microbial degradation of synthetic recalcitrant compounds - iupac

Biotech. Adv. Vol. 5, pp. 85-99, 1987 Printed in Great Britain. All Rights Reserved.

0734-9750/87 $0.00 + .50 C.pvri~hl ' Pergamon ~ournals Lid

MICROBIAL DEGRADATION OF SYNTHETIC RECALCITRANT COMPOUNDS BETSY FRANTZ, TERI ALDRICH and A. M. CHAKRABARTY Department

of Microbiology and I m m u n o l o g y . University of I l l i n o i s College of M e d i c i n e . Chicago. Illinois 60612. U S A

ABSTRACT Synthetic compounds, particularly highly chlorinated aromatics, comprise the bulk of the environmental pollutants that somehow must be removed from the environment.

Microbial degradation of such compounds is usually very slow, making them

highly persistent in nature. chlorination studies

are,

however,

demonstrate

Some synthetic compounds, with a lower degree of biodegradable;

biochemical,

genetic,

the evolution of new plasmid-encoded

and molecular

enzymatic

activities

specifically designed for the chlorinated substrates.

Nucleotide

sequences of

many of the genes encoding

activities

demonstrate

considerable

throughout

the molecules

homology

either

near

the

such enzymatic

active

sites

or

with

the

chromosomal genes encoding enzymes catalyzing analogous reactions. In some cases, unique

repeated

sequences,

reminiscent

of prokaryotlc

insertion sequence

ele-

ments, are present at or near the newly evolved genes. This suggests gene duplication and divergence as well as recombinational events mediated by transposable type elements as key ingredients in the evolution of new degradative functions. An understanding of such evolutionary processes is an essential feature for the development

of genetically-lmproved

bacteria

capable

of utilizing

and

thereby

removing highly chlorinated environmental pollutants from our environment.

KEY WORDS toxic chemicals, gene evolution, repeated sequences, plasmlds

INTRODUCTION Many synthetic compounds are released in the environment for household or industrial applications, resulting in a maze of governmental regulations to prevent the immediate toxic effects on human beings and animals (3,5).

85

The regulations

86

B E T S Y F R A N T Z et(21

are

usually

based

on

two major

toxicological properties

characteristics

and their fate

of

the

chemicals,

in the environment.

viz.

their

If a chemical

decidedly toxic, its release is highly regulated and seldom permitted.

is

Determin-

ation of the toxicity of a chemical, particularly its slow-acting effect on the human

immune

system

is,

however,

tricky,

variable

and

controversial,

adding

complexity to the regulatory maze (22) that often allows the chemical industry to continue

to dump

suspected

chemicals

into

the

environmental contamination is the accidental

environment.

Another

route

release of known hazardous

of

chem-

icals, demonstrated by the infamous Seveso and Bhopal incidences as well as the recent accident in Basle leading to the contamination of the Rhine river

(34).

Storage of toxic chemicals, usually as by-products of the chemical industry and the deliberate or accidental dumping of toxic chemicals in the environment have created

massive

pollution

problems

in

the

industrialized

countries

for which

there are no particular scientific answers or technologies available. 'Superfund'

program in the United States,

decontaminate from

the

toxic dump sites,

site

storage.

No

of

contamination

technology

is

aims only at hauling the to

some

presently

disposal of these toxic chemicals.

Even the

that is purportedly designed to help

EPA-designated

available

for

toxic chemicals 'safe'

the

site

safe

and

Since natural microorganisms

recycling all kinds of natural wastes but are usually unable

for

away their

economical

are known for

to attack highly

chlorinated synthetic chemicals, the toxic chemical pollution problem presents a major challenge to the biotechnology industry to develop innovative technologies for the recycling of some of these chemicals. discuss

how

new

degradative

mostly chlorinated aromatics, how

an understanding

of

functions

In this short article, we will

against

simple

chlorinated

compounds,

evolve in natural microflora and speculate as to

such

evolutionary

processes

may

help

us

develop

new

technologies for the recycling of similar, but more complex toxic chemicals from the environment.

It

is

known

that

while

simple

chlorinated

compounds

are

usually

biodegraded

rather rapidly, the highly chlorinated compounds in general are recalcitrant to microbial attack accomplished

(2,8).

either

by

Biodegradation of simple chlorinated compounds may be aerobic

or

anaerobic

microbial

consortia,

or

by

pure

cultures (13). It is often possible to isolate pure cultures capable of utilizing known

synthetic

compounds

such

as

2,4,5-trichlorophenoxyacetic

acid

(2,4,5-T),

1,4-dichlorobenzene or 6-aminonaphthalene-2-sulfonic acid from an initial mixed bacterial

community

through

prolonged

chemostatic

selection

chemical as the sole source of carbon and energy (19, 25, 35). bacterial

strains

capable of utilizing

a variety

with

the

target

While a number of

of xenobiotic

compounds have

been isolated this way, only a few bacterial cultures have been studied intensively with regard to the genetic and molecular basis of such degradation (27).

