Species and serovars of the genus Listeria lsolated from different sources in Brazil from 1971 to 1997

Species and Serovars of the Genus Listeria lsolated from Different Sources in Brazil from 1971 to 1997

Vol. 95(5): 615-620, Sep./Oct. 2000

Ernesto Hofer⁺, Rosmery Ribeiro, Deise Paranhos Feitosa

Laboratório de Zoonoses Bacterianas, Departamento de Bacteriologia, Instituto Oswaldo Cruz, Av. Brasil 4365, 21045-900 Rio de Janeiro, RJ, Brasil

Using phenotype techniques, characterization was made to species and serovar of 3,112 strains of Listeria, isolated from different sources of infection such as human (247-7.9%) and animals (239-7.6%), as well as from various routes of infection, including food (2,330-74.8%) and environmental constituents (296-9.5%), all coming from different regions of the country and collected during the period 1971-1997. The following species were recovered in the cultures analysed: L. monocytogenes (774-24.8%), L. innocua (2,269-72.9%), L. seeligeri (37-1.1%), L. welshimeri (22-0.7%), L. grayi (9-0.2%), and L. ivanovii (1-0.03%). L. monocytogenes was represented by ten serovars, the most prevalent being 4b (352-11.3%), 1/2a (162-5.2%), and 1/2b (148-4.7%). The predominant serovar in L. innocua was 6a (2,093-67.2%). Considerations about laboratory methods for diagnosis and epidemiological aspects are presented on the basis of the results obtained.

Key words: Listeria - species - serovars - human - animals - environment - Brazil

The constituents of the genus Listeria are widely distributed in nature, although with a limited number of species of medical and veterinary importance (Schuchat et al. 1991, Low & Donachie 1997, McLauchlin 1997). Currently, listeriosis is considered to be an emergent disease, with its agent, L. monocytogenes, being mainly transmitted by food (Farber & Peterkin 1991). In humans, the disease presents clinical forms of a certain severity with a high socioeconomic impact among the food borne diseases (Roberts & Pinner 1990).

Among the various surveillance systems used for monitoring the disease or its etiologic agent, laboratory analysis data predominates (Rocourt 1988, Rocourt et al. 1997). ln most cases, this information is limited to a sporadic event with no clinical or epidemiological notification (Rocourt et al. 1997).

The basic laboratory methods used are usually concentrated on the isolation and phenotypic identification of species of the genus Listeria, followed less frequently by antigenic characterization of serovars (Rocourt & Seeliger 1985, McLauchlin et al. 1986, Farber & Peterkin 1991).

Considering the lack of information available in Brazil concerning these aspects, the present study was undertaken to assess the distribution and frequency of Listeria species and serovars from different areas of Brazil, isolated from various sources of infection and transmission vehicles during the period from 1971 to 1997.

We studied a total of 3,112 Listeria strains maintained for 26 consecutive years in the form of a collection in properly sealed tubes containing a thick layer of semisolid medium (Bacto Tryptose Broth plus 0.4% of Bacto Agar, Difco) and stored at 4-8^oC.

The cultures were shipped from different parts of Brazil by public and private entities in order to establish a conclusive laboratory diagnosis, or were isolated by the laboratory itself in the various investigations performed (Table I).

Table I

Identification and characterization of Listeria strains - They were assigned to the genus Listeria and its species using morphological and staining characteristics (Gram method) and biochemical tests: fermentation of D-xylose, D-mannitol, L-rhamnose and a-methyl-D-mannoside, plus routine analysis of hemolytic activity by the CAMP tests with Staphylococcus aureus and Rhodococcus equi (Rocourt et al. 1983).

Serogroups and serovars determinations were performed with polyclonal crossed absorbed factor antisera raised against somatic and flagellar Listeria antigens in rabbits, according to the method described by Seeliger and Höhne (1979). The antisera were obtained from the Department of Bacteriology, Instituto Oswaldo Cruz.

During this period 3,532 cultures with presumptive original diagnosis of Listeria were analysed. Of these, 3,112 were morphologically and biochemically compatible with the genus Listeria, with 88.2% positivity.

