The amino acid sequence of Rhodobacter sphaeroides dimethyl sulfoxide reductase

ARCHIVES

Vol.

320,

OF BIOCHEMISTRY

No. 2, July

AND

BIOPHYSICS

10, pp. 266-275,

1995

The Amino Acid Sequence of Rhodobacter Dimethyl Sulfoxide Reductase’

sphaeroides

Michael J. Barber,*gtv2 Hillary Peter J. Neame,**tr$ and Neil

J. Trimboli,”

Van Valkenburgh,* R. Bastiang

Anthony

Veronica

V. Pollock,”

*Department of Biochemistry and Molecular Biology, College of Medicine, and tlnstitute for Biomolecular Science, University of South Florida, Tampa, Florida 33612; $Sh rtners Hospital for Crippled Children, Tampa, Florida 33612; and $Division of Infectious Diseases, University of Utah School of Medicine, Salt Lake City, Utah 84132

Received

February

6, 1995,

and

in revised

form

April

17, 1995

The complete amino acid sequence of the soluble, monomeric molybdenum-containing enzyme dimethyl sulfozide reductase from Rhodobacter sphaeroides f sp. denitrificans has been determined using a combination of gas-phase Edman sequencing of isolated pep tides and direct sequencing of PCR products generated from R. sphaeroides genomic DNA. The protein comprises 777 residues corresponding to an apoenzyme molecular weight of 84,748 Da. The amino acid sequence was rich in Ala and Gly residues which represented 21% of the protein’s composition. The DNA sequence was 67% rich in G and C nucleotides. The amino acid sequence contained 10 cysteine residues which were relatively evenly distributed throughout the sequence and featured regions of sequence corresponding to the prokaryotic molybdopterin-binding signatures 2 and 3. While exhibiting limited sequence similarity to the corresponding membrane-bound molybdenum-containing subunit (DmsA) of Escherichia coli dimethyl sulfozide reductase, the Rhodobacter sequence showed extensive sequence similarity to that of the E. coli molybdoprotein, trimethylamine N-ozide reductase (torA). Comparison with other related prokaryotic molybdenum-containing enzymes indicated the presence of two highly conserved cysteine residues (Cys-268 and Cys-616) which may function in molybde0 1995 Academic Press. Inc. num coordination.

Sequence data from this article have been deposited with the GenBank/EMBL Data Library under Accession No. U25037. i This work was supported by Grants GM 32696 (M.J.B.) from the National Institutes of Health, the Institute for Biomolecular Science (M.J.B.), University of South Florida and Shriners Hospital for Crippled Children (P.J.N.) and a Biomedical Research Support Grant from the University of Utah (N.R.B.). 2 To whom correspondence should be addressed at the Department of Biochemistry and Molecular Biology, University of South Florida, College of Medicine, Tampa, FL 33612.

The reduction of dimethyl sulfoxide (DMS013 to dimethyl sulfide, two important components of the global sulfur cycle (11, is catalyzed by the enzyme DMSO reductase. DMSO reductase has been identified and isolated from several bacterial species including Escherichia latus

coli (2), Proteus vulgaris (31, Rhodobacter capsu(4), and Rhodobacter sphaeroides (5). In all cases,

the enzymes have been shown to require MO for activity, while more detailed characterization of the E. coli and R. sphaeroides enzymes has demonstrated that MO is present in combination with molybdopterin guanine dinucleotide (MGD), forming the mopterin guanine dinucleotide4 cofactor (6). Little structural information is available concerning the primary sequences of these enzymes which may assist in understanding how mopterin guanine dinucleotide is bound to these proteins. The cDNA-deduced amino acid sequence has been reported for E. coli DMSO reductase (7). This membrane-bound protein comprises three individual subunits, DmsA, -B, and C. DmsA (M, = 87,350 Da), which corresponds to the mopterin guanine dinucleotide-containing subunit and DmsB (M, = 23,070 Da), which contains four 4Fe-4S clusters, comprises the catalytically active portion of the enzyme, while DmsC (M, = 30,789 Da) functions as the membrane-anchoring domain securing the enzyme to the inner membrane in E. coli. In contrast, DMSO reductase from R. sphaeroides (M, = 85 kDa) has been demonstrated to be a soluble, monomeric enzyme located in the periplasmic space (5) and which contains mopterin guanine dinucleotide as the sole prosthetic group (8). 3 Abbreviations used: DMSO, dimethyl sulfoxide; PCR, polymerase chain reaction; BV, reduced benzyl viologen; PAGE, polyacrylamide gel electrophoresis. ’ The term “mopterin guanine dinucleotide” is used to denote the complex of molybdenum, molybdopterin guanine dinucleotide and associated protein ligands to molybdenum.

266 Copyright 0 1995 All rights of reproduction

0003-9861/95

$12.00

by Academic Press, Inc. in any form reserved.

DIMETHYL

SULFOXIDE

REDUCTASE TABLE

Sequences

Note.

Determined

for Peptides Derived DMSO Reductase

Fraction No.