SYNTHETIC RECALCITRANT C O M P O U N D S

87

The compounds which are known to be degraded readily by pure cultures comprise the chlorinated benzoic and phenoxyacetic acids, more specifically 3-chlorobenzoate (3Cba) and 2,4-dichlorophenoxyacetate (2,4-D). Most of the genes required for the degradation of these two compounds are known to be borne on plasmids (10,13). There

are some

interesting

interrelationships

between

chromosomal

and plasmid

genes in allowing efficient biodegradation of these compounds. The mode of biodegradation of a naturally-occurring compound such as benzoate and two synthetic CHROMOSOME

i......... o~

pat27

DJP4

i coo.

cl

i............ i

.

............... i

O= ~0~ 0 ~ l~............. ~4 ............. ~.~ /

"]i"" .....

l ...........

@

O

c c, o oo ..

s.,.,..,.

,1 Fig. I.

Pathways for the degradation of benzoate, chlorophenoxyacetates and chlorobenzoates. The benzoate degradative genes are borne on the chromosome while the 2,4-D and 3Cba genes ere borne on plasmids pJP4 and pAC27, respectively. The plasmid-specified pyrocatechase II, cycloisomerase II and hydrolase II have high affinity for chlorinated substrates, while the corresponding chromosomally coded pyrocatechase I, cycloisomerase I and hydrolase I have high affinity for the non-chlorinated substrates with little or no activity towards the chlorinated compounds.

compounds vi__~z. 3-or 4-chlorobenzoate

(4Cba) or 2,4-D is shown in Fig.

i.

The

plasmid pJP4 is believed to encode the complete pathway for the degradation of 2,4-D (9,38), although plasmid mutations in all the catabolic steps have not been characterized so far (9).

In contrast, the plasmid pAC27 (Fig. i) encodes only a

partial pathway, vi_...~z,that of chlorocatechol. 3-chlorobenzoate

The first two enzymes that convert

to 3-chlorocatechol are specified by the chromosomal genes of

88

B E T S Y F R A N T Z etal

the host P. putida or A.

eutrophus cells.

Thus both the plasmid and the host

cell chromosomal genes are involved in the total degradation of 3-chlorobenzoate. Other plasmid genes may sometimes be needed for the degradation of other chlorinated compounds.

For example, for the degradation of 4Cba, where the chromosomal

benzoate oxygenase genes are of little value because of the limited substrate specificity

of

their

gene

products

(only

towards

3Cba but not

4Cba),

a TOL-

plasmld derived set of genes is needed for the conversion of 4Cba to 4-chlorocatechol, (7,15).

the

latter being

Consequently,

a

substrate

for

either chromosomal

the pAC27

plasmid encoded

enzymes

or other plasmid genes are sometimes

involved, in addition to a naturally-occurring degradative plasmid, for the total dissimilation of a chlorinated compound.

ORGANIZATION AND REGULATION OF THE CHLOROCATECHOL DEGRADATIVE GENES The pathways depicted in Fig.

i show some interesting characteristics regarding

the enzymes and the nature of their substrates.

For example, the pyrocateehase

II, eycloisomerase II, and the hydrolase II, encoded by plasmid pJP4 and involved in the degradation of chlorocatechols derived from 2,4-D, are also involved in the degradation of 3-chlorocatechol derived from 3Cba (9,38).

Thus, the presence

of pJP4 allows the host cells to utilize 3Cba, although the rate of such degradation

is very

genetic (14).

slow due

rearrangements

to

regulatory

before

plasmid

constraints, pJP4 will

which

allow

must

rapid

be

overcome

growth with

by

3Cba

Analysis of transposon-generated mutations in these genes (termed tfdC, D,

and E), has demonstrated the gene order tfdC, D, and E, similar to the order of the

enzymes

involved

in

the

2,4-D

degradative

pathway

(9).

An

interesting

question in this regard is how much homology the chlorocatechol degradativ~ tfdC, D, and E genes on pJP4 might have with the chlorocatechol genes present in the plasmld

pAC27.

It

should

be noted

that while

the

chlorocatechol

degradative

genes are present in a Pl-ineompatlbility 2,4-D degradative plasmid pJP4 to allow degradation of chlorocatechols derived from 2,4-dlchlorophenol,

the chlorocate-

chol genes borne on an entirely different plasmid pAC27 are expressed specifically to allow degradation of chlorocatechols derived from chlorobenzoates. theless,

homology

demonstrated

studies

homology

genes (13,14).

only

between

these

between

the

two

plasmids

fragments

(pJP4

harboring

and

the

Never-

pAC27)

have

chlorocatechol

Thus, the chlorocateehol genes may have a common ancestry, even

when present on two different plasmids under two different regulatory controls.

The pathways depicted in Fig. 1 also demonstrate the involvement of pyrocatechase I, cycloisomerase I and enol-lactone hydrolase of

non-chlorlnated

catechol.

Such

enzymes

(hydrolase I) for the degradation

are

known

to

be

specified

by

the

chromosomal genes, and have rather stringent specificity towards catecbol or its metabolltes.

