Science in Society Archive

Dangerous Field Test of Non-pathogenic GM Bacteria

The non-pathogenic GM bacteria not only carry antibiotic resistance marker genes, but the proposed field tests will also involve the release of the wild-type bacterium that is pathogenic to rice and may cause disease in human beings. Prof. Joe Cummins and Dr. Mae-Wan Ho

This report has been submitted to the USDA on behalf of ISIS, please circulate widely.

Bacterium causing rice panicle blight

The United States Department of Agriculture Animal and Plant Health Inspection Service (USDA/APHIS) carried out an Environmental Assessment  [1] in response to a permit application (06-111-01r) received from Dr. Milton Rush of Louisiana State University for a field test of two non-pathogenic, genetically engineered strains of Burkholderia glumae, and is available for public comment by 19 July 2007 at:

Burkholderia glumae Kurita et Tabei is a bacterial plant pathogen that causes bacterial panicle blight in rice, and is transmitted by infected seed. This bacterium was first described in Japan as the cause of grain rotting and seedling blight and is considered one of the most important rice pathogens in Japan. Epidemics of panicle blight occurred in the southern rice producing area of the United States during the 1995 and 1998 growing seasons, with yield losses in some fields estimated to be as high as 40 percent. Currently, there is no control method for panicle blight in the US, where most commercially grown rice varieties are susceptible to the disease. Field-testing non-pathogenic, transgenic strains of B. glumae is supposed to provide information on bacterial panicle blight infection of rice, and indicate potential routes for control of the pathogen.

Non-pathogenic transgenic bacteria contain two antibiotic resistance markers

B. glumae has been modified by disrupting the disease-causing gene, resulting in avirulent or non-pathogenic transgenic strains.

One virulence factor in B. glumae is the compound  toxoflavin , a yellowish substance that results in significant damage to rice in the infected plants. Toxoflavin is produced in Burkholderia by an operon (group of genes with a defined function) consisting of the tox gene cluster (toxABCDE) controlled by the toxR gene that is activated when the bacterium invades the rice plant. Disruption of the toxA gene (methyltransferase) results in mutants that do not produce toxoflavin. The cloning vector also contains two selectable markers, the gene (nptII) for neomycin phosphotransferase from Streptomyces kanamyceticus and the gene (bla) for beta-lactamase from Escherichia coli, providing resistance to kanamycin and ampicillin, respectively. The promoter for each of the genes is the Bacteriophage T7 promoter, and the terminator a synthetic TAA codon sequence. The donor DNA sequences are stably and irreversibly integrated into the bacterial genome, where they are maintained and inherited as any other genes of the bacteria cell [1].  The avirulent non-pathogenic strain therefore also carries stable resistance to the antibiotics kanamycin and ampicllin. The potential for horizontal gene transfer of the antibiotic resistance markers to soil bacteria is acknowledged in the USDA/APHIS assessment, but is presumed to have insignificant consequences. This presumption is not borne out by a wealth of evidence we have presented repeatedly to our regulators, the most recent in June 2007 [2] (GM Food Nightmare Unfolding in the Regulatory Sham , I-SIS scientific publication)

Pathogenic wild-type bacteria will be released in field-tests of non-pathogenic strains

Two experiments will be conducted; the first evaluates toxoflavin as a disease causing agent by challenging the rice plants with wild-type B. glumae, the second involves inoculating the rice with the transgenic avirulent bacterium followed by challenge with the virulent strain to see whether or not the presence of the avirulent strain will protect rice from B. glumae infection. These are obviously dangerous experiments to be carried out in the open fields; as the wild-type pathogen could easily spread from the experimental fields to other rice crops. The risks are unjustifiable, especially when there are other safer strategies.

An alternative approach to controlling B. glumae is via ‘quorum sensing’, a regulatory network influencing virulence based on the local density of bacteria that intercommunicate with one another. Quorum sensing can occur within a single bacterial species as well as between disparate species, and can regulate a host of different processes, essentially serving as a simple communication network. The bacteria signal to one another via special molecules. For example, toxoflavin is regulated by a quorum sensing mechanism that uses N-acyl homoserine lactones as signal molecules. A Burkholderia endophyte (a bacterium that lives inside the plant) was selected from rice and found to be non-pathogenic to rice and to inhibit pathogenic fungi. The endophyte, modified with a gene from Bacillus thuringiensis specifying N-acyl homoserine lactones, was found to prevent toxoflavin synthesis and virulence of B. glumae [3]. Genetic modification involving quorum sensing provides an alternative, also avoids use of the antibiotic résistance genes described above  because  toxoflavin can be detected by its fluorescence and its absence is readily detected.

