Science in Society Archive

GM Probiotic Bacteria in Gene Therapy

Prof. Joe Cummins and Dr. Mae-Wan Ho repeat their call for a ban on GM probiotics (Ban GM probiotics) as the first clinical trial has been carried out

Natural probiotic bacteria promote health, but GM probiotics are downright dangerous

Probiotic bacteria are beneficial bacteria living in the human gut that are now widely used as food additives for their health-promoting effects. These bacteria have co-evolved with their human host over millions of years. Their contributions to health and to the development of the host’s immune system depend on an intricate web of bacteria-bacteria and bacteria-host relationships that if thrown out of balance will most likely result in disease. For that reason alone, probiotics should never be subjected to genetically modification, let alone genetic modification for use directly on human beings.

There are other reasons that make GM probiotics particularly hazardous. The human gut is an ideal environment for horizontal gene transfer and recombination, the main route to creating dangerous pathogens. And pathogens created from probiotic bacteria will be pre-adapted to invade and colonize the human gut. We have published a paper in a scientific journal expressing our concerns [1] (GM probiotics should be banned).

Despite our warning, the first clinical trial using GM probiotic bacteria as gene therapy vector has taken place [2]. And supporters have written to the journal [3] criticizing our paper [1] for “sensationalism” and “lack of common sense”, and insist on defining probiotics in such a way as not to exclude GM strains. We took issue with their assertions  [4] (Reply to GM microbes for human health), especially in their attempt to blur the distinction between GM and natural probiotic bacteria, which is misleading and dangerous.

Let’s look at the first clinical trial with GM probiotics and the research leading up to it and beyond more closely here.

GM probiotic gene therapy uses an old bag of tricks

This first probiotic gene therapy makes use of a bag of tricks – thymineless death - that was first discovered 50 years ago.

Thymineless death is a peculiar cell death spasm that occurs when cells are deprived of the DNA base thymine. It happens in all organisms from bacteria to humans, but the most detailed information is available from the study of mutant bacteria strains lacking the ability to make thymine, an essential building block for DNA. When deprived of thymine, therefore, the cells accumulate both single and double strand DNA breaks. The double strand DNA breaks lead to cell death unless promptly repaired [5]. Thymine-deprivation is believed to activate a genetic suicide module leading to DNA degradation and death [6]. During thymine starvation, the cells rapidly lose viability. But lysates of the cells are nevertheless capable of transforming (genetically modifying) recipient cells [7], a kind of sex after death.

Mutant thymine-minus bacteria have been prepared as gene therapy vectors for delivering human genes to patients. In order to make sure that the bacterial vector would not regain its ability to synthesize thymine, the human therapy gene was inserted into the bacterial vector so as to disrupt a gene for thymine synthesis. Gene disruption is achieved by adding short DNA segments of sequences from the thymine gene to both ends of the human gene, so as to direct the foreign gene to the thymine gene where homologous recombination can take place to insert the human gene into the thymine gene. This allows the human gene to be expressed in the genetically modified bacterium in place of the thymine gene, and the disrupted thymine gene does not revert easily.

As human genes are not readily expressed in the bacterium because different DNA codons for the same amino acids tend to be used (codon bias), the disrupting gene inserted is a synthetic approximation of the human gene, with codons adjusted to suit the bacterium [8].

Mouse model inadequately investigated

Mouse colitis (gut inflammation) was treated using Lactococcus lactis modified with a mouse interleukin-10 (an anti-inflammatory cytokine) gene [9]. In that study the containment of the interleukin gene was not discussed except for a cursory mention. Pigs were treated with a synthetic interleukin-10 gene in a thymine-minus Lactococcus lactis. Both mixed bacterial cultures, or bacteria recovered from the pigs ileum were studied to determine whether or not there was mating to produce a thymine positive bacteria from the thymine-minus transgenic bacteria used to treat the pigs. There was no evidence that thymine positive strains were appearing due to reversion and loss of the interleukin gene.

