Developmental Brain Research, 67 (1992) 265-278 © 1992 Elsevier Science Publishers B.V. All rights reserved. 0165-3806/92/$05.00

265

BRESD 51460

A developmentally regulated plasmalemmal antigen present in synaptosomes but not in growth cones Ira Wallis a'l, Robert S. Lasher b, Leland Ellis a'2, Karen Siller a and Karl H. Pfenninger a'b aDepartment of Anatomy and Cell Biology, Columbia University, New York, NY (USA) and bDepartment of Celfular and Structural Biology, University of Colorado, Denver, CO (USA) (Accepted 11 February 1992)

Key words: Neuronal differentiation; Synaptogenesis; Presynaptic ending; Growth cone; Surface antigen; Monoclonal antibody

Monoclonal antibody 2L4 was generated against rat synaptosomes but does not cross-react with nerve growth cones. Expression of the 2L4 antigen is developmentally regulated in a manner that is, in part, the opposite of the expression of the 5B4-CAM antigen, a marker of neuronal outgrowth belonging to the N-CAM family. While 5B4-CAM appears and increases during sprouting and then decreases to reach low levels in the adult, the 2L4 antigen appears only late in development, when neuronal outgrowth ceases, at or around the time of synaptogenesis. Once expressed, the antigen is found on the entire plasmalemmal surface of the neuron, but seems to be enriched at synaptic endings, at least of some neuron types. Biochemical analyses involving blotting of non-denaturing gels and immunoaffinity chromatography identify the antigen as a pair of polypeptides with similar, basic isoelectric points. These polypeptides form a somewhat diffuse, probably glycosylated band at 67 kDa and may be part of a hetero-oligomeric complex. The localization, biochemical, and developmental results suggest that the 2L4 antigen is a plasmalemmal marker of maturing and/or mature neurons whose expression may be triggered by synaptogenesis. INTRODUCTION

Synaptogenesis entails dramatic morphological changes in the nerve growth cone 27'32, i.e., the differentiation of the growth cone into a presynaptic terminal. More difficult to characterize are the molecular and biochemical changes that take place during this transformation. A direct comparison of radiolabeled membrane proteins of isolated 'growth cone particles' (GCPs) and synaptosomes has been performed 3°. In addition, we have described a monoclonal antibody (Mab 5B4) that binds to the GCP fraction, a preparation of sheared-off pieces of nerve growth cones 25, but only weakly to synaptosomes 9'33. Mab 5B4 recognizes the cytoplasmic tails of ld (180 kDa polypeptide) and sd (140 kDa polypeptide) members of the N-CAM family 26. The 5B4 epitope, 5B4-CAM, is abundant in post-mitotic, sprouting neurons and enriched in growth cones, but only faintly detectable by immunofluorescence in the mature brain. In the developing spinal cord and the mature olfactory system, 5B4 has been shown to stain selectively the growing axons 33. The predominant antigen in growth cones is the polysialylated ld form of N-CAM whereas that in the adult brain is the polysialic-acid-free sd form 9'22'28.

In o r d e r to identify antigens that are expressed during neuronal m a t u r a t i o n , especially synaptogenesis, we have used analogous techniques to obtain a Mab (Mab 2L4) that differentially binds to synaptosomes r a the r than G C P s . In this p a p e r we characterize the 2L4 antigen and c o m p a r e its developmental regulation with that of 5 B 4 - C A M in the developing cerebellum. MATERIALS AND METHODS

Subcellular fractions and membrane preparations Subcellular fractions were prepared from various adult tissues or adult developing brains or cerebella of Sprague-Dawley rats. The gestational age of fetal rats is indicated counting the plug date as day 0. Crude membrane preparations were generated by homogenizing non-neural tissues in a glass-glass homogenizer and brains in a Teflon/glass homogenizer in approximately 6 volumes of 0.32 M sucrose containing 1 mM MgCl 2 and 1 mM Tris buffer (pH 7.3). Homogenates were spun at 1660 × gma, for 15 rain in a Beckman JA17 rotor, the pellet discarded and the supernatant pelleted at 100.000 × gmax for 60 min in a Beckman SW55Ti rotor. Protein concentrations were determined by the method of Bradford 4 (BioRad reagent), using bovine plasma globulin as standard. Synaptosomes ~ were prepared from cerebral cortex of the adult rat essentially according to the procedure of Cohen et al. 6 as modified by Ellis et al. 9. Synaptosomal membranes used for affinity purification of the 2L4 antigen were prepared according to Jones and Matus ~s.

