Substrate and Inhibitor of Dioxygenases

Transcription

1 JOURNAL OF BACTEROLOGY, Dec. 1975, p Copyright American Society for Microbiology Vol. 124, No. 3 Printed in U.S.A. Alternative Routes of Aromatic Catabolism in Pseudomonas acidovorans and Pseudomonas putida: Gallic Acid as a Substrate and nhibitor of Dioxygenases VELTA L. SPARNNS AND STANLEY DAGLEY* Department of Biochemistry, College of Biological Sciences, University of Minnesota, St. Paul, Minnesota Received for publication 2 September 1975 When 3,4-dihydroxyphenylacetic acid (homoprotocatechuic acid) was added to Pseudomonas acidovorans growing at the expense of succinate, enzymes required for degrading homoprotocatechuate to pyruvate and succinate semialdehyde were strongly induced. These enzymes were effectively absent from cell extracts of the organism grown with 4-hydroxyphenylacetic acid, and this substrate was metabolized by the catabolic enzymes of the homogentisate pathway. Two separate ring-fission dioxygenases for 3,4,5-trihydroxybenzoic acid (gallic acid) were present in cell extracts of Pseudomonas putida when grown with syringic acid, and gallate was degraded by reactions associated with meta fission. One of the two gallate dioxygenases also attacked 3-O-methylgallic acid; the other, which did not, was induced when cells were exposed to gallate. This organism possessed ortho fission enzymes, including protocatechuate 3,4-dioxygenase (EC ) and cis,cis-carboxymuconate-lactonizing enzyme (EC ), after induction with 3,4-dihydroxybenzoic acid (protocatechuic acid). Gallate was a substrate for protocatechuate 3,4-dioxygenase, with a Vmax about 3% of that of protocatechuate and with an apparent Km slightly lower. Gallate was a powerful competitive inhibitor of protocatechuate oxidation. The continuous operation of the earth's carbon cycle depends in part upon the ability of microorganisms to degrade chemical structures that are not attacked by other living forms. Many of these structures contain benzene nuclei. Fission of the nucleus is catalyzed by microbial dioxygenases acting upon substrates that contain, as a minimum, two phenolic hydroxyl groups. These may be attached to adjacent carbon atoms, as in catechols, and the nucleus can then be cleaved by ortho or meta fission enzymes, or the hydroxyls may be placed across the nucleus in para positions, as in gentisic and homogentisic acids. For any one of the naturally occurring catechols, the enzymes of its pathway of catabolism are usually substrate specific. Thus, the meta fission enzymes for catechol (2, 4, 20) do not degrade 4-carboxycatechol (protocatechuic acid; 5, 7), and neither pathway includes enzymes that degrade 4-carboxymethylcatechol (homoprotocatechuic acid; 14, 21). The ortho fission pathways for catechol and protocatechuate (15, 16, 18) and the degradative routes for gentisate (12, 13) and homogentisate (1, 3) likewise are each separate and distinct: different suites of enzymes are used for the compounds named, and they catalyze routes that bear little or no chemical resemblance to those of meta fission. This does not imply that all these enzymes are absolutely specific for their substrates. Thus, those involved in the meta fission route for catechol, and in the degradation of gentisate, will tolerate methyl group substituents suitably located on the nucleus (4, 12); moreover, it is found that a small fraction of the total activity of pure catechol 1,2-oxygenase (EC ) towards 3-methylcatechol is expressed as meta, rather than ortho, fission (8). However, it may be noted that the cresols and xylenols, which are metabolized by the enzymes used for catechol or gentisate, are compounds of industrial rather than biological origin and, not surprisingly, their degradation does not call for enzymes that are highly specific for methyl-substituted phenolics. An organism is usually able to employ several different catabolic pathways for dihydric phenols, and it may then appear that a choice of routes is available for various growth substrates that contain a single phenolic substituent. n practice, the organism is committed to one 1374

