Molecular Pathogenesis
The pathogenesis of primary CoQ10 deficiency is still not clear and the molecular basis of the locus heterogeneity of this group of disorders remains to be elucidated. Although the bioenergetic defect plays a crucial role in the pathophysiology of CoQ10 deficiency, CoQ10 carries out a number of fundamental functions in cells (it is a cofactor of other mitochondrial dehydrogenases, an essential antioxidant, and a modulator of apoptosis), suggesting that other mechanisms are involved.
In fact, it has been shown in cells that severe CoQ10 deficiency causes a marked reduction in ATP production without increased production of reactive oxygen species (ROS), while mild CoQ10 deficiency is associated with high ROS production without significant impairment of cellular bioenergetics [Quinzii et al 2010].
In addition, CoQ10 deficiency impairs de novo pyrimidine synthesis, further contributing to disease pathogenesis [López-Martín et al 2007].
Note: In this section the genes associated with primary CoQ10 deficiency are ordered by gene.
COQ2
Gene structure.
COQ2 consists of seven exons [Forsgren et al 2004]. The open reading frame contains four in-frame ATG initiation codons (termed ATG1-4 [López-Martín et al 2007]); the third one produces a transcript similar to yeast COQ2. Human COQ2 cDNA originating from ATG1, ATG2, and ATG3 (but not from ATG4) can complement the defective respiratory phenotype of yeast COQ2-null strains [Forsgren et al 2004, López-Martín et al 2007, Mollet et al 2007].
Note: The presence of multiple possible initiation codons has generated confusion in naming COQ2 pathogenic variants. The majority of reports consider the most 5’ ATG (ATG1) as the initiation codon and the longer transcript NM_015697.7 as reference. GeneReviews adheres to this nomenclature. However, changes to this convention are possible; it was recently proposed to transition from legacy nomenclature to nucleotide 1 corresponding to the A of ATG4 [Desbats et al 2016].
Benign variants. Multiple rare benign COQ2 variants have recently been associated with sporadic multiple-system atrophy [Mitsui et al 2013]; however, this finding is still under debate and further confirmation is needed [Mitsui et al 2013, Jeon et al 2014, Schottlaender & Houlden 2014, Sharma et al 2014].
The p.Val393Ala COQ2 variant, which is relatively common in the Japanese population, has not been found in European or North American populations.
Pathogenic variants.
COQ2 was the first gene found to be mutated in individuals with primary CoQ10 deficiency [Quinzii et al 2006]. COQ2 pathogenic variants include mainly missense alleles; truncating variants have also been reported (Table 3).
To date COQ2 pathogenic variants have been reported in ten families with primary CoQ10 deficiency [Quinzii et al 2006, Diomedi-Camassei et al 2007, Mollet et al 2007, Dinwiddie et al 2013, Jakobs et al 2013, McCarthy et al 2013, Mitsui et al 2013, Scalais et al 2013, Desbats et al 2015b, Desbats et al 2016].
Table 3.
Selected Pathogenic COQ2 Variants that Cause Primary CoQ10 Deficiency
View in own window
DNA Nucleotide Change (Alias 1) | Predicted Protein Change (Alias 1) | Reference Sequences |
---|
c.382A>G (232A>G) | p.Met128Val (Met78Val) |
NM_015697.7
NP_056512.5
|
c.437G>A | p.Ser146Asn |
c.545T>G | p.Met182Arg |
c.590G>A | p.Arg197His |
c.683A>G | p.Asn228Ser |
c.890A>G | p.Tyr297Cys |
c.905C>T | p.Ala302Val |
c.1159C>T | p.Arg387Ter |
c.1197delT (1198delT) | p.Asn401IlefsTer15 |
Variants listed in the table have been provided by the authors. GeneReviews staff have not independently verified the classification of variants.
GeneReviews follows the standard naming conventions of the Human Genome Variation Society (varnomen.hgvs.org). See Quick Reference for an explanation of nomenclature.
- 1.
Variant designation that does not conform to current naming conventions
Normal gene product.
COQ2 encodes a 421-amino acid para-hydroxybenzoate:polyprenyltransferase (NP_056512.5) required for the second step of the final reaction sequence of CoQ biosynthesis. COQ2 catalyzes the condensation of 4-hydroxybenzoate with polyprenyl-pyrophosphate, generating the first membrane-bound CoQ intermediate [Ashby et al 1992].
