Angiotensin receptor agonistic autoantibody-mediated soluble fms-like tyrosine kinase-1 induction contributes to impaired adrenal vasculature and decreased aldosterone production in preeclampsia

Abstract

Preeclampsia (PE) is a life-threatening hypertensive disorder during pregnancy associated with decreased circulating aldosterone levels. However, the molecular mechanisms underlying aldosterone reduction in PE remain unidentified. Here we demonstrate that reduced circulating aldosterone levels in preeclamptic women are associated with the presence of angiotensin II type 1 receptor agonistic autoantibody and elevated soluble Fms-like tyrosine kinase-1, 2 prominent pathogenic factors in PE. Using an adoptive transfer animal model of PE, we provide in vivo evidence that the injection of IgG from women with PE, but not IgG from normotensive individuals, resulted in hypertension, proteinuria, and a reduction in aldosterone production from 1377 ± 272 pg/mL to 544 ± 92 pg/mL (P<0.05) in pregnant mice. These features were prevented by coinjection with an epitope peptide that blocks antibody-mediated angiotensin type 1 receptor activation. In contrast, injection of IgG from preeclamptic women into nonpregnant mice induced aldosterone levels from 213 ± 24 pg/mL to 615 ± 48 pg/mL (P<0.05). These results indicate that maternal circulating autoantibody in preeclamptic women is a detrimental factor causing decreased aldosterone production via angiotensin type 1 receptor activation in a pregnancy-dependent manner. Next, we found that circulating soluble Fms-like tyrosine kinase-1 was only induced in autoantibody-injected pregnant mice but not nonpregnant mice. As such, we further observed vascular impairment in adrenal glands of pregnant mice. Finally, we demonstrated that infusion of vascular endothelial growth factor(121) attenuated autoantibody-induced adrenal gland vascular impairment resulting in a recovery in circulating aldosterone (from 544 ± 92 to 1110 ± 269 pg/mL; P<0.05). Overall, we revealed that angiotensin II type 1 receptor agonistic autoantibody-induced soluble Fms-like tyrosine kinase-1 elevation is a novel pathogenic mechanism underlying decreased aldosterone production in PE.

Figure 1. Reduced aldosterone levels in sera of women with PE are associated with the presence of AT₁-AA and elevation of sFlt-1 levels

(A) Aldosterone levels in the sera of normotensive pregnant women (NT) and women with preeclampsia (PE) were measured by EIA specific for aldosterone. Aldosterone levels were significantly reduced in PE sera compared to NT sera. (B) AT₁-AA levels in NT and PE sera were quantified by an NFAT-Luciferase bioassay that reflects AT1 receptor activation. AT₁-AA activity was significantly elevated in IgG prepared from PE compared to NT sera. (C) sFlt-1 levels in NT and PE sera were measured by ELISA. sFlt-1 levels were significantly increased in PE compared to NT sera. * P<0.05 versus NT, n=12 for NT and 15 for PE.

Figure 2. PE-IgG induced hypertension and proteinuria via AT₁R activation in pregnant mice

(A) Hypertension and (B) proteinuria, two key features of PE, are induced in the PE-IgG-injected pregnant mice. Both of these features were attenuated by co-injection with losartan or a 7aa epitope peptide that corresponds to a site on the second extracellular loop of the AT₁R. Blood pressure and proteinura represented here were measured on gestation day 18. * P<0.05 versus NT-IgG treatment. ** P<0.05 versus PE-IgG treatment. n=8-10 injected mice for each experimental group.

Figure 3. PE-IgG injection leads to decreased aldosterone levels in the serum of pregnant mice and increased aldosterone production in non-pregnant mice

(A) PE-IgG but not NT-IgG significantly reduced maternal circulating adolsterone concentration in the pregnant mice. Pregnant mice were injected with NT-IgG or PE-IgG on gestational days 13 and 14. On gestational day 18, aldosterone levels in the serum of PE-IgG injected pregnant mice were significantly reduced compared to NT-IgG injected pregnant mice. The 7aa epitope peptide co-injection attenuated the reduced aldosterone production in PE-IgG injected mice * P <0.05 compared to NT, ** P<0.05 compared to PE. n = 5-7 injected mice per experimental group. (B) PE-IgG but not NT-IgG significantly induced circulating adosterone concentration in non-pregnant mice. Non-pregnant mice were injected with NT-IgG or PE-IgG on two consecutive days. Five days following the first injection injection, aldosterone levels in the serum of PE-IgG injected non-pregnant mice were significantly induced compared to NT-IgG injected non-pregnant mice. The 7aa epitope peptide co-injection reduced the elevated aldosterone production in PE-IgG injected mice *P<0.05 compared to NT, **P<0.05 compared to PE. n = 6-8 injected mice per experimental group.

