Molecular Brain Research

MOLECULAR BRAIN RESEARCH ELSEVIER

Molecular Brain Research 24 (1994) 107-117

Research Report

De novo synthesis of GAP'43" in situ hybridization histochemistry and light and electron microscopy immunocytochemical studies in regenerating motor neurons of cranial nerve nuclei in the rat brain G. Palacios a,,, G. Mengod b, M. Sarasa c, j. Baudier a, J.M. Palacios b a Department of CellularBiology and Physiology, Subunit of Histology, Faculty of Medicine, UniversidadAutdnoma de Barcelona, Bellaterra, Barcelona 08193, Spain, b Department of Neurochemistry, CID-CSIC, Barcelona, Spain, c Department of Anatomy, Embryology and Genetics, Veterinary Faculty, Universityof Zaragoza, Zaragoza, Spain, a Laboratoire de Biologie Moleculaire, Cycle Cellulaire, INSERM U309, 38041 Grenoble, France

(Accepted 28 December 1993)

Abstract In order to investigate the modulation of the synthesis and the subcellular localization of the growth associated protein GAP-43 in neuronal cell bodies we have taken advantage of the well known regenerative properties of axotomized motor neurons of the facial and hypoglossal nuclei. Alterations in the levels of GAP-43 mRNA containing cells were studied by in situ hybridization histochemistry. The protein localization was examined using immunohistochemistry at the light and electron microscopic levels. Neurons from the control side showed undetectable levels of both GAP-43-1ike immunoreactivity and GAP-43 mRNA levels. Whereas axotomized neurons exhibited a marked increase in GAP-43 mRNA levels and in GAP-43-1ike immunoreactivity. Three to 50 days after axotomy, motor neurons ipsilateral to the lesion displayed a dense reticular or filamentous perinuclear distribution of the imrnunoreactivity in somata and proximal dendritic processes, corresponding to the location of the Golgi apparatus in these neurons. At the electron microscopic level the immunoreactivity was located in the cisternae of the Golgi complex and found to be associated with trans-side vesicles of these complexes. The myelinated fibers of the transectomized facial nerve also presented an intense GAP-43-1ike immunoreactivity. Twenty-one days after the axotomy a decay in the number of immunostained neurons and in the intensity of immunolabeled somata was observed. Our study reveals a rapid induction of GAP-43 mRNA and protein after axotomy. The localization of the newly synthesized GAP-43-1ike immunoreactivity to the Golgi apparatus seen in the present work suggests an early association of this protein with newly formed membranes prior to transport toward the terminals through the axons. Key words: GAP-43; Rat brain; Immunocytochemistry; mRNA; Axotomy; Facial nucleus

1. Introduction Growth-associated protein GAP-43 (B-50, F1, pp. 46 and neuromodulin) is a 24 kDa neuron specific phosphoprotein that has been implicated in the axogenesis and synaptogenesis during the development and regeneration of the nervous system [3,4,7,9,14,25,43,45,48, 49,51,53]. The levels of GAP-43 are increased during the postnatal period in the rat brain [4,14,22,40,41,53], whereas in the adult brain its concentration declines rapidly following the establishment of final synaptogen-

* Corresponding author. 0169-328X/94/$07.00 © 1994 Elsevier Science B.V. All rights reserved SSDI 0169-328X(93)00006-Z

esis [2,3,14,20,25,26,44,48,50]. However, the protein and its m R N A persist in moderated levels in certain locations of the mature CNS [5,6,12,14,16,36,42,43] and this has been correlated with a potential role of GAP-43 in synaptic remodeling and functional plasticity of these areas throughout life [4,14,53]. GAP-43 is a protein synthesized in the neuronal somata, rapidly transported throughout axons and localized in growth cones and synapses [15,27,39,52]. Immunoelectron microscopic studies have demonstrated that the localization of GAP-43 is predominantly within axon terminals, small unmyelinated and thin myelinated axons, as well as in dendritic spines [17,59,60]. Following axotomy of peripheral nerves (sci-

108

G. Palacios et aL / Molecular Brain Research 24 (1994) 107-117

atic and facial nerves) dramatic increases in GAP-43 expression have been described in axotomized neurons [4,8,13,24,29,45,50,53,56,57,63,64]. This increase in both synthesis and transport of GAP-43 has been correlated with the ability of axons to regenerate. Immunocytochemical studies of the axotomized neurons have also demonstrated that GAP-43 is associated with unmyeiinated and myelinated regenerating axons and with extraaxonal elements of peripheral nerves such as Schwann cells [11,56,66]. In two recent immunocytochemical studies [47,55] accumulation of GAP-43 has been described in neuronal somata of dorsal root ganglion after peripheral nerve crush or resection. In view of the evidence that GAP-43 may be associated with regenerative response, especially in extrinsic neurons of the nervous system, we have examined the distribution of this protein by light and electron microscopy in axotomized facial nucleus using well characterized monoclonal and polyclonal antibodies. To obtain a more complete picture we have also determined the location of cells containing mRNA coding for it by in situ hybridization histochemistry using radiolabeled oligonucleotides as probes.

32p]dATP (3000 C i / m m o l , NEN) and terminal deoxynucleotidyltransferase (Boehringer Manheim) to specific activities of 019-210× 104 C i / m m o l . Labeled probes were purified by chromatography through a NACS P R E P A C column (BRL).

