Neuroscience Vol. 111, No. 3, pp. 649^656, 2002 ß 2002 IBRO. Published by Elsevier Science Ltd All rights reserved. Printed in Great Britain PII: S0306-4522(02)00031-3 0306-4522 / 02 $22.00+0.00 www.neuroscience-ibro.com

A NON-CHOLINERGIC, TROPHIC ACTION OF ON HIPPOCAMPAL NEURONES IN VITRO: MOLECULAR MECHANISMS

T. DAY and S. A. GREENFIELD Department of , University of Oxford, Mans¢eld Road, Oxford OX1 3QT, UK

AbstractöIn this study neurite outgrowth from cultured hippocampal neurones was increased by addition of acetylcho- linesterase acting in a non- manner. Only monomeric acetylcholinesterase, a form of acetylcholinesterase dominant in development, increased neurite outgrowth (3^10 U/ml); moreover this e¡ect was not blocked by active site blockers ( and galanthamine) but was sensitive to the addition ofperipheral site blockers (fasciculinand BW284c51). It appears therefore that acetylcholinesterase has alternative, non-cholinergic functions, one of which could be in development, via a peripheral site. The possibility ofa causal relationship between neurite outgrowth and calcium in£ux was explored using a spectrum ofacetylcholinesterase variants, inhibitors and calcium channel blockers. Acetyl- regulation ofoutgrowth was shown to depend on an in£ux ofextracellular calcium speci¢cally via the L-type voltage-gated calcium channel. In summary, we propose that, independent ofits catalytic activity, a selective formofacetylcholinesterase has a role in the development ofhippocampal neurones via a selective voltage-gated calcium channel. ß 2002 IBRO. Published by Elsevier Science Ltd. All rights reserved.

Key words: neurite, outgrowth, neurotrophic, tissue culture, nimodipine.

Acetylcholinesterase (AChE, EC 3.1.1.7) is well known used to demonstrate that AChE promotes neurite out- for its role in the hydrolysis of (ACh). growth at least in part through an adhesive function However, it has been shown in several brain regions (Grifman et al., 1998; Sharma et al., 2001). that there is a disparity between levels ofAChE and These studies have demonstrated that AChE is actively acetyltransferase, a more reliable marker of cho- involved in neurite outgrowth since selective action of linergic neurones (Silver, 1974). Moreover, a soluble AChE inhibitors retards neuritic outgrowth and neuro- form of AChE has been identi¢ed that is secreted, an nal migration. Whilst a tetrameric form (G4) is dominant action which is beyond the requirements ofcholinergic within the adult CNS, lower weight forms (monomeric transmission (Green¢eld, 1991b). Hence AChE may have [G1] and dimeric [G2]) are most abundant in the devel- a further role distinct from its enzymatic function (Silver, oping CNS. Thus, in addition to the proposal that AChE 1974; Green¢eld, 1984, 1991a,b, 1992; Balasubramanian has a non-cholinergic developmental role, it has also and Bhanumathy, 1993; Appleyard, 1994; Darboux et been postulated that speci¢c developmental forms of al., 1996; Grifman et al., 1998; Tsigelny et al., 2000; AChE exist (Drews, 1975). In an attempt to identify Soreq and Seidman, 2001). More direct evidence for a selective forms of AChE responsible for these non-clas- non-cholinergic, developmental role for AChE has sical actions, this study compared for the ¢rst time the emerged from studies of cultured neurones (Gupta and actions ofmonomeric and tetrameric AChE on the Bigbee, 1992; Layer et al., 1993; Dupree and Bigbee, development ofthe hippocampal neurones in vitro. 1994; Jones et al., 1995; Small et al., 1995; Holmes et The early survival and development ofneurones is al., 1997). More recently, antisense technology has been dependent on levels ofcalcium (Ca 2þ) in£ux through voltage-gated calcium channels (VGCC)(Meir et al., 1999). Neurotransmitter-controlled Ca2þ in£ux can also *Corresponding author. Tel.: +44-1865-271627; direct remodelling in the mature CNS, whilst Ca2þ £ux is fax: +44-1865-271853. a necessary step in several signal transduction pathways E-mail address: [email protected] (T. Day). controlling neurite outgrowth (Mattson, 1992; Sa¡ell et Abbreviations: ACh, acetylcholine; AChE, acetylcholinesterase; al., 1992). Based on biophysical and pharmacological BChE, ; BW284c51, 1:5-bis(4-allydimethyl- data these channels can be divided into three distinct ammoniumphenyl)-pentan-3-one-dibromide; DIV, days in vitro; groups (Hagiwara et al., 1975; Usowicz et al., 1992; Fas, ; G1, monomeric AChE; G2, dimeric AChE; G4, tetrameric AChE; LDH, lactate dehydrogenase; MAP, microtu- Cherksey et al., 1991): (1) low-voltage-activated channels bule-associated protein; NGF, nerve growth factor; P, postnatal; blocked by nickel, (2) a set ofmoderate-conductance, PAS, peripheral anionic site; rAChE, recombinant AChE; high-voltage-activated channels (N, R, and P/Q) blocked VGCC, voltage-gated calcium channel; UV, ultraviolet. by funnel web spider peptide toxin, g-agatoxin VIA and 649

NSC 5485 6-5-02 650 T. Day and S. A. Green¢eld

Fig. 1. Hippocampal cell culture maintained in serum-free medium 14 DIV and processed for MAP-2 immunohistochemistry. This immunochemistry formed the basis of this investigation. Scale bar = 50 Wm.