2 VOL. 124, 1975 particular pathway, determined by the substrate specificity of the hydroxylase that introduces the second hydroxyl group required before the ring-fission reaction can be catalyzed. For example, Wheelis et al. (25) found 'that, when Pseudomonas testosteroni was grown with 3-hydroxybenzoate, enzymes for both the gentisate and protocatechuate pathways were induced to high levels. However, the protocatechuate route was the only one used, because the cells did not possess a 6-hydroxylase required to form gentisate, whereas a mutant lacking the 4-hydroxylase failed to grow with 3-hydroxybenzoate. Pseudomonas acidovorans takes the homogentisate pathway when grown with 4-hydroxyphenylacetic acid (11), whereas other species use the catabolic route through homoprotocatechuate (21). n the present work we show that P. acidovorans can also elaborate the enzymes for degrading homoprotocatechuate, but this strain does not appear to possess the 3-hydroxylase needed to take 4-hydroxyphenylacetate through this sequence. We also found that Pseudomonas putida degraded 4-hydroxybenzoate and protocatechuate by ortho fission reactions as anticipated (23); however, a section of the meta fission pathway for protocatechuate, as established for P. testosteroni (5), was utilized when P. putida degraded gallic acid. MATERALS AND METHODS Organisms and cell extracts. P. acidovorans (ATCC and British NCB 10013) was the strain AROMATC CATABOLSM N PSEUDOMONAS 1375 cco C)H CH2 O 4-HPC ()CH HPA CH2 02 O used in a previous investigation (11) of the properties and metabolic function of 4-hydroxyphenylacetic acid 1-hydroxylase (4-HPA 1-hydroxylase). P. putida, which was isolated by elective culture in media containing syringic acid, has been used to study the metabolism of syringate and gallate (24). Cultures of either organism were grown with aeration at 30 C in the basal medium of Sparnins et al. (21), with additions of the following organic acids to serve as single carbon sources (grams per liter): succinic, 1.0; 4-hydroxyphenylacetic (4-HPA), 0.5; 4-hydroxybenzoic, 0.5; and syringic, 0.5. Cell extracts were prepared in various buffer solutions from frozen cell pastes crushed in a Hughes bacterial press (6). The buffers used were: for extracts containing enzymes of the homoprotocatechuate (HPC) pathway in P. acidovorans (Fig. 1), 0.1 M K+-Na+ phosphate, ph 7.0 (21); and for those containing enzymes of the homogentisate pathway (Fig. 1) in P. acidovorans, 0.1 M K+-Na+ phosphate, ph 7.2, plus 1 mm dithioerythritol, 5 mm MgSO, and 10 mm flavin adenine dinucleotide (11). Cell extracts of P. putida were prepared in 80 mm phosphate buffer, ph 7.0, with additions of 20% glycerol and 5 mm L-cysteine hydrochloride to stabilize dioxygenases (24). The protein contents of extracts were determined by a modification of the biuret reaction (9). nduction of enzymes. The turbidity of a culture (1 liter in a shaken 2-liter Erlenmeyer flask) was monitored on a Beckman DBG spectrophotometer during exponential growth with succinate until an absorbancy of 0.3 at 540 nm was reached, when the inducing substrate was added to give a final concentration of 0.3 g per liter. Shaking was continued at 30 C for 3 h in the case of gallic and protocatechuic acids and for 6.5 h with HPC. During this time, measurements of turbidity indicated that all the succinate was consumed when the absorbancy HOOC FH2 HOOC CH2 CO H X + CO (2) CH 3 CH2 CH2 CM2 C-M2 CM2. CHO NAD COO Co2 CHH2 C2 (3) N (4) N %.N (5) + CM2 CM3 CUO C-0 HPC FG. 1. Catabolic reaction sequences for 4-HPA and HPC in Pseudomonas acidovorans. Numbers below arrows indicate enzymes that were assayed (see also Table 1).

