INFECrION AND IMMUNI, Dec. 1994, p. 5361-53660019-9567/94/$04.00+0Copyright X 1994, American Society for Microbiology

Brucella Ribosomal Protein L7/L12 Is a Major Component inthe Antigenicity of Brucellin INRA for Delayed-TypeHypersensitivity in Brucella-Sensitized Guinea Pigs

G. BACHRACH,1 M. BANAI,2* S. BARDENSTEIN,2 G. HOIDA,3 A. GENIZI,4 AND H. BERCOVIER1Department of Clinical Microbiology, The Hebrew University-Hadassah Medical School, Jerusalem,1 and Department

of Bacteriology, The Kimron Veterinary Institute,2 The Israeli Veterinary Services,3 and Department of Statistics,Agricultural Research Organization,' Beit Dagan 50250, Israel

Received 13 May 1994/Returned for modification 16 July 1994/Accepted 2 September 1994

A delayed-type hypersensitivity (DTH) reaction in the course of brucellosis in humans and animals can berevealed by the brucellin INRA (Brucellergen) skin test. Brucellergen is composed of more than 20 proteins ofdifferent molecular weights. A 12-kDa protein eliciting DTH in BruceUa melitensis Revl-sensitized guinea pigswas found to be a significant component for the allergenic properties of Brucellergen. Sequencing of the geneencoding this protein identified it as the L7/L12 ribosomal protein. The L7/L12 gene of B. melitensis wasamplified by PCR and subcloned in the Escherichia coli pQE30 plasmid. The resulting recombinant protein didnot produce a DTH reaction in sensitized animals. It was used to raise specific antibodies in a rabbit. Affinitychromatography with these antibodies was used to isolate a single protein from Brucellergen and from B.melitensis cytosol preparations which produced a DTH reaction in guinea pigs sensitized with B. melitensis Revl.N-terminal amino acid sequencing of the protein confirmed that it was the L7/L12 ribosomal protein. This isthe first complete report on the involvement of a defined bacterial ribosomal protein in the DTH response ofanimals infected with intracellularly multiplying bacteria.

Brucellosis is a zoonotic disease of great public health andeconomic importance throughout the world (14). Control ofthe disease depends on the one hand on the ability to identifyand remove contaminated animals and on the other hand onprophylactic measures such as vaccination. Identification ofbrucella infections in animals is based mainly on serologicaltests. These tests are sensitive and specific, but they are difficultto perform under field conditions (12, 13).As a facultative intracellular pathogen, a brucella elicits a

cellular immune response as well as an antibody response.Delayed-type hypersensitivity (DTH) is one of the expressionsof cellular immunity. This reaction can be measured in sensi-tized animals by the swelling and the erythema resulting fromthe intradermal injection of the pathogen's allergenic antigens(15, 17, 24).A skin test to detect DTH and identify brucella-contami-

nated animals is used widely. However, the use of this test islimited to surveillance of whole herds and does not extend tothe detection of the exposure of individual animals in a herd tothe pathogen. This is because of the lower sensitivity of the skintest (60 to 80% sensitivity) compared with that of the conven-tional serum tube agglutination test (80 to 85% sensitivity) andthat of the complement fixation test or the Rose Bengal test,which both have 90 to 95% sensitivities (12). Nevertheless, thespecificity of the skin test is superior to that of the seroagglu-tination test and equal to that of the complement fixation test(14, 27). Brucella protein preparations (such as brucellinINRA, also called Brucellergen) have proved to be able toreveal a DTH reaction in sensitized animals (14). The proteincomposition of these preparations has been poorly character-ized (18). Brucellergen is composed of 20 to 30 proteins. Onlyone fraction, composed of at least 3 not functionally defined

* Corresponding author. Mailing address: Department of Bacteriol-ogy, The Kimron Veterinary Institute, P.O. Box 12, Beit Dagan 50250,Israel. Phone: 972-3-9681698. Fax: 972-3-9681753.

proteins each with a molecular weight of 3 x 104, has beenshown to be immunogenic in the DTH reaction (7). For thereasons stated above, the standardization of the existing skintest has not been yet achieved. Recombinant brucella-aller-genic proteins could be good candidates for use in a skin testthat could be standardized. In addition, the use of recombinantproteins would reduce the risks due to exposure to brucelladuring production of such reagents.