Elution time (min)

Code

21 22 22 22 22 23 24 24 25 27 27

34.6 26.0 28.1 29.5 30.0 22.4 21.3 26.7 16.3 16.7 17.2

Dl D2 D3 D4 D5 D6 D7 D8 DS DlO Dll

27 27 27 27

18.8 19.7 21.2 23.5

D12 D13 D14 D15

29 29 29 29

13.9 15.1 16.8 18.3

D16 D17 D18 D19

29 29

18.8 22.1

D20 D21

29 29

22.8 25.1

D22 D23

31 31 31

15.3 16.4 17.2

D24 D25 D26

31

19.9

D27

31 31 31

20.7 22.7 23.5

D28 D29 D30

31 33

26.5 12.0

D31 D32

33

12.9

D33

33

14.8

D34

33

16.5

D35

33 33

17.7 18.8

D36 D37

33

22.2

D38

Lowercase

letters

identify

uncertain

residues.

AMINO

ACID

SEQUENCE

I

from Cleavage of Reduced with Endoprotease Asp-N

and Carboxymethylated

Sequence DLFAAYLTGESDGTPKTAEWAAEICGL DLVARELKRVQ DxFAVINP DGTPKTAEWAAEICGLPAAQIRELARSF DLFAAYLTGESDGTPKTAEWAAEICGLPAAQIRELA DLVARELKxVQESYGPTGTFGGSYGWREcGxFHHxSTxMRRLLNxA DDCPAHPT DLVAR DDCPAHPTWMEPAERLGGAGAKYPLHVVASHPKXRLHSQL DSIYSF’TRIKY DGMKAVEGAAVL DDCPAHFTWMEPAE DGLALGLIRDgiKAGPG EGLANxEVMSGCHWGVFKARVEgaa DLFAAY DPLFNPLGTP DYSTAAAQIIxxxVMGXTA DVGTSKLAQGNSGQTILA DHGAYAGMKALKEKGTRV EAAVKQAEFKVAMPSFE DSIYSPTPIKYPMVRREFLEKGVN DGLA DGGKAVEGAAWLSESGATSIPSARvG DYSTMAAQIIMPHVMfRLgVY DIFAALAERLGKGAEFTERGRR DRNRMLKAWEKLETFIVQg DPLFNPLGTPPGLIEIYSKkIEKMGY DEMLWISxFYxAxVKQAxxKxVxxPF DRGQILVGAKVSS DPLDPSEEGtK DIVLPATTSYERN DVGTSKLAQGNSGQTILA DPAVA DKLP DLYAVAGHEPclv DVEKYAGAPVtVINP DVKLAYgAGGNPFAHHQ DMLLNPG DYSTSAAVIIMPW DVALMLGMAHTLRSE DFWSEGIVEFPITEGANFVRY DFQWTATARHA DVLRVFN DIVLPATTSYERN DGDVLRVFN DPAPSHQLPGvWGs DFLARCTTGF DFLENF DVKLAYWAGE DLFAAYLTGG DAVMPGAIQIYEGGWY DAVMPGAIQIYEGGWY DVALMLGMAHTLrSEGW DFWSEGIV “x0 designates

unidentified

residues.

267

268

BARBER TABLE

ET

AL. II

Sequences Determined for Peptides Derived from Cleavage of Reduced and Carboxymethylated DMSO Redukase with Endoprotease Glu-C Elution time (min)

Fraction No.

Note.

Code

24 24 26 26 26

19.4 28.6 16.5 21.0 23.4

El E2 E3 E4 E5

28

12.9

E6

28

14.8

E7

30 30

13.9 15.1

E8 E9

30 33

19.9 15.8

El0 El1

Sequence KMGYDDCPAHPTWMEPAERLGGAGAKYPLHVVAHxxPKR LAxSFVAGRTMLAAGWSIQRMxGRAQASQHKF RLGGAGAKYPLxXWA VMSGCHWGVFKArVE SYGPTGTFGGSYGWke LYAVAHE LYAVAxFEPSPSDKDPADxAq DGRAVAExxxxxL GANFVRYADFRED LYAVAGHEPclin KGTRVICINPVRTE SIYSPTRIKYPMVRRE KGVNADRxTRGxV VKLAYxAGGNPFAHHQDRNRMLxAWE

QQTAQxW FKNVAMPSFEvMVRE

Lowercase

letters

identify

uncertain

residues.

“x”

designated

We have determined the complete amino acid sequence of the DMSO reductase isolated from R. sphaeroides f sp. denitrificans by a combination of direct peptide sequencing and PCR amplification of genomic DNA using oligonucleotide probes specific for selected portions of the internal amino acid sequence and have compared this sequence with those of other related molybdenum-containing enzymes to aid in identification of conserved structural features in this class of enzymes that may be critical to MO incorporation. MAW

