Hydroxylation of demethoxy-Q₆ constitutes a control point in yeast coenzyme Q₆ biosynthesis

Yeast strains and growth media

Yeast strains used in this study are shown in Table 1. Growth media for yeast was prepared as described [33] and included YPD (1% yeast extract, 2% peptone, 2% dextrose), YPD10 (YPD containing 10% glucose instead of 2%), YPG (1% yeast extract, 2% peptone, 3% glycerol), SDC (0.18% yeast nitrogen base without amino acids, 2% dextrose, 0.14% NaH₂PO₄, 0.5% (NH₄)₂SO₄, and complete amino acid supplement), and SD-Ura (SDC, but amino acid supplement lacks uracil). Yeasts were grown at 30 °C with shaking (200 rpm).

Plasmid pMCOQ4 was created by digesting the plasmid pTL4 [15] with BamHI and ligating the fragment to pRS426 that had been similarly treated. Plasmid pMABC1 was created by digesting the plasmid p3HN4 [16] with SalI and NotI and ligating the fragment with pRS426 that had been similarly treated. pMCOQ4 and pMABC1 plasmids were used to transform the CEN.MP3–1A yeast strain using the lithium acetate-PEG method [34].

Oxidative stress treatment

Yeast cells were cultured in YPD10 overnight (OD_600nm = 9–10), washed and resuspended in sterile 100 mM sodium phosphate buffer, pH 6.2, with 0.2% glucose, at 180 × 10⁶ cells/ml. Cell suspensions were incubated with 1 mM H₂O₂, 750 μM oleic acid, or 750 μM linolenic acid for four hours.

In vivo labeling of Q₆ and Q₆-intermediates

p-[U-¹⁴C]-hydroxybenzoic acid ([¹⁴C]-pHB) was synthesized from L-[U-¹⁴C]-tyrosine as previously described [3]. Yeast strains were grown in YPD at 30 °C with shaking (200 rpm) for 48 h in the presence of [¹⁴C]-pHB (17 nM or 10^{5 14}C-dpm/ml of medium).

Identification of Q₆ and Q₆-intermediates by HPLC/ECD

Lipid extractions of whole cells and quantification of Q₆ and Q₆-intermediates were performed as described [35]. Coenzyme Q₉ was used as an internal standard; 1 μg Q₉ was added per g wet weight cell pellet. Lipid components were separated with a Beckmann 166–126 HPLC system equipped with a 15 cm Kromasil C-18 column in a column oven set to 40 °C, with a flow rate of 1 ml/min and a mobile phase containing 88/24/10 methanol/ethanol/2-propanol and 13.4 mM lithium perchlorate. Quinones were quantified with an ESA Coulochem III electrochemical detector (ECD) and a 5010 analytical cell (E1, −500mV; E2, +500mV). Hydroquinones present in samples were oxidized with a pre-column cell set in oxidizing mode (E +500mV). Q₆ and Q₉ were quantified with external standards.

RPHPLC-APCI-MS/MS multiple reaction monitoring analysis

One gram wet weight pellets of yeast strains as indicated were harvested at 44 hours (14–26 OD/ml) harvested from YPgal media: (2% galactose, 0.1% dextrose) and collected by centrifugation at 1000 × g in 50 ml tared borosilicate glass conical tubes with Teflon-lined caps. These were stored at −20 °C until extraction. Samples were thawed on ice, and 183 pmol of internal standard was added (Q₄, Sigma) to samples and standards. This was followed by 1 ml water and 16 ml hexane/2-propanol (3/2, v/v) [36]. After shaking at 200 rpm for 10 minutes the samples were vortexed vigorously for ten seconds and the phases were allowed to separate while standing at room temperature. The upper phase was removed to a 50 ml amber glass vial. The lower phase was re-extracted with hexane:2-propanol; 7:2 (8 ml) twice and the pooled organic phases were concentrated by vacuum centrifugation and stored dry at −20 °C. The four point standard curve (7.8 fmol/μl to 1.01 pmol/μl) was extracted simultaneously with the cell pellet samples.

