The short-chain fatty acid propionate prevents ox-LDL-induced coronary microvascular dysfunction by alleviating endoplasmic reticulum stress in HCMECs

Abstract

Coronary microvascular dysfunction (CMD) is a critical pathogenesis of cardiovascular diseases. Lower endothelial nitric oxide synthase (eNOS) phosphorylation leads to reduced endothelium-derived relaxing factor nitric oxide (NO) generation, causing and accelerating CMD. Endoplasmic reticulum stress (ER stress) has been shown to reduce NO production in umbilical vein endothelial cells. Oxidized low-density lipoprotein (ox-LDL) damages endothelial cell function. However, the relationship between ox-LDL and coronary microcirculation has yet to be assessed. Short-chain fatty acid (SCFA), a fermentation product of the gut microbiome, could improve endothelial-dependent vasodilation in human adipose arterioles, but the effect of SCFA on coronary microcirculation is unclear. In this study, we found ox-LDL stimulated expression of ER chaperone GRP78. Further, we activated downstream PERK/eIF2a, IRE1/JNK, and ATF6 signaling pathways, decreasing eNOS phosphorylation and NO production in human cardiac microvascular endothelial. Furthermore, SCFA-propionate can inhibit ox-LDL-induced eNOS phosphorylation reduction and raise NO production; the mechanism is related to the inhibition of ER stress and downstream signaling pathways PERK/eIF2a, IRE1/JNK, and ATF6. In summary, we demonstrate that ox-LDL induced CMD by activating ER stress, propionate can effectively counteract the adverse effects of ox-LDL and protect coronary microcirculation function via inhibiting ER stress.

Fig 1. ox-LDL triggers ER stress activation and impairs HCMECs function.

(A) The protein levels of GRP78, eNOS(Ser1177) phosphorylation and total eNOS were detected by Western blot after ox-LDL treatment at various concentrations (0, 50, 100, or 150 μg/ml) for 24 hours. (B) Griess reaction observed nitric oxide (NO) production after ox-LDL treatment (0–150 μg/ml) for 24 hours. (C) The protein levels of GRP78, eNOS(Ser1177) phosphorylation, and total eNOS were detected after exposure to 100 μg/ml ox-LDL at various times (0, 12, 24, or 48 hours). (D) NO production after ox-LDL treatment (0–150 μg/ml) for 24 hours. The data were expressed as the mean ± SD, n = 3. Compared with control, *P<0.05; **P <0.01.

Fig 2. The role of the ER stress sensor PERK in ox-LDL-induced NO decrease in HCMECs.

(A-B) PERK siRNA effectively suppressed the mRNA and protein expression of PERK. HCMECs were transfected with PERK siRNA, pretreated with inhibitor or L-arginine for 6 hours, and then exposed to ox-LDL (100μg/ml, 24 hours). (C) The protein levels of PERK and its downstream effector eIF2α phosphorylation were detected by Western blot. (D) eNOS phosphorylation and total eNOS were detected by Western blot. (E) NO production was detected by Griess reaction. The data were expressed as the mean ± SD, n = 3. Compared with the control, *P<0.05; **P<0.01. Compared with ox-LDL, #P<0.05; ##P<0.01.

Fig 3. The role of the ER stress sensor IRE1 in ox-LDL-induced NO decrease in HCMECs.

(A-B) IRE1 siRNA effectively suppressed the mRNA and protein expression of IRE1. HCMECs were transfected with IRE1 siRNA, pretreated with inhibitor or L-arginine for 6 hours, and then exposed to ox-LDL (100μg/ml, 24 hours). (C) The protein levels of IRE1 and its downstream effector JNK phosphorylation were detected by Western blot. (D) eNOS phosphorylation and total eNOS were detected by Western blot. (E) NO production was detected by Griess reaction. The data were expressed as the mean ± SD, n = 3. Compared with the control, *P<0.05; **P<0.01. Compared with ox-LDL, #P<0.05; ##P<0.01.

Fig 4. The role of the ER stress sensor ATF6 in ox-LDL-induced NO decrease in HCMECs.

(A-B) siRNA effectively suppressed the mRNA and protein expression of ATF6. HCMECs were transfected with ATF6 siRNA, pretreated with inhibitor or L-arginine for 6 hours, and then exposed to ox-LDL (100μg/ml, 24 hours). (C) Immunofluorescence detected the protein expression of ATF6 (red) in HCMECs. (D) eNOS phosphorylation and total eNOS were detected by Western blot. (E) NO production was detected by Griess reaction. The data were expressed as the mean ± SD, n = 3. Compared with the control, *P<0.05; **P<0.01. Compared with ox-LDL, #P<0.05; ##P<0.01.

Fig 5. The impact of propionate on ox-LDL-induced ER stress and NO release in HCMECs.

HCMECs were treated with different concentrations (10 and 20 mM) of propionate for 2 hours and then exposed to ox-LDL (100μg/ml, 24hours). (A) The protein levels of GRP78, PERK, and IRE1 were detected by Western blot. (B) Immunofluorescence detected the activation of ATF6. (C) eNOS phosphorylation and total eNOS were detected by Western blot. (D) NO production was detected by Griess reaction. The data were expressed as the mean ± SD, n = 3. Compared with the control, *P<0.05; **P<0.01. Compared with ox-LDL, #P<0.05; ##P<0.01.

Fig 6. Proposed mechanism of propionate-mediated prevention of ox-LDL-induced downregulation of eNOS phosphorylation and NO production in HCMECs.

Ox-LDL upregulates the expression of GRP78, thereby activating the UPR pathway. Activating the PERK/eIF2a, IRE1/JNK and ATF6 pathways leads to downregulating eNOS phosphorylation and NO production in HCMECs. Propionate upregulates the ox-LDL-induced decrease of eNOS phosphorylation and NO production, possibly mediated by inhibiting the UPR pathway.