abortus protein - Science Direct

Gene, 63 (1988) l-9 Elsevier GEN 02292

The cloning, expression, abortus protein (Recombinant

DNA;

Bruceh

and nucleotide sequence of a gene coding for an immunogenic

fusion protein;

promoter;

signal sequence;

vaccine;

plasmid;

phage A vector)

John E. Mayfield”, Betsy J. Bricker b, Holly Godfrey”, Renae M. Crosby”, David J. Knight ‘, Shirley M. Hailing ’ *, Doris Balinsky ” and Louisa B. Tabatabai b a Department of Zoology, Iowa State University, Ames, IA 50011 (U.S.A.) Tel. (515)294-3316; b National Animal Disease Center, ARS, USDA, Ames, IA 50010 (U.S.A.) Tel. (515)239-8200, and e Molecular Genetics, Inc., Minnetonka, MN 55343 (U.S.A.) Tel. (612)935-7335

Received

5 August

Accepted

20 October

1987 1987

Received

by publisher

20 November

1987

SUMMARY

Bruceflu abortus is the causative agent for brucellosis test for the differentiation of vaccinated from infected vaccine are top priorities of the brucellosis research

in cattle and man. Development of a single diagnostic animals and the development of a nonviable ‘subunit’ program in the United States. Preliminary evidence

previously showed that a purified 3 1-kDa protein (thought to be localized at or near the bacterial cell surface) protects against experimental brucellosis in rodents. The gene for this 31-kDa protein has now been cloned in Escherichiu coli. The protein is expressed well, apparently from its native promoter, when placed in several different E. coli plasmids. The nucleotide sequence of the flanking and encoding sequences has been determined, and comparison with the N-terminal amino acid (aa) sequence of the mature protein indicates the presence of a putative 28-aa signal sequence. The availability of the 3 1-kDa protein free of Brucella contaminants now allows rigorous study of the immunological properties of this protein.

Correspondence to: Dr. J.E. Maytield,

Department

339 Science II, Iowa State University,

Ames, IA 50011 (U.S.A.)

of Zoology,

Abbreviations: found pairs(s);

Tel. (5 15) 294-6847.

aa, amino acid(s);

in the Bmcellu

* Present

address:

National

Animal

Ames, IA 50010 (U.S.A.)

Disease

Center,

ARS,

Tel. (515)239-8200.

otide(s);

0 1988 Elsevier

Science Publishers

B.V. (Biomedical

PAGE,

PBS, phosphate-buffered

polyacrylamide

buffer,

SDS, sodium dodecyl sulfate; USDA, of Agriculture;

protein;

0.15 M NaCl);

[ 1,designates

gel

saline (20 mM sodium

hydantoin;

Division)

pH 7.4,

bp, base

or 1000 bp; nt, nucle-

phosphate Department

0378-l 119/88/$03.50

frame;

extract;

CSP, cell-surface

G; kb, kilobase

ORF, open reading

electrophoresis;

1,3 1-kDa polypeptide

protein

BSA, bovine serum albumin;

IgG, immunoglobin USDA,

BCSP3

cell-surface

PTH,

phenylthioUnited

plasmid-carrier

States state.

2

INTRODUCTION

Brucellosis is a disease of cattle caused by B. adores and is characterized by abortion of the fetus and chronic infection of the mammary gland and lymph nodes (Subcommittee on Brucellosis Research, 1977). The disease is of major economic importance, and is targeted by the USDA for eventual elimination from the United States. Presently, no single test exists for the differentiation of infected from vaccinated cattle. Furthermore, the recommended vaccine in use is a live attenuated form, known as strain 19, which results in the shedding of organisms in a small percentage of vaccinated cattle, and is pathogenic for humans. Improved vaccines and/or diagnostic procedures would permit an accelerated eradication program and consequently save many years of effort and millions of dollars for both the public and the cattle industry. B. dmtus is a Gram-negative rod which persists in cattle as a facultative intracellular parasite of phagocytic cells (Collins and Campbell, 1982). The phylogenetic relationship of the organism to the enteric bacteria is unknown and prior to this study there was no indication of whether or not B. abortus promoter sequences would function in E. co/i cells. Because it is a Class-3 disease organism, direct handling of the organism is restricted and requires extreme care. As a result, study of biochemical and biolo~cal properties of specific ~~ceZ1~ proteins is greatly hampered. The ability to produce selected polypeptides in a nonpathogen, such as E. coli, provides a major opportunity for the study of the organism and the disease. It was previously reported that if methanol-killed B, abortus cells were incubated in a solution of 1 M NaCl plus 0.1 M sodium citrate, an extract (Bnncella CSP extract), was obtained which could be used to vaccinate lemmings (Tabatabai et al., 1979; Tabatabai and Deyoe, 1984), and mice (L&T., unpublished results). This extract contained a limited spectrum of polypeptides, as well as a small fraction of the cellular polysaccharides. The major protein in this extract (BCSP31) has an Mr of 31000 as determined by SDS-PAGE. Vaccination with the partially purified protein protects both lemmings and mice as effectively as the whole Bruce&z CSP extract (L.B.T., in preparation). These data make this protein particularly important to study further. The

