Vol. 173, No. 19

Molecular Cloning and Sequence of the thdF Gene, Which IsInvolved in Thiophene and Furan Oxidation by Escherichia coli

KISWAR Y. ALAMt AND DAVID P. CLARK*Department of Microbiology, Southern Illinois University, Carbondale, Illinois 62901-6508

Received 19 November 1990/Accepted 24 July 1991

Our previous work resulted in the isolation of mutant strains of Escherichia coli K-12 which were able tooxidize furans and thiophenes as a result of mutations in several novel genes. Some of the genes involved inthiophene oxidation were cloned into the multicopy vector pUCl9. The plasmid pKA10 carries a 3.8-kbchromosomal fragment which encodes a previously undiscovered gene involved in thiophene oxidation. Threeproteins with approximate molecular sizes of 48, 30, and 26 kDa were overproduced by cells carrying pKA10.Maxicell experiments and DNA sequence analysis indicated that the 48- and 26-kDa proteins are encoded bypKA10, whereas the 30-kDa protein is apparently chromosomally derived. A cassette specifying kanamycinresistance was inserted into various sites on pKA10. An insertion which abolished the 48-kDa protein alsoabolished thiophene oxidation. Chromosomal integration of pKA10::Kan allowed us to locate the chromosomalinsert of pKA10 at 84 min on the E. coli genetic map by transduction. Since no previously identified genesinvolved in thiophene metabolism are located in this region, we designated the gene for the 48-kDa protein asthdF. Sequencing of the 3.8-kb insert revealed an overlap of several hundred bases with the regulatory andstructural regions of the tnaA gene, which is also located at 84 min. The 26-kDa protein is probably truncatedtnaA protein. An open reading frame corresponding to the 48-kDa thdF protein was located next to the tnaAgene, which encodes tryptophanase, but was transcribed in the opposite sense.

The degradation of aromatic hydrocarbons by bacteria hasbeen extensively investigated, and plasmids carrying thetoluene or naphthalene degradation systems have been ana-lyzed genetically (14, 36). Much has been made of theecological importance of such degradative systems and theirpossible applicability in biodegradation and environmentalcleanup. However, many xenobiotics and pollutants are notmerely hydrocarbons but contain heteroatoms. Many dye-stuffs and detergents contain sulfonic acid groups, whereasmost coal and oil contains significant amounts of reducedsulfur moieties, including those in which the sulfur is anintegral part of the ring system (18). The simplest aromaticsulfur heterocycles are those based on the thiophene ringsystem. Although compounds containing thiophene ringsform a major fraction of the organic sulfur component offossil fuels such as crude oil and coal, they are rarely foundin living organisms (15, 26). One recently discovered excep-tion is caldariellaquinone, which contains a benzothiophenering system and is found in certain sulfur-metabolizingarchaebacteria (37).

Since sulfur dioxide emission from burning high-sulfurcoals is a major contributor to acid rain, it is important todevelop bacteria which are capable of efficiently removingthe sulfur from coal before combustion (11, 18, 26). Inor-ganic sulfur can be removed from coal by certain strains ofThiobacillus or Sulfolobus (15, 17, 18); however, the organicsulfur remains intransigent. Since high-sulfur Illinois coalstypically contain 60 to 70% of their sulfur in the form of theheterocyclic thiophene ring (3), we have started to investi-gate the biodegradation of derivatives of thiophene and thecorresponding oxygen heterocycle, furan (1, 9).Our previous work resulted in the isolation of a triple

* Corresponding author.t Present address: Department of Medicine, Medical College of

Ohio, Toledo, Ohio 43614.

mutant, NAR30, capable of oxidizing a range of furan andthiophene derivatives (1). Mutations at three novel loci,thdA, thdC, and thdD, were required together with consti-tutive mutations in the fadR (= "thdB") (28) and atoC (29)genes. However, NAR30 does not completely degradethiophenes or furans, and its oxidation of these compoundsis slow and inefficient. We decided to clone the thd genesboth to increase the efficiency of degradation and to inves-tigate the nature of the reactions involved.We cloned two different chromosomal fragments from

NAR30, both of which confer the ability to oxidizethiophenes and furans. One of these most likely carries thethdA gene itself (2). The other, described in this report,carries a novel chromosomal locus which when present inhigh copy mediates the oxidation of certain heterocyclicsubstrates. This new gene has been designated thdF.

MATERIALS AND METHODS

Bacterial strains and media. Bacterial strains are all deriv-atives of Escherichia coli K-12 and are described in Table 1.Rich broth contains (per liter) 10 g of tryptone, 5 g of NaCl,and 1 g of yeast extract. The minimal medium used was M9(25) for growth tests or medium E (34) for genetic manipu-lations. Sugars, succinate, glycerol, etc., were used at con-centrations of 0.4% as carbon sources, whereas aromaticand heterocyclic substrates were used at 0.1% (wt/vol), asmost of these are moderately toxic at higher concentrations.Amino acids (50 mg/liter) and vitamins (5 mg/liter) were usedwhen appropriate. Solid media contained 1.5% (wt/vol)Bacto Agar (Difco). The tetrazolium indicator plates usedwere modified from those of Bochner and Savageau (7) byusing M9 salts as the buffer and using aromatic substrates at0.1% final concentration. Antibiotics were used at the fol-lowing concentrations: ampicillin, 100 mg/liter; tetracycline,10 mg/liter; kanamycin, 25 mg/liter; and chloramphenicol, 30mg/liter.

