5.9—Plant genes don't move to gut microbes

Viral genes don’t transfer to gut bacteria

See Genetic Roulette’s False Claims at Bottom of Page

Analysis of Peer-Reviewed Research:

The scenarios introduced in this section of Genetic Roulette involve a number of highly unlikely speculations about events postulated to occur one after another but which are not likely to be harmful if they happened. They are largely in the realm of fantasy. Our scientific resources are much better used in responding to more probable events that do real harm already, such as existing malnutrition, hunger, and constantly emerging real infectious diseases and pest that damage crops. One issue raised is worth serious attention—and fortunately research biologists have got on the job and investigated it. This is the idea that DNA versions of RNA viruses inserted in transgenic plants open up new avenues of virus recombination. The results of this research are reassuring– the silencing mechanisms of virus-protected plants will act to prevent hybrid viruses from proliferating. Readers could benefit from being told about the extensive scientific efforts that has been devoted to analyzing the safety of virus-protected plum trees. Recent US deregulation documents for plum pox virus resistant plums provide excellent reading on this (APHIS 2007).

The rest of this section is about genes for virus proteins moving from plants to bacteria. The starting point for this chain of worries is just plain wrong. As was discussed in the previous section, plant DNA from food crops doesn’t move to bacteria at any detectable rate. Given this, all the worries raised by Smith about virus gene transfer to bacteria are inconsequential. Jeffrey Smith also worries about transgenic crop stability, but doesn’t properly discuss the field trials that have been which verify that transgenic virus protection in plum trees is stable in practice over 10 years of practical experience.

A thorough risk analysis of this situation should also weigh up the damage caused by doing nothing about crop diseases. This involves careful thinking about the damage done to agriculture by delaying disease resistant crops that can help manage the many existing pests and diseases that weigh down on our ability to produce more food with steadily decreasing resources of land and water.

See also

5.8 Genes may transfer to bacteria in the mouth or throat from many other microbes.



1. Uptake of a complete gene from a food plant by a bacterium has never been demonstrated. Section 5.9 of Genetic Roulette is based on wooly thinking by Smith about movement of genes to bacteria. We have explained his misunderstanding of those experiments in the previous section. They actually confirmed that bacteria have mechanisms to exchange DNA with one another but do not take up plant DNA from food crops DNA at any measurable rates. Repeatedly experiments that attempt to detect uptake of plant DNA by bacteria have been unsuccessful (Gerhard and Smalla 1998, Nielsen and others 1997, Schlüter and others 1995) unless the bacterium is deliberately engineered to already have the gene being tested, which increases its ability to incorporate the new genetic material. Frank Gebhard and Kornelia Smalla have summarised the many failures to detect this gene movement, and discuss the reasons why is hasn’t happened (Gerhard and Smalla 1998).  They point out that bacteria take up only small fragments of DNA — parts of genes. Bacteria have elaborate mechanisms to cut up DNA entering their cell, and this DNA fragmentation makes it very improbable they will incorporate a large enough fragment of DNA to comprise a complete gene. This all says that transfer of plum virus protein genes to gut bacteria is highly unlikely. Long-term establishment in human guts of bacteria burdened with the production of extra proteins that provide them no advantages is also unlikely.

2. Over evolutionary time periods movement of plant genes into bacteria is very rare compared with frequent movement of genes between different bacterial species, and even movement of bacterial genes to animals and fungi. Biologists who study evolution are fascinated by the movement of genes across biological kingdoms and between unrelated species. This movement occurs at such low rates that it cannot be easily detected occurring over the short time periods of say, one generation of human life. But it does have something interesting to say about the relative speed at which plant genes move to bacteria compared to the rates of exchange of genes between different bacterial species. A recent scientific review (Keeling and Palmer 2008) about gene movement between different organisms — the horizontal gene transfer as it is called– comments about the surprising lack of plant (and animal) genes found to taken up by bacteria in the face of their overt promiscuity in accepting foreign genes from other microbes. The promiscuity of bacteria is widely recognized and a typical microbe has numerous genes that it has scavenged from other species, but very rarely during the course of evolution have bacteria kept genes from plants. We don’t understand fully why this is so, but it is clear evidence that overall, movement of genes between bacteria is far more frequent than movement of genes from plants to bacteria. All this evidence is kept neatly stacked away on the evolutionary archive of gene movement of the bacterial genome that can be decoded quite easily using modern scientific techniques. Thus it is clearly imprinted on the evolutionary record that gene movement from plants to bacteria is not very likely over the long time scale of the evolutionary record. Gene movement from bacteria to animals? Or even phagocytic protists to bacteria? Well that’s a very different story (Citizendium 2007 , Gladyshev and others 2008, Keeling and Palmer 2008).

