THE JOURNAL OF BIOLOGICAL CHEMISTRYVol. 241, No. 16, Issue of August 25, pp. 3800-3810, 1966

Printed in U.S.A.

The Conversion of Catechol and Protocatechuate to-Ketoadipate by Pseudomonas putida

IV. REGULATION*

(Received for publication, March 4, 1966)

L. N. ORNSTONJ

From the Department of Bacteriology and Immunology, University of California, Berkeley, California 94720

SUMMARY

Study of the regulation of the syntheses of enzymes ofthe catechol and protocatechuate pathways in Pseudomonasputida has shown that two groups of enzymes are subject tocoordinate control. cis, cis-Muconate-lactonizing enzymeand muconolactone isomerase, which are uniquely asso-ciated with the catechol pathway, constitute the first co-ordinate block of enzymes. The synthesis of these enzymes,as well as that of catechol oxygenase (which is regulatedindependently), seems to be induced by cis, cis-muconate.

The second coordinate block of enzymes comprises 3-carboxy-cis, cis-muconate-lactonizing enzyme and y-car-boxymuconolactone decarboxylase, which are uniquely asso-ciated with the protocatechuate pathway, and -ketoadipateenol-lactone hydrolase, which is functional in both the proto-catechuate and the catechol pathways. This group ofenzymes seems to be induced by -ketoadipate or -keto-adipyl coenzyme A.

Moraxella woffii, which degrades protocatechuate andcatechol through identical step-reactions, regulates thesynthesis of the enzymes mediating these conversions by adifferent mechanism.

Benzoate and p-hydroxybenzoate can be used as sole sources ofcarbon and energy by many species of aerobic bacteria belongingto several different genera. Although other pathways for theirmetabolism occur in some bacteria, the most widespread mode ofbacterial attack on these two compounds involves conversion ofthe aromatic nucleus to f-ketoadipate, which is cleaved after ac-tivation to yield acetyl coenzyme A and succinate. The initialsteps are mediated by specific hydroxylases, which convert ben-zoate to catechol and p-hydroxybenzoate to protocatechuate. Adetailed analysis of the conversions of catechol and protocatechu-ate to 0-ketoadipate by Pseudomonas putida (1) has shown that

* This investigation was supported in part by Research GrantAI-1808 from the National Institutes of Health, United StatesPublic Health Service, to M. Doudoroff and R. Y. Stanier.

tNational Science Foundation Predoctoral Fellow, 1961 to1965. Present address, Department of Biochemistry, Universityof Leicester, Leicester, England.

the reactions proceed through two series of chemically analogousintermediates, metabolic convergence occurring with the forma-tion of the immediate precursor of f-ketoadipate, -ketoadipateenol-lactone. Two specific sets of enzymes mediate the respec-tive conversions of catechol and protocatechuate to this lactone;a common enzyme, 3-ketoadipate enol-lactone hydrolase, con-verts the lactone to -ketoadipate.

Specific assays have been developed for all of the enzymesoperative in the conversions of catechol and protocatechuate to0-ketoadipate (2, 3). Without exception these enzymes are in-ducible, so that the existence of these specific assays has made pos-sible a detailed study of the regulatory mechanisms controlling.their synthesis, described in the present paper. The regulationof the synthesis of some of these enzymes by other species ofbacteria that possess the ability to convert benzoate and p-hy-droxybenzoate to 0-ketoadipate has also been examined, and the.results will be compared with those obtained for P. putida.

EXPERIMENTAL PROCEDURE

Biological Materials-P. putida A.3.12 (ATCC 12633) was theorganism used for most of the experiments to be reported. Mu-tants of this strain were prepared by methods described below,except for mutant A202, which was obtained from Dr. J. Mandel-stam. P. putida Cl-A (ATCC 17452), a naturally occurring strainwhich possesses all enzymes necessary for utilization of p-hy-droxybenzoate except -carboxy-cis, cis-muconate-lactonizingenzyme, was obtained from Dr. I. C. Gunsalus. Other bacterialstrains used were Moraxella woffli (Vibrio 01) ATCC 11171;Pseudomonas multivorans ATCC 17759; Pseudomonas aeruginosaATCC 17503; and Hydrogenomonas eutropha ATCC 17697.

The composition of media and the conditions of growth aredescribed elsewhere (1).

Isolation and Characterization of llutants-Two general classesof mutants were isolated. The first class has lost the ability tosynthesize one or more of the enzymes operative in the catecholand protocateehuate pathways. Such mutants will be termedblocked mutants. The second class has acquired the ability (notpossessed by the wild type) to grow exponentially at the expensOof eis,cis-muconate. Such mutants will be termed permeabilitymutants, because the mutational change apparently involves saalteration of the cell membrane which permits the dicarboxyi0acid to enter the cell.

Blocked mutants of P. putida A.3.12 were obtained by treat

3800

by guest on April 11, 2018

http://ww

w.jbc.org/

Dow

nloaded from

TABLE I

Influence of various growth substrates on levels of enzymes of catechol and protocatechuate pathways in extracts fromwild type P. putida A..12

Cultures were grown at the expense of 20 mM acetate, 20 mM succinate, 10 mM p-hydroxybenzoate, 10 mM benzoate, or 10 mM 3-keto-adipate. The values given for cells grown at the expense of -ketoadipate were obtained from a single culture. Other values rep-resent the average of determinations on extracts prepared from at least three different cultures grown on the indicated substrate.

Enzyme

Catechol oxygenasecis,cis-Muconate-lactonizing enzymeMuconolactone isomeraseProtocateehuate oxygenasep-Carboxy-cis, cis-muconate-lactonizing

enzymey-Carboxymuconolactone decarboxylase#-Ketoadipate enol-lactone hydrolase

Physiological role

Catechol pathwayCatechol pathwayCatechol pathwayProtocatechuate pathwayProtocatechuate pathway

Protocatechuate pathwayBoth pathways

Specific activities in extracts from cells grown with

Acetate or e todipatsuccinate p-Hydroxybenzoate Benzoate Ketoadipate

nits/mg protein

<0.0002 0.005 (0.005-0.006)- 1.09<0.0002 <0.0002 0.39<0.02 <0.02 2.200.004 2.63 (2.42-2.85)0.02 0.95 (0.89-1.03) 0.68

0.05 2.82 (2.60-3.07) 2.020.03 1.49 (1.33-1.62) 1.03

-Numbers in parentheses represent range of values observed.

meant of the wild type withN-methyl-N'-nitro-N-nitrosoguanidine(nitrosoguanidine, purchased from Aldrich). Cultures growingexponentially with suceinate were harvested and resuspended toa final concentration of 1.0 X 108 viable cells per ml in a solutionof 0.005% nitrosoguanidine in 10 mM sodium citrate buffer, pH 5.5.The suspension was shaken at 300 for 1 hour. This treatmentkilled 99.9% of the cells. The suspension was diluted with 100mM disodium potassium phosphate buffer, pH 6.8, to a concentra-tion of approximately 103 viable cells per ml, and aliquots of 0.1 mlwere spread on plates of the basal synthetic agar medium con-taining succinate (1 mM) and either benzoate or p-hydroxyben-zoate (10 maM). Surviving cells of the wild type formed largecolonies on this medium; mutant cells, which were unable to uti-lize the aromatic substrate provided, formed small colonies.

