Re-Speciation of the Original Reference Strains of Serovars in the Citrobacter freundii (Bethesda- Ballerup Group) Antigenic Scheme of

Transcription

1 Microbiol. Immunol., 40(12), , 1996 Re-Speciation of the Original Reference Strains of Serovars in the Citrobacter freundii (Bethesda- Ballerup Group) Antigenic Scheme of West and Edwards Kanji Mild1, Kazumichi Tamura*,2, Riichi Sakazaki,2, and Yoshimasa Kosako3 of Clinical Pathology, Kobe City General Hospital, Kobe, Hyogo 650, Japan, 2Enterobacteriology Laboratory, National Institute of Health, Shinjuku-ku, Tokyo 162, Japan, and 3Japan Collection of Microorganisms, Riken, Wako, Saitama , Japan Received June 6, Accepted September 7, 1996 Abstract: The antigenic scheme for the Bethesda-Ballerup group of bacteria established by West and Edwards in 1954 has continued to be applied as a serotyping scheme for Citrobacter freundii. In 1993, however, the classification of the Citrobacter was drastically revised and the species C. freundii redefined by Brenner et al. Accordingly, to judge the propriety to continuously use a single antigenic scheme for the C. freundii complex, the 90 reference strains listed in the antigenic scheme for C. freundii by West and Edwards were characterized phenotypically and specified based on the revised classification. Of these 90 strains, two strains of Hafnia alvei and one of Escherichia coli were found. Among the remaining 87 reference strains, Citrobacter youngae was the predominant species (40 strains), followed by Citrobacter braakii (25 strains), Citrobacter werkmanii (13 strains), and the unnamed Citrobacter genospecies 10 of Brenner et al (six strains). Citrobacter freundii, as redefined, accounted for only three strains and ranked behind the other four species. No overlapping with most of the 42 O-groups and 82 H-antigens was recognized between species with few exceptions. O-groups 1-9 inclusive, which were estimated to represent more than 90% of the former C. freundii strains, occurred in strains of C. youngae and C. braakii; and all nine strains of O- group 29, formerly known as the Ballerup group, were identified as C. braakii. These findings suggest that further study of the serotyping system is needed for all H2S-producing Citrobacter species. Key words: Citrobacter freundii complex, Bethesda-Ballerup group Citrobacter freundii was first described by Braak (4) in 1928 as Bacterium freundii. Werkman and Gillen (26), in 1932, created the genus Citrobacter for the citrate-utilizing, lactose-fermenting coliform bacteria, and they classified B. freundii in this genus. However, their classification was not universally accepted, and these organisms were subsequently described under a variety of scientific and vernacular names such as Escherichia freundii (29), biotype of paracolon intermediate (24), Colobactrum freundii (3), Paracolobactrum intermedium (3), the Bethesda group (9), the Ballerup group (6) and the Bethesda-Ballerup group (27). In 1958, the International Subcommittee of Taxonomy of Enterobacteriaceae (19) adopted the term Citrobacter freundii for this group of organisms. Since *Address correspondence to Dr. Kazumichi Tamura, Enterobacteriology Laboratory, National Institute of Health, Toyama, Shinjuku-ku, Tokyo 162, Japan. address: Japan Institute of Biological Sciences, , Shinden, Ohme, Tokyo 198, Japan. then, all cultures of the hydrogen sulfide-producing Citrobacter species have been labeled C. freundii. Recently, Brenner et al (5) proposed significant changes in the classification of the genus Citrobacter on the basis of DNA relatedness. They designated six species, comprising C. freundii, C. youngae, C. braakii, C. werkmanii and two unnamed genospecies for hydrogen sulfide-producing Citrobacter strains, collectively forming the "C. freundii complex." Previously, in 1954, West and Edwards (27) established an antigenic scheme comprising 32 O-groups, extended later to 42 groups by Sedlak and Slajsova (23), which contained 167 serovars for the Bethesda-Ballerup group. This scheme continued to be applied for serotyping C. freundii strains even after the term Bethesda-Ballerup group was no longer used. Since this antigenic scheme for C. freundii is still in use, but may not be valid for the species as redefined, it is important that the scheme be reviewed in the light of current taxonomic changes. This study was accordingly carried out to determine 915