Present address: Genentech Inc., South San Francisco, CA, USA 2 Present address: Institute of Biosciences and Technology, Texas A&M University, Houston, TX 77030. USA. Correspondence: K.H. Pfenninger, Department of Cellular and Structural Biology, University of Colorado Health Sciences Center, 4200 East Ninth Avenue, B-Ill, Denver, CO 80262, USA. Fax: (1) (303) 270-4729.

266

Growth cone particles (GCPs) were prepared from 17- to 18-day fetal brains according to the two-step procedure described by Pfenninger et al. 2s. The resulting "A fraction' was diluted 3- to 4-fold with 0.32 M sucrose (with 1 mM MgCI2 and 1 mM TES buffer, pH 7.3) and pelleted at 39,800 × gmax for 30 rain at 4°C. This GCP pellet (PI) was re-suspended for use as described or processed further for the isolation of GCP membranes. In many situations (see text), GCPs and membranes were prepared in the presence of the following protease inhibitors: phenyl-methyl-sulfonylfluoride (1 raM), soybean trypsin inhibitor (type !-S: 100/~glm!), pepstatin A (1/~g/ml) and ieupeptin (30 ~M). Cell cuhure Cerebella from 19-day gestation fetal rats were dissected, cut into small pieces and then digested in trypsin solution for 8 minutes at ambient temperature9. Soybean trypsin inhibitor and DNase I were added. The dissociated cells were pelleted'~ and plated into 35 mm poly-D-lysine-coated plastic dishes. The cells were grown in FI2 medium containing insulin, transferrin, progesterone, putrescine, selenous acid, and estradiol (all from Sigma Chemical Co., St. Louis, MO), at 37°C with about 3% CO2 in humidified air 3 and was supplemented with 10% human placental serum. Cultures were maintained from 4 to 13 days. Cultured cells were fixed by slowly infusing (80 ~l/min), directly into the culture medium, a 2% solution of paraformaldehyde in phosphate buffer (0.1 M, pH 7.3). Following infusion, ten drops of fixative were added to the plates. After 5 rain the cultures were drained and fresh 2% paraformaldehyde-phosphate buffer was added. Following 15 rain in the fresh fixative, the cultures were washed carefully with phosphate buffer. Generation of monoclonal antibodies Monoclonal antibodies (Mabs) were prepared according to Kohler and Milstein ~6, as modified by Sharon et al. ~'9. BALB/c mice were given an initial i.p. injection of pelleted synaptosomes (100 l~g protein)in phosphate-buffered saline (PBS), mixed 1:1 with Freund's complete adjuvant. Over a period of months, mice were boosted with i.p. injections of synaptosomes, plus Freund's incomplete adjuvant, and then given a final i.v. injection with synaptosomes in PBS. Four days after the i.v. injection, splenocytes were prepared from the immunized mouse. Splenocyt0s were fused with a k-chain-secreting myeloma cell line (45.6.TO,1.7.5) using PEO-1000 (J.T. Baker Chemical, Co., Phillipsburg, NJ), at a ratio of 10 splenocytes to I mycloma cell. Maintenance and selection of fused cells was as described in Ellis et al. 9, Cultures or cells that secreted antibodies recognizing synaptosomes but not GCPs in a differential enzyme-linked immunosorption assay (ELISA: ref. 10) or in a dot-immunobinding assay i'~ were cloned in semi-solid agarose 2~ 1-2 times, grown in mass culture and then used for the preparation of ascites fluid. Dot immunobinding assays Dot immunobindin~, assays (DIAs) were carried out as described by Hawkes et ai. I"~, Briefly, samples were spotted onto a nitrocellulose filter or in a Millititre plate (Millipore Corp., Bedford, MA) in Tris-buffered saline (TBS, 50 mM Tris-HCI, pH 7.3, plus 150 mM NaCI). The filters were blocked with 3% bovine serum albumin (BSA) in TBS (TBSA) and then incubated for 2 h with Mab (as culture supernatant) diluted 2- to 4-fold in TBSA. Following washing with TBSA, the filters were incubated with an immunoaf. finity-purified goat anti-mouse.lgG conjugated to horseradish peroxidase (HRP). Filters were again washed with TBSA and reacted with the chromogenic substrate 4-chloro-l-naphthol. For differential screening assays I l~g of synaptosome or GCP protein was spotted onto nitrocellulose in 1 ~1 of TBS. For the tissue specificity experiments, 1 l~g protein of whole synaptosomes and of crude membranes prepared from brain and various non-neural tissues was spotted into individual Millititre wells in 4 #i TBS. The developmental DIA was done using crude brain membranes prepared from fetal rats of gestational ages 13, 16, 19 and 22 days and from adult

