PART 1 couvansion or LACTATE-CM AND mucosa .cH To PROPIONATE BY THE RUMEN MICROBIOTA PART 11 LACTATE Mausousm av Pmosmnococcus ELSDENH Thesis for ”1.2 Degree of ph. D. MICHIGAN STATE UNIVERSITY Ransom Leland Baldwin, Jr. 1962 This is to certify that the thesis entitled LACTATE METABOLISM Part I - Conversion 0f Lactate-Cl4 And Glucose-Cl4 To Propionate By The Rumen Microbiota Part II - Lactate Metabolism by Peptostreptococcus El d ii JEL— presented by Ransom Leland Baldwin, Jr. has been accepted towards fulfillment of the requirements for Ph.D degree in Biochemistry and Dairy ./) " /, fl -—~/~ , I {If ~ g ‘, _x / ' ' ,\ ’1 I“ /|, A "‘~‘; ‘.. L/ , ‘ -' ,7 " ' //I / /V / ”A //‘( k (I " / ' /‘ ,’,-/ . // I /" f ’L- —.—-— -/ 15/491"— ’ [/6 o L.“ ' ’.-4_ l‘r Major professor v / ,1 " “If A I / 1"- "’ Date ‘. gt 171/“ % A/P/s .1 . ,/ I"/ 0-169 LIBRARY Michigan State University ABSTRACT LACTATE METABOLISM PART 1 .. CONVERSION or LACTATE—c“ AND GLUCOSE-C” TO PROPIONATE BY THE RUMEN MICROBIOTA PART II -- LACTATE METABOLISM BY PEPTOSTREPTOCOCCUS ELSDENII by Ransom Leland Baldwin Jr. The problem outlined for investigation was divided into two parts. The first part involved an investigation of the relative contributions of the two alternate pathways (via symmetrical and non-symmetrical intermediates) for the conversion of lactate to propionate in a natural biological system. The second part was concerned with the isolation and characterization of the enzymes responsible for the conversion of lactate to propionate via the acrylate or the so-called non- randomizing pathway. The biological system chosen for the first phase of the investigation was bovine rumen fluid. Samples removed from cows fed a variety of diets were incubated in vitro with lactate-Z-C”, lactate-3-C”, glucose- l-CM, glucose-Z-C” and glucose-643”. The acetic and propionic acids were isolated and degraded and the specific activities of the individual carbon atoms determined. The results obtained in these experiments clearly indicated that (l) the acrylate pathway makes a significant contri- bution, (2) the contribution of the acrylate pathway increases with the starch content of the diet, and (3) the succinate (randomizing) pathway is the most important since it performs the major role during hexose utilization. Ransom Leland Baldwin Jr. Considerable progress has been made on the problem of identification, isolation and characterization of the enzymes of the ac rylate pathway: (1) acetyl—PO4 + CoASH Phosphotransacetylase acetyl-CoA + Pi ‘ CoA transphorase (2) acetyl-COA + lactate lactyl-GOA + acetate “ (3) lactyl-GOA iaCtYlCoA dehydrase acrylyl COA + HZO (4) acrylyl-GOA + 2e- aCYl'COA-dehydrogenase propionyl-COA \ CoA transphorase (5) propionyl-COA + acetate acetyl-COA + propionate \ Lactyl CoA dehydrase (3), the key enzyme of the pathway, has been identified in cell-free extracts of the rumen organism Peptostreptococcus elsdenii and an assay system developed for measuring its activity. Phosphotransacetylase (l) and CoA transphorase (2., 5) have been purified from extracts of the same organism. Acyl CoA dehydrogenase (4) has been purified 25-fold and partially characterized. The enzyme is freely reversible, has a pH optimum of 7. 5 and prefers four carbon to three carbon substrates. Several other enzymes observed in extracts of the organism are: a D lactic dehydrogenase (no pyridine nucleotide require- ment), a very stable phosphoroclastic system, and a diaphorase. The most important result of this work has been the demonstration of lactyl CoA dehydrase. This establishes the fact that the pathway pro- ceeds as postulated. LACTATE METABOLISM PART 1 -- CONVERSION OF LAOTATE—c“ AND GLUCOSE-c14 TO PROPIONATE BY THE RUMEN MICROBIOTA PART II -- LACTATE METABOLISM BY PEPTOSTREPTOCOCCUS ELSDENII BY Ransom Leland Baldwin Jr. A THESIS Submitted to Michigan State University in partial fulfillment Of the requirements for the degree of DOC TOR OF PHILOSOPHY Departments of Biochemistry and Dairy 1962 ACKNOWLEDGMENTS The author wishes to express his sincere appreciation to Drs. R. S.. Emery and W. A. Wood for their guidance and encouragenaentthroughoutthe course OfthlS‘NOIkm 'The support Of a National Science Foundation cooperative fellowship and an Atomic Energy Commission grant are gratefully acknowledged. ************ ii VITA Ransom Leland Baldwin Jr. was born in Meriden, Connecticut on September 21, 1935. He graduated from Me riden High School in June, 1953 and received the B. S. degree in agriculture from the University of Connecticut in June, 1957. He accepted a graduate research assi stantship in the Dairy department at Michigan State University and received the M. S. degree from that department in 1958 with a minor in physiology. At that time he arranged to do further graduate work jointly between the Departments of Dairy and Biochemistry with a minor in Bacteriology. The requirements for the Ph. D. degree will be completed in the fall Of 1962. During the tenure of his Ph. D. work he was supported by a National Science Foundation cooperative fellowship. Mr. Baldwin is married and has one child. He has accepted an appointment with the Animal Husbandry Department at the University Of California at Davis. iii TO MARY ELLEN iv TABLE OF CONTENTS Page INTRODUCTION ......................... 1 LITERATURE REVIEW ..................... 2 Organic Acid Formation in the Rumen .......... 3 Effect Of Diet Upon Volatile Fatty Acid Production 3 Importance Of Lactate ................ 4 Propionic Acid Metabolism ................ 5 The Dicarboxylic Acid Pathway. . ......... 5 Isotopic . ................... '7 Enzymatic ................... 8 The Direct Reductive Pathway ........... ll Oxidative Pathways ................. 13 Associated Enzyme Systems ......... 15 PART 1 «CONVERSION OF LACTATE-c“ AND GLUCOSE-c“ TO PROPIONATE BY THE RUMEN MICROBIOTA. . . . 16 METHODS AND MATERIALS . . .. . . ........... 17 RESULTS ......................... 20 DISCUSSION ........ . .............. 35 SUMMARY ......................... 39 PART II -- LACTATE METABOLISM BY PEPTOSTREPTO- COCCUS ELSDENII .................... 4O METHODS AND MATERIALS ............... 41 RESULTS ......................... 51 DISCUSSION ........................ 70 SUMMARY ......................... 73 LITERATURE CITED ...................... 74 TABLE 1. 2. 10. ll. 12. 13. 14. 15. 16. 17. LIST OF TA BLES Page Diets Of cows used as sources of rumen fluid ..... 21 Ratios Of acetate to propionate in rumen fluid . . 21 . Effect of lactate concentration upon the ratios of acetate tO prOpionate formed from lactate-C”. . . . ..... . Distribution Of label in acetate from lactate-Z--Cl4 and 1actate-3-C“ .................. . Distribution Of label in propionate from lactate-Z-C”. . Distribution of label in propionate from lactate-3-C“. . Ratios Of acetate to propionate in rumen fluid . . . . . . Distribution of label in acetate from glucose-LC“, glucose-Z-Cl‘ and glucose-€r-Cl4 .......... . Distribution of label in propionate from glucose-l-CM, glucose-Z-C14 and glucose-b-C“. . . . ........ Specific activities Of several enzymes present in rumen fluid extracts ...................... Phosphotransac etylase purification . . . ....... Coenzyme A transphorase purification. . ....... Lactic dehydrogenase purification . ........ Acyl coenzyme A dehydrogenase purification ...... Diaphorase purification .......... Specificity of COA transphorase . . ..... Dye reduction of several dehydrogenase systems . . . . vi 22 26 27 28 30 3O 31 32 45 47 47 49 49 66 67 FIGURE 1. Dicarboxylic acid pathway to propionate . . . ..... 2. The direct reductive pathway . . . . . . . ....... 3. Oxidative pathways . . . . . . ..... . ....... 4. Lactate concentration vs. rate of lactate utilization . . . 5. Sample preparation and degradation ......... 6. Conversions Of propionyl-Cu-pantetheine with cell free extracts of rumen fluid . . . . . . . .......... 7. Routes Of conversion to propionate in rumen fluid . . 8. Spectra Of thiol esters . . . . . . ...... 9. Lactyl COA utilization: chromatography of hydroxa- mates ........... ....... 10. Lactyl CoA dehydrase: requirements of the assay system...... ........ .. ........ 11. Lactyl CoA dehydrase assay: rate vs. enzyme concen- tration . . ........................ 12. Chromatography Of reaction product dinitrophenyl- hydrazone. ...... l3. Enzymatic determination of acrylyl COA formation . . . l4. Acyl COA dehydrogenase: reduction of acrylyl CoA . . . 