Male c57BL/6 mice (8 weeks old) were purchased from Beijing Wei Tong Li Hua Experimental Animal Technology Co. Ltd. (Beijing, China). Mice were fed in temperature-controlled cages with a 12-h light-dark cycle and received freely normal chow and water, as previously described (Ma et al., 2017). This study was carried out in strict accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health. The animal protocol was reviewed and approved by the University of Shandong Animal Care and Use Committee.

Three age-matched male WT mice cohorts received an intraperitoneal injection with Salmonella typhosa LPS in PBS, with or without VitB6 pre-treatment (20 mg/kg, 6 h) at 4 mg/kg for 24 h (Supplementary Figure S3A).

Mice were divided into control (n = 8), LPS (n = 9), and VitB6+LPS (n = 9) groups. Mice were pretreated with PBS or VitB6 for 6 h and then treated with LPS (4 mg/kg) for 24 h. Cardiac ultrasound was performed before sacrifice. Inhaled isoflurane was given to mice for volatile anesthesia and the chest hair was removed with a depilatory cream. Then, mice were fixed on the warmed imaging platform and wore with the coupling agent. The Vevo2100 imaging system, equipped with a 40-MHz high-frequency transducer (VisualSonics Inc., Toronto, Canada), was applied to perform non-invasive examinations. The M-mode echocardiogram at the parasternal long axis was used to obtain the ejection fraction (EF) of left ventricular and fractional shortening (FS).

The levels of cTnI, LDH, MDA, and SOD in serum, as well as MDA and SOD concentrations in the myocardial tissue, were detected using different kits. Total and lipid ROS were assayed by flow cytometry.

Protein levels in H9C2 cell lysates and myocardial tissue homogenates were analyzed by Western blot. BCA protein assay kit was applied to determine protein concentrations. Briefly, 20 µg of protein samples were separated by 10% SDS-PAGE and then electroblotted onto polyvinylidene fluoride membranes. The membrane was blocked in 5% bull serum albumin for 1.5–2 h at room temperature, then washed three times by TBST, and submerged in diluted primary antibodies (1:1,000) overnight at 4°C, followed by the secondary antibodies for 2 h at room temperature. Finally, protein-attached bands were visualized using the ECL chemiluminescence system (Millipore Corp. MA, United States).

A cosolvent, composed of DMSO, Tween 80, NMP, and PEG400 in different proportions, with or without ferrostatin-1 (Fer-1; 2 mg/kg), emricasan (2.5 mg/kg), and VitB6 (20 mg/kg) were given to mice by i.p. for 24 h and challenged i.p. with LPS (20 mg/kg). Mice were monitored three times daily for 7 days.

All quantitative results are expressed as means ± SEM. One-way ANOVA was used to compare multiple groups followed by Tukey post hoc tests. Statistical analyses were conducted using GraphPad Prism Plus software, and a p < 0.05 was considered statistically significant.

Previous studies have demonstrated that VitB6 could improve myocardial ischemia (Pham et al., 2003; Kandzari et al., 2005). In the current study, to investigate VitB6 protective effects on LPS-induced myocardial injury, mice were pretreated with VitB6 for 6 h and then challenged with LPS (4 mg/kg). The EF and FS significantly decreased in LPS group mice compared with controls, but cardiac dysfunction improved in the VitB6+LPS group (Figures 1A–C). No differences were detected in the left ventricular end-diastolic volume between groups. However, the left ventricular end-systolic volume was reduced in the VitB6+LPS group (Figures 1D,E). Serum cTnI and LDH, which indicated myocardial injury appearance, increased in LPS-challenged mice compared with control and VitB6-pretreated groups (Figures 1F,G). Myocardial tissue morphology stained with H&E is shown in Figure 1H. The myofibrillar structure was normal in the control group, but LPS-challenged myocardial tissue appeared swollen and fractured. However, the VitB6+LPS group showed a mild swollen. LPS and VitB6 did not affect collagen deposition (Figures 1I,J).

Ferroptosis is iron-dependent and characterized by the accumulation of intracellular ROS and lipid peroxidation products (Dixon et al., 2012; Li et al., 2021). To explore the influences of LPS and VitB6 on non-heme iron, we measured iron levels in serum and myocardial tissue. We found that LPS-treated mice presented notably higher iron levels in serum and heart tissue compared with the control group. We also found that this trend was reversed by VitB6 in LPS-induced mice (Figures 2A,B). These results indicated that VitB6 might mediate iron metabolism.

MDA is one of the most important membrane lipid peroxidation products and impairs the activities of key mitochondrial enzymes, finally promoting aging (Yang et al., 2019). SOD is a crucial antioxidant enzyme that can scavenge free radicals (Yang et al., 2019). Glutathione (GSH)—a glycine, glutamic acid, and cysteine tripeptide—is a vital antioxidant that can protect hemoglobin from oxidation. In LPS-treated mice, MDA levels increased and were significantly attenuated by VitB6 (Figures 2C,D). In LPS-induced mice, SOD activity (Figures 2E,F) and GSH levels (Figures 2G,H) markedly decreased. However, this decrease could be reversed by VitB6. Nrf2, a crucial transcriptional activator for antioxidative responses (Ci et al., 2017), can be mediated by different stimuli, including LPS (Xu et al., 2018). To determine whether VitB6 affected Nrf2 expression in LPS-stimulated myocardial tissue, we pretreated mice with VitB6 followed by LPS. We found that VitB6 increased Nrf2 expression (Figure 6C). In addition, Nrf2 mRNA levels in the myocardium were consistent with its protein levels (Figure 6D). Altogether, these results demonstrated that VitB6 inhibit LPS-caused oxidative stress and peroxidation via Nrf2 activation.

To further identify the alterations caused by LPS in vitro, we estimated SOD, ROS, and lipid peroxidation (lipid ROS) levels. Cells were pretreated with VitB6 for 2 h followed by LPS for 24 h. As predicted, in LPS-treated H9C2 cells, MDA, ROS, and lipid ROS levels increased, and SOD activity decreased. However, the above alterations were ameliorated when cells were exposed to VitB6 (Figures 3A–F). Therefore, these results are consistent with our in vivo results.

VitB6 inhibited erastin-induced ferroptosis in vitro.

Erastin is a small molecule that can trigger ferroptosis. To verify that VitB6 could suppress ferroptosis, H9C2 cells were pretreated with VitB6 for 2 h followed by erastin for 24 h. Erastin induced cells overt death and, in contrast, VitB6 suppressed erastin-induced cell death (Supplementary Figure S1). Thus, VitB6 inhibits ferroptosis induced by erastin as expected.

VitB6 decreased nonheme iron levels in serum and myocardial tissue LPS-induced in vivo. Iron homeostasis maintenance depends on the participation of different ferroregulatory proteins. For example, transferrin receptor (TFR) transports iron into cells and FPN1 exports iron out of cells (Gkouvatsos et al., 2012; Ward and Kaplan, 2012), thereby regulating cellular iron levels. Ferritin is an iron-binding protein and can be used to store iron (Huang et al., 2019). Thus, we explored the expression of iron regulatory proteins TFR, FPN1, and ferritin. Cells were pretreated with VitB6 for 2 h followed by LPS for 24 h. Then, lysates were subjected to Western blot. After LPS treatment, the levels of TFR and ferritin increased (Figures 4A–C), and the expression of FPN1 decreased (Figures 4A–D). In contrast, the synthesis of TFR, ferritin, and FPN1 was reversed after treatment with VitB6+LPS (Figures 4A–D). These results indicated that LPS may lead to an iron elevation in H9C2 cells, as necessary for ferroptosis.

Many studies have shown that Gpx4 inactivation can trigger ferroptosis (Ingold et al., 2018) and that upregulation of Nrf2 signaling can suppress ferroptosis and apoptosis in various diseases (Sun et al., 2016; Ishii et al., 2019). NQO1 and HO1 are crucial antioxidant-related enzymes and play a vital role in the Nrf2 signaling pathway (Ishii et al., 2000; Sun et al., 2016). To elucidate VitB6 effects on these proteins, we pretreated cells with VitB6 for 2 h followed by LPS for 24 h. Then, lysates were subjected to Western blot. The LPS treatment decreased Nrf2, Gpx4, NQO1, and HO1 expressions (Figures 5A–E). On the other hand, VitB6 significantly increased the expression of these proteins (Figure 5A). These results indicated that VitB6 altered the synthesis of proteins related to ferroptosis and apoptosis.

Studies have reported that VitB6 can protect the intestinal epithelium from ionizing radiation-induced apoptosis (Thotala et al., 2009). Thus, we used the TUNEL assay to investigate whether VitB6 inhibited LPS-induced apoptosis. The number of TUNEL-positive cells increased followed by LPS stimulation. However, in VitB6 pretreated myocardium, the number of positive cells was reduced (Figures 6A,B). In addition, LPS increased Bax protein levels and decreased Bcl2 expression of myocardial tissue, but VitB6 reversed this Bax/Bcl2 trend (Figure 6C). These data revealed that VitB6 can regulate LPS-activated intrinsic apoptotic pathway.

Cells seeded in 96-well plates were preprocessed with VitB6 (500 μM) or PBS for 2 h and then treated with different LPS concentrations for 24 h. Cell viability was assayed by Cell Counting Kit-8, and the cell viability percentage is depicted in Figure 7A. We observed that VitB6 reduced LPS-induced cell death at 100 and 1,000 ng/ml. To further investigate the VitB6 in vitro effects on apoptosis, cells were pretreated with VitB6 (500 μM) for 2 h and then stimulated with LPS (100 ng/ml) for 24 h. The TUNEL kit was used to elucidate the effect of VitB6 on LPS-induced cells. TUNEL-positive cells were significantly increased followed by LPS, and the numbers of TUNEL-positive cells of VitB6 pretreated lessened visibly. In response to LPS stimuli, the expression levels of apoptosis-related proteins, cleaved caspase3, and Bax increased but Bcl2 reduced. Pretreatment with VitB6 effectually repaired these alterations (Figures 7D–F).

Apoptosis is closely related to mitochondrial injury. The mitochondrial membrane of H9C2 cells in the control group was intact, including clear mitochondrial cristae (Supplementary Figure S2A). However, mitochondria swelled and its inner crest ruptured, and even disappeared, after LPS exposure (Supplementary Figure S2B). VitB6 improved some LPS-induced morphological alterations of mitochondria structure, but they remained a bit swollen (Supplementary Figure S2C). These outcomes showed that VitB6 can protect mitochondria from LPS-induced injury.

In vivo, Fer-1 and Emricasan mimicked pharmacological effects of VitB6 decreasing the mortality of LPS-induced mice.

We hypothesized that LPS may induce ferroptosis and apoptosis. Thus, we preconditioned mice with Fer-1 (2 mg/kg) (n = 14) and emricasan (2.5 mg/kg) (n = 5) for 24 h followed by LPS (20 mg/kg) and monitored them three times daily for over 7 days to further confirm this hypothesis. The survival percentages of mice preconditioned with Fer-1 and emricasan were higher than the LPS group (n = 15) (Supplementary Figure S3C). The protocol is depicted in Supplementary Figure S3A, B. VitB6 improve the survival percentages of mice stimulated with LPS (n = 15) (Supplementary Figure S3C).