Microarray Analysis For Expression Profiles of lncRNAs and circRNAs in Rat Liver after Brain-Dead Donor Liver Transplantation

The mechanisms underlying severe liver injury after brain-dead (BD) donor liver transplantation (BDDLT) remain unclear. In this study, we aimed to explore the roles of lncRNAs and circRNAs in liver injury after BDDLT. Rat liver injury was detected in the sham, BD, control, and BDDLT groups. We examined the expression profiles of lncRNAs and circRNAs in the livers of the BDDLT and control group using microarray analysis. The main functions of the differentially expressed genes were analyzed by gene ontology (GO) and KEGG pathway enrichment analysis. In addition, we used bioinformatic analyses to construct related expression networks. Liver injury was aggravated in the BD and BDDLT groups. We found various mRNAs, lncRNAs, and circRNAs that were differentially expressed in the BDDLT group compared with those in the control group. Coding-noncoding gene co-expression (CNC) network analysis showed that expression of the lncRNA LOC102553657 was associated with that of the apoptosis-related genes including HMOX1 and ATF3. Furthermore, competing endogenous RNAs (ceRNAs) network analysis revealed that the lncRNA LOC103692832 and rno_circRNA_007609 were ceRNAs of rno-miR-135a-5p targeting Atf3, Per2, and Mras. These results suggest that lncRNAs and circRNAs play important roles in the pathogenesis and development of liver injury during BDDLT.


Introduction
In current practice, most liver gra s are obtained from braindead (BD) donors [1].Brain death refers to the irreversible and permanent loss of all function throughout the entire brain, including the brainstem [2].Previous clinical and experimental studies have shown that a series of complex pathophysiological changes occur during brain death, including hemodynamic instability, increased leukocyte in ltration, hormonal imbalance, induction of cytokines and chemokines, and increased expression of heat shock proteins.ese changes reduce the tolerance of liver gra s to the further cold preservation/reperfusion injury and aggravate liver injury a er liver transplantation (LT), thus reducing the gra survival rate [3][4][5][6].Despite an increasing number of studies on brain death, to date, there have been few in-depth studies on its mechanism, especially at the genetic level.Furthermore, no precise biomarker has been identi ed which can predict liver injury following BD donor LT (BDDLT).
Noncoding RNAs are RNA molecules that normally do not encode proteins, but instead functionally regulate protein expression [7,8].Noncoding RNAs mainly include long noncoding RNAs (lncRNAs), circular RNAs (circRNAs), and microRNAs (miRNAs) [9].lncRNAs are transcripts longer than 200 nucleotides which, because of their stability, are ideal biomarkers for disease research.New evidence suggests that lncRNAs play a key role in regulating many cellular biological processes, including cell proliferation, differentiation, apoptosis, migration and invasion, and the cell cycle [10][11][12].
In recent years, their potential roles in liver injury have been recognized.e lncRNA Gm2199 can promote hepatocyte regeneration and provides protection against liver injury, while the lncRNAs HOTAIR, HULC, and uc003wbd are highly expressed in patients with liver cancer or hepatitis B [13][14][15][16].Recent studies have shown that circRNAs play a key role in the development of human diseases.CircRNAs are a type of RNA formed from a single-stranded, covalently closed loop, which serve mainly as a "sponge" for miRNAs [17,18].Additionally, circRNAs can compete with pre-mRNA splicing or affect gene transcription by their association with phosphorylated Pol II [19,20].Recently, an increasing number of studies have focused on the role of circRNAs in liver injury [21,22].
Although Catania et al. investigated changes in mRNA expression during brain death [23].However, to date, no study has explored the mechanisms by which noncoding RNAs participate in liver injury a er BDDLT.erefore, to investigate the role of noncoding RNAs in liver injury a er BDDLT, we measured the differential expression of lncRNAs, circRNAs, and mRNAs in liver tissues a er BDDLT, using microarray analysis.Significant differences in expression levels of lnc RNAs and mRNAs were further verified by quantitative realtime polymerase chain reaction (qRT-RCR) analysis.In addition, we analyzed related pathways and gene ontology (GO), and furthermore performed network analyses of coding-noncoding gene CNC (CNC), competing endogenous RNAs (ceR-NAs), lncRNAs' adjacent coding genes, and TF-lncRNAs enriched with mRNAs using bioinformatic methods.Our study aimed to elucidate the mechanisms of liver injury a er BDDLT to help identify biomarkers of liver injury and new therapeutic targets.