15. Lactyl COA dehydrase assay system . . .1 ..... LIST OF FIGURES vii 24 25 33 37 42 53 56 .58 60 63 69 70 INTRODUCTION The rumen fermentation and the symbiotic relationships between the organisms concerned have been the subject of many investigations in the past decade. Although significant strides have been made, the nutritionist is still unable to solve many of the problems whose solutions were foreseen when the initial studies were undertaken. One of these is the problem Of increasing ruminant efficiency. In this particular case, however, sufficient knowledge is available to point the way for further studies. The principle aim Of the investigations reported in this thesis was to contribute knowledge about several reactions which occur in the rumen and are related to efficient carbohydrate utilization by the rumi- nant. In order to carry out these investigations it was necessary to utilize methods and knowledge contributed by biochemists, bacteriologists and nutritionists. The problem outlined for investigation was divided into two parts. The first part involved an investigation of the relative contributions of the two alternate pathways (via intermediates behaving in a symmetrical and nonsymmetrical fashion with respect to enzyme action) for the con- version Of lactate to propionate in bovine rumen fluid. The second part is concerned with the identification, isolation and characterization of the enzymes responsible for the conversion Of lactate to propionate via the acrylate, or the so-called non- randomizing pathway. LITERATURE REVIEW ORGANIC ACID FORMATION IN THE RUMEN Carbohydrate utilization by the ruminant has been studied ex- tensively. Several reviews are available which summarize the general nutritional characteristics of the ruminant, the contributions of the rumen fermentation to the host animal, and the characteristics Of the rumen microflora (1-10). The purpose of this review is to report some of the developments which are directly related to the data presented in this the sis . . Effect of Diet Upon Volatile Fatty Acid Production The effect of diet upon the relative proportions of acetate and propionate formedin the rumen has received considerable attention because of the effect these proportions have'upon the efficiency of the ruminant (11, 12) and upon the composition of milk (13). Results tabu- lated by Annison and Lewis (7) show that the molar proportion of acetate to propionate varies from 1 to 4. These data also reveal that there is a high correlation between carbohydrate availability of the ration and the amount of propionate formed. Diets consisting of high proportions of complex (or less available) carbohydrates yield very little propionate while diets containing large amounts of readily available carbohydrates yield higher levels of propionate. . For example, Gray and) co-workers (14) observed acetatezpropionate ratios as high as 4 in the rumen con- tents of cows fed Wheaten hay, while Phillipson (15) reported a ratio Of 1 on a diet of flaked maize. Balch and Rowland (l6) conducted an intensive study of the diurnal variations in volatile fatty acid (VFA) production,‘ lactate concentration, and pH in the rumens of cows receiving a variety of diets. They found that as the carbohydrate availability of 3 the diet increased the pH decreased and the concentrations of the organic acids, especially propionate and lactate, increased. They concluded on p the basis Of these data that the highly acidic conditions existing in the rumen Of cows fed high starch diets encourage the proliferation of organisms which produce lower amounts of acetic acid. Recent advances in the development Of bacteriological techniques suitable for culturing a large number Of the rumen microorganisms should enable investigators to correlate changes in the microbial population with changes in diet (9,10,17,18). Importance of Lactate The importance of lactate as a precursor of the volatile fatty acids is not clear. The levels of lactate present in rumen fluid, although usually less than 1. 0 mg per cent, are highly correlated with the starch content of the diet and the amounts of propionate present (19, 20). Lactate concentrations as high as 0.270 per cent have been noted in cows fed diets high in flaked maize (16), but only for short intervals immediately after feeding. It has long been known that lactate is an inter- mediate in carbohydrate dissimilation in the rumen and is quickly degraded to acetate, propionate, butyrate, Hz and CO; (21). . Several groups have reported, on the basis of quantitative estimates of the volatile acids formed during lactate fermentation, that lactate is converted mainly to propionate (15, 22, 23). Heuter e1: a_,_l.. (24) fed Sodium lactate and Calcium lactate to cows consuming normal rations and found that lactate stimu- lated the formation of propionate and butyrate. They also reported that considerable amounts of lactate are absorbed through the rumen wall. Jayasuriya and Hungate (25) studied the fermentation of lactate-Z—C“ by rumen inocula removed from cows fed several diets. They Observed that only small amounts Of propionate -C14 were formed (less than 20% of VFA—CM) while large amounts Of acetate-C14 were formed (70 to 90% Of the VFA-C“). No apparent correlation existed between carbohydrate availability in the diet and the fate Of lactate. Bruno and Moore (26) reported similar results. It would seem that there is a discrepancy between conclusions based upon quantitative data and those based upon isotopic data. An explanation for these discrepancies will not be forth- coming until more knowledge of the kinetics Of carbohydrate fermentation and of the role Of lactate and other three-carbon intermediates is available. PROPIONIC ACID METABOLISM Studies carried out upon animal, plant and bacterial systems indi- cate that there are at least three distinctly different pathways for the formation and utilization of propionate. Many review articles concerned with various aspects Of the problem are available and they have been cited along with the original articles as sources of more detailed information. The three pathways, the dicarboxylic acid pathway, the direct reductive pathway, and the oxidative pathway(s) are treated separately, below. The Dicarboxylic Acid Pathway Studies on the mechanism of propionate formation in bacteria and propionate utilization in animals have revealed the existence of a common pathway (27). The various intermediates of the pathway are common to both organisms even though the system described in animals is concerned with the oxidation of propionate whereas the system described in bacteria involves the anaerobic production Of propionate. Although the steps involved are the same for both organisms, the characteristics of several of the enzymes are different. The bacterial pathway (Figure .1) involves a condensation Of pyruvate and CO; to form oxalacetate (reactions I-IV). H mmDOHh a: INOO m sq N8 m :0 .. .02 Io NIOIA mmEchoOOmfi: _* 0 1/585 1. 1K_ _ N _* N _ A v <00! m... xx...U lOeI N Ion. Iowm .NH O\\ N v _ EEO/e <8 #52 a Noon 2 :05 :N8 1. N . MID NH Ion. On 0 ell) 6 mIos Imoo N Is as 1mm... HH 2%.: E a Imow INOO INOO H «:0 Iw H mIo «I0 5 ~8~Im Odo: m iii Nil/ V .* In? oft. of- 100*: 2% 1an On 9.13 :30 RS INS 18.0 .2 .8 mhqzoiomd O... ><>>I._.xomm ? XI *| CH3 * 5 H (fOH \ XII EH XIV Ozch-COA CH3 CH2 < >*I 3H2 HZNHZ acrylate were rapidly metabolized, but extracts subjected to dialysis, even under anaerobic conditions, soon lost activity. After dialysis for five hours it was possible to reactivate the extracts by adding ATP, ADP, pyruvate or acetyl phosphate thus demonstrating that energy is required. Prolonged dialysis resulted in loss Of the ability Of the extracts to ferment lactate or acrylate alone under any conditions, however, when equimolar proportions Of the acids were added together both were fermented slowly. Employing hydroxylamine as a trapping agent and lactate-C” as the substrate, these workers found radioactive spots on two-dimensional chrOmatograms corresponding in mobility to the hydroxamates Of lactate, acrylate and propionate. The metabolism of threonine by this organism occurs via routes consistent with the mechanism postulated for lactate metabolism (96). Soon after Stadtman (97) proposed the pathway propionyl COA dehydrogenase(XIII) and acrylyl COA aminase (XIV) were identified in extracts of (_3_. Hopionicum (27, 98). Propionyl COA dehydrogenase has not been purified nor its properties studied. Acrylyl COA aminase is present in large amounts in extracts of cells grown on B-alanine and has been purified and fully characterized (99). Goldfine and Stadtman (94) studied p-alanine utilization in extracts of C. propionicum and found a l3 situation similar to that described by Ladd and Walker (93) in extracts subjected to prolonged dialysis, i. e. , lactate and/or an a-keto acid was required in order to cause the conversion Of acrylate to propionate. It would seem that the link between acrylate and lactate was lacking in their preparations. Vagelos e_t a_l. (100) have succeeded in demonstrating the presence of an acrylyl COA hydrase- (VII) in extracts Of pigeon breast muscle and in a pseudomonad isolated from propionate enrichment culture, but have been unable to demonstrate either the lactyl COA dehydrase reaction (X1) or any enzyme capable of utilizing the lactyl COA formed. Oxi dati ve Pathways Studies carried out on animal tissues, peanut cotyledons and Clostridium kluyveri indicate the existence of alternate oxidative pathways for the utilization Of propionate (27). Two general mechanisms have been proposed (Figure 3). The first two reactions (XV, XVI) involving a dehydrogenation Of propionyl COA and hydration of the acrylyl COA formed to yield B-hydroxypropionyl COA are common to all three organisms. The