Studies on the Serology of Flagellar Antigens of Yersinia Enterocolitica and Related Yersinia Species

Zbl. Bakt. Hyg. A 261,299-310 (1986)

Studies on the Serology of Flagellar Antigens of Yersinia enterocolitica and Related Yersinia Species S. ALEKSIC,]. BOCKEMOHL, and F. LANGE Medizinaluntersuchungsanstalt, Institute of Hygiene, Hamburg, Federal Republic of Germany

Received October 2, 1985 . Accepted December 5, 1985

Abstract A total of 1242 strains of Y. enterocolitica, 104 strains of Y. frederiksenii, 95 strains of Y. kristensenii and 85 strains of Y. intermedia were serotyped with antisera against 56 0 antigens and 19 H antigens according to the extended antigenic scheme of Wauters, and with additional antisera against 4 somatic and 19 flagellar antigens not previously described. H antigens of Y. frederiksenii, Y. kristensenii, and Y. intermedia turned out to be rather homogeneous without distinct subfactors. In these species the scope of identified serovars was narrow. Flagellar antigens of Y. enterocolitica were mostly composed of several subfactors, leading to a total of 117 serovars identified in the species. A number of cross-reactions between Yersinia H antigens were observed which could be avoided by absorption without significant lowering of the titre. Flagellar antigens of Yersinia were monophasic, and species specific. The antigens remained stable after storage in agar stabs and repeated subcultures. The epidemiological value of serotyping is demonstrated by strains from three different sources. It is suggested serotyping of Yersinia strains should be performed in three steps: - 0 typing of the prevailing enteropathogenic Y. enterocolitica serogroups in the medical routine laboratory; - 0 and H typing of Y. enterocolitica by National Reference Centres applying a typing scheme reduced to this species; and - 0 and H typing of Y. enterocolitica, Y. frederiksenii, Y. kristensenii and Y. intermedia by specialized International Centres using an extended typing scheme. The need for international standards comparable to those established for Salmonella is emphasized.

Zusammenfassung Insgesamt 1242 Starn me von Y. enterocolitica, 104 Stamrne von Y. frederiksenii, 95 Starnme von Y. kristensenii und 85 Srarnme von Y. intermedia wurden mit Antiseren gegen 56 O-Antigene und 19 H-Antigene nach dem erweiterten Antigenschema von Wauters sowie zusatzlichen Seren gegen 4 neue 0- und 19 neue H-Antigene typisiert. Die GeilSelantigene von Y. frederiksenii, Y. kristensenii und Y. intermedia erwiesen sich als recht einheitlich ohne ausgepragte Subfaktoren; das Serovar-Spektrum war dementsprechend relativ

300

S. Aleksic,

J. Bockemiihl, and

F. Lange

eng. Die H-Antigene von Y. enterocolitica waren dagegen iiberwiegend aus mehreren Unterfaktoren zusammengesetzt: insgesamt 117 Serovare wurden bei dieser Spezies unterschieden. Bei einigen Yersinia-H-Antigenen wurden Kreuzreaktionen nachgewiesen, die sich durch Absorption ohne nennenswerten Titerverlust beseitigen lielSen. Die YersiniaCeifclanrigene sind monophasisch und Spezies-spezifisch, Die Antigene aller vier Spezies erwiesen sich auch nach langerer Lagerung im Agarstich sowie nach mehreren Passagen als stabil. Die epidemiologische Bedeutung der Serotypisierung wird am Beispiel von Stammen von verschiedenen Erkrankungsfallen dargestellt. Wir schlagen eine Serotypisierung von Yersinia-Stammen auf drei Ebenen vor: - O-Typisierung der wichtigsten enteropathogenen Serogruppen von Y. enterocolitica im medizinischen Routinelaboratorium - 0- und H-Typisierung von Y. enterocolitica nach einem auf diese Spezies reduzierten Antigenschema in nationalen Referenzlaboratorien - 0- und H-Typisierung von Y. enterocolitica, Y. frederiksenii, Y. kristensenii und Y. intermedia nach einem erweiterten Antigenschema in wenigen internationalen Referenzzentren. Die Griinde hierfiir werden diskutiert, und die Notwendigkeit internationaler Standards vergleichbar den en fur die Serotypisierung von Salmonellen wird besonders hervorgehoben.