2.2.2. Tissue preparation Tissue sections from lesioned animals, sacrificed at several postoperation times, were all hybridized at the same time. At least two sections per level and animal were used. However, not in all cases was it possible to obtain exactly the same level of the hypoglossal nucleus (see Fig. 2). Frozen tissue sections were thawed to room temperature, air dried, and fixed by immersion for 20 min in 4% paraformaldehyde in PBS (2.6 m M KCI/1.4 m M K H 2 P O 4 / 8 m M N a 2 H P O 4 / 1 3 6 m M NaCI), washed once in 3 × P B S and twice in 1 ×PBS, 5 min each. They were incubated in a freshly prepared solution of pre-digested pronase at a final concentration of 24 U / m l in 50 m M Tris-HCI pH 7.5, 5 m M E D T A for 10 min. T h e proteolytic activity was stopped by immersion in a solution of 2 m g / m l g!ycine in

2. Materials and methods 2.1. Animals, surgery and tissue processing Adult, male Sprague-Dawley rats were used. The animals were deeply anaesthetized with sodium pentobarbital (60 m g / k g , i.p.). In one group of animals the right hypoglossal nerve was transected near the hypogh)ssal canal, close to the origin of the descending branch to the hypoglossal ansa. Animals were killed by decapitation 6, 14 h and 1, 2, 4, 7, 15, 30 and 60 days after the nerve transection and the brains were quickly removed, frozen on dry ice and stored at - 2 0 ° C until used for in situ hybridization histochemistry experiments. Coronal sections (20 /zm thick) from the brain stem were obtained in a microtome cryostat (Leitz 1720, Wetzlar, FRG), thaw-mounted onto gelatin-coated slides, and kept at - 2 0 ° C until use. In another group of animals the right facial nerve was cut near its exit from the stylomastoid foramen. At 3, 5, 7, 14 and 21 days following nerve transection the animals were either killed and their brains frozen at - 2 0 ° C for in situ hybridization experiments or were perfused transcardially u n d e r ether anaesthesia with 50-100 ml of 0.9% saline followed by 4% paraformaldehyde, 0.1% glutaraldehyde and 15% saturated picric acid in 0.1 M p h o s p h a t e buffer (PB) pH 7.4. Brains were removed and further fixed for 4 h at 4°C with 4% paraformaldehyde in 0.1M PB, and then were stored overnight in 5% sucrose in 0.1M PB at 4°C. Coronal sections (40 ~ m ) were obtained with a vibratome (Lancer) at the nucleus facialis level. T h e sections were then washed for 12-20h in 5% sucrose in 0.1 M PB at 4°C.

2.2. In situ hybridization procedures 2.2.1. Probes Oligonucleotide probes were synthesized on a 380B Applied Biosystems D N A synthesizer and purified on a 20% a c r y l a m i d e / 8 M urea preparative sequencing gel. T h e oligomers were complementary to bases 300-350 ( g a p 4 3 / I ) and 550-600 (gap43/II) of the rat GAP-43 c D N A [28]. They were labeled at their 3' end with [a-

B

d

Fig. 1. Expression of GAP-43 m R N A after 15 days of facial nerve transection. A: Toluidine-stained section. B: photomicrograph of an autoradiogram of a rat section hybridized with a 32P-labeled oligonucleotide complementary to the rat GAP-43 m R N A . Dark areas correspond to regions rich in hybridization signal. T h e granule cells of the cerebellum show an intense hybridization signal. The arrows point the lesioned nucleus facialis. The arrowheads point to the ipsilateral side. Observe the dramatic increase in the hybridization signal of the motor neurons of the nucleus facialis in the lesioned neurons. Bar -- 2 mm.

G. Palacios et aL / Molecular Brain Research 24 (1994) 107-117 PBS during 30 s. T h e slides were then rinsed twice in PBS for 30 s, and dehydrated in a graded series of ethanol (60-100%).

2.2.3. Hybridization Labeled probes were diluted to a final concentration of 0.4-0.8 p m o l / m l in 50% formamide, 600 m M NaCl, 10 m M Tris-HCl p H 7.5, 1 m M E D T A , 1 × D e n h a r d t ' s solution (0.02% Ficoll, 0.02% polyvinylpyrrolidone, 0.02% bovine serum albumin) and 500 /zg/,~l

109

yeast tRNA. Tissues were covered with 60 p.l of hybridization solution, overlaid with nescofilm coverslips, and hybridized for 17 h in a humid c h a m b e r at 42°C. T h e sections were then washed at 50°C in 600 m M NaCI, 10 m M Tris-HCl pH 7,5, 1 m M E D T A for 4 h with 4 changes of buffer. Tissues were dehydrated and either apposed to /3-max film ( A m e r s h a m ) or dipped into A m e r s h a m LM-1 nuclear track emulsion. Films were developed after 4 days and emulsion after 7 days in Kodak D-19. Tissues were stained with Giemsa,

Fig. 2. Effect of the transection of the hypoglossal nerve on the levels of GAP-43 m R N A at different times after lesion. Arrows point o u t the nucleus of the lesioned nerve. Pictures are photomicrographs from film autoradiograms generated from tissue sections hybridized in the s a m e experiment. The sections do not correspond to exactly the same level of the hypoglossal nuclei. Note the fast increase in the levels of G A P - 4 3 m R N A (hybridization signal can be observed already at 14 h) and the return to basal levels after survival times of 60 days. Bar = 2 m m .

