Journal of Neurocytology 4, 47-53 (I975)

Immunohistochemical demonstration of antigen associated with the post-synaptic lattice A. I. M A T U S Department of Anatomy, University College London, Gower Street, London WCIE 6BT

Received I2 March I974; revised 3 September 1974; accepted 17 September 1974

Summary The binding of antisera against synaptosomes from rat cerebral cortex to dissociated cells from both cerebral cortex and liver was assessed by immunofluorescent labelling. This showed that none of the liver cells bound antisynaptosome antibodies but that some of the cerebral cortex cells bound antibody on their surfaces. Immunofluorescent labelling showed that all the particles in the crude mitochondrial fraction from rat cortex bound antibodies present in the unadsorbed antisera. However, when the antisera were absorbed with purified mitochondria and myelin, only a proportion of the mitochondrial fraction particles then bound antibody. Isolated IgG from the adsorbed antisera was labelled with ferritin and incubated with the crude mitochondrial fraction. Examination in the electron microscope showed that the ferritin and hence the antisynaptosome antibody was bound to the postsynaptic thickenings of about 2o% of synaptosomes having their junctions in the plane of section.

Introduction The utility of immunohistochemical techniques for investigating the functional anatomy of the nervous system has already been established by studies of the distributions of particular molecules such as enzymes involved in transmitter metabolism (Hartman and Udenfriend, 1973), proteins associated with particular cell types (Bignami et al., I972) and cyclic nucleotides produced in response to physiological and pharmacological stimulation (Bloom et al., 1973). Potentially one of the most valuable applications of immunohistochemical methods to the nervous system is the localization in intact tissue of specific functional elements of the synaptic surface. Among these are presumed to be membrane-bound macromolecules subserving the release and reception of the various transmitter substances (Hall, 1972), molecular complexes responsible for the adhesion of the pre- and postsynaptic membranes (Pfenninger, 1973), and possibly others acting as markers for the establishment of specific synaptic connections (Barbera et al., I973). The availability of subcellular fractions from brain enriched in isolated nerve terminals 9 i975 Chapman and Hall Ltd. Printed in Great Britain

48

MATUS

or synaptosomes ( G r a y and Whittaker, I962 ) has afforded a simple means of raising antisera directed mainly against synaptic antigens (Mickey et al., I971 ; L i n and H s u , I97I ; H e r s c h m a n et al., 1972; Raiteri et aI., I973). T h e specificity claimed for such antisera has varied f r o m exclusive reaction with synaptic m e m b r a n e preparations ( H e r s c h m a n et al., I972) to extensive cross-reaction with other subcellular species such as synaptic mitochondria and myelin (Mickey et al., I97I). M o r e recently, delipidated extracts f r o m otherwise unfractionated brain tissue were used to raise antibody which was subsequently shown to be directed against two antigens, one soluble the other localized by immunohistochemical methods on the post-synaptic dense material of synaptosomes (Orosz et al., I973). I n the study reported here antisera raised against purified synaptosome preparations were investigated by immunohistochemical methods b o t h to establish the presence of antibodies specifically directed against synaptic antigens and to discover the localization of the latter within the structure o f t h e synaptosome.

Methods The crude mitochondrial fraction (P~) from cerebral cortex of rats (200-250 g bodyweight) was prepared by homogenization and differential centrifugation as described by Gray and Whittaker (i962). Synaptosomes were isolated from this crude fraction by centrifugation on a two-step density gradient of Ficoll (Pharmacia G.B.) in isotonic sucrose as described by Cotman and Matthews (I97z). Rabbits were immunized at fortnightly intervals by intravenous injection of synaptosome suspensions containing 5 mg of protein. Blood was collected on the Ioth day following the final injection and the serum heat-treated (56~ for z tl) and stored at --2o~ Absorption of both anti-serum and normal serum was performed by incubation for 3~ min at 37~ with a packed volume of mitochondria, myelin or P~ equal to half the volume of serum. Mitochondria were obtained as a P2 fraction from rat liver. Myelin was recovered from a two-step sucrose density gradient on which hypotonically shocked and sonicated P2 had been fractionated (Jones and Matus, z974)Dissociated cells were prepared from both liver and cortex of 7 day old rats by incubating the tissue in o.25% trypsin for 2o min at 37~ then dispersing the cells by repeated gentle pipetting and collecting them by centrifugation at 2o0 xg. Fluorescence histochemistry was performed by first resuspending packed cells or P~ in an equal volume of undiluted serum and incubating at 2o~ for 20 re_in. After washing three times with 2o vols.

