High-throughput Agrobacterium-mediated barley transformation

Abstract

Background: Plant transformation is an invaluable tool for basic plant research, as well as a useful technique for the direct improvement of commercial crops. Barley (Hordeum vulgare) is the fourth most abundant cereal crop in the world. It also provides a useful model for the study of wheat, which has a larger and more complex genome. Most existing barley transformation methodologies are either complex or have low (<10%) transformation efficiencies.

Results: A robust, simple and reproducible barley transformation protocol has been developed that yields average transformation efficiencies of 25%. This protocol is based on the infection of immature barley embryos with Agrobacterium strain AGL1, carrying vectors from the pBract series that contain the hpt gene (conferring hygromycin resistance) as a selectable marker. Results of large scale experiments utilising the luc (firefly luciferase) gene as a reporter are described. The method presented here has been used to produce hundreds of independent, transgenic plant lines and we show that a large proportion of these lines contain single copies of the luc gene.

Conclusion: This protocol demonstrates significant improvements in both efficiency and ease of use over existing barley transformation methods. This opens up opportunities for the development of functional genomics resources in barley.

Figure 1

Vectors used. (a) and (b): pBract vectors. (a). pBract203 destination vector containing the hpt gene under the control of a CaMV 35s promoter together with the chlormaphenicol resistance gene and the negative selection ccdB gene flanked by attR Gateway recombination sites. (b). pBract215 expression vector, created by cloning a luciferase cassette into pBract203 (via Gateway recombination). This construct contains the hpt gene under the control of a CaMV 35s promoter together with the luciferase gene under the control of a maize ubiquitin promoter (ubi1). (c): pSoup containing pSa-RepA (a component of the Agrobacterium replication origin), the replicase gene trfA, a tetracycline resistance gene (tetR), a broad-host range origin of replication (pRK2) and a multiple cloning site.

Figure 2

The effect of additional copper on the growth of callus from immature barley embryos. (a) and (b) show immature embryo derived callus after 27 days culture on callus induction medium. (a): standard callus induction medium. (b): Callus induction medium with 5 μM CuSO₄.

Figure 3

The effect of copper treatments on the number of shoots regenerated per embryo without transformation. Each bar represents the average number of shoots per embryo (recorded after 10 weeks in tissue culture) from 20–30 individual embryos. 'Standard' tissue culture regime: weeks 1–4 callus induction (CI) medium with no additional copper. 'Additional copper' tissue culture regime: weeks 1–4 CI + 5 μM CuSO₄. Callus from both treatments was transferred to standard transition media (which includes CuSO₄) after 4 weeks and standard regeneration medium after a further 2 weeks.

Figure 4

Copy numbers of the luciferase gene within transformed plants, determined by quantitative real-time PCR, given as a percentage of the 260 lines analysed. Note that '<1' refers to samples which gave a positive quantitative real-time PCR signal for luciferase and the internal control gene, but the luciferase signal occurred later than for single copy lines (a later signal normally implies less starting template) indicating there may be a problem with gene rearrangements.

Figure 5

Southern blot analysis of eight independently-transformed pBract216 lines. Genomic DNA was digested with BamHI and probed with a 1176 bp fragment of luciferase DNA. The lanes 1–8 correspond to lines 90-04-01, 90-06-01, 70-05-01, 70-07-01, 70-21-01, 84-07-01, 87-07-01, and 81-14-01 respectively. Lane 9 corresponds to a wild-type Golden Promise control.