After 2 days of incubation at 30 °, numerous small colonies werepicked and patched on plates containing 10 mM succinate. Thepatched plates were incubated for 30 hours at 30° and then repli-cated by the method of Lederberg and Lederberg (4) on threeplates, containing succinate, p-hydroxybenzoate, and benzoate,respectively, as sole carbon and energy source. Strains whichcould grow at the expense of succinate, but not of one or both ofthe aromatic acids, were restreaked on 1% yeast extract plates.A well isolated colony was picked, checked for identity as P.putida, and re-examined for the stability of the mutationalchange. Mutants were maintained on 1% yeast agar slants.

The site of the enzymic lesion in blocked mutants was deter-mined as follows. Cells were grown for at least three generationsat the expense of 10 mM glucose in the presence of either 10 mMbenzoate or 5 mM p-hydroxybenzoate, and were harvested andextracted. The extracts were then assayed for the enzymes ofthe eatechol and protocatechuate pathways.

Spontaneous revertants were selected by inoculating about 109viable mutant cells into basal medium containing either 5 mMbenzoate or 5 mM p-hydroxybenzoate. The culture was in-cubated with mechanical agitation at 30° until (after 2 or 3 days)a marked increase in turbidity was observed. It was thenStreaked on a plate of the homologous agar medium and an in-dividual colony was selected. Revertants were maintained on1% yeast extract slants.

Permeability mutants were obtained without mutagenesis byspreading about 108 wild type cells of P. putida A.3.12 on platesof basal agar medium containing 10 mM cis,cis-muconate. Twoor three colonies appeared on most plates after several days.One colony was picked from each plate and purified by restreak-ing on a plate of the same medium. The mutant stocks weremaintained on slants of cis, cis-muconate agar in order to counter-select possible revertants; however, they could be transferredseveral times on 1% yeast agar without losing the ability to growat the expense of cis,cis-muconate.

Enzyme Assays-Cell-free extracts were prepared and enzymeassays were performed by procedures described previously (1-3).Enzyme levels in crude extracts are always expressed as unitsper mg of protein, a unit of enzyme being defined as the amountnecessary to cause disappearance of 1.0 molee of substrate permin under the conditions of assay. Protein concentrations weredetermined with biuret reagent (5).

RESULTS

Patterns of Induction of Enzymes of Catechol and ProtocatechuatePathways in Wild Type P. putida A.3.12-Although P. putidaA.3.12 grows well at the expense of benzoate or p-hydroxyben-zoate, with generation times at 300 of 50 and 75 min, respectively,few of the subsequent intermediates can support growth of thewild type. Some of the intermediates are too unstable chemi-cally to serve as growth substrates (e.g. -carboxy-cis ,cis-mucon-

ate, y-carboxymuconolactone). Others are toxic (e.g. catechol),or do not readily enter the cell (e.g. cis,cis-muconate). How-ever, -ketoadipate can support slow growth of the wild type(with a generation time of 180 min at 30°), and is sufficientlystable chemically to be used in growth experiments. Accord-ingly, inductive patterns with respect to the enzymes of the cate-chol and protocatechuate pathways were determined on extractsof the wild type after growth at the expense of benzoate, p-hy-droxybenzoate, and f-ketoadipate. Basal levels of all these en-zymes were determined on extracts of cells grown with acetateand with succinate, which support generation times at 30° of 200and 45 min, respectively. The results are shown in Table I.

The basal levels of activity after growth with succinate and

(0.99-1.29)(0.38-0.40)(2.03-2.23)0.03(0.64-0).75)

(1.85-2.22)(0.89-1.16)

<0.002<0.002<0.02

0.80

2.481.58

ssue of August 25, 1966 L. N. Ornston 3801

<

by guest on April 11, 2018

http://ww

w.jbc.org/

Dow

nloaded from

Catechol and Protocatechuate Pathways in P. putida. IV Vol. 241, No. 16

2.0

c-C

VV -

) D0 0

o 0)1.0o -_

U -

a) C

U

0

2)

0

X0

0

0) Cb

C ~0 0)

o -E

0 :0O C:

0U_A

C0 0)

E N*C

0)

C

'0

00

0UZ

Q)

2CL

a

00)C(E

D

02

04

60 05

109 8 071All

J3-Ketoodipate Enol-Loctone HydroloseUnits/mg protein

FIG. 1. Coordinate regulation of the syntheses of 3-carboxy-cis,cis-muconate-lactonizing enzyme, -carboxymuconolactonedecarboxylase, and -ketoadipate enol-lactone hydrolase. En-zyme levels were determined in extracts of wild type P. putidawhich had been grown for at least seven generations in mineralmedium containing the following carbon sources: 1, 10 mm p-hy-droxybenzoate; 2, 10 mM benzoate; 3, 10 mM p-hydroxybenzoateand 10 mm glucose; 4, 10 mu benzoate and 10 mM glucose; 5, 10mM benzoate and 10 mM succinate; 6, 10 mm benzoate and 30 mMsuccinate; 7, 10 mm benzoate and 30 mm glucose; 8, 10 mma p-hy-droxybenzoate and 10 mm succinate; 9, 10 mm succinate; 10, 10mm glucose.

with acetate were essentially identical, and are shown in a singlecolumn of Table I. The three enzymes uniquely associated withthe catechol pathway (catechol oxygenase, cis,cis-muconate-lactonizing enzyme, and muconolactone isomerase) were unde-tectable by the assay methods used; the values given for themtherefore represent the maximal possible levels. The three en-zymes uniquely associated with the protocatechuate pathway(protocatechuate oxygenase, /3-carboxy-cis,cis-muconate-lacto-nizing enzyme, and y-carboxymuconolactone decarboxylase) andthe first enzyme common to both pathways -ketoadipate enol-

lactone hydrolase) were all present at very low but detectablelevels in acetate- and succinate-grown cells. Growth at the ex.pense of p-hydroxybenzoate elicits high levels of activity of thefour enzymes required for the conversion of protocatechuate tof3-ketoadipate. The activity of protocatechuate oxygenase in.creases 650-fold over the basal level; the activities of the otherthree enzymes, about 50-fold. There is no measurable inductionof cis, cis-muconate-lactonizing enzyme or muconolactone isomer-ase; the slightly increased level of catechol oxygenase activitymay be attributed to nonspecific catalysis by protocatechuate.oxygenase (2). Accordingly, the inductive response resultingfrom growth with p-hydroxybenzoate is physiologically specific.

Growth at the expense of benzoate elicits high levels of ac-tivity of the four enzymes required for the conversion of catecholto -ketoadipate. The activity of catechol oxygenase increasesat least 5000-fold over the basal level, the activity of cis,cis-muconate-lactonizing enzyme at least 2000-fold, and the activityof muconolactone isomerase at least 100-fold. The activity of/-ketoadipate enol-lactone hydrolase increases about 35-fold.However, the inductive response to benzoate is physiologicallynonspecific, since levels of two enzymes uniquely operative in theprotocatechuate pathway, -carboxy-cis,cis-muconate-lactoniz-ing enzyme and y-carboxymuconolactone decarboxylase, likewiseincrease about 35-fold. The levels of activity of the three latterenzymes in benzoate-grown cells are characteristically somewhatlower than in p-hydroxybenzoate-grown cells (Table I). Itshould be noted that the ability of extracts of benzoate-growncells to attack -carboxy-cis, cis-muconate and y-carboxymucon-olactone cannot be attributed to nonspecific catalysis by cis,cis-muconate-lactonizing enzyme and muconolactone isomerase, re-spectively, since it has been shown that purified preparations ofthese two enzymes have no activity on the carboxylated ana-logues of their natural substrates (3). Hence, it is evident thatan intermediate produced in the course of benzoate metabolismcan induce synthesis of two enzymes uniquely operative in theprotochatechuate pathway. Protocatechuate oxygenase, thethird enzyme unique to the protocatechuate pathway, is possiblyinduced to a very slight extent by growth with benzoate; its ac-tivity increases 8-fold over the basal level. This small increaseof specific activity might simply reflect the ability of catecholoxygenase to attack protocatechuate at a very low rate.

