The recognized sign of selenium (Se) deficiency, liver necrosis, was mostly discovered in rats (Burk et al., 2010; Schwarz and Foltz, 2010; Wang H. et al., 2020) and swines (Tang et al., 2020; Zhang Y. et al., 2020). However, in our previous report (Huang et al., 2011, 2015), we also found liver necrosis induced by Se and vitamin E (VE) deficiency in chicks. We have investigated the expression profiles of 14 selenoprotein genes (i.e., Gpx1, Gpx4, Selenof, Selenoi, Selenok, Selenon, Selenoo, Selenop, Selenos, Selenot, Selenow, Msrb1, Txnrd1, and Txnrd2) in the liver and muscles of chicks fed with Se-deficient (0.01 mg/kg) and adequate (0.3 mg/kg) diets at week 6 (Huang et al., 2011, 2015). Dietary Se deficiency decreased (P < 0.05) mRNA levels of seven common selenoprotein genes (i.e., Gpx1, Gpx4, Selenok, Selenon, Selenoo, Selenop, and Selenow) in liver and muscle. Nevertheless, these are the genetic changes far from after the onset of liver necrosis (Ren et al., 2018). Recently, 24–25 selenoprotein genes were postulated in avians (Mariotti et al., 2012; Li et al., 2018), so this study followed to investigate the changes of 25 selenoproteins in the liver of chicks responded to dietary Se and VE deficiencies. The fact reminds us that the expression between proteins and genes is not identical; sometimes, the expression of the proteins is closer to the physiological condition (Alan, 2015; Li and Sunde, 2016; Katarzyna et al., 2020). Therefore, we assayed seven key selenoproteins (i.e., GPX1, GPX3, GPX4, SELENON, SELENOP, SELENOW, and MSRB1) in liver. The activities and mRNA levels of two abundant selenoprotein (GPX1 and GPX4) genes are highly regulated by Se status in chicks (Huang et al., 2011; Li et al., 2018; Ren et al., 2018). Meanwhile, other five selenoprotein genes code for glutathione peroxidase 3 (GPX3) (Zhu et al., 2018), selenoprotein P (SELENOP) (Short et al., 2020; Gül-Klein et al., 2021), selenoprotein W (SELENOW) (Shin et al., 2019), selenoprotein N (SELENON) (Silwal et al., 2020), and methionine sulfoxide reductase B1 (MsrB1) were also regulated by Se status in chicks.

Both apoptosis and necrosis frequently occur together after death signals and toxic stresses (Xiao et al., 2015; Seremelis et al., 2019). The term, necrapoptosis, describes such a process of death that begins with a general stress or death signal (Amr et al., 2004). Necroptosis, a newly discovered pathway, regulated necrosis and mediated by the proteins of receptor-interacting protein kinase-1 (RIPK1), RIPK3, and mixed lineage kinase domain-like (MLKL) (Wang Y. et al., 2020). Also, the cleavage of RIPK1 and RIPK3 has been identified as the most important steps in the prevention of necroptosis (Yan et al., 2018; Iorga et al., 2021; Liu et al., 2021). Some studies have suggested that integrin activation of phosphatidylinositol-3-kinase (PI3-kinase), c-Jun NH2-terminal kinase (JNK), and extracellular signal-regulated kinase (ERK) signaling pathways require focal adhesion kinase (FAK) (Huang et al., 2015). The p38 mitogen-activated protein kinase (MAPK) and JNK pathways play an important role in ROS-induced apoptosis. It has been shown that prevention of p38 MAPK phosphorylation at Thr-180/Tyr-182 under hypoxia-induced oxidative stress in mouse embryonic fibroblasts due to the overexpression of GPX1 (Huang et al., 2018). The first report on liver necrosis caused by Se deficiency can date back to 1957. It was the first report showing that Se is essential for mammals, specifically rats (Burk et al., 2010; Schwarz and Foltz, 2010). Although liver injury and Se deficiency have been studied extensively, less is known about how these protein kinases are associated with avian liver necrosis. Therefore, we conducted this experiment to (1) compare the effects of Se and VE on mRNA expression profiles of 25 selenoproteins in liver of chicks at week 2 prior to the onset of liver necrosis, (2) explore how tissue selenoprotein levels of GPX1, GPX3, GPX4, SELENOP, SELENOW, SELENON, and MSRB1 in liver of chicks were related to liver necrosis, and (3) determine the responses of RIPK1/RIPK3/MLKL and MAPK signaling pathway proteins in chicks associated with liver necrosis.

