An evaluation of new and established methods to determine T-DNA copy number and homozygosity in transgenic plants.

Stable transformation of plants is a powerful tool for hypothesis testing. A rapid and reliable evaluation method of the transgenic allele for copy number and homozygosity is vital in analysing these transformations. Here the suitability of Southern blot analysis, thermal asymmetric interlaced (TAIL-)PCR, quantitative (q)PCR and digital droplet (dd)PCR to estimate T-DNA copy number, locus complexity and homozygosity were compared in transgenic tobacco. Southern blot analysis and ddPCR on three generations of transgenic offspring with contrasting zygosity and copy number were entirely consistent, whereas TAIL-PCR often underestimated copy number. qPCR deviated considerably from the Southern blot results and had lower precision and higher variability than ddPCR. Comparison of segregation analyses and ddPCR of T-1 progeny from 26 T-0 plants showed that at least 19% of the lines carried multiple T-DNA insertions per locus, which can lead to unstable transgene expression. Segregation analyses failed to detect these multiple copies, presumably because of their close linkage. This shows the importance of routine T-DNA copy number estimation. Based on our results, ddPCR is the most suitable method, because it is as reliable as Southern blot analysis yet much faster. A protocol for this application of ddPCR to large plant genomes is provided. Genetic transformation is being used increasingly in the public domain to test a range of hypotheses concerning gene action, not only in Arabidopsis, but now in a broad range of plants. A major challenge though, particularly with species with relatively long life cycles, is in identifying individuals that are homozygous for the insert or DNA modification at T2, which is necessary to provide homozygous lines. Southern blotting has been the traditional approach, but it is slow and requires considerable skill. Various PCR methods have been used to accelerate testing, but have not been as reliable. However, we show here that the recently developed digital droplet PCR is as effective as Southern blotting, yet faster and capable of high-throughput. A protocol for application of ddPCR is provided together with evidence of its efficacy.