Analysis of changes to lncRNAs and their target mRNAs in murine jejunum after radiation treatment

Abstract LncRNAs have been reported to play an important role in various diseases. However, their role in the radiation‐induced intestinal injury is unknown. The goal of the present study was to analyse the potential mechanistic role of lncRNAs in the radiation‐induced intestinal injury. Mice were divided into two groups: Control (non‐irradiated) and irradiated. Irradiated mice were administered 14 Gy of abdominal irradiation (ABI) and were assessed 3.5 days after irradiation. Changes to the jejuna of ABI mice were analysed using RNA‐Seq for alterations to both lncRNA and mRNA. These results were validated using qRT‐PCR. LncRNAs targets were predicted based on analysis of lncRNAs‐miRNAs‐mRNAs interaction. 29 007 lncRNAs and 17 142 mRNAs were detected in the two groups. At 3.5 days post‐irradiation, 91 lncRNAs and 57 lncRNAs were significantly up‐ and downregulated respectively. Similarly, 752 mRNAs and 400 mRNAs were significantly up‐ and downregulated respectively. qRT‐PCR was used to verify the altered expression of four lncRNAs (ENSMUST00000173070, AK157361, AK083183, AK038898) and four mRNAs (Mboat1, Nek10, Ccl24, Cyp2c55). Gene ontology and KEGG pathway analyses indicated the predicted genes were mainly involved in the VEGF signalling pathway. This study reveals that the expression of lncRNAs was altered in the jejuna of mice post‐irradiation. Moreover, it provides a resource for the study of lncRNAs in the radiation‐induced intestinal injury.

cancer. 8 Other disease-relevant roles (eg, aetiology and development) have been reported for lncRNAs, such as in infectious diseases, 9 ophthalmological 10 as well as neurodegenerative diseases, 11 and various cancers. 8 More than half of the cancer patients are currently treated with radiotherapy, but this approach is inexact and often damages surrounding healthy tissues. In particular, the epithelium of mammalian intestinal mucosa undergoes rapid and constant renewal throughout the life of an organism. Additionally, it acts as a physical barrier between the luminal microbiota and the rest of the body. 12 As its rapid renewal and does not have much natural protection, the intestinal epithelium is sensitive to ionising radiation. As such, the intestine is one of the most sensitive organs to radiation toxicity. 13 Radiotherapy of abdominal and pelvic tumours results in high radiation toxicity that can lead to the radiation-induced intestinal injury.
Radiation-induced tissue injury is a complex, pathophysiological process involving multiple and wide-ranging mechanisms, which are dependent on the radiation dose and time course. These mechanisms include DNA repair, cell death, inflammation, endothelial activation, angiogenesis and matrix remodelling. 14,15 Symptomatically, the radiation-induced intestinal injury mainly manifests as diarrhoea, dehydration, sepsis and intestinal bleeding, with eventual mortality within 10-15 days post-exposure. 16 Due to its severity, there is a tremendous need for therapeutic measures that can prevent or treat the radiation-induced intestinal injury.
It has been reported that some lncRNAs are involved in regulating the intestinal epithelial barrier. For instance, Su et al 17 found that miR-874 suppressed AQP3 expression, resulting in AQP3 down-regulation. This impaired intestinal barrier integrity and may have led to intestinal barrier dysfunction via the opening of the tight junction complex. 18 Interestingly, H19, a maternally expressed imprinted lncRNA, 19 may function as a competing endogenous RNA (ceRNA). This would allow for finer regulation of AQP3 expression by competing for miR-874 and may also improve intestinal barrier dysfunction.
Geng et al 20  Past work has also found lncRNAs play a role in DNA damage induced by ionising radiation. For instance, Michelini et al 22  PRE1 has also previously been shown to regulate the expression of CCND1. 24 Moreover, Betts also showed that CUPID1 and CUPID2 regulated the decision to engage either the non-homologous end joining or homologous recombination (HR) pathway.
Past work has found many lncRNAs that are involved in intestinal epithelial barrier and DNA damage induced by ionising radiation; however, there remain limited studies examining the role of lncRNAs in the radiation-induced intestinal injury. Currently, there is no accepted approach to either prevent or treat the radiation-induced intestinal injury. Given this, we sought to determine which lncRNAs were involved in the radiation-induced intestinal injury and provide a better understanding of this type of injury to guide better clinical treatment of the radiation-induced intestinal injury. In this study, we sequenced both the lncRNAs and mRNAs in murine jejuna, both at baseline and 3.5 days post-irradiation. We selected differently expressed lncRNAs and mRNAs, as lncRNAs can act as an endogenous "sponge" that regulates the target gene of miRNAs through competition with miRNAs, we predicted lncRNAs-targeted miRNAs and miRNAs-targeted mRNAs, and compared these predictions with our mRNAs sequencing results. Finally, we performed GO and KEGG signalling pathway analysis and illustrated lncRNA-miRNA-mRNA network; this was performed in order to find out which specific lncRNAs might be involved in the radiation-induced intestinal injury.
All mice were housed in a temperature-controlled, pathogen-free environment with a 12-hour light/dark cycle and allowed ad libitum access to water and standard chow. Mice had been divided into two groups: Control (non-irradiated) and irradiated. The control group included three mice that had not been irradiated, while the irradiated group contained three mice that received 14 Gy ABI. Abdominal irradiation (ABI) was performed on mice using a Cs137 γ-ray irradiator (Atomic Energy of Canada, Chalk River, ON, Canada). Lead shielding was used to protect other body parts from irradiation. Mice were exposed to 14 Gy at 1 Gy/min at room temperature. On day 3.5 post-irradiation, mice were killed and the jejuna were frozen in −80°C freezer. All experimental procedures and protocols were conducted according to the guidelines of our institutional animal care and use committee.

