Identification of novel circulatory microRNA signatures linked to patients with ischemic stroke

Identification of novel circulatory microRNA signatures linked to patients with ischemic stroke Abstract MicroRNAs (miRNAs) are involved in growth, development, and occurrence and progression of many diseases. MiRNA-mediated post-transcriptional regulation is poorly understood in vascular biology and pathology. The purpose of this is to determine circulatory miRNAs as early detectable peripheral biomarkers in patients with ischemic stroke (IS). MiRNAs expression levels were measured in IS serum samples and healthy controls using Illumina deep sequencing analysis and identified differentially expressed miRNAs. Differentially expressed miRNAs were further validated using SYBR-green-based quantitative real-time PCR (qRT-PCR) assay in postmortem IS brains, lymphoblastoid IS cell lines, oxygen and glucose deprivation/reoxygenation -treated human and mouse neuroblastoma cells, and mouse models of hypoxia and ischemia (HI)-induced stroke. A total of 4656 miRNAs were differentially expressed in IS serum samples relative to healthy controls. Out of 4656 miRNAs, 272 were found to be significantly deregulated in IS patients. Interestingly, we found several novel and previously unreported miRNAs in IS patients relative to healthy controls. Further analyses revealed that some candidate miRNAs and its target genes were involved in the regulation of the stroke. To the best of our knowledge, this is the first study identified potential novel candidate miRNAs in IS serum samples from the residents of rural West Texas. MiRNAs identified in this study could potentially be used as a biomarker and the development of novel therapeutic approaches for stroke. Further studies are necessary to better understand miRNAs-regulated stroke cellular changes. Introduction Stroke is a common neurological disease with diverse etiologies that occurs when the blood supply to the brain is interrupted, resulting in a shortage of oxygen and nutrients to brain tissue. Due to multifactorial nature, stroke may be classified as a syndrome, not as a single disease. Stroke is the second leading cause of death globally and third leading cause of disability-adjusted life years worldwide (1,2). An estimated 7.2 million Americans ≥20 years of age self-report having had a stroke and approximately 795 000 strokes occur in the United States every year. On an average, every 40 s, someone in the United States has a stroke, and on an average, every 4 min, someone dies of stroke. Prevalence of stroke in the United States increases with age in both men and women (3). Ischemic stroke (IS) is described as a lack of blood supply and oxygen availability to an area of the brain due to narrowed or blocked arteries leading to or within the brain and the most predominant type of stroke accounting for approximately 87% of stroke cases (4). Stroke doubles the risk for dementia (post-stroke dementia), and approximately 30% of stroke patients go on to develop cognitive dysfunction within 3 years (5,6). Biomarkers might be useful in identifying different diseases, such as stroke, cancer, diabetes and disease severity (7,8). Identification of biomarkers could inform researchers in their attempts to develop early detectable peripheral biomarkers and could contribute to a better understanding of the etiologies and mechanisms underlying particular diseases, such as stroke. Recent molecular biology discoveries have revealed that microRNAs (miRNAs) can detect changes in the bodily organs, including brain that may lead to IS. MiRNAs are important post-transcriptional regulators that connect with multiple target messenger RNAs coordinately regulating target genes. MiRNAs have also been found to be important regulators of leukocyte gene expression in acute IS cases (9). Many studies showed that miRNAs altered after central nervous system injury moderate processes that stimulate neuronal death with inflammation, apoptosis and oxidative stress (10,11). Furthermore, miRNAs can act as sensitive biomarkers of secondary brain damage. Studies also suggested that peripheral blood miRNAs and their profiles could be developed as diagnostic and prognostic biomarkers of IS, as well as serving as innovative targets in the treatment of this disease (12). Clinical approaches accessible for the diagnosis and prognosis of stroke were restricted to radiological imaging, which was with limited availability and higher cost. Diagnosis of early stage of stroke and its development could be improved through the finding of new biomarkers. MiRNA-mediated post-transcriptional regulation is poorly understood in vascular biology and pathology. Currently, there are no drugs/agents and peripheral biomarkers available that can delay and/or detect IS in humans. Hence, identification of blood-based early detectable miRNAs could contribute to a better understanding of the etiologies and mechanisms underlying IS. In the present study, we sought to determine miRNAs as early detectable biomarkers in serum samples from IS patients relative to healthy controls. We used miRNA deep sequencing method and validated differentially expressed miRNAs using quantitative real-time PCR (qRT-PCR). Further, we validated the trend of selected miRNAs using postmortem IS brains, lymphoblastoid IS cell lines, oxygen and glucose deprivation/reoxygenation (OGD/R)-treated human (SH-SY5Y) and mouse neuroblastoma (N2a) cells and hypoxia