Acetylcholinesterase in neuroblastoma and neuroblastoma × glioma hybrid cells: Cellular localization and molecular forms

hr. J. Devl. Neuroscience. Vol. 5, Nos 516. pp. 417-428. 1987.

0736-5748/87 503.00+0.00 Pergamon Journals Ltd. @ 1987 ISDN

Printedin GreatBritain.

ACETYLCHOLINESTERASE IN NEUROBLASTOMA AND NEUROBLASTOMA x GLIOMA HYBRID CELLS: CELLULAR LOCALIZATION AND MOLECULAR FORMS M. A. R. MELONE,+

A. LONGO, C. Tmtmt

and G. AUGUSTI-TOCCO$

*Istituto

SDipartimento

di Scienze Neurologiche, I” FacoltA di Medicina, Universita di Napoli; tlstituto di Anatomia Comparata. Universita di Bologna; and di Biologia Cellulare e dello Sviluppo. Universita di Roma “La Sapienza”, p. le Aldo 00185 Roma, Italy (Received

I2 February

1987; in revbed form

25 June 1987; accepted

Moro

5,

I9 July 1987)

Abstract-The cellular localization of acetylcholinesterase (AChE) was investigated at the electron microscope (E.M.) in a neuroblastoma and neuroblastoma x glioma hybrid line, which differ for their ability to establish synaptic contacts. Only cells of the latter line show association of AChE to the plasmamembrane, while in the former the activity is mainly intracellular. Sucrose sedimentation analysis of AChE molecular forms has shown no significant differences in the distribution of the two forms, G2 and G4, between the two cell lines. On the contrary a marked difference is observed in the ability of the cell to release the enzyme in the culture medium. In fact the cells lacking AChE on their surface release in the medium a much higher proportion of their enzyme, than the cells showing AChE association to their plasmamembrane. The possible role of two alternative fates for AChE, secretion or membrane insertion, in determining the observed differences of enzyme localization is discussed. Key words: Acetylcholinesterase,

Neuroblastoma,

Neuroblastoma

hybrid,

Neuronal

maturation.

AChE is an enzyme well known for its function in cholinergic synapses, where it inactivates acetycholine released into the synaptic cleft. The cellular localization of the enzyme is of primary importance for this function. AChE has in fact been demonstrated to be an integral membrane protein, whose active site is exposed on the outer plasmamembrane face.‘4.‘7*3’.35Furthermore it has been shown that AChE has an uneven distribution on the muscle cell surface, being accumulated in the endplate region. I3 AChE is present both in muscle and nerve cells in multiple molecular forms; they have been described as globular forms, resulting from the association of a monomer (G,) into dimeric (G,) and tetrameric (G4) structures, and asymmetric forms (A4, AS, A,*) arising from the association of l-3 tetrameric structures to a collagen-like tail. 3*23The various molecular forms display characteristic cellular distribution. Thus A,*, the heaviest among the asymmetric forms, which is accumulated in the muscle endplate region, I3 has been proposed to be associated with the basal lamina of neuromuscular junction. I2 Globular forms also have distinct cellular distribution; G4 is predominantly associated to the plasmamembrane as an integral protein, while the monomer G, is mainly found in the cytoplasm, either as a soluble or membrane-associated enzyme.s*‘2.‘7 AChE in order to exert its functional role in cholinergic transmission requires a complex regulation not only at the level of monomer biosynthesis, but also with respect to the formation of complex molecular structures resulting from monomer association, their transport through the internal membrane system and insertion into the plasmamembrane or excretion into extracellular space, and finally restriction of enzyme molecules to a functionally specialized district of the cell surface. The biosynthesis, molecular form distribution and cellular localization of AChE have been shown to be developmentally regulated in several cellular systems. 22.26.27,3s Studies on rat brain,22 neuroblastoma’“~2n*” and phaechromocytoma’“,2y have demonstrated an increase in AChE activity and a shift from smaller to more complex molecular forms as neuronal maturation progresses. On the other hand functional interactions of motor neurons with muscle cells is known to induce clustering of AChE, characteristic of the junctional area.” However, no direct data are as yet available on the role and possible modifications of neuronal AChE in relation to specific stages of neuronal maturation leading to the formation of mature functional synapses. A system to investigate this problem is represented by neuroblastoma cultures. In fact, in neuroblastoma 417

