Monoclonal antibodies to the optic tectum and cerebellum of the bullfrog, Rana catesbeiana.

Abstract

Amphibians such as the bullfrog, Rana catesbeiana, are often used in studies of developmental plasticity because the animals show a great capacity for central nervous system regeneration. However, immunohistochemical analyses employing amphibians have been hindered by a scarcity of reagents that recognize amphibian brain components. In this study, we generated a panel of monoclonal antibodies to brain antigens of Rana catesbeiana. Several of the antibodies reacted with antigenic sites that appeared to be localized in the extracellular spaces, yielding reactions that produced a strong "cell-negative" stain in which the neuronal somata were clear against a background of dense stain. Two of the most interesting antibodies localized principally to the cerebellum, while a third antibody reacted with an antigen abundant in the zone of the optic tectum known to receive ingrowing optic fibers. Through biochemical and immunological techniques, the antigen in the optic tectum was tentatively identified as the human natural killer cell antigen-1 (HNK-1) carbohydrate epitope of neural cell adhesion molecule (NCAM). Monocular enucleation of tadpoles resulted in visibly diminished HNK-1 staining of the stratum opticum in the denervated optic tectum contralateral to the enucleation. The unique staining patterns of the anti-HNK-I and additional antibodies we have generated should facilitate investigations of bullfrog brain maturation and regeneration by providing reagents suitable for tracking antigen expression in histological specimens.

Key words: Rana catesbeiana, optic tectum, cerebellum, neural cell adhesion molecule (NCAM), human natural killer cell antigen- 1 (HNK- 1), monoclonal antibodies

1. Introduction

Our laboratory has used the bullfrog, Rana catesbeiana, as a model for studying vertebrate brain development for several years. Most of our efforts have been directed toward analysis of cerebellar development (e.g., Uray and Gona, 1977; Uray, 1985; Uray and Gona, 1999), but recently our interests have come to include the bullfrog visual system. Attempts to supplement morphological studies with an immunohistochemical analysis of the maturation process in the bullfrog, using commercially available monoclonal antibodies (MABs) to relevant antigens of avian and mammalian origin, have been largely unsuccessful. Other researchers have noted a similar lack of cross-reactivity in amphibians when using antibodies generated against other taxa (Steen et al., 1989; Crowe and Piley, 1992). Some laboratories have circumvented this problem by generating their own MABs specific for R. catesbeiana antigens (Reh and Nagy, 1987; Anholt et al., 1990; Key and Akeson, 1990), a strategy we have employed here to obtain reagents that may be useful in tracking the development of the optic tectum and cerebellum in this species.

2. Materials and Methods

Frogs. Bullfrog (Rana catesbeiana Shaw) tadpoles were obtained from Grassy Forks Fisheries, IN, USA. The animals were reared under laboratory conditions in aerated, dechlorinated water and were fed canned spinach. Tadpoles that metamorphosed into froglets were reared for up to seven months and were fed Mazuri amphibian feed (PMI Feeds, Inc., St. Louis, MO, USA). All rearing and experimental protocols were conducted in accordance with Institutional Animal Care and Use Committee (IACUC) standards.

Tadpoles were anesthetized by immersion in a dilute solution of methanesulfonate salt (Sigma Chemical Co., St. Louis, MO, USA). Froglets were anesthetized by intraperitoneal injection of ethyl carbamate solution (Mallinckrodt Chemicals, Chesterfield, MO, USA). Following anesthesia, tissue containing the optic tectum and cerebellum was dissected from the crania of tadpoles and froglets and homogenized together in amphibian Ringer's solution. The homogenates were centrifuged at 14,000g for 2 min, and the protein concentrations of the supernatant and pellet fractions were determined using a dye binding method (Bradford, 1976). Protein concentrations were adjusted to 1 mg/ ml in 0.05 M phosphate buffered saline (PBS), pH 7.5, and the samples were stored at--80[degrees]C until used for the immunization of mice.

Monoclonal antibody production and screening. Young, female BALB/c mice (Mus musculus Linnaeus), purchased from Hilltop Labs (Scottsdale, PA, USA), were housed and used in accordance with IACUC guidelines. The animals were immunized by intraperitoneal injections containing 100 [micro]g of frog brain protein. For the first and second injections, the protein was emulsified in an equal volume of Freund's complete or incomplete adjuvant, respectively. All subsequent antigen preparations were made in PBS alone. Immunizations proceeded at 2- to 3-week intervals until serum antibody titers against brain homogenates reached a minimum of 1000 in immunoblots of denaturing polyacrylamide gels (described below). Four days prior to fusion, mice received booster injections containing 50 [micro]g of soluble brain protein administered via the tail vein. Splenocytes from the immunized mice were fused to the nonsecreting myeloma cell line SP2/O-Ag 14 with 1300-1600 MW polyethylene glycol (Sigma). Hybridomas were selected in hypoxanthineaminopterin-thymidine medium and maintained in vitro by methods modified from Van Deusen (1983) in Dulbecco's modified Eagle's medium supplemented with 15% horse serum (Sigma). Cell lines were cloned twice by the limiting dilution method (Campbell, 1984) before their use in antigen characterization.