(3) a large-conductance L-type channel blocked by the Drugs and solutions dihydropyridines, e.g. nimodipine. The present study investigated the e¡ect ofAChE and Cultures were maintained and all treatments were given in g prewarmed, fresh ¢ltered serum-free medium (Dulbecco’s mini- VGCC blockers, nimodipine, -agatoxin VIA, mal essential medium plus B-12 supplement) containing antimi- GVIA and soluble spider toxin (sFTX), in the develop- totic agents (Sigma-Aldrich) to prevent the over-proliferation of ment ofhippocampal neurones using the following glia cells and Fungizone0 (Life Technologies, Paisley, UK) to parameters: neurite outgrowth, lactate dehydrogenase minimise the risk ofinfection.At this point it is relevant to state (LDH) release, cell body area and primary neurite num- that B-12 is su⁄cient to replace serum in the medium. All inhib- itors were solubilised in water at 10 mM and stored in 50-Wl ber (Holmes et al., 1995, 1997; Heng et al., 1999). aliquots at 320‡C. Fasciculin (Fas; Latoxan, France) was solu- bilised in water to give a 10 WM stock. Nimodipine, sFTX 3.3 and g-agatoxin (CN Biosciences, Nottingham, UK) were made EXPERIMENTAL PROCEDURES up as 1 M stock solutions in 0.1% dimethyl sulphoxide and stored as aliquoted fractions at 320‡C. g-Conotoxin GVIA (Tocris-Cookson, Bristol, UK) was made as 1 mM stock and Puri¢cation of AChE maintained at 320‡C.

The puri¢cation oftetrameric AChE (G 4) was carried out by Dr Jane Loughlin (Department ofPharmacology, University of Hippocampal slice cultures Oxford), from a commercial stock of electric eel AChE (Sigma- Aldrich, Poole, UK; type V-S, 1000 U/ml) using procainamide The organotypic slice culture system was chosen to investigate ECH Sepharose a⁄nity chromatography (Novales-Li, 1994). the e¡ects ofcholinesterase due to the complexity ofthe prep- Mouse recombinant AChE (rAChE) was kindly provided by aration. These cultures resemble as closely as possible the phys- Professor Palmer Taylor (Department of Pharmacology, Uni- iological environment in the developing rat brain. versity ofCalifornia,San Diego, CA, USA). The enzymes Hippocampal slices were prepared from postnatal day 7 (P7) were produced from stable transfections of constructs from Wistar rats (Harlan UK, Oxford, UK) (Ga«hwiler, 1981, 1984). the mouse pcDNA-3 clone. All preparations were expressed in In brief, rats were decapitated by a quick scissor cut at the level a HEK 293 cell line and shipped to Oxford in foetal calf serum ofthe foramenmagnum. Excess cortex and adhering meninges with 20% glycerol on dry ice. were removed before the hippocampi were placed ventral sur- AChE puri¢cation was assessed using gel ¢ltration with ultra- face down on a McIlwain tissue chopper (The Mickle Lab., violet (UV) detection (SMART1, Fig. 1) and AChE activity for Engineering Company, Surrey, UK). Coronal sections (400 each molecular form was determined using the spectrophoto- Wm) were cut and attached to cleaned, sterile poly-D-lysine- metric Ellman assay (Ellman et al., 1961). coated coverslips by means ofa plasma clot, formedby mixing In brief, 25-Wl aliquots ofsample were added to 96-well a solution ofchicken plasma with bovine thrombin. microtitre plates, followed by the addition of 175 Wl Ellman Cultures were treated for 14 days in vitro (DIV) with serum- reagent. Absorbance was determined at 405 nm over a 10-min free medium supplemented with cholinesterase in the presence or time span using a Molecular Devices UV plate reader (Alpha absence ofcholinesterase inhibitors. Ca 2þ channel blockers were Laboratories, Hampshire, UK). All were added for 24 h prior to ¢xation. Cultures were changed biweekly expressed in U/ml. using supplemented serum-free medium, used in order to mini- The Ellman assay was used to determine cholinesterase levels mise the amount ofcholinesterase in the system. 14 DIV was in the culture system (culture medium and chicken plasma), to chosen as a time point prior to the cultures becoming less con¢rm the ¢nal concentration ofcholinesterase in the supple- healthy and cell necrosis occurring. mented culture medium and to con¢rm that the concentration of All experiments were conducted in accordance with the Ani- inhibitors added was su⁄cient to inhibit levels ofcholinesterase mals Scienti¢c Procedures Act (1986), approved by Her Majes- within the culture system. The Bradford method was used to ty’s Government Home O⁄ce. All e¡orts were made to determine protein content (Dang et al., 1997). minimise the number ofanimals used and their su¡ering.