Following basic studies on the serology of Yersinia enterocolitica by Knapp, Thai (10, lOa) and Winblad (17), Wauters and coworkers (15, 16) published a typing scheme of 34 0 antigens and 19 H antigens which subsequently was used by most workers. Later, Wauters (14) extended his scheme to a total of 56 0 antigenic factors. All typing schemes, however, except that of Knapp and Thai (10, lOa), included organisms formerly called "Yersinia-like" which were later defined as separate species by groups at the Centers for Disease Control, Atlanta/U.S.A., and the Institut Pasteur, Paris (2, 3, 4, 6, 7, 8, 13). Today, beside Y. pestis and Y. pseudotuberculosis, six Yersinia species are recognized: Y. enterocolitica, Y. frederiksenii, Y. kristensenii, Y. intermedia, Y. aldovae, and the fish-pathogenic Y. ruckeri. With a few exceptions in immunocompromised hosts, disease in humans and warmblooded animals is almost exclusively caused by strains of Y. enterocolitica. Therefore, for practical use, Aleksic and Bockemiihl (1) recently suggested a revised typing scheme, based on the antigenic scheme by Wauters et al. and including only antigens of the species Y. enterocolitica. Yersiniae are monophasic bacteria. Whereas flagellar antigens of Y. frederiksenii, Y. kristensenii and Y. intermedia are rather homogeneous, H antigens of Y. enterocolitica are mostly composed of several subfactors. In the typing scheme by Wauters et al. (14, 15, 16), only one combination of 0 und H antigens is given. However, among clinical and environmental isolates, a great diversity of H antigenic factors can be observed which has not been described previously. Therefore it appeared appropriate to report on the antigenic analysis of 1526 Yersinia strains, and to present the findings in a list of identified serovars.

Materials and Methods

Bacterial strains A total of 1526 Yersinia strains was examined. Of these, 169 cultures were gifts by colleagues from all parts of the world, especially H. H. Mollaret (Paris), G. Wauters (Brus-

Studies on the Serology of Flagellar Antigens of Y. enterocolitica

301

sets), S. Winblad (Lund), and R. Sakazaki (Tokyo). The remaining 1357 strains had been isolated in the Federal Republic of Germany, and most of them were sent to the National Reference Centre for Salmonella, Institute of Hygiene Hamburg, for identification. A total of 1242 strains (81.4%) belonged to Y. enterocolitica, 104 (6.8%) were identified as Y. frederiksenii, 95 (6.2 %) were Y. kristensenii, and 85 (5.6%) Y. intermedia. The cultures were conserved in nutrient agar stabs at room temperature. Species identification and biotyping was performed according to the methods of Bercouier et al. (2,3,4), Brenner et al. (7), and Ursing et al. (13). Production of antisera

o and H antisera were prepared in rabbits as described previously (1). Briefly, 0 antigens for immunization were prepared from cultures on nutrient agar incubated at 28°C for 48 h and heated at 100°C for 2.5 h. For H antigen production, the strains were passed several times through U shaped tubes with semisolid nutrient agar in order to induce sufficient motility. Then they were inoculated on plates containing soft agar according to Sven Card (5) and incubated at room temperature for 48 h. The culture was suspended in normal saline with 0.5% formalin, and used for immunization. Absorption of sera was performed by the method used for the preparation of Salmonella factor sera (11). 0 agglutinins were removed from H antisera by absorption with heated (100°C for 2.5 h) 0 antigens. Antisera were prepared against 560 antigens and 19 H antigens established by Wauters and his coworkers (14, 15, 16), using the strains recommended by these authors. Furthermore, antisera were prepared against 4 somatic and 19 flagellar antigens hitherto not considered by the available typing schemes. Serotyping