6018

JOURNAL OF BACTERIOLOGY, OCt. 1991, p. 6018-60240021-9193/91/196018-07$02.00/0

SEQUENCING OF E. COLI THIOPHENE GENE 6019

TABLE 1. Bacterial strains and plasmids

Strain or Relevant characteristics reference

E. coliDC625NAR11NAR30NAR420

NAR710NAR820NK6659GM5180SG20253NAR955NAR967NAR968NAR977JW355

A31318404JRG997BW6159DC679DF968W1485

Plasmids

gyrA fadR atoC adhCthdA of DC625thdA thdC thdD of DC625zba-300::TnlO thdA+ ofNAR30

srl::TnJO recA of DC625recA::Cam of NAR420srl::TnJO recArecA: :Camzba-300: :TnJOzic-SOl: :TnJOthdF::Kan of DC679metE: :TnJOthdF::Kan of HfrCasnB relA thyA thi-J deo

zic-501: :TnlOmetE: :TnJO trxA: :Kanzie-296::TnJOilv metB ampAilv: :TnJOfadR atoC adhCzig: :TnJOWild type

pUC19 Cloning vector, AprpKA10 3.8-kb PstI chromosomal

fragment in pUC19pKA15 0.9-kb PstI chromosomal

fragment in pUC19pKA20 Kanamycin cassette in

BglII site of pKA10pKA30 Kanamycin cassette in

HpaI site of pKA10pKA35 1.6-kb ClaI fragment

deleted from pKA10pKA37 1.7-kb BamHI fragment

of pKA10 subclonedinto pUC19

pKA39 1.3-kb BamHI fragmentof pKA10 subclonedinto pUC19

1

1, 21P1 (SG20253) x NAR30

P1 (NK6659) x DC625P1 (GM5180) x NAR420N. KlecknerG. WalkerS. GottesmanP1 (JW355) x W1485Integration of pKA20P1 (A313) x W1485P1 (NAR967) x HfrCJ. Wechsler

M. RussellC. GrossJ. GuestB. WannerLaboratory collectionD. FraenkelB. Bachmann

35Fig. 1

2

Fig. 1

Fig. 1

This study

This study

This study

Plasmids and recombinant DNA procedures. ChromosomalDNA was isolated and purified as described by Sato andMiura (32). Plasmids were isolated by alkaline lysis and thenethidium bromide-CsCl density centrifugation as describedby Maniatis et al. (22). The rapid plasmid isolation procedureof Birnboim and Doly (6) was used for screening clones byrestriction analysis. Ligations with T4 DNA ligase andrestriction enzyme digests were performed under conditionsrecommended by the manufacturer (Bethesda ResearchLaboratories). DNA fragments were separated by electro-phoresis on 0.7% agarose gels in 89 mM Tris-borate-89 mMboric acid-2 mM EDTA. Transformation procedures were asdescribed by Hanahan (13). Plasmids constructed are de-scribed in Table 1.DNA sequencing. DNA sequencing was performed as in

our previous analysis of the adhE gene (12). In brief, DNA ofplasmid pKA10 was isolated and then digested with appro-priate restriction enconucleases. Fragments were ligatedinto the correspondingly digested single-strand sequencingvectors M13mpl8 and M13mpl9 (24, 27). Both strands of

DNA were completely sequenced by the technique of Sangeret al. (31).

Gel electrophoresis of proteins. Sodium dodecyl sulfate-(SDS)-polyacrylamide gel electrophoresis was done by themethod of Laemmli (21). Whole-cell proteins were preparedby boiling in sample buffer for 5 min, and samples wereelectrophoresed on SDS-12.5% polyacrylamide gels. Molec-ular weight markers (97,400 to 14,400) were obtained fromBio-Rad.Plasmid-encoded proteins were labeled by the maxicell

technique of Sancar et al. (30). A 1-ml sample of culture(approximately 2 x 108 cells per ml) was labeled for 90 minwith 10 ,uCi of [35S]methionine. Proteins were separated onpolyacrylamide gels as above.

RESULTS

Cloning of thiophene degradation genes. ChromosomalDNA from the thiophene-degrading strain NAR490 wastotally digested with the restriction endonuclease PstI. Frag-ments were ligated into the PstI site in the polylinker of themulticopy vector pUC19 (35). The mixture of recombinantplasmids was transformed into strain NAR710, a recA deriv-ative of strain DC625, the fadR atoC parent of the furan-thiophene-degrading mutants. Ampicillin-resistant transfor-mants were selected on tetrazolium indicator platescontaining furfuryl alcohol. From among approximately1,000 transformants, we found 6 which gave a positiveresponse (i.e., a pink color) on the indicator plates. Aftersingle-colony purification, these were screened for the pres-ence of plasmids by agarose gel electrophoresis of minily-sates. Several plasmid-containing transformants were cho-sen for further work. These fell into two categories: thoseexemplified by pKA10 and those exemplified by pKA15.The two thiophene degradation-positive plasmids, pKA10

and pKA15, were purified by the standard CsCl procedure(see Materials and Methods) and subjected to restrictionanalysis. Plasmid pKA10 contained a 3.8-kb PstI insert,whereas pKA15 had a 0.9-kb PstI fragment. The detailedrestriction map ofpKA10 is shown in Fig. 1. Of the enzymesused, only BglI cut inside the 0.9-kb insert of pKA15. Sincethe restriction maps were not superimposable, it was clearthat the smaller chromosomal PstI fragment of pKA15 wasnot included within the larger PstI fragment of pKA10.

