JOURNAL OF CLINICAL MICROBIOLOGY, June 1988, p. 1142-11460095-1137/88/061142-05$02.00/0Copyright C 1988, American Society for Microbiology

Immunologic and Genetic Comparison of Streptococcus equiIsolates from the United States and Europe

JORGE E. GALANt AND JOHN F. TIMONEY*

Department of Veterinary Microbiology, New York State College of Veterinary Medicine, Cornell University,Ithaca, New York 14853

Received 28 December 1987/Accepted 14 March 1988

A series of isolates of Streptococcus equi from the United States and Europe were compared by thebactericidal test, immunoblotting, DNA restrictions, and Southern hybridization analysis. All isolates testedwere sensitive to the same bactericidal serum. In addition, immunoblotting revealed no differences in Mproteins prepared by acid or mutanolysin extraction. Immunoblotting of acid extracts of the isolates withmucosal nasopharyngeal mucus from a convalescent horse revealed the presence of the 41,000- and 46,000-Mrpolypeptide fragments of the M protein of S. èqui known to be important in stimulating mucosal nasopharyn-geal immune responses. DNA restriction analysis of total cell DNA digests, as well as Southern hybridizationsusing an S. equi M protein gene probe, did not detect any differences among these isolates. Our results,therefore, confirm the antigenic homogeneity of the M proteins of S. equi isolates and suggest that variation inthis antigen is not a reason for the failure of commercial vaccines in the field. Interestingly, the protoplast Mproteins of all isolates showed remarkable size homogeneity, in contrast to the size variation reported in Mproteins of group A streptococci.

Streptococcus equi causes strangles, a highly contagiousdisease of the upper respiratory tract and adjacent lymphnodes of horses and donkeys. Two antiphagocytic virulencefactors have been identified on the surface of S. equi, thehyaluronic acid capsule and the M-like protein (2, 4, 16, 27).In acid extracts, the protein is present in a series of definedfragments; the immunologically most reactive fragmentshave molecular weights of about 29,000, 41,000, and 46,000(9, 26). The determinant that elicits bactericidal (opsonic)antibodies is present on the 29,000-molecular-weight frag-ment (26). The native M protein of S. equi is found as atrimer with molecular weights between 56,000 and 60,000(9).Although protection against group A streptococcus infec-

tions in humans is associated with the presence of serumbactericidal antibodies (14), these antibodies in horses cor-relate poorly with protection against S. equi (22). Instead,studies in our laboratory have demonstrated that protectionagainst strangles is correlated with the presence of mucosalnasopharyngeal immunoglobulin G and immunoglobulin Aantibodies to acid-extracted M protein fragments with mo-lecular weights of about 46,000 and 41,000 (8, 9). Bothbacterin and M protein extract-type vaccines have beenwidely used but have given disappointing results in the field(22). These vaccines are effective in stimulating serum

bactericidal antibodies but are ineffective in protectinghorses against natural exposure either because of antigenicvariations in the M protein or because they do not stimulatemucosal nasopharyngeal antibodies.

Unlike group A streptococci, for which at least 74 M typeshave been described (6), S. equi has only one M type, as

precipitin analysis and passive protection studies in micesuggest (2, 16). The only phenotypic variation reported hasbeen matte-colony variants due to the presence of lysogenicbacteriophage that encodes or controls the production of

* Corresponding author.t Present address: Department of Biology, Washington Univer-

sity, St. Louis, MO 63130.

hyaluronidase (24). However, the presence of the phagedoes not affect the M protein of the host organism (25).

In this paper, we describe the results of a comparison of a

number of U.S. and European isolates of S. equi collectedover a period of 10 years. Using sensitivity to a bactericidalserum, immunoblotting, DNA restriction analysis, andSouthern hybridization with an S. equi M protein geneprobe, we have been unable to detect any differences in theM proteins of the isolates, nor have we been able to detectdifferences either in the restriction endonuclease pattern oftotal cell DNA digests or in the size of the restrictionfragments that carry the M protein gene. This finding isconsistent with the conclusion that there is only one M typeof S. equi.