Fig. I. One cell in a suspension of dissociated cells from rat cerebral cortex. (a) Immunofluorescence with antiserum to synaptosomes. Focused on cell surface. (b) Phase contrast focused to show nucleus (n). • 16oo. Fig. 2. Clump of cells in a suspension of dissociated cells from rat liver. (a) Immunofluorescence with antiserum to synaptosomes (b) phase contrast. • 7oo. Fig. 3. P~ fraction from rat cerebral cortex. (a) Immunofluorescence with antiserum to synaptosomes (b) phase contrast. • 9oo. Fig. 4. P~ fraction from rat cerebral cortex. (a) Immunofluorescence with antiserum to synaptosomes absorbed with myelin and mitochondria (b) phase contrast. • 9oo. Fig. 5. P2 fraction from rat cerebral cortex to show synaptosome with postsynaptic lattice (pl) labelled with ferritin-antisynaptosome IgG. (a) Uranyl acetate block stain, not lead stained. • 55 ooo. (b) Inset • 2ooooo (c) lead stained. • 55000 (d) ferritin molecules (ft) of conjugate bound m postsynaptic lattice. • 225000.

5~

MATUS

of IO% (w/w) sucrose, cells or P2 were resuspended in fluorescein-labelled goat anti-rabbit IgG (Wellcome Labs., Dagenham) and incubated and washed as before. IgG was isolated from normal serum and antiserum by ammonium sulphate precipitation and chromatography on DEAE-cellulose as described by DePetris and Raft (1972). The IgG was conjugated to ferritiu by glutaraldehyde and the conjugate isolated by density gradient centrifugation (De Petris and Raft, 1972). P2 was incubated with the conjugate for 20 min at 2o~ and washed as in the fluorescence experiments. The washed pellet was fixed for 1-2 h with 3.5% glutaraldehyde in o.I M phosphate buffer pH 7.4 followed by 1% OsO4 for I h. Blocks were stained with aqueous 1% uranyl acetate, dehydrated via graded alcohols and embedded in Araldite. Ultrathin sections were examined with or without o. I ~o lead citrate staining on the grid.

Results When dissociated cells from rat cerebral cortex were incubated with antisera raised against cerebral synaptosome preparations (anti-syn) and subsequently with fluorescein-labelled goat anti-rabbit IgG (G-anti-R-FL), fluorescent labelling was evident on the surface of many cells (Figs. Ia, b). The fluorescence was distributed over the cell surface in a reticular pattern among which were larger (I-2 ~zm) fluorescent dots. Fluorescence of this type was absent from the interior of the cells. Cells incubated with non-immune sera instead of antisyn were devoid of surface fluorescence. A further control was established by the concurrent incubation of dissociated liver cells from the same animal with anti-syn and G-anti-R-FL under identical conditions to those employed for the cerebral cortex cells. The results of this experiment are shown in Figs. 2a and b and it is evident that there is no fluorescent labelling of the liver cell surface as a result of incubation with anti-syn sera, the only fluorescence associated with the cells being a weak emission which the cytoplasm of all cells shows when examined under the fluorescence microscope. The crude mitochondrial fraction (P2) from cerebral cortex incubated with anti-syn and G-anti-R-FL showed the following features (Figs 3a, b). Clumps of small particles (circa I ~m) were rendered highly fluorescent while larger (up to 5 ~m) particles showed a distinct though less intense fluorescent ring. Although not well reproduced in photographs, many of the particles ill the clumps of smaller particles could be seen to exhibit a ring of fluorescence rather than a solid dot. In both cases this indicates binding of antibody to the particle surface and absence of bound antibody from the interior. On the smallest particles observed, a ring of fluorescence could not be distinguished but this may merely indicate that such a structure is below the limit of resolution for the microscopic system employed. The induction of the fluorescence described was specifically dependant upon incubation of the P2 fraction with anti-syn sera and was not visible in P2 concurrently incubated with nonimmuna serum under identical conditions. The anti-syn serum was successively absorbed wkh purified mitochondria and myelin and the absorbed antiserum tested against P2 as before. Fluorescence was now limited to a weak emission from some of the clumps of small particles (Figs. 4a, b). Many of the small particles and all the larger particles which were rendered fluorescent by incubation with the unabsorbed serum did not fluoresce after incubation with the absorbed sera. When the absorbed antiserum was further absorbed with purified cerebral synaptosomes it no longer induced any fluorescence in the P~ fraction.