S-hydroxypropionyl COA formed in animal (101) and plant tissue (102) is deacylated before further oxidation occurs (XX-XXII), while an alternate route is present in (_3_. kluyveri (103). This route involves a direct oxi- dation Of B-hydroxyprOpionyl COA (XVII) to malonyl COA. The significance of these pathways is unknown. Of some relevance are reports of a malonyl semialdehyde transaminase in kidney tissue (104) and a malonyl semialdehyde dehydrogenase in extracts Of Pseudomonas aeruginosa (105). l4 TNOO omo Tofu _ zmo _ zmo _ Nmn_o A :xx v NEW A Hxx v smw x ENOO TNOO moduo TONTHW NEW .mw u; wa A H>x w Two AIHWIY NEW msoo T~OO OTO Rimdno m3£umnu o>3mpfix0 .m oudwfih 15 Associated Enzyme Systems: The three pathways discussed above all have one feature in common, they all requirethe participation Of thiol esters. Bacteria carry out the required activation in three independent steps: M ++ acetate + ATP 5 acetyl phosphate + ADP (XXIII) acetokinase acetyl phosphate + COA 3 acetyl COA (XXIV) AM ‘pho sphotran sac etylas e acetyl COA + X —~ ‘ X-COA + acetate (XXV) VCOA transphorase Acetokinase (106), phosphot’ransacetylase (107) and COA transphorase (108) have been purified and their prOperties investigated. . Two enzymes, aceto-CoA—kinase and COA transphorase, carry out the same function in animals: COA + acetate + ATP ‘ ‘ acetyl - COA + AMP + PPi (XXVI) acetyl COA + X A acetate + X-COA (XXVII) \ (109,110,111). PART I CONVERSION OF LACTATE-C“ AND GLUCOSE-cl4 TO PROPIONATE BY THE RUMEN MICROBIOTA* The experiments to be described were undertaken to determine the degree Of participation Of each Of the pathways Of propionate production in the mixed population Of the bovine rumen. This was accomplished by examining the degree of randomization of C“ from lactate-C14 and glucose-C14 into carbon atoms 2 and 3 of propionate. In addition the selective effect upon the distribution of pathways (or distribution of species) exerted by the degree of carbohydrate availability in the diet was determined. *A substantial proportion Of these results have been presented at the 56th Annual Meeting of the American Dairy Science Association, Madison, Wisconsin, June 11-15, 1961, and published in The Journal of Bacteriology, Vol. 83, pp. 907-913, 1962. 16 METHODS AND MATERIALS . Bacteriological. Cows fitted with the screw cap, plastic fistula plugs described by Hentschl €_:_t 11' (112) were used as sources Of rumen micro- flora. The experimental rations were fed for at least two weeks prior to sampling. For incubation with lactate-C” and glucose-C14, samples were strained through 2 layers of cheesecloth and used directly. For the enzyme experiments, after a preliminary centrifugation at 2, 000x g to remove feed particles, the cells were collected by centrifugation at 18, 000 x g for 30 minutes, washed twice, and resuspended in 0.§02 M potassium phosphate buffer, pH 7.4. Cell-free extracts were prepared by exposing cell suspensions tO vibration in a 10 KC sonic oscillator for 35 minutes. The suspensions were centrifuged at 18, 000 x g for 10 minutes to remove cellular debris. The yellow-green extracts were stored at - 15C until used. Enzymatic. Succinic dehydrogenase activity was measured in both directions by the manometric and reduced dye methods of Singer and Kearney (113). The reduced dye employed was safranine. Acetokinase and phosphotransacetylase were estimated as described by Rose (114) and Stadtman (115). Coenzyme A transphorase activity was estimated by incubating, in a reaction volume of 1 ml, 10 moles Of acetyl phosphate, 0.01 mole coenzyme A, 50 urnoles of potassium lactate, excess phos- photransacetylase (10 units) and cell-free extract. After 5 minutes Of incubation at 38 C, neutral hydroxylamine (100 urnoles) was added and the sample allowed to stand at room temperature for an additional 10 minutes. The protein was precipitated with 25 ml. of 95% ethanol. . The samples were then taken to dryness on a steam bath, the hydroxamates extracted into ethanol and spotted on paper. The chromatogr‘ams were 17 18 developed with butanolezO (100:18), the spots located with a ferric chloride spray, eluted with a five per cent solution Of ferric chloride in 3 1:1 HCl, and the amount of lactyl hydroxamate determined (116). The assay was linear with enzyme concentration when the amount of lactyl hydroxamate formed during the incubation did not exceed 1 urnole. One unit was defined as the amount of coenzyme A transphorase which would form 1 pxnole Of lactyl hydroxamate under the conditions of the assay. Radiochemical. Radioactivity measurements were carried out in the Packard Tricarb Liquid Scintillometer. Two scintillation systems were employed; the first (toluene system) was made up of 0.4% PPO (2, 5-di- phenyloxazole) and 0. 005% POPOP (1, 4-bis-2-[5-phenyloxazolyl]benzene) in toluene; the second (gel system) consisted of 5 grams of PPO, 50 mgrams of aNPO (alphanaphthylphenyloxazOle), 80 grams of napthalene and 40 grams of cab-O-sil dissolved in 385 ml each of dioxane and xylene and 230 ml ethanol. The distribution of label in the volatile fatty acids and lactic acid was determined by adding 2 ml aliquots of the respective fractions eluted from a celite column (117) to 15 ml. of toluene scintil- lation fluid. Prior to degradation, acetic and propionic acids recovered from the column were further purified by 11 volume distillation. After addition of carrier, the acids were degraded by the Schmidt-azide reaction according to the method Of Phares (118) in the apparatus described by Krichevsky and Wood (119). The following modifications in the degradation procedure were employed: (a) the ethylamine recovered from the degradation of propionate was oxidized to acetate and then puri- fied by celite chromatography, (b) the carbon dioxide released in the degradations was collected in 10 ml of 0. 25 M Primene 81-R in methanol (120) or 0. 25 M NaOH and counted directly, (c) methylamine HCl derived from the degradation was counted directly in 8 ml ethanol plus 12 ml toluene scintillation fluid or 0. 5 ml aliquots were counted in 15 ml of the 19 gel system. From degradation Of acetate-l-CM, acetate-Z—CM, propionaterZ-C“, and propionate-3-C”, the cross-contamination between carbons were determined to be less than 1%. The specific activities Of the single labeled carbon Of these acids agreed within :1:2% with the measured specific activities of the starting acids. Since the distribution of C14 within a given sample was the only point Of interest, the yields during purification prior to degradation were not ascertained. Thus, the specific activities reported do not reflect the specific activities Of the acids in the incubation mixture, but the specific activities of the respective samples at the start of the degradation. Chemicals. Lactate-Z-C“, lactate-3-C“, acetate-l-C“, acetate-Z-C”, prOpionate-Z-C“, and prOpionate-3-Cl4 were purchased from Volk Radio- chemical Company. Glucose-l-C“, glucose-Z-C“, and glucose-6-Cl4 were purchased from New England Nuclear Corp. Pantethine was kindly supplied by Dr. 0. D. Bird Of the Park Davis Company. Coenzyme A was purchased from Pabst Laboratories. Primene 8l-R, kindly donated by Rohm and Haas Company, was vacuum distilled prior to use. Propionyl pantetheine was prepared by the anhydride method (121). Prior to use, the ester was purified by chromatography on Whatman 3 MM paper in butanOl-wate‘r (100: 18). Acyl hydroxamates were synthesized and chromatographed as described by Stadtman and Barker (116). Hydroxamates of acetate, propionate, lactate, and acrylate, used as standards, were prepared from ethyl acetate, ethyl propionate, ethyl lactate and the mixed anhydride Of ethyl carbonate and acrylate, respectively. RESULTS Stoichiometry Of lactate utilization: The diets indicated in Table l were selected so that relative to each other microbial activity would produce a graded increase in carbohydrate and/or lactate availability in the rumen. Previous studies (7) indicated that this increased availability would cause a decrease in the molar ratios of acetate to propionate in rumen fluid. In order to determine the amounts of acetate and propionate formed from lactate by the populations selected by these diets, 20 ml of strained rumen fluid were removed from the rumen 3 hours after feeding. The samples were added to 5 ml of 0.05 M phosphate buffer, pH 7. 