However, the experiment was not designed to detect ‘partial diploids’ that carry a functioning thymine gene on a plasmid, which could then enable the bacterium carrying the interleukin gene disupting the thymine gene on its chromosome to escape cell death. Plasmid exchange between the transgenic and a plasmid-bearing strain was studied [10], but the plasmid did not appear to carry a thymine-plus gene which would have complemented the thymine-minus trait to produce a partial diploid positive for both thymine production and interleukin-10 production, and thus capable of growing in an environment lacking thymine.

Phase 1 clinical trial not adequately followed up

The phase 1 human trial using Lactococcu lactis expressing the synthetic human gene for interleukin-10 inserted into the thymine gene to treat 10 people for Crohn’s disease was carried out in Holland [2]. A reduction in disease activity was reported and the interleukin-10 producing bacteria recovered in stools were found to be dependent on thymine for growth.

But as every microbiologist knows, the proportion of gut bacteria that can be cultured is very small, certainly not greater than 10 percent, and little attempt was made to recover partial diploids, or to test whether the lysate of dead transgenic therapy bacteria could transform other gut bacteria. Nevertheless, the investigators concluded that containment of the transgenic bacteria was complete. 

The thymine-minus trait is gaining popularity in GM bacteria as a means of ‘containing’ the transgene. It has also been used to construct a live attenuated cholera vaccine. A thymine gene mutated in vitro was cloned, and then returned to Vibrio cholera to produce the non-proliferative strain as a vaccine candidate [11]. A thymine-minus strain of Streptococcus thermophilus (a bacterium used to produce yogurt and cheese) was constructed as a vector to deliver transgenes for food production; in this case, the thymine-minus gene was a spontaneous mutant [12].

Transgene containment using modified thymine-minus suicide strains is dependent on two important assumptions, both of which are invalid. The first is that mutational reversion is unlikely in the disrupted gene strain, though it is possible in strains carrying a conventional thymine-minus mutation. In the event of recombination, the transgene has to be spliced out for reversion to occur, so the transgenic bacterium is no longer transgenic. However, the strain that has recombined with the transgenic bacterium and gained the transgene may not be isolatable by current culture techniques. This could result in a false negative indicating that the transgene has not escaped, especially if the interleukin gene is carried on a plasmid in a partial diploid bacterium. The thymine-plus trait can also be introduced into the transgenic bacterium itself on a plasmid or a transducing bacteriophage, resulting in a partial diploid thereby preventing cell suicide. These possibilities have not been considered or discussed by those promoting the use of the thymine minus trait for bacterial containment.

There are numerous Lacctococcus plasmids, one in particular, a thymine-plus plasmid is used as a selectable marker in place of an antibiotic resistance marker [13,14]. The Lactococcus lactis bacteriophage sk1 efficiently carries plasmids and transfers them into cells [15]. Lactococcus contains numerous lysogenic bacteriophages many of which are also capable of carrying and tranferring genes into cells. 

Another assumption is that dead cells will not engage in gene exchange, or that transgenic DNA from dead cells will not transfer horizontally to other bacteria.  In fact, lysis of transgenic bacteria will release the synthetic interleukin-10 DNA in the bowels or faeces where the DNA may transform a range of bacterial species. For example, Lactobacillus may be transformed at a relatively high frequency in the natural environment [16]. Food commensal bacteria have been implicated in the horizontal transfer of antibiotic resistance, and such transfer may equally spread the synthetic interleukin-10 gene. Living, dying and dead bacteria may all be sources of gene transfer.

Thymine-minus bacteria are being promoted as bacterial vectors for human gene therapy. Unfortunately, the experiments reported so far seem to have ignored the avenues for the spread of transgenes from the bacteria to the natural environment via well-known processes of horizontal gene transfer and recombination. 

Proponents are now describing microbial gene therapy as “probiotic” treatment, and actually making use of genetically modified probiotic bacteria. Probiotic treatment has a long and honourable history of effective and ethical medical treatment while microbial gene therapy is an extremely risky business, especially when it uses genetically modified probiotic bacteria.

We reiterate our call for a ban on GM probiotic bacteria.