Animals and Establishment of BD, Orthotopic LT
Model.Male Lewis rats weighing 250-280 g were purchased from Beijing Vital River Laboratory Animal Technology Co. (Beijing, China).All of the experimental procedures were approved by the Animal Ethics Committee of Zhengzhou University.
Brain death was induced by increasing intracranial pressure in a modified, slow, and intermittent fashion [24].A 3F Fogarty catheter was connected to a pressure pump and then inserted into the epidural space through a hole in the skull.An additional catheter was inserted into the saphenous artery to record the arterial blood pressure, and a blood pressure above 80 mmHg was considered normal.e liver was procured 6 h a er brain death induction according to the clinical practice [25,26].
Livers from BD and nonBD donors were perfused with the University of Wisconsin (UW) preservation solution.e rat models of LT were established using the two-cuff technique without liver artery reconstruction [27].e anhepatic phase lasted 17-20 min.
Experimental animals were randomized into four groups: sham-operated rats underwent only drilling of the skull and laparotomy, without any other procedure performed; in the BD group, brain death was maintained for 6 h a er induction; in the control group, the sham-operated rats were used as donors for orthotopic LT; in the BDDLT group, gra s for orthotopic LT were obtained 6 h a er brain death.Serum and tissue samples were obtained before liver harvesting and 12 h a er reperfusion.

TUNEL Fluorescence Assay.
From the TUNEL kit, appropriate volumes of reagent 1 (TdT) and reagent 2 (dUTP) were mixed at a ratio of 1 : 9 to cover the tissue.e mixture was added to the tissue with DAPI and then incubated at room temperature in the dark.e results were observed under a fluorescence microscope [24].

RNA Extraction and Quantitative Real-Time PCR
Validation.RNA was extracted from tissue using TRIZOL (Invitrogen, America).Synthesis of cDNA was performed with a reverse transcription kit (TaKaRa, Japan) using the Gene Amp PCR System 9700 (Applied Biosystems, America), real-time PCR was performed using SYBR Green PCR Master Mix (TaKaRa) on the Gene Amp PCR System 9700, and the primers used were as described below.e primers of the randomly selected lncRNAs, mRNAs, and internal parameters of GAPDH are shown in Table 1.