0 con- taining 1 ucurie Of either sodium lactate-Z-C14 or lactate-3-C“ and sufficient unlabeled sodium lactate to bring the final concentration to 2 moles per ml (specific activity Of 0. Ozjucuri'e/umole). A sample of 3 ml was removed at zero time and samples of 11 ml each were removed after 45 and 90 minutes incubation at 39 C under a nitrogen atmosphere. The fer- mentation was stopped by the addition of several drops of 50% HZSO‘. The percentages of the lactate remaining after 45 minutes of incubation were, respectively, 32, 51, 56, 24 and 18 in flasks A through E. The molar ratios of acetate to propionate in the rumen fluid at zero time are pre- sented in the first column of Table 2. The production of acetate and propionate from lactate-C“ as determined by amount of radioactivity in the isolated acetate and propionate appears in column 2. The latter data are expressed as total counts in acetate divided by total counts in propionate. Although the ratios expressing acetate and propionate pro- duction in general are considerably higher than the molar ratios, the same trend toward decreasing ratio with increasing dietary carbohydrate availability is evident. High concentrations of lactate decrease the 20 21 Table 1. Diets Of cows used as sources of rumen fluid. m Cow Diet Alfalfa pellets (18)* Alfalfa hay (20) Grain (14), alfalfa hay-(2) Cooked grain (10), grain (4), alfalfa hay (2) WUOUJ> .Grass silage (65), lactic acid (1) 9 "Numbers indicate lb. /day. Table 2. Ratios of acetate to propionate in rumen fluid. Diet Acetate/propionate ratio - urnole/mriolea dpm/dpmD A 3.1 10. 0 B 3.1 10. 8 C 2. 5 9.1 D Z. 5 5. 5 E Z 2 2.5 a . . Determined at zero time. After 45 min. Of incubation with labeled lactate. Average of values obtained with .lactate-Z-Cl‘ and lactate-3-C“; dpm = disintegrations per min. 22 isotope ratios (Table 3). The data presented in Figure 4 show the effect increasing concentrations of lactate have upon the rate Of lactate utilization. The maximum rates of lactate disappearance observed for the five samples ranged from 0.05 to 0. 09 umoles/min/ml. Table 3. Effect Of lactate concentration upon the ratios Of acetate to propionate formed from lactate-C14. Diet Lactate Added Acetate/Propionate Ratio ' umoles umole/prnole dpmfdpm B o 4. 2a 17. 3b 4 4. Z 10. 5 C 0 3.3 7 l 4 3. 3 5 5 8. . . Determined at zero time. After 45 min. Of incubation with labeled lactate. dpm = disintegrations per min. Pathways contributing to lactate utilization. Acetate and propionate isolated from the incubation mixtures after 45 and 90 minutes were purified and degraded as outlined in Figure 5. The distribution of radio- activity within these acids is presented in Tables 4, 5, and 6. The data in Table 4 show that the acetate formed from lactate-Z-Cl4 or -3-Cl‘ was, respectively, carboxyl- and methyl- labeled as should be expected. This indicates that randomization between carbon atoms 2 and 3 of lactate during incubation does not occur to the extent that it does in the experiments with propionibacteria described by Leaver, Wood and Stjernholm (38). It was necessary in these experiments to utilize acetate as a measure of lactate randomization because the levels of 23 Figure 4. Lactate concentration vs. rate of lactate utilization. The curve presented was Obtained from average values determined for each of the five diets A through E. . Incubation mixtures contain- ing 10 ml Of rumen fluid, five ml Of 0. 02 M potassium phosphate buffer, pH 7. 0, and the indicated amount of potassium lactate were gassed with nitrogen and incubated at 37 C for 45 minutes. The samples were deproteinized with trichloroacetic acid (5%) and the r emaining lactate determined . LACTATE UTILIZED / HOUR (ALMOLE/ML) 24 50 LACTATE CONCENTRATION \L§ ' RATE OF LACTATE UTILIZATION 4.0 /O"f ‘9 3.0 ”/0 2.0 /Q 1.0 9 0 2.0 4.0 6.0 8.0 10.0 INITIAL LACTATE CONCENTRATION m 1401.531 ML) m>§prm pwmpoDfioz >20 Dmom>0>joz \mczmz 3&6 nos _zocm_5_oz :Lyodfim 1 ad _ 0m_1_.q.m OIIOZ>AOmm>pI< _ . J pomifim IIIIIIJ 5820255 . , _ . _ o_m_::. “ o_m_j_._. _ omopmwoxfifim “ omo>mmo_x<_.5m _ m r E 02693:. “ Wes o>mmozw o>mmox4» HEN wfiv NJ» hie mm mm Q mm mm Cm Em m4: Ndm w.m m.m - em as 0 mm vs NE Dd c.w~ m.: N.m o4 mm 5 m m» 2. wd $6 oOm w.mm H6 EN Hm mm < .58 cm .55 me .88 oo .58 my .58 oo .38 me .58 oo 8:5 mm .58 cm .55 me $358 no no dU Hooch. $5 manages. mm 3 definite oflfioomm Thais... see 3.8.55 5 32 s nonsense .m .32. £83.80 mo Goflfippm node po~fi§u3ob 3051\8QOM 8 em I- on“; I- 2.5 I- We Nd «mm 8 m 2 on 5 Tom 9: 0.2 on m.~ To No .3 o as me on: 0.5. «.3 mg mg. m.~ m: S O E E oomw 0.3. Rum N8 is as am . S m o» 2. m8. oi. one 8: m4 m; S . 3 < .58 on .58 me .58 on .58 me .58 on .58 me .58 on .55 me .58 on .58 me $353 nO no .0 Demos oono bungee/w on m>fi>30m 03825 I IA {0-9.0382 Eons obeoonoono E Hones no noses—End .o oEoP 29 Pathways contributing to glucose utilization. The amounts Of acetate and propionate formed from glucose by the populations selected by three diets (hay, hay plus grain, and high grain) were determined. Twenty ml samples of strained rumen fluid were added to 5 ml of 0. 05M phosphate buffer (pH 7.0) containing 1 ucurie Of either glucose-l-C”, glucose-Z-Cl4 and glucose-6-C” and sufficient unlabeled glucose tO bring the final concentration to O. 5%. A sample Of 3 ml was removed at zero time and the remainder incubated for 45 minutes at 39C under a nitrogen atmos- phere. The fermentation was stopped with H2804. Glucose utilization was virtually complete in all cases. The molar and isotope ratios of acetate to propionate are presented in Table 7. Degradation Of the acetate (Table 8) indicated that glucose degradation occurs via the Emden- Meyerhof glycolytic pathway and that very little randomization of label occurred during the incubation. The results Obtained from degradation Of the propionate formed from glucose-l-C”, glucose-Z-Cl4 and glucose-6-Cl4 are presented in Table 9. Calculations Of the contribution Of the nonrandomizing (acrylate) pathway were based on the following equation for glucose-l-C14 and g1ucose-6-C”: sp. act. of carbon 3 minus __ sp. act. of carbon 2 ' ' - 100 % nonrandom1z1ng pathway sp. act. of carbon 3 x Similar calculations were employed for glucose-Z-C“. 'NO corrections for randomization were employed. The calculations show that the "acrylate" pathway contributes up to 40% of the propionate formed. Enzjmes for propionate formation in cell-free extracts Of rumen fluid. In support Of the isotope experiments, cell-free extracts of bovine rumen microflora were prepared by sonic oscillation and tested for the ability to catalyze certain reactions of the succinate and acrylate pathways to 30 Table 7. Ratios Of acetate tO propionate in rumen fluid. Ac etate/Propionate Ratio Diet a umole/urnole dpm/dpmb Hay (.20)C 4.0 3.1 Hay (10), grain (8) 2.6 1.4 Hay (2), grain (14) 3. 2 1. 8 a . . Determined at zero time. bAfter 45 min. Of incubation with labeled glucose. Average Of values obtained with glucose-l-CM, glucose-Z-C”, and g1ucose-6-Cl4; dpm = disintegrations per min. CNumbers indicate pounds per day. Table 8. Distribution Of label in acetate from glucose-l-CM, glucose- 2-Cl4 and glucose-6-Cl‘1'.a J’ = Diet Glucose-l-C” Glucose-Z-C14 Glucose-6-Cl4 C1 C2 C1 .C2 C1 C2 Hay 25 720 800 73 56 625 Hay, grain 25 500 760 11 30 600 Hay, grain 7 1,000 990 17 6 845 aValues expressed as dpm/mele. 31 Table 9. Distribution of label in propionate from glucose-l-CM, glucose-Z-C” and glucose-6-C”. a . Specific Activity % Acrylate b Dlet Su strate Total9 C1 C2 - C3 pathway Hay C-l-C” 87.0 2.7 42.0 44.5 0.6 Hay + grain 86.0 2.8 33.8 52.0 34.5 High grain 42.0 0.7 14.7 26.3 45.8 Hay C..2-C14 100.5 17.0 42.5 43.5 1 Hay-I- grain 115.5 6.8 64.5 41.5 36.0 High grain 138.0 1.7 84.0 50.5 40.5 Hay C-6-Cl4 77. 5 1.7 33.2 37.2 1 Hay + grain 114.0 1.7 39.5 68.0 41.0 High grain 32.0 0.2 11.2 18.4 39.0 a dpm/emole . bafter addition of carrier (lmmole). propionate. Table 10 gives the specific activities found for succinic dehydrogenase, acetokinase, phosphotransacetylase and coenzyme A transphorase. . Succinic dehydrogenase activity compares quite favorably with the activity found in X.alca1escens used as a positive control. The high rate of fumarate reduction observed in the control organism was not Observed in the extracts Of the rumen microflora. Acetokinase, phosphotransacetylase and coenzyme A transphorase were very active indicating that the activating system for organic acids is present. To demonstrate further the existence of the acrylate pathway, extracts were incubated with propionyl pantetheine and prOpionate-Z-CM. 32 3:: Table 10. Specific activities Of several enzymes present in rumen- fluid extracts. Source of Extract Enzyme Rumen Rumen Pgatostrep- \_/'_.alca1es- (grain) (hay) tococcus cens Acetokinase 0.72 0.75 2.5 —-- Phosphotransacetylase 3 . 4 3 . 5 10 . 6 - - - COA transphorase 1.0 0.81 1.8 -..- Succinic dehydrogenase succinate -> fumarate 0. 20 0 . 20 --- O. 19 fumarate -> succinate 0. 28 0 . 32 --- 6. 