Thyrotropin aggravates atherosclerosis by promoting macrophage inflammation in plaques

Abstract

Subclinical hypothyroidism is associated with cardiovascular diseases, yet the underlying mechanism remains largely unknown. Herein, in a common population (n = 1,103), TSH level was found to be independently correlated with both carotid plaque prevalence and intima-media thickness. Consistently, TSH receptor ablation in ApoE ^-/- mice attenuated atherogenesis, accompanied by decreased vascular inflammation and macrophage burden in atherosclerotic plaques. These results were also observed in myeloid-specific Tshr-deficient ApoE ^-/- mice, which indicated macrophages to be a critical target of the proinflammatory and atherogenic effects of TSH. In vitro experiments further revealed that TSH activated MAPKs (ERK1/2, p38α, and JNK) and IκB/p65 pathways in macrophages and increased inflammatory cytokine production and their recruitment of monocytes. Thus, the present study has elucidated the new mechanisms by which TSH, as an independent risk factor of atherosclerosis, aggravates vascular inflammation and contributes to atherogenesis.

Graphical abstract

Figure 1.

TSH correlated positively with both atherosclerosis and inflammation in the population. (A) After exclusion criteria were applied, the included subjects were divided into three groups: euthyroidism (0.27 ≤ TSH < 4.2), mild SH (4.2 ≤ TSH < 10), and severe SH (10 ≤ TSH). (B and C) The prevalence of carotid plaque (B) and CIMT (C) detected by ultrasonography in populations with different serum TSH levels. Sample size (n) is indicated in A. (D–H) Serum TNF-α (D) detected by ELISA and CCL2 (E), IL-1β (F), CX3CL1 (G), and CD40L (H) levels in each group. Subjects were selected randomly from the included population (n = 24 for each group). The data represent the mean ± SD. Difference in prevalence was compared with crosstab χ² or Fisher’s exact test when appropriate. In other cases, the differences were compared with one-way ANOVA and post hoc Dunnett t test using euthyroid group as control. *, P < 0.05; **, P < 0.01; ***, P < 0.001.

Figure 2.

TSHR knockout alleviates vascular inflammation and atherosclerosis in ApoE^−/− mice. (A) Schedule of the experiment. Diets of Tshr^−/−ApoE^−/− mice were supplemented with thyroxine (T4) after weaning. (B and C) Representative Oil Red O–stained whole aorta (B) and aortic root sections (C) of the four groups of mice. (D) F4/80 IHC showing macrophages in aortic root lesions. (E) qPCR analysis of inflammatory marker expression in the aorta. Fold change of the expression level of each marker compared with Tshr^+/+ ApoE^−/− mice (WT) is displayed. Independent experiments were repeated three times. (F and G) IHC of TNF-α (F) and IL-6 (G) in aortic root sections. (H and I) ELISA of serum IL-6 (H) and TNF-α (I) of the two genotypes. For D–I, data of mice fed WD for 16 wk are shown. Plaque area and stained area of IHC results were measured by Image-Pro Plus software. n = 10–12 (B–D, F, and G), 10–11 (H), or 13–14 (I). Data represent mean ± SD. Scale bars: 200 µm. Differences between two groups were analyzed by t test with Welch’s correction, except for E, which was compared with multiple t tests using Holm–Sidak correction. *, P < 0.05; **, P < 0.01; ***, P < 0.001.

Figure 3.

TSH promoted inflammation in macrophages in vitro. (A) TSHR immunofluorescence (red) in mouse peritoneal macrophages. Scale bar: 10 µm. (B and C) ELISA of IL-6 (B) and TNF-α (C) in the culture medium of mouse peritoneal macrophages, 12 h after escalating dose of TSH stimulation. (D) qPCR analysis of the effect of TSH on macrophage function in the presence of OxLDL. Tshr^+/+ or Tshr^−/− BMDMs were treated with or without TSH for 24 h and further treated with 25 µg/ml OxLDL for 24 h. Asterisk, significantly affected by TSH treatment. (E) The effect of TSH treatment on the chemoattractive ability of the BMDM-conditioned medium in the presence of 20 µg/ml of OxLDL was studied using Transwell assay. (F and G) The effect of TSH on F4/80 expression during monocyte maturation (F) and macrophage apoptosis (G) was evaluated by FCM. Experiments were repeated three times. Data represent the mean ± SD. In B and C, statistical analysis was done with one-way ANOVA and post hoc Dunnett t test, while in D, multiple t tests using Holm–Sidak method were used. Differences between two groups were analyzed by t test with Welch’s correction. **, P < 0.01; ***, P < 0.001. N.S., not significant.