Although 3-ketoadipate is the eventual common product of theaction of the enzymes of the catechol and protocatechuate path-ways, cells grown at its expense contain high levels of three ofthese enzymes: -carboxy-cis,cis-muconate-lactonizing enzyme,y-carboxymuconolactone decarboxylase, and #-ketoadipateenol-lactone hydrolase. Their levels are comparable to the lev-els resulting from growth at the expense of p-hydroxybenzoate.The other enzymes of the catechol and protocatechuate pathwaysare not measurably induced by growth at the expense of /-ketoadipate.

O3-Carboxy-cis, cis-muconate Coordinate Block-The absolutespecific activities of -carboxy-cis,cis-muconate-lactonizing en-zyme, y-carboxymuconolactone decarboxylase, and 3-ketoadi-pate enol-lactone hydrolase are different in extracts of cells growwith benzoate, p-hydroxybenzoate, and -ketoadipate, but theirrelative specific activities remain closely comparable (Table I)This fact suggested that their syntheses might be coordinateAll the enzymes of the catechol and protocatechuate pathwayare subject to catabolite repression by glucose and succinatowhich permitted a more rigorous test of their presumptive co-

3802

l

2.

·a.J vHe- -

1

by guest on April 11, 2018

http://ww

w.jbc.org/

Dow

nloaded from

L. N. Ornstonse of August 25, 1966

ordinate nature. A series of cultures was grown with varyingratios of a specific inducer (either benzoate or p-hydroxyben-zoate) and a compound which causes catabolite repression (eitherglucose or succinate). Extracts from each culture were assayedfor four enzymes of the catechol and protocatechuate pathways.The results (Fig. 1) show that strict proportionality is maintainedbetween the levels of -carboxy-cis, cis-muconate-lactonizingenzyme, y-carboxymuconolactone decarboxylase, and #-ketoadi-pate enol-lactone hydrolase over a wide range of specific activity.These enzymes therefore constitute a coordinate block, termedthe -carboxy-cis,cis-muconate block. There is, in contrast, noproportionality between the levels of -ketoadipate enol-lactonehydrolase and of catechol oxygenase. Catechol oxygenase is notinduced by growth with p-hydroxybenzoate, and is far more sen-sitive to catabolite repression than the enzymes of the 0-carboxy-cis,cis-muconate block.

cis,cis-Muconate Block-Experiments analogous to those de-scribed in the preceding section showed (Fig. 2) that two enzymesunique to the catechol pathway, cis,cis-muconate-lactonizing en-zyme and muconolactone isomerase, likewise constitute a co-ordinate block, termed the cis,cis-muconate block. As shown inFig. 2, the synthesis of these two enzymes is regulated inde-

011.0 32

~06~~ ~3or 0.8

a, ._ 0.4110012 04

0.2 130 09 050.6215° 8O 06

o I l

ENC

oL- L

a 0a, E

u a

U

:[

v._v-

Muconolactone IsomeraseUnits/mg protein

FIG. 2. Coordinate regulation of the syntheses of cis,cis-mnuconate-lactonizing enzyme and muconolactone isomerase.Enzyme levels were determined in extracts of wild type P. putidawhich had been grown at least seven generations in mineral me-diumn containing the following carbon sources: 1, 10 mM benzoate;2, 10 mM benzoate; 3, 10 mM benzoate; 4, 10 mM benzoate and 10mIm glucose; 5, 10 mM benzoate and 5 mM glucose; 6, 5 mM benzoateand 5 mM glucose; 7, 5 mM benzoate and 10 mM glucose; 8, 5 mMbenzoate and 20 mm glucose; 9, 10 mM benzoate and 20 mm glucose;10, 10 mM benzoate and 5 m succinate; 11, 10 m benzoate and10 m succinate; 12, 10 m benzoate and 20 m succinate; 13,5 ma benzoate and 5 mM succinate; 14, 5 mM benzoate and 10 mMSuecinate; 15, 5 mM benzoate and 20 mM succinate.

TABLE II

Levels of some enzymes of catechol and protocatechuate pathwaysin extracts of Strain CM3A

Extracts were prepared from cultures which had been grown atthe expense of 10 mM benzoate, 10 mM cis,cis-muconate, or 20 mmsuccinate.

Specific activities in extractsfrom cells grown with

Enzyme

Ben- sMu- Succinatezoate conate

units/mg protein

Catechol oxygenase ..................... 0.94 0.35 <0.002cis,cis-Muconate-lactonizing enzyme.... 0.39 0.41 <0.002Muconolactone isomerase .............. 2.60 2.29 <0.02-Carboxy-cis, cis -muconate-lactonizingenzyme . .............. ........... 0.69 0.58 0.02

y-Carboxymuconolactone decarboxylase.. 2.74 2.09 0.05t-Ketoadipate enol-lactone hydrolase.... 1.44 1.38 0.03

pendently of that of catechol oxygenase, the third enzyme uniqueto the catechol pathway. Whereas synthesis of enzymes of thecis,cis-muconate block is more sensitive to succinate (45 % aver-age repression) than to glucose (14% average repression), syn-thesis of catechol oxygenase is affected to an almost equal extentby both these compounds (about 75% average repression).

The experiments with the wild type of P. putida A.3.12 de-scribed above reveal in broad outline the control mechanismsgoverning the synthesis of the enzymes of the catechol and pro-tocatechuate pathways. The synthesis of catechol oxygenaseand protocatechuate oxygenase, the first enzymes of the catecholand protocatechuate pathways, respectively, are each individ-ually controlled. The second and third enzymes unique to thecatechol pathway, cis, cis-muconate-lactonizing enzyme andmuconolactone isomerase, constitute a coordinate block. Thesecond and third enzymes of the protocatechuate pathways, 8-carboxy-cis,cis-muconate-lactonizing enzyme and y-carboxymu-conolactone decarboxylase, together with 3-ketoadipate enol-lactone hydrolase, the first enzyme common to both pathways,constitute a second coordinate block. However, these experi-ments do not reveal the specific inducers operative in the fourseparately controlled inductive events. A clearer insight intothis question was obtained through experiments with mutants,described in the following sections.

Inductive Properties of cis, cis-Muconate-Succinate-growncells of the wild type of P. putida cannot grow at all with cis,cis-muconate; benzoate-grown cells grow arithmetically with it.However, spontaneous mutants that are capable of exponentialgrowth with cis,cis-muconate can be obtained by selection (see"Experimental Procedure"). One such mutant, CM3A, wasstudied in detail. After growth with benzoate, its enzymic con-stitution is indistinguishable from that of the wild type (Table II).However, benzoate-grown cells of CM3A consume oxygen rapidlywhen furnished with cis,cis-muconate, whereas benzoate-grownwild type cells do not (Fig. 3). These facts indicate that CM3Ais a mutant with altered permeability properties, which permitcis,cis-muconate to enter the cell more readily than it can enterthe cell of the wild type.