cloning and expression of this gene in E. co/iprovides the opportunity to examine the biological properties of BCSP31 removed from all other contaminating Brucelfa macromolecules, and also provides a sufficient source of this protein to permit testing in cattle. This report describes the cloning and expression, and gives the complete nucleotide sequence of the gene coding for BCSP31.

MATERIALS

(a) Enzymes,

AND METHODS

DNA p~ri~cation

and cloning

Enzymes were purchased from commercial sources and used according to the manufacturers’ recommendations. Plasmid DNA was isolated according to Birnboim and Doly (1979), followed by a phenol-extraction step. Cloning protocols used are described elsewhere (Maniatis et al., 1982). (b) Preparation of antisera

Rabbit antisera against the whole BruceIJu CSP extract or against purified BCSP3 1 were prepared by injecting 50 to 500 pg of protein in Freund’s complete adjuvant subcutaneously at multiple sites on the back followed by two to four boosts with the same protein at two-week intervals using Freund’s incomplete adjuvant. All antisera were preadsorbed with E. coli antigens either by incubating the serum directly with E. coli cells for one hour, or by passing each serum through a Sepharose 4B affinity column coupled with an E. coli lysate (Young and Davis, 1983). (c) Antibody screens

Bacteriophage plaques were screened by a procedure which combined elements of several previously reported methods (Young and Davis, 1983; Woo, 1979). Briefly, a dried nitrocellulose filter which had previously been dipped in a 1: 10 dilution of an overnight culture of 4358 cells (Karn et al., 1980) was placed on a petri plate of freshly developed plaques until wetted. After each filter was marked with a needle, it was removed to a fresh nutrient plate plaque-side up and incubated overnight. A bacterial

3

lawn with replica plaques developed directly on the nitrocellulose. The filter was then washed five times in PBS containing 0.05% Tween-80 to remove the cells. It was then blocked for 15 min in PBS-I% BSA, followed by incubation for 2 h with a 1: 200 dilution of rabbit antiserum PBS-BSA. The filters were then washed five times, blocked for another 15 min with PBS-BSA, and incubated for 2 h with horseradish-peroxidase-conjugated goat anti-rabbit IgG (Cooper Biomedical) diluted 1 : 200 (final concentration of 60 pg/ml of antibody protein). Following five more washes, the color was developed with a solution of 0.06% 4-chloro, 1-naphthol and 0.012% hydrogen peroxide in 80% PBS-20% ethanol. Bacterial colonies were screened for the potential expression for foreign antigens by a similar procedure. Colonies were either grown directly on nitrocellulose paper incubated on top of nutrient agar, or dry nitrocellulose was laid onto a nutrient plate cont~ning colonies of interest. The nitrocellulose filter with adhering colonies was then suspended in a tank of chloroform vapor for 30 min, after which it was shaken for 1 h in 20 ml of PBS-BSA plus 5 mM MgCl,, to which 20 @gof pancreatic DNase I and 1 mg of lysozyme were added (Helfman et al., 1984). The filters were then washed and successively incubated with antisera as described for the plaque assay. Western blotting was performed by standard procedures (Towbin et al., 1979). Proteins separated by SDS-PAGE (Laemmli, 1970) were electrophoretitally transferred from slab gels to nitroceliulose in buffer containing 12.4 mM Tris base, 76 mM glycine, 0.05% SDS, and 10% methanol using the Hoefer ‘Transphor’ device (Hoefer Scientific Instruments). The nitro~ellulose paper was then probed with antisera as described above for the plaque assay. (d) N-terminal amino acid sequence analysis The N-terminal amino acid sequence was determined from recombinant BCSP31 which had been electroeluted from SDS-PAGE gel slices as described (Hunkapillar et al., 1983) using an electroeluter device from C.B.S. Scientific (San Diego, CA). After the dialysis step in the electroeluter, the protein was precipitated from solution by the methanol-chloroform procedure of Wessel and