The strains were serotyped by slide agglutination with the growth of 24 h cultures from nutrient agar or Kligler iron agar (0 antigens), and soft agar according to Sven Card (H antigens). Cross-reacting antigens were analyzed quantitatively in the tube test. Tubes for 0 agglutination were incubated overnight at 50°C; those for H agglutination were incubated at 3rc. A total of 38 H antisera prepared from strains of Y. enterocolitica, Y. frederiksenii, Y. kristensenii, and Y. intermedia were tested, in addition to the 1526 strains described in this study, against the flagellar antigens of Y. aldovae, Y. ruckeri, Y. pseudotuberculosis (0 groups I-VI) and all recognized Salmonella H antigens. Results Besides the 56 0 and 19 H antigenic factors defined by the extended typing scheme according to Wauters (14) we found 4 additional 0 and 19 H antigens in single or various isolates, which were specifically associated with certain Yersinia species. The symbols given to these new antigens are set in brackets and should be considered as provisional in order to avoid confusion with the typing scheme by Wauters et al. As pointed out earlier (1), O-factors 1, 2a, and 2b are difficult to demonstrate and require incubation at lower temperatures (14). 0 antigens [A] and [C] were identified in Y. frederiksenii, [B] in Y. enterocolitica, and [D] in Y. kristensenii. H antigens [u], [v], [w], and [x] were associated with Y. enterocolitica, [pz] to [P6] and [Ps] to [Pll] with Y. frederiksenii, [oz] and [03] with Y. kristensenii, and [qz] to [qs] with Y. intermedia. All tested strains were monophasic. Flagellar antigens of Y. enterocolitica were in most cases composed of several antigenic factors. As shown in Table 1 a total of 117 serovars of Y. enterocolitica was identified.

302

S. Aleksic, J.Bockemiihl, and F. Lange

Table 1. Serovars of Y. enterocolitica

o antigen

H antigen

Biovar Origin

Number of strains

Representative strain

1, 2a, 3 2a, 2b, 3 3 3 3

a, b, c b, c a, b, c d, e, f, g, h [u]

3 5 4 1 3

chinchilla hare human, swine human human

1 1 313 1 33

IP 135 IP 178 IP 134, My 0 H 392/85 H 87/82

4,32 4,32 4,32 4, 32 4,32 4,32

a, b, c, i b, c, d, g b, e, f, i b, f, i c, d, e, g, i d, e, g, i

1 1 1 1 1 1

human surface water human human cheese human

5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5

a, b, c, d, i a, b, c, g a, c, g, i a, c, 1 b, c, d b, c, d, e, i b, c, e, i b, c, d, e b, c, d, i b, c, g b, c, i b, d, e, g, i b, d, i b, d, g, i c, d, i c, g g, i

2 1 1 2 1 1 1 1 1 1 1 1 2 1 1 1 1

1 human 5 human human 6 4 human surface water 5 bovine, human 43 1 human 8 human human, surface water 12 human 6 8 human 10 human 5 human 6 human surface water, sausage 7 4 human 4 human

H 364/82 H 315/83 H 63/80 H 86/82 H 84/80 IP 124, H 218/84 H 367/85 H 177/83 H 228/82 H45/81 H 212/80 H 98/83 H 252/81 H 85/81 H 59/80 H 18/81 H 222/82

5,27

a, b, c

5,27

b, c

2 3 2

human human dog, surface water

14 8 2

IP 47 H 427/84 IP 885, H 113/80

6,30 6,30 6, 30 6,30 6,30 6,30 6,30 6,30 6,30 6, 30 6, 30 6, 30 6,30 6,30

a, b, c, d, e, i a, b, d, g, i a, b, c, i b, c, d b, c, d, e, g, i b, c, d, e, i b, c, d, g, i b, d, e, f, g b, d, e, f, g, h b, d, e, g, i b, d, g b, d, g, i b, e, g, i b, g, h

1 1 1 1 1 1 1 1 1 1 1 1 1 1

human human, sewage human human human human human human human human human human human human

4 7 4 3 13 2 5 8 8 9 1 5 6 4

H 66/83 IP 102, H 502/83 H 96/80 H 153/83 H 94/83 H 97/83 H 86/81 H 92/83 H 102/83 H 67/83 H 138/84 H 233/82 H 57/83 H 286/84

2 3 4 3 3 2

H 61/81 H 89/80 IP 96 H 110/80 H 337/83 H 99/83

Studies on the Serology of Flagellar Antigens of Y. enterocolitica

303

Table 1. continued

o antigen

H antigen

Biovar Origin

6, 30 6, 30 6, 30 6,30 6, 30 6,30,47

b, g, i b, i d, e, f, g, h e, f, g, h, k g, i [x]

1 1 1 1 1 3

human human cheese human human drinking water

6,31 6,31 6,31 6,31 6,31 6,31 6,31 6,31 6,31 6,31 6,31 6,31 6,31 6,31

a, b, g a, c, i a, g, i b, c, d, e, i b, c, d, i b, c, i b, d, e, g, i b, d, g, i d, e, f, g, h d, f, g, h c, i f, g g, h g,l

