Skip to main content

Thank you for visiting You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

MtDNA population variation in Myalgic encephalomyelitis/Chronic fatigue syndrome in two populations: a study of mildly deleterious variants


Myalgic Encephalomyelitis (ME), also known as Chronic Fatigue Syndrome (CFS) is a debilitating condition. There is growing interest in a possible etiologic or pathogenic role of mitochondrial dysfunction and mitochondrial DNA (mtDNA) variation in ME/CFS. Supporting such a link, fatigue is common and often severe in patients with mitochondrial disease. We investigate the role of mtDNA variation in ME/CFS. No proven pathogenic mtDNA mutations were found. We then investigated population variation. Two cohorts were analysed, one from the UK (n = 89 moderately affected; 29 severely affected) and the other from South Africa (n = 143 moderately affected). For both cohorts, ME/CFS patients had an excess of individuals without a mildly deleterious population variant. The differences in population variation might reflect a mechanism important to the pathophysiology of ME/CFS.


Myalgic encephalomyelitis (ME), also known as chronic fatigue syndrome (CFS) is an illness characterized by post-exertional malaise, cognitive dysfunction, unrefreshing sleep, and other symptoms1. Severe fatigue is the defining clinical problem, and sleep or rest does not alleviate this fatigue. ME/CFS is a disease associated with substantial levels of disability in many patients, with a high social and economic cost2. Skeletal muscle fatigue, and more specifically post-exertional malaise, is a key feature of ME/CFS, and recent studies point to abnormalities of muscle function3. Patients suffering from ME/CFS show a reduced capacity to recover from acidosis on repeat exercise3,4, which is suggestive of a role for mitochondrial dysfunction in the pathophysiology of this disease. In vitro investigations have supported this hypothesis with studies showing functional mitochondrial abnormalities such as a lower mitochondrial respiratory function in ME/CFS patients5.

ME/CFS is clinically heterogeneous with the patients most severely affected being house- or even bed-bound for months and sometimes years. Recent findings have indicated that the house-bound group represents one quarter of the patient population2. The functional abilities of these patients are significantly more impaired in many aspects compared to ME/CFS patients who were not house-bound, including but not limited to fatigue, post-exertional malaise, sleep and pain, as well as neurocognitive, autonomic, neuroendocrine, and immune functioning2. However, the factors that make a patient less or more likely to fall into this severely affected group are unknown, as are the factors linked to recovery. Evidence suggests that recovery remains rare in the ME/CFS patient population regardless of disease severity6.

Clinically proven pathogenic mtDNA mutations are recognized as a cause of maternally-inherited disorders, with a minimum prevalence rate of 1 mutation in 5,000 (20 per 100,000) people7. Such mtDNA mutations frequently cause multisystem disorders with fatigue being prevalent in these patient groups8. Hundreds or even thousand of copies of mtDNA are present in a /cell, this is linked to the energetic demands of the cell. These copies can either be identical, a state called homoplasmy, or two or more species of mtDNA can be present in a state referred to as heteroplasmy. Thus, a possible approach is to investigate whether mtDNA mutations are present either at heteroplasmy levels sufficient to cause mtDNA disease or at sub-clinical levels that are too low to cause primary mtDNA disease, but are perhaps at sufficient levels to act as a risk factor or affect the course of a complex disease such as ME/CFS.

Beyond such recognised pathogenic mtDNA mutations, many studies have suggested a role for common mtDNA variants in complex diseases, with mtDNA variants either modulating susceptibility to a disease and/or affecting the course of the disease, including those where fatigue is an important feature of disease9,10. While many studies have reported a significant association of specific mtDNA haplogroups with a number of complex disorders, there is often substantial disagreement among different studies examining the same phenotype.

Another possibility is that rare mtDNA population variants might have a role in the disease process, since rare variants are predicted to be more mildly deleterious as such variants are removed by purifying selection over generations11. This is supported by recent works12,13, using a computational tool, MutPred, that has been widely used in the mitochondrial context14. Given this it might be expected to see a greater number of rare/mildly deleterious variants in any given patient group15, which may implicate a role in the disease process.

