Localization of the adenomatous polyposis coli tumour suppressor protein in the mouse central nervous system

Pergamon

PII:

Neuroscience Vol. 83, No. 3, pp. 857–866, 1998 Copyright  1998 IBRO. Published by Elsevier Science Ltd Printed in Great Britain. All rights reserved 0306–4522/98 $19.00+0.00 S0306-4522(97)00459-4

LOCALIZATION OF THE ADENOMATOUS POLYPOSIS COLI TUMOUR SUPPRESSOR PROTEIN IN THE MOUSE CENTRAL NERVOUS SYSTEM T. SENDA,*‡ S. IINO,* K. MATSUSHITA,* A. MATSUMINE,† S. KOBAYASHI* and T. AKIYAMA† *Department of Anatomy I, Nagoya University School of Medicine, 65 Tsuruma-cho, Showa-ku, Nagoya, Aichi 466, Japan †Department of Oncogene Research, Research Institute for Microbial Diseases, Osaka University, 3-1 Yamadaoka, Suita, Osaka 565, Japan Abstract––The adenomatous polyposis coli gene is mutated in familial adenomatous polyposis and in sporadic colorectal tumours. The adenomatous polyposis coli gene product is a 300,000 mol. wt cytoplasmic protein that binds to at least three other proteins; â-catenin, a cytoplasmic E-cadherinassociated protein; hDLG, a human homologue of the Drosophila discs large tumour suppressor protein and glycogen synthase kinase 3â, a mammalian homologue of the Drosophila ZESTE WHITE 3 protein. The adenomatous polyposis coli gene is highly expressed in the brain, suggesting that it may be involved in nerve function. Here we show that adenomatous polyposis coli is localized in the pericapillary astrocytic endfeet throughout the mouse central nervous system. Adenomatous polyposis coli is also localized in the astrocytic processes in the cerebellar granular layer, and displays concentrated expression in the terminal plexuses of the basket cell fibres around Purkinje cells. Adenomatous polyposis coli is further expressed in neuronal cell bodies and/or nerve fibres in the olfactory bulb, hippocampus, brain stem, spinal cord and dorsal root ganglia. Adenomatous polyposis coli is demonstrated to be co-localized with â-catenin and/or hDLG in neurons and nerve fibres, but not in astrocytes. From these results, adenomatous polyposis coli is suggested to participate in a signal transduction pathway in astrocytes which is independent of â-catenin and hDLG, and also in regulation of neuronal functions in association with â-catenin and hDLG.  1998 IBRO. Published by Elsevier Science Ltd. Key words: APC, â-catenin, hDLG, astrocyte, signal transduction, immunoelectron microscopy.

The adenomatous polyposis coli (APC) gene is localized to the human chromosome region 5q21–224,10,20 and mutated not only in familial adenomatous polyposis but also in sporadic colorectal tumours.4,8,13,17,30,34 Almost all germline and somatic mutations found to date are nonsense or frame-shift mutations that result in a carboxy-terminal truncation of the APC gene product. Germline APC mutations are known to induce extracolonic disorders. In Gardner’s syndrome, benign soft tissue and bone tumours, dental abnormalities, desmoid tumour and polyps in the upper gastrointestinal tract co-exist.6 Turcot’s syndrome includes familial adenomatous polyposis and brain tumours.44 The APC gene is also considered to be essential for normal development; development of the primitive ectoderm prior to gastrulation failed in mouse embryos carrying homozygous APC mutant alleles.26 The APC gene product is a 300,000 mol. wt cytoplasmic protein composed of 2843 amino acids.8,17,40 ‡To whom correspondence should be addressed. Abbreviations: APC, adenomatous polyposis coli; FITC, fluorescein isothiocyanate; hDLG, human homologue of the Drosophila discs large tumour suppressor protein; PBS, phosphate-buffered saline; PSD, postsynaptic density. 857

Overexpression of this protein blocks cell cycle progression from the G0/G1 to S phase.1 APC was reported to associate with â-catenin, a protein associated with the cell adhesion molecule E-cadherin.36,39,43 Furthermore, we demonstrated that APC is precisely co-localized with both á- and â-catenins in mouse intestinal epithelial cells.25,38 Recently, two other APC-binding proteins has been identified; glycogen synthase kinase 3â, a mammalian homologue of the Drosophila ZESTE WHITE 3 protein37 and hDLG, the human homologue of Drosophila discs large tumour suppressor protein.24 Since both APC and hDLG are highly expressed in the CNS,3,28 APC is assumed to play important roles in higher nervous functions. Thus, in the present study, we examined the cellular and subcellular localization of APC and co-localization of this protein with â-catenin and/or hDLG in the mouse CNS by immunohistochemical and immunoelectronmicroscopic analyses. EXPERIMENTAL PROCEDURES