AND

METHODS

Reagents. Proteolytic enzymes (endoprotease Asp-N, Glu-C, and Lys-C) were of sequencing grade and were obtained from either Boehringer-Mannheim (Indianapolis, IN) or Sigma Chemical Co. (St. Louis, MO). Cyanogen bromide was from Kodak (Rochester, NY). Reagents for DNA isolation were obtained from Promega (Madison, WI). Protein sequencing reagents and HPLC columns used for analysis of amino acid derivatives were obtained from Perkin-Elmer/Applied Biosystems Intl. (Foster City, CA). PCR reagents were obtained from Perkin-Elmer Cetus (Norwalk, CT), Boehringer-Mannheim and Promega Corp. GeneClean was obtained from BIOlOl (La Jolla, CA). pT7Blue was obtained from Novagen (Madison, WI). Growth ofbacterium. R. sphaeroides f sp. denitrificans IL 106 was grown photoheterotrophically using DMSO as the terminal electron acceptor as previously described (9). Enzyme purification. DMSO reductase was purified as previously described (9) with the addition of a final FPLC step using Mono-Q (0.5 x 5 cm) eluting the enzyme with a linear gradient of 0 to 0.3 M NaCl in 50 mM Tris buffer, containing 0.1 mM EDTA, pH 8.0. Fractions containing DMSO reductase and exhibiting BV:DMSO reductase activity (10) were pooled, concentrated by pressure filtration, and stored in liquid nitrogen until required.

unidentifies

residues.

Reduction, carboxymethylation, and peptide preparation. Samples of DMSO reductase (approximately 0.5-1.0 mg) were reduced and carboxymethylated as previously described (11). Proteolytic digests were performed as previously described (ll), using a variety of methods including endoprotease Asp-N, endoprotease Glu-C, and endoprotease Lys-C. In addition, cyanogen bromide was used as a specific chemical degradation method. Peptides were separated from the various digestion mixtures using a combination of FPLC gel filtration (Pharmacia Superdex 75, 1.2 x 30 cm) in 4 M guanidine hydrochloride or cation exchange (Pharmacia Mono-S, 0.5 x 5 cm) and reverse-phase HPLC (Vydac Cl8 column, 4.6 x 250 mm) in water/acetonitrile/trifluoroacetic acid. Utilization of two-dimensional chromatography provided both an estimation of length of the individual peptides and their acquisition in a solvent system compatible with direct application to the protein sequencer. Peptide sequencing and nomenclature. Gas-phase peptide sequencing was performed using both Applied Biosystems Intl. 477 and 473A peptide sequencers. Individual peptide sequences were referred to by an alpha-numeric designation which indicates the methodology used for protein cleavage and their elution order from the combined FPLC and reverse-phase HPLC separations (11). Thus for the peptide sequence defined as D21, the “D” indicates that the peptide was obtained from an endoprotease AspN digest of DMSO reductase. The numerical value “21” indicates that it was the 21st peptide sequence obtained from the individual FPLC fractions of the digest. Individual peptide sequences were aligned using commercial software (PC/GENE, version 7.0, Intelligenetics Inc., Mountainview, CA). In some cases, peptides appeared as a mixture of two or more sequences. These multiple sequences could usually be resolved into their individual components since one peptide was normally present in greater yield providing a dominant sequence or the sequence had been identified from a peptide obtained from a previous fraction. Oligonucleotide synthesis. Oligonucleotides were designed from the degenerate reverse-translated peptide sequences using commercial software (PC/GENE) and analyzed using the program “Primer Designer” (Scientific and Educational Software, State

DIMETHYL

SULFOXIDE

REDUCTASE TABLE

Sequences

Determined

Elution time (min)

Fraction No.

6 6 6 Note.

Lowercase

for Peptides Derived DMSO Reductase

letters

identify

Kl K2 K3 K4 K5 K6 K7

22.9

K8

24.3 15.6 24.6

K9 KlO Kll

29.7 19.1 15.5 17.7

K12 K13 K14 K15

19.1 22.1 24.1

K16 K17 K18 residues.

“x”

Note.

Fraction No.

Elution time (min)

24 27 27 29

23.4 23.1 26.1 27.1

Lowercase

letters

identify

uncertain

III

AVQ UQW TQEMVANRE YAGAPVTVFVFPT LAQGNTGQTILADVEK VSDAVMPGAIQIYEGGWYDPLDPSEEGTKDW DVNGDLSDc/qPILAFIPGTK VsDAVMPgA MGYDDCPAHPTWMEPAE YAGAPVTV DPAPSHQLPGDSIY AWEK EGLA DPAP GTRVIciNPVRTEsA YPLHV SWK EKGTREVICIK GTREVICIK YPLHVVASHPK ARVEdfra TNEIGWVIPDHGAYAGMK

designates

for Peptides Derived DMSO Reductase

unidentified

residues.

were purified using Geneclean and ligated into the vector pT7 blue. Sequencing was performed using fluorescent dye termination sequencing chemistry and analysis of the reaction products on an Applied Biosystems 373A DNA sequencer at the DNA Sequencing Core Laboratory at the University of Florida (Gainsville, FL). DNA sequences were analyzed using PC/GENE.

