serological differentiation between infected and ... · abortus by gel diffusion techniques....

Vol. 29, No. 2INFECTION AND IMMUNITY, Aug. 1980, p. 435-4410019-9567/80/08-0435/07$02.00/0

Serological Differentiation Between Infected and VaccinatedCattle by Using Purified Soluble Antigens from Brucella

abortus in a Hemagglutination SystemT. J. G. RAYBOULD* AND SHIREEN CHANTLER

Wellcome Research Laboratories, Langley Court, Beckenham, Kent, BR3 3BS England

Serological tests used in current brucellosis eradication schemes, such asbacterial tube agglutination, do not readily distinguish between infected animalsand those immunized with strain 19 or 45/20 Brucella abortus vaccines. In thisstudy, sera from naturally infected cattle were used to identify serologicallyimportant antigens in extracts of virulent B. abortus by gel diffusion techniques.Antisera from rabbits hyperimmunized with selected precipitation lines were usedfor purification by affinity chromatography of two precipitating and one non-precipitating antigen from crude bacterial extracts. A passive hemagglutinationtest using these antigens was developed. A number of characterized bovine serawere screened by passive hemagglutination and conventional bacterial tubeagglutination tests. A considerable improvement in discrimination between serafrom infected and vaccinated cattle was obtained with the hemagglutination testcompared with bacterial tube agglutination.

Current schemes for brucellosis eradicationrely on herd vaccination, with identification andslaughter of infected animals (2). Difficulties indifferentiating infected animals from those im-munized with strain 19 or 45/20 Brucella abor-tus vaccines can result in unnecessary slaughterof animals (1, 6, 7). This economic factor hashindered eradication schemes, and as a conse-quence several investigators have looked for se-rological differences between these two groupsbut have been unsuccessful in achieving gooddiscrimination (5, 9, 14, 20, 21, 26).

In this investigation, sera from naturally in-fected cattle were used to identify serologicallyimportant antigens in extracts of virulent B.abortus by gel diffusion techniques. Rabbitswere immunized with selected precipitationlines, and after removal of antibovine activity,the antisera were used for purification of anti-gens from crude bacterial extracts by affinitychromatography. Purified antigens and crudeextracts were coupled to fixed erythrocytes andused in a passive hemagglutination (HA) systemto screen serum samples from infected and vac-cinated cattle.

MATERIALS AND METHODSSerum samples. Eighty-eight sequential serum

samples taken from a total of six cows (numbers 1through 6) were used. Details of the history of eachcow are summarized: numbers 1 and 2 were vaccinatedwith Duphavac (N. V. Philips-Duphar, Amsterdam)(dead rough strain 45/20) on day 1; number 3 wasvaccinated with strain 19 (smooth, attenuated, living

strain) on day 1; number 4 was vaccinated with strain19 as a calf but was subsequently infected with B.abortus, and serum samples were taken from the dayafter birth of infected calf; number 5 was vaccinatedwith strain 19 as a calf but became infected with B.abcrtus during the course of the experiments (acuteinfection); number 6 was vaccinated with strain 19 asa calf but subsequently became infected with B. abor-tus (chronic infection).A pool of sera from vaccinated and from infected

animals was prepared from individual bleeds takenfrom animals 3 and 6, respectively. A pool of serumfrom noninfected, nonvaccinated cattle was also used.Antigen extracts. Sodium dodecyl sulfate (SDS)

extracts were prepared from a 10% (vol/vol) suspen-sion of B. abortus strain 544/W (19).Immunization ofrabbits. Immunodiffusion plates

were prepared by using 1% purified agar (Oxoid) in0.15 M saline. A series of troughs were cut 10 mmapart, and alternate troughs were filled with SDSextract of B. abortus strain 544/W and pooled infectedbovine serum. After incubation for 3 days at 40C, twodiscrete precipitation lines, designated B and C, werepresent between each pair of troughs. Each line wascut out of the agar and washed with frequent changesof 0.15 M saline for 2 days and then emulsified inFreund complete adjuvant to give a total volume of 3ml. New Zealand white rabbits were immunized intra-muscularly with 1 ml of emulsion and boosted withsimilar injections until satisfactory antisera were ob-tained.