and ischemia (HI)-induced stroke mouse model. Results Differentially expressed miRNA profile by deep sequencing Illumina deep sequencing analysis of serum samples provided a total of 484 651 777 raw RNA reads. Among these, 341 678 616 (70.5%) were mapped to miRNAs, and 39 890 853 reads were mapped to mRNA and 24 723 087 reads were mapped to other RNAs (RFam: rRNA, tRNA, snRNA, snoRNA and others) (Supplementary Material, Fig. S1). Based on the size distribution of all known miRNAs, 15–32 nucleotide (nt) reads were selected as ‘mappable reads’ for further analysis (Supplementary Material, Fig. S2). Of these reads, 87.6% of the small RNAs were 17–22 nt in size, which is typical miRNA sizes produced by RNA Dicer-digested products. The mappable reads sequences were subjected to advance bioinformatics analysis and to simplify the data from sequencing, all identical sequence reads were grouped and then assigned a unique sequence tag (Supplementary Material, Table S1). Our miRNA sequencing analysis revealed/detected a total of 4656 miRNAs in serum samples of IS patients versus healthy controls. Among them, 272 miRNAs were differentially deregulated (FC ±2, P ≤ 0.05) in IS patients, compared with healthy controls. Interestingly, 173 miRNAs were significantly upregulated, while 76 were found to be significantly down-regulated in IS patients. Hierarchical clustering performed with differentially expressed miRNAs, revealed that miRNA expression patterns were able to classify individuals according to their disease status. Among these miRNAs, we chose 16 miRNAs (Supplementary Material, Table S2) that were differentially expressed between the IS patients and healthy controls and where number of reads 10 in either IS patients or healthy controls were detected and at least a meaningful ±2-fold change between the group was identified (Fig. 1). Figure 1. View largeDownload slide Heat map of the hierarchical cluster analysis of differentially expressed miRNAs between IS patients and healthy controls detected by deep Sequencing. The color indicates the log 2-fold change from high (red) to low (green), as indicated by the color key. Figure 1. View largeDownload slide Heat map of the hierarchical cluster analysis of differentially expressed miRNAs between IS patients and healthy controls detected by deep Sequencing. The color indicates the log 2-fold change from high (red) to low (green), as indicated by the color key. Validation of candidate miRNAs in serum samples by real-time RT-PCR We validated 16 miRNAs using real-time RT-PCR analysis in same RNA samples that were used for deep sequencing analysis. A few known and several novel and previously unreported miRNAs were found in IS serum samples. Of the 16 miRNAs differentially expressed between IS patients and healthy controls in the discovery cohort, the validation studies found that four miRNAs [PC-3p-57664 (P = 0.01), PC-5p-12969 (P = 0.04), hsa-miR-122-5p (P = 0.01) and hsa-miR-211-5p (P = 0.03)] were significantly upregulated in IS patients compared with healthy controls. Whereas four miRNAs [hsa-miR-22–3p (P = 0.01), PC-3p-32463 (P = 0.0001), hsa-miR-30d-5p (P = 0.0009) and hsa-miR-23a-3p (P = 0.03)] were significantly down-regulated in the same comparison (Fig. 2A). Figure 2. View largeDownload slide (A) Validation of candidate miRNAs in serum samples by qRT-PCR. Significantly deregulated miRNA expression in IS versus the healthy controls. The y-axis depicts lnΔCq. P-values were determined by Mann-Whitney test. (B) Validation of serum miRNAs using postmortem IS brains by qRT-PCR. Box plots of lnΔCq values of significant serum miRNAs in IS brains compared to healthy control brains. Figure 2. View largeDownload slide (A) Validation of candidate miRNAs in serum samples by qRT-PCR. Significantly deregulated miRNA expression in IS versus the healthy controls. The y-axis depicts lnΔCq. P-values were determined by Mann-Whitney test. (B) Validation of serum miRNAs using postmortem IS brains by qRT-PCR. Box plots of lnΔCq values of significant serum miRNAs in IS brains compared to healthy control brains. Validation of serum miRNAs using postmortem IS brains We analyzed the expression of above-selected 16 miRNAs in the postmortem IS brains (n = 10) and control brains (n = 10) by real-time RT-PCR (Supplementary Material, Table S3). Analysis showed that four miRNAs [PC-3p-57664 (P = 0.04), PC-5p-12969 (P = 0.006), hsa-miR-122-5p (P < 0.0001) and hsa-miR-211-5p (P < 0.0001)] were consistently upregulated and three miRNAs [PC-3p-32463 (P = 0.01), hsa-miR-30d-5p (P = 0.01) and hsa-miR-23a-3p (P = 0.03)] were significantly down-regulated in the IS brains compared with control brains (Fig. 2B). The expression of PC-3p-57664, PC-5p-12969, hsa-miR-122–5p and hsa-miR-211-5p were the most significantly upregulated in both the IS serum and postmortem IS brains, suggesting that these upregulated miRNAs are relevant to IS—in terms of early detection and disease progression. Validation of serum miRNAs using lymphoblastoid IS cell lines To further validate the miRNA sequencing data, expression of 16 miRNAs were measured in lymphoblastoid cell line (LCL) strains by real-time RT-PCR (Supplementary Material, Table S4). Seven miRNAs [mmu-mir-6240-p5 (P = 0.007), ggo-miR-139 (P = 0.002), hsa-mir-760 (P = 0.001), PC-3p-57664 (P = 0.0009), PC-5p-12969 (P = 0.02), hsa-miR-122-5p (P = 0.03) and hsa-miR-211-5p (P < 0.0001)] were upregulated in IS LCL compared with control LCL