THE EFFECT OF BISULFITE 0N CERTAIN OXIDIZING ENZYMES Thesis for the Degree of Ph. D. MICHIGAN STATE UNIVERSITY RICHARD J. EMBS 1969 mail! This is to certify that the thesis entitled Effect of Bisulfite on Certain Oxidizing Enzymes presented by Richard J. Embs has been accepted towards fulfillment of the requirements for Ph.D. Food Science degree in Momée4 GUM/ca Major professor Dam March 282 1969 0-169 ABSTRACT THE EFFECT OF BISULFITE ON CERTAIN OXIDIZING ENZYMES BY Richard J. Embs Investigations were carried out on the effects of bisulfite on the following oxidizing enzymes: peroxidase, catalase, lipoxygenase, phenolase, and ascorbic acid oxidase. It was found that 0.05-0.1M bisulfite retards the inactivation of horseradish peroxidase solutions (0.5-10 purpurogallin units per ml) by weak acids. Spectral anal— ysis indicates that it accomplishes this by stabilizing the linkage between the iron-containing prosthetic group and the protein. Cyanide, azide, and fluoride, which form reversible complexes with peroxidase iron, exert a similar effect; thus, it is inferred that bisulfite also forms a complex with peroxidase iron. A kinetic method was used to calculate a dissociation constant of 0.02M for the bisulfite-enzyme complex. A solution of bovine liver catalase (2640 units per ml) was nearly inactivated by twelve days of contact " Richard J. Embs with 0.1M bisulfite. The apparent cause is denaturation of the catalase protein by bisulfite. The oxidation of linoleic acid by lipoxygenase in dried pea extract is slightly accelerated in the first 20— 25 minutes of the reaction by 1.6 x 10-4M bisulfite, and later inhibited. The enhancement may be due to the bisul— fite destruction of free radicals and hydroperoxides which may inhibit the enzyme. The bisulfite inhibition, which occurs later in the reaction, is probably caused by an attack on the enzyme by the bisulfite itself. The oxidation of monophenols by phenolase is stimu- lated by low levels (lo—SM) of catechol and other reducing ’5M) of bisulfite. agents and inhibited by low levels (10 This inhibition does not appear to be due to any enzyme— bisulfite interaction because the extent of the inhibition is nearly independent of enzyme concentration for low cresolase preparations. The bisulfite inhibition is over— come by reducing agents, and conversely, the stimulation of monophenol oxidation by reducing agents is diminished by bisulfite. This indicates that monophenol oxidation by phenolase involves a reaction not directly controlled by the enzyme, but in which a reducing agent participates and bisulfite interferes. A mechanism for this reaction is proposed. A crude preparation of ascorbic acid oxidase (0.0026 units per ml) was readily inhibited by 10—4M ' Richard J. Embs bisulfite, apparently because of denaturation of the enzyme protein by the bisulfite. ADDENDUM: The effect of 50—480 ppm 802 solutions on the growth of Saccharomyces cerevisiae was also investigated. Evidence is presented to show that the sulfurous acid mole- cule, and not bisulfite or sulfite, is the antiseptic agent. Its antiseptic potency depends on its rate of uptake by the yeast cells and not on the total amount taken up. THE EFFECT OF BISULFITE ON CERTAIN OXIDIZING ENZYMES BY Richard J; Embs A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Food Science 1969 ACKNOWLEDGMENTS The author wishes to express his sincere apprecia- tion to Dr. Pericles Markakis for his guidance, aid and encouragement during the research for this project and in the preparation of the manuscript. He is also indebted to Dr. Georg Borgstrom, Dr. J. Robert Brunner, Dr. Clifford Bedford, and Dr. John C. Speck for their advice and help in the preparation of this manuscript. ii TABLE OF CONTENTS LIST OF TABLES . . . . . . . . . . LIST OF FIGURES . . . . . . . . . CHAPTER I. INTRODUCTION . . . . . . . II. REVIEW OF THE LITERATURE . Peroxidase . . . . . . Catalase . . . . . . . Lipoxygenase . . . . . Phenolase . . . . . . . Ascorbic Acid Oxidase . III. METHODS AND MATERIALS . . IV. RESULTS AND DISCUSSION . . Peroxidase . . . . . . Catalase . . . . . . . Lipoxygenase . . . . . Phenolase . . . . . Ascorbic Acid Oxidase ADDENDUM: The Effect of Bisulfite on the Growth of Saccharomyces cerevisiae . . . . . REFERENCES . . . . . . . . . . . . iii Page iv \lU‘lsbuhk/o l3 13 23 24 26 43 47 61 LIST OF TABLES Page The effect of bisulfite on the activity of horseradish peroxidase solutions . . . . . 14 The effect of sulfate and bisulfite on the activity of HRPR solutions at pH 4.0 . . . 15 Loss of enzymatic activity of solutions A and B With time 0 O O O O O O O O O O O O 19 Effect of catechol on the browning rate of phenolase-tyrosine reaction mixtures . . . 28 Effect of bisulfite on the browning rate of phenolase-tyrosine-catechol reaction mixtures . . . . . . . . . . . . . . . . . 30 Effect of cuprous ions on the bisulfite inhibition of the browning rate of a phenolase-tyrosine-catechol reaction mixture . . . . . . . . . . . . . . . . . 32 Effect of cuprous ions on the cysteine inhibition of the browning rate of a phenolase—tyrosine-catechol reaction mixture . . . . . . . . . . . . . . . . . 33 Effect of catechol on the bisulfite in— hibition of browning of a phenolase— tyrosine reaction mixture . . . . . . . . 34 Effect of increasing enzyme concentration on the bisulfite inhibition of brown- ing by phenolase-tyrosine-reducing agent reaction mixtures . . . . . . . . . . . . 41 Growth of yeast cells after exposure to 200 ppm 802 at pH's 3, 4, 5, 6, and 7 . . 55 Growth of yeast cells after exposure to 480 ppm 802 at pH's 3, 4, 5, 6, and 7 . . 56 iv 10. LIST OF FIGURES Effect of bisulfite on the spectrum of HRPR O O O O O O O O O O O O O O O O O Mildvan-Leigh plot of the HRPR-H803 interaction 0 O O O O O O O O O O O 0 Effect of bisulfite on the oxidation of linoleic acid by lipoxygenase . . . Effect of varying concentrations of hi- sulfite on the rate of ascorbate oxidation by ascorbate oxidase in 0.1 ml of cucumber juice . . . . Effect of 30-min. preincubation with 1.5 x 10‘4M bisulfite on the ascorbate oxidase activity of 0.1 m1 cucumber peel juice . . . . . . . . . . . . Relative concentrations of H 803, HSOS, " ' I and 803 at various pH 5 . . . . . . Growth of yeast in the pH range 3-7 Uptake of sulfur by yeast cells exposed to 200 ppm 802 at pH's 3, 4, 5, 6, and 7 . . . . . . . . . . . . . Uptake of sulfur by yeast cells eXposed to 480 ppm 802 at pH's 3, 4, 5, 6, and 7 O O O O O O O O O O O O O O Uptake of sulfur by yeast cells exposed to 600 ppm 802 at pH's, 3, 4, and 5 Page 19 22 25 44 45 51 52 57 58 60 I. INTRODUCTION Sulfur dioxide, along with its salts, is probably the most versatile of all the chemical agents used in the food industry. It is used widely as a preservative to re- duce or prevent spoilage by microorganisms and as a selec- tive inhibitor of undesirable organisms in the fermentation industries. It is also used to inhibit enzymatic and non— enzymatic oxidative discoloration of many food products, and to protect ascorbic acid, carotene, and other biologi- cally active components. The pronounced bleaching action of SO2 has been employed in the preparation of specialty products such as Maraschino cherries, and for decolorizing sugar-cane and sugar—beet juices in sugar manufacture. Sulfiting of grapes reduces molding and may repel insects. Treatment of flour with sulfur dioxide modifies its baking characteristics by breaking disulfide bonds in the gluten molecules. The main disadvantages of sulfur dioxide are its unpleasant flavor and its destructive action in thiamine (vitamin B1). In spite of its wide commercial use, the chemical mechanisms involved in the various effects of 802 are not well understood. The purpose of this study was to clarify some of the mechanisms relative to the effect of SO2 on certain oxidizing enzymes important in the food field. Since there is no one common name for sulfur di- oxide and its salts, the term "bisulfite" will be used in describing the 802 