Cytochrome P450 subfamily 2J polypeptide 2 expression and circulating epoxyeicosatrienoic metabolites in preeclampsia

Abstract

Background: Preeclampsia is a multisystem disorder of pregnancy, originating in the placenta. Cytochrome P450 (CYP)-dependent eicosanoids regulate vascular function, inflammation, and angiogenesis, which are mechanistically important in preeclampsia.

Methods and results: We performed microarray screening of placenta and decidua (maternal placenta) from 25 preeclamptic women and 23 control subjects. The CYP subfamily 2J polypeptide 2 (CYP2J2) was upregulated in preeclamptic placenta and decidua. Reverse-transcription polymerase chain reaction confirmed the upregulation, and immunohistochemistry localized CYP2J2 in trophoblastic villi and deciduas at 12 weeks and term. The CYP2J2 metabolites, 5,6-epoxyeicosatrienoic acid (EET), 14,15-EET, and the corresponding dihydroxyeicosatrienoic acids, were elevated in preeclamptic women compared with controls in the latter two thirds of pregnancy and after delivery. Stimulating a trophoblast-derived cell line with the preeclampsia-associated cytokine tumor necrosis factor-α enhanced CYP2J2 gene and protein expression. In 2 independent rat models of preeclampsia, reduced uterine-perfusion rat and the transgenic angiotensin II rat, we observed elevated EET, dihydroxyeicosatrienoic acid, and preeclamptic features that were ameliorated by the CYP epoxygenase inhibitor N-(methylsulfonyl)-2-(2-propynyloxy)-benzenehexanamide (MsPPOH). Uterine arterial rings of these rats also dilated in response to MsPPOH. Furthermore, 5,6-EET could be metabolized to a thromboxane analog. In a bioassay, 5,6-EET increased the beating rate of neonatal cardiomyocytes. Blocking thromboxane synthesis reversed that finding and also normalized large-conductance calcium-activated potassium channel activity.

Conclusions: Our data implicate CYP2J2 in the pathogenesis of preeclampsia and as a potential candidate for the disturbed uteroplacental remodeling, leading to hypertension and endothelial dysfunction.

Figure 1

CYP2J2 expression and metabolites in uteroplacental unit and maternal circulation. (A) CYP2J2 expression was upregulated in placental and decidual tissue in preeclamptic patients (black bars) (n=23 each group; * p<0.001; †p<0.0001; Mann-Whitney U test; median ± IQR). (B) CYP2J2 (green) was strongly co-localized to the trophoblast marker cytokeratin 7 (CK7) (red) at both gestational states. For better localization in the diffuse structure of villous tissue at term, nuclei were stained by DAPI. Open arrows are indicating CYP2J2 positive cytotrophoblasts, black arrows indicate CYP2J2 positive extravillous trophoblasts. (C) CYP2J2 metabolite levels are shown in maternal plasma of a longitudinal study population. 5,6-(EET + DHET) levels (left panel) were elevated in preeclamptic plasma (black bars), compared to controls (open bars). 14,15-(EET + DHET) levels were also elevated in preeclampsia (n=10 each group; * p<0.05; † p<0.01; ‡ p<0.001; unpaired t-test).

Figure 2

Tumor necrosis factor α (TNFα) regulates CYP2J2 expression in a trophoblast-derived cell line. (A) CYP2J2 expression in a trophoblast derived cell line is shown on mRNA level. TNFα stimulated the CYP2J2 expression dose- and time-dependently (n=4; * p<0.05; † p<0.001; Mann-Whitney U test with Bonferroni correction; median ± IQR). (B) Western blot results confirmed upregulation of CYP2J2 by TNFα stimulation. Recombinant CYP2J2 served as control.

Figure 3

Activity and Inhibition of CYP2J2 metabolites. (A) Cytochrome P450-pathway is shown. The phospholipase A2 (PLA2) releases AA from the cell membrane, which is further metabolized by the enzyme CYP2J2 to its epoxyeicosatrienoic acids (EET’s) and by the soluble epoxide hydrolase (sEH) to the dihydroxyeicosatrienoic acids (DHET’s). MsPPOH inhibits cytochrome P450 epoxygenases. 5,6-EET can be metabolized by the cyclooxygenases 1 and 2 (COX1, COX2) and the thromboxane synthase (TXS) to the thromboxane analog 5,6-epoxy-thromboxane A1 (5,6-epTXA1) that can induce signaling via the thromboxane receptor. Furegrelate (TXS-Inhibitor) and the thromboxane A2-receptor-antagonist SQ29548 inhibit the cascade. (B) The effect of 5,6-EET on cardiomyocyte contraction bioassay is shown. On left panel, 5,6-EET significantly enhanced the beating rate of cardiomyocytes. This effect was blocked by SQ29548 and Furegrelate. The thromboxane analog U-49919 also enhanced beating rate, blocked by SQ29548. Right panel shows synergistic effect of 5,6 EET to Ang II. Losartan blocks the Ang II-dependent effect, but not the 5,6 EET effect (n=6–8; * p<0.05; ANOVA with Scheffe post-Hoc test; mean ± SEM). (C) The large-conductance calcium-activated potassium channel K_Ca1.1 is shown (left panel representative picture, right panel graph of 15 experiments). Activation of the K_Ca1.1 current in the whole cell mode by 0.5 μM Ca²⁺ led to a maximal current of 31.5 pA/pF at a holding potential of −100mV. Direct application of 5,6-EET (1μM) led to a down regulation of the channel activity by 73.7% compared to the control current. Furegrelate abolished the effect of 5,6-EET (n=15; *p<0.001; ANOVA with Dunnett-T3; mean ± SEM).

Figure 4

Cytochrome P450 inhibition in preeclampsia models. (A) Mean arterial pressure (MAP) measured on day 18 of pregnancies was enhanced in RUPP rats compared to normal pregnant rats (NP). MsPPOH and SC58236 abolished this effect (n=5–9; * p<0.05; † p<0.01; ANOVA with Tukey post hoc test; mean ± SEM). MsPPOH and SC58236 also abolished the effect of the RUPP procedure on pups. RUPP-group litters showed growth restriction compared to litters of NP-group (n=5–9; * p<0.05; † p<0.001 ANOVA with Dunnett-T3; mean ± SEM). (B) Cytochrome P450-metabolites are increased in RUPP-rats. This effect is lowered by MsPPOH administration (n=5–9; * p<0.01 ANOVA with Dunnett-T3; mean ± SEM). C) MAP measured by telemetry was enhanced in transgenic rats with preeclamptic phenotype (PE) compared to normal pregnant rats (SD). MsPPOH administration abolished this effect (n=4–6; * p<0.05; † p<0.01; Mann-Whitney U test). D) Albuminuria was also enhanced in PE compared to SD and normalized by MsPPOH administration (n=3–11; * p<0.05; † p<0.001; ANOVA with Tukey post hoc test; mean ± SEM). E) Ex vivo relaxation experiments of uterine artery rings of pregnant rats in response to ACh are shown. Uterine artery rings of PE rats showed a paradoxical contractile response to high dose of ACh (3×10⁻⁶ to 1×10⁻⁵ mol/L) compared to rings SD rats. MsPPOH (50 μM) enhanced the vasodilatory response to ACh in both groups compared to non-stimulated (n=5–6; * p<0.05 for SD vs. SD + MsPPOH and PE vs. PE + MsPPOH; Mann-Whitney U test).