Dried samples and standards organic extracts were resuspended in 400 μl neat ethanol. These crude lipids were separated (50 μl/injection) by a reverse phase column Thermo Hypersil (Keystone Scientific Operations) BetaBasic 18, 4.6 × 250 mm, 5 μ particle size) equilibrated in water/acetonitrile (50/50 (v/v); solvent A) and eluted (1 ml/min μl/m) with an increasing concentration of 100% acetonitrile; solvent B (min/% acetonitrile: 0/90; 3/90; 7/100; 30/100). The eluate was passed directly into an atmospheric pressure chemical ionization source (APCI, nebulizing gas, “zero”-grade air produced by a Zero Air Generator (Peak Scientific, Chicago, Il) at 2–3 L/s, nebulizer 450 °C, orifice 50 volts) attached to a Perkin-Elmer Sciex (Thornhill, Canada) API III triple quadrupole mass spectrometer operated in the tandem mass spectrometric (MS/MS) mode (99.999% argon collision gas at an instrumental GCT setting of 200). The mass spectrometer had been previously tuned and calibrated by flow injection (100 μl/min) of a mixture of polypropylene glycol (PPG) 425, 1000 and 2000 (3.3 × 10−⁵, 1 × 10−⁴ and 2 × 10−⁴ M, respectively) in water/methanol (1/1, v/v) containing 2 mM ammonium formate and 0.1% acetonitrile. Calibration across the mass range was made using the singly charged PPG/NH₄⁺ ions at m/z 58.99, 326.25, 906.67, 1254.92, 1545.13 and 1863.34 with instrument resolution set so the ¹³C isotope satellites were resolved with 10–40% valley.

Quantitative data was collected in the multiple reaction monitoring (MRM) mode by recording the MH⁺ → tropylium ion transitions (Q₆, m/z 591.4 → 197.0; Q₄, m/z 455.2 → 197.0 and DMQ₆, m/z 561.4 → 167.0) with dwell set to 1000 ms, a pause time of 0.3 ms and a total scan time of 4.0 seconds. Standards (DMQ₃ and Q₃) were fragmented and used to confirm the identity of the ions. Synthesis of DMQ₃ was as described [22].

MRM peak areas were measured by instrument manufacturer supplied software (MacSpec^™, version 3.3, PE Sciex, Ontario, Canada). The calibration curve from the extracted set of standards (ordinate, Q₆/Q₄ peak area ratio; abscissa, Q₆ amount) had correlation coefficient (r²) of 0.99 with slope = 4.68 (intercept forced to zero). Concentrations of Q₆ and DMQ₆ (in Q₆ equivalents) present in the extracts were directly interpolated from the calibration curve, and quantities in each fraction were calculated after correction for dilution and amount assayed. Lipid extractions and analyses were completed on independent replicate yeast pellets and were injected twice. The data presented represent independent replicate cell pellets and duplicate LC/MS-MS analyses.

RNA isolation and cDNA synthesis

Cells (200 × 10⁶) were harvested by centrifugation and washed twice in DEPC-treated water. Cells were resuspended in 200 μl 20 mM Tris-HCl, pH 7.4, with 30% glycerol, frozen in N₂(l) and stored at −80 °C until use. For total RNA isolation, cells were thawed and washed with DEPC-treated water. Each cell pellet was resuspended in 500 μl cell-wall-digestion buffer (50 mM potassium phosphate, pH 7.4, 10 mM DTT) and incubated 45 min at 30 °C in the presence of 10 μl Zymolyase 20T, 10 mg/ml (Seikagaku Corporation, Japan). Spheroplasts were collected by centrifugation and total RNA was isolated (Eppendorf Perfect RNA Eukaryotic Mini kit) and quantified with a Nanodrop ND-1000 spectrophotometer (NanoDrop Technologies, Wilmington, DL USA). cDNA was synthesized from 1 μg of total RNA in a 20 μl reaction containing Moloney Murine Leukemia Virus (M-MuLV) reverse transcriptase buffer, 40 U M-MuLV, 1 mM dNTP mix, 20 U RNase inhibitor (Fermentas Life Sciences, Vilnius, Lithuania) and 500 ng oligo(dT) from MWG, Ebersberg, Germany. Each reaction was performed for 1 h at 42 °C according to manufacturer’s directions.

