Gut metabolite Urolithin A mitigates ionizing radiation‐induced intestinal damage

Abstract Ionizing radiation (IR)‐induced intestinal damage is the major and common injury of patients receiving radiotherapy. Urolithin A (UroA) is a metabolite of the intestinal flora of ellagitannin, a compound found in fruits and nuts such as pomegranates, strawberries and walnuts. UroA shows the immunomodulatory and anti‐inflammatory capacity in various metabolic diseases. To evaluate the radioprotective effects, UroA(0.4, 2 and 10 mg/kg) were intraperitoneally injected to C57BL/6 male mice 48, 24, 1 h prior to and 24 h after 9.0Gy TBI. The results showed that UroA markedly upregulated the survival of irradiated mice, especially at concentration of 2 mg/kg. UroA improved the intestine morphology architecture and the regeneration ability of enterocytes in irradiated mice. Then, UroA significantly decreased the apoptosis of enterocytes induced by radiation. Additionally, 16S rRNA sequencing analysis showed the effect of UroA is associated with the recovery of the IR‐induced intestinal microbacteria profile changes in mice. Therefore, our results determinated UroA could be developed as a potential candidate for radiomitigators in radiotherapy and accidental nuclear exposure. And the beneficial functions of UroA might be associated with the inhibition of p53‐mediated apoptosis and remodelling of the gut microbes.

These studies suggested that gut metabolites could be alleviating IR-induced intestinal damage.
Urolithin A (UroA) is a gut metabolite of ellagitannin, a compound in fruits and nuts such as pomegranates, strawberries and walnuts. It shows various pharmacological effects, including regulating oestrogen receptor and androgen receptor, antioxidant, antiinflammation and anti-ageing ability and so on. UroA can restore the colon damage induced by high-fat diet and regulate the gut microbiota. 6 But the effects of UroA in IR-induced intestinal damage are still unknown.
In present research, we detected whether UroA ameliorates radiation-induced intestinal damage, and further explored the possible mechanism involved. The results proved UroA markedly elevated the survival rate and the average survival days of mice exposed by lethal dose of IR and improved the intestinal tissue morphology architecture and the regeneration ability of enterocytes. Therefore, the inhibition of p53-mediated apoptosis and the reshaping of gut microbes are indispensable for the beneficial effects of UroA. We demonstrated that UroA might be a potential candidate for radiomitigators in radiotherapy and accidental nuclear exposure.

| MATERIAL S AND ME THODS
Detailed materials and methods are provided in the Appendix S1 section.

| UroA improves the survival and intestinal morphology and function of mice after 9.0 Gy IR
To determine the antioxidant activity of UroA, the DPPH radical scavenging assay was performed. We applied melatonin as the positive control for its antioxidant capacity of driving the synthesis of superoxide dismutase and glutathione peroxidase. 7,8 As Figure 1A shown, the F I G U R E 1 UroA shows excellent antioxidant capacity and improves the survival and intestinal morphology and function of mice after 9.0 Gy TBI. (A) The antioxidant capacity of UroA and melatonin was compared and detected by DPPH. The DPPH free radical scavenging activities by 0.025(p < 0.001, 95% confidence interval: −41.86 to −34.63), 0.05(p < 0.001, 95% confidence interval: −54.13 to −46.34), 0.1(p < 0.001, 95% confidence interval: −61.26 to −53.75), 0.2(p < 0.001, 95% confidence interval: −65 to −57.98), 0.3(p < 0.001, 95% confidence interval: −65.12 to −57.27), 0.4 (p < 0.001, 95% confidence interval: −62.83 to −56.09) mg/ml of UroA were significantly higher than that of melatonin. n = 3 samples per group. * * * p < 0.001. (B) The survival of mice in three groups. Log-rank (Mantel-Cox) test suggested that compared with the IR group, all three doses of UroA (0.4, 2 and 10 mg/kg) significantly developed the survival of mice exposed by 9.0 Gy TBI (p = 0.0076, p < 0.001 and p = 0.0015). (C) The average survival days of mice in five groups. One-way ANOVA test suggested that compared with the control group, the average survival days of IR group had significantly reduced (p < 0.001, 95% confidence interval: −5.477 to −4.723); compared with the IR group, the average survival days of UroA (0.4, 2 and 10 mg/kg) group had significantly increased (p = 0.1649, 95% confidence interval: −1.566 to 0.166, p < 0.001, 95% confidence interval: −2.766 to −1.034 and p = 0.0009, 95% confidence interval: −2.166 to −0.434). The mice in the control group and the IR group were injected intraperitoneally with phosphatebuffered saline, and the mice in the IR+UroA group were injected intraperitoneally with 0.4, 2 or 10 mg/kg UroA at 48, 24, 1 h before and 24 h after the 9.0 Gy TBI. n = 10 mice per group. **p < 0.01, ***p < 0.001. (D) Representative haematoxylin and eosin staining images of small intestine. (E) The quantified results of crypts counting. Two-tailed unpaired t test suggested that compared with the control group, the number of crypts in IR group significantly reduced (***p < 0.001, 95% confidence interval: 4.746-10.50); compared with the IR group, the numbers of crypts in UroA group significantly increased (***p < 0.001, 95% confidence interval: −10.87 to −6.435). (F) Representative Villin immunostaining images of the small intestinal tract. (G) The quantified results of villous height. Two-tailed unpaired t test suggested that compared with the control group, the villous height in IR group significantly reduced (***p < 0.001, 95% confidence interval: 94.86-204.9); compared with the IR group, the villous height in UroA group significantly increased (p = 0.0378, 95% confidence interval: −91.17 to −2.734). n = 5 mice per group. *p < 0.05, **p < 0.01, ***p < 0. DPPH free radicals scavenging ability of UroA significantly increased in a dose-dependent manner between concentration range from 0.025 to 0.4 mg/ml, and it is much stronger than melatonin of the same concentration.
Three doses of UroA(0.4, 2 and 10 mg/kg) were intraperitoneally injected to mice 48, 24, 1 h prior to and 24 h after 9.0 Gy TBI, and monitored for 10 days. Compared with the IR group, all three doses of UroA could improve the survival rate of irradiated mice. But the 2 mg/kg UroA performed best (p < 0.001). We found that all mice were died in the IR group at 5 days after 9.0 Gy TBI ( Figure 1B), 10% survival in 0.4 mg/kg UroA group, 70% survival in 2 mg/kg UroA group and 40% survival in 10 mg/kg UroA group. As shown in Figure 1C, the average survival days of mice are 4.1 days (the IR group), 4.8 days (the IR + 0.4 mg/kgUroA group), 6.0 days (the IR + 2 mg/kgUroA group) and 5.4 days (the IR + 10 mg/kgUroA group). Results above indicated that UroA could improve the survival of mice exposed with lethal dose IR, and 2mg/kg performed best. Therefore, we selected 2 mg/kg UroA as the dose of following intestinal analyses.
As shown in Figure 1D-G, the images of haematoxylin and eosin staining showed the identical enterocytes death, the villus incompletion, crypts drop out and mucosal inflammation. At the 3rd day after 9.0 Gy TBI, the crypts numbers of the small intestine in IR group significantly decreased compared with the control group (p < 0.001).
Compared with the mice in the IR group, IR + UroA group showed more survival crypts (p < 0.001). The villin height of the IR group was also markable lower compared with the control group, while the UroA treatment could rescue the loss of villin height reduced by 9.0 Gy TBI. Collectively, we suggested that UroA can protect against the intestinal tract morphology damage of mice induced by IR.
Failure to maintain normal mucosal structure and function after IR indicates the defection of regeneration ability. After IR 3 days, the number of Lgr5 + positive cells was notably decreased in the small intestine, while the loss of Lgr5 + was relieved by UroA ( Figure 1H Figure 2E-F, IR could significantly increase the 8-OHdG content in the small intestine, and UroA could significantly alleviate the increase in 8-OHdG in small intestine of mice exposed with radiation.
Since p53 protein is the key molecule responding to the cellular DNA injury by triggering apoptosis or inducing a cell cycle arrest, we assessed the level of p53 protein by IF analysis. As shown in Figure 2G-H, IR induced the overexpression of p53, while the p53 in intestinal tissue was remarkably restored to the normal level by the administration of UroA. As shown in Figure 2A-D, the upregulation of caspase8 and caspase3 was observed in the small intestine of mice exposed with 9.0 Gy TBI, while the effect was distinctly prevented by the treatment of UroA. These results indicated that UroA could downregulate the p53-mediated apoptosis pathway downstream molecules to ameliorate the radiation-induced intestinal damage.

| UroA recovers the changes of gut microbiota in mice after TBI
Gut microbes play a crucial role in gastrointestinal pathology caused by IR, and UroA was studied previously to modulate the profile of gut microbes. Therefore, the effect of UroA in the gut of irradiated mice might be associated with the composition change of the gut microbes. Faecal samples of irradiated mice were harvested 3 days after IR, and 16S ribosomal RNA analyses of microbiota were performed. As shown in the flower plot ( Figure 2N), among the 4772 identified operational taxonomic units (OTUs), 107 of them were the core OTUs that exist in each sample. As shown in Figure 2Q, the microbes Euryarchaeota and Nitrospirae are rich in the IR + UroA group and lack in the IR group, which associate with the benefit mechanism of UroA on radioprotective property. The main composition of phylum Euryarchaeota is Methanogens which is flourish in inflammatory bowel disease patient compared with healthy individual. 10 The loss of Nitrospirae and the flourish of Actinobacteria are two signs associated with histological stages of gastric carcinogenesis. 11 In our study, IR obviously upregulated the abundance of Actinobacteria compared with control group; and UroA relieved this change of community structure. To determine the alpha diversity of the gut microbes, the Chao1 index, the Shannon index and the Simpson index ( Figure 2M) using pairwise comparisons using Wilcoxon rank-sum test were used. There was no significant difference among three groups, suggested species diversity of microbial composition between each group is limited. As for beta-diversity, the principal component analysis (PCA) ( Figure 2O) and the principal co-ordinates analysis (PCoA) ( Figure 2P) revealed that irradiation caused an obvious change in the structure of gut microbes compared with the control group, while UroA treatment was confirmed a distinct segregation in the gut microbes profile paralleled with the IR group.
In order to clarify the changes of IR and UroA on gut microbe composition of irradiated mice, we performed statistical analyses at levels of the class and family, and drew heatmaps based on the relative abundance of different species. As seen in Figure 2I, the results did not show obviously differences among three groups at classes level. Contrarily, As shown in Figure 2J, 9 Gy IR exposure increased the relative abundance of the family Flavobacteriaceae, Enterobacteriaceae, Bifidobacteriaceae and Erysipelotrichaceae.
However, UroA treatment ameliorated the bacteria differences induced by 9.0 Gy IR. Relative abundance of top 10 different gut microbes at the class and family level showed in Figure 2K-L, and the abundance of Enterobacteriaceae and Erysipelotrichaceae in IR group was upregulated compared with others (p < 0.05).
Recently, the association between gut microbes and host health has been one of the most popular research topics. Studies have found that microbial metabolites promote the proliferation and regeneration of ISCs. 12 Owing to the effect of anti-inflammation property in protecting against radiation-induced intestinal damage, we wonder whether UroA to some extent benefits organism by the influence on gut microbe composition. To scrutinize the effect of UroA on intestinal tract gut microbe composition, our study presents 16S rRNA sequencing analyses for three groups. The results showed that IR could cause gut microbes imbalance and a downregulation of beta-diversity, accordance to the clinical counterpart. 13 However, the administration of UroA could significantly alleviate the disturbance of gut microbes and the decrease of gut microbes abundance caused by radiation damage, indicated that reshape of gut microbes is also a mechanism of UroA ameliorate radiation-induced intestinal damage. Escherichia shigella could mediate the transfer of virulence protein into host cytoplasm, destroy epithelial integrity and functions and subsequently promote cell death. 14 Our result showed that radiation flourished the abundance of Escherichia shigella, and UroA administration produced a significantly remission ( Figure 2R). Since Proteobacteria plays an important role in facilitating inflammatory response by regulating IL-10/IL-17 balance and the production of TGFβ, the upregulation of Proteobacteria level in gut reflects dysbiosis or a tender status of gut microbial community. 15 In present study, we found that radiation exposure leaded the bloom of Alphaproteobacteria, and UroA administration sharply relieved it ( Figure 2S-T). Erysipelotrichaceae is associated with the host lipid metabolism, positively correlated with TNFα and cause inflammation in gut. We found that irradiation remarkedly increased the abundance of Erysipelotrichaceae, however, UroA administration slightly ameliorated this change. According to results above, we suggested the radioprotective capacity of UroA is associated with the remodelling of gut microbes.
In summary, we demonstrated that UroA improves the maintenance of the intestinal homeostasis and the ability of regeneration exposed with radiation. These benefits of UroA might be associated with the inhibition of p53-mediated apoptosis and remodelling of the gut microbes. Therefore, UroA could be a potential target for radiomitigators in radiotherapy and accidental nuclear exposure.

CO N FLI C T O F I NTE R E S T
The authors declare that there is no conflict of interest.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data used to support the findings of this study are available from the corresponding author upon request.