solutions referred to in this thesis, because bisulfite is the dominant form among the various molecular species of 802 in the mildly acidic pH range of most interest in the food field. II. REVIEW OF LITERATURE Peroxidase Peroxidase (Doner: H202 oxidoreductase 1.11.1.7) is a hemiprotein which catalyzes the reaction: ROOH + H2 x (or 2xm%—————9 ROH + H20 + x (or 2xm+l) where H2x is an oxidizable substrate such as ascorbic acid, a phenol, a short chain alcohol, etc. and xm is a metallic ion (Mahler and Cordes, 1966). Therefore, its presence in fruit and vegetables can result in discoloration and loss of vitamin C. The effect of sulfurous acid and its salts on per- oxidase has not been extensively studied. Klebanoff (1961) reported that sulfite activated the nonperoxidative oxida- tion of NADH and NADPH as catalyzed by peroxidase. Chmiel- nicka (1963, 1964) and Monikowski and Chmielnicka (1964) found that 802 inhibited peroxidase activity both in pure systems and in raw vegetables and their products. Maehly (1952) noted that acids in general split the iron—containing prosthetic group of horseradish peroxidase from its protein, thereby inactivating the enzyme. The half—life of the active enzyme in acidic solution varies inversely with the pH of the solution and with the ionic strength. The purpose of our study was to observe and ex- plain some of the effects of bisulfite on the enzymatic activity of horseradish peroxidase. Catalase Catalase (H202:H202 oxidoreductase 1.11.1.6) is an iron—containing protein which catalyzes the following reaction: 2H202—-—> ZHZO + 02 Catalase is quite similar to peroxidase in that it contains a protohematin prosthetic group and catalyzes a similar type of oxidation-reduction reaction. Also, the linkage between the iron atom of the prosthetic group and the protein is susceptible to rupture by acid (Lewis, 1954). The question then arises as to whether bisulfite can exert a protective effect on catalase in acid media as it does with peroxidase. To clarify this issue, a study was made on the effect of bisulfite on the enzymatic activ— ity of catalase. Lipoxygenase Lipoxygenase, or lipoxidase, (1.99.2.1) catalyzes the oxidation of linoleic, linolenic, and arachidonic acids. The reaction proceeds as follows: -CH=CH-CH2-CH=CH- + 02—9 -CH=CH-CH=CH-CH(OOH)- cis cis cis trans A complex series of free radicals and hydroperoxides are formed in this reaction, and some of these may inhibit the enzyme even as it produces them. Evidence for this may be seen in the fact that hydrogen peroxide readily inhibits lipoxygenase (Mitsuda, et_§l., 1967). Lipoxygenase occurs in green peas and may be re- sponsible for chlorophyll and flavor deterioration in this product (Eriksson, 1967). Therefore, a method for inhibit- ing lipoxygenase may have some practical value. In this work, bisulfite was investigated as a possible inhibitor of this enzyme. Phenolase Phenolase (o-diphenol: O oxidoreductase 1.10.3.1) 2 is a copper-containing enzyme which catalyzes the reaction: 2 o—diphenol + O ——-——9 2 o—quinone + 2 H O 2 2 However, it also appears to catalyze the o—hydroxylation of monOphenols: monophenol + 2e_ + O -—————9 o-diphenol + O: 2 although there is some evidence that the monophenol is converted directly to the o-quinone (Dressler and Dawson, 1960): monOphenol + O -—————9 o-quinone + H O 2 2 The copper moiety of the enzyme is absolutely essential for enzymic activity and no other element can substitute for it (Kubowitz, 1938). The hydroxylating function of phenolase is re- ferred to as its "cresolase" activity while the oxidative function is its "catecholase" activity. The relationship between these activities is unknown and is a matter of some dispute. The following mechanisms have been proposed: 1. The monophenol is hydroxylated nonenzymatically by the highly reactive o-quinones produced by enzymic oxi- dation of an o-diphenol (Kertesz and Zito, 1962). However, this could not involve an oxidation-reduction reaction be- tween the monophenol and quinone as in the equation: monophenol + o-quinone + H O —————> 2 o—diphenol 2 because Mason et_gl. (1955) have shown that the hydroxyl group added to the monophenol molecule is derived from molecular oxygen and not from water. 2. The enzyme possesses one c0pper-containing active site. The copper catalyzes the oxidation of an o-diphenol while being itself reduced from the cupric to the cuprous state. The cuprous copper then catalyzes the hydroxylation of a monophenol (Mason, 1956). 3. The enzyme possesses two separate, c0pper- containing active sites, one for oxidizing diphenols to o~quinones and the other for the hydroxylation of mono- phenols (Dressler and Dawson, 1960). 4. Phenolase is a mixture of two different enzymes, one for the oxidation of o-diphenols and the other for the hydroxylation of monophenols (Macrae and Duggleby, 1968). None of these theories satisfactorily accounts for all the enzymatic properties of phenolase. Theory 1 fails to explain why the enzyme produces only o—diphenols (or o-quinones) when nonenzymatic hydroxylations are always randomly directed. Theory 2 would have difficulty showing, if there were only one active center, why bisulfite has a much more adverse effect on the cresolase activity than on the catecholase activity (Markakis and Embs, 1966; Muneta, 1966). Theories 3 and 4 do not account for the fact that minute amounts of an o-diphenol added to a phenolase-monophenol mixture greatly enhance the rate of monophenol hydroxylation. In the present work we investigated the effect of bisulfite on the phenolase-monophenol reaction and at- tempted to construct a hypothesis which would account for this effect and also elucidate some aspects of the pheno- lase-monophenol reaction. Ascorbic Acid Oxidase Ascorbic acid oxidase (1—ascorbate:O2 oxidoreductase 1.10.3.3) is a copper-containing protein that catalyzes the aerobic oxidation of L—ascorbic acid. It also catalyzes the oxidation of a limited number of phenolic compounds, such as 2,6-dichloroindophenol (Dawson, 1965). Means of preventing vitamin C loss by this enzyme may have practical significance for fruit and vegetable processors. Therefore, bisulfite was investigated as a possible inhibitor of this enzyme. III. METHODS AND MATERIALS The peroxidase used in this work was "Type II" horseradish peroxidase from the Sigma Chemical Company. Its activity was 110 purpurogallin units per mg. Enzymatic activity was determined by adding 0.1 m1. of the peroxidase solution, properly diluted (usually 1:40), to a mixture of 1 ml of 0.5M phosphate buffer, pH 6.5, 1 ml of 20 mM guaiacol, 0.1 ml of 10mM H202, ml H20. The reaction mixture was in a l-cm-path Beckman cuvette and the enzyme solution was added with a square and 0.8 teflon plunger provided with a groove and three orifices and connected to a stainless steel handle. Color formation in the solution was measured at 470 nm with a Beckman DU spectrophotometer connected to a Ledland log-converter and a Sargent SR recorder. The rate of reaction was measured by the tangent at the origin of the curve obtained with the recorder. Spectra were obtained with a Bausch & Lomb Spec— tronic 505 recording spectrophotometer. Twice crystallized bovine liver catalase from the Sigma Chemical Company was used. Catalase activity was determined spectrophotometri- cally by incubating 1 m1 of 0.059M H O with 2 ml of catalase 2 2 9 10 solution, properly diluted, in a 1-cm-path Beckman cuvette. The reaction was followed at 240 nm. The rate of reaction was taken as the tangent at the origin of the curve ob- tained with the recorder, in the same manner as in the peroxidase work. Pea lipoxygenase extracts were made by grinding dried, split peas in a Wiley mill with a 40—mesh sieve, washing the powder with petroleum ether to remove lipids, and then shaking a five gram quantity of the powder with 50 ml of 0.05M phosphate buffer, pH 7.0, for thirty minutes. The extract was filtered with No. 1 filter paper before use. Linoleic acid solution was made up as follows: 0.5 ml of linoleic acid and 0.5 m1 of Tween 80 were shaken vigorously with 10 m1 of (DJfld borate buffer, pH 9.0. Then 1.3 ml of 1M NaOH was added and the solution shaken until clear. Finally, 90 ml of 0.05M phosphate buffer, pH 7.0 were added and the solution diluted to 200 ml with demin- eralized water. This method of solubilizing linoleic acid was taken from Surrey (1964). Lipoxygenase activity was determined manometrically by adding 2.0 ml of the linoleic acid solution and 0.5 m1 of water (or bisulfite solution) to the main compartment of a Warburg flask and 0.5 ml of the pea extract to the side arm. A Warburg apparatus thermostated at 30° C was used in the subsequent measurements of oxygen uptake. 