When mutant CM3A is grown at the expense of cis,cis-mu-conate, all the enzymes of the catechol pathway are induced, to-

3803

by guest on April 11, 2018

http://ww

w.jbc.org/

Dow

nloaded from

Catechol and Protocatechuate Pathways in P. putida. IV

gether with Pf-carboxy-cis,cis-muconate lactonizing enzyme andy-carboxymuconolactone decarboxylase (Table II). These in-ductions cannot be explained by invoking an endogenoussynthesis of catechol from cis, cis-muconate through the action ofcatechol oxygenase; this enzyme, like other enzymes that mediateoxygenations, catalyzes an essentially irreversible reaction. Thefailure of p-hydroxybenzoate to induce significant levels of thethree enzymes uniquely associated with the catechol pathway inthe wild type (Table I) shows that neither 3-ketoadipate enol-lactone nor subsequent intermediates common to both pathwayspossess inductive function for the enzymes in question. The re-sults presented in Table II accordingly show that either cis,cis-muconate or (+)-muconolactone (or possibly both compounds)can induce the three enzymes unique to the catechol pathway.

Growth with cis,cis-muconate elicits lower levels of catecholoxygenase in the permeability mutant than does growth withbenzoate (Table II), but this result does not necessarily meanthat a precursor of cis,cis-muconate is essential for full inductionof this enzyme. Benzoate-grown cultures of the mutant oxidizebenzoate more rapidly than cis,cis-muconate (Fig. 3), indicatingthat the mutation has not made the cells fully permeable to the

a

0

CCa

0,X0,0

'5

$

Minutes

FIG. 3. Rates of oxidation of cis,cis-muconate and benzoate bywild type P. putida and permeability mutant CM3A. Cells weregrown at the expense of 10 mM benzoate, washed with 0.02 Mdisodium potassium phosphate buffer, pH 7.0, and 6 X 108 cellswere suspended in flasks containing 100 moles of disodium potas-sium phosphate buffer, pH 7.0, in a final volume of 1.6 ml. Thecenter well contained 0.2 ml of 10% NaOlH. Substrate (10 molesin 0.2 ml) was added from a side arm at 0 min, and the rate ofoxygen consumption at 30° was followed in a Gilson differentialrespirometer.

dicarboxylic acid. This limitation is reflected in a relatively klogeneration time (75 min) when the mutant is growing at the ,pense of cis,cis-muconate. Since the supply of exogenous ,cis-muconate limits the rate of growth, the intracellular coneeotration of this compound under such growth conditions is probeably considerably lower than in benzoate-grown cultures. cis,cis-muconate is a less efficient inducer of catechol oxygenasthan of the cis,cis-muconate block of enzymes, the low intracellular levels of cis,cis-muconate during growth with this corpound may elicit only partial induction of catechol oxygenase, bhfull induction of the subsequent enzymes. In this light,should be noted that the synthesis of catechol oxygenase is fatmore sensitive to catabolite repression than subsequent enzymesof the catechol and protocatechuate pathways (Figs. 1 and 2). Ahigh intracellular concentration of cis ,cis-muconate may be neeessary to overcome catabolite repression; and this may not beobtainable with cis, cis-muconate as an exogenous substrate.

Induction of Mutants Specifically Blocked in Catechol Pathway-Three independent mutants of P. putida A.3.12 which had lo4the ability to grow on benzoate, while retaining the ability togrow on p-hydroxybenzoate, were studied. Cultures of thesemutants and of the wild type were induced by growth for threegenerations in a medium containing 10 mM glucose and 10 mbenzoate. As shown by the data in Table III, none of the threemutants contained detectable amounts of cis,cis-muconate-lac.tonizing enzyme; failure to grow on benzoate is, accordingly, ex.-plainable by loss of the ability to synthesize this enzyme. Never.theless, growth of two of these mutants (DLM8 and NG22) inthe presence of benzoate elicits the synthesis of both catecholoxygenase and muconolactone isomerase. Since neither straincan form ( +)-muconolactone from benzoate, ( +)-muconolactoneis not a necessary inducer of catechol oxygenase and of the cis, cis-muconate coordinate block. Taken in conjunction with the dataon the permeability mutant CM3A, these results suggest that cis,cis-muconate is the metabolite which induces synthesis both ofcatechol oxygenase and of the cis, cis-muconate coordinate block.None of the experiments described excludes the possibility thatcatechol may also possess inductive function for the enzymesspecific to the catechol pathway; but in view of the generallyhigh steric specificity of inducers (6), it seems unlikely that twomolecules as dissimilar in structure as catechol and cis,cis-mu-conate share a common inductive function.

One of the mutants blocked in the catechol pathway (NG4)can synthesize neither cis,cis-muconate-lactonizing enzyme normuconolactone isomerase (Table III). This mutant readily re-verts to growth at the expense of benzoate; and reversion restoresthe ability to synthesize both enzymes. Mutant NG4 thus ap-pears to have undergone a single mutation, not a deletion or a

TABLE III

Levels of enzymes of catechol pathway in extracts of wild type P. putida A. 3.12 and of benzoate-negative mutants, grown at expense of 10mm glucose with benzoate (10 mM) as inducer

Wild type DLM8 NG22 NG4Enzyme

Inducer present Inducer absent Inducer present Inducer present Inducer present

units/nig protein

Catechol oxygenase ........................ 0.31 <0.002 0.26 0.15 0.13cis, cis-Muconate-lactonizing enzyme........ 0.14 <0.002 <0.002 <0.002 <0.002Muconolactone isomerase .................... 1.19 <0.02 1.07 2.69 <0.02s-Ketoadipate enol-lactone hydrolase ........ 0.33 0.03 0.06 0.04 0.03

3804 Vol. 241, No.

by guest on April 11, 2018

http://ww

w.jbc.org/

Dow

nloaded from

Issue of August 25, 1966

double mutation. This pleiotropic effect further supports theconclusion that cis,cis-muconate-lactonizing enzyme and mu-conolactone isomerase share regulatory linkage.

Induction in Strains with Blocks in Protocatechuate Pathway-

Nearly every wild type strain of P. putida can grow at the expenseof both benzoate and p-hydroxybenzoate; however, an extensivenutritional survey of this species (6) revealed one strain, P. pu-

tida CI-A (ATCC 17452), which could grow at the expense ofbenoate, but not of p-hydroxybenzoate. A mutant of P. putidaA.3.12 with the same nutritional phenotype (strain A202) wasalso available. Both these strains, when grown with benzoate,synthesize normal levels of y-carboxymuconolactone decar-boxylase and ji-ketoadipate enol-lactone hydrolase, but do notsynthesize any detectable a-carboxy-cis,cis-muconate-lactonizingenzyme (Table IV). Their inductive responses to early membersof the protocatechuate pathway are somewhat difficult to deter-mine, since both p-hydroxybenzoate and protocatechuate aregrowth-inhibitory. However, both strains can grow slowly in amedium containing 10 mM glucose and 5 mM protocatechuate.Table V shows their inductive patterns after growth for threegenerations in this medium, compared to the inductive pattern ofthe wild type of P. putida A.3.12 grown under the same condi-tions. It is evident that both P. putida C1-A and mutant A202can synthesize appreciable amounts of protocatechuate oxygen-ase; but the enzymes of the -carboxy-cis, cis-muconate block arenot induced. The growth-inhibitory effects of p-hydroxyben-zoate and protocatechuate can probably be interpreted as the re-sult of an intracellular accumulation of endogenously generatedf-carboxy-cis,cis-muconate. This experiment shows that -

TABLE IV

Levels of enzymes of 3-carboxy-cis, cis-muconate coordinate blockin extracts of several strains of P. putida after growth at

expense of 10 m.r benzoate

P. putida P. putidaEnzyme A.3.12, A.3.12, p

wild type mutant A202 wi

units/mg protein

-Carboxy- cis, cis -muconate-lactonizing enzyme.......... 0.68 <0.008 <0.008

Y-Carboxymuconolactone de-carboxylase ................. 2.02 1.15 1.87

O-Ketoadipate enol-lactone hy-drolase ..................... 1.03 0.78 0.91

TABLE V

Enzyme levels in extracts of several strains of P. puticgrowth on 10 mM glucose in presence of 5 mr protocatec

Enzyme

Protocatechuate oxygenase ....f -Carboxy cis, cis -muconate-

lactonizing enzyme..........T-Carboxymuconolactone de-

earboxylase ...............f-3etoadipate enol-lactone hy-

drolase ....................