RJNJLTS

Peptide sequencing. Direct sequencing of the purified reduced and carboxymethylated DMSO reductase for 25 cycles of Edman degradation yielded the Nterminal sequence EGLANGEVMSGCHWGVF’KARV-

Iv

from Cleavage with Cyanogen

of Reduced bromide

Code

“x”

designates

and Carboxymethylated

Sequence

Ml M2 M3 M4

residues.

and Carboxymethylated

Sequence

TABLE

Determined

269

SEQUENCE

from Cleavage of Reduced with Endoprotease Lys-C

Line, PA). Oligonucleotides were codon-optimized using a R. sphaeroides codon usage table developed for the related molybdenum-containing enzyme biotin sulfoxide reductase (12) and were synthesized at the DNA Synthesis Laboratory at the University of Florida (Gainsville, FL). DNA isolation and PCR. Total genomic DNA was isolated from R. sphoeroides as previously described (13). Standard PCR reactions contained (100 ~1 total volume) 10 ~1 of 10 x PCR buffer with 1.5 mM MgCl,, 0.2 mM each dNTP, 1 pM each primer, 5- 10 pg genomic DNA, 5% DMSO, and 3 units of Tao DNA Polymerase. Cycles were performed with an initial 1 min denaturation step at 95°C followed by 30 cycles of 94°C for 1 min, 42°C for 1 min, 72°C for 2 min each and a final extension at 72°C for 7 min. Products of the PCR amplification were isolated by agarose gel electrophoresis. Excised bands

Sequences

ACID

Code

8.2 19.6 20.2 22.2 22.7 26.2 26.5

uncertain

AMINO

IGQIGLPGGGFGLSYHYSNGGSPT LLNPGGEFQFNGATATYPDVKLAYpAGGNPFA KKVVDPLYEARSDYDIFAALAERLGxGAEF RRLLNLAGGFVNSVGVYNTD GTAEVYE unidentified

residues

270

BARBER

0’+ -Al L I

21

I

24

I f

023

ET

AL.

by sequence analysis and alignment of overlapping peptides, obtained following digestion of the reduced and carboxymethylated protein with endoprotease AspN, Lys-C, and Glu-C, with the E. coli biotin sulfoxide reductase template sequence as shown in Fig. 2. A total of 106 peptide sequences were determined, resulting in the identification of 632 distinct amino acid residues. The N-terminal amino acid sequence of DMSO reductase was identified both from the undigested protein (Nl) and from peptide sequences obtained from AspN (D13) and Lys-C (Kll) digests. In addition, the Cterminal sequence, FVFPT, was also obtained from a peptide derived from the Lys-C (K4) digest. The majority of the peptide sequences that were obtained from the various proteolytic and chemical diges-

E. co/i biotin sulfoxide reductase residue number 1 N~I +pg ASQ-N

$

sequence

template

50 100 150 200 250 300 350 400 450 500 550 600 650 700 726 I I I I I I,, I,, , , , g ;7& q ~23~~g f3gg

;o

A % >

$

l

26

+

+>p

l

>

Elution Time (min) HPLC chromatography of gel filtration fracFIG. 1. Reverse-phase tions obtained from an endoprotease Asp-N digest of DMSO reductase. Reduced and carboxymethylated DMSO reductase (0.8 mg total protein) was digested with endoprotease Asp-N (0.8 pg) for 10 h at 37°C. The digest was fractionated on a Superdex 75 gel filtration column and individual fractions (0.5 ml) were analyzed using reverse-phase HPLC. Individual peptides were collected, lyophilized, and sequenced. The peptide maps obtained for FPLC fractions 21, 24, 29, and 31 are shown as representative examples. Individual peak labels correspond to the sequence identities given in Table I.

ENGR. A search of all the available protein sequences (GenBank or NBRF-Protein Identification Resource data bases) originally indicated this to be a unique sequence with little structural similarity to the known primary structures of either E. coli DmsA or any other molybdenum-containing enzyme. In contrast, proteolytic cleavage of the protein followed by isolation and analysis of individual peptides (the various sequences obtained are given in Tables I-IV while an example of the peptide maps produced using two-dimensional chromatography is shown in Fig. 1) indicated a significant degree of structural similarity to the amino acid sequence of E. coli biotin sulfoxide reductase (bisC) (14). The majority of the protein sequence was obtained

1117914 LY+c>>..

$y

10 .+

6 3;6

5;

PCR

FIG. 2. Peptide and direct PCR sequencing strategy from which the DMSO reductase sequence was obtained. The locations of the individual peptide sequences (Tables I-IV) that were used to derive the majority of the R. sphaeroides DMSO protein sequence are shown by solid arrows immediately under the line representing the E. coli biotin sulfoxide reductase template sequence. They are grouped into sequence obtained from the N-terminus of the intact protein (top arrow) and sequence obtained from peptides produced by digestion of the reduced and carboxymethylated protein with either endoprotease Asp-N (second group of arrows), endoprotease Glu-C (third group of arrows), endoprotease Lys-C (fourth group of arrows), and CNBr (fifth group of arrows). The PCR primers discussed in the text and shown in Table II and the PCR products whose sequences were used to complete the sequence and resolve any residue identity ambiguities are shown below (dashed arrows) with their corresponding sizes. In all cases, the length of the arrows approximates to the length of the individual peptide sequence obtained or the length of the PCR fragment.

