Differential localization of the high- and low-molecular weight variants of MAP2 in the developing retina

Developmental Brain Research, 38 (1988) 313-318 Elsevier

313

BRD60249

Differential localization of the high- and low-molecular weight variants of MAP2 in the developing retina Richard P. Tucker 1, Lester I. B i n d e r 2 and A n d r e w I. Matus 1 IFriedrich Miescher-lnstitut, Basel (Switzerland) and 2Department of Cell Biology, University of Alabama, Birmingham, A L 35294 (U. S.A.) (Accepted 27 October 1987) Key words: Microtubule-associated protein; MAP2; MAP2c; Retina; Development; Neuronal differentiation; Immunohistochemistry

Microtuhule-associated protein 2 (MAP2) occurs in developing mammalian neuronal tissue as both high (280 kDa)- and low (70 kDa)-molecular weight forms with temporally regulated expression. We have studied the developing avian retina with a monoclonal antibody that recognizes both the high- and low-molecular weight forms of MAP2 and a second monoclonal antibody that recognizes only high-molecular weight MAP2. The developmentally regulated, low-molecular weight protein, MAP2c, has a more widespread distribution in the embryonic avian retina than high-molecular weight MAP2. Our results suggest that MAP2c is the first form of MAP2 to appear in differentiated embryonic retinal neurons, and that the high-molecular weight isoforms of MAP2 appear only later when they may confer stability to neuronal processes.

Brain microtubule-associated proteins (MAPs) are a complex group of proteins that are isolated from brain homogenates together with tubulin following repeated cycles of temperature-dependent microtubule polymerization and depolymerization 17"18. Because of their distribution in adult brain, as determined by immunocytochemistry l'6'TAl't9,22, and changes in abundance and form during development, as determined by immunoblotting 2'5"21"22, the MAPs have been implicated in several roles in brain morphogenesis and function. For example, MAP2, the most plentiful M A P in adult mammalian brain, is much more abundant in cell bodies and dendrites than in axons 6'7'11, and during embryonic and early postnatal periods a 70 kDa form, MAP2c, is found 9"21. MAP2c is scarce in adult brain homogenates, where two high-molecular weight (280 kDa) isoforms, MAP2a and MAP2b, predominate. Recently, we have shown that mammalian MAP2c is highly homologous to a stretch of high-molecular weight MAP2 that includes the tubulin-binding domain 9. This similarity has made it difficult to obtain

monoclonal antibodies to MAP2c that do not crossreact with MAP2a and MAP2b, making the immunohistochemical characterization of MAP2c with MAP2c-specific antibodies impossible. In this communication, we report patterns of immunostaining in the embryonic avian retina using two well-characterized monoclonal antibodies to MAP2. Monoclonal antibody C (MAb/C) recognizes both the low- and high-molecular weight forms of MAP2 (refs. 11, 21), whereas monoclonal antibody AP14 (refs. 4, 6) recognizes an epitope found only on the larger isoforms. The differences in the staining patterns of these antibodies correspond to the pattern of MAP2c, allowing us to localize this protein in the absence of a MAP2cspecific antibody. Both monoclonal antibodies, which were raised against mammalian proteins, recognize epitopes that are conserved in the quail. This permits us to use the avian retina, with its large, welldefined neurons and synaptic layers, as a histological system to study the distribution of the developmentally regulated forms of MAP2. We have localized MAP2c in both embryonic and juventile retinas: the

Correspondence: R.P. Tucker, Friedrich Miescher-Institut, P.O. Box 2543, CH-4002 Basel, Switzerland. 0165-3806/88/$03.50 © 1988 Elsevier Science Publishers B.V, (Biomedical Division)

314 former tissue is a b u n d a n t in MAP2c and the staining pattern with AP14 is, correspondingly, a limited subset of the M A b / C staining pattern; in the latter tissue,

AB C D

which contains relatively less MAP2c, the staining

t

patterns with the two antibodies are more similar.

o

The eyes from Japanese quail (Coturnix coturnirt embryos (embryonic day 14, E l 4 ) or juveniles (11)

x

days after hatching, P10) were removed and fixed for 4 h in cold 4% paraformaldehyde in (/. 1 M potassium phosphate buffer. The eyes were then rinsed, crvoprotected in 25% sucrose, and e m b e d d e d and frozen