In addition to acting as a susceptibility factor to disease, it has also been suggested that mtDNA variants might modify the course of common complex diseases16. In line with this a recent study of 193 ME/CFS patients and 196 age- and gender-matched controls, reported haplogroups J, U and H, as well as eight mtDNA SNPs are significantly associated with particular ME/CFS symptoms in patients (course of disease), but not with increased susceptibility to ME/CFS (onset of disease)10.

In the current study, using mtDNA sequence data from ME/CFS patients from both the United Kingdom (UK) and South Africa (RSA), we ask whether mtDNA population variants alter susceptibility to ME/CFS. Due to the maternal inheritance pattern of mtDNA, linking mtDNA variation to common complex traits has been recognized as a difficult task, with models used in nuclear association genetics proving unsuitable17,18. There have been a number of improvements in the design of haplogroup association studies over the last 10 years, such as the importance of a replicate cohort being recognized15,19. However, there are still many doubts as to whether this simple methodology is the correct approach to assess the role of mtDNA variation in complex disorders. Indeed, a number of prior studies have applied the haplogroup association approach in complex diseases where fatigue is an important clinical feature, such as multiple sclerosis9,20, but the results were inconclusive. Here, we applied an improved approach which focuses on variants predicted computationally to be mildly deleterious, most of which are rare. This approach takes advantage of the advances in bioinformatics21,22 using the mtDNA-server23 and MutPred tools14,23.


Heteroplasmy analysis

To investigate the possibility that ME/CFS patients harbour clinically proven mtDNA mutations, either above the threshold required for mtDNA disease or at a sub-threshold level, the complete mitochondrial genomes of the (n = 89 moderate; 29 severely affected) patients from the UK and (n = 143) South African ME/CFS patients were analysed. Only two mutations associated with clinically proven mitochondrial disease were seen, m.3337G>A24 in a patient and m.11778G>A25 in a control participant, both in the South African cohort. However, these mutations were heteroplasmic at the 5% and 16.5% level respectively, and as a result these mutations are unlikely to have a phenotypic role. No other clinically proven mtDNA mutations were detected in any of the patient or control groups. It should be noted that such frequencies of low level pathogenic mtDNA mutations are entirely consistent with large population studies considering this question26, as well as prior studies on this phenotype10,27.

Haplogroup distributions

Human mtDNAs can be assigned to one of several haplogroups. This traditional classification system was originally based on the presence or absence of one or a small number of likely benign polymorphisms28, rather than mutations with functional consequences. However, the advent of large-scale sequencing has led to the identification of a vast number of haplogroup-specific polymorphisms, allowing haplogroup classifications to be expanded into more and more subgroups29. To put our analysis into the context of prior studies, a simple haplogroup distribution analysis was performed.

To ensure a robust comparison between the groups from the UK and the RSA, only the frequencies of the nine haplogroups of European origin (HVUKTJIXW) from the RSA ME/CFS cohort were used, as the group from the UK (North East England) was of lower diversity. The haplogroup distributions for the UK and RSA cohorts – ME/CFS and control - are given in Table 1. Comparing the haplogroup distributions between the UK ME/CFS patients and controls with a Monte Carlo-based approach, no significant difference was observed (ns, p = 0.31). Similarly, there was no significant difference in haplogroup distribution between RSA patients and controls (ns, p = 0.55). Thus as in the work of Billing-Ross et al., mtDNA haplogroup was not seen to affect the susceptibility to ME/CFS10.

Table 1 Haplogroup distribution for each group in the two cohorts.

Network analysis

Functional networks were produced for the UK cohort (Fig. 1a) and the RSA cohort (Fig. 1a). These networks represent all possible least complex phylogenetics trees, based on the variants included as described in Methods. Sequences that are very similar cluster together in smaller or larger nodes. Nodes are ordered along a phylogenetic tree with links indicating the variants by which a connected node deviated from another30. In both cohorts, controls were more abundant in nodes that were separated by several variants from the central, haplogroup H-dominated nodes. These peripheral nodes frequently contained mtDNAs that could be assigned to haplogroups T, K and U, although as noted in the previous section, no significant haplogroup associations were found.