2 916 K. MIKI ET AL Table 1. Serovars and species assigned to the reference strains of the Bethesda-Ballerup group and their sources

3 REFERENCE SEROVAR STRAINS OF CITROBACTER FREUNDH 917 Table 1. (Continued) the species identity of the original reference strains of the serovars designated by West and Edwards and to clarify the antigenic relationships among the species included in the C. freundii complex. Materials and Methods Bacterial strains. A total of 90 reference strains described by West and Edwards (27) and Sedlak and Slajsoya (23), and listed in Table 1 was used. Of these, 80 strains for O-groups 1-32 inclusive were obtained originally from West in 1955 and the remaining 10 for O - groups inclusive were from the Czechoslovakian Collection of Microorganisms in 1987 held by the Enterobacteriology Laboratory of the National Institute of Health, Tokyo, where they were kept lyophilized. The antigenic formula of each strain was previously reconfirmed by Sakazaki et al (21), who serotyped a number of C. freundii isolates using antisera produced with the reference strains. Prior to characterization, all strains were removed from storage, reconstituted and passed twice on 5% sheep blood agar. Characterization. Initially, 52 biochemical tests comprising 47 described by Farmer (11) and an additional six for activities of Ĉ-glucuronidase, Ĉ-xylosidase, L-pyrrolidonyl peptidase and acid production from sorbose, 2- ketogluconate and 5-ketogluconate were carried out. The methods used for these tests were those described by Ewing (10) and Cowan (8). With few exceptions, all test preparations were incubated at 37 C as follows: four days for sugar fermentation, gelatin liquefaction and citrate and acetate utilization; two days for production of indole, H2S, acetoin, urease, Tween 80 esterase and amino-acid decarboxylases, motility, malonate utilization, growth in KCN broth, and esculin hydrolysis; and one day for phenylalanine deaminase production and nitrate reduction. DNase tests were incubated at 25 C for four days. The activities of Ĉ-galactosidase, Ĉ-glucuronidase, Ĉ-

4 918 K. MIKI ET AL xylosidase and L-pyrrolidonyl peptidase were tested with prepared paper discs dipped in a heavy suspension of the test cultures and incubated at 37 C for 4 hr. Identification. After preliminary characterization, 33 tests that were useful in separating each species as redefined by Brenner et al (5) and in differentiating between Citrobacter and other species of the family Enterobacteriaceae were selected. These tests included motility, Voges-Proskauer, production of indole, H2S, urease, lysine and ornithine decarboxylases, arginine dihydrolase, phenylalanine deaminase, DNase, Tween 80 esterase, three Ĉ-glycosidase and L-pyrrolidonyl peptidase, utilization citrate, malonate and acetate, esculin hydrolysis, gas production from glucose, and acid production from cellobiose, lactose, melibiose, raffinose, rhamnose, sucrose, adonitol, D-arabitol, dulcitol, glycerol, sorbitol, sorbose and a-methyl-d-glucoside. Identification was performed by computer using the method of Willcox et al (28) for calculation from the test results of the score values for the likelihood of match with those obtained from the identification database. The database for these calculations was prepared from the results cited by Farmer (11) and our own data. The Likelihood score and the Most Typical Likelihood score values for each reference strain was calculated, and the former then divided by the latter to obtain the Modal Likelihood score. After each calculation, a database taxon that exceeded 0.9 of the Normalized Likelihood score and 0.1 of the Modal Likelihood score was regarded as a match with the reference strain tested. Results Prior to the study on the 90 reference strains, calculation using our database was carried out on the 11 species based on the probabilities of each test described by Brenner et al (5) to confirm the reliability of identification by this system. For