Strains

P. putida P. putida PA.3.12, A.3.12,

wild type mutant A202

units/mg protein

0.88

0.23

0.72

0.45

0.10

<0.005

0.06

0.03

la after

TABLE VI

Enzyme levels in extracts of wild type P. putida A .3.12 and ofmutant A 14 after growth at expense of 10 mM 1-ketoadipate

Strain

Enzyme

Wild type Mutant A14

units/mg protein

,-Carboxy-cis, cis-muconate-lactonizingenzyme............................. 0.80 0.89

7-Carboxymuconolactone decarboxyl-ase ................................. 2.48 0.29

O-Ketoadipate enol-lactone hydrolase ... 1.58 <0.02

TABLE VII

Enzyme levels in extracts of wild type P. putida A .S.12 and mutantA14 after growth at expense of 10 mM glucose in presence of

10 mM benzoate

Wild type A14

EnzymeInducer Inducer Inducerpresent absent present

units/mg protein

cis, cis-Muconate-lactonizing enzyme... 0.14 <0.002 0.26Muconolactone isomerase ............. 1.19 <0.02 1.97O-Carboxy-cis, cis-muconate-lactonizing

enzyme ............................ 0.20 0.02 0.02y-Carboxymuconolactone decarboxyl-

ase ............................... 0.51 0.05 <0.02Os-Ketoadipate enol-lactone hydrolase.. 0.33 0.03 <0.02

TABLE III

Enzyme levels in extracts of wild type P. putida A.3.12 and ofmutant A14 after growth with 10 mim glucose in presence of

5 mrn. protocatechuate

StrainEnzyme

Wild type I Mutant A14

units/mg protein

Protocatechuate oxygenase............. 0.88 0.19O-Carboxy-cis, cis-muconate-lactonizing

enzyme.............................. 0.23 0.03y-Carboxymuconolactone decarboxyl-

ase .................................. 0.72 <0. 005t3-Ketoadipate enol-lactone hydrolase ... 0.45 <0. 005

huate carboxy-cis,cis-muconate is not an inducer of the enzymes of the8f-carboxy-cis,cis-muconate block, even though it is the primarysubstrate for this block of enzymes.

Inductive Patterns of Mutant A 14-One of the induced mutants. putida 76,wild type isolated from P. putida A.3.12, mutant A14, is unable to grow at

the exnense of either benzoate or -hvdroxvbenzoate. although itcan grow, with a generation time similar to that of the wild type,

0.27 at the expense of -ketoadipate. Growth of this mutant on,8-ketoadipate results in full induction of -carboxy-cis,cis-mu-

<0.005 conate-lactonizing enzyme and induction of y-carboxymuco-nolactone decarboxylase to a level of about 10% of that charac-teristic of the wild type; no activity of enol-lactone hydrolase is

0.03 detectable (Table VI). Mutant strain A14 has therefore under-gone a Dleiotroic mutation. in which the activity of one enzyme

L. N. Ornston 3805

" "

.~

by guest on April 11, 2018

http://ww

w.jbc.org/

Dow

nloaded from

Catechol and Protocatechuate Pathways in P. putida. IV

TABLE IX

Enzyme levels in revertants of mutant A14 after growth at expense of 10 mM benzoate

Revertant strainsEnzyme Wild type

A14R1 A14R2 A14R3 A14R4

units/mg protein

l-Carboxy-cis, cis-muconate-lactonizing enzyme ................ 0.62 0.61 0.64 0.66 0.56?-Carboxymuconolactone decarboxylase ........................ 2.06 1.93 1.62 1.97 1.57/3-Ketoadipate enol-lactone hydrolase .......................... 1.08 0.250 0.320 0.35a 0.36

a Enzyme unstable under conditions of assay.

TABLE X

Influence of growth substrate on enzyme levels in P. aeruginosa

Growth substrate

Enzyme p-Hy-Succin- Benzo- drox Adip-

ate ate ybenzo- ateate

units/mg protein

-Carboxy-cis ,cis-muconate-lactonizingenzyme ............................. 0.012 0.56 0.50 0.86

,-Carboxymuconolactone decarboxyl-ase ................................ .008 2.14 1.73 2.79

-Ketoadipateenol-lactonehydrolase.. 0.04 1.10 0.66 1.21

TABLE XI

Influence of growth substrate on enzyme levels in P. multivorans

Enzyme

-Carboxy-cis, cis -muconate-lactonizingenzyme ............................

y-Carboxymuconolactone decarboxyl-ase ................................

/-Ketoadipate enol-lactone hydrolase .

Growth substrate

p-Hy-Succin- Benzo- dox- Adip-

ate ate ybenzo- ateate

0.02

0.020.04

units/mg protein

0.74 0.69 0.33

1.46 1.66 0.720.90 1.02 0.47

in a coordinate block has been completely eliminated, and theactivity of another enzyme in the same block severely reduced.The failure of this mutant to grow at the expense of either ben-zoate or p-hydroxybenzoate is a consequence of its completeinability to synthesize f-ketoadipate enol-lactone hydrolase.

The inductive patterns resulting from exposure of mutant A14to compounds early in the protocatechuate and catechol path-ways were determined by growing this strain at the expense of 10mM glucose in the presence of 5 mM protocatechuate or 10 mMbenzoate. Cultures of the wild type grown under the same con-ditions served as controls. The results of these experiments areshown in Tables VII and VIII. Grown in the presence of ben-zoate, mutant A14 synthesizes the two enzymes of the cis, cis-mu-conate block at levels comparable to those of the wild type; butthere is no induction of any enzymes of the 3-carboxy-cis,cis-muconate block under these conditions (Table VII). It has beenshown elsewhere that growth of mutant A14 in the presence ofbenzoate causes the accumulation of an equilibrium mixture ofcis,cis-muconate, (+)-muconolactone, and -ketoadipate enol-

lactone (1). The total absence of induction of any of the en.zymes of the 0-carboxy-cis,cis-muconate coordinate block undeithese circumstances therefore shows that -ketoadipate enol.

lactone cannot serve as an inducer of the coordinate block,Grown in the presence of protocatechuate, mutant A14 synthesizes protocatechuate oxygenase; but there is again no inductionof the enzymes of the 0-carboxy-cis, cis-muconate block, althoughthese are substantially induced in wild type cells grown under thesame conditions (Table VIII). However, it should be noted thatmutant A14 has the normal basal level of -carboxy-cis,cis.muconate-lactonizing enzyme, and is therefore capable, followinginduction with protocatechuate, of a slow endogenous synthesiof y-carboxymuconolactone. The absence of induction deither -carboxy-cis,cis-muconate-lactonizing enzyme or y-carboxymuconolactone decarboxylase under these circumstancetherefore indicates that y-carboxymuconolactone is also unableto serve as an inducer for the enzymes of the 0-carboxy-cis,cismuconate coordinate block.