NSC 5485 6-5-02 AChE increases neurite outgrowth via voltage gated calcium channels 651

Fig. 2. Serum-free medium supplemented with 3 U/ml G1 rAChE signi¢cantly increased neurite outgrowth ofhippocampal cultures, maintained for 14 DIV when compared to a medium control group. Addition of Fas to the AChE peripheral site blocker alone had no e¡ect on any parameter studied. Fas added with G1 rAChE (3 U/ml) blocked the neurotrophic action. **P 6 0.01 using a Dunnett’s multiple comparisons t-test. n = 150. Results are expressed as mean neurite outgrowth (Wm) þ S.E.M.

Measurements and analysis and by 28.7 þ 0.4% (206.9 þ 6.6 Wm, mean þ S.E.M., n = 150, P 6 0.01, Dunnett’s multiple comparisons On completion ofthe incubation period, immunohistochem- ical staining for microtubule-associated protein (MAP-2) was t-test) respectively. Interestingly, G1 rAChE had no e¡ect performed using the avidin^biotin^peroxidase method (Hsu et on any other parameter measured including LDH al., 1981). MAP-2 was the antibody ofchoice to measure the release, primary neurite number and cell body area, rel- dendritic growth ofpyramidal cells in culture. We did not want ative to control cultures. The addition ofG 4 AChE or to distinguish between dendrites and axons in this system; there- fore only one measure of neurite growth was employed. Mor- butyrylcholinesterase (BChE) (Sigma-Aldrich) had no phogenic analysis was by camera lucida drawing. Dendrite e¡ect on any ofthe parameters measured in this culture length was measured from cell body to dendrite tip and system. expressed as percentage ofcontrol cultures. The quantitative determination ofLDH kinetics was via spec- The e¡ect of cholinesterase inhibitors on neurones of the trophotometry. During the reduction ofpyruvate by LDH, an equimolar amount ofNADH is oxidised to NAD þ. The oxida- hippocampus in vitro tion ofNADH results in a decrease in the light absorbance at 340 nm wavelength. The rate ofthis decrease during this reac- The stimulatory e¡ect observed by an increase in neu- tion is directly proportional to LDH activity in the sample. rite outgrowth following the addition of G1 rAChE-sup- plemented medium was blocked by the addition of Experimental design and statistical analysis AChE inhibitors that blocked the peripheral site of Experiments were repeated a minimum ofthree times using AChE. Fas (a potent peripheral site blocker) alone had di¡erent litters. Each treatment group contained a minimum of no e¡ect on neurite development in vitro (Fig. 2). How- ¢ cultures, with a minimum of10 cells counted per culture ever, when co-applied with G1 rAChE, Fas blocked the (n v 150). Results were expressed as percentage mean þ S.E.M. trophic e¡ect (20^100 nM). Increasing the concentration ofa medium only control group. Statistical analysis was by analysis ofvariance followedby Dunnett’s multiple comparisons ofFas (500 nM) caused a signi¢cant decrease in neurite t-test, used to compare treatment groups with control groups. outgrowth (82.7 þ 2.9 Wm, mean þ S.E.M., P 6 0.01, Dunnett’s multiple comparisons t-test, n = 150), when compared to control cultures. Although at this concen- tration there was a tendency for Fas (500 nM) to RESULTS increase LDH release into the medium, this was not sig- ni¢cant and no higher concentrations were tested. The e¡ect of cholinesterase on neurones of the BW284c51 (Sigma-Aldrich), a potent AChE inhibitor hippocampus in vitro that binds to both the active site and the peripheral site ofAChE, alone caused toxic e¡ects on hippocampal cell Control cultures maintained in serum-free medium had in vitro. At 100 WM, BW284c51 caused a signi¢cant a mean neurite length of160.6 þ 0.1 Wm (mean þ S.E.M., decrease in outgrowth (83.7 þ 2.7 Wm, mean þ S.E.M., n = 330), a mean cell body area of85.3 þ 2.4 Wm2 P 6 0.01, Dunnett’s multiple comparisons t-test, (mean þ S.E.M., n = 330) and a mean primary neurite n = 160) (Fig. 3) and a signi¢cant decrease in cell body number of4.9 þ 0.3 (mean þ S.E.M., n = 150). Addition area (56.5 þ 7.5 Wm2, mean þ S.E.M., P 6 0.05, Dunnett’s ofG 1 rAChE (3 and 10 U/ml, protein content 1.34 and multiple comparisons t-test, n = 160). A signi¢cant 4.47 Wg/ml, respectively) signi¢cantly increased neurite increase in LDH release was seen at 10^100 WM length by 26.4 þ 2.6% (203.1 þ 10.2 Wm, mean þ S.E.M., BW284c51 (6.1 þ 2.9 U/ml, P 6 0.01, and 5.2 þ 1.7 n = 150, P 6 0.01, Dunnett’s multiple comparisons t-test) U/ml, P 6 0.05 respectively, mean þ S.E.M., Dunnett’s