MATERIALS AND METHODS

Organisms. S. equi isolates are described in Table 1. Eachisolate came from a different outbreak and was stored at-70°C in Todd-Hewitt broth (THB) immediately after isola-tion. For antigen preparation, cultures were grown at 370C inTHB. Escherichia coli 1089 (lambda gtll/SEM7), carryingthe S. equi M protein gene, has been described elsewhere(9).

Bactericidal test. The sensitivity of 14 isolates of S. equi(Table 2) to a high-titer bactericidal serum was studied byusing a modification of the tests described by Lancefield (12)for group A streptococci and by Timoney and Eggers (22) forS. equi. Briefly, each strain was incubated overnight in

THB, standardized to give a Klett reading of 150 U, and thendiluted 1:100 in THB. Samples (60 ,ul) of this dilution of eachstrain were then separately added to equal quantities of a

strongly bactericidal serum from a horse (PP) with chronic S.equi pleuritis and incubated at 37°C for 30 min. Nonspecificphagocytosis was measured by treating samples (60 ,ul) ofeach strain with nonbactericidal serum from five weanlingfoals raised in isolation with no known exposure to S. equi.Both of the sera used in the test were heat inactivated for 30min at 560C before use in the assay. Following incubationwith serum, 100 ,ul of each of the treated bacterial suspen-

1142

Vol. 26, No. 6

on March 20, 2020 by guest

http://jcm.asm

.org/D

ownloaded from

IMMUNOLOGIC AND GENETIC COMPARISONS OF S. EQUI

TABLE 1. Origins and dates of isolation of S. equi isolates

Strain Origin Yr

Boldmani Ireland 1983CF32 New York 1980Battisti Ohio 1980F43 U.S. vaccine strainCrafty Red New York 1982V1331 SwedenMelody New York 1983Ireland 22 Ireland 1983Mitchell Barn New York 1982Sprinkle Misty New York 1981Headache New York 1983Lady Margaret Ireland 1983Top Sun New Jersey 1985e23 New York 1975elOa New York 1976e24 New York 1976e27 New York 1976e29 New York 1976e80-1 New York 1980e80-5 New York 1980e80-7 New York 1980

sions, containing about 3 x 104 CFU, was added to 0.9 ,ul offresh heparinized equine blood in a sterile polypropylenetube and rotated end over end in a tube rotator (BBLMicrobiology Systems, Cockeysville, Md.) for 120 min at37°C. The pony used as a source of blood had been raised inisolation with no known exposure to S. equi and had nodetectable antibody to S. equi M protein.Samples (100 ,ul) of the blood-bacteria mixtures were

removed at 120 min and diluted 1:10, 1:100, and 1:300 inTHB. These dilutions were then drop plated (25 ,uI per drop)in triplicate on dry chocolate blood agar plates. Colonycounts were performed following incubation at 37°C for 24 h.

Antigen preparations. The procedure of Lancefield andPerlmann (13) was used to prepare acid extracts of the S.equi cultures. Native M protein was obtained as describedelsewhere (9). Briefly, S. equi strains were grown overnightin 200 ml ofTHB, and cells were recovered by centrifugationat 7,000 x g for 15 min and washed twice with 20 mM

TABLE 2. Sensitivity of 14 isolates of S. equi tobactericidal serum

CFU of S. equi per mlS. equi strain Origin (yr) after 120 min"

Bactericidal serum Normal serum

MB'82 New York (1982) 2.0 x 103 15.7 x 103CF32 New York (1980) 7.3 83.0elOa New York (1975) 2.3 18.0e23 New York (1975) 6.7 110.0e24 New York (1976) 10.3 121.0e27 New York (1976) 12.0 102.0e29 New York (1976) 2.5 43.3e80-1 New York (1980) 5.0 58.7e80-5 New York (1980) 0 79.3e80-7 New York (1980) 1.7 14.0F43 4.0 49.3Battisti Ohio (1981) 3.7 43.7V1331 Sweden 0.3 50.7A89772 New York (1980) 2.3 24.0