418

M. A. R. Melone et al.

cells it is possible to modulate the ability to grow fibers by changes in the culture conditions;’ on the other hand cell lines are available which differ for their ability to establish, or not, synaptic contacts with target cells.’ Thus in this system, using different cell lines and culture conditions, one can attempt to distinguish among cellular events related to the two main phases of neuronal maturation, that is fiber growth and synaptogenesis. In this paper we report the results of a study on the ultrastructural localization and biochemical characterization of molecular forms of AChE in a neuroblastoma x glioma hybrid line, which is able to form synaptic contacts with muscle cells24 and the parental neuroblastoma line, defective for synaptic contacts. Both cell lines have been examined under growth conditions allowing, or not, elongation of fibers. MATERIALS

AND METHODS

Cell culture N18TG2 and 108CC5 lines were kindly supplied by Dr B. Hamprecht. Cells were routinely grown as monolayer cultures in a 5% CO2 atmosphere. The growth medium was Dulbecco’s modified minimal essential medium (DMEM) with 1.24 g/i sodium bicarbonate and 15 mM HEPES, supplemented with 10% fetal calf serum (FCS). 108CC5 hybrid cells were grown in the presence of 0.13 mM hypoxanthine, 0.011 mM aminopterine and 0.02 mM thymidine. For experimental analysis cells with an immature morphology, characterized by the absence of fiber outgrowth, were obtained by subculturing an appropriate cell number into Petri dishes. On the other hand to enhance fiber formation in monolayer cultures, serum-supplemented medium was substituted 3-4 days after plating by a chemically defined medium, as described by Bottenstein and Sate,’ and 3 x lo-’ M prostaglandin E, (PGE,) was added. Ultrastructural

analysis

For ultrastructural analysis cells were grown in 60 mm Contur-Permanox dishes. After repeated washing in phosphate-buffered saline (PBS) the cells were fixed in situ with 2.5% glutaraldehyde in 0.1 M cacodilate buffer and postfixed in 2% colloidine buffer osmium tetroxide, dehydrated in ethanol and embedded in Epon 812. The ultrathin sections, obtained with a diamond knife at a LKB ultratome III, were observed under a Philips 301 electron microscope. AChE activity was revealed following the Karnowski procedure, as modified by Rash and Ellisman.*‘” All experiments were performed in the presence of lo-’ M ethopropazine as inhibitor of butyrylcholinesterase. Control experiments were also run in the presence of 10B4 M eserine. Samples were incubated in the reaction mixtures for 20 min in all cases. Cell extracts and AChE

determination

For all biochemical analyses cells were grown in chemically defined medium in 100 mm tissue culture Falcon dishes. At appropriate times after plating the cultures were washed with PBS, the cells mechanically removed with a rubber policeman and homogenized with a glass-teflon homogenizer in 10 mM Ttis-HCI, pH 7.2, 1 mM EDTA, 1 M NaCl, 1% Triton X-100. One mM benzamidine, 25 U/ml aprotinin and 1 mg/ml bacitracin were present during homogenization. Cells in suspension culture were collected and washed by low speed centrifugation and homogenized as above. Homogenates were centrifuged at 31,000 g for 30 min; aliquots of the supernatants diluted in Triton homogenizing buffer were used for AChE determination, according to the Ellman procedure.“’ Protein content of the extracts was determined following the Lowry procedure. ” A sequential extraction, yielding first soluble and then ‘membrane-bound’ enzyme was also adopted; cells were first homogenized omitting Triton X-100 from the buffer, the homogenates were centrifuged and the resulting pellets extracted in the presence of the detergent. To determine the extent of AChE release, the culture medium was changed 24 hr prior to cell harvesting. When the cells were collected for AChE activity assay, the medium was saved and aliquots were used for enzyme activity determination after low speed centrifugation. All enzyme assays were run in the presence of 10d5 M ethopropazine.