Filters bearing a 12-kDa protein isolated from Brucellaabortus were able to reveal a DTH reaction in sensitized guineapigs (5). This protein was characterized as an immunodomi-nant protein by Brooks-Worrell and Splitter using a T-lympho-cyte Western blot (immunoblot) technique (9). The N-terminalamino acid sequence of this protein was reported by Brooks-Worrell and Splitter (10). An oligonucleotide corresponding tothe N-terminal amino acid sequence of the B. abortus 12-kDaprotein was used as a probe to screen a Brucella melitensisgenomic library. The gene encoding the 12-kDa B. melitensisprotein was identified, sequenced, and found to be similar tothe Escherichia coli ribosomal protein L7/L12 (2) and identicalto the B. abortus L7/L12 protein (22).

In the work presented here, we describe the isolation andthe identification of the brucella L7/L12 protein from brucellinINRA (Brucellergen). We also characterize the antigenicproperties of this protein in the DTH reaction of brucella-sensitized guinea pigs.

MATERUILS AND METHODS

Bacterial strains and DNA preparation. B. melitensis Revl,B115, and 16M were grown on tryptic soy agar supplementedwith glucose serum (1). The strains were from freeze-driedbacterial stocks maintained at the Israeli Brucellosis ReferenceCenter, Kimron Veterinary Institute. The DNA of strain 16Mwas isolated as previously described (6).

5361

Vol. 62, No. 12

on August 20, 2020 by guest

http://iai.asm.org/

Dow

nloaded from

5362 BACHRACH ET AL.

Expression and purification in E. coli of the brucella 12-kDaprotein. The primers p512 (GGGGGATCCGCTGATCTCGCAAAGATTGTTFGAA) and p312 (CCAGGTACCTCCAAACITAC1T'GAGTTCAACCT) were selected from the B.melitensis L7/L12 sequence (GenBank L27819) and synthe-sized with the Applied Biosystems 380 B DNA synthesizer.These primers included a BamHI site at the 5' end (p512) ofthe gene and a KpnI site at the 3' end (p312). B. melitensis 16MDNA encoding the L7/L12 protein was amplified by PCR(VentR DNA polymerase; New England Biolabs, Beverly,Mass.). After amplification, the PCR product was digestedwith KpnI and BamHI simultaneously and the resulting DNAfragment was purified by agarose gel separation. Purificationwas followed by gene clean extraction of the DNA (GenecleanII kit; Bio 101, La Jolla, Calif.). The purified L7/L12 gene was

ligated to the E. coli pQE30 plasmid expression vector (Qia-gen; Diagen, Hilden, Germany) predigested by KRnI andBamHI. After ligation (2 h at 20°C with 2 U of T4 ligase, IBI,USA), the resulting plasmid containing the L7/L12 sequence,named pQE3012, was used to transform E. coli SG13009. Theplasmid pQE3012 encodes for six histidines which are taggedto the NH2 terminus of the expressed protein. These histidinesenabled easy purification, under denaturation conditions, ofthe recombinant protein by affinity chromatography with im-mobilized nickel chelate (nickel-nitrilotriacetic acid [Ni-NTA]column; patent by Hoffmann-La Roche, Basel, Switzerland).The product of the cloned gene was expressed in high quanti-ties by IPTG (isopropyl-4-D-thiogalactopyranoside) inductionand purified on an Ni-NTA column as specified by themanufacturer's instructions (Diagen). The purified protein wasdialyzed against saline and subjected to sodium dodecyl sul-fate-polyacrylamide gel electrophoresis (SDS-PAGE) (19) andWestern blot analysis (26) with anti-L7/L12 hyperimmunerabbit serum (described below) at a dilution of 1/2,000.

Preparation of rabbit anti-recombinant brucella L7/L12antiserum. Two hundred micrograms of the recombinantL7/L12 protein (1 ml in a 0.9% NaCl solution) was emulsifiedin 1 ml of Freund complete adjuvant. The mixture was

administered to a New Zealand White rabbit by subcutaneousinjection (neck) and intramuscular injection (hind legs).Booster injections of the recombinant L7/L12 protein (200 ,ug)mixed with Freund incomplete adjuvant were given on day 14and day 28. Blood was collected on days 38 and 43. The serum

was separated by centrifugation (3,000 x g for 15 min) andstored at 4°C with 0.02% sodium azide.

