Molecular characterization of an Arabidopsis gene encoding hydroperoxide lyase, a cytochrome P-450 that is wound inducible

Abstract

Hydroperoxide lyase (HPL) cleaves lipid hydroperoxides to produce volatile flavor molecules and also potential signal molecules. We have characterized a gene from Arabidopsis that is homologous to a recently cloned HPL from green pepper (Capsicum annuum). The deduced protein sequence indicates that this gene encodes a cytochrome P-450 with a structure similar to that of allene oxide synthase. The gene was cloned into an expression vector and expressed in Escherichia coli to demonstrate HPL activity. Significant HPL activity was evident when 13S-hydroperoxy-9(Z),11(E),15(Z)-octadecatrienoic acid was used as the substrate, whereas activity with 13S-hydroperoxy-9(Z),11(E)-octadecadienoic acid was approximately 10-fold lower. Analysis of headspace volatiles by gas chromatography-mass spectrometry, after addition of the substrate to E. coli extracts expressing the protein, confirmed enzyme-activity data, since cis-3-hexenal was produced by the enzymatic activity of the encoded protein, whereas hexanal production was limited. Molecular characterization of this gene indicates that it is expressed at high levels in floral tissue and is wound inducible but, unlike allene oxide synthase, it is not induced by treatment with methyl jasmonate.

Figure 1

Biochemical pathway illustrating the enzymatic activity of HPL. HPL cleaves either HPOT (A) or HPOD (B), forming 12-oxo-trans-10-dodecenoic acid and cis-3-hexenal or hexanal, respectively. An isomerization factor (IF) interconverts cis-3-hexenal to trans-2-hexenal in vivo.

Figure 2

Nucleotide and deduced protein sequence of Arabidopsis HPL. The arrowhead denotes the start of the sequence used in protein-expression studies. Cyt P-450 domains A to D have lines above the protein sequence. An internal EcoRI site is boxed.

Figure 3

Protein-sequence comparison between Arabidopsis HPL (94J16) and related proteins. A, Comparison of the four Cyt P-450 domains (A–D) of Arabidopsis HPL (94J16) with those of green pepper HPL (CA-HPL), as well as those of AOS from Arabidopsis (AT-AOS), flax (FX-AOS), and rubber plant (RP-AOS). Asterisks in Cyt P-450 domain D denote residues conserved in HPL sequences that are dissimilar from corresponding residues in published AOS sequences. B, Structural comparison of the full-length sequences of 94J16, green pepper HPL (CA-HPL), and Arabidopsis AOS (AT-AOS). The figure divides the protein sequence into three regions: TP, putative transit peptide; N-Term, N-terminal region; and P-450, Cyt P-450 region, indicating the four Cyt P-450 domains (A–D). Percentage figures above the individual regions indicate the degree of similarity at the protein level between 94J16 and CA-HPL or AT-AOS.

Figure 4

SDS-PAGE analysis of fusion proteins produced in E. coli. Protein M_r marker (lane 1), purified GST (lane 2), and purified GST-94J16 fusion protein (lane 3) were run on a 12.5% polyacrylamide gel and the bands visualized by staining with Coomassie blue. The arrow indicates the full-length GST-94J16 fusion protein.

Figure 5

Total ion-count chromatogram of volatile compounds from the headspace of E. coli extracts containing GST protein alone (A and B) or 94J16 as a GST fusion protein (C and D). In A and C, HPOD was used as the substrate, and in B and D, HPOT was used as the substrate.

Figure 6

Analysis of HPL gene-expression patterns by northern-blot and RT-PCR analysis. A, Northern-blot analysis of HPL mRNA accumulation in different tissues of Arabidopsis, including inflorescence, green silique, mature leaf, and root tissue. Total RNA was extracted from each tissue and 10 μg was separated on a formaldehyde gel before blotting onto nylon membrane and hybridization with a 94J16 (HPL) gene-specific probe. B, RT-PCR analysis of HPL, AOS, and β-ATPase expression after wounding of leaf tissue for 15, 30, or 60 min. Total RNA was extracted from intact or wounded tissue, treated with DNase, and subjected to a RT reaction with random hexanucleotide primers. Equal quantities of the RT products were PCR amplified for 20 cycles with gene-specific oligonucleotides corresponding to each gene. An aliquot of the PCR product was separated on an agarose gel, blotted, and hybridized with probes specific for each gene. C, RT-PCR analysis of HPL, AOS, and β-ATPase expression after exposure to MeJA for 4 or 24 h. RT-PCR conditions were as in B.