PARP1 utilizes NAD+ as substrate and attaches ADP-ribose units to amino acids on various acceptor proteins. The nuclear PARP1/PAR was discovered more than fifty years . In the nucleus, PARP1 activity is mainly targeted towards PARylation of lysine, arginine, and cysteine residues, and it participates in nuclear DNA repair through PARylation of histone proteins . Recent studies show that major histone proteins, H2A and H2B involved in nuclear DNA repair, are also present in the inner mitochondrial membranes , and intra-mitochondrial localization of PARP1 and PARP1 activity were increased in cultured cells exposed to genotoxic stimuli (e.g. ROS) . However, investigators have largely ignored studying the role of mtPARP1/PAR in health and disease. In this study, we have found that the expression, activity, and intra-mitochondrial localization of PARP1/PAR were increased in cardiac biopsies of chagasic mice and in human cardiomyocytes infected by Tc. While the expression of a majority of the components of the mtDNA replisome machinery was not significantly altered in chagasic (vs. control) mice, PARP1 binding to POLG and mtDNA was increased and associated with a loss in mtDNA content, mtDNA-encoded gene expression, and OXPHOS capacity, and an increase in mitochondrial ROS production in chronically infected murine myocardium. Importantly, Inhibition of PARP1 by genetic deletion or treatment with a chemical antagonist had protective effects against cardiac hypertrophy and LV dysfunction in chagasic heart. The beneficial effects of PARP1 inhibition were not delivered via its direct effect on parasite persistence that is suggested to be the major cause for the development of chronic cardiomyopathy (reviewed in [29,30]). Instead, PARP1 depletion preserved the POLG-dependent mtDNA content, mitochondrial function, and antioxidant/oxidant balance in chagasic myocardium and human cardiomyocytes infected by T. cruzi. Taken together, we have demonstrated that the mitochondrial transport of PARP1/PAR adversely impacts the mtDNA maintenance by POLG replisome, and exacerbates the mitochondrial dysfunction, oxidative stress and cardiac remodeling in Chagas disease. We propose that PARP1 inhibitors will be beneficial in preserving the mitochondrial health and LV function in chronic cardiomyopathy of chagasic (and potentially other) etiologies.
It is logical to expect that mitochondrial PARP1, via its capacity to carry out PARylation, can modulate mitochondrial function. A proteomic study has indeed reported an increase in PARylation of a variety of mitochondrial proteins in response to traumatic brain injury in rodents , and Modis et al  have found mitochondrial respiratory deficiency in A549 human epithelial cells overexpressing mitochondria targeted PARP1. In our in vivo model of Chagas disease, cytosolic, nuclear, as well as mitochondrial PARP/PAR were increased in the myocardium of infected mice. The organelle-specific deletion or overexpression of PARP1 in vivo is not yet feasible, and thus it is difficult to dissect the role of nuclear (vs. mitochondrial) PARP1/PAR in health and disease. Nevertheless, we noted that the nuclear-DNA encoded proteins of the respiratory complexes as well as majority of the proteins of POLG replisome were not altered in expression in chagasic (vs. control) heart. This observation suggests that nuclear PARP1/PAR is either beneficial or not detrimental in Chagas disease. Instead, the increase in intra-mitochondrial PARP1/PAR was associated with mitochondrial electron transport chain dysfunction, and decreased the CII-complex supported ATP synthesis. Further, the dysregulation of mitochondrial electron transport chain resulted in mitochondrial uncoupling, and produced a secondary increase in ROS by mitochondria. PARP1/PAR inhibition, in a dose dependent manner, was beneficial in rescuing the mitochondrial electron transport chain activity and OXPHOS capacity in chagasic heart. Sirtuin 1 is a highly conserved member of the family of NAD+-dependent Sir2 histone deacetylases , and it competes with PARP1 for NAD+ substrate. Bai et al showed that PARP1 or PARP2 deficiency enhanced the mitochondrial oxidative metabolism, energy expenditure, and protection against diet-induced obesity, and this outcome was achieved, at least partially, via increased availability of NAD+ for SIRT1 activation [34,35]. However, we did not observe the beneficial effects of treatment with a SIRT1 agonist (SRT1720) in improving the mitochondrial function in chagasic mice . Thus, it is unlikely that benefits of PARP1 inhibition in rescuing mitochondrial oxidative metabolism in chagasic mice were delivered via increase in SIRT1 activity. Our finding of increased PARP1/PAR suggest that mitochondrial PARP1 may have had a direct effect on OXPHOS capacity through PARylation of various complexes of the electron transport chain in chagasic WT mice, as was observed in ovarian epithelial cancer cells .