Figure 1

Shows functional networks of UK participants (a) and RSA (a) participants. Ellipses indicate clusters of branches and nodes that fall within similar haplogroups. Small red nodes indicate potential median vectors. Pie-chart nodes indicate the proportion of individuals per phenotype classification, as well as the number of participants per node. mtDNA variants responsible for links are listed. Transversions are shown in italics.

Individuals with a potentially mildly deleterious mtDNA variant

Haplogroup association studies are confounded by a number of factors, including population stratification, with this and other factors resulting in elevated type 1 error18,31. These limitations, coupled with the relatively low statistical power of haplogroup association studies32, underscore the fact that new methods are required for discerning the pathogenic and/or etiological role of mtDNA mutations in a common complex disorder21,22. We determined the number of individuals with mtDNA sequences containing a variant with a MutPred score over the 0.5 threshold, such variants being considered “actionable hypothesis” or candidates for having a functional effect14, here referred to as “mildly deleterious”.

The number of controls and patients in each population that harbours none, one or more mildly deleterious variants are summarised in Table 2. Figure 2 illustrates these numbers as percentages of each group.

Table 2 Number of individuals that harbour high MutPred scored variants.
Figure 2

Pie-charts indicating the percentage of controls and ME/CFS patients that harbour variants with MutPred scores above 0.5 for the (a) UK cohort and (b) South African cohort. ME/CFS: Myalgic encephalomyelitis/chronic fatigue syndrome.

For the cohorts collected in the North East of England, 49 (55%) of the moderate ME/CFS patients were without a mildly deleterious variant. Comparing this result to the controls from the United Kingdom, only 17 (27%) of these individuals had no such variants. Conversely 45% of moderate ME/CFS patients had one or more mildly deleterious variants, compared to 72% of controls from the UK who harboured one or more such variants (see Fig. 2). These differences were found to be highly significant (p = 0.0008) with a Fisher’s Exact test. This observation suggests that individuals with ME/CFS are less likely to have an mtDNA variant that is predicted to have a functional impact.

As mentioned above, as many as 25% of those with ME/CFS are severely affected, being house or even bed bound. A cohort of such patients was recruited in the North East of England (UK). Of the 29 severe ME/CFS patients 15 (52%) were without a mildly deleterious variant. Comparing these numbers with those of UK controls using a Fisher’s Exact test, a significant difference (p = 0.03) was found. Thus even the most severely affected patients had fewer mildly deleterious variants than the controls. It is important to note that when the moderately and severely affected patients are compared using a Fishers exact test no difference is observed (p = 0.83). The analysis was repeated in the RSA cohort, including only those ME/CFS cases that fall within the nine haplogroups designated as European (HVUKTJIXW) to ensure that differences in lineage diversity could not impact upon the results. Conducting the same analysis with the ME/CFS 143 patients from the RSA, there were 80 (56%) individuals with predicted mildly deleterious variants; considering the healthy controls from the RSA, 40 (41%) individuals had no such variants. That leaves 59% of controls from South Africa compared to only 44% of ME/CFS patients who harbour one or more mildly deleterious mtDNA variants. A Fisher’s exact test again found these differences to be statistically significant (p = 0.03). Our analyses thus show, for both UK and RSA cohorts, those with ME/CFS have fewer mildly deleterious variants than controls.


We have investigated a possible role of mtDNA variants in ME/CFS. This was done firstly by reviewing the sequence data of each individual for the presence of clinically proven pathogenic mtDNA mutations associated with primary mitochondrial disease. We ruled out the possibility that previously identified pathogenic mtDNA mutations are contributing to ME/CFS, a result that was not unexpected10,27.

Secondly, we considered the possible role of population variants in ME/CFS, with a focus on variants that are predicted to be mildly deleterious by computational methods. These variants are frequently rare at the population level resulting in an absence of population stratification, and thus in a lower chance of false association or type 1 error. Such false associations are believed to be common in mtDNA association studies applying the haplogroup association model18. The current approach has been utilized in two prior studies21,22 that considered Alzheimer and cardiovascular disease respectively.

