The stoichiometry and kinetics of the inducible cysteine desulfhydrase from Salmonella typhimurium.

Abstract Studies using highly purified cysteine desulfhydrase from Salmonella typhimurium reveal that only a small fraction of the cysteine utilized by the enzyme appears as pyruvate. The isolation of 2-methyl-2, 4-thiazolidinedicarboxylic acid from reaction mixtures offers an explanation for this unusual stoichiometry. The relative amounts of pyruvate and thiazolidine produced during a reaction depend upon the cysteine concentration, pH, and the presence of a protein termed Fraction B, which prevents the formation of the thiazolidine. We propose that 2-aminoacrylate may be an intermediate in the formation of 2-methyl-2, 4-thiazolidinedicarboxylic acid. Substrate velocity curves for cysteine desulfhydrase reveal positive cooperativity with an n value of 1.9 and a Km, for l-cysteine of 0.17 to 0.21 mm. The product, sulfide, inhibits the reaction with a Ki of 0.010 mm. Sulfide inhibition is of the linear competitive type at high cysteine concentrations, but it becomes nonlinear and more pronounced at low cysteine concentrations.

The unstable intermediate, 2-aminoacrylate, is assumed to decompose rapidly to ammonia and pyruvat,e even without enzymatic intervention (2). Since this equation predicts equimolar yields of pyruvate, ammonia, and sulfide, it is puzzling that several studies on the stoichiometry of the reaction have found lesser yields of pyruvate than could be accounted for on the basis of sulfide production or cysteine depletion (8-10). Previous investigators have suggested that this unusual stoichiometry might be due to peculiarities of the pyruvate assay used (8) or the presence of transaminases in the crude extracts in which the enzyme has been studied (11).
We have recently reported the purification of an inducible cysteine desulfhydrase from Xalmonella typhimurium to a state of near homogeneity (12). Using highly purified enzyme, we find that the yield of pyruvate is still only a fraction of that expected on the basis of sulfide or ammonia production. This report details the isolation and characterization of 2-methyl-2,4thiazolidinedicarboxylic acid as a product of the cysteine desulfhydrase reaction.
The results of studies on certain kinetic properties of the purified enzyme are also presented. d  I  I  I  I  I  I  I  I  I  I  I  3  5  7  9 Cysfeme (mM) FIG.
1. Dependence of apparent ~650 on cysteine concentration in the sulfide assay. Sulfide was assayed as described under "Experimental Procedure" using solutions containing 0.05 M sodium sulfide and varying concentrations of L-cysteini in 0.1 RI Tris-HCl, pH 8.6. convenient to measure sulfide production when purifying the enzyme and for kinetic studies. Reactions are carried out at 23" in capped test tubes (10 X 75 mm) containing 2.0 ml of a given concentration of L-cysteine in 0.1 M Tris-HCl, pH 8.6. The reaction is started by the addition of a small volume of enzyme diluted in 0.1 M Tris-HCl, pH 7.6, containing 0.5 mg per ml of bovine serum albumin, and terminated by the addition of 0.2 ml of 0.02 &I N,N'-dimethyl-p-phenylenediamine sulfate in 7.2 N HCl followed immediately by 0.2 ml of 0.03 M FeC13 in 1.2 N HCl (15). The tube is then recapped, vigorously shaken for a few seconds, and, after storage in the dark for 15 to 20 min, the absorbance at 650 nm is determined in a spectrophotometer.
The apparent ~~50 for sulfide is dependent upon the cysteine concentration, and for kinetic studies in which cysteine concentrations were varied, the curve shown in Fig. 1 was used to calculate the amounts of sulfide formed.
At the 2.0 mM cysteine concentration used for enzyme purification and other routine assays, the apparent ~50 for sulfide is 1.56 x lo4 M-l cm-l.
Due to inhibition of the enzyme by the product sulfide, a plot of sulfide production per given period of time uersus enzyme concentration is not linear.
We have previously shown (12) that the initial velocity of the reaction, Vi, can be calculated from the expression vi =++ K,Q' 2K,1(K,tA) , t>o (II where & is the sulfide concentration at time t, A is the initial cysteine concentration, which is assumed not to vary significantly during the course of the reaction, K, is the Michaelis constant for cysteine, K, is an inhibition constant for sulfide, and & = 0 at t = 0. At 2.0 mM L-cysteine K,/BK,(K, + A) is equal to 5 mM+. One unit of enzyme is defined as that amount which gives a Vi of 1 pmole of sulfide per min under these standard conditions.