Preparation of immunoadsorbents. Immuno-adsorbents were prepared by coupling appropriatematerial to Sepharose CL4B activated with sodiummetaperiodate (18). The following immunoadsorbentswere used.

(i) Sepharose CL4B-normal bovine serum.435

on April 29, 2019 by guest

http://iai.asm.org/

Dow

nloaded from

http://iai.asm.org/

436 RAYBOULD AND CHANTLER

Sepharose CIAB-normal bovine serum was used forthe removal of antibovine activity from rabbit antiseraraised against precipitation lines B and C before eval-uation in two-dimensional immunoelectrophoresis(2D-IE), using an intermediate gel technique (4). Cou-pling was performed at a concentration of 25 mg ofprotein per g of gel.

(ii) Sepharose CL4B-immunoglobulin G frac-tion of rabbit anti-B or -C. Sepharose CL4B-im-munoglobulin G fraction of anti-B or -C was used forpurification of antigens from crude 544/W extracts.Five milligrams of protein per gram of gel was usedfor coupling.

(iii) Sepharose CL4B-purified antigens. Seph-arose CL4B-purified antigens were prepared with ap-propriate concentrations of affinity chromatography-purified antigens from the 544/W extract.

Immunoadsorbents ii and iii were packed into col-umns over an equal volume of Sephadex G-25 tofacilitate rapid separation of eluted antigens fromeluent buffer during purification by affinity chroma-tography.

Affinity chromatography. Removal of antibo-vine activity from rabbit antisera was achieved byapplying samples to Sepharose-normal bovine serumcolumns and washing through with 10mM phosphate-buffered saline, pH 7.5. Antigen and antibody purifi-cation was performed by passing the crude 544/Wextract or infected bovine serum pool through appro-priate Sepharose-antibody or Sepharose-antigen im-munoadsorbent columns. After emergence of unad-sorbed material, the columns were eluted with 2 Msodium thiocyanate in 10 mM phosphate buffer (pH6.6) to remove bound components. Both washes andeluted material from each column were collected, con-centrated, and examined by the techniques describedin Results.

Gel diffusion techniques. (i) 2D-IE. The specific-ity of rabbit antisera and the purity of antigens andantibodies eluted from immunoadsorbent columnswere assessed by 2D-IE, using 1% agarose in 50 mMVeronal buffer (25). For preliminary investigations ofantigen purity, separation in the first dimension wasfollowed by electrophoresis in the second dimensioninto agarose containing 12.5% pooled infected bovineserum.

(ii) 2D-IE with intermediate gel. Identificationof individual precipitation peaks obtained with differ-ent antigen preparations and evaluation of the speci-ficity of rabbit antisera were performed by using anintermediate gel technique (4). In this method, antigenpreparations were electrophoresed in the second di-mension through an intermediate gel containing thetest antiserum into the "reference" gel containingpooled infected bovine serum, thus enabling investi-gation of immunological relationships between theantigen preparation and antibodies present in eachgel.

(iii) Immunodiffusion. Immunological identitybetween components in crude SDS extracts and puri-fied antigen, and between different antisera, was in-vestigated by Ouchterlony double immunodiffusion in1% purified agar containing 50mM Veronal buffer, pH8.6.

Sensitization of SRBCs. Sheep erythrocytes

(SRBCs) were initially fixed with pyruvic aldehydeand formaldehyde, using the method of Prince et al.(16) but substituting 10 mM phosphate-buffered sa-line, pH 7.6, for 5% sodium citrate buffer. The cellswere further treated with tannic acid followed bysensitization with an appropriate dilution of antigenpreparation or with the immunoglobulin fraction ofpooled infected bovine serum for detection of antibodyor antigen, respectively.HA tests. Test sera or antigen preparations were

titrated from an initial dilution of 1:5 in V-well micro-titer plates (Dynatech Ltd.), using 0.1 M phosphate-buffered saline (pH 7.2) containing 1.5% normal rabbitserum preabsorbed with 10% fixed SRBCs. An equalvolume of a 1% suspension of the appropriately sensi-tized SRBCs in the same buffer was added. Aftergentle agitation, the plates were covered and allowedto stand at room temperature for 18 h before endpointagglutination titers were recorded.