1 1 1 1 1 1 1 1 1 1 1 1 1 1

human, surface water 2 human 4 human 6 human, water 10 human 8 surface water 6 human, sewage water 23 human 7 human 7 human 4 human, sewage 6 human 8 human' 3 human 8

H 78/80 H 437/80 H 19/81 IP1477,H140/84 H 217/81 H 135/80 H 91183 H 230/82 H 231182 H 116/82 H 438/80 H 435/80 H 1/81 H 436/80

7,8 7,8 7,8 7,8 7,8 7,8 7,8 7,8 7,8 7,8 7,8 7,8 7,8 7,8 7,8 7,8 7,8

a, b, i a, b, e, g, i b, c, d, e, i b, c, d, k b, c, i b, d, e, g, i b, e, f, i b, f, i b, g, i b, i c, d, i d, e, f, g, h, d, g, i d, i d, g, i, k g, i n

1 1 1 1 1 1 2 1 1 1 1 1 1 1 1 1

dog human human unknown human human, guinea pig human human human, subsoil water surface water, dog human guinea pig human human human human human

2 4 6 1 1 15 5 2 6 10 5 1 4 2 1 1 2

H 369/82 H 313/84 H 286/82 IP 589 H 80/85 H 343/82, IP 106 H 343/82 H 353/82 H 316/83 H 344/82 H 441/80 IP 106 H 440/80 H 439/80 H 60/80 H 306/83 H 267/84

8 8

b, e, f, i b, e, g, i

1 1

human human

16 1

IP 161 H 130178

9

a, b

2

human, swine

10 10, K1 10 10

b, c, e, g, i b, C, d, e, i b, c, e, f, k b, d, e, f, i

1 1 1 1

ice cream unknown human sewage

21

Zbl. Bakt. Hyg. A 26113

Number of strains 9 3 2 7 3 72

127 1 4 3 3

Representative strain H H H H H H

77/83 302/83 336/83 330/82 150/80 275/84

IP 383, My079b IP 500 IP 474, IP 679 H 101183 H 484/83

304

S.Aleksic, ]. Bockemiihl, and F. Lange

Table 1. continued

o antigen

H antigen

Biovar Origin

10 10, K]

b, e, f, k b, f, k

1 1

13

a, b, i

13,7 13,7 13,7 13,7

a, c, g d, g, i g, i n

2 1 1 1

human human human human, drinking water

14 14 14 14

b, c, d, e, g b, d, i m [v]

1 1 1 1

human human human human

15

(b), c

4

human

18 18 18

b, e, f, i b, d, e, g, i b, g, i

1 1 1

human human sewage

1 4 1

IP 846 H 256/85 H 697/83

19, 8

c

4

IP 842

19, 8 19, 8

d, e, g, i n

human, sewage, surface water human human, sewage

2 3

H 239/83 H 711/83

20

b, e, f, i

human

3

IP 845

21 21

b, e, f, i b, i

human human

9 1

H 404/85 IP 1110

22

b, d, e, i, k

human

3

IP 1367

36

n

oyster

IP 2222

41,42

b, d, e, g, i

human

IP 2223

41,43 41,43 41,43

a, b, c, i b, d, e, g, i n

1 1 1

human human human

47

[w]

3

human

[B]

[x]

3

drinking water

Total strains IP: H:

Number of strains

Representative strain

human human, sewage

5 7

IP 1501 IP 551

human

2

Atlanta 1568, 1211 H 434/80 H 236/80 H 433/80 IP 553

1 1 1 13 4 1 3 2

H 168/84 H 107/85 IP 480 H412/83

IP 614

2 2 4

H 251/85 H 243/83 IP 3235

IP 7184 129

253/84

1242

Culture collection, National Reference Centre for Yersinia, lnstitut Pasteur Paris Culture collection, National Reference Centre for Salmonella, Institute of Hygiene, Hamburg Proposed designations for new antigens are set in []. Factors in () are weakly or irregularly expressed