AChE in neuroblastoma

and neuroblastoma

x glioma hybrid cells

419

Sucrose gradient sedimentation analysis Sedimentation through sucrose gradient was performed to separate AChE molecular forms. ’ Aliquots of the extracts, obtained as described in the previous section, were loaded on 11 ml linear sucrose gradient (5-20%) in complete homogenizing buffer. Usually the same volume was loaded on each gradient. However the enzyme concentration was not always constant, depending on the number of cells used in each case. Gradients were centrifuged in the SW 41 rotor of a L5-65 Beckman ultracentrifuge at 40,000 rpm for 21 hr. Approximately 40 fractions were collected from each gradient and assayed for AChE activity. Sedimentation coefficients for the AChE peaks were assigned by comparison with the sedimentation of standard P-galactosidase (16 S) and alkaline phosphate (6.1 S). RESULTS E-M. studies show that AChE activity in the parental line N18TG2 cells is localized in the cytoplasm (Figs 1 and 2). When the cells are grown in suspension, the activity appears in the perinuclear space and in the cytoplasm (Fig. 1). In monolayer cultures perinuclear localization of AChE was never observed; on the other side the cytoplasm appears filled with enzymatic activity both in the perikaryon (Fig. 2a) and in the fibers (Fig. 2b). The association of the enzymatic activity to vesicles is clearly evident only in a few cases. Very little activity was found on the cell surface under both growth conditions. When cells of the hybrid neuroblastoma x glioma line were examined a striking difference in AChE localization was observed. In fact in this case, as shown in Fig. 3, the enzyme is preferentially associated to the plasmamembrane. Figure 3a shows a cluster of cells grown in suspension; the enzymatic activity is clearly localized at the cell surface in the contact area between adjacent cells. In cells grown as monolayer cultures also, AChE activity is preferentially associated to the plasmamembrane both on the cell body and fibers (Fig. 3b, c). Analysis of molecular forms Previous data on rat brain,” neurobIastoma’h*“4 and phaechromocytomat4 have shown that G4 AChE is preferentially associated to the membrane compartment. Thus on the basis of the E.M. obse~ations described in the previous section, one may expect a different dist~bution of AChE molecular forms in the two cell lines studied. The data reported in Fig. 4 show that in both cell lines only two globular forms of AChE, G, and G4, are present, as previously reported in various neuroblastoma lines. ‘s~‘7~30 The relative position of the peaks on the gradients shows some variation due to differences in the number and volume of fractions collected. The molecular forms of both membrane-bound (Fig. 4a, b) and soluble AChE (Fig. 4c, d) were separately analysed after sequential extraction of the cell without and with detergent. Soluble enzyme is only represented by G, (Fig. 4c, d), while in the detergent extracted fractions G, and G4 are present, the latter being the predominant form (Fig. 4a, b). Data presently available do not show significant differences in the relative amount of the two enzyme forms in the total extracts of the two cell lines (Table 2). Similar results were obtained in the analysis of AChE extracted from cells grown in suspension cultures (data not shown). Thus these data do not suggest any correlation between AChE localization on the cell surface and the dist~bution of its molecular forms. They also show that in the parental line there is a significant amount of AChE associated to membranes; however in these cells it remains confined to a cytoplasmic compartment and never reached the cell surface, as shown by the ultrastructural studies. A quantitative evaluation of AChE extracted in the two fractions, soluble and membrane-bound, is reported in Table 1. This shows that both in the hybrid and parental line, 40% of the enzyme activity is extracted in the absence of detergent. The data described in this section thus suggest that the absence of AChE on the cell surface of N18TG2 cells, demonstrated by E.M., is not dependent on alterations in the molecular assembly of the protein monomer into larger complexes and/or its association to cytomembranes. Release Release of AChE by nerve and muscle cells is a well-known phenomenon. It has been described in several culture systems,“*1’.N’.2sincluding neuroblastoma,” as well as in intact DN 5:5/6-n

M. A. R.

420

Melone

et al.