For the determination of pyruvate 1.0 ml of reaction mixture is incubated in an uncapped test tube (13 X 100 mm), and the reaction is terminated by the addition of 0.5 ml of 1.0 N H&Oh. After 5 min, 0.5 ml of 3 mM 2,4-dinitrophenylhydrazine in 1.5 N HCl is added, followed 15 min later by 0.5 ml of 7.1 M KOH. After an additional 10 min the absorbance of the 2,4-dinitrophenylhydrazone at 540 nm is measured. Early in the course of this work it was found that one of the products of the cysteine dcsulfhydrase reaction is a derivative of pyruvate, which gives no appreciable color reaction u-ith the 2,4-dinitrophenylhydrazine reagent unless pretreated u-ith an acidic solution of mercuric ion. 11-e refer to t'he pyruvate detectable in the absence of mercuric ion as free pyruvate, while the pyruvate which is measured after treatment with mercuric ion is referred to as total pyrurate.
Total pyruvate is the sum of free pyrurate and the pyruvatc present as the derivative.
To measure the total pyrurate produced in a reaction, 0.01 M HgS04 in 1.0 N HzS04 is substituted for the 1.0 r HzS04 used in the free pyruvate assay. This results in the formation of a precipitate after the addition of the KOH reagent, u-hich must be removed by centrifugation before determining the absorbance at 540 nm. Using sodium pyruvate solutions standardized by the lactate dehydrogenase method (16), I\-e find that this modified 2,4-dinitrophenylhydrazine assay gives an c540 of 4.4 x lo3 h1-l cm-l without HgS04 (free pyruvate) and 4.6 X lo3 nz+ cnl-1 ITit HgS04 (total pyrurate).
The rate of pyruvate production can also be determined in a continuous spectrophotometric assay, utilizing NADH and a large excess of lactate dehydrogenase.
For this purpose the basic reaction mixture is supplemented with 0.2 mM NADH and 5 units per ml of lactate dehydrogcnase, and the loss of absorbance at 340 nm is followed \\-ith time.
Initial reaction \ elocities are measured in a rerording spectrophotometer, and are linearly proport'ional to the amount of cysteine desulfhydrase added. This procedure measures only the rate of free pyruvate production, since lactate dehydrogenase does not react with the pyruvate derivative.
Stoichicmetry-In experiments designed to determine the stoichiometry of the cysteine desulfhydrase reaction, 5 ml of a solution containing 2.0 rnM L-cysteine in 0.1 1~ Tris-HCl, pH 8.4, in a Thunberg tube (150 X 18 mm) [I-ere deaeratcd by bubbling nitrogen for 5 min through an aperture specially fitted to the bottom of the tube. Approximately 0.3 unit of purified cysteine desulfhydrase in a small volume of solution was then added to start t'he reaction, which was carried out for 15 min at 23" while nitrogen was continuously bubbled through the incubation misture. Hydrogen sulfide n-as collected by directing the gas outlet stream through 70 ml of a solution containing 0.5 gram of zinc acetate and 0.75 gram of sodium acetate in a loo-ml volumet,ric flask. The reaction was terminated by the addition of 1.0 ml of 1.0 N HzS04, and the remaining hydrogen sulfide was collected for an additional 10 min. Control experiments, using standard solutions of sodium sulfide, sho\T-ed that 95 to 100% of the added sulfide could be collected in this manner.
To the solution of zinc acetate were then added in rapid succession 10 ml of 0.02 hl N ,N'-dimethgl-p-phenylenediamine sulfate in 7.2 N HCI and 10 ml of 0.03 %f ferric chloride in 1.2 N HCl. The flask was quickly stoppered, shakenvigorouslyfor 1 min, and the total volume was adjusted to 100 ml with water.
After storage for 15 min in the dark, the absorbance at 650 nm was determined.
The ~650 for sulfide under these conditions was found to be 2.67 X lo4 M-' cm-l.
The acidified reaction mixture was then diluted with water to a volume of 25 ml and assayed for cysteine with 5,5'-dithiobis(Znitrobenzoic acid) (17), for ammonia by t)he glutamate dehydrogenase method of Su et al. (18), and for free and total pyrurate. xc'0 detectable sulfide remained in the reaction mixture.
ORDl and ultraviolet spectra were obtained using automatic recording spectrophotometers, Cary models 60 and 15, respectively.
Protein determinations were performed by the biuret method (19), and autoradiography was done as previously described (20).