Studies on the Serology of Flagellar Antigens of Y. enterocolitica

305

As described previously (1), H factors a to k, m, n, as well as the provisional antigens [u], [v], [w], and [x], exclusively occurred in Y. enterocolitica. The serogroups 3 and 9 which in most parts of the world prevail in human intestinal yersiniosis usually were associated with antigenic factors a, b and a, b, c, whereas the enteropathogenic American strains of serogroup 8 have the complex flagellar antigen b, e, f, i. On the other hand, H factor d in various combinations was almost indicative of the non-pathogenic biovar 1 strains. The flagellar antigens remained stable after storage for several months in agar stabs and repeated subculturing. Complete serotyping proved to be of epidemiological value in distinguishing strains from different sources. During an outbreak of nosocomial infections in Northern Germany (Wuthe, pers. communication) 33 strains of serogroup 3 were isolated which could, however, be distinguished from typical cultures of this group by a new flagellar antigen [u] (Table 1). Likewise, the 0 group 8 isolate from a case in the Federal Republic of Germany, i.e. 8 : b, e, g, i, was shown to be different from the reference strain 8: b, e, f, i OP 161). Finally in 1982, a foodborne outbreak in the U.S.A. (9, 12) had been caused by Y. enterocolitica 13 : a, b, i which was different from the reference culture 13, 7: n OP 553) by both 0 and H antigens. Within the species Y. frederiksenii 12 serovars were identified (Table 2). Whereas in the typing scheme by Wauters et al. only one flagellar antigen p is given, additional 9H antigens were found in the present study and tentatively labelled [P2] to [P6] and [Pal to [Pld. All these antigens were specific for Y. frederiksenii.

Table 2. Serovars of Y. frederiksenii

o antigen

H antigen

Origin

3 4,33 9 16 17 35 38 39 44 44, 45 [AJ

[P2] P [P4] P

human, water human, water sewage human, swine, water unknown swine swine human human human water human

[C]

[Ps]

[P6J [PaJ [P9] [PlOJ [PlOJ [Pll] [P1J

Total strains

Number of strains 3 20 1 65 1 2 2 1 1 1 6 1

Representative strain H 176/80 H 682/83 H 538/83 IP 867 H 279/83 IP 3842 IP 7175 IP 7142 IP 7146 IP 7210 H 44/80 H 168/82

104

Proposed designations for new antigens are set in [ ]

As presented in Table 3, a total of 18 serovars could be distinguished in Y. kristensenii. The H antigens I, 0, r, s, and t were specific for this species (1). Two more flagellar antigens, provisionally designated [02] and [03], were identified.

306

S. Aleksic, J. Bockemiihl, and F. Lange

Table 3. Serovars of Y. kristensenii

o antigen 11,23 11,23 11,23 11,24 11,24 12,25 12,25 12,26 12,26 12, 26 12,26 16 28 28 46 50,51 52 [0]

H antigen

t r r t 0 r

I

[oJ] r t 0 s t [02] s [oJJ t

Origin

Number of strains

Representative strain

human, swine unknown water, sausage human, sausage, surface water animal hare, poultry, human, swine human, animal sheep, water, animal human human, water unknown water, poultry, swine water human, water mouse rodent human water

5 1 6 9 2 16 9 5 10 6 1 8 1 12 1 1 1 1

IP 105 H 19/83 H 23/80 IP 841 H 242/82 IP 490 H 93/83 IP 103 H 460/85 H 289/84 IP 1494 IP 1475 IP 1474 H 88/80 IP 7230 IP 7229 IP 7209 H 24/80

95

Total strains Proposed designations for new antigens are set in [ ]

Twelve serovars were distinguishable within Y. intermedia (Table 4). Besides H antigen q defined by Wauters et aI., four additional flagellar antigens, tentatively labelled [q2] to [qs] were identified. Table 4. Serovars of Y. intermedia

o antigen

H antigen

Origin

Number of strains

Representative strain

4, 33 17 17 19,8 37 40 48 49,51 52,53 52,54 55 57

q q [g2] q g q q q [gJ] q [g4] [gs]

human, bovine, water human, water human human human human human unknown water beaver water human

32 36 1 8 1 1 1 1 1 1 1 1

IP 1476 IP 955 H 266/84 H 207/84 IP 7224 IP 2677 IP 3960 IP 7231 IP 2842 IP 2835 IP 7554 IP 7569

Total strains Proposed designations for new antigens are set in [ ]