3 1376 SPARNNS AND DAGLEY reached 0.6, whereas the added substrate provided an additional increase of about 0.3. The cells were centrifuged and washed with buffer, and extracts were prepared. Assays of enzymatic activities. Homoprotocatechuate 2,3-dioxygenase (enzyme 3, Fig. 1) was assayed by following the increase of absorbancy at 380 nm due to appearance of the ring-fission product (e = 38,000 at ph 7.5). n the absence of added nicotinamide adenine dinucleotide (NAD), removal of the ring-fission product by its dehydrogenase was very slow, but, when measurable, a correction for this decrease was made. Cuvettes contained, in 1.0 ml, 100 Mmol of 0.1 M K+-Na+ phosphate buffer, ph 7.5, and 0.02 Mmol of HPC; reactions were started by adding sufficient cell extract (0.115 to 1.45 mg of protein) to provide a convenient rate of increase in absorbancy. The NAD-dependent dehydrogenase for the ring-fission product (enzyme 4) and 4-hydroxy- 2-ketopimelate aldolase (enzyme 5) were assayed as described previously (21). Homogentisate 1,2-dioxygenase (enzyme 1, Fig. 1) was assayed by observing the increase in absorbancy at 330 nm due to appearance of maleylacetoacetate ( ( = 10,800 at ph 7.2). Cuvettes (1.0 ml) contained 100,pmol of phosphate buffer, ph 7.2, 0.05 gmol of homogentisate, and cell extract. For maleylacetoacetate isomerase (enzyme 2), cuvettes contained the same mixture and, after 6 min, when the absorbancy had ceased to increase, 0.25 Mmol of reduced glutathione was added to activate the isomerase, and the rate of decrease in absorbancy at 330 nm was measured. Dioxygenase activities against 3-O-methylgallic acid (enzyme 1, Fig. 2), gallic acid (enzyme 2), and PC (enzyme 3) were measured with an oxygen electrode operated as described previously (5, 21). The substrates (0.1 to 0.3 gmol) were added to 1.5 ml of 0.05 M tris(hydroxymethyl)aminomethane-hydrochloride buffer, ph 8.0, and reactions were started by adding cell extract. The oxygen electrode was also used for enzyme kinetic studies with pure, crystalline protocatechuate 3,4-dioxygenase. cis-cis-carboxymuconate-lactonizing enzyme (enzyme 4, Fig. 2) was assayed by the method of Ornston (15). The production of pyruvate from gallate by cell extracts was determined essentially as previously described (21) by following the decrease in absorbance at 340 nm due to oxidation of reduced NAD. The sources of chemical compounds were those given in previous publications (11, 21, 24). Ultrogel AcA was trom LX nstruments, nc. 6-Carboxycis,cis-muconic acid and pure, crystalline protocatechuate 3,4-dioxygenase were, respectively, the generous gifts of Peter J. Chapman and John D. Lipscomb. RESULTS Metabolism of 4-hydroxyphenylacetic acid and homoprotocatechuic acid by P. acidovorans. Rates of oxidation, using 5 gmol of each substrate and 4 mg (dry weight) of cells, were measured and compared for P. acidovorans when (i) grown with 4-HPA and (ii) induced to HPC as described above. n experiment (i) the J. BACTEROL. rates (microliters of 02 per min) were: 4-HPA, 7.1 and HPC, 1.0. n (ii) rates were: 4-HPA, 1.5 and HPC, 9.0. The specific activities of enzymes present in extracts of these cells, and also succinate-grown cells, were measured (Table 1). Comparison shows that activities of enzymes 3, 4, and 5 (Fig. 1) were much higher in extracts from cells exposed to HPC than in those from cells grown with succinate or 4-HPA as sole carbon sources. However, although induction to HPC did not increase levels of enzymes 1 and 2 (Fig. 1), these enzymes were strongly induced by 4-HPA in accordance with previous findings that this organism uses the homogentisate pathway for degrading 4-HPA (11). Metabolism of protocatechuic and gailic acids by P. putida. Protocatechuate and gallate were added separately to cultures of P. putida during exponential growth with succinate, and enzymes present in cell extracts that catalyzed reactions shown in Fig. 2 were assayed. Rates of formation of pyruvate from gallate were also determined, and these various enzymatic activities were compared with those present in cells grown, respectively, with succinic, 4-hydroxybenzoic, and syringic acids as sole carbon sources (Table 2). 4-Hydroxybenzoate-grown cells contained at least 60 times more protocatechuic 3,4-dioxygenase (enzyme 3, Fig. 2) and 13 times more cis,cis-carboxymuconate lactonizing enzyme (enzyme 4) than those grown with succinic or syringic acids, or adapted to gallic acid. These enzymes were also induced, though less strongly, by exposure to protocatechuate for the period of time used. Only cells grown with syringic acid or adapted TABLE 1. Enzyme assayed nduction of enzymes in Pseudomonas acidovorans Growth substrate (sp act, nmol/min per mg of protein) Suc- Succinate cinate 4-HPA HPC Homogentisate 2.3 < ,2-dioxygenase (1)a somerase (2) HPC 2, dioxygenase (3) Dehydrogenase (4) Aldolase (5) anumbers in parentheses denote enzymes identified in Fig. 1.