---200

in O C T c o m p o u n d (Miles). Serial frozen sections (25 urn) were collected on gelatin-coated slides, airdried, and rinsed in Tris/EDTA buffer (50 mM "I'risHC1, 0.1 mm E D T A , pH 8.0). Alternate sections were treated with 100/~g/ml bacterial alkaline phosphatase (Type I l l - R , Sigma) or 40 U of calf intestinal alkaline phosphatase (Boehringer) in T r i s / E D T A buffer with l mM phenylmethyl sulfonylfluoride (PMSF) for 5 h at 37 °C or in T r i s / E D T A / P M S F buffer alone u n d e r the same conditions ca. Identical resuits were obtained with both enzymes. After enzyme or control treatment, sections were blocked in 0.5% bovine serum albumin (BSA) in PBS, and incubated overnight in diluted monoclonal antibody in the form of hybridoma culture supernatants. The production and characterization of monoclonal antibodies tau-1 (refs. 3, 4), AP14 (refs. 4, 6). and M A b / C ll have been detailed elsewhere. After incubation in primary antibody, the slides were rinsed thoroughly in PBS and stained with rhodammelabelled rabbit anti-mouse secondary antibody (Dakopatts) for 2 h. Sections incubated in BSA/PBS instead of primary antibody and then treated similarly were completely unstained. Stained sections were

" - 116 "--95 i

"-66

"--45

Fig. 1. Immunoblots of El4 quail brain microtubules, Nitrocellulose strips were cut from the same lane of transferred proteins after separation with SDS-PAGE. l,ane A: control nitrocellulose strip incubated with secondary antibody alone. Lane B: strip stained with tau-I recognizes 4 maior bands with an average apparent molecular weight of 60,(10(I. I aae C: strip stained with MAb/C stains both high (M, -2f)(I.tX)0)- and low (M~ -65,000)-molecuhlr weight torms ol MAP2. Lane D: strip stained with API4 recognizes only lhc high-molecular weight form of MAP2.

Fig. 2. Photomicrographs of stained serial frozen sections of an El4 quail retina. Sections incubated in secondary antibody alone were completely unstained. A,B: sections (the same field shown in C,D) counterstained with the fluorescent Hoechst nuclear dye show the different layers of the El4 retina, gcl, ganglion cell layer; ipl, inner plexiform layer; acl, amacrine cell layer; bcl, bipolar cell layer; opl, outer plexiform layer; pcl, photosensitive cell layer; pet, periocular connective tissue. The optic fiber layer is just internal to the ganglion cell layer. C: tau-1 stains the ganglion cell layer faintly in the El4 retina. D: after digesting the section in alkaline phosphatase, tau-1 stains the ganglion cell layer intensely as well as a single lamina in the external portion of the inner plexiform layer. The section shown in C was stained, photographed, and printed using identical conditions as the enzyme-treated section in D. E: MAb/C stains ganglion, amacrine, and photosensitive cells in the El4 retina, as welt as laminae in the inner plexiform layer and the outer ple×iform layer. F: the MAb/C-staining pattern is unchanged following digestion of the section with alkaline phosphatasc. G: an adjacent section showing the same region of the retina as in E stained with APt4, which recognizes only the high-molecular weight form of MAP2. Ganglion cells and a subset of large amacrine cells adjacent to the inner plexiform layer are stained. H: following alkaline phosphatasc digestion, the AP14-staining pattern is unchanged. The non-neuronalperiocular connective tissue is unstained by all 3 MAP antibodies.