Bacterial tube agglutination tests. Tube agglu-tination tests were performed by the method of Altonet al. (3), using B. abortus strain 99-stained suspension(Wellcome Reagents Ltd.). Endpoint agglutination ti-ters were recorded, avoiding the need for scoring de-grees of reaction.

RESULTSReactivity of crude extracts in gel diffu-

sion techniques. Analysis by 2D-IE of 544/Wextract against pooled infected bovine serumproduced three separate precipitation peaks.These peaks were designated a, ,B, and y (Fig. 1).No peaks were obtained in similar tests againstvaccinated or negative bovine serum pools.Ouchterlony immunodiffusion tests on 544/Wextract against pooled infected bovine serumproduced at least three precipitation lines, butwhen parallel troughs were cut in gel 10 mmapart and filled alternately with extract or in-fected serum, only two discrete lines, designatedB and C, were observed.Analysis of rabbit antisera. Antisera from

rabbits hyperimmunized with precipitation linesB and C gave strong reactions in gel diffusionagainst 544/W extract and normal bovine serum.The activity against bovine immunoglobulinspresent in the precipitation lines used for im-munization was removed by affinity chromatog-raphy on the Sepharose CL4B-normal bovineserum immunoadsorbent. Reactivity against544/W extract was retained after this treatment.Immunodiffusion tests with selected absorbedrabbit antisera (B5 and C9) showed clear noni-dentical precipitation lines, confirming that theycontained antibody against two immunologicallydistinct antigens present in 544/W extract.To identify the antigens involved, antisera B5

and C9 were tested in intermediate gel 2D-IE;the results are shown in Fig. 1. A reduction inthe size of a single peak and an extension of theprecipitation line forming this peak into the

INFECT. IMMUN.


http://iai.asm.org/

Dow

nloaded from

http://iai.asm.org/

HA WITH PURIFIED B. ABORTUS ANTIGENS 437

I

a

0

b d

0.FIG. 1. Wells contained 5441W extract. Reference gels contained a 12.5% pool of infected bovine serum.

Intermediate gels contained: (a) 10% absorbed antiserum against 8 line (rabbit B5); (b) 30% absorbedimmunoglobulins from anti-y serum (rabbit C9); (c, d) normal rabbit serum.

intermediate gel indicate that the intermediateand reference gels contain antibody of identicalspecificity. These tests show that antisera B5and C9 contain specific reactivity against theantigens forming peaks ,8 and y, respectively.Purification of antigens. Purification of I?

and y antigens from 544/W extract was per-formed by affinity chromatography, using im-munoadsorbents prepared from antisera B5 andC9 as described above. Similar elution profileswere obtained with both columns. Some frac-tionation of nonadsorbed components in 544/Wextract occurred during passage through the col-umn (wash), presumably due to the molecular-sieve effect of the Sephadex G-25. The peak ofpurified antigen obtained after elution with so-dium thiocyanate was also separated from theeluent buffer by passage through the SephadexG-25 portion of the column.2D-IE using intermediate gels containing rab-

bit antisera B5 and C9 was performed on con-centrated washes and eluates from each columnin order to identify the components present. Theimmunoadsorbent columns effectively adsorbedthe corresponding antigens as the eluate peakscontained selectively purified immunologicallydistinct antigenic components ,B and y (Fig. 2).2D-IE tests on the washes from each column

showed variable results depending on the vol-ume and concentration of the sample applied.Application of large samples resulted in detect-

able levels of specific antigens appearing in thefirst peak of the wash, indicating overloading ofthe column. If the sample volume was too small,specific antigen was not detectable in the eluate,emphasizing the need for optimizing the volumesused. The optimal sample volume in this systemwas approximately 2% of the column volume.HA tests with affinity chromatography-

purified antigens. SRBCs sensitized with af-finity chromatography-purified ,8 and y antigenpreparations (SRBC-,B, SRBC-y) and crude 544/W extract (SRBC-544/W) were tested for HAactivity against titrations of infected and vacci-nated bovine serum pools and B5 and C9 rabbitantisera. The results in Table 1 show that similardiscrimination was obtained between infectedand vaccinated bovine serum pools irrespectiveof the antigen preparation used for sensitization.Minimal titers were observed with B5 and C9antisera even against SRBCs sensitized with thehomologous antigen, indicating that insignifi-cant levels of ,B and y antigens were coupled andas a consequence were unlikely to contributetowards the serological differences observed.This apparent dissociation of precipitating anti-gen (,8 and y) activity and HA activity wasconfirmed by absorption experiments.Samples of B5 and C9 antisera were exhaus-

tively absorbed with SRBCs sensitized with ,Band y antigen preparations and crude 544/Wextract and then tested for HA activity and

-- ...- -!. ... -.-.

