Integrated Analysis Of LncRNA,miRNA And MRNA Profiles Reveals Potential LncRNA Functions During Early HIV Infection

Objective:Early human immunodeficiency virus（Human Immunodeficiency Virus,HIV）infection reflects approximately six months post infection following viral entry during which viral load of plasma increases rapidly to a high level,while many proinflammatory cytokines are produced by innate immune cells and coincide with mounting adaptive immune responses,and then the viral load of plasma decreases to a stable viral load level.The interactions between host and virus during early HIV infection influence the viral transmission rates,the course of disease progression,HIV-related morbidity and mortality.To date,the pathogenesis of early HIV infection remains unclear.Therefore,exploring the pathogenesis and host antiviral response of early HIV infection will help us to provide new ideas for screening disease-related biomarkers and developing effective HIV treatment strategies.Long noncoding RNAs（lncRNAs）,which are longer than 200 nt and do not encode proteins,have attracted much attention in recent years.LncRNAs play vital roles in a number of biological processes,can act as guides,scaffolds,decoys,signal molecules or act as cis or trans regulatory elements to target genomic,or through antisense interference to target genes.Moreover,lncRNAs bind to specific miRNAs and regulate their functions by acting as "miRNA sponges" or are known as competing endogenous RNAs（ceRNAs）.Several HIV-related transcriptome analyses have explored the potential roles of lncRNAs in HIV infection from different aspects,and it has found that lncRNAs can impact HIV virus replication,the establishment and reversal of latent HIV infection and the relationships of disease progression.However,the potential roles of lncRNAs during early HIV infection have not been systematically investigated.In this study,we obtained the lncRNA,miRNA,and mRNA expression profiles from 3 HAART-naive early HIV-infected patients and 3 healthy controls（HCs）using RNA-seq and miRNA-seq,and the differentially expressed lncRNAs,miRNAs and mRNAs were identified.The potential roles of lncRNAs were analysed based on the prediction of lncRNA cis-regulatory target mRNAs and by constructing lncRNA-miRNA-mRNA ceRNA networks.The results suggested that lncRNAs may be involved in HIV-1 replication and immune activation during early HIV infection.The above results will help to further understand the pathogenic mechanism of HIV and propose new potential targets for therapeutic strategies.Methods:1.Study subjectsThe HAART-naive early HIV-infected patients and healthy controls in our study were recruited from a large-scale,prospective high-risk men who had sex with men（MSM）cohort of in Liaoning,China.The stage within 180 days post HIV infection was defined as EHI.Three HAART-naive early HIV-infected patients and 3 healthy controls（HCs）were enrolled for RNA-seq and miRNA-seq.The HCs were recruited from the same HIV-negative MSM cohort.None of the above subjects were infected with herpes simplex virus,hepatitis B virus and hepatitis C virus.The subjects used for differentially expressed genes（mRNA,lncRNA,miRNA）qRT-PCR verification included 8 HAART-naive early HIV-infected patients and 10 healthy controls.2.RNA isolation and quality controlTotal RNA was isolated from PBMCs of 3 HAART-naive early HIV-infected patients and 3 HCs using TRIzol reagent.The concentrations and quality of RNA were measured with a NanoDrop ND-2000 instrument/NanoDrop ND-1000 instrument for subsequent sequencing.3.RNA sequencing and analysisRibo-Zero rRNA removal kits were used to exclude ribosomal RNA molecules（rRNA）form total RNA.RNA was then pretreated with the TruSeq Stranded Total RNA Library Prep Kit to construct a sequencing library.Sequencing for 150 cycles on an Illumina HiSeq 2500 sequencing system,and then paired-end reads were acquired.High-quality clean reads were obtained using Cutadapt software（v1.9.3）and were compared with