3. The stability of gene silencing in virus resistant plum tree (C5 plum) has been carefully checked in field trials. Smith speculates that the C5 plum trees that have been deliberately selected to not make plum pox virus protein will revert to making virus protein. The mechanism of gene silencing (Hily and others 2005) that prevents plum pox virus proteins being produced in genetically engineered C5 plum tree tissues does indeed make it possible that the gene silencing might revert, and allow virus protein to be produced as the genes are not silenced by irreversible mutation. The scientists that developed these trees however, have subjected the plum pox virus resistant plum trees to lengthy field trials to verify their stability. Documents from the US agency APHIS accompanying the deregulation of this transgenic plant mention that “10 years of field tests of C5 trees, in multiple locations and environments, have shown the virus resistance to be stable.”  These trials are reported in the peer-reviewed scientific literature too (Hily and others 2004) and they included deliberate exposure of the trees to plum pox virus and comprehensive testing that the transgenes genes remain silent during the full term of the field trial. As in earlier greenhouse tests, the transgenic trees retained their virus resistance over the full time period of the trial. In contrast, 95% of sensitive non-transgenic trees were infected with virus after four years. The 2004 and earlier publications about field trials verifying stability of virus gene silencing in field trials are not mentioned in Genetic Roulette.

4. For plum pox virus resistant plums (C5 plums), the USDA Animal and Plant Health Inspection Service has assessed the likelihood new viruses emerging with novel or altered properties is very low. Jeffrey Smith quotes biologist Jonathan Latham’s opinion that virus resistant plum trees open up new ways for viruses to evolve. Other biologists contest this, but readers of Genetic Roulette are not told about that. The possibility that C5 would decrease the circulation of viruses among plum trees and thus decrease the possibilities of virus interaction (as they have with papaya ringspot virus resistance transgenic papaya in Hawaii) does not seem to have occurred to Smith.

In issuing a “Finding of no significant impact and decision notice” about plum trees resistant to plum pox virus (C5 plum), the United States Animal and Plant Health Inspection Service have reached a different conclusion to Jonathan Latham and Jeffrey Smith. APHIS has fully documented their assessment of virus interactions with the gene in C5 transgenic plum trees with extensive references to the scientific literature (APHIS 2007). They have concluded

“In assessing risks posed by viral interactions, APHIS considered the potential for recombination, heterologous encapsidation and synergy. Extensive scientific knowledge is available about plum pox virus, and other members of the potyvirus group, based upon research performed around the world. Analysis of all available scientific information suggests that the likelihood of development of new viruses, or viruses with novel/altered properties is very low to non-existent. The low likelihood of risk posed by viral interactions suggests the lack of a plant pest risk in C5 ‘HoneySweet’ plum.”

APHIS notes that it is common in nature to find plants that are infected with multiple viruses, and the presence of more than one virus in infected plants theoretically provides the opportunity for same virus recombination events that worry Smith. “However, based on the published literature and unpublished data collected by researchers, natural development of functional the continent viruses producing new and different diseases are not common.” according to APHIS. In other words, the possibilities of new virus combinations are not novel to transgenic virus protected plum trees, and they do not appear to represent an appreciable hazard to disease experts. After citing extensive scientific discussion of the risks that new viruses might be generated from C5 plum (none of which are mentioned in Genetic Roulette) APHIS finishes with “therefore, based on available scientific information, it appears very unlikely that widespread deployment of the C5 plum trees will increase the likelihood for recombination occurred between the plum pox virus coat protein and other viruses. It is even less likely that any resulting recombinant would cause new or different disease characteristics.” (APHIS 2007).

