transfer from and survival of genetically modified lactic acid bacteria
Lactic acid bacteria (LAB) are widely used in the food industry for making
beer, wine, cheese and various other foods, and have been for hundreds,
or even thousands of years. As a consequence, they are the focus of intensive
investigations and one of the aims of this work is to modify their genetic
constitution so that they are better adapted to agro-food industrial purposes.
Prior to the use of any such modified strains, it is necessary to assess
the risks associated with the transfer of genes from modified LAB to other
organisms. This project aimed to improve our understanding and knowledge
about such transfer and its consequences.
Approach and methodology
In addition to standard laboratory conditions, three model systems were
used to investigate gene transfer between LAB. They were fermenters, cheese
and the mouse digestive tract. These three model systems correspond to
the three major stages through which industrial LAB pass: industrial production
of starter cultures in fermenters, cheese-making, and consumption (by
man). The transfer between strains of antibiotic-resistance genes carried
on non-transmissible plasmids, self-transmissible plasmids or the bacterial
chromosome was studied. Molecular genetics techniques were used to construct
and analyse bacterial strains, and microbiological techniques were used
to evaluate the stability and transfer of the marker genes in each of
the three model systems.
Main findings and outcome
Under optimal conditions in the laboratory, the frequency of transfer
from LAB of genes on self transmissible plasmids was below 10-2.
The transfer of genes carried on the chromosome was so infrequent as to
be undetectable. Similarly, the transfer of genes on non-transmissible
plasmids was undetectable. However, if the strain carried a self-transmissible
plasmid in addition to a non-transmissible plasmid, genes on the non-transmissible
plasmid could be transferred at a frequency of about 10-5.
Transfer in fermenters and in cheese was much less efficient and could
only be detected for genes on self-transmissible plasmids, for which the
frequency was less than 10-7. L. lactis,
a major LAB, did not establish itself durably in the intestine of axenic
mice. Nevertheless, a low level of transfer of self-transmissible high
copy number plasmids could be demonstrated from L. lactis to E.
faecalis (a normal constituent of the gut flora). However, the transconjugants
(created by the transfer) were rapidly eliminated from the digestive tract.
No transfer of low copy number plasmids or chromosomal markers was detected.
The frequency of gene transfer from L. lactis, which is a widely
used LAB, is either very low or so low as to be undetectable (below 10-11)
in the three model systems used to represent the various stages in the
life of a foodstuff produced with LAB. This suggests that it is unlikely
that there is a large risk associated with gene transfer from genetically
modified LAB used in the agro-food sector. However, more extensive work
is required to confirm this conclusion.
January 1989 - December 1990
Thiverval Grignon (FR)