Integrative multi-omics analysis of radionuclide-induced intestinal injury reveals the radioprotective role of L-citrulline through histone H3-mediated Cxcl3

Abstract

Background

The widespread application of nuclear technology has markedly heightened the risk of extensive, uncontrolled exposure to radiation. Nevertheless, in contrast to external irradiation, the biological impacts and countermeasures against internal irradiation from radionuclides remain inadequately characterized.

Methods

Mice were administered yttrium-90 (Y90) carbon microspheres via gavage at different dosages (0–5.0 mCi) to establish a radionuclides exposure model. A multi-omics analysis was employed to access alterations in gut microbiota, fecal and colonic metabolites profiles, and intestinal mRNA expression post-irradiation. The function of significant metabolite was validated at both cellular levels and organismal levels. Additionally, ChIP-Seq and RNA-Seq techniques were utilized to investigate the molecular mechanism underlying the actions of key metabolite.

Results

Exposure to Y90 resulted in intestinal damage and hematological impairment. Multi-omics analysis revealed significant alternations of gut microbiota, fecal metabolites, colonic metabolites, and intestinal mRNA expression following internal radiation exposure. Notably, L-citrulline was identified as a metabolite with changes observed in both fecal and colonic tissues, demonstrating radioprotective properties in vitro and in vivo. Mechanistically, L-citrulline facilitated the citrullination of histone H3 at the 17th site (H3Cit17), and multiple mRNAs including C-X-C motif chemokine ligand 3 (Cxcl3), were transcriptionally regulated by H3Cit17 post L-citrulline treatment. Furthermore, Cxcl3 conferred protective effects for intestinal epithelial cells against ionizing radiation.

Conclusions

The research offers critical perspectives on the intestinal and gut microbiota’s reaction to radionuclides exposure. It underscores the promise of L-citrulline as a radioprotective compound, which may have substantial ramifications for the formulation of strategies to mitigate radiation exposure.

Background

Nuclear technology is integrated into daily life globally [1], with applications extending beyond energy production to encompass fields such as medicine, industry, and agriculture [2]. Nevertheless, the complex and volatile global environment heightens apprehensions regarding nuclear accidents and terrorism, as evidenced by the use of uranium munitions in armed conflicts and the release of Fukushima wastewater. In the natural environment, radionuclides dissolved in water exhibit inherent mobility [3, 4]. If these radionuclides are inadequately managed, they may permeate aquatic ecosystems, posing a risk of contamination of both river systems and groundwater supplies. This will inevitably increase the risk of human exposure to radionuclides [5]. The potential hazards associated with public exposure to radiation arise not only from external irradiation (originating from sources outside the body), but also from internal irradiation (from source within the body). Radionuclides can enter the human body primarily through three pathways, leading internal irradiation: ingestion of food or liquids containing radioactive substances; inhalation of radioactive gases or aerosolized particles; and dermal absorption or contamination through wounds [6]. Internal irradiation presents a hazard to human health by subjecting organs and tissues to prolonged exposure to radioactive substance [7]. Furthermore, the administration of radiopharmaceuticals may also result in adverse toxic effects across multiple organ systems [8].

Intestinal tissue, known for its extraordinary sensitivity to ionizing radiation, is susceptible to degeneration and necrosis when humans experience high-dose (> 10 Gy) whole-body external radiation exposure [9]. Regarding internal irradiation, intestinal damage could result from ingesting radionuclides. The intestinal injury caused by internal irradiation is characterized in intestinal barrier integrity, dysbiosis of the gut microbiota and immune system dysfunction, all of which contribute to a multifaceted pathophysiological response. This damage can lead to diarrhea, vomiting and abdominal pain, potentially progressing to inflammation or malignancy within the intestine tract. Animal research models have demonstrated a significant finding: after the ingestion of polonium-210 (Po210, Po²¹⁰, 210Po, ²¹⁰Po), contrary to the expected absorption by the intestinal mucosa, there is actually adhesion occurring on the luminal surface [10]. Furthermore, it has been observed that rats administered ²¹⁰Po orally experienced severe damage to their intestinal epithelium [11]. However, the cellular and molecular mechanisms associated with internal radiation pathology remain poorly understood, and there is a scarcity of studies that have clarified these intricate processes.

Additionally, the intestine is inhabited by a diverse array of microorganisms, including Firmicutes, Bacteroidaceae, Muribaculaceae, Lachnospiraceae and Prevotellaceae [12]. Recent studies indicate that ionizing radiation can alter the gut microbiota and associated metabolites, potentially affecting the repair mechanism of radiation-induced intestinal damage [13]. However, existing research has predominantly focused on external irradiation, leaving a gap in our understanding of internal radiation exposure on these biological processes. It is imperative to investigate the impact of internal irradiation on gut microbiota and metabolites to develop innovative strategies and solutions for radiation protection against internal exposure.

With the development of modern high-throughput multi-omics measurement platforms, it has become essential for biomedical studies to undertake an integrative approach to fully utilize these data to gain insights into biological systems [14]. Data derived from diverse omics domains, including genomics, proteomics, and metabolomics, can be integrated to unravel the intricate working of systems biology through machine learning-based predictive algorithms [15]. In this study, in vivo internal radiation models were established utilizing Y90 carbon microsphere-based gavage (0–5.0 mCi dosage). We then evaluated the changes of intestinal microbiota, metabolites, and intestinal tissue mRNA following internal irradiation through multi-omics analysis. By employing a multi-omics analytical framework, we focused on L-citrulline and subsequently examined its radioprotective properties. L-citrulline confers radioprotection in intestinal epithelial cells and murine intestine, which has the potential to be developed as a protective drug against internal irradiation.

