In many annual plant species, transgene inactivation occurs most often when multiple incomplete/complete copies of the transgene are present in a genome. The expression of single-copy transgene loci may also be negatively influenced by the flanking plant DNA and/or chromosomal location (position effect). To understand transgene silencing in a long-lived tree system, we analyzed several wild (Populus tremula L.) and hybrid (P. tremula L. x P. tremuloides Michx.) aspen lines transgenic to the rolC phenotypical marker system and grown under in vitro, greenhouse and field conditions. The morphological features of the 35S-rolC gene construct were used to screen lines with altered transgene expression, which was later confirmed by Northern experiments. Molecular analyses of hybrid aspen revealed that transgene inactivation was always a consequence of transgene repeats. In wild non-hybrid aspen, however, multiple-insertion-based altered or loss of rolC expression was observed only in three out of six lines showing transgene inactivation. Sequencing analysis revealed AT-rich patches at the transgene flanking genomic regions of some of the wild aspen transgenic lines. One wild aspen line showing variable rolC expression revealed characteristic integration of the transgene into genomic regions containing a high AT content (85% or more). In the remaining two wild aspen transgenic lines unstable for rolC expression, single-copy integration and non-AT-rich or repeat-free transgene flanking regions were found. A partial suppression of rolC was observed in some plants of one of the field-grown wild aspen transgenic lines. In the other wild aspen transgenic line an additional mutant phenotype along with transgene inactivation was found. This indicates that the host genome has some control over expression of a transgene, and the possible role of AT-rich regions in defense against foreign DNA.

Site-directed nucleases (SDNs) used for targeted genome editing are powerful new tools to introduce precise genetic changes into plants. Like traditional approaches, such as conventional crossing and induced mutagenesis, genome editing aims to improve crop yield and nutrition. Next-generation sequencing studies demonstrate that across their genomes, populations of crop species typically carry millions of single nucleotide polymorphisms and many copy number and structural variants. Spontaneous mutations occur at rates of ;10 28 to 10 29 per site per generation, while variation induced by chemical treatment or ionizing radiation results in higher mutation rates. In the context of SDNs, an off-target change or edit is an unintended, nonspecific mutation occurring at a site with sequence similarity to the targeted edit region. SDN-mediated offtarget changes can contribute to a small number of additional genetic variants compared to those that occur naturally in breeding populations or are introduced by induced-mutagenesis methods. Recent studies show that using computational algorithms to design genome editing reagents can mitigate off-target edits in plants. Finally, crops are subject to strong selection to eliminate off-type plants through well-established multigenerational breeding, selection, and commercial variety development practices. Within this context, off-target edits in crops present no new safety concerns compared to other breeding practices. The current generation of genome editing technologies is already proving useful to develop new plant varieties with consumer and farmer benefits. Genome editing will likely undergo improved editing specificity along with new developments in SDN delivery and increasing genomic characterization, further improving reagent design and application. PLANT GENETIC VARIABILITY Genetic differences between individuals are the basis of adaptation and evolution. Plant breeding, as a form of directed evolution, has a long history of using genetic diversity for crop improvement. During the process of crop domestication, humans selected individual plants with favorable traits that resulted from novel mutations or standing variation in the ancestral species. The process of selecting plant varieties with favorable characteristics for cultivation and consumption continues to the present day. Modern plant breeding is a more directed process than the crop improvement that occurred through the history and prehistory of most

CRISPR-Cas9 enabled genome engineering has great potential for improving agriculture productivity, but the possibility of unintended off-target edits has evoked some concerns. Here we employ a three-step strategy to investigate Cas9 nuclease specificity in a complex plant genome. Our approach pairs computational prediction with genome-wide biochemical off-target detection followed by validation in maize plants. Our results reveal high frequency (up to 90%) on-target editing with no evidence of off-target cleavage activity when guide RNAs were bioinformatically predicted to be specific. Predictable off-target edits were observed but only with a promiscuous guide RNA intentionally designed to validate our approach. Off-target editing can be minimized by designing guide RNAs that are different from other genomic locations by at least three mismatches in combination with at least one mismatch occurring in the PAM proximal region. With well-designed guides, genetic variation from Cas9 off-target cleavage in plants is negligible, and much less than inherent variation.

SummaryTo obtain insight into the mechanism of transferred DNA (T-DNA) integration in a long-lived tree system, we analysed 30 transgenic aspen lines. In total, 27 right T-DNA/plant junctions, 20 left T-DNA/plant junctions, and 10 target insertions from control plants were obtained. At the right end, the T-DNA was conserved up to the cleavage site in 18 transgenic lines (67%), and the right border repeat was deleted in nine junctions. Nucleotides from the left border repeat were present in 19 transgenic lines out of 20 cases analysed. However, only four (20%) of the left border ends were conserved to the processing end, indicating that the T-DNA left and right ends are treated mechanistically differently during the T-DNA integration process. Comparison of the genomic target sites prior to integration to the T-DNA revealed that the T-DNA inserted into the plant genome without any notable deletion of genomic sequence in three out of 10 transgenic lines analysed. However, deletions of DNA ranging in length from a few nucleotides to more than 500 bp were observed in other transgenic lines. Filler DNAs of up to 235 bp were observed on left and/or right junctions of six transgenic lines, which in most cases originated from the nearby host genomic sequence or from the T-DNA. Short sequence similarities between recombining strands near break points, in particular for the left T-DNA end, were observed in most of the lines analysed. These results con®rm the well-accepted T-DNA integration model based on single-stranded annealing followed by ligation of the right border which is preserved by the VirD2 protein. However, a second category of T-DNA integration was also identi®ed in nine transgenic lines, in which the right border of the T-DNA was partly truncated. Such integration events are described via a model for the repair of genomic double-strand breaks in somatic plant cells based on synthesis-dependent strandannealing. This report in a long-lived tree system provides major insight into the mechanism of transgene integration.

No abstract