Preparation of the anti-L7/L12 affinity column. One millili-ter of Affi-Gel 10 (Bio-Rad Laboratories, Hercules, Calif.) wasfiltered, washed with distilled water, and equilibrated with 10volumes of phosphate-buffered saline (PBS). The activatedbeads were added to a mixture containing 1 ml of serum

(rabbit anti-recombinant brucella L7/L12) and 3 ml of 0.167 MMOPS (3-[N-morpholino]propanesulfonic acid) (pH 7). Themixture was agitated for 4 h at 4°C. The remaining free bindingsites on the beads were blocked by the addition of 0.3 ml of 1M ethanolamine (pH 7), and then the mixture was agitated at4°C for 1 h. The agarose beads were transferred to a disposableplastic column apparatus and washed with 10 volumes of PBSbefore use.

Purification of the L7/L12 antigen from brucellin INRA andfrom B. melitensis. Tryptic soy broth (500 ml) was inoculatedwith B. melitensis B115 and incubated for 48 h at 37°C. Cellswere harvested by centrifugation at 5,000 x g for 15 min,washed with PBS, and suspended in 15 ml of lysis buffer (0.44M sucrose, 0.03 M MgCl2, 25 pug of DNase per ml). Sand grains(12.5 g) were added, and the cells were crushed with a Braunhomogenizer in five cycles of 50 sec each, with 10-sec inter-

ruptions between cycles. The sand grains were removed bycentrifugation (5,000 x g for 5 min). Whole cells and largeaggregates were removed by centrifugation (14,000 x g for 20min), and the supernatant was collected. The supernatant wasloaded on the anti-L7/L12 affinity column. The column wasthen washed with 10 volumes of PBS followed by 3 volumes ofNaCl solution (1 M) at a rate of 1 ml/min. The L7/L12 antigenwas released from the column by the addition of 0.2 M glycinebuffer (pH 2.3). Fractions of 1.35 ml were collected in tubescontaining 150 RI of 2 M Tris (pH 9) neutralizing buffer.Samples (10 ,u) from each fraction were loaded on SDS-15%polyacrylamide gels (19) to determine the purity of eachfraction and were subjected to Western blot analysis as de-scribed above. The L7/L12 protein was also isolated fromBrucellergen (batch 96 G 091; Rhone-Merieux, Lyon, France)by affinity column purification and subjected to Western blotanalysis in accordance with the protocols described above. Inaddition, immunoaffinity-purified L7/L12 was further purifiedby SDS-PAGE. The band corresponding to the 12-kDa protein(the L7/L12 protein as identified by Western immunoblotanalysis) was excised from the gel. The resulting highly purifiedL7/L12 protein was released from the gel by electroelution andthen dialyzed against saline. The protein concentration wasdetermined according to the technique of Bradford (8).Amino acid sequencing. The purified L7/L12 protein iso-

lated from the brucellin INRA with the affinity purificationcolumn was subjected to N-terminal amino acid sequencing bythe automatic Edman degradation procedure with an AppliedBiosystems 475A gas-phase protein sequencer coupled with asynchronized high-performance liquid chromatography sys-tem.

Skin reaction test. Forty male Hartley guinea pigs (250 geach at the beginning of the experiment) were sensitized bysubcutaneous injection of 106 B. melitensis Revl organisms peranimal (1). Three weeks later, recombinant L7/L12, affinity-purified L7/L12, Brucellergen (Rhone-Merieux), and L7/L12-depleted Brucellergen were injected intradermally at variousdoses (in 100 RI). The L7/L12 protein (150 ng) purified on theimmunoaffinity column and then subjected to SDS-PAGE andelectroelution was injected into three sensitized guinea pigsand one control guinea pig. Diameters (in millimeters) of theareas of erythema were read 24 h later. The immediatenonspecific response was checked after 4 h.The statistical analysis was done by the SAS (1990) gener-

alized linear models procedure.