110

G. Palacios et aL /Molecular Brain Research 24 (1994) 107-117

dehydrated and mounted with Eukitt. Sections were examined using light- and dark-field microscopy (Leitz, Orthoplan). The specificity of the oligonucleotide probes was assessed by different criteria (Sarasa et al., in preparation). By Northern blot analysis on total RNA extracted from several regions of the rat brain, a single mRNA species of approximately 1400 nucleotides was detected with a regional distribution similar to the one observed in the in situ hybridization experiments. Two oligonucleotides complementary to different regions of the GAP-43 mRNA were used independently as hybridization probes in consecutive sections from the same animal and showed an identical pattern of hybridization. Cohybridization of a labeled oligonucleotide with a 20-fold excess of the same unlabeled probe resulted in the complete abolishment of the hybridization signal. The signals were not affected when the oligonucleotide added in excess was complementary to a different region of the same mRNA. The thermal stability of the hybrids was determined by washing at increasing temperatures: a sharp decrease in the hybridization signal was observed at a temperature consistent with the theoretical value of melting temperature (Tm) [38] of the hybrids formed.

2.3. Immunocytochemical procedures

2.3.2. lmmuno-light and -electron microscopy Tissue sections were first incubated with either GAP-43 monoclonal antibody (diluted 1:1000) or GAP-43 polyclonal antibody (diluted 1:400) for 17 h, rinsed and incubated with the reagents of the avidin-biotin-peroxidase kits (mouse Vectastain ABC and rabbit Vectastain ABC kits respectively, Vector Laboratories) using recommended dilutions. Peroxidase activity was finally visualized by incubating the sections with 0.05% 3,3'-diaminobenzidine and 0.01% H 2 0 2 in PB for 5-10 min. Some sections were processed with 1-naphtol solution for 2 min [35]. For electron microscopic observations, blocks of the facial nucleus obtained with punches of the monoclonal antibody treated material were postfixed in 1% OsO 4 for 1 h, block-stained in 1% uranyl acetate veronal buffer, dehydrated and flat-embedded in Durcupan (Fluka). Semithin sections (1 ~m) were cut on an LKB Ultratome III and stained with 1% Toluidine Blue for light-microscope photography. Ultrathin sections were also obtained and stained with alcoholic uranyl acetate for I0 min or with uranyl acetate and lead citrate for 5 min. They were later examined and photographed with a Hitachi H-7000 electron microscope.

3. Results

2.3.1. Antibodies GAP-43 monoclonal antibody was obtained from Boehringer Manheim (Germany) and the GAP-43 polyclonal antibody was obtained by immunizing rabbits with GAP-43 purified from bovine brain. The antibody production, purification and characterization of the latter antibody have been described previously [1,34]. The avidin-biotin-peroxidase pre-embedding technique was used for light and electron microscopic localization of both GAP-43 antibodies.

3.1. In situ hybridization histochemistry The

oligonucleotide probes

u s e d in t h e s e e x p e r i -

ments, revealed a distribution of GAP-43 mRNA t h r o u g h o u t t h e r a t b r a i n w h i c h is in c o m p l e t e a g r e e m e n t w i t h p r e v i o u s p u b l i c a t i o n s [9,31]. H i g h l e v e l s o f

Fig. 3. Cellular localization of GAP-43 mRNA in the hylgoglossal nucleus 2 days after the nerve transection. The autoradiographic image is presented in A as a dark-field photomicrograph from emulsion-dipped tissue section in which autoradiographic grains are seen as bright points. Only lesioned neurons express GAP-43 mRNA. The arrows point out the same neuron in both fields. B is a bright-field image of the section shown in A. Bar = 200/zm.

G. Palacios et al. / Molecular Brain Research 24 (1994) 107-117

h y b r i d i z a t i o n signal w e r e s e e n in c o r t e x , h i p p o c a m p u s , some thalamic and hypothalamic nuclei, locus coeruleus, dorsal and medial raph6, granule cells of the

111

c e r e b e l l u m , a n d n u c l e i o f m e d u l l a o b l o n g a t a ( S a r a s a et al, m a n u s c r i p t in p r e p a r a t i o n ) . F u r t h e r m o r e , c o n t r o l experiments (see Materials and methods section)

Fig. 4. Light microscopic immunoreactivity in facial nuclei using a polyclonal antibody to GAP-43. A,B: 3 days following axotomy. C,D: 2l days following axotomy. A: vibratome section of the ipsilateral facial nucleus showing a intense immunoreactivity in the somata and proximal dendrites of motor neurons (arrows). B: the motor neurons of the contralateral facial nucleus are devoid of immunolabeling (arrows). C: axotomized motor neurons at higher magnification showing a perinuclear dense and reticular distribution of the immunoreactivity (arrows). Other neurons exhibited a reduced intensity of the immunolabeling (arrow heads). Immunostained axons and fiber bundles can be seen in the neuropil. Note the ausence of terminals on the plasmalemma of axotomized neurons. D: contralateral facial motor neurons are devoid of immunoreactivity. Note the immunostained axons and fiber bundles in the neuropil. Immunolabeling boton terminals can be observed in contact with the plasmalemma of motor neurons (small arrows). Bar = 100 ~zm in A, B; 20/xm in C, D.