Microarray Assay.
Microarray analysis was performed by KangChen Bio-tech, Shanghai, PR China.Briefly, ree pairs of liver tissues were harvested 12 h a er LT in the BDDLT group and the control group.Differential expression levels of lncRNAs, circRNAs, and mRNAs were measured using microarray analysis techniques [28].Arraystar Rat LncRNA Microarray V2.0 was used.Sample labeling and array hybridization were performed according to the Agilent One-Color Microarray-Based Gene Expression Analysis protocol (Agilent Technology) with minor modifications.Briefly, mRNA was purified from total RNA a er removal of rRNA (mRNA-ONLY™ Eukaryotic mRNA Isolation Kit, Epicentre).en, each sample was amplified and transcribed into fluorescent cRNA along the entire length of the transcripts without 3' bias utilizing a random priming method (Arraystar Flash RNA Labeling Kit, Arraystar).e hybridized arrays were washed, fixed, and scanned using the Agilent DNA Microarray Scanner (part number G2505C).Meanwhile, the Arraystar Rat circRNA Array V1 analysis of the liver samples was conducted.Total RNAs were digested with Rnase R (Epicentre, Inc.) to remove linear RNAs and enrich circular RNAs.en, the enriched circular RNAs were ampli ed and transcribed into uorescent cRNA utilizing a random priming method (Arraystar Super RNA Labeling Kit; Arraystar).e labeled cRNAs were hybridized onto the Arraystar Rat circRNA Array V1 (8 × 15 K, Arraystar).A er having washed the slides, the arrays were scanned by the Agilent Scanner G2505C.Agilent Feature Extraction so ware (version 11.0.1.1)was used to analyze acquired array images.Statistically signi cant di erences in the expression of lncRNAs, circRNAs, and mRNAs were evaluated in the BDDLT group and the control group based on -values and FDR ltering.A fold change ≥ 2.0, < 0.05, and FDR < 0.05 were considered statistically signi cant.We deposited the microarray data and additional information to the GEO database with accession number GSE14520 and GSE119748, respectively.2.6.Gene Ontology (GO) and KEGG Pathway Analysis.Gene ontology (GO) analysis provides a controlled vocabulary to describe gene and gene product attributes in any organism.We use topGO to analyze di erential genes in order to infer their molecular functions (http://www.geneontology.org).e 10 items with the smallest -values were selected for graphical display; in the case that fewer than 10 items met the statistical criteria, all the items were displayed.We performed KEGG pathway analysis to harvest pathway clusters covering our knowledge on the molecular interaction and reaction networks in di erentially regulated gene pro ling.Again, the 10 items with the smallest -values were selected for graphical display or all the items if fewer than 10 met the statistical criteria.We also used di erentially expressed genes for pathway analysis to infer the pathways they participate in.Pathway analysis is a functional analysis mapping genes to KEGG pathways.e -value (EASE-score, Fisher--value or Hypergeometric--value) denotes the signi cance of the Pathway correlated to the conditions.Lower the -value, more signi cant is the Pathway.( e recommend -value cut-o is 0.05.)Signi cant pathways contain some genes that di erently, pathway-gene network was constructed by the relationship between pathway and gene.usually to baseline level or below, and a at-lined EEG (Supplementary Figure 1).We rst measured the serum levels of transaminases (AST and ALT).Compared with those in the control group, AST and ALT levels were signi cantly increased in the BD group and the BDDLT group.Based on the ELISA results, the BD and BDDLT groups showed greater aggravated oxidative stress and in ammatory injury than the control group, characterized by signi cantly increased levels of MDA, MPO, and 3-notrotyrosine and decreased SOD levels in the BDDLT group (Figure 1(a)).In addition, liver tissue cell apoptosis was detected by TUNEL staining; compared with the control group, the BDDLT group had increased cell apoptosis (Figure 1(b)).