RESULTS

Antigenicity of the B. melitensis L7/L12 protein expressed inE. coli. The DNA of B. melitensis 16M encoding the L7/L12protein was amplified by PCR with the primers p512 and p312.The resulting product had a size of 375 bp (data not shown),compatible with the expected size of the gene. It was subclonedin plasmid pQE30 to give plasmid pQE3012, which subse-quently was transformed into E. coli SG13009. The six histi-dines tagged at the N terminus of the recombinant proteinenabled purification of the product with the Ni-NTA column.As can be seen in Fig. 1 (lane b), the recombinant protein wasexpressed in E. coli at a high level (8 mg of recombinantprotein per liter of E. coli culture) and was purified tocomplete homogeneity (6 mg of pure recombinant protein perliter of bacterial culture) by passage of the E. coli lysatethrough the Ni-NTA column (Fig. 1, lanes d to h).

This purified L7/L12 protein was used to immunize a rabbit.The rabbit antiserum prepared against the recombinant L7/L12 protein was collected twice on days 38 and 43 of the

INFECT. IMMUN.

on August 20, 2020 by guest

http://iai.asm.org/

Dow

nloaded from

BRUCELLA L7/L12 PROTEIN ANTIGENICITY IN DTH 5363

a b c d e f 9 h a b c d e

kDa

46 -

30

21.5-14.3-6.52.4 -

FIG. 1. Expression and purification of recombinant brucella L7/L12 protein. E. coli SG13009 cells harboring plasmid pQE3012 were

grown to an optical density at 600 nm of 0.8. IPTG (1 mM) was addedto induce the recombinant protein. Recombinant L7/L12 was purifiedfrom the E. coli lysate by affinity purification on an Ni-NTA column(see Materials and Methods). Samples were analyzed on SDS-14%polyacrylamide gels and stained with Coomassie blue. Lanes: a,molecular mass markers; b, total lysate of induced cells (10 p.1); c, totallysate of noninduced cells (10 p.1); d to h, fractions 1 to 4 ofrecombinant L7/L12 from induced lysate purified on an Ni-NTAcolumn.

immunization treatment. The sera reacted identically againstthe recombinant protein. A titer of 1:2,000 with the recombi-nant protein as the sole antigen was found optimal for Westernblot analysis (data not shown). The serum was tested againstBrucellergen and the brucella lysate, and as shown in Fig. 2(lanes b and d, respectively), it was highly specific and sensitivefor the detection of the L7/L12 protein.

Identification and purification of L7/L12 from Brucellergenand brucella lysate. Brucella L7/L12 protein was purified fromBrucellergen and from the B. melitensis B115 lysates with therabbit antibodies bound to the Affi-Gel column. Figure 3depicts the various stages of purification of the protein fromBrucellergen. Purification from whole-cell lysates producedthe same results (data not shown). A single protein band wasobtained after elution of the column and was subjected toN-terminal amino acid sequencing for definitive identification.The resulting sequence confirmed that this protein was theL7/L12 protein; the sequence was identical for the first 9 aminoacids (ADLAKIVED) to the amino acid sequences deduced

a b c d

12 kDa_ _ _

!oboFIG. 2. Analysis of rabbit anti-L7/L12 serum specificity. A rabbit

was immunized with the purified recombinant brucella L7/L12 protein(see the legend to Fig. 1). Brucellergen (lanes a and b) and B.melitensis lysate (lanes c and d) were subjected to SDS-PAGE and thento Western blotting (lanes b and d) as described in Materials andMethods. The hyperimmune anti-L7/L12 serum (1:2,000) was used todetect the presence of the L7/L12 protein in Brucellergen (lane b) andin the B. melitensis lysate (lane d). The total protein (Coomassie bluestain) profiles of Brucellergen (7 ,ug) and the B. melitensis lysate (5 ,ug)are depicted in lanes a and c, respectively.

FIG. 3. Chromatography purification of L7/L12 from Brucellergen.Brucellergen was loaded on an anti-L7/L12 immunoaffinity column.The bound L7/L12 was released from the column as described inMaterials and Methods. Fractions were collected, analyzed on SDS-14% polyacrylamide gels, and stained with Coomassie blue. Lanes: a,molecular mass markers; b, Brucellergen (7 p.l); c, unbound Brucel-lergen (7 p.1) after passage through the immunoaffinity column; d to h,glycine elution fractions 1 to 5. The purified brucella L7/L12 ribosomalprotein is depicted in lane f as a single 12-kDa band.

from the L7/L12 B. melitensis and B. abortus DNA sequences.It was also identical to the N-terminal amino acid sequence ofthe 12-kDa immunodominant protein (10).