Thus, Knackmuss, Reineke, and their associates

(32,33) have shown

SYNTHETIC RECALCITRANT C O M P O U N D S

89

the relative activity of pyrocatechase I towards 3-chlorocatechol at less than i% (relative to catechol) and of cycloisomerase I towards 2-chloromuconate at less than i% (relative to ci___.ss,cis--muconate). Enol-lactone hydrolase has very little activity towards the dienelactone. Such high specificity of the catechol degradarive enzymes readily explains the inability of natural catechol degrading strains to utilize

chlorocatechols.

In contrast,

the plasmid-specffied

degradative enzymes pyrocatechase II and cycloisomerase active

for the chlorinated

3-chloromuconate

substrates,

viz.

3- or 4-chlorocatechol

still retain substantial activity

chlorinated catechol or cis,cis-muconate.

chlorocatechol

II, while being highly

(50% or more)

and

2- or

for the non-

Dienelactone hydrolase, however, has

very little activity towards the enol-lactone

(32,33).

Such enzymatic activity

studies appear to indicate that the plasmid-coded pyrocatechase II and cycloisomerase II might have evolved from the respective analogous chromosomally-coded enzymes, while dienelactone hydrolase may have evolved independently.

To

obtain

an

degradative

insight

regarding

genes present

kilobase pair

the mode

of

on plasmid pAC27,

(kb) segment

evolution Ghosal

from plasmid pAC27

cells to utilize 3Cba slowly.

of

et al.

the

chlorocatechol

(14) cloned a 4.2

that allows the host P.

putida

This gene cluster was shown to have appreciable

homology with the i0 kb BamHI-EcoRl fragment of plasmid pJP4 that also harbors the chlorocatechol degradative genes.

Rapid growth with 3Cba ensured only when

the gene cluster was amplified to a copy number of 7 or 8 on the plasmid, and the amplification was found to be dependent on the recA + function (13). Amplification was

postulated

sequences

to

be

necessary

on this fragment.

due

to

the

A 4.3 kb Bglll

absence

of

the

regulatory

fragment containing

gene

two adjoining

Bglll fragments with a 385 base pair (bp) segment with the promoter sequences and the

larger

fragment

harboring

the

sequenced by Frantz and Chakrabarty

chlorocatechol (12).

genes

have

been

completely

Some of the interesting features of

this 4.3 kb BglIl fragment derived from plasmid pAC27 are given in Fig. 2. The steps involved in the conversion of chlorocatechol to maleylacetic acid (which is believed to be converted to 8-ketoadipate, presumably by a chromosomally coded maleylacetate reductase enzyme) require the participation of 3 key chlorocatechol (clc) degradative genes clcA, clcB, and clcD, encoding respectively the pyrocatechese If, the cycloisomerase II and the dienelactone hydrolase Cloning

of

this 4.3 kb

fragment

under

the

ta_..~cpromoter

(hydrolase II).

in broad host

range

plasmid pMMB22 followed by transfer and subsequent activation of the tac promoter in Escherichia coli resulted in the appearance of all three enzymes in E. coli. The enzymes were subsequently purified in the laboratory of Dr. L.N. Ornston, the N-terminal amino acid sequence determined, and the relative position of the clcA, B, and D genes deduced on the fragment by comparison of the N-terminal amino acid sequences derived

from purified enzymes with those predicted from the DNA se-

90

BETSY F R A N T Z et al.

cl

~>

~

o

(OnF3)

/\

**=,,r...* ......, Fig. 2.

The pathway and genes for chlorocatechol degradation. The steps A, B and D are mediated by ehlorocatechol degradative (cl___c_c) genes clcA, clcB, and clcD, respectively, encoding pyrocatechase II, cycloisomerase II and hydrolase II, on the 4.3 kb Bglll fragment of the plasmid pAC27. The location of the promoter upstream of the structural genes is shown by the arrow, and its homology with other promoter sequences of a number of operons known to be under positive control is shown at the bottom. PRCS represents positively regulated conserved sequences at the -I0 and -35 region present upstream of the structural genes of the positively-regulated operons. The ATG of the clcB gene overlaps with the TGA of the clcA gene as shown on the intersection of these two genes. The various restriction sites on the fragment are marked.

quences (11,24). As shown in Fig. 2, the 4.3 kb fragment contains, in addition to the

clcA,

B,

and D

genes,

another

function is presently unknown.

putative

open

reading

frame

(ORF3)

whose

The clcA and B genes overlap by 4 bp, and this is

believed to result in reduced expression of the clcB gene. The entire clcABD gene cluster appears to be regulated as a single unit, since hyperproduction of both clcA and clcD gene products occurs in E. coli or ~. putida only on activation of the ta___~cpromoter by isopropyl-B-D-thiogalactoside.

S1 nuclease mapping experi-

ments have demonstrated the site of transcription initiation on the adjoining 385 bp Bglll fragment, and the upstream sequences of the clcABD gene cluster show a good deal of homology with the putative promoter sequences of xylCAB and xylDEFG operons, which are under positive control.

