B14007: Food safety implications of potentially pathogenic clostridia

Study Duration: December 2002 to February 2006

Contractor: University of Reading

The microbiological safety of food depends on ensuring that pathogenic microbes are eliminated from food or prevented from growing. For most of the well-known microbial pathogens the conditions necessary to achieve this have been worked out in considerable detail over many years. However from time to time new hazards emerge and information about the properties of the new organisms becomes necessary.

In recent years certain strains of Clostridium butyricum and Clostridium barati have been isolated that produce neurotoxins very similar to those produced by Clostridium botulinum types E and F respectively. Initially these organisms were associated with infant botulism but C. butyricum has also been implicated in foodborne illness. A different example of a possible new hazard is illustrated by a product recall based on high numbers of Clostridium tertium and Clostridium bifermentans in pate. Though neither of these organisms is associated with foodborne illness the counts were sufficiently high to cause concern. With all three organisms there is a need for more information about their growth characteristics and properties in relation to food safety.

The aims of this project were:

• To establish the proportion of C. butyricum strains isolated from natural sources that carry the botulinum neurotoxin gene
• To determine whether C. tertium and C. bifermentans produce toxins
• To define the conditions of temperature, pH and solute concentration that will prevent growth of C. butyricum, C. tertium, C. bifermentans and C. barati.

The experimental plan was to develop and use a selective enrichment medium to isolate C. butyricum from food and environmental samples, and then to screen isolates for the presence of the type E toxin gene using the polymerase chain reaction (PCR). Testing for possible toxicity in C. tertium and C. bifermentans was based on screening culture filtrates for toxic effects against cultured Vero cells. Growth limiting pH values, solute concentrations (sodium chloride, sucrose) and temperatures for growth were determined by inoculating replicate tubes of growth medium with spore inocula of test organisms derived from a cocktail of non-toxigenic strains or toxigenic ones.

978 food samples were examined for Clostridium butyricum by enrichment in a minimal medium with lactate and acetate as a source of carbon and energy. Antibiotics were present in the medium to select for growth of C. butyricum. The foods examined were mainly fresh vegetables but also included milk, cream, yoghurt and pâté. 93 isolates were tested for the presence of the gene encoding type E botulinal toxin by the polymerase chain reac tion (PCR) - no toxin positive samples were detected. A nested PCR approach was tested as a means of distinguishing between C. butyricum and the physiologically similar Clostridium beijerinckii but the primers were found to be sufficiently specific for the unequivocal identification of C. butyricum.

When culture supernatants of Clostridium tertium and Clostridium bifermentans were tested against Vero cells three out of five C. tertium strains and five out of six C. bifermentans strains showed a very weak toxic effect but, in most cases, it could only be demonstrated by incubating Vero cells for 48h rather than 24h and was quickly lost on moderate dilution.

Growth limits for C. butyricum, C. barati, C. tertium and Clostridium bifermentans were determined in broth incubated anaerobically at 30°C for up to 42days. The minimum pH values permitting growth depended on the acidulant and strain. Contrary to expectations organic acids were more effective at inhibiting growth than hydrochloric acid (HCl). In general the toxigenic strains appear to be less tolerant of acid conditions than the non-toxigenic ones.

The lowest pH values at which growth of toxigenic and non-toxigenic strains of C. butyricum was observed in broth acidified with HCl were 4.1 and 4.2 respectively. The minimum water activities for growth of toxigenic and non-toxigenic strains of C. butyricum were 0.95 and 0.96 respectively. The lowest pH values for growth of C. bifermentans and C. tertium in broth with HCl as acidulant were 4.1 and 4.2 respectively whilst in the presence of organic acids the minimum pH for both species was between 4.4 and 4.9. The minimum water activity for growth of C. tertium, C. bifermentans and C. baratiwas 0.95 for all. The minimum growth temperatures of the toxigenic strains of C. butyricum (ca 10 -11°C) was somewhat higher than for non-toxigenic ones (7-8°C). The minimum growth temperature for the other clostridia was about 8°C.

Growth of C. butyricum, C. bifermentans, C tertium and C. barati was prevented by water activities that will prevent growth of proteolytic strains of C. botulinum. However these organisms can grow at pH values below those that prevent growth of C. botulinum. Control of C. butyricum in the food industry needs to allow for the greater pH tolerance of this species compared with proteolytic C. botulinum, C. tertium and C. bifermentans showed some toxic activity against Vero cells but the effect was so weak that it was concluded that growth of these organisms in food is unlikely to be a significant risk factor in food safety.

The final report is available from the Agency's Information Centre.
To obtain a copy, please contact the Enquiry Desk, Information Services, Food Standards Agency (tel: 020 7276 8181/8182 or email: infocentre@foodstandards.gsi.gov.uk)

Contact: For any enquiries concerning this research project, please contact the relevant Programme contact or email: science@foodstandards.gsi.gov.uk