NSC 5485 6-5-02 652 T. Day and S. A. Green¢eld

Fig. 3. Serum-free medium supplemented with 3^10 U/ml G1 rAChE signi¢cantly increased neurite outgrowth ofhippocampal cultures, maintained for 14 DIV when compared to a medium control group. Addition of BW284c51 (BW), which blocks both the AChE active site and peripheral anionic site, caused a signi¢cant decrease in neurite outgrowth at the maximum concentration tested (100 WM). BW284c51 applied with G1 rAChE (3 U/ml) blocked the neurotrophic action. **P 6 0.01, *P 6 0.05 using a Dunnett’s multiple comparisons t-test. n = 160. Results are expressed as mean neurite outgrowth (Wm) þ S.E.M. multiple comparisons t-test, n = 150). This result was not AChE is vast and many good reviews exist (Green¢eld, disregarded due to reports ofa similar nature showing a 1991a,b; Soreq and Seidman, 2001). For these studies toxic action ofhigh BW284c51 concentrations (Holmes endogenous cholinesterases in the preparation might et al., 1997). At all concentrations, BW284c51 blocked have been accountable for the e¡ects observed. In an the trophic e¡ect ofG 1 rAChE (Fig. 3). Addition ofthe attempt to minimise cholinesterase levels in the culture active site inhibitors galanthamine (Sigma-Aldrich) and system, tissue was maintained in a serum-free medium. echothiophate and the BChE inhibitor tetraisopropyl However, despite this strategy, cholinesterase would still pyrophosphoramide (Sigma-Aldrich) had no e¡ect be present in the chicken plasma used to secure the tissue when applied alone or when co-applied with G1 section. However, this background cholinesterase was rAChE on this slice culture system. predominantly BChE which was shown to be ine¡ective There is a possibility that due to the puri¢cation pro- at increasing neurite length in this system. None the less, cedure used, procainamide may bind to the anionic site, inclusion ofa group with no additional exogenous cho- even in the case ofthe ionic site inhibitors, thereby linesterase adequately controlled for the presence of inducing a block attributable to the peripheral anionic endogenous cholinesterase in the culture system. site (PAS), rather than to the actual anionic site. How- ever, given that these inhibitors did not have an e¡ect A neurotrophic role for monomeric AChE and, indeed, were di¡erent from those peripheral site blockers, such a possibility seems unlikely. The addition ofG 1 rAChE (3 U/ml) (protein content, 1.34 Wg/ml) to developing and regenerating neurones of The e¡ect of calcium channel blockers on neurones of the the hippocampus within the slice culture system signi¢- hippocampus in vitro cantly increased the length ofMAP-2 positive neurones. The concentrations ofG 1 rAChE, which elicited a stim- The stimulatory e¡ect observed by an increase in neu- ulatory response (3^10 U/ml, protein content of1.34^ rite outgrowth following the addition of G1 rAChE-sup- 4.47 Wg/ml), were relatively high when compared to plemented medium was blocked by the addition of doses at which other neurotrophic factors are active. nimodipine (1^100 nM) (Fig. 4), the L-type VGCC The survival promoting e¡ect ofnerve growth factor blocker. This e¡ect was not observed with the addition (NGF) on GABAergic cells ofthe hippocampus can be ofagatoxin VIA (N-type), GVIA (P-type) or sFTX 3.3 seen at 100 ng/ml (Arimatsu and Miyamoto, 1991). The (N-, P-, T-, R-type). The addition ofthese agents, which more modest potency ofG 1 rAChE could be due to the selectively block Ca2þ in£ux through voltage-dependent fact that the hippocampal slice culture was embedded in membrane channels, had no e¡ect alone on hippocampal a plasma clot. Di¡usion experiments have showed that cells in culture. AChE was capable ofpermeating the plasma clot. How- ever, it is likely that the concentration reaching the neu- rones in the cultured hippocampus would be greatly DISCUSSION reduced. In contrast tetrameric AChE (G4) (3^10 U/ml) had no Cholinesterases within the organotypic culture system signi¢cant e¡ect on the growth ofhippocampal neurites in vitro. One possible explanation for this would be the The literature surrounding a ‘non-classical role’ for species di¡erence. The monomeric AChE used was

NSC 5485 6-5-02 AChE increases neurite outgrowth via voltage gated calcium channels 653