Physiological properties of pKA10 and pKA15. BothpKA10 and pKA15 were transformed into several recA hoststrains; the transformants were selected by their resistanceto ampicillin. Both plasmids conferred the ability to oxidizefurfuryl alcohol, thiophene-2-carboxylic acid, and severalother heterocyclic substrates on the fadR atoC strainNAR710. These plasmids were also tested in strain NAR820,a derivative of NAR30 which has the wild-type (i.e., degra-dation-negative) version of the thdA gene but still carries thethdC and thdD mutations. Whereas pKA10 had relativelylittle effect, pKA15 restored the degradative ability ofNAR820 to that of NAR30. The responses of these variousconstructs to a variety of heterocylclic substrates appear inTable 2. As indicated previously (2), it seems likely thatpKA15 carries the thdA gene. Here we are concerned withthe insert of pKA10, which apparently carries a novel geneinvolved in thiophene metabolism.

Insertion analysis of pKA10. The 3.8-kb insert of pKA10 isabout three times as long as an average gene. We thereforewished to locate the thd gene(s) more accurately in thissegment of DNA. Our approach was to cut open the plasmidat single restriction sites and insert a DNA cassette specify-

VOL. 173, 1991

6020 ALAM AND CLARK

Ii1\ Hinci Hp.!KpnI

Sma I Sm I

pKA30

FIG. 1. Restriction map of pKA10. Plasmid pKA10 was digestedwith a variety of restriction endonucleases, and the fragments wereanalyzed by agarose gel electrophoresis. The chromosomal insert isfrom coordinates 0 to 3800, and the vector (pUC19) is from 3800 to6600. The insertions of the kanamycin (Kixx) cassette to givepKA20 and pKA30 are shown.

ing resistance to kanamycin. This procedure was repeatedfor a selection of restriction sites around the target plasmid.The insertion derivatives were transformed back into therecA strain, NAR710, which is wild type for thiophenemetabolism, and transformants were selected by resistanceto ampicillin (on the original plasmid) plus kanamycin (on theinsert). The insertion derivatives were then examined for thepresence of an active thd gene(s).We opened up pKA10 using the enzymes BamHI, BglII,

and HpaI. Digested pKA10 was run on an agarose gel. Theband corresponding to linearized pKA10 was cut out, andthe opened-up plasmid was removed from the gel by theGene-Clean (Bio 101) procedure. The kanamycin resistance

TABLE 2. Responses of plasmid-bearing strains

Host Geno- . Substrate'Hostraa Gpen Plasmidstrain type FfOH TmOH TCA TMSO TMSO2

NAR710 Wild type None - - - - -pKA10 + ++ ++ ++ ++pKA20 - + - - -pKA30 ++ +++ ++ ++ ++pKA15 ++ +++ ++ + ++

NAR11 thdA None +++ +++ +++ +++ +++NAR30 thdACD None +++ +++ +++ +++ ++NAR820 thdCD None ++ + ++ + +

pKA10 ++ +++ + + ++pKA15 +++ +++ +++ +++ +++

a Strain NAR710 is a recA derivative of DC625, and NAR820 is a recAderivative of NAR420. All strains are derived from DC625 and contain thefadR, atoC, and adhC mutations.

b FfOH, furfuryl alcohol; TmOH, thiophene methanol; TCA, thiophene-2-carboxylic acid; TMSO, tetramethylene sulfoxide; TMSO2, tetramethylenesulfone. Responses: -, no reaction; +, ++, +++, increasing color ontetrazolium indicator medium, with the specified substrate.

cassette was cut out of plasmid pUC4::Kixx (5, 23) bydigestion with BamHI or SmaI. The cassette was thenligated into the linearized pKA10. After transformation andselection for ampicillin and kanamycin resistance, we iso-lated several derivatives of pKA10 carrying the kanamycincassette (Kan). Insertions into pKA10 opened by the restric-tion enzyme BamHI resulted in derivatives carrying Kanmostly in the BamHI site between Kpn and Pst at coordinate3550 (Fig. 1). These derivatives retained full activity of thethd genes. In addition, derivatives were found which had lostone or more of the small Bam fragments (coordinates 1650 to2250) and which were defective in thd activity. We alsoinserted the kanamycin cassette into the unique BgllI site (atcoordinate 1700) and isolated a plasmid, pKA20, with no thdactivity. This plasmid was subjected to restriction analysis toconfirm its structure.The kanamycin cassette of pKA20 could not be removed

by either BamHI or BglIl digestion since its construction hadproduced hybrid Bam-Bgl sites. Digestion with PstI gave acharacteristic 930-bp fragment from the interior of the Kan(Kixx) element together with the other expected fragments,i.e., a 2,700-bp vector fragment and two pieces of 2,750 and1,750 bp, each containing part of the chromosomal fragmentand an end portion of the Kan (Kixx) fragment. Since pKA20contains a single insertion which inactivates the thd gene,this must lie around the BglII site, approximately in themiddle of the chromosomal portion of pKA10. Insertions ofKan into the unique HpaI site (at coordinate 3000) to createpKA30 did not inactivate the thd gene(s) of pKA10. Table 2summarizes the responses of host cells with various plasmidderivatives toward furans and thiophene substrates.