the calculation, tests indicating 0-50% probability in the table of Brenner et al were entered as minus and those showing % probability were input as plus. The Normalized Likelihood score and the Modal Likelihood score values exceeded 0.9 and 0.5, respectively, for each taxon, and there was no instance in which a database taxon matched more than two species with adjacent score values. Of the 90 reference strains for serovars of the Bethesda-Ballerup group, 87 were identified as members of the Citrobacter freundii complex by computer-assisted numerical analysis, and the remaining three strains were assigned to Hafia alvei (two strains) and Escherichia coli Table 2. Species identity and designated serovars of the reference strains of Bethesda-Ballerup group

5 REFERENCE SEROVAR STRAINS OF CITROBACTER FREUNDII 919 (one strain). The two strains of H. alvei gave typical results for this species, such as positive Voges-Proskauer, lysine and ornithine decarboxylases and malonate utilization tests, and a negative reaction for H2S production in TSI agar butt and sorbitol fermentation. The biochemical characteristics of the single strain identified as E. coli were rather atypical in indole production and some sugar reactions, and the Normalized and Modal Likelihood score values for this identification were much lower than 0.9 and 0.1 respectively. However, this strain showed Ĉ-glucuronidase activity and no L-pyrrolidonyl peptidase activity, which are typical enzymatic results for E. coli. The 87 strains of the C. freundii complex were divided into C. youngae (40 strains), C. braakii (25 strains), C. werkmanii (13 strains), genospecies 10 (six strains) by Brenner et al (5) and three strains of the redefined species C. freundii. Based on the original description by West and Edwards (27) and by Sedlak and Slajsova (23), the antigenic formula and species assigned to each strain and their sources are shown in Table 1. In most of the 42 O- groups no overlapping of O antigens among species was recognized, although five O-groups were found in more than one species: O-group 5 was present in C. braakii and Hafnia alvei; O-groups 7 and 8 were present in C. youngae and C. braakii; O-group 9 was present in C. youngae and genospecies 10; O-group 12 was present in C. werkmanii and genospecies 10, and O-group 13 was present in C. werkmanii, genospecies 10 and Hafnia Table 3. Biochemical characteristics of the reference strains of the Bethesda-Ballerup group

6 920 K. MIKI ET AL alvei. All nine strains of O-group 29, formerly known as the Ballerup group, were included in C. braakii. On the other hand, no overlapping of H-antigens was observed between species except for H-35,37; H-35,(36),28 and H- 40,41, but many minor antigens were found common among species. H-antigens designated for the H. alvei and E. coli strains were not found in the C. freundii complex. Although H-antigens of the strains for O - groups inclusive were not designated, it was determined previously that they have no relationship with any other H-antigens. Discussion From this study, it is clear that the antigenic scheme for C. freundii originally established for the Bethesda- Ballerup group by West and Edwards (27) is related to heterologous species. Of the 90 strains obtained originally from the author (West) as representative of each serovar, C. youngae was the most common species followed by C. braakii, C. werkmanii, and the unnamed genospecies 10; only three redefined strains of C. freundii were included. No strains of genospecies 11, which is defined as H2S positive, were found among the 90 strains. As C. freundii was formerly defined as indole negative, only such strains were, of course, included in the antigenic scheme by West and Edwards. Similar reasons were applied to the absence of genospecies 11 in the scheme, which is defined as indole-positive. It is not surprising that H. alvei and E. coli were included among the reference strains of the C. freundii complex, because, in 1954, some important tests such as amino-acid decarboxylases and KCN tolerance, or differentiating between species or genus of the family Enterobacteriaceae had not been developed. Moreover, when H. alvei was