Microarray Analysis Reveals Di erentially Expressed
lncRNA and circRNA Pro les.Microarray probes successfully detected thousands of transcripts in the liver tissue in the BDDLT group and in the control group.Compared with those in the control group, 228 di erentially expressed lncRNAs and 330 di erentially expressed mRNAs were detected in three liver tissues in the BDDLT group.Among these di erentially expressed transcripts, there were 181 upregulated lncRNAs, among which LOC102553824 showed the largest di erence (fold change: 7.86); there were 47 down-regulated lncRNAs, among which AABR06018038.2 showed the largest di erence (fold change: 4.32).ere were 164 and 166 up-and down-regulated mRNA transcripts, respectively (fold change ≥ 2.0, < 0.05, and FDR < 0.05).In addition, a total of 5,501 up-regulated and 5,329 downregulated circRNAs were detected in the liver tissue chips from both the BDDLT and control groups.Coding gene expression pro ling revealed 19 mRNAs with a more than 5-fold di erence, among which 12 were up-regulated, with the largest fold change being that for Acpp (fold change: 12.22), and 7 were down-regulated, with the largest fold change being that for Serpina7 (fold change: 7.71).Clustering analysis showed that the expression patterns of lncRNAs, mRNAs, and circRNAs signi cantly di ered between the two groups.
us, expression levels of the identi ed lncRNAs, cirRNAs, and mRNAs in the liver tissue a er BDDLT were signi cantly di erent from those in the control group (Figure 2(a)).
ese lncRNAs are widely distributed on all chromosomes including the X chromosome and the Y chromosome.Circos images show the distribution of lncRNA in rat chromosomes.e outermost layer of Circos shows the distribution of rat chromosomes.
e black and white bands represent the cytobands of the chromosomes.e two circles in the middle represent the up-down distribution of lncR-NAs.e more intense the red color, the higher the up-down multiple, and the more intense the blue color, the higher the down-down multiple.e intermediate outer ring represents the di erence in all detected lncRNAs in the chip, and the intermediate inner ring represents the di erence in different lncRNAs (fold change is greater than 2.0, < 0.05) (Figure 2(b)).
e abnormally expressed lncRNAs were divided into six categories according to their relationships with protein-encoding genes: exon-overlapping (10.9%), transcripts according to their expression levels.e criteria for signi cant e ects were set to PCC ≥ 0.90, < 0.01, and FDR < 0.01.e normalized signal intensities of speci c mRNA and lncRNA expression levels were used to construct coexpression network.According to the expression data of coding and noncoding RNA, we calculated the co-expression relationship (Pearson correlation coe cient, PCC) between coding and noncoding RNA to construct the network.By analyzing the functions of these known RNA molecules, we can associate unknown RNA molecules with speci c signaling pathways or diseases to predict their functions and mechanisms.e network was visualized using Cytoscape so ware (version 2.8.3) [29].By analyzing the functions of these known RNA molecules, we can associate unknown RNA molecules with speci c signaling pathways or diseases to predict their functions and mechanisms.
As is known to us, the ceRNA regulatory mechanism is very important between the mRNA and ncRNAs, including miRNA, lncRNA, and circRNA.Possible miRNA response elements were searched for based on the lncRNA, circRNA, and mRNA sequences.We predicted target mRNAs and lncRNAs (circRNAs) of these miRNAs based on TargetScan 7.2 and compared these target genes to the di erentially expressed genes that were identi ed in our array results.en the overlapping of the same miRNA seed sequence binding site both on the lncRNAs/circRNAs, and the mRNA predicted lncRNA/circRNA-miRNA-mRNA interaction.
2.8.Cis-and Trans-Regulation Prediction.lncRNAs whose expression showed signi cant correlations with the cisand trans-regulation of mRNAs were processed.For cisregulation, previous study de ned that a cis-regulator is the one that exerts its function on a neighbouring gene located at the same chromosome, so we search for the adjacent genes less than 300 kb upstream or downstream of lncRNAs may serve as potential targets for cis-regulation of lncRNA [28].For trans-regulation, we were more interested in the way lncRNAs exert their e ects via transcription factors (TFs).e TFs were predicted by the TRANSFAC8.3database based on promoter regions of lncRNA.LncRNA binding transcription factors were predicted by scoring and E value <0.01.erefore, we enriched those mRNAs co-expressed with lncRNAs that signi cantly overlapped with the target genes of a given TF and constructed an lncRNA-TF-mRNA network.e enrichment and connectivity was based on Position Frequency Matrix (PFM) performed as described previously [30,31].
2.9.Statistical Analysis.All statistical analyses were performed in SPSS 21.0 and GraphPad Prism so ware.Data are expressed as mean ± standard deviation (means ± SD).Statistical signi cance was de ned as value less than 0.05 using twotailed unpaired Student's t-test.

Brain Death Promotes Early Liver Gra Injury in
Rats.Brain death is characterized by a rapid increase in blood pressure (autonomic storm), followed by a decrease, BioMed Research International overlapping of relationships and the numbers indicate the lncRNA counts (Figure 2(d)).

GO and KEGG Pathway Analysis
. GO enrichment analysis of signi cantly di erentially expressed mRNAs can reveal the role of di erentially regulated lncRNAs.Up-regulated mRNAs have been found to be primarily involved in single-organism metabolic processes and activities of the germ cell nucleus and hydrolase.Furthermore, down-regulated transcripts intron-overlapping (8.7%), natural antisense (3.8%), and intron antisense (9.3%), and there were a number of crossover events among these four categories, with 60.7% being intergenic and 6.6% being bidirectional.Furthermore, intergenic lncRNAs were the largest category among all of the di erentially expressed lncRNAs (60.7%), with both the up-regulated intergenic lncRNAs (45.9%) and down-regulated intergenic lncRNAs (14.8%) accounting for large proportions (Figure 2(c)).
e Venn diagram presents the LncRNA by bioinformatics speculation were selected for qRT-PCR analysis to verify the microarray results, using 10 pairs of liver tissues obtained a er LT in the BDDLT group and the control group.e analysis revealed that expression of LOC102555648, LOC103692056, LOC103692832, Ablim3-XR_596973, Ablim3-XR_596974, and LOC102554711 decreased in the liver tissue of the BDDLT group, consistent with the microarray results (Figure 4(a)).Similarly, consistent with the microarray results, eight mRNAs showed altered expression compared with those in the control group, with Per2, Inhba, and Rb1 up-regulated, but Angptl4, Atf3, Hmox1, Ndrg1, and Trib3 down-regulated (Figure 4(b)).