3 3:: Units/mg protein under standard assay conditions for the respective enzymes. After 5 min. of incubation, 1 ml of 2M hydroxylamine was added and allowed tO react. The samples were chromatographed on paper and the hydroxamates located by spraying with ferric chloride; radioactivity was located on duplicate Chromatograms by counting 1 cm strips immersed in 15 ml of scintillation fluid. . Figure 6 shows that 4 radioactive peaks were Obtained. The peaks corresponded in mobility to known acetyl, propionyl, lactyl, and acrylyl hydroxamates. The spots located with the ferric chloride corresponded to the radioactive peaks. A demonstration of the reverse reaction was accomplished by incubating the cell-free extracts with lactate-Z-C“, ATP, and acetate. The hydroxamates were prepared and chromatographed. A distinct radioactive peak corresponding to acrylyl hydroxamate was Observed and the FeC13 spray located spots corresponding to acrylyl, acetyl, lactyl, and propionyl hydroxamates. The radioactivity in the areas Of the latter three acids could not be resolved into discrete peaks presumably because 33 00. o MMDOHLH am om . on 2. oo on on on ON 0. u q — _ u H q q - 319213317 319213111 HIV—IAHOV 31VNOIdOHd J 051$ 293?. do mko> MZEIPMPZQn. I...U...._>zo_aomo_ .5 nZD_nru>ZDu IA\L CV Om ON. 00. OON ovm omN —\\‘~ 00M ovm . BIONIW 83d SINOOO 34 the activity of the acetate and lactate peaks was too high. However, a distinct radioactive area for propionyl hydroxamate: was found. DISCUSSION The isotope experiments establish the fact that two mechanisms of propionate formation exist in the natural ecological system of the bovine rumen. It is interesting to note that the lesser known, non- randomizing route, predominates in lactate utilization and performs a 7 significant role in glucose utilization. Opposite situations prevail with lactate and glucose as substrates, in that the dicarboxylic acid route plays a minor role in lactate metabolism and a major role in glucose metabolism. The feeding regimes were selected to present varied amounts Of available carbohydrate on the assumption that higher carbohydrate availability promotes the growth of organisms which produce higher levels of propionate (7). The molar and isotope ratios of acetate to propionate reported are consistent with the conclusion that as the carbohydrate or lactate availability of the diet increases, a higher prO- portion of propionate is produced. Apparently under conditions of increasing substrate availability organisms possessing the nonrandomiz- ing pathway are favored because the contribution Of this pathway increases with the grain content Of the diet. This conclusion is supported by the studies carried out on five diets with lactate-Cl4 and on three diets with glucose-C“. The contrast between the glucose-Cl4 experiments wherein the molar and isotope ratios of acetate to propionate agree and the lactate- C14 experiments where there is a sharp discrepancy between the molar and isotope ratios, suggests that lactate does not equilibrate with the three-carbon intermediates arising from hexose fermentation. 3 Consideration of this Observation and the fact that the pathways of pro- pionate formation differ quantitatively with respect tO the two substrates 35 36 and with diet led to the concept presented. in Figure 7. Two routes Of glucose fermentation are indicated: the first involves a direct con- version Of glucose to propionate and acetate by a single organism such as a propionibacterium; the second requires the participation Of two types of organisms--a lactate former (Streptococcus bovis) and a lactate utilizer such as _P. elsdenii. If diets high in starch promote the growth of homolactic species such as S. bovis and the increased lactate formed * by these organisms in turn selectively increases the population Of lactate fermenters--a supposition amply supported by direct evidence (7, 9, 18)--- the apparent result would be a relative increase in the participation of the two step sequence. Further, if the lactate utilizers selected convert lactate to propionate via acrylate, and lactate does not equilibrate with other 3-carbon intermediates arising from glucose the expected result would be an increase in the contribution of the acrylate pathway. These effects have been clearly demonstrated and are consistent with the postulate presented in Figure 7. Lack Of knowledge Of the many types of organisms isolated from lactate. enrichment cultures precludes critical examination of the suggestion that the bulk of these organisms convert lactate to propionate via the nonrandomizing pathway. However, under the conditions employed in the experiments with uniquely labeled lactate the acrylate route was predominant. It should be recognized that the experimental conditions were not strictly parallel to the conditions existing in the rumen: i. e. , less COZ tension and a higher pH (7. 0 experimental vs. 6. 2-6. 8 for rumen fluid). The degree to which these differences influence the distribution of pathways is not known. The dicarboxylic acid pathway--long considered to require a CO; fixation step--is now known to involve a transcarboxyl- ation between methylmalonyl COA and pyruvate (66). . Hence the dependence Of this route upon exogenous COZ, although unknown, is probably not very high. Similarly, the pH differential could favor the 37 Figure 7. Routes of Conversion to Propionate in Rumen Fluid. Dietary "Glucose" GluCose Homolactic species (Streptococcus bovis) Lactate I Added Lactate I Lactate Pool I Lactate I Peptostreptococcus other 3 I Acrylyl COA Propionate I Propionate pool JG Ix Glucose NI Propionibacterium Veillonella V 9 Pyruvate Succinate Propionate 38 acrylate route if the pH Optima of the two systems differed in a way so as to favor the acrylate route at pH 7. 0. The effect Of this shift in pH is not known. Ample evidence is available to indicate that the dicarboxylic acid cycle Operates in the rumen (7). The nonrandomizing pathway has been partially characterized in g. propionicum and f. elsdenii (27, 93). The isotopic, enzymatic and chromatographic data Obtained have consequently been interpreted against this background, without unequivocal proof that the same reaction sequences occur. The incubation of cell-free extracts with (a) propionyl- pantethine and prOpionate-C14, and (b) lactate-Z-CM, ATP and acetate, yielded results consistent with the routes proposed for (_3. propionicum and _P. elsdenii. Measurement of succinic dehydrogenase, acetokinase, phospho- transacetylase and coenzymeA transphorase activity in cell-free extracts Of rumen microflora selected on several diets indicated no discernible trend. Errors due to sampling and cell preparation were relatively large, and attempts to correlate the activity of these enzymes with diet were therefore not feasible. SUMMARY Rumen microflora enriched on diets calculated to present increas- ing carbohydrate or lactate availability were employed to determine the contribution of the randomizing (succinate) and nonrandomizing (acrylate) routes to propionate with lactate-Z-CM, lactate-3-Cl4, glucose-l-CM, glucose-Z-C” and glucose-6-Cl4 as substrates. On the lactate substrates the propionate was labeled as though 70-90% was formed via the non- randomizing route,while on the glucose substrates the contribution Of this pathway ranged from 0-40%. The contribution Of the nonrandomizing pathway was highest on diets containing high levels Of readily available carbohydrate or lactate or both. . Evidence for the presence Of succinic dehydrogenase, acetokinase, phosphotransacetylase and coenzyme A transphorase was Obtained with cell-free extracts. PrOpionate--C14 and lactate-C14 were converted by extracts to the activated derivatives Of acrylate, lactate, propionate and acetate. 39 PART II LACTATE METABOLISM BY PEPTOSTREPTOCOCCUS ELSDENII The primary aim Of the investigations reported in this section was to provide unequivocal proof for existence Of the direct reductive or acrylate pathway. This was accomplished by identifying, purify- ing and partially characterizing the enzymes of the pathway. 40 METHODS AND MATERIALS Bacteriological. . Peptostreptococcus elsdenii, strain B-159, kindly provided by Dr. M. P. Bryant and employed throughout these studies, was grown on the corn steep liquor (Staley) sodium lactate medium described by Ladd and Walker (93). Cells, grown in deep culture at 38 C, were collected in a Sharples centrifuge and washed twice in O. 02 M phosphate buffer, pH 7. 5, prior to disruption. " Cell-free extracts were prepared by exposing thick cell suspensions to sonic oscillation (Raytheon 10 KC) for 5 to 10 minutes. The suspensions were centrifuged at 18, 000 x g for 15 minutes to remove cellular debris. Extracts were stored at -14 C. Chemicals. Lactate-l-C“, lactate-Z-C” and propionate-3-C14 were purchased from Volk Radiochemical Company. . Safranine, cytochrome c, 2, 6 dichlorophenol indophenol (DCP), methylene blue (MB), 2., 3, 5 triphenyl tetrazolium- HC1(TTZ), diphosphopyridine dinucleotide (DPN), and 3-p-nitrophenyl-2-p-iodophenyl tetrazolium chloride (INT) were commercial preparations. Pantetheine (Pa), purchased from the California Corporation for Biochemical Research, was reduced with potassium borohydride. Coenzyme A (COA) was purchased from'Pabst Laboratories. Thiol esters of acetate, prOpionate, butyrate and acrylate were prepared by the anhydride'method (121). Pantetheine esters were purified prior to use by extraction with ethyl acetate and chromatography on acid washed Whatman 3 MM paper in water-saturated butanol. Typical spectra for acetyl COA, propionyl Pa and acrylyl Pa are pre- sented in Figure 8. , Lactyl COA was prepared from the mixed anhydride (94) of lactate and ethyl formats by the thiol ester exchange method 41 ABSORBANCY 42 0.7 ' 0.6 0.5 0.4 0.3 0.2 0.