Computational protocol: Effects of over-expressing a native gene encoding 5-enolpyruvylshikimate-3-phosphate synthase (EPSPS) on glyphosate resistance in Arabidopsis thaliana

Similar protocols

Protocol publication

[…] Three additional experiments (Experiments A, B, and C) were conducted to confirm results from the dose-response experiment and test for consistent relative resistances when glyphosate was applied to rosettes vs. flowering plants (). These experiments included 6 OX lines, 4–7 EV lines, and the wild-type line, all of which were subjected to discriminating dosages of either 1.0x or 0.5x glyphosate (). Three of the original OX lines were not used because two had low resistance to glyphosate (labeled OX8, OX9; ), and a third (OX7) produced plants that were severely deformed, with abnormal rosettes and very short internodes at flowering. We also eliminated two randomly chosen EV lines (labelled EV8, EV9; ) because these were deemed unnecessary.Seeds were germinated in moistened Sun Gro Metro-Mix 360 soil, with one seedling per pot (4.5 or 6 cm square pots), and one pot per line arranged in random positions within trays. The experiments included 15–24 plants per line (i.e., 15–24 trays; ). All trays were irrigated from the bottom every 2 or 3 days and were rotated weekly to account for variation in environmental conditions. Experiments A and B were started in a growth room with 8 hrs of light for one month and then transferred to a greenhouse with supplemental light as above and ambient day lengths (13–15 hrs). For Experiment B, the pots were fertilized twice using 25 mL of nutrient solution (180 ppm 20:10:20; and then were sub-irrigated with this nutrient solution approximately every 3 weeks. Supplemental light was used to provide 16 hours of daylight. Fertilizer was not used in Experiments A or C. Plants in Experiment C were started over four consecutive days to stagger the timing of leaf samples used for EPSPS gene expression (see below).Plants in Experiments A and B were sprayed at 44 days after planting, prior to flowering, while those in Experiment C were sprayed at 35 days when they had started flowering (). Glyphosate treatments of 0.5x or 1.0x were applied as described above. Resistance to glyphosate was recorded 21 days after spraying in all experiments. Visual damage scores were “0” for dead plants, “1” for almost dead, “2” for likely to die, “3” for likely to survive, “4” for partly damaged, and “5” for those that were mostly green and had green meristems (center of rosette). This scoring system was deemed to be more efficient and easier to carry out than that used in the dose-response experiment, which involved scores of 1–10. For Experiment A, we also weighed the fresh, above-ground, living biomass of each surviving plant. A one-way ANOVA with Tukey’s multiple range tests was performed to compare above-ground biomass or visual damage scores among the OX, EV, and wild-type lines, using the software IBM SPSS Statistics ver. 19.0 for Windows (SPSS Inc., IBM Company Chicago, IL, USA, 2010). [...] Quantitative real-time PCR was used to estimate gene expression of EPSPS relative to the native Actin7 gene (e.g., []). We compared EPSPS expression levels among 6 OX lines, 3 EV lines, and wild-type plants from Experiment C, before glyphosate was applied (). Because gene expression is likely to vary over time, we staggered the planting and RNA sampling of these lines over four days. Thus, on days 1–4, we planted six plants per line each day, for a total of 24 plants per line. Three plants per line were sampled from rosettes, 28 days after planting, and the other three were sampled at the flowering stage, 35 days after planting, to obtain composite samples (see below). This provided independent estimates of gene expression from different plants at two distinct developmental stages. Then, all 24 plants per line were sprayed with glyphosate at 35 days after planting, as described above.On each day of sampling, we collected one healthy, young leaf (~1 cm2 leaf area, ~75–100 mg) per plant and combined the three leaves per line into a composite sample. Thus, for each line and each stage of development, we obtained four composite samples (replicates), which were frozen immediately in liquid nitrogen and used for RNA extraction. RNA was extracted using RNeasy Plant Mini Kit (Qiagen), quantified using a Nano drop spectrophotometer (ND-1000, Thermo Scientific), and treated by DNase I (Invitrogen). The cDNA was created by a Reverse-transcriptase (RT) reaction using Promega Reverse-Transcription System. For the real-time PCR, all primers were designed using PRIMER 3 (version 0.4.0) ( and synthesized by Eurofins MWG Operon LLC. Primer sequences for EPSPS were 5’-TCGTGCTGTAGTTGAAGGATG-3’ (forward) and 5’-GCGGTAAGTGGACGCATT-3’ (reverse); Primer sequence for Actin7 were 5’- CAGTGTCTGGATCGGAGGAT-3’ (forward) and 5’- TGAACAATCGATGGACCTGA-3’ (reverse). Three reaction mixes of 25ul for real-time PCR was prepared for each composite sample with iQ™ SYBR® Green Supermix from BioRad proportioned according to its user guide. A two-step program on the real-time PCR cycler (Bio-Rad CFX96 Touch™) with annealing temperature at 60 C was used with each composite sample. Efficiencies for the EPSPS (96.1%) and Actin7 (104%) primer pairs were determined by slopes of dilution curves with R2>0.99. The lack of primer dimers and specificity of each primer was demonstrated through gel analysis, melt curves, and sequencing of products.The Relative Normalized Quantification of EPSPS in each composite sample was calculated based on the modified 2-ΔΔCT method [,]. Wild-type samples were used as the control and EPSPS and Actin 7 were the target and reference (ref) genes, respectively. To calculate the relative quantification, the modified equation used is: (Etarget+1)ΔCqtarget(Eref+1)ΔCqref where ΔCqtarget=Cqtargetcontrol−Cqtargetsample, ΔCqref=Cqrefcontrol−Cqrefsample, E indicates primer efficiency, and Cq is the quantification cycle. To summarize, the sampling procedure entailed 10 lines x 2 developmental stages x 4 days on which we collected one independent composite sample per line, for a total of 80 composite samples.The final Relative Normalized Quantification of each OX, EV, or wild-type line at each growth stage (vegetative vs. flowering) was calculated by averaging the values of composite samples from different days and weighting the values such that the wild-type line had a value of 1.0. In some lines, no usable data were obtained for one of the four days of sampling, resulting in final sample sizes of 3–4, except for one flowering line, OX1, which had two samples. For each growth stage, the average expression levels of the 6 OX, 3 EV, and the wild-type lines were compared using a two-way ANOVA to account for differences among sampling dates (blocks), with Tukey’s multiple range tests (IMB SPSS Statistics ver. 19.0 for Windows). Although our final sample sizes were lower than planned, differences in gene expression associated with the OX construct were clear and consistent, as shown below. […]

Pipeline specifications

Software tools SPSS, Primer3
Applications Miscellaneous, qPCR
Diseases Musculoskeletal Diseases
Chemicals Shikimic Acid