Another danger from the transgenic B. glumae proposed for release is that the genus contains serious pathogens for humans: B. cepacia is a potent pathogen [4] (Bio-remediation Without Caution, SiS 230; B. thailandensis  caused pneumonia and septicemia [5]; B. dolosa  is pathogenic for people with cystic fibrosis [6]; B. gladioli  caused ocular keratitis in an individual with diabetes, and is also found in other diseases [7]; and a number of other Burkholderia species are associated with human infections.  It is not at all surprising, therefore, that a B. glumae infection was observed in an infant with chronic granulomatous disease [8]. Further investigation of the clinically isolated strain of B. glumae showed that the bacterium caused severe disease symptoms in rice, and a quorum sensing regulated secreted lipase was implicated in the pathogenesis of the clinical strain [9].

The USDA/APHIS assessment did not consider human infection by B. glumae a serious matter [1] based on the single human case, nor did it recommend precaution for those working with the pathogen, who will most likely take the pathogen to their homes, families and neighbours. The dangers of the transgenic B. glumae itself as a potential pathogen armed with two antibiotic resistance marker genes that could further transfer horizontally to other known Burkholderia pathogens appear to have completely escaped the notice of USDA/APHIS. Both USDA/APHIS and the scientists involved should be held responsible for any harm caused to people and crops, should they allow this field release to go ahead.

Article first published 16/07/07


  1. U.S. Department of Agriculture Animal and Plant Health Inspection Service Biotechnology Regulatory Services USDA APHIS Environmental Assessment In response to a permit application (06-111-01r) received from Dr. Milton Rush of Louisiana State University for a field test of two non-pathogenic, genetically engineered strains of Burkholderia glumae.
  2. Ho MW, Cummins J and Saunders P.  GM food nightmare unfolding in the regulatory sham. Microbial Ecology in Health and Disease 2007,19, 2, 66 – 77.
  3. Cho HS, Park SY, Ryu CM, Kim JF, Kim JG and Park SH. Interference of quorum sensing and virulence of the rice pathogen Burkholderia glumae by an engineered endophytic bacterium FEMS Microbiol Ecol. 2007 Apr;60(1):14-23.
  4. Cummins J and Ho MW. Bio-remediation without caution Science in Society 23, 40, 2004
  5. Glass MB, Gee JE, Steigerwalt AG, Cavuoti D, Barton T, Hardy RD, Godoy D, Spratt BG, Clark TA and Wilkins PP. Pneumonia and septicemia caused by Burkholderia thailandensis in the United States. J Clin Microbiol. 2006, 44(12),:4601-4.
  6. Caraher E, Duff C, Mullen T, Mc Keon S, Murphy P, Callaghan M and McClean S.
  7. Invasion and biofilm formation of Burkholderia dolosa is comparable with Burkholderia cenocepacia and Burkholderia multivorans. J Cyst Fibros. 2007, 6(1):49-56.
  8. Ritterband D, Shah M, Cohen K, Lawrence J and Seedor J. Burkholderia gladioli keratitis associated with consecutive recurrent endophthalmitis. Cornea 2002, 21(6), 602-3.
  9. Weinberg JB, Alexander BD, Majure JM, Williams LW, Kim JY, Vandamme P and LiPuma JJ. Burkholderia glumae infection in an infant with chronic granulomatous disease. J Clin Microbiol. 2007, 45(2), 662-5.
  10. Devescovi G, Bigirimana J, Degrassi G, Cabrio L, Lipuma JJ, Kim J, Hwang I and  Venturi V. A clinical isolate of Burkholderia glumae causes severe disease symptoms in rice; involvement of a quorum sensing regulated secreted lipase. Appl Environ Microbiol 2007 Jun 8; [Epub ahead of print] doi:10.1128/AEM.00105-07

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 tripod?