11 The "low cresolase" enzyme was mushroom tyrosinase from the Worthington Biochemical Corporation. The "high cresolase" enzyme was mushroom tyrosinase from the Sigma Chemical Company. Its activity was listed as 1050 units per mg. (1 unit = 0.001 absorbancy increase per min. at 280 nm, in 0.16M phosphate buffer, pH 6.5, 25° C, contain— ing 3 x 10—4 M L—tyrosine). Browning rates were determined at 470 nm with a Beckman DU Spectrophotometer, log converter, and recorder in the same manner as in the peroxidase and catalase work. The browning rate was taken as the tangent at the origin of the curve obtained with the recorder, and was expressed in arbitrary units. In paper chromatography, No. l Whatman paper was employed, with n-butanol-acetic acid-water (25:6:25) as solvent, in descending irrigation at 25° C. The juice squeezed from fresh cucumber peel was used as the source of ascorbic acid oxidase. Ascorbic acid oxidase activity was determined manometrically by adding 1.5 m1 of buffer (0.2M phosphate— 0.1M citrate, pH 5.7), 0.5 m1 of gelatin solution (750 mg gelatin in 150 ml water), 0.5 ml of ascorbic acid solution (250 mg of ascorbic acid and 50 mg of metaphosphoric acid in 50 m1 of water), and 0.9 ml of water (or bisulfite solu- tion) in the main compartment of a Warburg flask and 0.1 m1 of the peel juice in the side arm. Subsequent measurements 12 of oxygen uptake were carried out with a Warburg apparatus thermostated at 30° C. IV. RESULTS AND DISCUSSION Peroxidase When horseradish (HRPR) solutions were held for thirty hours at room temperature at the pH range 4.0 - 7.0, there was a loss of enzymatic activity at the acidic end of the range (Table l), presumably because of the loss of the prosthetic group as described by Maehly (1952). With sodium bisulfite present in the solutions, however, there was a considerable retention of activity at the lower pH levels. The opposite effect would be expected as the in- crease in ionic strength, caused by the addition of bisul— fite, should accelerate the detachment of the prosthetic group by acid. This is illustrated in Table 2, where it can be seen that the addition of sodium sulfate hastens the loss of enzymatic activity at pH 4.0, in contrast to the addition of bisulfite. The protective effect of bisulfite was also ob- served at pH levels below 4.0 (3.5 and 3.0), where the inactivation of peroxidase is much more rapid, taking place in only a few hours. The protective effect also exists in the absence of oxygen (Table 1). This indicates that the effect is not due to the antimicrobial action of 13 14 Table l.--The effect of bisulfite on the activity of horseradish peroxidase solutions. pH Without Bisulfite Containing 0.067M Bisulfite 4 37 87 5 80 92 6 100 96 7 100 98 Each solution contained 0.5 mg HRPR and 0.66M citrate-phosphate buffer in a total volume of 3 ml. The figures are the percent enzymic activity remaining after 30 hours incubation under anaerobic conditions. 15 Table 2.--The effect of sulfate and bisulfite on the activity of HRPR solutions at pH 4.0. Time (hr) NO SO4 or HSO With 0.05M SO With 0.05M HSO 3 4 3 0 100 100 100 1.5 91 69 94 3.0 80 45 86 4.5 72 29 82 6.5 59 17 78 8.5 43 8 72 10.5 34 0 71 Each solution contained 0.03 mg HRPR in 6 m1 0.05M citrate buffer, pH 4.0. The figures are the percent activ- ity remaining after varying periods of incubation. l6 bisulfite, in that the bisulfite preserves the enzyme by discouraging the growth of microorganisms (e.g., mold) in the solutions. A similar protective effect on the Soret absorption band (360—450 nm) was observed when bisulfite was present in the HRPR solutions at pH 4.0. Line 1 of Figure 1A represents the normal absorbance spectrum of HRPR in the Soret region, with a peak at 403 nm. Line 2 represents the spectrum of the same material six hours later: the absorbance had fallen off because a portion of the enzyme had lost its protohematin prosthetic group. Line 3 repre- sents the spectrum after twelve hours and line 4 after twenty—four hours of incubation. By that time the enzyme had become nearly colorless and inactive. Figure 1B illustrates the spectral curve of a sim— ilar solution of peroxidase at pH 4.0 with bisulfite added. It can be seen that the curve did not fall with time and after twenty—four hours the only change was a shift of the peak from 403 nm to about 395 nm. Both solutions were dialyzed for twenty—four hours at 5°C against demineralized water and their Soret spectra then compared. (Figure 1C). The peroxidase which had been treated with bisulfite had an almost normal spectrum, with its peak shifted back to 402 nm, whereas the untreated peroxidase had a flattened spectrum. l7 A Figure l.--Effect of bisulfite on the spectrum of HRPR. (A) Spectra of a solution containing 1 mg HRPR in 5 ml of 0.04M citrate buffer, pH 4.0. Lines 1, 2, 3, and 4 correspond to the spectra at 0, 6, 12 and 24 hours of incubation, respectively. (B) Spectra of same solution containing 0.1M bisulfite. (C) Line a is the spectrum of solution A, and b of B, both after 24-hour incubation followed by 24-hour dialysis. 18 The enzymatic activities of these solutions at the times their spectra were determined are given in Table 3. These findings indicate that bisulfite retards the loss of the prosthetic group by acid. The reversible shift in the spectral peak caused by bisulfite suggests a reac- tion between the bisulfite and the protohematin group, which stabilizes the linkage between the iron atom and the protein. This reaction could be a reversible complex for- mation between bisulfite and the enzymic prosthetic group. To test this hypothesis, four peroxidase solutions ‘ I with the same HRPR and buffer compositions as those in Table 1 were held for twenty-four hours at pH 4.0. Solu- tions 1, 2, and 3 also contained 0.033 m KCN, 0.033M NaN3, and 0.033M NH4F, respectively; the fourth solution was a control. Cyanide, azide, and fluoride all form spectro- sc0pically distinctive complexes with peroxidase (Keilin and Hartree, 1951). After twenty-four hours incubation the four solu- tions were dialyzed for thirty-six hours against demin- eralized water at 5°C. With the removal of the complexing agents the normal Soret spectrum of each peroxidase solu- tion was restored, but the spectrum of the control solution was very flat in comparison. When the enzymatic anitivities of the dialyzed solutions were determined, it was found that solution 1 retained 81% of its original activity; solution l9 Table 3.-—Loss of enzymatic activity of the solutions A and B of Figure l with time. Incubation Time Solution A Solution B 0 100 100 6 45 96 12 25 80 24 15 75 Figures are percent activity remaining after incubation. 