4 VOL. 124, 1975 AROMATC CATABOLSM N PSEUDOMONAS 1377 Syringic a. Gallic a. CO Protocatechuic a. CH30 OCH3 H l (2)102 /* // (3) 02 CH3O0$} HH NADCOO- C ('1,02 4K,H20 HOOC H CH30 C. O HOOC COC ~~HOOC CH3 CO CO CO 2 CO <'A CH2 CH2 COOCH3 2 Pyruvic a. C (44 C41 CO 4O l CO co C~)O HOOC / Acetyl CoA + Succinate FG. 2. Catabolic reaction sequences for syringic, gallic, and protocatechuic acids in Pseudomonas putida. Numbers beside arrows refer to enzymes that were assayed (see also Table 2). The dotted arrows show reactions catalyzed by Pseudomonas testosteroni but not by Pseudomonas acidovorans. TABLE 2. nduction of enzymes in Pseudomonas putida Growth substrate (sp act, nmollmin per mg of protein) Enzyme assayeda Succinate Succinate 4-Hydroxy- Succinate plus gallate Srnae benzoate plus PCb 3-O-Me-gallate dioxygenase (1) Gallate ,030 50c 17c dioxygenase (2) Pyruvate from gallate < 1 < 1 Protocatechuate , ,4-dioxygenase (3) CM-lactonizing , enzyme (4) a Numbers in parentheses denote enzymes identified in Fig. 2. CM, b cis,cis-carboxy-muconate. PC, Protocatechuic acid. c Activity due to protocatechuate 3,4-dioxygenase (see text).

5 1378 SPARNNS AND DAGLEY to gallic acid gave extracts that degraded gallate to pyruvate. Syringate-grown cells contained almost three times the gallate dioxygenase activity (enzyme 2, Fig. 2) of those grown with succinate and adapted to gallate, and more than 40 times that of cells grown with any of the other substrates used. t was also evident that syringate-grown P. putida contained an oxygenase for 3-O-methylgallate (enzyme 1) that was not detected in cells adapted to gallate. Gallic acid as substrate and inhibitor of protocatechuate 3,4-dioxygenase. Extracts of P. putida grown with 4-hydroxybenzoate were found to oxidize gallate at a low but significant rate (Table 2). However, no pyruvate was produced, and a ring-fission product accumulated that showed the same spectral characteristics as the compound formed by the action of protocatechuate 4,5-dioxygenase on gallate (24). The oxygen electrode was used to measure activities against gallate, over a range of ph values, for two cells extracts, the first obtained from cells grown with 4-hydroxybenzoate and the second from cells induced to gallate. The ph profiles obtained were entirely different (Fig. 3); that obtained with gallate as substrate for the first extract resembled the effect of ph on protocatechuate oxidation found by Stanier and ngraham (22) for PC 3,4-dioxygenase. These results suggested that oxidation of gallate by extracts of 4-hydroxybenzoate-grown cells was due to PC 3,4-dioxygenase, a conclusion supported by observations made with the pure, crystalline enzyme. Thus, the spectra of the ring-fission products using either pure enzyme or extract were identical, and kinetic studies using the oxygen electrode provided the following data. Apparent Km values were 1.7 x 10- M gallate, 2.3 x 10-' M protocatechuate; and the Vmax (micromoles per minute per milligram of enzyme) was 0.4 for gallate and 12.0 for protocatechuate. Gallate was a competitive inhibitor of protocatechuate oxidation (Fig. 4) with a K, of 5 x 10-6 M gallate. This compound is therefore a strong inhibitor, but the present data are not sufficient to establish that gallate is bound at the same site as protocatechuate on this two-substrate enzyme. Gallate dioxygenases induced in P. putida by gallic and syringic acids. Cells grown with syringic acid contained two enzymes that oxidized gallate; these were separated by fractionation. The cell extract was brought to 60% saturation with ammonium sulfate (previously neutralized with ammonia), and the precipitate was removed by centrifuging and redissolved in a buffer containing 0.08 M K+-Na+ phosphate, c E 0 E cq C a- D cm ph FG. 3. Effects of ph on gallic acid dioxygenase activities for two cell extracts of Pseudomonas putida. The cells were induced to metabolize 4-hydroxybenzoic acid (0) or gallic acid (0). Enzymatic activities, against