CNC and Functional
Prediction of lncRNAs/mRNAs.We selected seven coding genes with signi cant di erential expression to construct the CNC network based on the mainly participate in drug metabolism, extracellular space, and immunoglobulin receptor binding (Figure 3(a)).Our data revealed ten pathways associated with the up-and down-regulation of mRNAs.KEGG enrichment analysis of the up-regulated protein-coding genes showed that the signaling pathway responsible for fatty acid elongation was the most signi cantly enriched.In contrast, KEGG analysis of the down-regulated transcripts revealed that the PI3K-Akt signaling pathway was the most signi cantly enriched pathway (Figure 3(b)).According to the relationship between enrichment pathway and mRNA, the corresponding pathwaygene network map was constructed (Figure 3(c)).

Veri cation of Di erentially Expressed lncRNAs and mRNAs.
In our results, mRNAs reported to be involved in hepatic ischemia-reperfusion injury, and their corresponding   mRNA).Based on the coexpression, we further explored how the abnormally regulated lncRNAs could achieve cis-or transregulation of mRNAs.Ten di erentially expressed lncRNAs were selected to conduct searches for their adjacent coding genes.e coexpressed protein-coding genes were de ned as cis-regulated genes with one di erentially expressed lncRNA within 300 kb on the same chromosome.e number of adjacent coding genes di ers among lncRNAs.For example, LOC103692721 has up to three adjacent coding genes, whereas other lncRNAs have only one target gene.
e adjacent gene pattern showed that the down-regulated IGH-6 and ABR06046430.3were cis-regulated by lncRNA LOC103692721 and involved in liver injury a er LT from BD donors.e upregulated lncRNAs-TFs network consisted of 15 TFs and 95 correlated lncRNAs, and they were connected by 172 edges.e down-regulated lncRNAs-TFs network consisted of 13 TFs and 26 correlated lncRNAs, and they were connected by 43 edges (Figures 6(a Assuming that lncRNAs have transregulatory functions, we analyzed the mRNAs coexpressed with these lncRNAs and mRNAs regulated by TFs.We found that the up-regulated-lncRNA-TF-gene network consisted of 83 lncRNAs, 15 TFs, and 6 correlated genes.e down-regulated-lncRNA-TF-gene network consisted of 22 lncRNAs, 15 TFs, and 6 correlated genes.

Construction of the ceRNA Network.
e ceRNA network was constructed based on microarray results from liver tissues a er LT in both the BDDLT group and the control group.Six lncRNAs and 11 circRNAs with signi cant di erential expression were selected.Some of these shared miRNA response elements (MREs); for example, lncRNAs LOC103692832 and rno_circRNA_007609 are ceRNAs of rno-miR-135a-5p targeting Atf3, Per2, and Mras, whereas lncRNAs Ablim3 and rno_circRNA_013693 are ceRNAs of rno-miR-322-5p targeting Per2, Mapkapk3, and Trim59 (Figure 8).