Proteins of plasmid pKA10. We ran SDS-polyacrylamidegels to analyze the proteins produced by cells carryingplasmid pKA10 and its derivatives pKA20: :Kan and pKA30::Kan. Several gels were run, and we found several consistentdifferences between the parental strain and plasmid carryingderivatives. The gel shown (Fig. 2) compares NAR710 (arecA derivative of the parental strain) with the same straincarrying pUC19 (the vector), pKA10, and several deriva-tives of pKA10. Cells carrying plasmid pKA10 expressedthree prominent protein bands (a, b, and d in Fig. 2) whichwere largely absent in NAR710 or NAR710(pUC19).The a band is approximately 48 kDa. It was present in

pKA10 and pKA30 (both Thd+) but absent in pKA20(Thd-), implying that it is a product of a gene required forthiophene degradation. The 48-kDa protein was producedonly in late-exponential and early-stationary-phase cells(i.e., as in Fig. 2) but was not observed in early-exponential-phase cells (data not shown). The 26-kDa protein, band d,was highly expressed by cells carrying pKA10 and pKA37.Plasmid pKA37 carries the BamHI fragment from coordi-nates 1 to 1675 subcloned into pUC19 and expresses proteind but not a, b, or c. As shown below, this region of our DNAsequence overlaps the previously sequenced tnaA geneencoding typtophanase (10). It seems likely that protein d isa truncated tryptophanase gene product. Protein c (29 kDa)was observed only when those plasmids carrying the Kanelement were present and is presumably the neomycin-kanamycin phosphotransferase protein, the product of thenpt gene. The b protein, of approximately 30 kDa, provedsomewhat puzzling. This protein was especially prominentin cells carrying pKA35 and pKA39, which both contain theregion between coordinates 2250 and 3550, as well as thosewith the complete pKA10 insert. However, as discussedbelow, it does not appear to be plasmid encoded.To establish which proteins were actually encoded by the

J. BACTERIOL.

SEQUENCING OF E. COLI THIOPHENE GENE 6021

1 2 3 4 5 6 7 8 9 10

a-

b-c-d-

FIG. 2. Electrophoresis of proteins. Cultures of E. coli NAR710carrying pUC19, pKA10, and derivatives were grown overnight inLB containing ampicillin. Cells were lysed in sample buffer andelectrophoresed on an SDS-12.5% polyacrylamide gel. Proteinbands were visualized by Coomassie blue staining. Lane 5, molec-ular size markers: phosphorylase B, 97.4 kDa; bovine serum albu-min, 66.2 kDa; ovalbumin, 42.7 kDa; carbonic anhydrase, 31.0 kDa;soybean trypsin inhibitor, 21.5 kDa; lysozyme, 14.0 kDa. Lanes: 1,pUC19; 2, pKA10; 3, pKA30; 4, pKA20; 6, pKA39; 7, pKA37; 8,pKA35; 9, pKA15; 10, NAR710 with no plasmid. Band a, 48-kDathdF protein; band b, 30-kDa protein; band c, 29-kDa npt protein(neomycin-kanamycin phosphotransferase); band d, 26-kDa protein.

plasmid, we performed maxicell experiments. As shown(Fig. 3), only the prominent 48-kDa band (a) and a muchfainter 26-kDa band (d), together with the plasmid-derivedP-lactamase (31 kDa), were observed from cells carryingpKA10. Plasmid pKA20 showed ,-lactamase, npt protein(c), and enhanced production of protein d (26 kDa). The30-kDa protein (b) observed in extracts of unirradiated cellswas not observed in maxicell experiments and is presumably

4 3 2 1

a-

d- _

FIG. 3. Protein analysis by maxicell technique. Plasmid-encodedproteins were labeled with [35S]methionine as described in theMaterials and Methods. Lanes: 1, NAR710 (pKA10) labeled for 24h; 2, NAR710 (pKA10) labeled for 90 min; 3, NAR710 (pUC19)labeled for 90 min; 4, NAR710 (pKA20) labeled for 90 min. Band a,48-kDa thdF protein; band b, P-lactamase (31 kDa); band c, nptprotein (29 kDa); band d, 26-kDa protein.

chromosomally encoded. The sequence data support theseobservations (see below).

Location of thdF on chromosome. To decide whetherpKA10 actually carried thdA or another thd gene controlledby thdA, it was necessary to find where on the E. colichromosome the DNA inserted in pKA10 had come from.The plasmid pKA20 contains Kan inserted into the thd

gene of plasmid pKA10, causing loss of the thiophene-positive phenotype and the 48-kDa protein band. We there-fore recombined pKA20 with the chromosome to insert theKan marker at the original location of the gene(s) carried onpKA10-pKA20. Plasmid pKA20 was transformed into therecA+ gyr+ strain DC679, and transformants were selectedfor resistance to both ampicillin and kanamycin. ThepKA20-bearing derivative was then grown for many gener-ations in rich broth in the presence of kanamycin but withoutampicillin. At appropriate time intervals, aliquots were di-luted and plated on rich broth agar. A hundred colonies werepicked and tested versus kanamycin and ampicillin. Afterapproximately 100 generations, 9% of the colonies had lostthe plasmid, as evidenced by sensitivity to ampicillin yetretained kanamycin resistance. After 200 generations, 93%were Amps and Kanr. The Kan-carrying fragment had pre-sumably inserted into the host chromosome by reciprocalrecombination between the chromosomally derived se-quences surrounding Kan on pKA20 and their chromosomalhomologs (19). Several such Kanr derivatives of strainDC679 were purified and crossed with P1 grown onSG20253, which carries the zba::TnJO insertion close tothdA. No cotransduction was observed between zba::TnJOand Kan. We also tried TnWO insertions highly linked to thdCand thdD and again found no cotransduction.

