Table 2.
Summary of the in silico and in vitro analyses in MDH2 missense and deletion VUS
| VUS | V-l p.Ser3Phe | V-2 p.Argl04Gly | V-5 p.Glnl30Arg | V-6 p.Vall60Met | V-9 p.Ala256Thr | V-10 p.Lys314del | |
|---|---|---|---|---|---|---|---|
| SNP ID ExAC allele freq. gnomAD* allele freq. 1000 Genomes* | not described | not described | not described | rsl38541865 1,66–10−4; 0 hom | rsl47655350 4,69–10−5; 0 hom | not described | |
| Evolutionary conservation Interpretation (a) | Conserved position | Conserved position | Conserved position | Conserved position | Conserved position | Conserved position ¤ | |
| Functional domain | NA (Low complexity region) | Ldh_l_N domain (PF00056) | Ldh_l_N domain (PF00056) | Ldh_l_N domain (PF00056) | Ldh_l_C domain (PF02866) | Ldh_l_C domain (PF02866) | |
| Protein function prediction Interpretation (a) | Inconclusive (5/9) | Impaired (10/10) | Neutral (3/10) | Impaired (8/10) | Impaired (9/10) | NA | |
| 3D structural annotations (Structure-PPi) | Transit peptide to Mitochondria (aa 1–24) | NAD-binding site | α-helix | NA | α-helix | α-helix | |
| 3D structural prediction Interpretation (a) | No effect | Destabilize (4/5) | Inconclusive (3/5) | Destabilize (5/5) | Destabilize (5/5) | NA | |
| Neighbor positions mutated in other cancers (b) (Structure-PPi) | NA | Malignant melanoma, Ovary carcinoma, Colon carcinoma, Endometrium carcinoma | Breast carcinoma | Stomach carcinoma | Endometrium carcinoma, Head and neck squamous cell carcinoma | Endometrium carcinoma (p.Ser310Ser), Prostate carcinoma (p.Ser317Leu), Oral carcinoma (p.Ala319Ser) | |
| RBP1 expression (c) | +++ | + | NA | + | NA | + | |
| LOH | no | no | NA | no | NA | Yes | |
| MDH2 subcelular localization (d) | M | M | M | M | M | NA | |
| Enzymatic experiment | with saturating [NAD+] and [malate] (e) | +++ | + | +++ | +++ | +++ | NA |
| co-transfection experiment (e) | NA | ++ | NA | NA | NA | NA | |
| Affinity assay: with lower [NAD+] and [malate] (f) | NA | Impaired for malate and NAD+ | NA | Likely impaired for malate | Likely impaired for malate and NAD+ | NA | |
| Molecular dynamics simulations | NA | NA | NA | Substrate binding site likely affected | Dimerization likely affected | NA | |
| Classification of VUS | LIKELY BENIGN | PATHOGENIC | LIKELY BENIGN | LIKELY PATHOGENIC | LIKELY PATHOGENIC | PATHOGENIC | |
Results in bold indicate major alterations. V-variant, NA not available, Ldh_1_N domain lactate/malate dehydrogenase, NAD-binding domain (Pfam accession: PF00056), Ldh_1_C domain lactate/malate dehydrogenase, alpha/beta C-terminal domain (Pfam accession: PF02866), SNP single-nucleotide polymorphism, VUS variant of unknown significance, M mitochondrial
See Table S4 for breakdown of the different computational predictors. The number of tools predicting a damaging effect is in brackets
See Fig. S2 for mapping of the pathogenic variant onto the crystal structure
RBP1 messenger RNA (mRNA) expression relative to tumors with pathogenic variants in non-Krebs cycle genes: +++ corresponds to nonsignificant RBP1 expression reduction, + corresponds to significant (p < 0.05) RBP1 expression reduction
M: mitochondrial subcelular localization
From in vitro MDH2 enzymatic activity (Fig. 2a, b): +++ corresponds to MDH2 activity >75% compared with control, ++ 25–75%, and + 0–25%
http://gnomad.broadinstitute.org/ and http://www.internationalgenome.org/ (Last accessed 7 June 2017)
Despite the fact that the variant p.K314del is not annotated in the dbNSFP database, we infer the evolutionary conservation of this position considering the annotated variants at this position (i.e., p.K314Q, p.K314E, p.K314*, p.K314T, p.K314R, p.K314M and p.K314N). For each method, the lowest and the highest scores are included