A sequence comparison between these

operons, and the E. coli consensus positively-regulated conserved sequences (31) show

interesting

conservation

of

upstream

sequences,

Pseudomonas positively-regulated regulatory sequences

characteristic

of

(Fig. 2). Since the clcABD

gene cluster has previously been inferred to be under positive control

(13,14),

it would be interesting to find out if all positively-regulated degradative gene clusters sequences.

in

Pseudomonas

will

demonstrate

the

presence

of

these

conserved

SYNTHETIC RECALCITRANT C O M P O U N D S

91

EVOLUTION OF GENES FOR THE DEGRADATION OF SYNTHETIC C0b~0UNDS The determination of complete nucleotlde sequences for the clcABD gene cluster raises the interesting question of how much sequence identity such genes have with chromosomal genes encoding analogous reactions.

It is interesting to note

that the order of the clcABD genes is the same as the order of the degradatlve steps in the pathway, similar to the chlorocatechol genes (tf~CDE) pathway encoded by plasmid pJP4.

The nucleotide

sequences

in the 2,4-D

of the tfdCDE

gene

cluster have, however, not been determined as yet. It should be reemphasized that while clcA and clcB gene products have activities towards catecho]

and cis,cis-

muconate, which are normal substrates for catA and catB gene products, the clod gene product appears to catalyze a novel reaction.

How do such genes evolve in

nature? Since oxygenases similar to pyrocatechase II have previously been purified and their amino acid sequences determined (29), it is possible to make comparative studies on sequence homologies between pyrocatechase II and another chromosomally-coded oxygenase such as protocatechuate 3,4-dloxygenase, which is formed by self-assoclatlon of equal amounts of nonidentical a and 8 subunits. tion,

Aldrich

et

al.

(i) have

recently

cloned

and

completely

In addi-

sequenced

the

chromosomal catB gene from P. putida which has made a comparative study with the plasmid-borne clcB gene possible. amino

acid

sequence

homology

Frantz et al.

near

a catalytic

(ii) have also determined the site

Cys

residue

between

the

chromosomally encoded enol-lactene hydrolase and the plasmid-coded dlenelactone hydrolase.

The results of such comparative studies are shown in Fig. 3.

subunlts of protocatechuate

catechase II suggesting a common ancestry among them. the fact

The two

3,4-dioxygenase share extensive homology with pyro-

that pyrocatechase

II retains appreciable

This is also reflected by activity

towards

catechol.

Similarly, the 52% homology at the nucleotlde level between the chromosomal catB and the plasmid borne clcB gene (I) may indicate the evolutionary origin of clcB gene from an ancestral catB gene, with divergence throughout the gene segment. Such

divergence

may

still

allow

the

evolved

towards the non-chlorlnated cis,cls-muconate.

enzyme

to

retain

some

activity

In contrast, the hydrolases appear

to have diverged widely, since the N-terminal amino acid sequences of the enzymes are dissimilar. Both the hydrolases are, however, amounts

of ~-chloromercuribenzoate,

cysteinyl sequence closely

side

chain

lying at or near

surrounding resembles

hydrolase

suggesting

one

the

of

amino

the acid

the

cysteine sequence

inactivated by stoichiometric

that each hydrolase

active residues

site in

neighboring

(ii). The enol-lactone Cys-60

contains amino

a

acid

hydrolase

in dienelactone

(Fig. 3), suggesting that the catalytic region alone might have been

constrained

against

divergence because

of

its critical

role

in the enzymatic

catalysis. Various models have been proposed for the evolution of genes involved in the degradation of aromatic compounds and their metabolites in Pseudomonas and other

BETSY FRANTZ el al.

92

~; ~ ; ,~; ~': ~,~,~~ 7 , ~ h' i ~ : '. ~~~ t :~'[:j :~

Fig. 3.

Amino acid sequence homology between pyrocatechase II (PYRII) and protocatechuate dioxygenase subunits ~ and (PD= ,PDB), cycloisomerase II (MLE II) and cycloisomerase I (MLE I), and dienelactone hydrolase (DLH) and enol-lactone hydrolase (ELH). The amino acid sequence for PYRII, MLEII, MLEI and DLH has been determined from the DNA nucleotide sequences while that for PD=, PDB and ELH has been determined from the purified proteins.

soil bacteria (8,28,29).

Clear evidence for gene duplication followed by diverg-

ence has been presented for the evolution of nylB gene involved in the degradation of xenobiotic

compounds

such

as nylon

(23,26).

generally believed to be the result of divergent

Dissimilatory

evolution,

genes

are

and both clcA and

clcB genes might have evolved from the corresponding chromosomal genes by such an evolutionary process.

One of the models proposed for generating gene divergence

involves repeated recombinational events between misaligned DNA segments (28). In this model, double crossovers allow base pair substitutions within genes without altering substantially the translational reading frame or the length of the protein.

They result in nontandem direct repeats and rapid, irreversible divergence

from the ancestral sequence provided subsequent recombination with other genomes does not occur.

This model permits more rapid and drastic gene variation than

could be obtained extent Aldrich

by multiple

of divergence et al.