85

Studies on the Serology of Flagellar Antigens of Y. enteroco litica

307

Th ere exis ted antige nic relat ionships bet ween som e Yersinia H antigens which had been ana lyzed by cross -absorptio n tests. Thus, H ant igen 0 which was formerly believed to occur in Y. k ristensenii, Y. frederiksenii and Y. intermed ia (1) was shown to sha re only partial antigen s with q, [P2] , and [q4]' In Tabl e 5, cross -rea ctio ns bet ween flagellar ant igens of Yersinia, detected in the present study, are summa rized. All cro ssreactio ns co uld be removed by abso rp tion with the coa gglutinat ing antigen without a significant loss of the homologou s titre. Th e sera did not agglutina te th e H ant igens of reference stra ins of Y. aldovae and Y. ruckeri. Antise ra aga inst H : I of Y. k ristensenii cross-reac ted with th e flagellar ant igen of Y. pseudot uberculosis VI, and sera agai nst H : g of Y. enterocolitica (strain IP 102) agglutina ted the H : a antigen of Salmo nella paratyphi A.

Discussion Serotyping of Y. enterocolitica strain s is of practical importance for both bacterial diagn osis and epidemiological studies. 0 typing of certain sero gro ups should be available in every diagnostic laboratory as stra ins enteropathogenic for hum an s are limited to a few 0 groups (0 : 3; 0 : 9; 0 : 8; 0: 5, 27). On the other hand, related Yersinia species (Y. frederiksenii, Y. k ristensenii, Y. intermedia, Y. aldovae) are widel y distribut ed in th e environment but rarely cau se disease in humans or wa rm-blooded ani mals. Oc cassion ally, an 0 antige n typical of pathogenic Y. enteroco litica, as for instanc e 0 : 3 or 0: 9, ma y be encountered in o ne of the other species (Table 2). Such crossreaction s, which are not unu sual between species and genera of Ent erobacteriaceae, may ca use confusion if the prelim inary bacterial diagn osis is based on sero logical methods only. H antigens, on the other hand, are specifically associated with certain species, an d the ir identificat ion might be a valua ble diagnostic tool. Th e pr esent stu dy confirmed our pr eviou s ob servati on (1) th at H antige ns of Y. enterocolitica, Y. frederiksenii, Y. kristensenii and Y. intermedia were spec ies specific. Thus, H antigens a to k, m, n, as well as th e provision al antige ns [u], [v], [w], and [x], occurred onl y in Y. enterocolitica, flagellar antigens p, [P2] to [P6], and [Ps] to [PId in Y. [rederiksenii, 1,0, r, s, t, [02] an d [03] in Y. kristensenii, and H ant igens q, [q2] to [qs] in Y. intermedia. The symbols [0. ..], [p...] and [q. . .] given to our provision al new antigens do not indicate- a close relationship of H antigens within the species with which they are associated. In fact, they were markedly distinct, and our design ati on s were chosen to have these provisional antigens fit into the extended typing scheme of Wauters (14). On th e other hand, as shown in Table 5, cross-reactions of a number of flagellar antigens with unabsorbed H anti sera occur red. Such cross-reaction s could be removed by abs orption without markedly lowering the homologous titre . Unab sorbed sera, however, can cause false-positive slide agglutinations, so th e necessity of proper specificity testing of antiser a is emphas ized aga in. The flagellar ant igens of Y. frederiksenii, Y. k ristensenii and Y. intermedia turned o ut to be rather homogeneous, with out distinct subfactors. Thus, the sco pe of identified serovar s was narrow (Ta bles 2-4). Y. enterocolitica, on the other hand, was cha rac terized by a great diversity of its H an tigens, whic h were mostly composed of several antigenic factors. Th is diversity was especiall y pronounced in 0 gro ups 5 (17

IP 7230 IP 7569 IP 178 H 87/82 IP 490 IP 955 IP 383 IP 1474 IP 383 IP 490 IP 955 IP 106 IP 1497

q [Pl] [q4]

n

[01] [qs]

b, c [u]

0 q

a, b

s

a, b 0 q

d, e, f, g, h

0

[03]

P

[Pl]

[P3]

[P4]

[P6]

[PIO]

[P11]

IP 995 H 168/82 IP 7554

IP 614 IP 106

(,;.J

q

q [u]

lwl [x]

m

0

[v]

[u]

IP 955

IP 955 H 87/82

IP 480

IP 490

IP 480 H 266/84

d, e, f, g, h [Pll]

r

m [ql]

IP 106 H 44/80

P

IP 3842

IP 867

[qs]

[P6]

3

IP 955

q

[q4]

s

n

IP 7209 IP 955

(]'Q t1l

::l

O'

c-

:-n

::l 0.-

O'

F=

~,

~ t1l

0

I:/j

'--<

s-

[

r>

)-

en

[°1] q

00

a

[ql]

IP 2842 IP 7554 IP 7184 H 275/84

Strain No.