Flugge (1984) adapted to accommodate a O.l-ml protein sample. The protein pellet was dissolved in 0.1 “i, NH,HCO, and 1 nmol was sequenced on a model 470A gas-phase protein sequenator (Applied Biosystems, Foster City, CA) using a double coupling protocol. PTH-amino acids were detected by high performance liquid chromatography on a model 120A on-line PTH-amino acid analyzer (Applied Biosystems, Foster City, CA).

RESULTS

(a) Gene identification and cloning A nucleotide sequence library was constructed by partially digesting B. abortus strain 19 DNA with the restriction endonuclease Sau3A. The resultant DNA fragments were fractionated on a sucrose gradient and fra~ents of 1.5-20 kb were ligated to the BamHI arms of phage I 1059 DNA (Karn et al., 1980). The DNA was packaged in vitro (Promega, Madison, WI) and amplified by plating the primary library at approx. 1000 plaques per 100-cm petri plate. The amplified library contained 20000 independent recombinant 1 phages at a titer of lo6 phages per ml. When screened with rabbit antisera prepared against the Brucella CSP extract, l-2% of plaques produced a positive response. Assuming that most of the proteins are expressed during ;t 1059 infection, this result is consistent with four to eight antigenic proteins in the Brucella CSP extract. Twenty positive plaques were screened by Westernblot probing with the same anti-Bnrcella CSP antiserum used to screen plaques. One recombinant /z clone produced a 31-kDa Bruceila protein which comigrated during SDS-PAGE with the major 3 1-kDa polypeptide in the Brucella CSP extract. The recombinant 1 DNA was subsequently cleaved with EcoRI and the resultant fragments inserted into the plasmid pKK223-3 (purchased from Ph~macia). Colonies were screened with anti-CSP antisera, and plasmids purified from positive clones. The plasmids from each positive colony contained a 4.2-kb EcoRI fragment. One of these was called pBA31-R7. A restriction map of this fragment is shown in Fig. 1. The fragment was subsequently subcloned into both pUC8 and pBR325 (Vieira and Messing, 1982;

4

H

R t

730 1000

I

H I

Hf

425 1250 1

Hf

400

1

Hf

Fig. 1. Restriction

map ofthe 4.2-kb Brucella DNA fragment

Hf, Hi&l; lengths

are given in bp.

Bolivar,

1978). Results of Coomassie

stained

gels and

indicated

that

antibody-probed expression

was

brilliant Western

not

I

840

which expresses

I

I

22001200

Hf

a 31-kDa Bruceila protein in E. coli. R, EcoRI; H, HindHI,

blue-

This band

blots

protein

most

probably

containing

represents

an uncleaved

a precursor

signal

sequence

appreciably

affected by the orientation of the fragment or the plasmid vector used. Thus, expression of this protein is apparently under the control of a Brucella DNA sequence which functions as a promoter in E. coli. Subclones of the fragments generated by Hind111 digestion showed no expression from the 2200-bp Hind111 or 730-bp HindIII-EcoRI fragment but weak expression from the 1250-bp Hind111 fragment. Full expression could be restored by rejoining the 1250-bp Hind111 fragment with the 730-bp HindIIIEcoRI fragment in the original orientation, but not in the reverse orientation (not shown). The protein expressed from the 1250-bp Hind111 fragment exhibited the same M= as that encoded by the full 4.2-kb EcoRI fragment. These results strongly suggested that the entire gene was located within the 1250-bp fragment and that the leftmost Hind111 site shown in Fig. 1 was located within upstream control sequences. This conclusion was later confirmed by nucleotide sequence analysis. (b) Protein expression and N-terminal sequence

355

Hf

JB

4

a-

amino acid

Comparison of total protein from E. coli cells containing the plasmid pBA3 l-R7 with control cells by SDS-PAGE clearly showed that the 31-kDa protein was one of the most prominent proteins in the cells (not shown). Furthermore, expression was dependent on the age of the culture with little expression in log-phase cells and increasing expression as the culture aged. Also, 24-h cultures exhibited variable amounts of 3 1-kDa protein in the medium. When Western blots were probed with anti-CSP antiserum, a fainter band at about 34 kDa was seen in whole-cell lysates, but not in growth medium (Fig. 2, lane 4).

