sign in sign up
<<
Home , Aspartic acid

Vol. 564 417

The Enzymic Reaction of Amino with

By F. J. R. HIRD AND P. H. SPRINGELL Department of and School of Agriculture, University of Melbourne, Australia (Received 4 August 1953) It has been shown that kidney extracts will bring (3) The estimation of cysteinylglycine and about the exchange of L-amino acids for cysteinyl- by the Lugg (1933) modification of the in reduced glutathione (GSH) to form the Sullivan method (1929). Initially this was thought corresponding y-glutamyl (Hanes, Hird & to be suitable but it was found that during the Isherwood, 1950, 1952). The enzymic reaction of an estimation procedure GSH gave rise to significant amino , e.g. L- with GSH, and the amounts of cysteinylglycine and/or cysteine. products formed in the system studied can be The method was abandoned in favour of the represented diagrammatically as follows: following. (4) The estimation of cysteine. The enzymic y-Glutamylcysteinylglycine + L-Alanine preparations contained an active peptidase which y-Glutamyl hydrolysed cysteinylglycine to cysteine and glycine. Preliminary experiments showed that the addition y-Glutamyl-L-Alanine + Cysteinylglycine of certain amino acids to the -GSH system Peptidase brings about the liberation of cysteine in excess of that liberated by the control without . Cysteine + Glycine The amount of cysteine produced in excess of that In the previous investigations the evidence for liberated by the control varies with the different a reaction was the appearance of a new on amino acids and, with certain qualifications, can be chromatograms of digest samples and was of a taken as an index of their reactivity in the system. qualitative nature. A quantitative comparison of the reactivities ofthe various amino acids with GSH EXPERIMENTAL in the system under investigation involves the Estimation of cysteine estimation of either a peptide or an am,ino acid, and the estimation must be made in the presence of The method used is a modification of that developed by other peptides and amino acids. Nakamura & Binkley (1948). It depends on the blue colour Several possible methods for the estimation of produced specifically by cysteine in the presence of brucine and persulphate under acid conditions. Difficulties were the reactivity of an amino acid with GSH were con- encountered with the method. The colour intensity was sidered. (1) The estimation of the appropriate y- found to be dependent not only on the concentration of glutamyl peptide formed. Large variations in the cysteine but was also influenced by the concentrations of physical and chemical properties of these peptides brucine, persulphate and ions, as well as by time would make each estimation an individual problem, and temperature. However, these conditions can be especially since such estimations would have to be standardized. made in the presence of amino acids and closely We found that and (cf. Nakamura related peptides. However, in some cases roughly & Binkley, 1948) seriously interfered with colour develop- ment and the method could not be used in their presence. In quantitative estimations can be made by comparing addition, ethionine was found to inhibit the colour pro- the colour intensity ofpeptide spots after separation duction. The concentration of glycine was doubled to by paper chromatography. eliminate the small depression of colour production which (2) The estimation of residual GSH by use of was caused by amino acids. In the absence of cysteine the glyoxalase. In the system being studied, two reagents produce a reddish colour, the absorption spectrum molecules of GSH would react to form the tetra- ofwhich is reproduced in Fig. 1. This colour production falls peptide y-glutamylglutathione which might not off with increased cysteine concentration as shown. The act as a coenzyme for glyoxalase. Ifthis were so, one absorption due to the red complex is negligible at 660 m/.., molecule of cysteinylglycine would be formed but and optical densities have accordingly been measured at this two molecules of GSH would 'disappear'. Thus wavelength. Absorption due to the red colour was not increased by addition of chloride which is known to react at there would be a disparity between residual GSH a higher temperature to give a similar colour (Binkley, and the exchange reaction. This has indeed been 1948). To enable low cysteine concentrations to be shown by Waelsch (1952). Further, an amino acid measured, this competing reaction, leading to the red reacting with y-glutamylglutathione could set free complex, was prevented by adding 0x2 mg. cysteine to a molecule of GSH and no cysteinylglycine. every estimation. Biochem. 1954, 56 27 418 F. J. R. HIRD AND P. H. SPRINGELL I954 represents cysteine and GSH concentrations corresponding to the amounts resulting from the progressive breakdown of GSH (1F012 mg.). This curve, in the range in which we are interested, is nearly linear, unlike the cysteine calibration curve. Because of the depression of colour production by GSH, the background cysteine enables smaller amounts of cysteine to be measured when GSH is present. Nakamura & Binkley (1948) do not record any interference by GSH. Their finding that in enzymic experiments GSH inhibits at higher H concentrations has an alternative explanation in terms of j &05mg.S>H \ img. CySI the colour depression described. C CySH Reagents CySH The reagents used were all of analytical quality. 8 2 t0075mg Glycine-sulphuric acid. Glycine (5 g.) in 100 ml. of CySH -2* 12-5 % (v/v) sulphuric acid. This reagent is stable for months. Solution for background cy8teine. Cysteine (0.2 mg., or 00 equivalent amount of cysteine HCI) in 1 ml. water. This reagent is freshly made up daily. Brucine 8olution. Brucine (1 g.) in 100 ml. 5% (v/v) sulphuric acid. This reagent is made up weekly. 400 500 600 700 Potasgium persulphate 8Olution. Potassium persulphate Wavelength (my.) (0.5 g.) in 100 ml. water. This reagent is stable for at least 2 weeks. Fig. 1. Absorption curve ofcysteine-brucine complex in the Trichloroacetic acid (TCA). 50 % (w/v). presence of varying amounts of cysteine. Recorded optical densities of solutions of cysteine-brucine complex Procedure a blank. were read against water The colour was developed in test tubes which had a small process on the inside, just above the 8 ml. mark, to facilitate 0.15r pipette drainage. The reagents were added in the following order: 2 ml. glycine/sulphuric acid reagent; 1 nI. cysteine reagent; 2 ml. brucine reagent; 2 ml. TCA filtrate containing V. from 0.0 to 0 4 mg. cysteine, or 2 ml. water in the control (the TCA and phosphate from the digest did not affect LUJ o10o colour development); 1 ml. potassium persulphate reagent. Optical densities were measured at 660 m,. on a Beckman spectrophotometer, Model DU. The reaction resulting in development of colour begins when the persulphate is added. The persulphate was there- -0 fore added to successive tubes at 1 min. intervals. After 0.05F mixing of the contents, the tubes were immersed and the colour developed in a water bath at 30±0-10 for 45 mi. The tubes were then removed and placed in an ice bath for 2 min. The blue solutions were transferred to the spectro- photometer cells and 4 min. after removal from the constant bath the blank was read off a temperature cysteine against 0 0'1 0-2 0.3 0.4 water (as a check on the reagents). For 0-2 mg. cysteine, Cysteine (mg.) log Is/l at 660 mp. was about 0-170. The background ...... s , cysteine solution was then set at zero (5 min. after removal 0 25 50 75 100 from bath) and the unknowns were read at minute intervals, GSH breakdown (%' all 45 min. after the persulphate additions. The amount of or GSH broken down was then read Fig. 2. Cysteine and cysteine-GSH calibration curves. cysteine present the Optical densities recorded were read against 0°2 mg. from the graph (Fig. 2). The method is reproducible to cyeine blank. Incysteine-GSH solutioun8bQIne.hproupuriwors-ssrs4. ±30+/° o*ieach were such as arise from GS]H by . QFtical densities were measured at 660 my. E8timation of cysteine producedfrom GSH by kidney extracts in the presence of amino acid8 JGSH does not itself react with bruciine to give a blue To enable the reactivities of a wide range of amino acids colour, but it does markedly reduce the9 colour intensity with GSH to be tested, it was convenient in most cases to produced by cysteine (Fig. 2). To estimate cysteine in the use an enzyme concentration that brought about 35-40% presence of GSH, a separate calibration curve is therefore GSH breakdown during the incubation period of 30 min. at necessary. Onthecysteine-GSHcalibration curve eachpoint 300, i.e. in the absence of any added amino acid. Extracts VoI. 56 REACTION OF AMINO ACIDS WITH GSH 419 prepared in the following manner were suitable: 3 g. of runs; up to 100 hr. for long runs. In the longer runs, papers frozen sheep kidney cortex were disintegrated in 20 ml. of had a thick pad of filter paper clipped to the bottom of the ice-cold 0-2M potassium phosphate buffer, pH 7-4, in a glass paper strip. homogenizer for 30 sec. at 1400 . min.-l. After centri- fugation for 15 min. at 1500 g to remove the coarse fraction, RESULTS the supernatant was stored at - 100 in 2 ml. portions until Trial experiments showed that with the concentra- needed. Under these conditions the enzyme was stable for tion of GSH chosen, 20-6 tanoles/5 rnl., and an months, and it was therefore possible to compare a large amino acid concentration of 132 ,umoles/5 ml., range of amino acids using a single enzymic preparation. increases in cysteine liberated over a 30 min. A small amount of cysteine and of GSH was present in the incubation period could be observed. At this stage enzymic preparations. After incubation, cysteine (0-010- 0-015 mg./2 ml. sample) was found in the six preparations of the work the effect of the concentration of amino tested. In any one enzymic preparation the endogenous acids on the liberation of cysteine had not been cysteine was constant and therefore the order of reactivities determined. This aspect is discussed in the following of the amino acids was not altered. section. Table 1 gives the excess of cysteine liber- The systems and samples for cysteine analysis were pre- ated from GSH by a wide range of amino acids and pared as follows. To 2 ml. enzymic preparation in 0-2m some of their derivatives. Estimations of the potassium phosphate buffer, pH 7-4, were added 2 ml. amino liberated cysteine were carried out in duplicate and acid solution (neutralized when necessary) and 1 ml. GSH always with a GSH control without added amino solution (6.32 mg./ml., 20-6umoles) in this order. This acid. quantity of GSH corresponds on complete hydrolysis to 0-4 mg. cysteine in the final 2 ml. TCA supernatant. The The results given in Table 1 show that L-x-amino GSH solution was prepared afresh before each incubation acids differ markedly in their abilities to liberate and dissolved in a vessel in which the air had been displaced cysteine from GSH. The DL-amino acids, where by N.. The reaction was carried out in stoppered test tubes tested, gave approximately half the values of the under N2 in a water bath at 300 for 30 min. At the end of this time, 4 ml. were transferred to a glass-stoppered 10 ml, Table 1. Comparative reactivity of variou8 centrifuge tube containing 5 ml. of water and 1 ml. of 50 % amino acids with GSH TCA. The tube was stoppered, the contents were mixed, (System: as described in Experimental with 2 ml. allowed to stand for 5 min. and then centrifuged lightly; enzyme preparation, 2 ml. amino acid solution containing 2ml. ofthe supernatant was then usedfor analysis. Cysteine, 132,umoles (control: 2 ml. water in place of amino acids) when added to the enzymic preparations and incubated and 1 ml. solution containing 20-6 ,umoles GSH. Incu- under the conditions used, could be recovered within bation: 30 min. under N, at 300. Figures represent ad- experimental error, showing the absence of significant ditional cysteine liberated in excess of control by amino cysteine-desulphurase activity. acids and are expressed in terms of % GSH breakdown. Figures in brackets calculated for L- acids from DL- acids Chromatographic procedure x 2. Reactivities measured with a single preparation have The system was on a scale of one-tenth of that used for been listed separately to enable more accurate comparisons in to be made.) cysteine estimations. To 0-2 ml. enzymic preparation GSH breakdown in 0-02M potassium phosphate buffer, pH 7 4, were added excess of control (%) 2,umoles neutralized GSH in 0-1 ml., and 0-2 ml. amino acid solution (neutralized when necessary). Incubation was Results carried out in stoppered 9 x 75 mm. test tubes in an atmos- In a single from other phere of N.. The enzymic preparation was dialysed in a Amino acid preparation preparations moving bag against 0-02M potassium phosphate buffer for Glycine 33 26, 27, 28, 32 4 hr. before use to remove amino acids present in the L-Alanine 24 (28) 17 (10) enzymic preparation. After 30 min. incubation at 300, DL-c-Amino-siobutyric acid 0 1-0 ml. of warm ethanol containing 2-4jumoles N-ethyl- DL-cx-Amino-n-butyric acid 12 6 maleimide was added to precipitate and block L-Vahine 2 (-I) -SH groups (Hanes et al. 1950). Cysteinylglycine when DL-Norvaline 22 L- 2 (0) coupled with N-ethylmaleimide in this manner is capable of L- 21 16 (16) being detected on chromatograms at a concentration of DL-Norleucine 21 20 0-001 M. Under the conditions used, this represents a figure L- - 21, 25, 26 of 5% GSH breakdown. After centrifugation, the super- DL-Serne 20 17 natant was collected, the precipitate washed once with DL-ThreoniDe 3 3,6 0-2 ml. 66% ethanol (v/v) and the combined supernatant L- 2 -1, -4 liquids evaporated in vacuo over NaOH in a small evapor- L-Asparaglne 27 23 ating cup. The residue was then taken up in 0-05 ml. water L- 21 15, 16 and 0-002 ml. were for L- 37 42, 43, 44 portions used chromatography. L- 36 19, 21, 23, 24 Solvents used were phenol saturated with water, butanol: L- 17 water: glacial acetic acid (50:40: 10, v/v), propanol:water L-Citrulline 37 (80:20, v/v) as such, and with 0-5 ml. NH, (sp.gr. 0,880) L- 33 18 added to the chromatography jar. Time of development L- 22 (26) 20, 21, 24 varied according to separation required: 15 hr. for short L-Methionine sulphoxide 43 27-2 420 F. J. R. HIRD AND P. H. SPRINGELL I954 Table 2. Comparative reactiKvity of the more reactive liberated, and in some cases it fell below the control amino acids upith GSh with GSH alone, e.g. DL-. The increase in (System: as described in IExperimental with 2 ml. cysteine liberation with increasing substrate con- enzymic preparation, 2 ml. aminio acid solution containing centration was probably due to increased saturation 132 or 264 ,umoles (control: 2 mLI. water in place of amino of the active centre of the enzyme. This increase in acids) and 1 ml. solution conttaining 206,umoles GSH. cysteine liberation was usually paralleled by an Incubation: for varying periods3 under N2 at 30°. Figures increase in corresponding y-glutamyl peptide forma- represent additional cysteine lib erated, in excess of control, tion as judged from the chromatograms. Fig. 6 by amino acids and are expreased in terms of % GSH shows that at higher concentrations of glycine breakdown. The controls corresIponded to 12, 24 and 43 % where cysteine liberation fell off, y-glutamylglycine GSH breakdown for 5, 10 anc1 20 mini incubations re- spectively.) formation fell offalso. At these high concentrations Time of incubation (min.) the residual GSH was high. This means that GSH 5A10 20 was not being acted upon by the enzyme system. It GSH breakdown in is possible that the amino acids compete with GSH Concentration excess of control (%) for the active centre of one enzyme, as seems to be Amino acid (,umoles/5 ml.), A__ _ the case with L- (Fig. 4). A decrease in L-Methionine 132 36 62 41 cysteine liberation was also observed when the sulphoxide L-Glutamine 132 28 53 45 L-Citrulline 132 28 42 35 DL-Norleucine 264 11 26 DL-Norvahne 264 9 26 - L-form; there is thus no pronounced inhibition by the D-form. Amino acids substituted on the amino group such as , L- and L-hydroxy- 0- proline were not reactive. Tryptophan, methionine and ethionine interfered with the colour production -o by cysteine, and so quantitative data for these -0 arnino acids cannot be given. However, chromato- I graphic evidence has been obtained to show that these amino acids react with GSH in the system. was too insoluble for comparisons with other amino acids to be made. , ammonium chloride, 8-hydroxyquinoline and chloramphenicol had no effect on the system. With the more reactive amino acids, over 80 % of the GSH was broken down by the end ofthe reaction period. This resulted in a reduction ofthe amount of Fig. 3. GSH available as a substrate and in progressive slowing of the reaction; thus, these amino acids are more reactive than indicated in Table 1. As the

rangeofreactivitiesof the amino acids was too great 0- for them to be compared on the same time scale, the most reactive ones were tested over shorter s0-h periods of time. The results are given in Table 2. V .0tv

Effect of amino acid concentration on the I liberation of cysteine Figs. 3 and 4 show the effect on the liberation of ,.le x ne cysteine from GSH of increasing concentrations of various amino acids. It can be seen that as the con- 200 400 600 800 1000 centrations of L-aspartic acid, L-valine and L- Concentration ml.) isoleucine were increased, the liberation of cysteine (Pmoles/5 decreased. In contrast, the other L-ao-amino acids, Fig. 4. up to a certain concentration depending on the Figs. 3, 4. The effect of varying concentrations of amino amino acid concemed, caused an increase in the acids on release of cysteine from GSH expressed in terms amount of cysteine liberated. At higher concentra- of % GSH breakdown. Conditions as in Table 1 but with tions, however, there was a decline in the cysteine varying concentrations of amino acids. Vol. 56 REACTION OF AMINO ACIDS WITH GSH 421 reacting amino acid was replaced by similar con- preparations, as well as catalysing the transfer of centrations of sodium sulphate and glucose. The y-glutamylgroupsto amino acids, also formglutamic effect of high concentrations of amino acids could acid from GSH, presumably by hydrolysis. Fig. 6 therefore be a non-specific one. As any point on the shows that as the glycine concentration was in- amino acid concentration curves must represent creased up to a concentration of 67 jamoles/0.5 ml. a combination of reaction and inhibition, com- digest, y-glutamylglycine formation increased and parison of the relative reactivities of amino acids in glutamic acidformation decreased. This competition the system is only possible on the ascending slope of between water and glycine for the y-glutamyl the concentration curve. group in GSH mnakes it likely that the two effects, hydrolysis and aminolysis (transpeptidation) of the Paper chromatography as the index y-glutamyl , are properties of one of transpeptidation enzyme. Ifthis were so, the initial transfer reaction Hanes et al. (1950, .1952) have shown that L- would involve water or the amino group of GSH as glutamic acid, L-glutamine, L-phenylalanine, L- an acceptor. In the case ofwater as an acceptor, all leucine, L-valine, L-tryptophan, L-cysteine and three amino acids in GSH are liberated. In the case glycine would react with certain y-L-glutamyl of an amino acid (including GSH) as an acceptor, peptides, including GSH, to form new spots running cysteine and glycine only are liberated (the y- in the expected positions for the y-glutamyl glutamyl linkage being preserved in the new y- peptide corresponding to the added amino acid. In glutamyl peptide formed). As the reaction proceeds many cases these spots were eluted, hydrolysed and there is a progressive increase in glutamic acid, re-chromatographed to establish their composition. cysteine and glycine and these amino acids can then In the present work other amino acids have been participate in further transpeptidation reactions. tested for their reaction with GSH by the presence or The addition of other amino acids, with few ex- absence of new spots on chromatograms after con- ceptions, brings about the release of extra cysteine centration of digest samples. Positive evidence has (due to extra y-glutamyl peptide formation) at the been obtained for the reactivity of L-argimiine, L- expense of GSH. This is shown in the time/progress , L-lysine, L-histidine, L-methionine, DL- curve in Fig. 5 with glycine as the added amino acid. ethionine, DL-norleucine, DL-norvaline, L-alanine, It is probable that each y-glutamyl peptide formed DL- and DL-threonine. y-L-Glutamyltyrosine serves as a donor peptide in further transpeptidation has been shown to serve as a donor peptide (Hanes reactions. et al. 1952), and it is probable that tyrosine will For liberated cysteine to be taken as an index of react with other y-glutamyl peptides in the system. transpeptidation activity, it is necessary that a With the exception of L-isoleucine and L-valine the relationship be established between cysteine libera- reactivity of amino acids as followed by the tion and y-glutamyl peptide formation. Four appearance of a new spot on a chromatogram was aspects of the enzymic reactions that are concerned always paralleled by a positive figure for cysteine in such a relationship have been examined. liberation by the amino acid concerned. Similarly, failure to detect a reaction by chromatography was lOOr paralleled by a zero figure for cysteine liberation in the case ofL-aspartic acid, L-proline and L-hydroxy- 80 proline. .