FIG. 3. R. sphaeroides DMSO reductase consensus sequence. The DNA sequence obtained from the PCR products line of sequence information while the second line, shown in boldface, represents the consensus sequence obtained the DNA and peptide sequencing. ‘Ix” in the DNA sequence indicates unidentified bases.

is shown as the from a combination

first of

DIMETHYL

SULFOXIDE

REDUCTASE

AMINO

ACID

271

SEQUENCE

1 1

XXXXXXXXXXXXXXXXXXXXXXXGGGTGCCACTGGGGCGTGTTCAAGGCCCGG EGLANGBVMSGCHWGVPKAR

60 20

61 21

GTCGAGAACGGCCGCGCCGTGGCCTTCGAGCCCTGGGACACCCGCGCCGTCGCAC VBNGRAVAFE PWDKDPAPSH

120 40

121 41

CAGCTGCCGGGCGTGCTCGATTCGATCTATTCGCCCACGCCCGATGGTG QLPGVLDS IYSPTRIKYPMV

180 60

181 61

CGCCGCGAGTTCCTCGAGAGGNGTGAACGCCGCCGATCGCTCGACCCGCGG~CGGCGAT RRBFLEKGVNADRSTRGNGD

240 80

241 81

TTCGTCCGCGTCACCTGGNATGAGGCGCTCGACCTCGTGGCCCGCGAGCTG~GCGCGTT FVRVTWKEALDLVARBLKRV

300 100

301 101

CAGGAGAGCTACGGCCCCACCGGCACCTTTGGCGGCTTT~CGGCTCCTACGGCTGGAGNAGCCC~GC TGTFGGSYGWRSPG QBSYGP

360 120

361 121

CGGCTGCACAATTGTCAGGTCCTCATGCGCGCCGCGCGCTG~TCTGGNGGGCGGGTTCGTG RLHNCQVLMRRALNLAGGFV

420 140

421 141

AACTCNGTCGGGGANTATTCGACGGCGGGCGCGCAGATCAC NSVGDYSTAGAQI

480 160

481 161

AXXCTCGAGGTCTACGAGCAGCAGACCGCCGCCTGGCCCGTGGTGGTGGAG~~CCGATCTT TLEVYEQQTAWPVVVENTDL

540 180

541 181

ATGGTCTTCTTGGCCGCCGACCCGATGAAGACCAACGAC~CGAGATCGGCTGGGTGATCCCCGAC MVFWAADPMKTNBIGWVIPD

600 200

601 201

CATGGAGCCTATGCCGGCATGAAGGCGCGCTG~GGAG~GGG~C~GGGT~TCTG~TC HGAYAGMKALKEKGTRVICI

660 220

661 221

AACCCCGTGCGCACCGAGACGGCCGACTATTTCGACTATTTCGGCGCCGACGTCGTGTCGCCCCGGCCG NPVRTETADYFGADVVSPRP

720 240

721 241

CAGACCGACGTGGCGCTGATGCTCGGCATGGCGCACACGCC QTDVALML GMAHTLYSEDLH

780 260

781 261

GACAAGGACTTCCTCGAGAACTGCACCACGGGCTTCGGGCTTCGACCTCTTCGCGGCCTACCTGACC DKDFLENCTTGFDLFAAYLT

840 280

841 281

GGCGAGAGCGACGGCACGCCCAAGACGGCCGACGGCCG~TG~CCGCCGAGATCTGCGGCCTGCCG GESDGTPKTABWAAEICGLP

900 300

901 301

GCCGAGCAGATCAGGGAGCTCGCCCGCAGCTTCGTGGCCGGCTCGCCGCG RELARSFVAGRTMLAA A E Q I

960 320

961 321

GGCTGGTCGATCCAGCGGATGCACATGGCGCAACAGGCGCGCTCGTGACTCTG GW S I Q RMHMAQQAHWMLVTL

1020 340

1021 341

GCCTCGATGATCGGCCAGATCGGTCTGCCGGGCGGCGGCCTTCGGCCTCAGCTAC~CTAT ASMIGQ IGLPGGGFGLSYHY

1080 360

1081 361

TCGAACGGCGGCTCGCCCACGAGCGACGGCCCGGCGCTGGCTCGGACGGCGCC SNGGSPTSDGPALGGISDGA

1140 380

1141 381

AAGGCGGTCGAAGANGGCGCGGCCTGTCTGTCGGAGAGCGGCGCGACCTCGATCCCCTGC KARVBGAACL SBSGATSIPC

1200 400

1201 401

GCCCGCGTGGTAGACATGCTGCTCAATCCGGGCGGCGAGTTC~GTT~CGGCGC~CG FQFNGAT ARVVDMLLNPGGB

1260 420

1261 421

GCGACCTATCCCGACGTGACTGGCCTACTGGGCGGGCGG~CCCCTTCGCG~C~C ATYPDVKLAYWAGGN

IMPYVMG

P

F

A

H

H

1320 440

272

BARBER

ET

AL.