We compared the number of individuals in both the ME/CFS patient and control groups with variants classified as mildly deleterious, to the number of individuals without such variants. In both the UK and RSA cohorts, there was a significant difference between the patients and the controls, with the ME/CFS patients having a higher percentage of patients without a variant predicted to be mildly deleterious. Although surprising, this observation was seen in two independent cohorts, suggesting that these differences are not due to chance. One of the reasons for this observation might have been that deleterious variants confer disease susceptibility in only severely affected ME/CFS patients or modulate fatigue severity among ME/CFS patients. However, when comparing patients that were severely affected with controls in the UK cohort, again ME/CFS patients had fewer variants predicted to be mildly deleterious. Furthermore, there was no difference in the number of patients with such variants between the moderate and severely affected groups. Taken together, these data suggest that our observation is not the result of a simple patient stratification effect, although additional replication cohorts of moderate and severely affected ME/CFS cohorts are needed to confirm this.

It has been proposed that those with ME/CFS might not have a problem with ATP production33, but rather with ATP utilization. Therefore, we consider this genetic difference in mtDNA variation as an accurate observation which may prove to have a biological relationship to the function and regulation of the OXPHOS system and subsequent wide-ranging immediate and downstream consequences on energy metabolism34,35. It should be considered that the wider homeostasis and/or responsiveness of the various elements of energy metabolism are affected in ME/CFS, rather than merely single “segments” of energy pathways such as ATP production.

A number of mitochondrial abnormalities in ME/CFS have previously been reported, indicating that mitochondrial dysfunction may play a role in the pathogenesis of disease, at least in a sub-set of patients36. Previously ME/CFS patients have been shown to have significantly lower mitochondrial function than healthy controls5. Detection of these mtDNA variants has the potential to be used as a tool for measuring mitochondrial dysfunction in ME/CFS. Another avenue of investigation in the mitochondrial field is copy number analysis. Mitochondrial DNA copy number (mtDNAcn) has been reported as an indirect representative of mitochondrial function and as a biomarker of disease with studies. As an example, in Parkinson’s disease (PD), mtDNAcn was elevated in the pedunculopontine nucleus (PPN), which is a brainstem region associated with progression of motor and non-motor symptoms of PD37. Other neurodegenerative conditions in which alterations in mtDNAcn have been reported include Multiple Sclerosis. The results in this phenotype are conflicting, with some showing evidence for reduced copy number38, while others indicate an increase39. Additionally. MtDNAcn was shown not to be associated with fatigue status in Primary Sjögren’s Syndrome30. Taken together these papers demonstrate that in heterogeneous diseases with a variable course small single time point studies will produce data that is difficult to interpret and is likely to conflict between studies.

In conclusion, this is the first paper to demonstrate mitochondrial genetic differences between ME/CFS patients and controls. It also demonstrates the power of mtDNA analysis focused on variants likely to be of a functional effect to detect differences between case and control cohorts where the traditional haplogroup association method frequently fails to do so. Future studies need to include larger cohorts from multiple centres, within and between nations, with standardized sample handling. These studies need to take a multi-disciplinary approach linking genetics, including mtDNA copy number analysis and bioenergetics. Given the changing nature of the disease, longitudinal studies would seem to be essential to further understanding by allowing us to determine how mtDNA varation and mitochondrial dysfunction relates to fluctuations in symptom severity.


Ethical approval and informed consent

This study was conducted in two independent cohorts, one from England and the other from South Africa. All experimental protocols were approved by the corresponding intuitional committees, namely the CRN National Coordinating Centre (CRNCC) | NIHR Clinical Research Network (CRN) in England (UK ME/CFS - IRAS ID 221364) and the Health Research ethics committee (HREC) of the North-West University in South Africa (SABPA: NWU 00036-07-S6, CFS: NWU 00102-12). All study procedures carried out in accordance with relevant guidelines and regulations of Newcastle University and North-West University. All participants gave informed consent and were over the age of 18.