315 rinsed thoroughly in PBS, counterstained with Hoechst nuclear dye (1 /~g/ml bisbenzimidine H 33258; Reidel-de HaEn), coverslipped in glycerol,

and viewed with a Zeiss photomicroscope equipped with appropriate filters. Microtubule proteins made from El4 brains were

316 used for immunoblotting. Once-cycled microtubules were made using previously described methods re, and separated on a 3.5-15~'/- SDS-polyacrylamide gradient gel according to the methods of kaemmli H Proteins were electrophoretieally transferred to nitrocellulose 26, and stained with monoclonal antibodies using previously described methods i('. On immunoblots, monoclonal antibody tau-1 recognizes the 4 major, closely spaced bands characteristic of tau 4 (Fig. 1, lane B). When El4 retinas are stained with monoclonal antibody tau-l there is faint staining in the optic fiber and ganglion cell layers (Fig. 2C). After alkaline phosphatase digestion, however, the staining in the optic fiber layer and ganglion cell layer becomes intense, and staining in the external portion of the inner plexiform layer is also seen (Fig. 2D). These results confirm earlier findings that the tau-1 epitope can be masked by phosphorylation> and indicate that the enzyme digestions used in the present study were suitable for unmasking phosphorylated epitopes. The staining pattern with MAb/C, which recognizes both the high- and lowmolecular weight forms of MAP2 on immunoblots of quail microtubule proteins (Fig. 1, lane C), is intense and unchanged by alkaline phosphatase treatment (Fig. 2E,F). The optic fiber layer, ganglion cell layer, several laminae in the inner plexiform layer, amacrine cells, the outer plexiform layer, and the photosensitive cells are stained with MAb/C. In contrast, the staining pattern with AP14, which recognizes only high-molecular weight MAP2 (Fig. 1, lane D), is

-. ........

~>

..........

~ ,

..................

.~

,..,~-~...~,~..:,.,~.~

~-<<>

.

~

-

~

=

~

+

nlore restricted. Only ganglion celt bodies, some ganglion cell dendrites, and a subset ot large amacrine cells found adjacent to the inner ptexiform layer and the branching processes of the latte~ ;ire stained (Fig. 2G). tn some sections, punctate staining corresponding to the microtubule organizing centers M T O C ) of the photosensitive cells are also stained with API4. After treatment with alkaline phosphatase, the APt 4 staining pattern is unchanged (Fig, 2H), indicating thai the limited distribution of A P I 4 staining is nut the result of epitope masking by ph~sphorylation. In the PI0 retina, when this tissue ha,,, reached a more mature form, there are still differences between the staining patterns with MAb/C and API 4, but they arc less pronounced than in the embryo (Fig. 3A,B). MAb/C stains 3 broad, distinct layers in the inner plexiform layer, as well as some ganglion cell and amacrine cell bodies. The celt bodies of bipolar cells are also stained in P10 retinas, albeit faintly. Staining m the photosensitive cells is still present but is greatly reduced when compared to the embryonic pattern. AP14 stains the same 3 laminae ill the inner plexiform layer as MAb/C, though less completely. Subsets of the ganglion and amacrine cell populations are also stained, and the M T O C staining is more pronounced. As in the embryonic retina, the MAP2 staining patterns were unchanged following treatment with alkaline phosphatase, ~hough the tau-I staining pattern was enhanced (restilts not shown). In both E l 4 and the Pt0 eyes, all 3 M A P monochmal antibodies stained neurons exclusively.

'

~

.....

7"',~

¸: . . . . . . .

". . . .

~

~ H - - -

7,'

~

~ , ~ ' ~ : ' - ~

:~=:!r

!~{~

..........

~r

T~:7'

~

Fig. 3. Frozen sections of the retina of a P10 (juvenile) quail stained with monoclonal antibodies to MAP2. A: MAb/C-staining pattern. B: AP14-staining pattern. In contrast to staining in the embryonic retina, the staining of the P10 retina is similar with bothantibodies. This reflects the reduction in the amount of MAP2c found in adult tissues. Ganglion cells and a subset of amacrine cells arc stained, as well as 3 broad laminae in the inner plexiform layer (small arrows, see Fig. 2 for description of neuronal layers)~ Bipolar cells are stained faintly by MAb/C, and the microtubule organizing centers of the photosensitive cells (large arrow) are stained with API4.