Synapse-specific antigen

51

To determine the ultrastrucmral localisation of the synapse-specific antigen IgG isolated from anti-syn serum after absorption with mitochondria and myelin was labelled with ferritin (anti-syn-FT) and then incubated with P2. When examined in the electron microscope all the bound ferritin was associated with the post-synaptic dense material (lattice) of about 2o% of the synaptosomes (Figs. 5a, b, c). As a control, P2 from the same preparation was incubated with ferritin-labelled IgG from similarly absorbed non-immune serum. In this case electron microscopic examination showed that ferritin was not bound to post-synaptic lattice. Discussion

The immunofluorescence assay shows that antisera raised against cortical synaptosomes contain antibodies which bind to many components of the crude mitochondrial fraction (P2) from cerebral cortex. The major components of the Pz fraction are synaptosomes, mitochondria, and myelin and accordingly the antiserum was sequentially absorbed with purified preparations of these species. Absorption with purified myelin selectively removed antibody binding to the large (circa 5 ~m) particles in the P2 fraction which are seen by electron microscopic examination to be fragments of myelinated processes. It is thus dear that antibodies directed against myelin are present in the antiserum. Absorption with mitochondria greatly reduced the number of small (circa I ~m) particles binding antibody and so it is concluded that antibody against mitochondria is also present. Finally absorption with purified synaptosomes removed all binding of antibody to P~. indicating the presence of antibody directed against synapse-specific antigens in the antiserum. Results of similar antibody absorption experiments have lead Mickey et al. (I97 I) to describe an elaborate set of shared antigens, but it seems more likely that this apparent crossreactivity is caused by antibody raised to independant myelin and mitochondrial antigens since both species are inevitably present in small quantities as contaminants of synaptosome preparations. It is also noteworthy that Herschrnan et al. (I97 z) have raised antiserum to synaptosome preparations made in an identical manner to those employed in this study, and found it to react exclusively with synaptosomes and synaptic membrane when assayed by the complement fixation procedure. Attempts to repeat the above absorption experiment with ferritin-labelled IgG isolated from the unabsorbed antiserum were unsatisfactory because the ferritin-conjugated unabsorbed IgG bound to a minor proportion of all the subcellular species in the P2 fraction, and after absorption with mitochondria and myelin exhibited extremely low activity, binding to the postsynaptic lattice of a very few synaptosomes. However, when the anti-syn IgG was absorbed prior to ferritin conjugation, the resulting conjugate showed much enhanced activity binding to about zo% of the postsynaptic sites examined. The possibility of nonspecific interaction being responsible for the observed binding of the ferritin-IgG conjugate was eliminated by control experiments in which ferritin conjugated to non-immune IgG was found not to bind. It is remarkable that of the many polypeptides present in the synaptic membranes (Jones, Matus and Walters, unpublished observations), antibodies raised against synaptosomes or delipidated homogenates of cerebral cortex (Orosz et al., 1973) should be directed only