Activity of the ribosomal L7/L12 protein in sensitized guineapigs. The antigenicity of the various preparations of the L7/L12ribosomal protein in the DTH reaction was compared with thatof Brucellergen. The sensitization of guinea pigs inoculatedwith live B. melitensis Revl was checked 3 weeks after inocu-lation with Brucellergen. A typical reaction in these guineapigs, which were injected with a 1-to-15 dilution of Bruceller-gen (7 ,ug of protein in 100 ,ll), was an area of erythema witha mean diameter of 19 mm. As shown in Fig. 4, furtherdilutions of Brucellergen led to reductions in the sizes of theareas of erythema. The regression of the sizes of the areas oferythema on log (dose) calculated from the results up to thedose of 2 ,ug was not significant (P = 0.1367). One outlyingresult (an area of erythema with a diameter of 24.5 mm after a

protein dose of 0.41 ,ug) was excluded from the calculation.Doses of 16 and 6 jig of the recombinant L7/L12 protein

produced in E. coli and purified with the Ni-NTA column (asdescribed above) were intradermally injected into five guineapigs sensitized to B. melitensis and into two naive controlguinea pigs. No detectable DTH reaction to the recombinantprotein was observed in sensitized guinea pigs, whereas thesame guinea pigs reacted to Brucellergen injected into thecontralateral side, as described above.

B. melitensis L7/L12 proteins purified from either Bruceller-gen or whole-cell brucella lysate as described above wereintradermally injected into sensitized guinea pigs. As can beseen in Fig. 4, the L7/L12 protein purified from Brucellergenelicited a strong DTH response at amounts as small as 62.5 ng.The DTH response of the guinea pigs to the L7/L12 proteinwas dose dependent up to the 1-,ug dose (P = 0.0001), with a

slope much steeper than that obtained for Brucellergen. Theresult obtained with 2 jig of the L7/L12 protein was notincluded. The 500-ng dose of L7/L12 gave a reaction similar tothat for 2,500 ng of Brucellergen, demonstrating that theL7/L12 protein was a potent DTH-revealing antigen. Similarresults (data not shown) were obtained with the L7/L12protein isolated from whole-cell brucella lysate. The injectionof the L7/L12 protein (150 ng) purified by immunoaffinity and

kDa

463021.5

14.3

f g h

_E ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~I.!_ f~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~i

_ i~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

__ _ _ -' I~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

I r~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

460M.

M. q-

4.mn* -Wftbob

WAMW -A.

VOL. 62, 1994

on August 20, 2020 by guest

http://iai.asm.org/

Dow

nloaded from

5364 BACHRACH ET AL.

20-

E 18-

G) 14-

° 12-

10L.-i-- 1 0-

E 8-C'a6-

a) 4-

2-

0.01 0.1 1 10

Dose injected (,ug protein)FIG. 4. DTH activity of affinity-purified brucella L7/L12 ribosomal protein. DTH reactions of brucella-sensitized guinea pigs were elicited with

different doses (in micrograms) of the L7/L12 purified protein (diamonds), Brucellergen (squares), and Brucellergen from which the L7/L12protein has been depleted (circles). The number of guinea pigs tested with each dose is given inside each symbol. The arithmetic means of thediameters (in millimeters) of the areas of erythema at each point were plotted with a corresponding regression line (regression calculations weredone as described in Results). Solid line, L7/L12 protein; hatched line, Brucellergen; broken line, Brucellergen from which the L7/L12 protein hasbeen depleted.

then subjected to SDS-PAGE and electroelution resulted alsoin a significant DTH reaction (erythema areas with diametersof 10 mm ± 1 mm versus no reaction in the control guinea pig).In order to evaluate the relative activity of the L7/L12 proteinin Brucellergen, whole Brucellergen and Brucellergen fromwhich the L7/L12 has been depleted by the affinity columnmethod were intradermally injected into sensitized guinea pigs.Depletion of the L7/L12 antigen from Brucellergen resulted ina reduction of the allergenic activity of the preparation. Asignificant regression on log (dose) was demonstrated (P =

0.038). There was no significant difference between the slopes(P = 0.817) of the lines of the two preparations (regular ordepleted Brucellergen). Therefore, a model with a common

slope (P = 0.011) was adopted, and the lines in Fig. 4 weredrawn accordingly. A constant difference of 2.24 mm betweenthe DTH responses elicited by the two preparations was found.