The antioxidant effect of Se is mainly realized through selenoprotein, and an appropriate amount of Se can reduce the concentration of free radicals and thus protect organisms from lipid peroxidation. Furthermore, VE as a non-specific biological antioxidant could compensate and coordinate with Se in protecting cell membranes from oxidative damage (Huang et al., 2011). Previous studies have found that the lack of Se and VE in diet could lead to liver necrosis in chicks. In this study, we successfully replicated liver necrosis in chicks by feeding the Se-deficient diets. The signs of liver necrosis were observed, namely, varying degrees of swelling, purple, red or yellow color, fragility, and pinpoint bleeding in the liver (Sun et al., 2016). No incidence of liver necrosis was observed at any time point in the +Se chicks. Dietary Se concentrations exerted the intended impacts on the Se status of chicks, as shown by the Se concentrations in liver and GPX activities (Taylor et al., 2018). The lack of Se and VE in chicks leads to hepatic necrosis and increased peroxidation (Guillermo et al., 2018). Lipid peroxidation may cause membrane destruction and cell death, thereby triggering the process of necrosis (Burk et al., 1980; Zhang Y. et al., 2020). Although the mechanistic basis of mitochondrial injury may vary in different settings, mitochondrial permeability and dysfunction are common events that lead to cell apoptosis and necrosis (Guicciardi et al., 2013). Generally, apoptosis and necrosis are considered to be separate entities, but an alternate view is emerging that both are frequently a result of the same initiating factors and signaling pathways. Furthermore, apoptosis and necrosis in their pure form may represent extremes of a continuum of cell death rather than being separate entities (Zhang Y. et al., 2020). However, the protection of VE against the incidence of liver necrosis in the Se-deficient chicks was seemingly independent of muscle GPX and SOD activities. It is unclear whether these two biomarkers (i.e., GPX and SOD) were not sufficiently sensitive to reflect the effects of the relatively low dose of VE (50 mg/kg) supplementation or if VE protection was mediated by other mechanisms.

The available information on selenoprotein genes of novel chicken enabled us to design primers for qPCR to complete the systematic analysis of the effects of dietary Se on 25 selenoprotein genes expression profiles of chicken (Li et al., 2018; Sun et al., 2018). Dietary Se deficiency decreased (P < 0.05) the mRNA levels of Gpx1, Gpx3, Gpx4, Selenof, Selenoh, Selenok, Selenom, Selenoo, Selenop, Selenou, and Selenow in liver at week 2. And six common genes (Gpx1, Gpx4, Selenok, Selenoo, Selenop, and Selenow) were downregulated by dietary Se deficiency in liver and muscle at weeks 2 and 6 (Huang et al., 2015; Sun et al., 2018). In addition to the analogous response of these selenoprotein genes in other animals (Huang et al., 2015; Sunde et al., 2016; Sun et al., 2018; Chen et al., 2019; Zhang Y. et al., 2020), these genes can be used as indicators of Se status of the body in chicks. In addition, further functional assessments should be conducted to determine whether the downregulation of these genes is associated with hepatic necrosis. Moreover, the responses of liver Selenoi and muscle Selenon mRNA levels were different between the time points (weeks 2 and 6). SELENOI (encoded by Selenoi) is an evolved selenoprotein specific to vertebrates and is essential for murine embryogenesis (Avery et al., 2020). However, this membrane selenoprotein has no known function. Recent studies have identified selenoprotein N (encoded by Selenon) as an important protein that protects cells against oxidative stress and redox-related calcium homeostasis (Shin et al., 2019). SELENON is an endoplasmic reticulum calcium sensor that links luminal calcium levels to a redox activity (Pozzer et al., 2019).

At week 6, dietary VE deficiency upregulated hepatic Selenoi, Selenos, Txnrd1, and Txnrd2 mRNA levels at week 2. While three hepatic selenoprotein genes (i.e., Selenoi, Txnrd1, and Txnrd2) show the same trend at two time points, the expression of Gpx1 and Selenos was affected by VE supplementation at different time points in the liver. Overall, the selenoprotein gene mRNA levels have an inverse relationship with VE status, and further protein assays should be conducted in the future. In the liver, three additional selenoprotein genes (i.e., Selenon, Selenot, and Dio1) were downregulated by dietary Se deficiency at week 2. However, among the 25 genes assayed, only Selenoi mRNA level was decreased by dietary Se supplementation. Additionally, liver mRNA levels of Selenon were decreased in Se-adequate chicks by dietary VE supplementation. The gene code for protein is associated with neuroprotective function and cytosolic calcium regulation (Selenom) (Huang et al., 2016). In conclusion, responses of selenogenome expression in the liver of chicks to dietary Se and VE deficiencies differed between the time period before and the time period after the onset of liver necrosis. These differences may offer us a new clue on the role of selenoproteins in the pathogenesis of liver necrosis.

Although Se deficiency affects gene expression in multiple tissues (Liu et al., 2012; Yao et al., 2013; Sun et al., 2018), the selenogenome that are regulated is diverse, tissue-specific, and depends on their functions in the tissue (Joseph and Peter, 2018). The liver, in addition to its target organ of liver necrosis, has many important functions in performing and regulating diverse metabolic processes (Zhang Y. et al., 2020). In order to better understand the impact of selenoprotein influenced by liver necrosis, this study revealed novel regulations of protein production of GPX1, GPX3, GPX4, SELENOP, SELENON, SELENOW, and MsrB1 in the liver of chicks by dietary Se and/or VE at weeks 2 and 4. The tissue protein levels of these seven selenoproteins responded to dietary Se concentrations in three shapes. First, the levels of four selenoproteins (i.e., GPX1, GPX4, SELENOW, and SELENON) in the liver showed similar downregulations by dietary Se deficiency at two time points. At present, the functions of GPX1, GPX4, SELENOW, and SELENON are key redox enzymes in oxidative defense. Second, the abundance of SELENOP in the liver was elevated by dietary Se deficiency at two time points. SELENOP accounts for >50% of the total plasma Se content and exists in the blood as at least two isoforms with molecular weights of 50 and 60 kDa, respectively (Huang et al., 2015). Studies on transgenic mice lacking the Selenop gene have shown that this protein has an important function in delivering Se from the liver to extra-hepatic tissues, such as the brain and testis (Evangelos et al., 2018; Saito, 2019). The opposite influence of SELENOP between mRNA and protein by Se supplementation is hard to explain. Finally, GPX3 and MsrB1 showed no change of protein levels in the liver at two time points. GPX3, also known as plasma or extracellular GPX, is a selenoprotein mainly secreted by the proximal tubule cells of the kidney. MsrB1 (also called selenoprotein R) was first identified as a selenoprotein through bioinformatics methods (Lee et al., 2020). MsrB1 deficiency exacerbates the hepatotoxicity caused by acetaminophen via increased oxidative damage (Kim et al., 2017). Among the seven selenoproteins assayed, only the GPX4 level was decreased by VE supplementation in the liver of +Se chicks. Downregulation of muscle GPX4 protein by VE supplementation in the +Se chicks may be partially explained by a mutually sparing mechanism due to their overlap function of protecting against lipid peroxidation. As mentioned above, tissue Gpx4 mRNA, GPX4 activity, and GPX4 protein levels were affected by dietary Se in chicks. Another unique feature of selenoprotein abundance in broilers is that GPX4 protein levels were ~83–99% lower in the liver of –Se chicks, compared with those of +Se chicks. GPX4 is the only GPX that accepts phospholipid hydroperoxidase as an oxidation substrate in membranes, as well as uses protein-thiol groups as a reduction substrate in the absence of GSH (Ding et al., 2021). Since early embryonic lethality of GPX4-knockout mice limits further in vivo studies on the molecular and cellular mechanisms of GPX4, we have speculated that avian GPX4 could function as a redox sensor (Huang et al., 2015). In this study, the incidences of hepatic necrosis in chicks were completely prevented by dietary Se supplementation, and the necrotic mechanism induced by Se deficiency was still unknown. In this study, we identified appropriate antibodies from multiple sources to assay several major cell signaling pathway proteins that may be related to RIPK1/RIPK3/MLKL and MAPK pathway (see below). Compared with those of the Se-adequate chick, the Se-deficient chick had greater hepatic protein levels of RIPK1, RIPK3, MLKL, p38, p-p38, JNK, p-JNK, ERK, and p-ERK. Previous studies have shown that cell necroptosis was determined by increasing the levels of RIPK1, RIPK3, and MLKL and decreasing the expression of caspase 8 (Guida et al., 2017; Cui et al., 2020). Activation of RIPK1 triggers mitochondrial damage, which in turn promotes higher levels of oxidative stress and ultimately leads to cell death. It is suggested that the apoptotic pathway may partially develop into necrosis (Yan et al., 2018). Se deficiency leads to MLKL-dependent necroptosis and mitochondrial apoptosis in Se-deficient animal models (Li et al., 2019; Wang H. et al., 2020). The ubiquitination of RIPK1, RIPK3, and recruitment of the downstream mediator MLKL resulted in a necrotic-like cell death process called necroptosis (Shi et al., 2019). The previous study demonstrated that arbutin protects retinal ganglion cells against oxidative injury induced by H₂O₂ via inhibiting the p38MAPK and MEK/ERK signaling pathways (Zhao et al., 2019). Induction of phosphorylated p38, JNK, and ERK, the active forms of the protein kinases, occurs due to Se deficiency in chicks. Compared with the wild type, the two knockout mice (i.e., SOD^−/−and GPX1^−/−) had higher levels of phospho-p38 MAPK (Wang et al., 2011). Meanwhile, there were significantly increased levels of phosphorylated ERK1/2 in the GPX4^+/−mice. JNK is a group of mitogen-activated protein kinases activated by cytokines or environmental stress that participates in a variety of signaling pathways, including apoptosis pathways. Interestingly, the low activity of GPX1 (4% of the adequate level) in Se-deficient mice was sufficient to attenuate low-dose pro-oxidant-mediated JNK phosphorylation at Thr-183/Tyr-185 (Huang et al., 2018). Overall, these signaling proteins were first reported in hepatic necrosis induced by Se and/or VE deficiency, and the postulated mechanism was that the RIPK1/RIPK3/MLKL and MAPK signaling pathways were activated by Se deficiency to protect the liver from necrosis.

In summary, our data revealed the differential regulation of dietary Se deficiency on 25 selenoprotein genes, 7 key selenoproteins, and the RIPK1/RIPK3/MLKL and MAPK signaling pathway proteins in chicks and provided new molecular clues for understanding the pathogenesis of liver necrosis. The protective effects of Se are probably exerted via peroxide scavenging, controlling oxidative stress, and regulating subsequent stress signaling (Miyata et al., 2020). As shown in Figure 5, the main stress signaling proteins were activated, including p38 MAPK, JNK, and ERK. Additionally, the RIPK1/RIPK3/MLKL signaling pathway was involved in mediating the protection conferred by the five selenoproteins against the onset of liver necrosis (Wang H. et al., 2020).

The original contributions presented in the study are included in the article/Supplementary Material, further inquiries can be directed to the corresponding authors.

The animal study was reviewed and approved by China Agricultural University.

J-QH provided ideas for the experiment. TL and JZ completed the experiment and wrote the manuscript. Z-WZ and P-JW completed the parts of the figures. All authors contributed to the article and approved the submitted version.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.