Antibodies Anti-APC and anti-hDLG antibodies were generated by immunizing rabbits with synthetic peptides corresponding

858

T. Senda et al.

to the COOH-terminal 14 amino acids of APC25 and to the NH2-terminal region of hDLG containing amino acids 6–205,24 respectively. The specificity of both antibodies was checked by immunoprecipitation.24,25 Monoclonal mouse anti-neurofilament 200,000 mol. wt protein antibody and monoclonal mouse anti-â-catenin antibody were purchased from Sigma and Transduction Laboratories, respectively. Fluorescence immunohistochemistry Normal male ddY mice aged eight to 10 weeks (Japan SLC, Inc.) were used in the present study. Under anaesthetization with pentobarbital, the animals were perfused transcardially for 5 min with 4% paraformaldehyde in 0.1 M phosphate buffer (pH 7.4). The brains were removed, and immersed in the same fixative for 2 h. After washing overnight with sucrose solution, the brains were embedded in O.C.T. compound (Miles), and frozen. Frozen sections (7 µm-thick) were incubated with anti-APC antibody diluted 1:50 with phosphate-buffered saline (PBS) at 4C for 24 h, and subsequently with fluorescein isothiocyanate (FITC)-labelled anti-rabbit IgG antibody at room temperature for 1 h. The sections washed in PBS were mounted in glycerol, and examined with a fluorescence microscope (Nikon). For double-immunostaining of APC and neurofilament 200,000 mol. wt protein, frozen sections were incubated with polyclonal rabbit anti-APC antibody diluted 1:50 and monoclonal mouse anti-neurofilament 200,000 mol. wt protein antibody diluted 1:50, and subsequently with Texas Red-labelled anti-rabbit IgG and FITClabelled anti-mouse IgG1. For double-immunostaining of APC and â-catenin, frozen sections were incubated with polyclonal rabbit anti-APC antibody diluted 1:50 and monoclonal mouse anti-â-catenin antibody diluted 1:50, and subsequently with Texas Red-labelled anti-rabbit IgG and FITC-labelled anti-mouse IgG1. Because the anti-APC and hDLG antibodies were both generated in rabbits, simultaneous double-staining on the same sections was technically difficult. Instead, serial frozen sections were immunostained alternately with one antibody then the other, followed by incubation with FITC-labelled antirabbit IgG. Control sections were incubated with normal rabbit serum and normal mouse serum in place of the rabbit polyclonal antibodies and mouse monoclonal antibodies, respectively. In addition, some of the control sections were incubated with anti-APC antibody preabsorbed with the antigen peptide. Avidin–biotin–peroxidase immunohistochemistry Frozen sections of the paraformaldehyde-fixed mouse brain, spinal cord and dorsal root ganglia were incubated with anti-APC antibody diluted 1:500 at 4C for 24 h, and subsequently with biotinylated anti-rabbit IgG antibody at room temperature for 1 h. The immunoreaction was visualized by incubating with avidin–biotin–peroxidase complex for 30 min and 0.01% H2O2-containing diaminobenzidine solution for 10 min. The sections were examined with a light microscope (Nikon). Control sections were incubated with normal rabbit serum instead of anti-APC antibody, or with anti-APC antibody preabsorbed with the antigen peptide. Golgi’s silver staining The mouse cerebellum was fixed for one week with 2.5% potassium dichromate and 4% formaldehyde, and stained for one week with 1% silver nitrate. Then, the cerebellum was dehydrated in ethanol, and embedded in celloidin. The sections (200 µm-thick) were cut with a microtome, and examined with the light microscope. Post-embedding immunoelectron microscopy Mice were perfused intracardially for 5 min with 4% paraformaldehyde and 0.1% glutaraldehyde in 0.1 M

phosphate buffer (pH 7.4). The cerebellum was removed, and immersed in the same fixative for 2 h. Then, the cerebellum was dehydrated in ethanol, and embedded in Lowicryl K4M resin. Ultrathin sections (0.1 µm-thick) were incubated at room temperature for 2 h with antiAPC antibody diluted 1:50, and consequently with 10 nm colloidal gold-conjugated goat anti-rabbit IgG antibody (Amersham) diluted 1:30. The sections were stained with uranyl acetate and lead citrate, and examined with an electron microscope (Hitachi). Control sections were incubated with normal rabbit serum instead of anti-APC antibody. Pre-embedding immunoelectron microscopy Frozen sections of the 4% paraformaldehyde-fixed mouse cerebellum were immunostained with anti-APC antibody, and visualized by incubation with biotinylated goat antirabbit IgG antibody, avidin–biotin–peroxidase complex and diaminobenzidine as described above. The sections were postfixed with 1% osmium tetroxide for 5 min, dehydrated in ethanol, and embedded in Epon resin. Ultrathin sections cut with an ultramicrotome were stained with uranyl acetate and lead citrate, and examined with the electron microscope. Control sections were incubated with normal rabbit serum instead of anti-APC antibody. RESULTS