1321 441

CAGGACCGCAACCGGATGCTCAAGGCCTGGGAAAAGCTCGAGACCTTCATCGTGCAGGAC QDRNRMLKAWEKLETPIVQD

1380 460

1381 461

TTCCAGTGGACCGCAACCGCGCGCCACGCCGCCGA~TCGTCCTGCCGGCGACGACCTCCTAC PQWTATA RHADIVLPATTSY

1440 480

1441 481

GAGCGCAACGACATCGAGTCGGTGGGCGACTATTCGACTATTCG~CCGCGC~TCCTCGCGATG~G ERNDIESVGDYSNRAILAMK

1500 500

1501 501

AAGGTGGTCGATCCCCTCTACGAGGCCCGGTCGGACTACGCC KVVDPLYEARSDYDIPAALA

1560 520

1561 521

GAGCGCCTGGGCAAGGGCGCCGAATTCACCGAAGGGCGCGC ERLGKGAEPTEGRDEMGWIS

1620 540

1621 541

TCGTTCTACGAGGCTGCGGTGAAGCAGGCGGCGGAGTT~G~CGTGGCGATGCCGTCGTTC SPYEAAVKQAEFKNVAMPSF

1680 560

1681 561

GAGGATTTCTGGTCGGAAGGGATCGTCGAATTCCCGATCATTCGTC EDFWSEGIVEFPITEGANPV

1740 580

1741 581

CGCTATGCCGACTTCCGCGAGGATCCGCTGTTCCGCTGTT~CCCGCTC~~CGCCCTCG~CCTG RYADFREDPLFNPLGTPSGL

1800 600

1801 601

ATCGAGATCTACTCGAAGAACATCGAGAG~GATGGGCTATGATGATTGCCC~CC~TCCG IEIYSKNIEKMGYDDCPAHP

1860 620

1861 621

ACCTGGATGGAACCGGCCGAGCGTCTCGGCGGCGGCG~G~CG~TATCCGCTC~TGTC TWWEPAERLGGAGAKYPLHV

1920 640

1921 641

GTGGCGAGCCATCCGAAGTCGCGGCTGCACTG~CTGA VASHPKSRLHSQLTGT

1981 661

CTCTATGCGGTGGTGGGGCACGAACCCCTGCCT~T~CCCCGCCGATGCGGCCGCGCGC LYAVAGHEPCLINPADAAAR

2040 680

2041 681

GGCATCGCGGACGGCGATGTGCTGCGGGTGTTCAACGACCATCCTCGTGGGG GIADGDVLRVFNDRGQILVG

2100 700

2101 701

ACGAAGGTCAGCGACGCGGTGATACCGGGCGCGCGATCCAGATCTAC~G~CGGCTGGTAC TKVSDAVI PGAIQIYEGGWY

2160 720

2161 721

GATCCGCTCGATCCCTCGGAGGAGGGCACGCTCGCTCGA~GTACGGCGACGTG~CGTGCTG DPLDPSEEGTLDKYGDVNVL

2220 740

2221 741

TCGCTCGATGTCGGCACCTCGAAGCTGGCTGGCGCAGGA~CTGCGGCCAGAC~TCCTCGCG SLDVGTSKLAQDNCGQTILA

2280 760

2281 761

GATGTCGAGAAATATGCGGGCGCGCCGGTGACGGTGACCGTGDVPKYAGAPVTVFVFPT

2331 777

FIG.

S

L

R

D

1980 660

3-Continued

tions, with the exception of the 25 amino terminal residues, could be aligned using the E. coli biotin sulfoxide reductase sequence as the template. In combination, these peptide sequences yielded approximately 84% of the projected DMSO reductase sequence, shown in Fig. 3, based upon the length of the E. coli biotin sulfoxide reductase template. However, while many of these peptide sequences yielded regions of overlapping sequence, a number of gaps together with regions of sequence uncertainty were readily apparent which would have been potentially difficult to resolve using further pro-

tein digestion techniques. Thus, to determine these regions of missing sequence and resolve regions of residue uncertainty, PCR was utilized to identify the unknown residues and complete the sequence assembly. PCR strategy and DNA sequencing. To determine the regions of missing sequence, a number of PCR primers, shown in Table V, were designed to specifically amplify the majority of the DNA sequence using the minimum number of fragments and to produce regions of sequence overlap, as shown in Fig. 2. Regions of very low sequence similarity to both E. coli (14) and

DIMETHYL

SULFOXIDE

REDUCTASE

AMINO

ACID

273

SEQUENCE

TALtLFzV Oligonucleotide

Primers

Used

to Derive Sequence

Primer DMSORl

5' 5' 5' 5'M 5' 5' 5' 3' 5' 5' 5' 3'M 5' 3'P 5' 3'V

DMSORB DMSOR3 DMSORI DMSOR5 DMSORG DMSOR7 DMSOR8

Note. The primer to the sense strand

sequences of DMSO

are given reductase

PCR (IUPAC protein

Products

indicate attachment codons synthesized

R. sphaeroides biotin sulfoxide reductase (121, such as the N-terminal portion of the protein’s sequence, were used for primer construction. A limited number of PCR experiments using various primer combinations resulted in the production of DNA fragments of the anticipated size, following agarose gel electrophoresis. For example, the use of the oligonucleotides DMSOl and DMS05 as the sense and antisense primers, respectively, resulted in the amplification of an approximately 1.2-kb DNA fragment that was sequenced in both directions to yield the N-terminal portion of the sequence. Utilization of other appropriate oligonucleotide pairs (Table VI resulted in the amplification of approximately 0.8-, 0.6-, and 0.9-kb DNA fragments, respectively, that yielded the majority of the protein sequence up to within three amino acid residues of the carboxyl terminus. Assembly of these individual PCR fragment nucleic acid sequences generated the composite DNA sequence shown in Fig. 3. Thus, four PCR fragments produced 2292 bp of sequence corresponding to approximately 98% of the total DNA sequence. Some ambiguities may result from the direct sequencing of PCR products. However, in all cases, these could be resolved by either analyzing the sequence from the reverse direction, comparing the sequence with regions of overlap obtained from other PCR fragments or by comparison with specific peptide sequences. DISCUSSION