strains. PC-3p-32463 (P = 0.01) and hsa-miR-30d-5p (P = 0.01) were significantly down-regulated in the IS LCL strains (Fig. 2C). These results further confirmed the significant response of these four miRNAs (PC-3p-57664, PC-5p-12969, hsa-miR-122-5p and hsa-miR-211-5p) in IS pathogenesis. Candidate miRNAs expression in OGD-treated cells Human neuroblastoma cells (SH-SY5Y) To determine the involvement of candidate miRNA expression in hypoxic-ischemic-induced neuronal death, OGD-stimulated human neuroblastoma cells was monitored. We selected 16 miRNAs for analysis based on a number of factors including, their expression levels in IS serum, postmortem IS brains and IS LCL strains. PC-5p-211 (P = 0.006), ggo-miR-139 (P = 0.001), hsa-mir-760 (P = 0.02), hsa-miR-96 (P = 0.0007), hsa-miR-99a-5p (P = 0.0004), PC-3p-57664 (P = 0.01), PC-5p-12969 (P = 0.01), hsa-miR-122–5p (P = 0.0006), hsa-miR-211–5p (P = 0.001) were increased significantly in human neuroblastoma cells following OGD/R exposure compared with control cells (Fig. 3A). Mmu-miR-5124a (P = 0.03), PC-3p-32463 (P = 0.0003) were significantly down-regulated in the OGD-treated cells. Figure 3. View largeDownload slide (A) MicroRNAs expression in OGD/R-treated human neuroblastoma cells (SH-SY5Y) by qRT-PCR. Data are presented as the mean ± SD of three independent experiments. (B) MicroRNAs expression in OGD/R-treated mouse neuroblastoma cells (N2a) by qRT-PCR. Figure 3. View largeDownload slide (A) MicroRNAs expression in OGD/R-treated human neuroblastoma cells (SH-SY5Y) by qRT-PCR. Data are presented as the mean ± SD of three independent experiments. (B) MicroRNAs expression in OGD/R-treated mouse neuroblastoma cells (N2a) by qRT-PCR. Mouse neuroblastoma (N2a) cells We further evaluated the miRNA expression profiles of the OGD/R-activated N2a cells. To test the hypothesis, we selected 12 miRNAs. In these, nine miRNAs that exhibited significantly altered expression levels between the hypoxic and normoxic conditions. PC-3p-57664 (P = 0.04), PC-5p-12969 (P = 0.005), mmu-miR-122-5p (P = 0.002) and mmu-miR-211-5p (P = 0.04) were upregulated significantly in mouse neuroblastoma cells following OGD exposure compared with normoxia-treated cells (Fig. 3B). Mmu-miR-5124a (P = 0.01) and PC-3p-32463 (P = 0.002) were significantly down-regulated in the OGD-treated cells. Differential expression of miRNAs in the brain of hypoxia and ischemia (HI)-induced neonatal mice To verify the accuracy of miRNA sequencing results, we selected 11 miRNAs for further validation in brains of HI-induced mouse models. We studied four different brain regions, including hippocampus, striatum, cerebral cortex and cerebellum from HI-induced neonatal and control, naive mice. Out of 11 miRNAs, the following five miRNAs [mmu-miR-211-5p (P = 0.003), PC-5p-211 (P = 0.02), PC-3p-57664 (P = 0.02), PC-5p-12969 (P = 0.0005) and mmu-miR-122-5p (P = 0.002)] were significantly upregulated in the hippocampus of HI mice when compared with that of naïve control mice (Fig. 4). Details of 11 miRNA expressions in other brain regions are given in Supplementary Material, Figure S3A–C. Figure 4. View largeDownload slide Quantitative RT-PCR analysis of miRNAs in hippocampus region of stroke hypoxia ischemia model. Fold change was calculated by 2-ΔΔCT method. Significant difference among groups were calculated by paired t-test with two-tailed P < 0.05 is considered significant. Figure 4. View largeDownload slide Quantitative RT-PCR analysis of miRNAs in hippocampus region of stroke hypoxia ischemia model. Fold change was calculated by 2-ΔΔCT method. Significant difference among groups were calculated by paired t-test with two-tailed P < 0.05 is considered significant. Receiver operating characteristics (ROC) curve analysis Expression of four miRNAs (PC-3p-57664, PC-5p-12969, hsa-miR-122-5p and hsa-miR-211–5p) consistently upregulated throughout the validation analysis. Therefore, we evaluated the diagnostic value of these four miRNAs by plotting ROC curve in IS serum, postmortem IS brains, LCL strains and HI stroke mouse models. The curves were plotted based on the ΔCt value of candidate miRNAs expression in different sources. Upon analysis, PC-3p-57664 (AUROC = 0.76; 95% CI: 0.571–0.953; P = 0.01), PC-5p-12969 (AUROC = 0.80; 95% CI: 0.6053–0.996; P = 0.006), hsa-miR-122–5p (AUROC = 0.72; 95% CI: 0.569–0.874; P = 0.03) and hsa-miR-211–5p (AUROC = 0.72; 95% CI: 0.533–0.919; P = 0.04) showed significant area under curve in IS serum samples compared with the healthy controls. The same trend was observed in postmortem IS brains, IS LCL strains as well in the HI stroke mice (Fig. 5). Thus, ROC analysis confirmed that the profile of the four serum miRNAs could be a simple, specific and non-invasive molecular biomarker for diagnosing IS. Figure 5. View largeDownload slide ROC curve analysis of serum miRNAs as diagnostic biomarkers differentiating IS patients from healthy controls. (A) Serum, (B) postmortem IS brains and (C) IS lymphoblastoid IS cell lines. (D) HI stroke mouse model (hippocampus). Figure 5. View largeDownload slide ROC curve analysis of serum miRNAs as diagnostic biomarkers differentiating IS patients from healthy controls. (A) Serum, (B) postmortem IS brains and (C) IS lymphoblastoid IS cell lines. (D) HI stroke mouse model (hippocampus). Discussion The overall objective of our study was to identify early detectable peripheral biomarkers for IS in the residents of rural West Texas. MiRNAs have been