The COQ2 enzyme is highly conserved throughout evolution. The human protein contains a N-terminal mitochondrial leader sequence, two conserved putative substrate-binding domains (which are rich in aspartic acid residues) and six predicted trans-membrane helices [Forsgren et al 2004].
For information on yeast studies, see Coenzyme Q10 Deficiency – Model Organisms, COQ2.
Abnormal gene product. All coenzyme Q10 deficiency-related COQ2 pathogenic variants reported to date act through a loss-of-function mechanism, reducing the polyprenyl-transferase activity, as proved by the lack of complementation in yeast strains harboring deletion in the COQ2 ortholog [Mollet et al 2007] or by a reduced incorporation of radiolabeled substrates into CoQ10 [Quinzii et al 2006]. Although genotype-phenotype correlations are still unclear, most COQ2 pathogenic variants behave as hypomorphic alleles, retaining residual activity that may contribute to the phenotype.
All known COQ2 pathogenic variants affect highly conserved amino acid residues. The variant c.890A>G changes a highly conserved tyrosine to cysteine at amino acid 297 within the third predicted transmembrane domain. Variants p.Ser146Asn and p.Arg197His are located in the putative substrate-binding site (UbiA), whereas p.Asn228Ser is located in the first putative transmembrane domain.
COQ4
Gene structure.
COQ4 spans a region of about 12 kb and has two transcript variants (details in Table A, Gene, COQ4). The longer transcript NM_016035.4 has seven exons. An alternate transcript (NM_001305942.1) is shorter and has four exons.
COQ4 is ubiquitously expressed, with higher levels in liver, lung, and pancreas [Casarin et al 2008].
Pathogenic variants.
COQ4 pathogenic variants have been reported in eleven affected individuals from eight unrelated families [Brea-Calvo et al 2015, Chung et al 2015].
A patient with haploinsufficiency of COQ4 due to a de novo 3.9-Mb deletion of chromosome 9q34 and documented CoQ10 deficiency in fibroblasts had encephalomyopathic manifestations [Salviati et al 2012].
Table 4.
Selected Pathogeneic COQ4 Variants that Cause Primary CoQ10 Deficiency
View in own window
DNA Nucleotide Change | Predicted Protein Change | Reference Sequences |
---|
c.155T>C | p.Leu52Ser |
NM_016035.4
NP_057119.2
|
c.190C>T | p.Pro64Ser |
c.421C>T | p.Arg141Ter |
c.433C>G | p.Arg145Gly |
c.521_523delCCA | p.Thr174del |
c.718C>T | p.Arg240Cys |
Variants listed in the table have been provided by the authors. GeneReviews staff have not independently verified the classification of variants.
GeneReviews follows the standard naming conventions of the Human Genome Variation Society (varnomen.hgvs.org). See Quick Reference for an explanation of nomenclature.
Normal gene product.
COQ4 transcript NM_016035.4 encodes coq4 isoform 1, which consists of 265 amino acids, localizes to mitochondria, and is required for CoQ10 biosynthesis since it efficiently restores both growth in glycerol and CoQ content when expressed in a COQ4-null yeast strain.
An alternate transcript (NM_001305942.1) encodes coq4 isoform 2, which has 100 amino acids and unknown function; it lacks the first 24 amino acids that specify the predicted mitochondrial targeting sequence [Casarin et al 2008].
The precise function of ubiquinone biosynthesis protein COQ4 in CoQ10 biosynthesis is still unknown: the protein lacks enzymatic activity but in yeast it is thought to organize proteins encoded by other genes involved in the synthesis of CoQ10 into a multi-enzymatic complex [Marbois et al 2009].
Abnormal gene product. Missense COQ4 pathogenic variants expressed in yeast failed to complement a COQ4null yeast strain [Brea-Calvo et al 2015].
COQ6
Gene structure.
COQ6 transcript variant 1 has 12 exons.
Among the 18 putative isoforms resulting from alternative splicing, two full-length transcript variants NM_182476.2 and NM_182480.2 (designated transcript variants 1 and 2, respectively) were found to be expressed in several tissues including kidney; however, the longer transcript variant 1 is more abundant than variant 2. The two isoforms differ in the use of alternative exon 1a or 1b and the splicing of exon 3 (absent in isoform b) [Heeringa et al 2011, Doimo et al 2014] (see details in Table A, Gene, COQ6).