Figure 4. Autoantibody-induced cellular and vascular impairment in adrenal glands of pregnant mice can be prevented by 7aa epitope peptide and VEGF₁₂₁

(A) Histological analysis of adrenal glands, assessed by H&E staining, indicated that the arrangement of the cellular components within the zona glomerulosa (ZG) layer of the adrenal glands displayed a disorganized and non-uniform pattern in the PE-IgG group compared to NT-IgG group (20X, Scale bar 50 μm). (B) The cellular components, including nuclei, were found to be irregularly placed in patches, shrunken and clustered in PE-IgG injected mice (100X, Scale bar 10 μm). (C) H&E staining showed that the corticomedullary junction area (CA) of the adrenal glands of PE-IgG injected pregnant mice displayed substantially diminished capillaries with less branching compared to the NT-IgG injected mice (100x, Scale bar 10 μm). n=5-7. M: Medulla, C: cortex, ZG: zona glomrulosa; S: sinusoid.

Figure 5. Antibody-induced vascular impairment in adrenal glands of the pregnant mouse was attenuated by 7aa epitope peptide or VEGF₁₂₁

Adrenal gland vascularity was assessed by CD34 immunostaining. (A) CD34 immunostaining showed that CD34 was specifically expressed in the endothelium of the blood vessels (microvessels, MV) in the sinusoidal areas (S) between the cortex and medulla (M). (100x, scale bar 10 μm). Microscopic examination revealed the presence of microvessels at the corticomedullary junction of the adrenal glands. The incidence of microvessels in large areas that branched deeply into the cortex was evident by the CD34 staining in the NT-IgG injected mice. CD34 staining was decreased in the PE-IgG injected pregnant mice and 7aa epitope peptide co-injection or VEGF₁₂₁ infusion attenuated vascular impairment in these mice. (B) An arbitrary histological quantification of the of CD34 staining. *P<0.05 compared to NT, **P<0.05 compared to PE.

Figure 6. VEGF₁₂₁ infusion prevented the PE-IgG-induced hypertension and reduction of aldosterone production in pregnant mice

(A) sFlt-1 levels in the serum were significantly increased in pregnant mice compared to non-pregnant mice and further induced in the PE-IgG-injected pregnant mice (PE) compared to NT-IgG injected pregnant mice (NT). The 7aa epitope peptide coinjection (PE+7aa) attenuated PE-IgG-mediated sFlt-1 induction in the serum of pregnant mice. *P<0.05 compared to non-pregnant mice, **P<0.05 compared to NT. ***P<0.05 compared to PE. n = 6 or 7 injected mice per experimental group. (B) VEGF₁₂₁ infusion significantly attenuated PE-IgG-induced hypertension in pregnant mice. Pregnant mice were injected with either PE-IgG or NT-IgG on gestation days 13.5 and 14.5. Some of the PE-IgG injected mice were continuously infused with VEGF₁₂₁. Blood pressure was determined immediately prior to the initial injection and at 3 and 5 days following the initial injection. *P<0.05 compared to NT, **P<0.05 compared to PE. n = 5 or 6 mice per experimental group. (C) Aldosterone levels in the serum of PE-IgG injected pregnant mice were significantly reduced compared to NT-IgG injected pregnant mice. VEGF₁₂₁ infusion attenuated the reduction in aldosterone production in PE-IgG injected mice *P<0.05 compared to NT, **P<0.05 compared to PE. n = 5-7 injected mice per experimental group. (D) Working model of AT₁-AA in preeclampsia. Our findings suggest that AT₁-AA contributes to reduced aldosterone production via elevated sFlt-1 production which contributes to adrenal gland vascular impairment and reduced aldosterone production. AT₁-AA-mediated induction of sFlt-1 and the resulting inhibition of VEGF signaling also has detrimental effects on renal and vascular function leading to proteinuria and hypertension. Overall, AT₁-AA is a detrimental factor contributing to multiple features associated with PE in a sFlt-1-dependent manner that is based on interference with VEGF signaling. Thus, interfering with AT₁-AA-mediated AT₁R receptor activation or neutralizing the effects of increased sFlt-1 by the infusion of VEGF₁₂₁ represent important therapeutic possibilities for PE.