VOL. 29, 1980

I \


http://iai.asm.org/

Dow

nloaded from

http://iai.asm.org/


0....;,.. ...

INFECT. IMMUN.

I,

FIG. 2. Reference gels contained a 12.5% pool of infected bovine serum. (a) Well 1, Eluate from B5 column(ft antigen); well 2, 5441W extract. Intermediate gel was 10% absorbed B5 antiserum. (b) Well 1, Eluate fromC9 column (y antigen); well 2, 5441W extract. Intermediate gel was 30% absorbed C9 immunoglobulins.

TABLE 1. HA studies using SRBCs sensitized withaffinity chromatography-purified 13 and -y antigen

preparations and crude 5441W extract

Reciprocal of HA titer againstSRBCs sensitized with:

Antiserum 8i anti- y anti- 544/W No an-gen gen extract tigen

prepn prepn

Infected bovine 20,480 20,480 20,480 <5serum pool

Vaccinated bovine 80 80 80 <5serum pool

B5 antiserum 40 40 40 <5C9 antiserum 5 5 5 <5Nonimmune rab- <5 <5 <5 <5

bit serum

precipitating antibody activity in 2D-IE with544/W extract. Absorption removed all HA ac-

tivity, but strong precipitating activity against,8 and y antigens remained in the antisera, thusconfirming the failure of 8 and y antigens to bindto SRBCs.

Similar experiments were performed by usingimmunoadsorbent columns containing Sepha-rose coupled to affinity chromatography-puri-fied ,8 and y antigen preparations. Samples of B5and C9 rabbit antisera and the infected bovineserum pool were absorbed through each column.Washes and eluates were tested by 2D-IE

against 544/W extract and for HA activityagainst SRBC-,1, SRBC-y, and SRBC-544/W.The results in Table 2 show that precipitatingactivity against/ or y antigens was confined tocolumn washes whereas HA reactivity was pres-ent predominantly in column eluates. Thesedata provide evidence that ,8 and y antigens wereneither coupled to Sepharose beads nor respon-sible for the discriminatory HA reactivity.Antigens responsible for discriminatory

HA activity. The presence of additional anti-gen(s), specifically bound by both B5 and C9Sepharose immunoadsorbents and tannedSRBCs, was investigated by passing /3 and yantigen preparations through the heterologousC9 and B5 immunoadsorbent columns, respec-tively. The washes and eluates were tested forHA activity against SRBCs sensitized with im-munoglobulins from the infected bovine serumpool. Most of the HA activity present in theapplied sample was bound by both B5 and C9immunoadsorbent columns and present in theeluate, showing that the antigenic componentsresponsible for HA discrimination between poolsof bovine sera from infected and vaccinated an-imals were bound reversibly by both B5 and C9immunoadsorbent columns and were thereforepresent in both affinity chromatography-puri-fied /, and y preparations. These componentswere designated antigen X. Since this antigenadsorbed equally to B5 and C9 antibody-Seph-


http://iai.asm.org/

Dow

nloaded from

http://iai.asm.org/


arose columns, attempts were made to deter-mine whether the interaction was due to im-munological or non-immunological binding. Im-munological binding might have been due to thepresence of "nonspecific" antibodies in B5 andC9 antisera, and non-immunological bindingmight have been due to direct adsorption of thisantigen by Sepharose gel.