identified as circulating biomarkers in several diseases, including IS (13–15). Our Illumina deep sequencing and further validation analysis revealed that 16 circulating miRNAs that distinguishes between IS patients and healthy controls. Of the 16 miRNAs differentially expressed between IS patients and healthy controls, 12 miRNAs are previously reported in stroke and other diseases and four miRNAs are novel and unreported miRNAs. As of now, no study is reported to validate the expression of these miRNAs in post-mortem IS brains, IS LCL strains, OGD/R-treated human and mouse neuroblastoma cells and HI stroke mice. For the first time, we identified the novel potential candidate miRNAs (PC-3p-57664, PC-5p-12969) in IS serum samples from the residents of rural West Texas and verified their expression in mice as well. MiRNAs are cell specific, interestingly these novel candidates were consistently upregulated in all stroke sources and showed a strong correlation with stroke pathology. Hence, these miRNAs might provide the possibilities of unique biomarker candidate for stroke. Chen and Zhang identified the variant rs2507800 in the 3′-untranslated region of angiopoietin-1 that might reduce the risk of stroke by interfering with hsa-miR-211 binding site (16). Interestingly in our current study, hsa-miR-211-5p was upregulated in IS patients. Hsa-miR-122 was identified to be related to human stroke based on the Human MicroRNA Disease Database (17). Another study investigated miRNA expression profile and found that miR-122 was down-regulated (18). Jickling et al. identified that miR-122 were decreased in acute IS patients compared with controls (9). Hsa-miR-23a and hsa-miR-22 were significantly down-regulated in stroke patients (19). MiR-23a levels differed in male and female ischemic brains, providing evidence for sex-specific miRNA expression in stroke (20). Hypertension is a well-established risk factor for stroke. Several studies showed that miRNAs were known to impact the state of hypertension directly or indirectly. In another research, miR-30d was down-regulated known to be involved in hypertension (21). Interestingly, in our study also miR-30d was down-regulated, which clearly meant there was a link between stroke and hypertension. A study by Long et al. identified that circulating miR-30a was markedly down-regulated in all patients with IS until 24 weeks (22). Postmortem human brain tissue was being used for quantifying cellular and molecular markers of neural courses with the area of improved understanding the variations in the brain caused by neurological diseases (23). However, the miRNA expression levels and molecular characterizations were not investigated using postmortem IS brains. To the best of our knowledge, our study was first to validate the miRNAs using postmortem IS brain specimens. MiRNAs PC-3p-57664, PC-5p-12969, hsa-miR-122-5p and hsa-miR-211-5p were consistently upregulated and PC-3p-32463, PC-5p-211, ggo-miR-139, hsa-miR-30d-5p, mmu-mir-6240-p5 and hsa-miR-23a-3p were significantly down-regulated in the IS brains. A recent study reported that a decrease of brain miRNA-122 level was deleterious and could be considered as an early marker of stroke in the stroke-prone spontaneously hypertensive rat (24). Elevating miR-122 improves stroke outcomes and this occurred via down-regulating miR-122 target genes in blood leukocytes (25). Down-regulation of miRNA-30a improves ischemic injury through enhancing beclin 1-mediated autophagy in N2a cells and cultured cortical neurons after OGD, and mouse brain with MCAO-induced IS (26). LCLs are the biological resources that have been used in various research fields related to human genetics, pharmacogenomics and immunology (27,28). LCLs have the potential to disclose at least a subset of brain-related miRNAs implicated in stroke. Hypoxia induces time-dependent alteration of the expression levels of miRNAs suggesting their involvement in the cellular response to ischemic injury (29). In this study, we performed miRNA expression in IS LCL strains and OGD/R on human and mouse neuroblastoma cells to mimic ischemia in vitro. To our knowledge, this is the first study to examine the roles of miRNA expression variations in IS LCLs. MiRNAs have essential roles in brain function, including neurogenesis, neural development and cellular responses leading to changes in synaptic plasticity. They are also implicated in neurodegeneration and neurological disorders, in responses to HI, and in ischemic tolerance induced by ischemic pre-conditioning (30). Expression levels of few miRNAs could be differently modulated in both in vivo and in vitro experimental models (25,31). We assessed the expressions of 11 miRNAs using a HI-induced in post-natal day nine (P9) C57BL/6J mice. Hippocampal region of the HI-induced neonatal mouse brain showed the most consistent differential expression of miRNA compared with other regions. A recent global expression of miRNAs in a P10 rat model of cerebral HI found that miR-30d-5p was one of the most deregulated miRNAs in neonatal brains in response to HI. Collectively, these results indicated that miR-30d-5p modulated survival programs of neural cell by regulating autophagy and apoptosis (32). It is possible that miRNAs identified in this study may have implications for both consequences and risk factors of stroke. In this study, we investigated serum samples from IS patients and we strongly feel that differentially expressed miRNAs are consequences of disease process, and these differentially miRNAs can also be used for the development of novel therapeutic targets for IS. However, further studies are necessary to better understand miRNAs-regulated stroke not only for risk but also for consequences of stroke. Based on our findings, we can only say that observed miRNAs in IS patients are different from healthy controls. In summary, miRNA sequencing analysis of IS serum samples showed significant deregulation of 16 miRNAs. Among 16 miRNAs, four miRNAs (PC-3p-57664, PC-5p-12969, hsa/mmu-miR-122–5p and hsa/mmu-miR-211–5p) were almost consistently upregulated in human IS serum samples, human post-mortem IS brain specimens, human lymphoblastoid IS cell lines, OGD/R-treated human and mouse neuroblastoma cells and HI stroke mouse models. ROC curve analysis in serum and postmortem brain also confirmed their diagnostic potential for stroke. Further, GO and KEGG pathway analysis showed the regulation of many stroke-related genes and pathways by these miRNAs. Based on intense analysis, we conclude that circulatory levels of PC-3p-57664, PC-5p-12969, miR-122-5p, miR-211-5p might be the potential biomarkers for the diagnosis of IS. Materials and Methods Enrollment of study samples For this, 34 IS patients (13 males, 21 females: mean age of 62.88 ± 11.94 years) and 11 healthy controls (5 males, 6 females: mean age: 62.63 ± 6.6 years) were used as the study group. Sera samples were collected from patients and healthy controls under Facing Rural Obstacles to healthcare Now Through Intervention, Education & Research (FRONTIER) project based at Garrison Institute on Aging (GIA), Texas Tech University Health Sciences Center. The Institutional Review Board (IRB) protocol was approved for Project FRONTIER (IRB#L06–028). All the bio-specimens were stored at the GIA. Information on demographic characteristics, medical history, biochemical profile and established risk factors were recorded by using a standardized questionnaire (Supplementary Material, Table S5). RNA extraction, small RNA library construction RNA was isolated from 1.5 ml of serum using Plasma/Serum RNA purification Midi Kit as per manufacturer’s instructions (Cat. No: 56100; Norgen Biotek Corp., Thorold, ON, Canada). All RNA samples were processed and analyzed by LC Sciences (Houston, TX). The quality and quantity of the RNA samples were tested using an Agilent 2100 Bioanalyzer (Agilent). A small RNA library was generated using the Illumina Truseq™ Small RNA Preparation kit according to Illumina’s TruSeq™ Small RNA Sample Preparation Guide [(15004197 C), Illumina Inc., Part # 1004239 Rev. A, 2008; Cat. No. RS-930-1012, Part No. 15004197 Rev. B, January 2011]. Primary screening by deep sequencing and data analysis The purified cDNA library was used for cluster generation on Illumina’s Cluster Station and then sequenced on Illumina GAIIx following vendor’s instruction for running the instrument. Raw sequencing reads (40 nts) were obtained using Illumina’s Sequencing Control Studio software version 2.8 (SCS v2.8) following real-time sequencing image analysis and base-calling by Illumina’s Real-Time Analysis version 1.8.70 (RTA v1.8.70). The extracted sequencing reads were stored and then a proprietary pipeline script, ACGT101-miR v4.2 (LC Sciences), was used for sequencing data analysis. After the raw sequence reads, or sequenced sequences (sequ seqs) were extracted from image data, a series of digital filters (LC Sciences) were employed to remove various un-mappable sequencing reads. Impurity sequences due to sample preparation, sequencing chemistry and processes, and the optical digital resolution of the sequencer detector were also removed. Remaining sequ seqs with lengths between 15 and 32 bases were grouped by families (unique seqs), and were used to map with the reference database files. Various ‘mappings’ were performed on unique seqs against pre-miRNA (mir) and mature miRNA (miR) sequences listed in the latest release of miRbase (v21.0; ftp://mirbase.org/pub/mirbase/CURRENT/; specific species: hsa; selected species: ggo, ppa, ptr, ppy, ssy, age, lla, sla, pbi, mml, mne, lca, cgr, mmu, rno, cfa, ocu, efu, aja, eca, mdo, sha, meu, oan, bta, chi, oar, tch, ssc) (33–35) or genome based on the public releases of appropriate species (V37.1; ftp://ftp.ncbi.nih.gov/genomes/H sapiens/). Mappings were also done on mirs of interest against genome sequence. Mappable unique seqs were mapped to other defined databases, such as mRNA, RFam and Repbase (V37.1; ftp://ftp.ncbi.nih.gov/genomes/H sapiens/RNA/). Methods and criteria used for various mappings were documented in the ACGT-101 User’s Manual. Sequences were mapped against reported miRNA, species’ genomes, and other RNA databases (e.g. RFam, repase, mRNA) and were classified as follows (Supplementary Material, Fig. S4): Mappable reads mapped to selected mirs in miRbase Mirs mapped to species specific genome (Homo sapiens) Mirs are of specific species (Homo sapiens) (group1a) Mirs are of selected species (Mammalia) (group1b) Reads mapped to other locations too and Reads mapped only to the same locations in the genome as that of mirs (group 1c) Mirs un-mapped to species specific genome Reads mapped/un-mapped to species specific genome Extended sequences potentially form hairpins (group 2a) Extended sequences potentially cannot form hairpins (group 2b) Reads mapped to miRs of selected species (group 3a) Reads unmapped to miRs of