Fig. 4. Serum-free medium supplemented with 3^10 U/ml G1 rAChE signi¢cantly increased neurite outgrowth ofhippocampal cultures maintained for 14 DIV, when compared to a medium control group. Nimodipine (Nim; L-VGCC blocker) was added to the culture system for 24 h prior to tissue ¢xation. Alone nimodipine had no signi¢cant e¡ect on neurite outgrowth. However, when applied with G1 rAChE (3 U/ml) nimodipine blocked the neurotrophic action. **P 6 0.01 using a Dunnett’s multiple comparisons t-test. n = 150. Results are expressed as mean neurite outgrowth (Wm) þ S.E.M. recombinant mouse whereas the tetrameric form was eel. Implications for a neurotrophic role of monomeric AChE There is 91% identity between rat and mouse AChEs, whereas the identity between eel and mouse is only Under the current culture conditions it would appear 59%. However, Holmes et al. (1995) used organotypic that the stimulatory e¡ect ofG 1 rAChE was mediated cultures ofthe substantia nigra to show that there is via the peripheral site as Fas and BW284c51, both of an e¡ect ofeel G 4 on rat neurones in culture. Clearly, which block the peripheral site, attenuated this stimula- therefore, eel AChE in a similar preparation can exert a tory e¡ect. This e¡ect was not replicated by the addition signi¢cant bioactive e¡ect. Hence the discrepancy seen ofthe active site blockers echothiophate or galanth- here, where no e¡ect was observed, is more likely to be amine, further con¢rming a non-cholinergic action of attributable to a di¡erence in the brain region (hippo- AChE. In the case ofBW284c51, at the highest concen- campus rather than substantia nigra), than in the species. tration (100 WM), there was a non-speci¢c toxic action as It has been demonstrated that AChE is released from the re£ected by an increase in LDH release and a decrease in dendrites ofneurones in the substantia nigra, indepen- neurite outgrowth. However, in the remaining range of dently ofcholinergic transmission, almost exclusively BW284c51 used, since it had no e¡ect on its own, its from the dopaminergic containing cells of the pars com- blockage ofthe AChE should be interpreted as a speci¢c pacta (Green¢eld, 1991a,b). The dopaminergic cells of blocking action. Analysis ofthe molecular composition the substantia nigra may therefore be more sensitive to ofembryonic AChE reveals that an increase in the ratio the e¡ects of‘non-cholinergic’ AChE. The e⁄cacy of ofG 4 :G1/G2 occurs during development (Inestrosa et al., PAS inhibitors, and hence the involvement ofthe PAS 1994). in this non-cholinergic action ofAChE, might also A cell adhesion function of AChE has also been cor- explain the enhanced stimulatory e¡ect ofG 1 over G4. related with the peripheral site (Jones et al., 1995; De- Bourne et al. (1999) have shown that there is a partial Farrari et al., 1998; Sharma and Bigbee, 1998; Johnson occlusion ofthe peripheral site in the tetramer that pre- and Moore, 1999). Experiments on rat spinal cord cul- sumably would not apply in the monomeric form. tures, PC12 cell lines and dopaminergic pedal ganglion In comparison addition ofBChE (3^10 U/ml) had no cells also suggest that the polypeptide domains, which signi¢cant e¡ect on the survival ofneurones in this slice form the peripheral site microdomain, mediate the tro- culture system. This ¢nding provides further proof that phic e¡ects ofAChE (Srivatsan and Peretz, 1997; this action ofAChE is not via the hydrolysis ofACh. Johnson and Moore, 1999; Mun‹oz et al., 1999). None the less, BChE has been postulated to have a role Srivatsan and Peretz (1997) demonstrated a neurotro- in development with an involvement in cell proliferation phic action for AChE in Aplysia hemolymph. Using cul- (Layer, 1991; Willbold and Layer, 1994; Mack and tured dopaminergic neurones from the Aplysia, Robitzki, 2000). AChE and BChE are expressed at di¡er- exogenously applied hemolymph caused an increase in ent stages ofdevelopment and their expression is mutu- neurite outgrowth which was subsequently blocked ally exclusive: thereforethe lack ofan e¡ect ofBChE on using AChE inhibitors ofthe catalytic and peripheral this slice culture system does not preclude a developmen- sites but not by a catalytic site blocker alone. Grifman tal role for BChE elsewhere or at other times in develop- et al. (1998) used antisense suppression ofAChE on ment (Drews, 1975; Layer, 1991; Mack and Robitzki, PC12 cells to demonstrate redundant neurite out- 2000). growth-promoting activities for AChE and implicate