Undiluted hybridoma supernatants were screened for MABs reactive with brain components by an indirect enzyme-linked immunosorbent assay (ELISA) (Lenz and Greenstone, 1988) in which soluble frog brain protein (5 [micro]g/100 [micro]l/well in 0.01 M PBS, pH 7.4) was used as the plate-coating antigen. Reactive MABs were detected with 1:1000 peroxidase-conjugated goat antibodies specific for mouse IgG, IgM and IgA (Sigma). Wells were developed with 2,2'-azinobis(3-ethylbenzthiazoline-6-sulfonic acid) (ABTS, Sigma), and absorbances were determined at 410 nm in a BioTek ELISA reader (Winooski, VT, USA). Hybridoma supernatants that yielded absorbances at least 5 times greater than negative controls (spent medium from myeloma cells) were subsequently assayed against tissue sections of R. catesbeiana brain.

Immunohistochemistry. A total of 20 premetamorphic and prometamorphic tadpoles and froglets were sacrificed by immersion in methanesulfonate. The crania were dissected free and fixed for 2 h in Bouin's fixative, rinsed in running tap water for 30 min, and then refrigerated overnight. The next day, the brains were dissected from the skulls and embedded in paraffin. Sections 8 [micro]m thick were mounted on Probeon Plus slides (Fisher Scientific, St. Louis, MO, USA). Slides were processed and stained according to the method of Reed et al. (1992), using the capillary gap technology of the MicroProbe system (Fisher). In each reaction, undiluted hybridoma supernatant was used as the primary antibody, and peroxidase-labeled goat anti-mouse immunoglobulins conjugate (Sigma) diluted 1:1000 was used as the secondary antibody. Immunoreactive sites were detected with 3,3'-diaminobenzidine (DAB). The sections were covered with water-miscible mounting medium (Crystal Mount, Fisher), examined with an Olympus microscope, and photographed with an Olympus PM-10 camera.

Effect of monocular enucleation on the staining pattern obtained with MAB 1H12. In a series of 12 R. catesbeiana of various developmental ages, the right eye was removed under general anesthesia. Six animals were sacrificed after 2 weeks of survival, and six animals of matched ages at enucleation were sacrificed at 14 weeks. For each animal, the entire brain was removed, lightly fixed in formalin, paraffin embedded, sectioned and reacted histochemically against the 1H12 antibody (generated during this study) using DAB as the chromagen. Mounted sections were examined microscopically and photographed. Because the retinotectal projection in the bullfrog is entirely contralateral, comparison of the extent of the 1H12 histochemical reaction could be made in each animal by comparing the normal and enucleated tecta.

Immunoblot analysis of R. catesbeiana antigens. Bullfrog brain antigens (100 [micro]g/lane) solubilized in Laemmli (1970) sample buffer were resolved by electrophoresis through 20-cm denaturing gradient gels of 5-15% or 5-20% polyacrylamide. Antigens were electrotransferred from the gels to supported nitrocellulose membranes (Durablot, Sigma) for immunoblot analysis (Partanen et al., 1983). Membranes were incubated for 30 min at room temperature in blocking buffer [5% nonfat dry milk in Tris-buffered saline (TBS) containing 20 mM Tris-HC1, 500 mM NaC1, pH 7.5], followed by a 2-h incubation in hybridoma supernatant diluted 1:4 in blocking buffer containing 0.05% Tween 20. The membranes were washed extensively in TBS-0.05% Tween 20 (TTBS), and then incubated for 1 h in peroxidase-conjugated goat antiserum specific for mouse IgG, IgA and IgM, diluted 1:1000 in TTBS-1% nonfat dry milk. After further washes, the membranes were developed in 4-chloro1-naphthol/3,3' diaminobenzidine (CN/DAB, Pierce). Molecular masses of the immunoreactive components were calculated using Bio-Rad (Hercules, CA, USA) broad-range standards, and are reported in this paper as the average determined from 3-5 blots for each band.

