ERα levels were negatively associated with iron content in human plaques.

(A) Ferritin (Ft-L), revealed by immunohistochemistry, and iron content, revealed by DAB-enhanced Prussian blue staining, in plaque paraffin sections of 8 postmenopausal patients. The upper panel: the early postmenopausal (EPM) group (P1-P4, < 65 years old); the lower panel: the late postmenopausal (LPM) group (P5-P8, > 65 years old). (B) Serum iron in EPM (blue) and LPM (magenta) patients, measured by using an autochemical analyser (Beckman Coulter AU5421, CA). n=10/group, ***p < 0.001. (C) Serum ferritin levels detected by ELISA. n=10/group, ***p < 0.001. (D) Serum total cholesterol (left) and total triglyceride (right) levels. n=10/group, *p < 0.05, ** p < 0.01. (E) ERα and Ft-L expression in plaques measured by Western blotting. The samples are the same as in (A). (F) The plotted and calculated Pearson correlation coefficient (r = -0.7205) between the plaque Ft-L and ERα levels (n=8, p = 0.0438).

Atherosclerosis was aggravated in E2-treated late postmenopausal ApoE-/- mice with lower ERα expression.

(A) Flow diagram of mouse modeling. Early E2-treatment group: OVX at 8 weeks old, one-week recovery, E2 treatment for 8 weeks; late E2-treatment group: OVX at 8 weeks old, E2 treatment from 21 weeks old to 29 weeks old for 8 weeks. Saline is vehicle control. Mice were fed high-fat chow from 9 weeks old. (B) Oil red O-stained aortic lesions in ApoE-/- mice after E2 treatment for 8 weeks in the EPM or LPM group. (C) Statistical analysis of the area of atherosclerotic plaque in the aorta. n = 4/group, *p < 0.05, ** p < 0.01. (D) The expression of iron-related or ERα-targeted proteins in the aorta, detected by Western blotting. (E) Protein expression in peritoneal macrophages detected by Western blotting. Macrophages were isolated from 4 mouse groups (early/late ± E2, details see Materials and Methods). (F) Serum total cholesterol and total triglyceride levels in the 4 mouse groups. Pre: serum samples before OVX as a control group. n=6/group, *p < 0.05, ** p < 0.01. (G) Iron content in aorta and liver, detected by ferrozine assays. n = 6/group, ****p < 0.0001, **p < 0.01, *p < 0.05. (H) Serum iron in different groups, detected by using an autochemical analyser (Beckman Coulter AU5421). n = 6/group, ***p < 0.001, **p < 0.01. Student’s t-test analysis was used for C, F, G, and H.

E2-triggered ERα deficiency was observed in a genetic iron overload mouse model at postmenopausal age.

(A) Flow diagram of mouse modeling. Early groups: OVX at 16 weeks old, one-week recovery, ± E2 treatment for 8 weeks; late groups: OVX at 40 weeks old, one-week recovery, ± E2 treatment for 8 weeks. Saline is vehicle control. The mice were fed with normal chow. (B) Oil red O-stained aortic lesions in ApoE-/-and ApoE-/- Fpn1LysM/LysM mice after E2 treatment for 8 weeks in the EPM or LPM groups as indicated. (C) The lesion area in the aorta. n = 4/group, **p < 0.01, *p < 0.05. (D) The expression of iron-related or ERα-targeted proteins in the aorta, detected by Western blotting. (E) The iron content of the aorta and liver detected by ferrozine assays. n = 6/group, ***p < 0.001, **p < 0.01. (F) Serum iron level in different groups. n = 6/group, ***p < 0.001, **p < 0.01. (G) Serum total cholesterol and total triglyceride levels. n=6/group, *p < 0.05, ** p < 0.01. The samples for D-G were from 49-week-old ApoE-/- and ApoE-/- Fpn1LysM/LysM mice. Student’s t-test analysis was used for C, E, F, and G.

E2 treatment potentiates iron-induced downregulation of ERα in vitro.

(A-B) ERα expression in the presence or absence of E2 under different iron concentration conditions (A) or in the time course (B). Quantification was carried out using ImageJ. Two-way ANOVA was used. (C) The rescue effect of iron chelation on the downregulation of ERα by FAC or FAC plus E2. (D) The intracellular iron content in J774a.1 under different iron-concentration conditions in the presence or absence of E2, detected by ferrozine assays. n = 4, *p < 0.05. (E, F) ERα expression in peritoneal macrophages (E) and HUVECs (F) under the indicated iron and E2 conditions. A, B, C, E, and F are data from Western blotting. Numbers indicate the relative intensity of ERα n=4. (G) Oil red O-stained J774a.1 cells after treatment with FAC and/or E2 (left) with the quantified droplets (right). scale bar = 25 μm, n = 4, ***p < 0.001. (H) HUVEC angiogenesis assays, revealed by the number of branch points (left) and capillary length (right). n = 4, *p < 0.05, **p < 0.01. (I) eNOS level in HUVEC, assessed by ELISA. n = 4, *p < 0.05, **p < 0.01, ***p < 0.001. The student’s t-test analysis was used for G - I.

