Eur. J. Biochem. 219, 155-159 (1994) 0 FEBS 1994

Trp279 is involved in the binding of quaternary ammonium at the peripheral site of Torpedo marmorata acetylcholinesterase Isabelle SCHALK', Laurence EHRET-SABATIER', FranGoise BOUET2,Maurice GOELDNER' and Christian HIRTH'

' Laboratoire de Chimie Bio-organique, URA

1386 CNRS, UniversitC Louis Pasteur Strasbourg, FacultC de Phamacie, France

* Service de Biochimie, Laboratoire d'inginierie des protCines, Centre d'Etudes NuclCaires Saclay, Gif-sur-Yvette, France (Received July 6/0ctober 7, 1993) - EJB 93 1005/2

Specific photoaffinity labelling of purified acetylcholinesterase from Torpedo murmorutu by pN,N-['H]dirnethylamino benzenediazonium and p-N,N- ['HH]dibutylamino benzenediazoniurn derivatives was demonstrated. This occured at the active site of the enzyme for lower concentrations of the probes and at the peripheral ammonium binding site for higher concentrations. The affinities and the rate constants of alkylation for each probe on both sites have been established. Specific labelling at the peripheral site of the enzyme with both probes allowed the identification of radiolabelled peptides having the common sequence K270PQELIDVEW. The radioactivity was always associated with the residue Trp279 indicating the preferential ammonium complexation with this aromatic residue.

The recent description of the three-dimensional structure of the Torpedo caEifomicu acetylcholinesterase (AChE) [11 has aided our understanding of the functioning of this enzyme. In particular, the catalytic triad Ser200/His440/Glu327 has been suggested to be localized at the bottom of a 2-nmdeep cleft named the aromatic gorge since its lining is mainly composed of aromatic amino acid residues. The anionic subsite, i.e. the binding site of the quaternary ammonium of acetylcholine, was located in close proximity to Trp84 and Phe330. This localization was in full agreement with former irreversible labelling studies which identified both Trp84 [2] and Phe330 [3,4] as being involved in the binding of quaternary ammonium. Recent results showing the crystal structures of three enzyme complexes with edrophonium, tacrine and decamethonium [ 5 ] , confirm this picture, showing a close interaction of the different ammonium groups with both Trp84 and Phe330. Interestingly, the only major conformational difference between the three complexes is the orientation of the phenyl ring of Phe330. The decamethonium complex showed an additional interaction of the second ammonium group with another aromatic residue, Trp279, located at the top of the aromatic gorge. The labelling of Trp279, in addition to Phe330, by the aromatic ['H]dimethylamino benzenediazonium salt has already been described [4]. The observed mutually exclusive labelling of these two residues demonstrated a functional connection between these two ammonium-binding sites. The localization of Trp279 within the peripheral regulatory ammonium-binding site corroborates other labelling experiments which identified His280 [6] and Correspondence to L. Ehret-Sabatier, Laboratoire de Chimie Bio-Organique, UniversitC Louis Pasteur, FacultC de Pharmacie, 74 route du Rhin, BP 24, F-67401 Illkirch Cedex, France Fax: +33 88 67 88 91. Abbreviations. AChE, acetylcholinesterase; Me,NF'hN,-R, dimethylamino benzenediazonium probe; Bu,NPhN,-R, dibutylamino

benzenediazonium probe. Enzyme. Acetylcholinesterase (EC 3.1.1.7).

peptide fragments in the vicinity of this residue [7, 81. The present work describes the photoaffinity labelling of residue Trp279 of the purified torpedo AChE by using tritiated p-dialkylamino benzene diazonium probes dimethylamino (Me,NPhN,-R) and dibutylamino (Bu,NPhN,-R) derivatives. Furthermore, it demonstrates that residue Trp279 belongs to the peripheral site and more precisely participates in ammonium-binding within this domain.

MATERIALS AND METHODS Chemicals 5,5'-Dithiobis 2-nitrobenzoic acid, dithiothreitol, tetramethylammonium bromide and acetylthiocholine iodide were purchased from Aldrich Chemicals, edrophonium chloride and propidium iodide from Sigma, sequencing-grade trypsin from Boehringer. Dimethylamino and dibutylamino salts and [3H]Me,NPhN,-R and [3H]Bu,NPhN,-R were synthesized as described [9].

AChE preparation AChE was obtained from electric organs of Torpedo marmorutu (Marine Biological Station, Arcachon), solubilized by a mild proteolytic procedure and purified by affinity chromatography [9]. Activity was assayed spectrophotometrically on acetylthiocholine [101. Concentration of protein in the AChE preparation was estimated using the Biorad assay [ll].Active and peripheral sites were titrated spectrofluorimetrically using N-methylacridinium 1121 and propidium [ 131, respectively.

Photolabelling of AChE AChE was diluted in 50 mM potassium phosphate, pH 7.2, in the presence of various concentrations of Me,WhN,-

156 R (3-200 pM) or Bu,NPhN,-R (5-500 pM). The final assay volume was 1 ml and the protein concentration 50 pg/ ml. The mixture was introduced into a quartz cell and irradiated at 295 nm (50 pV) at 10°C with gentle stirring. Aliquots of 5 pl were taken at various times and diluted (2000X) for enzyme activity determination using Ellman’s assay. Rates of inactivation were corrected for the loss of activity of a solution of AChE irradiated without any inhibitor (control experiment). To study the specific labelling of the peripheral site of AChE, a constant concentration of 5 pM edrophonium was added to the incubation mixture. In this case the control experiment was a solution of AChE irradiated in the presence of 5 pM edrophonium. Inactivations were analyzed according to the kinetics scheme described earlier [9]. To study the edrophonium protection pattern of AChE against the labelling by Me,NPhN,-R, the enzyme was irradiated in the presence of 40 pM Me,NPhN2-R and various concentrations of edrophonium (from 50 nM to 1 &). The affinities, Kd’,of edrophonium for the protected sites were determined as described [14, 151.

[3H]Bu,NPhN,-R-labelled enzyme was denaturated, alkylated with iodoacetamide and cleaved by CNBr [161.Dried CNBr digests were resuspended in 4 M urea containing 0.1% trifluoroacetic acid, 3% CH,CN and 6% propanol and loaded (up to 500 pg) onto a reverse-phase HPLC column (Brownlee RP 300 Aquapore, 250 mmX4.6 mm). Peptides were eluted at a flow rate of 1 mVmin by a gradient of 0.1% trifluoroacetic acid (solution C) to 30% CH,CN, 60% propanol, 0.1% trifluoroacetic acid (solution D) as follows: from 5% to 65% solution D in 60min then to 100% in 10min. Fractions containing radioactive peptides were collected and lyophilized. Residues were redissolved in 0.1 M Tris/HCl pH 8.5 (final concentration 2.5 mg/ml) and submitted to proteolysis using trypsin. After 24 h incubation, digests were loaded directly onto the Vydak C,, column. Peptides were eluted at a flow rate of 1 ml/min by a gradient of 0.06% trifluoroacetic acid (solution E) to 80% CH,CN, 0.052% trifluoroacetic acid (solution F j as follows: from 5% to 35% solution F in 10 min, then to 80% solution F in 71 min.

Protein sequencing Residual peripheral sites titration of AChE labelled by Me,NPhN,-R AChE diluted (500 pg/ml) in 50 mM potassium phosphate, pH 7.2 was irradiated at 295 nm, E of 100 pV at 10°C in the absence (control experiment) or presence of 0.1 mM Me,NPhN,-R. Inactivation was monitored using Ellman’s assay. After 20 min irradiation the solutions were dialyzed at 4°C in the dark against 50 mM Tris/HCl, pH 8.0, to remove the excess of Me,NPhN,-R. The fluorescence of propidium (0.2-5.0 pM) was then measured in this buffer (final volume 600 p1) at 20°C in the presence of 175 pg dialyzed AChE.

HPLC fractions were lyophilized and redissolved in 70% formic acid. Automated Edman degradation was carried out on Applied Biosystems ABI 477 A liquid-phase sequencer, using standard cycle programs provided by the manufacturer. Aliquots of the phenylthiohydantoin-amino acids released from the sequencer at each cycle were analyzed on-line using a ABI 120A reverse-phase HPLC system. The remaining sequencer output from each cycle was used to determine radioactivity.

RESULTS Titration of active and peripheral sites

Peptide maps of AChE labelling by [3H]Me,NPhN,-R and [3H]Bu,NPhN,-R AChE diluted (300 pg/ml) in 50 mM potassium phosphate pH 7.2, was irradiated at 295 nm (100 pV) and 10°C in the presence of 0.1 mM [“H]Me,NPhN,-R (S.OCi/mmol) for 20 min or 20 pM [’H]Bu,NPhN,-R (0.8Ci/mmol) for 30min, in a final volume of 2ml. These operations were repeated in the presence of a protective agent (0.1 M NMe,Br, 5 pM or 20 pM edrophonium). After irradiation, dithiothreitol was added to a final concentration of 10 mM to destroy unreacted [’H]Me,NPhN,-R or [’H]Bu,NPhN,-R. Extensive dialysis against 50 mM potassium phosphate, pH 7.2 was followed by lyophilization. rH]Me,NPhN,-R labelled enzyme (1.1 mg proteidml) was denaturated in 2 M uredO.1 M Tris/HCl, pH 8.5, and submitted to proteolysis for 24 h at 37°C with a 1 : 20 (by mass) trypsidAChE ratio. Tryptic digests (up to 1.1 mg) were loaded directly onto a reverse-phase HPLC column (Vydak C,8, 250mmX4.6mm). Peptides were eluted by a linear gradient from 0 to 80% CH,CN in 0.1 % trifluoroacetic acid for 80 min, at a flow rate of 1.5 ml/min. Fractions containing radioactive peptides were collected and lyophilized. Residues were resuspended in 0.1 % heptafluorobutyric acid, 4.5% propanol, 9% CH’CN (solution A) and subjected separately to reverse-phase HPLC on the same column equilibrated in (solution A). Peptides were fractioned by a linear gradient of 0-70% of solution B (0.1% heptafluorobutyric acid, 30% propanol-1, 60% CH,CN) in 140min, at a flow rate of 1 ml/min (total recovery of injected radioactivity).

The homogeneity of the AChE preparation was ascertained, before the labelling, by titration of both the active and the peripheral sites The use of N-methylacridinium yielded 9.6pmol active sites/pg purified AChE (with Kd of 3.3 lO-’M) while the use of propidium yielded 10.3 pmol peripheral sites/pg purified AChE (with Kd of 2.4 lO-’Mj. These titrations establish that the binding sites of AChE were not altered during the extraction and purification procedures and in particular that the active and the peripheral sites remain present in a 1 : 1 ratio.

Photolabelling of the peripheral site Me,NPhN,-R and Bu,NPhN,-R are efficient photoaffinity ligands for the ammonium-binding sites of Torpedo AChE [9]. This photolabelling can be directed either towards the active (near the catalytic serine) or the peripheral binding site. In particular, the use of edrophonium as a protective agent allowed for discrimination between these two sites in terms of the affinity of edrophonium for them: Kd of 2.5 lO-’M for the active site and Kd greater than 10-4M for the peripheral site [13]. Table 1 shows the affinities (Kd)and the rate constants ( k ) , related to both sites, deduced from the labelling experiments using Me,NPhN,-R and Bu,NPhN,-R as photoprobes for the purified AChE. Me,NPhN,-R showed a fivefold higher affinity for the active site than for the peripheral site (2.0 10-5M and 1.1 10-4M, respectively), while Bu,NPhN,-R exhibited only twofold difference (3.5 M and 7.8 lo-”). The rate constants indicate a faster labelling

157 Table 1. Kinetics constants of AChE photoaffinity labelling by Me,NPhN,-R and Bu,NPhN,-R. Purified AChE was irradiated at 295 nm (50 pV) and 10°C with different concentrations of Me,NPhN,-R or Bu,NPhN,-R. The specific labelling of the peripheral site of AChE was studied in the presence of a constant concentration of 5 pM edrophonium. Inactivations were analyzed according to the kinetics scheme described earlier [9] to determine the kinetics constants related to each site. Site labelled

K, for

k for

Me,NPhN,-R

BuZNPhNZ-R

Me,NPhN,-R

M

min-’

2.0 10-5 1.1 10-4

Active site Peripheral site

3.5 10-5 7.8 10-5

Table 2. Peripheral-site labelling. AChE was irradiated at 295 nm (SO pV) in the presence of 5 pM edrophonium with different concentrations of Me,NPhN,-R or Bu,NPhN,-R and for 20min or 30 min, respectively. ~

Probe concentrations

Bu,NPhN,-R

0.212 0.093

04 OV5

0.099 0.073

‘

Extent of peripheral site labelling with Me,NPhN,-R

Bu,,NPhN,-R ~-

%

7.0 M 4.0 lo-’ M 1.7 10-4 M

0 20 62

22 36 81

of the active site by Me,NPhN,-R (0.212 min-l and 0.093 min-’), while Bu,NPhN,-R labels both sites at similar rates (0.099 min-’ and 0.073 min-l). The relative extent to which the peripheral site is irreversibly labelled as a function of the concentration of the Me,NPhN,-R and Bu,NPhN,-R is shown in Table 2 . These values were deduced by conducting the labelling experiments in the presence of 5 pM edrophonium, a concentration at which a complete but selective protection of the active site is observed. We previously reported that the pattern of edrophonium protection against the inactivation of AChE was monophasic for Me,NPhN,-R (10 pM) [9] and biphasic for Bu,NPhN,-R (20 pM) [15]. We now demonstrate, according to the concentration dependence (Table 2) that the protective effect of edrophonium is also biphasic (data not shown) for Me,NPhN,-R at the higher concentration of 40 pM, with apparent dissociation constants for edrophonium of 1.5 lO-’M and 3.8 M. These values are close to the described affinity of edrophonium for the active site and the peripheral site, respectively [131. Up to now, the photolabelling of AChE peripheral site has been monitored by the loss of hydrolytic activity (Ellman’s assay) which occurs at the active site. To measure the direct effect at the peripheral site, we studied the propidiumbinding properties of the enzyme. Fig. 1 shows the progressive saturation of AChE peripheral sites by propidium. The difference observed between the control level PI (AChE irradiated without ligand, no loss of hydrolytic activity) and the level P, (AChE photolabelled by Me,NPhN,-R) establishes that 39% of peripheral sites were alkylated. This value can be compared to the 33% of peripheral site photoinactivation obtained with Me,NPhN,-R at an identical concentration. This result shows that the labelling of the peripheral site

0

1

2 3 [propidium] (pM)

4

5

Fig. 1. Residual peripheral site titration of AChE labelled by Me,NPhN,-R. AChE was irradiated for 20 min at 295 nm in the absence (0)or presence (W) of 0.1 mM Me,NPhN,-R. After dialysis the fluorescence of propidium was measured in the presence of 175 pg irradiated AChE.

involves a loss of propidium-binding ability concomitant with the loss of acetylthiocholine-hydrolytic activity.

Analysis of [3H]Me,NPhN,-R and [3H]Bu,NPhN,-R labelled peptides Purified Torpedo AChE was irradiated on a larger scale in the presence of [3H]Me,NPhN,-R or [3H]Bu,NPhN,-R, then cleaved by trypsin and/or CNBr (see Material and Methods) and analyzed by HPLC. Figs 2 and 3 show the HPLC profiles obtained with 100 pM [3H]Me,NPhN,-R and 20 pM [’H]Bu,NPhN,-R, respectively. The addition of 5 pM edrophonium or 0.1 M NMe,Br to the irradiation medium allows the discrimination of radiolabelled peptides belonging to the active or to the peripheral site. The HPLC profile resulting from the [3H]Me,NPhN,-R labelling experiment (Fig. 2) shows, besides radioactivity incorporation due to the residual probe (quoted 0), eight radioactive peaks (1 -7, Fig. 2A), totally protectable by NMe,Br. Among them, radioactivity incorporation was not prevented by 5 pM edrophonium in peaks 4 and 5 (Fig. 2B) and this labelling can thus be attributed to peptides belonging to the peripheral site. In fact, a detailed study using increasing concentrations of [’H]Me,NPhN,-R (from 3.6 pM to 100 pM) clearly correlated the extent of labelling of the peripheral site with the progressive appearance of peaks 4 and 5 in the HPLC profiles (data not shown). Finally, from the labelling experiment using [‘H]Me,NPhN,-R at 100 pM, we purified peaks 4 and 5 using a different chromatographic elution system, then identified the labelled amino acid residue(s) by sequencing.

158 80

0

20

40

60

RO

100

u

120

140

160

1

0

F

10

20

40

30

A

50

60

120

70

B

5

0

20

40

60

80 100 fraction number

120

140

160

Fig.2. Reverse-phase HPLC profile of trypsin digests of [3H]MezNPhNz-R-labelled AChE. AChE was irradiated at 295 nm in the preswith 100 pM [3H]Me,NPhN,-R (5ci/mmol; 2A -) Laence of 5 pM edrophonium (2B) or 0.1 M NMe,Br (2A -). belled AChE was then dialyzed and proteolyzed using trypsin. Resulting peptides (9Opg protein) were analyzed by HPLC using a Vydak C,, column equilibrated in 0.1% trifluoroacetic acid. The column was eluted at a flow rate of 1.5 mumin by a linear gradient from 0 to 80% CH,CN in 80 min. Aliquots (200 pl) from 750-p1 fractions were taken for radioactivity measurements. Recoveries of injected radioactivity were 90-95% in the three cases.

Fig. 3A shows the profile obtained after a CNBr cleavage of [3H]Bu,NPhN,-R labelled AChE and exhibits several large radioactive peaks (A-F). Peptides belonging to the peripheral site can be localized in two peaks, D (unaffected by edrophonium) and F (partially protected). Only peak F was purified and submitted to trypsinolysis. The analysis of the peptide map (Fig. 3B) showed three peaks belonging specifically to the peripheral site (Fa, F% and Fc) which were purified and sequenced. For all the peptides belonging to the peripheral site and purified either from [3H]Me,NPhN,-R or from [’H]Bu,NPhN,-R labelled AChE, the common sequence JCPQELIDVEW corresponding to residues 270 -279 was obtained. In all cases, the radioactivity was associated with the residue Trp279.

DISCUSSION Photoaffinity labelling experiments constitute valuable tools to investigate the structure of receptor-binding sites by identifying the labelled amino acids. Two structurally related aryldiazonium salts (compounds Me,NPhN,-R and Bu,NPhN,-R) have been previously described as efficient photoaffinity probes for Torpedo AChE by use of an energytransfer-mediated photoactivation process [9]. To take advantage of these probes to localize the ammonium-binding sites on the enzyme, we used purified enzyme from the electric

0

20

40

60

80

100

120

140

160

fraction number

Fig.3. Reverse-phase HPLC of CNBr and trypsin digests of [3H]BuJVl”PN,-R labelled AChE. (A) AChE was irradiated at 295 nm with 20 pM [3H]Bu,NPhN,-R (0.8Ci/mmol; ---) in the presence of 20 pM edrophonium (-) or 0.1 M NMe,Br (-). After dialysis and CNBr treatment the resulting peptides (80 pg protein) were analyzed by HPLC using a Brownlee RP300 reversephase column eluted at a flow rate of 1 mVmin by a gradient of 0.1% trifluoroacetic acid to 0.1% trifluoroacetic acid, 60% propanol, 30% CH,CN as described in Materials and Methods. Fractions of 1ml were taken for radioactivity measurements. The recoveries of injected radioactivity were 8 5 9 5 % in the three cases. (B) Peak F was proteolyzed using trypsin. The resulting peptides [irradiation : (- --), without protective ligand; (-), with edrophonium] were loaded directly onto a Vydak C,, column and eluted at a flow rate of 1 mumin by a gradient of 0.06%trifluoroacetic acid to 80% CH,CN, 0.052% trifluoroacetic acid as described in Materials and Methods. Fractions of 0.5 ml were taken for radioactivity measurements.

organ of 7: marmorata with active and peripheral sites functional and present in an identical ratio. Kinetic analyses of the photochemical inactivation experiments allowed the determination of the affinities and the rate constants for Me,NPhN,-R and Bu,NPhN,-R, for both active and peripheral sites (Table 1). The concentration dependence of the labelling of the peripheral site for both probes (Table 2) show that at lower concentrations only Bu,NPhN,-R alkylates this site. These analyses could be achieved by using as a protective ligand 5 pM edrophonium, a concentration at which a complete but selective occupancy of the active site was observed. These results indicate that Me,NPhN,-R is better able to label the active site, as indicated by its better affinity and a higher rate constant, while Bu,NPhN,-R is less discriminative between the two sites. Finally, the effect of edrophonium concentration on the inactivation showed a biphasic relationship for both probes, reflecting an alkylation process occuring at two different sites. This interaction could be demonstrated to occur at the active site or at the peripheral site, according to the deduced apparent dissociation constants of edrophonium. As direct evidence for the peripheral-site labelling in the presence of higher concentration of Me,NPhN,-R M), we deter-

159 mined by fluorescence measurements the loss of propidium binding (39%, Fig. 1). The concomittant loss of hydrolytic activity (33%) indicates that the alkylation at the peripheral site by Me,NPhN,-R prevents both the binding of propidium at this site and the binding of substrate at the active site. The irreversible labelling study with the diazonium probes was further pursued by identifying the labelled amino acids in order to localize the peripheral site on the primary structure of the enzyme. This study was possible by performing appropriate protection experiments, i.e. by using 0.1 M tetramethylammonium for a total protection of both sites and 5 pM edrophonium for selective protection of the catalytic site. The peptide analyses of the labelled fragments belonging to the peripheral site, by either Me,NPhN,-R or Bu,NPhN,-R, (Figs 2 and 3) resulted in a common decamer sequence, 270 -279, KPQELIDVEW, carrying the radioactivity on the last amino acid, Trp279. The position of this amino acid at the top of the catalytic gorge [l]was described recently to be in a proximal position to the second ammonium group of decamethonium [5],corroborating this labelling study. Aromatic diazonium salts constitute photosensitive analogs of acetylcholine and have been extensively used to characterize the binding site of the cholinergic ammonium moiety from the AChE active site [3,4] or from the nicotinic receptor [ 16, 171. Again, an aromatic hydrophobic residue seems best adapted to complex quaternary ammonium groups [18]. We have established in this study, that the labelling of Trp279 is not the result of energy-transfer-mediated photoactivation by using a standard photoactivation methodology, which resulted in an identical labelling pattern of peptides (data not shown). The labelling of an unique residue within this peripheral acetylcholine-binding site contrasts clearly the observed labelling of the nicotinic receptor with the same Me,NPhN,-R. This study indicates that,within the AChE peripheral site, a preferential interaction occurs between the ammonium moiety of acetylcholine and Trp279. Such a close spatial interaction between the quaternary ammonium and Trp279 was in fact observed in the crystal stntcture of the complex with decamethonium [ 5 ] . This residue is likely to play a relevant role in the functionning of the enzyme as indicated by mutagenesis experiments, i.e. the point mutation of Trp279Ala in Torpedo [19] or human [20] cholinesterase reduced its sensitivity to propidium while the inhibitory effect of edrophonium was not affected. In conclusion, the present labelling study demonstrated not only the localization of Trp279 to the peripheral site but also defines it as the preferential residue at which ammonium binds within this domain. This work was supported by the Centre National de la Recherche Scientifque, the Ministire de la Recherche et de la technologic, the Association Franco-Isradlienne pour la Recherche Scientijique et Technologique, the Association pour les Myopathes, the

Fondation pour la Recherche MLdicale and the Fondation de France.

REFERENCES 1. Sussman, J. L., Harel, M., Frolow, F., Oefner, C., Goldman, A., Toker, L. & Silman, I. (1992) Science 253, 872-879. 2. Kreienkamp, H. J., Weise, C., Raba, R., Aaviksaar, A. & Hucho, F. (1991) Proc. Natl Acad. Sci. USA 88, 6117-6121. 3. Kieffer, B., Goeldner, M., Hirth, C., Aebersold, R. & Chang, J. Y. (1986) FEBS Lett. 202, 91-96. 4. Schalk, I., Ehret-Sabatier, L., Bouet, F., Goeldner, M. & Hirth, C. (1992) in Multidisciplinary approaches to cholinesterase functions (Shafferman, A. & Velan, B., eds) pp. 117-120, Plenum Press, New York. 5. Harel, M., Schalk, I., Ehret-Sabatier, L., Bouet, F., Goeldner, M., Hirth, C., Axelsen, P., Silman, I. & Sussman, J. L. (1993) Proc. Nut1 Acad. Sci. USA 90,9031-9035. 6. Haas, R., Adams, E. W. & Rosenbeny, M. A. (1992) in Multidisciplinary approaches to cholinesterase functions (Shafferman A. & Velan, B., eds) pp. 131-139, Plenum Press, New York. 7. Amitai, G. & Taylor, P. (1991) in Cholinesterases: structure, function, mechanism, genetics, and cell biology (MasoulliC, J., Barnard, E. A., Chatonnet, A., Doctor, B. P. & Quinn, D. M., eds) p. 285, American Chemical Society, Washington D. C. 8. Weise, C., Kreienkamp, H. J., Raba, R., Pedak, A., Aaviksaar, A. & Hucho, F. (1990) EMBO J. 9, 3885-3888. 9. Ehret-Sabatier, L., Schalk, I., Goeldner, M. & Hirth, C. (1992) Eu,: J. Biochem. 203, 475-481. 10. Ellman, G. L., Courtney, K. D., Andres, V. & Featherstone, M. R. (1961) Biochem. Pharmacol. 7, 88-95. 11. Bradford, M. M. (1976) Anal. Biochem. 72, 248-254. 12. Mooser, G., Schulman, H. & Sigman, D. S. (1972) Biochemistry 11, 1595-1602. 13. Taylor, P. & Lappi, S. (1975) Biochemistry 14, 1989-1997. 14. Ehret-Sabatier, L., Goeldner, M. & Hirth, C. (1991) Biochim. Biophys. Acta 1076, 137-142. 15. Ehret-Sabatier, L., Goeldner, M., Hirth, C. & Rousseau, B. (1991) in Cholinesterases: structure, function, mechanism, genetics, and cell biology (Masoullit, J., Barnard, E. A., Chatonnet, A,, Doctor, B. P. & Quinn, D. M., eds) pp. 245-251, American Chemical Society, Washington D. C. 16. Dennis, M., Giraudat, J., Kotzyba-Hibert, F., Goeldner, M., Hirth, C., Chang, J. Y., Lazure, C., Chritien, M. & Changeux, J. P. (1988) Biochemistry 27, 2346-2357. 17. Galzi, J. L., Revah, F., Black, D., Goeldner, M., Hirth, C. & Changeux, J. P. (1990) J. Biol. Chem. 265, 10430-10437. 18. Dougherty, D. A. & Stauffer, D. A. (1990) Science 250,15581560. 19. Harel, M., Sussman, J. L., Krejci, E., Bon, S., Chanal, P., Massoulii J. & Silman, I. (1992) Proc. Nut1 Acad. Sci. USA 89, 10827- 10831. 20. Shafferman, A., Velan, B., Ordentlich, A., Kronman, C., Grosfeld, H., Leitner M., Flashner, Y., Cohen, s., Barak, D. & Ariel, N. (1992) EMBO J. 11, 3561-3568.