| High-throughput sequencing
High-throughput, whole transcriptome sequencing and subsequent bioinformatics analysis were all performed with Cloud-Seq Biotech (Shanghai, China). The following steps were used: Total RNA was used and rRNAs were removed using Ribo-Zero rRNA Removal Kits

| lncRNAs sequencing analysis
After image and base recognition, the original reads were harvested from an Illumina HiSeq sequencer. 3′ adaptor-trimming and low-quality removal was performed with cutadapt software, 25 after which the resulting high-quality clean reads were used for lncRNA analysis.
Clean reads were aligned to the mouse reference genome (UCSC MM10) using hisat2 software. 26

| lncRNAs identification and their differential expression
Cuffdiff software 27 was used to calculate differentially expressed lncRNAs. lncRNAs that exhibited fold changes ≥2.0 with P < 0.05 and FPKM value ≥0.1 in the least in one sample from a group were classified as having a significant and differentially expressed lncRNA.

| Experimental validation of lncRNAs
Quantitative real-time polymerase chain reaction (qRT-PCR) was used to validate lncRNA expression. Two up-regulated, two down-regulated lncRNAs and mRNAs were selected for validation. The housekeeping gene GAPDH was used as a reference for normalisation. All primers used are presented in Table 1 groups were presented for differential expression. KEGG pathway analysis is a process that maps molecular data sets in genomics, transcriptomics, proteomics and metabolomics onto the KEGG pathway map. This mapping allows for the interpretation of biological function of these molecules. The KEGG pathway analysis was performed with differentially expressed lncRNAs and targeted mRNAs; the analysis speculated as to the pathways involved in these lncRNAs and targeted mRNAs. P-value <0.05 was used as our threshold for statistically significant enrichment.
For mRNAs, 282 were only detected in the control group, 346 were only detected 3.5 days post-irradiation and 16 514 were detected in both groups (Figure 2A). Of the 17 142 total mRNAs, 752 were significantly up-regulated and 400 were significantly down-regulated 3.5 days post-irradiation when compared with controls (P < 0.05, fold change ≥2) ( Figures 2B, C and 3A).
Using the differently expressed lncRNA data (top 20 for each of up-regulated and down-regulated lncRNAs are listed in Table 2), we next predicted their respective sponge miRNA and target mRNA. We also compared these predictions with our mRNA sequencing results.
As shown in Table 3 and Figure 3B

| Validation of lncRNAs and mRNAs expression using qRT-PCR
We selected two up-regulated, two down-regulated lncRNAs and mRNAs at 3.5 days post-irradiation for validation purposes. Our qRT-PCR results indicated that up-and downregulated lncRNAs and mRNAs were consistent with our sequencing results (Figure 4).

| Functional analysis of miRNA target genes
We next performed a GO analysis to better understand the functional association of target genes with the differentially expressed lncRNAs ( Figure 5). Our GO analysis included three parts: MF, BP and CC. As shown in Figure 5A,

| Prediction of miRNA binding sites and lncRNA-miRNA-mRNA network analysis
We selected up-regulated lncRNAs and their 26 targeted mRNAs as well as down-regulated lncRNAs and their 25 targeted mRNAs to construct our lncRNAs-miRNAs-mRNAs network ( Figure 6). As indicated, the network is complex-one lncRNA can associate with multiple miRNAs and one miRNA can inhibit multiple mRNAs. The miRNA mmu-miR-5110 combined the most lncRNAs and target mRNAs. Critically, some of these targeted mRNAs have been reported to be involved in the radiation-induced intestinal injury or repair. Ephb2 is an intestinal stem cell marker, and it has been studied extensively in colorectal cancer. 29 It has been proposed that EXO1 acts in the excision step during mismatch-repair. After mismatch recognition in prokaryotes and eukaryotes, 30 EXO1 is also involved in the 5′ to 3′ end resection at DSB (double-strand breaks) ends to initiate HR in both yeast and mammalian systems. 31 Moreover, Kobayashi et al 32 found that Pgap1 was highly expressed by the follicle-associated epithelium; given this, it may be associated with intestinal mucosal immunity. A better understanding of lncRNA function, as well as their potential intestinal target genes will need further study.

| DISCUSSION
Here, we provide a basic study of lncRNAs function during the radiation-induced intestinal injury. Using a sequencing approach, we

CONFLI CT OF INTEREST
We declare that we have no financial and personal relationships with other people or organizations that can inappropriately influence our work. There is no professional or other personal interest of any nature or kind in any product, service and/or company that could be construed as influencing the position presented in, or the review of, the manuscript entitled. F I G U R E 6 lncRNA-miRNA-mRNAs regulatory network analysis of lncRNAstargeted mRNAs. The quadrilateral represents lncRNAs, arrowheads represent miRNAs, hexagons represent mRNAs, red represents up-regulated, and green represents down-regulated