Fig. 2. SDS-PAGE ofmedium

profile

and whole-cell

and

10% gels were run according cedures

are described

corresponding

Western

to Laemmli

in MATERIALS

AND

black-stained

SDS-polyacrylamide

(B) Duplicate

to panel A reacted with anti-CSP

and stained

with horseradish-peroxidase-conjugated

pro-

METHODS,

section c. (A) Naphthol-blue-amido from

cells.

(1970). Transfer

following

electrotransfer

blot

protein from JM105[pBA31-R7]

nitrocellulose gel.

rabbit antiserum goat anti-

rabbit IgG. Lanes 1 and 3, medium from 24-h culture; lanes 2 and 4, whole cells from same culture boiled in Laemmli (1970) sample buffer. The migration

distance

cated beside each panel.

of a 31-kDa polypeptide

is indi-

(Michaelis and Beckwith, 1982). The presence of a signal sequence is also clearly indicated by the nucleotide and amino acid sequences. A small amount of the mature recombinant BCSP31 protein was purified by electroeIution after SDS-PAGE and subjected to N-terminal amino acid sequence analysis on an Applied Biosystems model 470A gas phase protein sequencer. The resultant partial amino acid sequence agrees per-

A

B

(c) Nucleotide sequence The entire 1250-bp Hind111 fragment and 566 bp of the 730-bp ~j~dIII-~cuRI fragment from pBA31-R7 were sequenced by the methods of Maxam and Gilbert (1980) and Sanger et al. (1977) as outlined in Fig. 4. Primers for sequencing were synthesized using phosphoramidite chemistry (Beaucage and Caruthers, 1981) and a Microsyn1450 DNA Synthesizer (Systec, Inc. Minneapolis). The nucleotide sequence is given in Fig. 5 along with the predicted amino acid sequence. Examination of the sequence reveals a large ORF beginning with an ‘ATG’ 49 nt from the Hind111 site previously predicted to lie in the promoter region. The predicted protein has 329 aa and matches perfectly with the experimentally determined N-terminal amino acid sequence, provided the primary transcription product contains a 28-aa signal sequence. The predicted M, of the mature protein is 31521, in excellent agreement with the SDS-PAGE measurements. The predicted protein is slightly basic, contains 24% alanine plus glycine, and has no major hydrophobic sequence blocks except for the signal sequence.

66

29

Fig. 3. SDS-PAGE

fectly with the nucleotide sequence, provided a 2%aa signal sequence is allowed. The first 6 aa of this sequence are shown aligned with the nucleotide sequence and predicted amino acid sequence in Fig. 5. Rabbit antiserum prepared against BCSP31 puritied from the Bruce/la CSP extract (L.B.T. and B.L. Deyoe, in preparation) reacts strongly and specifically with both the recombinant 3 1-kDa protein and the major 31-kDa band in the Bmcella CSP extract (Fig. 3).

profile and corresponding Western blot of

Bruce& CSP extract and partially purified BCSP31 from E. coli cells. Gel and transfer conditions are the same as in Fig. 2. M,

standards: BSA, 66000; egg albumin, 45 000; glyceraldehyde3-phosphate dehydrogenase, 36000; carbonic anhydrase, 29 000; trypsinogen, 24000; trypsin inhibitor, 20 100; a-lactalbumin, 14200. (A) Naphthol-blue-black-stained nitrocelln~ose following electrotransfer from SDS-polyacrylamide gel. First lane, M, standards (scale given in kDa); second lane, Brucelh CSP extract; third lane, BCSP31 from E. cofi cells. (B) nitrocellulose with proteins electrotransferred from a gel duplicate to that shown in panel A probed with rabbit antisera made against BCSP31 purified from B. abortus followed by horseradishperoxidase-conjugated goat anti-rabbit IgG. First lane, Bruce& CSP extract; second lane, recombinant E. colt’extract.

DISCUSSION

A 3. amorousgene coding for a 3 1-kDa protein has been cloned in E. co/i. This gene expresses very well under the control of its own promoter when placed in a multicopy plasmid. The migration rate of this protein by SDS-PAGE is indistinguishable from that of BCSP31, a protein which can be isolated from a high salt extract of meth~ol-killed B. aborter cells. Rabbit antiserum prepared against the 3 I-kDa

6

A

0

566

.2!8,“1’.