It would appear that pKA10 carries a gene which gives athiophene-positive phenotype, at least when present in mul-tiple copies, but which is distinct from thdA, -C, or -D. Tomap this novel gene, designated thdF, we grew P1 onNAR967, a thdF::Kan derivative of DC679, and transducedthe thdF: :Kan insertion into HfrC. The thdF: :Kan derivativeof HfrC was conjugated with JC1552, a multiply marked,streptomycin-resistant, F- strain. Recombinants were se-lected for Strr and Kanr and tested for retention of theauxotrophic markers of JC1552. The results showed thatthdF lay closest to metB at 89 min on the chromosome (datanot shown). We then used P1 to transduce TnWO insertions inthis neighborhood of the chromosome into NAR967thdF::Kan. The thdF::Kan insertion showed 27% cotrans-duction with zie::TnJO of strain 18404, 16% cotransductionwith ilv::TnJO of strain BW6159, and 30% cotransductionwith the ilv mutation of strain JRG997. No cotransductionwas observed between thdF: :Kan and zic: :TnJO (strainNAR955), metE::TnJO (strain NAR968), or zig::TnJO (strainDF968). The results indicated a location close to the ilvGcluster at 84.8 min on the chromosome (4). It is clear fromthis that the thdF gene on pKA10 is quite distinct from thethdA gene which maps at 11 min (1, 9).Sequence of thdF region. DNA of plasmid pKA10 was

isolated and digested with suitable restriction endonu-cleases, and the fragments were ligated into correspondinglydigested M13mpl8 or M13mp19 (24). Both strands weresequenced by the method of Sanger et al. (31). The 3.8-kbchromosomal insert of pKA10 proved to be 3,774 bp long,and part of this sequence is shown in Fig. 4. When thissequence was compared with known sequences present inGenBank, we found an overlap of approximately 400 bp withthe tryptophanase (tnaA) gene of E. coli. The PstI site atcoordinate 0 of our insert corresponds to the PstI site at

VOL. 173, 1991

6022 ALAM AND CLARK

430 440 450 460 470 480 490 500* * * * * * * *

CATTACATAATCCTTCATTTATTTTAATTACAGTGATCCCTGTGAATATTACATCTGCTATACCGATTAATTCGCCAGAT<<<TnaA start codon

510 520 530 540 550 560 570 580* * * * * * * *

ATATATGCAAGCACCATCATTTACTAAAACAAGTAAATAAACGAGAGATGACCTTTGCAAAAGGCAAATTAAGATATAG

590 600 610 620 630 640 650 660* * * * * * * *

CAAACAAATAGAAC,ACATTGAATCAATCGGTTTGGCTCTGGTG&TGGCTACAGAAGGGCAAATCAAGGGCGGTGATCGAC

670 680 690 700 710 720 730 740* * * * * * * *

AATTTTATTGTCAATATTGAACCATTTTGAGGTCAACACATATATGTAAGATATTCATAATGCACTTATCCTCGCAAGAC

750 760 770 780 790 800 810 820* * * * * * * *

ACGAGGCTACATTATTACACTCAGAGTTTTTGTCTATTCTGAAATTGTTTAAATGTGAATCGAATCACAATCGTTCGGGG

830 840 850 860 870 880 890 900* * * * * * * *

AGC3AATATTAAATGCTTTTTTAA TTTAACG=GTTGAGGAATAACAG

910 920 930 940 950 960 970 980* * * * * * * *

GAGCGGTAACAATGAAGCGAAGCGCATCTGACAGTCAGAATGCGGCTGTATGCGCGGTTTACTTGATGTCTTCGACCTCT

990 1000 1010 1020 1030 1040 1050 1060* * * * * * * *

TTTACGCGGGGTCGTCCAGCAGACTTCATTTAAGGGGCCACTAAATGGAATTGCAAGACGACACGGTCTGCGTTGAGAAGMetGluLeuGlnAspAspThrValCysValGluLys

1070 1080 1090 1100 1110 1120 1130 1140* * * * * * * *

CGGTCGTCAAGTGGACGGGTCCGAGGGTTGTCAACGCGAAACGGGACACATTCCGGTCCTTCGGCGGAAGGTACAACCACArgSerSerSerGlyArgValArgGlyLeuSerThrArgAsnGlyThrHisSerGlyProSerAlaGluGlyThrThrTh

1150 1160* *

1170 1180 1190 1200* *

AGTTTCGGGTACGAGACAAACTCTACCAATCGTCGTGCAGGTGCGGAAGTGGTCAGGAACGGCTCTCTGCTTAATTGCGCrValSerGlyThrArgGlnThrLeuProIleValValGlnValArgLysTrpSerGlyThrAlaLeuCysLeuIleAlaH

1230 1240 1250 1260 1270 1280 1290 1300* * * * * * * *

ACTGGCAAGTGAAGTGCGTAAGGGTCGCAAGCGGCCACTACTGGTCATCGCCGGAACTACGCGTACGTGCTGCGTGCGCAisTrpGlnValLysCysValArgValAlaSerGlyHisTyrTrpSerSerProGluLeuArgValArgAlaAlaCysAla

1310 1320 1330 1340 1350 1360 1370 1380* * * * * * * *

.TATCCACATAGCCGGAATGCCGCTGCATATCATCGATACCGCCGGGCTACCTGCAGCCAGCGACGAAGTAGAACGTATTGTyrProHisSerArgAsnAlaAlaAlaTyrHisArgTyrArgArgAlaThrCysSerGlnArgArgSerArgThrTyrTr

1390 1400 1410 1420 1430 1440 1450 1460* * * * * * * *

GTATCGAGCGACGCGTGGCAGGAAATTGAACAGGCCGACCGCGTGCTGTTTATGGTCGATGGCACCACAACAGACGCCGTpTyrArgAlaThrArgGlyArgLysLeuAsnArgProThrAlaCysCysLeuTrpSerMetAlaProGlnGlnThrProT

FIG. 4. Sequence of thdF and intragenic region. The sequence of the whole chromosomal insert of pKA10 is shown from the start codonof the tnaA gene (10) through the intragenic region and the thdF open reading frame (coordinates 1031 to 2346). The putative stem-and-loopstructure (coordinates 840 to 889) is underlined.