Primers and DNA competitors

Primers were designed with Primer Premier 5.00 software (Premier Biosoft International, Palo Alto, CA, USA) and the available sequence information from SGD yeast database (Table 2). For each gene analyzed, a competitor fragment was generated with recombinant PCR. The competitor fragment was amplified from genomic DNA [33] with the competitor primers (i.e. ABC1-ComF and ABC1-ComR). These primers contain the 5′ and 3′ specific ends of the sequence used as competitor and a tail corresponding to the sequence of primers used to amplify the target gene (i.e. ABC1-U210 and ABC1-L1260). The competitor was cloned in a T-hang vector such as pMBL (Dominion-MBL SL, Córdoba, Spain) and sequenced to confirm the construct (MWG, Ebersberg, Germany). The target gene fragment was also cloned in the same vector. For analyses of the COQ genes, the ACT1 gene was used as a competitor. For analyses of the ACT1 gene (used as calibrator) the COQ5 gene was used as competitor. The competitor required for expression analyses was amplified by PCR with the appropriate target primers, extracted from agarose (Eppendorf Perfect Gel Cleanup), and quantified by absorbance spectrophotometry. The designated competitor DNA products were prepared and purified freshly for each set of experiments.

Amplification kinetics and determination of the optimal concentration of the competitor DNAs

To test the quality of competitor, each target-competitor pair (0.1 attomol) were used as template for the PCR in the presence of target primers. Samples were removed at the specified cycle number (from cycles numbered 23 to 30), and resolved by agarose gel electrophoresis (Fig. 2A). Both competitor and target specific DNA fragments were quantified with a gel analyzer (Gel Doc 1000, Bio-Rad Laboratories, Hercules, CA, USA). Data obtained were subjected to linear regression analysis validating the effectiveness of each competitor (Fig. 2B). In order to determine the optimal amount of competitor 10-fold serial dilutions of competitor (100–0.1 attomol) were amplified together with constant aliquots (2 μl) of cDNA from the wild-type strain CEN.PK2-1C with target primers and PCR samples were resolved by agarose gel electrophoresis (Fig. 2C). Band intensities were quantified and expressed as arbitrary units (AU). After a correction depending of the fragment size, the base-10 logarithm of the target/competitor ratio was calculated. The ratios were plotted against the logarithm of the amount of competitor added (Fig. 2D). This plot was used to calculate the equivalence point, the point in which the logarithm of the ratio target/competitor is equal to 0 (target concentration equal to competitor concentration). The measured concentration of the competitor determined the concentration to be used in the expression analysis experiments.

Competitive RT-PCR expression analyses

Samples of yeast cDNA (2 μl from a 20 μl reaction containing 1 μg of total yeast RNA) were amplified together with constant amounts of competitor in a volume of 25 μl. PCR reaction mixtures were resolved by agarose gel electrophoresis and the band intensities were used to calculate the logarithm of the ratio target/competitor. Differences in these ratios indicate differences in expression. Competitive RT-PCR expression method was performed as was described with minor modifications [37, 38].

DMQ₆ is a predominant Q₆-intermediate in log phase yeast cultures

To study the regulation of Q₆ biosynthesis in yeast, we first analyzed the concentrations of both DMQ₆ and Q₆ throughout the growth in both non-fermentable (YPG) (Fig. 3A) and fermentable (YPD) media (Fig. 3B). The Q₆/DMQ₆ ratio was significantly decreased (from 4.0 to 0.5) during the first 10 h culture in YPG culture, due to the transient accumulation of DMQ₆ (Fig. 3A and 3C). The amount of DMQ₆ was surprisingly persistent; even after 24 and 48 hours of culture the content of DMQ₆ remained high. Yeast grown in YPD showed also a transient increase in DMQ₆ content; the Q₆/DMQ₆ ratio decreased from 4.4 to 2.0 over the first 10 h. This trend was then reversed by the increase in Q₆ content yielding a ratio of 5.0 by 24 h (Fig. 3B).

Changes in COQ RNA levels as a function of growth phase and culture

Results showed above suggest the existence of a biochemical mechanism through yeast cells could regulate Q₆ biosynthesis by modulation of DMQ₆ to Q₆ transformation rate. To further investigate this process mRNA levels of several genes involved in Q₆ biosynthesis pathway (COQ genes) were determined. Seven of the nine COQ genes products in yeast (Coq3-Coq9) are required for ring modification of Q₆ molecule [20, 21, 23] and Coq7p is the enzyme involved in DMQ₆ hydroxylation (Fig. 1). Therefore, mRNA levels of COQ3, COQ4, COQ5, COQ7 and ABC1 genes were measured in yeast grown in YPG and YPD media (Fig. 3A and 3B). In general COQ RNA levels are higher in yeast grown in YPG (Fig. 4A). COQ7, ABC1/COQ8, COQ5, and COQ3 RNA levels tend to increase (from 2- to 7-fold) with time in culture. Changes in COQ RNA levels as a function of growth phase in YPD media are fairly modest (Fig. 4B), however, COQ7 RNA levels were significantly elevated as cells transit from log to stationary phase. These results suggest that there are at least two phases in Q₆ biosynthesis; one leading to DMQ₆ biosynthesis driven by a slow increase of the expression of COQ genes, and a second phase, driven by the expression of COQ7, where the final ring hydroxylation and methylation steps produce Q₆.

