Evaluating oxidative stress targeting treatments in in vitro models of placental stress relevant to preeclampsia

Abstract

Background: Preeclampsia is a complex pregnancy disorder characterized by the new onset of hypertension and organ dysfunction, often leading to significant maternal and fetal morbidity and mortality. Placental dysfunction is a hallmark feature of preeclampsia, which is often caused by inappropriate trophoblast cell function in association with oxidative stress, inflammation and/or pathological hypoxia. This study explores the role of oxidative stress in trophoblast cell-based models mimicking the preeclamptic placenta and evaluates potential therapeutic strategies targeting these mechanisms.

Methods: Uric acid (UA) and malondialdehyde (MDA) concentrations were measured in human plasma from women with preeclampsia (n = 24) or normotensive controls (n = 14) using colorimetric assays. Custom-made first trimester trophoblast cell line, ACH-3P, was exposed to various preeclampsia-like stimuli including hypoxia mimetic (dimethyloxalylglycine or DMOG, 1 mM), inflammation (tumour necrosis factor or TNF-α, 10 ng/mL) or mitochondria dysfunction agent, (Rhodamine-6G or Rho-6G, 1 μg/mL), ± aspirin (0.5 mM), metformin (0.5 mM), AD-01 (100 nM) or resveratrol (15 µM), for 48 h. Following treatments, UA/MDA, proliferation (MTT), wound scratch and cytometric bead, assays, were performed.

Results: Overall, MDA plasma concentration was increased in the preeclampsia group compared to healthy controls (p < 0.001) whereas UA showed a trend towards an increase (p = 0.06); when adjusted for differences in gestational age at blood sampling, MDA remained (p < 0.001) whereas UA became (p = 0.03) significantly correlated with preeclampsia. Our 2D first trimester trophoblast cell-based in vitro model of placental stress as observed in preeclampsia, mimicked the increase in UA concentration following treatment with DMOG (p < 0.0001), TNF-α (p < 0.05) or Rho-6G (p < 0.001) whereas MDA cell concentration increased only in the presence of DMOG (p < 0.0001) or Rho-6G (p < 0.001). Metformin was able to abrogate DMOG- (p < 0.01), Rho-6G- (p < 0.0001) or TNF-α- (p < 0.01) induced increase in UA, or DMOG- (p < 0.0001) or TNF-α- (p < 0.05)induced increase in MDA. AD-01 abrogated UA or MDA increase in the presence of TNF-α (p < 0.001) or Rho-6G (p < 0.001)/DMOG (p < 0.0001), respectively. The preeclampsia-like stimuli also mimicked adverse impact on trophoblast cell proliferation, migration and inflammation, most of which were restored with either aspirin, metformin, resveratrol, or AD-01 (p < 0.05).

Conclusion: Our 2D in vitro models recapitulate the response of the first trimester trophoblast cells to preeclampsia-like stresses, modelling inappropriate placental development, and demonstrate therapeutic potential of repurposed treatments.

Keywords: aspirin; metformin; oxidative stress; placenta; preeclampsia; pregnancy; resveratrol; trophoblast cells.

FIGURE 1

UA and MDA are increased in plasma from women with preeclampsia and cell lysates from 2D first trimester trophoblast in vitro models of preeclampsia. (A, B) UA and MDA concentrations were measured in plasma samples from individuals with preeclampsia or normotensive controls. Absorbances were recorded at 570 nm. (C) Metformin or (D) AD-01 treatment abrogated the increase in UA and MDA under certain preeclampsia-like conditions. ACH-3P cells were exposed to (DMOG, 1 mM) or (Rho-6G, 1 μg/mL) or (TNF-α, 10 ng/mL) to mimic hypoxia or mitochondrial dysfunction or inflammatory condition, respectively, and treated with metformin (0.5 mM), or AD-01 (100 nM), for 48 h. Untreated cells were used as controls. (C, D) UA and (E, F) MDA concentration was measured in ACH-3P cell lysates following the addition of treatments. (A, B) The data was plotted as mean ± SD; n ≥ 13; unpaired student’s t test; (C–F) The data was analyzed by one-way analysis of variance (ANOVA) with Sidak’s post-hoc test; and expressed as mean ± SEM; n = 3; *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001.

FIGURE 2

Metformin, aspirin and AD-01 improve cell proliferation in the presence of hypoxia and oxidative stress, whereas resveratrol only improves cell proliferation in hypoxic conditions. ACH-3P cells were treated with DMOG (1 mM) or Rho-6G (1 μg/mL) to emulate hypoxia or oxidative stress, respectively, ± (A–D) PBS ± (A) metformin (0.5 mM), or (B) aspirin (0.5 mM), or (C) AD-01 (100 nM) or (D) resveratrol (15 µM) for 48 h. MTT assay was performed as per manufacturer’s instructions and absorbance recorded at 565 nm. Data was analyzed by one-way ANOVA with Sidak’s post-hoc test; and expressed as mean ± SEM; n = 3; *p < 0.05; **p < 0.01; **p < 0.001; ****p < 0.0001.

FIGURE 3

Hypoxia reduced trophoblast cell migration only at 48 h time point, and metformin, aspirin, AD-01 restore it to normal. (A) Representative images of the wound scratch assay with ACH-3P cells, following treatment with DMOG (1 mM) to mimic hypoxia, and metformin (Met) (0.5 mM), aspirin (asp) (0.5 mM), or AD-01 concentration (100 nM) at t = 0 h, t = 24 h and t = 48 h after wound scratch. Untreated cells were used as controls. The scale bar indicates 400 µm. (B, C) Percentage area of wound closure from 0 h to 24 h and 0 h to 48 h following treatment with various metformin doses (0.5 mM, 1 mM and 5 mM) under hypoxia. (D, E). Percentage area of wound closure from 0 h to 24 h and 0 h to 48 h following treatment with two different doses of aspirin (0.5 mM, 0.1 mM) under hypoxia. (F, G) Percentage area of wound closure from 0 h to 24 h and 0 h to 48 h following treatment with AD-01 (100 nM) under hypoxia. The data is analyzed by one-way ANOVA with Sidak’s post hoc test and expressed as mean ± SEM; n = 3; ***p < 0.05, **p < 0.01, ***p < 0.001.

FIGURE 4

Pro-inflammatory cytokines were increased following induction of mitochondrial dysfunction and metformin showed some potential at abrogating this increase in the first trimester trophoblast cells. (A–E) ACH-3P cells were treated with DMOG (1 mM) or Rho-6G (1 μg/mL) to mimic hypoxia, or mitochondrial dysfunction, respectively, ± metformin (0.5 mM), or AD-01 (100 nM), for 48 h. Cell supernatants were collected to determine (A–D) the pro-inflammatory (IL-1β, IL-6, IL-8, IFN-α) and (E) anti-inflammatory cytokine (IL-10) concentrations by Cytometric Bead Array. Untreated cells were used as control groups. Pro-inflammatory cytokines (A) IL-1β, (B) IL-6, (C) IL-8, (D) IFN-α levels were increased following mitochondrial dysfunction induction and IL-1β and IL-6 increase was abrogated by metformin treatment in ACH-3Ps. (E) Anti-inflammatory IL-10 cytokine was also increased in the presence of mitochondrial dysfunction, which was borderline reduced with metformin. The data was analyzed by one-way analysis of variance (ANOVA) with Sidak’s post-hoc test and expressed as mean ± SEM; n = 5; *p < 0.05; **p < 0.01.