coordinate 1850 (approximately) on the restriction map ofDeeley and Yanofsky (10) and lies about one-third of the waythrough the tnaA gene. Since tnaA lies close to 84 min on thegenetic map (4), this confirms our cotransductional mappingand also serves to locate our 3.8 kb ofDNA on the physicalmap of Kohara et al. (20).We found a reading frame, of the correct size to encode a

protein of approximately 48 kDa, centered about the BglIIsite (approximately coordinate 1700 bp) used to generate thethdF::Kan insertion of pKA20. The presumed thdF readingframe runs from coordinate 1031 to 2346 and is the gene nextto tnaA but is transcribed in the opposite direction (Fig. 4).No homology to any sequences in GenBank was found withthe thdF region or the other portion of the chromosomalinsert of pKA10, except for the overlap with tnaA already

mentioned. No open reading frame of sufficient size toencode a 30-kDa protein (i.e., band b) was found in the2250-to-3550 region of the insert. The sequence of this latterregion has been omitted from Fig. 4.

DISCUSSION

Our previous work resulted in the isolation of a series ofmutants of E. coli with increasing ability to oxidize furan andthiophene derivatives. Mutations in three novel genes wereinvolved, thdA, thdC, and thdD, together with mutations infadR ("thdB") and atoC (1, 4, 9). To investigate this noveldegradative system further, we cloned some of the genesinvolved. We isolated two plasmids carrying chromosomalfragments from a thiophene-degrading strain both of which

1210 1220

J. BACTERIOL.

SEQUENCING OF E. COLI THIOPHENE GENE 6023

1470 1480 1490 1500 1510 1520 1530 1540* * * * * * * *

GGATCCGGCAGAGATCTGGCCGGAATTTATTGCCCGTCTGCCAGCGAAACTGCCGATCACCGTGGTGCGCAATAAAGCCGrpIleArgGlnArgSerGlyArgA3nL.uLeuProValCy3GlnArgAsnCysArgSerProTrpCy3AlaIleLySPro

1550 1560 1570 1580 1590 1600 1610 1620* * * * * * * *

ATGGATCCACAGCTTGTGGTAACTGGCTCTCATATCATCATCACCTTATGCTTGCTGCATCATGTACCCGCTGACCAAAGMetAspProGlnLeuValValThrGlySerHisIleIleIleThrLeuCysLeuLeuHisHiSValProAlaASpGlnAr

1630 1640 1650 1660 1670 1680 1690 1700* * * * * * * *

GCCAGTACACCTCCATGGCGAAGATGCGTATGTTGCAGCCGAAGATTCAGGCAATGCGCGAGCGTCTGGGCGATCACAAAgProValHi3LeuHi3GlyGluAspAlaTyrValAlaAlaGluAspSerGlyASnAlaArgAlaSerGlyArgSerGlnT

1710 1720 1730 1740*

1750 1760*

1770 1780

CAGCGTATCAGCCAGGAAATGATGGCGCTGTACAAAGCTGCGAAGGTCAACCGCTGGGCGGCTGCTTCCCGCTGCTGATChrAlaTyrGlnProGlyAsnAspGlyAlaValGlnSerCysGluGlyGlnProLeuGlyGlyCySPheProLeuLeuIle

1800 1810 1820 1830 1840 1850 1860* * * * * * * *

CAGATGCCAATCTTCCTGGCGTTGTACTACATGCTGATGGGTTCCGTTGAACTGCGTCAGGCACCGTTTGCACTGTGGATGlnMetProIlePheLeuAlaLeuTyrTyrMetLeuMetGlySerValGluLeuArgGlnAlaProPheAlaLeuTrpIl

1870 1880 1890 1900 1910 1920 1930 1940* * * * * * * *

CCAAGTAAAATTCGCTGGGGTGGAGCACGTGGGGGTTCATACGTTAATAGGCCTATTCCGCTACCGTAACCGAGGGGGCGeGlnValLysPheAlaGlyValGluHisValGlyValHisThrLeuIleGlyLeuPheArgTyrArgAsnArgGlyGlyV

1960 1970*

1980 1990*

2000 2010 2020* *

TATATATCTCAGTCCTTGTTCGATACCGCTATGTGGTCGGCTTACGCGGATTCAAAGGCCGCTATCGCTCGTACTATCAGalTyrIleSerValLeuValArgTyrArgTyrValValGlyLeuArgGlyPheLy3GlyArgTyrArgSerTyrTyrGln

2030 2040* *

2050 2060 2070 2080 2090 2100

CGGTTATGCTCCCTCATATCCGATACGCAGTACCAGAACGAGTCATTCTTCATGCGAGTAGTGCGAACGGGCTGTTCCGAArgLeuCysSerLeuIleSerAspThrGlnTyrGlnAsnGluSerPhePheMetArgValValArgThrGlyCysSerGl

2110 2120 2130 2140 2150 2160 2170 2180*, * * * * * * *

ACTGCGTCGACAAGATCGCAATCGCCTCGGCCTGGCTAAGTCAAGTTTATGCTTAAGAAACGCGCGTTCGGAAAACTCGCuLeuArgArgGlnAspArgAsnArgLeuGlyLeuAlaLysSerSerLeuCysLeuArgA3nAlaArgSerGluAsnSerH

2190 2200 2210 2220 2230 2240 2250 2260* * * * * * * *

ACCAGGGCGAGCAATTCCGCAGGCCGGGAATGGTCAGAATGCGTTTTAGACAGCAGGTCGAGGATCACCGGACCGCCATGisGlnGlyGluGlnPheArgArgProGlyMetValArgMetArgPheArgGlnGlnValGluAspHisArgThrAlaMet

2270 2280* *

2290 2300* *

2310 2320 2330 2340

ACCTTGCAGTTCCAGCACATCTTCGCCGGTGCGAGTTCGGGCCAGGGAACCACAGCGTCTTTATACGGAAGATACTCGTCThrLeuGlnPheGlnHisIlePheAlaGlyAlaSerSerGlyGlnGlyThrThrAlaSerLeuTyrGlyArgTyrSerSe

2350

GTAGCr

conveyed the phenotype of a thdA mutation upon a recAderivative of the wild-type strain DC625 (2). However,although there is only one thdA gene, the two clones carrieddistinct DNA segments.