Gels not used for blotting were stained with Coomassie Brilliant Blue R-250 or a silver stain kit (Bio-Rad). Alternatively, some gels were stained with Stains-All (ICN Biochemicals Inc., Costa Mesa, CA, USA) (Goldberg and Warner, 1997), a cationic carbocyanine dye that stains sialoglycoproteins and phosphoproteins blue, most other proteins red, and phospholipids yellow-orange (Campbell et al., 1983).

Identification of the antigen recognized by MAB 1H12. Additional biochemical and immunological investigations were performed to ascertain the specificity of MAB 1H12. In the first of these investigations, the 1H12 antigen was assayed for protein determinants by pronase digestion (Sharma et al., 1983). Brain homogenate protein in 0.1 M Tris/0.02 M Mg[Cl.sub.2] buffer, pH 7.2, was incubated for 2 h at 37[degrees]C with pronase (protease type IV from Streptomyces griseus) (Sigma) at a ratio of 5 [micro]g of homogenate to 1 [micro]g of enzyme. The samples were dried by vacuum evaporation, solubilized by boiling for 5 min in Laemmli sample buffer, and then subjected to electrophoresis and immunoblot analysis with MAB 1H12 as described previously.

To assay the 1H12 antigen for carbohydrate residues, the antigen was immunoprecipitated from frog brain homogenates, and aliquots of the immunoprecipitate were incubated in the presence or absence of deglycosylation enzymes before being probed with antibody. For the immunoprecipitation procedure, 25 [micro]g of MAB 1H12 were incubated for 2 h at 37[degrees]C with 3.3 mg of paramagnetic beads (Dynal, Inc., Lake Success, NY, USA) pre-coated with goat anti-mouse IgM. The beads were collected with a magnet and washed in several changes of PBS containing 0.05% Tween 20. Subsequently, 100 [micro]g of frog brain protein were added to the beads, and incubation was allowed to proceed for an additional 2.5 h at 37[degrees]C. After further washes, the immunoprecipitated antigen was incubated with 1 [micro]l each of NANase II, O-glycosidase DS, and PNGase F, either individually or together, using a Bio-Rad deglycosylation kit. After deglycosylation, the samples were dried by vacuum evaporation, reconstituted in electrophoresis sample buffer and used for electrophoresis and immunoblot analysis with MAB 1H12.

To assay the antigen recognized by MAB 1H12 for sialic acid residues in particular, brain homogenates were incubated for 18 h at 37[degrees]C with neuraminidase type V from Clostridium perfringens (Sigma) at a ratio of 5 [micro]g of protein to 1 mU of enzyme in 0.05 M sodium phosphate/0.01 M EDTA buffer, pH 7.2 (Backstrom et al., 1995). Samples dried by vacuum evaporation were reconstituted in electrophoresis sample buffer and probed by immunoblotting with MAB 1H12.

To test the antigen recognized by MAB 1H12 for cross-reactivity with antibodies specific for neural cell adhesion molecule (NCAM) and the carbohydrate epitope HNK-1, immunoprecipitated antigen was electrophoresed and transferred to nitrocellulose as described above. The nitrocellulose strips were incubated for 90 min at room temperature with 10 [micro]g/ ml of an affinity-purified polyclonal rabbit antiserum raised against purified chicken NCAM (Cat. No. AB 5032, Chemicon, Temecula, CA, USA). Rabbit antibodies bound to the blot were detected with 1 : 1000 sheep anti-rabbit IgG-AP (Sigma) and visualized with nitroblue tetrazolium/5-bromo-4-chloro-3-indolyl phosphate (NBT/BCIP). Additional nitrocellulose strips were incubated for 2 h at room temperature with 10 [micro]g/ml of mouse monoclonal antibody VC1.1 (Sigma) specific for HNK-1. The binding of VC1.1 to the blot was detected with 1:2000 anti-mouse IgM-AP, followed by NBT/BCIP.