' Counts are the means of three replications. Bactericidal serum (PP) wasfrom a horse with chronic pleuritis caused by S. equi. Normal serum was apool from five weanling foals with no known exposure to S. equi.

phosphate buffer (pH 7). Cells were suspended in 3 ml of 0.5M phosphate buffer (pH 6) containing 10 mg of lysozyme(Sigma Chemical Co., St. Louis, Mo.) per ml, 30 itg ofN-acetylmuramidase SG (Seikagaku Kogyo Co., Ltd., To-kyo, Japan) per ml, and 0.5 M sucrose, and the mixture wasincubated at 37°C for 1 h. Finally, protoplasts were sepa-rated by centrifugation at 8,000 x g, and supernatantscontaining the M protein were saved for further analysis.Immunoblot analysis. Acid-extracted or protoplast M pro-

teins from different S. equi isolates were separated bysodium dodecyl sulfate (SDS)-polyacrylamide gel electro-phoresis (PAGE) and transferred to nitrocellulose sheets (8).Blots of acid extracts were developed as described else-where (8) with nasopharyngeal mucus obtained from aconvalescent horse following experimental infection with S.equi CF32. Blots of protoplast M protein preparations andacid extracts were developed with antiserum preparedagainst purified M protein of S. equi CF32 (26).DNA restriction analysis. Total cell DNAs from S. equi

strains were obtained as follows (9). Cells were grownovernight in 200 ml of THB, recovered by centrifugation at10,000 x g, and washed twice with TE buffer (10 mM Tris[pH 7.6], 1 mM EDTA). Cells were suspended in 2 ml of TEbuffer containing 20 mg of lysozyme (Sigma) per ml and 50,ug of N-acetylmuramidase SG (Seikagaku Kogyo) per mland were incubated at 37°C for 1 h. TE (1 ml) containing 10mg of preincubated pronase was added, and the mixture wasincubated for 30 min at 37°C. Cells were lysed with 1 ml of20% SDS solution, adjusted to 50 mM NaCI, phenol ex-tracted three times, chloroform extracted twice, and etherextracted several times. DNA (2 ,ug) of each strain wastreated with 10 U of the appropriate enzyme in 50 ,ul ofrestriction buffer according to the instructions of the manu-facturer (Bethesda Research Laboratories, Inc., Gaithers-burg, Md., or Boehringer Mannheim Biochemicals, India-napolis, Ind.). Electrophoresis of DNA was carried out inhorizontal gels of 0.7% agarose in TBE buffer (89 mMTris-borate, 89 mM boric acid, 0.5 mM EDTA) in a subma-rine unit (Hoefer HE99; Hoefer Scientific Instruments, SanFrancisco, Calif.) at 15.5 V/cm for 18 h. Gels were stainedwith ethidium bromide and visualized by illumination withlong-wave UV light.

Southern hybridization. DNA digests were transferredfrom agarose gels to nitrocellulose membranes as describedby Southern (19). Cloned S. equi DNA fragments wereisolated from low-temperature gel agarose (SeaPlaque; FMCCorp., Marine Colloids Div., Rockland, Maine) as describedby Struhl (20) and were 32P labeled by nick translation.Hybridization was performed as described elsewhere (9).

RESULTSSensitivity of S. equi strains to bactericidal serum. All

isolates of S. equi were phagocytosed and killed in muchgreater numbers following treatment with the bactericidalserum (PP) than with the normal control serum (Table 2).Counts (CFU) after 120 min were about 8 to 190 times higherfor the normal serum than for the bactericidal serum.Western blot (immunoblot) analysis. Blots of acid extracts

from different S. equi isolates showed no differences whendeveloped with nasopharyngeal mucus from a convalescenthorse (Fig. 1). An acid extract of S. equi CF32, used toexperimentally infect the horse from which convalescentnasopharyngeal mucus was obtained, showed no differenceswhen compared with acid extracts of other isolates (Fig. 2).