Reaction mixtures to be analyzed for alanine were first oxidized with performic acid (21), lyophilized, and then adsorbed to a column (5 cm x 1 cm%) of Dowex 50W-Hf (X8,200 to 400 mesh) at pH 2. After washing the column with water to remove cysteic acid, alanine and other amino acids were eluted with 1 N KHIOH, concentrated by lyophilization, and analyzed on a Beckman model 121 amino acid analyzer.
Recoveries subsequent to performic acid oxidation and prior to amino acid analysis were estimated by adding a small amount of [lJ4C]glycine to each sample.

RESULTS
Initial attempts to quantify the products of the cysteine desulfhydrase reaction revealed an unusual stoichiometry, which varied with the stage of enzyme purification (see below). The data in Table I show that, using purified enzyme, the molar yields of sulfide and ammonia are equal, while the amount of free pyruvate detected is less than 10 y0 of that expected from the accumulation of the former two products.
Furthermore, the disappearance of cystcine from the reaction mixture is greater than can be accounted for by the yield of any one of these three products. As measured after preincubation of the reaction mixture with acidic mercuric ion, the yield of total pyruvate nearly equals that of ammonia or sulfide.
In addition, that portion of the total pyruvate which is not detectable as free pyruvate is approximately equal to the amount of cystcinc not accounted for by the sum of the total pyruvate formed and the cysteine remaining at the end of the reaction.
We hare accounted for this unusual stoichiometry by identifying a mercuric ion-labile conjugate of cystcine and pyruvate, provisionally designated Compound CP, as a product of the cysteine desulfhydrase reaction.
Preparation of Compound CP-The enzyme used was a fraction which had been purified through the first ammonium sulfate step (12) and then desalted at room temperature by gel filtration through a Sephadex G-50 column, equilibrated with 0.1 M Tris-HCl, pH 8.4. The specific activity of this preparation was 1 unit per mg, representing a 4-fold purification from the crude extract.
Three hundred units (12 ml) of cysteine desulfhydrase were added to 380 ml of a solution containing 40 mmoles of L-cysteine (free base) adjusted to pH 8.4 with 5 N NaOH.
During the entire course of the reaction, the mixture was stirred vigorously at room temperature in an open beaker, and the pH was kept at 8.0 to 8.4 by the addition of NaOH.
An additional 40 mmoles of dry L-cysteine were added after 3 hours of incubation, at which time the total pyruvate concentration was 0.034 M. Four hours later, total pyruvate was 0.054 RI and another 175 units of enzyme were added.
After an additional 16 hours of incubation, the total pyruvate concentration had reached 0.079 M (32 mmoles), and Dhe free pyruvate concentration was 0.006 M. The solution was adjusted to pH 7.6 with glacial acetic acid and was filtered through Whatman No. 1 paper.
After the addition of 4 volumes of cold absolute ethanol, the filtrate was chilled to -20" and refiltered.
Following concentration to a volume of 100 ml in a rotary evaporator at a temperature not exceeding 1 The abbreviation used is: ORD, optical rotatory dispersion.  I Stoichiometry of cysteine desulfhydrase reaction Reactions and assays were carried out as described under "Experimental Procedures," using 5 ml of reaction mixture containing 2.0 mM rJ-cysteine at pH 8.4 as substrate.

6189
Highly purified cysteine desulfhydrase was added at a concentration of approximately 0.06 unit per ml to start the reaction, which was terminated 15 min later by the addition of acid. 45", the solution was titrated to $1 7.6 and filtered. Four volumes of absolute ethanol were added, and the solution was again filtered and reconcentrated to a volume of 40 ml. The pH of the concentrate was adjusted to 7.6, and then it was filtered first through Whatman No. 1 paper and then through a Millipore (0.45 ~1) membrane filter.
The addition of 19 volumes of icecold absolute ethanol to this solution resulted in the formation of a white, gel-like precipitate.
After storage at -20" overnight, the precipitate was collected by filtration, washed with cold absolute ethanol, and dried in vacua.
The yield was 5.8 g. The dried material was dissolved in water at room tcmperature (200 mg per ml), and, after the addition of 4 volumes of cold absolute ethanol, the turbid, yellow solution was clarified by passage through a Millipore membrane filter. Absolute ethanol was added to a final concentration of 95%,, and after 2 hours at -20" the resultant precipit,ate was collected by filtration and dried in uacuo. The yield was 4.5 g. A second reprecipitation gave Compound CP in 3.1 g yield.