G. Palacios et al. /Molecular Brain Research 24 (1994) 107-117

112

A

G. Palacios et aL /Molecular Brain Research 24 (1994) 107-117

proved the specificity of the hybridization signal obtained with the labeled oligonucleotide probe (data not shown). In unoperated, sham operated and in the side contralateral to the lesioned side, GAP-43 mRNA was undetectable in both facial and hypoglossal nerve nuclei. Fifteen days after unilateral transection of the facial nucleus, the concentration of GAP-43 mRNA in the lesioned motor neurons increased in a dramatic manner (Fig. 1B) when compared to the basal levels seen in the contralateral unlesioned side. As early as 14 h after transection, GAP-43 mRNA can be detected by in situ hybridization (not shown). Unilateral axotomy of the hypoglossal nerve resulted also in a prominent increase in the levels of GAP-43 mRNA when compared to the contralateral unlesioned side (Fig. 2). This rise was already evident 14 h after the lesion (Fig. 2) The highest levels were observed from 48 h onwards, beginning to decline by 30 days onwards, presumably when reinnervation is taking place, to reach control levels 60 days after the lesion.

3.2. Cellular localization of GAP-43 mRNA Nuclear emulsion autoradiography shows that the labeled hybrids are located exclusively in the lesioned motor neurons of the facial (not shown) and hypoglossal (Fig. 3) nuclei 2 days after the nerve transection.

113

of immunolabeling in some of them was, however, reduced 21 days after nerve transection (Fig. 4C). At higher magnification the staining pattern in the cell bodies of the axotomized neurons differed slightly with the two antibodies used. With the polyclonal antibody, the cytoplasm exhibited a dense homogeneous or reticular distribution of the immunoreactive product surrounding the unstained nucleus (Fig. 4C). The monoclonal antibody revealed thin filamentous patches of the immunoreactive product distributed in a perinuclear manner, in agreement with the well-known distribution of the Golgi apparatus in these motor neurons (Fig. 5C). The neuropil in the contralateral facial nucleus showed immunostained nerve processes and nerve terminals. Most of these synaptic terminals were observed covering the perikarya and main dendrites of motor neurons (Figs. 4D and 5D). On the transected side, the motor neurons in the facial nucleus appeared free of these synaptic contacts probably due to the fact that the surface area of motor neurons was occupied by microglial activated cells which displaced these synaptic boutons (synaptic stripping) (Figs. 4C and 5C). The facial nerve in the ipsilateral side showed an intense immunoreactivity in regenerating myelinated fibers (Fig. 5A). In contrast, the contralateral facial nerve appeared unstained. (Fig. 5B).

3.3. Immunocytochemistry: light microscopy

3.4. Immunocytochemistry: electron microscopy

The two antibodies used in the present study displayed a high degree of immunoreactive sensitivity. This immunoreactivity was predominantly restricted to the neuropil and axon bundles of the brainstem zone and facial nuclei while the neuronal somata were unstained (Figs. 4B,D and 5D). The immunostaining signal in the neuropil was stronger when the monoclonal antibody was used (Fig. 5C,D). Three days after unilateral transection of the right facial nerve, the motor neurons and their proximal dendritic processes of the contralateral facial nucleus appeared unstained and surrounded by a homogeneous dense labeling of the neuropil (Fig. 4B). In contrast, an increase in GAP-43-1ike immunoreactivity in motor neurons was seen in the ipsilateral facial nucleus (Fig. 4A). In the following days examined (5-15 days) the vast majority of the motor neurons in the ipsilateral side showed GAP-43-1ike immunoreactivity in their somata. The number of labeled cells and the intensity

The results of electron microscopic immunocytochemistry were in accordance with previous light microscopic studies of semithin sections (Fig. 6A,B). These sections revealed a perinuclear pattern of dense patches restricted to motor neurons in the ipsilateral side (Fig. 6A). The electron microscopic study showed that the immunoreactivity was associated with the dispersed Golgi complexes (rete dispersion) only in the cytoplasm of axotomized neurons (Fig. 6C). These injured motor neurons were covered by activated microglial cells (Fig. 6C). At higher magnification, the immunoreaction product was scattered between the cisternae stacks throughout the Golgi complex (Fig. 6D,E). Some lightly immunoreactive vesicles were also seen in the trans-side of the Golgi complex (Fig. 6E). Labeling of other cellular organelles in these injured neurons was not observed. Attempts to further localize GAP-43 using immunogold techniques, which has been mainly used in cultured neurons, failed in our tissues.

Fig. 5. Light microscopic immunoreactivity in facial nuclei using a monoclonal antibody to GAP-43, 7 days following axotomy. A: vibratome section of the ipsilateral facial nerve exhibiting a intense immunoreactivity in myelinated fibers comparing with the absence of immunoreactivity in contralateral facial nerve shown in B. C: axotomized motor neurons shows perinuclear filamentous immunoreactive patches in the citoplasm corresponding to the Golgi apparatus localization (small arrows). D: contralateral facial motor neurons are devoid of immunoreactivity (arrows). Note the dense immunoreaction product in the neuropil and terminals surrounding these neurons. Bar = 40/zm in A, B; 25/zm in C, D.

G. Palacios et al./Molecular Brain Research 24 (1994) 107-117

L4

,J N

? -

C

!