[qJ] [q4] [w] [x]

Cross-reaction with antigen

q

Antiserum

" This serum furthermore cross-reacted with flagellar antigen of Y. pseudotuberculosis VI Proposed designations f(}l"~newamigens are set in [ 1

t

IP 2222

b, c d, e, f, g, h

n

IP 102 IP 106

a, b, d, g, i d, e, f, g, h

m

H 110/80 H 137/84

Strain No.

b, f, i b, d, e, i

Cross-reaction with antigen

I"

Antiserum

Table 5. Cross-reactions between flagellar antigens of Yersinia

Studies on the Serology of Flagellar Antigens of Y. enterocolitica

309

serov ars), 0 : 6, 30 (20 sero vars), and 0: 6, 31 (14 serova rs), respecti vely, with a total o f 117 serova rs identified in th e species (Ta ble 1). Complete seroty ping of 0 a nd H ant igens may be a valua ble adjunct to epid emi ological studies. Through decad es, the appl icati on of the Kauffm ann-Wh ite scheme for Salmonella has provided convincing evid ence of this, and th e examples of cases of disease due to Y. enterocolitica quoted in th e present paper do so as well . The prepar ation of specific antisera is lab ori ous , and the reliability of a set of typin g sera can only be warr ant ed by its co ntin uo us use and specifici ty testing. We th erefore suggest sero typing Yersinia str ains to be performed in three stages: - The medical microbiological laboratory dealing with th e ro utin e diagnosis of inte stinal diseases sho uld dispose of a small set of 0 typing sera for enteropath ogen ic Y. enteroco litica pre va iling in their region (e.g. in central Euro pe fo r the diagnosis of serogro ups 0: 3, 0: 9, and 5, 27). - National refer ence cent res sho uld mak e ava ilable th e serotyping by 0 a nd H ant igens of Y. enteroco litica, since Y. [rederiksenii, Y. k ristensenii and Y. int erm edia ar e as a rule ap athogenic for hum an s and wa rm-bl ooded animals. For this purpose, the revised typ ing scheme by W auters et al. as suggested by Aleksic and Bock emuhl (1) might be applied. - A few specia lized internat io nal centres sho uld o ffer th e complete 0 and H serotyping of Y. enterocolitica, Y. frederi kse nii, Y. k ristensenii and Y. intermedia on th e basis o f the extended ant igenic scheme of Wa uters (14), supplemente d by additional anti gens (e.g. th ose suggested by th e pr esent autho rs). As such co mplete sero typ ing is laboriou s, st ra ins sho uld be submitte d through the national reference centres. The diversity of Yersinia strains in clini cal and environmen tal material requires the initiat ive of an interna tio na l wo rking gro up o r an inte rna tiona l reference cent re to defin e new sero vars or ant igens, to set up stan dards as to the preparation and specificity testin g of ant isera, and to make avail able th e informati on to the scientific world by yearly or occasional publications. For decad es, this procedure has been pr act iced with unquestionabl e success in the field of Salm on ella serology, and technical information is regularly published to ensure a uniform sta nda rd of the diagnostic all ove r th e world (11, 18). Acknowledgement. We are grateful to G. Wauters, Brussels, for his valuab le criticism.

References 1. Aleksi c, S. and j. Bockemiibl: Proposed revision of the Wauters er al. anti genic scheme for serot yping of Yersinia enterocolitica. ] . Clin. Microb ial. 20 (1984) 843-845 2. Bercovier, H., D. j. Brenner, j. Ursing, A . G. Steigerwalt, G. R. Fanning, J. M. Alonso, G. A . Carter, and H. H. Mol/aret: Characterization of Yersinia enterocolitica sensu stricto. Curr. Microbial. 4 (1980) 201-206 3. Bercouier, H., j. Ursing, D. j. Brenner, A. G. Steigerwalt, G. R. Fanning, G. A. Carter, and H. H. Mo l/aret: Yersinia kristensenii: a new species of Enterobacteriacae comp osed of sucrose-negative strains formerly called Yersinia enterocolitica or Yersinia enterocolitica·like. Curr. Microb iol. 4 (1980) 2 19-224 4. Bercovier, H., A. G. Steigerwalt, A. Guiyo ule, G. P. Carter, and D. j. Brenner: Yersinia aldovae (formerly Yersinia enterocolitica-like group X l ): new species of Enterobacteriaceae from aqu atic ecosystems. Int.]. system. Bact. 34 (1984) 166-178

