Plant transformation with Agrobacterium

Using a soil bacterium as a gene transporter

For twenty years now people have been using Agrobacteria to transport new genes into plant cells. Although the procedure is now routine, there is still room for further improvement.

Agrobacterium-mediated transformation is one of the most commonly used methods for transporting new genes into plant cells and for ensuring their stable integration into the genome. It is now used for many types of crop.

Agrobacteria: ‘Natural’ genetic engineering

The Agrobacterium tumefaciens used for this method is a soil bacterium which occurs in cultivated an uncultivated soils and can cause tumours at plant wound sites (crown galls). The cause of this is the natural ability of the Agrobacteria to introduce the genetic information for the crown gall (tumour) formation into the plant genome. Instead of being situated on the Agrobacteria’s chromosome, this T-DNA (=transferred DNA) is found on a plasmid (Ti plasmid = tumour-inducing plasmid), flanked by two border areas, the right and left border (RB and LB).

Agrobacteria manage to modify plants genetically by ‘natural’ means – and, astonishingly, they succeed even if the original T-DNA containing the information for tumour formation is removed from the plasmid and replaced with foreign DNA. In this way, the Agrobacterium can be used as a ‘transport vehicle’ to introduce new genes. Since an indicator is needed so that transformed cells can be identified, in the past a marker gene has often been placed on the Ti plasmid alongside the target gene.

Successfully transformed plant cells can be regenerated into new plants.

Binary vectors: two from one

To prevent the T-DNA introduced into the Ti plasmid from spreading uncontrollably, binary vectors were developed.

Normally, the Ti plasmid also contains ‘virulence genes’ (vir genes), which are necessary for the transfer of the T-DNA. In the case of the binary vectors, the vir genes are removed and the desired DNA can be integrated between the borders. These ‘disarmed’ binary plasmids have several advantages: they are easy to propagate in laboratory bacteria like e.g. E. coli. A binary plasmid is also much smaller than a normal Ti plasmid, so technically easier to handle.

The vir genes needed for transfer to the plant are arranged on a second plasmid, from which the T-DNA and both borders are removed. With a binary vector, the target gene (T-DNA) and virulence gene, which are naturally joined on one plasmid ring, are split between two plasmids. One transfers the target gene, while the other helps with the transformation through the activity of the vir genes, without the information for the transfer itself being integrated into the plant as well.

First step: Integration of foreign DNA (green) into a Ti plasmid without virulence genes (vir genes). This vector is propagated in E.coli bacteria.

Second step: The two plasmids are joined together in an Agrobacterium strain.

Third step: The T-DNA is transferred to plant cells. The Agrobacteria and plant leaf pieces are cultivated together, followed by regeneration. The untransformed cells are selected with the help of the marker gene.

Cotransformation – transferring two independent T-DNAs

Several biosafety research projects have focused on optimising binary vectors. One of the key aims was to develop vectors that are suitable for the simultaneous transfer of two T-DNAs. The objective was to enable the target gene and marker gene to be integrated independently of one another into the plant genome, and to be able to separate them again at a later point in time using segregation to obtain transgenic, marker-gene-free plants.

For cotransformation using Agrobacteria various arrangements of marker gene (red) and target gene (green) are possible. The helper Ti plasmids are not shown.

Diagram: Different ways of transferring two independent T-DNAs using cotransformation. The helper Ti plasmids are not shown.

  • Two different strains of Agrobacteria, each with a transformation vector; one vector contains the marker gene, the other the target gene (top diagram). This arrangement is particularly successful with barley.
  • One bacterial strain with two vectors, each with one gene (centre). This arrangement is the most promising for oilseed rape and tobacco.
  • One bacterial strain with one vector, on which the two genes are located at separate sites (bottom). This arrangement is particularly successful with e.g. Arabidopsis (Thale cress).

Further developments in the area of binary vectors have optimised various features:

  • Removal of unnecessary sequences: To be able to transfer two T-DNAs with one plasmid, the vector has to be made even smaller. It was possible to remove DNA sequences that do not have to be located on the plasmid, thereby reducing the size of the vector by around half (smaller than 5 kb).
  • "Leaky" left border. Transfer and integration of T-DNA into the plant genome start at the RB (right border) and end at the LB (left border). For a long time, the left border was considered ‘leaky’. Sequences outside the T-DNA – and sometimes the entire plasmid – were frequently transferred at the same time. Depending on the extent and type of the transferred sequences beyond the ‘leaky’ LB, the release or breeding of this kind of transgenic plant could be problematic, since not all transferred genes/DNA sequences are desirable. Vectors were developed which were designed to achieve more accurate integration of the T-DNA. For this, the termination sequence (‘stop signal’) used as the left border of the T-DNA was employed twice or even four times.