Several independent revertants of mutant A14, selected for theability to grow at the expense of benzoate, all regained simultanoously the ability to grow at the expense of p-hydroxybenzoataAfter growth with benzoate, these revertants contained y-caboxymuconolactone decarboxylase at the normal wild type levelHowever, the level of -ketoadipate enol-lactone hydrolase wasin each case substantially lower than that characteristic of thewild type (Table IX). In the revertants, this enzyme is mudless stable than the wild type enzyme, and its rapid loss of ativity during extraction and assay probably explains the rehltively low values shown in Table IX.

The experiments with mutants described in this and precedingsections indicate that none of the substrates for the enzymes 4the j3-carboxy-cis,cis-muconate block possesses inductive funstion. Since growth at the expense of -ketoadipate elicits fulinduction of the block (Table I), the inducer is evidently normallYproduced through the action of the enzymes of the block. Sinelthese enzymes are not induced in either succinate- or acetategrown cells (Table I), either fi-ketoadipate or 3-ketoadipyl-Comust be the intermediate which elicits induction.

Preliminary Observations on Synthesis of Enzymes of Protocatchuate Pathway in Other Species of Bacteria-Since many differelAbacteria can metabolize benzoate and p-hydroxybenzoate throuO#-ketoadipate, the enzymes of the catechol and protocatechuatlpathways provide valuable material for the comparative analystof regulatory mechanisms. We have examined the synthesis 0

-carboxy-cis, cis-muconate-lactonizing enzyme, -y-carboxymucOnolactone decarboxylase, and -ketoadipate enol-lactone hydrOlase by four other species, in order to determine whether the coordinate control observed in P. putida is characteristic of bactern

---

3806 Vol. 241, No.-I

by guest on April 11, 2018

http://ww

w.jbc.org/

Dow

nloaded from

Issue of August 25, 1966

TABLE XII

Influence of growth substrate on enzyme levels in H. eutropha

Enzyme

o-Carboxy-cis, cis-muconate-lactonizing enzyme...........?-Carboxymuconolactone decarboxylase ...................g-Ketoadipate enol-lactone hydrolase ...................

Growth substrate

Succinate

0.0050.010.02

Benzoate p-Hydroxybenzoate

units/mg protein

0.002 0.610.01 1.341.38 0.68

TABLE XIII

Influence of growth substrate on enzyme

Enzyme

,-Carboxy-cis, cis-muconate-lactotnizinig en-zyme ...............................

,-Carboxymuconolactone decarboxylase....^ - Z I - I A A Al_ A - A A A |_ --- - --trA AX

levels

SucCa

0.0.n

-etoaolpate enol-ltacUole nyylroulase ...... v.

Two other species of the genus Pseudomonas, fP. multivorans, were studied. Both differ fromany phenotypic respects (6), and the molar pelnine plus cytosine in their DNA is also significanAs shown in Tables X and XI, 0-carboxy-cis,ctonizing enzyme, y-carboxymuconolactone dec#-ketoadipate enol-lactone hydrolase are inducedlevels in both species by growth at the expense ofor p-hydroxybenzoate. Both basal and induce(enzymes are similar to those observed in P. putifore probable that their synthesis is coordinatelythree species, and elicited by the same metabolic

P. aeruginosa and P. multivorans can grow onputida cannot. As shown in Tables X and XI, adiof these two species contain levels of B-carboxy-cilactonizing enzyme, y-carboxymuconolactone dec#-ketoadipate enol-lactone hydrolase comparalbenzoate or p-hydroxybenzoate-grown cells. Altway of adipate dissimilation has not been deterlikely metabolite common to the catechol, protoadipate pathways is -ketoadipyl-CoA. Accoisuits shown in Tables X and XI suggest that B-elicits the synthesis of the f-carboxy-cis,cis-muenzymes, and is possibly the sole inducer of themfunction for -ketoadipate cannot, however, be etion of the -carboxy-cis,cis-muconate block inpseudomonads might be caused by endogenousketoadipate, through hydrolysis of the thiol ester

The two other species examined, H. eutropha annot closely related to P. aeruginosa, P. multivoransIH. eutropha, although possessing a DNA base c

ilar to that of P. putida (7), is peritrichously flagan entirely different nutritional spectrum.' M.flagellated bacterium and its DNA has a far

I D. Davis, personal communication.

guanine plus cytosine than either the pseudomonads or H. eu-; in M. Iwoffli tropha (7). As shown in Tables XII and XIII, the syntheses of

p-carooxy-cs s, cs-muconate-tactomzing enzyme, y-carDoxymu-conolactone decarboxylase, and fi-ketoadipate enol-lactone hy-

B p-Hy- drolase are not coordinately controlled in H. eutropha and 3l.ti ae- 2ybeo- Iwoffi. All three enzymes are induced in these species as a re-

ate suit of growth with p-hydroxybenzoate, but only i-ketoadipateenol-lactone nyarolase activity s induced to a signlhcant extentunits/rag proteinby growth with benzoate. Although the biochemistry of the

04 0.15 1.87 catechol and protocatechuate pathways in M. Icoffli is identical02 0.05 0.72 with that in P. putida (1), these data show that the regulatory11 0.82 0.78 control of these pathways is entirely different in the two species.

DISCUSSION

.aeruginosa and Conclusions concerning the regulation of the synthesis of en-m P. putida in zymes of the catechol and protocatechuate pathways in P. putidarcentage of gua- are summarized in Fig. 4. These findings provide some insighttly different (7). into the sequence of inductive events that takes place when unin-is-muconate-lac- duced cells are exposed to these diphenolic compounds or formarboxylase, and them endogenously from metabolic precursors.I to comparable Two successive inductive events permit the synthesis of theeither benzoate four enzymes that convert protocatechuate to 0f-ketoadipate; in-

I levels of these duction of protocatechuate oxygenase, and induction of the threeda. It is there- enzymes of the -carboxy-cis,cis-muconate coordinate block.controlled in all The inducer of protocatechuate oxygenase has not been iden-

te inducer. tified; it appears to be either the substrate or one of two subse-adipate, but P. quent intermediates in the protocatechuate pathway. This can

ipate-grown cells be inferred from the fact that the synthesis of protocatechuate5s,cis-muconate- oxygenase is elicited by growth with protocatechuate, but not bymarboxylase, and growth with benzoate. The enzymes of the 3-carboxy-cis,cis-

ble to those in muconate block, which convert -carboxy-cis,cis-muconate tohough the path- /3-ketoadipate, are present in uninduced cells at about 2% ofmined, the only fully induced levels. Consequently, the product of protocatech-catechuate, and uate oxygenase, 0-carboxy-cis,cis-muconate, can be converted tordingly, the re- /-ketoadipate without induction. Since /3-ketoadipate or --ketoadipyl-CoA ketoadipyl-CoA elicits the synthesis of this coordinate block, 3-conate block of ketoadipate formation triggers induction. The regulation of the

.An inductive synthesis of the enzymes that convert B-ketoadipate to succinateexcluded; induc- and acetyl-CoA has not been studied.i adipate-grown The synthesis of the enzymes that convert catechol to -ke-formation of #- toadipate requires three separately controlled inductive events.

The synthesis of catechol oxygenase is elicited by its product,id M. wofii, are cis,cis-muconate. There is no experimental evidence which ex-s, and P. putida. eludes catechol itself as an inducer, but it seems improbable thatomposition im- a diphenolic compound and a dicarboxylic acid could act inter-ellated, and has changeably as inducers. If cis,cis-muconate is in fact the solehIoffii is a non- inducer,, the extremely low levels of catechol oxygenase in unin-

)wer content of duced cells (less than 0.02% of the fully induced level) must besufficient to permit an effective endogenous generation of inducer.