Patient cohorts

We used two well-characterised cohorts of ME/CFS patients, one from the North East of England (n = 89 moderately affected; 29 severely affected) and the other from South Africa (n = 143 moderately affected). Both cohorts met the Fukuda diagnostic criteria1. Potentially confounding causes of fatigue, including depression, were excluded in all patients. The two control cohorts were regionally matched and had been collected for two prior studies. The controls from the North East of England (n = 64) were used previously for a variant load study on Alzheimer’s disease15. The control cohort from South Africa (n = 98) was comprised of healthy high school teachers assembled previously for a study on hypertension and diabetes22.


Sequencing of the ME/CFS samples from both the UK and RSA was carried out at Source Bioscience using a fluidigm technology. The sequencing methodology for the controls has been described previously for the UK controls15 and for the RSA controls22. The reference sequence used in all datasets was the revised Cambridge reference sequence (rCRS).

Network analysis

After selective pruning of the total mtDNA variants to those in (a) protein-encoding genes with MutPred pathogenicity scores above 0.5, and (b) rRNA and tRNA variants, a functional maximum parsimony (MP) network analysis was performed with the NETWORK (version software package ( Transversions, being chemically less likely to occur, were weighted three times more than transitions. Star contractions were performed using a maximum radius of 5. This step was followed by reduced median (RM)40 processing (reduction threshold r was 1). The “Frequency >1” criterion was activated to exclude sequences which are unique to the dataset. Network figures were produced using NETWORK Publisher (version, PowerPoint (Microsoft Office 365) and Adobe Photoshop Elements (version 10.0).

Data analyses

Sequencing data was processed using the online mtDNA-server ( tool. With this tool, homo- and heteroplasmy variants were identified. The pathogenicity status of heteroplasmic variants were assessed using various clinical and online databanks e.g. MitoMap, and the application of accepted scoring criteria41,42. For all other analyses described below, only homoplasmic variants and those with a heteroplasmy level above 90% were used, thus only inherited and not somatic variants were considered. Haplogroups were assigned using the online Haplogrep 2.0 tool ( The control data from the UK is Sanger Sequence data, the processing of which is described in44.

Classification and selection of mtDNA variants, as compared by group

Analysis using variants predicted to be mildly deleterious are less prone to the effects of population stratification because they typically analyse rare variants that are not stratified between geographical locations. The variants are assessed using the MutPred program, which assigns a “pathogenicity” score between 0–1 (an amino acid change with 0 is predicted to be perfectly benign). A score above 0.5 for an amino acid change is classified as an “actionable hypothesis”14. Variants with pathogenicity scores below the “actionable hypothesis” threshold (0.5) are considered less likely to be deleterious or to have an impact on protein function; instead, they are more likely to be common population variants. The inclusion of these more numerous but low-scoring, low-impact variants in the analyses could be problematic, especially because the resultant “noise” differs greatly among different population groups45. Therefore, as in some of our previous studies, we have not included variants scoring below 0.5 in the current analyses22.

Statistical analyses

All statistical analyses were performed using SPSS Statistics (Version 25), Prism (Version 23) or GraphPad (Version 7). Haplogroup distribution between ME/CFS patient cohorts and their corresponding control were performed using a Monte Carlo based approach, this methodology is part of the standard package but more accurate than the Chi Square estimation. Fisher’s exact tests were utilised to compare the number of patients and controls in each cohort that have mtDNA variants with MutPred scores above 0.5 (mildly deleterious) with those who do not.

Data Availability

Whole mtDNA sequences for the South African controls have been deposited in the sequence read archive (SRA) of NCBI under BioProject accession number PRJNA403942 ( The UK controls can be found as electronic supplemental material.


  1. 1.

    Fukuda, K. et al. The chronic fatigue syndrome: a comprehensive approach to its definition and study. International Chronic Fatigue Syndrome Study Group. Ann Intern Med. 121, 953–959 (1994).

    CAS  Article  Google Scholar 

  2. 2.

