Agrobacterium-mediated transient transformation of Flaveria bidentis leaves: a novel method to examine the evolution of C4 photosynthesis

Abstract

The genus Flaveria has been studied extensively as a model for the evolution of C₄ photosynthesis. Thus far, molecular analyses in this genus have been limited due to a dearth of genomic information and the lack of a rapid and efficient transformation protocol. Since their development, Agrobacterium-mediated transient transformation protocols have been instrumental in understanding many biological processes in a range of plant species. However, this technique has not been applied to the genus Flaveria. Here, an efficient protocol for the Agrobacterium-mediated transient transformation of the leaves of the C₄ species Flaveria bidentis is presented. This technique has the distinct advantages of rapid turnaround, the ability to co-transform with multiple constructs, and the capacity to assay coding and non-coding regions of Flaveria genomes in a homologous context. To illustrate the utility of this protocol, the quantitative transcriptional regulation of phosphoenolpyruvate carboxylase, the primary carboxylase of C₄ plants, was investigated. A 24 bp region in the ppcA1 proximal promoter was found to elicit high levels of reporter gene expression. The Agrobacterium-mediated transient transformation of F. bidentis leaves will accelerate the understanding of the biology and evolution of C₄ photosynthesis in the genus Flaveria as well as in other C₄ lineages.

Keywords: Flaveria; C4 photosynthesis; Evolution of C4 photosynthesis; Phosphoenolpyruvate carboxylase; Transient transformation.

Fig. 1

Schema of the Agrobacterium-mediated transient transformation of Flaveria bidentis leaves. Green boxes denote plant preparation steps, purple boxes denote Agrobacterium preparation and subsequent transformation steps. Arrows indicate direction of workflow with the duration between steps noted beside each arrow and represented in either hours (hr), days (d), or weeks (wk). LB, Luria–Bertani medium. See ‘Methods’ for details

Fig. 2

Flaveria bidentis leaves suitable for transient transformation. A A mature Flaveria bidentis plant with leaf pairs labelled at the nodes, starting at the first fully expanded leaf pair 1. Leaves labelled with a red asterisk are suitable for transient transformation. B F. bidentis leaf suitable for transformation viewed from the abaxial surface. The mid-vein (MV) and lateral veins (LV) are marked with red arrowheads. The leaf area that yields the greatest and most reproducible transformation efficiency is highlighted in red. This area is bounded by the LVs and bisected by the MV. Regions X and Y are marked in this area on either side of the MV. No discernible transfer of infiltrate across the MV has been observed; therefore, regions X and Y can be infiltrated with different Agrobacterium suspensions

Fig. 3

Optimization of the transformation efficiency of the Agrobacterium-mediated Flaveria bidentis leaf transient transformation protocol. A and F Flaveria bidentis leaves transformed with an Agrobacterium suspension harboring a vector containing a gene encoding p19 under the control of the cauliflower mosaic virus 35S (CaMV) promoter. B–D and G–I F. bidentis leaves transformed with an Agrobacterium co-suspension of one strain containing a gene encoding the green fluorescent protein (GFP) under the control of the CaMV promoter, and a strain harboring the p19 vector. A–E Efficiency of the Agrobacterium-mediated F. bidentis leaf transient transformation system at different leaf developmental zones; negative control using the upper-middle zone and representative of all three leaf zones (A), the proximal (B), upper-middle (C), and distal (D) leaf zones. (E) Quantification of transformation efficiency along the developmental gradient. Three leaves were transiently co-transformed. For each developmental zone, the number of transformed and untransformed cells in three fields of view were counted, and transformed cells were expressed as a percentage of total cells. Each data point (blue) represents the mean transformation efficiency from one developmental zone of a single leaf, as calculated from the mean of three fields of view. Columns represent the mean transformation efficiency of the developmental zone from three leaves. F–J Efficiency of the Agrobacterium-mediated F. bidentis leaf transient transformation system at different acetosyringone concentrations. Each Agrobacterium co-suspension contained a different concentration of acetosyringone in the leaf infiltration buffer; 0.1 mM (G), 0.5 mM (H), and 1.0 mM (I). An acetosyringone concentration of 1.0 mM was used in the transformation control leaf represented in (F). J Each data point (yellow) represents the mean transformation efficiency of a single leaf. Columns represent the mean transformation efficiency conferred by each acetosyringone concentration. For all experiments, leaf disks were excised 3 days post-infiltration and were visualized via brightfield (BF) and epifluorescence microscopy using a GFP filter set (GFP; excitation: 454–490 nm, emission collection: 500–540 nm). Scale bar = 30 μm