H

_, _,425_, _,

;,I:, I:, I’/ I’/ /‘,/‘, 1 M

987

4% 145 , Hf I,

364

BHf

H bp

,\,‘,\,‘,~,‘,-\,‘,px,‘, J

S

Fig. 4. Strategy to determine the nucleotide sequence of the gene encoding BCSP31. (A) The bar represents the DNA flanking and encoding BCSP31; the coding region is shaded. Letters appearing above the line indicate restriction endonuclease sites. Symbols for the enzymes are: B, BssHII; H, HindIII; Hf, Hit&I. Numbers above the bar indicate length, in bp between restriction sites. The number below the bar is the number of bp encoding BCSP31. The M and S, which appear below the bar, indicate the start and stop codons, respectively. (B) The arrows indicate the length and direction of sequences determined by the method of Maxam and Gilbert (1977) after labeling at the indicated restriction sites. In some cases, fragments were subcloned before sequencing. (C)The arrows indicate the length and direction of sequences determined by dideoxy termination (Sanger et al., 1977). DNA primers were synthesized (Beaucage and Caruthers, 1981), annealed, and extended to determine the nucleotide sequence across the Hf sites (Sanger et al., 1977).

protein isolated from recombinant E. coli cells reacts well with both this protein and with BCSP31 from B. abortus. By Western-blot analysis an antigenically related 34-kDa minor protein is detected in whole E. coli lysates which is probably a precursor to the protein. The complete nucleotide sequence together with the N-terminal amino acid sequence of the 3 1-kDa protein predicts the existence of a precursor polypeptide which contains a 28-aa signal sequence. The mature protein, but not the precursor, is found in the growth medium of stationary-phase cultures. This observation suggests that the signal peptide is removed prior to export from E. coli cells as might be expected (Michaelis and Beckwith, 1982). The signal sequence does not appear to be unusual (Watson, 1984). It consists of a very basic 9-aa N te~inus followed by 19 uncharged and largely nonpolar aa which are predicted to have a strong tendency toward the formation of a-helical structure when analyzed by the method of Gamier et al. (1978). An alanine precedes the cleavage site. Examination of the putative promoter region

(Fig. 6) reveals a high degree of similarity to published E. coli promoters and ribosome-binding sequences (Shine and Dalgarno, 1974; Youderian et al., 1982; Siebenlist et al., 1980; McClure, 1985; Gold et al., 1981). The presence of this sequence strongly suggests, but does not prove, that it also serves as a promoter in Brucella cells. This point is important because it may suggest that other genes from this organism can also be expressed in E. cd without the use of special expression vectors. This finding may also indicate that B. abortus is more closely related to the enteric bacteria than has previously been supposed, since genes from other distantly related Gram-negative bacteria such as Rhizobium and Agrobacteriwn frequently express poorly in E. coli. Further examination of the sequence reveals what appears to be a Rho-independent type terminator sequence (Platt, 198 1) at nt positions 128-152. This is 350 nt upstream from the BCSP31 promoter raising the possibility of a substantial spacer preceding the BCSP3 1 tr~scription unit. A possible terminator is also found downstream