NSC 5485 6-5-02 654 T. Day and S. A. Green¢eld

AChE-like proteins as potential mediators ofcytoarchi- ing ofmRNA and ligand occupation ofthe dihydropyr- tectural changes supporting neuritogenesis. These papers idine receptor. Later studies attempted to ascertain the support the current data presented in this study and we importance ofthe signalling mechanism during muscle further demonstrate a possible mechanism for the action development (Luo et al., 1994, 1996). Such studies indi- ofAChE in neurite outgrowth. cate that L-VGCC play an important role in the regula- tion ofAChE expression during skeletal muscle The role of the calcium channel and neurite outgrowth development in vivo. Since AChE enhances calcium entry (Webb et al., Critical developmental processes such as appearance 1996) and since calcium can subsequently promote ofneurotransmitters and ion channels, neurite out- AChE mRNA expression (Luo et al., 1994, 1996; Decker growth, synaptogenesis and intrinsic ¢ring patterns all and Berman, 1990), it is possible that there could be a depend in part on developmental changes in voltage-de- feed-forward mechanism involving neurite outgrowth, pendent Ca2þ conductances (Lipton and Kater, 1989; calcium and AChE. Heng et al., 1999). Regulation ofneuronal Ca 2þ channels is important because entry ofCa 2þ through VGCC is a major mechanism by which changes in membrane poten- CONCLUSION tial can in£uence cellular processes (Shitaka et al., 1996). Ca2þ a¡ects the assembly states ofboth microtubules The transient expression ofAChE during the develop- and micro¢laments and is a prime candidate as an intra- ment ofmany brain regions has been correlated with cellular regulator ofneurite elongation and motile stages ofneurite outgrowth. (Dupree and Bigbee, 1994; growth cone structures (¢lopodia and lamellipodia). Karpel et al., 1996; Holmes et al., 1997; Bataille et al., Many neurotrophins such as NGF, brain-derived neu- 1998; Sternfeld et al., 1998). AChE has also been postu- rotrophic factor and basic ¢broblast growth factor have lated in a role ofaxonal path¢nding and synaptogenesis been shown to exert their trophic action by their ability (Anderson and Key, 1999), while high expression of to regulate intracellular Ca2þ via the VGCC (Black, AChE in migrating germ cells is lost after their target 1999; Defazio et al., 2000; Gysbers et al., 2000; Kang is reached (Drews, 1975). and Schuman, 2000). AChE has been implicated in dif- In conclusion, the current ¢ndings provide further evi- ferentiationand survival ofneuronal cells ofvarious dence for a non-classical trophic role for AChE and brain regions. Ca2þ is a critical signal in neuronal devel- suggests a role for the exposed peripheral site on the opment and electrophysiological studies suggest that G1 form of AChE. Moreover, for the ¢rst time this AChE may cause Ca2þ in£ux (Webb et al., 1996), study demonstrates that only the L-type VGCC channel which could in turn result in altered growth cone activity has a role in this action, adding powerful support to the during development (Kater and Mills, 1991). In our idea that the trophic action ofAChE is via Ca 2þ in£ux. hands the trophic action ofAChE was attenuated by We still have much to understand concerning the co-application ofAChE with nimodipine, the L-type non-cholinergic action ofAChE, yet identi¢cation of VGCC blocker. Application ofAChE with other respec- the mechanism as provided here might lead to more tive blockers ofother VGCC failedto block this selective manipulation ofthe phenomenon, and hence non-classical e¡ect. Hence it appears that AChE exerts insights into its physiological and pathological implica- a non-cholinergic trophic action, selectively via L-type tions. VGCC. This ¢nding is in agreement with that ofLuo et al. (1994). Early studies on avian pectoral muscle cultures have AcknowledgementsöWe wish to thank Drs Palmer Taylor and shown that the dihydropyridine Ca2þ channel blocker Joan Kanter (Department ofPharmacology, Univeristy ofCal- nifedipine caused a reduction in AChE secretion rate ifornia, San Diego, CA, USA) for proving the puri¢ed recombi- nant monomeric AChE and Dr Jane Loughlin (Open (Decker and Berman, 1990). Results indicated a reduc- University, School ofBiological Science, UK) forthe tetrameric tion in AChE biosynthesis and a link in skeletal muscle AChE. This project was supported by funds from Synaptica between transcription and post-transcriptional process- Limited.

REFERENCES

Anderson, R.B., Key, B., 1999. Role ofacetylcholinesterase in the development ofaxon tracts within the embryonic vertebrate brain. Int. J. Dev. Neurosci. 17, 787^793. Appleyard, M.E., 1994. Non-cholinergic functions of acetylcholinesterase. Biochem. Soc. Trans. 22, 749^755. Arimatsu, Y., Miyamoto, M., 1991. Survival-promoting e¡ect ofNGF on in vitro septohippocampal neurons with cholinergic and GABAergic phenotypes. Dev. Brain Res. 58, 189^201. Balasubramanian, A.S., Bhanumathy, C.D., 1993. Noncholinergic functions of cholinesterases. FASEB J. 7, 1354^1358. Bataille, S., Portalier, P., Coulon, P., Ternaux, J., 1998. In£uence ofacetylcholinesterase on embryonic spinal rat motoneurones growth in cultur e: a quantitative morphometric study. Eur. J. Neurosci. 10, 560^572. Black, I.B., 1999. Trophic regulation ofsynaptic plasticity. Neurobiology 41, 108^118. Bourne, Y., Taylor, P., Bougis, P.E., Marchot, P., 1999. Crystal structure ofmouse acetylcholinesterase. A peripheral site-occluding loop in a tetrameric assembly. J. Biol. Chem. 274, 2963^2970.

NSC 5485 6-5-02 AChE increases neurite outgrowth via voltage gated calcium channels 655