Characterization of Compound CP-To 40 ml of a solution containing 500 mg (1.96 mmoles of total pyruvate) of Compound CP were added 20 ml of 0.5 M HgClz in 1.0 N I-ICI. A white precipitate formed which, after adjustment of the solution to pH 3.0 with concentrated NH40H, was collected by filtration, washed with two 5-ml portions of water, and set aside for further analysis.
Excess mercuric ion was removed from the filtrate and washes by passage through a column (20 cm x 2.5 cmz) of Dowex 50W-H+ X8, following which the column was eluted with water. The eluate was assayed by both the lactate dehydrogenase and 2,4-dinitrophenylhydrazine methods and was found to have a total of 2.3 and 2.0 mmoles of pyruvate, respectively, in a volume of 96 ml. The 2,4-dinitrophenylhydrazone derivative II-as prepared from this solution (22), and after recryst'allization from 955; ethanol-ethyl acetate a yield of 197 mg (0.73 mmole) was obtained.
This material was characterized as the derivative of pyrurate by its meking point (222-223O; authentic 221-222"; mixed 221-223"), and by its mobility in three thin layer chromatography solvent systems.
The precipitate from the initial acidic HgC!& step T\-as dissolved in 80 ml of 1.0 1\~ HCl and treated as previously described for the isolation of cystine (20). After two reprecipitat'ions, a yield of 145 mg (equivalent to 1.21 mmoles of half-cystinc) was obtained. The product IT-as identified as L-cystine by its mobility in three thin layer chromatography systems, an [a]:' of -205" (0.1 y0 in 1 x HCl; authentic L-cystine, -203"), and its ability to serve, after reduction with dit'hiothreitol, as a subskate for purified cysteine desulfhydrase.
The compound, 2-methyl-2,4-thiazolidinedicarboxylic acid, is a conjugate of pyruvate and cysteine, which by analogy with ot,her thiazolidines might be predicted to decompose in the presence of mercuric ion, giving as produck free pyrurate and the mercuric mercaptide of cysteine (23). The free acid of this thiazolidine derivative was chemically synthesized from pyruvic acid and L-cysteine by the method of Schubert (24), and recrystallized from hot water.
il 0.8 M solution was t'itratcd t'o pH 9 with concentrated KaOH, and the disodium salt was precipitated by the addition of 19 volumes of cold ethanol.
After two reprecipit,ations the dried product was compared Ivith Compound CP.
Both compounds I\-erc found t,o contain negligible amounts of free pyruvate, thiol, sulfide, and ammonium ion, while giving 1 mole of total pyrwatr per 245 to 255 g of material.
The results of elemental analyses were as folloTT-s: Two moles of sodium \vere found per mole of total pyruvate, indicating that both compounds are disodium salts. The two compounds camlol be distinguished from each other by thin layer chromatography in t,hree solvent' systems, and have identical infrared spectra. A comparison of the ORD spectra from 225 to 400 nm for t'hc disodium salts of both products shows them to be virtually identical TTith a single Cotton effect noted at 253 nm (Fig. 2). Ultraviolet spectra of both compounds are also similar kth a shallow shoulder at 250 to 255 am and an ~~~0 of 150 X1-l cm-l.
The free acid of Compound CT was prepared by adding 0.7 ml of 11.6 s HCl to 4 ml of a solut'ion containing 800 mg of the disodium salt, followed by crystallizat,ion in the cold. The resultant crystals were collected by filtration and recrystallized from hot water giving a yield of 78 mg. The melting point of the free acid of Compound CP is 163-164"; authentic 2-methyl-2,4thiazolidinedicarboxylic acid, 163-164"; mixture 163-164".