Immunohistochemical staining using anti-APC antibody revealed immunoreactivity around the blood vessels and just inside the pia mater throughout the mouse CNS (Fig. 1A,B). In the cerebral cortex, no significant immunoreactivity was seen except for that associated with blood vessels and pia mater (Fig. 1A). In the cerebellar cortex, weak diffuse staining in the molecular layer and intense reticular staining in the granular layer were also observed (Fig. 1B). No immunoreactivity was detected in control sections incubated with both normal rabbit serum in place of anti-APC antibody (Fig. 1C,D) and anti-APC antibody preabsorbed with the antigen peptide (Fig. 1E,F). To further clarify the precise localization of APC, we performed immunoelectron microscopy. The granular layer of the cerebellum was most suitable for immunoelectron-microscopic analyses as it exhibited the highest expression of APC. Post-embedding immunoelectron microscopy using a colloidal gold-conjugated second antibody revealed that gold particles were accumulated in the layer between the blood capillary endothelial cells and various neuronal components (neuronal cell bodies, nerve fibres and nerve terminals) and in processlike structures between the neuronal components (Fig. 2A). Thus, APC seemed to be confined to the astrocytic processes and their endfeet surrounding blood capillaries. To confirm this, we employed pre-embedding immunoelectron microscopy. Electron-dense immunoprecipitation was localized exclusively in the pericapillary astrocytic endfeet (Fig. 2C,D) and astrocytic processes between the neuronal components (Fig. 2E), but never in the cell body of astrocytes. Neuronal components, blood capillary endothelial cells and pericytes were not labelled at all in the granular layer of the cerebellum.

APC localization in mouse CNS

Fig. 1. Expression and localization of APC in mouse brain detected by fluorescence immunohistochemistry. (A) In the cerebral cortex, APC immunoreactivity was associated with blood vessels (arrows) and the pia mater (arrowheads). (B) In the cerebellar cortex, in addition to the blood vessel-associated immunoreactivity (arrows), the molecular layer (M) also exhibited weak diffuse staining while the granular layer (G) exhibited intense reticular staining. (C–F) No immunoreactivity was seen in control sections of the cerebral (C,E) and cerebellar (D,F) cortices incubated with normal rabbit surum (C,D) or preabsorbed anti-APC antibody (E,F). Scale bar=20 µm.

859

860

T. Senda et al.

Fig. 2. Immunoelectron-microscopic localization of APC in mouse cerebellar cortex by post- (A,B) and pre- (C–F) embedding techniques. (A,C–E) APC was localized in the pericapillary astrocytic endfeet (arrows in A,C,D) and in the astrocytic processes between neuronal components (arrowheads in C,E). (B,F) No immunostaining was seen in control sections. L, capillary lumen; PE, pericyte; NT, nerve terminal; G, granular cell. Scale bars=100 nm.

APC localization in mouse CNS

861

Fig. 3. Localization of APC in the terminal plexus of basket cell fibres around Purkinje cells. (A,B) APC (A) and neurofilament 200,000 mol. wt protein (B) were doubly immunostained on the same sections. Most of the APC immunoreactivity around Purkinje cells overlapped with that of the neurofilament 200,000 mol. wt protein (arrows). (C) Basket cell fibres surrounding a Purkinje cell (P) revealed by Golgi’s silver staining. As compared with the profile by Golgi staining, monoclonal anti-neurofilament 200,000 mol. wt protein antibody apparently stains the basket cell fibres around Purkinje cells (arrows in B). (D) Post-embedding immunoelectron-microscopic demonstration of the APC localization. Gold particles were localized in basket cell fibres and their terminals near a Purkinje cell (P). Scale bar=15 µm (A–C), 200 nm (D).