the human reference genome（UCSC hg19）using hisat2 software（v2.0.4）.Then,the expression profiles of mRNAs and lncRNAs were obtained using cuffdiff software.4.miRNA sequencing and analysisTotal RNA was used to prepare miRNA sequencing library.A HiSeq 4000 sequencing system was used for sequencing.Cutadapt software（v1.9.3）was used to obtain the adaptor-trimmed reads（>=15 nt）.The miRDeep2 software was used to predict novel miRNAs.The miRbase and NovoAlign software were used to align trimmed reads to get the known and newly predicted pre-miRNAs.The primal expression levels of the miRNA were determined via the numbers of tags on each mature miRNA,and the TPM（tag counts per million aligned miRNAs）method was used for the standardization of read counts.5.Analysis of differentially expressed lncRNA cis-regulatory target mRNAsThe LncRNA Disease database was used to identify overlapping mRNAs of non-intergenic differentially expressed lncRNAs.Cis-regulatory target mRNAs that were involved in HIV infection were determined by the HIV Interaction Database,LncRNA Disease database and Genecard database.Moreover,the neighbouring differentially expressed mRNAs were screened within a genomic distance of 300 kb upstream or downstream of each differentially expressed lncRNA transcription start and stop site.6.Construction of ceRNA networksLncLocator and iLoc-LncRNA were used to predict lncRNAs which located in the nucleus in both predictors and were excluded from ceRNA network analysis.DIANA tools-LncBasev.2 and miRDB databases were used to predict the potential differentially expressed target miRNAs of differentially expressed lncRNAs.TargetScan and miRDB databases were used to predict the potential differentially expressed mRNAs targets of differentially expressed miRNAs.According to the above predicted interactions,the ceRNA network was constructed and visualized using Cytoscape software.In addition,miRTarBase database was used to query the experimentally verified miRNA-mRNA relationships.7.GO and KEGG enrichment analysis of differentially expressed mRNA in ceRNA networkThe differentially expressed mRNAs in ceRNA network were analyzed by GO（Gene ontology）and KEGG（Kyoto Encyclopedia of Genes and Genomes）using DAVID 6.8 database（Database for Annotation,Visualization and Integrated Discovery）,p<0.05 was considered to be significantly different.8.Verification of differential expressed lncRNAs,mRNAs and miRNAsTotal RNA was isolated with the miRNeasy Mini Kit.Total RNA was reverse transcribed with the PrimpScript（?）RT reagent Kit for validation of lncRNAs and mRNAs and with the Mir-XTM miRNA First-Strand Synthesis Kit for validation of miRNAs.TB Green（?）Premix Ex TaqTM Ⅱ was used for qRT-PCR.GAPDH and U6 were employed as endogenous controls for lncRNA/mRNA qRT-PCR and miRNA qRT-PCR,respectively.The relative expression of lncRNAs,mRNAs and miRNAs was calculated by the 2-ΔΔCt method.9.Statistical analysisA nonparametric Mann-Whitney U test was performed to investigate differentially expressed genes.Data were analysed,and graphs were created using GraphPad Prism v8.0.Two-tailed p-value of less than 0.05 was considered statistically significant.Results:1.Analysis of differentially expressed lncRNAs,mRNAs and miRNAs between HAART-naive early HIV-infected patients and HCsAccording to the RNA-seq results of PBMC from 3 early HIV-infected patients and 3 healthy controls,a total of 17235 lncRNA transcripts were detected.Among these lncRNAs,a total of 242 lncRNAs with fold change above 2（p<0.05）,including 36 up-regulated lncRNAs and 206 down-regulated lncRNAs.At the same time,we identified 16410 mRNA transcripts.1240 mRNAs were significantly differentially expressed,including 344 up-regulated mRNAs and 896 down-regulated mRNAs.According to the results of miRNA-seq,a total of 1090 mature miRNAs were detected.A total of 21 differentially expressed miRNAs were identified,including 7 up-regulated miRNAs and 14 significantly down-regulated miRNAs.2.Verification of the results of RNA-seq