Control columns were prepared with nonim-mune rabbit immunoglobulin coupled to acti-vated Sepharose CL4B and with unactivatednoncoupled Sepharose CL4B overlaid on Seph-adex G-25. Passage of 544/W extract throughthese columns showed that similar amounts ofHA-reactive material (antigen X) were boundand eluted to that obtained from B5 and C9immunoadsorbent columns. These results indi-cate that antigen X was bound by non-immu-nological adsorption directly to SepharoseCL4B. Precipitating antigens ft and y were notbound by these control columns.These results indicate that discrimination be-

tween infected and vaccinated bovine serum

pools by HA was not due to the precipitable a,A, or y antigen, but rather to another componentin the 544/W extract, designated antigen X,which was also present in affinity chromatogra-phy-purified ,8 and y antigen preparations. Thedata in Tables 1 and 2 indicate that tannedSRBCs and activated Sepharose beads selec-tively bind antigen X from crude extract andaffinity chromatography-purified antigen prep-arations. If this were so, then similar HA activitywould be seen with SRBC-,f, SRBC-y, andSRBC-544/W. Ninety-two characterized bovinesera from infected and vaccinated animals weretherefore coded and tested by HA against thesesensitized SRBCs.

Figure 3 shows that excellent discriminationbetween sera from infected and vaccinated cattlewas obtained irrespective of the antigen prepa-ration used for erythrocyte sensitization, con-firming the major involvement of antigen X inthe discriminatory ability of this system.Correlation between HA and bacterial

tube agglutination tests. The 92 bovine sera

TABLE 2. HA andprecipitating activity in bovineand rabbit antisera before and after passage through Sepharose-antigen immunoadsorbent columns'

Serum adsorbed

Sample B5 C9 Infected bovine pool

HA titer Lines in 2D- HA titer Lines in 2D- HA titer Lines in 2D-HAtiter IE HAttr IE HAttr IE

544/W control 40 / 5 y 10,240 a, /3, y,8 column wash <5 /3 <5 y <10 a,/, y,/ column eluate 20 None <5 None 2,560 Noney column wash <5 /3 <5 y 2,560 a,/, yy column eluate 20 None <5 None 2,560 None

a Sera were passed through Sepharose-/B antigen and Sepharose-y antigen immunoadsorbent columns. Thewash containing unadsorbed antibody and the eluate containing affinity chromatography-purified antibodywere tested in HA and 2D-IE. HA reactivity was tested with SRBCs sensitized with crude extracts and with ,/or y antigens. Analogous titers were found with each group of sensitized SRBCs.

_X--* _ _ m~hm_srcs-s4/w SqWc - A

0

i1_ ~~~~~~o

-- m -

0 owouwm~sSo

cx~ooscoo s*

00im 0

SRBCs 6 RC wTMPAGGLflNArm

FIG. 3. Sera from infected (x), strain 19-vaccinated (0), or strain 45/20-vaccinated (0) cattle.

910

x

I

VOL. 29, 1980

000

I


http://iai.asm.org/

Dow

nloaded from

http://iai.asm.org/


examined in the HA system were also tested bybacterial tube agglutination. Figure 3 shows thatwhereas HA titers were consistently low withsera from vaccinated animals, these sampleswere distributed throughout the titer range inbacterial agglutination tests, showing lack of cor-relation between the two test procedures.

DISCUSSIONSeveral authors have demonstrated precipi-

tation lines, similar to those reported in thisstudy, with a variety of Brucella extracts (10,12, 13, 23, 24). The differential precipitationresponse seen with infected and vaccinated cat-tle indicated that the use of selected antigens ina diagnostic system might enable improved dis-crimination between these two groups of sera(19). Purification of selected antigens from SDSextracts of B. abortus was attempted by usingimmunoadsorbents prepared from rabbit anti-sera raised against individual precipitation lines.

Affinity chromatography enabled purificationof antigens which were reactive in gel precipi-tation tests and, when bound to SRBCs, enabledexcellent differentiation of infected from vacci-nated bovine sera by passive HA. Subsequentexperiments showed a disassociation of precipi-tating and HA activity, indicating that the fi andy antigens were not involved in the discrimina-tory ability of the HA system. 2D-IE and HAtests on bovine antibodies separated by affinitychromatography confirmed that the / and yantigens were not coupled to sensitized SRBCsor activated Sepharose beads. The differentialreactivity of bovine sera in both HA and im-munofluorescence (19) systems was thereforedue to another component in the 544/W extractwhich was selectively bound by tanned SRBCsand activated Sepharose beads. The presence ofthis component, designated antigen X, in affinitychromatography-purified /3 and y preparationswas due to its non-immunological adsorptionfrom the crude extract (by Sepharose CL4B)and elution by sodium thiocyanate. Antigen Xhas not been described previously, probably be-cause most authors have used immunoprecipi-tation procedures for examination of antigeniccomponents in extracts. Antigen X did not forma precipitation line in gel tests with bovine an-tisera despite its extremely high activity in pas-sive HA systems. Non-precipitation may havebeen due to the presence of SDS, which canprevent the occurrence of immunological reac-tions (8, 11, 17), although this seems unlikely asSDS neither interfered with the reactivity ofantigen X in HA systems nor prevented precip-itation of other antigens in the 544/W extract.Furthermore, no differences were observed be-tween the precipitation characteristics of /f and