cortex. These membrane preparations were spotted onto nitrocellulose in amounts varying from 12.5 to 1000 ng of protein (in 1 #1 TBS).

Cryotomy Adult brain tissue was obtained after perfusing the rat with 2% formaldehyde (prepared fresh from paraformaldehyde powder) in phosphate buffer (0.1 M, pH 7.3). The dissected tissue was immersed in the same fixative for 1 to 2 h at room temperature. It was then transferred into 30% sucrose in 0.1 M phosphate, pH 7.3, and kept at 4°C until the sample sank (usually'overnight). Following equilibration with 30% sucrose, tissue piece~, were embedded in O.C.T. (Lab-TEK. Miles, Naperville, IL) and frozen by slow submersion in liquid nitrogen. The frozen blocks were kept at -80°C until used for sectioning. Sections were cut in a cryostat to a thickness of approximately 10-15 #m. Ultracryotomy Cerebellar cortex, fixed as described above, was cut into small blocks, washed 2 x 5 min with 0.1 M phosphate buffer, pH 7.3, and then immersed in 1.6 M sucrose in buffer for 1 h at room temperature. Individual pieces of cortex were subsequently mounted on ultracryotome studs and quickly frozen in liquid nitrogen-cooled Freon 22. The frozen tissue blocks were sectioned on a Reichert ultracryotome to a thickness of approximately 0.5 ~m. Sections were mounted on poly-o-lysine-coated glass slides. lmmunofluorescence For indirect immunofluorescence, frozen sections and fixed cultures were first incubated in PBS containing 1% BSA (PBSA) for 30 rain to quench any residual aldehyde groups. Cultures to be labeled with Mab 5B4 were permeabilized first with 0.01% saponin in PBSA. Mab 2L4 or 5B4 (ascites fluids) and, as a control, my. eloma ascites were used for the initial antibody incubations (1 h at room temperature for sections, 2 h for cultures), diluted in PBSA, 250.fold for labeling sections, 200.fold for cultures. After washing with PBSA (3 × 5 rain) incubation with fluorescent secondary antibody (goat anti-mouse IgO-FITC, diluted 40-fold in PBSA) w a s for 1 h (sections) or 2 h (cultures) at room temperature, After washing, cultures were mounted under coverslips in PBS containing 50% glycerol. Frozen sections were sealed between glass coverslips in the same solution and then cemented to slides with Permount (Fisher, Springfield, NJ). For the semithin ultracryotome sections a modified labeling procedure was followed. The sections mounted on the glass slides were washed in fresh PBS, and then incubated in 2 drops of 2% paraformaldehyde in phosphate buffer for 5-10 rain, followed by another PBS wash. Residual aldehyde groups were quenched with 50 mM NH4C! in PBS for 10 rain. The sections were washed again with PBS for 5 rain and then incubated in PBS plus 0.2% gelatin for 5 rain. The same immunoreagents used for the cryotome sections were used here with some differences in the protocol, All immunoreagents and washes were in PBS plus 0,2% gelatin, pH 7,3. The primary antibody incubation was for 20 rain, and washes were for 3 × 5 rain. Labeling with the secondary antibody was for 45 rain followed by washing and mounting the sections in 90% glycerol in PBS. Immunofluorescence microscopy was done using epifluorescence optics. For comparison of fluorescence intensities in different sections and anatomical regions, exposure times were held constant. Likewise, negatives were printed using equal exposure times and conditions of paper development. hnmuno.electron microscopy The binding of Mab 2L4 to synaptosomes fixed at higher concentrations of glutaraldehyde (without formaldehyde) was greatly inhibited. Thus synaptosomes were la~eled unfixed using the following protocol (antibody-sucrose density gradient). Pelleted synaptosomes were resuspended in 0.32 M sucrose containing 1% BSA and Mab 2L4 (diluted 4-fold, fina~concentration) and incubated on ice for 60 rain. This solution (:~ ml) was then layered as the load