    Pendergrast, T. et al. Housebound versus nonhousebound patients with myalgic encephalomyelitis and chronic fatigue syndrome. Chronic Illn 28, 1742395316644770 (2016).

    Google Scholar 

  3. 3.

    Jones, D. E. et al. Loss of capacity to recover from acidosis on repeat exercise in chronic fatigue syndrome: a case-control study. Eur J Clin Invest. 42, 186–194, Epub02011 Jul 02512 (2012).

  4. 4.

    Brown, A. E., Jones, D. E., Walker, M. & Newton, J. L. Abnormalities of AMPK activation and glucose uptake in cultured skeletal muscle cells from individuals with chronic fatigue syndrome. PLoS One. 10, e0122982, 0122910.0121371/journal.pone.0122982. eCollection0122015 (2015).

  5. 5.

    Tomas, C. et al. Cellular bioenergetics is impaired in patients with chronic fatigue syndrome. PLoS One. 12, e0186802, 0186810.0181371/journal.pone.0186802. eCollection0182017 (2017).

  6. 6.

    Cairns, R. & Hotopf, M. A systematic review describing the prognosis of chronic fatigue syndrome. Occup Med (Lond). 55, 20–31, (2005).

    CAS  Article  PubMed  Google Scholar 

  7. 7.

    Gorman, G. S. et al. Prevalence of nuclear and mitochondrial DNA mutations related to adult mitochondrial disease. Ann Neurol. 77, 753–759, 710.1002/ana.24362. Epub22015 Mar 24328 (2015).

  8. 8.

    Gorman, G. S. et al. Perceived fatigue is highly prevalent and debilitating in patients with mitochondrial disease. Neuromuscul Disord. 25, 563–566, Epub2015 Apr 1023 (2015).

  9. 9.

    Ban, M. et al. Investigation of the role of mitochondrial DNA in multiple sclerosis susceptibility. PLoS One. 3, e2891, (2008).

    ADS  CAS  Article  PubMed  PubMed Central  Google Scholar 

  10. 10.

    Billing-Ross, P. et al. Mitochondrial DNA variants correlate with symptoms in myalgic encephalomyelitis/chronic fatigue syndrome. J Transl Med. 14 : 19, (2016).

  11. 11.

    Elson, J. L., Turnbull, D. M. & Howell, N. Comparative genomics and the evolution of human mitochondrial DNA: assessing the effects of selection. Am J Hum Genet. 74, 229–238. Epub 2004 Jan 2007 (2004).

  12. 12.

    Pereira, L., Soares, P., Radivojac, P., Li, B. & Samuels, D. C. Comparing phylogeny and the predicted pathogenicity of protein variations reveals equal purifying selection across the global human mtDNA diversity. Am J Hum Genet. 88, 433–439, Epub 2011 Mar1031 (2011).

  13. 13.

    Soares, P. et al. Evaluating purifying selection in the mitochondrial DNA of various mammalian species. PLoS One 8, e58993, Epub0052013 Mar 0058922 (2013).

  14. 14.

    Li, B. et al. Automated inference of molecular mechanisms of disease from amino acid substitutions. Bioinformatics. 25, 2744–2750, Epub2009 Sep 2743 (2009).

  15. 15.

    Elson, J. L. et al. Does the mitochondrial genome play a role in the etiology of Alzheimer’s disease? Hum Genet. 119, 241–254. Epub 2006 Jan 2012 (2006).

  16. 16.

    Bregman, J. A. et al. Mitochondrial Haplogroups Affect Severity But Not Prevalence of Diabetic Retinopathy. Invest Ophthalmol Vis Sci. 58, 1346–1351, (2017).

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  17. 17.

    Elson, J. L., Majamaa, K., Howell, N. & Chinnery, P. F. Associating mitochondrial DNA variation with complex traits. Am J Hum Genet. 80, 378–382; author reply 382-373 (2007).

  18. 18.

    Salas, A. & Elson, J. L. Mitochondrial DNA as a risk factor for false positives in case-control association studies. J Genet Genomics. 42, 169–172, Epub 2015 Mar1017 (2015).

  19. 19.

