Identification and functional analysis of two P450 enzymes of Gossypium hirsutum involved in DMNT and TMTT biosynthesis

Abstract

The homoterpenes (3E)-4,8-dimethyl-1,3,7-nonatriene (DMNT) and (E,E)-4,8,12-trimethyl-1,3,7,11-tridecatetraene (TMTT) are major herbivore-induced plant volatiles that can attract predatory or parasitic arthropods to protect injured plants from herbivore attack. In this study, DMNT and TMTT were confirmed to be emitted from cotton (Gossypium hirsutum) plants infested with chewing caterpillars or sucking bugs. Two CYP genes (GhCYP82L1 and GhCYP82L2) involved in homoterpene biosynthesis in G. hirsutum were newly identified and characterized. Yeast recombinant expression and enzyme assays indicated that the two GhCYP82Ls are both responsible for the conversion of (E)-nerolidol to DMNT and (E,E)-geranyllinalool to TMTT. The two heterologously expressed proteins without cytochrome P450 reductase fail to convert the substrates to homoterpenes. Quantitative real-time PCR (qPCR) analysis suggested that the two GhCYP82L genes were significantly up-regulated in leaves and stems of G. hirsutum after herbivore attack. Subsequently, electroantennogram recordings showed that electroantennal responses of Microplitis mediator and Peristenus spretus to DMNT and TMTT were both dose dependent. Laboratory behavioural bioassays showed that females of both wasp species responded positively to DMNT and males and females of M. mediator could be attracted by TMTT. The results provide a better understanding of homoterpene biosynthesis in G. hirsutum and of the potential influence of homoterpenes on the behaviour of natural enemies, which lay a foundation to study genetically modified homoterpene biosynthesis and its possible application in agricultural pest control.

Keywords: Gossypium hirsutum; DMNT and TMTT; EAG recording; behavioural response; cytochrome P450; enzymatic activity assay; expression profile.

Figure 1

GC‐MS analysis of DMNT and TMTT emitted from cotton plants. (a) I, gas chromatogram of authentic DMNT; II, volatiles from cotton plants infested with H. armigera; III, volatiles from cotton plants infested with A. lucorum; IV, volatiles from control cotton plants; V, volatiles from an empty glass jar; (b) mass spectrum of peak 1; (c) I, gas chromatogram of authentic TMTT; II‐V are the same as described in (a) II–V; (d) mass spectrum of peak 2. 1, DMNT; 2, TMTT.

Figure 2

Alignment of deduced amino acid sequences of candidate CYPs with AtCYP82G1. PRD, proline‐rich domain; PD, PERF domain; HBD, heme‐binding domain. SRS1‐6 indicates substrate recognition sites as predicted for AtCYP82G1. Residues with an asterisk represent the key substrate‐interacting residues identified in AtCYP82G1.

Figure 3

GC‐MS analysis of DMNT and TMTT produced by recombinant proteins expressed in S. cerevisiae with C₁₅ or C₂₀ substrates. (a) I, gas chromatogram of authentic DMNT; II–VIII, volatiles produced by recombinant proteins CotAD_38483, CotAD_50571‐1, CotAD_50571‐2, CotAD_50575, CotAD_66393, CotAD_58474 and empty vector pYES2, respectively, with (E)‐nerolidol; (b) I, gas chromatogram of authentic TMTT; II–VIII, volatiles produced by recombinant proteins CotAD_38483, CotAD_50571‐1, CotAD_50571‐2, CotAD_50575, CotAD_66393, CotAD_58474 and empty vector pYES2, respectively, with (E,E)‐geranyllinalool. 1, DMNT; 2, TMTT.

Figure 4

Effect of CPR on the enzymatic activity of recombinant CotAD_50571 proteins. (a) I, gas chromatogram of authentic DMNT; II–VII, volatiles produced by recombinant CotAD_50571‐1 protein co‐expressed with CPR, CotAD_50571‐1 without CPR, CotAD_50571‐2 with CPR, CotAD_50571‐2 without CPR, empty vector pYES2 with CPR and empty vector pYES2 without CPR, respectively, when (E)‐nerolidol was used as substrate; (b) I, gas chromatogram of authentic TMTT; II–VII, volatiles produced by recombinant CotAD_50571‐1 protein co‐expressed with CPR, CotAD_50571‐1 without CPR, CotAD_50571‐2 with CPR, CotAD_50571‐2 without CPR, empty vector pYES2 with CPR and empty vector pYES2 without CPR, respectively, when (E,E)‐geranyllinalool was used as substrate. 1, DMNT; 2, TMTT.

Figure 5

Phylogenetic relationships of P450s of the plant CYP 82 family. Multiple alignments were performed with Clustal W, and the tree was generated with MEGA 6.0. The two sequences marked with dots were used for homoterpene biosynthesis in G. hirsutum. Am, Ammi majus; At, Arabidopsis thaliana; Cp, Carica papaya; Ec, Eschscholzia californica; Gm, Glycine max; Ms, Medicago sativa; Nt, Nicotiana tabacum; Ps, Pisum sativum; Pt, Populus trichocarpa.

Figure 6

Transcript abundance of CYP82L genes in different organs of G. hirsutum. Data represent the means ± SE. Asterisks indicate significant differences between treatments (**P < 0.01, *P < 0.05).

Figure 7

EAG responses of M. mediator and P. spretus to DMNT and TMTT. (a) MmF, females of M. mediator; MmM, males of M. mediator; (b) PsF, females of P. spretus; PsM, males of P. spretus.

Figure 8

Behavioural responses of M. mediator and P. spretus to DMNT and TMTT. MmF, females of M. mediator; MmM, males of M. mediator; PsF, females of P. spretus; PsM, males of P. spretus. The percentage of wasps that chose mineral oil (white bars) versus DMNT (grey bars) or mineral oil (white bars) versus TMTT (tawny bars) are shown in figure. The numbers on each bar represent the number of wasps that made a choice. Sixty wasps in total were tested in each treatment. Asterisks indicate a significant difference with a choice test: **P < 0.01, *P < 0.05, and ns indicates no significant difference.