The preceding results represent the first complete amino acid sequence of a soluble, periplasmic-localized, molybdenum-containing DMSO reductase. The com-

Sequencing

code) direction/Coded sequence

GGITGYCAYTGGG 3' G C H WG ATGAAGGCI CTSAAG 5' K A L K GARTTCCAGTTYAA 3' E F Q F N GTTGRACTGGAACTC 3' E F Q F N GACATCGAGTCGGTGGGCGAC D I E S V G ACSACCTTCTTCAT 3' K K V V TTCCATCCA IGTSGG 3' T W M E ACGAASACSGTSAC 3' T V F V

such that 5’ primers DNA. Degenerate

for

of DMSO

Reductase Location (bp) 31 619

1237 1450

D

TAT Y

3'

1237 1495 1857

2308 to the anti-sense strand and 3’ primers indicate were Y(C/TT), S(C/G), RCA/G), and I (inosine).

attachment

plete protein sequence has been assembled from a combination of results obtained from two complementary techniques. The majority of the primary sequence has been obtained from direct gas-phase amino acid sequencing of the native protein and a large number of individual peptides generated using a variety of proteolytic cleavage methods. Results obtained from these methods have facilitated the identification of both the amino-terminal and carboxyl-terminal sequences in addition to extensive portions of internal sequence. To resolve any individual residue ambiguities and to determine the portions of sequence that were not identified by direct sequencing, PCR has been utilized to obtain the majority of the corresponding nucleic acid sequence using total genomic R. sphaeroides DNA. A minimal number of codon-optimized oligonucleotide primers have been synthesized and used which were derived from specific regions of the protein sequence that showed little structural similarity to the related sequence of R. sphaeroides biotin sulfoxide reductase (12) to avoid potential sequencing conflicts. This combination of amino acid and direct PCR sequencing has enabled identification of the protein’s sequence with minimal redundancy. The native protein has been demonstrated to comprise 777 amino acid residues corresponding to a calculated apoprotein molecular weight of 84,748 Da, while the full-length DNA sequence was calculated to comprise 2331 bp, including the N-terminal and C-terminal regions that were not directly identified. The calculated molecular weight of the monomeric protein was in excellent agreement with the value of 85 kDa, previously obtained from PAGE analysis (5).

274

BARBER

ET

AL.

RSDMSOREG-------LANGEVMSGCHWGVFKARVENGRAVAFEPWDKDPAFSHQLP BCTMNOR ECBSOR RSBSOR RSDMSOR ECl'MNOR ECBSOR RSBSOR

AAQA~.TDAVISF.EGILTGSH~~GAIRA~RFVAAKPFELDKYPS~IA MENS---------------------------------------~---LQS

-----Ho

---------MITTRVPHCSHWGAYTLLVDEGRIVGVEPFAHDPAPSELIH

VSWDEALDL

RSDMSOR BCTMNOR ECESOR RSBSOR RSDMSOR ECTMNOR ECBSOR RSBSOR RSDMSOR ECTMNOR ECBSOR RSBSOR RSDMSOR ECTMNOR BCBSOR RSBSOR

sig.

2----

RSOMSOR ECTMNOR ECBSOR RSBSOR RSDMSOR ECTMNOR ECBSOR RSBSOR RSDMSOR ECTMNOR ECBSOR RSBSOR RSDMSOR ECTMNOR ECBSOR RSBSOR RSDMSOR ECTMNOR ECBSOR RSBSOR

RSDMSOR ECTMNOR ECBSOR RSBSOR

RSDMSOR ECTMNOR ECBSOR RSBSOR

RSDMSOR ECTHNOR ECBSOR RSBSOR

RSDMSOR ECTMNOR ECBSOR RSBSOR

RSDMSOR ECTMNOR ECBSOR FSBSOR

RSOMSOR ECTKNOR ECBSOR RSBSOR

DP KG LD

KYAGAP-VTVEV YNGTV-EQVTA HGW-NTT-GDAGDAVR-

FIG. 4. Comparison of DMSO reductase and related MO-containing enzyme sequences. The amino acid sequences of R. sphaeroides DMSO reductase (RSDMSOR) and biotin sulfoxide reductase (RSBSOR) (12) and E. coli trimethylamine N-oxide reductase (ECTMNOR) (16) and biotin sulfoxide reductase (ECBSOR) (14) were aligned for maximum similarity. Regions of sequence conservation (identical and similar residues) in these four molybdoproteins are shown in boxes. The two conserved cysteine residues are indicated by ‘I*“, while the locations of the prokaryotic molybdopterin-binding signatures 2 and 3 are shown.