In control sections, no immunolabelling was seen by both kinds of immunoelectron-microscopic techniques (Fig. 2B,F). APC in pericapillary astrocytic endfeet was generally localized throughout the mouse CNS while that in astrocytic processes was confined to the granular layer of the cerebellum. Immunohistochemical staining revealed intense immunoreactivity around the Purkinje cells (Figs 1B, 3A), suggesting that APC is localized to some structural components other than the astrocytic processes and endfeet. As revealed by Golgi’s silver staining, a Purkinje cell body is surrounded by a terminal plexus of basket cell fibres (Fig. 3C). We next carried out double-labelling immunohistochemical staining using anti-APC antibody and anti-neurofilament 200,000

mol. wt protein antibody; the latter strongly stained the basket cell fibres (Fig. 3B). Most of the APC immunoreactivity around the Purkinje cells overlapped with that of the neurofilament 200,000 mol. wt protein (Fig. 3A,B). Furthermore, immunoelectron microscopic analysis revealed significant labelling of APC in nerve fibres and terminals around the Purkinje cells (Fig. 3D). APC was also expressed in neurons of some nervous tissues. Pyramidal cells in the hippocampus were weakly stained (Fig. 4A). In the pons and spinal cord gray matter, neuronal cell bodies and nerve fibers exhibited strong immunoreactivity; for example, APC-positive neurons in the pontine reticular nucleus (Fig. 4B) and APC-positive motor

862

T. Senda et al.

Fig. 4. Localization of APC in neurons in the mouse hippocampus (A), pons (B), spinal cord (C) and dorsal root ganglion (D). Arrows indicate APC-immunopositive neurons; pyramidal cells of the hippocampus (A), neurons in the pontine reticular nucleus (B), motor neurons in the anterior horn of the spinal cord (C) and dorsal root ganglion cells (D). Blood vessel-associated APC immunoreactivity (arrowheads) was also seen in the hippocampus (A), pons (B) and spinal cord (C), but not in the dorsal root ganglion (D), where there exists no astrocytes. Both neuronal cell bodies and nerve fibres were immunopositive in the pons (B) and spinal cord (C) while only cell bodies were positive in the hippocampus (A) and dorsal root ganglion (D). The cytoplasm was diffusely stained in these APC-positive neurons. Scale bar=10 µm.

neurons in the spinal anterior horn (Fig. 4C) are shown. Dorsal root ganglion neurons exhibited high expression of APC, while their nerve fibres did not (Fig. 4D). Neuronal cell bodies and nerve fibres were stained with equal intensity in the olfactory bulb (data not shown). In each APC protein-expressing neuron, the cytoplasm was stained diffusely, while there was little staining in the nucleus. Though the hippocampus, pons, spinal cord and olfactory bulb exhibited also blood vessel-associated APC immunoreactivity (Fig. 4A–C), the dorsal root ganglion did

not (Fig. 4D). Most likely, this is because the dorsal root ganglion, included in the peripheral nervous system, contains no astrocyte. No immunoreactivity was seen in both kinds of the control experiments (data not shown). We next investigated by double-labelling immunohistochemistry whether APC co-localizes with â-catenin and hDLG in the mouse CNS. We found that APC immunoreactivity around the Purkinje cells overlapped completely with that of â-catenin (Fig. 5A,B). However, around the blood vessels APC

APC localization in mouse CNS

863

Fig. 5. Co-localization of APC (A,C) with â-catenin (B) and/or hDLG (D) in neurons but not in astroglial endfeet. (A,B) APC (A) and â-catenin (B) were doubly immunostained on the same sections of the cerebellar cortex. APC and â-catenin co-localized in the terminal plexus of basket cell fibres around the Purkinje cells (arrows). APC was also associated with blood vessels (arrowheads in A), but â-catenin was not. (C,D) APC (C) and hDLG (D) were immunostained independently on serial sections of the pons. Co-localization of APC with hDLG was observed in neurons in the pons (arrows). Blood vessel-associated immunoreactivity was seen only in APC-stained sections (arrowheads in C). Scale bar=15 µm.

staining was frequently observed (Fig. 5A), whereas â-catenin staining was negative (Fig. 5B). Quite similar results were observed in the pons (data not shown). Some of APC-positive neurons in the pons also showed hDLG immunoreactivity, whereas no hDLG staining was observed around blood vessels where APC staining was positive (Fig. 5C,D). These results suggest that APC is co-localized with â-catenin and/or hDLG in neurons, but not in astrocytic endfeet and processes.

