Summary

Bacillus thuringiensis og fødevareforgiftninger

The bacterium Bacillus thuringiensis is the active microorganism in a number of plant protection products, which are used for the control of insects. It is the active ingredient in about half of the microbial plant protection products which are used in Denmark and constitutes about 90% of the use of microbial plant protection products on the world market.

B. thuringiensis is known as an insectpathogenic organism with activity against a number of insect orders. During the sporulation the bacterium produces specific d-endotoxins, which are toxic for insect larvae. The specific toxins are not expected to be toxic for humans. Except for these insect toxic properties is it hard to distinguish B. thuringiensis from B. cereus. B. cereus is known as a food born pathogen for humans, who can cause diarrhoea and emesis. Diarrhoea is caused due to a number of different enterotoxins, hemolysins and enzymes produced by B. cereus during vegetative growth in the intestine. The genes coding for these products are known, and therefore it has been possible to examine, whether strains of B. thuringiensis possess these genes by PCR-based techniques and other kinds of molecular techniques. These techniques have shown that a number of B. thuringiensis strains have characteristics which make them potential food borne pathogens. It has also been shown that these strains at optimal growth conditions can express these characteristics in the laboratory. Among these B. thuringiensis strains, which possess these potential pathogenetic properties, are strains, which are the active ingredient in microbial pest control agents. However, B. thuringiensis strains seem only to have been involved in disease in a very few cases.

However, neither the practice of physicians nor the methods used for detection of food borne pathogens distinguish between B. cereus and B. thuringiensis as the causative agent. It is therefore unknown whether and to what extent B. thuringiensis has been involved in food borne pathogenesis, which has been ascribed to B. cereus.

The objectives of this project are to:

To develop and use methods for the quantification of B. cereus and B. thuringiensis in faeces from humans, with special attention to B. thuringiensis strains used in microbial pest control agents.
To estimate the occurrence of B. cereus and B. thuringiensis in faces from healthy individuals, including the possible occurrence of B. thuringiensis strains from microbial pest control agents
Further is should be evaluated whether it is possible to identify at least 100 samples of faeces with a high number of B. cereus group bacteria from people with gastroenteritis among a realistic number of samples. The basis for the 100 samples is that the occurrence of B. thuringiensis in environmental samples is low, only a few percent of the total number of B. cereus group bacteria. Thus it was evaluated that 100 samples was a necessity to identify samples containing B. thuringiensis in high numbers.

For that purpose a standard collection of faeces from 420 healthy people was investigated. Bacteria from the B. cereus group were found in 4.8% of the samples. Among one fifth of the persons being positive for B. cereus group bacteria, one or more of the isolates could be identified as B. thuringiensis (0.95% of the persons). One half of these B. thuringiensis isolates has bipyramidal crystals and possesses the genes for cry1 and cry2, as strains used as microbial plant protection products. By RAPD-PCR analysis it was shown that four types of these bacteria was isolated. In one person the isolates were identical with B. thuringiensis kurstaki HD1, a common strain in plant protection products, based on the RAPD profile, but as the strain do not possess a flagellin gene, which can be detected by the used PCR-method, these strains are not identical with B. thuringiensis kurstaki HD1. However, it is likely that these isolates are identical with a strain from the serotype B. thuringiensis aizawai. In Denmark there are no products on the marked based on this serotype, but products are in use in other countries.

Conclusions:

The methods used were found useful for the detection of B. cereus group bacteria in faeces, to differentiate between B. cereus and B. thuringiensis and for the characterization of B. thuringiensis at the level of strains.
B. cereus group bacteria were found in 4.8% of the faeces samples from the persons in the Danish standard collection.
B. thuringiensis was identified in 0.95% of the samples.
No isolates could be connected unambiguously to B. thuringiensis kurstaki HD1, thus no isolates could be referred to a microbial plant protection product with certainty.
On the basis of the relatively low numbers of samples used in this investigation, seems neither B. cereus nor B. thuringiensis to occur in faeces from healthy Danes with a high frequency, although most Danes are exposed for these two species of bacteria during the consumption of food, notable vegetables.

B. cereus group bacteria were found in 10.2% of the samples from patients with gastroenteritis. In four samples the number exceeded 1000 cfu/g, in all other samples the number was below this level.

Isolates, which could be identified as B. thuringiensis, were found in 2.3% of the samples. This occurrence was approximate twice the occurrence in the samples from healthy people, however the results are not direct comparable, so this might be a difference by chance.

All the B. thuringiensis isolates have bipyramidal crystals and 12 isolates possess genes for cry1, cry2 and flagellin. Four of these isolates from four patients have a RAPD-profile, which are identical with the profile of B. thuringiensis kurstaki HD1, thus it is very likely, that these isolates derive from a microbial pest control agent. The other B. thuringiensis isolates have also several characteristics in common with strains used as microbial pest control agents, but they cannot be connected to a specific strain in use as microbial pest control agent.

B. thuringiensis was identified in faeces from eight persons with gastroenteritis, it is not possible to determine whether B. thuringiensis has been involved in the disease, but we found it not very likely, as the bacteria was found in low concentrations as spores.

It is most likely, that these persons have been eating food contaminated with B. thuringiensis as spores. These spores might have been sprayed on these food items as microbial plant protection products. Isolates identical with B. thuringiensis kurstaki HD1 have in earlier studies been identified on e.g. tomatoes, cucumbers and cabbage.

Conclusions:

B. cereus group bacteria were found in 10.2% of the faeces samples from persons suffering from gastroenteritis.
B. thuringiensis was found in 2.3% of the samples from persons suffering from gastroenteritis.
Four isolates from four patients was identical with the strain B. thuringiensis kurstaki HD1 and it is therefore very likely that these isolates derive from a microbial pest control agent.
It seems not to be possible with a manageable number of samples to identify 100 faeces samples containing a high number of B. cereus group bacteria. The demand will be around 10 000 samples.

Opinion from the Danish Environmental Protection Agency

Based on the relatively few samples included in this study, it can be concluded, that it is not likely that B. cereus or B. thuringiensis occurs in high number in the faeces of healthy Danish population. The Danish EPA believes that based on these results it should be anticipated that B. thuringiensis only seldom is found in faeces from patients with gastroenteritis, and that the organisms in these few cases hardly have been the cause of the disease. However, since the faeces samples probably have been provided to the doctors several days after the symptoms have started, it can still not be ruled out that B. thuringiensis has been the cause of gastroenteritis in a few patients.

The Danish EPA assess, that based on the present study, there is not reason to assume that the use of microbial plant protection products based on B. thuringiensis has caused gastroenteritis in the Danish population.