Ph alk

L

0b

_-A-

10

30

50 Fracfions

0

0d

Fig. 4. Sedimentation profiles of AChE activity in cellular extracts. Cells were grown in IOU mm dishes as described in Materials and Methods. After harvesting, soluble and membrane-bound AChE was sequentially extracted by a first homogenization in the absence of detergent, followed by a second extraction with the addition of 1% Triton X-100. Aliquots of the extracts were run on S-IO% sucrose gradients as described. N18TG2 (a) and 108CC5 (b) extracts in the presence of Triton X-100; N18TG2 (c) and 108CC5 (d) extracts without detergent. AChE activity is expressed as O.D. at 412 urn developed in the assay of each fraction of the gradients.

AChE

Fig. 1. N18TG2

in neuroblastoma

and neuroblastoma

X glioma

hybrid cells

cells grown in suspension. AChE activity appears mainly localized in the perinuclear space and inside the cytoplasm, as indicated by arrowhead. X 17,000.

421

M. A. R. Melone et

al.

Fig. 2. N18TG2 cells grown in monolayer cultures. (a) Part of the perikaryon showing AChE acttvity inside the cytoplasm. x 27,600. (b) Portion of a fiber with AChE activity localized among neurotubules and neurotilaments. x 27,MW).

AChE in neuroblastoma

and neuroblastoma

x glioma hybrid cells

Fig. 3. 10%X5 cells. (a) A group of ceils grown in suspension, showing AChE activity on the cell surface, as indicated by arrowheads. x 13,600. (b) Perikaryon of a cell grown on monolayer, showing enzymatic activity associated to the cell surface. x i12.500. (c) A fiber of a cell in monolayer culture showing AChE activity on the cell surface. x 33,750.

423

AChE

in neuroblastoma Table

and neuroblastoma

1. Cellular

Cellular fraction

distribution

AChE

x glioma

of AChE

hybrid cells

activity

activity

N18TG2

108ccs

Soluble

% Total activity Specific activity

40 0.209

38 0.214

Membrane-bound

% Total activity Specific activity

60 0.802

62 0.752

AChE activity was determined obtained after sequential extraction 1% Triton X-100, as described in activity in each fraction is expressed activity. Specific activity is expressed

425

on two cellular fractions, of the cells, without and with Materials and Methods. The as percentage of total cellular as mU/mg protein.

Although its biological role is still obscure, evidence obtained in cultured muscle cells3 points to AChE release as an active process, independent on the enzyme insertion into the plasmamembrane. It thus appeared of interest to examine the ability to release AChE in the culture medium by the two cell lines used in this study and differing for the presence of AChE on their surface. Figure 5 shows the sedimentation pattern of AChE activity released in the medium over a 24-hr period. As previously reported for neuroblastoma” and other cells” G, is selectively released in both cell lines. Although G, is also present in a significant amount, the predominant molecular species found in culture medium is in fact G,; on the contrary the cell-associated activity has a higher content of G, than G4 (Table 2). Furthermore, while the ratio G4/GI of the cell-associated AChE is very similar in the two cell lines, the parental neuroblastoma cells differ considerably from the hybrid ones as far as the released enzyme is concerned. As shown in Table 2, not only after 24 hr is release of AChE by the parental neuroblastoma cells about three-fold higher than in the hybrid cells, but also the relative amount of G, with respect to G, is considerably higher in the released activity. tissue.4.”

Table 2. Cellular

and released AChE

Released activity

Cellularactivity

U/dish N18TG2 108CC5

2.17 I.58

G/G, 0.6 0.8

activity

U/dish

G,/G,

released cellcontent

4.9 1.07

4.21 1.7

225 68

Y a

Cells were grown in the absence of serum, as indicated in Materials and Methods. Cells and media, 24 hr after medium change, were collected and AChE activity determined on aliquots of total homogenates and media. Samples were run on sucrose gradients and relative amounts of G4 and G, estimated from the area of each peak.