DISCUSSION

Involvement of adenomatous polyposis coli in the signal transduction system in astrocytes One of the major findings in this study is that APC is localized in the perivascular astrocytic endfeet in

the mouse CNS. The topology of astrocytes in the nervous tissue and their spatial relationship with other neural components are important when considering the roles of APC. Astrocytes, one of the glial cells, are distributed throughout the CNS, and their cytoplasmic processes extend along neuronal and glial elements, such as neuronal cell bodies, nerve fibres, nerve terminals, glial cell bodies and their processes, finally reaching the blood capillary endothelial cells and pia mater. Thus, the astrocytic processes and endfeet come into contact with all nervous tissue components. Furthermore, all the blood capillaries in the CNS are covered by sheaths of the astrocytic endfeet. The product of the APC gene is a 300,000 mol. wt cytoplasmic protein, and expressed ubiquitously in various tissues, especially in the CNS and colon at higher levels in adult animals.3,25,38,40

864

T. Senda et al.

Overexpression of APC blocks cell cycle progression, suggesting its involvement in the regulation of cellular proliferation.1 The binding of APC to â-catenin36,39,43 further suggests that APC may participate in cell adhesion and in the transmission of cell-to-cell signals leading to cell cycle regulation. It has recently been reported that GSK3â binds to APC-â-catenin complex and regulates its assembly.37 Furthermore, we have recently found that hDLG is also associated with APC-â-catenin complex.24 In intestinal epithelial cells, APC is actually co-localized with á- and â-catenins and hDLG.24,25,38 From these investigations, APC is believed to be involved in the signalling pathway regulating cell proliferation.32 The present study showed that APC is localized in astrocytic processes and endfeet, but, unlike in neurons and intestinal epithelial cells, not associated with â-catenin and hDLG in astrocytes. This suggests the possible existence of an APC proteindependent, but â-catenin- and hDLG-independent, signal transduction system in astrocytes. Since astrocytic processes and endfeet are in contact with all nervous tissue components as described above, they can in theory monitor all signals transmitted through the nervous tissue. Thus, this APC proteindependent signal transduction system in astrocytes may respond to these extrinsic signals and regulate the proliferation of astrocytes. In this sense, it is of interest that astrocytes regulate the concentration of potassium ions released from neurons, and take up neurotransmitters released from neuronal synapses.16,45 High surface area provided by multiple astrocytic processes is convenient for these functions of astrocytes. The finding that neuronal contact induces the elongation of astrocytic processes suggests that some signals from neurons are transmitted to astrocytes resulting in the morphological changes of astrocytes.9,22 Adjacent astrocytes and their endfeet are combined by the gap junctions and adherens junctions, resulting in the formation of an astrocytic functional syncytium.23 The gap junctions are also seen between astrocytes and oligodendrocytes.27 Thus, glial cells are likely to function synchronously and cooperatively under intercellular signals transmitted through these cell-to-cell junctional complexes. As revealed in intestinal epithelial cells, APC tends to localize to the cell-to-cell contact sites with junctional complexes.25,38 It is therefore suggested that the astrocytic processes and endfeet act as the sites for cell adhesion and cell-to-cell interactions, and that APC localized to these sites is involved in transduction of the intercellular signals. Recently, attention has been focused on the signal transduction system in glial cells, and our knowledge of the neurotransmitter receptors and transporters and ion channels expressed in glial cells has been growing. Glutamate is the major excitatory neurotransmitter, and its resting extracellular concentration is kept low due to the action of sodium-

dependent glutamate transporters.2,12,14,21,29 Two kinds of glutamate transporters, GLT-133 and GLAST,42 were reported to be localized to the plasma membranes of astrocytes, especially of astrocytic processes and endfeet.19,35 This can be interpreted as that glutamate is a signal transmitter from neurons to astrocytes as well as to neurons. Indeed, glutamate is known to induce calcium waves in cultured astrocytes, suggesting involvement of astrocytes in long-range signalling within the brain.5 In addition, we are interested in the striking similarity of localization between APC and these two glutamate transporters. The functional relationship between APC and these glial transporter proteins is an intriguing problem to be investigated. Involvement of adenomatous polyposis coli in neuronal functions In the present study, neurons and nerve fibres in some nervous tissues exhibited the expression of APC, while cerebral cortex neurons did not express APC at all. In the cerebellum, though nerve fibres and terminals in the granular layer were negative for APC, those in the molecular layer were weakly positive, and terminal plexuses of the basket cell fibres around Purkinje cells showed the concentrated APC localization. We have also shown that APC is localized in the cell body and synaptic sites in cultured hippocampal neurons.24 Thus, APC expression and subcellular localization in neurons is variable depending on their topology in the brain. Co-localization of APC with â-catenin and hDLG in the neuronal components was demonstrated in the present study. Like intestinal epithelial cells, neurons seem to have a signal transduction system which depends on APC, â-catenin and hDLG. Interestingly, APC was found to be concentrated in the terminal plexuses of the basket cell fibres around Purkinje cells. The presynaptic processes within this terminal plexus are known to have specialized membrane junctions, the morphology of which is very similar to the septate junctions in insects.7,31,41 This kind of intercellular junction is not found in any other regions of the mammalian CNS. The concentrated localization of APC in the terminal plexuses of the basket cell fibres may be related to these specialized intercellular junctions. The postsynaptic density protein PSD-95, which is very closely related to hDLG and also concentrated to the terminal plexuses of the basket cell fibres,11 was shown to interact and co-localize with the COOHterminal portion of the N-methyl--aspartate receptor subunits.18 Subunits of the voltage-gated K+ channel were also demonstrated to associate with PSD-95 and hDLG, and co-localization of the K+ channel and PSD-95 was found in the terminal plexuses of the basket cell fibres.15 These interactions are mediated by the DLG homology repeat domains of PSD-95 and hDLG. The interaction between APC

