Profiling Analysis Of Long Non-coding RNAs In Early Postnatal Mouse Hearts

Cardiovascular disease (CVD) occupies 50% morbility and the highest lethality among the chronic non-infective diseases. The incidence of CVD is progressively increasing though traditional preverntions and treatments are down. Recently, increasing evidences indicate that CVD is closely related to the abnormal cardiovascular development, therefore,studying cardiovascular development is crucial for revealing novel pathological mechanism of CVD. Postnatal mammalian heart will enter a critical period during which cardiomyocytes undergo transition of growth, mitochondrial dynamics, energy utilication and other processes. These transitions are critical in establishing normal cardiac structure and function. Among these processes, cardiomyocytes undergo a hyperplastic-to-hypertrophic growth transition in early postnatal life, which is critical for self-recovery of heart from cardiac damage within postnatal 7 days. This transition may be controlled by numerous growth factors and signal pathways. However, the precise mechanism is still largely unknown. LncRNAs are non-coding RNAs with more than 200 nucleotides length and exert important biological function such as cell growth, differentiation, apoptosis and autophagy,etc. due to their various regulatory mechanisms. Recently, lncRNAs are widely reported to be involved in cardiac development and diseases, however, the role of lncRNAs in this transition remains unclear. The current study is aimed at exploring the role of lncRNAs in this critical transition.Methods and results:1. Profiling analysis of lncRNAs in early postnatal mouse hearts.1.1 Characterization of lncRNA profile in early postnatal mouse hearts.Three litters of postnatal day (P) 1,7 and 28 C57BL/6J mouse hearts were randomly selected and then total RNA was extracted for further lncRNA microarray analysis. Totally,We identified 18,158 signals of lncRNA from University of California Santa Cruz, Refseq,Ensembl, etc. by using lncRNA microarray, among which lncRNAs with length between 200 and 3000 muceotides constitute the majority.LncRNAs can be divided into different forms according to their positions,among which intergenic lncRNAs are over 10,000.We identified 765 differentially expressed lncRNAs (fold change > 2, FDR < 0.05) between P1 and P7 mouse hearts, 4,855 differentially expressed lncRNAs between P7 and P28 mouse hearts. After ANOVA analysis of raw data, we identified 1,146 differentially expressed IncRNAs between P1, P7 and P28 mouse hearts. Among these 1,146 differentially expressed transcripts, 253 lncRNAs were gradually upregulated, while 487 lncRNAs were gradually downregulated. As compared to the PI mouse heart, 315 lncRNAs were upregulated at P7 and then downregulated at P28; while only 91 lncRNAs were downregulated at P7 and then upregulated at P28.1.2 Bioinformatic analysis of differentially expressed lncRNAs in early postnatal mouse hearts.By utilizing the series test of cluster (STC) analysis of the 1,146 and 1,358 differentially expressed lncRNAs and mRNAs, respectively, we identified 16 expression patterns. Each expression pattern represents a trend and contains lncRNAs or mRNAs with similar trend. We identified 7 significant expression patterns in both lncRNAs and mRNAs.Surprisely, the 7 lncRNA expression patterns also exist in the 7 mRNA expression pattern.In addition, lncRNAs and mRNAs in the most significant pattern are both stable from P1 to P7,but rapidly decreased from P7 to P28. The similar trend indicates the possible interactions between lncRNAs and mRNAs. To further explore the role of lncRNA-mRNA interactions in early cardiac development, we constructed co-expression network of differentially expressed lncRNAs and mRNAs and found the core genes are mostly involved in regulating early cardiac development and cell cycle regulation.879 neighboring genes of the 1,146 lncRNAs were detected in microarray, among which 59 differentially expressed neighboring genes were identified. Among the 59 lncRNA-neighboring gene pairs, 35 pairs are positive correlated and 11 pairs are negative correlated. These correlations indicate lncRNAs may regulate expression of their neighboring genes. Some of these neighboring genes are involved in early postnatal heart development such as Ccbel and Vcan. Additionally, we also performed Gene ontology (GO)and pathway analysis of these differentially expressed neighboring genes based on KEGG.The neighboring genes are largely involved in cell cycle-associated processes such as DNA replication and transcription and cell proliferation-hypertrophy transition-associated pathways such as AMP-activated protein kinase (AMPK). These data suggest that differentially expressed IncRNAs may be involved in early postnatal heart development via regulating expression of neighboring genes.We discussed the upstream mechanisms that regulate the expression of differentially expressed IncRNAs by using transcription factors (TFs). Therefore, we selected 100 IncRNAs and mRNAs with the highest degrees from the co-expression network and analyzed their potential TF binding sites and TFs. Then, we constructed feedforward loops according to these TFs, lncRNAs and mRNAs. Each TF is able to regulate several IncRNAs and mRNAs and each lncRNA and mRNA can also be regulated by several TFs. These data indicate that the complex regulatory relationship between the three molecules constitute molecular network regulating early postnatal heart development.2.Validation, screening and functional analysis of lncRNAs in early postnatal mouse hearts.2.1 LncRNA validation and tissue specificity determination in microarray.10 constantly increased and declined IncRNAs with the highest fold changes from P1 to P7 were selected and validated by real time qRT-PCR. The data acquaired from qRT-PCR were consistent with that in microarray. 