Five control naive guinea pigs which were inoculated witheither the L7/L12 preparations (400 ng in 0.1 ml), the wholeBrucellergen, or the L7/L12-depleted Brucellergen did notshow any reactions after 4, 24, and 48 h.

DISCUSSIONRibosomal vaccines against different diseases have been

proposed, but their rationale has been controversial (16, 20, 25,31-34). It has been shown that brucella (11) and mycobacte-rium (3, 4, 23) ribosomes can elicit a DTH response wheninjected into sensitized animals. However, what exactly in theribosomes induced the DTH was not definitively identified.The results reported in our present work clearly identifybrucella ribosomal protein L7/L12 as an antigen capable ofeliciting DTH in brucella-sensitized guinea pigs. We showedthat this protein is present in Brucellergen (Fig. 2). The

depletion of this protein from Brucellergen reduced the anti-genicity of the preparation in the DTH test by a considerable(15%) but not significant amount (P = 0.16). The purifiedL7/L12 protein elicited a DTH response in sensitized guineapigs, and its activity was dose dependent, with a slope of theregression line much steeper than that observed with Brucel-lergen. This probably indicates that the purified L7/L12 proteinhas a higher specific activity in the DTH reaction than theBrucellergen composed of a mixture of proteins. In a recentreport, a partially purified L7/L12 protein from Mycobacteriumbovis BCG was shown to be antigenic by the DTH reaction ofguinea pigs sensitized with either live M. bovis BCG or killedM. tuberculosis (29). It seems therefore that during infection byintracellularly multiplying bacteria such as brucellae and my-

cobacteria, some of the bacterial structural proteins, such as

the L7/L12 protein, are involved in the cellular immuneresponse developed in the host. The abundance of the L7/L12protein, which can reach as much as 45% of the ribosomalmass in E. coli (21), may be the reason for its recognition ininfected guinea pigs. However, the highly conserved sequencesof this protein among bacteria and even eukaryotes (21) raisequestions about the nature of the antigenic motif that isrecognized by the immune response of the infected host.Ribosomal proteins L7/L12 are encoded by the same gene

and differ only by acetylation of the NH3 terminus of the L12protein, which then becomes the L7 protein (21, 30). Whichone of these two proteins, if not both, is responsible for theDTH reaction remains to be determined.The reason(s) why the recombinant L7/L12 protein was not

antigenic in the DTH reaction is not yet understood. The lackof brucella-specific posttranslational modification in E. colicould be one explanation. It seems more reasonable to suggest

9©El

INFEC'T. IMMUN.

on August 20, 2020 by guest

http://iai.asm.org/

Dow

nloaded from

BRUCELLA L7/L12 PROTEIN ANTIGENICITY IN DTH 5365

that the six histidines added to the protein N terminus in thepQE3012 construct might have interfered with protein antige-nicity. We are now building new recombinant proteins lackingthese histidines and trying specific polypeptides for DTH-inducing activity.

It has been suggested that the immunological properties ofribosomal proteins depend on contaminating molecules suchas rRNA or cellular RNA (11). The purified B. melitensisL7/L12 protein was treated with RNase (500ng of L7/L12 in0.1 ml with 5 mg of RNase for 1 h at 37°C) (data not shown),and it did not lose its DTH activity after treatment. In contrast,the DTH reaction was nearly nullified when the L7/L12 proteinwas treated with proteinase K (500 ng of L7/L12 in 0.1 ml with5,ug of proteinase K for 1 h at 50°C, followed by inactivationat 80°C for 10 min). The DTH reaction obtained with theelectroeluted L7/L12 protein isolated after immunoaffinitycolumn purification and SDS-PAGE of Brucellergen rules outthe possibility of contamination by an additional protein.These preliminary data showed that only the protein L7/L12and not other molecules attached to it was responsible for theDTH activity.So far, a major drawback in the use of Brucellergen has been