We conclude from t,hese data that Compound CP is the disodium salt of 2-methyl-2,4-thiazolidincdicarbosylic acid. Furthermore, although such an analysis cannot completely rule out the possibilitJy of stereochemical differences between the two compounds, the ORD and ultraviolet specka, together with the identical melting points of the free acids, constitute excellent evidence for the stcreochcmical identity of Compound CP kth Since both syntheses started with L-cysteine as a reactant, and the chemical synthesis utilizes pyruvate, we feel that both arc probably equal mixtures of diastereomcrs at C-2 with the configuration of the o( carbon atom of L-cysteine at C-4. Fraction B-Using 2.0 rnM L-cystcinc at pH 8.6, the portion of enzymatically produced total pyruvate appearing as free pyrurate varies ITith the stage of enzyme purification from a total pyrurate to free pyruvate ratio of 2, using a crude extract, to a ratio of approximately 6, using highly purified enzyme. Thus, if t'he progress of enzyme purification is follow-cd using the usual types of pyrwate assays rather than the tot'al pyruvate or sulfide assays, an apparent large loss of activity occurs after the first ammonium sulfate precipitation step (12). The greater re1at'iv-e yields of free pyruvate noted with crude preparations of cystcinc desulfhydrase can be attributed to the presence of a factor which we have designated Fraction B. Preparations of this substance can be obtained which have no appreciable cysteinc desulfhydrase activity, but nhkh, when added to reaction mixtures cont'aining pure cystciric desulfhydrase, increase the yields of free pyruvate without affecting the rates of total pyruvate or sulfide product'ion (Table I). Fraction B does not convert purified Compound CP to pyruvate or sulfide, even in the presence of purified cysteine desulfhydrase, but the addition of this factor to a cysteine desulfhydrase reaction misture, in I&i& Compound CI' has already accumulated, results in a decrease in the total pyrurate to free pyrurate ratio of products formed after such addition.
Thus the action of Fraction B seems to be to prevent the formation of Compound CP during the cyst'eine desulfhydrase reaction rather than to degrade it.
In our attempts to devise a quantit'ative assay for Frartion B rre have found that a linear relationship exists between the total pyruvate to Compound CP ratio (where Compound CP is assumed to be t'he difference bctlwcn total pyrm-ate and irec pyruvatc) and the amount of Fraction B added to a cysteinc desulfhydrase reaction mixture (Fig. 3). Thus our standard assay consists of adding Fraction B to 1.0 ml of a standard incubation mixture containing 0.05 unit per ml of purified cysteine desulfhydrasc and measuring the total pyruvate to Compowd CP ratio after a 5-min incubation.
A control in l&ich Fraction B is omitted is also run and the diffcrcnce in the total pyruvate to by guest on March 24, 2020 http://www.jbc.org/

Downloaded from
Compound CP ratio is dctc~rmincd.
One unit of activity is defined as that amount of Fraction B which CHUSCS an increasr in the total pyruvatr to ('ompound CP I,atio of 1.0 under these standard conditions. 'fh(~ assay is wcful between the limits of 0.3 to 5 units of Fractioil B activity per ml of reaction mixture. Fraction B was purified from frozen cells of S. typhimurium, LT2, grown in minimal salts-~lucoar media containing cithcr 1.0 mM L-cystinc or 0.5 nw L-djrnkolate as the sole sulfur source (20). We find that levels of Fraction B activity are independent of the sulfur source used for growth, and for that reason djew kolate-grown cells were used in the preparation described hcrr to eliminate the l)ossibility of cwntamination of Fraction B with cystcine desulfhydrase.
Following ceIltrifugation at 40,000 X g for 30 min, the supernatant was rrmovrd and treated with 0.4 volume of 10% streptomycin sulfa@ pH 7.0. After 10 min of stirring at room temperature, t,hc Ilrecipitate was removed by centrifuaation and ammonium sulfate, 210 mg per ml, was slowly added to the supernatant with stirring. Following: centrifugation, the supcrnatant horn this step was heated to 90" in a boiling lx-ate1 bath, and, aftrr 1 min at that temperature, cooled in an ice bath. Coagulated protein was removed by centrifugation, and Fraction B activity was preripitatcd by the addition of an additional 280 mg per ml of ammonium sulfate to the supernatant.
This precipitatc was dissolved in 0.1 JI Tris-HCl, pH 7.6, 0.5 M XaCl, and dialyzed at 4" against the same buffer.
This prorrdurc results in a 25 to 30.fold purification with a 45 to 500/, yield (Table II).
The ammonium sulfate and heat steps remove all cysteine desulfhydrase activity, whether the ~11s are growl on djrnkolate or cystine as a sole sulfur source. Fracation B activity is resistant to treatment with RNase and DiYase but is rapidly illactivated by treatment with small amounts of trypsin, which, after subsequent dilution, have no effect on the cysteine dcsulfhydrasc assay itself. The purified material is relatively stable when stored frozen, losing approximately 1O70 of its activity prr mo;itli at -20".

Other Factors
InJluencing Synthesis oj' Compound Cl'-The cysteine desulfhydrase-mediated synthesis of Compound CP is markedly dependent upon $1 and cysteine caoncentration.