gallic acid in each case, were measured with an oxygen electrode. x 10-4(M-1) J. BACTEROL. FG. 4. Competitive inhibition of protocatechuate 3,4-dioxygenase by gallic acid. Concentrations of inhibitor: 13.3 AM (0); 26.7 AM (A); 53.3 gm (A); 66.7 AM (0); and no inhibitor (0). One unit of activity is the amount of enzyme that catalyzes the uptake of lmol of oxygen per min. ph 7, 20% glycerol, and 5 mm L-cysteine hydrochloride. The solution was then applied to a column of Ultrogel AcA 34 (1.6 by 108 cm) connected in series to a similar column (1.6 by 92 cm), both of which had been equilibrated with the buffer in which the precipitate was dissolved. Fractions (4.5 ml) were eluted with this buffer and assayed for activity against gallic acid and 3-O-methylgallic acid (Fig. 5).

6 VOL. 124, 1975 AROMATC CATABOLSM N PSEUDOMONAS 1379 E FRACTON NUMBER FG. 5. Separation of gallate dioxygenase activities present in an extract of Pseudomonas putida grown with syringic acid. Fractions from a column of Ultrogel AcA 34 were assayed for activity against gallic acid (x) and 3-0-methylgallic acid (A). Protein eluted in tubes catalyzed oxidation of both substrates; that in tubes oxidized gallate alone. The enzymes also differed in heat stability; thus, the fraction that catalyzed only the oxidation of gallate retained 92% of its activity when held at 53 C for 3 min, whereas the activity of the second fraction against 3-O-methylgallic acid was virtually abolished (<1% remaining) after this treatment. The activity of both dioxygenases was increased by adding ferrous ammonium sulfate to reaction mixtures, but this compound was not used in assays since addition of Fe'+ ions significantly increased uptake of 0, by extracts in the absence of substrates. DSCUSSON We have already referred to the existence of several separate and distinct pathways of aromatic catabolism, initiated by specific dioxygenases that cleave the benzene nuclei of dihydric phenols. There is mounting evidence that any one strain, taken from a range of species, may be capable of utilizing most of these pathways, or at least segments of them and that the particular catabolic route selected for a monohydric phenol is often determined by the specificity of the monohydroxylase available for that substrate. n the present investigation we found that P. acidovoran possesses the enzymes of the homoprotocatechuate pathway (21) which certain other organisms use when they grow with 4-hydroxyphenylacetate. However, this species does not use that particular catabolic sequence for the same growth substrate; instead, hydroxylation at C-1 occurs to give homogentisate. We suggest that P. acidovorans, which grows readily with L-tyrosine as a carbon source, has possessed the enzymes for utilizing this amino acid from early times, and that the enzyme 4- hydroxyphenylacetic acid 1-hydroxylase was a later acquisition which, by forming homogentsate, enabled the organism to use this pre-existing pathway for the purpose of catabolizing 4- hydroxyphenylacetic acid. We found that P. putida possesses enzymes that catalyze ortho and meta fission pathways for closely related phenolic compounds. Thus, we confirm that dioxygenases present in syringate-grown cells (24) attack gallate to give a metabolite that is also formed by meta fission reactions when P. testosteroni degrades protocatechuate to pyruvate (Fig. 2). Previous work with P. putida ascribed the oxidation of gallate to a single enzyme (24), but we found that two gallate dioxygenases were present in extracts of syringate-grown P. putida: one of these enzymes also oxidized 3-0-methygallic acid but the other, which could be induced specifically by exposing cells to gallate, did not attack the methyl ether. Neither of these dioxygenases was found to attack protocatechuate and, when P. putida was grown with 4-hydroxybenzoate, protocatechuate was degraded by ortho fission. By contrast, P. testosteroni degrades protocatechuate by meta fission. The enzyme is