Pathogenic variants. Two homozygous pathogenic missense variants, c.763G>A and c.1058C>A, and two heterozygous pathogenic nonsense variants, c.1341G>A and c.1383delG, were found in four different families with steroid-resistant nephrotic syndrome (SRNS) [Heeringa et al 2011].
Variant c.763G>A was found in one family from northern Lebanon and one from southern Turkey, suggesting a possible founder effect [Heeringa et al 2011].
Two pathogenic nonsense variants, c.484C>T and c.564G>A, were found as single heterozygous pathogenic variants in two individuals with cyclosporine A-dependent nephrotic syndrome and diffuse mesangial sclerosis, respectively [Heeringa et al 2011].
The pathogenic missense variant c.1235A>G was found in the heterozygous state in another individual with SRNS [Doimo et al 2014].
Table 5.
Selected Pathogenic COQ6 Variants that Cause Primary CoQ10 Deficiency
View in own window
DNA Nucleotide Change | Predicted Protein Change (Alias 1) | Reference Sequences |
---|
c.484C>T | p.Arg162Ter |
NM_182476.2
NP_872282.1
|
c.564G>A | p.Trp188Ter |
c.763G>A | p.Gly255Arg |
c.1058C>A | p.Ala353Asp |
c.1235A>G | p.Tyr412Cys |
c.1341G>A | p.Trp447Ter |
c.1383delG | p.Ile462LeufsTer18 (Gln461fsTer478) |
Variants listed in the table have been provided by the authors. GeneReviews staff have not independently verified the classification of variants.
GeneReviews follows the standard naming conventions of the Human Genome Variation Society (varnomen.hgvs.org). See Quick Reference for an explanation of nomenclature.
- 1.
Variant designation that does not conform to current naming conventions
Normal gene product. COQ6 protein is a flavin-dependent monooxygenase involved in CoQ10 synthesis [Ozeir et al 2011]. COQ6 transcript variant 1 encodes isoform a (NP_872282.1), a 468-amino acid protein (54 kd) containing a mitochondrial import sequence. Transcript variant 2 encodes isoform b (NP_872286.2), a 443-amino acid protein (51 kd).
The human COQ6 isoform a localizes to mitochondria when overexpressed in several cell lines including podocytes. Under endogenous conditions it is expressed in glomeruli but not in tubules and localizes within cellular processes and Golgi apparatus [Heeringa et al 2011].
A pathogenic variant that reduced COQ6 expression (knockdown) in podocytes caused mitochondrial depolarization and increased the apoptotic rate through the intrinsic pathway, leading to growth defect. This phenotype was rescued by treating cells with CoQ10 [Heeringa et al 2011].
For information on yeast studies, see Coenzyme Q10 Deficiency – Model Organisms, COQ6.
Abnormal gene product. Alleles p.Trp447Ter, p.Gly255Arg, and p.Tyr412Cys did not rescue the respiratory deficiency of the COQ6-null yeast strain as did the wild-type, and p.Ala353Asp, and p.Ile462LeufsTer18 [Doimo et al 2014]. However, in vitro experiments suggest that all the alleles, with the exception of the nonsense allele p.Trp447Ter, are thought to be hypomorphic, because modeling of the human pathogenic variant on the correspondent yeast amino acid residue did not completely abolish the respiratory growth of the yeast strain. Finally, the phenotype of yeast expressing the human pathogenic alleles recovers after addition of vanillic acid or 3,4 dihydroxybenzoic acid [Doimo et al 2014].
The pathogenic variants p.Tyr412Cys and p.Ala353Asp affect an amino acid located at the flavin adenine dinucleotide (FAD) binding domain and may negatively interfere with COQ6 binding to FAD. The p.Gly255Arg variant, which affects a residue located in the active site pocket, and the p.Trp447Ter and p.Ile462LeufsTer18 variants, affecting residues located at the C-terminal tail, may cause perturbation of the active site [Doimo et al 2014].
COQ7
Gene structure.