DISCUSSION The reported data show that at the E.M. AChE appears differently localized in the two neuronal lines investigated. In the hybrid cells it appears mostly on the cell surface, while in the parental cells an intracellular localization is predominant. While in the parental cells grown in suspension the enzyme is found inside the perinuclear cisternae, in the cells grown in monolayer the association to vesicular structures in most cases cannot be clearly established. This may be dependent on the small size of vesicles, which can be hidden by the product of enzyme reaction. However the existence of at least a fraction of the enzyme associated to cytoplasmic matrix cannot be excluded. As a matter of fact it would be in agreement with previous findings, such as the presence in several systems of a soluble AChE form, which does not require detergent for extraction.l”.l2.‘7.23” Also in neuroblastoma cells data on subcellular distribution of AChE indicate the presence of some activity not associated to subcellular structures.s*32a

M. A. R. Melone et al.

426 OD

0a

I

/

30

50

Froctlonr

0b

10 Fig. 5. Sedimentation

profiles of AChE

I

30

50

__-

Fractlans

activity released in the culture medium in 24 hr (a) IOWCS (b) Nl8TG2 cells.

cells,

Since association to plasmamembrane is preferential for G4 molecular forms, the predominant cytoplasmic localization of AChE in Nl8TG2 cells could be ascribed to an altered assembly of the monomeric protein into tetrameric structures. However the analysis of molecular forms in the two cell lines has not shown significant differences in the relative amount of G, and G,. Moreover the percentage of AChE present in the cells as a soluble protein is about 40% in both cell lines. This rules out the possibility that in N18TG2 cells newly synthesized enzyme molecules carry some modifications, which impair their association to internal membranes, where posttranslational modifications of the enzyme have been shown to occur.3’ It has been recently proposed that soluble and membrane-bound AChE may be coded by different genes.“.33..UsA differential expression of the two genes in parental and hybrid cells could take into account the reported differences in enzyme localization. However the presence in both cell lines of a similar fraction of AChE requiring detergent for extraction, suggests an equal expression of the genes coding for the two enzyme forms. The data of enzyme release in the culture medium suggest an alternative interpretation for the absence of AChE activity on the cell surface of N18TG2. In several cell types (neurons, muscle,

AChE in neuroblastoma and neuroblastoma x glioma hybrid cells

427

gland) AChE release has been demonstrated; it has also been reported that release of AChE by muscle cells in culture does not involve its previous association to the plasmamembrane.“2 One can then envisage that in our system the preferential fate for AChE (membrane insertion or secretion) may be different in the two cell lines. In fact N18TG2 cells, which show very little association of the enzyme to the cell surface at the E.M., also show a considerably higher level of AChE in their culture medium, as compared with the hybrid cells, suggesting that the relative fraction of enzyme secreted and inserted into the plasmamembrane varies in the two cell types. The ratio of G$G, in the culture medium is very different from that of the cellular enzyme, indicating that AChE release, also in our case, is an active process and does not result from passive leakage of cellular components through the cell membrane. Whether the higher release of AChE in N18TG2 cells depends on some as yet unknown signal present on the enzyme molecule, which is capable of forwarding AChE to a secretory pathway, or may be a consequence of the inability of plasmamembrane to retain the enzyme molecules in its structure, cannot be established on the basis of the data presented here. The ability of hybrid celb to establish synaptic function contacts,24 as well as previously reported differences in the distribution of Concanavalin A receptors between the two cell lines, ’ indicate that composition and/or architecture of membrane components are different. These considerations may support the second hypothesis; however, further work is in progress to investigate these aspects in more detail. adrenal

~e~ffo~~e~gemenr~-~e authors express their thanks to Prof. G. Ghiara and to the technical staff of the Centro di Studio di Microscopia Elettronica. Universita di Napoli. where parts of this work were carried out. Many thanks are also due to Prof. G. Toschi, S. Biagioni and G. Scarsella for their advice in the early phase of this work and for critical reading of the manuscript. Work supported by grants from the MPI 40% and by the CNR P.F. “Medicina Preventiva e Riabilitativa”, contract no. 86.01657.56.