G. Palacios et al. /Molecular Brain Research 24 (1994) 107-117

4. Discussion

The present study shows that a rapid increase in GAP-43 mRNA levels and accumulation of GAP-43 in motor neurons cell bodies and their proximal axon segments occur after nerve axotomy. This initial accumulation remained at a high level until 3 to 4 weeks after lesion, when it began to decline. By immunoelectron microscopy we have been able to show that motor neurons in the lesioned side exhibited considerable GAP-43-1ike immunoreactivity in the cell bodies and that the newly synthesized protein was localized in the Golgi apparatus. While increases in GAP-43 mRNA in spinal motor neurons [9,32] and in the nucleus facialis [46] have been already shown, this is the first study in hypoglossal neurons. Furthermore, we show a quite early increase, demonstrable as short as at 14 h after lesion. The fact that both GAP-43 mRNA and protein are below detection levels in the resting 'normal' neurons and increase dramatically after axotomy is consistent with an induction of the de novo synthesis and transport of the protein throughout injured axons, events that can be modulated by the well-known regenerative potential of the facial and hypoglossal axotomized neurons [30]. Facial neurons and other PNS neurons have the capacity to regenerate while the CNS neurons do not. This has been correlated with the incapacity of these CNS neurons to express regeneration-associated proteins like GAP-43 [53,57]. However, in some central neurons such as rubrospinal and in retinal ganglion neurons, GAP-43 mRNA can be expressed after axotomy [18,20,33,57]. GAP-43 has been proposed as a protein normally disposed on the cytoplasmic surface of axonal membranes [3,21,37,58]. The binding of this protein to the vesicular elements in the Golgi apparatus can facilitate the transport to the distal axon segments and their attachment to the cytoplasmic side of membranes. In this sense, other studies have demonstrated a subcellular localization of GAP-43 in vesicle and plasma membranes [17]. GAP-43 has been ultrastructurally localized in unmyelinated and thinly myelinated axons of the adult pyramidal tract [21]. In the rat neostriatum, GAP-43 was distributed in discrete patches throughout fine

115

caliber unmyelinated axons and in axospinous asymmetric synapses [17]. GAP-43 immunoelectron reactivity has been also found in the neostriatal neuropil distributed in cytoplasmic patches in dendrites, dendritic protrusions and thin spines [17]. Recent light microscopic immunocytochemical studies showed accumulation of GAP-43 in the cell bodies of dorsal root ganglion neurons (DRG) after sciatic nerve crush or resection [47,55]. 100% of L 4 and L 5 DRG neurons expressed and accumulated detectable amounts of GAP-43 in their cell bodies 1 week after sciatic nerve injury [47]. Previous studies [23], have suggested that GAP-43 is concentrated in the Golgi apparatus during the developmental stages of hippocampal neurons, by comparing the immunocytochemical distribution of this protein with that of wheat germ agglutinin, a lectin that selectively labels this organelle. Meiri and coworkers [40] have also noted a perinuclear concentration of GAP-43 in cultured sympathetic neurons compatible with its localization to the Golgi apparatus. The majority of electron microscopy studies with GAP-43 have focused on the location of this protein in growth cones, axons and terminals in the adult animal [17,21,58,65]. These studies have not provided evidence as to where in the cell body this protein could be synthesized. Biochemical studies have proposed that GAP-43 is actively synthesized as a soluble protein which rapidly associates with membranes [10,54]. As GAP-43 is not a glycoprotein, fatty acylation has been proposed as the mechanism for its membrane attachment [10,54]. More recently, GAP-43 immunoreactivity has been located using immunogold probes, on the cytosolic face of electron-lucent putative transport vesicles in the trans region of the Golgi apparatus, in neurites and in growth cones of cultured hippocampal neurons [61]. Quantification of the density of GAP-43 on the plasma membrane of hippocampal pyramidal neurons demonstrated no differences between growth cones and neuritic processes, which may indicate that in in vitro conditions, GAP-43 does not play a selective role in growth cones [62]. The results in this work show that in vivo and in neurons of the rat cranial nerve, with a well known regenerative potential, GAP-43 is concentrated in cell bodies, indicating that newly synthesized GAP-43 could

Fig. 6. A,B: light microscopic immunoreactivity in semithin sections of facial nuclei using a monoclonal antibody to GAP-43, 3 days after axotomy. A: injured motor neurons display immunoreactive product dispersed in discrete patches in a perinuclear position (small arrows) B: contralateral motor neurons appeared with unlabeled cytoplasm. In both sides the fibers in the neuropil exhibited intense inmunoreactivity. C,D: electron microscopic immunoreactivity in thin sections of the same injured neurons stained with alcoholic uranyl acetate. C: axotomized motor neuron showing dispersed immunoreactive Golgi complex in the cytoplasm (arrows). Note the unlabeled nucleus (N), nucleolus (No) and microglial activated cell (M) attached to neuron plasmalemma. D: immunoreactive Golgi complex in injured motor neurons (arrows). The reaction product appears between stack cisternae (arrows). Note an activated microglial cell (M). E: axotomized motor neuron stained with uranyl acetate and lead citrate showing immunostained round vesicles located in the trans side of the Golgi complex (arrows). Bars = 20/xm in A, B; 1/zm in C; 0.5/xm in D, E.

116

G. Pa lacios et al. / Molecular Brain Research 24 (1994) 107-117

be associated with Golgi sacules and vesicles prior to transport throughout the axons to synaptic terminals. The localization of GAP-43 to the Golgi apparatus, the organelle involved in the formation of a variety of transport vesicles and in the traffic of those vesicles from the Golgi to the plasma membrane [19] suggests an important role of the protein in axonal regeneration.