APC localization in mouse CNS

and hDLG also requires the DLG homology repeat domains.24 These facts suggest that APC, N-methyl-aspartate receptor and K+ channel compete with one another for binding to PSD-95/hDLG, and are thus involved in the synaptic transmission between basket cell fibres and Purkinje cells. In addition, â-catenin, which was found in the present study to be co-localized with APC in these terminal plexuses, may also be involved in synaptic transmission. CONCLUSIONS

In the present study, we found that APC is localized in the pericapillary astrocytic endfeet throughout the mouse CNS, and in the astrocytic processes in the cerebellar granular layer. APC also displayed concentrated expression in the terminal

865

plexuses of the basket cell fibres around Purkinje cells. APC was further expressed in neuronal cell bodies and/or nerve fibres in the olfactory bulb, hippocampus, brain stem, spinal cord and dorsal root ganglia. We demonstrated that APC was co-localized with â-catenin and/or hDLG in neurons and nerve fibres, but not in astrocytes. From these results, it is suggested that APC may participate in a signal transduction pathway in astrocytes which is independent of â-catenin and hDLG, and also in regulation of neuronal functions in association with â-catenin and hDLG.

Acknowledgements—This study was supported by grants from Nagoya University Foundation and Aichi Cancer Research Promotion Society.

REFERENCES

1. 2. 3. 4. 5. 6. 7. 8.

9. 10. 11. 12. 13.

14. 15. 16. 17.

18. 19.