designated protocatechuate 4,5-dioxygenase (EC ) and the substrate named was the first one to be investigated (5), but cells grown with 4-hydroxybenzoate, and extracts made from them, oxidize protocatechuate and gallate equally well (24). Therefore, it is conceivable that this enzyme was evolved to metabolize gallate in the first instance, and then, by recruiting one enzyme (an aldehyde dehydrogenase for the ring-fission product), protocatechuate was brought into the meta fission pathway that functioned for gallate (Fig. 2). The latter compound is abundant in nature, either free or combined, and we find that gallate, and other 3,4,5-trihydroxy-substituted aromatic acids or their methyl esters, are readily utilized as carbon sources by various soil organismas. These compounds may be formed in vast amounts during fungal degradation of lignin, the second most abundant polymer in nature, although their rapid utilization as microbial carbon sources would prevent any substantial accumulation. The chemical bonds between units of this polymer are not readily subject to hydrolysis, as are those for other biopolymers, and bond cleavage most probably entails attack by oxygenases, probably acting extracellularly. The monomers thus liberated would be polyphenolic

7 1380 SPARNNS AND DAGLEY aromatic acids, with hydroxyl groups most commonly occupying positions 3,4, and 5 of the nucleus relative to a side-chain substituent. t may be noted that ring cleavage of a 3,4,5-trihydroxy aromatic acid such as gallate may be designated as either intradiol or extradiol. However, subsequent reactions of the catabolic pathway must be similar in type to those encountered in the meta fission route for dihydric phenols since a pyruvate residue inevitably arises when the benzene ring of such a polyphenol is cleaved between either adjacent pair of the three hydroxyl groups. Our observations are pertinent to speculations which propose that many catabolic pathways may have evolved sequentially, following attack by dioxygenases upon simple polyphenolic compounds that are formed by the breakdown of lignins and other large molecules of biological origin. Thus, gallate is attacked by the intradiol enzyme, protocatechuate 3,4-dioxygenase, whose only known present function is to introduce its substrate into a,8-ketoadipate catabolic pathway. n the same organism, P. putida, the catabolism of gallate itself can be initiated by two separately inducible dioxygenases which do not attack protocatechuate. We suggest, therefore, that in P. putida a primitive gallate dioxygenase, when modified by mutations, might have provided an effective intradiol dioxygenase for protocatechuate. By contrast, P. testosteroni already possessed a gallate dioxygenase that rapidly attacked protocatechuate, and the effective degradation of this substrate could then be accomplished by other means, namely by recruiting a suitable dehydrogenase for the product of ring fission. Enzymes for the,-ketoadipate pathway can be induced in P. testosteroni and P. acidovorans, but Ornston and Ornston (17) have argued convincingly that they serve the purpose of metabolizing,-carboxy-cis,cis-muconate found in nature. Similarly, P. acidovorans synthesizes inducibly all the enzymes needed to convert catechol intof-ketoadipate, and this pathway is likewise believed to operate for utilizing muconic acids, rather than aromatic acids (19). The most important natural precursor of catechol is usually assumed to be benzoic acid, which is degraded by enzymes of the gentisate pathway in the strain of P. acidovorans investigated (25). The significance of this organism's capacity to metabolize catechol in addition to muconates therefore is not evident, although catechol, like other dihydric and trihydric phenolic compounds and muconates, may be formed from natural products by other microorganisms but not utilized by them instantly. Thus, Klebsiella aerogenes forms catechol by decarboxylating protocatechuate (10). ACKNOWLEDGMENTS Generous gifts of materials were made by John D. Lipsomb and Peter J. Chapman, with whom we also had many fruitful discussions. Skilled technical assistance in assays and chromatography of enzymes was provided by.john Almquist. This investigation was supported by Public Health Service grant ES A from the National nstitute of Environmental Health Sciences. LTERATURE CTD J. BACTROL. 