30 60 90 GCGCCTATACGCGCGACCTGATCCAGTCTTTTGCATCGCGGTGCCATGT~CGTCTGGTGGGGGCCGACG~GCTGGCGGC~ATTGCCGAAG 120 150 180 CGCATATTCGCGGTGAAAGiTTCGATGAGGCGCTGGTCAiCGCACCGATA 210 240 270 TCGTGGTGCiTGCCTGTACiCATTATCCGTTTCTCGTCAGACCCGGCGG 300 330 360 AAGCCATIGCACGACGCATGAAATCGCTTTTGCCTGCGCGAAGTGACGA~GATGAATTTCATTCTCAAG~TGACCTAGC~TTCTTCACAT 390 420 450 CCAGGAAACCCGACTATGCCATTCGCCGCCTGATGCAGGGTTTTGGGCT~CGTTTTTAATCGTTTCAGT~GGCTCTGGC~GCGCTTGTCG 480 510 540 GCAAGCGAT~GTATTCTTTEGGAAAAT~~~GAATAATGGATGCGGTGG~TGAC~T~~~CCTCAAGCT~CCTATGGTT~TCGGCAT~~~ CTATGCGGGAAGAGGACTGiTATTATGAAATTCGGAAGC~A~TCCGTC~CTTGGCTGTTGCGGCGGTG~CGGGCGCGA~TGCGTTGGGA MetLysPheGlySerLysIleArgArgLeuAlaValAlaAlaValAlaGlyAlaIleAlaLeuGly 660 690 720 GCGAGCTTTGCGGTTGCACjGGCCCCGACATTTTTCCGT~TCGGCACTG~CGGCACAGCCGGAACCTAT~ATCCGATTG~TGGTCTGATC AlaSerPheAlaValAlaGlnAlaProThrPhePheArgIleGlyThrGlyGlyThrAlaGlyThrTyrTyrProIleGlyGlyLeuIle GlnAlaProThrPhePhe... BCSP31 N-term: 750 780 810 GCGAACGCGATTTCCGGCGCAGGCGAAAAGGGCGTGCCGGGTCTCGTCG~GACGGCCGTTTCGTCGAAT~GCTCGGTTG~CAATATCAAT AlaAsnAlaIleSerGlyAlaGlyGluLysGlyValProGlyLeuValAlaThrAlaValSerSerAsnGlySerValAlaAsnIleAsn 840 870 900 GCGATCAAGiCGGGCGCTCjGGAGTCCGGCTTTACGCAGiTGATGGCAAG AlaIleLysSerGlyAlaLeuGluSerGlyPheThrGlnSerAspValAlaTyrTrpAlaTyrAsnGlyThrGlyLeuTyrAspGlyLys 960 930 990 GGCAAGGTGGAAGATTTGCGCCTTCTGGCGACGCTTTACCCGGAAACGA~CCATATCGTTGCGCGTAAG~ATGCAAACA~CAAATCGGTC GlyLysValGluAspLeuArgLeuLeuAlaThrLeuTyrProGluThrIleHisIleValAlaArgLysAspAlaAsnIleLysSerVal 1020 1050 1080 GCAGACCTGAAAGGCAAGCGCGTTTCGCTGGATGAGCCGGGTTCTGGCA~CATCGTCGATGCGCGTATC~TTCTTGAAG~CTACGGCCTC AlaAspLeuLysGlyLysArgValSerLeuAspGluProGlySerGlyThrIleValAspAlaArgIleValLeuGluAlaTyrGlyLeu

1110

1140

1170

ACGGAAGACGATATCAAGGCTGAACACCTGAAGCCGGGACTTTCTTTGTG ThrGluAspAspIleLysAlaGluHisLeuLysProGlyProAlaGlyGluArgLeuLysAspGlyAlaLeuAspAlaTyrPhePheVal 1200 1230 1260 GGCGGCTATCCGACGGGCGCAATCTCGGAACTGGCCATCiCGAACGGTA~TTCGCTCGTTCCGATCTCC~GGCCGGAAG~GGACAAGATT GlyGlyTyrProThrGlyAlaIleSerGluLeuAlaIleSerAsnGlyIleSerLeuValProIleSerGlyProGluAlaAspLysIle 1290 1320 1350 CTGGAGAAAiATTCCTTCTiCTCGAAGGATGTGGTTCCTGCCGGAGCCT~TAAGGACGTGGCGGAAACA~CGACCCTTG~CGTTGCCGCA LeuGluLysTyrSerPhePheSerLysAspValValProAlaGlyAlaTyrLysAspValAlaGluThrProThrLeuAlaValAlaAla 1380 1410 1440 CAGTGGGTGACGAGCGCCAAGCAGCCGGACGACCTCATCiATAACATCA~CAAGGTTCTCTGGAACGAG~ATACACGCA~GGCACTCGAT GlnTrpValThrSerAlaLysGlnProAspAspLeuIleTyrAsnIleThrLysValLeuTrpAsnGluAspThrArgLysAlaLeuAsp 1470 1500 1530 GCGGGCCATGCGAAGGGCAlGCTCATCAAGCTCGATAGT~CGACGAGCA~CCTCGGTATTCCGCTGCAT~CCGGCGCAG~ACGCTTTTAC AlaGlyHisAlaLysGlyLysLeuIleLysLeuAspSerAlaThrSerSerLeuGlyIleProLeuHisProGlyAlaGluArgPheTyr 1590 AAGGAAGCGGGCGTGCTGAlATAATC~~~~AATGATCGG~TCCTGATAT~TTATTCCGAATTGAAGGGT~ACATTGCGG~AGCTCG~~~~ LysGluAlaGlyValLeuLys 1680 1710 GCGCGCTGC;GCGCTCCCGiTTTCCA~~~~GGTTCCGGTiCT~TTTGTTTTT~CGCATTATCC 1740 1770 1800 GACGCAAAACCGTTTCACGCTTTTGCTGGAAATGCTCTAGTGCACGACCGGCAAAGTGG~GTTGGCCCG~ATGACAGAAG AACAAAATGCAAAGCTT Fig. 5. Nucleotide

sequence

* of 1817 nt flanking

the nucleotide

sequence,

site predicted

to lie in the promoter

-35 promoter

sequences

Last digits of the numbers

and compared

and encoding

the gene for BCSP31.