Similarly, native M protein preparations of different S.equi isolates showed no differences when reacted in Western

1143VOL. 26, 1988

on March 20, 2020 by guest

http://jcm.asm

.org/D

ownloaded from

1144 GALAN AND TIMONEY

a b c d e f g h i k a b c d e f g h i97

68

%4m oe W

iw lu » M "

-29

18

FIG. 1. SDS-PAGE and immunoblot analysis of acid extractsfrom different S. equi isolates. The blot was developed with naso-

pharyngeal mucus obtained from a horse convalescent from stran-gles. Isolates are described in Table 1. Lanes: a, Boldmani; b, CF32;c, Battisti; d, F43; e, Crafty Red; f, V1331; g, Melody; h, Ireland 22;i, Mitchell Barn; j, Sprinkle Misty; k, Headache. Numbers on theright indicate molecular weights in thousands.

blots with antiserum to the M protein of S. equi CF32 (Fig.3). The 58,000- to 60,000-molecular-weight dimer of thenative M protein was present in all strains analyzed. Largerbands, presumably the result of association of the M proteinwith other cell wall components (6, 9), were also detectedand were the same in all strains tested. Native M proteinfrom CF32, the isolate used as a source of antigen forpreparation of M protein antiserum, showed no differenceswhen compared with native M proteins from other S. equistrains (Fig. 3).

Blots developed with normal rabbit serum and with mucusfrom horses negative for antibodies to S. equi showed noreactivities with acid extracts and native M protein prepara-tions (data not shown).

a b C d e f g h i i k m n

-97|l A Z _ ~-68__ 4* -

e,à u 4

*I."IIisIII.!1 2929* _

FIG. 3. SDS-PAGE and immunoblot analysis of native M pro-teins from different S. equi isolates. The blot was developed with S.equi M protein antiserum. Strains are described in Table 1. Lanes:a, Boldmani; b, CF32; c, Battisti; d, F43; e, Crafty Red; f, V1331; g,Melody; h, Ireland 22; i, Mitchell Barn. Numbers on the rightindicate molecular weights in thousands.

DNA restriction analysis. The restriction analysis ofHindIII digests of total cell DNAs from several S. equistrains is shown in Fig. 4. No differences were detected inthe restriction profiles of all the strains tested despite thesubstantial number of clearly resolved bands. Similar resultswere obtained with EcoRI restriction digests (data notshown).

st A B C E F C H 1

,*. 91 - 1 a`âE.`- 'L_P

I." *_#

FIG. 2. SDS-PAGE and immunoblot analysis of acid extractsfrom different S. equi isolates. The blot was developed with S. equiM protein antiserum. Isolates are described in Table 1. Lanes: a,Boldmani; b, CF32; c, Battisti; d, F43; e, Crafty Red; f, V1331; g,Melody; h, Ireland 22; i, Mitchell Barn; j, Sprinkle Misty; k,Headache; 1, e23; m, Top Sun; n, Lady Margaret. Numbers on theright indicate molecular weights in thousands.

FIG. 4. HindIII restriction analysis of total cell DNAs fromdifferent isolates of S. equi. Isolates are described in Table 1. Lanes:st, standard; A, Boldmani; B, CF32; C, Battisti; D, F43; E, CraftyRed; F, VI331; G, Melody; H, Ireland 22; I, Mitchell Barn.

Uuil*mM 45

4ff_29

-- 97

68

- 45

J. CLIN. MICROBIOL.

4p -mm r

4"

4m»

on March 20, 2020 by guest

http://jcm.asm

.org/D

ownloaded from

IMMUNOLOGIC AND GENETIC COMPARISONS OF S. EQUI

abcdef ghi j. a`.