317 W e have shown previously that M A b / C stains neurons in the early developing retina at a time just following their differentiation and the onset of process formation, i.e. in a roughly internal to external wave, with ganglion cells and their axons staining at E4, amacrine cells staining at E l 0 , and photosensitive cells staining between E l 0 and E14 (ref. 27). In the current study, these observations have been extended to later periods of d e v e l o p m e n t , when M A b / C staining begins to diminish and resemble the staining pattern of a high-molecular weight M A P 2 specific antibody. The loss of M A P 2 c staining in the rods and cones of photosensitive cells between E l 4 and P10 corresponds with the completion of the development of these structures, and the a p p e a r a n c e of M A P 2 c in P10 bipolar cells corresponds with the late maturation of this population of neurons 1°. Thus, in the developing retina, M A P 2 c is found consistently in differentiating neurons before the a p p e a r a n c e of high-molecular weight M A P 2 , and M A P 2 c appears to be lost from these cells once they attain their final configuration. High-molecular weight M A P 2 staining was seen in some but not all amacrine and ganglion cells. Similar M A P 2 staining patterns have been obtained in the adult rat retina by others 7. The laminations stained by M A P 2 monoclonal antibodies are similar to patterns stained by L-glutamate decarboxylase antibodies, i.e. G A B A e r g i c amacrine cells 13. Since one of the roles that has been p r o p o s e d for high-molecular

weight isoforms of M A P 2 is to confer stability to neuronal processes 18'23, it is possible that a subpopula-

1 Bernhardt, R., Huber, G. and Matus, A., Differences in the developmental patterns of three microtubule-associated proteins in the rat cerebellum, J. Neurosci., 5 (1985) 977-991. 2 Binder, L.I., Frankfurter, A., Kim, H., Caceres, A., Payne, M.R. and Rebhun, L.I., Heterogeneity of microtubule-associated protein 2 during rat brain development, Proc. Natl. Acad. Sci. USA., 81 (1984) 5613-5617. 3 Binder, L.I., Frankfurter, A. and Rebhun, L.I., The distribution of tau in the mammalian central nervous system, J. Cell Biol., I01 (1985) 1371-1378. 4 Binder, L.I., Frankfurter, A. and Rebhun, L.I., Differential localization of MAP-2 and tau in mammalian neurons in situ, Ann. NYAcad. Sci., 466 (1986) 145-166. 5 Burgoyne, R.D. and Cumming, R., Ontogeny of microtubule-associated protein 2 in rat cerebellum: differential expression of the doublet polypeptides, Neuroscience, 11 (1984) 157-167. 6 Caceres, A., Binder, L.I., Payne, M.R., Bender, P., Rebhun, k.I. and Steward, O., Differential subcellular localization of tubulin and the microtubule-associated protein

MAP 2 in brain tissue revealed by immunocytochemistry with monoclonal hybridoma antibodies, J. Neurosci., 4 (1984) 394-410. 7 De Camilli, P., Miller, P.E., Navone, F., Theurkauf, W.E. and Vallee, R.B., Distribution of microtubule-associated protein 2 in the nervous system of the rat studied by immunofluorescence, Neuroscience, 11 (1984) 817-846. 8 De Camilli, P., Moretti, M., Donini, S.D., Walter, U. and Lohmann, S.M., Heterogeneous distribution of the cAMP receptor protein RII in the nervous system: evidence for its intracellular accumulation on microtubules, microtubule organizing centers, and in the area of the golgi complex, J. Cell Biol., 103 (1986) 189-203. 9 Garner, C.C., Brugg, B. and Matus, A., A 70 kDa microtubule-associated protein (MAP), related to MAP2, J. Neurochem., in press. 10 Griin, G., The development of the vertebrate retina: a comparative survey, Adv. Anat. Embrvol. Cell Biol., 78 (1982) 1-85. Huber, G. and Matus, A., Differences in the cellular distributions of two microtubule-associated proteins, MAP 1 and

tion of amacrine cell and ganglion cell processes are more stable than others. The a p p e a r a n c e of M A P 2 in the M T O C of photosensitive cells is correlated with the presence of the regulatory subunit of c A M P - d e p e n d e n t protein kinase (RH), which is known to bind to M A P 2 (refs. 15, 25, 28), in this structure s. W e have used polyclonal antibodies to R H to stain E l 4 quail retinas and have found the photosensitive cell M T O C to be RH-positive (Tucker and H e m m i n g s , unpublished results). Previous studies using the adult rat central nervous system have shown that M A P 2 is more a b u n d a n t in dendrites than in axons 6"7'11. Our observation of M A b / C staining of the optic fiber layer and in the optic nerve 27, and the absence of AP14 staining in these regions, suggests that M A P 2 is found in axons in the avian embryo, whereas the high-molecular weight form of M A P 2 is not. MAP2c appears to have a widespread distribution in embryonic, differentiated neurons during the period of process formation, whereas the highmolecular weight forms of M A P 2 are found in more mature processes. Perhaps the function of M A P 2 c is to compete for tubulin binding sites with the highmolecular weight isoforms of M A P 2 , preventing the latter from conferring stability to the growing and branching process.

