Because mtDNA content was improved in PARP1-/- (vs. WT) chagasic mice, we considered that PARP1/PAR constitute a risk factor by adversely affecting the mtDNA replication. The mRNA levels for the components of the mtDNA replication/repair complex, including TOP1mt, POLRMT, and SSPB1, as well as the genes of the mitochondrial transcription machinery including TFB1M and TFB2M were slightly decreased or not changed in the myocardium of WT, PARP1+/- and PARP1-/- mice chronically infected with T. cruzi (vs. matched controls, Fig 3A–3F). Only POLG exhibited 35% decline at mRNA level (Fig 3D) and a significant decline at mitochondrial protein level in the myocardium of WT.Tc (vs. WT) mice (Fig 3G & 3H, p<0.05). The POLG mRNA level was preserved to normal level and mitochondrial POLG protein level was substantially preserved in PARP1-/-.Tc (vs. WT.Tc) mice (Fig 3D, 3G & 3H). No significant decline in the mitochondrial levels of other proteins of the mtDNA biogenesis machinery, including LIG3, TWNK, RNASEH1, TFAM, POLG2, and TFB2M was observed in cardiac mitochondria of chronically infected WT and PARP1-/- (vs. matched controls) mice (Fig 3I & 3J). Immunoprecipitation/Western blotting (IP/WB) showed that the direct binding of POLG to PARP1 (but not to TWNK and SSBP1) was increased in cardiac mitochondria of chagasic WT mice (Fig 3K). The POLG was pulled down with high efficiency as is evidenced by WB finding of all of the POLG in binding fraction and none in unbinding fractions (Fig 3K, bottom panel). We also monitored PARP1/POLG interaction in HeLa cells infected with T. cruzi. Cross-IP/WB showed that POLG pull-down of PARP1 (and vice versa) was increased in mitochondria of infected (vs. normal) HeLa cells (Fig 3L). Treatment of HeLa cells with 50 ?M H2O2 did not enhance POLG/PARP1 interaction to the same extent as was noted with Tc infection (Fig 3L). Chromatin immunoprecipitation with anti-PARP1 antibody followed by PCR showed that PARP1 binding to mtDNA sequence (7S near D-loop) and nuclear sequence (COIV) was also increased in infected HeLa cells (Fig 3M). No changes in input 7S mtDNA and GAPDH nuDNA were noted (Fig 3M, bottom panels). Together, the results presented in Fig 3, along with those presented in Fig 2, suggest that a) nuclear PARP1 does not affect the expression of a majority of the components of the mtDNA replication machinery in chagasic myocardium. However, b) mitochondrial PARP1 binding to POLG and mtDNA was increased in the myocardium of chronically infected mice, and c) genetic depletion of PARP1 was beneficial in preserving the POLG-dependent mtDNA content in cardiac mitochondria of chagasic mice.