Baeg G. H., Matsumine A., Kuroda T., Bhattacharjee R. N., Miyashiro I., Toyoshima K. and Akiyama T. (1995) The tumour suppressor gene product APC blocks cell cycle progression from G0/G1 to S phase. Eur. molec. Biol. Org. J. 14, 5618–5625. Balcar V. J. and Johnston G. A. R. (1972) The structural specificity of the high affinity uptake of L-glutamate and L-aspartate by rat brain slices. J. Neurochem. 19, 2657–2666. Bhat R. V., Baraban J. M., Johnson R. C., Eipper B. A. and Mains R. E. (1994) High levels of expression of the tumor suppressor gene APC during development of the rat central nervous system. J. Neurosci. 14, 3059–3071. Bodmer W. F., Baily C. J., Bodmer J., Bussey H. J. R., Ellis A., Gorman P., Lucibello F. C., Murday V. A., Rider S. H., Scambler P., Sheer D., Solomon E. and Spurr N. K. (1987) Localization of the gene for familial adenomatus polyposis on chromosome 5. Nature 328, 614–616. Cornell-Bell A. H., Finkbeiner S. M., Cooper M. S. and Smith S. J. (1990) Glutamate induces calcium waves in cultured astrocytes: long-range glial signaling. Science 247, 470–473. Gardner E. J. and Richards R. C. (1953) Multiple cutaneous and subcutaneous lesions occuring simultaneously with hereditary polyposis and osteomatosis. Am. J. hum. Genet. 5, 139–147. Gobel S. (1971) Axo-axonic septate junctions in the basket formations of the cat cerebellar cortex. J. Cell Biol. 51, 328–333. Groden J., Thliveris A., Samowitz W., Carlson M., Gelbert L., Albertsen H., Joslyn G., Stevens J., Spirio L., Robertson M., Sargeant L., Krapcho K., Wolff E., Burt R., Hughes J. P., Warrington J., McPherson J., Wasmuth J., Le Paslier D., Abderrahim H., Cohen D., Leppert M. and White R. (1991) Identification and characterization of the familial adenomatous polyposis coli gene. Cell 66, 589–600. Hatten M. E. (1985) Neuronal regulation of astroglial morphology and proliferation in vitro. J. Cell Biol. 100, 384–396. Herrera L., Kakati S., Gibas L., Pietrzak E. and Sandberg A. (1986) Brief clinical report: Gardner syndrome in a man with an interstitial deletion of 5q. Am. J. med. Genet. 25, 473–476. Hunt C. A., Schenker L. J. and Kennedy M. B. (1996) PSD-95 is associated with the postsynaptic density and not with the presynaptic membrane at forebrain synapses. J. Neurosci. 16, 1380–1388. Johnston G. A. R. (1981) Glutamate uptake and its possible role in neurotransmitter inactivation. In Glutamate: Transmitter in the Central Nervous System (eds Roberts P. J., Storm-Mathisen J. and Johnston G. A. R.), pp. 77–87. Wiley, Chichester. Joslyn G., Carlson M., Thliveris A., Albertsen H., Gelbert L., Samowitz W., Groden J., Stevens J., Spirio L., Robertson M., Sargeant L., Krapcho K., Wolff E., Burt R., Hughes J. P., Warrington J., McPherson J., Wasmuth J., Le Paslier D., Abderrahim H., Cohen D., Leppert M. and White R. (1991) Identification of deletion mutations and three new genes at the familial polyposis locus. Cell 66, 601–613. Kanner B. I. and Schuldiner S. (1987) Mechanism of transport and storage of neurotransmitters. CRC Crit. Rev. Biochem. 22, 1–38. Kim E., Niethammer M., Rothschild A., Jan Y. N. and Sheng M. (1995) Clustering of Shaker-type K+ channels by interaction with a family of membrane-associated guanylate kinases. Nature 378, 85–88. Kimelberg H. K. and Norenberg M. D. (1989) Astrocytes. Scient. Am. 260, 66–76. Kinzler K. W., Nilbert M. C., Su L. K., Vogelstein B., Bryan T. M., Levy D. B., Smith K. J., Preisinger A. C., Hedge P., McKechnie D., Finniear R., Markham A., Groffen J., Boguski M. S., Altschul S. F., Horii A., Ando H., Miyoshi Y., Miki Y., Nishisho I. and Nakamura Y. (1991) Identification of FAP locus genes from chromosome 5q21. Science 253, 661–665. Kornau H. C., Schenker L. T., Kennedy M. B. and Seeburg P. H. (1995) Domain interaction between NMDA receptor subunits and the postsynaptic density protein PSD-95. Science 269, 1737–1740. Lehre K. P., Levy L. M., Ottersen O. P., Storm-Mathisen J. and Danbolt N. C. (1995) Differential expression of two glial glutamate transporters in the rat brain: quantitative and immunocytochemical observations. J. Neurosci. 15, 1835–1853.

866 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30.

31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42. 43. 44. 45.