Cherksey, B.D., Sugimori, M., Llinas, R.R., 1991. Properties ofcalcium channels isolated with spider toxin, FTX. Ann. NY Acad. Sci. USA 635, 80^89. Dang, M.H., Birchler, F., Wintermantel, E., 1997. Toxicity screening ofbiodegradable polymers (2). Evaluation ofcell culture test with medium extract. J. Environ. Polymer Degrad. 5, 49^56. Darboux, I., Barthalay, Y., Piovant, M., Hipeau-Jacquotte, R., 1996. The structure function relationships in Drosophila neurotactin show that cholinesterasic domains may have adhesive properties. EMBO J. 15, 4835^4843. Decker, M.M., Berman, H.A., 1990. Dihydropyridine receptor regulation ofAChE biosynthesis. J. Biol. Chem. 265, 11796^11803. De-Farrari, G.V., Bernhardi, R.V., Caldero¤n, F.H., Luza, S.C., Inestrosa, N.C., 1998. Responses induced by in neuronal and non-neuronal cell lines. J. Neurosci. Res. 52, 435^444. Defazio, R.A., Pong, K., Knusel, B., Walsh, J.P., 2000. Neurotrophin-4/5 promotes dendritic outgrowth and calcium currents in cultured mesencephalic dopamine neurons. Neuroscience 99, 297^304. Drews, U., 1975. Cholinesterase in embryonic development. Prog. Histochem. Cytochem. 7, 1^52. Dupree, J.L., Bigbee, J.W., 1994. Retardation ofneuritic outgrowth and cytoskeletal changes accompany acetylcholinesterase inhibitor treatmen t in cultured rat dorsal ganglion neurons. J. Neurosci. Res. 39, 567^575. Ellman, G.L., Courtney, K.D., Andres, V., Featersone, K.M., 1961. A new and rapid colorimetirc determination ofacetylcholinesterase activity. Biochem. Pharmacol. 7, 88^95. Ga«hwiler, B.H., 1981. Organotypic monolayer cultures ofnervous tissue. J. Neurosci. Methods 4, 329^342. Ga«hwiler, B.H., 1984. Development ofthe hippocampus in vitro: cell types, synapses and receptors. Neuroscience 11, 751^760. Green¢eld, S.A., 1984. Acetylcholinesterase may have novel functions in the brain. Trends Neurosci. 7, 364^368. Green¢eld, S.A., 1991a. A non-cholinergic function for AChE in the substantia nigra. In: Massoule, J., Doctor, B., Bacou, F. (Eds.), Cholines- terases, Structure, Function, Mechanism, Genetics and Cell Biology. Plenum, Dordrecht, pp. 366^370. Green¢eld, S.A., 1991b. A non-cholinergic action ofacetylcholinesterase (AChE). In the brain: fromneuronal secretion to the generation of movement. Cell. Mol. Neurobiol. 11, 55^77. Green¢eld, S.A., 1992. Acetylcholinesterase as a modulatory neuroprotein and its in£uence on motor control. In: Sha¡erman, A., Velan, B. (Eds.), Multidisciplinary Approaches to Cholinesterase Functions. Plenum, Dordrecht, pp. 233^241. Grifman, M., Galyam, N., Seidman, S., Soreq, H., 1998. Functional redundancy of acetylcholinesterase and neuroligin in mammalian neurito- genesis. Proc. Natl. Acad. Sci. USA 95, 13935^13940. Gupta, J.J., Bigbee, J.W., 1992. Substratum-induced modulation ofacetylcholinesterase activity in cultured dorsal root ganglion neurons. J. Neurosci. Res. 31, 454^461. Gysbers, J.W., Guarnieri, S., Mariggio, M.A., Pietrangelo, T., Fano, G., Rathbone, M.P., 2000. Extracellular guanosine 5P triphosphate enhances nerve growth factor-induced neurite outgrowth via increases in intracellular calcium. Neuroscience 96, 817^824. Hagiwara, S., Ozawa, S., Sand, O., 1975. Voltage clamp analysis oftwo inward current mechanisms in the egg cell membrane ofa star¢sh. J. Gen. Physiol. 65, 617^644. Heng, J.E., Zurakowski, D., Vorwerk, C.K., Grosskreutz, C.L., Dreyer, E.B., 1999. Cation channel control ofneurite morphology. Dev. Brain Res. 113, 67^73. Holmes, C., Jones, S.A., Budd, T.C., Green¢eld, S.A., 1997. A non-cholinergic, trophic action ofrecombinant acetylcholinesterase on mid-brain dopaminergic neruons. J. Neurosci. Res. 49, 1^12. Holmes, C., Jones, S.A., Green¢eld, S.A., 1995. The in£uence oftarget and non-target brain regions on the development ofmid-brain dopamin- ergic neurons in organotypic slice culture. Dev. Brain Res. 88, 212^219. Hsu, S.M., Raine, L., Fanger, H., 1981. Use ofavidin biotin peroxidase complex (ABC) in immunoperoxidase techniques. J. Histochem. Cytochem. 29, 577^580. Inestrosa, N.C., Moreno, R.D., Fuentes, M., 1994. Monomeric amphiphilic forms of acetylchoilnesterases appear early during brain development and may correspond to biosynthetic precursors ofthe amphiphilic G 4 forms. Neurosci. Lett. 173, 155^158. Johnson, G., Moore, S.W., 1999. The adhesion function of acetylcholinesterase is located at the peripheral anionic site. Biochem. Biophys. Res. Commun. 258, 758^762. Jones, S.A., Holmes, C., Budd, T.C., Green¢eld, S.A., 1995. The e¡ect ofacetylcholinesterase on the outgrowth ofdopaminergic neurones in organotypic slice culture ofthe rat mid-brain. Cell Tissue Res. 279, 323^330. Kang, H., Schuman, E.M., 2000. Intracellular calcium signalling is required for neurotrophin-induced potentiation in the adult hippocampus. Neurosci. Lett. 282, 141^144. Karpel, R., Sternfeld, M., Ginzberg, D., Guhl, E., Graessmann, A., Soreq, H., 1996. Overexpression of alternative human acetylcholinesterase forms modulate process extensions in cultured glioma cells. J. Neurochem. 66, 114^123. Kater, S.B., Mills, L.R., 1991. Regulation ofgrowth cone behaviour by calcium. J. Neurosci. 11, 891^899. Layer, P.G., 1991. Cholinesterases during development ofthe avian nervous system. Cell. Mol. Neurobiol. 11, 7^33. Layer, P.G., Weikert, T., Alber, R., 1993. Cholinesterases regulate neurite growth ofchick nerve cells in vitro by means ofa non-enzymatic mechanism. Cell Tissue Res. 273, 219^226. Lipton, S.A., Kater, S.B., 1989. Neurotransmitter regulation ofneuronal outgrowth, plasticity and survival. Trends Neurosci. 12, 265^270. Luo, Z.D., Fuentes, M., Taylor, P., 1994. Regulation ofacetylcholinesterase mRNA stability by calcium during di¡erentiation frommyoblasts to myotubes. J. Biol. Chem. 269, 27216^27223. Luo, Z.D., Pincon-Raymond, M., Taylor, P., 1996. Acetylcholinesterase and nicotinic expression diverge in muscular dysgenic mice lacking the L-type calcium channel. J. Neurochem. 67, 111^118. Mack, A., Robitzki, A., 2000. The key role ofbutyrylcholinesterase during neurogenesis and neural disorders: an antisense-5 P butyrylcholinester- ase-DNA study. Prog. Neurobiol. 60, 607^628. Mattson, M.P., 1992. Calcium as sculptor and destroyer ofneural circuitry. Exp. Gerontol. 27, 29^49. Meir, A., Ginsburg, S., Butkevich, A., Kachalsky, S.G., Kaiserman, I., Ahdut, R., Demirgoren, S., Rahamimo¡, R., 1999. Ion channels in presynaptic nerve terminals and control oftransmitter release. Physiol. Rev. 79, 1019^1088. Mun‹oz, F.J., Aldunate, R., Inestrosa, N.C., 1999. Peripheral binding site is involved in the neurotrophic activity ofacetylcholinesterase. Neuro- Report 10, 3621^3625. Novales-Li, P., 1994. Comparison ofstandard chromatographic procedures forthe optimal puri¢cation ofsoluble human brain acetylcholinester- ase. Biomed. Chromatogr. 8, 259^266. Sa¡ell, J.L., Walsh, F.S., Doherty, P., 1992. Direct activation ofsecond messenger pathways mimics cell adhesion molecule-dependent neurite. J. Cell Biol. 118, 663^670. Sharma, K.V., Bigbee, J.W., 1998. Acetylcholinesterase antibody treatment results in neurite detachment and reduced outgrowth from cultured neurons: further evidence for a cell adhesive role for neuronal acetylcholinesterase. J. Neurosci. Res. 53, 454^464.