1. Adachi, K., Y. wayama, H. Tanioka, and Y. Takeda Purification and properties of homogentisate oxygenase from Pseudomonas fluorescens. Biochim. Biophys. Acta 118: Bayly, R. C., and S. Dagley Oxoenoic acids as metabolites in the bacterial degradation of catechols. Biochem. J. 111: Chapman, P. J., and S. Dagley Oxidation of homogentisic acid by cell-free extracts of a vibrio. J. Gen. Microbiol. 28: Collinsworth, W. L., P. J. Chapman, and S. Dagley Stereo8pecific enzymes in the degradation of aromatic compounds by Pseudomonas putida. J. Bacteriol. 113: Dagley, S., P. J. Geary, and J. M. Wood The metabolism of protocatechuate by Pseudomonas testosteroni. Biochem. J. 101: Dagley, S., and D. T. Gibson The bacterial degradation of catechol. Biochem. J. 96: Dennis, D. A., P. J. Chapman, and S. Dagley Degradation of protocatechuate in Pseudomonas testosteroni by a pathway involving oxidation of the product of meta-fission. J. Bacteriol. 113: Fujiwara, M., L. A. Golovleva, Y. Saeki, M. Nozaki, and 0. Hayaishi Extradiol cleavage of 3-substituted catechols by an intradiol dioxygenase, pyrocatechase, from a pseudomonad. J. Biol. Chem. 250: Gornall, A. G., C. J. Bardawill, and M. M. David Determination of serum proteins by means of the biuret reaction. J. Biol. Chem. 177: Grant, D. J. W Kinetic aspects of the growth of Klebsiella aerogenes with some benzenoid carbon sources. J. Gen. Microbiol. 46: Hareland, W. A., R. L. Crawford, P. J. Chapman, and S. Dagley Metabolic function and properties of 4-hydroxyphenylacetic acid 1-hydroxylase from Pseudomonas acidovorans. J. Bacteriol. 121: Hopper, D. J., P. J. Chapman, and S. Dagley The enzymic degradation of alkyl-substituted gentisates, maleates and malates. Biochem. J. 122: Lack, L The enzymic oxidation of gentisic acid. Biochim. Biophys. Acta 34: Leung, P.-T., P. J. Chapman, and S. Dagley Purification and properties of 4-hydroxy-2-ketopimelate aldolase from Acinetobacter. J. Bacteriol. 120: Ornston, N. L The conversion of catechol and protocatechuate to B-ketoadipate by Pseudomonaas putida. H. Enzymes of the protocatechuate pathway. J. Biol. Chem. 241: Ornston, N. L The conversion of catechol and protocatechuate to,-ketoadipate by Pseudomonas putida.. Enzymes of the catechol pathway. J. Biol. Chem. 241: Ornston, M. K., and L. N. Ornston The regulation of the,-ketoadipate pathway in Pseudomonas acidovorans and Pseudomonas testosteroni. J. Gen. Microbiol. 73: Omston, L. N., and R. Y. Stainier The conversion of catechol and protocatechuate to,-ketoadipate by Pseudomonas putida.. Biochemistry. J. Biol. Chem. 241:

8 VOL. 124, 1975 AROMATC CATABOLSM N PSEUDOMONAS Robert-Gero, M., M. Poiret, and R. Y. Stanier The function of the s-ketoadipate pathway in Pseudomonas acidovorans. J. Gen. Microbiol. 57: Sala-Trepat, J. M., K. Murray, and P. A. Williams The metabolic divergence in the meta cleavage of catechols by Pseudomonas putida NCB Physiological significance and evolutionary implication. Eur. J. Biochem. 28: Sparnins, V. L., P. J. Chapman, and S. Dagley Bacterial degradation of 4-hydroxyphenylacetic acid and homoprotocatechuic acid. J. Bacteriol. 120: Stanier, R. Y., and J. L. ngraham Protocatechuic acid oxidase. J. Biol. Chem. 210: Stanier, R. Y., N. J. Palleroni, and M. Doudoroff The aerobic pseudomonads: a taxonomic study. J. Gen. Microbiol. 43: Tack, B. F., P. J. Chapman, and S. Dagley Metabolism of gallic and syringic acids by Pseudomonas putida. J. Biol. Chem. 247: Wheelis, M. L., N. J. Palleroni, and R. Y. Stanier The metabolism of aromatic acids by Pseudomonas testosteroni and P. acidovorans. Arch. Mikrobiol. 59: