Placental mitochondrial adaptations in preeclampsia associated with progression to term delivery

Abstract

Preeclampsia is a devastating pregnancy disorder. Severity varies widely, and while severe preeclampsia often requires pre-term delivery, women with mild preeclampsia may reach term with minor interventions. The mechanisms that mediate disease severity are poorly understood, but may include adaptive processes by the placenta. We aimed to establish whether in pregnancies that reached term and those that delivered pre-term, the placental response to preeclampsia was intrinsically different, and explore potential adaptive mechanisms. Hydrogen peroxide production and antioxidant activity were increased in term preeclamptic placentae, whereas pre-term preeclamptic placentae had reduced hydrogen peroxide production and reduced function of the antioxidant system superoxide dismutase compared to control placentae. Markers of mitochondrial fission/fusion, apoptosis and the expression level of mitochondrial complexes were differentially disrupted in term compared to pre-term preeclamptic placentae. Mitochondrial respiration and content were increased in term preeclamptic placentae, but mitochondria had a lower respiratory reserve capacity. Mitochondrial respiration and hydrogen peroxide production were increased in healthy term placentae after in vitro hypoxia/reoxygenation. Placentae from preeclamptic pregnancies that reached term showed multiple adaptions that were not present in pre-term preeclamptic placentae. Increased antioxidant activity, and expression of markers of mitochondrial fusion and apoptotic suppression, may relate to salvaging damaged mitochondria. Increased mitochondrial respiration may allow ongoing tissue function even with reduced respiratory efficiency in term preeclamptic pregnancies. Response after in vitro hypoxia/reoxygenation suggests that disruption of oxygen supply is key to placental mitochondrial adaptations. Reactive oxygen species signalling in term preeclamptic placentae may be at a level to trigger compensatory antioxidant and mitochondrial responses, allowing tissue level maintenance of function when there is organelle level dysfunction.

Fig. 1. Total antioxidant activity increases in term preeclamptic placentae and superoxide dismutase function is reduced in pre-term preeclamptic placentae.

a Hydrogen peroxide (H₂O₂) level was affected by preeclampsia (p = 0.0295), gestation (p = 0.0186) and interaction factor (p = 0.0074), with levels decreased in pre-term preeclamptic placentae compared to pre-term controls (p = 0.0014, fold change 0.244). b Total antioxidant activity was affected by preeclampsia (p = 0.0024) and gestation (p < 0.0001), with activity increased in term preeclamptic placentae compared to term controls (p = 0.0171, fold change 1.130). c Soluble superoxide dismutase 1 (SOD1) mRNA expression was affected by gestation (p = 0.0130), and the interaction of preeclampsia/gestation (p = 0.0329). d Total superoxide dismutase (SOD) activity over all cellular compartments was affected by gestation (p < 0.0001). e Mitochondrial superoxide dismutase 2 (SOD2) mRNA expression was affected by preeclampsia (p = 0.0310) and gestation (p = 0.0001). f SOD activity in enriched mitochondrial fractions was affected by preeclampsia (p = 0.0305) and gestation (p = 0.0100), and was reduced in pre-term preeclamptic placentae compared to pre-term controls (p = 0.0277, fold change 0.794). g Catalase activity was affected by gestation (p = 0.0010). h Glutathione peroxidase (GPx) activity was affected by gestation (p < 0.0001) and increased in term preeclamptic placentae compared to term controls (p = 0.0357, fold change 1.244). i Nuclear factor kappa-light-chain-enhancer of activated B cells (NF-κB) was affected by preeclampsia (p = 0.0164), gestation (p < 0.0001), and interaction factor (p = 0.0095), and was increased in pre-term preeclamptic placentae (p = 0.0077, fold change 1.599). j mRNA expression of ADAM metallopeptidase domain 10 (ADAM10) was not different between groups. βAbeta actin. Representative western blot images. N = 5–20; *P < 0.05; **P < 0.01; ***P < 0.001

Fig. 2. Mitochondrial fission/fusion signalling is disrupted differentially in term and pre-term preeclamptic placenta.

a Diagram of proteins involved in mitochondrial fusion/fission processes. b mRNA expression, (c) total protein and (d) mitochondrial protein fractions of pro-fission dynamin-1-like protein (mRNA = DNM1L; protein = DRP1) were unchanged between control and preeclamptic placentae, with DRP1 affected by gestation (p = 0.0122). e Mitochondrial fission 1 protein (FIS1) was affected by preeclampsia (p = 0.0145) and gestation (p = 0.0008), and was decreased in term preeclamptic placentae (p = 0.0384, fold change 0.406). f mRNA expression of pro-fusion mitochondrial dynamin like GTPase (OPA1 /OPA1) was affected by the interaction of preeclampsia and gestation (p = 0.0288). g The ratio of the OPA1 long isoform to short isoform (L-OPA1/S-OPA1) was affected by gestation (p = 0.0013) and interaction factor (p = 0.0147), and was increased in term preeclamptic placentae (p = 0.0207, fold change 1.617) and unchanged in pre-term preeclamptic placentae. h Mitofusin 1 (MFN1) was affected by preeclampsia (p = 0.0436) and gestation (p = 0.0001) and interaction factor (p = 0.0099), and was increased in term preeclamptic placentae (p = 0.007, fold change 1.778), (i) with no difference in mitofusin 2 (MFN2). MAMmitochondria associated membrane; βAbeta actin; CVmitochondrial complex V. Representative western blot images. N = 6–7; *P < 0.05; **P < 0.01; ***P < 0.001