52

MATUS

against antigen associated with the postsynaptic lattice. Evidently the other components of the exposed synaptic surface are highly conserved, i.e. the immune response to them is at least partially supressed perhaps because of their ubiquitous distribution both in animals supplying antigen and in animals used to raise antisera. The situation regarding potential antigens within the synaptic cleft must be regarded as undecided both because such antigens may be unable due to their location to interact with lymphocyte receptors necessary to stimulate antibody production, or because the antigens are inaccessible to the antibodyferritin conjugate even when such antibody be raised. No definitive conclusion is possible at present regarding the labelling of the surface of dissociated neurons by the anti-syn serum. This labelling being on the surface of the cells does not bear any clear relationship to the ferritin-labelled antibody binding to the postsynaptic lattice - a cytoplasmic component. However, it is noteworthy that there are antigens present on the surface of cells from the cerebral cortex of 7 day old rats which bind antibody against mature synaptosomes and that these antigens survive exposure to trypsin during dissociation of the cells. These antigens are limited to the cell surface where they are distributed in a systematic manner, the same pattern of reticular fluorescence and fluorescent dots appearing on all antibody-binding cells. The binding of antibody is not only specific to the anti-syn serum but the same anti-serum does not bind to the surface of liver cells which have been shown by immunofluorescence methods to possess their own specific surface-bound antigens (Sheffield and Emmelot, 1972). To obtain antisera against other synapse-specific surface components obviously demands more rigorous immunization procedures than those employed in the present study. To this end an immunization programme is in progress with highly purified synaptic membrane preparations available in high yield via a recently developed fractionation procedure (Jones and Matus, 1974). References BAR B ERA, A. J., M ARC H E S v., R. B. a n d R o x H, S. (1973) A d h e s i v e r e c o g n i t i o n a n d retinotectal specificity. Proceedings of the National Academy of Sciences (U.S.) 7o, 2482-86. BIGNAMI, A., ENG, L. Y., DAHL, D. a n d UYEDA, C. T. (1972) Localization of t h e glial fibrillary acidic p r o t e i n i n astrocytes b y i m m u n o f l u o r e s c e n c e . Brain Research 43, 429-35. BLOOM, F. r., WEDNER, H. J. a n d VARKER, C. W. (I973) T h e use of antibodies to s t u d y cell s t r u c t u r e a n d m e t a b o l i s m . Pharmacological Reviews 25, 343-58. COTMAN, C. W. a n d MATTHEWS, D. A. (1971) I s o l a t i o n a n d c h a r a c t e r i z a t i o n o f synaptic p l a s m a m e m b r a n e s f r o m r a t brain. Biochimica et Biophysica Acta 249, 38o-94. DE PETRIS, S. a n d RAFF, M. C. (I972) D i s t r i b u t i o n of I m m u n o g l o b u l i n o n t h e surface o f m o u s e l y m p h o i d cells as d e t e r m i n e d b y i m m u n o f e r r i t i n electron microscopy. Europeanffournal of Immunology

2, 523-35. GRAY, E. 6. and WHITTAKER,V. r. (1972) The isolation of nerve endings from the brain. Journal of Anatomy (London) 96, 79-88. HALL, Z. (1972) Release of neurotransmitters and their interaction with receptors. Annual Review of Biochemistry 4I, 925-52. HARTMAN, B. K. and UDENFRIEND, S. (1972) The application of immunological techniques to the study of enzymes regulating catecholamine synthesis and degradation. Pharmacological Reviews 24, 3II-3O.

Synapse-specific antigen

53

H E R S C H M A N N , H. R.~ COTMAN~ C. W. and M A T T H E W S , D. A. (1972) Serological specificities of brain

subcellular organelles. 1. Antisera to synaptosomal fractions. Journal of Immunology lO8, 1362-9. JONES, D. H. and MATUS, A. I. (1974) Isolation of synaptic plasma membrane from rat brain by combined flotation-sedimentation centrifugation. Bioehimica et Biophysica Acta (Submitted for publication). LIM, R. and Hstr, L. (1971) Studies on brain-specific membrane proteins. Biochimica et Biophysica Aeta 249 , 569-82. M I C K E Y , D. D., M C M I L L A N , P. N.~ A P P E L , S. H. and DAY, E. D. (1971) The specificity and crossreactivity of antisynaptosome antibodies as determined by sequential absorption analysis. Journal of Immunology lO7, 1599-161o. OROSZ, A., H A M O R I , ~., FALUS, A., M A D A R A S Z , E., LAKOS, I. and ADAM, G. (1973) Specifc antibody fragments against the postsynaptic web. Nature New Biology 245, 18-19. PFENNINGER, K. (1973) Synaptic morphology and cytochemistry. Progress in Histochemistry and Cytochemistry 5, No. I. R A I T E R I , M., A N G E L I N I , F., B E R T O L L I N I , A. and F E D E R I C O , R. (1973) Quantitative evaluation of antigen-antibody interactions at the external surface of the nerve ending membrane. Journal of Neurobiology 3, 225-31. SHEFFIELD, I. B. and EMMELOT, V. (1972) Tissue specific antigens in the liver cell surface. Experimental Cell Research 71, 97-1o5.

Immunohistochemical demonstration of antigen associated with the post-synaptic lattice.

The binding of amtosera against synaptosomes from rat cerebral cortex to dissociated cells from both cerebral cortex and liver was assessed by immunof...
1MB Sizes 0 Downloads 0 Views