Competitive ELISA. A competitive ELISA was performed to measure the ability of MAB VC1.1 to inhibit binding of MAB 1H12 to frog brain antigen. For this assay, an aliquot of MAB 1H12 was conjugated to horseradish peroxidase (HRP) using a maleimide-activated labeling kit (Pierce). Falcon Probind ELISA plates (Fisher) were coated by overnight incubation at 4[degrees]C with 5 [micro]g/well of brain homogenate in 100 [micro]l/ well of 0.01 M PBS, pH 7.4. The wells were blocked by a 15-min incubation at 37[degrees]C in 250 [micro]l/well of PBS containing 20% w/v nonfat dry milk. Subsequently, the wells were incubated for 1 h at 37[degrees]C with 100 [micro]l/well of diluent consisting of PBS containing 0.05% Tween 20 and 0.5% bovine serum albumin (no competitor), or with 10 [micro]g/well of MAB VC1.1 as a competitor. As controls, antigen-coated wells were incubated with 10 [micro]g/well of unlabeled MAB 1H12 (positive control for inhibition), or with 10 [micro]g/well of the irrelevant IgM MAB 9A5 (negative control for inhibition) in place of MAB VC1.1. The plates were washed five times with tap water containing 0.05% Tween 20, and then incubated for 45 min at 37 [degrees] C with 100 [micro]l/well of 1H12-HRP conjugate diluted 1:500. After further washes, the plates were developed by a 5-min incubation at room temperature in ABTS substrate solution. Color development was stopped by the addition of sodium azide (final concentration, 0.0085%), and absorbances ([A.sup.410]) were read at 410 nm. All assays were run in triplicate. The ability of each competitor to inhibit subsequent binding by 1H12-HRP conjugate was calculated using the equation,

Percent inhibition=[1 - [A.sup.410] of competitor sample/[A.sup.410] of diluent sample] x 100

The software program SigmaStat 2.0 (San Rafael, CA) was used to analyze the data by one-way analysis of variance (ANOVA), followed by an all pairwise multiple comparison procedure (Tukey test) to isolate differences in mean inhibition values.

3. Results

Monoclonal antibody production and screening. Supernatants from 79 wells supporting hybridoma growth yielded ELISA absorbance values at least five times greater than negative control supernatants. MABs from all 79 lines were used to probe bullfrog brain tissue sections. Of these, six MABs (1B3, 4B8, 1H12, 7E6 12C4, and 12D11) were selected for further analysis based on their tissue staining patterns.

Immunohistochemistry. The antigens recognized by the selected MABs showed variation in tissue distribution in all developmental ages of R. catesbeiana that were studied. For the purposes of illustration, the focus of the description here will be on the localization of the MABs in the fully-metamorphosed animal, the froglet. Several of the MABs reacted strongly with antigenic sites that appeared to be localized in the extracellular space in both the optic tectum (Fig. 1, A-F) and cerebellum (GL). These reactions produced a strong "cell-negative" immunostain in which the neuronal somata were clear against a background of dense stain. This reaction allowed easy recognition of nuclei or cell laminae, as collections of somata were easily distinguished as unstained relative to the darkly reacted extracellular matrix. In most of these stains, the neuronal dendrites or axons were not identified; thus, we could not determine whether the reaction was against only elements of the extracellular space, or might include processes of neurons and/or glial cells.

[FIGURE 1 OMITTED]

Although similar, the cell-negative staining reactions yielded by MABs 1H12 (Fig. 1, C and I), 12C4 (D and J), 12D 11 (E and K), and 7E6 (F and L) were not completely identical. Reaction products stained with MABs 1 B3 (A and G) or 4B8 (B and H) were considerably lighter than the other antibody reactions. Additionally, 4B8 reacted against endothelial cells of the capillary bed, a reaction not seen in the other tissues. It should be emphasized that the sections illustrated in Fig. 1 are all from the same animal, reacted at the same time under identical conditions, varying only in primary antibody.

For our purposes, one of the most interesting antibodies was MAB 1H12. As shown in Fig. 1C, reaction with MAB 1H 12 produced a dense, granular reaction product in the outer layer (the stratum opticum) of the developed portion of the optic tectum. This discretely organized reaction product was found in precisely the site where the ingrowing fibers from the retina were located. The appearance of the optic tectum stained with MAB 1H12 was that of a strongly laminated structure with layers of neuronal cell bodies alternating with relatively somata-free zones. The unstained, clear areas in the deep layers of the optic tectum were composed principally of densely-packed cell bodies, and the darkly-stained layers were composed of neuronal processes, mainly axons. In the more superficial layers, individual neuronal somata were seen as clear against a relatively uniform, densely immunostained reaction product.

As in the optic tectum, the principal reaction products recognized by the antibodies in the cerebellum were elements outside the neuronal somata (Fig. 1, G-L). Cell bodies free of reaction product were seen as clear against a dark background. The cerebellum was recognized more avidly than the optic tectum by MABs 1B3 (A and G) and 4B8 (B and H). Within the cerebellum, the staining patterns were regionalized, implying a relatively unique compartmentalization of the antigen or antigens recognized by these two antibodies. MABs 1H12 (I), 12C4 (J), and 12D11 (K) bound across all cell laminae in the cerebellum, but were especially well seen in the molecular layer; these antibodies yielded alternating lightly- and darkly-stained stripes that were in register with each other. Another interesting pattern in the cerebellum was seen for MABs 12C4 (J), 12D11 (K), and 7E6 (L), wherein the wide, sinusoidal-shaped area containing the Purkinje cell layer was particularly obvious because of the lack of any reaction product.