(i).

seen These

point mutations.

throughout

the

catB

two genes have

Such a model may and

numerous

clcB

genes

as

explain

the

reported

by

short duplicated

ranging from 6 to 13 base pairs throughout their lengths.

segments

The nontandem direct

repeats which have been identified in the first 200 bp of each sequence are shown in Fig. 4.

The repeats usually occur 2 to 4 times within their respective genes.

They are generally unique to the gene in which they are located indicating that they

were

acquired

by

intra-

rather

than

inter-gene

recombinations.

These

observations support the model presented above which could account for much of the sequence divergence among B-ketoadipate pathway genes.

SYNTHETIC RECALCITRANT C O M P O U N D S

93

ROLE OF REPEATED SEQUENCES IN THE EVOLUTION OF DEGRADATIVE GENES Repeated sequences have been discovered

in many prokaryotic

systems

including

Gram-negative and Gram-positive eubacteria as well as the archaebacteria.

Com-

pelling evidence for the role of such sequences in the evolution of degradative functions against synthetic compounds such as nylon oligomers has been provided by Negoro et al. (23) and Okada et al. (26). These authors characterized a plasmid in Flavobacterium sp. K172 which encodes the degradation of 6-aminohexsnoic acid cyclic dimer, a by-product of nylon manufacture, through elaboration of two newly evolved enzymes. and RS-II.

The plasmid contains two kinds of repeated sequence, RS-I

One of the two RS-II sequences, RS-IIA, contains the nylB gene while

the other, RS-II B, contains a homologous gene encoding an enzymatically-nonfunctional protein.

RS-I, which appears 5 times on the plasmid, is thought to be

involved in the rearrangement of the plasmid to translocate the proto-nylB gene, in the same way as insertion sequences mediate gene rearrangements. nylA

gene,

encoding

extremely

low enzymatic

activity

A duplicate

is also believed

to be

present on the plasmid, again suggesting gene duplications and further divergence for the evolution of this gene. Circumstantial evidence for the role of repeated sequences in the evolution of genes for the degradation of synthetic compounds comes from the characterization of such an element in the 2,4,5-T degrading strain of ~.

cepacia ACII00

(36). Lessie and Gaffney (21) have recently described the presence of a number of

P. putida catB ATGACAAGTGTGCTGATTGAACGTATCGAGGCAATTATTGTGCATGACCTGCCGACCA TTCGTCCGCCGCACAAGCTGGCGATGCACACCATGCAGACGCAGACCC~GT~-rGATTC GTGTTCGCTGCAGTGATGGCGTGGAAGGCATGGGCGAGGCCACCACCATGGCCGGCCTGG CCTATGGCTACCAAACGCCGGA

P. purida clcB i A~GAACA~CGAA~CATCGAT~GACGCZ~T~ACG~CCCACC~CCCGTCCCArCC

AGATGTCGTTTACCACGGTGCAGAAGCAGAGCTATGCGATCGTGCAGATCCGTGCGGGCG GGCTTGTCGGCATCGGCGAGGGCAGCAGCGTAGGTGGGCCGACTTCGAG~FTCCGAATGCG CTGAAACCATCAAGGTCATCAT

Fig. 4.

Nontandem direct repeats found in the P. putida catB (encoding cycloisomerase I) and clcB (encoding cycloisomerase II) genes. The first 200 bp of the coding strand of each sequence is shown beginning with the ATG start codons. Arrows above the sequences designate nontandem direct repeats. Letters above the arrows indicate pairs of repeats.

BETSY FRANTZ eta].

94

insertion sequence elements

in the genome of a strain of P.

cepacia, which

is

well known for its catabolic versatility. They have postulated that some of these insertion sequence elements may be involved not only in recombinational events, but

also

in

facilitating

gene

expression

by

providing

appropriate

promoter

sequences. The 2,4,5-T degrading strain of [. cepacia ACIlO0 was isolated after prolonged selection in the chemostat in presence of 2,4,5-T as its major source of carbon and energy (19).

It not only is able to utilize 2,4,5-T and 2,4,5-tri-

chlorophenol as its sole source of carbon and energy, but can completely mineralize a number of chlorophenols,

including pentachlorophenol,

although regulatory

constraints do not allow it to utilize pentachlorophenol as its sole source of carbon and energy (18). In an effort to identify and localize ACII00 genes associated with

2,4,5-T degradation,

transposon

insertion mutagenesis with

used to generate mutants blocked in the 2,4,5-T degradative pathway.

Tn5 was

One mutant,

PT88, was studied further since it produced a dark brown color in the culture medium

characteristic of

presence of 2,4,5-T.

chlorocatechol

or ortho-chloroquinone

accumulation

in

EcoR1-restricted plasmid and chromosomal DNA from PT88 were

probed with Tn5 DNA and the insertion was localized on the chromosome.

Cloning

of the kanamycln-resistance marker of Tn5 from this mutant allowed the isolation of a 6 kb Sall fragment that contained half of the Tn5 sequence kanamycin resistance gene and the other half of chromosomal

including the

sequences flanking

the insertion site (36). The flanking chromosomal DNA present on this fragment is believed to carry the insertion-lnactivated 2,4,5-T degradatlve gene, presumably coding for an enzyme involved in chlorocatechol metabolism.