References [1] Baudier, J., Bronner, C., Kligman, D. and Cole, R.D., Protein kinase C substrates from bovine brain. Purification and characterization of neuromodulin, a neuron-specific calmodulin-binding protein, J. BioL Chem., 264 (1989) 1824-1828. [2] Benowitz, L.I., Shashona, V.E. and Yoon, M., Specific changes in rapidly transported proteins during regeneration of the goldfish optic nerve, J. Neurosci., 1 (1981) 300-307. [3] Benowitz, L.I. and Lewis, E.R., Increased transport of 44,000-to 49,000- dalton acidic proteins during regeneration of the goldfish optic nerve: A two-dimensional gel analysis, J. Neurosci., 3 (1983) 2153-2163. [4] Benowitz, L.I. and Routtenberg, A., A membrane phosphoprorein associated with neural development, axonal regeneration, phospholipid metabolism and synaptic plasticity. Trends Neurosci., 10 (1987) 527-531. [5] Benowitz, L.I., Apostolides, P.J., Perrone-Bizzozero, N.I., Finklestein, S.P. and Zwiers, H., Anatomical distribution of the growth-associated protein GAP-43/B-50 in the adult rat brain, J. Neurosci., 8 (1988) 339-352. [6] Benowitz, L.I., Perrone-Bizzozero, N.I., Finklestein, S.P. and Bird, E.D., Localization of the growth-associated phosphoprotein GAP-43 in the human cerebral cortex, J. Neurosci., 9 (1989) 990-995. [7] Benowitz, L.I., Rodriguez, W.R. and Neve, R.L., The pattern of GAP-43 immunostaining changes in the rat hippocampal formation during reactive synaptogenesis, Mol. Brain Res., 8 (1990) 17-23. [8] Bisby, M.A., Dependence of GAP-43 (B-50, F1) transport on axonal regeneration in rat dorsal root ganglion neurons, Brain Res., 458 (1988) 157-161. [9] Chong, M.S., Fitzgerald, M., Winter, J., Hu-Tsai, M., Emson, P.C., Wiese, U. and Woolf, C.J., GAP-43 mRNA in rat spinal cord and dorsal root ganglia neurons: developmental changes and re-expression following peripheral nerve injury, Eur. J. Neurosci., 4 (1992) 883-895. [10] Coggins, P.J. and Zwiers, H., B-50 (GAP-43): biochemistry and functional neurochemistry of a neuron-specific phosphoprotein. J. Neurochem., 56 (1991) 1095-1106. [11] Curtis, R., Stewart, H.J.S., Hall, S.M., Wilkin, G.P., Mirsky, R. and Jessen, K.R., GAP-43 is expressed by nonmyelin-forming Schwann cells of the peripheral nervous system, J. Cell. BioL, 116 (1992) 1455-1464. [12] Curtis, R, Averill, S., Priestley, J.V. and Wilkin, G.P., The distribution of GAP-43 in normal rat spinal cord, J. NeurocytoL, 22 (1993) 39-50. [13] Curtis, R., Green, D., Linssay, R.M. and Wilkin, G.P., Up-regulation of GAP-43 and growth of axons in rat spinal cord after compression injury, J. Neurocytol., 22 (1993) 51-64. [14] Dani, J.W., Armstrong, D.M. and Benowitz, L.I., Mapping the development of the rat brain by GAP-43 immunocytochemistry. Neuroscience, 40 (1991) 277-287.

[15] De Graan, P.N.E., Van Hooff, C.O.M., Tilly, B.C., Oestreicher, A.B., Schotman, P. and Gispen, W.H., Phosphoprotein B-50 in nerve growth cones from fetal rat brain, Neurosci. Lett., 61 (1985) 235-241. [16] De la Monte, S.M., Federoff, H.J., Shi-Chung, Ng., Grabczyk, E.D. and Fishman, M.C., GAP-43 gene expression during development: persistence in a distinctive set of neurons in the mature central nervous system, Dev. Brain Res., 46 (1989) 161-168. [17] Di Figlia, M., Roberts, R.C. and Benowitz, L.I., Immunoreactive GAP-43 in the neuropil of adult rat neostriatum: localization in unmyelinated fibers, axon terminals, and dendritic spines, J. Comp. Neurol., 302 (1990) 992-1001. [18] Doster, S.K., Lozano, A.M., Aguayo, A3. and Willard, M.B., Expression of the growth-associated protein GAP-43 in adult rat retinal ganglion cells following axon injury, Neuron, 6 (1991) 635 -647. [19] Farquhar, M.G. and Palade, G.E., The Golgi apparatus (complex)-(1954-1981)-from artifact to center stage, Z Cell Biol., 91 (1981) 77s-103s. [20] Freeman, J.A., Bock, S., Deaton, M., McGuire, B., Norden, J.J. and Snipes, G.J., Axonal and glial proteins associated with development and response to injury in the rat and goldfish optic nerve, Exp. Brain Res. (SuppL), 13 (1986) 34-47. [21] Gorgels, T.G.M.F., Van Lookeren Campagne, M., Oestreicher, A.B., Gribnau, A.A.M. and Gispen, W.H., B-50/GAP-43 is localized at the cytoplasmic side of the plasma membrane in developing and adult rat pyramidal tract, J. Neurosci., 9 (1989) 3861-3869. [22] Goslin, K., Schreyer, D.J., Skene, J.H.P. and Banker, G., Development of neuronal polarity: GAP-43 distinguishes axonal from dendritic growth cones, Nature, 336 (1988) 672-674. [23] Goslin, K., Schreyer, D.J., Skene, J.H.P. and Banker, G., Changes in the distribution of GAP-43 during the development of neuronal polarity, J. Neurosci., 10 (1990) 588-602. [24] Hoffman, P.N., Expression of GAP-43, a rapidly transported growth-associated protein, and class II beta tubulin, a slowly transported cytoskeletal protein, are coordinated in regenerating neurons, J. Neurosci., 9 (1989) 893-897. [25] Jacobson, R.D., Virag, I. and Skene, J.H.P., A protein associated with axon growth, GAP-43, is widely distributed and developmentally regulated in rat CNS, Z Neurosci., 6 (1986) 18431845. [26] Kalil, K. and Skene, J.H.P., Elevated synthesis of an axonally transported protein correlates with axon outgrowth in normal and injured pyramidal tracts, J. Neurosci., 6 (1986) 2563-2570. [27] Katz, F., Ellis, L. and Pfenninger, K.H., Nerve growth cones isolated from fetal rat brain. III. Calcium-dependent protein phosphorylation, J. Neurosci., 5 (1985) 1402-1411. [28] Karns, L.R., Ng, S.-C., Freeman, J.A. and Fishman, M.C., Cloning of complementary DNA for GAP-43, a neuronal growth-related protein, Science, 236 (1987) 597-600. [29] Knyih~ir-Csillik, E., Csillik, B. and Oestreicher, A.B., Light and electron microscopic localization of B-50 (GAP43) in the rat spinal cord during transganglionic degenerative atrophy and regeneration, J. Neurosci. Res., 32 (1992) 93-109. [30] Kreutzberg, G.W., Graeber, M.B. and Streit, W.J., Neuron-glial relationship during regeneration of motor neurons. Metabolic Brain Dis'., 4 (1989) 81-85. [31] Kruger, L., Bendotti, C., Rivolta, R. and Samanin, R. Distribution of GAP-43 mRNA in the adult rat brain, J. Cornp. Neurol., 333 (1993) 417-434. [32] Lind,, H., Piehl, F., Dagerlind, ,~, Verge, V,M.K., Arvidsson, U., Cullheim, S., Risling, M., Ulfhake, B. and H6kfelt, T., Expression of GAP-43 mRNA in the adult mammalian spinal cord under normal conditions and after different types of lesions, with special reference to motoneurons, Exp. Brain Res., 91 (1992) 284-295.

