Complementation of the Arabidopsis pds1 mutation with the gene encoding p-hydroxyphenylpyruvate dioxygenase

Abstract

Plastoquinone and tocopherols are the two major quinone compounds in higher plant chloroplasts and are synthesized by a common pathway. In previous studies we characterized two loci in Arabidopsis defining key steps of this biosynthetic pathway. Mutation of the PDS1 locus disrupts the activity of p-hydroxyphenylpyruvate dioxygenase (HPPDase), the first committed step in the synthesis of both plastoquinone and tocopherols in plants. Although plants homozygous for the pds1 mutation could be rescued by growth in the presence of homogentisic acid, the product of HPPDase, we were unable to determine if the mutation directly or indirectly disrupted HPPDase activity. This paper reports the isolation of a cDNA, pHPPD, encoding Arabidopsis HPPDase and its functional characterization by expression in both plants and Escherichia coli. pHPPD encodes a 50-kD polypeptide with homology to previously identified HPPDases, including 37 highly conserved amino acid residues clustered in the carboxyl region of the protein. Expression of pHPPD in E. coli catalyzes the accumulation of homogentisic acid, indicating that it encodes a functional HPPDase enzyme. Mapping of pHPPD and co-segregation analysis of the pds1 mutation and the HPPD gene indicate tight linkage. Constitutive expression of pHPPD in a pds1 mutant background complements this mutation. Finally, comparison of the HPPD genomic sequences from wild type and pds1 identified a 17-bp deletion in the pds1 allele that results in deletion of the carboxyterminal 26 amino acids of the HPPDase protein. Together, these data conclusively demonstrate that pds1 is a mutation in the HPPDase structural gene.

Figure 1

The plastoquinone and α-tocopherol biosynthetic pathway in higher plants. For clarity, not all biosynthetic steps are shown and only the HPPDase reaction is shown in detail. The CO₂ lost and molecular oxygen introduced by HPPDase are indicated with a larger font and asterisks, respectively. The conjugated rings of HPP and HGA are numbered to indicate rearrangement of the side chain. The locations of the pds1 and pds2 mutations in the pathway are indicated.

Figure 2

Nucleotide and deduced amino acid sequence of the Arabidopsis HPPDase cDNA pHPPD. The protein sequence is shown in boldface underneath the nucleotide sequence (accession no. AF000228). The nucleotide sequence of the originally identified, truncated expressed sequence tag (accession no. T20952) is indicated by a single underline. Alignments were performed to13 other HPPDase proteins (accession nos. AJ000693, D64004, L38493, U11864, U87257, S69666, M59289, M59429, Z50016, X72389, D29987, M18405, and D13390). Arabidopsis HPPDase amino acid residues showing identity in 9 of the other 13 HPPDase proteins are indicated with shaded boxes. Amino acid residues identical in all 14 HPPDase sequences are denoted with black boxes. The five conserved Tyr and His residues postulated to form the HPPDase ferric iron center are indicated by filled dots. The location of the single 107-bp intron in the HPPDase genomic sequences of Ws and pds1 is denoted by an inverted, filled triangle. The 17-bp deletion in the HPPDase gene in pds1 is denoted by a boldface, italic DNA sequence and two overhead lines. The Ws and pds1 HPPDase gene and protein sequences are identical up to the deletion. The consequence of this mutation at the protein level is indicated in the box below the deletion.

Figure 3

Expression of Arabidopsis HPPDase cDNA in E. coli. A, HPLC analysis of a HGA standard in Luria-Bertani broth is shown in the top plot. The middle and bottom plots are cell-free extracts from cultures of E. coli harboring the pET-HPPD construct and the pET15b construct, respectively. B, Absorption spectra of peaks 1 and 2 from A. Peak 1, HGA standard and co-migrating peak in medium of pET-HPPD; peak 2, unidentified compound in pET15b.