COQ7 has two transcript variants each comprising six exons. They differ in the first exon; the longer transcript (NM_016138.4) encodes a 217-amino acid long protein (NP_057222.2), whereas the shorter transcript (NM_001190983.1) uses an alternate 5’ exon, resulting in a downstream AUG start codon with a shorter N-terminus resulting in a 170-amino acid protein (NP_001177912.1). See Table A, Gene, COQ7 for a detailed summary of gene and protein information.
Pathogenic variants. A single affected individual born to consanguineous parents has been reported to date [Freyer et al 2015] harboring a homozygous c.422T>A missense variant. The patient manifested mild learning disabilities, muscular hypotonia, and hearing and visual impairment.
Table 6.
Selected Pathogenic COQ7 Variants that Cause Primary CoQ10 Deficiency
View in own window
Variants listed in the table have been provided by the authors. GeneReviews staff have not independently verified the classification of variants.
GeneReviews follows the standard naming conventions of the Human Genome Variation Society (varnomen.hgvs.org). See Quick Reference for an explanation of nomenclature.
Normal gene product. COQ7 is a mitochondrial di-iron oxidase responsible for the penultimate step of CoQ synthesis, hydroxylating 5-demethoxyubiquinol (DMQH2) in the presence of NADH.
Abnormal gene product. The variant p.Val141Glu likely affects enzymatic function by impairing iron binding. Of note, supplementation of fibroblasts from the affected individual with 2,4-dihydroxybenzoic acid resulted in increased CoQ10 content and restored the combined activities of Complex I+III and II+III [Freyer et al 2015].
COQ8A
Gene structure.
COQ8A (previous symbols: ADCK3, COQ8, CABC1) comprises 15 exons. Alternatively spliced transcript variants have been found; however, their full-length nature has not been determined. The gene is ubiquitously expressed, with greater abundance in heart and skeletal muscle [Iiizumi et al 2002].
Pathogenic variants.
COQ8A pathogenic variants causing autosomal recessive ataxia associated with CoQ deficiency have been described extensively [Lagier-Tourenne et al 2008, Mollet et al 2008, Anheim et al 2010, Gerards et al 2010, Horvath et al 2012, Terracciano et al 2012, Mignot et al 2013, Blumkin et al 2014, Liu et al 2014, Barca et al 2016, Hikmat et al 2016].
More than 20 pathogenic variants have been reported, including missense, nonsense, and frameshift variants and a multiexon deletion (from exon 3 to exon 15).
To date all pathogenic variants reported are private and no founder effect has been identified.
Table 7.
Selected Pathogenic COQ8A Variants that Cause Primary CoQ10 Deficiency
View in own window
DNA Nucleotide Change (Alias 1) | Predicted Protein Change | Reference Sequences |
---|
c.637C>T (636C>T) | p.Arg213Trp |
NM_020247.4
NP_064632.2
|
c.811C>T | p.Arg271Cys |
c.815G>A | p.Gly272Asp |
c.815G>T | p.Gly272Val |
c.895C>T | p.Arg299Trp |
c.993C>T 2 | p.Leu314_Gln360del |
c.1042C>T | p.Arg348Ter |
c.1081-1_1082dupGTA | -- |
c.1136T>A | p.Leu379Ter |
c.1398+2T>C 3 | |
c.1523T>C | p.Phe508Ser |
c.1541A>G | p.Tyr514Cys |
c.1645G>A | p.Gly549Ser |
c.1651G>A (1655G>A) | p.Glu551Lys |
c.1750_1752delACC | p.Thr584del |
c.1813dupG (1812_1813insG) | p.Glu605GlyfsTer125 |
c.1844dupG (1844_1845insG) | p.Ser616LeufsTer114 |
c.1844G>A | p.Gly615Asp |
g.227150977_227195656del44680 4 | |
NC_000001.10
5
|
Variants listed in the table have been provided by the authors. GeneReviews staff have not independently verified the classification of variants.
GeneReviews follows the standard naming conventions of the Human Genome Variation Society (varnomen.hgvs.org). See Quick Reference for an explanation of nomenclature.
- 1.
Variant designation that does not conform to current naming conventions
- 2.
- 3.
- 4.
- 5.
Normal gene product.
COQ8A encodes a 647-amino acid protein that belongs to the UbiB protein kinase-like family and contains the conserved kinase motif in the region responsible for ATP binding and phosphotransfer reaction, but lacks the conserved kinase C-term motif. Moreover, it presents an N-terminal domain that is absent in the other proteins of the kinase family and it appears to be specifically related to ubiquinone metabolism [Stefely et al 2015].