The close proximity to the electron transport chain exposes the mtDNA to endogenous ROS, and thus, mtDNA may accumulate 20-100-fold higher level of oxidative adducts and DNA breaks than might be noted in nuclear DNA under similar conditions. Indeed, we have noted increased mtDNA damage in human cardiomyocytes in vitro infected with T. cruzi and in cardiac biopsies of chagasic patients . Two mitochondrial enzymes, POLG and ExoG, carry out DNA base excision repair/single-strand breaks repair (BER/SSBR) of mtDNA . POLG also assembles the mtDNA replisome, and carries the burden of mtDNA replication . Whether PARP1 independently, or in conjunction with POLG and ExoG, is involved in mtDNA repair and replication in health or disease has not been studied so far. The first evidence that PARP1/PAR may exert a negative effect on mtDNA was provided by the observation of lower number of mtDNA mutagenic lesions in PARP1-/- (vs. PARP1+/+) cells . Szabo and coworkers showed that PARP1 depletion by shRNAi led to faster recovery of mtDNA integrity after initial oxidative insult in A549 epithelial cells . In this study, we provide the first in vivo evidence that overexpression of PARP1 was detrimental to mtDNA maintenance and contributed to increased severity of chagasic cardiomyopathy. The PARylation of POLG or other proteins of the POLG replisome was not increased in chagasic heart, as was speculated to be the cause for deficiency of mtDNA repair in A549 cells exposed to oxidative stress . PARP1 binding to LIG3 or other proteins of the DNA repair machinery or change in the expression of the components of the mtDNA replication and transcription machinery was also not noted in chagasic myocardium. Instead, we noted that PARP1 direct binding to POLG and mtDNA was increased and the levels of POLG and intact mtDNA were decreased in cardiac mitochondria of chagasic WT mice. Treatment of chagasic mice with PJ34 (PARP1 chemical inhibitor) or genetic depletion of PARP1 restored the POLG level and intact mtDNA content in chagasic heart. Further studies will be required to delineate if the PARP1 depletion restored the mtDNA repair or mtDNA replication activities of the POLG complex. Yet, our studies allow us to surmise that chemical inhibition of PARP1 by PJ34 treatment or genetic depletion of PARP1 was beneficial in controlling the mtROS and cardiac remodeling and restoring the mitochondrial oxidative metabolism as well as LV function in chagasic heart.
The pathologic features of the chronic Chagas cardiomyopathy also include focal areas of inflammation constituted by T cells and macrophages with a few eosinophils, plasma cells, and mast cells (reviewed in ). Parasites are rarely observed by conventional microscopic analysis of cardiac biopsies or random sections of explants or autopsy specimens during chronic Chagas disease . It is, therefore, imperative that research community pays attention to host-derived factors as modulators of inflammatory responses in chronic Chagas disease. In this context, it is important to note that PARP1/PAR have been implicated in driving inflammation induced by cytotoxic agents (e.g. arsenite ). PARP1 suppression by genetic deletion or pharmacological inhibitors have been shown to be beneficial in reducing LPS-induced lung inflammation  and gut inflammation  in mice and humans. We have shown PARP1-dependent post-translational modification of Rel A (p65)-interacting nuclear proteins facilitated the assembly of NF?B transcription complex and cytokine gene expression in cardiomyocytes infected by T. cruzi . Further, Sirtuin 1 (SIRT1), a highly conserved member of the family of NAD+-dependent Sir2 histone deacetylases, is known to compete with PARP1 for NAD+ substrate, and integrate mitochondrial metabolism and inflammation [45,46,47]. Treatment of mice with the SIRT1 agonist SRT1720 suppressed NF?B-dependent transcription and inflammatory stress in T. cruzi-infected cells and chagasic mice . Thus, it is feasible that the observed benefits of PARP1 depletion in improving cardiac outcomes are not only due to improvement of mitochondrial function but also due to control of chronic inflammation in chagasic mice. We propose that while mitochondrial PARP1 is detrimental to maintaining the mtDNA integrity as observed in this study, nuclear PARP1 contributes to chronic inflammation in Chagas disease. Further studies will be required to delineate the potential role of nuclear PARP1 in driving chronic inflammation in Chagas disease.
In summary, we have used in vitro and in vivo models of T. cruzi infection and Chagas disease, and demonstrated that mitochondrial PARP1/PAR disturbs the POLG-dependent mtDNA integrity, and contributes to loss in mitochondrial function. Inhibition of mitochondrial PARP1/PAR offers a novel therapy in preserving the mitochondrial and LV function in chronic Chagas disease.