the difficulty in standardization of the antigen (1). The identi-fication of a single protein with DTH antigenicity will contrib-ute to the improvement of a skin test which can be standard-ized for the diagnosis of brucellosis. This would result in a skintest with better diagnostic and epidemiological values than theexisting one. However, the correlation in DTH reactionsbetween guinea pigs and large animals must be assessed (28)before the L7/L12 protein can be used as a diagnostic tool.Finally, a new strategy to develop a vaccine which wouldenable a skin test that could differentiate between vaccinatedand infected animals is now conceivable. This could beachieved by the alteration of the L7/L12 protein in brucellastrains used for vaccines, such as B. melitensis Revl and B.abortus S19, in such a way that the DTH reaction of theanimals towards this protein would be abolished.

ACKNOWLEDGMENTS

This research was supported in part by funds provided by the Agencyfor International Development under the Middle East regional coop-

eration program managed by the U.S. Department of Agriculture,trinational animal health program, and by grant IS-1434-89 from theU.S.-Israel BARD fund.

REFERENCES

1. Alton, G. G., L. M. Jones, R D. Angus, and J. M. Verger (ed.).1988. Techniques for the brucellosis laboratory. Institut Nationalde la Recherche Agronomique, Paris.

2. Bachrach, G., D. Bar-Nir, M. Banai, and H. Bercovier. 1994.Identification and nucleotide sequence of Brucella melitensisL7/L12 ribosomal protein. FEMS Microbiol. Lett. 120:237-240.

3. Baker, R. E., W. E. Hill, and C. L. Larson. 1972. Delayedhypersensitivity reactions provoked by ribosomes from acid-fastbacilli. Infect. Immun. 6:258-265.

4. Baker, R. E., W. E. Hill, and C. L. Larson. 1973. Ribosomes ofacid-fast bacilli: immunogenicity, serology, and in vitro correlatesof delayed hypersensitivity. Infect. Immun. 8:236-244.

5. Banai, M., S. Bardenstein, G. Hoida, H. Bercovier, G. Bachrach,

and G. Splitter. Submitted for publication.6. Bercovier, H., A. G. Steigerwalt, A. Guiyoule, G. Huntley-Carter,

and D. J. Brenner. 1984. Yersinia aldovae (formerly Yersiniaenterocolitica-like group X2): a new species of Enterobacteriaceaeisolated from aquatic ecosystems. Int. J. Syst. Bacteriol. 34:166-172.

7. Bhongbhibhat, N., S. Elberg, and T. H. Chen. 1970. Characteriza-tion of Brucella test antigens. J. Infect. Dis. 122:70-82.

8. Bradford, M. M. 1976. A rapid and sensitive method for thequantitation of microgram quantities of protein utilizing theprinciple of protein-dye binding. Anal. Biochem. 72:248-254.

9. Brooks-Worrell, B. M., and G. A. Splitter. 1992. Sodium dodecylsulfate- and salt-extracted antigens from various Brucella speciesinduce proliferation of bovine lymphocytes. Infect. Immun. 60:2136-2138.

10. Brooks-Worrell, B. M., and G. A. Splitter. 1992. Antigens ofBrucella abortusS19 immunodominant for bovine lymphocytes asidentified by one- and two-dimensional cellular immunoblotting.Infect. Immun. 60:2459-2464.

11. Corbel, M. J. 1976. The immunogenic activity of ribosomalfractions derived from Brucella abortus. J. Hyg. Camb. 76:65-74.

12. Dohoo, L. R., P. F. Wright, G. M. Ruckerbauer, B. S. Samagh, F. J.Robertson, and L. B. Forbes. 1986. A comparison of five serolog-ical tests for bovine brucellosis. Can. J. Vet. Res. 50:485-493.

13. Fensterbank, R. 1977. Diagnostic allergique de la brucellosebovine. II. Utilisation du test allergique dans les troupeauxinfectes. Ann. Rech. Vet. 8:195-201.

14. Fensterbank, R. 1987. Brucellosis in cattle, sheep and goats:diagnosis, control and vaccination, p. 9-35. In Brucellosis in cattle,sheep and goats. OIE technical series no. 6. Office Internationaldes Epizooties, Paris.

15. Fensterbank, R., and P. Pardon. 1977. Diagnostic allergique de labrucellose bovine. I. Conditions d'utilisation d'un allergene pro-teique purifie: la brucelline. Ann. Rech. Vet. 8:187-193.