Article first published 03/07/06


  1. Cummins J and Ho MW. Genetically modified probiotics should be banned. Microbial Ecology in Health and Disease 2005, 17, 66-68.
  2. Braat H, Rottiers P, Hommes DW, Huyghebaert N, Remaut E, Remon JP, van Deventer SJ, Neirynck S, Peppelenbosch MP, Steidler L  A Phase I Trial With Transgenic Bacteria Expressing Interleukin-10 in Crohn’s Disease. Clin Gastroenterol Hepatol. 2006; May 19 [Epub ahead of print]
  3. Reid G, Gibson G, Gill H, Klaenhammer T, Rastall R, Rowland I and Sanders M. Use of genetically modified microbes for human health, letter to the editor. Microbial Ecology in Health and Disease 2006 (in press).
  4. Cummins J and Ho MW. Reply to GM microbes for human health. Microbial Ecology in Health and Disease 2006 (in press).
  5. Ahmad SI, Kirk SH and Eisenstark A. Thymine metabolism and thymineless death in prokaryotes and eukaryotes. Annu Rev Microbiol. 1998, 52, 591-625.
  6. Sat B, Reches M and Engelberg-Kulka H. The Escherichia coli mazEF suicide module mediates thymineless death. J Bacteriol. 2003, 185(6),1803-7
  7. Bousque JL and  Sicard N. Size and transforming activity of deoxyribonucleic acid in Diplococcus pneumoniae during thymidine starvation. J Bacteriol. 1976, 128(2), 540-8.
  8. Steidler L, Rottiers P and Remaut E. Self containing Lactobacillus strain United States Patent Application  20050276788
  9. Steidler L, Hans W, Schotte L, Neirynck S, Obermeier F, Falk W, Fiers W and  Remaut E. Treatment of murine colitis by Lactococcus lactis secreting interleukin-10. Science 2000, 289, 1352-5.
  10. Steidler L, Neirynck S, Huyghebaert N, Snoeck V, Vermeire A, Goddeeris B, Cox E, Remon JP and Remaut E. Biological containment of genetically modified Lactococcus lactis for intestinal delivery of human interleukin 10. Nat Biotechnol. 2003, 21(7), 785-9.
  11. Valle E, Ledon T, Cedre B, Campos J, Valmaseda T, Rodriguez B, Garcia L, Marrero K, Benitez J, Rodriguez S and  Fando R. Construction and characterization of a nonproliferative El Tor cholera vaccine candidate derived from strain 638. Infect Immun. 2000, 68(11), 6411-8.
  12. Sasaki Y, Ito Y and Sasaki T. ThyA as a selection marker in construction of food-grade host-vector and integration systems for Streptococcus thermophilus.  Appl Environ Microbiol. 2004, 70(3), 1858-64.
  13. Mills S, McAuliffe OE, Coffey A, Fitzgerald GF and Ross RP. Plasmids of lactococci - genetic accessories or genetic necessities? FEMS Microbiol Rev. 2006, 30(2), 243-73.
  14. Ross P, O'Gara F and Condon S. Thymidylate synthase gene from Lactococcus lactis as a genetic marker: an alternative to antibiotic resistance genes. Appl Environ Microbiol. 1990, 56(7), 2164-9.
  15. Chandry PS, Moore SC, Davidson BE and Hillier AJ. Transduction of concatemeric plasmids containing the cos site of Lactococcus lactis bacteriophage sk1. FEMS Microbiol Lett. 2002, 216(1), 85-90.
  16. Lorenz M and Wackernagel W. Bacterial gene transfer by natural genetic transformation in the environment. Microbial Reviews 1994, 58, 563-602.
  17. Wang HH, Manuzon M, Lehman M, Wan K, Luo H, Wittum TE, Yousef A and Bakaletz LO. Food commensal microbes as a potentially important avenue in transmitting antibiotic resistance genes. FEMS Microbiol Lett. 2006, 255(2), 328. 

Got something to say about this page? Comment

Comment on this article

Comments may be published. All comments are moderated. Name and email details are required.

Email address:
Your comments:
Anti spam question:
How many legs on a duck?