Methods

Yttrium-90 carbon microsphere

Y90 carbon microsphere is produced and provided by Chengdu New Radiomedicine Technology Co., Ltd. (Chengdu, China). Y90 carbon microsphere is obtained by loading radioactive Y90 on the carbon microsphere through chemical adsorption, sodium phosphate curing treatment, and then sterilized. The loading rate of Y90 is over 99% and the leaching rate of Y90 is less than 1/1,000,000. The particle size of Y90 carbon microsphere is 20–45 μm. The absorbed dose of intestine in mice after gavage is calculated through formula [16] in the supplemental materials. All animal experiments related to Y90 microspheres are conducted at Chengdu New Radiomedicine Technology Co., Ltd., which possesses an experimental animal platform and testing technology platform capable of conducting preclinical research and testing of radioactive drugs.

Animals and treatments

Male C57BL/6 mice that were 6–8 weeks old and weighed 22–25 g were purchased from the Chengdu dossy experimental animals Co. Ltd. (Chengdu, China). Animals were housed under controlled conditions of 20–22 °C and relative humidity of 50%, a fixed 12 h light/dark cycle, and ad libitum access to food and water. Especially, in the experiment of establishing an internal irradiation model through Y90 gavage, since the 16s RNA sequencing of the mice will be carried out subsequently, to ensure the accuracy and reliability of the experimental results, the mice were housed in the SPF-level environment provided by Chengdu New Radiomedicine Technology Co., Ltd. In this environment, all items entering the breeding area, including feed, bedding and various experimental equipment, were subjected to a strict sterilization process to prevent external microorganisms from interfering with the experimental environment. For relevant operators involved in the experiment, before entering the mouse breeding environment, they must strictly follow the hygiene and disinfection regulations. Specifically, they needed to change into special work clothes and shoe covers, and wear masks and hats in a standardized manner. This effectively prevents the introduction of external microorganisms into the breeding environment, providing a strong guarantee for the stability of the intestinal flora of the mice. This work has received approval for research ethics from the Animal Experimentation Ethics Committee at Sichuan University (Chengdu, China) and a proof of approval is available upon request. The detailed methods were described in the supplementary materials.

Multi-omics analysis

To investigate the changes of fecal and colonic metabolites, gut microbiota and colon mRNA, we conducted multi-omics analysis, incorporating techniques such as metabolomics, microbiome profiling, and transcriptomics. The multi-omics analysis was performed in the laboratory of OE Biotechnology Company (Shanghai, China). The detailed methods were described in the supplementary materials.

RNA-sequencing (RNA-Seq)

Details of comparative analysis of RNA-Seq were reported previously [17,18,19]. IEC-6 cells were pretreated with or without L-citrulline and then exposed to X-rays (10 Gy). Cell samples were subjected to conduct RNA-Seq. RNA-Seq was performed and analyzed by Igenebook Biotech Co., Ltd. (Wuhan, China). The detailed methods were described in the supplementary materials.

Chromatin immunoprecipitation sequencing (ChIP-Seq)

IEC-6 cells were pretreated with or without L-citrulline for 24 h and then exposed to X-rays 48 h post-irradiation, cells were collected for ChIP-Seq. ChIP-Seq was performed at Igenebook Biotech Co., Ltd. (Wuhan, China). Briefly, the cells were cross-linked with formaldehyde, lysed and sonicated to produced chromatin fragments of 200–500 bp, which were then immunoprecipitated with anti-Histone H3 (citrulline R17) (#ab219407; Abcam, Cambridge, UK), normal rabbit IgG (2.5 μg; #A21020, Abbkine, Wuhan, China) was included as a control. The detailed methods were described in supplementary materials.

Statistical analysis

An unpaired 2-tailed Student’s t test was used to compare the means of two groups. Multiple comparisons were performed with one-way ANOVA with post hoc tests by Tukey’s test. A P value less than 0.05 was considered statistically significant for all tests.

Other methods are detailed in the supplementary materials.

Results

Internal radiation exposure reduces the survival and disrupts the intestine of mice

To explore the biological consequences of internal radiation exposure, an internal irradiation murine model was established by administering different concentration of Y90 microspheres. The radioactivity levels for the different groups were set at 0.2, 0.5, 1.0 and 5.0 mCi. Then, animal survival was monitored for a duration up to 70 days. during which colon tissue and feces were collected for multi-omics analysis (Fig. 1A). For different Y90 activity groups, a formula was employed to calculate the absorbed dose in intestinal tissue, with specific dose illustrated in Fig. 1B. However, the above calculation method assumes that all radiation doses are delivered to a single point, given the extensive surface area of the intestine and its dynamic peristaltic motion, the actual dose absorbed by intestine is much lower than that calculated by the formula. The survival of the animals was tracked over a maximum period of 70 days. The results indicated that gavage with 5 mCi of Y90 led to 75% lethality rate among C57BL/6J mice within the initial 6 days, followed by 100% lethality within 8 days post-treatment. Mice administered 1 mCi of Y90 via gavage exhibited a 25% lethality rate within 12 days, whereas the groups administered with 0.2 mCi and 0.5 mCi of Y90 demonstrated a 100% survival rate over the same timeframe (Fig. 1C). Compared to the control group, the irradiated group displayed lethargy, reduced mobility, and a tendency to cluster, indicating that internal irradiation a measurable degree of harm on the mice.

Fig. 1

Internal irradiation induces the death and intestinal injury in mice. (A) Schematic showing internal irradiation model in C57BL mice and the procedures for 16sRNA sequencing, metabolic analysis (LC-MS and GC-MS) and transcriptome sequencing. (B) Dose estimation after gavage of different doses of Y90 nuclide. (C) Kaplan-Meier analysis of male mice treated with radionuclides of different dose (0, 0.2, 0.5, 1.0 or 5.0 mCi). (D) Representative H&E-stained sections of the small intestines. Scale bar = 500 μm. (E) The height of small intestine villus and the number of crypts pre mm in different group. (F) and (G) Intestines of mice in different Y90 gavage groups. The data are shown as the mean ± SEM for three independent experiments. *P < 0.05; **P < 0.01