Using 2.0 mM L-cysteine and purified cysteine desulfhydrase, the total pyruvate to free pyruvatc ratio illcwases from a value of 2 at $1 7.2 to a value of about 6 at 111-I 8.6 ( Fig. 4A). At a constant 1)1-I of 8.6, the total pyruvatc to free pyruvate ratio is directly but not linearly, proportional IO x>-cysteille concentration, and cstrapolates to a value of 1 at, zero cyst&e concentration (Fig. 4B). Other investigat,ors have prwiously postulated (8, 10) and demonstrated (25) the Irollenzymatic formation in aqueous solutions of addwts b&wcII cysteillc and certainly carbonyl compounds.
Therefore, studies wore performed to evaluate the extent to which the Ilollc~tlzymatic formation of Compound CI' occurs. Fig. 5 s11ows the results of rspcriments in which L-cysteinc and sodium pyruvate at several different concentrations were incubated ill 0.1 M 'I'ris-IICl, pH 8.6, at 23" for varying periods of time. Usilla 2.0 ~JI I,-c)-stcine and 0.2 mM pyruvate, no appreciable loss of free pgruvatc could be detected even after 90 min of incubation.
At hi&w concentrations of both substratcs, however, significant losses of free pyruvate were noted  stant. Under these wnditions the half-life of flee pyruvatc~ is 60 min at 20 mar L-cystrinr, 2.0 ~RI pyruratc, and 13 inin at 100 mM L-cysteine, 10 mu pyruvate.
Fl,action B has no effwt on the i,ate of iioiieilzymatic~ formation of mercuric ioli-labile pyruvatc.
Although no appreciable nonenzymatic loss of free pyruvatcs ocrurs at the cysteine and pyl'uvate collc~elltratiorls presrnt' in our routine cysteine dcsulfhydrasc assay, all of our analytical data have been obtained on ('ompound CP which leas prepared using 0.1 M L-cysteinc as substxrte.
l'hereforr it is likely that at, least a portion of our enzymatically produced material was fol,rnc>d by a non-enzyme-depend& reaction bctn-ten L-cystcine and ~JTUvate. To establish the identity of Compound CP x\ ith the mercuric ion-labile pyl,nyatc made in the ~~KXXKT of cysteillr desulfhydrase at 10~ cystcine concentrations, reaction mirturca containing 2.0 11131 Q"S]cystcGne as substrate were analyzed f'ol radiolabeled Compound CP. Small llortions of thrse reaction mixtures wwc spotted on T'i'hatmaii Ko. 3811\1 I:apcr, and, after electrophoresis in 0.025 31 sodium citratr, $1 5.8, for 1 hour at 20 volts per cm, the 1:ositions of rlirlll~drin-llositive carrier conpounds wcrc compared with the locations of radiolabrl as tic,tected by autoradiography.
The arcas on the paper correspondwith time, wllile total pgrllvate concentrations remained con-ing to cysteine and Coml~ound Cl' wre then cut out and countc,d in a scintillation counter to obtain more quantitative results. When 0.4 mM pyruvate (approximately the amount produced in the enzymatic reaction) was included in a reaction mixture lacking cysteine desulfhydrase, only 1.77; of the total radiolabel was incorporated into material with the electrophoretic mobility of Compound CP. Fraction B had no effect on this nonenzymatic reaction.
In contrast, after incubation with cysteine desulfhydrasr, 12.2yb of the total radiolabel migrated with Compound CP, and the addition of Fraction B at a concentration of 20 units per ml decreased this incorporation to a level of 2.2% of the total rndiolabcl.
Autoradiogral)hy revealed a radioactive spot which esactly superimposed over the faintly ninhpdrin-positive area corresponding to added carrier Compound CP. These data substantiate the notion that under our usual assay,conditions the formation of 2-methyl-2,4-thiazolidincdicarboxylic acid is de-I'cndent upon the cysteine desulfhydrase reaction. Subsfrate Spec$city and pH Optinluni-Among potential substratcs thus far tested, purified cysteine desulfhydrase is quite sprcific for I>-cysteine. lncubntion of the enzyme with L-cys-oL---+ IO 20 Cysfelne (m&4) total pyruvate concentration of approximately 0.2 mM after 5 min of incubation. B, varying concentrations of L-cysteine were incubated for 5 min with cysteine desulfhydrase in 0.1 M Tris-HCl, pH 8.6.
The other compounds tested for substrate activity inhibit the enzyme less than 15y0 at concentrations of 0.5 to 2.0 MM.
