Coenzyme Q (CoQ), ubiquinone or 2,3-dimethoxy-5-methyl-6-polyprenyl-1,4-benzoquinone is a two-part molecule composed of a benzoquinone ring, which has redox active sites, and a long polyisoprenoid lipid chain that positions the molecule in the mid-plane of a membrane bilayer. The length of the chain depends on the species, i.e., 10 isoprene units (CoQ10) in humans and S. pombe, eight units (CoQ8) in Escherichia coli, six units (CoQ6) in Saccharomyces cerevisiae and nine units (CoQ9) in rodents. Nevertheless, some species have more than one CoQ form, e.g., human cells and tissues also contain CoQ9 and rodent cells and tissues also contain CoQ10 as minor forms.
The synthesis of CoQ in eukaryotes mainly occurs in mitochondria by a set of nuclear-encoded CoQ proteins through a biochemical pathway that has not been fully defined. Moreover, some authors have suggested that CoQ may also be produced outside of mitochondria, mainly in the endoplasmic reticulum-Golgi
[1,2][1][2]. On the other hand, recent studies suggest that the endoplasmic reticulum mitochondria encounter structure (ERMES) might promote CoQ biosynthesis and the movement of CoQ and its precursors and intermediates between these organelles
[3]. Thus, the mitochondrial production of CoQ could contribute to extramitochondrial pools of CoQ. The known steps required for CoQ production were first identified in
S. cerevisiae and later confirmed in mammals. Specifically,
S. cerevisiae CoQ (CoQ1–CoQ9) mutants, but also
E. coli ubi (ubiA-J and ubiX) mutants, have allowed identifying the CoQ genes and biosynthetic intermediaries of the biosynthetic pathway of CoQ
[4,5,6,7,8,9][4][5][6][7][8][9]. At least 11 genes (ubiA, B, C, D, E, F, G, H, I, J, and X) in
E. coli and 13 genes (YAH1, ARH1, and CoQ1–CoQ11) in
S. cerevisiae are involved in CoQ biosynthesis. Human and mouse orthologues for almost all of these genes have already been identified (with the possible exception of
CoQ11). In the cases of
CoQ8 and
CoQ10, humans have two orthologues for each one:
ADCK3 (
CoQ8A) and
ADCK4 (
CoQ8B), and
CoQ10A and
CoQ10B, respectively. In
Homo sapiens the polypernyl diphosphate synthase is a heterotetramer formed by the dimer of PDSS1 and PDSS2 proteins
[10,11][10][11]. The expression of the human homologs (
CoQ1–CoQ10) for yeast genes restores CoQ
6 biosynthesis in the corresponding
S. cerevisiae CoQ mutants, indicating the profound functional conservation in human cells
[12]. Structurally, CoQ biosynthesis could be divided into four stages: quinone synthesis, isoprenoid tail synthesis, both molecules condensation and ring modification.
1.1. Quinone Synthesis
The primary aromatic precursor of the benzoquinone ring is 4-hydroxybenzoic acid (4HB). For the biosynthesis of 4HB, human cells utilize phenylalanine or tyrosine as ring precursors. Many of the steps involved in the generation of 4HB from tyrosine are not completely understood and the responsible enzymes are unidentified
[13,14][13][14]. Current studies have identified the first and last reactions of this pathway and the enzymes involved in these reactions in yeast. Thus, 4HB synthesis starts with the deamination of tyrosine to 4-hydroxyphenylpyruvate by Aro8 and Aro9 and finishes with the oxidation of 4-hydroxybenzaldehyde to 4HB by Hfd1. Human Hdf1 homolog ALDH3A1 plays a role in CoQ biosynthesis and is able to rescue CoQ deficiency in yeast with the inactivated
HFD1 gene
[15,16][15][16]. Aro8 and Aro9 human homolog remain unidentified, but α-aminoadipate aminotransferase (AADAT) and the tyrosine aminotransferase (TAT) have been postulated as candidates
[17].
Although 4HB is depicted as the main precursor of CoQ, other aromatic ring precursors may be incorporated into CoQ biosynthesis. The 2,4-dihydroxybenzoic acid (2,4-diHB), 3,4-dihydroxybenzoic acid (3,4-diHB) and vanillic acid are analogs of 4HB that may allow the stimulation and/or bypass of CoQ biosynthesis defects in yeast
CoQ6 and
CoQ7 mutants, as well as in
CoQ7 conditional knockout mice,
CoQ9R239X mice and human skin fibroblasts from patients with mutations in
CoQ7 and
CoQ9 [18,19,20,21,22,23][18][19][20][21][22][23]. Para-coumarate and resveratrol also can be used as CoQ ring precursors in
E. coli,
S. cerevisiae, and human and mouse cells
[24]. Moreover, in mouse and human kidney cells, kaempferol is able to increase CoQ levels. This phenolic compound apparently acts as a biosynthetic ring precursor competing with 4HB
[25].
1.4. Next Steps of CoQ Biosynthesis-Ring Modifications
Then, 3-decaprenyl-4HB or 3-nonaprenyl-4HB undergoes subsequent modifications of the ring through a set or reactions of methylations, decarboxylation and hydroxylations to finally generate the molecule of CoQ10 or CoQ9.