Table I shows a 96% predominance of pathological processes in human isolates, 78.5% of them involving patients from the States of São Paulo and Rio de Janeiro. Most animal strains (96%) were originated from apparently normal cattle slaughtered in an abattoir in the State of Rio de Janeiro (Hofer 1972, 1983).

Among the strains obtained from foods consisting of dairy products (9.7%), meat products submitted or not to industrial processing (89.9%) and vegetables (0.25%), particularly the samples from Goiás, with a predominance of isolations from chicken carcasses and meat (98.9%). ln a similar way, among isolations from the environment, most strains analysed were obtained from Goiás (88.8%), isolated from the effluents of meat-processing plants, strains from Rio de Janeiro (11.1%) from studies of different soils (Hofer & Póvoa 1984) and also material entering from a sewage treatment plant (Hofer 1975).

ln the characterization of species of the genus Listeria (Table II), there was a prevalence of L. innocua, which represented 72.9% of the strains analysed, obtained mostly from foods, sewage, meat-processing plants, and apparently healthy carrier animals (cattle).

Table II

Among the strains of human origin, there was a prevalence of L. monocytogenes over the other species, being the latter always associated with temporary carrier status and fecal excretion.

lt is interesting to point out the fortuitous occurrence of L. ivanovii isolated from the feces of an apparently normal cow, and the detection of L. seeligeri and L. welshimeri only in foods: L. seeligeri was more frequent in milk and in poultry, while L. welshimeri was predominantly in sausages and poultry.

Analysis of serovar distribution (Table III) showed that L. monocytogenes was represented by ten serotypes, although it was concentrated in three antigenic types in 85.4% of the cases: 4b (45.4%), 1/2a (20.9%) and 1/2b (19.1%). No isolation source revealed all L. monocytogenes serovars or exhibited a significant difference in the number of serovars identified

Table III

Some data listed in Table III concerning L. monocytogenes serovars are worthy of note, such as the predominance of serotype 4b in specimens from humans and meat products; the higher frequency of serovar 4a in animals, particularly healthy carriers; of 1/2a in dairy products; and of 1/2b in the environment. ln general, serovar 4b was the most frequently serovar isolated among the antigenic representatives of the five Listeria species identified.

For the other species (Table III), we emphasize the high prevalence of serovar 6a of L. innocua (67.2%), as well as the occurrence of serovar 4ab in L. innocua of animal origin and of serovar 1/2b in L. welshimeri from meat products.

An immediate effect of the intensification of studies concerning various aspects of Listeria over the past few decades has been an increased detection of Listeria from different sources, especially foodstuffs. This also holds good for Brazil, as can be seen in Table I, representing 26 years of uninterrupted analyses.

This clearly implies a necessity for the bacteriologic support of prospective epidemiological studies by the phenotypic characterization of isolates, indicating the genus, the species and the respective serovars (Rocourt & Seeliger 1985). Considering these aspects, the first step is the unequivocal identification of L. monocytogenes, responsible for human and animal listeriosis, in order to differentiate it from the remaining species, particularly L. innocua. Secondly, more accurate processes are used for somatic and flagellar antigenic characterization in combination with phagetyping and various molecular techniques (McLauchlin et al. 1986, Boerlin & Piffaretti 1991, Brosch et al. 1994, Pereira et al. 1994).

On the basis of the present results, we emphasize certain aspects such as the relatively easy identification of isolates of the genus Listeria, with an 88.2% rate of accuracy, although in 59.7% of this sample (1,860 cultures) there had been no definition of species or recognition of serogroups and serovars, or even mention that this analysis had been done. lt should also be pointed out that of the 1,374 cultures with species definition, we were able to confirm the original identification for 1,210 (88%). These cultures consisted of strains mostly from human sources and, also from animal sources, as explained by the fact that in most cases the bacteria were isolated from clinical specimens with a defined pathology (Hofer 1971, 1972, 1983, Hofer et al. 1984, 1998).