The pH optimum of cysteine dcsulfhydrase in 0.1 M Tris-HCl is 8.6, with a rather sharp decline in activity at pH levels below 8.3. The activity at pH 7.0 is less than 5y0 that observed at 8.6.
Kinetic Studies-Kinetic studies of cysteine desulfhydrase have been complicated by the potent inhibition of the enzyme by its product, sulfide.
One approach to this problem has been to measure rates of pyruvate production in uncapped reaction tubes, which allow diffusion of hydrogen sulfide from the reaction mixtures (9). Due to the quantitatively uncertain extent of sulfide diffusion under such conditions and the lesser sensitivity of the pyruvate assay, u-e have chosen to measure rates of sulfide production and to analyze our results after correcting for sulfide inhibition.
Substrate-velocity studies \vere carried out by measuring the accumulation of sulfide ad a function of time at different concentrations of L-rysteine.
Initial velocities of reaction, Vi, were then estimated graphically using the t050 for sulfide appropriate for each L-cysteine concentration (see Fig. 1). As shown in Fig. 6, a plot of T/i versus z-cysteine concentration gives a sigmoidshaped curve with a half-maximum Vi at 0.21 mM L-cysteine.
A plot of l/Vi versus l/S reveals that at L-cysteine concentrations greater than 0.3 mM a straight line is obtained, and that at points corresponding to lower substrate concentrations the slope of the line increases (Fig. 6, inset). The apparent K, for L-c>-steine calculated from the linear portion of the double reciprocal plot corresponding to higher cyst&e concentrations is 0.17 mM. The Vi at each cysteine concentrat,ion was determined as the initial rate of sulfide production as described under "Results." The inset shows the plot of l/Vi WTSWS l/S. Treating the data according to the method of Hill (26), a plot of la [Bi/(V, -Vi)] versus In X gives a straight line with a slope of 1.9 (Fig. 7). Thus the dependence of the reaction rate on L-cysteine concentrations shows positke cooperativity with an n value of almost 2 at substrate concentrations less than 0.03 m&r. The rate of sulfide production is unaffected by sodium pyruvate, Compound CP, or ammonium sulfate when added either singly or in various combinations at concentrations of 0.025 InM or 0.5 mM. Preincubation of cysteine desulfhydrase with 0.02 mM sodium sulfide, however, leads to a partial inhibition of activity which is unrelated to the time of preincubation for at least 10 min. Removal of the sulfide by dilution results in a loss of inhibition to a level expected by the lower concentration of sulfide. Since the inhibitioll appears to be very rapid and reversible, we have endeavored to describe it in terms based on the assumptions of steady state, rapid equilibrium kinetics. Due to the difficulties involved in estimating initial velocities by the sulfide method in solutions to which exogeneous sulfide has already been added, we have carried out our inhibition studies by measuring the time-dependent accumulation of endogeneously formed sulfide in the absence of added sulfide. Fig. 8 shows product (sulfide) versus time curves at five different concentrations of L-cysteine using a constant amount of enzyme. The shapes of these curves indicate that the percentage of inhibition at a given sulfide concentration is markedly dependent on the cysteine concentration, and is greater at lower substrate concentrations.
Fraction B at a concentration of 10 units per ml has no effect on the shape of such curves.
We find that under certain conditions sulfide inhibition appears to be of the linear competitive type (27) which can be described by : which holds only when Q = 0 at t = 0. Thus when the sulfide concentration is measured as a function of time under conditions where the change in cysteine concentration is small (less than 10% in our experiments) and the sulfide concentration is zero at time zero, a plot of Q versus t/Q should give a straight line with a y intercept equal to -2K, obtained to sulfide concentrations as high as 0.09 m&l (Fig. 9). Using lower substrate concentrations a straight line can be drawn through points corresponding to low sulfide concentrations, but at higher sulfide concentrations and longer incubation times the points describe lines which becomc concave downward.
Since prrincubation studies show no time-dependent effect of sulfide on the enzyme activity, we conclude that, under the combined conditiolls of low qxteine concentration (less than 1 IrIM) and high sulfide conrcntrntion (greater than 0.05 m&f) the inhibition of cysteille desulfhydrase by sulfide is nonlinear.