Fig. 4

Efficiency of the Agrobacterium-mediated Flaveria bidentis leaf transient transformation system. A Leaf transformed with an Agrobacterium suspension harboring a vector containing a gene encoding p19 under the control of the cauliflower mosaic virus 35S (CaMV) promoter. B Leaf transformed with an Agrobacterium co-suspension of one strain containing a gene encoding the green fluorescent protein (GFP) under the control of the CaMV promoter, in addition to a strain harboring the p19 vector. Leaves were visualized 3 days post-infiltration via brightfield (BF) and epifluorescence microscopy with a GFP filter set (GFP; excitation: 454–490 nm, emission collection: 500–540 nm). Scale bar = 30 μm. C Quantification of transformation efficiency. Three leaves were transiently co-transformed on three different occasions (leaves 1–3, 4–6, and 7–9). The number of transformed and untransformed cells in three fields of view were counted and the transformed cells expressed as a percentage of the total. Each data point (green) represents the transformation efficiency in one field of view, and columns represent the mean transformation efficiency per leaf

Fig. 5

Transformability of Flaveria bidentis leaf cell types. A Flaveria bidentis leaf infiltrated with buffer containing no Agrobacterium. B F. bidentis leaf infiltrated with an Agrobacterium suspension harboring a gene encoding the green fluorescent protein (GFP) under the control of the upstream region of the gene encoding F. bidentis β-carbonic anhydrase 3 (Fbca3). C F. bidentis leaf infiltrated with an Agrobacterium suspension containing a gene encoding GFP fused to the 5′- and 3′-untranslated regions (UTR) of F. bidentis ribulose-1,5-bisphosphate carboxylase/oxygenase small subunit 1 and under the control of the cauliflower mosaic virus promoter (CaMV). Leaves were visualized 3 days post-infiltration via brightfield (BF) and epifluorescence microscopy with a GFP filter set (GFP; excitation: 454–490 nm, emission collection: 500–540 nm). Scale bar = 30 μm

Fig. 6

Quantification of promoter strength using Flaveria bidentis leaf transient transformation. A Maps of Flaveria trinervia C₄-associated phosphoenolpyruvate carboxylase (FtppcA1) truncated proximal promoter constructs. Constructs are depicted in beige and segmented based on truncation sites. The coordinates of the truncation sites are shown above the promoter maps and are relative to the first nucleotide upstream of the F. trinervia PPCA1 translational start site (ATG), shown in green, and referred to as –1. Truncation sites correspond to the name of the fragments; FtppcA1(–570), FtppcA1(–543), FtppcA1(–472), and FtppcA1(–362). The proximal promoter is segmented into nucleotide regions (R1-4), defined by the truncation sites. The position of the predicted TATA box and transcriptional start site (TSS) are denoted by arrowheads above the constructs. B Activity of F. trinervia ppcA1 promoter-reporter constructs in transiently transformed F. bidentis leaves. Mean ratios of firefly luciferase to Renilla luciferase are depicted as columns. All means are normalized to the mean of the FtppcA1(–570) construct. The normalized mean ratios of the technical triplicates for each transformed leaf disk are plotted in purple. Significance (α = 0.05) is denoted by letters above the data

Fig. 7

The promoter region controlling quantitative expression of Flaveria trinervia phosphoenolpyruvate carboxylase. A Maps of Flaveria trinervia C₄-associated phosphoenolpyruvate carboxylase (FtppcA1) truncated proximal promoter constructs. Constructs are depicted in beige and segmented based on truncation sites. The coordinates of the truncation sites are shown above the maps and are relative to the first nucleotide upstream of the F. trinervia PPCA1 translational start site (ATG), shown in green, and referred to as –1. Truncation sites correspond to the name of the fragments; FtppcA1(–448), FtppcA1(–421), and FtppcA1(–396). The construct FtppcA1(–472) is described in Fig. 6 and is depicted here for clarity. The position of the predicted TATA box and transcriptional start site (TSS) are denoted by arrowheads above the constructs. B Activity of F. trinervia ppcA1 promoter-reporter constructs in transiently transformed F. bidentis leaves. Mean ratios of firefly luciferase to Renilla luciferase are depicted as columns. All means are normalized to the mean of the FtppcA1(–570) construct (Fig. 6). The normalized mean ratios of the technical triplicates for each transformed leaf disk are plotted in orange. Significance (α = 0.05) is denoted by letters above the data

Fig. 8

Comparison of phosphoenolpyruvate carboxylase proximal promoters from Flaveria trinervia and Flaveria pringlei. A Nucleotide sequence alignment of the Flaveria trinervia C₄-associated phosphoenolpyruvate carboxylase (FtppcA1) proximal promoter region 3a to the homologous region of the F. pringlei ortholog (FpppcA). Asterisks indicate nucleotide conservation. Numerals indicate distance, in base pairs, from the translational start site. B Activity of F. trinervia ppcA1 and F. pringlei ppcA promoter-reporter constructs in transiently transformed F. bidentis leaves. Mean ratios of firefly luciferase to Renilla luciferase are depicted as columns. All means are normalized to the mean of the FtppcA1(–570) construct (Fig. 6). The normalized mean ratios of the technical triplicates for each transformed leaf disk are plotted in green. Significance (α = 0.05) is denoted by letters above the data