In humans there are five paralogs belonging to the aarF domain-containing protein kinase (ADCK1-5); among them, COQ8A and COQ8B are highly similar and both are involved in CoQ10 biosynthesis [Lagier-Tourenne et al 2008, Ashraf et al 2013]. COQ8A localizes in mitochondria.
Computational and in vitro analyses prove that COQ8A forms homodimers after dimerization at the level of the transmembrane alpha-helices [Khadria et al 2014] and that the kinase motif displays magnesium (Mg(2+))-dependent ATPase activity [Wheeler & Jia 2015].
For information on yeast studies, see Coenzyme Q10 Deficiency – Model Organisms, COQ8A.
Abnormal gene product. The variants p.Arg213Trp, p.Gly272Val, p.Gly272Asp, and p.Glu551Lys predict changes in highly conserved amino acids of the protein, although none are in the kinase motifs [Mollet et al 2008].
The p.Tyr514Cys allele affects a residue proximal to the aspartates that bind the magnesium ions chelated by ATP [Lagier-Tourenne et al 2008].
The 1-bp frameshift insertion c.1813dupG results in the formation of a longer abnormal product (728 amino acids) and it is thought to modify an important domain of the protein altering the putative interaction or regulation between COQ8A and COQ9 [Mollet et al 2008].
The homozygous frameshift pathogenic variant p.Ser616LeufsTer114 causes the loss of the stop codon, leading to a 81-amino acid longer protein. The patient had significant CoQ10 deficiency and reduced mitochondrial respiratory chain enzyme activity.
The two pathogenic nonsense variants p.Arg348Ter and p.Leu379Ter cause a premature stop codon that triggers nonsense-mediated mRNA decay, leading to complete absence of functional COQ8A protein. Due to its regulatory role and to the presence of at least another ADCK protein with similar function (although patients with mutation of COQ8A do not have COQ8B up-regulation), the complete lack of residual functional protein is compatible with life [Gerards et al 2010].
The c.1081-1_1082dupGTA pathogenic variant does not alter the splicing of the transcript but causes insertion of three nucleotides, resulting in a stop codon [Mignot et al 2013].
The p.Phe508Ser variant is localized in one motif of the kinase domain [Mignot et al 2013].
The p.Gly549Ser and p.Gly615Asp pathogenic variants are in the C-terminal domain common to COQ8A and COQ8B [Lagier-Tourenne et al 2008, Mignot et al 2013].
COQ8B
Gene structure.
COQ8B spans 12 kb. Among the hypothetic 17 putative alternative splicing variants, the longest transcript NM_024876.3 contains 15 exons; exon 1 is noncoding [Ashraf et al 2013].
Pathogenic variants. Recessive loss-of-function pathogenic variants in COQ8B have been described in patients with steroid-responsive nephrotic syndrome (SRNS) associated with primary CoQ10 deficiency [Ashraf et al 2013].
Table 8.
Selected Pathogenic COQ8B Variants that Cause Primary CoQ10 Deficiency
View in own window
DNA Nucleotide Change | Predicted Protein Change | Reference Sequences |
---|
c.101G>A | p.Trp34Ter |
NM_024876.3
NP_079152.3
|
c.532C>T | p.Arg178Trp |
c.645delT | p.Phe214LeufsTer14 |
c.857A>G | p.Asp286Gly |
c.954_956dup | p.Thr319dup |
c.958C>T | p.Arg320Trp |
c.1027C>T | p.Arg343Trp |
c.1199dupA | p.His400AsnfsTer11 |
c.1356_1362delGGGCCCT | p.Gln452HisfsTer261 |
c.1430G>A | p.Arg477Gln |
c.1447G>T | p.Glu483Ter |
Variants listed in the table have been provided by the authors. GeneReviews staff have not independently verified the classification of variants.
GeneReviews follows the standard naming conventions of the Human Genome Variation Society (varnomen.hgvs.org). See Quick Reference for an explanation of nomenclature.
Normal gene product. Transcript variant NM_024876.3 encodes isoform a, which is a 60.1-kd protein that contains a helical domain, an ABC1 domain, and a kinase domain [Ashraf et al 2013]. COQ8B is one of the five ADCK paralogs and is highly similar to COQ8A, a putative kinase involved in CoQ10 biosynthesis [Lagier-Tourenne et al 2008]. It is conserved in several species and displays high sequence similarity with S.
cerevisiae Coq8/Abc1 protein [Ashraf et al 2013].