G. Palacios et al. /Molecular Brain Research 24 (1994) 107-117 [33] Lozano, A.M., Doster, S.K., Aguayo, A.J. and Willard, M.B., Immunoreactivity to GAP-43 in axotomized and regenerating retinal ganglion cells of adult rats, Soc. Neurosci. Abstr., 13 (1987) 1389. [34] Masliah, E., Mallory, M., Hansen, L., Alford, M., Albright, T., De Teresa, R., Terry, R,, Baudier, J. and Saitoh, T., Patterns of aberrant sprouting in Alzheimer's disease, Neuron, 6 (1991) 729-739. [35] Mauro, A., Germano, 1., Giaccone, G., Giordana, M.T. and Schiffer, D., 1-Naphtol basic dye (1-NBD). An alternative to diaminobenzidine (DAB) in immunoperoxidase techniques, Histochemistry, 83 (1985) 97-102. [36] Mc Guire, C.B., Snipes, G.J. and Norden, J.J., Light-microscopic immunolocalization of the growth-associated protein GAP-43 in the developing brain, Dec. Brain Res., 41 (1988) 277-291. [37] Mc Intosh, H., Parkinson, D., Meiri, K., Daw, N. and Willard, M., A GAP-43-1ike protein in cat visual cortex, l/'ts. Neurosci., 2 (1989) 583-591. [38] Meinkoth; J. and Wahl, G., Hybridization of nucleic acids immobilized on solid supports, Anal Biochem., 138 (1984) 267284. [39] Meiri, K.F., Pfenninger, K.H. and Willard, M.B., Growth-associated protein, GAP-43, a polypeptide that is induced when neurons extend axons, is a component of growth cones and corresponds to pp. 46, a major polypeptide of a subcellular fraction enriched in growth cones, Proc. Natl. Acad. Sci. USA, 83 (1986) 3537-3541. [40] Meiri, K.F., Willard, M. and Johnson, M.I., Distribution and phosphorylation of the growth-associated protein GAP-43 in regenerating sympathetic neurons in culture, J. Neurosci., 8 (1988) 2571-2581. [41] Moya, K,L., Jhaveri, S., Schneider, G.E. and Benowitz, L.I., Immunohistochemical localization of GAP-43 in the developing hamster retinofugal pathway, J. Comp. Neurol., 288 (1989) 51-58. [42] Neve, R.L., Finch, E.A., Bird, E.D. and Benowitz, L.I., The growth-associated protein GAP-43 (B-50, F1) is expressed selectively in associative regions of the adult human brain, Proc. Natl. Acad. Sci. USA, 85 (1988) 3638-3642. [43] Oestreicher, A.B. and Gispen, W.H., Comparison of the immunocytochemical distribution of the phosphoprotein B-50 in the cerebellum and hippocampus of immature and adult rat brain, Brain Res., 375 (1986) 267-279. [44] Perry, G.W., Burmeister, D.W. and Grafstein, B., Fast axonally transported proteins in regenerating goldfish optic nerve, J. Neurosci., 7 (1987) 792-806. [45] Redshaw, J.D. and Bisby, M.A., Proteins of fast axonal transport in the regenerating hypoglossal nerve of the rat, Can. J. Physiol. PharmacoL, 62 (1984) 1387-1393. [46] Saika, T., Kiyama, H., Tohyama, M. and Matsunaga, T., GAP-43 mRNA expression in facial motoneurons during regeneration: in situ hybridization histochemistry using an alkaline phosphataselabelled probe, Acta Otolaryngol. Suppl., 501 (1993) 80-84. [47] Schreyer, D.J. and Skene, J.H.P., Fate of GAP-43 in ascending spinal axons of DRG neurons after peripheral nerve injury: delayed accumulation and correlation with regenerative potential, J. Neurosci., 11 (1991) 3738-3751. [48] Skene, J.H.P. and Willard, M., Changes in axonally transported proteins during axon regeneration in toad retinal ganglion cells, J. Cell Biol., 89 (1981) 86-96. [49] Skene, J.H.P. and Willard, M., Characteristics of growth-associated polypeptides in regenerating toad retinal ganglion cell axons, J. Neurosci., 1 (1981) 419-426. [50] Skene, J.H.P. and WiUard, M., Axonally transported proteins associated with growth in rabbit central and peripheral nervous system, Z Cell Biol., 89 (1981) 96-103.