Ilow W.- _ *` A&

* u4 M

e* :d ::

;

FIG. 5. Southern hybridization analysis of HindIlI restrictiondigests of total cell DNAs from different isolates of S. equi. Isolatesare described in Table 1. Lanes: a, Boldmani; b, CF32; c, Battisti; d,F43; e, Crafty Red; f, V1331; g, Melody; h, Ireland 22; i, MitchellBarn; j, Sprinkle Misty.

Southern hybridization analysis. Southern blots ofHindIII-digested total cell DNAs of several strains of S. equi probedwith the [32P]dCTP-labeled EcoRI fragment of the recombi-nant phage lambda gtll/SEM7 that contained the M proteingene of S. equi (9) failed to detect any differences in therestriction patterns of all the strains tested (Fig. 5). HindIIIwas selected because it cuts several times within the clonedfragment, thereby increasing the possibility of detectingrestriction differences within that fragment among the dif-ferent S. equi strains.

DISCUSSION

Strangles caused by S. equi continues to be one of themost serious bacterial diseases of horse populations through-out the world. Protection against the disease is more closelycorrelated with the presence of mucosal nasopharyngealimmunoglobulin G and immunoglobulin A antibodies toepitope(s) on acid hydrolytic M protein fragments withmolecular weights of 41,000 and 46,000 than with serum

bactericidal antibodies, as in the case ofgroup A streptococ-cus infections in humans (7, 8, 12). Commercial vaccinepreparations consisting of heat-killed organisms or M pro-

tein-rich extracts seem to afford little protection, althoughthey stimulate a good serum bactericidal response (22). Ourdata suggest that vaccine failure is not likely due to varia-tions in the M antigen.

Earlier precipitin and mouse passive protection studieshave suggested that, unlike group A streptococcus, S. equihas only one serotype (2, 16). Our studies using opsoniza-tion, Western blot, restriction endonuclease, and Southernblot analyses support this conclusion and confirm the anti-genic homogeneity of S. equi isolates.The bactericidal test commonly used by investigators

working on the immune responses to streptococcal M pro-

teins in fact measures an antibody that is opsonic. The termbactericidal became widely used because the assay of op-

sonic antibody is most conveniently and reliably performed

in a system in which the antibody activity is measured byreduction in counts of viable streptococci following incuba-tion with neutrophils in whole blood or in tissue culturemedium. The traditional bactericidal assay is therefore ameasure of the end result of both opsonizing activity andsubsequent killing in neutrophils (11). The assay used in ourstudy compared the numbers of viable S. equi followingtreatment with a strongly bactericidal equine serum (PP) or anonbactericidal serum pool free of S. equi antibody andsubsequent incubation in horse blood for 120 min. Differ-ences in counts between the two serum treatments for astrain of S. equi were taken as evidence of neutralization ofthe antiopsonic effect of the M protein of that strain by thetype-specific antibody in the bactericidal serum. All of thestrains in our study showed much lower counts after 120 minwhen pretreated with the bactericidal serum (PP) than whenpretreated with the normal serum pool, suggesting that theyall carried the same M protein antigen. This result wasconsistent with the findings of Bazeley (2), who demon-strated that antiserum produced against a single strain of S.equi protected mice against challenge by each of 32 strains ofthe organism in his collection. Two strains (e80-5 andV1331), however, in our study were phagocytosed and killedmore effectively than the others. This result is unexplainedbut could be due to reduced amounts of M protein orcapsular hyaluronic acid that resulted in greater phagocyto-sis of these strains. In addition, some strains may be moreresistant to intracellular killing than others. However, stud-ies to substantiate this have never been reported for group Cstreptococci.The 46,000- and 41,000-Mr acid hydrolytic fragments of

the M protein known to be important in stimulating mucosalnasopharyngeal immune responses were demonstrated in allthe isolates tested (Fig. 1). In addition, rabbit antiserum tothe M protein of S. equi CF32 gave identical reactions withacid-extracted and protoplast M proteins of all the isolatessurveyed (Fig. 2 and 3). This finding is of great relevance tothe potential success of vaccines that will stimulate localprotective immune responses, such as the avirulent strain ofS. equi developed in our laboratory (23). This strain isprotective in challenge studies and stimulates production ofmucosal nasopharyngeal antibodies to the 41,000- and46,000-Mr polypeptide fragments of the M protein of S. equi(23). This or a similar vaccine therefore should be able toprotect against most or all isolates of S. equi, since thesepolypeptide fragments are the same in S. equi isolates in theUnited States and Europe.