In this study, we demonstrated that the chromosomalinsert ofpKA10 derives from the 84-min region of the E. colichromosome and that it encodes a novel gene designatedthdF. Plasmids carrying an intact thdF gene confer a

thiophene oxidation-positive phenotype, whereas disruptionof thdF by inserting a kanamycin cassette at a unique BglIIsite abolishes the thiophene phenotype. A protein of approx-imately 48 kDa was observed both in standard cell extractsand by the maxicell technique. This protein is abolished bydisruption of the thdF open reading frame and is of theappropriate size. A 30-kDa protein was also observed in cellscarrying pKA10 but was not seen in maxicells nor was there

an appropriate open reading frame found upon sequencingthe insert of pKA10. We presume this protein is somehowinvolved in the thiophene response but is chromosomallyencoded. A third protein (26 kDa) was encoded by pKA10and is presumably a truncated version of the tnaA geneproduct.Sequence analysis showed an overlap of approximately

400 bp with the tnaA gene (10), serving to locate our segmenton the E. coli physical map (20). The thdF open readingframe (coordinates 1031 to 2346) is next to tnaA but tran-scribed in the opposite direction. Tryptophanase convertstryptophan to indole, releasing pyruvate and ammonia (33).Conceivably, thdF is an evolutionary relic of a disusedpathway for the further degradation of indole. Perhaps thethdF gene has been recruited for the breakdown of sulfurrather than nitrogen heterocycles in our thd mutants. Note

VOL. 173, 1991

1790

1950

6024 ALAM AND CLARK

that our thd mutants do not grow on indole nor do they showsignificant oxidation of indole itself or derivatives such asindole-2-carboxylate or indole-3-acetate when assessed intetrazolium indicator medium (data not shown). However,the thiophene-positive strains were more sensitive to indoletoxicity than wild-type strains, suggesting possible partialmetabolism of indole to some toxic intermediate(s).The intercistronic region between tnaA and thdF contains

the regulatory elements for tnaA (10). A putative Shine-Dalgarno sequence, TAAGGGG (consensus sequence TAGGAGG), for thdF can be seen at coordinates 1016 to 1022. Italso contains a possible stem-and-loop structure with a 20-bpstem, located at coordinates 840 to 889, approximatelyhalfway between tnaA and thdF. We found that both thethdF protein and the 30-kDa protein, as observed on poly-acrylamide gels, are much more prominent in cultures growninto the early stationary phase. Moreover, we have recentlyshown that oxidation activity toward thiophene derivativesin our E. coli strains is also greatest in the early stationaryphase (16). Exponential cultures, or those left in the station-ary phase for more than 24 h, show little activity (16). It hasrecently been demonstrated that the nah operon of somenaphthalene-degrading Pseudomonas strains shows a similarregulatory pattern (8). The regulatory significance of theputative stem-and-loop structure and the possible evolution-ary relationship of thdF to other aromatic degradative sys-temns requires further investigation.

ACKNOWLEDGMENTS

This work was supported by contracts DE-FG22-87PC79912 andDE-FC01-83FE60339 from the U.S. Department of Energy.

REFERENCES1. Abdulrashid, N., and D. P. Clark. 1987. Isolation and genetic

analysis of mutations allowing the degradation of furans andthiophenes by Escherichia coli. J. Bacteriol. 169:1267-1271.

2. Alam, K. Y., M. J. Worland, and D. P. Clark. 1990. Analysisand molecular cloning of genes involved in thiophene and furanoxidation by E. coli. Appl. Biochem. Biotechnol. 24/25:843-855.

3. Attar, A. J., K. Krebs, and P. B. Andersen. 1985. Organic sulfurfunctional groups in Ohio, Kentucky, Illinois and Iowa bitumi-nous coals, p. 12-16. In N. Berkowitz (ed.), First InternationalConference on Processing and Utilization of High Sulfur Coal.Elsevier Science Publishers, Amsterdam, The Netherlands.

4. Bachmann, B. J. 1990. Linkage map of Escherichia coli K-12,edition 8. Microbiol. Rev. 54:130-197.

5. Beck, E., E. A. Ludwig, A. B. Reiss, and H. Schaller. 1982.Nucleotide sequence and exact localization of the neomycinphosphotransferase gene from transposon Tn5. Gene 19:327-336.

6. Birnboim, H. C., and J. Doly. 1979. A rapid alkaline extractionprocedure for screening recombinant plasmid DNA. NucleicAcids Res. 7:1513-1523.

7. Bochner, B. R., and M. A. Savageau. 1977. Generalized indica-tor plate for genetic, metabolic and taxonomic studies withmicroorganisms. Appl. Environ. Microbiol. 33:434 444.

8. Burlage, R. S., G. S. Sayler, and F. Larimer. 1990. Monitoring ofnaphthalene catabolism by bioluminescence with nah-lux tran-scriptional fusions. J. Bacteriol 172:4749-4757.