20 2, 72%; and solution 3, 87%; the control retained only 14% of its original activity. It appears, therefore, that complexing agents in general stablilize the enzyme against attack by weak acids, and it may be inferred that bisulfite, since it exerts a similar stabilizing effect, is also a complexing agent for peroxidase. The fact that the bisulfite-peroxidase complex is active, whereas the other complexes are not, may be due to the rapid oxidation of the bisulfite upon addition of ; ‘ H202 for the assay. ' 1 Assuming the bisulfite—peroxidase complex is exactly as active as free peroxidase, its dissociation con- stant could be determined by the kinetic method of Mildvan and Leigh (1964). This method is applicable in cases where an enzyme is gradually inactivated by some inhibitory agent and a cofactor or other substance combines with the enzyme and retards the inactivation. The key equation is as follows: 1 ____ _1_ + [113] k k k K app d where Kd is the dissociation constant of the ligand-enzyme complex and [lig] is the molar concentration of the com- plexing agent. k and kapp are, respectively, the actual and apparent rate constants for the inactivation of the enzyme. kapp is defined by the equation: 21 [E0] _ kappt [I] log “EFT " 2.3 where [E0] is the total enzyme concentration and [E] is the active enzyme concentration after time t of incubation with an inhibitor present in concentration [I]. Kd is determined by plotting 1/kapp against [lig] and the intercept on the x axis has the value of -Kd. In this study, four HRPR solutions containing vary- ing amounts of bisulfite were deoxygenated by passing nitro— gen through them and incubated for twenty-four hours at pH 4.0 at room temperature. The activities of the solu- tions were determined both at zero time and after incubation. The solutions were then held for an additional twenty-four hours and their activities again determined. These data were used to determine a simplified ka , to be designated PP as k' , in the following manner: the value of Eo/E was app taken as the ratio of the activity at zero time to that after a given incubation time and log Eo/E was considered equal to k'app; this simplification is based on the fact that incubation time, t, and acid concentration, [I], do not enter into the final value of Kd' Figure 2 gives the plot of l/k'app vs [H803] for both the twenty-four hour and forty-eight hour periods. The intercept on the X axis indicates a value of about 0.02M as the dissociation constant of the bisulfite-HRPR complex. 22 .GOHDMQSOGH mo munch me one wm pm venommoa m63 mpfl>flpom oenmfiwuco one .HE v mo mESHo> HMDOD m ca .opflmasmfln mo mmfipflpcwsv mafiaum> cam o.v mm .Hommsn m mnuao Smo.o .mmmm mo 08 m.o pocfimuaoo musuxwfi :ofipommu comm .H_wmmH m> mam.x\a mo uon zmfloqlcm>paflzll.m ousmflm Hmbma: mo.o v0.0 mo.o No.0 Ho.o Ho.OI No.0I _ _ . a a _ PT mam 23 This determination was based on the assumption that the bisulfite ion did not inhibit the enzyme directly, e.g., by breaking disulfide bridges in the enzymic protein. How— ever, there is evidence of such an inhibition, since Chmiel— nicka (1963) reported that 0.48 mg of 802 in solution at pH 4.5 will inhibit the activity of one microgram of horse— radish peroxidase after preincubation for 10 hours. But for the purpose of our work, this loss was considered negligible. Catalase One milliliter of commercial beef liver catalase preparation was added to each of three 250—m1 volumetric flasks. The first flask was made to volume with demin- eralized water, the second with 0.1M Na2S04, and the third with 0.1M NaHSO3. No buffer was added, but the pH of each solution was about 5.0 because of dissolved carbon dioxide. The mixtures were allowed to stand at room tempera- ture for twelve days. The water suspension retained a constant degree of turbidity during that time, as judged by visual observation, while the sodium sulfate solution remained clear; apparently the catalase was "salted in." The NaHSO3 solution, however, developed an increasingly heavy turbidity and ultimately a white precipitate ap- peared on the bottom of the flask. On the twelfth day of preincubation, the enzymatic activities of the catalase solutions were determined. The 24 catalase treated with sulfate had 85.5% of the activity of the catalase in water suspension. This loss could be due to a splitting of the protohematin prosthetic group from the catalase protein by the acid (H2C03) in the solution; this process would be accelerated by the presence of sodium sulfate. The catalase treated with bisulfite had only 9.5% of the water-treated enzyme. The bisulfite obviously caused extensive denaturation of the catalase protein, and therefore no "protective effect" of bisulfite on catalase could be observed. Lipoxygenase The aerobic oxidation of linoleic acid by pea lipoxygenase was studied manometrically in the presence of zero and 1.6 x lO-4M bisulfite. The results are depicted in Figure 3. Other bisulfite concentrations studied were: 1.6 x 10'5M, 3.3 x 10'5M, 3.3 x 10—4M, and 1.6 x 10‘3M. For purposes of clarity, however, reaction curves corres— ponding to these concentrations were not included in Figure 3 because they would crowd the space around the two curves already illustrated. Two conclusions can be drawn from these results: (a) bisulfite slightly enhances the lipoxygenase reaction during the first 20—25 minutes of the oxidation; and (b) it inhibits the reaction after that point. It may be argued that the enhancement is only apparent, that the bisulfite 25 120 _ 100 p 80“ ul O2 absorbed 40 — 20 D I I l I I I 5 10 15 20 25 30 Time (Min.) Figure 3.——Effect of bisulfite on the oxidation of linoleic ‘ acid by lipoxygenase. Line A: No bisulfite in reaction mixture. _4 Line B: Reaction mixture contained 1.6 x 10 M NaHSO . Each reaction mixture contained 0.1% linoleic acid at pH 7.0, and 3% crude pea extract. 26 autooxidizes and contributes to the oxygen uptake of the system. However, control experiments show that very lit— tle of the bisulfite autooxidizes in thirty minutes under the conditions of the reaction. An alternative explanation is that the bisulfite acts as a scavenger, destroying free radicals and hydro- peroxides and relieving the enzyme of some reaction in- hibition. Ultimately, however, the bisulfite attacks the enzyme itself, presumably by breaking disulfide bonds in the enzymatic protein, and in the long run inhibits the oxidation reaction. Phenolase When our "low cresolase" enzyme was incubated with a quantity of tyrosine in a solution buffered at pH 6.5, there was a long period of time (exceeding an hour) during which the mixture remained colorless and after which a brown color slowly appeared. However, when a small quantity of catechol was included in the reaction mixture, the colorless lag period was reduced to only a few seconds and the browning reaction (melanin formation) proceeded at a greatly accelerated rate. This catalytic effect of o-diphenols (and other reducing agents, such as ascorbic acid) has been observed by several investigators, and is explained by Bright (1963) as a triggering effect of the reducing agent on the enzyme, in which the reducing “hm 27 agent furnishes the electrons necessary to initiate the conversion of monophenol to diphenol. If this were true, then adding catechol should shorten the lag period and enhance the monophenolase reaction until enough catechol was added to reduce the enzyme fully, after which no fur- ther increase in reaction rate would occur. The shortening of the lag period by o—diphenols has been noted by Mason (1956), and the increase in brown- ing rate caused by catechol is illustrated in Table 4, where it can be seen that the browning rate of a phenolase— 1 4 tyrosine mixture is strictly dependent on the amount of catechol initially present. It should be emphasized that the concentrations of catechol used in these reactions are at catalytic levels, and the brown products derived from the catechol are negligible in amount compared with those from tyrosine (as measured by absorbance at 470 nm). A difficulty with Bright's explanation arises from a simple mathematical calculation: the reaction mixtures in Table 4 each contained 0.05 mg tyrosinase, and assuming that the enzyme contained 0.2% copper (Kertesz and Zito, 1965), each mixture would contain 1.6 x 10—9 equivalents of copper. The lowest concentration of catechol used would then be about thirty times the minimum amount necessary to reduce all the enzymic copper, and consequently the reac— tion should proceed at its maximum rate. However, Table 4 shows that additions of larger amounts of catechol increase the rate of reaction. 