The interactive effects of iron overload and E2 treatment on ERα downregulation are mediated by the E3 ligase MDM2.

(A) Evaluation of ERα proteasome-dependent degradation in J774a.1 cells by western blotting. MG132: 10 μM. n=4. (B) ERα turnover rate in J774a.1 cells under FAC or E2+FAC conditions, detected by western blotting after 20 μM cycloheximide (CHX) treatment. *p < 0.05 using two-way ANOVA (C) Ubiquitination of ERα, evaluated by western blotting (anti-ubiquitin) following immunoprecipitation against ERα antibody. n = 3, *p < 0.05. (D) Relative Mdm2 mRNA expression in J774a.1 cells, assessed by qPCR, n = 4, **p < 0.01. (E) The protein levels of ERα in the presence of FAC or FAC plus E2 in J774a.1 cells after treatment of Nutlin-3, a specific antagonist of Mdm2. n=3. (F) The protein levels of ERα in the presence of FAC or FAC plus E2 in HUVECs after treatment of Nutlin-3. n=3. (G) Mdm2 protein expression in the aortas of mice in the EPM or LPM stage, as detected by western blotting. n=3/group. (H) MDM2 protein levels in patient plaques, detected by western blotting and quantified with ImageJ. n=4/group, ***p < 0.001. Student’s t-test analysis was used for B, E, and H.

Iron restriction therapy restored ERα levels and attenuated E2-triggered progressive atherosclerosis in late postmenopausal mice.

(A) Flow diagram of mouse modeling. The mice were ovariectomized at 8 weeks old and E2 or E2+DFP treated from 21 weeks old to 29 weeks old for 8 weeks. Saline is vehicle control. Mice were fed high-fat chow one week after OVX. Thirteen weeks post-OVX is considered as late post-menopause. (B) Oil red O-stained aortic lesions in ApoE-/- mice treated with E2 or E2+DFP as indicated. (C) The quantified lesion area of atherosclerotic plaques in the aorta from B. n=4, **p < 0.01. (D) Serum total cholesterol and total triglyceride levels. n=6, *p < 0.05, ** p < 0.01. (E) The iron content in the aorta and liver, detected by ferrozine assays. n = 6, **p < 0.01, *p < 0.05. (F) Determination of serum iron in different groups. n = 6, ***p < 0.001, **p < 0.01. (G) Protein expression in the aorta, detected by western blotting (left) and quantified with ImageJ (right). ***p < 0.001, **p < 0.01, *p < 0.05. The student’s t-test analysis was used for C - G.

Schematic model for the effects of postmenopausal iron accumulation with or without HRT on AS severity through modulating ERα expression.

Iron accumulation occurs naturally and gradually after menopause. In EPM, iron retention was mild, and ERα was responsive to HRT application to achieve protective effects (A). However, when iron overload is significant in LPM, Mdm2 is upregulated along with ERα downregulation (B). This negative correlation is potentiated by the application of HRT and iron accumulation with aging. Therefore, HRT use avails to aggravate the progression of AS in the LPM period. Iron chelation, however, reverses the adverse effect of HRT and attenuates the accelerated development of AS (A), suggesting a protective role of appropriate iron restriction in the LPM stage.

E2-triggered ERα deficiency was observed in ApoE-/- Fpn1LysM/LysM at early post-menopause (25 weeks old).

(A) Iron content of aorta and liver detected by ferrozine assays. n = 4, ****p < 0.0001. (B) ABCA-1, ERα, VEGF and Ft-L protein expression of aorta were detected. (F) Serum iron in different groups. n = 6, *p < 0.05.

No significant oxidative-stress was raised by application of E2 and iron within the indicated concentration.

(A) SOD2 protein levels of J774a.1 post treatments with FAC/DFP in the presence/absence of E2. (B) The enzymatic activity of catalase in J774a.1. n=4. (C) Relative ERα mRNA expression of J774a.1 treated with different concentrations of FAC in the presence/absence of E2, assessed by qPCR. n = 5.

E3-ligase responses to iron and E2 treatment in different cell types.

(A) BRCA1 and AHR protein expressions in J774a.1 were detected. (B) The protein expression of ERα and its related E3 ligase, BRCA1 and AHR, was detected in MCF-7 cell line. n=4