    Chinnery, P. F., Elliott, H. R., Syed, A. & Rothwell, P. M. Mitochondrial DNA haplogroups and risk of transient ischaemic attack and ischaemic stroke: a genetic association study. Lancet Neurol. 9, 498–503, Epub72010 Mar 70031 (2010).

  20. 20.

    Yu, X. et al. mtDNA nt13708A variant increases the risk of multiple sclerosis. PLoS One. 3, e1530, (2008).

    ADS  CAS  Article  PubMed  PubMed Central  Google Scholar 

  21. 21.

    Pienaar, I. S., Howell, N. & Elson, J. L. MutPred mutational load analysis shows mildly deleterious mitochondrial DNA variants are not more prevalent in Alzheimer’s patients, but may be under-represented in healthy older individuals. Mitochondrion 7, 30194–30195 (2017).

    Google Scholar 

  22. 22.

    Venter, M., Malan, L., van Dyk, E., Elson, J. L. & van der Westhuizen, F. H. Using MutPred derived mtDNA load scores to evaluate mtDNA variation in hypertension and diabetes in a two-population cohort: The SABPA study. J Genet Genomics. 44, 139–149, Epub2016 Dec 1026 (2017).

  23. 23.

    Weissensteiner, H. et al. mtDNA-Server: next-generation sequencing data analysis of human mitochondrial DNA in the cloud. Nucleic Acids Res. 44, W64–69, Epub2016 Apr 1015 (2016).

  24. 24.

    Zifa, E. et al. A novel G3337A mitochondrial ND1 mutation related to cardiomyopathy co-segregates with tRNALeu(CUN) A12308G and tRNAThr C15946T mutations. Mitochondrion. 8, 229–236, Epub2008 May 1027 (2008).

  25. 25.

    Hudson, G. et al. Clinical expression of Leber hereditary optic neuropathy is affected by the mitochondrial DNA-haplogroup background. Am J Hum Genet. 81, 228–233. Epub 2007 Jun 2004 (2007).

  26. 26.

    Elliott, H. R., Samuels, D. C., Eden, J. A., Relton, C. L. & Chinnery, P. F. Pathogenic mitochondrial DNA mutations are common in the general population. Am J Hum Genet. 83, 254–260, (2008).

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  27. 27.

    Schoeman, E. M. et al. Clinically proven mtDNA mutations are not common in those with chronic fatigue syndrome. BMC Med Genet. 18, 29, (2017).

    Article  PubMed  PubMed Central  Google Scholar 

  28. 28.

    Torroni, A. et al. A signal, from human mtDNA, of postglacial recolonization in Europe. Am J Hum Genet. 69, 844–852. Epub 2001 Aug 2021 (2001).

  29. 29.

    van Oven, M. & Kayser, M. Updated comprehensive phylogenetic tree of global human mitochondrial DNA variation. Hum Mutat. 30, E386–394, (2009).

    Article  PubMed  Google Scholar 

  30. 30.

    De Menezes, E. C. S. et al. Mitochondrial DNA copy number is not associated with fatigue status in Primary Sjögren’s Syndrome. Fatigue: Biomedicine, Health & Behavior 6, 123–131, (2018).

    Article  Google Scholar 

  31. 31.

    Herrnstadt, C. & Howell, N. An evolutionary perspective on pathogenic mtDNA mutations: haplogroup associations of clinical disorders. Mitochondrion. 4, 791–798. Epub 2004 Oct 2001 (2004).

  32. 32.

    Samuels, D. C., Carothers, A. D., Horton, R. & Chinnery, P. F. The power to detect disease associations with mitochondrial DNA haplogroups. Am J Hum Genet. 78, 713–720. Epub 2006 Feb 2017 (2006).

  33. 33.

    Lawson, N., Hsieh, C. H., March, D. & Wang, X. Elevated Energy Production in Chronic Fatigue Syndrome Patients. J Nat Sci 2(10), e221 (2016).

    PubMed  PubMed Central  Google Scholar 

  34. 34.