267 Antibody - 8ucroiI Density Gr,~dlnnt

~

Input (0.32MI1%BSAlidCA) O.gOMI1~ B8A

U

O.8OMII% B8AIGAM-IoG-HRP 1.'~! ~IL B8A

t

1080g-1~0'Ot4~C

P Resuspendin O.3gMBLears:n I 100000g-30' P Fix. gnnctwithDAB.Proce|sfor EM

Fig. 1. Flow diagram of the antibody-sucrose density gradient procedure. Molarity is for sucrose solution; MCA, monoclonal antibody; BSA, bovine serum albumin; GAM-IgG-HRP, goat antimouse IgG conjugated to HRP. For details, see Materials and Methods.

on a 3-step sucrose density gradient (3 ml/step) as illustrated in Fig. 1. The gradient was centrifuged in a swinging-bucket rotor (Beckman JS-7.5) at 1090 × gay for 2 h at 4°(2. The antibody-labeled, pelleted synaptosomes were resuspended in solution A (see synaptosome protocol) and centrifuged at 100,000 x gay for 30 min at 4°C, in a swinging-bucket rotor (Beckman SW55Ti). The pellet was washed briefly (2 x 5 min with 1 ml of 0.2 M sodium cacodylate buffer, pH 7.3) and fixed using cacodylate-buffered 1.5% glutaraldehyde. First, 300 ~! of the aldehyde was added to the second cacodylate buffer wash; after 10 min, this solution was exchanged with 1,5% glutaraldehyde in cacodylate buffer (for 30 min). The pellet was washed 3 × 10 min with cacodylate buffer and 3 × 5 rain with 0.1 M Tris-HCI buffer, pH 7.6. Subsequently, the pellet was reacted with saturated (750 /~g/ml) 3,Y-diaminobenzidine (DAB) in 0.05 M Tris-HC! buffer, pH 7.6. The incubation was carried out first without H2O2 for 30 rain and then in the presence of 0.003% H20 2 for 45 min at 4°C. The reaction was stopped by washing the pellet 3 × 5 min with 0.1 M Tris-HC! buffer, pH 7.3. Finally, the pellet was processed for electron microscopy as described in Pfenninger et al. 25. The ultrathin sections were not stained. Control samples were prepared as descri',ed above, using a second gradient and, instead of 2LA, an irrelevant monoclonai antibody in the load. For electron microscopy of cerebellar tissue, anesthetized rats were perfused transcardially with a fixative consisting of 4% paraformaldehyde +0.05% glutaraldehyde +15% saturated picric acid in 0.I M Sorenson's phosphate buffer, pH 7.4, containing 0.08 M sucrose. The cerebellum was removed, cut into three parts, and fixed for an additional 4 h. Vibratome sections were cut at 50 ~m, rinsed well in PBS +0.08 M sucrose (SS), treated for 30 min with 1% NaBH4 in SS and rinsed again in SS. The binding of 2L4 antibody prepared from hybridoma culture medium was localized with an indirect ABC immunoperoxidase technique using a biotinylated rat anti-mouse IgG (Jackson ImmunoResearch) followed by a complex of streptavidin plus biotin-HRP and by subsequent incubation of sections in Tris-buffered saline (pH 7.6) containing 0.05% diaminobenzidine-4HCI +0.03% n 2 o 2 + 0.01 M imidazole + 10/~M nickel ammonium sulfate. The sections were treated with 1% OsO4 in SS, rinsed in SS, dehydrated in ethanol, followed by

2-hydroxypropylmethacrylate and EMbed 812 (EM Sciences). UItrathin sections without counterstaining were examined in a Philips CM-10 electron microscope at 60 KV.