The major source of error in the primary phase of identification has been the conflicting interpretation of the bacterioscopic examination, especially in the presence of Gram-positive cocci arranged in pairs often simulating cocco-baciliary forms or, to a lesser extent, irregularly arranged Gram-positive or labile rods, or, even more rarely, Gram-negative bacilli. Most cultures not confirmed as Listeria (416) were excluded by bacterioscopy. During this phase, a search for catalase was originally rarely performed, although it represents the basic differentiation between Listeria and the genera Enterococcus and Streptococcus, with Enterococcus predominating in food and environmental isolates. We also draw attention to the possibly incorrect execution and reading of the motility test, sometimes performed with inadequate media associated, or not, with inappropriate inoculation procedure, temperature and incubation time. lt is interesting to note that most of the 416 cultures not confirmed as Listeria (383 or 92%) were isolated from food, and 271 of them were non motile at room temperature.

Another problem observed in this investigation was the past use of countless biochemical tests, some of them being of no differential or specific importance. We emphasize the routine use of the scheme proposed by Rocourt et al. (1983), which is simple and accessible to most laboratories. lt clearly differentiates species of the genus Listeria, with the fundamental requirement of detection of hemolysis either by direct observation on blood agar or, better still, by the CAMP test. The exclusion of this detail or an inadequate reading may have been the most frequent cause of error in the differentiation between L. monocytogenes and L. innocua in the past. ln the present this identification can be easily obtained by a miniaturized process containing ten substrates (Bille et al. 1992).

Analysis of Table II supports the views discussed above, considering that 1,648 (99.3%) cultures of the 1,658 strains isolated from clinical specimens, food and environmental material in the State of Goiás did not present previous species identification.

Table II shows that most Listeria species were detected in food, especially L. innocua and L. monocytogenes in a similar way, in the environment and, also in animal sources. This result was obtained by analysis of feces from cows acting as apparently healthy carriers, but passing the pathogenic species L. monocytogenes and L. ivanovii into the external environment (Hofer 1972, 1983).

L. monocytogenes predominated in human origin strains (95.9%), frequently involved in pathological processes in the central nervous system and/or with systemic distribution, reaching defined age groups and more rarely detected in other conditions or in healthy carriers (Hofer 1974, Nojimoto et al. 1997).

The most prevalent serovars of L. mono-cytogenes were usually limited to 4b, 1/2a and 1/2b, and, less frequently, 4a, 1/2c and 4ab (Table III). This result demonstrates that the antigenic characterization, represented by 13 serotypes, has limitations for epidemiological analyses due to the prevalence of a discrete number of serovars in most isolation sources, and due to the fact that this property is of cosmopolitan nature. This, of course, does not invalidate the process of epidemiological investigation, since it is a fundamental point for the determination of the phenotypic profile, favoring the adoption of other more discriminatory methods such as phagetyping. Even so, there are some obstacles such as the large proportion of non-typable L. monocytogenes strains of serogroup 1/2 (McLauchlin et al. 1986), despite the addition of new phage preparations to the scheme, as opposed to the relative success of phagetyping for serovar 4b (Estela & Sofos 1993). These problems were actually observed in a part of the present sampling (107 cultures), mostly from human source origin, 79 of which (73.8%) permitted the recognition of lysotypes, with serovar 4b and 1/2a making up 78.3% and 63.1% of the typable samples, respectively (Nunes 1990).

Epidemiologically, the distribution of L. monocytogenes serovars (Table III) demonstrates the prevalence of 4b serotypes from human origin and from food, differently from data reported by Nicolas et al. (1989), Farber and Peterkin (1991) and McLauchlin (1997). These investigators reported that most part of Europe the serogroup 1/2 predominates in foods of any kind, with the emergence of serovar 1/2c (Rocourt 1988). Paradoxically this event has no major effect on cases of human listeriosis, especially during outbreaks, in which serovar 4b prevails (Jacquet et al. 1995). ln contrast, in the United States of America and in Canada (Schuchat et al. 1991, Farber & Peterkin 1991) comparison of serotypes isolated from human and from food reveals a certain identity, with serovars 1/2b and 1/2c being most commonly characterized, although in four of the five outbreaks of toxinfection that occurred in these regions, involving 254 persons, serovar 4b was the responsible agent (Farber & Peterkin 1991).