16. Gregory, R. L. 1986. Microbial ribosomal vaccine. Rev. Infect. Dis.8:208-217.

17. Jones, L. M., and D. T. Berman. 1975. Antibody-mediated anddelayed-type hypersensitivity reactions to brucella skin test anti-gens in guinea pigs. Infect. Immun. 11:360-364.

18. Jones, L. M., R. Diaz, and A. G. Taylor. 1973. Characterization ofallergens prepared from smooth and rough strains of Brucellamelitensis. Br. J. Exp. Pathol. 54:492-508.

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

20. Loge, R. V., W. E. Hill, R. E. Baker, and C. L. Larson. 1974.Delayed hypersensitivity reactions provoked by ribosomes fromacid-fast bacilli: physical characteristics and immunological as-pects of core ribosomal proteins from Mycobactenum smegmatis.Infect. Immun. 9:489-496.

21. Noller, H. F., and M. Nomura. 1987. Ribosomes, p. 104-125. InF. C. Neidhardt, J. L. Ingraham, K. B. Low, B. Magasanik, M.Schaechter, and H. E. Umbarger (ed.), Escherichia coli andSalmonella typhimunium: cellular and molecular biology. AmericanSociety for Microbiology, Washington, D.C.

22. Oliveira, S. C., Y. Zhu, and G. A. Splitter. 1994. Sequence of therpIJL operon containing L10 and L7/L12 genes from Brucellaabortus. Gene 140:137-138.

23. Ortiz-Ortiz, L., E. B. Solarolo, and L. F. Bojalil. 1971. Delayedhypersensitivity to ribosomal protein from BCG. J. Immunol.107:1023-1026.

24. Ottosen, H. E., and N. Plum. 1949. A nonantigenic allergic agentfor intradermal brucellosis tests. Am. J. Vet. Res. 10:5-11.

25. Paterson, J. S., N. W. Pirie, and A. W. Stableforth. 1947. Protectiveantigens isolated from Br. abortus. Br. J. Exp. Pathol. 28:223-236.

26. Sambrook, J., E. F. Fritsch, and T. Maniatis. 1989. Molecularcloning: a laboratory manual, 2nd ed. Cold Spring Harbor Labo-ratory Press, Cold Spring Harbor, N.Y.

27. Stemshorn, B. W. 1984. Recent progress in the diagnosis ofbrucellosis. Dev. Biol. Stand. 56:325-340.

28. Sterne, M., G. Trim, and E. S. Broughton. 1971. Immunisation oflaboratory animals and cattle with non-agglutinogenic extracts ofBrucella abortus strain 45/20. J. Med. Microbiol. 4:185-194.

29. Tantimavanich, S., S. Nagai, H. Nomaguchi, M. Kinomoto, N.Ohara, and T. Yamada. 1993. Immunological properties of ribo-somal proteins from Mycobacterium bovis BCG. Infect. Immun.61:4005-4007.

30. Terhorst, C., B. Wittmann-Liebold, and W. Moller. 1972. 50Sribosomal proteins. Peptide studies on two acidic proteins, Al andA2, isolated from ribosomes of Escherichia coli. Eur. J. Biochem.25:13-19.

VOL. 62, 1994

on August 20, 2020 by guest

http://iai.asm.org/

Dow

nloaded from

5366 BACHRACH ET AL.

31. Thomas, D. W., and E. Weiss. 1972. Response of mice to injectionof a ribosomal fraction from group B Neisseria meningitidis. Infect.Immun. 6:355-363.

32. Thompson, H. C. W., and I. S. Snyder. 1971. Protection againstpneumococcal infection by a ribosomal preparation. Infect. Im-mun. 3:16-23.

33. Venneman, M. R., and N. J. Bigley. 1969. Isolation and partialcharacterization of an immunogenic moiety obtained from Salmo-nella typhimurium. J. Bacteriol. 100:140-148.

34. Youmans, A. S., and G. P. Youmans. 1965. Immunogenic activityof a ribosomal fraction obtained from Mycobacterium tuberculosis.J. Bacteriol. 89:1291-1298.

INFECT. IMMUN.

on August 20, 2020 by guest

http://iai.asm.org/

Dow

nloaded from