Coq6 hydroxylase: unmasked and bypassed

Summary

Coenzyme Q is a polyisoprenylated benzoquinone lipid essential in cellular energy metabolism. Ozeir et al. (2011) show that an enzyme, Coq6, is required for the coenzyme Q C5-ring hydroxylation, and that defects in Coq6 can be bypassed by providing alternate ring precursors.

Coenzyme Q (ubiquinone or Q) is an essential lipid quinone in respiratory electron transport. Q accepts electrons and protons from Complex I and Complex II, and the reduced hydroquinone (QH₂), is oxidized by Complex III. Q also serves as an essential redox cofactor in the oxidation of fatty acids, glycerol-3-phosphate, dihydroorotate, sulfide, choline, and other amino acid-derived metabolites. QH₂ functions as a lipid soluble chain breaking antioxidant, and also functions as a co-antioxidant maintaining vitamin E in its reduced state (Turunen, et al., 2004). Yet unlike vitamin E, QH₂ is synthesized de novo by human and other animal cells.

Studies with the yeasts Saccharomyces cerevisiae and Schizosaccharomyces pombe have identified many of the genes required for Q biosynthesis (Kawamukai, 2009; Tran and Clarke, 2007). However, the functional roles of several of the COQ gene products (including Coq4, Coq6, and Coq9) remain mysterious. Previous work identified the COQ6 gene as encoding a flavin-dependent hydroxylase required for synthesis of Q in yeast (Tran and Clarke, 2007). Although the yeast coq6 null mutant accumulated 3-hexaprenyl-4-hydroxybenzoic acid (HHB; Fig. 1), this same early intermediate was found to accumulate in each of the coq null mutants (coq3-coq9), and thus was not diagnostic of the blocked step (Tran and Clarke, 2007). Many of the yeast coq null mutants (including the coq6 null mutant) were shown to lack other Coq polypeptide partner proteins, due to the destabilization of a large Coq polypeptide complex required for Q biosynthesis.

Figure 1. Many paths to Q in yeast.

S. cerevisiae utilize 4-hydroxybenzoic acid (4-HB) and para-aminobenzoic acid (pABA) as ring precursors in the synthesis of Q₆H₂ (Marbois, et al., 2010; Pierrel, et al., 2010). Coq2 attaches the polyisoprenyl tail (designated as R; in S. cerevisiae hexaprenyl-diphosphate), generating 3-hexaprenyl-4-hydroxybenzoic acid (HHB) and 3-hexaprenyl-4-aminobenzoic acid (HAB). The 4-HB and pABA pathways are speculated to converge at the point of 4-amino-DMQ₆H₂ to demethoxy-Q₆H₂ (DMQ₆H₂) (Marbois, et al., 2010). The studies of (Ozeir, et al., 2011) show that 3-hexaprenyl-4-aminophenol (4-AP) and 3-hexaprenyl-4-hydroxyphenol (4-HP) accumulate in yeast coq6 and yah1 mutants fed pABA and 4-HB, respectively (compounds denoted by red astericks). The coq6 or yah1 defect in Q biosynthesis can be bypassed by feeding the alternate ring precursors, 3,4-dihydroxybenzoic acid (3,4-diHB) or vanillic acid (VA) (compounds denoted in green).

In this issue, Ozeir et al. (2011) identify the function of the Coq6 polypeptide as required for the C5-hydroxylation in yeast coenzyme Q biosynthesis. The authors use a combination of yeast genetics, lipid biochemistry and clever feeding experiments of alternate ring precursors to characterize the defective step in yeast coq6 mutants. The authors capitalize on the observations that several of the yeast coq null mutants harboring multi-copy COQ8 have restored steady state levels of the Coq3 and Coq4 polypeptides (Zampol, et al., 2010), and in the case of the coq7 null, accumulate DMQ₆, an intermediate just two steps removed from Q (Padilla, et al., 2009). Ozeir et al. (2011) show: (1) the yeast coq6 null mutants harboring the COQ8 gene on a multi-copy plasmid produce intermediates lacking the C5-ring hydroxyl group; and (2) yeast coq6 null mutants expressing inactive Coq6p with two amino acid substitution mutations (Coq6-G386A,N388D) also accumulate the same two intermediates lacking the C5-ring hydroxyl, namely, 3-hexaprenyl-4-aminophenol (4-AP) and 3-hexaprenyl-4-hydroxyphenol (4-HP) (Fig. 1). These intermediates are shown to accumulate when the yeast are supplied with either para-aminobenzoic acid (pABA) or 4-hydroxybenzoic acid (4-HB) as the respective aromatic ring precursors.

The authors identified 4-AP and 4-HP previously, and showed that they accumulate in two other yeast mutants deficient in either ferredoxin (Yah1) or ferredoxin reductase (Arh1) (Pierrel, et al., 2010). In the present study, the coq6, yah1, arh1 and flx1 (FLX1 encodes a mitochondrial FAD transporter) deficient yeast mutants are connected by their identical accumulation of 4-AP and 4-HB. Thus, unlike most flavin-dependent monooxygenases, the electrons from NADPH are funneled indirectly to yeast Coq6 via ferredoxin reductase and ferredoxin (Fig. 2).

Figure 2. The S. cerevisiae Coq6 monooxygenase requires an additional electron transport system.

Coq6 is shown to work in conjunction with Yah1 (ferredoxin) and Arh1 (ferredoxin reductase).

The tour de force of the study are the bypass feeding experiments, where alternate aromatic ring precursors are supplied that rescue the yah1 deficient mutant and the coq6 null mutant harboring multi-copy COQ8. These precursors include 3,4-dihydroxybenzoic acid (3,4-diHB) and vanillic acid (VA) (Fig. 1). By supplying either of these precursors, the authors demonstrate that the coq6 and yah1 deficient yeast mutants now acquire the ability to grow on non-fermentable carbon source and synthesize Q₆. These bypass experiments convincingly demonstrate that the defect in Q biosynthesis in the coq6 yeast mutant is due to the lack of C5-ring hydroxylation.

A recent report identified Q-deficiencies in patients with mutations in the human homolog of Coq6 (Heeringa, et al., 2011). Expression of the human Coq6 polypeptide in yeast coq6 null mutants was shown to rescue growth on non-fermentable carbon source and to restore synthesis of Q₆. Together with the findings of Ozeir et al. (2011), these results indicate that both human and yeast Coq6p function at the C5-ring hydroxylation step. In a prescient review discussing potential alternate pathways of coenzyme Q biosynthesis, Olson and Rudney noted that 3,4-diHB and VA could supply ring precursors for prenylation via metabolism of tyrosine and norepinephrine (Olson and Rudney, 1983). Now, as suggested by Ozeir et al. (2011), it is possible that the deficiency in Q biosynthesis in certain human patients could be corrected by administering such alternate ring precursors.

The use of pABA by human or animal cells as a metabolic precursor in Q biosynthesis is possible, but so far has not been shown experimentally. It will be interesting to determine whether the Coq6-defective patient-derived cells or cell lines accumulate 4-AP or 4-HP as Q-intermediates. It is not yet known whether 4-AP and/or 4-HP are productive Q-intermediates. If so, then not only are there multiple pathways to Q, but these pathways may be branched, with flexibility in the metabolic sequence of steps similar to that of bile acid synthesis.