In humans, COQ8A expression exceeds COQ8B in several tissues with the exception of kidney. COQ8B is highly expressed in podocyte cell bodies and primary processes and, to a lesser extent, in renal glomeruli and in proximal tubules and collecting ducts. Analysis of subcellular fractions from cultured podocytes reveals that COQ8B resides both in mitochondria and cytosol, suggesting a localized function at ruffles and foot processes of podocytes besides its role in CoQ biosynthesis [Ashraf et al 2013].
See also Coenzyme Q10 Deficiency – Model Organisms, COQ8B.
Abnormal gene product. All the reported pathogenic missense variants affect conserved residues. Patients have reduced levels of CoQ10 in both primary skin fibroblasts and lymphoblastoid-derived cells.
All individuals with biallelic pathogenic variants in COQ8B have SRNS; however, the phenotype depends on the genotype [Ashraf et al 2013]:
The patient
homozygous for the p.His400AsnfsTer11 truncating variant had the earliest onset and developmental delay.
The patient
homozygous for the p.Arg178Trp amino acid change had diffuse glomerulosclerosis.
Homozygosity for the p.Gln452HisfsTer261
pathogenic variant was found in two sibs of Indian ancestry with renal histology and collapsing focal segmental glomerulosclerosis (cFSGS). Notably, cFSGS is common in individuals with mutation of
COQ8B as well as in the pdss2 kd/kd mouse model [
Saiki et al 2008,
Ashraf et al 2013].
COQ9
Gene structure.
COQ9 has nine exons. No alternative splicing variants are known.
Pathogenic variants. One patient of Pakistani origin with multiple-system disease characterized by intractable seizures, developmental delay, hypertrophic cardiomyopathy, and renal tubular dysfunction was homozygous for the c.730C>T pathogenic variant in exon 7 resulting in a premature stop codon (p.Arg244Ter) [Duncan et al 2009].
The homozygous loss-of-function variant c.521+1del was reported in a child of Turkish origin with fatal neonatal lactic acidosis and encephalopathy [Danhauser et al 2016].
Table 9.
Selected Pathogenic COQ9 Variants that Cause Primary CoQ10 Deficiency
View in own window
DNA Nucleotide Change | Predicted Protein Change | Reference Sequences |
---|
c.521+1delG | |
NM_020312.3
NP_064708.1
|
c.730C>T | p.Arg244Ter |
Variants listed in the table have been provided by the authors. GeneReviews staff have not independently verified the classification of variants.
GeneReviews follows the standard naming conventions of the Human Genome Variation Society (varnomen.hgvs.org). See Quick Reference for an explanation of nomenclature.
Normal gene product.
COQ9 encodes a 318-amino acid protein that is involved in the synthesis of CoQ10 [Duncan et al 2009].
The crystal structure of the human protein reveals that COQ9 is homologous to the TetR family of transcriptional regulators but does not retain any DNA binding ability. It is organized as a homodimer and contains a hydrophobic pocket, responsible for binding of lipid molecules (likely CoQ10 or CoQ10 precursor) and a binding surface crucial for protein-protein interaction with Coq7 [Lohman et al 2014].
See also Coenzyme Q10 Deficiency – Model Organisms, COQ9.
Abnormal gene product. The c.730C>T pathogenic variant is presumed to cause nonsense-mediated mRNA decay, as no transcript was detected in patient fibroblasts.
The c.521+1del pathogenic variant affects splicing with the skipping of exons 4 and 5 (p.Ser127_Arg202del), as shown by sequencing of the COQ9 transcript in the patient’s fibroblasts, with consequent degradation of the truncated protein [Danhauser et al 2016].
See also Coenzyme Q10 Deficiency – Model Organisms, COQ9.
PDSS1
Gene structure.
PDSS1 spans more than 49.14 kb and comprises 12 exons. There is only one coding transcript, which is 1,679 bp long.
Pathogenic variants.
PDSS1 pathogenic variants have been identified in only two families with primary CoQ10 deficiency to date:
Two sibs with encephalopathy, peripheral neuropathy, optic atrophy, cardiac valvulopathy, and mild lactic acidosis were
homozygous for the c.924T>G
missense variant in
exon 10 [
Mollet et al 2007].
