Neofunctionalization of Duplicated P450 Genes Drives the Evolution of Insecticide Resistance in the Brown Planthopper

Abstract

Gene duplication is a major source of genetic variation that has been shown to underpin the evolution of a wide range of adaptive traits [1, 2]. For example, duplication or amplification of genes encoding detoxification enzymes has been shown to play an important role in the evolution of insecticide resistance [3-5]. In this context, gene duplication performs an adaptive function as a result of its effects on gene dosage and not as a source of functional novelty [3, 6-8]. Here, we show that duplication and neofunctionalization of a cytochrome P450, CYP6ER1, led to the evolution of insecticide resistance in the brown planthopper. Considerable genetic variation was observed in the coding sequence of CYP6ER1 in populations of brown planthopper collected from across Asia, but just two sequence variants are highly overexpressed in resistant strains and metabolize imidacloprid. Both variants are characterized by profound amino-acid alterations in substrate recognition sites, and the introduction of these mutations into a susceptible P450 sequence is sufficient to confer resistance. CYP6ER1 is duplicated in resistant strains with individuals carrying paralogs with and without the gain-of-function mutations. Despite numerical parity in the genome, the susceptible and mutant copies exhibit marked asymmetry in their expression with the resistant paralogs overexpressed. In the primary resistance-conferring CYP6ER1 variant, this results from an extended region of novel sequence upstream of the gene that provides enhanced expression. Our findings illustrate the versatility of gene duplication in providing opportunities for functional and regulatory innovation during the evolution of an adaptive trait.

Keywords: Nilaparvata lugens; P450; duplication; imidacloprid; neofunctionalization; resistance.

Figure 1

Characterization of CYP6ER1 Variants in BPH Populations (A) Number and type of nucleotide polymorphisms in different CYP6ER1 variants relative to CYP6ER1vL, the variant observed in the lab-susceptible strain. (B and C) Relative expression of CYP6ER1 variants in imidacloprid-resistant BPH field strains (NLF1-8) and a susceptible strain (NLS), as determined by cDNA cloning and sequencing (B), and variant-specific QPCR (C). In (C), letters above bars are used to denote significant (p = < 0.01 in all cases) differences in expression between variants within each strain as assessed by one-way ANOVA with post-hoc Tukey HSD. Error bars in (C) indicate 95% confidence intervals (n = 4). (D) Metabolism of imidacloprid by recombinantly expressed CYP6ER1 variants. NADPH-dependent conversion of imidacloprid to 4/5-hydroxy imidacloprid (IMI-OH) and 6-chloronicotinic acid (6-CNA) is shown. Error bars indicate 95% confidence intervals (n = 3). See also Figure S1 and Tables S1 and S2.

Figure 2

Modeling the Active Site of CYP6ER1 Reveals the Impact of Amino Acid Alterations on Imidacloprid Binding (A) Amino-acid alignment of CYP6ER1 variants highlighting substitutions and deletions within substrate recognition sites four and five (boxed in red). (B–D) Protein homology modeling for 3 different CYP6ER1 variants (upper row), showing key residues surrounding the catalytic site. Amino-acid positions T318/A376 in CYP6ER1vL (B), S318/G376 in CYP6ER1vA (C), and S318/A376 in CYP6ER1vB (D) are in close proximity (spacefill representation). Imidacloprid docking into the active site is illustrated (lower row) by colored volumes, constituting an envelope around an ensemble of possible binding poses. See also Figure S1.

Figure 3

Genomic Analyses of the CYP6ER1 Locus (A–F) Coverage plots of DNA-seq reads from the NLS (imidacloprid susceptible) strain and NLF7 and NLF2 (imidacloprid resistant) strains mapped to the coding sequence of two reference single-copy genes: the voltage-gated sodium channel (VGSC) (A and D) and CYP6AY1 (B and E), and CYP6ER1 (C and F). (G and H) Copy number of CYP6ER1 in the NLF7 (G) and NLF2 (H) strains relative to NLS determined by qPCR. Error bars indicate 95% confidence intervals (n = 4). ^∗∗∗p < 0.001; one-way ANOVA with post hoc Tukey HSD. (I) Number of sequenced colonies obtained of each CYP6ER1 variant after cloning and sequencing PCR products amplified from either genomic DNA or mRNA of individuals of the NLF2 and NLF7 strains. Error bars indicate 95% confidence limits (n = 3). NS, not significant. ^∗∗p < 0.01; paired t test. (J) Assembly of gene capture long reads reveals marked variation in intron size between different variants. The position of exons is shown by blue arrow heads, with the partially duplicated exon in CYP6ER1vB highlighted in red. Gaps illustrate the position of assembly gaps. (K) Alignment of different putative promoter variants of CYP6ER1 upstream of the translation start codon. Gray regions indicate similarity between sequences, and black regions indicate sequence differences. Indels are indicated by gaps in the sequences. The identity plot above the alignment displays the identity across all sequences for every position. Green indicates that the residue at the position is the same across all sequences. Yellow is for less than complete identity, and red refers to very low identity for the given position. The position of the breakpoint observed in CYP6ER1vA is illustrated with an arrow. (L) Reporter gene activity (normalized to renilla fluorescence) of CYP6ER1 promoter variants. Letters above bars indicate significant differences, p < 0.001; one-way ANOVA with post-hoc Tukey HSD. Error bars indicate 95% confidence limits (n = 3). See also Figures S2, S3, and S4.