-y antigens in 544/W extract which contained0.5% SDS and affinity chromatography-purifiedantigen preparations which contained no freeSDS.The preferential binding of antigen X from

crude 544/W extract or affinity chromatogra-phy-purified /3 and y preparations indicates thatit is the major component responsible for theimproved discriminatory activity seen both inthe immunofluorescence system described pre-viously (19) and in this HA test. The mechanismof antigen binding to fixed, tanned SRBCs is notknown. However, the reactions involved in cou-pling to sodium metaperiodate-activated Seph-arose gels have been described by Sandersonand Wilson (22). Aldehyde groups created byoxidation of sugar subunits of the agarose gelcondense with amino groups present on the an-tigen molecule. Thus, a high concentration ofamino groups on antigen X might encouragepreferential binding by activated Sepharose.The presence of amino groups might also beresponsible for reversible adsorption of this an-tigen to nonactivated Sepharose and activatedSepharose coupled to nonimmune rabbit im-munoglobulins, as all cross-linked Sepharosegels contain a small number of charged groupswhich can cause adsorption of basic proteins(15).HA tests enabled accurate identification of

the infection status of animals in 95% of bovinesera tested, a considerable improvement overcurrent bacterial tube agglutination or comple-ment fixation tests (6, 7). Only five sera gaveinconclusive HA titers; two of these were sam-ples taken from an animal 14 and 28 days afterstrain 19 vaccination, and the other three werefrom an acute case taken within 6 weeks ofactual contact with another actively infectedanimal. These results suggest that serologicaldiscrimination between infected and vaccinatedcattle should be possible by systems utilizingantigen X from 6 weeks after inoculation withthe virulent or vaccine organism.The improved discrimination obtained with

serological systems involving antigen X may bedue to differential distribution of this componenton the surface of virulent and vaccine strains ofB. abortus or differences in the immune re-sponse to infection with brucellosis and vacci-nation against the disease. Studies are in prog-ress to clarify which of these mechanisms isresponsible.

ACKNOWLEDGMENTS

We thank J. K. Miller and J. I. Kelly of the VeterinaryInvestigation Centre, Midlothian, Scotland and D. J. O'Reillyand P. Reilly of the Veterinary Research Laboratory, Thorn-dale, Dublin, Ireland, for the generous supply of bovine sera.

INFECT. IMMUN.


http://iai.asm.org/

Dow

nloaded from

http://iai.asm.org/


LITERATURE CIMD1. Allan, G. S., R. J. Chappel, P. Williamson, and D. J.

MeNaught. 1976. A quantitative comparison of thesensitivity of serological tests for bovine brucellosis todifferent antibody classes. J. Hyg. 76:287-298.

2. Alton, G. G. 1978. Recent developments in vaccinationagainst bovine brucellosis. Aust. Vet. J. 54:551-557.

3. Alton, G. G., L. M. Jones, and D. E. Pietz. 1975.Laboratory techniques in brucellosis, 2nd ed. W.H.O.Monogr. Ser. No. 55.

4. Axelson, N. H. 1973. Intermediate gel in crossed and infused rocket immunoelectrophoresis. Scand. J. Immu-nol. 2(Suppl. 1):71-78.

5. Beh, K. J. 1974. Quantitative distribution of brucellaantibody amongst immunoglobulin classes in vacci-nated and infected cattle. Res. Vet. Sci. 17:1-4.

6. Chappel, R. J., D. J. McNaught, J. A. Bourke, and G.S. Allan. 1978. Comparison of the results of someserological tests for bovine brucellosis. J. Hyg. 80:365-371.

7. Chappel, R. J., D. J. McNaught, J. A. Bourke, and G.S. Allan. 1978. The diagnostic efficiency of some sero-logical tests for bovine brucellosis. J. Hyg. 80:373-384.

8. Crumpton, M. J., and R. M. E. Parkhouse. 1972. Com-parison of the effects of various detergents on antigen-antibody interactions. FEBS Lett. 22:210-212.

9. Diaz, R., P. Garatea, L M. Jones, and I. Moriyon.1979. Radial immunodiffusion test with a Brucella pol-ysaccharide antigen for differentiating infected fromvaccinated cattle. J. Clin. Microbiol. 10:3741.

10. Diaz, R., L M. Jones, D. Leong, and J. B. Wilson.1968. Surface antigens of smooth Brucellae. J. Bacte-riol. 96:893-901.

11. Dimitriadis, G. J. 1979. Effect of detergents on antibody-antigen interactions. Anal. Biochem. 98:445451.

12. Freeman, B. A., J. R. McGhee, and R. E. Baughn.1970. Some physical, chemical and taxonomic featuresof the soluble antigens of Brucellae. J. Infect. Dis. 121:522-527.

13. Hinsdill, R. D., and D. T. Berman. 1967. Antigens ofBrucella abortus. I. Chemical and immunoelectropho-retic characterization. J. Bacteriol. 93:544-549,

14. Ianelli, D., RI Diaz, and T. Manlio-Bettini. 1976. Iden-tification of Brucella abortus antibodies in cattle serumby single radial immunodiffusion. J. Clin. Microbiol. 3:203-205.

15. Pharmacia Fine Chemicals, Inc. 1979. Sepharose CL.In Gel filtration: theory and practice, p. 22. PharmaciaFine Chemicals, Inc., Piscataway, N.J.

16. Prince, A. M., H. Ikram, D. Chicot, R. Wright, J.Vnek, R. Neurath, A. Lippin, and S. Swiss. 1975. Anew reversed passive haemagglutination test for detec-tion of HBsAg. Vox Sang. 29:319-329.

17. Qualtiere, L F., A. G. Anderson, and P. Meyers. 1977.Effects of ionic and non-ionic detergents on antigen-antibody reactions. J. Immunol. 119:1645-1651.

18. Raybould, T. J. G., and S. Chantler. 1979. Comparisonof activation and blocking procedures in the use ofdefined antigen substrate spheres (DASS) for quanti-tative and visual assessment of conjugate activity. J.Immunol. Methods 27:309-318.

19. Raybould, T. J. G., and S. Chantler. 1979. Serologicaldifferentiation between infected and vaccinated cattleby visual and quantitative immunofluorescence usingBr. abortus antigen coupled Sepharose beads. J. Im-munol. Methods 30:37-46.

20. Rice, C. E., D. Cochrane, and J. Tailyour. 1966. Elec-trophoretic studies of sera from cattle vaccinated ornaturally infected with Br. abortus. Can. J. Comp. Med.Vet. Sci. 30:161-168.

21. Rice, C. E., D. Cochrane, and J. Tailyour. 1966. Ultra-centrifugal studies of sera from cattle vaccinated ornaturally infected with Br. abortus. Can. J. Comp. Med.Vet. Sci 30:270-278.

22. Sanderson, C. J., and D. V. Wilson. 1971. A simplemethod for coupling proteins to insoluble polysaccha-rides. Immunology 20:1061-1065.

23. Schurig, G. G., L M. Jones, S. L Speth, and D. T.Berman. 1978. Antibody response to antigens distinctfrom the smooth lipopolysaccharide complex in Bru-cella infection. Infect. Immun. 21:994-1002.

24. Stemshorn, B., and K. Nielsen. 1977. Bovine immuneresponse to Br. abortus: water soluble antigen precipi-tated by sera of some naturally infected cattle. Can. J.Comp. Med. 41:152-159.

25. Weeke, B. 1973. Crossed immunoelectrophoresis. Scand.J. Immunol. 2(Suppl. 1):47-56.

26. Wood, W. A., and M. J. Corbel. 1973. Concentrations ofbovine serum protein classes in relation to reactivity inserological tests for brucellosis. J. Comp. Pathol. 83:143-150.

VOL. 29, 1980


http://iai.asm.org/

Dow

nloaded from

http://iai.asm.org/

serological differentiation between infected and ... · abortus by gel diffusion techniques....

Documents