    Reinecke, F., Smeitink, J. A. & Van der Westhuizen, F. H. OXPHOS gene expression and control in mitochondrial disorders. Biochim Biophys Acta. 1792, 1113–1121, Epub 2009 Apr 1121 (2009).

  35. 35.

    Bothma K, van der Westhuizen, F. H. & Louw, R. Metabolomics of Mitochondrial Disease. Mitochondrion 97–110 (2017).

  36. 36.

    Tomas, C. & Newton, J. Metabolic abnormalities in chronic fatigue syndrome/myalgic encephalomyelitis: a mini-review. Biochem Soc Trans. 46, 547–553, Epub 20172018 Apr 20170517 (2018).

  37. 37.

    Bury, A. G. et al. Mitochondrial DNA changes in pedunculopontine cholinergic neurons in Parkinson disease. Ann Neurol. 82, 1016–1021, Epub22017 Dec 25094 (2017).

  38. 38.

    Lowes, H., Pyle, A., Duddy, M. & Hudson, G. Cell-free mitochondrial DNA in progressive multiple sclerosis. Mitochondrion 8, 30101–30106 (2018).

    Google Scholar 

  39. 39.

    Varhaug, K. N. et al. Increased levels of cell-free mitochondrial DNA in the cerebrospinal fluid of patients with multiple sclerosis. Mitochondrion. 34:32–35, Epub2016 Dec 1023 (2017).

  40. 40.

    Bandelt, H. J., Forster, P., Sykes, B. C. & Richards, M. B. Mitochondrial portraits of human populations using median networks. Genetics. 141, 743–753 (1995).

    CAS  PubMed  PubMed Central  Google Scholar 

  41. 41.

    Yarham, J. W. et al. A Comparative Analysis Approach to Determining the Pathogenicity of Mitochondrial tRNA Mutations. Human Mutation 32, 1319–1325, (2011).

    CAS  Article  PubMed  Google Scholar 

  42. 42.

    Yarham, J. W., McFarland, R., Taylor, R. W. & Elson, J. L. A proposed consensus panel of organisms for determining evolutionary conservation of mt-tRNA point mutations. Mitochondrion. 12, 533–538, Epub2012 Jul 1017 (2012).

  43. 43.

    Weissensteiner, H. et al. HaploGrep 2: mitochondrial haplogroup classification in the era of high-throughput sequencing. Nucleic Acids Res. 44, W58–63, Epub2016 Apr 1015 (2016).

  44. 44.

    Herrnstadt, C. et al. Reduced-median-network analysis of complete mitochondrial DNA coding-region sequences for the major African, Asian, and European haplogroups. Am J Hum Genet. 70, 1152–1171. Epub 2002 Apr 1155 (2002).

  45. 45.

    Behar, D. M. et al. A “Copernican” reassessment of the human mitochondrial DNA tree from its root. Am J Hum Genet. 90, 675–684, (2012).

    CAS  Article  PubMed  PubMed Central  Google Scholar 

Download references


This work received funding support from Action for ME, The ME Association and the North-West University, South Africa.

Author information




J.L.E., I.S.P., N.H. and M.V. conceived the strategy. J.L.E. and F.H.v.W. directed the project. M.V., C.T. and F.H.v.W. designed the experiments. V.S., E.E., W.F.N. and J.L.N. gave clinical oversight, M.V., C.T. and I.S.P. conducted experiments. M.V., F.H.v.W. and J.L.E. analysed data. M.V. and J.L.E. wrote the manuscript. C.T., I.S.P., W.F.N., N.H. and F.H.v.W. revised the manuscript. All Authors read and approved the final manuscript.

Corresponding author

Correspondence to Joanna L. Elson.

Ethics declarations

Competing Interests

The authors declare no competing interests.

Additional information

Publisher’s note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Venter, M., Tomas, C., Pienaar, I.S. et al. MtDNA population variation in Myalgic encephalomyelitis/Chronic fatigue syndrome in two populations: a study of mildly deleterious variants. Sci Rep 9, 2914 (2019).

Download citation

Further reading


By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.


Quick links

Nature Briefing

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing