Mitochondrial DNA: Innocent in Plasma, but Guilty in Urine?

To the Editor: With interest we read the paper by Jansen et al. (1), who report that, although plasma mtDNA levels are elevated in patients with noninfectious SIRS, they do not correlate with markers of systemic inflammation or renal injury. These data are corroborated by subsequent murine experiments, where administration of mtDNA or a cocktail of mitochondrial DAMPs (MTD) neither induced systemic inflammation or organ injury, nor worsened renal function. These results are in line with findings of our group, demonstrating that mtDNA plasma levels are elevated in critical illnesses, e.g., sepsis (2), trauma (3), and cardiac arrest (4), but do not correlate with markers of inflammation, immune dysfunction, and organ damage. Therefore, we share the authors’ conclusion that plasma mtDNA does not seem to play an important causative role in these conditions, which could possibly be explained by RBC-mediated DNA scavenging mechanisms limiting mtDNA availability in the circulation (5). So is there still a role for mtDNA outside of the plasma, for instance, in the kidney? DAMPs are filtered in the glomeruli and pattern recognition receptors are expressed in the tubules (6). Interestingly, Jansen et al. show that urinary mtDNA levels are elevated in SIRS patients with AKI and, more importantly, correlate with the severity of renal dysfunction and inflammation. Subsequently, the authors performed in vitro studies in primary tubular epithelial cells (pTECs) aimed at establishing a causative link between mtDNA and renal inflammation. Their key experiment reveals that stimulation of pTECs with necrotic supernatants from mtDNA-bearing pTECs induces production of the chemokine KC; however, this was not the case for the chemokine MCP-1. Furthermore, the used methodology does not permit us to draw definitive conclusions for several reasons. First, KC production, albeit to a lesser degree, was also observed when pTECs were stimulated with necrotic supernatants from a mtDNA-depleted fibroblast cell line. Second, apart from the presence/absence of mtDNA, there are many more differences between mtDNA-bearing pTECs and fibroblasts that could explain the observed effects. It would have been more straightforward and relevant if the authors would have stimulated pTECs with the purified mtDNA or MTD they used in mice, or used necrotic supernatants from mtDNA-bearing fibroblasts instead of necrotic supernatants from pTECs. In conclusion, Jansen et al. once more highlight that mtDNA in plasma seems to be innocent as a key player involved in the inflammatory cascade in critical ill patients, but may be guilty of inflicting renal injury. Therefore, the important question that Martin and Thiemermann also pose in their accompanying editorial (7) needs to be answered: is there a causative role for mtDNA in renal injury or is it merely another urinary marker reflecting kidney damage? In that respect, it would be interesting to measure urinary levels of other DAMPs in the patient cohort of Jansen et al., such as nuclear DNA or HMGB1, to shed more light on whether mtDNA truly plays a detrimental role in the kidneys, or if it proves to be an innocent bystander once again.