Discussion
Our study corroborates previous research demonstrating that the complex pathophysiological changes occurring during brain death have a signi cant impact on gra function and survival [32,33].e "autonomous storm" during brain death is considered the main cause of hemodynamic instability, which can lead to insu cient organ perfusion, activate proinammatory factors, and thus promote liver injury soon a er transplantation [34,35].We found that the levels of transaminases were elevated a er BDDLT, along with activation in ammatory factors and an increased number of apoptotic cells.However, the mechanisms underlying liver injury a er BDDLT remain unclear.correlation degree.
ese mRNAs are involved in many biological processes such as the cell cycle, apoptosis, epithelialmesenchymal transition (EMT), and angiogenesis.e network revealed that the up-regulated lncRNA AABR06081886.1 was negatively correlated with Atf3, Hmox1, Angptl4, and Trib3, whereas the down-regulated lncRNA LOC100911923 was positively correlated with these four protein-coding genes, which are mainly involved in cell apoptosis (Figure 5).Relative expression level analyzed for the rst time the changes in expression of lncR-NAs, circRNAs, and mRNAs in liver tissues a er BDDLT.We obtained the expression pro les of lncRNAs and circRNAs by microarray analysis.A large number of di erentially expressed lncRNAs and circRNAs were found in liver tissues a er BDDLT, with about 60% of the down-regulated lncRNAs being intergenic.Six down-regulated lncR-NAs and eight di erentially expressed mRNAs were further veri ed by qRT-PCR.Overall, the qRT-PCR veri cation results were consistent with the microarray results.e differentially expressed genes were subsequently classi ed according to the data using thermograms and cluster maps.We also found that the di erentially expressed lncRNAs were distributed across all the chromosomes, suggesting that every chromosome in the liver tissue, including the X and Y e traditional view of gene regulation focuses on protein-coding genes; however, this view has changed since the discovery of many noncoding genes such as lncRNAs, circRNAs, and microRNAs [9].e roles that lncRNAs can play in the diagnosis and treatment of liver diseases have been recognized [36].Abnormal lncRNA expression levels have been found in di erent types of liver injury, suggesting that lncRNAs are involved in the pathogenesis of liver injury [37,38].MEG3 and GAS5 can inhibit liver brosis, while TUG1 can protect liver gra s during cold preservation [39][40][41].A potential relationship between liver injury and circRNA has been previously established, suggesting that circRNA inhibits hepatic steatosis [42].A large number of di erentially expressed circRNAs were also found in hepatic ischemia-reperfusion injury models [21].In this study, we many lncRNAs were signi cantly associated with protein-encoding genes through correlation of expression levels.erefore, we established a CNC network to predict the relationship between lncRNAs and mRNAs.e network revealed that the up-regulated lncRNA AABR06081886.1 was negatively correlated with Atf3, Hmox1, Angptl4, and Trib3, whereas the down-regulated lncRNA LOC100911923 was positively correlated with these four protein-coding genes, which are mainly involved in cell apoptosis.

Prediction of Cis-and Trans-Regulatory Functions of lncRNAs (Adjacent Coding Genes of lncRNAs and lncRNA-TF-
It has been reported that lncRNA can a ect the expression of adjacent or distal genes via cis-regulation and trans-regulation [45][46][47].We studied the relationships between ten signi cantly di erentially expressed lncRNAs and their adjacent genes, in an attempt to provide a new theoretical mechanism for liver injury a er BDDLT.We found that the number of chromosome, was abnormal a er BDDLT.We found that Hmox1, Ndrg1, and Atf3 were signi cantly down-regulated in liver tissues a er BDDLT; thus, these genes may be involved in inducing apoptosis a er BDDLT to cause liver injury. KEGG pathway analysis of di erentially expressed mRNAs revealed that multiple pathways (primarily the fatty acid elongation pathway and the PI3K-Akt signaling pathway) were involved in liver injury a er BDDLT.GO analysis revealed that the main mechanisms of liver injury a er BDDLT included single-organism metabolism, drug metabolism, and immunoglobulin receptor binding.At present, the functions of most lncRNAs are unknown, and constructing a CNC network is one method for predicting lncRNA functions [43,44].We found that the expression levels of and 26 correlated lncRNAs, and they are connected by 43 edges.Furthermore, the lncRNA-TF-mRNA diagram showed that lncRNA LOC102553657 trans-regulated HMOX1 via HNF-3beta TF. us, we speculate that these TFs participate in liver injury a er BDDLT by regulating the transcription of lncRNAs and mRNAs.However, the relationships between lncRNAs and TFs needs further investigation.Pandol et al. introduced a hypothesis called the ceRNA mechanism, which proposed that transcripts such as mRNAs, pseudogenes, and lncRNAs can serve as natural miRNA sponges by competitively binding to miRNA response elements (MREs) to suppress their expression and function [55].Indeed, ceRNAs have important in uences on gene expression at the post-transcriptional level, and several studies have reported that lncRNA is involved in the pathogenesis of liver injury through ceRNA [56][57][58][59].However, the roles of lncRNA-and circRNA-related ceRNAs in liver injury have not been studied.In the current study, we constructed for the rst time the ceRNA network with lncRNA-miRNA-circRNA-mRNA associations based on microarray results.Our results reveal a speci c ceRNA network underlying liver injury a er BDDLT.We found that the lncRNAs LOC103692832 and rno_circRNA_007609 were ceRNAs of rno-miR-135a-5p targeting Atf3, Per2, and Mras, whereas the lncRNAs Ablim3 and rno_cir-cRNA_013693 were ceRNAs of rno-miR-322-5p targeting Per2, Mapkapk3, and Trim59.us, lncRNAs and circRNAs harbor MREs and thus play key roles in liver injury a er BDDLT.adjacent coding genes di ered among lncRNAs; for example, LOC103692721 had up to three adjacent coding genes, whereas other lncRNAs had only one target gene.In addition, the adjacent gene pattern showed that the down-regulated gene IGH-6 and ABR06046430.3were cis-regulated by lncRNA LOC103692721 and involved in liver injury a er BDDLT.
Furthermore, based on our results, we constructed the lncRNA-TF network.Transcriptional regulation by lncRNAs can be predicted from the regulation of TFs, which they express [48][49][50].e combination of TFs and cis-acting elements can also regulate the expression levels of target genes in the promoter region [51,52].It has also been reported that TFs participate in the occurrence and development of liver injury [53,54].erefore, a comprehensive analysis of TFs and di erentially coexpressed genes can improve our understanding of liver injury a er BDDLT.A number of lncRNAs may be involved in speci c pathways for TF regulation.erefore, assuming that lncRNAs regulate transcription, we analyzed mRNAs coexpressed with lncRNA and TF-regulated mRNAs.Using < 0.01 and FDR < 0.01 as the statistical criteria, we found that each lncRNA was associated with between one to over a dozen TFs, and that each lncR-NA-TF pair resulted from the enrichment of several genes, providing key data for subsequent studies.In the liver tissues from the BDDLT group, we found that the up-regulated lncRNAs-TFs network consists of 15 TFs and 95 correlated lncRNAs, and they are connected by 172 edges.e down-regulated lncRNAs-TFs network consists of 13 TFs  regulated lncRNAs and circRNAs, which could be promising markers of the pathogenesis of liver injury a er BDDLT and provide new directions for predicting and diagnosing this condition.Our research paves the way for further