310

S. Aleksic, J. Bockemiihl, and F. Lange

5. Bockemuhl, j. , S. Alek sic, R. Leimbeck , and H. E. Mul/er: Isolierung und Identifizierung von Enterobacteria ceae. Verfahrensrichtlinien der Deutschen Gesellschaft fur Hygiene und Mikrobiologie. Zbl. Bake Hyg., 1. Abr. Or ig. A 254 (1983 ) 1-25 6. Brenner, D. j.: Speciation in Yersinia. Contrib. Microb iol. Immunol. 5 (1979) 33-43 7. Brenner, D. j. , H . Bercovier, j. Ursing, j. M. Alonso, A . G. Steigerwalt, G. R. Fanning, G. A . Carter, and H. H. Mol/aret: Yersinia intermedia: a new species of Enteroba cteriaceae composed of rhamnose-positive, melibiose-posit ive, raffinose-positive strain s (formerly called atyp ical Yersinia enterocolitica or Yersinia enterocolitica-like). Curr o Microbiol. 4 (1980) 207-212 8. Brenn er, D. j. , j. Ursing, H. Bercouier, A . G. Steigerwalt, G. R. Fann ing, j.M. Alonso , and H. H. Mol/aret: Deoxyribonucleic acid relatedn ess in Yersinia enterocolitica and Yersinia enterocolitica-like organisms. Curr . Microb iol. 4 (1980) 195-200 9. Centers for Disease Control: Mult i-state outbreak of yersiniosis. Morb id. Mortal. Wkly Rep. 31 (1982) 505- 506 10. Knapp, W. and E. Thai : Untersuchungen iiber die kulturell-biochemischen, serologischen, tierexperimentellen und immunologischen Eigenschaften einer vorlaufig "Pasteurella X" benannten Bakterienart . Zbl. Bake, 1. Abt, Orig. 190 (1963) 472-484 lOa Knapp, W. and E. Thai: Differentiation of Yersinia enterocolitica by biochemical reactions. Contrib. Microbiol. Immunol. 2 (1973) 10-16 11. Le Minor , L. and R. Rohde: Guidelines for the preparation of Salmonella antisera. World Health Organization BAC 78.1 Rev. 1 (unofficial document), Geneva (1982) 12. Toma , S., G. Wauters, H . M. Mc Clure, G. K. Morris, and A . S. W eissfeld: 0: 13a, 13b, a new pathogenic serotype of Yersinia enterocolitica. J. Clin. Microbiol. 20 (1984) 843-845 13. Ursing, j. , D. J. Brenner, H. Bercouier, G. R. Fanning, A . G. Steigerwalt, j. Brault , and H. H. Mol/aret: Yersinia frederiksenii: a new species of Enterobacteriaceae composed of rhamnose-positive strains (formerly called atypical Yersinia enterocolitica or Yersinia enterocolitica-like). Curt. Microbiol. 4 (1980) 213-21 7 14. Wauters, G.: Antigens of Yersinia enterocolitica. In: E. j. Bottone (cd.), Yersinia enterocolitica, pp. 41-53. CRC Press Inc., Boca Raton (1981) 15. Wauters, G., L. LeMinor, and A. M. Chalon: Antigenes sornatiques er flagellaires des Yersinia enterocolitica. Ann. lost. Pasteur 120 (1971) 631-642 16. W auters, G., L. LeMin or, A . M. Cbalon, and j. Lassen: Supplement au scheme antigenique de Yersinia entero colitica. Ann. Inst. Pasteur 122 (1972) 951-956 17. W inblad, S.: Studies on 0 ant igen factors of Yersinia enterocolitica. Progr. lmmun obiol. Stand . 9(1968) 337-342 18. W HO Collabo rating Centre for Reference and Research on Salmon ella: Antigenic formulae of Salmonella, 4th Rev. lnstitut Pasteur, Paris (1984)

Priv. Doz. Dt. S. Aleksic, Prof. r», j. Bockemubl, F. Lange, Medizinaluntersuchungsansralt, Nat ionale Salmonella-Zentrale, Hygienisches Institut, Marckmannstr. 129 a, 0-2000 Hamburg 28