and miRNA-seqWe randomly selected 3 lncRNAs（ENST00000488545,ENST00000571722 and ENST00000602507）,3 mRNAs（ATF3,FOS and CDCA7）and 3 miRNAs（miR-31-5p,miR-484 and miR-3174）from differentially expressed mRNAs,lncRNAs and miRNAs for qRT-PCR verification.The results of qRT-PCR were consistent with RNA-seq and miRNA-seq（p<0.05）.It shows that the above sequencing results are reliable and can be used for subsequent analysis.3.Features of differentially expressed lncRNAsDifferentially expressed lncRNAs were categorized into six groups.The length range of differentially expressed lncRNAs was from 131 to over 6000 nt.The majority of the differentially expressed lncRNAs were distributed in 3 intervals:501-1000,1001-2000 and 2001-3000.Differentially expressed lncRNAs were located on 22 autosomes and the X chromosome.4.lncRNAs impact early HIV infections through cis-regulatory mechanismsWe found that 4 differentially expressed mRNAs which might be associated with HIV infection were regulated by 4 differentially expressed lncRNAs by investigating the overlapping mRNAs of nonintergenic lncRNAs.In addition,three up-regulated intergenic lncRNAs（ENST00000364880,ENST00000571722,ENST00000577988）were predicted to regulate down-regulated mRNA-GRAP by predicting the potential mRNAs targets near the differentially expressed lncRNAs.And the downregulated intergenic lncRNAs ENST00000492960 and TCONS₀0006930 were predicted to target the downregulated mRNAs-ZAP70 and-BTLA,respectively,it is speculated that the above lncRNAs may be related to T cell activation and HIV-1 replication.5.LncRNAs impact early HIV infection through ceRNA regulatory networksTo further elucidate the potential interactions of differentially expressed lncRNAs,miRNAs and mRNAs during early HIV infection,lncRNA-miRNA-mRNA networks were constructed.A total of 160 differentially expressed lncRNAs,21 differentially expressed miRNAs and 175 differentially expressed mRNAs were selected to construct 2 ceRNA networks:one contained 136 downregulated lncRNAs,7 upregulated miRNAs and 97 downregulated mRNAs,and the other contained 24 upregulated lncRNAs,14 downregulated miRNAs and 78 upregulated mRNAs.Moreover,31 experimentally validated miRNA-mRNA interactions were found in the ceRNA networks and 134 differentially expressed lncRNAs might performed functions through above experimentally validated miRNA-mRNA interactions.To explore the potential roles of differentially expressed lncRNAs and their associated ceRNA networks in early HIV infection,175 differentially expressed mRNAs involved in the ceRNA networks were selected for GO and KEGG enrichment analysis.Most of the top ten GO terms are related to transcription.In addition,5 significantly enriched KEGG pathways were identified.Five upregulated mRNAs,HIF1A,TCF7L2,FOS,FOSB and JUN occurred in multiple pathways which play important roles in HIV replication and immune activation.Further analysis found that seventeen differentially expressed lncRNAs were predicted to share the MRE with hsa-miR-548ah-3p and in turn regulate expression of HIF1A and TCF7L2,and a total of 20 differentially expressed lncRNAs were predicted to share MREs with four differentially expressed miRNAs,miR-101-3p,miR-139-5p,miR-584ah-3p and miR-27b-3p to regulate FOS,JUN and FOSB as ceRNAs.The regulation and their roles of the above lncRNAs on their target genes in early HIV infection need to be further verified.Conclusions:In this study,we systematically analyzed the changes of lncRNA,miRNA and mRNA expression profiles in early HIV infection,and we found that some lncRNAs regulated the expression of ZAP70,BTLA and GRAP through cis-regulation mechanism,and some lncRNAs regulated the expression of HIF1A,TCF7L2,FOS,FOSB and JUN through ceRNA network,in turn affecting HIV virus replication and cell activation.These results need to be verified by further experiments.Our findings contribute to further understanding of the pathogenesis of HIV and suggest new potential targets for treatment strategies.