NEPHGE immunoblot Non-equilibrium pH gradient gel electrophoresis (NEPHGE) was clone essentially as described by O'Farrell etal. 24. However, slab gels rather than tube gels were used. The slab gels (70 mm long × 80 mm wide × 0.75 mm thick) were cast using a mini-gel apparatus (Bio-Rad). Each well was loaded with 100/~g of synaptosome protein solubilized 1:1 with lysis buffer (8 M urea, 2% [w/v] NP-40 or Triton X-100, 2% ampholytes pH 3.5-10 or 1.6% ampholyte pH 6-8 plus 0.4% ampholyte pH 3-10). For some experiments, synaptosomes (approx. 100/~g protein) were pretreated with 25/~g/ml each of leucine aminopeptidase and proteinase K in 10 mM NH4HCO 3 buffer (pH 8.0), for 1 h at room temperature. Following incubation with the enzymes some samples were centrifuged in a Microfuge to separate pellets from supernatants for separate electrophoresis. The gels were run using 0.01 M phosphoric acid as the upper chamber buffer and 0.02 M sodium hydroxide in the lower chamber at a constant 200 V for 2.5 h at ambient temperature. Following the run, the gel was equilibrated in 0.7% acetic acid for 20 rain and then electroblotted onto a sheet of nitrocellulose filter paper (Schleicher and Schuell). Blotting was done for 1 h at 300 mA in 0.7% acetic acid3~, whereby transfer occurred in the direction of the cathode. The blots were quenched in 3% TBSA overnight at 4°C and then reacted with a 500-fold dilution of 2L4 ascites fluid in TBSA for 2 h, at room temperature. Controls were incubated with an equivalent protein concentration of myeloma ascites fluid (1500-fold dilution) in TBSA. Excess antibody was removed by washing the immunoblots 3 × 5 min with TBSA (220 ml/wash). The secondary antibody was a goat anti-mouse IgG conjugated to HRP (immunoaffinity isolated, diluted 250-fold in TBSA). Incubation was for 2 h at room temperature. The immunoblots were washed as described above and developed using 4.chloro-l-naphthol as substrata s3. The reaction was stopped usually within 15 min by exhaustive washing with HaG.

lmmunoaffinity chromatography 2L4 antibody, partially purified from ascites using DEAE-AffiGel Blue (BioRad), was coupled to protein O-Sepharose 4B (3.5 mg/1.2 ml packed beads; Pharmacia). Antibody was then crosslinked to the beads using dimethyl-pimelimidateI'. SDS-polyacrylamide gel electrophoresis (SDS-PAGE) of beads boiled in sample buffer indicated that only a small amount of light chain was released after cross-linking. Synaptosomal plasma me~m'0ranes (SPMs) or GCPs were solubilized in PBS plus 0.5% Triton X.100 (pH 7.5; PT buffer). Approximately 400/~g of either GCPs or SPMs was mixed with 2L4 beads (about 70/~1 packed beads) in PT buffer (0.4 mi) in microcentrifuge tubes on a rotator at room temperature for 2 h. After a brief centrifugation, the supernatant ('void') was removed, and the beads were rinsed three times with PT buffer, and then washed with two changes of 0.1 M glycine-HCI, pH 2.5 (190 ;d e.a.), over 4 min. The combined eluate was rapidly neutralized in 20/~1 of 1 M Sorenson's phosphate buffer, pH 8.4. BSA was not used in these procedures. The various fractions were analyzed by SDS-PAGE as described below.

SDS-PAGE SDS-PAGE was performed according to Laemmli17. Stacking gels were 3% polyacrylamide and 2 cm in length. Resolving gels were 17 cm long and contained a 5-15% linear gradient of polyacrylamide or pre-cast gels where indicated. Samples were boiled in Laemmli buffer containing 2% SDS and 5% fl-mercaptoethanol prior to electrophoresis. Molecular weight standards (from BioRad, Richmond, CA) were myosin (200 kDa), fl-galactosidase (116 kDa), phosphorylase B (97 kDa), bovine serum albumin (66 kDa), ovalbumin (43 kDa), carbonic anhydrase (31 kDa), chymotrypsinogen (26 kDa). Gels were stained with Coomassie blue or silver ac-

268

B •

H

~':,~" . . . . .

~:.;~;~

-,"

,,~7;t,~;~;,;~;:::;'::~7{ < f'.

,-.