T. Senda et al. Leppert M., Dobbs M., Scambler P., O’Connell P., Nakamura Y., Stauffer D., Woodward S., Burt R., Hughes J., Gardner E., Lathrop M., Wasmuth J., Lalouel J. M. and White R. (1987) The gene for familial polyposis coli maps to the long arm of chromosome 5. Science 238, 1411–1413. Logan W. J. and Snyder S. H. (1972) High affinity uptake systems for glycine, glutamic and aspartic acids in synaptosomes of rat central nervous tissues. Brain Res. 42, 413–431. Mason C. A., Edmondson J. C. and Hatten M. E. (1988) The extending astroglial process: development of glial shape, the growing tip, and interactions with neurons. J. Neurosci. 8, 3124–3134. Massa P. T. and Mugnaini E. (1982) Cell junctions and intramembrane particles of astrocytes and oligodendrocytes: a freeze-fracture study. Neuroscience 7, 523–538. Matsumine A., Ogai A., Senda T., Okumura N., Satoh K., Baeg G. H., Kawahara T., Kobayashi S., Okada M., Toyoshima K. and Akiyama T. (1996) Binding of APC to the human homolog of the Drosophila discs large tumor suppressor protein. Science 272, 1020–1023. Miyashiro I., Senda T., Matsumine A., Baeg G. H., Kuroda T., Shimano T., Miura S., Noda T., Kobayashi S., Monden M., Toyoshima K. and Akiyama T. (1995) Subcellular localization of the APC protein: immunoelectron microscopic study of the association of the APC protein with catenin. Oncogene 11, 89–96. Moser A. R., Shoemaker A. R., Connelly C. S., Clipson L., Gould K. A., Luongo C., Dove W. F., Siggers P. H. and Gardner R. L. (1995) Homozygosity for the Min allele of Apc results in disruption of mouse development prior to gastrulation. Devl Dynam. 203, 422–433. Mugnaini E. (1986) Cell junctions of astrocytes, ependyma, and related cells in the mammalian central nervous system, with emphasis on the hypothesis of a generalized functional syncytium of supporting cells. In Astrocytes (eds Fedoroff I. S. and Vernadakis A.), pp. 329–371. Academic, Orlando. Mu¨ller B. M., Kistner U., Veh R. W., Cases-Langhoff C., Becker B., Gundelfinger E. D. and Garner C. C. (1995) Molecular characterization and spatial distribution of SAP97, a novel presynaptic protein homologous to SAP90 and the Drosophila discs-large tumor suppressor protein. J. Neurosci. 15, 2354–2366. Nicholls D. and Attwell D. (1990) The release and uptake of excitatory amino acids. Trends pharmac. Sci. 11, 462–468. Nishisho I., Nakamura Y., Miyoshi Y., Miki Y., Ando H., Horii A., Koyama K., Utsunomiya J., Baba S., Hedge P., Markham A., Krush A. J., Petersen G., Hamilton S. R., Nilbert M. C., Levy D. B., Bryan T. M., Preisinger A. C., Smith K. J., Su L. K., Kinzler K. W. and Vogelstein B. (1991) Mutations of chromosome 5q21 genes in FAP and colorectal cancer patients. Science 253, 665–669. Palay S. L. and Chan-Palay V. (1974) The basket cell. In Cerebellar Cortex, pp. 180–216. Springer, New York. Peifer M. (1996) Regulating cell proliferation: as easy as APC. Science 272, 974–975. Pines G., Danbolt N. C., Bjøra˚s M., Zhang Y., Bendahan A., Eide L., Koepsell H., Seeberg E., Storm-Mathisen J. and Kanner B. I. (1992) Cloning and expression of a rat brain L-glutamate transporter. Nature 360, 464–467. Powell S. M., Zilz N., Beazer-Barclay Y., Bryan T. M., Hamilton S. R., Thibodeau S. N., Vogelstein B. and Kinzler K. W. (1992) APC mutation occur early during colorectal tumorigenesis. Nature 359, 235–237. Rothstein J. D., Martin L., Levey A. I., Dykes-Hoberg M., Jin L., Wu D., Nash N. and Kuncl R. W. (1994) Localization of neuronal and glial glutamate transporters. Neuron 13, 713–725. Rubinfeld B., Souza B., Albert I., Muller O., Chamberlain S. H., Masiarz F. R., Munemitsu S. and Polakis P. (1993) Association of the APC gene product with â-catenin. Science 262, 1731–1734. Rubinfeld B., Albert I., Porfiri E., Fiol C., Munemitsu S. and Polakis P. (1996) Binding of GSK3â to the APC-â-catenin complex and regulation of complex assembly. Science 272, 1023–1026. Senda T., Miyashiro I., Matsumine A., Baeg G. H., Monden T., Kobayashi S., Monden M., Toyoshima K. and Akiyama T. (1996) The tumor suppressor protein APC colocalizes with â-catenin in the colon epithelial cells. Biochem. biophys. Res. Commun. 223, 329–334. Shibata T., Gotoh M., Ochiai A. and Hirohashi S. (1994) Association of plakoglobin with APC, a tumor suppressor gene product, and its regulation by tyrosine phosphorylation. Biochem. biophys. Res. Commun. 203, 519–522. Smith K. J., Johnson K. A., Bryan T. M., Hill D. E., Markowitz S., Willson J. K. V., Paraskeva C., Petersen G. M., Hamilton S. R., Vogelstein B. and Kinzler K. W. (1993) The APC gene product in normal and tumor cells. Proc. natn. Acad. Sci. U.S.A. 90, 2846–2850. Sotelo C. and Llina´s R. (1972) Specialized membrane junctions between neurons in the vertebrate cerebellar cortex. J. Cell Biol. 53, 271–289. Storck T., Schulte S., Hofmann K. and Stoffel W. (1992) Structure, expression, and functional analysis of a Na+-dependent glutamate/aspartate transporter from rat brain. Proc. natn. Acad. Sci. U.S.A. 89, 10,955–10,959. Su L. K., Vogelstein B. and Kinzler K. W. (1993) Association of the APC tumor suppressor protein with catenins. Science 262, 1734–1737. Turcot J., Despres J. P. and St Pierre F. (1959) Malignant tumors of the central nervous system associated with familial polyposis of the colon: report of two cases. Dis. Colon Rectum 2, 465–468. Walz W. (1989) Role of glial cells in the regulation of the brain microenvironment. Prog. Neurobiol. 33, 309–333. (Accepted 30 July 1997)