Conclusions
We found that brain death promotes liver injury a er LT.Furthermore, we identi ed a number of speci c abnormally

F 1 :
Biochemical and ELISA measurements of liver damage markers; TUNEL staining of liver tissue.(a) ALT, AST, MDA, 3-nitrotyrosine, and MPO levels were signi cantly increased, and the SOD level was signi cantly decreased in the BD and BDDLT groups.All data are shown as mean ± SD. e levels of statistical signi cance are indicated as * < 0.05, * * < 0.01 statistically signi cant.For statistical analysis, a two-tailed Student's t test was used.(b) TUNEL staining of liver tissue showed more apoptotic cells in the BDDLT group (scale bar: 50 µm).

F 2 :
Heat maps showing expression pro les of lncRNAs, circRNAs, mRNAs, and identi cation of di erentially expressed lncRNAs in the liver tissues of the control and BDDLT group.(a) Heat maps show normalized expression values of signi cantly changed lncRNAs, circRNAs, and mRNAs with a fold change ≥ 2.0, < 0.05, and FDR < 0.05.C: control group tissues.B: BDDLT group tissues.(b) Circos images show the distribution of lncRNA in rat chromosomes, the outermost layer of Circos is the distribution of rat chromosomes, the black and white bands are the cytobands of chromosomes, the middle two circles are the up-down distribution of lncRNA, the intermediate outer ring represents the di erence of all detected lncRNAs in the chip, and the intermediate inner ring represents the di erence of di erent lncRNAs (fold change is greater than 2.0, < 0.05).(c) Types and counts of di erentially regulated lncRNAs detected by microarray (fold change ≥ 2.0, < 0.05, and FDR < 0.05).e lncRNAs are classi ed into 6 types according to the relationship and genomic loci with their associated coding genes.(d) e Venn diagram represents the overlapping of relationships and the numbers indicate the lncRNA counts.