REFERENCES 1. Bon S., Vigny M. and Massoulie J. (1979) Asymmetric and globular forms of AChE in mammals and birds. froc. natn. Acad.

Sci.,

U.S.A.

76, 2546-2550.

la. Bon S. and Massoulie J. (1980) Collagen-tailed and hydrophobic components of acetylcholinesterase in Torpedo marmorata electric organ. Proc. narn. Acad. Sci., U.S.A. 77, 44644X& 2. Bottenstein J. E. and Sato G. H. (1979) Growth of a rat neuroblastoma cell line in serum-free supplemented medium. Proc. narn. Acad. Sci., U.S.A. 76, 514-517. 3. B~mijoin S. (1983) Molecular forms of AChE in brain, nerve and muscle: nature, localization and dynamics. Prog. Neurobiol. 4.

5. 6. 7.

8. 9. IO.

ii. 12. 13. 14. IS.

21, 291-322.

Carter G. L. and Brimijoin S. (1981) Effects of acute and chronic denervation on release of AChE and its molecular forms in rat diaphragms. 1. Neurochem. 36, 1018-1025. Chang C. H. and Blume A. J. (1976) Heterogeneity of AChE in neuroblastoma. J. Neurochem. 27, 1427-1435. Chubb I. W., Goodman S. and Smith A. D. (1976) Is AChE secreted from central neurons into the cerebrospinal fluid? Neuroscience I, 57-62. Denis-Donini S. and Augusti-Tocco G. (1980) Molecular and lectin probes analysis of neuronal differentiation. Curt. Top. dept. Biot. 16, 324-348. Denis-Donini S., Estenoz M. and Augusti-Tocco G. (1978) Cell surface modifications in neuronal maturation. Celf Diff. 7, 193-201. Doctor B. P., Camp S., Gentry M. K., Taylor S. S. and Taylor P. (1983) Antigenic and structural differences in the catalytic subunits of the molecular forms of AChE. froc. mm. Acad. Sci., U.S.A. 80, 5767-5771, Ellmann G. L., Courtney K. D., Andres, V., Jr and Featherstone R. M. (1961) A new and rapid calorimetric determination of AChE activity. Biochem. Pharmac. I, 89-95. Gisiger V. and Vigny M. (1977) A specific form of AchE is secreted by rat sympathetic ganglia. FEBS Letr. 84.253256. Grassi J., Vigny M. and Massoulit J. (1981) Molecular forms of AChE in bovine caudate nucleus and superior cervical ganglia: solubihty properties and hydrophobic character. J. Neurochem. 38, 457-469. Hall ZT W. (1973) Multiple forms of AChE and their distribution in endplate and non-endplate regions of rat diaphragm muscle. J. Neurobiol. 4, 343-361. lnestrosa N. C., Reiness C. G., Reichardt L. F. and Hall 2. W. (1981) Cellular localization of the molecular forms of AChE in rat phaechromocytoma PC12 cells treated with nerve growth factor. J. Neurosci. 1, 1260-1267. Kimhi Y.. Mahler A. and Saya D. (1980) AChE in mouse neuroblastoma cells; intracellular and released enzyme. J. Neurochem.

34, 554-557.

16. Lanks K. W., Dorwin J. M. and Papirmeister B. (1974) Increased rate of AChE synthesis in differentiating neuroblastoma cells. J. Cell Biol. 63, 824-830. 17. Lazar M. and Vigny M. (1980) Modulation of the distribution of AChE molecular forms in a murine neuroblastoma x sympathetic ganglion hybrid cell line. 1. Neurochem. 35, 1067-1079. 18. Lomo T. and Slater C. R. (1980) Control of junctional AChE by neural and muscular influences in the rat. J. Physiol.. Land. 303, 191-202. 19. Lowry 0. H., Rosebrough N. J., Farr A. L. and Randall R. J. (1951) Protein measurements with the Folin phenol reagent. 1. biol. Chern. 193, 2-275.

