Nature Biotechnology 22, 1349 - 1350 (2004)

Reply to 'Monitoring horizontal gene transfer'

Jack A. Heinemann and Terje Traavik reply:

Our article points out that the relevance of transgene transfer to biosafety has been dismissed, as if its irrelevance were experimental fact. Davison misses the point by confusing the example of antibiotic-resistance gene transfer from transgenic plants to bacteria with our illustration of resistance evolution by gene transfer. We showed that based on experimental observations of resistance evolution, sound biosafety experiments have been impossible. Either the biosafety claims are experimentally verified or they are based on intuition and educated guesses. The standards of the science behind claims of safety should not be adjusted in the name of transgenic organisms simply because the claim is too difficult to verify.

Antibiotic-resistance evolution arose because of undetectable frequencies of horizontal gene transfer (HGT) that grew into discernable threats. This exemplifies the general uncertainty over how the many different transgenes, including those for antibiotic resistance, will affect the characteristics of organisms that subsequently receive them. Resistance genes and antibiotics were not rare before human use of antibiotics1, as Davison says, but they were effectively nonexistent among the organisms that cause disease in people. A change in human practices influenced the transition of these genes into “horizontal transfer machines,” leading ultimately to their introgression into new species. The lesson here is that different applications of genetic engineering may create new opportunities in evolution, some of which are worth anticipating before we invite them.

For example, the US National Research Council (Washington, DC, USA) recently considered 'suicide genes' for containing recombinant microorganisms2. We predict that such genes easily become horizontal transfer machines3, 4, making their use as a containment tool a potential mistake.


Unless we have a basis for predicting when genes make the transition to horizontal reproduction5, we cannot know a priori that particular transgenes will be uncompetitive, contrary to what Davison suggests. We similarly cannot assume that transgenes are equivalent in their abilities to make this transition, particularly as they are neither the same DNA sequence nor in the same context as their natural counterparts. They are also not released into the same world that their natural counterparts evolved in, because they are maintained through human-assisted breeding programs and very often reach novel concentrations in the environment through the use of co-technologies (e.g., herbicides) that may have unanticipated effects on the evolution of these genes by HGT.

Contrary both to Davison and to Nielsen and Townsend, we do not assume that the only biosafety-relevant outcome of HGT is introgression. We also question the latters' assertion that an “overwhelming majority of HGT events in nature are known to be deleterious to the bacterial transformant” (our emphasis) on the basis of only two laboratory studies using Escherichia coli. First, there is no evidence that the majority of HGTs result in genomic insertion. Second, not all biosafety risks are from the growth and spread of organismal recombinants and transgene introgression into organismal lineages. Some horizontal transfer machines pose relevant risk without having introgressed into a genome, as viruses and plasmids and some transposable elements demonstrate6.



  1. Baquero, F. & Blazquez, J. Trends Ecol. Evol. 12, 482–487 (1997). | Article | ISI |
  2. Committee on the Biological Confinement of Genetically Engineered Organisms, NRC. Biological Confinement of Genetically Engineered Organisms. (National Academies Press, Washington, DC, 2004).
  3. Cooper, T.F. & Heinemann, J.A. Proc. Natl. Acad. Sci. USA 97, 12543–12648 (2000).
  4. Cooper, T.F. & Heinemann, J.A. Selection for plasmid post-segregational killing depends on multiple infection: evidence for the selection of more virulent parasites through parasite-level competition. Proc. R. Soc. Lond. B Biol. Sci., in press.
  5. Heinemann, J.A. & Silby, M.W. in Multiple Drug Resistant Bacteria. (ed. Amábile-Cuevas, C.F.) 161–178 (Horizon Scientific Press, Wymondham, 2003).
  6. Syvanen, M. & Kado, C.I. (eds.) Horizontal Gene Transfer, edn. 2 (Academic Press, San Diego, 2002).


These links to content published by Nature Research are automatically generated.

Extra navigation