Evolution of the isoprene biosynthetic pathway in kudzu

Abstract

Isoprene synthase converts dimethylallyl diphosphate, derived from the methylerythritol 4-phosphate (MEP) pathway, to isoprene. Isoprene is made by some plants in substantial amounts, which affects atmospheric chemistry, while other plants make no isoprene. As part of our long-term study of isoprene synthesis, the genetics of the isoprene biosynthetic pathway of the isoprene emitter, kudzu (Pueraria montana), was compared with similar genes in Arabidopsis (Arabidopsis thaliana), which does not make isoprene. The MEP pathway genes in kudzu were similar to the corresponding Arabidopsis genes. Isoprene synthase genes of kudzu and aspen (Populus tremuloides) were cloned to compare their divergence with the divergence seen in MEP pathway genes. Phylogenetic analysis of the terpene synthase gene family indicated that isoprene synthases are either within the monoterpene synthase clade or sister to it. In Arabidopsis, the gene most similar to isoprene synthase is a myrcene/ocimene (acyclic monoterpenes) synthase. Two phenylalanine residues found exclusively in isoprene synthases make the active site smaller than other terpene synthase enzymes, possibly conferring specificity for the five-carbon substrate rather than precursors of the larger isoprenoids. Expression of the kudzu isoprene synthase gene in Arabidopsis caused Arabidopsis to emit isoprene, indicating that whether or not a plant emits isoprene depends on whether or not it has a terpene synthase capable of using dimethylallyl diphosphate.

Figure 1.

The isoprene biosynthetic pathway in plants. The isoprene biosynthetic pathway in plants consists of the MEP pathway (modified from Rodriguez-Concepción and Boronat [2002]) and isoprene synthase. Shown are the intermediates of the pathway: G3P, glyceraldehyde 3-phosphate; DXP; MEP; CDP-ME, 4-diphosphocytidyl-2C-methyl-d-erythritol 2-phosphate; CDP-MEP; ME-cPP, 2C-methyl-d-erythritol 2,4-cyclodiphosphate; HMBPP, 1-hydroxy-2-methyl-2-(E)-butenyl 4-diphosphate; IPP; DMAPP. The enzymes in the pathway are indicated in bold: DXS; DXR; CMS, CDP-ME synthase; CMK, CDP-ME kinase; MCS, ME-cPP synthase; HDS, HMBPP synthase; IDS, IPP/DMAPP synthase; IspS, isoprene synthase.

Figure 2.

Phylogenetic analysis of DXS genes. A rooted phylogram was generated using a ClustalX alignment of the nucleotide coding region of DXS genes using the neighbor-joining method. Bootstrap analysis was done with 1,000 replicates and some of the bootstrap values are indicated. The tree was rooted with C. reinhardtii DXS.

Figure 3.

Phylogenetic analysis of DXR genes. A rooted phylogram was generated using a ClustalX alignment of the nucleotide coding region of DXR genes using the neighbor-joining method. Bootstrap analysis was done with 1,000 replicates and some of the bootstrap values are indicated. The tree was rooted with monocot DXR sequences.

Figure 4.

Phylogenetic analysis of Idi genes. A rooted phylogram was generated using a ClustalX alignment of the nucleotide coding region of Ipi genes using the neighbor-joining method. Bootstrap analysis was done with 1,000 replicates and some of the bootstrap values are indicated. The tree was rooted with C. reinhardtii Idi. To improve clarity, not all genes used in the analysis are shown.

Figure 5.

Gene structure of the isoprene synthase genes. The gene structure for both isoprene synthase genes was predicted based on comparison with the cDNA sequence. The lengths of exons and introns are given in Table I.

Figure 6.

Isoprene synthase activity as a function of DMAPP concentration. Isoprene synthase with a 6× N-terminal His tag was grown in E. coli. Activity was assayed in sealed vials for 15 min at 35°C. A small amount of the headspace gas was then withdrawn and injected into a gas chromatograph for quantitation by photoionization. The line is modeled data assuming K_m = 7.7 mm, V_max = 75 nmol mg⁻¹P min⁻¹, and a Hill coefficient (H) of 4.1 using the equation v = V_max S^H/(S^H + K_m^H).

Figure 7.

The detection of isoprene synthase protein in transgenic plants. A, The specificity of the anti-serum against the kudzu isoprene synthase protein was determined using western-blot analysis of 7.5 μg soluble protein extracted from kudzu leaves treated with high light and high temperature. PI, Preimmune serum; I, immune serum. The Coomassie Blue, stained blot is shown to indicate equal loading of protein. B, Western-blot analysis of 7.5 μg soluble protein extracted from wild-type (wt) and transgenic Arabidopsis (ISP).

Figure 8.

Phylogenetic analysis of Tps-b and isoprene synthase. A rooted phylogram was generated using a ClustalX alignment of the nucleotide coding region of Tps-b and the isoprene synthase sequences with PAUP and maximum likelihood analysis. Included in the alignment were Tps-g and Tps-a coding sequences. (The Tps-a genes are not shown in this tree to simplify the presentation.) The isoprene synthase genes formed a group that appeared to be beside the Tps-b group. The known isoprene synthase sequences are indicated by bold lines.

Figure 9.

Model of the active site of 5EAS (A) and similar model with amino acids specific to isoprene synthase (B). The amino acid sequence of 5EAS was modified as follows: I294F, Y404G, C440F, and I443C. This modified sequence was submitted to SWISS-MODEL and the predicted structure was examined. Selected residues in the active site of 5EAS were displayed using DeepView-Spdbv 3.7 (http://www.expasy.org/spdbv/; Guex and Peitsch, 1997). The following residues were adjusted by changing rotamer conformation: F294, W273, and F440.

Figure 10.

Pictures of Arabidopsis transformed with kudzu isoprene synthase and gas chromatograph traces. Leaves of transgenic and untransformed plants were sealed in vials for 30 or 60 min at room temperature. Head space gas was withdrawn by syringe and injected into a gas chromatograph with detection by photoionization. The rate of isoprene production in the light was 1.32 nmol m⁻² s⁻¹ with a range from 0.88 to 1.7 for 3 plants derived from 1 transformation event. The peak on the top trace (for the transgenic plant) is 0.39 nmol isoprene and the same settings were used to test headspace air from the wild-type plant (bottom trace).