Breeding causes more changes in plants than genetic engineering

"The impact of transgenes is basically limited to their immediate function"

It is often maintained that genetic interventions may have unintended consequences for the metabolism of modified plants and by implication for human health and the environment as well. A recently completed research project compared gene expression and plant substances in different conventional and transgenic barley lines. GMO Safety discussed the findings with Uwe Sonnewald, one of the project leaders.

Prof. Uwe Sonnewald from the University of Erlangen-Nuremberg.

Extract from a microarray used to analyse gene expression. Each dot corresponds to a single gene. In total, the activity of around 32,000 genes was investigated.

Most significant research results

  • Research material: transgenic barley plants of the "Golden Promise" and "Baronesse" varieties and the conventional parent lines
  • Comparison of conventional and transgenic "Golden Promise": no significant differences
  • Comparison of conventional and transgenic "Baronesse": Differences in the expression of 22 genes and 4 metabolites. The researchers were able to demonstrate that these differences were due to a previous crossing with a different variety.
  • Comparison of the two conventional parent lines: Differences in the expression of around 1,600 genes

GMO Safety: What were the aims of your research project?

Uwe Sonnewald: In our joint project with the University of Giessen and the State University Washington we aimed to test the extent to which different varieties of barley differ from one another, the extent to which transgenic and non-transgenic barley plants differ from one another and the impact of environmental factors. To this end we studied gene expression in the plants and the composition of the metabolites, i.e. the metabolic products. We originally intended to study the leaves and the grains but as it turned out we were unable to complete the grain studies due to repeated crop vandalism in Giessen.

GMO Safety: What are your most important findings?

Uwe Sonnewald: Firstly, we found virtually no differences between the genetically modified barley plants under investigation and their non-transgenic parent lines, either in terms of metabolites or gene expression. We then found that colonisation of the barley plants by mycorrhizal fungi produced virtually no changes in gene expression, but did change the metabolites. This demonstrates the value of combining both approaches. We also discovered that differences between conventional varieties can be considerable. Approximately 1,600 genes in the two conventional varieties that we compared are differentially regulated. We don’t even know the function of many of these genes.

GMO Safety: The following statement was deduced from your findings: Conventional breeding causes more changes in plants than the introduction of a single transgene. Can you make such a generalisation? After all, you only looked at barley. Have comparable studies been carried out on other genetically modified crops?

Uwe Sonnewald: As far as I know, this was the first time that both methods had been used in a simultaneous investigation. Researchers have studied either gene expression or plant substances in wheat, potatoes and maize and have come to very similar conclusions. The impact of transgenes is basically limited to their immediate function. For example, if I insert a gene for fructan biosynthesis in potatoes, it is hardly surprising that these potatoes then produce fructan and so differ in this way from their parent lines. But only negligible additional differences were found. I know of no instance where a more significant change in gene expression has been caused by a single transgene. However, great variability exists between individual varieties of all the crops mentioned and the obvious explanation for this is that often the breeding objective is to create resistance to external stress factors, and this involves a large number of genes.

GMO Safety: Do you consider that it would be useful to use the profiling methods that you have just applied to assess the safety of genetically modified crops in the future?

Uwe Sonnewald: First you would have to perform a kind of cataloguing, in other words you would have to determine the gene expression and plant substances of different existing varieties of each crop type. Only on this basis can you then decide whether a change in a new plant gives any cause for concern. This type of cataloguing is currently available only for metabolites, and then only for the main substances, which is too crude.

Metabolic profiling could be useful in principle if a genetic modification had been carried out with the aim of modifying specific metabolic processes. Since different metabolic pathways are interconnected, it is certainly useful to establish whether anything other than the intended modification has changed in other metabolic pathways - for example, whether a plant produces too few vitamins because it produces increased amounts of a specific substance.

Studying gene expression is even more problematic. Important differences emerge even from a comparison of conventional varieties and in most cases I am unable to say what these changes signify because I don’t even know the function of the genes. Therefore it would not be useful to make this type of investigation a prerequisite for approving a plant.

GMO Safety: Thank you for talking to us.