R. sphaeroides DMSO reductase contained 10 cysteine residues, at positions 12, 125, 219, 268, 297, 389, 400,616,670, and 754, that were relatively evenly dispersed throughout the sequence. Composition analysis indicated that the amino acid sequence of R. sphaeroides DMSO reductase contained a relatively high concentration of A and G amino acid residues which together comprised approximately 21% of the protein’s amino acid content. In contrast, the corresponding DNA sequence was relatively rich in G and C, these two bases accounting for approximately 67% of the experimentally determined DNA sequence. Analysis of the R. sphaeroides DMSO reductase sequence indicated the presence of a number of features that have been identified by characteristic signature sequences (15). Thus, the sequence contained regions that have been identified as prokaryotic molybopterinbinding oxidoreductase signatures 2 and 3, that corres-

ponded to residues 464 to 481 (TATARHADIVLPATTSYE) and 677 to 704 (AAARGIADGDVLRVFNDRGQILVGAKVS), respectively. Surprisingly, the R. sphaeroides DMSO reductase shared little sequence similarity with the primary structure of the other known DMSO reductase from E. coli (7). Comparison of the R. sphaeroides sequence with that of the MO-containing subunit (DmsA) of the E. coli enzyme indicated that only approximately 35% of the residues were conserved. However, comparison of the amino acid sequence of R. sphaeroides DMSO reductase with those of other known prokaryotic Mocontaining enzymes, indicated that the greatest degree of sequence similarity, corresponding to 59.9% identical and conserved amino acid residues, was to trimethylamine N-oxide reductase of E. coli (torA) (16), as demonstrated by the sequence alignment show in Fig. 4. In addition, the sequence was also similar to the se-

DIMETHYL

SULFOXIDE

REDUCTASE

quences of biotin sulfoxide reductase from both E. coli (14) and R. sphaeroides (121, though the similarity decreased to 58.3 and 51.0%, respectively. Multiple alignments indicated that the major region of sequence dissimilarity in these related proteins was confined to the amino terminus. Thus, R. sphaeroides DMSO reductase sequence contained a unique 26-residue aminoterminal extension sequence. Sequence alignments have facilitated the identification of two conserved cysteine ligands, at positions 268 and 616 in R. sphaeroides DMSO reductase, that are also conserved in both E. coli trimethylamine N-oxide reductase and biotin sulfoxide reductase and in R. sphaeroides biotin sulfoxide reductase. While the precise role of these highly conserved cysteine residues in the function of any of these proteins has not been established, they may potentially be involved in molybdenum coordination and the binding the MGD prosthetic group (17). The presence of two highly conserved cysteine residues has also been demonstrated in the molybdenum-binding domains of sulfite oxidase and assimilatory nitrate reductase (18). Determination of the R. sphaeroides DMSO reductase sequence has allowed us to extend this two conserved cysteine motif to a group of prokaryotic MGD-containing enzymes which may indicate a common mode of molybdenum coordination. REFERENCES 1. Lovelock, J. E., Maggs, 237,452-453.

R. J., and Rasmussen,

R. A. (1972)Nature

AMINO 2. Bilous, P. 1155. 3. Styrvold, 74-78. 4. McEwan, Ferguson, 5. Satoh, T., 6. Rajagopalan, 10199-10202.

ACID

275

SEQUENCE

T., and

Weiner,

0. B., and

Strom,

J. H. (1985)

J. Bacterial.

A. R. (1984)

Arch.

162,

1151-

Microbial.

140,

A. G., Wetzstein, H. G., Meyer, O., Jackson, J. B., and S. J. (1987) Arch. Microbial. 147, 340-345. and Kurihara, F. N. (1987) J. Biochem. 102,191-197. K. V., and Johnson, J. L. (1992)5. Biol. Chem. 267,

7. Bilous, P. T., Cole, S. T., Anderson, W. F., and (1988) Mol. Microbial. 2, 785-795. a. Johnson, J. L., Bastian, N. R., and Rajagopalan, Proc. Natl. Acad. Sci. USA 87, 3190-3194. 9. Bastian, N. R., Kay, C. J., Barber, K. V. (1991) J. Biol. Chem. 266,45-51. 10. Daniels, L., and Wessels, D. (1984) 237. 11. Neame, P. J., 20894-20901.

and

Barber,

12. Pollock, V. V., and Barber, 318, 322-332.

M.

M.

J., and

Weiner,

J. H.

K. V. (1990) Rajagopalan,

Anal.

Biochem.

141,

232-

J. (1989)

J. Biol.

Chem.

264,

M. J. (1994)

Arch.

Biochem.

Biophys.

13. Ausubel, F. M., Brent, R., Kingston, R. E., Moore, D. D., Seidman, J. G., Smith, J. A., and Struhl, K. (1990) Current Protocols in Molecular Biology, Vol. 1, Wiley, New York. 14. Pierson, D. E., and Campbell, A. (1990) J. Bacterial. 172, 21942198. 15. Bairoch, A. (1993) Nucleic Acids Res. 21, 3097-3103. 16. Mejean, V., Lobbi-Nivol, C., LePelletier, M., Giordano, G., Chippaux, M., and Pascal, M. C. (1994) Mol. Microbial. 11,11691179. 17. Barber, M. J., and Neame, P. J. (1990) J. Biol. Chem. 265, 20912-20915. 18. Garrett, R. M., and Rajagopalan, K. V. (1994) J. Biol. Chem. 269, 272-276.