NSC 5485 6-5-02 656 T. Day and S. A. Green¢eld

Sharma, K.V., Koenigsberger, C., Brimijion, S., Bigbee, J.W., 2001. Direct evidence for an adhesive function in the noncholinergic role of acetylcholinesterase in neurite outgrowth. J. Neurosci. Res. 63, 165^175. Shitaka, Y., Matsuki, N., Saito, H., Katsuki, H., 1996. Basic ¢broblast growth factor increases functional L-type calcium channels in fetal rat hippocampal neurons: Implications for neurite morphogenesis in vitro. J. Neurosci. 16, 6476^6489. Silver, A., 1974. The Biology ofthe Cholinesterase. North Holland, Amsterdam. Small, D.H., Reed, G., White¢eld, B., Nurcombe, V., 1995. Cholinergic regulation ofneurite outgrowth fromisolated chick sympathetic neurons in culture. J. Neurosci. 15, 144^151. Soreq, H., Seidman, S., 2001. Acetylcholinesterase ^ new roles for an old actor. Nat. Rev. Neurosci. 2, 294^302. Srivatsan, M., Peretz, B., 1997. Acetylcholinesterase promotes regeneration ofneuritis in cultured adult neurons of Aplysia. Neuroscience 77, 921^ 931. Sternfeld, M., Ming, G., Song, H., Sela, K., Timberg, R., Poo, M., Soreq, H., 1998. Acetylcholinesterase enhances neurite growth and synapse development through alternative contributions ofits hydrolytic capacity, core protein, and variable C termini. J. Neurosci. 18, 1240^1249. Tsigelny, I., Shindyalov, I.N., Bourne, P.E., Sudhof, T.C., Taylor, P., 2000. Common EF hand motifs in cholinesterases and neuroligins suggest a role for Ca2þ binding in cell surface associations. Protein Sci. 9, 180^185. Usowicz, M., Sugimori, M.M.M., Cherksey, B., Llinas, R., 1992. P-type calcium channels in the somata and dendrites ofadult cerebellar Purkinje cells. Neuron 9, 1185^1199. Webb, C.P., Nedergaard, S., Giles, K., Green¢eld, S.A., 1996. Involvement ofthe NMDA receptor in a non-cholinergic action ofacetylcholin- esterase in guinea-pig substantia nigra pars compacta neurons. Eur. J. Neurosci. 8, 837^841. Willbold, E., Layer, P.G., 1994. Butyrylcholinesterase regulates laminar retinogenesis ofthe chick embryo in vitro. Eur. J. Cell Biol. 64, 192^199.

(Accepted 16 January 2002)

NSC 5485 6-5-02