Figure 4.

Myeloid-specific TSHR knockout suppressed atherosclerosis in ApoE^−/− mice. (A and B) En face Oil Red O staining of aortas (A) and H&E, Oil Red O staining or IHC of F4/80, IL-6, and CCL2 in the aortic root sections (B) from myeloid-specific TSHR knockout mice (TSHR^MKO) and their littermates (LC), after 12 or 16 wk of WD. Scale bar: 200 µm. Plaque size or stained area was quantified with Image-Pro Plus software. n = 11 for mice of 12 wk and n = 9 or 10 for 16 wk. Data represent the mean ± SD. Differences between two genotypes were analyzed by t test with Welch’s correction. *, P < 0.05; **, P < 0.01; ***, P < 0.001.

Figure 5.

Myeloid-specific TSHR knockout suppressed monocyte recruitment in atherosclerosis. (A) Representative FCM plots and statistics showing CFSE⁺ cells in the aorta of myeloid-specific TSHR knockout mice (TSHR^MKO) and their littermates (LC) 24 h after adoptive transfer of 10⁶ CFSE-labeled monocytes to each mouse. (B) Representative FCM plots and statistics showing frequencies of CD45⁺ cells in whole aorta detected by FCM (n = 7 or 8). (C) Representative FCM plot of aortic CD45⁺ CD11b⁺ Ly6G⁻ cells showing how macrophages and Ly6C^hi monocytes were gated. (D and E) Frequencies of monocytes (D) and macrophages (E) in whole aorta detected by FCM. n = 7 or 8 (A–E). Data represent the mean ± SD. Differences between two genotypes were analyzed by t test with Welch’s correction. *, P < 0.05; **, P < 0.01.

Figure 6.

TSH promoted inflammation in macrophages via IκB/NFκB and MAPKs pathway. (A) Cellular distribution of NF-κB p65 (shown in red) 30 min after TSH treatment was visualized by immunofluorescence in mouse peritoneal macrophages. Nuclei were stained with DAPI (shown in blue). Scale bar: 20 µm. (B and C) Dose-dependent effects of TSH at 30 min on NF-κB pathway (B) or MAPKs (ERK1/2, p38 MAPK, and JNK) pathways (C) in RAW264.7 were examined by Western blotting. GAPDH and LAMB were used as internal references for the cytoplasmic and nuclear fractions, respectively. (D) The effect of TSHR silencing on the activating effects induced by TSH was evaluated by Western blotting. Control or si-Tshr RNA was transfected into RAW264.7 36 h before TSH treatment. For the above experiments, PBS and LPS were used as the negative and positive control, respectively. Experiments were repeated three times. Intensity of optical density of Western blotting results were measured by Image-Pro Plus software. Data represent the mean ± SD. Multigroup differences were tested with one-way ANOVA and post hoc Dunnett t test. *, P < 0.05 vs. control; #, P < 0.05 vs. 1 ng/ml TSH; †, P < 0.05 vs. corresponding treatment of scramble siRNA control.

Figure 7.

TSH promotes atherosclerosis by acting on macrophages. Left: By binding to macrophage TSHR, TSH activated IκB/NF-κB and MAPK pathways, induced the expression and secretion of chemokines (CCL2 and CX3CL1) and other inflammatory cytokines (IL-6 and TNF-α). Right: The increase in chemokine concentration in the plaque promoted monocyte recruitment, and the increase in inflammatory cytokines led to more propagated and sustained inflammation.