A group of researchers from Pioneer Hi-bred International recently produced lines of maize with resistance to the herbicide Lightning (a mixture of imazethapyr and imazapyr)1. On the surface, this is an unremarkable event. After all, resistance to these and similar herbicides has previously been produced in maize and other crops using either conventional breeding or genetic engineering strategies. However, what is unusual about the new maize lines is not the fact that they are resistant to these herbicides, but rather the method that was used to confer the resistance: a precise replacement of a single amino acid in the endogenous herbicide target enzyme.

The herbicides imazethapyr and imazapyr are members of the imidazolinone herbicide family, which, along with the sulfonylureas (and a few newer herbicide families), act on plants through the inhibition of the enzyme acetohydroxy-acid synthase (AHAS). AHAS is a key enzyme in the synthesis of the branched chain amino acids_valine, leucine, and isoleucine; so disruption of the activity of this enzyme leads to metabolic disruption and death of plants. However, a very slight alteration in the amino acid sequence of the AHAS enzyme can prevent herbicidal inhibition while preserving its normal catalytic function, a feature that has been exploited by scientists to create crops resistant to AHAS-inhibiting herbicides.

The Conventional Method For Engineering Resistance

The conventional method of generating herbicide resistant plants involves isolating a gene of interest (encoding a protein that confers herbicide resistance) and manipulating it in a bacterial-derived plasmid vector. The coding sequence of the gene may have its normal endogenous promoter, or be fused to another regulatory element that will cause it to have a new pattern of expression. This gene is then linked to a selective marker, for example, a gene that encodes antibiotic resistance, which will aid in the selection of cells transformed with the new gene. This entire construct is then introduced into plant cells resulting in its incorporation into the plant genome. Cells that have been transformed with the construct are then selected by their ability to grow in the presence of the antibiotic and surviving plants are regenerated.

This conventional method has two problematic results. First, the transgenic crop plant still contains the antibiotic resistance marker, left over as an artifact of the transformation/regeneration process, which has led some people to worry about the consequences of having antibiotic resistance genes widely expressed in transgenic crops. The second concern is that, because integration of foreign genes into the genome is generally random, the number and location of the transgene insertions into the crop genome cannot be controlled. Transgene insertion can disrupt the function of genes into which they might insert, and the expression of the transgene itself can be influenced by its location in the genome. Some areas of the genome appear to be more actively transcribed than others, and multiple lines transformed with an identical construct can vary widely in their levels of transgene expression. Such issues of transgene copy number, insertion location, and gene stability and expression make the job tougher for plant breeders who are charged with efficiently incorporating the new trait into the latest crop varieties.

An Alternative Approach

Enter the Pioneer group, led by Chris Baszczynski, who generated herbicide resistant maize lines that avoid these potential problems by specifically changing a single amino acid in the AHAS gene using a chimeric RNA/DNA oligonucleotide. This technique has been used in mammalian systems as a tool for gene therapy, and is now shown to have value in plants. The approach uses an oligonucleotide made up of a combination of DNA and RNA bases, with a 32-base section having nearly exact homology to the target sequence of the endogenous plant gene, except that there is a single base mismatch at the point of the desired mutation. The chimeric oligonucleotide is delivered into target cells using microprojectile bombardment, where it aligns with the endogenous homologous sequence. In certain cases the normal DNA repair mechanism reads the chimeric oligonucleotide as the template gene and "corrects" the mismatched base in the endogenous gene. The result is a direct change in a specific nucleotide in the target gene.

Making herbicide resistant maize by this method is no simple task, however, and requires a lot of information before even starting. First, it was important to identify the specific amino acid to change, which in maize was to change the serine (coded by AGT) at position 621 to asparagine (AAT). It was also known that maize contains two families of the AHAS gene, with multiple members in each family, so extensive sequencing of various AHAS genes from maize was conducted in order to verify that the region around the target amino acid was not polymorphic. This allowed the researchers to design a chimeric oligonucleotide that would be homologous to all AHAS genes. It was also important to know if genes in addition to those for AHAS had sufficient homology to this sequence that there would be a chance of introducing an unintended mutation in other genes. Extensive database searching of known maize expressed sequences revealed no sequences that matched the oligonucleotides as well as the AHAS genes.

Cultured maize cells were bombarded with the chimeric oligonucleotides. Previous experiments2 using oligonucleotides tagged with a fluorescent dye had demonstrated a preferential accumulation of the oligonucleotides in cell nuclei within one hour of bombardment, but the oligonucleotides were degraded rapidly and did not persist beyond 24 hours. Cells that were able to grow and develop callus in the presence of the herbicide imazethapyr were characterized to determine whether the desired mutation had occurred.

Plants were regenerated from nine separate transformation events and tested for susceptibility to Lightning. Of these, three plants were resistant to the herbicide at four times the normal field dose, four others were able to tolerate the normal field dose, and were only slightly injured by the four-fold rate, while two plants (and the untransformed control) were severely injured by the normal field dose of the herbicide. AHAS genes from these plants were sequenced and the highly resistant ones contained the predicted guanine to adenine change that would result in the amino acid conversion. Other regenerated plants had nucleotide conversions different from those predicted, but the mutations were all at, or within a few bases of, the target site. The herbicide susceptible plants did not have the predicted change. When resistant plants were back-crossed to wild-type plants, approximately half of the resulting progeny inherited the resistance, as would be predicted for a dominant trait.

Although the use of chimeric oligonucleotides for the engineering of plants has great potential, it also has some limitations. One problem is the requirement that the trait of interest must be conferred by the alteration of a single amino acid and produce a selectable phenotype to allow regeneration of putative transformants. This is relatively simple to do with herbicide resistance, but it may not be as easily applied to other traits. Another limitation is the need for extensive sequence information on the target gene and crop of interest, which may not be easily performed in less well-studied crops. Finally, the frequency of transformation (10-4) is lower than for conventional transformation events. Nevertheless, the elegance and directness of the technique results in a targeted change in a specific gene, with little or no confounding alterations in the expression of the gene. This technique may become a powerful tool in crop improvement and could allow investigations into the effects of subtle changes in single plant genes.

Sources

1. Zhu T, Mettenburg K, Peterson DJ, Tagliani L, and Baszczynski CL. 2000. Engineering herbicide-resistant maize using chimeric RNA/DNA oligonucleotides. Nature Biotechnology 18:555-558.

2. Zhu T, Peterson DJ, Tagliani L, St. Clair G, Baszczynski CL, and Bowen B. 1999. Targeted manipulation of maize genes in vivo using chimeric RNA/DNA oligonucleotides. Proceedings of the National Academy of Science 96:8768-8773.