with the experimentally

region is doubly underlined

as shown in Fig. 6. The inverted

repeats

(nt) are aligned with given nucleotide.

determined

and falls between discussed

The predicted

N-terminal

sequence

sequences

amino acid sequence

which are homologous

in the text as possible

is listed under

of BCSP3 1 (bold print). The Hind111 terminator

to E. coli -10 and

sequences

are underlined.

Promoter sequence 490 . . . . ~TGCGGTGG~~$~~~...(17)...~~XGG~..(6)..~CAX...

B. abortus

. . . . A .. .. .. ..T-fEACA...(17)...TATAAT.(4-7).CAT...

E. coli

-45

_-_- ->

-10

Ribosome binding sequence -10

. . . AGAGGACTGGTATTm...

BCSP31

. ..AUUCCUCCACUAG...

16s rRNA (E. coli)

****

Fig 6. Comparison

of the presumed

1985); and the homology the arrow

indicates

commonly

occurring

*

BCSP31

promoter

sequence

at -10 from the start of translation

the typical

start point of transcription

and the consensus

E. coli promoter

(Siebenlist

with E. coli 16s ribosomal

RNA. The asterisks

in E. coli, and the capital

letters

et al., 1980; McClure,

indicate

homologous

in the E. co/i promoter

bases,

specify the most

bases.

from the gene. A stem-loop structure, located at nt 1646-1685, contains a perfect 11-bp inverted repeat closely followed by a T-rich region. In this case, however, the inverted repeat is separated by 18 bp which is unusually long, and the runs of Ts are not immediately adjacent to the last base pair of the predicted stem structure. At the present time the cellular function of BCSP31 is uncertain. The expression of the protein does not seem to have a major effect on the viability of E. coli cells. A computer search of the NIH and EMBL sequence data banks has not revealed any persuasive sequence homologies. Methods are currently being developed for the purification of milligram or larger quantities of BCSP31 for vaccine, diagnostic, and biological studies.

REFERENCES Beaucage,

S.L. and Caruthers,

phoramidites

-

a new

deoxypolynucleotide

synthesis.

phos-

intermediates

Tetrahedron

H.C. and Doly, J.: A rapid

cedure for screening

for

Lett. 22 (198 1)

recombinant

alkaline

plasmid

extraction

pro-

DNA. Nucl. Acids

Res. 7 (1979) 1513-1523. Bolivar,

F.: Construction

vehicles, EcoRI

and characterization

III. Derivatives

of EcoRI-generated

Garnier,

Vet. Immunol.

J., Osguthorpe,

accuracy

S.G.:

Immunity

Immunopathol.

structure

to intracellular

3 (1982) 5-66.

D.J. and Robson,

and implications

the secondary

unique

recombinant

Gene 4 (1978) 121-136.

F.M. and Campbell,

bacteria.

of new cloning

of plasmid pBR322 carrying

sites for selection

DNA molecules. Collins,

B.: Analysis

of simple methods

ofglobular

proteins.

of the

for predicting J. Mol. Biol. 120

(1978) 97-120. and

Stormo,

D., Schneider,

Helfman,

D.M., Fermisco,

Hughes, libraries.

Focus

Isolation

Karn,

J.R., Fiddes,

S., Singer, B.S. in prokaryotes.

J.C., Thomas,

screening

ofcDNA

G.P. and expression

6 (1984) 1-5.

M.W., Lujan,

E., Ostrander,

of microgram

quantities

acrylamide Enzymol.

initiation

35 (1981) 365-403.

S.H.: Immunological

Hunkapillar,

T., Shinedling,

G.: Translational

Annu. Rev. Microbial.

This material is based in part upon work supported by the Competitive Research Grants Office, U.S. Department of Agriculture under Agreement No. 85-CRCR-1-1842. This work also utilized the BIONET resource which is supported by the NIH Division of Research Resources Grant No. RR0186505.

