1. Chromosomes and Gene Expression
  2. Genetics and Genomics
Download icon

Nuclear genetic regulation of the human mitochondrial transcriptome

  1. Aminah T Ali
  2. Lena Boehme
  3. Guillermo Carbajosa
  4. Vlad C Seitan
  5. Kerrin S Small
  6. Alan Hodgkinson  Is a corresponding author
  1. King’s College London, United Kingdom
Research Article
Cite this article as: eLife 2019;8:e41927 doi: 10.7554/eLife.41927
3 figures, 2 tables, 4 data sets and 7 additional files

Figures

Variation in the expression of mitochondrial-encoded genes across datasets.

(A) Hierarchical clustering of median expression levels per gene across all datasets where WBL = Whole Blood, SAD = Subcutaneous Adipose, LCL = Lymphoblastoid cell lines, SKN = Non sun exposed skin, SKE = Sun exposed skin, VAD = Visceral omentum adipose, ADG = Adrenal gland, AOR = Aorta, CAR = Coronary artery, TAR = Tibial artery, ACB = Anterior cingulate cortex (BA24) (Brain), CGB = Caudate basal ganglia (Brain), CHB = Cerebellar Hemisphere (Brain), CEB = Cerebellum (Brain), COB = Cortex (Brain), FCB = Frontal cortex (BA9) (Brain), HIB = Hippocampus (Brain), HYB = Hypothalamus (Brain), NAB = Nucleus accumbens (basal ganglia) (Brain), PBB = Putamen basal ganglia (Brain), BRE = Breast mammary tissue, SCO = Sigmoid colon, TCO = Transverse colon, GEJ = Gastroesophageal junction, EMC = Esophagus mucosa, EMS = Esophagus Muscularis, AAH = Atrial appendage (Heart), LVH = Left ventricle (Heart), LUN = Lung, SMU = Skeletal muscle, TNV = Tibial Nerve, PAN = Pancreas, SFI = Transformed fibroblasts, STO = Stomach, TES = Testes and THY = Thyroid, Multi-dataset tissues on the x-axis are shown in red (whole blood), orange (subcutaneous adipose), green (lymphoblastoid cell lines) and blue (non-sun exposed skin). (B) Standardized expression levels of each mitochondrial-encoded gene across all independent datasets, (C) Coefficient of variation across individuals for the expression levels of mitochondrial encoded genes and the top 1000 most highly expressed nuclear genes in all datasets. Range of coefficient of variation is restricted to between 0 and 1.5 as this contains the majority of the data.

https://doi.org/10.7554/eLife.41927.003
Figure 2 with 1 supplement
Associations between the expression of MTND1 and rs2304694 in whole blood data.

(A) Genome-wide association analysis for the expression of MTND1 in whole blood data from the discovery datasets (meta-analysis of CARTaGENE, TwinsUK and GTEx data), (B) Genome-wide association analysis for the expression of MTND1 in whole blood data from the replication dataset (NIMH data), (C) Expression of MTND1 (Log10(TPM +1)) versus non-reference allele frequency of rs2304694 in the four independent whole blood datasets.

https://doi.org/10.7554/eLife.41927.004
Figure 2—figure supplement 1
QQ plots for associations between nuclear genetic variants and mitochondrial gene expression for discovery associations that replicate at the nominal 5% level.
https://doi.org/10.7554/eLife.41927.005
Replication and validation of significant associations between nuclear genetic variants and the expression of mitochondria-encoded genes.

(A) Discovery versus replication beta estimates for significant associations between nuclear genetic variation and mitochondrial gene expression detected in discovery data at FDR 5%, (B) Validation of the association between rs2304694 and the expression of MTND4 using quantitative PCR in LCLs. MTND4 mRNA expression levels are normalised to GAPDH (theoretical quantities).

https://doi.org/10.7554/eLife.41927.007

Tables

Table 1
Associations where a suggestive causal nuclear gene is implicated.