selected species (group 3b) Mappable reads un-mapped to selected mirs in miRbase Reads un-mapped to mRNA, Rfam and repbase Reads mapped to species specific genome Extended sequences potentially form hairpins (group 4a) Extended sequences potentially cannot form hairpins (group 4b) Reads un-mapped to species specific genome (no hit) Reads mapped to mRNA, Rfam, or repbase (others) Validation of differently expressed serum miRNAs using quantitative real-time RT-PCR To support the data obtained from the deep sequencing results, qRT-PCR analysis was performed to validate further. One µG of total RNA was reverse transcribed using miRNA First-Strand cDNA synthesis kit (Agilent Technologies Inc., CA), following manufacturer’s instructions. Resulting cDNAs were diluted with 20 μl of RNase-free water and stored at −80°C for further analysis (36). Primers for 16 miRNAs were synthesized commercially (Integrated DNA Technologies, Inc., Iowa) (Supplementary Material, Table S2). U6, one of the uniformly expressed small RNAs, was used as the internal control for real-time RT-PCR. Briefly, 1 μl of miRNA-specific forward primer (10 μM), 1 μL of a universal reverse primer (3.125 μM) (Agilent Technologies Inc.), 10 μl of 2× SYBR® Green PCR master mix (Applied Biosystems, NY) and 1 μl of cDNA were mixed. To this mixture RNase-free water was added up to a 20 μl of final volume. The reactions were amplified for 5 min at 95°C, followed by 40 cycles of 95°C for 10 s, 60°C for 15 s and 72°C for 25 s at 7900HT Fast Real Time PCR System (Applied Biosystems). All reactions were performed in triplicate, and the controls (no template and no RT) were included for each gene, and the data were expressed as the mean ±SD. Fold change of each miRNAs were calculated as described previously (37). The threshold cycle (CT) values were automatically determined by the instrument, and the fold change of each miRNAs were calculated using the following equation: the formula (2-ΔΔCt), where ΔCt was calculated by subtracting Ct of U6snRNA from the Ct of particular miRNAs target, and ΔΔCt value was obtained by subtracting ΔCt of particular miRNAs target in the controls from the ΔCt of miRNAs target in the IS. Postmortem brains from stroke patients and controls In the present study, 20 postmortem brain samples were investigated which consisted of 10 IS [5 males and 5 females: Age ranged from 57–91 years (78.3 ± 11.89) and postmortem interval (PMI) varied from 4–23.9 h (average 16.48 h)] and 10 normal control subjects [5 males and 5 females: age ranged from 67 to 91 years (76.9 ± 8.62) and PMI varied from 11.8 to 25 h (average 17.99 h)] were obtained from the Human Brain and Spinal Fluid Resource Center (Los Angeles, CA) and Harvard Brain Tissue Resource Center (HBTRC) through NIH NeuroBiobank. These brain banks were responsible for obtaining subject consent and the unidentifiable coding of subject information. The study protocol was approved by the Institute Ethical Committee at TTUHSC (IBC protocol number: 14013). Lymphoblastoid cell lines Epstein-Barr Virus (EBV) transformed LCLs from 20 IS patients (10 males, 10 females: mean age of 66.9 ± 6.13 years) and 10 unrelated healthy subjects (5 males, 5 females: mean age of 62.8 ± 4.91 years) were obtained from the Coriell Cell Repository. These samples had been collected and anonymized by National Institute of Neurological Disorders and Stroke (NINDS), and all subjects had provided written consent for their experimental use. LCLs were cultured in Roswell Park Memorial Institute Medium 1640 (RPMI 1640 medium) with 2 mM l-glutamine (Gibco, Carlsbad, CA, #11875) supplemented with 15% heat-inactivated fetal bovine serum (FBS). Induction of neonatal hypoxia and ischemia HI was induced in post-natal day nine (P9) C57BL/6J mice. The pups were anesthetized with isoflurane (Butler Schein Animal Health Supply, Reno, NV) (5% for induction, 2–3% for maintenance) in 30% oxygen mixed with nitrous oxide. The body temperature of the pups were maintained at 36°C using a heated surgical table (Molecular Imaging Products, Bend, OR). Under a surgical microscope (Nikon SMZ-800 Zoom Stereo, Nikon, Melville, NY), a midline skin incision was made and the trachea was visualized through the muscle overlying it. The left common carotid artery was freed from the left common jugular vein and left vagus nerve by blunt dissection, electrically cauterized and cut. The incision was injected with 0.5% bupivacaine and closed with a single 6.0 silk suture. Animals were returned to their dams and monitored continuously for a 2 h recovery period. To induce unilateral ischemic injury, the animals were placed in a hypoxia chamber (BioSpherix Ltd, Redfield, NY) equilibrated with 10% O2 and 90% N2 at 36°C for 50 min. This is a well-characterized model of neonatal HI and results in reproducible brain injury ipsilateral (IL) to the electrocauterized left common carotid artery (38–41). Naïve mice were not exposed to HI or any surgical intervention. Mice were decapitated for collection of fresh tissue of the regions of striatum, hippocampus, cortex and cerebellum on day 3 (P12) post-HI (42). After extraction, the fresh tissues were kept in −20°C for 1 day and in −80°C for long-term storage. All procedures on animals were carried out in adherence with NIH Guide for the Care and use of Laboratory Animals and approved by the Institutional Animal Care and Use Committee at the University of Wisconsin-Madison. Cell cultures The human neuroblastoma