Expression Profile Analysis Of Long Non-coding RNAs Involved In Metformin-inhibited Gluconeogenesis

Background and ObjectiveWith the development of modern economy and the change of people’s living style,the incidence of diabetes,especially type 2 diabetes mellitus（T2DM）,is reaching epidemic level.Diabetes mellitus has been one of the major causes for mortality and morbidity.Thus,exploring the exact mechanism of diabetes and novel therapeutic strategies has great implications.The liver organ,which participates in the energy metabolism,maintains the blood glucose level in a narrow range through the opposing action of insulin and glucagon.The insulin level is increased and hepatic gluconeogenesis is inhibited under feeding state.When starved,glucagon enhances gluconeogenesis so that the blood glucose level remains normal.The process of gluconeogenesis in type 2 diabetes is enhanced,which results in hyperglycemia.Consequently,elucidating the mechanism underlying gluconeogenesis may provide a new perspective for treating diabetes.Long non-coding RNAs（lnc RNAs）are defined as transcripts longer than 200 nucleotides that don’t encode proteins.Increasing evidence demonstrates that lnc RNAs play a critical role in metabolic tissue development and function,such as muscle,adipose tissues,and islets.However,it remains unknown about the role of lnc RNAs in hepatic gluconeogenesis.Metformin is now widely prescribed for treating T2 DM.Previous researches on both diabetic patients and animal experiments showed that metformin lowered glucose level through inhibiting gluconeogenesis.However,the exact mechanism remains controversial.To identify whether lnc RNAs are involved in metformin-mediated inhibition of gluconeogenesis,we performed a systematic analysis of lnc RNAs expression profile in primary hepatocytes treated with c AMP and metformin.Then we selected eight lnc RNAs to validate their expressions with RT-q PCR.Our study aims to investigate the potential roles of lnc RNAs in gluconeogenesis and metformin’s glucose-lowering effect.Materials and Methods1.Primary hepatocytes of mice were isolated by a modified version of the collagenase method.2.After cell attachment to the plates,the medium was replaced with low glucose DMEM with dexamethasone prestimulation for 12 to 16 hours.Then the cells were incubated with c AMP in the presence or absence of metformin containing sodium lactate and sodium pyruvate for eight hours,termed control group,c AMP group,and metformin group,respectively.3.Total RNA was collected from hepatocytes under the three conditions,and then one of the key gluconeogenic enzymes PEPCK was measured by RT-q PCR.4.The global lnc RNAs expression analysis on hepatocytes obtained from three experiments was performed using Arraystar mouse lnc RNA microarray version 2.0.After extraction,the RNA was transcribed into c DNA.c DNA was labeled and further hybridized onto the mouse 8′60 K Lnc RNA Arrays.After washing,Agilent Scanner was used to scan the fluorescent intensities.Subsequently,the array images were analyzed and quantile normalization and data processing were performed by relevant software.Significantly differential genes were defined as an absolute log value of fold change over 2.0 and a P value less than 0.05.5.Several lnc RNAs were selected to validate the microarray result by RT-q PCR.6.Bioinformatic analysis was performed.Results1.Compared with control group,c AMP dramatically increased the m RNA level of PEPCK,which was suppressed by metformin treatment.2.Totally,22,016 lnc RNAs were detected in primary mouse hepatocytes.These lnc RNAs were carefully collected from the most authoritative databases,such as Ref Seq,UCSC,En Sembl,NRED,and other influential literatures.3.The protein-coding genes which located near those differentially expressed lnc RNAs were subjected to GO analysis and KEGG analysis,while the former was composed of molecular function,biological process,and cellular component.GO analysis and KEGG analysis provided a basis for understanding the general functions and possible pathways involving these differentially expressed lnc RNAs.4.Compared with control group,c AMP upregulated 456 lnc RNAs,among which 189 was downregulated by metformin.Similarly,c AMP downregulated 409 lnc RNAs whereas 167 lnc RNAs were reversed by metformin.The 356 metformin-reversed lnc RNAs had some general expression signatures.There were 39 sense lnc RNAs,78 antisense lnc RNAs,25 bidirectinal lnc RNAs,and 214 intergenic lnc RNAs.The length was mainly between 400 and 3600 nucleotides.They were transcribed from different chromosomes.5.From those metformin-inhibited lnc RNAs upregulated by c AMP,we selected eight lnc RNAs（AK133602,ENSMUST00000138573,ENSMUST00000129953,uc009 njr.1,AA591058,NR₀27710,NR₀30715,ENSMUST00000145208）to verify their expressions by RT-q PCR.Consequently,their expressions were in accordance with the microarray results.6.By bioinformatic analysis,we got the sequences of those validated lnc RNAs.Specifically,NR₀27710 and ENSMUST00000145208 were associated with protein-coding genes,peroxisome proliferative activated receptor γ coactivator 1α（PGC-1α）and G protein-coupled receptor 155（Gpr155）,respectively.7.PGC-1α and Gpr155 exhibited similar expression patterns to their associated lnc RNAs by RT-q PCR.c AMP upregulated their expressions which were suppressed by metformin（P<0.05）.ConclusionsMetformin suppressed gluconeogenesis via inhibiting the expression of key gluconeogenic enzyme genes.The microarray results were of high quality and were reliable for further research.Differentially expressed lnc RNAs exerted their biological functions and participated in various metabolic pathways through affecting their nearby protein-coding genes.Especially,NR₀27710 and ENSMUST00000138573 are associated with PGC-1α and Gpr155,which may be involved in the process of metformin-inhibted gluconeogenesis.Lnc RNAs are potentially involved in hepatic gluconeogenesis and metformin may exert its glucose-lowering effect by regulating a subset of lnc RNAs.