Restriction endonuclease analysis of total cell DNA hasbeen very useful in epidemiological studies and in straindifferentiation of several DNA and RNA viruses, as well asof bacteria (3, 14, 15, 17, 18). By this technique, differenceshave been detected in strains of bacteria that were homoge-neous by immunological analysis (21). Our restriction endo-nuclease (Fig. 4) and Southern hybridization (Fig. 5) analy-ses, using two different restriction enzymes, failed to detectany differences among the isolates of S. equi and so corrob-orate the immunological findings. In contrast to S. equi,Streptococcus zooepidemicus, a closely related group Cstreptococcus, has been shown to possess different M pro-tein types (M. M. Muhktar and J. F. Timoney, unpublisheddata). In addition, Skjold et al. (18) have recently found agreat variety of DNA fingerprints among strains of S. zoo-epidemicus of human and animal origin.Of interest is the finding of remarkable size homogeneity

in the protoplast M proteins of S. equi isolates (Fig. 3).Fischetti et al. (6) have reported a substantial size variation

1145VOL. 26, 1988

4b 164ba a M a

on March 20, 2020 by guest

http://jcm.asm

.org/D

ownloaded from

1146 GALÂN AND TIMONEY

in the M proteins of Streptococcus pyogenes not only amongdifferent M serotypes but also within the same M type. Morerecently, Hollingshead et al. (10) have postulated that sizevariation is due to duplication and deletions within the Mprotein gene, since it has extensive repeated domains knownto be hot spots for such genetic events (1, 5). Examination ofthe nucleotide sequence of the cloned M protein gene (9)may offer some clues as to why the S. equi M protein is soantigenically stable and shows no size variation.

LITERATURE CITED

1. Albertini, A. M., M. Hoefer, M. P. Calos, and J. H. Miller. 1982.On the formation of spontaneous deletions: the importance ofshort-sequence homologies in the generation of large deletions.Cell 29:319-328.

2. Bazeley, P. L. 1947. Studies with equine streptococci. 4. Crossimmunity to Streptococcus equi. Aust. Vet. J. 18:189-197.

3. Buchman, T. H., B. Roizman, G. Adams, and B. H. Stover. 1976.Restriction endonuclease fingerprinting of herpes simplex virusDNA: a novel epidemiological tool applied to nosocomial out-breaks. J. Infect. Dis. 138:488-498.

4. Erickson, E. D., and N. L. Norcross. 1975. The cell surfaceantigens of S. equi. Can. J. Comp. Med. 39:110-115.

5. Farabaugh, P. J., U. Schmeissner, M. Hoefer, and J. H. Miller.1978. Genetic studies of the lac repressor. VII. On the molecularnature of spontaneous hotspots in the lac 1 gene of E. coli. J.Mol. Biol. 126:847-863.

6. Fischetti, V. A., K. F. Jones, and J. R. Scott. 1985. Size variationof the M protein in group A streptococci. J. Exp. Med.161:1384-1401.

7. Fox, E. N. 1974. M proteins of group A streptococci. Bacteriol.Rev. 38:57-86.

8. Galan, J. E., and J. F. Timoney. 1985. Mucosal nasopharyngealimmune response of horses to protein antigens of Streptococcusequi. Infect. Immun. 47:623-628.

9. Galan, J. E., and J. F. Timoney. 1987. Molecular analysis of theM protein of Streptococcus equi and cloning and expression ofthe M protein gene in Escherichia coli. Infect. Immun.55:3181-3187.

10. Hollingshead, S. K., V. A. Fischetti, and J. R. Scott. 1986.Complete nucleotide sequence of type 6 M protein of the groupA streptococcus. J. Biol. Chem. 261:1677-1686.

11. Lancefield, R. C. 1957. Differentiation of group A streptococciwith a common R antigen into three serological types withspecial reference to the bactericidal test. J. Exp. Med. 106:525-544.

12. Lancefield, R. C. 1962. Current knowledge of the type-specificM antigen of group A streptococci. J. Immunol. 89:307-313.

13. Lancefield, R. C., and G. E. Perlmann. 1952. Preparation and

properties of type-specific M antigen isolated from group A type1 haemolytic streptococcus. J. Exp. Med. 96:71-82.

14. Marshall, R. B., B. E. Wilton, and A. J. Robinson. 1981.Identification of Leptospira serovars by restriction-endonu-clease analysis. J. Med. Microbiol. 14:163-166.

15. McClenaghan, M., A. J. Herring, and I. D. Aitken. 1984.Comparison of Chlamydia psittaci isolates by DNA restrictionendonuclease analysis. Infect. Immun. 45:384-389.

16. Moore, B. O., and J. T. Bryans. 1970. Type specific antigenicityof group C streptococci from diseases of the horse, p. 231-238.In J. T. Bryans and H. Gerber (ed.), Proceedings of the SecondInternational Conference on Equine Infectious Diseases, Paris,1969. S. Karger, Basel.

17. Robinson, A. J., P. Ramadass, A. Lee, and R. B. Marshall. 1982.Differentiation of subtypes within Leptospira interrogans sero-vars hardjo, balcanica and tarassovi by bacterial restriction-endonuclease DNA analysis (BRENDA). J. Med. Microbiol.15:331-338.

18. Skjold, S. A., P. G. Qure, L. A. Fries, M. Barnham, and P. P.Cleary. 1987. DNA fingerprinting of Streptococcus zooepi-demicus (Lancefield group C) as an aid to epidemiologic study.J. Infect. Dis. 155:1145-1150.

19. Southern, E. M. 1975. Detection of specific sequences amongDNA fragments separated by gel electrophoresis. J. Mol. Biol.98:503-517.

20. Struhl, K. 1985. A rapid method for creating recombinant DNAmolecules. Biotechniques 3:452-453.

21. Thiermann, A. B., A. L. Handsaker, S. L. Moseley, and B.Kingscote. 1985. New method for classification of leptospiralisolates belonging to serogroup pomona by restriction endonu-clease analysis: serovar kennewicki. J. Clin. Microbiol. 21:585-587.

22. Timoney, J. F., and D. Eggers. 1985. Serum bactericidal re-sponses to S. equi of horses following infection or vaccination.Equine Vet. J. 17:306-310.

23. Timoney, J. F., and J. E. Galan. 1985. The immune response ofthe horse to an avirulent strain of S. equi, p. 294-295. In Y.Kimura, S. Kotami, and Y. Shiokawa (ed.), Recent advances instreptococci and streptococcal diseases. Reedbooks, Bracknell,Berkshire, England.

24. Timoney, J. F., L. Pesante, and C. Ernst. 1982. Hyaluronidaseassociated with a temperate bacteriophage of Streptococcusequi, p. 145-146. In D. Schlessinger (ed.), Microbiology-1982.American Society for Microbiology, Washington, D.C.

25. Timoney, J. F., P. J. Timoney, and K. L. Strickland. 1984.Lysogeny and immunologically reactive proteins of Streptococ-cus equi. Vet. Rec. 115:148-149.

26. Timoney, J. F., and J. Trachman. 1985. Immunologically reac-tive proteins of Streptococcus equi. Infect. Immun. 48:29-34.

27. Woolcock, J. B. 1974. Purification and antigenicity of an M-likeprotein of Streptococcus equi. Infect. Immun. 10:116-122.

J. CLIN. MICROBIOL.

on March 20, 2020 by guest

http://jcm.asm

.org/D

ownloaded from