On the other hand, serovar 1/2a played a more effective role in the so-called sporadic cases of toxinfection recorded in countries of the northern hemisphere, as well as in Australia and New Zealand (McLauchlin 1997). ln a world calculation involving 29 events of outbreaks and of sporadic cases of toxinfection, McLauchlin (1997) reported the presence of serovar 4b and serogroup 4 in 72.4% of the events, as in opposition to a 27.5% frequency of serotypes 1/2a and 1/2b.

No reference to the above problem has been made in Brazilian studies, but the wide dissemination of the more incident serovars in the various sources (Table III) permits us to assume that the transmission of listerias by food, including fish (Hofer & Ribeiro 1990), represents one of the most important mechanisms of transmission.

With respect to the environment, and especially the studies on soil (Hofer & Póvoa 1984), the most frequent serovars were quite similar to those reported by Weis (1975). The data also demonstrated the resistance of Listeria in soil and the occurrence of contamination from animal hosts, especially in effluents from abattoirs.

ln summary, the results obtained till now agree with those reported in several parts of the world (Rocourt & Seeliger 1985, Schuchat et al. 1991, Farber & Peterkin 1991, Low & Donachie 1997, McLauchlin 1997, Rocourt et al. 1997), demonstrating that members of the genus Listeria are widely disseminated in various ecological systems and epidemiological cycles, in which food represents a common vehicle of transmission to countless sources of infection. This agrees with Jacquet et al. (1995) who emphasized the importance of the information obtained rapidly and in a relativeiy easy manner by phenotyping, and showed that this method is indispensable and should precede more refined analyses of a genotypic nature.

To colleagues from public and private institutions for providing the bacterial strains; to Dr Zéa C Lins Lainson for critical reading of the manuscript and to Evaldo Soares da Silva, Sérgio Alves Azevedo and Darcília Maria de Andrade for technical assistance.

Table I | Table II | Table III

The authors dedicate this paper in honor of the Oswaldo Cruz lnstitute on the occasion of the centenary of its foundation, 25 May 1900.