proposed by Moller in 1954 (17), he suggested that many strains of this species may give negative reactions to the Voges-Proskauer test in cultures incubated at 37 C. It is not clear from this study if the overlapping of some O- and H-antigens over two or three species indicates an antigenic relationship between the species or whether a single antigenic scheme is adquate for the serotyping of the C. freundii complex as the number of strains investigated was limited. Although the settlement of these points requires further study, the distribution of among the redefined species is of particular interest. From the results given by West and Edwards (27), it was estimated that more than 90% of the strains studied by them belonged to O-groups 1-9 inclusive, and 29. Similar results were also reported by Sakazaki et al (21). In this study, the reference strains for 1-9 inclusive denoted C. youngae and C. braakii. Furthermore, all nine strains of the O-group 29 were identified as C. braakii. 29 was originally assigned to organisms belonging to the Ballerup group. The first strain of this group was isolated from a patient with enteritis by Kauffmann and Moller (16), who proposed the name Salmonella ballerup because the organism resembled Salmonella in its sugar fermentation patterns, in producing H2S and in possessing a similar Vi-antigen to that of Salmonella serovar Typhi. Sakazaki (20) also reported the isolation of S. ballerup from a patient with severe gastroenteritis. Monteverde (18) described a strain with the Vi-antigen and the same O-antigen as that of S. ballerup but with a different H-antigen and named this serovar Salmonella hormaechei. Although the original strain of Salmonella hormaechei was isolated from the overy of a hen, another strain of this serovar was isolated from a patient with gastroenteritis (7). Approximately 20% of the total 506 strains employed in the serological study on the Betheda-Ballerup group by West and Edwards (27) were isolates from diarrheal illness, though it may be open discussion whether all of them were actual causative agents in reported cases or not. Strains of the C. freundii complex are usually regarded as commensal intestinal bacteria, and the status of the organisms as a causal agent of enteric diseases remains unclear. In recent years, however, reports of some strains of the C. freundii complex as well as those of H. alvei, which is also considered a normal intestinal inhabitant, are appearing in literature as a possible cause of enteric disease (1, 2, 12-14, 22, 25). Most of these reports were substantiated by molecular and epidemiological techniques. These findings suggest that some species or serovars of C. freundii complex may be enteropathogenic or that the possibility that diarrheagenic mechanisms required for disease production in these organisms may be acquired from traditional pathogens through genetic exchange. Further speciation and serotyping of this complex is needed to clarify these questions. Although only three refind strains of C. freundii were found among the 90 reference strains examined; therefore, this species may well be the predominant species in the clinical isolates of the complex. Janda et al (15) reported that although C. freundii ranked fourth behind C. youngae, C. braakii and C. werkmanii in frequency in their laboratory, it ranked first in the routine fecal isolates of the C. freundii complex. They considered that this difference may be related to the way in which the strains were selected or to the properties of strains referred to their laboratory. This aspect of species distribution also requires further study.

