We welcome the comments from Finsterer (1) on our article (2) proposing a new perspective on the mechanisms underlying the development of diabetes in maternally inherited diabetes and deafness (MIDD) with a focus on both β-cell function and insulin sensitivity.
First, although it may seem intuitive, we disagree that impaired function of complexes I, III, and IV may be sufficient to lead to β-cell dysfunction. There are numerous other primary oxidative phosphorylation deficiencies featuring severe reduction of ATP that do not commonly result in diabetes. As appropriately noted by Finsterer (1), the variability of the MIDD phenotype likely results from the presence of additional factors, whether environmental or genetic. Hyperstimulation by elevated blood glucose, poor antioxidant defenses, and other factors may also affect the fate of the β-cell. In addition, despite overlap between the phenotype of MIDD and mitochondrial encephalomyopathy, lactic acidosis, and strokelike episodes (MELAS), MELAS is rarely associated with diabetes. The severe phenotype of MELAS is thought to be at least in part secondary to a greater percentage of heteroplasmy. If the hypothesis proposed by Finsterer (1) was correct, would we not expect that oxidative phosphorylation would be severely affected in the β-cell and that diabetes would be systematically present in MELAS?
Second, we do not believe that the mTORC1 hyperactivity reported in cells affected by m.3243A>G goes against our hypothesis. We agree that the role of mTORC1 activity in the β-cell is complex and incompletely understood (3). We suggest that the higher heteroplasmy driven by mTORC1 activation that was reported in vitro by two groups of authors (4,5) may stimulate mitophagy of the most severely affected mitochondria and may represent a defense mechanism against the deleterious effects of the phenotypic variant. This needs to be confirmed in vivo in the context of additional modifiers (see below). Central to our hypothesis is the dual role of β-cell dysfunction and decreased insulin sensitivity in the development of diabetes in MIDD. The role of mTORC1 hyperactivity, which was also shown to decrease insulin sensitivity (6), thus should be considered in the context of the overall underlying mechanism of glucose metabolism dysfunction. This may include additional, mTORC1-independent pathways, as suggested by Yang et al. (7). Our suggestion that diabetes in MIDD results from a more severe heteroplasmy in the skeletal muscle and a less severe heteroplasmy in the β-cell of the pancreas mostly comes from the very limited published human data listed in our article.
Third, although the term “haplotype” was not specifically used, the role of other factors was made very clear in our proposed model section, including “interaction between the mitochondrial and nuclear genomes . . . concomitant nuclear risk factors for type 2 diabetes, additional nuclear or mitochondrial variants, mtDNA copy number, age, sex, and environmental factors, decreased physical activity . . . influence of the paternal genome, and metabolic characteristics” of the individual. In vitro experiments offer the advantage of investigating a specific consequence of the phenotypic variant but have the disadvantage of not considering other potential modifiers, such as the ones mentioned in our article. Integrating the in vivo role of these modifiers is difficult, but we agree that these factors likely explain the extreme phenotypic variability of this monogenic condition.
We fully agree with the fourth point raised by Finsterer (1). Diabetes and hearing impairment are only two of the many potential manifestations of MIDD. Some of the manifestations are shared with MELAS, which is associated with the same phenotypic variant. This has been described in many articles and was made clear in our article. Our perspective was, however, fully dedicated to diabetes, which may be the only MIDD manifestation consistent with a two-hit model (β-cells and insulin-sensitive tissues). The rationale is that diabetes in MIDD is often misdiagnosed as type 1 or type 2 diabetes in individuals with MIDD. This is particularly important when considering that cardiovascular disease and kidney disease are present independently in both mitochondrial disease and diabetes. Diabetes in MIDD and its complications may therefore benefit from a different management approach compared with type 1 and type 2 diabetes.
Article Information
Duality of Interest. No potential conflicts of interest relevant to this article were reported.
