VOL. 12 (1953) CHOLINE ACETYLASE SPECIFICITY 317 



phosphate i, L-cysteine 20, CoA 0.05, acetyl phosphate 40, KjHPO^ buffer of pH 7 10. To the mixture 

 were added 0.15 ml of transacetylase of an ammonium sulfate fractionation from E. coli extracts of 

 60% saturation, and 0.05 ml of choline acetylase to be assayed. The total volume was i ml, the 

 temperature was 32 °C. After 30 min the reaction was stopped by addition of i ml of o.i A^^ HCl. After 

 destroying the excess acetyl phosphate by boiling for 4 minutes at pH 4.5 the acethydroxamic acid 

 formation was determined as usual. The protein concentration of the different enzyme fractions was 

 determined by the biuret method developed for serum protein by Gornall^* and standardized against 

 the Kjeldal method for determining protein nitrogen. 



Preparation of acyl coenzymes 



Acyl coenzymes were prepared by adding the corresponding acid anhydrides to the coenzyme in 

 slightly alkaline medium as described by Simon and SheminI^. Benzoyl CoA was kindly supplied by 

 Dr John Taggart. The amount of acyl CoA formed was determined by two methods: (i) the hydrox- 

 amic acid formation at neutral pH, at which presumably only the acyl group on the SH is tested, 

 and (ii), the nitroprusside test for free SH groups described below. The hydroxamic acid test was 

 calibrated against acetylthiocholine iodide solutions. According to both methods 95 to 100% of the 

 CoA was acylated. 



Assay of the reactions mediated by choline acetylase by the nitroprusside test for free SH groups 



If the reaction mixture contains acetyl CoA and choline as substrates and choline acetylase as 

 enzyme, the enzymic activity may be tested either by bioassay of the acetylcholine formed, using the 

 frog's rectus muscle, or by determining the remaining acetyl CoA by the use of hydroxylamine in 

 neutral solution or by determining the appearance of free SH groups of CoA. Of the two chemical 

 tests the latter is more sensitive and was used by adapting the method described by Grunert and 

 Philips^^ to our special conditions. 



The reaction mixture contained only acetyl CoA, choline and enzyme in buffer (o.i M KgHPO^ 

 of pH 7). The total volume was i.o ml. An aliquot of this solution (0.2 ml) was added to o.i ml of 

 0.033 N HCl and 1.2 ml of saturated sodium chloride solution. Solid NaCl (0.2 gram) was added to 

 keep the system saturated. Following the inactivation of the enzyme, 0.1 ml of 0.033 ^ NaOH was 

 added. The protein precipitate was removed by centrifugation ; at this stage the mixture remains 

 stable. An aliquot (0.8 ml) was used for spectrophotometric determination. To the aliquot were added 

 0.1 ml of 0.067 M sodium nitroprusside and 0.1 ml of a solution containing 1.5 sodium carbonate and 

 0.067 M NaCN. The optical density was determined in the Coleman Junior Spectrophotometer at 

 520 m/i exactly 30 seconds after the addition of the base. To correct for slight current changes in the 

 spectrophotometer control tubes containing saturated NaCl, the nitroprusside and cyanide-carbonate 

 reagents were tested between every 3 or 4 assays. If the assay mixture was not saturated with NaCl, 

 a significant decrease in color was observed, whereas the presence of metaphosphoric acid was not 

 found to be necessary for the experimental conditions used. There is a slight decrease in optical den- 

 sity if the phosphate concentration in the acid system exceeds 0.0 1 M , probably due to the effect of 

 the buffer on the final pH, since the optimal pH of the colorimetric test is 10. 



The nitroprusside test was calibrated against standard glutathione solution under identical con- 

 ditions and with the same constituents as in the tests themselves. If the color developed per SH group 

 is assumed to be the same in both cases, i mg of the Pabst CoA preparation would contain about 

 740 ^g CoA, or close to i ^M. In the calculations i mg of the CoA preparation was assumed to contain 

 I nM of CoA. 



Acyl CoA breaks down at pH 10, liberating SH groups. There is, therefore, an increase in colour 

 between the time of addition of the alkali and the reading 30 seconds later. This increase is proportional 

 to the amount of acyl CoA present at the time of the addition of alkali. Curves which indicate the 

 amount of colour produced by various acyl CoA concentrations were used in two successive approxi- 

 mations to correct for this factor. 



The stability of the different acyl Coenzymes differs considerably. Butyryl and acetyl CoA have 

 about the same stability, but propionyl CoA breaks down almost twice as fast and benzoyl CoA even 

 more rapidly. 



Since the quantitative measurement of acetylation was based upon the appearance of mercaptan 

 groups during the course of the reaction, it is necessary to establish the validity of the test. If, as is 

 believed, the colour is produced by a complex involving the mercaptan, then the molar extinction 

 coefficient should depend upon the precise nature of the mercaptan and colorimetric assay would re- 

 quire calibration with CoA. This latter procedure requires a sample of known composition and this 

 was not readily available to us. However, it appeared likely that the molar extinction coefficient for 

 glutathione might not differ greatly from that of CoA and could be used for the assays. Consistent 

 results have been obtained by others on this assumption. Fortunately, this question could be tested 

 by measuring the acetylation in an independent way, namely by assaying for acetylcholine directly : 

 This latter comparison is necessary for establishing the validity of the method in general even aside 

 from the question of extinction coefficients. The following figures compare the nitroprusside mercaptan 



References p. 324. 



