FQR1, a novel primary auxin-response gene, encodes a flavin mononucleotide-binding quinone reductase

Abstract

FQR1 is a novel primary auxin-response gene that codes for a flavin mononucleotide-binding flavodoxin-like quinone reductase. Accumulation of FQR1 mRNA begins within 10 min of indole-3-acetic acid application and reaches a maximum of approximately 10-fold induction 30 min after treatment. This increase in FQR1 mRNA abundance is not diminished by the protein synthesis inhibitor cycloheximide, demonstrating that FQR1 is a primary auxin-response gene. Sequence analysis reveals that FQR1 belongs to a family of flavin mononucleotide-binding quinone reductases. Partially purified His-tagged FQR1 isolated from Escherichia coli catalyzes the transfer of electrons from NADH and NADPH to several substrates and exhibits in vitro quinone reductase activity. Overexpression of FQR1 in plants leads to increased levels of FQR1 protein and quinone reductase activity, indicating that FQR1 functions as a quinone reductase in vivo. In mammalian systems, glutathione S-transferases and quinone reductases are classified as phase II detoxification enzymes. We hypothesize that the auxin-inducible glutathione S-transferases and quinone reductases found in plants also act as detoxification enzymes, possibly to protect against auxin-induced oxidative stress.

Figure 1

Deduced amino acid sequence of FQR1 and its relationship to the WrbA gene family as aligned by ClustalW in MacVector. The sequences shown in this alignment were selected from those that have a high degree of sequence similarity with FQR1; preference was given to those sequences for which biochemical information is available. FQR1 is displayed on the top line. TvQR2 is quinone oxidoreductase from T. versicolor (accession no. AAG53945; Matvienko et al., 2001), QR is 1,4-benzoquinone reductase from P. chrysosporium (accession no. AAD21025; Akileswaran et al., 1999), WrbA is Trp repressor-binding protein from E. coli (accession no. P30849; Yang et al., 1993), p25 protein is from S. pombe (accession no. P30821; Toda et al., 1992; Turi et al., 1994), and flav. is flavodoxin from C. acetobutylicum (accession no. A38177; Santangelo et al., 1991). Sequence identity with the archeal protein (from Archaeoglobus fulgidus), not shown because of its lower level of similarity, is 32%. Identical residues are indicated with dark gray shading, and similarities are indicated with light gray shading. The N-terminal region matching the flavodoxin signature sequence is underlined.

Figure 2

FQR1 fusion protein binds a flavin. Protein was isolated from IPTG-induced BL21 (DE3) E. coli harboring the His-tagged FQR1 expression vector and purified on a column containing an Ni²⁺-charged agarose matrix. A, Flavin was dissociated from purified protein under acidic conditions, and the spectrum of the isolate (dotted line) was compared with that of a flavin (FMN) standard (solid line). B, Thin layer chromatographic analysis of FQR1-associated flavin harvested from an E. coli overexpression system. The isolated protein was dialyzed in phosphate-buffered saline and precipitated using 5% (w/v) TCA. The flavin-containing supernatant was dried, resuspended in 65% (v/v) ethanol, and loaded onto silica plates. Approximately 1 μg of each standard was added. The experiment was performed once each in two different solvent systems. The data shown here are from 3:1:1 butanol:acetic acid:water. Rf, Riboflavin.

Figure 3

FQR1 functions in electron transport. A, Absorption spectrum of FQR1 fusion protein. B, Absorption spectrum of FQR1 fusion protein from A to which NADH has been added. The loss of A₄₅₀ is indicative of flavin reduction. C, Absorption spectrum of FQR1 fusion protein-NADH mixture from B to which the electron acceptor DCPIP has been added. At this point, all three components are present at 50 μm. D, Absorption spectrum of FQR1 fusion protein mixed with DCPIP. The high absorbance in the blue spectrum is indicative of the oxidized state of DCPIP. E, Absorption spectrum of FQR1 fusion protein mixed with DCPIP from D to which NADPH has been added. At this point, all three components are present at 50 μm.

Figure 4

FQR1 has QR activity. A, Specific activity of FQR1 as determined by the formation of reduced MTT in the QR assay. Numbers for the first four samples reflect the averages of two trials. Protein concentrations for each trial were determined by averaging the results of triplicate Bradford assays. B, Coomassie Brilliant Blue-stained gel showing the relative amount of FQR1 present in the E. coli extracts used in A. The additional lane on the far left contains M_r markers. The arrow indicates the position of the FQR1 fusion protein.

Figure 5

Northern blots were hybridized with probe made from the 550-bp fragment of the FQR1 cDNA. Total RNA (10 μg) was loaded per lane. A, Increase in FQR1 mRNA accumulation with time after auxin treatment. RNA was isolated from cultured roots to which 30 μm IAA was added at t = 0. B, Incubating roots with cycloheximide leads to increased levels of FQR1 mRNA. RNA was extracted from root cultures treated with 10 μm cycloheximide and/or 10 μm IAA for 2 h. “C,” Control; “CHX,” cycloheximide. C, Effect of increasing concentrations of IAA on accumulation of FQR1 mRNA. RNA was isolated from cultured roots 1 h after IAA application. D, Effect of multiple auxin and non-auxin compounds. Hormones were applied to cultured roots for 2 h at a concentration of 30 μm. “C,” Control. rRNA bands from the ethidium bromide-stained gels, shown below each northern blot, served as loading controls.

Figure 6

Western blot showing that anti-FQR1 antibody recognizes both fusion and native proteins. Lanes 1, 2, and 3: 10, 5, and 1 ng of the FQR1 fusion protein, respectively; lanes 4, 5, and 6, 8 μg of total protein extract from wild-type shoots, roots, and root cultures. As expected, the molecular mass of the fusion protein is approximately 5 kD more than the native protein. The blot was incubated with a 1:10,000 dilution of immune serum. Incubation with preimmune serum did not reveal any bands. The lower arrow indicates the native protein. The positions of comparatively faint bands of larger molecular mass are indicated with the upper arrows.

Figure 7

Levels of FQR1 mRNA, FQR1 protein, and QR activity in untreated plants transformed with CaMV35S::FQR1 as compared with wild type. All five of the transformed lines (C1K4, B1C2, C1C3, CIB6, and B1B1) carry the same construct. A, Northern blot probed with FQR1 cDNA. Samples of leaf and stem tissue were taken from 21-d-old plants. RNA (10 μg) was loaded per lane. rRNA bands shown below the northern blot served as loading controls. B, Western blot shows that FQR1 protein levels correlate with FQR1 mRNA abundance. Lane 1, 10 ng of His-tagged FQR1 purified from E. coli. Lane 2 to 7, 8 μg of protein extract from 2-week-old soil-grown plants. The Coomassie Brilliant Blue-stained gel, shown below the western blot, was loaded with equivalent amounts of the samples used in the western blot and served as a loading control. C, Specific activity of extracts from CIB6 transgenic plants (n = 5 isolations) is greater than that of wild-type (Col-0) plants (n = 6). The error bars represent sds. Results are statistically significant at P = 0.01.