If one assumes that the linear portions of the curves obtained by plotting Q versus f,/Q reflect' cwnditiolls where sulfide inhibition is of the linear competitive Our tiata ii&ate that during the cyst&e desulfliydrase rcxtion, a l,ortion of the total desulfuratcd cystcinc, as measured by t,he production of sulfide, reacts with additional cysteine to give 2-met,hy-2,4-thiazolidinedicarboxylic acid. Thus for every mole of thiazolidine formed, 2 lnoles of cysteiuc are consumed, releasing 1 mole of sulfide, 1 mole of ammonia, and no free pgruvate. This s&me fits wc~ll with the stoichiomrtric data presented in Table  I  and free pyruvate confirm these earlier obserl-ations and favor the notion that the product of this reaction is the thiazolidine. Several investigators have suggested that the immediate products of the cyst&e desulfhydrase reaction are sulfide and 2aminoacrylate, and that the latter compound, being unstable in aqueous solution, is spontaneously alld rapidly hydrolyzed to ammonia and pyruvate (2, 3). Thus, while the nonenzymatic fol,mationof mercuric ion-labile pyruvatc from free pyruvate and cystcGne is low to undetectable under the conditions of our assay, it is possible that a reaction between 2-aminoacrylate and cys-t&w might, occur readily.
The immcdiatc product would be a thiolwmiketamine, which could then cyclize to the thiazolidine, liberating ammonia in the process (Fig. 11). Our data showing that the total pyruvatc to free pyruvate ratio estrapolates to a value of 1.0 at zero cystcine concentration (Fig. 4B) are consistent with the prediction of this model that the amount of Compound CP formed should br directly proportional to the cgsteine concentration.
In an experiment desigwd to demonstrate the existence of 2-amilroacrylatc as an intermediate in the cysteine dcsulfhydrase reaction, sodium borohydride n-as added to reaction mixtures on the assumption that any 2-aminoacrylate lpresent would be reduccd to alanine.
The data presented in Table III show that when a complctc reaction mixture was treated with borohydride, the amount of alanine recoxwcd was considerably more than that found in control mixtures lacking enzyme or cysteine.
The lesser yield of alanine noted with the higher concentration of borohydride is probably explained by the fact that while cysteine desulfhydrase retains aplxoximately 50yc of its activity in the lxrscnce of 0.2 mM borohydride, the enzyme is rapidly and complet,ely inactivated by 10 mu borohydride.
Therefore the ala- ;H2 -bH-CO; Although the in vitro formatiolI of 2.methyl-2,4-thiazolidinedicarbosylic acid under the conditions of our standard assay is pertinent to an undcrst'anding of the enzymology of cysteine desulfhydrasc, thr in viva significance of this reaction is problematic. rnder the conditions of pH and cysteine concentration which one would expect to find irz Go, it is unlikely that very much thiazolidine would bc made, particularly in the presence of Fractioll B.
The mechanism by which Fractioll B prevents thiazolidine formation is c~ompl&ly unknown.
Presumably it is an enzyme 6195 which reacts catalytically with an intermediate in the reaction to give free pyruvate. If 2-aminoacrylate is in fact a precursor of the thiazolidine, Fraction B might facilitate its hydrolysis to pyruvate and ammonia, either directly or by catalyzing an eneamine tautomerization to give that tautomer which is less reactive with cysteine or more readily hydrolyzed.
Alternatively Fraction B might hydrolyze the thiohemiketamine before cyclization takes place or perhaps even interact with cysteine desulfhydrase itself in such a way as to enable it to release pyruvate and ammonia directly.
The exact role of Fraction B in cellular metabolism is also unclear, since it is found at the same concentrations in cells either grown 011 cysteine or starved for sulfur by growth on djenkolate.
Perhaps Fraction B is an enzyme of general usefulness to the cell rather than being limited to a single function related to cysteine catabolism.
The results of our kinetic studies are essentially in agreement with those of Collins (9), who, using a part'ially purified preparation of enzyme and a different, assay, found cooperative kinetics for cysteine with a K, of 0.22 m&l and an n value of 1.9. Collins also studied the inhibition of the enzyme by sulfide and found evidence for mixed inhibition kinetics with a Ki (K 4 in our terminology) for sulfide of 0.007 m&f.
It may be of some significance that sulfide inhibition shows a greater deviation from linearity at lower cysteine concentrations M-hero the dependence of the reartiou rate upon cysteine concentration is positively cooperative.
The csaggeration of product inhibition at these low cyst,eine concentrations may be related to the ability of sulfide to interfere with cooperativity either by competing with cysteine for an allosteric site or by otherwise preventing the enhancement of enzyme activity related to an allosteric event.