The conversion of DMQ₆ to Q₆ responds to oxidative stress

Previous studies have demonstrated antioxidant properties of coenzyme Q. Yeast strains that are unable to synthesize Q₆ or producing only DMQ₆ are very sensitive to oxidative stress induced by H₂O₂ and linolenic acid [32, 39]. Therefore, yeast cells under oxidative stress conditions could potentially activate Q₆ synthesis in order to increase protection from oxidative damage. To test this hypothesis, in yeast subjected to oxidative stress we have analyzed the content of Q₆ and DMQ₆ in the wild-type CEN.PK2–1C yeast strain treated with H₂O₂, oleic acid, or linolenic acid. This wild-type strain was highly resistant to treatment with either 1 mM H₂O₂ or 750 μM linolenic acid [32].

Figure 5A shows that the Q₆/DMQ₆ ratio was increased (from 0.3 to 0.8) in yeast treated with linolenic acid for 4 h. That increase was due mainly to a shift of DMQ₆ intermediate to Q₆, because total amount of quinones remained unchanged. H₂O₂ did not affect the Q₆/DMQ₆ ratio and even produced a decrease in total quinone content. It is likely that the protective response elicited by H₂O₂ treatment is distinct from that elicited by linolenic acid. Oleic acid was also tested because its provision as a carbon source has been demonstrated to increase the expression of COQ5. The promoter of the COQ5 gene contains an oleic acid response element (ORE) that may boost Q₆ synthesis [40]. However, when provided at the same concentration as linolenic acid, oleic acid treatment does not produce significant changes in the Q₆ and DMQ₆ levels.

To investigate whether the oxidative stress treatments affected COQ gene expression, we analyzed the steady state COQ RNA levels. Linolenic acid, but neither H₂O₂ nor oleic acid, increased both COQ7 and COQ3 RNA levels following a 4 h treatment relative to control (Fig. 5B). Other COQ RNA levels examined did not change significantly, including COQ5, in response to oleic acid treatment (Fig. 5B). These results suggest that treatment with linolenic acid acts to increase COQ7 and COQ3 RNA levels, and acts to enhance the conversion of DMQ₆ to Q₆ as a mechanism to provide antioxidant protection.

DMQ₆ can be synthesized in yeast in the absence of Coq7p

Yeast strains harboring null mutations in one of the COQ genes (COQ3 to COQ9) accumulate HHB (3-hexaprenyl-4-hydroxybenzoic acid), an early Q-intermediate that is the product of the reaction carried out by Coq2p [13, 26, 41, 42]. Several lines of physical evidence now indicate the existence of a biosynthetic Coq-polypeptide complex inside mitochondria [22]. Supplementation of growth media with exogenous Q₆ has been shown to rescue mitochondrial respiration in the yeast coq mutants [43] and to preserve steady state levels of Coq3 and Coq4 polypeptides [27] of a coq7 null mutant. In addition, it has been shown that Q₆ and DMQ₆ are associated with a polypeptide complex defined by Coq3p and Coq4p products [22]. This latter observation suggests that interaction between Q₆ and COQ genes products could be important for the biosynthetic complex stabilization and regulation of Q₆ biosynthesis. To determine whether the functional interaction among Coq polypeptides depends on the presence of Q₆, we analyzed the effect of addition of exogenous Q₆ on coq null mutants growing in YPD. The analysis of lipid extracts from each of the coq mutants (coq3-abc1/coq8) indicated that a product co-eluting with DMQ₆ was produced only in the coq7 mutant strain cultured in the presence of exogenous Q₆ (Fig. S3).