M. A. R. Melone et al.

428

20. Lucas C. A. and Kreutzberg G. W. (1985) Regulation of AChE secretion from neuronal cell cultures-l. Actions of nerve growth factor, cytoskeletal inhibitors and tunicamycin. Neuroscience 14, 349-360. 21. McMahan U. J., Sanes J. R. and Marshall L. M. (1978) Cholinesterase is associated with the basal lamina at the neuromuscular junction. Nulure 271, 172-174. 22. Marchand A., Chapoutier G. and Massoulie J. (1977) Developmental aspects of AChE activity in chick brain. FEBS Len. 78, 233-236. 23. Masoulie J. and Bon S. (1982) The molecular Rev. Neurosci. 5, 57-106.

23a.

Massoulie

J.. Bon S., Lazar

Cholinesrerases,

Fundamental

M., Grassi

forms of cholinesterase

J., Marsh

and Applied

D., Meflah

Aspects (eds Brzin

and acetylcholinesterase

K., Toutant M., Barnard

in vertebrates.

Ann.

J. P., Vallette F. and Vigny M. (1984) E. A. and Sket D.), pp. 73-97. W. de

Gruyter. Berlin. 24. Nelson P., Christian C. and Nirenberg M. (197.5) Synapse formation between clonal neuroblastoma x glioma hybrid cells and striated muscle cells. Proc. narrr. Acad. Sci.. U.S.A. 73, 123-127. 25. Oh T. H., Chyu J. Y. and Max S. R. (1977) Release of AChE by cultured spinal chord cells. J. Neurubiol. 8,469-476. 26. Pannese E. (i974) The histogenesis of spinal ganglia. Adv. An. Embr. &II Biol. 47, l-97. 27. Poiana G.. Scarsella G., Biagioni S. and Augusti-Tocco G. (1985) Molecular forms of AChE during development of chick dorsal root ganglia. In Molecular Aspecrs of Neurobiology, Florence, p. 45. 27a. Rash J. E. and Ellisman M. H. (1974) Studies of excitable membranes. I. Macromolecular specializations of the neuromuscular junction and the non-junctional sarcolemma. J. CeII Biol. 63, 567-586. A., Benda P. and Vigny M. (1976) Molecular forms of AChE: their de novo synthesis in 28. Rieger F.. Faivre-Bauman mouse neuroblastoma cells. J. Neurochem. 27, 1059-1063. 29. Rieger F.. Shelanski M. L. and Greene L. A. (1980) The effects of nerve growth factor on AChE and its multiple forms in cultures of rat PC12 phaechromocytoma cells: increased total activity and appearance of the 16s molecular form. Devl Biol. 76, 238-243. 30. Rieger F. and Vigny M. (1976) Solubilization and physicochemical characterization of rat brain AChE: development and maturation of its molecular forms. J. Neurochem. 27, 121-129. R. L. (1984) Asymmetric AChE is assembled in the Golgi. Proc. nom. Acad. Sci.. U.S.A. 81, 479-483. 31. Rotundo R. L. and Fambrough D. M. (1980) Synthesis, transport and fate of AChE in cultured chick embryo muscle 32. Rotundo cells. Cell 22, 583-594. 32a. Scarsella G., Biagioni S. and Augusti-Tocco G. (1987) Cellular distribution of acetylcholinesterase in neuroblastoma and neuroblastoma x glioma cells. IBRO 2” World Congress of Neuroscience, Budapest 1987. 33. Sikorav J., Vallette F., Grassi J. and Massoulie J. (1985) Isolation of a cDNA clone for a catalytic subunit of Torpedo marmorata AChE. FEBS Len. 193, 159-163. 34. Vimard C., Jeantet C., Netter Y. and Gros F. (1976) Changes in the sedimentation properties of AChE during neuroblastoma differentiation. Biochimie 58, 473-478. 34a. Witzemann V. and Boustead C. (1983) Structural differences in the catalytic subunits of AChE forms from the electric organ of Torpedo marmorata. EMBO J. 2, 87>878. 35. Ziskind-Conhaim L., lnestrosa N. C. and Hall Z. W. (1984) AChE is functional in embryonic rat muscle before its accumulation at the sites of nerve-muscle contacts. Devl Biol. 103, 369-377.