sensory n e u r o n s . Some n e u r o n s begin to c h a n g e from fucosyl G M l - p o s i t i v e to negative a r o u n d l m o . Some n e u r o n s c h a n g e from U E A - l - n e g a t i v e to positive in the p e r i n a tal period. T h e s e changes might be associated with the differentiation of i m m a t u r e n e u r o n s to m a t u r e ones conveying a specific modality of sensation, either
Fig. 3. B-month-old.a: the anti-fucosyl GM1 antibody immunostained some small to intermediate neurons and some nerve fibers. Representative immunostained neurons and nerve fibers are indicated by arrows and arrowheads, respectively. Bar = 0.1 ram. b: the satellite cells (arrowheads) surrounding the fucosyl GMl-positive neuron (N) were also stained by the anti-fucosyl GM1 antibody. Bar = 0.05 mm. c: UEA-1 lectin bound to some small neurons and nerve fibers, but no satellite cells. Representative immunostained neurons and nerve fibers are indicated by arrows and arrowheads, respectively. Bar = 0.1 mm.
79
GM1
(-,,
fGM1
1
2
8B g GTlh 4
3
5
6
7
Fig. 4. Detection of fucosyl GM1 in the lipid fraction from lmo and 6mo rabbit DRGs by thin-layer chromatography (TLC) immunostaining procedure. Lanes 1 and 6: purified fucosyl GM1 (0.6/xg), lanes 2 and 4: lipid fraction from lmo-rabbit DRG (corresponding to 2 mg wet weight); lanes 3 and 5: lipid fraction from 6mo rabbit DRG (2 mg wet weight); lane 7: standard glycolipids (0.7/xg each). Lanes 1-3 were immunostained by the anti-fucosyl GM1 antibody, and lanes 4-7 were stained with orcinol. The arrow shows an immunostained band of fucosyl GMI, the arrowhead shows weak cross-reaction with GM1.
myelinated or not. Fucosyl GM1 is negative in the spinal cord of E25 but becomes positive in the dorsal horn of the spinal cord of lmo. Thus, it is possible that they are involved in the organization of the projection of dorsal root afferent fibers to the dorsal horn of the spinal cord. It is possible that such glycoconjugates may interact with complementary lectins on neighboring neurons or interstitial tissues. The immunostaining with the anti-fucosyl GM1 antibody appeared to be most intense at the cell surface. Fucosyl GM1 antigen was also expressed in the satellite ceils and the interstitial tissue surrounding the fucosyl GMl-positive neurons. Fucosyl GMl-positive neurons in DRG appeared to be aggregated. From
TABLE 1 DEVELOPMENTAL CHANGES OF THE STAINING PATTERN BY ANTI-FUCOSYL GM1 ANTIBODY AND UEA-1 LECTIN IN RABBIT DRG
antifucosyl GM1 antibody UEA-I lectin
E25
1month
6month
neuron Schwann cells
+ + + +
+ + +
+ +
q- -[- -1- q-
-t- -b -1-
_1_ a
neuron Schwann cells
-
+ + -
+ + -
+ + + +, all cells recognized; + + +, almost all cells recognized; + + , about half the cells recognized; +, only a few cells recognized; - , none of the cells recognized. a Only the satellite cells surrounding the fucosyl GMl-positive neurons were recognized.
these facts, it is possible that fucosyl GM1 is directly involved in cell adhesion in DRG. Since the staining pattern of the neurons and the satellite ceils in the rabbit DRG resembles that in the human DRG (Kusunoki et al., 1989, 1991), further analysis on the physiological roles of the fucosylated glycoconjugates in the rabbit DRG should provide us with useful information for understanding those in the human DRG.
Acknowledgments This work was supported in part by a Grant-in-Aid for scientific research from the Ministry of Education, Science and Culture of Japan (02259203, 02770445), and a grant (61-A) from the National Center of Neurology and Psychiatry (NCNP) of the Ministry of Health and Welfare of Japan. The authors wish to thank Ms. M. Sagishima for technical assistance.
References Chou, D.K.H., Dodd, J., Jessell, T.M., Costello, C.E. and Jungalwala, F.B. (1989) Identification of a-galactose (a-fucose)-asialoGM1 glycolipid expressed by subsets of rat dorsal root ganglion neurons, J. Biol. Chem., 264: 3409-3415. Dodd, J. and Jessell, T.M. (1985) Lactoseries carbohydrates specify subsets of dorsal root ganglion neurons projecting to the superficial dorsal horn of rat spinal cord, J. Neurosci., 5: 3278-3294. Dodd, J., Solter, D. and Jessel, T.M. (1984) Monoclonal antibodies against carbohydrate differentiation antigens identify subsets of primary sensory neurons, Nature, 311: 469-472. Iwamori, M., Shimomura, J., Tsuyuhara, S., Mogi, M., Ishizaki, S. and Nagai, Y. (1983) Differential reactivities of fucosyl GM1 and GM1 gangliosides on rat erythrocyte membrane revealed by analysis with anti-fucosyl GM1 and GM1 antisera, J. Biochem., 94: 1-10. Kunemund, V., Jungalwala, F.B., Fischer, G., Chou, D.K.H., Keilhauer, G. and Schachner, M. (1988) The L2/HNK-1 carbohydrate of neural cell adhesion molecules is involved in cell interactions, J. Cell Biol., 106: 213-223. Kusunoki, S., Craft, J.E., Roach, B., Hardin, J.A. and Yu, R.K. (1987) A human IgM M-protein in a patient with unknown bleeding disorder binds to /3-galactosyl and /3-glucosyl residues, Arch. Biochem. Biophys., 255: 226-232. Kusunoki, S., Inoue, K., Iwamori, M., Nagai, Y. and Mannen, T. (1989) Discrimination of human dorsal root ganglion cells by anti-fucosyl GM1 antibody, Brain Res., 494: 391-395. Kusunoki, S., Inoue, K., lwamori, M., Nagai, Y. and Mannen, T. (1991) Fucosylated glycoconjugates in human dorsal root ganglion cells with unmyelinated axons, Neurosci. Lett., 126: 159-162. Lowe, J.B., Stoolman, L.M., Nair, R.P., Larsen, R.D., Berhend, T.L. and Marks, R.M. (1990) ELAM-l-dependent cell adhesion to vascular endothelium determined by a transfected human fucosyltransferase cDNA, Cell, 63: 475-484.
80 Mori, K. (1986) Lectin Ulex europaeus agglutinin I specifically labels a subset of primary afferent fibers which project selectively to the superficial dorsal horn of the spinal cord, Brain Res., 365: 404408. Nakagawa, F., Schulte, B.A. and Spicer, S.S. (1986) Lectin cytochemical evaluation of somatosensory neurons and their peripheral and central processes in rat and man, Cell Tissue Res., 245: 579-589. Omlin, F.X., Matthieu, J.M., Philippe, E., Roch, J.M. and Droz, B. (1984) Expression of myelin-associated glycoprotein by small neurons of the dorsal root ganglion in chickens, Science, 227:1359 1360.
Plenderleith, M.B., Cameron, A.A., Key, B. and Snow, P.J. (1988) Soybean agglutinin binds to a subpopulation of primary sensory neurons in the cat, Neurosci. Lett., 86: 257-262. Scott, S.A., Patel, N. and Levine, J.M. (1990) Lectin binding identifies a subpopulation of neurons in chick dorsal root ganglia, J. Neurosci., 10: 336-345. Streit, W.J., Schulte, B.A., Balentine, J.D. and Spicer, S.S. (1986) Evidence for glycoconjugate in nociceptive primary sensory neurons and its origin from the Golgi complex, Brain Research, 377: 1-17.