Deoxynucleoside

of key

1859-1862. Birnboim,

Gold, L., Pribnow,

ACKNOWLEDGEMENTS

M.H.: class

F. and Hood,

of proteins

gels for amino acid sequence

analysis.

from

L.E.: poly-

Methods

91 (1983) 227-236.

J., Brenner,

bacteriophage

S., Barnett, lambda

L. and

Cesareni,

G.: Novel

cloning vector. Proc. Natl. Acad. Sci.

USA 77 (1980) 5172-5176. Laemmli,

U.K.:

Cleavage

of structural

proteins

during

the

9 assembly

of the head of bacteriophage

T4. Nature

227 (1970)

680-685. Maniatis,

A Laboratory

E.F. and Sambrook,

Manual.

Spring Harbor,

J.: Molecular

Cold Spring Harbor

Cloning.

Laboratory,

Cold

NY, 1982, pp. 363-402.

A.M. and Gilbert,

with base-specific

W.: Sequencing

chemical

cleavages.

cell

W.R.: Mechanism

end-labelled

Methods

and control

DNA

Enzymol.

Microbial.

of transcription

Annu. Rev. Biochem.

S. and Beckwith, envelope

J.: Mechanism in Escherichia

proteins

65

initia-

54 (1985) 171-204. of incorporation coli. Annu.

of Rev.

36 (1982) 435-465.

Platt, T.: Termination tryptophan

and its regulation

in the

of E. coli. Cell 24 (198 1) 10-23.

S. and Coulson,

chain-terminating

inhibitors.

Dalgarno,

A.R.: DNA sequencing

with

Proc. Natl. Acad. Sci. USA 74

L.: The

sense triplets

and ribosome

3’-terminal

sequence

RNA: complimentarity binding

of

to non-

sites. Proc. Natl. Acad.

U., Simpson,

polymerase moters.

R.B. and

Gilbert,

homologously

fractions

evaluation.

on Brucellosis Report

W.: E. coli RNA

with two different

National

Press, Washington,

Research:

Brucellosis

of the

Subcommittee

Academy

of Sciences,

DC, 1977.

pro-

research:

an

on Brucellosis National

A.E.: Isolation

from

T. and Gordon,

and

Brucelfa abortus.

J.: Electrophoretic

from polyacrylamide

trans-

gels to nitrocellulose

and some applications.

Proc. Natl. Acad.

Sci. USA 76 (1979) 4350-4354. Vieira,

J. and Messing,

derived

system

Watson,

J.: The pUC

for insertion universal

M.E.E.:

primers.

Compilation

plasmids,

mutagenesis Gene

an M13mp7and sequencing

19 (1982) 259-268.

of published

Wessel,

D. and

recovery Woo,

Flugge,

of protein

signal

sequences.

S.L.C.:

A sensitive

and rapid

Methods

activity.

Academy

in the presence for recombinant

68 (1979) pp. 389-395. M.M.: Sequence

with antibody

probes.

Communicated

by D.T. Denhardt.

deter-

Cell 30 (1982) 843-853.

Young, R.A. and Davis, R.W.: Yeast RNA polymerase isolation

of

138 (1984) 141-143.

method

Enzymol.

P., Bouvier, S. and Susskind, of promoter

for the quantitative

solution

and lipids. Anal. Biochem.

phage screening. Youderian,

V.I.: A method in dilute

Cell 20 (1980) 269-281.

Subcommittee Research,

interacts

Bruceliu

from

56 (1984) 199-211.

26 (1979) 668-679.

sheets: procedure

minants

Sci. USA 71 (1974) 1342-1346. Siebenlist,

and biological

preparations

B.L. and Ritchie,

of toxic

H., Staehelin,

detergents

Escherichia coli 16s ribosomal

Biol. Stand.

L.B., Deyoe,

Infect. Immun. Towbin,

B.L.: Biochemical

protein

Nucl. Acids Res. 13 (1984) 5145-5164.

(1977) 5463-5467. J. and

Tabatabai,

with synthetic

of transcription

operon

Sanger, F., Nicklen,

Shine,

abortus. Develop.

fer of proteins

tion in prokaryotes. Michaelis,

of soluble

characterization

(1980) 499-560. McClure,

L.B. and Deyoe,

properties

T., Fritsch,

Maxam,

Tabatabai,

II genes:

Science 222 (1983) 778-782.