7 REFERENCE SEROVAR STRAINS OF CITROBACTER FREUNDII 921 References 1) Albert, Mi., Adam, K., Islam, M., Montanaro, J., Rahman, A. S.M.H., Haider, K., Hossain, M.A., Kibriya, A.K.M.G., and Tzipori, S Hafnia alvei, a probable cause of diarrhea in humans. Infect. Immun. 59: ) Albert, Mi., Faruque, S.M., Ansuruzzaman, M., Islam, M.M., Haider, K., Alam, K., Kabir, I., and Robins-Browne, R.M Sharing of virulence-associated properties at the phenotypic and genotypic levels between enteropathogenic Escherichia coli and Hafnia alvei. J. Med. Microbiol. 37: ) Borman, E.K., Stuart, C.A., and Wheeler, K Taxonomy of the family Enterobacteriaceae. J. Bacteriol. 48: ) Braak, H.R Onderzoekingen over Vergisting van Glycerine. Thesis, W.D. Meinema-Utitgeurer, Delft. 5) Brenner, D.J., Grimont, PAD., Steigerwalt, A.G., Fanning, G.R., Ageron, E., and Riddle, C.F Classification of citrobacteria by DNA hybridization: designation of Citrobacter farmeri sp. nov., Citrobacter youngae sp. nov., Citrobacter braakii sp. nov., Citrobacter werkmanii sp. nov., Citrobacter sedlakii sp. nov., and three unnamed genospecies. Int. J. Syst. Bacteriol. 43: ) Bruner, D.W., Edwards, P.R., and Hopson, A.S The Ballerup group of paracolon bacteria. J. Infect. Dis. 85: ) Carlquist, P.R., and Conte, L.M A new type of Salmonella. Bull. U.S. Army Med. Dept. 6: ) Cowan, S.T Cowan and Steel's manual for the identification of medical bacteria, 2nd ed, Cambridge Univ. Press, London. 9) Edwards, P.R., West, M.G., and Bruner, D.W Antigenic studies of a paracolon bacteria (Bethesda group). J. Bacteriol. 55: ) Ewing, W.H Edwards and Ewing's identification of Enterobacteriaceae, 4th ed, Elsevier, New York. 11) Farmer, J.J., III Enterobacteriaceae, p In Murray, P.R., Baron, E.J., Pfaller, M.A., Tenover, F.C., and Yolken, R.H. (eds), Manual of clinical microbiology, 6th ed, American Society for Microbiology, Washington, D.C. 12) Frankel, G., Candy, D.C.A., Everest, P., and Dougan, G Characterization of the C-terminal domains of intiminlike protein of enteropathogenic and enterohemorrhagic Escherichia coli, Citrobacter freundii, and Hafnia alvei. Infect. Immun. 67: ) Guarino, A., Capano, G., Malamisura, B., Alessio, M., Guandalin, S., and Rubino, S Production of Escherichia coli STa-like heat-stable enterotoxin by Citrobacter freundii isolated from humans. J. Clin. Microbiol. 25: ) Guarino, A., Gianella, R., and Thompson, M.R Citrobacter freundii produces as 18-amino-acid heat-stable enterotoxin identical to the 18-amino-acid Escherichia coli heat-stable enterotoxin (ST Ia). Infect. Immun. 57: ) Janda, J.M, Abbott, B.L., Chung, W.K.M., and Hanson, D.F Biochemical identification of citrobacteria in the clinical laboratory. J. Clin. Microbiol. 32: ) Kauffmann, F., and Moller, E A new Salmonella, S. ballerup with Vi antigen. J. Hyg. 40: ) Moller, V Distribution of amino acid decarboxylases in Enterobacteriaceae. Acta Pathol. Microbiol. Scand. 35: ) Monteverde, J A new type of Salmonella genus. Nature 154: ) Report of the Enterobacteriaceae Subcommittee of the Nomenclature Committee of the IAMS Recommended biochemical methods for group differentiation within Enterobacteriaceae. Int. Bull. Bact. Nomencl. Taxon. 8: ) Sakazaki, R Gastroenteritis caused by Salmonella ballerup. Med. Biol. 18: ( in Japanese). 21) Sakazaki, R., Namioka, S., Osada, A., and Yamada, C A problem on the pathogenic role of Citrobacter of enteric bacteria. Jpn. J. Exp. Med. 30: ) Schauer, D., and Falkow, S Attaching and effacing locus of a Citrobacter freundii biotype that causes transmissible murine chronic hyperplasia. Infect. Immun. 61: ) Sedlak, J., and Slajsova, M Antigenstruktur und Antigenbeziehungen der Gattung Citrobacter. Zbl. Bakt. Hyg., Abt. I. Orig. 200: ) Stuart, C.A., Wheeler, K., Rustigian, R., and Zimmerman, A Biochemical and antigenic relationships of the paracolon bacteria. J. Bacteriol. 45: ) Tschape, H., Prager, R., Streckel, W., Fruth, A., Tietze, E., and Bohme, G Verotoxigenic Citrobacter freundii associated with severe gastroenteritis and causes haemolytic uraemic syndrome in a nursery school: green butter as the infection source. Epidemiol. Infect. 114: ) Werkman, C.H., and Gillen, G.F Bacteria producing trimethylene glycerol. J. Bacteriol. 23: ) West, M.G., and Edwards, P.R The Bethesda-Ballerup group of Paracolon Bacteria. U.S. Public Health Serv., Monograph No. 22. U.S. Government Printing Office, Washington D.C. 28) Willcox, W.R., Lapage, S.P., and Holmes, B A review of numerical methods in bacterial identification. Antonie Van Leeuwenhoeck 46: ) Yale, M.M Genus I. Escherichia Castellani and Charmers, p In Breed, R.S., Murray, E.G.D., and Hitchens, A.P. (eds), Bergey's manual of determinative bacteriology, 5th ed, The Williams and Wilkins, Baltimore.