When this fragment

was used as a probe against ACII00 chromosomal and plasmid restriction digests to identify the functional gene, a large number of bands lighted up, suggesting the presence of a repeated sequence at or near this gene (36).

This highly repeated

sequence (with at least 20 copies on the chromosome and 9 copies on the plasmlds) was

further

localized

to

a

1.27

kb

Sall-Pvull

restriction

fragment

which

no

longer contained the Tn5 sequence. Most interestingly, this repeated sequence did not hybridize with chromosomal DNA isolated from P. number of [. cepacia strains

aeru~inosa, P. putida or a

(36), suggesting that it is unique to the 2,4,5-T

degrading strain of P. cepacia ACt100. Is the presence of such a repeated sequence,

termed RSII00-1,

the context of evolution of the 2,4,5-T degradative genes?

relevant in

Although located near

one of the 2,4,5-T degrading genes, there are at least 20 copies of this element on the chromosome of ACII00 and therefore its location near a 2,4,5-T gene may be just coincidental. It is nonetheless important to note that the 2,4,5-T degrading ability

evolved

after months

of

selective

pressure

in the

chemostat,

and

the

repeated sequence has been undetectible in a large number of Pseudomonas species isolated from soil (37). and

the

repeated

It is therefore not unlikely that both the 2,4,5-T gene

sequence

element

may

have

nonpseudomonal

origin

and

genetic

SYNTHETIC RECALCITRANT COMPOUNDS

95

rearrangements may have involved transfer of genes from nonpseudomonal bacteria through transposable elements. sequence

element,

Tomasek

et

In order to define the nature of the repeated al.

(37) have

completely

sequenced

a number

of

chromosomal fragments harboring the repeat, and determined the sequence of the common element

(RSII00-I).

The structure of this element is shown in Fig. 5.

There is an 8 bp direct repeat, which is due to duplication of the host chromosomal target site. This is followed by a 39 bp terminal inverted repeat (38 bp in the other orientation) with a few mismatches.

There is an open reading frame of

1314 bp immediately bounded by the terminal inverted repeats interposed sequences).

(with 9 and 5 bp

The element has therefore all the structural character-

istics of a transposable element (20) which may also explain the large number of copies

of RSII00-I

on the chromosome

and plasmids

of ACII00.

The 8 bp host

chromosome duplication at the site of insertion is reminiscent of similar target duplications (17).

generated

upon

integration

of

prokaryotic

transposable

elements

That RSII00-I may indeed be a transposable element is seen from the amino

acid sequences predicted from the 1314 bp open reading frame.

A part of the 48

kilodalton polypeptide contains the typical DNA-blndlng domain as demonstrated by the presence

of a helix-turn-helix motif,

and the similarity of the critical

amino acid sequence with those encoded by the transposase gene (MuA) of phage Mu (16) and the lacR gene (30) is shown in Fig. 5. reading frame encodes element,

It is thus likely that the open

a transposase enzyme for efficient

translocatlon of the

although the origin of such an element in AC1100 and its role in the

recruitment of 2,4,5-T degradative genes remains undefined.

ebp

Fig. 5.

Im

19

2

,

I

Structure of a repeated sequence element present in multiple copies on the chromosome of [. cepacia ACII00. The open arrows represent the duplication of the host chromosome target site while the solid arrows represent the terminal inverted repeats with the number of bases designated on top. The thin lines represent sequences which are mismatches with the number of bases shown. The bottom part shows the amino acid sequence homology of the DNA-blndlng polypeptldes containing the helix-turn-hellx motif.

BETSY FRANTZ e t a l .

96

CONCLUDING REMARKS The

manufacturing,

essential

in

environmental remedied.

a

modern

distribution industrial

pollution,

Toxicological

that

not

studies,

and

use

of

synthetic

society,

have

created

only must

be

recognized,

which

have

been

the

chemicals,

severe but

focus

while

problems also

of

must

the

of be

chemical

industry and the government regulatory agencies simply help recognize the extent of the problem and seeking ACII00,

a

its causative

solution.

Laboratory

agent(s),

but do not address the problem of

selected

microorganisms

such

as

[.

cepacia

have been shown to be capable of removing large concentrations of the

toxic chemical (2,4,5-T) from the contaminated soil, and such cells appear to die off quickly once the target chemical is gone (13). Microbial degradation thus represents the major route to solving this problem, if only the problem of microbial recalcitrance to synthetic compounds is understood and overcome.

This short

review emphasizes the mode of evolution of new degradative functions in bacteria with the understanding that such studies will lead t2 the development of newer strains against newer synthetic chemicals through enhanced evolution of degradative genes under highly selective conditions in the chemostat. emphasized

(6) the roles played by microbial products

biopolymers in pollution control.