Adipate

<0.002<0.005<0.005

L. N. Ornston 3807

. . . . . . . . .

" `' ''

by guest on April 11, 2018

http://ww

w.jbc.org/

Dow

nloaded from

Catechol and Protocatechuate Pathways in P. putida. IV

Inducer Enzymes

cis, cis-Muconate Catechol Oxygenase

cis, cis-MuconateLactonizing Enzyme

cis, cis-Muconate

MuconolactoneIsomerase

Metabolites

aOOH -OH

I I·co7C

CO7

C CO2

.J

C = -

° coCO7

co

,0 o ,,0CH3C-SCOA CH SCoA

CHC-SCoA + ,CH2 C-SCoACH 5- C07

Enzymes Inducer

Protocatechuate Oxygenase

,-Carboxy-cis, cis-MuconateLactonizing Enzyme

y-CarboxymuconolactoneDecarboxylose

-:-Ketoadipate Enol-LactoneHydrolase

FIG. 4. Inductive mechanisms operative in the catechol and protocatechuate pathways in P. putida. The syntheses of enzymes thenames of which are enclosed in brackets are coordinately regulated. Inducers are the intermediates that are believed to elicit the synthesis of enzymes most directly.

t-Tryptophan

3 enzymes2 inductions

Anthranilate + L-alanine

1 enzyme1 induction

Catechol

3 enzymes2 inductions

19-Ketoadipate enol-lactone

3 enzymes*1 induction

/3-Ketoadipate

FIG. 5. The sites and nature of sequential inductions in then-mandelate and L-tryptophan pathways in fluorescent pseudo-monads. A shift of inducers occurs following the formation ofeach named intermediate. Also shown are the number of enzymessynthesized and the number of inductions in each sequential step.Only one of the enzymes in the group indicated by the asteriskis functional in the pathway shown.

cis,cis-Muconate also elicits coodinate synthesis of the two en-zymes that catalyze its conversion to -ketoadipate enol-lactone.Thus a single intermediate plays a dual inductive role in the cate-chol pathway; it accelerates the rate of synthesis of the enzymethat produces it, as well as the rates of synthesis of the enzymesthat decompose it.

The metabolic product of the cis,cis-muconate coordinateblock, 3-ketoadipate enol-lactone, lies at the point of metabolicconvergence with the protocatechuate pathway. Induction ofthe first enzyme common to the two pathways, -ketoadipate

enol-lactone hydrolase, can take place only after its substrate hasbeen converted to -ketoadipate, since this enzyme is a memberof the -carboxy-cis,cis-muconate coordinate block. The othertwo enzymes of the block, although they have no function in thecatechol pathway, are induced in cells metabolizing catechol orone of its precursors.

Control mechanisms in two metabolic sequences convergent oncatechol (Fig. 5) have also been studied in fluorescent pseu-domonads (8-10). The syntheses of the five enzymes that con,vert D-mandelate to benzoate are coordinate; either mandelate orthe product of the second enzyme in the sequence, benzoylformnate, can act as the inducer (8). A sequential inductive stepoccurs at the levels of benzoate (9). The syntheses of the threeenzymes that convert L-typtophan to anthranilate are induced byL-kynurenine, the product of the second enzyme of the pathway.The first and second enzymes of the L-tryptophan pathway (pres-ent at low levels in uninduced cells) mediate synthesis of the in-ducer when L-tryptophan is furnished exogenously. L-Kynurenine (like cis, cis-muconate) has a double inductive role; it acts asan inducer for the coordinate synthesis of the two enzymes thaiform it, as well as for synthesis of the enzyme that converts it toanthranilate and L-alanine. The enzyme that converts anthran-ilate to catechol is induced sequentially (10).

It is evident that many of the enzymes operative in the dissimilation of D-mandelate and L-tryptophan by fluorescent pseddomonads are coordinately induced. Synthesis of such coordinate blocks of enzymes requires a single inductive event. 1is also clear that enzymes the synthesis of which is governed independently may be induced by the same metabolite. Hencetwo inductive events may be triggered simultaneoulsy. At cttain points in these catabolic pathways a sequential inductistep, defined as a shift in the nature of the inducer, occurs (Fig. 5

Sequential inductive steps can occur at sites where a compoundthat is an intermediate in a catabolic pathway can also serve asO

3.-Ketoadipate or/-Ketoadipyl CoA

Succinyl CoA: -KetoadipoteCoA Transferose

,-Ketoadipyl CoA Thiolase

D-Mandelate

5 enzymes 1I induction

Benzoate

1 enzyme\1 induction ,

3808 Vol. 241, No.

by guest on April 11, 2018

http://ww

w.jbc.org/

Dow

nloaded from

L. N. OrnstonIssue of August 25, 1966

primary growth substrate; benzoate and anthranilate are ex-

amples of such compounds. Benzoate elicits the synthesis of the

enzymes requisite for its utilization without inducing the enzymes

that catalyze its formation from mandelate. Similarly, anthran-

ilate induces only enzymes associated with its dissimilation.

Sequential steps also occur at sites where intermediates

enter a catabolic sequence from a convergent metabolic

pathway. Catechol, for example, may be produced endoge-

nously by the degradation of either benzoate or anthranilate. Se-

quential induction at the level of catechol permits the synthesis

of the enzymes that degrade it without the synthesis of the en-

zymes that form it from the aromatic acids. Thus, sequential

induction usually permits the utilization of a wide variety of car-

bon sources with minimal synthesis of nonfunctional enzymes.

In this light, it is interesting to note that a sequential inductive

step does not occur at the site of convergence of the catechol and

protocatechuate pathways in P. putida. In order to metabolize

further f-ketoadipate enol-lactone produced through the catechol

pathway, cells must synthesize (as a consequence of regulatory

linkage) two enzymes uniquely associated with the protocatechu-

ate pathway.One explanation for this bizarre regulatory mechanism is that

-carboxy-cis, cis-muconate, y-carboxymuconolactone, and B-

ketoadipate enol-lactone do not possess chemical structures

which permit them to function as inducers. If this were the case,

the synthesis of the enzymes that mediate their dissimilation

would have to be induced either by a metabolic precursor or by a

metabolic product. In P. putida, the enzymes are induced by

their product, -ketoadipate (or -ketoadipyl-CoA). In con-

trast, the syntheses of -carboxy-cis,cis-muconate-lactonizing

enzyme and y-carboxymuconolactone decarboxylase are not in-

duced by these compounds in 3M. woffi. The syntheses of all

four of the enzymes that convert protocatechuate to -ketoadi-

pate are coordinately regulated in M. wofii; protocatechuate is

the probable inducer of the coordinate block.2 This control sys-

tem does not provide a mechanism for the synthesis of /3-ke-

toadipate enol-lactone hydrolase by cells utilizing catechol pre-

cursors. This difficulty is circumvented in M. woffi by the

synthesis of an isofunctional enzyme, which is specifically induced

by growth with benzoate. 2 Hence, the control mechanisms in

M. woffli permit strict economy of induced enzyme synthesis, but

require two structural genes governing synthesis of isofunctional

enzymes.Pseudomonads are not the only bacteria in which some en-

zymes can be induced by their products rather than by their sub-

strates. L-Histidine is degraded to L-glutamate by four enzymesin Aerobacter aerogenes; these enzymes are induced by the productof the first enzyme of the pathway, urocanate (11, 12). Since

only two of the four enzymes are subject to coordinate control,

this intermediate triggers three simultaneous inductive events

(13). Two cases of product induction have been reported in

Escherichia coli: a-glycerophosphate induces glycerokinase (14);

and -galactosidase is induced only after it catalyzes the trans-

formation of lactose to another /-galactoside (15). The occur-

rence of product induction in diverse bacterial systems indicates

that it may have selective value.