| SPECTRA 0F THIOL ESTERS I ACRYLYL PANTETHEINE PROPIONYL PANTETHEINE I ACETYL COA 2I5 235 1 l L 255 275 295 WAVELENGTH - MILLIMICRONS FIGURE 8 BIS 43 described by Wieland (122). Lactyl COA and acrylyl COA were prepared enzymatically by incubating acetyl COA (10 moles) with potassium lactate or acrylate (100 prnoles) and purifiedeoA transphorase (100 units) for 15 minutes at 37 C. The protein was removed by alcohol precipi- tation and the product purified by chromatography on acid-washed Whatman 3 MM paper in ethanol, sodium acetate (50:50) as described by Stadtman (121). In several cases the esters synthesized by this method were used without prior purification by paper chromatography. All thiol esters were assayed by the hydroxylamine method (121) and tested for purity by hydroxamate chromatography (116). _ Coenzyme A esters were'assayed enzymatically by arsenolysis with purified phosphotransacetylase and COA transphorase (121). Acetic, propionic, butyric, acrylic and lactic acids were identified by hydroxamate chromatography in two solvent systems according to the method of Seubert (123). Lactic and acrylic acids were separated prior to identification either by column chroma- tography (117) or by distillation. Standard techniques were employed for the isolation and identification of pyruvate (124-126): samples were deproteinized with metaphosphoric acid, reacted with dinitrophenylhydrazine for five minutes, and the hydrazone purified for chromatography by extraction serially into xylene, l M sodium carbonate, and chloroform: ethanol (80:20), respectively. The purified hydrazones were separated by paper chromatography in methanol:benzenezn-butanolezO (4:2:2:2). Spectrophotometric measurements were carried out with a Beckman DU monochromator fitted with the cuvette positioning and automatic recording equipment described by Wood and Gilford (127, 128). Protein was determined by the Lowry method (129). . Radioactivity was determined by the procedures outlined in Part I. 44 Enzymatic. . Several enzymes were identified, purified and characterized in the course of this work. Since each enzyme required the application of different techniques, the methods employed will be described separately. Phosphotransacetylase was identified by the methods described by Stadtman (115). The arsenolysis assay, preferred by earlier workers (115), was employed in some of the studies, but was later discarded because large amounts of enzyme were required, reproducibility was poor and the assay was not linear with enzyme concentration. Thepre- ferred assay system involved measurement Of the rate Of formation Of acetyl COA (XXIV) at 232 mu. acetyl phosphate + COASH ‘ acetyl COA + Pi (XXIV) V The incubation mixture (0. 20 ml) contained 25 moles of Tris buffer, pH 8. O, 0. 05 moles of COA, 5 moles of acetyl phosphate and enzyme. One unit was defined as the amount of enzyme required to form 1. 0 urnole Of acetyl COA per minute under standard assay conditions. . Phosphotransacetylase was purified 92-fold (Table 11).‘ The initial heat treatment at 55 C for six minutes served to coagulate residual cellular debris and some protein. The nucleic acids were removed by adjusting the protein content Of the extract to approximately 25 mg/ml adding 0. 20 volumes of a 2% protamine sulfate Solution, and removing the precipitate by centrifugation. Solid ammonium sulfate was added to the protamine sulfate supernatant to 45% saturation (calculated at 0 C, (130)). The precipitate, containing the enzyme, was‘resuspended in O. 05' M phosphate buffer, pH 7. 5. The enzyme was then precipitated with ethanol at -15 C. After the addition of an equal volume Of ethanol and centrifugation to remove the precipitate, the enzyme was precipi- tated by adding a second volume Of ethanol. The enzyme was resus- pended in phosphate buffer and treated with a saturated solution of 45 Table 11. Phosphotransacetylase purification. Stepa Total Units Sp. Act. . Fold Crude 189,000 19.8 Heat 66,500 31.1 1.6 Protamine 67, 000 34. 2 1. 7 Ammonium sulfate I 65, 000 87. 0 5 . 8 Ethanol 58, 500 265 13.4 Ammonium sulfate 11 51, 500 680 34. 3 Heat 50, 000 1, 820 92. O a . See text for details. ammonium sulfate, pH 8. 2. The ammonium sulfate concentration was increased in a stepwise manner until all the activity was recovered. The preparations having the highest specific activities were combined, the protein concentration adjusted to 10 mg per ml, and heated at 58 C for three minutes. The preparations were stored frozen overnight at -14 C, thawed and centrifuged. The purified enzyme was very stable when stored at -14 C. NO measurable contaminants were present. Coenzyme A transphorase was assayed by the method of Stadtman (115) (XXV, XXVIII) . butyryl COA + acetate ————-> acetyl COA + butyrate (XXV) acetyl COA + arsenate —-———> acetyl arsenate + COASH (XXVIII) The only change involved a ten-fold decrease in volume. The 0. 25 ml incubation mixture contained 7. 5 prnoles of potassium acetate, 50 urnoles potassium arsenate, pH 8.0, l. 0 unit of phosphotransacetylase, and COA transphorase. The assay is not linear with time, however, if initial 46 rates are considered, the velocity is linear with enzyme concentration when the rate does not exceed four. times the endogenous rate due to phosphotransacetylase. One unit Of enzyme converts 1 mole per minute under standard assay conditions. Coenzyme A transphorase has been purified 87-fold (Table 12). The initial heat and protamine steps employed were the same as those described for phosphotransacetylase. Ammonium sulfate steps I and II were the same, the protein precipitating between 55 and 70% saturation was collected in both cases. The ethanol precipitation was not very reprOducible and was run in a stepwise fashion, only the fractions con- taining high specific activity were retained. This step was carried out at --15 C. The terminal heat treatment involved holding the preparation at 65 C for four minutes. The purified enzyme was stored in 0. 02 M phosphate buffer, pH 7. 5, at -14 C. The heat treatment inactivated all the re sidual phosphotransac etyla s e . D-Lacti‘c dehydrogenase was identified by the methods described by Nygaard (51). The standard assay employed during purification Of D-lactic dehydrogenase involved measurement of the rate of 2, 6-dichloro- phenol indophenol (DCP) reduction at 600 mu (XXIX). DL lactate + dye_———-> pyruvate + reduced dye (XXIX) The assay mixture contained 30 moles Of phosphate buffer, pH 7. 0, 0.002 moles of dye, 5.0 muscles of D.-lactate, pH 7.0, and enzyme. The final volume was 0. 251ml. One unit was defined as the amount of enzyme causing an absorbancy change of 1. 0 in five minutes at 600 mu under standard conditions. Lactic dehydrogenase was purified 25 fOld (Table 13). In order to avoid precipitation of the enzyme during protamine treatment, the salt concentration of the extract was adjusted to 0. 2 M with solid 47 Table 12. Coenzyme A transphorase purification. Stepa . Total Units Sp. Act. Fold Crude 48, 000 10. 7 Heat 48,000 22.4 2.1 Protamine 48, 000 24. 6 2. 3 Ammonium sulfate I 46, 500 60. 8 5. 7 Ethanol Ammonium sulfate 11 28, 800 266 24. 8 Heat 25,000 930 87.0 a See text for details. Table 13. . Lactic dehydrogenase purification. Stepa . Total Units Sp. Act. Fold Crude 1, 140, 000 490 Protamine sulfate 1, 100, 000 510 1. 1 Ammonium sulfate I 1, 030, 000 795 1. 6 Calcium phosphate gel 500, 000 3, 920 8. 0 Ammonium sulfate II 324, 000 12, 000 24. 5 a - See text for details. 48 ammonium sulfate and then 0. 2 volumes Of 2% protamine sulfate was added. The first ammonium sulfate step involved collection of the protein precipitated between 60% and 80% Of saturation and the second between 55% and 65% Of saturation (obtained with solid ammonium sulfate, 130). The calcium phosphate gel absorbtion was run directly on the resuspended ammonium sulfate precipitate. The enzyme preparation was treated with increasing aliquots of the gel suspension until adsorption occurred (usually at a gel/protein ratio of 0. 1), and the several batches of gel containing the bulk of the activity were combined. The gel was washed with water and the enzyme was eluted with O. 5 M phosphate buffer, pH 7. 5. The purified enzyme was yellow-green in color, relatively unstable during storage and was contaminated with diaphorase and acyl COA dehydrogenase. Acyl COA dehydrOjenase was purified 23-fold (Table 14), by a) precipi- tation with protamine sulfate (0. 2 volume of a 4% solution) and resus— pension in 0. 05 M ammonium sulfate, b) stepwise absorption on alumina C 7 gel and elution with 0. 5 M phosphate buffer, pH 7. 5, and c) precipi- tation with ammonium sulfate (65 to 75% saturation). The purified enzyme was stable in 0. 2 M phosphate buffer, pH 7. 0, but very unstable at low salt concentrations. The standard assay procedure employed was the same as the one employed for lactic dehydrogenase except that butyryl COA was the substrate (XXX) ._ butyryl COA + dye ————> crotonyl COA + reduced dye (XXX) The same definition for one unit Of activity was employed. Diaphorase activity was Very high in crude extracts (Table 15). Since this enzyme was employed in assays for other enzymes, the primary aim during fractionation was removal of contaminating activities, not 49 Table 14. Acyl coenzyme A dehydrogenase purification. Stepa Total Units Sp. Act. . Fold Crude 55,000 33.5 Protamine sulfate 52, 000 99. 0 3 . 0 Alumina Cy gel 32, 600 168. 0 5. 0 Ammonium sulfate 12, 000 775 23 a I See text for details. Table 15. Diaphorase purification. Stepa Total Units Sp. Act. Fold Crude 640 0. 21 Protamine 5 10 0. 36 1. 7 Ammonium sulfate I 340 O. 50 2. 4 Ethanol 260 1. 10 5. 2 Ammonium sulfate 11 185 3. 10 14. 6 a See text for details. 50 purification. With the exception Of the protamine step, which was the same as that described for lactic dehydrogenase, all of the methods employed during fractionation were run in the stepwise manner described earlier and only the fractions containing high diaphorase‘ activity and low contaminating activity were retained; hence, the recovery was very poor but the final preparation was entirely free of measurable contaminants. The standard 2, 6 dichlorophenol assay procedure described by Mahler El: a_l. (131) was employed. The definition of unit and specific activity reported are the same as those described for lactic dehydrogenase and acyl COA dehydrogenase. RESULTS Demonstration of lactyl COA dehydrase. The enzyme of particular interest in these studies was lactyl COA dehydrase because, unlike the other enzymes of the pathway, this dehydrase had not been demonstrated in an organism capable of converting lactate to propionate via a non- randomizing route (94). Vagelos gt a_._l.. (100) reported an irreversible acrylyl COA hydrase in extracts of a pseudomonad and in pigeon breast muscle. . Presumably, this enzyme could not function in propionate formation. Early evidence, indicating that lactyl CoA dehydrase was present in crude extracts of the organism, was obtained as described in Part I. Lactyl-CM-COA or propionyl-CM-COA was incubated with crude extracts, the thiol esters converted to hydroxamates, separated by chromotography, and counted. The results, presented in Figure 9, curve I, show that lactyl-CM-COA is converted to the activated derivatives, presumably the COA esters, of acetate, acrylate, propionate, and B-hydroxypropionate, suggesting that the following reactions occur: lactyl-Cu-COA —————\ acrylyl-CM-COA (XI) ‘— acrylyl-CM-COA —-——-5- propionyl-Cu-COA (XIII) ,‘_____ acrylyl-C”-CoA————§_ B-hydroxypropionyl COA (XVI) lactyl-CM-COA —————¥ lactate-Cl4 + CoA (XXXI) -—-—-—— lactate-C,“ ———-\ pyruvate—C“ (XXIX) pyruvate-C“ + CoASH ———-> acetyl—COA + CO2 (XXXII) Although highly suggestive, these data do not constitute proof for the existence of the dehydrase. The desired proof required purification of the dehydrase and isolation'of the products. Toward this end a Spectrophotometric assay system based upon the measurement of lactic acid formation was devised as follows: 51' 52 Figure 9. Lactyl COA utilization: chromatography Of hydroxamates. Curve I -- crude enzyme Curve II -- after (NI-1,)ZSO4 treatment Curve III -- purified enzyme Curve IV -- Purified enzyme plus acyl CoA dehydrogenase 53 6.0. x ”152:2 mun mszaoo T s m x E _.M.. E M T A M E mEzoioma -mP<._>mO< A We mpqzoiomn U m 1 M T N R N m m. NM 34564 nlv 0 MW. 3503 m M E453 F T R H l A l A D m mZzoEoma m m —I/l. VI R P L L H E Y I F W m T A U 0 A w 0 C P ”1220an - 4.554 E L M w. M Y 1 Y G Y z Eggnog-33304 a o. o. a A M n 5452 0 L E W w <53 m E593 R m R I H W C BezoaoE-crxomo»: Q I.) _ _ _ Ir L h 6 4 2 0 Q 8 4 0.6 0.8 0.4 0.6 0.4 R; VALUE 54 acrylyl COA . > lactyl COA (XII) lactyl COA + acetate ————e-> lactate + acetyl COA (XXV) lactate + DPN ———-—> pyruvate + DPNH (XXXIII) DPNH + dye ————> DPN + reduced dye (XXXIV) The incubation mixture (0. 3 ml) contained 15 prnoles of tris~acetate buffer, pH 8. 0, 0.0004% INT, 1. 0 mole of DPN, 0.15‘umole of acrylyl COA, 0.02 ml of muscle lactic dehydrogenase .(1 to 10 diluti'on'of Worthington slurryof crystals), 10 units each of diaphorase and COA transphorase and varying amounts of lactyl COA dehydrase. The rate of INT reduction was followed at 485 mu. This procedure could not be employed to assay crude extracts because of lactate contamination, therefore, an alternate assay based on hydroxamate chromatography (as described above) was employed to locate the activity in an ammofiium sulfate fraction (60-90% of saturation), and this preparation was used to standardize the Spectrophotometric assay. The data presented in Figure 10 show that substrate, DPN, dehydrase and lactic dehydrogenase are required for dye reduction. . Neither acrylyl Pa, acetyl COA nor potassium acrylate substituted for acrylyl COA, however, acetyl COA and potassium acrylate in combination did partially replace acrylyl COA. The flavin-linked D-lactic dehydrogenase isolated from the extracts (Table 13) did not substitute for the muscle L-lactic dehydro- genase. Hence, it appears that lactyl COA dehydrase is specific for L-lactate. The assay was linear with respect to enzyme concentration only at very low rates (Figure 11) but is sufficiently accurate for puri- fication studies. It was found that freshly prepared diaphorase, which was free of phosphoroclastic enzymes (Table 15),. contained appreciable amounts of lactyl COA dehydrase. This preparation was employed in experiments designed to accumulate sufficient pyruvate in the assay mixture for isolation, .and to demonstrate the interconversion of lactyl COA and ac rylyl CoA. 55 Figure 10. Lactyl COA dehydrase: requirements of the assay system. Different components Of the assay system as described in the text were omitted as indicated. 1.2 1. 1 0 E In (D V 0.8 >_. 0 Z 4 m 0.6 a: O (D 3 0.4 0.2 0 56 LACTYL CoA DEHYDRASE ACRYU’L COA ——’ LACTYL CoA LACCYI. COA ACEIYL COA ACETATE LACTATE DPNH + are ——> DPN wry-1, O/9 LACTATE + DPN—b PYRUI/ATE + DPNH /9 I 2 3 4 MINUTES /0/ / .1 /Q 0 WITHOUT EITHER DPN, _— LACTIC DEHYDROGENASE, 0 OR ACRYLYL COA I I @O—L : l L 0...? 5 57 Figure 11. Lactyl CoA dehydrase assay: rate vs. enzyme concentration. The standard assay system described in the text was employed. A ABSORBANCY AT 485 m,u./5 MIN. 0 58 LACTYL CoA DEHYDRASE ASSAY ACRYLYL co A + Hgo-etACTYL COA O / / / / o\o 1! 0.005 0.0I 0.0l5 0.02 DEHYDRASE CONC. (ML) 3 0.025 59 ‘ Pyruvate was isolated from incubation mixtures as its dinitro- phenylhydrazone and identified by paper chromatography. The incubation mixture (2 m1) contained 250 prnoles of Tris-HCl buffer, pH 8.0, 0. 5 mole of DPN, 0. 1 ml of muscle lactic dehydrogenase diluted 1 to 10, 0‘. 005% INT, dehydrase and either two moles of acrylyl COA or two umoles of acetyl COA plus 50 prnoles of potassium acrylate. The reaction mixtures were incubated at 390C for 15 minutes. The samples were reacted with 2, 4-dinitrophenyhydrazine and the hydrazones were purified and separated on paper Chromatograms. Spots corresponding in mobility to dinitrophenylhydrazones prepared from standard pyruvate solutions were observed when either acrylyl CoA or acetyl COA plus potassium acrylate served as the substrate. . No spots were observed when either substrate, enzyme or lactic dehydrogenase was omitted from the reaction mixture. The spots were more intense with ac rylyl COA as substrate. A further check on the identity of the phenylhydrazone spots was accomp- lished by adding sodium lactate-B-C“ (0. Olucuries) to one duplicate containing the complete reaction mixture. The pyruvate-Cm dinitrophenyl- hydrazones formed in the reaction mixture chromatographed identically to the spots observed in a duplicate without lactate-C“ added (Figure 12). The pH optimum of the assay is 8. 0. Versene is slightly stimula- tory in some cases. The reaction is inhibited by divalent cations, protamine sulfate, both sodium and potassium phosphate and potassium acrylate. Several experiments were carried out to demonstrate the conversion of lactyl COA to acrylyl COA. The first method involved incubation of the dehydrase with lactyl-CM-COA, reaction with hydroxylamine, paper chromatography and counting the chromatogram. The results presented in Figure 9, curve 111, show that acrylyl—CM-COA was formed. * The second method employed acyl COA dehydrogenase and reduced safranine as a linking system to determine acrylyl COA formation. COUNTS PER MINUTE x 10-3 6O CHROMATOGRAPHY 0F REACTION PRODUCT DINITROPHENYLHYDRAZONE PAPERGRAM . . a. UV QUENCH 4%?“ l2.0 ' to O 0 f0 \ O 5E301N02 0 HYDRAZON E-lI I g e. '2‘ ‘2’ 6.0 - E a E 1'3 T.- O § § 23 D ‘23 0: 0 ml 3-0‘ a / E a O ° \ Q Q—fi'Q/O 1 1 O 0.4 0.5 0.6 0.7 0.8 RF VALUE FIGURE 12 gt 61 Attempts to employ the system in a continuous assay for reduced safranine oxidation were unsuccessful. lactyl COA ——-———> acrylyl CoA (XI) acrylyl COA + leucosafranine ————> propionyl COA + safranine (XIII) However, when lactyl-CH-COA was preincubated with lactyl COA dehydrase prior to the addition Of acyl COA dehydrogenase, oxidation of the reduced dye occurred (Figure 13). Isolation of the hydroxamates formed in this reaction mixture after the addition Of hydroxylamine indicated that acetyl-CM-COA and either acrylyl-CM-COA or propionyl- C14-COA or both were formed (Figure 9, curve IV). Apparently the purified acyl-COA dehydrogenase preparation also contained the phos- phoroclastic enzymes. Theoretically, when acetyl CoA, lactate-C“, COA transphorase, and lactyl COA dehydrase are incubated together the following reactions would occur: I lactyl-Cu-COA ————> acrylyl-CM-COA + H30 (XI) acrylyl-Cu-COA + acetate ——> acrylate-CHM- acetyl CoA (XXV) acetyl-COA + lactate-C“ —--> lactyl-Cu-COA + acetate (XXV) This should result in a net synthesis of acrylate-C” through continuous regeneration of substrate. . Less than stoichiometric amounts of COASH are needed because of recycling. Reaction mixtures contained 250 prnoles of Tris-HCl buffer, pH 8.0, 150 prnoles of potassium lactate, . 0. 5 ucuries of sodium lactate, 10 moles of acetyl phosphate, 50 (units each of phosphotransacetylase and COA transphorase, 1.0 prnole of lactyl-Cu-COA and dehydrase. Incubation periods ranging from five to thirty minutes were employed.. Five to ten per cent of the original substrate was converted to acrylate-C“. Apparently, the COA moiety was not recycled back to form new lactyl-Cu-COA. 62 Figure 13. . Enzymatic determination Of acrylyl COA formation. The incubation mixtures consisted of two parts. The first contained 100 moles of Tris.HCl buffer, pH 8.0, 3 moles of lactyl-Cu-COA (sp. act. approximately 2, 000 dpm per mole), and 0. 1 ml of lactyl COA dehydrase (equivalent activity in standard assay was 100. 0 optical density units per 5 minutes) in a total volume of 1. 5 ml.' The second part, the analytical system, contained 0. 015% safranine and 500 units of acyl COA dehydrogenase in a total volume of l. 1 ml. The analytical system was added at zero time or after the preincubation period, as indicated. The reactions were run in Thunberg tubes which were flushed with nitrogen prior to adding 0.4 ml Of sodium hydrosulfite (0.12% sol'n) to reduce the safranine. The products were reacted with hydroxylamine, separated on paper, and counted (see Figure 9, curve 4). 63 ENZYMATIC DETERMINATION OF ACRYLYL COA FORMATION 0 4 /O ,O o—eégfl‘ 8 I2 MINUTES LACTYL 60.4 ——> ACRYLYL CoA ACRYLYL coA PROPIONYL cm 0.75 4' ——" + LEUCOSAFRANl/VE SAFRANINE '1 ' | ——O ’ E 0/0 O v 0.60 0/ K) 23" PREINCIUBATED/ ‘ E 0.45 9 (D m / 8 o m 4 0.30 / /O 0.15 f NOT PREINCUBATED I 0—40—0-1-‘9—40 l6 20 64 Lactyl COA dehydrase has been purified 15 fold through the appli- cation of the following techniques: Protamine precipitation and elution with either 0. 1M ammonium sulfate or 0. 1M potassium phosphate buffer, pH 7. 5; adsorption on calcium phosphate gel and elution with 0. 1M potassium phosphate buffer; pH 7. 5; precipitation with ammonium sulfate (SO-70% saturation); adsorption on hydroxyapatite gel and elution with 0. 5M potassium phosphate buffer, pH 7. 5; and precipitation with ammonium sulfate (SS-75% saturation). The yields during, purification are in the range of 10-15 per cent. Sometimes the protamine step was omitted. Lactyl-COA dehydrase is relatively stable in crude extracts but rapidly loses activity during purification. Attempts to stabilize the purified enzyme with a variety of buffers over a pH range of 6. 0-8. 0, reducing agents, AMP, ADP, lactate, acrylate, boiled crude extracts, and bovine serum albumin have been unsuccessful. Purified preparations of the enzyme react slowly with acrylyl Pa and the reaction cannot be followed spectrophotometrically at 263 mp. 1 because the rates hardly exceed the rate of spontaneous degradatiOn of this substrate. However, lactyl hydroxamate was isolated from a reaction mixture containing five prnoles of acrylyl Pa, 25 moles of tris acetate buffer pH 8. 0, and enzyme. Characterization Of phosphotransacetylase and COA transphorase. Phosphotransacetylase and COA transphorase were identified, purified and partially characterized in extracts of the organism. Phosphotrans- acetylase is specific for the COA moiety, its pH optimumis 8. 0 and it is strongly inhibited by sodium ions. Propionyl COA and butyryl CoA will substitute for acetyl COA in the arsenolysis assay to the extent. of 18% and 1. 5%, respectively of the rates observed with acetyl COA. 65. acetyl COA+ arsenate ————-> acetyl arsenate + COASH (XXIV) acetyl arsenate ———> arsenate + acetate (XXXV) Coenzyme A transphorase uses formate, propionate, butyrate, lactate, a-ketobutyrate, n-hydroxybutyrate and acrylate but not succinate as acceptors (Table 16) according to the following reaction: acetyl COA + X -——--— X-COA + acetate (XXV) Coenzyme A transphorase is specific for the COA moiety (Table 16), and its pH Optimum lies between 8. 0 and 8. 5. ' Characterization of several electron transport enzymes. Three enzymes concerned with electron transport in this organism have been identified, partially purified and characterized: acyl COA dehydrogenase (Table 14), lactic dehydrogenase (Table 13), and diaphorase (Table 15). All three enzymes readily reduce a number Of dyes, have no cofactor requirement, and possess many characteristics in common with comparable enzymes reported by earlier workers (51-53). Data pertaining to the dye specificity of these dehydrogenases is summarized in Table 17. The standard assay procedure employed in these studies contained 25 moles of phosphate buffer, sufficient substrate to attain maximum velocity, partially purified enzyme ( not the most active preparations in all cases) and the indicated amount of dye. The final volume of the incubation mixtures was adjusted to 0. 25 ml and the reaction started by addition of enzyme. The same enzyme preparations were employed for all the determinations. The purified lactic dehydrogenase is specific for D—lactate and requires no cofactors. The pH Optimum Of this enzyme is 7.0. Attempts to reverse the reaction, i. e. , reduce pyruvate, with either DPNH or reduced safranine have been unsuccessful. . No other enzymes possessing lactic dehydrogenase activity appear to be present in the crude extracts. 66 Table 16. Specificity of COA transphorase. a Acceptor Donor . Products Formed DL lactate acetyl-P04, COASH lactyl hydroxamate DL lactate acetyl-COA (Zumoles) lactyl COA (1.6 umoles) DL lactate acetyl-Pa (5 moles) none propionate acetyl-P04, COASH propionyl hydroxamate butyrate acetyl-P04, COASH butyryl hydroxamate acrylate acetyl-P04, COASH acrylyl hydroxamate acrylate acetyl COA (2 moles) acrylyl COA (1. 3 moles) succinate acetyl-P04, COASH none formats acetyl-P04, COASH formyl hydroxamate a-ketobutyrate acetyl-P04, COASH a-ketobutyryl hydroxamate o-hydroxybutyrate acetyl-P04, COASH d-hydroxybutyryl hydrox- amate aThe assay systems employed contained 50 moles Of Tris buffer, pH 8. 0, 50 moles of the acceptor, 1 unit Of purified COA transphorase, l. 0 pinoles of reduced glutathione and either 1 unit of phosphotrans- acetylase, 10 moles of acetyl-P04, 0. 1 prnoles of COASH and 25 moles of neutral hydroxylamine or acetyl COA or acrylyl Pa, as indicated. The hydroxamates were identified by paper chromography and the COA esters were assayed enzymatically. 67' Table 17. Dye reduction Of several dehydrogenase systems. Dehydrogenase and Substrate Dye Lactic Acyl COA Diaphorase "Clastic" D-lactate Propionyl Butyryl‘. DPNH Pyruvate COA COA Cth 2.5031 .35 2.0 26.5 25.0 2,6 DCP 1.45 3.50 4.9 157.0 140.0 INT .65 0 0.68 45.0 50.0 INT + DPN -—- 0.6 1.08 3.4 “— TTZ .15 0 0.4 13.0 5.0 MB 2.4 0 1.4 --- 10. 2 Ferricyanideb 2.2 --- 3.0 --- --- Safraninec --- 0.5 --- --- --- a'Arbitrary units based upon change in absorbancy per five minutes with respective dyes per 0. 1 mg of protein. b . . . . (Extremely inhibitory in the two cases tested. c . . Reverse direction. Acyl COA dehydrogenase is readily reversible and is specific for the COA moiety (Figure 14). It has a pH. optimum of 7. 5 and is unstable in solution unless the salt concentratiOn exceeds one-tenth molar. The enzyme reduces a number of dyes quite readily when butyryl-COA serves as the substrate but not with propionyl CoA as substrate (Table 17). The addition of DPN to the reaction mixture tends to stimulate propionyl COA dehydrogenase activity . 68 Figure 14. Acyl CoA dehydrogenase: reduction of acrylyl CoA. The incubation mixtures contained 150 umoles potassium phosphate, pH 7. 5, 50 units of acyl CoA dehydrogenase, 0.015% safranine and 2 moles of substrate as indicated. The reactions were run in Thunberg tubes which were gassed with nitrogen prior to reduction of the safranine with sodium hydrosulfite. ABSORBANCY, 540 m0 '0 .0 00 .0 m .0 4:. 0.2 69 ACYL COA DEHYD ROGEN ASE ACRYL YL CoA PROPIONYL 00.4 + —> + LEUCOSAFRAN/NE SAFRANINE 1 j o 3 A : ACRYLYL COA f) o E o 1 / O / A = ACRYLYL PANTETHEINE / Pa = LACTYL (309. O 0.: NO ENZYME 0 = NO SUBSTRATE MINUTES DISCUSSION The direct reductive pathway (Figure 2) has been postulated-for some time (85, 98), however, until now investigators have been unable to provide unequivocal proof for the existence of the pathway (94). Early workers succeeded in demonstrating all of the reactions except the dehydration mediated by lactyl CoA dehydrase (27). The primary aim of the present study, therefore, was to identify, purify and characterize this enzyme. Considerable progress towards this goal has been achieved through the development of an assay procedure for the enzyme (Figure 15). Figure 15. Lactyl CoA dehydrase assay system. acrylyl CoA > lactyl CoA i F acetate I CoA transphorase acetyl CoA acrylate ‘ L-lactate LDH > pyruvate DPN DPNH aga>1