‘Missense mutation’ denotes that the nuclear genetic variant associated with the expression of a mitochondrial-encoded gene is a missense mutation, ‘Mediation (Mitochondrial Gene)’ denotes that the expression of a nearby nuclear gene known to play a role in mitochondrial processes explains a significant proportion of the association between a nuclear genetic variant and the expression level of a mitochondrial encoded gene, and ‘Mediation (other nuclear gene)’ denotes a similar result whereby the nuclear gene identified is thought to have no known role in mitochondrial processes (see Materials and methods).

https://doi.org/10.7554/eLife.41927.006
TissuePeak SNPMT geneMissense mutationMediation (Mitochondrial Gene)Mediation (other nuclear gene)
Whole Bloodrs7558127MTND6NAPNPT1NA
Whole Bloodrs6973982MTCO2NATBRG4NA
Whole Bloodrs11085147MTCO2LONP1NANA
Whole Bloodrs2304693MTCYBTBRG4NANA
Whole Bloodrs74025341MTCYBNANASLC7A6OS,ZFP90
Whole Bloodrs7158706MTND2NANAPPP2R3C
Whole Bloodrs10172506MTND5NAPNPT1NA
Whole Bloodrs74863981MTCO1NANAUBOX5,TGM3,LZTS3
Whole Bloodrs76125482MTND3NAFASTKD1NA
Whole Bloodrs6973982MTND4NANARP4-647J21.1,CCM2
Whole Bloodrs11008009MTND4NAMTPAPNA
Whole Bloodrs2304694MTND1TBRG4NANA
Whole Bloodrs1692120MTND1NANAMYRF
Whole Bloodrs6973982MTATP6NANACCM2
Whole Bloodrs589809MTATP6NANAFLT1
Whole Bloodrs375640557MTCO3NANACCDC104
Whole Bloodrs6973982MTCO3NATBRG4NA
Whole Bloodrs10165864MTRNR2NAPNPT1NA
Whole Bloodrs66892251MTRNR2NAMTPAPNA
Whole Bloodrs61988269MTRNR1NAMRPP3NA
Subcutaneous Adiposers2304694MTND6TBRG4NANA
Subcutaneous Adiposers2304694MTND5TBRG4NANA
Subcutaneous Adiposers2304694MTND1TBRG4NANA
Subcutaneous Adiposers12579998MTND1NAMRPS35NA
Subcutaneous Adiposers2304693MTCO3TBRG4NANA
Skin (Not sun exposed)rs2304693MTCO2TBRG4NANA
Skin (Not sun exposed)rs2304693MTCO3TBRG4NANA
LCLsrs7559561MTCO2NALRPPRCNA
LCLsrs2304694MTCO2TBRG4NANA
LCLsrs1047991MTND3MTPAPNANA
LCLsrs2304694MTND4TBRG4NANA
LCLsrs10205130MTND1NALRPPRCNA
LCLsrs35739334MTND1NATBRG4NA
LCLsrs2304694MTCO3TBRG4NANA
LCLsrs2304694MTRNR2TBRG4NANA
LCLsrs2304694MTND4LTBRG4NANA
Table 2
Significant associations between nuclear genetic variants and the expression levels of genes encoded in the mitochondrial genome that replicate in independent tissue matched datasets.

Point-wise permutation P values were generated by extrapolating from the underlying beta distribution (see Materials and methods).