We have previously

such as surfactants

and

Selective evolution not only allows evolution

of the degradative genes, but also the ancilliary genes for the production of emulsifiers and surfactants that might prove to be important in enabling the host microorganisms to take up and utilize highly hydrophobic synthetic compounds (4). An understanding of the mode of evolution of new genetic functions in bacteria is an

essential

strains

prerequisite

capable

for

of enhanced

the

construction

degradation of

and

toxic,

development

of

microbial

synthetic pollutants

in the

environment.

ACKNOWLEDGEMENT This work was supported by a Public Health Service grant ES 04050 from the National Institute of Environmental Health Science and in part by a grant from the National Science Foundation (DMB-8514671).

REFERENCES i.

T.L.

Aldrich,

Cloning encoding

and

B.

Frantz,

complete

J.F.

nucleotide

cis,cis-muconate

Gill,

J.J.

sequence

lactonizing

Kilbane

and

determination

enzyme,

Gene 48,

A.M. of

Chakrabarty,

the

000-000

catB

gene

(1987),

in

press. 2.

M. Alexander, Biodegradation of chemicals of environmental concern, Science 211, 132-138 (1981).

3.

J.L. Badaraeco, Jr., Loading the Dice.

A Five-Country Study of Vinyl Chlor-

ide Regulation, Harvard Business School Press, Boston (1985).

SYNTHETIC RECALCITRANT COMPOUNDS 4.

S.

Banerjee,

S. D u t t a g u p t a

and A.M.

Chakrabarty,

97

Production

of

emulsifying

agent during growth of Pseudomonas cepacia with 2,4,5-trichlorophenoxyacetic acid, Arch. Microbiol. 5.

R. Brickman,

13__~5, 110-114 (1983).

S. Jasanoff and T. llgen, Controlling Chemicals.

of Regulation

in Europe

and the United

States,

Cornell

The Politics

University

Press,

Ithaca, NY (1985). 6.

A.M.

Chakrabarty,

Genetically-manipulated

microorganisms

and their products

in the oil service industries, Trends in Biotechnol. !, 32-38 (1985). 7.

D.K.

ChatterJee

specifying

and

A.M.

Chakrabarty,

total degradation

Genetic

rearrangements

in plasmids

of chlorinated benzoic acids, Mol. Gen. Genet.

188, 279-285 (1982). 8.

P.R.

Clarke and J.H.

Pseudomonas,

in The

Slater,

Evolution of enzyme structure and function in

Bacteria,

Vol.

X,

The

Biology

of

Pseudomonas,

J.R.

Sokatch (Ed.), Academic Press, NY, 71-144 (1986). 9.

R.H.

Don,

A.J.

mutagenesis

and

Weightman, cloning

dichlorophenoxyaeetic

N.-J.

acid

and

JMPI34(pJP4), J. Bacteriol. 161, 10. B. Frantz and A.M. Chakrabarty, Bacteria,

Knackmuss

analysis

and

K.N.

of the pathways 3-chlorobenzoate

Timmis,

Transposon

for degradation in

Alcali~enes

of

2,4-

eutrophus

85-90 (1985). Degradative plasmids

Vol. X, The Biology of Pseudomonas,

J.R.

in Pseudomonas, Sokatch

in The

(Ed.), Academic

Press, NY, 295-323 (1986). II. B. Frantz, K.-L. Ngai, D.K. Nucleotide hydrolase

sequence

and

Chatterjee,

expression

gene from Pseudomonas

of

L.N. Ornston and A.M. clcD,

a plasmid-borne

sp., J. Bacteriol.

Chakrabarty, dienelactone

169, 000-000

(1987),

in

press. 12. B. Frantz and A.M. mination

Chakrabarty,

of a gene cluster

Organization and nucleotide

involved

in 3-chlorocatechol

sequence deter-

degradation,

Proc.

Natl. Acad. Sci. USA, in press (1987). 13. D. Ghosal, I.-S. You, D.K. Chatterjee and A.M. Chakrabarty, Microbial degradation of halogenated compounds, Science 228, 135-142 (1985). 14. D. Ghosal, I.-8. You, D.K. ChatterJee and A.M. Chakraharty, Genes specifying degradation of 3-chlorobenzoic

acid in plasmids pAC27 and pJP4, Proc. Natl.

Acad. Sci. USA 82, 1638-1642 (1985). 15. J. Harmann, W. Reineke and H.-J. Knackmuss, Metabolism of 3-chloro, 4-chloroand 3,5-dichlorobenzoate

by a pseudomonad, Appl. Environ. Microbiol. 37, 421-

428 (1979). 16. R.M.

Rarshey,

E.D.

Getzoff,

D.L.

Baldwin,

Primary structure of phage Mu transposase:

J.L.

Miller

homology

and

G.

Chaconas,

to Mu repressor,

Proc.

Natl. Acad. Sci. USA 82, 7676-7680 (1985). 17. S. lida, J. Meyer and W. Arber, Prokaryotic

IS Elements,

Elements, J. Shapiro (Ed.), Academic Press, NY, 159-221

in Mobile Genetic

(1983).

BETSY F R A N T Z eI(~I

98

18. J.S. Karns, J.J. Kilbane, S. Duttagupta and A.M. Chakrabarty, Metabolism of halophenols

by

2,4,5-triehlorophenoxyacetic

cepacia, Appl. Environ. Microbiol. 4_~6, 1176-1181

acid-degrading

Pseudomonas

(1983).