318 MAP 2, in rat brain, J. Neurosci., 4 (1984) 151-160. 12 Karr, T.L., White, H.D. and Purich, D,L., Characterization of brain microtubule proteins prepared by selective removal of mitochondrial and synaptosomal components. J. Biol. Chem., 254 (1979) 6107-6111. 13 Karten, H.J. and Brecha, N., Localization of neuroactivc substances in the ,'ertebrate retina: evidence for lamination in the inner plexiform layer, Vision Res, 23 (1083) 1197-1205. 14 Laemmli, U.K., Cleavage of structural proteins during the assembly of the head of the bacteriophage T4, Nature (Lond.), 227 (1970) 680-685. 15 Lohmann, S.M., De Camilli, P., Einig, I. and Walter, U., High-affinity binding of the regulatory subunit (RII) of cAMP-dependent protein kinasc to microtubule-associated and other cellular components, Proc. Natl. Acad. Sci. USA, 81 (1984) 6723-6727. 16 Matus, A., Pehling, G., Ackermann, M. and Maeder, J., Brain postsynaptic densities: their relationship to glial and neuronal filaments, J. Cell Biol., 87 (1980) 346-359. 17 Nunez, J., Differential expression of microtubule components during brain development, Dev. Neurosci., 8 (1986) 125-141. 18 Olmsted, J.B., Microtubule-associated proteins, Annu. Rev. Cell Biol., 2 (1986) 421-457. 19 Papsozomenos, S. Ch., Binder, L.I., Bender, P.K. and Payne, M.R., Microtubule-associated protein 2 within axons of spinal motor neurons: associated with microtubu[es and neurofilaments in normal and fl,fl'-iminodipropionitrile-treated axons, J. Cell Biol., 101)(1985) 74-85. 20 Papsozomenos, S. Ch. and Binder, k.l., Phosphorylation

21

22

23

24

25

26

27

28

determines two distinct species of tau m the central nerwms system, Cell Motil. Cytoskel., in press. Riederer, B. and Matus, A., Differential expression of distinct microtubule-associated proteins during brain development, Proc. Natl. Acad. Sci. USA, 82 (1985)6006-6009. Riederer, B., Cohen, R. and Matus, A,. MAP5: a novel brain microtubule-associated protein under strong developmental regulation, J. Neurocytol., 15 (1986) 763-775. Schliwa, M., Euteneuer, U., Bulinski, ,t.(. and Izant, J.G., Calcium lability of cytoplasmic microtubules and its modulation by microtubule-associated proteins, Proc. Natl, Acad. Sci. USA, 78 (1981) 1037-104I, Sternberger, L,A, and Sternberger, N,|t.. Monockmal antibodies distinguish phosphorylated and nonphosphorylated forms of neurofilaments in situ. P r ~ Natl. Acad. 5ci. USA, 80 (1983) 6126-6130. Theurkauf, W.E. and Vallee, R.B,. Molecular characterization of the cAMP-dependent protein kinase bound to microtubule-associated protein 2, J. Biol. Chem,. 258 (1982l 7883-7886. Towbin, H.. Stehelin, T. and Gordon, l., Elcctrophoretic transfer of protein from polyacrylamide gels to nitrocellulose sheets: procedure and some application. Proc, Nail. Acad. Sci. USA, 76 (1979)4354-4356. Tucker, R.P. and Matus, A.I., Developmental regulation of two microtubule-associated proteins (MAP2 and MAP5) in the embryonic avian retina, Development, in press. Vallee, R.B., Debartolomeis, M.J. and Theurkauf, W.E., A protein kinase bound to the projeclion portion of MAP2 (microtubule-associated protein 2). J ('ell Biol,, 90 (t981) 568-576,