To identify the material co-eluting with DMQ₆, we performed RP-HPLC/MS-MS analysis of the lipid extracts prepared from the wild-type parental strain and the coq7 null mutant cultured in the presence and absence of exogenously added Q₆. As described in experimental procedures, quantitative data was collected in the multiple reaction monitoring (MRM) mode. This method separates lipid extracts, producing specific retention times for the compounds of interest; the detection of the lipids involves selection of a specific precursor (parent) molecule, which is fragmented for quantification of a specific product (daughter) molecule. By these criteria the Q₆ and DMQ₆ content in the CEN.PK2–1C wild-type yeast were 62.73 ± 15.13 and 4.22 ± 1.04 pmol/gm wet weight, respectively. Neither Q₆ nor DMQ₆ were present at measurable amounts in the coq7 null yeast strain. However, when cultured in the presence of exogenous Q₆, the coq7 null mutant contained quantities of both Q₆ (8.05 ± 3.16 pmol/gm wet weight) as well as DMQ₆ (0.24 ± 0.10 pmol/gm wet weight). To determine whether the DMQ₆ produced by the coq7 mutant represented de novo synthesis, several of the coq null mutant strains were grown in the presence of [¹⁴C]-pHB, the aromatic ring precursor, and in the presence or absence of exogenously added Q₆. Supplementation of the growth media with exogenously added Q₆ did not affect the incorporation of [¹⁴C]-pHB into either DMQ₆ or Q₆ in the wild-type strain (Fig. S4). Similar incubations of the coq3 and abc1 null mutant strains showed that neither produced detectable DMQ₆ or Q₆, and each produced only a more polar radiolabeled peak eluting at about 6 min, as predicted for HHB (Fig. 6A and C). However, a significant radiolabeled peak corresponding to [¹⁴C]-DMQ₆ was observed in lipid extracts prepared from the coq7 null mutant strain following incubation with [¹⁴C]-pHB plus exogenous Q₆ (Fig. 6B). Because growth of the coq7 null mutant in the presence of exogenous Q₆ leads to uptake of Q₆ [44], these results indicate that Q₆ itself supports interaction of the Coq proteins and enables de novo synthesis of DMQ₆. Thus, Q₆ is a crucial lipid component of the Coq polypeptide biosynthetic complex, and its presence may be required to facilitate formation of, and/or stabilize, the Coq polypeptide biosynthetic complex. COQ RNA levels were quantified in the CEN MP3.1A (coq7 null) strain in response to culture in the presence or absence of exogenous Q₆ (Fig. 7). Addition of exogenous Q₆ did not produce a change in COQ3, COQ4, COQ5, or ABC1/COQ8 RNA levels, indicating that the de novo synthesis of DMQ₆ is not driven by changes in RNA levels. Recent studies show that exogenously added Q₆ to coq7 null mutant cultures enhanced the steady state levels of the Coq3 and Coq4 polypeptides [27]. Hence, it is likely that the addition of Q₆ acts to stabilize the Coq polypeptides.

Abc1p and Coq4p regulate DMQ₆ synthesis

To further understand whether there are more factors involved in the stabilization of Coq polypeptides, we analyzed whether an alteration of the expression of certain Coq proteins would lead to complex stabilization and DMQ₆ synthesis in a coq7 null mutant strain. Of the COQ genes participating in the synthesis of DMQ₆, Coq4p and Abc1/Coq8p do not have an assigned function, although Abc1/Coq8p has recognized sequence motifs that identify it as a potential Hank’s Ser/Thr protein kinase [17, 45]. To study the possible functions of Coq4p and Abc1/Coq8 in DMQ₆ biosynthesis, the yeast coq7 null mutant strain was transformed with a multicopy plasmid harboring either COQ4 or ABC1 coding sequence including the natural promoter. The over-expression of COQ4 or ABC1 in the coq7 null mutant does not produce significant changes in the expression of other COQ genes (Fig. 8A). Lipid extracts of these cells were analyzed for content of Q₆ and DMQ₆ by HPLC-ECD. Multicopy expression of COQ4 produces the synthesis of a small amount of DMQ₆ as compared with the coq7 null mutant strain; however, the multicopy expression of ABC1/COQ8 increases dramatically the amount of DMQ₆, to levels much higher than detected in wild-type cells (Fig. 8B). These results support a regulatory function for Abc1/Coq8p; and indicate that over-expression of Abc1/Coq8p can facilitate the formation of a Coq7p-independent complex for DMQ₆ biosynthesis in S. cerevisiae.