19. J.J. Kilbane, D.K. Chatterjee, J.S. Karns, S.T. Kellogg and A.M. Chakrabarty, Biodegradation

of

2,4,5-trichlorophenoxyacetic

acid

by

a pure

culture

of

Pseudomonas cepaeia, Appl. Environ. Mierobiol. 44, 72-78 (1982). 20. N. Kleckner, Transposable genetic elements in prokaryotes, Ann. Rev. Genet. 15, 341-404 (1981). 21. T.G. Lessie and T. Gaffney, The

Bacteria,

Vol.

X,

Catabolic potential of Pseudomonas cepacia,

The

Biology

of

Pseudomonas,

J.R.

Sokatch

in

(Ed.),

Academic Press, NY, 439-481 (1986). 22. E. Marshall, Immune system theories on trial, Science 234, 1490-1492 (1986). 23. S. Negoro, S. Nakamura and H. Okada, DNA-DNA hybridization analysis of nylon oligomer-degradative

plasmid pOAD2:

ogous

oligomer

to the nylon

identification of the DNA region anal-

degradation

gene,

J. Bacteriol.

158,

419-424

(1984). 24. K.-L. Ngai, B. Frantz, D.K. Chatterjee, L.N. Ornston and A.M. Chakrabarty, Organization and characterization of the plasmid-borne chlorocateehol degradative genes cleA and clcB from Pseudomonas, manuscript in preparation, 1987. 25. B. Nortemann, J. Baumgarten, H.G. Rast and H.-J. Knaekmuss, Bacterial communities degrading

amino- and hydroxynaphthalene-2-sulfonates,

Appl.

Environ.

Microbiol. 52, 1195-1202 (1986). 26. H. Okada,

S. Negoro, H. Kimura and S. Nakamura,

plasmid-eneoded

Evolutionary adaptation of

enzymes for degrading nylon oligomers,

Nature 306,

203-206

(1983). 27. G.S. Omenn and A. Hollaender,

Genetic Control of Environmental

Pollutants,

Plenum Press, NY (1984). 28. L.N.

Ornston

and

W.-K.

Yeh,

Origins

of metabolic

divergence by sequence repetition, Proe. Natl. Acad.

diversity:

evolutionary

Sci. USA 76,

3996-4000

(1979). 29. L.N.

Ornston

metabolic

and

W.-K.

evolution,

Yeh,

Recurring

in Biodegradation

themes

and

repeated

and Detoxifieation

sequences

in

of Environmental

Pollutants, A.M. Chakrabarty (Ed.), CRC Press, Boca Raton, 105-126 (1982). 30. C.O.

Pabo and R.T.

Sauer,

Protein-DNA recognition,

Ann.

Rev.

Biochem.

53,

293-321 (1984). 31. O. Raibaud and M. Schwartz, Positive control of transcription initiation in bacteria, Ann. Rev. Genet. 18, 173-206 (1984). 32. W. Reineke and H.-J. Knackmuss, Hybrid pathway for chlorobenzoate metabolism in Pseudomonas sp. BI3 derivatives, J. Baeteriol. 142, 467-473 (1980).

SYNTHETIC RECALCITRANT COMPOUNDS 3 3 . W.

Reineke,

Microbial

degradation

of

halogenated

99

aromatic

compounds,

in

Microbial Degradation of Organic Compounds, D.T. Gibson (Ed.), Marcel Dekker, NY, 319-360 (1984). 34. V. Rich, Rhine Poilution: death of Europe's sewer?, Nature 324, 201 (1986). 35. G. Schraa, M.L. Boone, M.S.N. A.J.B.

Jetten,

A.R.W.

Van Neerven, P.J. Colberg and

Zehnder, Degradation of 1,4-dichlorobenzene by Alcali~enes sp. strain

A175, Appl. Environ. Microbiol. 5_~2, i374-1381

(1986).

36. P.H. Tomasek, B. Frantz, D.K. Chatterjee and A.M. Chakrabarty,

Genetic and

molecular basis of the microbial degradation of herbicides and pesticides, in Biotechnology Danforth

and

for M.R.

Solving Bakst

Agricultural (Eds.),

Problems,

Martinus

NiJhoff

P.C.

Augustine,

Publishers,

H.D.

Dordrecht,

355-368 (1986). 37. P.H. Tomasek, B. Frantz and A.M. Chakrabarty, Characterization and nucleotide sequence determination of a repeated sequence element isolated from a 2,4,5-T degrading strain of Pseudomonas cepacia, manuscript in preparation (1987). 38. A.J. Weightman, R.H. Don, P.R. Lehrbach and K.N. Timmis, The identification and cloning of genes encoding haloaromatic

catabolic enzymes and the con-

struction of hybrid pathways for substrate mineralization, in Genetic Control of

Environmental

Pollutants,

Press, NY, 47-80 (1984).

G.S.

Omenn

and A.

Hollaender

(Eds.),

Plenum