3 differentially expressed neighboring genes of the 10 IncRNAs were acquaired from microarray and showed similar trend via qRT-PCR.Next,we determined mRNA levels of the 10 selected lncRNAs in heart, kidney, thoracic aorta,skeletal muscle, liver and small intestine for screening of cardiac-specific lncRNA by using qRT-PCR and fortunately found a cardiac-enriched lncRNA, ENSMUST00000117266.2.2 Co-expression network analysis of ENSMUST00000117266.We selected differentially expressed lncRNAs and mRNAs with Pearson’s correlation coefficient > 0.97 and constructed co-expression subnetwork of ENSMUST00000117266.In the network, totally 81 molecules are included among which 44 are coding genes. These genes are mostly involved in cardiac development. For example, Hmga2 is involved in embryonic cardiogenesis and postnatal heart development. Ccbel is associated with cardiac progenitor differentiation. Cited1 is critical to exbryonic cardiac development and survival.These data indicate ENSMUST00000117266 is likely to participate in molecular regulatory network in early postnatal heart.2.3 ENSMUST00000117266 is sensitive to adverse stimuli including hupoxia,paraquat and myocardial infarction (MI).The mRNA level of ENSMUST00000117266 significantly declined after 6 hours of hypoxic exposure in neonatal mouse cardiomyocytes.Neonatal mice were subcutaneously injected with reactive oxygen series (ROS) inductor paraquat for three continuous days.The mRNA level of ENSMUST00000117266 in paraquat-treated cardiac tissue also decreased. In addition, we utilized a MI model and found ENSMUST00000117266 mRNA expression in heart was significantly increased after 2 days but soon decreased after 3 days of MI.These data suggest ENSMUST00000117266 is sentitive to adverse stimuli in early postnatal heart.2.4 ENSMUST00000117266 is involved in regulating cardiomyocyte proliferative activity.To further explore the role of ENSMUST00000117266 in early postnatal cardiac hyperplastic-to-hypertrophic growth transition, we analyzed cardiomyocyte cell cycle by flow cytometry after 48 hours of ENSMUST00000117266 siRNA transfection.We found the percentage of cardiomyocytes in G0/G1 phase of cell cycle was significantly increased but the percentage of cardiomyocytes in G2/M phase was significantly decreased after ENSMUST00000117266 knockdown. We also found the mRNA level of proliferation-associated gene Ccndl significantly declined after ENSMUST00000117266 knockdown. In addition, after 6 days of rhBmp10 treatment incardionyocytes, ENSMUST00000117266 knockdown by siRNA significantly reversed up-regulation of Ccndl, Myh10 and TGFβ-associated genes Nkx2.5,Mef2c by rhBmp10. These results indicate ENSMUST00000117266 is involved in regulating cardiomyocyte proliferative activity.2.5 ENSMUST00000117266 regulates expression of its neighboring coding gene Sap301.We used qRT-PCR to detect whether ENSMUST00000117266’s location and found ENSMUST00000117266 was equally located at nucleus and cytosol. Then we used siRNA and antisense oligonucleotide (ASO) strategy for48 hours of transfection in primary mouse cardiomyocytes and determined the mRNA levels of neighboring genes of ENSMUST00000117266 Hand1, Larp1 and Sap301. The mRNA expression of Sap301 significantly declined after ENSMUST00000117266 knockdown via siRNA or ASO,indicating ENSMUST00000117266 may regulate Sap301 expression either transcriptionally or post-transcriptionally.Conclusions:(1) We identified a large group of differentially expressed IncRNAs in early postnatal mouse hearts by utilizing IncRNA microarray. STC analysis suggest that differentially expressed IncRNAs can interact with differentially expressed genes. GO and Pathway analysis further showed differentially expressed IncRNAs can affect heart development via regulating corresponding neighboring genes. We also identified potential TFs regulating the expression of these differentially expressed IncRNAs.(2) We firstly found a novel cardiac specific IncRNA ENSMUST00000117266, the co-expression network of which indicates ENSMUST00000117266 is involved in modulating heart development-associated genes. ENSMUST00000117266 is sensitive to adverse stimuli such as hopoxia, paraquat and MI. We also confirmed that ENSMUST00000117266 was involved in regulating cardionyocyte proliferative activity.Sap301, the neighboring gene of ENSMUST00000117266, might be involved in regulation of ENSMUST00000117266 on cardiomyocyte proliferative activity.