Table 10.
Selected Pathogenic PDSS1 Variants that Cause Primary CoQ10 Deficiency
View in own window
DNA Nucleotide Change (Alias 1) | Predicted Protein Change | Reference Sequences |
---|
c.661_662insT (661C>CT) | p.Arg221LeufsTer16 |
NM_014317.3
NP_055132.2
|
c.924T>G (977T>G) | p.Asp308Glu |
c.1108A>C (1108A>AC) | p.Ser370Arg |
Variants listed in the table have been provided by the authors. GeneReviews staff have not independently verified the classification of variants.
GeneReviews follows the standard naming conventions of the Human Genome Variation Society (varnomen.hgvs.org). See Quick Reference for an explanation of nomenclature.
- 1.
Variant designation that does not conform to current naming conventions
Normal gene product.
PDSS1 encodes decaprenyl-diphosphate synthase subunit 1 (previously reported as DPS1) which is required for the synthesis of the polyisoprenoid chain of the appropriate length, the first step in CoQ10 biosynthesis. The protein is composed of 415 amino acids.
It is an ortholog of Schizosaccharomyces pombe Dps1. Unlike in S. cerevisiae where the ubiquinone side chain is synthesized by the monomeric enzyme COQ1, in S.
pombe and in mammals the PDSS1 polypeptide interacts with the product of PDSS2 forming a heterotetramer that is responsible for the elongation of the prenyl side chain of CoQ10 and determines the isoprenoid chain length of ubiquinone [Saiki et al 2005].
Abnormal gene product. In the absence of PDSS1, decaprenyl-diphosphate synthase is not functional and does not produce CoQ10.
For information on yeast studies, see Coenzyme Q10 Deficiency – Model Organisms, PDSS1.
PDSS2
Gene structure. The gene has at least two different transcript variants that share the first three exons; only the longest (NM_020381.3), which has eight exons, is believed to encode a functional subunit of the decaprenyl diphosphate synthase [Saiki et al 2005].
Pathogenic variants. To date PDSS2 pathogenic variants have been reported in two families; the phenotypes ranged from fatal Leigh syndrome and nephrotic syndrome to infantile-onset encephalomyopathy with ataxia, deafness, retinitis pigmentosa, and kidney disease [Rötig et al 2000, López-Martín et al 2007, Rahman et al 2012]. The patient reported by López-Martín et al [2007] was compound heterozygous for two novel variants, c.964C>T and c.1145C>T.
Table 11.
Selected Pathogenic PDSS2 Variants that Cause Primary CoQ10 Deficiency
View in own window
DNA Nucleotide Change | Predicted Protein Change | Reference Sequences |
---|
c.964C>T | p.Gln322Ter |
NM_020381.3
NP_065114.3
|
c.1145C>T | p.Ser382Leu |
Variants listed in the table have been provided by the authors. GeneReviews staff have not independently verified the classification of variants.
GeneReviews follows the standard naming conventions of the Human Genome Variation Society (varnomen.hgvs.org). See Quick Reference for an explanation of nomenclature.
Normal gene product. The protein product of PDSS2 (previously reported as DLP1) is the second subunit of decaprenyl diphosphate synthase, which is required for the elongation of the prenyl side chain of CoQ10. The PDSS2 protein consists of 399 amino acids.
Unlike S
cerevisiae, the prenyl diphosphate synthase in humans acts as a heterotetrameric complex, formed by two protein subunits encoded by PDSS1 and two protein subunits encoded by PDSS2 [Saiki et al 2005]. The same heterotetrameric complex is also found in mice and S
pombe.
Abnormal gene product. In the absence of PDSS2, decaprenyl-diphosphate synthase is not functional and does not produce CoQ10. The PDSS2 pathogenic variants reported by López-Martín et al [2007] act through a loss-of-function mechanism, as suggested by substrate incorporation experiments showing a CoQ10 biosynthetic defect in fibroblasts from an affected individual when incubated with radioactive para-hydroxybenzoate (PHB), compared with normal synthesis in cells incubated with radiolabeled PHB and decaprenyl-PP.
For information on mouse studies, see Coenzyme Q10 Deficiency – Model Organisms, PDSS2.