Effect of monocular enucleation on the staining pattern obtained with MAB 1H12. MAB 1H12 immunostained the stratum opticum of R. catesbeiana at all ages investigated, including the premetamorphic (Fig. 2A) and prometamorphic tadpole (Fig. 2B) and the froglet (Fig. 3). Removal of the right eye, followed by a 2- or 14-week survival period, resulted in visibly diminished immunostaining by MAB 1H12 in the denervated optic tectum contralateral to the enucleation at each developmental stage. Differences in immunostaining between the normal and denervated sides were especially apparent 14 weeks after enucleation at the froglet stage (Fig. 3).

[FIGURES 2-3 OMITTED]

Immunoblotting. Each MAB, with the exception of 7E6, reacted with brain antigens in immunoblots of denaturing polyacrylamide gels (Fig. 4). MABs 1B3 (A) and 4B8 (B) both recognized a major component of 201 kilodaltons (kDa). MAB 12C4 (D) recognized several antigens between 31 and 38 kDa, while MAB 12D11 (E) bound to a discrete band of 15 kDa.

[FIGURE 4 OMITTED]

The major antigen detected by MAB 1H12 in immunoblots of bullfrog brain homogenates was a component of 185 kDa (Fig. 4C). Additional immunoreactive products seen in some blots included those of 201, 168, 145, 120, 110, and 100 kDa. Immunoblotting was the best method of visualizing the 1H12 antigen, since the antigen stained only lightly with a silver stain kit and not at all with Coomassie Blue or Stains-All (data not shown), even after the antigen was concentrated by immunoprecipitation.

Determination of MAB 1H12 specificity. The sensitivity of the antigen recognized by MAB 1H12 to pronase digestion provided evidence of its proteinaceous nature (Fig. 5A). When the antigen was deglycosylated with PNGase F, it was no longer recognized by MAB 1H12, a result that demonstrated the antibody's specificity for an N-linked carbohydrate determinant on the 1H12 glycoprotein antigen (Fig. 5B). Sialic acid was eliminated as the carbohydrate's possible identity, because treatment of the antigen with a neuraminidase capable of cleaving [alpha] (2,3), [alpha](2,6) and [alpha](2,8) sialic acid linkages (Powell et al., 1987) did not alter recognition by MAB 1H 12 (Fig. 5C). In contrast, immunoblot analysis with MAB VC1.1 showed that the HNK-1 carbohydrate was present on the antigen immunoprecipitated with MAB 1H12 (Fig. 6C), and should be included among the potential targets of antibody specificity. In addition, the antigen cross-reacted with a polyclonal antibody specific for NCAM (Fig. 6B), a glycoprotein known to express HNK-1 epitopes on some of its glycoforms (Naegele and Barnstable, 1991).

[FIGURES 5-6 OMITTED]

Competitive ELISA. To analyze more closely whether MAB 1H12 is specific for the HNK-1 epitope, a competitive ELISA was performed to measure what effect pre-incubation of frog brain antigen with various reagents had on subsequent recognition by 1H12-HRP conjugate (Table 1). The mean absorbance value obtained by pre-incubating the antigen with 10 [micro]g/well of MAB 9A5 was not significantly different from that obtained by pre-incubation with diluent alone (P = 0.511); hence, MAB 9A5 was used as a negative control in the competition assay. Pre-incubating the brain homogenate with MAB VC1.1 blocked subsequent binding by 1H12-HRP by 56.6%, a value that was significantly different from the inhibition observed for MAB 9A5 (P<0.001, Tukey test). These results imply that MABs 1H12 and VC1.1 compete for the same antigenic determinant, HNK-1.

4. Discussion

Frogs and other amphibians exhibit a low degree of morphological differentiation of the brain compared to other vertebrates, while at the same time showing a remarkable capacity for central nervous system regeneration (Becker et al., 1993b). For this reason, amphibians are often used in studies of developmental plasticity, but immunohistochemical analyses employing these animals have been hampered by a lack of reagents that recognize amphibian brain components. Our initial attempts to probe the cerebellum and optic tectum of the bullfrog, R. catesbeiana, employed commercially available monoclonal antibodies and polyclonal sera to such antigens as Leu-4 (found on rodent Purkinje cells), glial fibrillary acidic protein, neuron-specific enolase, and microtubule-associated proteins from various mammalian and avian species. The disappointing results prompted us to generate our own panel of monoclonal antibodies specific for R. catesbeiana brain components.

The antigens recognized by the antibodies generated during this study are similar in molecular mass to a number of axonal guidance molecules, adhesion molecules and extracellular matrix proteins known to be associated with developmental events. In particular, the similarity of the 1H12 antigen to previously published descriptions of neural cell adhesion molecule (NCAM) led us to probe the identity of this antigen more closely. NCAM is perhaps the most widely studied of the cell surface glycoproteins involved in the process of cell recognition (Sanes, 1989; Schachner and Martini, 1995). Through its homophilic binding, the molecule contributes to the histogenesis of the embryonic CNS and facilitates later stages of intercellular recognition such as synapse formation and stabilization (Naegele and Barnstable, 1991). Evidence that MAB 1H12 recognizes NCAM includes: 1) the presence of the immunoreactive 1H12 antigen within neural tissue; 2) the similarities in antigen molecular masses, generated by SDS-PAGE of brain tissue, that are recognized by MAB 1H12 and the anti-NCAM antibodies described by others (Reh and Nagy, 1989; Key and Akeson, 1990, 1991); 3) recognition of the immunoprecipitated 1H12 antigen by an affinity-purified polyclonal serum specific for highly purified chicken NCAM; and 4) the presence of the HNK-1 carbohydrate on both the 1H12 antigen and NCAM, as discussed below. Conclusive identification of the 1H12 antigen awaits tandem mass spectrometry or amino acid sequence analysis of the immunoprecipitated glycoprotein.

NCAM exists as several different glycoforms that can be differentiated by their reactivities with lectins and monoclonal antibodies specific for carbohydrate determinants (Key and Akeson, 1990, 1991). Two of the best-characterized carbohydrates expressed on some subsets of NCAM molecules are polysialic acid (PSA) (Hoffmann and Edelman, 1983) and HNK-1 (Schwarting et al., 1987), both of which are bound to the protein portion of NCAM through an N-linkage. PSA is a negatively charged [alpha](2,8)-linked sialic acid homopolymer that helps maintain plasticity during embryonic development and the regeneration of specific areas in the adult brain by weakening the strength of NCAM-NCAM interactions (Hoffman and Edelman, 1983; Rutishauser et al., 1988; Becker et al., 1993a,b). HNK-1 is a sulfated glucuronic acid-containing carbohydrate found on glycolipids, proteoglycans, and several glycoproteins in addition to NCAM (Jungalwala, 1994); the latter include J1, myelin-associated glycoprotein and L1 (Kruse et al., 1985), as well as cytotactin/tenascin (Hoffman et al., 1988). HNK-1 binds to the G2 domain of laminin and to a number of additional substrates to mediate neuron-extracellular matrix adhesion (Schmidt and Schachner, 1998). Its expression is temporally and spatially regulated in the developing nervous system (Schwarting et al., 1987).

Because deglycosylation with PNGase F revealed that the epitope recognized by MAB 1H12 is an N-linked carbohydrate, we examined the 1H12 antigen for the presence of both PSA and HNK-1. Incubation of the antigen with a neuraminidase capable of cleaving [alpha](2,8) linkages between sialic acid residues (Powell et al., 1987) did not abrogate antibody recognition by MAB 1H12, thus eliminating PSA as the possible epitope for which the antibody was specific. In contrast, we showed that HNK-1 is present on the immunoprecipitated 1H12 antigen by immunoblot analysis employing MAB VC1.1, an antibody specific for HNK-1. The ability of VC1.1 to block subsequent recognition of frog brain homogenates by peroxidase-labeled MAB 1H12 indicated that the antibodies shared specificity for the HNK-1 carbohydrate.

The 1H12 antigen is of particular interest to us because it is located in that zone of the optic tectum known to receive optic fibers. It has been shown repeatedly that the ingrowth of optic fibers plays a crucial role in maintaining the normal growth of the amphibian optic tectum (Kollros, 1988). Monocular enucleation of tadpoles, followed by a 2- or 14-week survival period, resulted in visibly diminished immunostaining of the stratum opticum in the denervated optic tectum contralateral to the enucleation for all developmental stages studied, results that imply a relationship between the 1H12 antigen and the optic fibers. In the goldfish, HNK-1 has been implicated in the activity-driven sharpening of the retinotopic map formed by regenerating retinal fibers (Schmidt and Schachner, 1998). While it appears that the 1H12 antigen plays a similarly prominent role in visual synaptic plasticity in the bullfrog, elucidation of its exact role awaits further studies.

For the additional antibodies we generated, the localization of any particular MAB differed across the entire froglet brain, reflecting local variations in the presence of each corresponding antigen. We are interested in determining what, if any, relation local anatomical variations in antigen concentrations have to normal development. For example, the abundance of the 1B3 and 4B8 reaction products in the froglet cerebellum, but not the optic tectum, may indicate a burst of development in the cerebellum at this age. Alternatively, these antigens may be consistently high in the cerebellum relative to the optic tectum at all developmental ages. With the development of monoclonal antibodies to the bullfrog brain, we now have the reagents that will allow us to pursue these investigations in the amphibian model.

5. Acknowledgments

This research was supported by grant No. AA07537

from the NIAAA to N.J.U., No. 900-306 from ATSU to M.K.S., and No. 501-593 from the Warner/ Fermaturo Fund to L.C.T.

6. Literature Cited

Anholt, R.R.H., A.E. Petro and A.M. Rivers, 1990. Identification of a group of novel membrane proteins unique to chemosensory cilia of olfactory receptor cells. Biochemistry 29:3366-3373.

Backstrom, J. R., R.S. Westphal, H. Canton and E. Sanders-Bush, 1995. Identification of rat serotonin 5-HT2C receptors as glycoproteins containing N-linked oligosaccharides. Brain Res. Mol. Brain Res., 33:311-318.

C.G. Becker, T. Becker and G. Roth, 1993a. Distribution of NCAM- 180 and polysialic acid in the developing tectum mesencephali of the frog Discoglossus pictus and the salamander Pleurodeles waltl. Cell Tissue Res. 272:289-301.

T. Becker, C.G. Becker, U. Niemann, C. NaujoksManteuffel, R. Gerardy-Schahn and G. Roth, 1993b. Amphibian-specific regulation of polysialic acid and the neural cell adhesion molecule in development and regeneration of the retinotectal system of the salamander Pleurodeles waltl. J. Comp. Neurol. 336:532-544.

Bradford, M.M., 1976. A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal. Biochem. 72:248-254.

Campbell, A.M., 1984. Laboratory Techniques in Biochemistry and Molecular Biology, Vol. 13 (ed. R. H. Burdon and P. H. van Knippenberg), pp. 157-160. New York: Elsevier Science.

Campbell, K.P., D.H. MacLennan and A.O. Jorgensen, 1983. Staining of the [Ca.sup.2+]-binding proteins, calsequestrin, calmodulin, troponin C, and S-100, with the cationic carbocyanine dye "Stains-all". J. Biol. Chem. 258:11267-11273.

Crowe, M.J. and S.K. Pixley, 1992. Species differences in amphibian olfactory neuron reactivity to a monoclonal antibody. Brain Res. Bull. 28:785-788.

Goldberg, H.A. and K.J. Warner, 1997. The staining of acidic proteins on polyacrylamide gels: enhanced sensitivity and stability of "Stains-all" staining in combination with silver nitrate. Anal. Biochem. 251:227-233.

Hoffman, S., K.L. Crossin and G.M. Edelman, 1988. Molecular forms, binding functions, and developmental expression patterns of cytotactin and cytotactin-binding proteoglycan, an interactive pair of extracellular matrix molecules. J. Cell Biol. 106:519-532.

Hoffman, S. and G.M. Edelman, 1983. Kinetics of homophilic binding by embryonic and adult forms of the neural cell adhesion molecule. Proc. Natl. Acad. Sci. USA 80:5762-5766.

Jungalwala, F.B., 1994. Expression and biological functions of sulfoglucuronyl glycolipids (SGGLs) in the nervous system-a review. Neurochem. Res. 19:945-957.

Key, B. and R.A. Akeson, 1990. Olfactory neurons express a unique glycosylated form of the neural cell adhesion molecule (N-CAM). J. Cell Biol. 110:729-1743.

Key, B. and R.A. Akeson, 1991. Delineation of olfactory pathways in the frog nervous system by unique glycoconjugates and N-CAM glycoforms. Neuron 6:381-396.

Kollros, J.T., 1988. Developmental Neurobiology of the Frog. New York: Alan R. Liss, Inc.

Kruse, J., G. Keilhauer, A. Faissner, R. Timpl, and M. Schachner, 1985. The J1 glycoprotein: a novel nervous system cell adhesion molecule of the L2/ HNK-1 family. Nature 316:146-148.

Laemmli, U.K., 1970. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 256:495-497.

Lenz, C.J. and M.H. Greenstone, 1988. Production of a monoclonal antibody to the arylphorin of Heliothis zea. Arch. Insect Biochem. Physiol. 9:167-177.

Naegele, J.R. and C.J. Barnstable, 1991. A carbohydrate epitope defined by monoclonal antibody VC1.1 is found on N-CAM and other cell adhesion molecules. Brain Res. 559:118-129.