https://doi.org/10.7554/eLife.41927.008
TissuePeak
nuclear
SNP
ChrPositionA1MAFMT
gene
P
value
P value
(FDR
corrected)
P value
(Point-wise
Permutation)
BetaReplication P valueReplication
beta
Nuclear gene
annotation
Nuclear SNP
annotation
LCLsrs10205130244151572C0.426MTND11.23E-106.40E-051.18E-100.0235.70E-050.015LRPPRCIntronic
Whole
Blood
rs10172506255865469C0.062MTND54.40E-402.00E-331.39E-390.0582.00E-080.037PNPT1Intronic
Whole
Blood
rs7558127255866605G0.064MTND65.26E-328.30E-265.29E-320.0708.70E-070.043PNPT1Intronic
LCLsrs2627775255877113T0.458MTCO11.33E-094.10E-049.95E-100.0176.30E-040.013PNPT1Intronic
LCLsrs62165226255878339C0.484MTCO32.79E-087.60E-032.71E-08−0.0186.80E-06−0.018PNPT1Intronic
Whole
Blood
rs6973982745143892G0.148MTCO27.53E-318.60E-251.12E-30−0.0282.30E-08−0.023TBRG4Intronic
Whole
Blood
rs6973982745143892G0.148MTCO31.23E-256.40E-201.05E-25−0.0281.40E-04−0.019TBRG4Intronic
Skin
(Not
sun
exposed)
rs2304693745148667A0.201MTCO32.84E-122.10E-052.93E-12−0.0271.80E-03−0.026TBRG4Missense
Whole
Blood
rs2304694745148773A0.148MTND12.34E-176.10E-123.19E-170.0151.60E-080.024TBRG4Missense
LCLsrs2304694745148773A0.197MTND42.35E-183.90E-112.22E-18−0.0291.40E-04−0.019TBRG4Missense
Subcutaneous
Adipose
rs125799981227861446G0.174MTND12.48E-236.50E-171.09E-23−0.0323.20E-05−0.032-Intergenic
Subcutaneous
Adipose
rs125799981227861446G0.174MTND42.20E-121.70E-061.56E-12−0.0224.00E-04−0.027-Intergenic
Skin
(Not
sun
exposed)
rs110491031227862081A0.179MTND17.72E-122.10E-056.68E-12−0.0228.10E-04−0.026-Intergenic
Whole
Blood
rs74863981203109875A0.099MTCO1
4.24E-115.40E-065.25E-11−0.0246.60E-05−0.022UBOX5Intronic

Data availability

Anonymized processed mitochondrial encoded gene expression matrices are available in Supplementary file 2 and from the Gene Expression Omnibus under accession GSE125013. Data from CARTaGENE (Awadalla et al., 2013) and NIMH sequencing data from the Depression Genes and Networks study (Battle et al., 2014) used in this work are available through request. Requests for access first need to be approved by a data access committee (further information can be found here https://www.cartagene.qc.ca/en/researchers/access-request and here https://www.nimhgenetics.org/request-access/how-to-request-access).

The following data sets were generated
  1. 1
    NCBI Gene Expression Omnibus
    1. AT Ali
    2. L Boehme
    3. G Carbajosa
    (2019)
    ID GSE125013. Nuclear Genetic Regulation of the Human Mitochondrial Transcriptome.
The following previously published data sets were used
  1. 1
    European Genome-Phenome Archive
    1. A Buil
    2. AA Brown
    3. T Lappalainen
    4. A Vinuela
    5. MN Davies
    (2015)
    Gene-gene and gene-environment interactions detected by transcriptome sequence analysis in twins.
  2. 2
    NCBI dbGaP
    1. Consortium GTEx
    (2017)
    Common Fund (CF) Genotype-Tissue Expression Project (GTEx).
  3. 3
    European Nucleotide Archive
    1. T Lappalainen
    (2013)
    ID ERA169774. Geuvadis Project.

Additional files

Supplementary file 1

Significant associations between nuclear genetic variants and the expression levels of genes encoded in the mitochondrial genome.

See Materials and methods for a description of the calculation of different permutation P values.

https://doi.org/10.7554/eLife.41927.009
Supplementary file 2

Details of trans-genome associations when calculating TPM values for mitochondrial-encoded genes using the total number of reads mapping to the mitochondrial genome.

https://doi.org/10.7554/eLife.41927.010
Supplementary file 3

Replication P-values and Beta Coefficients for mitochondrial trans-genome eQTLs in all tissue types.

https://doi.org/10.7554/eLife.41927.011
Supplementary file 4

Gene expression matrices for all dataset used in the study.

Values show log-transformed (TPM +1) values. After transformation, outlier values were removed (three inter-quartile ranges above/below the upper/lower quartiles respectively).

https://doi.org/10.7554/eLife.41927.012
Supplementary file 5

Population, sex and allele information for cell lines used in qPCR validation experiments.

https://doi.org/10.7554/eLife.41927.013
Supplementary file 6

Forest plots for each peak variant detected in association analyses.

Each plot contains Beta estimates and confidence intervals for each of the datasets and tissues considered in the study.

https://doi.org/10.7554/eLife.41927.014
Transparent reporting form
https://doi.org/10.7554/eLife.41927.015

Download links

A two-part list of links to download the article, or parts of the article, in various formats.

Downloads (link to download the article as PDF)

Download citations (links to download the citations from this article in formats compatible with various reference manager tools)

Open citations (links to open the citations from this article in various online reference manager services)