28 Table 4.--Effect of catechol on the browning rate of phenolase-tyrosine reaction mixtures. Concentration of Catechol Browning Rate 1.66 x 10’5M 6 3.32 x 10‘5M 12 6.64 x 10'5M 22 9.96 x 10‘5M 31 3 Each reaction mixture contained 3.3 x 10- M tyro- sine, 0.05 mg of "low cresolase" enzyme, 0.16M phosphate buffer, ph 6.5, and varying quantities of catechol, in a total volume of 3 m1. Browning rates are expressed in arbitrary units. 29 A further difficulty arises with respect to bisul— fite. Since this compound is capable of reducing cupric ions to the cuprous state, it would also be expected to catalyze the cresolase reaction. However, previous work in this laboratory (Embs and Markakis, 1965; Markakis and Embs, 1966) has shown that while bisulfite in small quan— tities has little or no effect on the catecholase activity of phenolase, it profoundly inhibits cresolase activity as measured by 02 uptake. This is also true for the browning reaction as shown in Table 5. It is interesting to note that bisulfite influences the rate of browning even though it disappears before the brown products appear; i.e. it combines with the enzymati- cally produced quinones to form colorless addition products, and when the supply of bisulfite is exhausted, brown pro- ducts appear (Embs and Markakis, 1965). Of course it may be suggested that the bisulfite addition products them- selves inhibit the cresolase reaction, though to a lesser extent than does bisulfite. There is no evidence for this, in theory at least, since the addition products are prob— ably cyclic sulfates and sulfonic acids whose molecular structures resemble those of the smaller melanin molecules. Another possibility is that bisulfite forms a re- versible complex with the enzymic c0pper, as is the case with cyanide or cysteine, and somehow inhibits the cresolase activity without disturbing the catecholase. This 30 Table 5.--Effect of bisulfite on the browning rate of phenolase-tyrosine-catechol reaction mixtures. Concentration of Bisulfite Browning Rate 0 10 —5 1.66 x 10 M 6 2.49 x 10‘5M 4 3.32 x 10‘5M o Each reaction mixture had the same composition as those in Table 4, except that each contained 3.32 x 10‘5M catechol and varying amounts of bisulfite. Browning rate is expressed in arbitrary units. 31 hypothesis is suggested by the fact that the bisulfite in- hibition is a temporary one, that the cresolase reaction proceeds very slowly until the bisulfite is consumed by the quinones and is no longer an obstacle to the reaction. Some of the bisulfite is also probably oxidized by the di- sulfide bridges of the enzyme. This hypothesis was tested by determining the re- action rates of a phenolase-catechol—tyrosine-bisulfite mixture in the presence of varying amounts of cuprous ions. The results are shown in Table 6. If bisulfite forms a complex with copper, excess cuprous ions should relieve the enzyme of some of the inhibition. However, it can be seen that there was no significant increase in reaction rate resulting from the added copper. When excess pheno- lase was added instead of excess copper, the reaction in- creased somewhat, but fell short of what would be expected if bisulfite were simply a complexing agent. As a comparison, the same experiments were repeated with cysteine, a known copper binding agent. The results in Table 7 show that the addition of cuprous ions readily overcomes the cysteine inhibition. Table 8 shows that the addition of increasing amounts of catechol to a phenolase-tyrosine-bisulfite re- action mixture abolishes the inhibition by bisulfite. Therefore, it would seem, from all the foregoing evidence, that bisulfite does not react with the enzyme itself but 32 Table 6.——Effect of cuprous ions on the bisulfite inhibition of the browning rate of a phenolase-tyrosine- catechol reaction mixture. Concentration of Cu Browning Rate 0 9 1.66 x 10'5M 12 3.32 x 10'5M 12 4.98 x 10'5M 13 Each reaction mixture had the same enzyme, tyrosine, buffer, and catechol concentration as those in Table 5. Also, each mixture contained 3.3 x 10‘5M NaHSO3. The brown- ing rate without bisulfite was 22. 33 Table 7.--Effect of cuprous ions on the cysteine inhibition of the browning rate of a phenolase—tyrosine— catechol reaction mixture. Concentration of Cu Browning Rate 0 11 1.66 x 10‘5M 31 Each reaction mixture was the same as those in Table 6 except that each contained 6.7 x 10'5M cysteine instead of bisulfite. Apparently once the copper relieved the cysteine inhibition, the cysteine enhanced the brown- ing rate in its capacity as a reducing agent. 34 Table 8.—-Effect of catechol on the bisulfite inhibition of browning of a phenolase—tyrosine reaction mixture. Concentration of Catechol Browning Rate 3.33 x 10'5M o 6.7 x 10'5M 13 1.0 x 10'4M 22 1.3 x 10'4M 31 Each reaction mixture had the same composition as_5 those of Table 4 except that each also contained 3.3 x 10 M bisulfite and varying amounts of catechol. 35 interferes with some intermediate reaction crucial for the hydroxylation process, and that an o-diphenol (or some other reducing agent) is involved in this reaction. Also, this intermediate reaction is not a reduction of the cop- per by the reducing agent. The only known reaction to occur among the reac- tants: phenolase, catechol, and bisulfite, is an oxidation of the catechol to o—quinone followed by a reaction with bisulfite to form the colorless addition product or pro- ducts. Thus a very simple explanation of the bisulfite effect can be proposed if the nonenzymatic hypothesis for the cresolase action is accepted (see Review of the Litera— ture): The O-quinone is responsible for the hydroxylation, and the bisulfite interferes by converting it to the in- active addition product. This would explain why bisulfite and O-diphenol act in Opposite directions in the cresolase reaction. How- ever, a mechanism must be proposed to account for the pos— sibility of hydroxylation caused by quinone, in which molecular oxygen is inserted into the monophenol molecule. One such mechanism could be suggested by the work of Hamilton et_al. (1966), who found that a ferric ion- catechol mixture will catalyze the hydroxylation of aromatic compounds by peroxide. The active agent is thought to be: 36 in which the oxygen atom on the right hand side of the iron ion is derived from peroxide and is the hydroxylating agent. One could imagineaisimilar species composed of o—quinone, copper ions, and molecular oxygen performing the same func- tion with monophenols in a cresolase reaction. The quinone, forming on the enzyme molecule, would draw copper ions from the enzyme, complex with molecular oxygen, and attack a monophenol molecule which would be held in place by the en- zyme. The result could be a diphenol or quinone. Put in schematic form: 1. E-Cu + 02 >>E—Cu - O 2 2. E-Cu - 02 + 213—9E-Cu—2Q + 21120 3. E-Cu - 2Q + 02—91: + Q-Cu-02 + Q 4. Q-Cu — 02 + M ———-> Q-Cu + Q + H20 5. Q-Cu + Q >me1anin where E = enzyme, Q = quinone, D = diphenol, and M = monOphenol. The following observations lend indirect support to this hypothesis: 37 1. Reaction of phenolase with catechol causes partial inactivation of the enzyme, and this inactivation is believed to be a reaction between quinone and enzyme (Brooks and Dawson, 1965). 2. Phenolase loses copper during reaction with crdiphenols (Dressler and Dawson, 1960). 3. The oxidation of catechol in the presence of cupric ions results in a dark purple color rather than the usual yellow color of the melanins. The purple color does not form if the cupric ions are added after the oxidation has taken place. This indicates that copper forms a com- plex with some early product of the oxidation, probably a quinone. This hypothesis suggeststhat the hydroxylation could be carried out in the absence of enzyme. According— ly, a solution containing 2 x 10-4M catechol, 5 x 10—3M p-coumarate, 4 x 10-4 bisulfite, and 2 x 10-4M cuprous ions was allowed to stand at room temperature for 3-4 hours at pH 6.5. A control solution, containing the same ingre— dients but also a small amount of phenolase, was allowed to stand for the same length of time. Afterwards the two solutions were streaked on chromatographic paper which was then irrigated overnight and dried. Under ultraviolet light the area of the paper corresponding to the control solution exhibited a series of fluorescent bisulfite addi- tion products of characteristic sizes, colors, and 38 distribution. The pattern of fluorescent products closely resembled that derived from a solution of caffeate and bi— sulfite under oxidative conditions, since caffeate is the o-dihydroxy analog of the monophenol p-coumarate (Embs and Markakis, 1965). The test solution (without phenolase) also yielded addition products, but of a different pattern than those of the control. These products were fewer in number, rather uniformly blue in color, and considerably fainter in in- tensity. A repetition of the above experiment using a test solution without cuprous ions yielded the same pattern, but only after four days of standing. However, solutions containing any combination of ingredients other than those mentioned above (e.g. cuprous-p-coumarate-bisulfite, cuprous- catechol—p-coumarate, etc.) did not yield detectable fluor- escent products of any kind. The atypical products from the nonenzymatic reaction could have been derived from nonspecific hydroxylation of the p-coumarate, resulting in a mixture of di, tri, or even tetrahydroxy compounds, some of which may not be fluores— cent. If this is true, it would seem that the enzyme is necessary to insure that the cresolase reaction produces only 3,4 dihydroxy compounds from their 4-hydroxy parents. The enzyme could hold the monophenol molecule in position and labilize the proton ortho to the hydroxyl group for attack by the quinone-COpper-oxygen complex. 39 The fact that addition products were also obtained from the reaction without either copper or enzyme indicates a possible contamination by traces of copper ions, or even that the quinones, derived from autooxidation of catechol, alone are able to cause a hydroxylation reaction, as Ker— tesz asserts. Of course it may be argued that the atypical bands are actually catechol addition products which have absorbed some p-coumarate and became fluorescent. Since the compo— sition of the products is still unknown, this argument can- ’ i not be entirely discounted. However, it may be mentioned that when o-coumarate, a brilliantly fluorescent monophenol, is allowed to react with catechol and bisulfite under oxi- dative conditions, no fluorescent products can be separated from the solution by paper chromatography, with either enzyme or copper present in the reaction mixture. There- fore, if the catechol addition products do not derive fluorescent from crcoumarate, they probably would not from p—coumarate. Two problems remain to be discussed: the fact that nonphenolic reducing agents, such as ascorbic acid, also stimulate the cresolase reaction and overcome inhibition by bisulfite (Markakis and Embs, 1966); and also that cer- tain mushroom phenolase preparations are able to carry out the cresolase reaction very efficiently in the absence of any apparent reducing agent. 40 Kertesz and Zito (1962) suggested that a phenolase- ascorbate combination could carry out the cresolase reaction with the mechanism proposed by Udenfriend et_al. (1954), in which an ascorbate - H202 reaction product hydroxylates aromatic compounds, with the help of ferric ions. This would be analogous to the quinone-copper—oxygen mechanism mentioned above, except that phenolase does not oxidize ascorbic acid. It may be argued that H 0 forms sponta- 2 2 neously from the autooxidation of ascorbic acid, and that it rids the system of bisulfite and promotes hydroxylation of monophenols. However, our previous work (1966) has shown that fluorescent addition products form in a pheno- lase-p-coumarate-bisulfite—ascorbate reaction mixture, indicating that the bisulfite was not oxidized in the re- action by peroxide. Obviously the effect of ascorbic acid on the cresolase reaction is complex and needs further study. Highly purified mushroom phenolase preparations will catalyze the rapid oxidation of tyrosine and other monophenols without the addition of reducing agent. These "high cresolase" preparations are sensitive to inhibition by bisulfite, but unlike the low cresolase type, addition of larger concentrations of enzyme readily overcomes the inhibition (Table 9). Therefore, it appears that the high cresolase enzyme is equivalent to a mixture of low creso— lase enzyme and a reducing agent, and it may be suggested 41 Table 9.--Effect of increasing enzyme concentration on the bisulfite inhibition of browning by phenolase— tyrosine-reducing agent reaction mixtures. A. Low cresolase enzyme Amount of Enzyme in Mixture Browning Rate 0.05 mg 0 0.10 1.7 0.15 0 0 20 2 0.25 3 Each reaction mixture had the same composition as those in Table 4 except that each had 3.3 x 10’5M bisulfite, 3.3 x 10‘5M catechol, and varying amounts of enzyme. B. High cresolase enzyme Amount of Enzyme in Mixture Browning Rate 0.05 mg 2.5 0.10 6 0.15 9 0.20 14 0.25 18 Each reaction mixture contained: 1.67 x 10—3M tyrosine, 0.16M phosphate buffer, H 6.5, 3.3 x 10-6M dihydroxyphenylalanine, 3.13 x 10' M bisulfite, and vary- ing amounts of phenolase in a total volume of 3 ml. 42 that the mushrooms phenolase molecule contains a residue which acts as a reducing agent, possibly dihydroxyphenyl- alanine (dopa) or even dopa-quinone, if quinones are neces- sary for hydroxylation. In terms of the cresolase mechanism mentioned above it may be suggested that high cresolase enzymes con- tain quinone residues which hold copper ions and oxygen molecules, and carry out the cresolase reaction. If this is true, the enzyme should not lose its c0pper as it does during the catecholase reaction, and this indeed was found to be the case by Dressler and Dawson (1960) with their high cresolase preparations. The fact that a high creso- lase preparation can be converted to a low cresolase state by partial denaturation (e.g. by heating or prolonged con— tact with bisulfite) may be explained as a disruption of the spatial relationship between the quinone—copper-oxygen complex, and the binding site of the monophenol molecule on the enzyme. In such a case, a trace of added o-diphenol, oxidized to o-quinone, would substitute for the enzyme-bound quinone. The foregoing discussion is admittedly largely speculative. Further lines of research into this problem may be suggested: (1) a better elucidation of the molecu- lar structure of phenolase to reveal the existence of a quinone residue; (2) an investigation of the possibility of a quinone-copper complex as a hydroxylating agent; 43 (3) an investigation of the effect of bisulfite on the iron— ascorbate hydroxylating system; and (4) a repetition of the work of Dressler and Dawson using low cresolase enzyme, a monophenol, and a trace of o—diphenol, to see if the enzyme loses its copper. If the above mechanism is correct, it would; and if Dressler and Dawson‘s two—site theory is cor- rect, it would not. Ascorbic Acid Oxidase When 0.1 ml amounts of cucumber peel juice, which contains ascorbic acid oxidase, wereincubated with solu- tions containing 4.8 x 10_3M ascorbic acid and varying amounts of bisulfite at pH 5.7 (volume of each solution was 3.0 ml), a loss of enzymatic activity in direct pro— portion to the amount of bisulfite present was observed (Figure 4). However, in contrast to its effect on the cresolase function of phenolase, the bisulfite did not cause a lag period followed by increased activity, but merely a general reduction in the rate of ascorbic acid oxidation. This indicates a direct inhibition of the enzyme by the bisulfite. To substantiate this conclusion, 0.1 ml of cucumber juice was preincubated with 0.1 ml of 10_2M NaHSO3 solution for 30 minutes and then its ascorbic acid oxidase activity was determined in the same manner as in Figure 4. The results, shown in Figure 5, indicate that the bisulfite caused an extensive loss of enzymatic activity. H2 02 absorbed 60 50 40 30 20 10 44 no HSO- 3 _. -4 _ 3X10 M HSO3 _ -4 _ 7x10 M H803 -3 - _ 10 M HSO3 l I l I l l 5 10 15 20 25 30 Time (Min.) Figure 4.-—Effect of varying concentrations of bisulfite on the rate of ascorbate oxidation by ascorbate oxi— dase in 0.1 m1 cucumber peel juice. Each reaction mixture contained 14.4 micromoles ascorbate at pH 5.7. pt 02 absorbed 70 60 50 4O 30 20 10 45 Time (Min.) f no HSO3 -4 - 3x10 M HSO3 r— I J H_ I I 5 10 15 20 25 30 4 Figure 5.-—Effect of 30 min. preincubation with 1.5 x 10_ M bisulfite on the ascorbate oxidase activity of 0.1 m1 cucumber peel juice. For composition of reaction mixtures, see 4. Figure 46 This loss of activity is quite similar to the gradual inhibition of phenolase by bisulfite (Embs and Markakis, 1965). In both cases the cause is apparently a denaturation of the enzymic protein by breakage of di— sulfide linkages. ADDENDUM: The Effect of Bilsulfite on the Growth of Saccharomyces cerevisiae Review of Literature Sulfur dioxide solutions are widely used in the food industry to control the growth of yeasts, molds, and bacteria. Their antiseptic potency is strongly pH depend- ent, in that they are powerfully inhibitive to microbial growth under acidic conditions and ineffective at neutrality. This could possibly be explained by the assumption that the nonionic form of the compound (H2803), which exists at lower pH levels, more rapidly penetrates the cell wall of the mi- 3, quently greater amounts can enter the cell and interfer with croorganism than the ionic forms (HSO 50:) do, and conse- its life processes (Bosund, 1962). Rehm and Wittmann (1963) studied the effect of 802 solutions on yeast growth at various pH levels, and con— cluded that the bisulfite ion, as well as the undissociated H2803, had an appreciable antiseptic effect on yeast. How- ever, their conclusions are dubious because they used an obsolete pK2 value for sulfurous acid in calculating the variation of bisulfite concentration with pH. In the present work attempts were made to recheck the work of Rehm and Wittmann using the proper pK and 2! 47 48 also to determine the accuracy with which Bosund's theory describes the mechanism of inhibition of yeast growth by 802° Methods and Materials All yeast used in these experiments was Saccharo- myces cerevisiae taken from yeast cake, Red Star brand. No attempt was made to use an individual variety of this yeast. The yeast was allowed to reach an actively grow- ing stage in growth media before use. All growth media used was distilled water contain— ing 0.67% yeast nitrogen base, 2% sucrose, and 0.05M citrate- phosphate buffer. The pH was adjusted with 5N NaOH using a pH meter. Fifty ml portions of the media were used in 125 ml Erlenmeyer flasks plugged with cotton. Steriliza- tions were done with a steam autoclave, and after inocula— tion each flask was allowed to rock gently on a mechanical shaker overnight. Growth assays were determined either by plating with a yeast nitrogen base agar mixture or by determining the turbidities of the growth media 24 hours after inocu- lation, using a Beckman DU Spectrophotometer. Turbidities were expressed as per cent transmittance at 600 nm. 35 Radioactive 802 was prepared by boiling H2 504, with copper, trapping the radioactive SO in 0.1N NaOH, 2 49 then distilling it after acidification with H 804, and re— 2 ceiving it in cold 0.1N NaOH. 802 uptake studies were performed by adding 0.35 -0.4g of actively growing yeast cells to 30 ml of 0.01M citrate—phosphate buffer containing radioactive 802. At measured time intervals, 4 m1 aliquots of the well mixed liquid were withdrawn and added to 25 m1 polyethylene scintillation bottles. The bottles were immediately placed in a Servall high speed centrifuge and the yeast precipi- tated by centrifugation. The supernatant liquid was poured off and the yeast washed three times with distilled water, centrifuging between washings. The yeast cells were then suspended with Cab-o—sil in a toluene based scintillation medium (6 g PPO and 0.2g POPOP in one liter toluene) and their radioactivity counted with a Packard Tri-Carb Liquid Scintillation Spectrometer. In the growth studies, aseptic techniques were em- ployed throughout. At measured time intervals 0.1 ml ali- quots were withdrawn from the buffered 35802 solutions and added to media for overnight growth. In the experiment involving 600 PPM 35802, each scintillation bottle received 5 ml of 0.03% H202 before the radioactive aliquots were added. The purpose of the H202 was to oxidize the 802 to 80:, which would be less readily absorbed by the yeast. 50 Results and Discussion Figure 6 shows the distribution of the three spe- cies, H2803, H803, and 80:, over the pH range 0—8. It can be seen that the nonionic form, H2803, predominates at the lower end of the pH scale, where 802 solutions have their greatest antiseptic potency. According to the theory of Bosund the H2803 form is antiseptic because, being non- charged, it is the only one of the three species which can penetrate the lipo-protein barrier in the cell wall. However, Rehm and Wittmann (1963) reported that bisulfite (H803), though charged, also had antiseptic properties. Therefore, the first step in this study was to check the validity of this report by investigating the effect of SO2 solutions at various pH levels on the growth of yeast cells. Equal numbers of yeast cells were innoculated into flasks of growth media which had been adjusted to pH 3, 4, 5, 6, and 7, and containing 50 to 400 ppm 802. Saccharo- myces cerevisiae was used because it grows throughout the pH range 3-7, though not equally well at every pH (Figure 7). The following observations were made: at pH 3 and 4, no yeast growth was evident even if the flasks were allowed to stand for several days with 50 ppm SO at pH 6 and 7 27 the cells grew as well as the controls when observed 18 hours after innoculation at 400 ppm 802; at pH 5 the cells grew after undergoing a lag period the length of which 51 Percent 100 50‘ I l I I J I I 0 1 2 3 4 5 6 7 8 pH Figure 6.-—Variation of the ionic species, H2803, H803, and SOS, with pH. 52 ) «>100‘ Number of cells pet M1 (x10 (180,000) I 3 4 5 6 pH Figure 7.--Growth of yeast in the pH range 3-7. (Media buffered at pH's 3, 4, 5, 6, and 7 were inocu- lated with S: Cerevisiae. After overnight growth, cell count was determined by plating.) 0 53 was proportional to the 802 concentration. At 50 ppm 802 the lag was a few hours in length; at 100 ppm, it was about 40 hours long; at 200 ppm, 60—70 hours; at 300 ppm about 4 days; and at 400 ppm about 5 days. Once the yeast recovered from the lag period, its growth was as full as that of the controls. On the basis of these results, the following con— clusions can be drawn: 1. The bisulfite ion has very little antiseptic potency. At pH 6, 88% of the total 80 is in the bisul— 2 fite form, and since even 400 ppm 802 will not prevent yeast growth at that pH, it means that yeast can grow without difficulty in a solution containing 0.044% H803. 2. The sulfurous acid molecule is an extremely potent antiseptic. At pH 4 the yeast cells would not grow at 50 ppm 80 , which at that pH constitutes a solution of 2 0.0062% H80; and 0.000044% H2803. Since at 50 ppm so2 the solution does not contain enough bisulfite to affect yeast growth, all its antiseptic potency must be derived from the sulfurous acid molecule. 3. A comparison of the figures, 0.000044% H280 and 3 0.044% H80; indicates that H2803 is well over one thousand 3O antiseptic potency at all. Therefore, the report of Rehm times more potent than H80 The latter may even have no and Wittmann is in error, probably because of their use of an obsolete pK2 value in calculating the bisulfite content of their systems. 54 If the sulfurous acid molecule is the only form that actually enters the cell, then a direct relationship should exist between the amount of sulfur absorbed by the cell and the antiseptic potency of the 802 solution at a given pH. This possibility was investigated by exposing yeast cells to labeled sulfur dioxide solutions (35802) at various pH's, times, and 802 concentrations in the follow- ing manner: yeast cells were grown in a suitable medium overnight, then collected and washed asceptically and added 35 to buffered sterile solutions of 80 At periodic time 2. intervals aliquots were taken and added to fresh media for growth and simultaneously other aliquots were taken for radioactive assay. This was done over the pH range 3—7; at time intervals of less than 5 minutes, 2, 4, and 8 hours; and at concentrations of 200 and 480 ppm 802. The results are shown in Table 10 and 11, and in Figures 8 and 9. It can be seen that uptake of labeled sulfur occurred at every pH, but that the uptake was more extensive at lower pH's. Tables 10 and 11 showed that the yeast cells died rapidly when exposed to 80 at pH 3; 2 at pH 4 and above they withstood the full 8 hours exposure with apparent ill effect. In spite of this, less 802 was absorbed at pH 3 than at pH 4 or 5. This indicates that the antiseptic power of an 80 solution does not depend on 2 the total amount of 802 absorbed by the microorganism but perhaps on the rate of absorbance. It may be suggested 55 Table 10.——Growth of yeast cells after exposure to 200 ppm 802 at pH's 3, 4, 5, 6, and 7. pH of Exposure Time of Exposure (Hr.) 3 4 5 6 7 0 3 3 2 2 2 2 22 3 2 2 2 4 86 5 3 2 2 8 85 3 2 2 2 The figures represent the transmittance values of the growth media at 600 nm. Figures below 10 represent substantial growth and figures above 80 represent no growth in 24 hours after inoculation. 56 Table ll.—-Growth of yeast cells after exposure to 480 ppm 802 at pH's 3, 4, 5, 6, and 7. pH of Exposure Time of Exposure (Hr.) 3 4 5 6 7 0 2 2 l l 1 2 77 2 l l l 4 74 2 1 l l 8 78 3 l l l These figures represent transmittance values of growth media, at in Table 4. 57 N .e one .6 .m .e .m m.me um Om 2mm com OD pomomxo mHHoo pmmmm ma Homasm mo oxmumDII.m madman A.Hmv oEHB omsmomxm o in co HIV. N I o.e mm .IIIIIIIII 0.6 mm 0.... mm HSMHSm Hopou mo pcmo.uom 58 .e pee .p .m .e .m m.me um Now 2mm owe 0p pmmomxo maamo pmmom ma HDMHSm mo oMMDmDII.m ousmflm A.Hmv mEHB musmomxm m 6 OK me e w 06 mm \III‘IIII-lll‘ Com mm x.) v L e.- o.m mm o.v mm HSMHDm HM#O# Mo Dame Mom .Io.m 59 that pH 3 the SO floods the cell so rapidly that its vital 2 functions are paralyzed in the first few minutes of expo— sure, after which it absorbs no more 802. At pH 4 and 5 the yeast apparently can absorb and store large quantities of 802, and even expel some back into solution after hold— ing it for a period of time. In an attempt to substantiate these points, the experiments described above were repeated with the exception that aliquots were taken only for radioactive assay, and none for growth studies. This meant that aseptic procedures were not needed and consequently aliquots could be taken much more rapidly than before, thus making possible a greatly reduced time scale. The results were given in Figure 10. Here it can be easily seen that at pH 3 the yeast almost instantly reaches its maximum load of 802 while at pH's 4 and 5 the build—up is gradual. To summarize: the antiseptic potency of an 802 solution toward yeast depends on the rate of uptake of 802 by the cell and not on the total amount absorbed. A more gradual rate of uptake means a longer survival time for the cell. Lower pH's cause higher uptake rates probably because of the greater proportions of H2803 molecules, which can easily penetrate the cell wall. 60 m .m use .4 .m m_me pm Om 2mm oom ou pomomxo maaoo ummoh >9 MSMHSm mo oxmumDII.oa ousmflm A.cflzv oEflB madmomxm ow om ma m o _ 1 _ _ mSMHSm HMDOD mo ucoo mom REFERENCES Bosund, J. 1962. The action of benzoic and salicylic acids on the metabolism of microorganisms. Adv. Food Research 11, 331. Bright, H. J., B. J. B. Wood, and L. L. Ingraham. 1963. Copper, tyrosinase, and the kinetic stabil— ity of oxygen. Ann. N. Y. Acad. Sci. 100, 965. Brooks, D. W., and C.R. Dawson. 1965. Aspects of tyrosinase chemistry. From "The Biochemistry of COpper," J. Peisach, P. Aisen, and W. E. Blumberg, ed. Academic Press, New York, 1966, p. 343. Chmielnicka, J. 1963. Influence of sulfurous acid on L—ascorbic acid oxidation by peroxidase in the presence of H202. Roczniki Zakladu Hig. 14, 393. Chmielnicka, J. 1964. Effects of sulfur dioxide on the ascorbic acid content and on the activity of peroxidase in strained pulp from tomatoes. Rocz- niki Zakladu Hig. 15, 13. Dawson, C. R. 1965. Ascorbic oxidase. From "The Biochemistry of Copper," J. Peisach, P. Aisen, W. E. Blumberg, ed. Academic Press, New York, 1966, p. 305. Dressler, H. and C. R. Dawson. 1960. On the nature and mode of action of the copper protein, tyrosinase. II. Exchange experiments with radioactive copper and the functioning enzyme. Biochim. Biophys. Acta. 45, 515. Embs, R. J. and P. Markakis. 1965. The mechanism of sulfite inhibition of browning caused by poly- phenol oxidase. J. Food Sci. 30, 753. Eriksson, C. E. 1967. Pea lipoxidase, distribution of enzyme and substrate in green peas. J. Food Sci. 32, 438. 61 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 62 Keilin, D. and E. F. Hartree. 1951. Purification of horse— radish peroxidase and comparison of its prOperties with those of catalase and methemoglobin. Biochem. J. 49, 88. Kertesz, D. and R. Zito. 1962. Kinetic studies of the polyphenoloxidase action: kinetics in the presence of reducing agents. Biochem. Biophys. Acta 84, 153. Kertesz, D. and R. Zito. 1965. Mushroom polyphenol oxidase. I. Purification and general prOperties. Biochim. Biophys. Acta 88, 447. Klenbanoff, S. J. 1961. Sulfite-activated oxidation of reduced pyridine nucleotides by peroxidase. Biochim. BiOphys. Acta 48, 93. Kubowitz, F. 1938. Resynthesis of phenol oxidase from protein and c0pper. Biochem. J. 296, 443. Lewis, U. J. 1954. Acid cleavage of heme proteins. J. Biol. Chem. 206, 109. Macrae, A. R. and R. G. Duggleby. 1968. Substrate and inhibitors of potato tuber phenolase. Phyto- chem. Z, 855. Maehly, A. E. 1952. Splitting of horseradish peroxi- dase into prosthetic group and protein as a means of studying the linkages between hemin and protein. Biochim. Biophys. Acta 8, l. Mahler, H. R. and E. H. Cordes. 1966. "Biological Chemistry," Harper and Row, Inc., New York, p. 590. Markakis, P. and R. J. Embs. 1966. Effect of sul— fite and ascorbic acid on mushroom phenol oxidase. J. Food Sci. 84, 807. Mason, H. S. 1956. Structures and functions of the phenolase complex. Nature 177, 79. Mason, H. 8., W. L. Fowlks, and E. J. Peterson. 1955. Oxygen transfer and electron transport by the phenolase complex. J. Amer. Chem. Soc. 11, 2914. Mildvan, A. S. and R. A. Leigh. 1964. Determination of cofactor dissociation constants from the kine- tics of inhibition of enzymes. Biochim. Biophys. Acta 88, 393. 23. 24. 25. 26. 27. 28. 63 Mitsuda, H., K. Yasumoto, and A. Yamamoto. 1967. In- activation of lipoxygenase by H 02, cysteine, and some other reagents. Agri. BioI. Chem. 84, 853. Monikowski, R. and E. Chmielnicka. 1964. Le role de l'acide sulfureux dans le processus d'oxydation peroxydasique de l'acide l'ascorbique en presence d'eau oxygenee. Rev. Ferment Ind. Aliment. 48, 113. Muneta, P. 1966. Bisulfite inhibition of enzymic blackening caused by tyrosine oxydation. Amer. Potato J. 43, 397. Rehm, H. J. and H. Wittmann. 1963. Antimicrobial action of sulfurous acid. II. Action of disso- ciated and undissociated acid on various micro— organisms. Z. fiLebensmittel- 'Untersuchen- Forsch. £88, 465. Surrey, K. 1964. Spectrophotometric method for de- termination of lipoxidase activity. Plant Physiol. 39, 65. Undenfriend, 8., C. T. Clark, J. Axelrod, and B. B. Brodie. 1954. Ascorbic acid in aromatic hydroxy— lation. J. Biol. Chem. 208, 731. IIIIHIIIIII 611