Partanen, P., H.J. Turunen, R. Paasivuo, E. Forsblom, J. Suni and P.O. Leinikki, 1983. Identification of antigenic components of Toxoplasma gondii by an immunoblotting technique. FEBS Lett. 158:252-254.

Powell, L.D., S.W. Whiteheart and G.W. Hart, 1987. Cell surface sialic acid influences tumor cell recognition in the mixed lymphocyte reaction. J. Immunol. 139:262-270.

Reed, J.A., L.J. Manahan, C.S. Parke and D.J. Brigati, 1992. Complete one-hour immunocytochemistry based on capillary action. Biotechniques 13:434-443.

Reh, T.A. and T. Nagy, 1987. A possible role for the vascular membrane in retinal regeneration in Rana catesbeina tadpoles. Dev. Biol. 122:471-482.

Reh, T.A. and T. Nagy, 1989. Characterization of Rana germinal neuroepithelial cells in normal and regenerating retina. Neurosci. Res. Suppl. 10: S151-162.

Rutishauser, U., A. Acheson, A.K. Hall, D.M. Mann and J. Sunshine, 1988. The neural cell adhesion molecule (NCAM) as a regulator of cell-cell interactions. Science 240:53-57.

Sanes, J.R., 1989. Extracellular matrix molecules that influence neural development. Annu. Rev. Neurosci. 12:491-516.

Schachner, M. and R. Martini, 1995. Glycans and the modulation of neural-recognition molecule function. Trends Neurosci. 18:183-187.

Schmidt, J.T. and M. Schachner, 1998. Role for cell adhesion and glycosyl (HNK-1 and oligomannoside) recognition in the sharpening of the regenerating retinotectal projection in goldfish. J. Neurobiol. 37:659-671.

Schwarting, G.A., F.B. Jungalwala, D.K. Chou, A.M. Boyer and M. Yamamoto, 1987. Sulfated glucuronic acid-containing glycoconjugates are temporally and spatially regulated antigens in the developing mammalian nervous system. Dev. Biol. 120:65-76.

Sharma, S.D., J. Mullenax, F.G. Araujo, H.A. Erhich and R.S. Remington, 1983. Western blot analysis of the antigens of Toxoplasma gondii recognized by human IgM and IgG antibodies. J. Immunol. 131:977-983.

Steen P., L. Kalghatgi and M. Constantine-Paton, 1989. Monoclonal antibody markers for amphibian oligodendrocytes and neurons. J. Comp. Neurol. 289:467-480.

Uray, N.J., 1985. Early stages in the formation of the cerebellum in the frog. J. Comp. Neurol. 232:129-142.

Uray, N.J. and A.G. Gona, 1977. The cerebellum of the bullfrog tadpole. J. Comp. Neurol. 176:559-573.

Uray, N.J. and A.G. Gona, 1999. Calbindin immunoreactivity in the auricular lobe and interauricular granular band of the cerebellum in bullfrogs. Brain Behav. Evol. 53:10-19.

Van Deusen, R.A., 1983. Hybridoma Technology in Agricultural and Veterinary Research (ed. N. J. Stern and H. R. Gamble), pp. 15-25. Beltsville, MD: Rowman and Allanheld.

Melissa K. Stuart (1), Patricia S. Sexton (2), Betty J. Cox (1), Nandor J. Uray (2) and Lex C. Towns (2)

(1) Department of Microbiology/Immunology, and (2) Department of Anatomy, Kirksville College of Osteopathic Medicine, A.T. Still University of Health Sciences, 800 West Jefferson Street, Kirksville, MO 63501, USA

Table 1. Results of the competitive ELISA

                                                    Mean Percent
                            Mean [A.sub.410]         Inhibition
Competitor                  [+ or -] SD (1)        [+ or -] SD (2)

Diluent (no competitor)   1.815 [+ or -] 0.012           --
MAB 9A5
  (negative control)      1.774 [+ or -] 0.055    2.3 [+ or -] 3.0
MAB 1H12
  (positive control)      0.274 [+ or -] 0.034   84.9 [+ or -] 1.9 *
MAB VC1.1
  (specific for HNK-1)    0.787 [+ or -] 0.022   56.6 [+ or -] 1.2 *

(1) Mean absorbance at 410 nm [+ or -] standard deviation, n=3.

(2) Calculated by the equation
[1- ([A.sub.410] of competitor sample/[A.sub.410] of diluent
sample)]x100

* Pre-incubation of frog brain antigen with this antibody
significantly inhibited subsequent binding by 1H12-HRP,
compared to negative control MAB 9A5 (P<0.001, Tukey test).