-- 0-R GSH Further investigations on the transfer reaction 3 0 60 The positive results given by chromatography are not consistent with the negative results given by 40 I- estimation of cysteine liberated from GSH by L- I isoleucine and L-valine. The figures given for excess I-, of cysteine liberated (Table 1) in these cases do not 20F measure the relative reactivities of the amino acids with GSH. Accordingly, we have examined the enzymic reaction further in an attempt to determine 0 30 60 90 120 the relationship between extra cysteine liberated by Time (min.) acids and relative reactivities. the amino their Fig. 5. Time/progress curve for % GSH breakdown with The liberation of cysteine from GSH is a two-step and without added glycine. System: 2 ml. enzyme pre- process. First, cysteinylglycine is liberated as a paration, 2 ml. glycine solution containing 660 jtmoles, or result ofa transfer reaction involving the y-glutamyl 2 ml. water and 1 ml. solution containing 20 6,umoles linkage. Secondly, the cysteinylglycine so formed is GSH. Incubation: varying intervalsX of time at 30° hydrolysed to cysteine and glycine. The enzymic under N2. 422 F. J. R. HIRD AND P. H. SPRINGELL I954 (1) Hydroly8i8 of GSH. If the two transfer the same direction with each amino acid, a com- reactions, hydrolysis and transpeptidation, are parison of reactivities can be made. The effect of properties of a single enzyme, then there is likely amino acids on enzymic hydrolysis of GSH is at to be competition between water and amino acids present being investigated further. for activated y-glutamyl linkages. Therefore, the (2) Hydrolys8is of cysteinylqlycine. For cysteine to higher the amino acid concentration the less should be an accurate measure of reactivity of amino acids be the hydrolysis. The depression of hydrolysis by with GSH, the cysteinylglycine formed must be added glycine as gauged by glutamic acid formation hydrolysed completely and as rapidly as it is formed. is shown in Fig. 6. The cysteine produced in the The enzymic preparations used, however, did not control therefore includes a greater proportion of carry the hydrolysis to completion. On concen- cysteine which has its origin in GSH hydrolysis than trating the breakdown products ten times, it was the cysteine produced from GSH in the presence of possible to detect traces of cysteinylglycine on added amino acid. This means that amino acids are chromatograms. The addition of large amounts of more reactive in transpeptidation than is shown by cysteinylglycinase (prepared by the method of the figures for extra cysteine liberated (Table 1). It Binkley, 1952) to the digests did not result in either is likely, however, that the amino acids most an increase in cysteine (measured spectrophoto- reactive in transpeptidation are also the best metrically) or a decrease in the cysteinylglycine competitors with water. For instance, Fig. 7 shows present on chromatograms. It is possible, therefore, that glycine depresses glutamic acid formation that the traces of this peptide which were present in more than does L-valine at the same concentration. the digests represented a final equilibrium mixture It is also a stronger reagent in transpeptidation, as of cysteinylglycine, cysteine and glycine. Further, measured by y-glutamyl peptide formation (Fig. 7) the trace amount of cysteinylglycine present in and by the liberation ofextra cysteine (Table 1). As digests in the presence of various amino acids long as the two effects, transpeptidation and de- (132 j.moles/5 ml.) was constant as judged from pression of hydrolysis, work proportionally and in chromatograms.

.1 -~. - to a a E. W'"

0 M.U) UL E) 1 A.

>I. +I +: + + --. + + I I I I I :I :1 I I Uf) CA 4A 4 43 Q Origin Origin

y-Glu.Gly y-Glu.Gly Glu Glu GSH GSI4 }yGIlu.CySH y-Glu.CySH

y-Glu.Val Gly Gly

Fig. 6. Chromatogram showing effect of glycine concentra- Fig. 7. Chromatogram showing effect of glycine and L- tion on the enzymic reaction with GSH. System: 0-2 ml. valine at two concentrations on the enzymic reaction with enzymic preparation (dialysed), 2 Lmoles GSH and GSH. System: 0-2 ml. enzyme preparation (dialysed), 0, 13.2,67 and 1330 umoles glycine, respectively. Solvent: 2jmoles GSH and 0, 13-2 and 10Omoles of glycine or propanol: water (80:20). Development: 60hr. Thehighest L-valine, respectively. Incubation: 30 min. at 30° concentration of glycine was obtained by weighing this under N2. Solvent: propanol: water (80:20). Develop- amino acid into the incubation vessel. ment: 60 hr. VOI. 56 REACTION OF AMINO ACIDS WITH GSH 423 (3) y-Glutamyl peptide formation. Some of the liberated cysteine reacted with GSH to form y- Table 3. The influence of L-vaiine on the liberation glutamylcysteine and as such was not estimated. of cy8teine from GSH by L-glutamine This might be the reason for the curve flattening off at about 80 % GSH breakdown (Fig. 5). The amount (System: as described in Experimental with 2 ml. ofcysteine liberated would therefore be greater than enzymic preparation, 1 ml. containing 132,umoles L-gluta- the amount estimated, and the latter cannot be mine, 1 ml. water and 1 ml. solution containing 20-6pmoles a precise index of the reactivity of an amino acid. G(SH. The valine was weighed directly into the vessel. The error will be greatest with the most reactive Control: 2 ml. water in place of amino acids. Incubation: amino 30 min. under Ns at 30°. Figures represent additional acids, but would allow comparisons of cysteine liberated, in excess of control, and are expressed in reactivity to be made. terms of % GSH breakdown.) We have observed on chromatograms that an GS]iH breakdown in increase in the concentration of a reactive amino L-Valine exxess of control acid brought about the formation of additional y- (,umoles/5 ml.) (%) glutamyl peptide of the amino acid concerned and, 0 31 by cysteine estimation, the release ofextra cysteine. 132 30 500 17 However, this is not true ofL-valine and L-isoleucine. 1000 3 An increase in concentration of these amino acids led to a decrease in cysteine liberation (Fig. 4) but, as measured from chromatograms, corresponding y-glutamyl peptide formation increased (Fig. 7). Incubation Incubation At a low concentration (132 pamoles/5 ml.), L-valine 15 min. 90 min. was less active than DL-norvaline, as measured by the formation of the corresponding y-glutamyl a 0 -. peptide (Fig. 8). Similarly, L-isoleucine was less z z > reactive than L-leucine and DL-norleucine in y- + + + + a glutamyl peptide formation. Qualitatively, the a., I colour reactions of different y-glutamyl peptides _ n Origin

> z + + I-Glu.Glu

u) n Ifl Origin Glu

Glu GSH y-GIu.Val I -Glu.CySH y-Glu.Val y-GIu.Nor.Val

Gly Val

Nor-Val 7-Glu.Nor.Val

Fig. 9. Chromatogram showing effect of L-valine and DL- norvaline on the enzymic reaction with y-L-glutamyl-L- Fig. 8. Chromatogram showing effect of L-valine and DL- glutamic acid. System: 0-2 ml. enzyme preparation norvaline on the enzymic reaction with GSH. System: (dialysed), 0-1 ml. solution containing 2 pemoles neutralized 0-2 ml. enzyme preparation (dialysed), 2 jumoles GSH and y-L-glutamyl-L-glutamic acid, 0-2 ml. solution containing 13-2 and 26.4pAmoles L-valine and DL-norvaline, re- 13-2 and 264 Aimoles L-vahne and DL-norvahne, re- spectively. Incubation:30min.at30°underNs. Solvent: spectively. Incubation: 15 and 90 min. at 300 under Ns. propanol:water (80:20) +0-5 ml. (sp.gr. 0-880) Solvent: phenol saturated with water. Development: in the chromatography jar. Development: 60 hr. 20 hr. 424 44F. J. R. HIRD AND P. H. SPRINGELL I954 with ninhydrin resembled the colour given by that ofconjugase (Laskowski, 1950). Both glutamic acid and not by the amino acid linked to it. are known to act on the y-glutamyl linkage. Dakin Further, it would be expected that the reaction of & Dudley (1913) reported the presence of an anti- y-glutamyl peptides with ninhydrin would involve glyoxalase in extracts of pancreas, and Woodward, the free cx-amino group of glutamic acid. It is, Munro & Schroeder (1935) found that this enzyme therefore, likely that differences in colour intensity was also present in extracts ofkidney. The action of of y-glutamyl peptides on chromatograms indicate antiglyoxalasewas foundto be due to thedestruction differences in amounts of these peptides. It seems of GSH, the coenzyme in the glyoxalase system justifiable then to conclude that at low substrate (Woodward et al. 1935; Salem & Crook, 1950). It is concentration L-isoleucine and L-valine are less possible that these three enzymes are identical. reactive than their structural isomers, and that the Although there has been some tentative speculation figures given for the liberation of cysteine by the about a possible role in synthesis (Fruton, amino acids tested indicate their comparative 1950, Hanes et al. 1950, 1952) no function has been reactivities. ascribed to the enzyme. The effect of increasing the concentration of L- Unpublished observations by Hird, Neville & valine which was added to a digest containing GSH Springell have shown by use of differential centri- and glutamine is shown in Table 3. At low substrate fugation that y-glutamyl transferase in sheep kidney concentration, L-valine brought about very little and ox pancreas is particle bound and that by reduction in cysteine liberated by L-glutamine, and treatment with butanol (Morton, 1950) it can be this indicates that the apparent inertness ofL-valine liberated from the particles. Such preparations in the system GSH + valine was not due to inhibition would still transfer the y-glutamyl group to both of cysteinylglycinase by this amino acid. At high water and amino acids. concentrations ofL-valine, the liberation ofcysteine Precise interpretation of the results obtained in by glutamine was reduced considerably. This the reaction of amino acids with GSH is difficult. reduction is consistent with: (i) L-valine inhibiting However, using the indices established in the the attachment of the donor peptide, GSH, to the present investigation, broad comparisons can be enzyme, (ii) L-valine, as a weak reactor, occupying made, especially within the groups of amino acids. the position of the acceptor amino acid on the Within the aliphatic series the most reactive amino enzyme and so reducing the number oftranspeptida- acids were those with long unbranched side chains, tion reactions, (iii) L-valine, at high concentration, e.g. methionine sulphoxide, glutamine and citrulline inhibiting cysteinylglycinase. had the greatest reactivity; norleucine and nor- In a further test of the reactivity of valine, y-L- valine had greater activity than leucine, isoleucine glutamyl-L-glutamic acid was used instead of GSH, and valine. Amidation ofthe distal carboxyl groups as it gave a simpler chromatographic pattem. The of L-aspartic acid and L-glutanic acid, respectively, chromatogram (Fig. 9) showed that residual y-L- markedly enhanced reactivity. This has also been glutamyl-L-glutamic acid was present in greater noted by Binkley (1952) and Fodor, Miller & amounts in the presence of L-valine than in the Waelsch (1952). presence of DL-norvaline. This result suggests that The inhibitory effect of the ,8-methyl group in L- L-valine at low substrate concentration inhibited valine and L-isoleucine has previously been shown cysteine liberation for one or both of the reasons by other workers. Fox & Winitz (1952) found that outlined in (i) and (ii). these amino acids were also unreactive in the en- (4) Effect of concentration of acceptor amino acids. zymic synthesis of anilides. Fox, Pettinga, Halver- Above a certain concentration characteristic ofeach son & Wax (1950) have shown that benzoyl-L-valine added amino acid, cysteine liberation decreased. amide is resistant to enzymic hydrolysis, and Smith, For the amino acids tested, but with the exceptions Spackman & Polglase (1952) have shown this for the of L-aspartic acid, L-isoleucine and L-valine, the amides of isoleucine and valine. These results are in concentration of 132 umoles/5 ml. represented a contrast to those given by the corresponding point on the concentration curves (Figs. 3, 4) on or derivatives of leucine. Valyl peptides are also below the point at which reactivity began to fall off. resistant to hydrolysis by acid (cf. Sanger, 1952). Such a concentration was therefore suitable to The preliminary reports by Binkley (1952) and compare reactivities. Fodor et al. (1952), working with y-glutamyl transferase from pig kidney, gave figures for DISCUSSION several amino acids and their results are in agree- ment with those reported in the present paper. The y-glutamyl transfer reaction has been shown to In a previous publication (Hanes et al. 1952), it be catalysed by extracts ofkidney and pancreas but was found that L-arginine did not react with y-L- not by extracts ofliver tissue (Hanes et at. 1952). The glutamyl-L-glutamic acid to form the corresponding ,distribution ofy-glutamyl transferase coincideswith y-glutamyl peptide. In the present work, however, Vol. 56 REACTION OF AMINO ACIDS WITH GSH 425 it has been shown that L-arginine is quite reactive aspartic and L-glutamic acids to react with GSH was with GSH. The reaction of L-arginine with GSH has enhanced by arnide formation as in L-asparagine also been reported by Kinoshita & Ball (1953). and L-glutarnine, respectively. Binkley (1951) and Binkley & Olson (1951) have 4. With the exceptions of L-aspartic acid, L- given a special role to L-glutamine in GSH hydrolysis isoleucine and L-valine, an increase in amino acid and suggest the following reaction mechanism: concentration brought about an increase in the cysteine liberated from GSH. At higher concentra- Enzyme COONa + glutamine -+ Enzyme CONH2 tions reactivity fell off. Increases in the concentra- + Na glutamate, tion of L-aspartic acid, L-isoleucine and L-valine Enzyme CONH2 + GSH -+ Enzyme COOH caused a steady decrease in cysteine liberation. + glutamine + cysteinylglycine. 5. Chromatographic evidence has now been obtained for transpeptidation with most of the The evidence for this hypothesis is that L-glutamine common, naturally occurring amino acids, with the increases the 'hydrolysis' of GSH without itself exceptions of L-aspartic acid, L-proline and L- being hydrolysed in the process. The results ofthese hydroxyproline. workers can be explained by the formation of y- glutamylglutamine with the liberation of cysteinyl- We wish to thank Mr A. M. Gallacher for preparing the glycine followed by hydrolysis of the latter. The glutathione. production of a new peptide (almost certainly y- glutamylglutamine) when L-glutamine is incubated REFERENCES with GSH has been reported (Hanes et al. 1952). In Binkley, F. (1948). J. biol. Chem. 173, 403. the present work L-glutamine has also been shown Binkley, F. (1949). Renal Function, Transactions ofthe Firdt to be very reactive. The range of amino acids tested Conference, p. 63. New York: Josiah Macy. by Binkley (1949) and Binkley & Olson (1951) was Binkley, F. (1951). Nature, Lond.,167, 888. not extensive and the smaller effects of other Binkley, F. (1952). Exp. Cell Re8. Suppl. 2, p. 145. New amino acids was thought to be due to the removal of York: Academic Press. inhibitory ions. As can be seen from Tables 1 and 2 Binkley, F. & Olson, C. K. (1951). J. biol. Chem. 188,451. there are many amino acids which are as active as Dakin, H. D. & Dudley, H. W. (1913). J. biol. Chem. 15,463. so. Fodor, P. J., Miller, A. & Waelsch, H. (1952). Nature, Lond., L-glutamine and some even more These do not 170, 841. have the amide configuration. This means that the Fox, S. W., Pettinga, C. W., Halverson, J. S. & Wax, H. hypothesis of L-glutamine as a coenzyme in the (1950). Arch. Biochem. 25, 21. hydrolysis of GSH lacks unequivocal supporting Fox, S. W. & Winitz, M. (1952). Arch. Biochem. Biophys. 35, evidence. 419. Fruton, J. S. (1950). Yale J. Biol. Med. 22, 263. SUMA:RY Hanes, C. S., Hird, F. J. R. & Isherwood, F. A. (1950). 1. A modification of the method of Nakamura & Nature, Lond., 166, 288. Binkley (1948) for the estimation of cysteine in the Hanes, C. S., Hird, F. J. R. & Isherwood, F. A. (1952). Biochem. J. 51, 25. presence of glutathione (GSH) is described. Kinoshita, J. H. & Ball, E. G. (1953). J. biol. Chem. 200,609. 3. Sheep-kidney y-glutamyl transferase, which Laskowski, M. (1950). Annu. Rev. Biochem. 1.9, 21. catalyses peptide-bond transfers between GSH and Lugg, J. W. H. (1933). Biochem. J. 27, 668. amino acids, has been investigated. With certain Morton, R. K. (1950). Nature, Lond., 166, 1092. qualifications, cysteine liberated by these transfer Nakamura, K. & Binkley, F. (1948). J. bidl. Chem. 173,407. reactions can be taken as an index of the com- Salem, H. M. & Crook, E. M. (1950). Biochem. J. 40, xxxvii. parative reactivity of amino acids in the system. Sanger, F. (1952). Advanc. Protein Chem. 7, 1. 3. The relative reactivities of L-a-amino acids Smith, E. L., Spackman, D. H. & Polglase, W. J. (1952). has been determined. The most reactive amino acids J. biol. Chem. 199, 801. Sullivan, M. X. (1929). U.S. Publ. Hlth Rep. 44, no. 27,1599. in the aliphatic series were those with long, un- Waelsch, H. (1952). Symposium sur la bioggndse des pro- branched side chains. The presence of a ,-methyl teines, 2e Congris international de Biochimie, p. 26. group, as in L-valine and L-isoleucine, interfered Paris: Sedes. with the reactivity of the amino groups of these Woodward, G. E., Munro, M. P. & Schroeder, E. F. (1935). amino acids. The ability of the amino groups of L- J. biol. Chem. 109, 11.