5. Conversion of an RNA virus into DNA copy is not novel to transgenic plants, and transgenic virus protection provides inbuilt protection against novel hybrid viruses. This section of Genetic Roulette mentions some interesting speculation by biologist Jonathan Latham who noted that protection against RNA viruses involves creation of a DNA copy of part of the RNA virus that is inserted in the plant genome. Latham suggests that this might allow a virus to recombine with a new spectrum of organisms. This is not novel to plants that are genetically engineered to be virus resistant. During virus evolution DNA copies of RNA viruses are often created and inserted in the host genomes, so although creation of a DNA copy this may open up different avenues for evolution, they are avenues that have been open for many viruses to exploit in the past (Davidson and Silva  2007,  Tanne and Sela  (2005). Experiments to evaluate whether this route of virus evolution is likely to give rise to successful new hybrids have been reported. The reassuring findings are that hybrids between dissimilar viruses will have a selective disadvantage (Chung and others 2007). Virus silencing mechanisms engineered into the transgenic plant will act against any hybrid viruses that are formed, preventing their proliferation.


APHIS (2007). Approval of USDA-ARS Request (04-264-01P) Seeking a Determination of Non-regulated Status for C5 Plum Resistant to Plum Pox Virus (FONSI and Decision Notice, Response to Comments, and Final Environmental Assessment) APHIS-2006-0084-1731.

Chung BN, Canto T, Palukaitis P (2007). Stability of recombinant plant viruses containing genes of unrelated plant viruses. Journal of General Virology 88:1347-1355 “These data indicate that such hybrid viruses, formed in resistant transgenic plants from a transgene and an unrelated virus, would be at a selective disadvantage, first by being targeted by the resistance mechanism and second by not being competitive with the parental virus.”

Citizendium (2007). Horizontal gene transfer encyclopedia article. en.citizendium.org/wiki/Horizontal_gene_transfer accessed Dec 30 2008

Davidson I and Silva RF (2007). Creation of diversity in the animal virus world by inter-species and intra-species recombinations: lessons learned from poultry viruses.Virus Genes. 2008 Feb;36(1):1-9.

Gerhard F and Smalla K (1998). Transformation of Acinetobacter sp. strain BD413 by transgenic sugar beet DNA. Applied and Environmental Microbiology 64, 1550–4. Gene fragments from plants taken up to repair a damaged gene in genetically manipulated bacteria.

Gladyshev EA, Meselson M and Arkhipova IR (2008). Massive horizontal gene transfer in bdelloid rotifers. Science 320:1210-1213. Movement of genes across kingdoms in the ocean. You are what you eat.

Hily J-M and others (2004). Stability of gene silencing-based resistance to Plum pox virus in transgenic plum (Prunus domestica L.) under field conditions. Transgenic Research 13(5): 427.

Hily JM, Scorza R, Malinowski T, Zawadzka B and Ravelonandro M (2005). Accumulation of the long class of siRNA is associated with resistance to Plum pox virus in a transgenic woody perennial plum tree. Molecular Plant-Microbe Interactions: MPMI 18(8): 794

Keeling PJ, and Palmer JD (2008). Horizontal gene transfer in eukaryotic evolution. Nature Reviews Genetics 9:605-618. The state of play on gene movement between different species showing that gene movement–for example between different plants that are widely unrelated to one another –has happened many times in during evolution of different organisms.

Nielsen KM, Gebhard F, Smalla K,Bones AM van Elsas JD (1997). Evaluation of possible horizontal gene transfer from transgenic plants to the soil bacterium Acinetobacter calcoaceticus BD413. Theor. Appl. Genet. 95:815–821. The bacterium does not take up non-homologous plant DNA at appreciable frequencies.

Schlüter K, Fütterer J and Potrykus I (1995). ‘Horizontal’ gene transfer from a transgenic potato line to a bacterial pathogen (Erwinia chrysanthemi) occurs—if at all—at an extremely low frequency. Biotechnology 13:1094–8.

Tanne E and Sela I (2005). Occurrence of a DNA sequence of a non-retro RNA virus in a host plant genome and its expression: evidence for recombination between viral and host RNAs. Virology.332(2):614-22.

Genetic Roulette Falsely Claims:
Transfer of viral genes into gut microorganisms may create toxins and weaken viral defenses.

1. As discussed earlier, proteins produced from viruses can be toxic and disable viral defenses

2. If viral genes from GM crops transferred to gut microorganisms, they might produce large quantities of potentially harmful proteins

3. Viral transgene characteristics may transfer to gut microorganisms much more likely

Section 5.9 of Genetic Roulette is an extension of 3.9, and continues a discussion of how disease resistant crops may be risky. It discusses recombination of viruses, and speculates about how plum pox virus resistant plum trees that produce no viral proteins may accidentally change when they are grown in orchards to produce virus proteins in plums.