Palleroni and Stanier (10) proposed that product induction ofthe enzymes of the L-tryptophan pathway by L-kynurenine pre-

vents induction of the catabolic enzymes by endogenously syn-

' J. L. CAnovas, personal communication.

thesized L-tryptophan. This would be the case if the Km of thefirst enzyme of the catabolic pathway, L-tryptophan pyrrolase,

were substantially higher than both the internal pool size of L-

tryptophan and the Km of tryptophanyl ribonucleic acid syn-

thetase. Schlesinger, Scotto, and Magasanik (12) concluded

that the hypothesis of Palleroni and Stanier might provide a

satisfactory explanation of their own observations on the regula-

tory mechanisms governing the synthesis of the enzymes that

degrade L-histidine in A. aerogenes. The Km of histidinyl ribo-

nucleic acid synthetase is far lower than the Km of histidine am-

monia-lyase in this organism. Hence, the enzymes that degrade

L-histidine are not synthesized in the absence of a relatively high

concentration of exogenously supplied L-histidine (12).

The hypothesis of Palleroni and Stanier does not provide an

interpretation for the other cases of product induction that have

been reported. Catechol, forexample, does not playa knownbio-

synthetic role in P. putida. There is no reason to believe that

catechol is produced endogenously in the absence of its metabolic

precursors. Some other factor must place selective pressure in

favor of product induction of catechol oxygenase. One such

factor might be the economy of protein synthesis that results

from the specific induction of enzymes that degrade compounds

with structures related to catechol. Product induction of cate-

chol oxygenase decreases the probability that the synthesis of

this enzyme would be stimulated by chemical analogues of cate-

chol because both catechol and cis,cis-muconate, widely dispa-

rate in chemical properties, must interact with biological systems

of high specificity before increased synthesis of this enzyme is

initiated. Even those diphenolic compounds that are cleaved by

catechol oxygenase cannot elicit the synthesis of the enzyme un-

less their products are dicarboxylic acids with structures closely

related to cis,cis-muconate.Coordinate regulation in the catechol and protocatechuate

pathways cannot yet be explored on the genetic level, since no

mechanism of recombination has been discovered in P. putida.

The enzymes of the cis,cis-muconate and -carboxy-cis,cis-

muconate coordinate blocks may be under the control of two

complex operons; but the only available evidence bearing on this

question, the properties of mutants, is difficult to interpret. The

revertible mutant NG4 (Table III), which has lost simultane-

ously the ability to synthesize both enzymes of the cis, cis-mucon-

ate coordinate block, could be the consequence of a polarity mu-

tation in one structural gene of a complex operon (16), but it is

also explainable as the result of a mutation in a regulatory gene

governing the functions of two separate operons (17). Mutant

A14, which is impaired in the synthesis of enzymes of the f-car-

boxy-cis,cis-muconate coordinate block (Table V) provides

stronger evidence for control by a complex operon. This re-

vertible mutant makes no 6-ketoadipate enol-lactone bydrolase,

but synthesizes y-carboxymuconolactone decarboxylase at a low

rate and -carboxy-cis,cis-muconate-lactonizing enzyme at a

normal rate. A pleiotropic effect of this nature cannot be ex-

plained by mutation in a regulatory gene, but is fully compatible

with the assumption of a polarity mutation in one structural gene

of a complex operon (18, 19). The properties of revertants

(Table IX) point to the structural gene for -ketoadipate enol-

lactone hydrolase as the site of the mutation: in all these revert-

ants, the capacity to synthesize both affected enzymes is re-

stored; but -ketoadipate enol-lactone hydrolase is far less stable

than in the wild type. The reversion can therefore be interpreted

as the consequence of a further amino acid substitution at (or

3809

by guest on April 11, 2018

http://ww

w.jbc.org/

Dow

nloaded from

3810 Catechol and Protocatechuate

near) the site originally changed by mutation in the -ketoadipateenol-lactone hydrolase of the wild type.

Acknowledgments-I am indebted to Dr. R. Y. Stanier for themany helpful suggestions he offered during the preparation ofthis manuscript. I am also grateful to Dr. G. D. Hegeman,Miss M. Leroux, and Mrs. M. Krawiec for assistance in theisolation of mutants.

REFERENCES

1. ORNSTON, L. N., AND STANIER, R. Y., J. Biol. Chem., 241,3776 (1966).

2. ORNSTON, L. N., J. Biol. Chem., 241, 3787 (1966).3. ORNSTON, L. N., J. Biol. Chem., 241, 3795 (1966).4. LEDERBERG, J., AND LEDERBERG, E. M., J. Bacteriol., 63, 399

(1952).5. WEICHSELBAUM, T. E., Am. J. Clin. Pathol., 16, Tech. Sect., 10,

40 (1946).6. STANIER, R. Y., PALLERONI, N. J., AND DOUDOROFF, M., J.

Gen. Microbiol., 43, 159 (1966).

IPathways in P. putida. IV Vol. 241, No.

7. MANDEL, M., J. Gen. Microbiol., 43, 273 (1966).8. HEGEMAN, G. D., J. Bacteriol., 91, 1161 (1966).9. STANIER, R. Y., HEGEMAN, G. D., AND ORNSTON, L. N.

Colloq. Intern. Centre Natl. Rech. Sci. (Paris), 227 (1965).10. PALLERONI, N. J., AND STANIER, R. Y., J. Gen. Microbiol., 35,

319, (1964).11. SCHLESINGER, S., AND MAGASANIK, B., J. Biol. Chem., 240,

4325 (1965).12. SCHLESINGER, S., SCOTTO, P., AND MAGASANIK, B., J. Biol

Chem., 240, 4331 (1965).13. MAGASANIK, B., LUND, P., NEIDHARDT, F. C., AND SCHWARTZ

D. T., J. Biol. Chem., 240, 4320 (1965).14. COZZARELLI, N., HAYAISHI, S., AND LIN, E. C. C., Federatioa

Proc., 24, 417 (1965).15. MUELLER-HILL, B., RICKENBERG, H., AND WALLENFELS, K.,

J. Mol. Biol., 10, 303 (1964).16. JACOB, F., AND MONOD, J., J. Mol. Biol., 3, 318 (1961).17. MAAS, W. K., AND CLARK, A. J., J. Mol. Biol., 8, 365 (1964).18. AMES, B. N., AND MARTIN, R. G., Ann. Rev. Biochem., 33, 23

(1964).19. NEWTON, W. A., BECKWITH, J. R., ZIPSER, D., AND BRENNER,

S., J. Mol. Biol., 14, 290, (1965).

by guest on April 11, 2018

http://ww

w.jbc.org/

Dow

nloaded from

L. N. OrnstonREGULATION

: IV.Pseudomonas putida-Ketoadipate by βThe Conversion of Catechol and Protocatechuate to

1966, 241:3800-3810.J. Biol. Chem. 

  http://www.jbc.org/content/241/16/3800Access the most updated version of this article at

 Alerts:

  When a correction for this article is posted• 

When this article is cited• 

to choose from all of JBC's e-mail alertsClick here

  http://www.jbc.org/content/241/16/3800.full.html#ref-list-1

This article cites 0 references, 0 of which can be accessed free at

by guest on April 11, 2018

http://ww

w.jbc.org/

Dow

nloaded from