117

[51] Skene, J.H.P, Growth-associated proteins and the curious dichotomes of nerve regeneration, Cell, 37 (1984) 697-700. [52] Skene, J.H.P., Jacobson, R.D., Snipes, G.J., McGuire, C.B., Norden, J.J. and Freeman, J.A., A protein induced during nerve growth (GAP-43) is a major component of growth cone membranes, Science, 233 (1986) 783-786. [53] Skene, J.H.P., Axonal growth-associated proteins. Annu. Reu. Neurosci., 12 (1989) 127-156. [54] Skene, J.H.P. and Virag, I., Post-translational membrane attachment and dynamic fatty acylation of a neuronal growth cone protein, GAP-43, J. Cell Biol., 108 (1989) 613-625. [55] Sommervaille, T., Reynolds, M.L. and Woolf, C.J., Time-dependent differences in the increase in GAP-43 expression in dorsal root ganglion cells after peripheral axotomy, Neuroscience, 45 (1991) 213-220. [56] Tetzlaff, W., Zwiers, H., Lederis, K., Cassar, L. and Bisby, M.A., Axonal transport and localization of B-50/GAP-43-1ike immunoreactivity in regenerating sciatic and facial nerves of the rat, Z Neurosci., 9 (1989) 1303-1313. [57] Tetzlaff, W., Alexander, S.W., Miller, F.D. and Bisby, M.A., Response of facial and rubrospinal neurons to axotomy: changes in mRNA expression for cytoskeletal proteins and GAP-43, J. Neurosci., 11 (1991) 2528-2544. [58] Van Lookeren Campagne, M., Oestreicher, A.B., Van Bergen En Henegouwen, P.M.P. and Gispen, W.H., Ultrastructural immunocytochemical localization of B-50/GAP-43, a protein kinase C substrate, in isolated presynaptic nerve terminals and neuronal growth cones, J. Neurocytol., 18 (1989) 479-489. [59] Van Lookeren Campagne, M., Oestreicher, A.B., Van Bergen En Henegouwen, P.M.P. and Gispen, W.H., Ultrastructural double localization of B-50/GAP43 and synaptophysin (p. 38) in the neonatal and adult rat hippocampus, J. Neurocytol., 19 (1990) 948-961. [60] Van Lookeren Campagne, M., Oestreicher, A.B., Buma, P., Verkleij, A.J, and Gispen, W.H., Ultrastructural localization of adrenocorticotrophic hormone and the phosphoprotein B50/Growth-Associated Protein 43 in freeze-substituted, lowicryl HM20-enbedded mesencephalic central gray substance of the rat, Neuroscience, 42 (1991) 517-529. [61] Van Lookeren Campagne, M., Dotti, C,G., Verkleij, A.J., Gispen, W.H. and Oestreicher, A.B., B-50/GAP-43 localization on membranes of putative transport vesicles in the cell body, neurites and growth cones of cultures hippocampal neurons, Neurosci Lett., 137 (1992) 129-132. [62] Van Lookeren Campagne, M., Dotti, C.G., Jap Tjoen San, E.R.A., Verkleij, A.J., Gispen, W.H. and Oestreicher, A.B., B-50/GAP-43 localization in polarized hippocampal neurons in vitro: an ultrastructural quantitative study, Neuroscience, 50 (1992) 35-52. [63] Van der Zee, C.E.E.M., Nielander, H.B., Vos, J.P., Lopes da Silva, S., Verhaagen, J., Oestreicher, A., Schrama, L.H., Schotman, P. and Gispen, W.H., Expression of growth-associated protein B-50 (GAP-43) in dorsal root ganglia and sciatic nerve during regenerative sprouting, J. Neurosci., 9 (1989) 3505-3512. [64] Verge, V.M.K., Tetzlaff, W., Richardson, P.M. and Bisby, M.A., Correlation between GAP-43 and nerve growth factor receptors in rat sensory neurons, J. Neurosci., 10 (1990) 926-936. [65] Verhaagen, J., Oestreicher, A.B., Edwards, P.M., Veldman, H., Jennekens, F.G.I. and Gispen, W.H., Light- and electron-microscopical study of phosphoprotein B-50 following denervation and reinnervation of the rat soleus muscle, J. Neurosci., 8 (1988) 1759-1766. [66] Woolf, C.J., Reynolds, M.L., Chong, M.S., Emson, P., Irwin, N. and Benowitz, L.I., Denervation of the motor endplate results in the rapid expression by terminal Schwann cells of the growth-associated protein GAP-43, J. Neurosci., 12 (1992) 3999-4010.