9. Clark, D. P., K. Y. Alam, N. Abdulrashid, and B. Klubek. 1988.Successive mutation of E. coli for improved thiophene degra-dation. Appl. Biochem. Biotechnol. 18:393-401.

10. Deeley, M. C., and C. Yanofsky. 1981. Nucleotide sequence ofthe structural gene for tryptophanase of Escherichia coli K-12.J. Bacteriol. 147:787-796.

11. Detz, C. M., and G. Barvinchak. 1979. Microbial desulfurizationof coal. Mining Congr. J. 65:75-82.

12. Goodlove, P. E., P. R. Cunningham, J. Parker, and D. P. Clark.1989. Cloning and sequence analysis of the fermentative alcohol

dehydrogenase encoding gene of Escherichia coli. Gene 85:209-214.

13. Hanahan, D. 1983. Studies on transformation of Escherichia coliwith plasmids. J. Mol. Biol. 166:557-580.

14. Harayama, S., and K. N. Timmis. 1989. Catabolism of aromatichydrocarbons by Pseudomonas, p. 151-174. In D. A. Hopwoodand K. F. Chater (ed.), Genetics of bacterial diversity. Aca-demic Press, Inc. (London), Ltd., London.

15. Hoffman, M. R., B. C. Faust, F. A. Panda, H. HI. Koo, and H. M.Tsuchiya. 1981. Kinetics of the removal of iron pyrite from coalby microbial catalysis. Appl. Environ. Microbiol. 42:259-271.

16. Juhl, M. J., and D. P. Clark. 1990. Thiophene-degrading Esch-erichia coli mutants possess sulfone oxidase activity and showaltered resistance to sulfur-containing antibiotics. Appl. Envi-ron. Microbiol. 56:3179-3185.

17. Kargi, F. 1982. Enhancement of microbial removal of pyriticsulfur from coal using concentrated cell suspensions of T.ferroxidans and an external carbon dioxide supply. Biotechnol.Bioeng. 24:749-752.

18. Kargi, F. 1982. Microbial coal desulfurization. Enzyme Microb.Technol. 4:13-19.

19. Kiel, J. A. K. W., J. P. M. J. Vossen, and G. Venema. 1987. Ageneral method for the construction of Escherichia coli mutantsby homologous recombination and plasmid segregation. Mol.Gen. Genet. 207:294-301.

20. Kohara, Y., K. Akiyama, and K. Isono. 1987. The physical mapof the whole E. coli chromosome: application of a new strategyfor rapid analysis and sorting of a large genomic library. Cell50:495-508.

21. Laemmli, U. K. 1970. Cleavage of structural proteins during theassembly of the head of bacteriophage T4. Nature (London)227:680-685.

22. Maniatis, T., E. F. Fritsch, and J. Sambrook. 1982. Molecularcloning: a laboratory manual. Cold Spring Harbor Laboratory,Cold Spring Harbor, N.Y.

23. Mazodier, P., P. Cossart, E. Giraud, and F. Gasser. 1985.Completion of the nucleotide sequence of the central region ofTnS confirms the presence of three resistance genes. NucleicAcids Res. 13:195-205.

24. Messing, J. 1983. New M13 vectors for cloning. MethodsEnzymol 101:20-78.

25. Miller, J. H. 1972. Experiments in molecular genetics. ColdSpring Harbor Laboratory, Cold Spring Harbor, N.Y.

26. Monticello, D. G., and W. J. Finnerty. 1985. Microbial desulfu-rization of fossil fuels. Annu. Rev. Microbiol. 39:371-389.

27. Norrander, J., T. Kempe, and J. Messing. 1983. Construction ofimproved M13 vectors using oligo-deoxynucleotide-directedmutagenesis. Gene 26:101-106.

28. Nunn, W. D. 1986. A molecular view of fatty acid catabolism inEscherichia coli. Microbiol. Rev. 50:179-192.

29. Pauli, G., and P. Overath. 1972. ato operon: a highly induciblesystem for acetoacetate and butyrate degradation in Escherichiacoli. Eur. J. Biochem. 29:553-562.

30. Sancar, A., A. M. Hack, and W. D. Rupp. 1979. Simple methodfor identification of plasmid-coded proteins. J. Bacteriol. 137:692-693.

31. Sanger, F., S. Nicklen, and A. R. Coulson. 1977. DNA sequenc-ing with chain-terminating inhibitors. Proc. Natl. Acad. Sci.USA 74:5463-5467.

32. Sato, H., and K. I. Miura. 1%3. Preparation of transformingdeoxyribonucleic acid by phenol treatment. Biochim. Biophys.Acta 72:619-629.

33. Snell, E. E. 1975. Tryptophanases: structure, catalytic activitiesand mechanisms of action. Adv. Enzymol. 42:287-333.

34. Vogel, H. J., and D. M. Bonner. 1956. Acetylornithinase inEscherichia coli. J. Biol. Chem. 218:97-103.

35. Yanisch-Perron, C., J. Vieira, and J. Messing. 1985. ImprovedM13 phage cloning vectors and host strains: nucleotide se-quences of the M13mpl8 and pUC19 vectors. Gene 33:103-119.

36. Yen, K.-M., and C. M. Serdar. 1988. Genetics of naphthalenecatabolism in pseudomonads. Crit. Rev. Microbiol. 15:247-268.

37. Zhou, D., and R. H. White. 1989. Biosynthesis of caldariellaqui-none in Sulfolobus spp. J. Bacteriol. 171:6610-6616.

J. BACTERIOL.