Evidence of oxidative stress-induced senescence in mature, post-mature and pathological human placentas

Abstract

Introduction: Premature ageing has been implicated in placental dysfunction. Senescence can be activated by oxidative stress, a key intermediary in the pathophysiology of pre-eclampsia. We examined senescence markers across normal gestation, and in pathological and post-mature pregnancies. Inducers of oxidative stress were used to mimic senescence changes in term explants.

Methods: Placental samples were collected with ethical approval and informed consent: first and second trimester samples from surgical terminations; term and pre-term controls, and early-onset pre-eclampsia samples from caesarean deliveries. Paraffin and EM blocks of post-mature placentas were from an archival collection. Term explants were subjected to hypoxia-reoxygenation (HR) or hydrogen peroxide (H2O2).

Results: p21 was increased significantly in term homogenates compared to first and second trimester samples, and was significantly higher in PE compared to term controls. Immunostaining revealed nuclear localisation of p21 and phosphorylated histone γH2AX in syncytiotrophoblast, with abundant foci in pathological and post-mature placentas. Abnormal nuclear appearances were observed in post-mature placentas. Sudan-Black-B staining demonstrated abundant lipofuscin, an aggregate of oxidised proteins, lipids and metals, in post-mature and pathological placentas. The percentage of nuclei positive for 8-hydroxy-2'-deoxy-guanosine, a marker of oxidised DNA/RNA, was increased in pathological placentas compared to age-matched controls. These changes could be mimicked by challenge with HR or H2O2.

Discussion: Senescence markers increase in normal placentas with gestational age, and are exaggerated in post-mature and pathological cases. Oxidative stress triggers equivalent changes in explants, and may precipitate senescence in vivo. The consequent pro-inflammatory senescence-associated secretory phenotype may contribute to the pathophysiology of pre-eclampsia.

Keywords: Oxidative stress; Senescence; Syncytiotrophoblast.

Copyright © 2018. Published by Elsevier Ltd.

Figures

Fig. 1

Evidence of senescence across gestation in first, second trimester and term samples collected by chorionic villus sampling technique (A: first trimester (7–8 wk; N = 5), second trimester (13–17 wk; N = 6), term (TC; 39 wk; N = 5)), in pathological placentas (B: term (N = 6, 39 wk) and pre-term controls (N = 5, 29 wk), and early-onset PE (N = 6, 30 wk); C: early-onset PE (N = 9, 30 wk), and IUGR placentas (N = 6, 31.5 wk)) obtained from elective caesarean deliveries. Placental tissue homogenates were probed with anti-p21, anti-p16, or anti-cGAMP antibodies and the signal was quantified. A) Quantification of Western blots revealed significant increase in p21 protein in term placental homogenates compared to first and second trimester samples, while p16 increased significantly in the second trimester but showed no further difference at term. B) The p21 and p16 expression was compared in term and preterm control placentas with samples from early-onset PE (PE). Levels of p21 were lowest in term control placentas, and significantly higher in both preterm controls and PE. There were no significant differences in p16 protein among the three groups. C) Levels of p21 were significantly higher in placentas from PE, and again there was no significant difference in p16. Data are presented as mean ± SD. *p t-test where only two groups of samples were compared (C).

Fig. 2

Evidence of senescence in post-mature placentas. Archival sections from term healthy placentas (N = 5) and post-mature placentas (delivered 7–20 d after due date, N = 6) were stained with Sudan-Black-B (SBB) to detect lipofuscin, which is an aggregate of oxidised proteins, lipids and metals, known to accumulate in aged tissues (A), with p21 (A, C) or cGAMP (A, C). Compared to term controls, which only showed sporadic SBB staining, the post-mature placentas showed abundant cytoplasmic staining with SBB (A). B) Archival EM blocks of control and post-mature placentas were examined. Images show dense heterochromatin in a placenta that was delivered 20 days after the due delivery date. C) The number of p21 positive trophoblast nuclei was quantified and expressed as a percentage of the total number of trophoblast nuclei counted (total number of trophoblast nuclei was counted in five fields of view per sample at ×20 magnification, using Image J). Similarly, the staining intensity of the cytoplasmic cGAMP staining was quantified in term and post-term placentas using Image J. * Significant differences (P t-test).

Fig. 3

Staining with Sudan-Black-B (SBB) to detect lipofuscin, an aggregate of oxidised proteins, lipids and metals, in control and pathological placentas. Immunostaining revealed abundant nuclear localisation in the syncytiotrophoblast of the pathological placentas.

Fig. 4

Immunolocalisation of p21 (A) and modified histone, γH2AX (B), in control and pathological placentas. Immunostaining revealed abundant nuclear localisation of p21 (A) and modified histone γH2AX (B) in the syncytiotrophoblast of the pathological placentas.

Fig. 5

Evidence of senescence in placental explants challenged with oxidative stress of hypoxia-reoxygenation in vitro. Paraffin sections from term placental explants (N = 4), which were cultured under normoxia (N, 10% O2) or hypoxia-reoxygenation (HR, 1–10% O2) for 24 h were stained with the lipofuscin stain Sudan-Black-B (SBB; A), or immunostained with p21 (B) or modified histone, γH2AX (C), antibodies. Starting material (T0) was also stained to show the unstressed conditions at the start of the culture.

Fig. 6

Evidence of senescence in placental explants challenged with oxidative stress of hypoxia-reoxygenation (A) or H2O2 (C–D) in vitro. A) Lysates from term placental explants (n = 4), which were cultured under normoxia (N, 10% O2) or hypoxia-reoxygenation (HR, 1–10% O2) for 20 h were immunoblotted with antibodies against p21 and p16. β-actin expression served to normalize gel loading. Normalized results (±SEM) are plotted, expressing normoxic controls as 100%. * Significant differences (P < 0.05, t-test). B-D) Immunoreactivity for 8-hydroxy-2′-deoxy-guanosine (8-OHdG), a marker of oxidised DNA, was examined in pathological placentas and in placental explants in vitro. B) Percentage of 8OHdG-positive syncytiotrophoblast nuclei was quantified in pathological and control placentas. Placentas from early onset IUGR (N = 5) and PE (N = 7) have increased proportions of 8OHdG-positive STB nuclei compared to gestational age-matched controls (pre-term controls, N = 5). Immunohistochemistry was used to detect 8OHdG-positive STB nuclei. Bar represents the group mean (Tukey's post hoc ***p < 0.0001). C-D) Quantification of 8OHdG-positive syncytiotrophoblast nuclei after H2O2 treatment (n = 3, 4). Term placental explants were cultured with 0–1000 mM H2O2 for 24–48 h, and fixed for immunohistochemical detection of 8OHdG. A higher proportion of 8OHdG-positive STB nuclei were observed in explants cultured in 1000 mM H2O2 when compared to control explants. Unpaired t-test between treatments showed statistically increased percentages of 8OHdG-STB in 100 mM and 1000 mM H2O2 after both 24 h culture, compared to the 0 mM 0hr control (*p < 0.05). There was an increase in the proportion of positive STB nuclei in 1000 mM after 48 h compared to the 0 mM 48hr control (**p < 0.001). (VS, villous stroma; IVS, intervillous space; arrows indicate positive STB nuclei; scale bar 20 μm).