Fig. 3. Mitochondrial apoptotic signalling is disrupted differentially in term and pre-term preeclamptic placenta.

a Diagram of proteins involved in mitochondrial apoptotic signalling. b Pro-apoptotic cleaved caspase 3 (CASP3) was affected by gestation (p = 0.0014) and interaction factor (p = 0.0403). c Pro-apoptotic Bcl-2-associated X protein (BAX) was affected gestation (p = 0.0008) and interaction factor (p = 0.0101), and was reduced in term preeclamptic placentae (p = 0.0356, fold change 0.5394). d Anti-apoptotic B-cell lymphoma 2 (BCL2) was increased in term preeclamptic placentae (p = 0.0156, fold change 1.526). e The BAX/BCL2 ratio was affected by preeclampsia (p = 0.0093), gestation (p < 0.0001) and interaction factor (p < 0.0001), and was decreased in term preeclamptic placentae (p < 0.0001, fold change 0.4044) and increased in pre-term preeclamptic placentae (p = 0.0334, fold change 1.763). f Pro-apoptotic cleaved caspase 9 (CASP9) did not change. g Representative western blot images. h mRNA expression of CASP3 was not different between groups. mRNA expression of (i) X-linked inhibitor of apoptosis protein (XIAP), (j) caspase 8 (CASP8), and (k) BH3 interacting-domain death agonist (BID) were affected by gestation (p = 0.0172, 0.0004, 0.0136, respectively). BAP31B-cell-receptor-associated protein 31; Cacalcium; CASP10caspase 10; CYTCcytochrome c; βAbeta actin. N = 6–8; *P < 0.05; **P < 0.01; ***P < 0.001

Fig. 4. Mitochondrial complex levels and respiration are dysregulated in preeclamptic placentae.

a Protein levels of mitochondrial complexes II and III were increased in preeclamptic placentae from term pregnancies compared to control placentae from term pregnancies (p = 0.0360, 0.0369, fold change 1.576, 1.413, respectively). Complex II was affected by gestation (p = 0.0125) and interaction factor (p = 0.0478), complex III was affected by gestation (p = 0.0401), and complex V (ATP synthase) was affected by gestation (p = 0.0370). There was no change in the protein level of complexes I, IV and V in placentae from term pregnancies, or all complexes in placentae from pre-term pregnancies. b Representative western blot images of mitochondrial complexes. c Citrate synthase activity was affected by gestation (p < 0.0001, fold change 0.495 pre-term compared to term pregnancies). d Oxidative phosphorylation through mitochondrial complexes I (OXPHOS I) and I + II, and non-phosphorylating LEAK respiration, were increased in term preeclamptic placentae (p = 0.0084, p = 0.0199, p = 0.0310, fold change 1.188, 1.232, 1.150, respectively). Maximum capacity of the respiratory system (ETS), oxygen consumption through complex II (CI-CII) and OXPHOS IV were not different. OXPHOS I as a proportion of maximal respiratory capacity (FCF CI) was increased in term preeclamptic placentae (p = 0.0026, fold change 1.126). Reserve capacity was reduced in term preeclamptic placenta (p = 0.0042, fold change 0.745). GAPDH = Glyceraldehyde 3-phosphate dehydrogenase. N = 6–10; *P < 0.05; **P < 0.01, ***P < 0.001

Fig. 5. In vitro hypoxia/reoxygenation of placental tissue leads to increased mitochondrial respiration and increased production of hydrogen peroxide during LEAK state respiration.

a Oxidative phosphorylation through mitochondrial complexes I + II (OXPHOS I + II), non-phosphorylating LEAK respiration, and maximum capacity of the respiratory system (ETS), were increased in healthy placental tissue after hypoxia/reoxygenation (p = 0.0065, p = 0.0091, p = 0.0060, fold change 1.401, 1.516, 1.527, respectively). Reserve capacity was not changed. b Hydrogen peroxide production was not changed during OXPHOS I + II, and was increased in LEAK respiration in healthy placental tissue after hypoxia/reoxygenation (p = 0.0020, fold change 1.945). N = 11; **P < 0.01

Fig. 6. Summary diagram showing proposed mechanisms of mitochondrial-related adaptations in preeclamptic placentae that reach term delivery.

Increased reactive oxygen species (ROS) production leads to a compensatory increase in antioxidant levels, potentially protecting mitochondria and other cellular organelles. Mitochondria show evidence of damage, with lower mitochondrial respiratory efficacy and changes in the relative proportions of mitochondrial electron transfer system complexes. To compensate for this damage, mitochondrial dynamics is shifted to increased fusion, allowing the salvage of damaged mitochondria, and there is mitochondrial suppression of apoptotic signalling. Although mitochondria show evidence of dysfunction, salvaged mitochondria contribute to ongoing tissue function, allowing for pregnancy continuation