F 3 :
e c t io n a l E x o n s e n s e -o v e r la p p in g I n t e r g e n ic I n t r o n s e n s e -o v e r la p p in g I n t r o n ic a n t is e n s e N a t r u a l a n t is e n s Gene Ontology (GO) and KEGG pathway analysis with top 10 enrichment score.(a) GO analysis of up-and down-regulated mRNAs with top ten enrichment score covering domains of biological processes, cellular components, and molecular functions.(b) KEGG pathway enrichment analysis of up-and down-regulated mRNAs with top ten enrichment score.(c) e link and overlapping of associated molecules among signi cant pathways.
se to st re ss [G O :0 00 69 50 ] R es po se to ex te rn al st im ul us [G O :0 00 96 05 ] lin co m pl ex ,c ir cu la ti ng [G O :0 04 25 71 ] E xt ra cl ee ul ar re gi on [G O :0 00 55 76 ] E xt ra cl ee ul ar re gi on pa rt [G O :0 04 44 21 ] E xt ra cl ee ul ar sp ac e[ G O :0 00 56 15 ] la to r ac ti vi ty [G O :0 06 11 35 ] E nd op ep ti da se in hi bi to r ac ti vi ty [G O :0 04 86 6] G ro w th fa ct or bi ng in g[ G O :0 01 98 38 ] C at al yt ic ac ti vi ty [G O :0 00 38 24 ] Pe pt id as e re gu la to r ac ti vi ty [G O :0 06 11 34 ] Pe pt id as e in hi bi to r ac ti vi ty [G O :0 03 04 14 ] E nz ym e nh ib it or ac ti vi ty [G O :0 00 48 57 ] In su lin -l ik e gr ow th fa ct or 1 bi nd in g[ G O :0 03 19 94 ] Im m un og lo bu lin re ce pt or bi nd in g[ G O :0 signaling pathway Biosynthesis of amino acids HIF-1 signaling pathway Glycine,serine and threonine metabolism FoxO signaling pathway Breast cancer One carbon pool by folate Proximal tubule bicarbonate reclamation Aminoacyl ) and 6(b)).

F 4 :
Validation for the expression of signi cant transcripts by qRT-PCR.e relative expression levels of six lncRNAs (a) and eight mRNAs (b) are shown in the control and BDDLT group liver tissues.All data are shown as the mean ± SD. e levels of statistical signi cance are indicated as * < 0.05, * * < 0.01.For statistical analysis, a two-tailed Student's t test was used.

F 5 :
Coding noncoding gene co-expression networks of 7 signi cant mRNAs with their associated lncRNAs.e network is based on the Pearson correlation coe cient (the absolute value of PCC ≥ 0.90, -value <0.01, and FDR < 0.01).Solid lines indicate positive correlations, and dashed lines indicate negative correlations.e size of the lncRNA dot represents the number of genes co-expressed with the lncRNA.

F 6 :
Cis regulation of lncRNAs to nearby coding genes and the network of enrichment transcription factors and query lncRNAs.(a) lncRNAs and their potential cis regulated nearby genes are shown in the network and the distances between lncRNAs and their cis regulated genes are presented.(b) According to the prediction of transcription factor regulation in the promoter region of lncRNA, the transcription factors of rats in the TRANSFAC8.3database were predicted.e up-regulated lncRNAs-TFs network consists of 15 TFs and 95 correlated lncRNAs, and they are connected by 172 edges.e down-regulated lncRNAs-TFs network consists of 13 TFs and 26 correlated lncRNAs, and they are connected by 43 edges.

F 7 :
e lncRNA-TF-gene trans regulation network.Network diagram based on the binary analysis of TF-lncRNA using the co-expressed mRNA analysis of lncRNA and mRNA regulated by these transcription factors, using the threshold of abs (PCC) ≥ 0.9 and FDR ≤ 0.05.(a) e up-regulated lncRNAs-TFs-genes network consists of 83 lncRNAs, 15 TFs, and 6 correlated genes.(b) e down-regulated lncRNAs-TFs-genes network consists of 22 lncRNAs, 15 TFs, and 6 correlated genes.

F 8 :
Competing endogenous RNA network.e competing endogenous RNA network is based on lncRNA/miRNA, circRNA/miRNA, and miRNA/mRNA interactions.e edges represent sequence matching, and lncRNAs or circRNAs connect expression correlated with mRNAs via miRNAs.