Therefore we believe that HIF mediated effects on

Therefore, we believe that HIF-mediated effects on CSC markers, apoptosis are mediated through HO-1. Our plan is to look for more mechanistic studies to understand the role of HO-1 degradation products (CO, Fe, and bilirubin) in cellular apoptosis, stemness under hypoxia. The model of role of HO-1 in pancreatic cancer has been modified accordingly and added as Fig 7.
Acknowledgments
Introduction Deoxynivalenol (DON) is one of the main kinds of trichothecenes. DON contaminates cereals and their by-products and cause a loss to agricultural economy worldwide. It is known that DON cannot be totally heat-inactivated during food processing due to its stable nature under high temperature (150–170 °C) and many harsh conditions (Chemists, 1983; Tanaka et al., 2009; Mishra et al., 2014a). Therefore, DON or DON-contaminated foods can induce different toxic effects on human and animals that include inhibition of translation, protein synthesis and mitochondrial function at molecular level (Pestka, 2010) as well as gut dysfunction, alteration of immune system and endocrine/brain systems through production of proinflammatory cytokines and of gut-correlated hormones (Maresca, 2013; Maresca et al., 2018; Terciolo et al., 2018). Particularly, according to previous studies on human, DON could trigger gastroenteritis and potentially induce esophageal cancer (Luo et al., 1990). Liver is responsible for the detoxification and formation of DON-glucuronide in both human and animals. In our previous review publication, we have systematically analyzed and summarized different mechanisms of DON-induced liver damages (hepatotoxicity) (Peng et al., 2017). For instance, levels of interleukin-1 beta (IL-1β), interleukin-6 (IL-6) and tumor necrosis factor-alpha (TNF-α) were elevated in the adult mouse by different concentrations of DON treatments (Amuzie et al., 2008). DON altered expressions of apoptosis-related genes in liver tissues of mice and different cell lines (Li et al., 2014; Deng et al., 2016; Ren et al., 2017). Tardivel and co-workers also found that persistent low DON concentration (25 μg/kg bw/day for 30 days) could trigger pro-inflammatory genes expression in brain, liver, duodenum and adipose tissues (Tardivel et al., 2015). In addition to above-mentioned effects, whether there are other mechanisms involved in DON-induce liver damages call our attention. The gut microflora is a complex ecosystem, consisting of numerous species of microorganisms, and the microbial gene copies are about 100 times more than that of the host while the ratio of bacteria versus human csf1r is much close to 1:1 (Sender et al., 2016). Therefore, the gut microbiota has been regarded as an “invisible organ” that interacts with the host in a broad spectrum ranging from colonization resistance, nutrient utilization to immune maturation. Recently, it has been considered as a new and critical player in the pathophysiology of many diseases as well (Lynch and Pedersen, 2016). Liver closely interrelates with the intestinal tract through the portal system and the bile secretion system (Cesaro et al., 2011). It has been found that the alterations of gut microbiota were associated with liver disorders (liver cirrhosis, alcoholic and non-alcoholic liver disease, hepatic encephalopathy and primary sclerosing cholangitis) (Qin et al., 2014; Tabibian et al., 2014; Minemura and Shimizu, 2015; Haque and Barritt, 2016), which suggests a “liver-gut-axis”. Several lines of clinical and laboratory evidences also demonstrated the critical role of gut microbiota in liver diseases. For instance, typical complications of liver disease, such as hepatic encephalopathy, were efficiently treated by various prebiotics, probiotics and antibiotics (Tilg et al., 2016). It is also controversial as to whether DON could affect gut microbiota. For example, during the DON treatment (2.8 mg/kg, 4 weeks), the colony forming units (CFUs) of Aerobic Mesophilic Bacteria were significantly increased during the first 7 days and then reached a plateau (days 7 to 14) (Wache et al., 2009). DON could also increase CFUs of Bacteroides and Prevotella, but showing no effect on Lactobacillus, Leuconostoc, pediococcus and Clostridium leptum in rats (Saint-Cyr et al., 2013). On the contrary, Payros and coworkers demonstrated that DON did not impact the composition of gut microbiota (Payros et al., 2017). These conflicting observations need to be further investigated. Meantime, recent findings supported that host gene mutations or different expression patterns could shape gut microbiota (bidirectional response) (Goodrich et al., 2014; Bonder et al., 2016; Hall et al., 2017; Kurilshikov et al., 2017). Heme oxygenase (HO) enzyme system provides an essential enzymatic activity by catalyzing degradation of heme to produce biliverdin-Ixα (BV), ferrous iron and carbon monoxide (CO) (Tenhunen et al., 1968; Tenhunen et al., 1969). Since heme serves as an important mediator of diverse vital biological processes including oxygen transport and delivery to cells and tissues, peroxide metabolism, xenobiotic detoxification, HO enzymes may exert a crucial role in metabolism by regulating heme bioavailability and turnover in cells and tissues (Maines, 1988; Ryter and Choi, 2016). Three isoforms of heme oxygenase are known, among which heme oxygenase 1 (HO-1) is an inducible isoform in response to various stress such as oxidative stress, hypoxia, cytokines, etc. (Alam et al., 1989; Keyse and Tyrrell, 1989). HO-1 may function as a pleiotropic modulator in inflammatory signaling programs and immunomodulatory effects (Ryter and Choi, 2016). Specifically, previous studies reported that HO-1 could attenuate various complications of liver diseases and other pathological conditions (Shibahara et al., 2002; Takahashi et al., 2004). In addition, HO-1 exerts an influence on gut microbiota to a certain degree, as recent evidence has shown that HO-1 could be induced by the enteric microbiota, thereby promoting bacterial clearance by carbon monoxide and macrophages in human inflammatory bowel diseases (IBD) (Onyiah et al., 2013; Onyiah et al., 2014). These previous studies could provide a new idea for further finding relationships about gut microflora, HO-1 and DON-induced liver damages.