Received 30 August 1999

Accepted 22 March 2000

• Bille J, Catimel B, Bannerman E, Jacquet C, Yersin MN, Caniaux 1, Monget D, Rocourt J 1992. API Listeria, a new and promising one-day system to identify Listeria isolates. Appl Environ Microbiol 58: 1857-1860.
• Boerlin P, Piffaretti JC 1991. Typing of human, animal, food, and environmental isolates of Listeria monocytogenes by multilocus enzyme electrophoresis. Appl Environ Microbiol 57: 1624-1629.
• Brosch R, Chen J, Luchansky JB 1994. Pulsed-field fingerprinting of Listeriae: identification of genomic divisions for Listeria monocytogenes and their correlation with serovar. Appl Environ Microbiol 60: 2584-2592.
• Estela LA, Sofos JN 1993. Comparison of conventional and reversed phage typing procedure for identification of Listeria spp. Appl Environ Microbiol 59: 617-619.
• Farber JM, Peterkin PI 1991. Listeria monocytogenes, a foodborne pathogen. Microbiol Rev 55: 476-511.
• Hofer E 1971. Presença de Listeria monocytogenes em material encefálico de bovino. Arq lnst Biológico (Săo Paulo) 38: 285-287.
• Hofer E 1972. Studies on a strain of Listeria (L. grayi) isolated from bovine nasal secretion. Rev Microbiol (Săo Paulo) 3:101-102.
• Hofer E 1974. Pesquisa sobre a ocorręncia de Listeria monocytogenes em fezes humanas. Rev Soc Bras Med Trop 3:109-116.
• Hofer E 1975. Isolamento e caracterizaçăo de Listeria monocytogenes em água de esgoto. Mem lnst Oswaldo Cruz 73: 31-38.
• Hofer E 1983. Bacteriologic and epidemiologic studies on the occurrence of Listeria monocytogenes in healthy cattle. Zbl Bakt Hyg A 256: 175-183.
• Hofer E, Póvoa MM 1984. Pesquisa de Listeria monocytogenes em solos. Mem lnst Oswaldo Cruz 79: 45-93.
• Hofer E, Ribeiro R 1990. Ocorręncia de espécies de Listeria em camarăo industrializado. Rev Microbiol (Săo Paulo) 21: 207-208.
• Hofer E, Nascimento RS, Oliveira MA 1998. Meningite por Listeria monocytogenes. Relato de casos em pacientes do Distrito Federal. Rev Soc Bras Med Trop 31: 173-177.
• Hofer E, Pessôa GVA, Melles CEA 1984. Listeriose humana. Prevalęncia dos sorotipos de Listeria monocytogenes isolados no Brasil. Rev lnst Adolfo Lutz 44: 125-131.
• Jacquet Ch, Catimel B, Brosch R, Buchrieser C, Dehaumont P, Goulet V, Lepoutre A, Veit P, Rocourt J 1995. lnvestigations related to the epidemic strain involved in the French listeriosis outbreak in 1992. Appl Environ Microbiol 61: 2242-2246.
• Low JC, Donachie W 1997. A review of Listeria monocytogenes and listeriosis. Vet J 153: 9-29.
• McLauchlin J 1997. The pathogenicity of Listeria monocytogenes: a public health perspective. Rev Med Microbiol 8: 1-14.
• McLauchlin J, Andurier A, Taylor AG 1986. Aspects of the epidemiology of human Listeria monocytogenes infections in Britain 1967-1984; the use of serotyping and phagetyping. J Med Microbiol 22: 367-377.
• Nicolas JA, Espaze EP, Catimel B, Vidaud N, Rocourt J, Courtieu AL 1989. Isolation of Listeria from French meat products. Zbl Bakt 272: 242-247.
• Nojimoto ITI, Souza SR, Valadăo LM 1997. Ocorręncia de Listeria spp. em crianças da cidade de Goiânia-Goiás. Rev Bras Anal Clin (Rio de Janeiro) 29: 73- 74.
• Nunes ZG 1990. Avaliaçăo de Marcadores Fenotípicos Associados ŕ Patogenicidade e a Epidemiologia do Gęnero Listeria, MS Thesis, Instituto Oswaldo Cruz, Rio de Janeiro, 85 pp.
• Pereira ML, Brosch R, Catimel R, Rocourt J, Hofer E 1994. Characterization of Brazilian Listeria monocytogenes strains using DNA macrorestriction patterns. Rev Microbiol (Săo Paulo) 25:144-148.
• Roberts T, Pinner R 1990. Economic impact of disease caused by Listeria monocytogenes. ln AJ Miller, JL Smith, GA Somkuti (eds), Foodborne Listeriosis, Elsevier Sci Publ, Amsterdam, p. 137-149.
• Rocourt J 1988. Identification and typing of Listeria. In Foodborne Listeriosis, Proceedings of a Symposium, Behr Verlag, Wiesbaden, p. 9-27.
• Rocourt J, Seeliger HPR 1985. Distribution des espčces du genre Listeria. Zbl Bakt Hyg A 259: 317-330.
• Rocourt J, Jacquet Ch, Bille J 1997. Human Listeriosis 1991-1992. WHO/FNU/FOS/97.1, World Health Organization, Geneva, 47 pp.
• Rocourt J, Schrettenbrunner A, Seeliger HPR 1983. Differénciation biochimique des groupes génomiques de Listeria monocytogenes (sensu lato). Ann Microbiol (Inst Pasteur) 134 A: 65-71.
• Schuchat A, Swaminathan B, Broome CV 1991. Epidemiology of human listeriosis. Clin Microbiol Rev 41: 169-183.
• Seeliger HPR, Höhne K 1979. Serotyping of Listeria monocytogenes and related species. ln T Bergan, JR Norris (eds), Methods in Microbiology, Academic Press, London, p. 31-49.
• Weis J 1975. The incidence of L. monocytogenes on plants and in soil. ln M Woodbine, Problems of Listeriosis, Leicester University Press, Leicester, England, p. 61-65.

⁺

Corresponding author. Fax: + 55-21-270.6565

Publication Dates

• Publication in this collection
  10 Oct 2001
• Date of issue
  Oct 2000

History

• Accepted
  22 Mar 2000
• Received
  30 Aug 1999