To investigate the effects of internal irradiation on intestinal tissues, mice were administered varying activities of Y90 microsphere via gavage and subsequently dissected to assess morphological changes in the intestine. The results showed that the degree of intestinal congestion, swelling and bleeding gradually increased in day 1 post internal irradiation with increasing irradiation dose. However, by day 3 post internal irradiation, the intestinal injury in the 0.5 mCi and 1 mCi groups was notably more pronounced (Fig. 1F and G). And small intestines of various groups were collected and observed by H&E staining (Fig. 1D). Histological examination of H&E-stained sections showed a normal crypt-villus architecture in the unirradiated intestines, whereas significant structural abnormalities were evident in the irradiation groups A shorten of the villi and reduction in the number of crypts were observed in the irradiation group, especially in the 5 mCi gavage group, suggesting a pronounced influence of internal irradiation on intestinal morphology (Fig. 1E). Hematological changes were observed between the internal irradiation group and control group. A complete blood count was performed on samples collected from the first to third day post-irradiation to assess the immediate effects of internal irradiation on the hematopoietic system of the mice. Results indicated that following gavage with varying activities of Y90, the white blood cell (WBC) counts decreased significantly on the second and third days after irradiation, dropping from approximately 10 × 10^9/L to about 2 × 10^9/L, while red blood cell (RBC) counts remained relatively stable (Figure S1). This phenomenon may be attributed to the difference in the sensitivity of white blood cells and red blood cells to radiation, compounded by the shorter lifespan of white blood cells. The more rapid response of white blood cells to radiation exposure results in more pronounced fluctuations in their numbers, whereas red blood cells, with their longer lifespan, exhibit a delayed response to radiation effects in the short term. All above results indicated that internal irradiation could cause various damages to the body.

Internal irradiation modulates the composition and diversity of gut microbiota

To investigate the characteristics of gut microbial community in mice subjected to internal irradiated mice, we compared the relative taxon abundance of microbiota in the Y90 irradiated group with that of the control group. A total of 6320 operational taxonomic units (OTUs) were annotated for subsequent analysis including 24 phyla, 59 classes, 133 orders, 208 families, 376 genera of gut microbes. The Venn diagram reflecting the distinctions between the groups (Fig. 2A), which exhibited 306, 186 and 253 differential OUTs in control, 0.5 mCi and 1.0 mCi groups, respectively. At the phylum level, Bacteroidata, Firmicutes and Proteobacteria were the predominant phyla identified in both internal irradiation groups compared with control group, contributing to 97.12% (0.5 mCi) and 97.48% (1.0 mCi) of the gut microbiota, respectively. While these phyla comprised 98.12% of the microbiota in the control group (Fig. 2B and D). The internal irradiated groups exhibited a significantly reduced abundance of Bacteroidota, alongside a relatively higher abundance of Firmicutes and Proteobacteria compared to the control group, especially in the 0.5 mCi group. The ratio of Firmicutes to Bacteroidota (F/B ratio) was elevated in the internal irradiated group, which is often used as an indicator of gut health and can be influenced by various factors.

Fig. 2

Internal irradiation modulates the composition and diversity of gut microbiota. (A) Venn diagrams shared OTUs between different groups. (B)-(D) The taxonomic profile demonstrated the OTUs are assigned to prevalent microbiome of Firmicutes, Bacteroidetes and Proteobacteria at phylum level. (E) The comparison of relative abundant microbiome at family level between internal irradiation group and control group. (F)-(H) Taxonomic summary of the gut microbiota of three groups at genus level. (I)-(L) The comparison of gut microbiota α diversity between each group, including species richness (represented by Chao1, observed species) and evenness (represented by Shannon and Simpson index). The result indicated that the α diversity of gut microbiota in 1.0 mCi group was higher than that of control group in terms of Shannon and Simpson index, while the Shannon and Simpson index in 0.5 mCi group is lower than control group. The richness index includes Chao1 and observed of internal irradiation group were lower than control group. Starred samples (*) were used to demonstrate the significant difference between the group. Statistically significant differences are indicated: Wilcoxon rank sum test. The top and bottom boundaries of each box indicate the 75th and 25th quartile values, respectively, and lines within each box represent the 50th quartile (median) values. Ends of whiskers mark the lowest and highest diversity values in each instance. (M) The heatmap of gut microbiota among 0.5 mCi, 1.0 mCi and control group. *P < 0.05; **P < 0.01

At the family level, there was a notable presence of higher and more dominant taxa of internal individual variation in taxonomic composition across each group. Among the total families identified within the gut microbiota, 203 dominant families were detected in the 0.5 mCi group, 206 in the control group and 1.0 mCi group. Muribaculaceae, Lachnospiraceae, Prevotellaceae, Ruminococcaceae, Bacteroidaceae, Rikenellaceae represented six most relative abundant microbiome in all groups (Fig. 2E). Lachnospiraceae and Oscillsopiraceae were increasing in internal irradiation group, while Muribaculaceae and Prevotellaceae were significantly decreasing in two internal irradiation group.

Studying the structure of microbiomes at the family level is unable to provide a deep understanding of the potential alternation caused by internal irradiation. For instance, either Muribaculaceae or Lachnospiraceae emerged as the most predominant families in three groups, while the analysis of genus level revealed a clear separation of genera belonging to this families (Fig. 2F-H). At the genus level, Muribaculaceae, Bacteroides and Faecalibacterium were the dominant genera in both control group and 0.5 mCi group. On the other hand, the microbiomes of the 1 mCi group were generally inhabited with Muribculaceae, Lachnospiraceae NK4A136 group and Bacteroides, indicating that high dose internal radiation led to an increase in the abundance of the Lachnospiraceae NK4A136 group. Additionally, the abundance of other genera was increased post internal irradiation, including Oddribacter, Lactobacillus and Escherichiia Shigella. These findings suggest that internal radiation exposure alters the composition of gut microbiota, necessitating further studies to identify the characteristic flora associated with intestinal damage due to internal irradiation.

To access the composition and diversity of microbial community, multiple indices have been evaluated. α diversity indices are regularly adopted in ecology to estimate the richness of microbial species and quantitatively describe microbial species diversity in a community. The larger of Shannon value and the smaller of Simpson value, the greater of Chao1 index and observed species, the richer of species in a community, the higher of diversity in a community [20]. To evaluate the alteration in microbiota community structure across different groups, the α diversity of microbiota was quantified as shown in Fig. 2I and J. The within sample α diversity analysis revealed a slight reduction in gut microbial community diversity following 0.5 mCi Y90 gavage compared to the control group, as assessed by the Shannon and Simpson indices. While the α diversity showed a slight increase in 1 mCi group relative to the control group, although this result was not statistically significant. The richness indices including Chao1 and observed species of 0.5 mCi group were lower than the control group (Fig. 2K and L). We have visually presented the changes in the top 30 microbial communities across the different groups using a heatmap. The pronounced differences depicted in the heatmap clearly demonstrate that the abundance of microbial communities experienced significant changes following internal radiation treatment (Fig. 2M), it is evident that the abundance of the microbial communities underwent significant changes after internal radiation treatment. Based on the above results, internal irradiation had an impact on composition and diversity of gut microbiota, with varying doses of internal irradiation exerting distinct effects on the microbiota. Generally, internal irradiation profoundly influences the intestinal microbiota, indicating that the metabolomic profiles associated with the gut microbiota will undergo alteration as well.

Multiple metabolites are dysregulated in mice fecal specimens after internal irradiation, which were associated with gut microbiota

To detect the alteration of fecal metabolite profiles, we adopted Gas Chromatography-Mass Spectrometry (GC-MS) and Liquid Chromatography-Mass Spectrometry (LC-MS) to detect the level of fecal metabolites. Compared with the control group, we identified 428 unique fecal metabolites in the 0.5 mCi group, and 209 unique fecal metabolites in the 1 mCi group, with 103 shared metabolites between the two irradiated conditions. (Fig. 3A, Table S1 and Table S2).

Fig. 3

Fecal metabolites are altered after internal irradiation and are associated with gut microbiota. (A) Venn diagram showing common and specific metabolites detected LC-MS in feces of irradiated mouse. (B)-(C) The correlation between gut microbiota (Top 30) and metabolites (Top 30) of 0.5 mCi group and 1.0 mCi. Correlation analysis was performed using Spearman’s ranks test; *P < 0.05, **P < 0.01. (D) Bar plot of the number of metabolites in different categories of 0.5 mCi vs. control group. (E) Venn diagram of the number of enriched metabolic pathways from metabolic pathway enrichment analysis (MPEA) in 0.5 mCi vs. control group. (F) Bar plot of the number of metabolites in different categories of 1.0 mCi vs. control group. (G) Venn diagram of the number of enriched metabolic pathways from MPEA in 1.0 mCi vs. control group. *P < 0.05; **P < 0.01

A spearman analysis was performed between the top 30 gut microbiota and the top 30 metabolites from the fecal samples. In the 0.5 mCi group, there was only LysoPE (16:1(9Z)/0:0) exhibited no significant correlation with any bacterial genus, while some metabolites that showed significant correlations with at least 10 genera, including PG (19:1(9Z)/0:0), citrulline, norvaline, saccharopine, etc. (Fig. 3B). In the 1.0 mCi group, there were three metabolites were not significantly correlated with any genera, including 3-α, 12-α-Dihydroxy-5-β-chola-7,14-dien-24-oic Acid, Arachidonic acid and LysoPE (18:0/0:0). A proportion of metabolites was only significantly correlated with one genus, while other metabolites were significantly correlated with at least two genera. Notably, L-citrulline, an α-amino acid obtained from watermelon, showed significant correlations with two genera (positively related with Colidextribacter, negatively related with Parabacteroides) (Fig. 3C).

Subsequently, we employed MetOrgin to analyze the gut microbiome and metabolome of fecal specimens. Compared to the control group, 218 metabolites identified in the 0.5 mCi group from TwinsUK study were categorized into four distinct groups: 3 host specific metabolites, 15 bacterial metabolites, 45 bacteria-host cometabolites, and 155 others (Fig. 3D). And metabolic pathway enrichment analysis (MPEA) identified two host specific pathways (ubiquinone and other terpenoid-quinone biosynthesis, metabolism of xenobiotics via cytochrome degradation); six pathways were specific to the microbial community (Fig. 3E). In 1 mCi group, a total of 151 identified metabolites from the TwinsUK study were initially classified into four groups: 1 host-specific metabolite, 9 bacterial metabolites, 41 bacteria-host co-metabolites and 100 others (Fig. 3F). The MPEA identified 1, 7 and 48 metabolic pathways from the host, microbiota and shared by both. Among them, fatty-acid biosynthesis pathway exhibited a significant correlation with the host, while ascorbate and aldarate metabolism, limonene and pinene degradation were specific to microbiota alone (Fig. 3G). Taken together, our data highlighted internal irradiation modulated the gut microbiota and their metabolites, and some of these metabolites may be co-metabolized by both the host and the microbiota.

The metabolites and mRNAs in colon tissues are altered by internal irradiation

To investigate the metabolic alterations of colon tissue, we employed GC-MS and LC-MS to quantify metabolite level in colon tissue. In the colon tissue, a total of 1057 metabolites were detected in both the 0.5 mCi group and 1 mCi group, respectively. Among them, 111 metabolites exhibited significant changes in the 0.5 mCi group, whereas 96 metabolites showed significant changes in the 1 mCi group compared to the control group. Moreover, 50 metabolites were found to be shared in both 0.5 mCi group and 1 mCi group relative to the control group. (Fig. 4A and Table S3).

Fig. 4

Internal irradiation changes the metabolites and mRNA profiles of mouse colon tissues. (A) Venn diagram showing common and specific metabolites detected LC-MS in colon tissue of 0.5 mCi, 1.0 mCi irradiation and control group of mice. (B) Kyoto Encyclopedia of Genes and Genomes (KEGG) analysis of the dysregulated metabolites in colon tissue of irradiated mice. 9 metabolic pathways were found to be enriched by differential metabolites in the 0.5 mCi irradiated group. (C) 14 metabolic pathways were found to be enriched by differential metabolites in the 1.0 mCi irradiated group. (D) Heatmap of differentially expressed mRNAs in 0.5 mCi, 1.0 mCi irradiated and control group. (E) Veen diagrams showing common and specific mRNAs detected by RNA-Seq. (F) The volcano plot of mRNAs between 0.5 mCi irradiated group and control group in colon tissue. (G) The volcano plot of mRNAs between 1.0 mCi irradiated group and control group in colon tissue. (H) and (I) GO analysis of the target genes of differentially mRNAs in 0.5 mCi Vs control group and 1.0 mCi Vs control group. (J) Specific data display of common differential mRNA in 0.5 mCi Vs control group and 0.5 mCi Vs control group. (K)-(M). GO analysis of the mRNAs that are differentially expressed in both treatment groups (0.5mCi and 1mCi) as compared to the control group in molecular function, biological process and cellular component

Furthermore, an analysis utilizing Kyoto Encyclopedia of Genes and Genomes (KEGG) database indicated 14 enriched metabolic pathways in the 0.5 mCi group (Fig. 4B) and 9 enriched metabolic pathways in the 1.0 mCi group (Fig. 4C) compared to the control group. The KEGG pathway analysis identified four common metabolic pathways that were significantly enriched across both internal irradiation groups relative to the control group.

We next investigated the RNA profiles of irradiated and non-irradiated colon tissues utilizing microarray analysis. The raw array measures were accessible at SUB11126291 in National Center for Biotechnology Information (NCBI). Messenger RNA profiling revealed 446 unique mRNAs with at least 1.5-fold change (P < 0.05) in the 0.5 mCi group compared with normal tissue, and 74 unique mRNAs were detected in the 1.0 mCi group (Fig. 4E). Notably, 120 mRNAs were found to be dysregulated in both the 0.5 mCi and 1.0 mCi group (Fig. 4E and J). RNA sequence analysis was performed to assess the gene expression profile of internal irradiation colon tissues. And the heatmap illustrated the collective changes in mRNA with at least 2.0-fold change (P < 0.05) following different doses of internal radiation exposure (Fig. 4D). As shown in the Fig. 4F, a total of 566 mRNAs with changes in the expression levels were identified in 0.5 mCi group, among them, 215 mRNAs were upregulated, and 351 mRNAs were downregulated and a total of 194 mRNAs with changes were identified in 1.0 mCi group (Fig. 4G), among them, 40 mRNAs were upregulated, and 154 mRNAs were downregulated. The differentially expressed mRNAs were clustered and further classified into molecular function, biological process and cellular components according to the standard Gene Ontology (GO) terminology (Fig. 4H and I). Furthermore, we conducted a meticulous GO enrichment analysis on the mRNAs that were differentially expressed at least 2.0-fold change (P < 0.05) in both treatment groups (0.5 mCi and 1 mCi) as compared to the control group. This analysis aims to delve into the biological significance behind the common changes of these genes. Within the molecular function category, mRNAs regulated by internal irradiation were associated with DNA binding, RNA polymerase II-specific, among others. About the category of biological process, the most prominent category was the transcription by RNA polymerase II and positive regulation of transcription by RNA polymerase II, which accounted for 18%, followed by negative regulation of transcription by RNA polymerase II (14%), positive regulation of gene expression (12%), and rhythmic process (9%). In terms of cellular components, internal irradiation primarily induced expression changes of genes located in the nucleus region when compared with the control cells (Fig. 4K and M). These findings indicated that internal irradiation significantly affects the mRNA expression of colon tissue.

L-citrulline confers radioprotection in vitro and in vivo

Among the metabolites present in fecal samples and colon tissue, we identified a total of seven shared metabolites (Fig. 5A and B), which include saccharopine, L-citrulline, aconitic acid, L-cysteine-glycine, dodecanol, lignoceric acid and resveratrol. Previous studies have showed that L-citrulline is a highly promising biomarker candidate accessing enterocyte integrity in various small intestinal disorder [21, 22], promoting further investigation of L-citrulline in this study.

Fig. 5

L-citrulline confers protection of intestinal epithelial cells against ionizing radiation. (A) Veen diagram showing unique and shared metabolites in feces and colon tissue. (B) Table showing common metabolites in feces and colon tissue. (C) Correlation network of microbes and metabolites from arginine biosynthesis, which included L-citrulline. (D) The STA Sankey Network for arginine biosynthesis. Aserisks (*) indicate statistically significant correlations with metabolites. The pink/blue color of nodes indicate up/downregulation. The red/green bands indicate the positive/negative correlations with metabolites. (E)-(H) The effects of L-citrulline treatment on cell viability and lactate dehydrogenase (LDH) release of intestinal cells (HIEC and IEC-6). *P < 0.05, **P < 0.01, compared with the control group. (I) A schematic timeline of treatment in intestinal cells, HIEC and IEC-6 cells were pretreated with L-citrulline, then irradiated, followed by ROS detecting, apoptosis detecting and clonogenic survival assay. (J)-(K) Effect of L-citrulline on ROS production after different dose of irradiation as determined by DCFH-DA staining in HIEC and IEC-6. (L)-(O) Effect of L-citrulline on radiation sensitivity as determined by a colony formation assay followed different dose of radiation. (P)-(Q) Effect of L-citrulline on cell apoptosis determined by AV/PI staining in HIEC and IEC-6 cells. (R) Schematic showing the protective effect of L-citrulline against radiation-induced intestinal injury. (S) The body weight of mice after different treatment. (T) Representative H&E-stained sections of the small intestines in different group. Scale bar = 200 μm. (U) The height of small intestine villus in different group. (V) Western blotting analysis of H3Cit17 in IEC-6 cells. (W) Immunofluorescence assay of H3Cit17 distribution in IEC-6 cells. Scale bar = 10 μm. The data are shown as the mean ± SEM for three independent experiments. (X) Mean fluorescence intensity of H3Cit17 in IEC-6 cells. The data are shown as the mean ± SEM for three independent experiments. *P < 0.05; **P < 0.01

As a crucial metabolite in arginine biosynthesis, we complied and integrated the differential metabolites associated with arginine biosynthesis and their related bacteria taxa within a network to elucidate the interplay between the microbiome and metabolome. This network analysis revealed a positive correlation between L-citrulline and Lactobacillus, Desulfovibrio and Roseburia, and which exhibiting negative correlation with 16 genera. Furthermore, we employed MetOrigin to explore the biological and statistical correlations between the microbiota and the metabolites using STA-Sankey networks. In the STA-Sankey network, Firmicutes and Bacteroidota were identified as major phyla closely associated with L-citrulline (Fig. 5C and D). In summary, our finding suggest that L-citrulline may function as a co-metabolite involved in the interactions between the host and the microbiota.

To explore the radioprotective effect of L-citrulline in vitro, the HIEC cell line and IEC-6 cell line were used to study the underlying mechanisms which contribute to L-citrulline mediated radioprotection. We first performed CCK-8 assays and LDH release assays to evaluate the effect of L-citrulline on cell viability and cytotoxicity. Treatment with L-citrulline at concentration exceeding 50 μM showed a stimulatory effect on cell proliferation while mitigating cytotoxicity in HIEC and IEC-6 cells (Fig. 5E and H).

To further determine the further principle underlying the protective effects of L-citrulline against radiation-induced injury (Fig. 5I). We observed that treatment with L-citrulline effectively inhibited the IR-induced ROS generation in HIEC and IEC-6 cells (Fig. 5J and K). We further performed clonogenic survival assays to assess the influence of L-citrulline on the colony-forming efficiency of HIEC cells and IEC-6 cells following irradiation (Fig. 5L and O). The results demonstrated that cells pretreated with L-citrulline exhibited a higher clonogenic survival rate compared to the control group. Furthermore, L-citrulline treatment significantly decreased cell apoptosis following irradiation (Fig. 5P and Q). Taken together, our data demonstrated that L-citrulline confers radioprotective effects on intestinal epithelial cells. However, we found that L-citrulline did not provide protection against damage induced by other chemotherapeutic agents (Figure S2), indicating its specific efficacy in mitigating radiation-indued injury.

We subsequently explored the radioprotective properties of L-citrulline in vivo (Fig. 5R). Mice subjected local intestinal irradiation exhibited weight loss after the exposure, however, those treated with 5 g/kg L-citrulline in conjunction with local intestinal irradiation demonstrated an increase in body weight four days post-irradiation (Fig. 5S). Then we observed the morphological changes of mice intestine subjected to various treatment at different time points after irradiation. The results from H&E staining showed that L-citrulline of high concentration (5 g/Kg) effectively maintained the crypt-villus structure in the small intestine and significantly increased villus height at three days post irradiation. Conversely, lower concentrations of citrulline failed to mitigate intestinal damage, likely due to insufficient levels to reach the protective threshold against radiation-induced injury (Fig. 5T and U). It has been demonstrated that low doses of valeric acid fail to provide radioprotective effects, whereas high concentrations of valeric acid possess significant radioprotective properties, with such effects only manifesting when the concentration of the metabolite reaches a certain level [23]. The above results indicate that L-citrulline confers a protective effect against radiation-induced intestinal injury, as evidenced by both in vivo and in vitro investigations.

To investigate the mechanism of L-citrulline in radioprotection, we found that pretreatment with L-citrulline promoted the citrullination of histone H3 at the 17th site (H3Cit17) (Fig. 5V and X, Figure S3). Histone citrullination is a posttranscriptional modification catalyzed by peptidyl arginine deiminase (PAD) enzymes, whereby the amino acid peptidyl-arginine is replaced by peptidyl-citrulline. Our findings indicate a significant increase in H3Cit17 levels. Although no specific enzymes that facilitating histone citrullination by regulating citrulline have been discovered to date, existing literature suggests that increasing the substrate can facilitate post-translational modification [24]. Further research is needed to understand the role of citrulline in promoting histone citrullination. In conclusion, L-citrulline plays its role in radioprotection by promoting citrullination of histone, potentially serving as a novel mechanism of radioprotection.

L-citrulline promotes gene expression through H3Cit17

To elucidate the radioprotective mechanisms of H3Cit17, we utilized a combination of ChIP-Seq and RNA-Seq to identify alterations in the H3Cit17-binding sequence and expression profile of IEC-6 cells following L-citrulline treatment (Fig. 6A). In RNA-Seq, PCA analysis showed a significant degree of separation between the two groups of samples, reflecting substantial differences in RNA between the two groups of samples (Fig. 6B). Profile of mRNA detected 116 mRNAs with significant differential expression levels with over 1.5-fold change (P < 0.05) in IEC-6 cells, comprising 77 upregulated and 39 downregulated mRNAs (Fig. 6C and D). KEGG pathway analysis revealed that mRNAs specifically responsive to L-citrulline were enriched in biological pathways associated with inflammation, including tumor necrosis factor (TNF) signaling pathway and interleukin-17 (IL-17) signaling pathway (Fig. 6E). To investigate how modified H3Cit17 regulates gene expression, we preformed ChIP-Seq analysis. The results showed that there were 20,396 different peaks in IEC-6 cells treated with L-citrulline (Fig. 6F). We observed an increased abundance of H3Cit17 near the transcription start site (TSS) upon L-citrulline treatment in IEC-6 cells (Fig. 6G). A genome-wide analysis of H3Cit17 peak localization indicated that among the upregulated peaks, 4.01% were located in promoters, 1.55% were located in coding exons and 43.25% were located in intron. In the downregulated peaks, 4.27% were located in promoter, 1.96% were located in exon and 28.71% were located in intron (Fig. 6H). To identify the H3Cit17-binding motifs, we used HOMER software, which scores a list of motifs within ChIP-Seq peaks by computing the enrichment of motifs with background sequences. The top ten motifs identified through HOMER-based de novo motif enrichment analysis revealed strong enrichment for YML081W, Sf3b4 and MYB88 (Fig. 6I). Combined data of ChIP-Seq and RNA-Seq indicated that 31 mRNAs were associated with changes in H3Cit17 binding after L-citrulline treatment (Fig. 6J), indicating that the expression levels of these mRNAs may be modulated by H3Cit17, with details information on the 31 mRNAs were shown in the Table S4. The several mRNA levels were increased in IEC-6 cells after L-citrulline treatment (Fig. 6K, Figure S4 A-D). Among them, Cxcl3 has been studied to be associated with the development of intestinal disease [25, 26]. Cxcl3 is a 10-kDa secreted protein that interacts with its receptor, C-X-C chemokine receptor type 2 (CXCR2). And the overexpression of Cxcl3 has been shown to affect the expression of genes associated with extracellular signal-regulated kinase (ERK) signaling pathway, including ERK1/2, Bcl-2 and Bax [27]. Next, to further explore the transcription factors that regulating the transcription of Cxcl3, bioinformatic tool Find Individual Motif Occurrences (FIMO) was used to predict potential transcriptional factors within the Cxcl3 promoter [28]. The results revealed that several putative binding sites for the transcriptional factors were present, including HNRNPK, NRF, GCM1-2 and PLAGL1. The presence of the transcription factors binding sequences in the immunoprecipitated chromatin was analyzed by PCR using three pairs of primers to targeting the H3Cit17 binding regions (Fig. 6L). The results of the ChIP assay showed H3Cit17 specifically binds to the − 1417 to -1196 bp region but not other sites (Fig. 6M and Figure S4 E-G). Additionally, we found that recombinant Cxcl3 protein significantly enhanced the proliferation of intestinal epithelial cells while concurrently decreasing LDH release (Fig. 6N), and the expression of Cxcl3 was found to be upregulated by L-citrulline treatment (Fig. 6O). Furthermore, Cxcl3 was found to reduce the radiosensitivity of intestinal epithelium cells (ER = 0.88) (Fig. 6P and Q). These results suggest that L-citrulline may contribute to radioprotection through the H3Cit17-Cxcl3 signaling pathway.

Fig. 6

L-citrulline promotes CXCL3 expression through H3Cit to exert the radioprotective effect. (A) The Schematic graph of ChIP-Seq and RNA-Seq. (B) The PCA analysis of IEC-6 cells treated with L-citrulline compared with control group. (C) The volcano plot of mRNAs between L-citrulline treated group and control group in IEC-6 cells, fold-change > 1.5, P < 0.05. (D) The heatmap of mRNAs between L-citrulline treated group and control group. (E) KEGG analysis of the differential mRNAs in L-citrulline treated Vs control group. (F) Specific peaks of genomic binding were identified by H3Cit17-ChIP-Seq in IEC-6 cells. (G) Distribution of H3Cit17 ChIP-Seq reads from control (blue) and L-citrulline-treated (pink) IEC-6 cells with ± 10 kb of TSS. (H) Localization of H3Cit17-bound peaks from ChIP-Seq analysis in IEC-6 cells treated with L-citrulline. (I) HOMER de novo motif enrichment analyses of H3Cit17 binding peaks. The top 10 significantly enriched binding motifs and their matched TF family were presented. (J) Combining the data from ChIP-Seq and RNA-Seq, 31 mRNA were differentially expressed with H3Cit17 enrichment. (K) The expression changes of H3Cit17-target genes (Aebp1, Cxcl3, Herc6, Slc16a6) in IEC-6 cells treated with L-citrulline. (L) A diagram indicates the relative positions of transcription factors of Cxcl3 gene, three pair of primers were used. (M) ChIP analysis of transcription factors binding site (primer #2) in Cxcl3 gene. (N) The effects of CXCL3 treatment on cell viability with or without radiation. (O) Western blotting analysis of Cxcl3 in IEC-6 cells. (P) Effect of Cxcl3 on radiation sensitivity as determined by a colony formation assay followed different dose of radiation. (Q) The clonogenic survival curve of IEC-6 cells was measured. The data are shown as the mean ± SEM for three independent experiments. *P < 0.05; **P < 0.01

Discussion

The internal irradiation caused by radionuclides remains a serious concern. Nonetheless, the molecular mechanisms and protective strategies for it are currently not well understood. Previous studies have focused on the toxicological effects of radionuclides. For example, the radiation-induced toxicity of ²¹¹At has been evaluated using detailed hematological, biochemical, and histological analyses in a murine model with intravenous tail injection of NaAt (²¹¹At) [29]. Another study observed alterations in metabolites in body fluids through intravenous tail injection of ¹³⁷Cs and ⁹⁰Sr [30]. Additionally, there have been studies simulating lung damage caused by the inhalation of radionuclides (such as ⁵⁶Mn) [31, 32]. Regarding intestinal injury, previous studies indicated that following the ingestion of ²¹⁰Po, the intestinal lumen exhibits a phenomenon of radionuclide adhesion, rather than the expected mucosal adsorption [11]. However, due to the limitations of experimental conditions and technology in the early stages, no studies have described the multi-omics landscape of intestinal injury resulting from radionuclide ingestion. Accordingly, we established a specific internal irradiation intestinal injury model via gavage with Y90 nuclide and used multi-omics sequencing to analyze the alterations in related microbiota, metabolites, and mRNA expression profiles after internal irradiation, and explored the protective measures against radiation-induced intestinal injury.

Radiation exposure can lead to alterations in the microbiota, with certain microbial communities contributing to radiation protection during radiation-induced injuries [33,34,35,36,37]. Relevant studies have shown that in radiation survivors, the abundance of Lachnospiraceae and Enterococcaceae significantly increases [13]. Consistent with these findings, in our two internal irradiation groups, we also observed an increase in the abundance of Lachnospiraceae. Lachnospiraceae not only enhances hematopoietic function but also alleviates gastrointestinal damage, thereby playing a crucial role in the recovery from intestinal injury [13, 38]. Furthermore, additional studies have indicated that radiation results an increase in the abundance of Lactobacillaceae, Staphylococcaceae, Bacteroidaceae, and Oscillospiraceae, while causing a decrease in the abundance of Clostridiaceae, Lachnospiraceae, and Ruminococcaceae [39]. Our study also observed similar phenomena, with an increase in the abundance of Oscillospiraceae and a decrease in the abundance of Clostridiaceae. Notably, Oscillospiraceae are recognized as beneficial bacteria that can protect the intestine from damage [40]. Consequently, interventions like probiotic supplementation or fecal microbiota transplantation (FMT) can significantly mitigate intestinal damage [41, 42]. Moreover, radiation can also induce changes in the microbial communities of other organs in the body. For instance, in radiation-induced skin injury, the Firmicutes phylum is predominant [43, 44], whereas in internal radiation-induced intestinal injury, the Bacteroidetes phylum is more commonly observed. Specifically, in radiation-induced skin injury, the proportion of the Firmicutes phylum increases, while the proportions of Proteobacteria, Bacteroidetes, and Actinobacteria decrease [43]. Our study also confirmed this observation, showing that in internal irradiation-induced intestinal damage, the proportion of Firmicutes increased, while the proportion of Bacteroidetes decreased. Contrary to expectations, the proportion of Proteobacteria increased. Research indicates that the ratio of Firmicutes to Bacteroidetes (F/B ratio) is negatively correlated with intestinal integrity [45, 46]. An increase in the F/B ratio following internal irradiation suggests that the intestinal barrier function has been compromised. Changes in the microbiota are influenced by various factors, including the type of radiation, housing conditions, and the site of injury [47, 48]. Interventions targeting specific microbial communities, such as the use of antibiotics, fecal microbiota transplantation, or probiotics, can help protect against radiation-induced damage.

Metabolites produced by microorganisms and host cells contribute positively to radiation protection [49]. External irradiation can induce alterations in the concentrations of short-chain fatty acids (SCFAs), including propionic acid, butyric acid, and valeric acid. Research indicates that exogenous SCFAs can mitigate radiation-induced damage [23, 49, 50]. For instance, propionic acid has been shown to significantly enhance the survival rate of bone marrow stem cells, increase mucus thickness, prolong crypt length, reduce the depletion of radiation-induced granulocyte-macrophage progenitor cells, and decrease intracellular reactive oxygen species (ROS) levels [13, 51]. In the metabolites resulting from the concurrent changes in cells and microbiota triggered by internal irradiation, the levels of long-chain fatty acids, such as lignoceric acid and aconitic acid, are primarily affected. Research has found that bacteria containing high levels of lignoceric acid exhibit antioxidant activity [52]. Lignoceric acid can also serve as a potential radioprotective agent. In addition, external irradiation can also lead to changes in the levels of amino acids and their metabolites. After exposure to radiation, the levels of tryptophan metabolites indole-3-acetaldoxime (I3A) and kynurenic acid (KYNA) are elevated [13]. These metabolites can suppress the response of pro-inflammatory factors, thereby preventing the occurrence of acute radiation syndrome (ARS) [42]. Furthermore, indole-3-propionic acid (IPA) can activate the intestinal PXR/ACBP signaling pathway, providing protection against gastrointestinal toxicity [53]. In our study, we observed that internal irradiation leads to increased levels of citrulline and saccharopine. Research has indicated that citrulline has antioxidant properties and may serve as a potential biomarker for radiation-induced intestinal injury. Our findings further confirm that L-citrulline possesses radioprotective effects both in vivo and in vitro.

Citrulline has emerged as a potential biomarker for evaluating gastrointestinal toxicity in patients receiving chemotherapy and radiotherapy [54]. The changes in urinary citrulline concentrations in response to various radiation modalities are variable. Specifically, the administration of ¹³⁷Cs and ⁹⁰Sr via tail vein induces internal radiation exposure, which is associated with a reduction in urinary citrulline levels. Conversely, exposure to low-dose rate (LDR) X-rays does not elicit any significant alterations in urinary citrulline concentrations, while exposure to high-dose rate (HDR) X-rays is associated with an elevation in urinary citrulline levels [30]. In our study, internal irradiation induced by Y90 gavage led to an increase in citrulline levels in feces and intestinal tissues. The aforementioned differences may arise from the distinct types of radiation and the differences in the organs affected by the radiation [55, 56]. Citrulline, as a key component of the nitric oxide system, possesses antioxidant and vasodilatory properties [57]. Moreover, citrulline can enhance intestinal barrier function by regulating iron metabolism and mitochondrial function, thereby providing protection against intestinal injury [58]. Consistent with this, our finding indicates that L-citrulline can exert antioxidant functions and inhibit cell apoptosis, thus providing protection against intestinal damage caused by internal irradiation. Subsequent investigations have demonstrated that citrulline can facilitate the citrullination of histones, leading to enhanced transcription of CXCL3 and increased expression, which in turn confers a radioprotective effect. CXCL3 is generally regarded as a pro-inflammatory mediator, and its elevated expression is typically associated with inflammation and intestinal damage resulting from internal irradiation [59, 60]. Nevertheless, our findings indicate that CXCL3 possesses a radioprotective role. This phenomenon may involve the interaction of CXCL3 with certain intracellular signaling pathways, the specific mechanisms of which still need to be further elucidated and are worthy of further research. Overall, L-citrulline can directly alleviate radiation-induced intestinal damage and is a potential candidate for radioprotective agents. In addition to citrulline, our study also observed that internal irradiation leads to an increase of other metabolites. Future research should investigate whether these compounds also exhibit radioprotective effects.

Conclusions

In conclusion, our work unraveled previously unexplored metabolic alterations associated with internal radiation exposure that facilitate the repair of intestinal injury. We identified a mechanism involving H3Cit17-dependent Cxcl3 expression in intestinal cells (Fig. 7). Taken together, our findings introduce a novel candidate with radioprotective potential. Considering its promising potential, L-citrulline is expected to emerge as an effective protective agent against radiation induced intestinal injury.

Fig. 7

Schematic representation of the discovery mechanism of the radioprotective ability of L-citrulline. L-citrulline increases the citrullination of histone H3 (R17), which regulates the expression of Cxcl3. Cxcl3 has a radio-protective effect in intestinal cells. L-citrulline plays its role in radioprotection via H3Cit17-Cxcl3 axis