Glyphosate-resistant goosegrass. Identification of a mutation in the target enzyme 5-enolpyruvylshikimate-3-phosphate synthase

Abstract

The spontaneous occurrence of resistance to the herbicide glyphosate in weed species has been an extremely infrequent event, despite over 20 years of extensive use. Recently, a glyphosate-resistant biotype of goosegrass (Eleusine indica) was identified in Malaysia exhibiting an LD(50) value approximately 2- to 4-fold greater than the sensitive biotype collected from the same region. A comparison of the inhibition of 5-enolpyruvylshikimate-3-phosphate synthase (EPSPS) activity by glyphosate in extracts prepared from the resistant (R) and sensitive (S) biotypes revealed an approximately 5-fold higher IC(50)(glyphosate) for the (R) biotype. Sequence comparisons of the predicted EPSPS mature protein coding regions from both biotypes revealed four single-nucleotide differences, two of which result in amino acid changes. One of these changes, a proline to serine substitution at position 106 in the (R) biotype, corresponds to a substitution previously identified in a glyphosate-insensitive EPSPS enzyme from Salmonella typhimurium. Kinetic data generated for the recombinant enzymes suggests that the second substitution identified in the (R) EPSPS does not contribute significantly to its reduced glyphosate sensitivity. Escherichia coli aroA- (EPSPS deficient) strains expressing the mature EPSPS enzyme from the (R) biotype exhibited an approximately 3-fold increase in glyphosate tolerance relative to strains expressing the mature EPSPS from the (S) biotype. These results provide the first evidence for an altered EPSPS enzyme as an underlying component of evolved glyphosate resistance in any plant species.

Figure 1

Basal and induced EPSPS expression levels in glyphosate-sensitive and -resistant goosegrass individuals. Two clones were generated from each representative plant; (S) clones were spray-treated with 0.5 kg acid equivalent (a.e.) ha⁻¹ glyphosate and (R) clones were treated with 2 kg a.e. ha⁻¹ glyphosate. Clones were harvested immediately (time [t] = 0 h) or 48 h post-treatment (t = 48 h), then independently analyzed for EPSPS activity and mRNA levels. a, EPSPS activity levels. Extracts prepared from crown tissues were radiometrically assayed for EPSPS activity (see “Materials and Methods”). Specific activities were calculated based on extract protein concentration and all assays were performed in triplicate. Each bar represents the mean activity observed within five individuals from each biotype. Error bars indicate sds. b, RNA-blot analysis. Ten micrograms of total RNA, isolated from crown tissues, was loaded per lane, then size fractionated on 1.0% (w/v) agarose gels containing 0.66 m formaldehyde and transferred to nylon membranes. Blots were hybridized with an L. rigidum ³²P-labeled EPSPS cDNA. S1 through S5 samples were isolated from duplicate clones derived from five different (S) individuals; R1 through R5 samples were isolated from duplicate clones derived from five different (R) individuals.

Figure 2

DNA-blot analysis of glyphosate-sensitive and -resistant goosegrass individuals. Ten micrograms of Genomic DNA from five sensitive individuals (samples S1–S5) and five resistant individuals (samples R1–R5) were digested with either EcoRV, SphI, SspI, or ScaI as indicated above, then size fractionated on 0.8% (w/v) agarose gels and transferred to nylon membranes. Blots were then hybridized with an L. rigidum ³²P-labeled EPSPS cDNA, washed at high stringency (see “Materials and Methods”), then subjected to autoradiography.

Figure 3

IC₅₀ (glyphosate) determinations for the (R) and (S) biotype EPSPS activities. Radiometric EPSPS assays were performed on (R) and (S) biotype crown tissue extracts in the presence of glyphosate at concentrations ranging from 0.1 to 100 μm. Inhibition curves and IC₅₀ (inflection point) values were generated by nonlinear regression analysis using the GraFit software package (Leatherbarrow, 1998).

Figure 4

Sequence comparison of EPSPS cDNAs isolated from the (S) and (R) goosegrass biotypes. The nucleotide sequence of the (S) EPSPS is indicated in the line below only where nucleotide differences occur between the two biotypes. The translation termination codon (TAA) is marked with an asterisk. Single nucleotide differences and corresponding amino acid substitutions are indicated in bold face. PCR primer sites are indicated by arrows. The deduced amino acid sequence of the (R) EPSPS mature protein coding region is shown above the nucleotide sequence, which has approximately 98% identity with the EPSPS enzyme from corn (Zea mays; GenBank accession no. X63374) and a predicted M_r of 47,402. A motif conserved in all plant and most bacterial EPSPS enzymes is boxed in black. Amino acid assignments referred to in the text are based on the plant EPSPS numbering system used by Padgette et al. (1996). EMBL accession numbers for the (R) and (S) EPSPS sequences are AJ417033 and AJ417034, respectively.

Figure 5

DNA cassettes containing different EPSPS mature protein coding regions for E. coli expression studies. P-lac and lac-O represent the native promoter and operator sequence from the E. coli lac operon. The Shine/Dalgarno sequence is indicated by the letters S/D. The predicted EPSPS mature coding region from the resistant biotype is indicated by (R) and the sensitive biotype EPSPS by (S). P106S is an engineered variant of (R) containing a Pro at position 381; P381L is a an engineered variant of (R) containing a Pro at position 106; “ter” represents the rho-independent transcriptional terminator element of the E. coli trpA gene (Sato et al., 1987).

Figure 6

Glyphosate sensitivity of pMON45364- and pMON45365-transformed cell lines. Overnight cultures of E. coli SR481 (aroA−) cell lines harboring pMON45364 [containing the EPSPS cDNA from the (R) biotype], and pMON45365 [containing the EPSPS cDNA from the (S) biotype], were used to inoculate 3.0-mL cultures containing minimal media supplemented with either 0.0, 0.5, 1.5, or 5.0 mm glyphosate. Aliquots were removed at t = 0, 24, 28, 32, and 48 h and optical densities were monitored at 595 nm. Before inoculation, optical density measurements were taken from the original overnight cultures to confirm similar cell densities. Each data point represents the mean of three replicates; error bars indicate sd. Where not visible, the error is enclosed within the data symbol. C, Growth rates of cells harboring pMON45337 (“empty” vector control). a, Growth rates of cells expressing the mature (S) EPSPS from pMON45365. b, Growth rates of cells expressing the mature (R) EPSPS from pMON45364.