Visual selection allows immediate identification of transgenic rice calli efficiently accumulating transgene products

Abstract

In genetic transformation systems, antibiotic resistance genes are routinely used as powerful markers for selecting transformed cells from surrounding non-transformed cells. However, simultaneous use of the gene encoding green fluorescent protein (GFP) and an antibiotic resistance gene facilitates the selection process, since it allows visible selection of transformed cells. Here, we report the development of a visual selection system for transformed cells using a GFP marker without selection against antibiotics after Agrobacterium-mediated transformation in rice. Both GFP protein levels and GFP fluorescence in calli isolated by visual selection were higher than in calli selected on hygromycin (Hyg), suggesting that transgenic calli hyper-accumulating GFP were efficiently obtained by selection using GFP fluorescence itself rather than Hyg resistance. Furthermore, gfp transcripts in calli isolated by visual selection were more abundant than under Hyg selection; in contrast, transcript levels of hpt in calli selected visually were comparable to those obtained under Hyg selection. These results suggest that there was no correlation between hpt and gfp expression levels, despite the fact that they are aligned in tandem on an integrated locus after selection by either GFP fluorescence or Hyg resistance. This fact indicates that positional effects can influence the expression of each transgene differently, even when they are located in tandem at the same locus. In summary, based on our results, we discuss a model system for rice cell culture transformation for the production of recombinant proteins using visual selection.

Highly efficient visual selection of transgenic rice plants using green fluorescent protein or anthocyanin synthetic genes

“…In our previous report, 2.4 copies of T-DNA, on average, were integrated in rice plants selected on Hyg in Kasalath (Saika and Toki 2010), suggesting that the integrated T-DNA copy number in regenerated plants selected visually is comparable to that in plants obtained by Hyg selection in Kasalath. In addition, on average, 2.3 T-DNA copies were integrated in rice plants of Nipponbare selected visually with GFP (Saika and Toki 2009), suggesting that the integrated T-DNA copy number in regenerated plants of Kasalath selected by visual selection is comparable to that of Nipponbare. An inverse correlation between T-DNA copy number and gene expression has been commonly reported in a wildtype Arabidopsis background where gene silencing is active, although a positive correlation is observed in a gene silencing mutant background (Luo and Chen 2007).…”

“…Using the visual selection system, we succeeded in achieving clonal propagation of transformed calli of Kasalath and regeneration of transformed plants. The efficiency of clonal propagation (the ratio of the number of clonal lines propagated successfully to the number of Agrobacterium-infected calli) obtained by visual selection in Kasalath was estimated at 62.2-85% (Table 1), compared to 9.4-20.5% achieved in Nipponbare (Saika and Toki 2009). We compared the details of the Figure 2A, B.…”

“…In our previous report, we succeeded in achieving clonal propagation of transformed calli at a frequency of 9.4-20.5% in Nipponbare using visual selection with GFP (Saika and Toki 2009). In the clonal propagation step in Nipponbare, cells emitting green fluorescence were not observed in ϳ50% transformed callus (10 of 24 primary calli) grown on medium without Hyg at 10-14 days after the onset of Agrobacterium elimination (Supplemental Table 1), suggesting that an improvement of transformation efficiency could increase clonal propagation efficiency when using visual selection in rice.…”

“…Thus, visual selection without drugs can be applied not only for the production of lower risk genetic engineering for crop but also for the establishment of improved plant cell culture systems to produce recombinant proteins. However, it had been pointed out that the overall transformation efficiency using visual selection remains low due to the technical difficulties involved in clonal propagation of transformed cells following visual selection in rice, sugarcane and wheat (Elliott et al 1999;Jordan 2000;Saika and Toki 2009;Vain et al 1998).…”

“…To date, visual selection using a GFP marker without antibiotics in rice has been established in a transformation system using particle bombardment of immature embryos and Agrobacterium-mediated transformation of callus derived from mature seeds (see Supplemental Figure 1) (Saika and Toki 2009;Vain et al 1998). In our previous report, we succeeded in achieving clonal propagation of transformed calli at a frequency of 9.4-20.5% in Nipponbare using visual selection with GFP (Saika and Toki 2009).…”