The first reaction is the C5-hydroxylation, catalyzed by CoQ6, to produce 3-decaprenyl-4,5-dihydroxybenzoic acid and 3-nonaprenyl-4,5-dihydroxybenzoic acid. In yeast, this step requires two enzymes, ferredoxin Yah1p and ferredoxin reductase Arh1p, which provide electrons for CoQ6 activity. It is unknown whether the mammalian orthologs, ferredoxin reductase (FDXR) and ferredoxin 2 (FDX2), respectively, act in mammalian CoQ biosynthesis. CoQ6 is characterized as a flavin-dependent monooxygenase peripherally associated with inner mitochondrial membrane on the matrix face
[26,27][26][27].
C5-hydroxylation is followed by O-methylation catalyzed by CoQ3 to form 3-decaprenyl-4-hydroxy-5-methoxybenzoic acid and 3-nonaprenyl-4-hydroxy-5-methoxybenzoic acid. CoQ3 is an S-adenosylmethionine-dependent methyltransferase that is involved in two O-methylation reactions in the CoQ biosynthetic pathway. This polypeptide is also peripherally associated with the mitochondrial inner membrane on the matrix face
[28,29][28][29].
The proteins involved in the subsequent steps, C1-decarboxylation and C1-hydroxylation, have not been identified in the eukaryotes. However, in prokaryotes the decarboxylation and hydroxylation in prokaryotes are catalyzed by the 3-Octaprenyl-4-hydroxybenzoate decarboxylase UbiD and the 2-octaprenyl-6-methoxyphenol hydroxylase UbiH, respectively
[30,31][30][31]. In prokaryotic decarboxylation, the flavin prenyltransferase UbiX is important for synthesizing prenylated flavin, which is used as a cofactor for UbiD
[32]. Nevertheless, there is no sequence homolog for UbiD or UbiX in humans, suggesting that the C1-hydroxylation could be part of the decarboxylation mechanism.
The ring is then further modified by C2-methylation in a reaction catalyzed by CoQ5. In this reaction, 2-decaprenyl-6-methoxy-1,4-benzenediol and 2-nonaprenyl-6-methoxy-1,4-benzenediol is converted to 2-decaprenyl-3methyl-6-methoxy-1,4-benzenediol and 2-nonaprenyl-3methyl-6-methoxy-1,4-benzenediol (DMQH
2)
[33]. CoQ5 is an S-adenosyl methionine (SAM)-dependent methyltransferase (SAM-MTase) that is peripherally associated with the mitochondrial inner membrane on the matrix face
[34]. DMQH
2 is converted to 2-decaprenyl-3methyl-6methoxy-1,4,5-benzenetriol and 2-nonaprenyl-3methyl-6methoxy-1,4,5-benzenetriol (DMeQH
2) in a C6-hydroxylation catalyzed by CoQ7, which is a carboxylate bridged diiron hydroxylase peripherally associated with the mitochondrial inner membrane on the matrix side
[35].
The final step of the CoQ biosynthetic pathway is an O-methylation also catalyzed by CoQ3.
1.5. Other Mitochondrial Proteins Involved in CoQ Biosynthesis
There are additional proteins needed for CoQ biosynthesis since their dysfunctions induce a decline in the levels of CoQ. However, their molecular exact functions are yet unclear. These are the cases of CoQ8A, CoQ8B, CoQ9, CoQ10A, CoQ10B and CoQ4.
CoQ8A (=ADCK3) and CoQ8B (=ADCK4) have been related with CoQ
10 biosynthesis and mutations in these proteins are associated with CoQ
10 deficiency
[36,37,38,39][36][37][38][39]. These two proteins are considered the human orthologues of yeast CoQ8p because CoQ
6 biosynthesis is rescued by the expression of human CoQ8A and CoQQB in CoQ8-mutant yeast strains
[39,40][39][40]. However, the precise biochemical activity and the role of CoQ8 in CoQ biosynthesis remains unknown. Although initially CoQ8 was hypothesized to be a protein kinase that may phosphorylate other CoQ biosynthetic proteins, recent studies have demonstrated that CoQ8A shows ATPase activity that is essential for CoQ biosynthesis. In any case, it seems that the activity of CoQ8A could stabilize Complex Q
[41].
CoQ9 is a lipid-binding protein associated with the inner mitochondrial membrane, on the matrix face
[42]. This protein binds aromatic isoprenes with high specificity, including CoQ intermediates that likely reside within the bilayer. Therefore, CoQ9 seems to present the intermediates of the CoQ biosynthetic pathway directly to CoQ enzymes
[43]. CoQ9 is also necessary for the stability and activity of CoQ7, so
CoQ9 mutant mice and human cells show reduced levels of CoQ7 and accumulation of DMQH
2, the substrate of the reaction catalyzed by CoQ7
[23,27,44,45,46][23][27][44][45][46].
CoQ
10A and CoQ
10B in humans
[47] are homologues of the yeast CoQ
10, a protein that contains a lipid-binding domain that binds CoQ and CoQ intermediates, and that is required for efficient biosynthesis of CoQ
6 and electron transport in the mitochondrial respiratory chain. The molecular basis for these effects is unknown, but CoQ
10 probably directs to CoQ from synthesis site to function sites
[48,49,50][48][49][50].
Finally, the function of the CoQ4 is still unknown, although some studies suggest that it is required for the assembly and stability of Complex Q, as it is associated with CoQ3, CoQ5, CoQ6 and CoQ9
[28,51,52][28][51][52]. Also, it is speculated that CoQ4 could bind to CoQ and CoQ intermediates and it could organize the enzymes that perform the ring modifications. In humans, there are two different isoforms of the CoQ4 polypeptide but only one is localized in mitochondria
[53].