cell line SH-SY-5Y and mouse neuroblastoma cell line N2a were grown in Dulbecco’s modified Eagle’s medium/F12 (1/1) (DMEM/F12, Gibco Life Technologies, Carlsbad, CA, #11320), minimum essential media respectively containing 10% FBS, penicillin (100 U/ml) and streptomycin (100 µG/ml). The cultures were maintained at 37°C in 95% air, 5% CO2 in a humidified incubator. After the cells were seeded, they were allowed to grow for 24–48 h or until 80% confluence in 6-well plates then they were used for experimentation. Oxygen and glucose deprivation/reoxygenation protocol OGD/R is a well-established in vitro model to study the pathology and pharmacology of ischemic damage, OGD/R was achieved using methods published earlier (43–46). Culture media were replaced with deoxygenated, no glucose DMEM (Gibco, Carlsbad, CA, #11966) and placed in hypoxia chamber (Biospherix, ProOx model 110) with 95% N2 and 5% CO2 for 4 h at 37°C to represent OGD condition. After the appropriate time, the culture plates were removed from the hypoxia chamber. Prior to re-oxygenation, cells were washed in PBS and the medium was replaced with complete culture medium then placed in a humidified incubator at 37°C for 20–24 h to represent reoxygenation. MiRNAs extraction from tissues and cell pellets and qRT-PCR Total RNA was isolated from the 80 mg of tissues and cell pellets using the TriZol RT reagent (Ambion, USA) as per manufacturer’s instructions. MiRNAs extraction and cDNA synthesis were followed as described earlier (36). The quality and quantity of the RNA were analyzed by NanoDrop analysis. The value of absorbance of each RNA sample (A260/A280) was 1.8–2.0. cDNA was synthesized from 1 μG of RNA using miRNA First-Strand cDNA synthesis kit (Agilent Technologies Inc.). Target gene prediction and enrichment analysis MiRNAs of interest were selected based on the statistical significance, fold change difference and biological rationales. Target gene prediction was performed by using TargetScan 6.0 (www.targetscan.org). For human, the program searches was run to match the miRNA seed regions and orthologous at 3' UTRs of human genes (47). Gene function enrichment and biological pathway analysis were performed using DAVID online tool suite (https://david.ncifcrf.gov/) (48). Validated and predicted target genes were uploaded at Gene List Manager according the tool instructions. Gene function enrichment and pathway analysis results were obtained and given in Supplementary Material, Figure S5A–C. Statistical Analysis Data were presented as means ± SD for other variables. The qRT-PCR validation analysis was based on the 2-ΔΔCT value of genes in each sample from IS and healthy controls. The Cq values of miRNAs were displayed as LnΔCq for statistical analysis (49). P-value was calculated, based on the paired and unpaired t-tests for analyzing two groups. MiRNAs levels between the probable IS and healthy controls were analyzed using a two-sided nonparametric Mann-Whitney test. Sensitivity and specificity of measured variable for IS biomarker were examined using a ROC curve analysis under a nonparametric approach. P-value < 0.05 was considered to be statistically significant. All analyses were performed by GraphPad Prism (version 6.0; GraphPad Software, La Zolla, CA). Supplementary Material Supplementary Material is available at HMG online. Acknowledgements We sincerely thank Ms. Annette Boles, Ms. Kathy Hudson and all field coordinators for their support in collecting data and blood samples for this study. We also thank all the participants of Project FRONTIER for providing blood specimens. Post-mortem brain tissues were obtained from the Human Brain and Spinal Fluid Resource Center, Los Angeles and Harvard Brain Tissue Resource Center (HHSN-271–2013-00030C). We are thankful to the Coriell Institute of Medical Research for providing IS and control LCLs. Conflict of Interest statement. A provisional patent is in progress with the scientific content of our paper. Funding The research presented in this article was supported by the National Institute of Health grants AG042178, AG47812 and NS105473, and the Garrison Family Foundation and CH Foundation (PHR), NIH-NINDS K08 NS088563 (PC) and NIH P30 HD03352 (Waisman Center). References 1 Lozano R., Naghavi M., Foreman K., Lim S., Shibuya K., Aboyans V., Abraham J., Adair T., Aggarwal R., Ahn S.Y. et al.   ( 2012) Global and regional mortality from 235 causes of death for 20 age groups in 1990 and 2010: a systematic analysis for the Global Burden of Disease Study 2010. Lancet , 380, 2095– 2128. Google Scholar CrossRef Search ADS PubMed  2 Murray C.J., Vos T., Lozano R., Naghavi M., Flaxman A.D., Michaud C., Ezzati M., Shibuya K., Salomon J.A., Abdalla S. et al.   ( 2012) Disability-adjusted life years (DALYs) for 291 diseases and injuries in 21 regions, 1990–2010: a systematic analysis for the Global Burden of Disease Study 2010. Lancet , 380, 2197– 2223. Google Scholar CrossRef Search ADS PubMed  3 Benjamin E.J., Blaha M.J., Chiuve S.E., Cushman M., Das S.R., Deo R., de Ferranti S.D., Floyd J., Fornage M., Gillespie C. et al.   ( 2017) American Heart Association Statistics Committee and Stroke Statistics Subcommittee. Heart disease and stroke statistics-2017 update: a report from the American Heart Association. Circulation , 135, e146– e603. Google Scholar CrossRef Search ADS PubMed  4 Rosamond W., Flegal K., Furie K., Go A., Greenlund K., Haase N., Hailpern S.M., Ho M., Howard V., Kissela B. et al.   ( 2008) American Heart Association Statistics Committee and Stroke Statistics Subcommittee. Heart disease and stroke statistics–2008 update: a report from the American Heart Association Statistics Committee and Stroke Statistics Subcommittee. Circulation , 117, e25– e146. Google Scholar CrossRef Search ADS PubMed  5 Pendlebury S.T., Rothwell P.M. ( 2009) Prevalence, incidence, and factors associated with pre-stroke and post-stroke dementia: a systematic review and meta-analysis. Lancet Neurol ., 8, 1006– 1018. Google Scholar CrossRef Search ADS PubMed  6 Allan L.M., Rowan E.N., Firbank M.J., Thomas A.J., Parry S.W., Polvikoski T.M., O'Brien J.T., Kalaria R.N. ( 2011) Long term incidence of dementia, predictors of mortality and pathological diagnosis in older stroke survivors. Brain , 134, 3716– 3727. Google Scholar CrossRef Search ADS PubMed  7 Schwarzenbach H., Nishida N., Calin G.A., Pantel K. ( 2014) Clinical relevance of circulating cell-free microRNAs in cancer. Nat. Rev. Clin. Oncol ., 11, 145– 156. Google Scholar CrossRef Search ADS PubMed  8 Jung H.J., Suh Y. ( 2014) Circulating miRNAs in aging and age related disease. J. Genet. Genom ., 41, 465– 472. Google Scholar CrossRef Search ADS   9 Jickling G.C., Ander B.P., Zhan X., Noblett D., Stamova B., Liu D. ( 2014) MicroRNA expression in peripheral blood cells following acute ischemic stroke and their predicted gene targets. PLoS One , 9, e99283. Google Scholar CrossRef Search ADS PubMed  10 Vijayan M., Reddy P.H. ( 2016) Peripheral biomarkers of stroke: focus on circulatory microRNAs. Biochim. Biophys. Acta , 1862, 1984– 1993. Google Scholar CrossRef Search ADS PubMed  11 Nieto-Diaz M., Esteban F.J., Reigada D., Muñoz-Galdeano T., Yunta M., Caballero-López M., Navarro-Ruiz R., Del Águila A., Maza R.M. ( 2014) MicroRNA dysregulation in spinal cord injury: causes, consequences and therapeutics. Front. Cell. Neurosci ., 8, 53. Google Scholar CrossRef Search ADS PubMed  12 Ouyang Y.B., Stary C.M., Yang G.Y., Giffard R. ( 2013) MicroRNAs: innovative targets for cerebral ischemia and stroke. Curr. Drug. Targets , 14, 90. Google Scholar CrossRef Search ADS PubMed  13 Lorenzen J.M., Thum T. ( 2012) Circulating and urinary microRNAs in kidney disease. Clin. J. Am. Soc. Nephrol ., 7, 1528– 1533. Google Scholar CrossRef Search ADS PubMed  14 Fichtlscherer S., Zeiher A.M., Dimmeler S. ( 2011) Circulating microRNAs: biomarkers or mediators of cardiovascular diseases? Arterioscler. Thromb. Vasc. Biol ., 31, 2383– 2390. Google Scholar CrossRef Search ADS PubMed  15 Laterza O.F., Lim L., Garrett-Engele P.W., Vlasakova K., Muniappa N., Tanaka W.K., Johnson J.M., Sina J.F., Fare T.L., Sistare F.D., Glaab W.E. ( 2009) Plasma MicroRNAs as sensitive and specific biomarkers of tissue injury. Clin. Chem ., 55, 1977– 1983. Google Scholar CrossRef Search ADS PubMed  16 Cheng Y., Zhang C. ( 2010) MicroRNA-21 in cardiovascular disease. J. Cardiovasc. Transl. Res ., 3, 251– 255. Google Scholar CrossRef Search ADS PubMed  17 Yuan Y., Kang R., Yu Y., Liu J., Zhang Y., Shen C., Wang J., Wu P., Shen C., Wang Z. ( 2016) Crosstalk between miRNAs and their regulated genes network in stroke. Sci. Rep ., 6, 20429. Google Scholar CrossRef Search ADS PubMed  18 He W., Chen S., Chen X., Li S., Chen W. ( 2016) Bioinformatic analysis of potential microRNAs in ischemic stroke. J. Stroke Cerebrovasc. Dis ., 25, 1753– 1759. Google Scholar CrossRef Search ADS PubMed  19 Sepramaniam S., Tan J.R., Tan K.S., DeSilva D.A., Tavintharan S., Woon F.P., Wang C.W., Yong F.L., Karolina D.S., Kaur P. et al.   ( 2014) Circulating microRNAs as biomarkers of acute stroke. Int. J. Mol. Sci ., 15, 1418– 1432. Google Scholar CrossRef Search ADS PubMed  20 Siegel C., Li J., Liu F., Benashski S.E., McCullough L.D. ( 2011) miR-23a regulation of X-linked inhibitor of apoptosis (XIAP) contributes to sex differences in the response to cerebral ischemia. Proc. Natl. Acad. Sci. U.S.A ., 108, 11662– 11667. Google Scholar CrossRef Search ADS PubMed  21 Da Costa Martins P.A., De Windt L.J. ( 2012) MicroRNAs in control of cardiac hypertrophy. Cardiovasc. Res ., 93, 563– 572. Google Scholar CrossRef Search ADS PubMed  22 Long G., Wang F., Li H., Yin Z., Sandip C., Lou Y., Wang Y., Chen C., Wang D.W. ( 2013) Circulating miR-30a, miR-126 and let-7b as biomarker for ischemic stroke in humans. BMC Neurol ., 13, 178. Google Scholar CrossRef Search ADS PubMed  23 Schmitt A., Parlapani E., Bauer M., Heinsen H., Falkai P. ( 2008) Is brain banking of psychiatric cases valuable for neurobiological research? Clinics (Sao Paulo) , 63, 255– 266. Google Scholar CrossRef Search ADS PubMed  24 Stanzione R., Bianchi F., Cotugno M., Marchitti S., Forte M., Busceti C., Ryskalin L., Fornai F., Volpe M., Rubattu S. ( 2017) A decrease of brain microRNA-122 level is an early marker of cerebrovascular disease in the stroke-prone spontaneously hypertensive rat. Oxid. Med. Cell. Longev ., 2017, 1. Google Scholar CrossRef Search ADS   25 Liu da Z., Jickling G.C., Ander B.P., Hull H., Zhan X., Cox C., Shroff N., Dykstra-Aiello C., Stamova B., Sharp F.R. ( 2016) Elevating microRNA-122 in blood improves outcomes after temporary middle cerebral artery occlusion in rats. J. Cereb. Blood Flow Metab ., 36, 1374– 1383. Google Scholar CrossRef Search ADS PubMed  26 Wang W., Sun G., Zhang L., Shi L., Zeng Y. ( 2014) Circulating microRNAs as novel potential biomarkers for early diagnosis of acute stroke in humans. J. Stroke Cerebrovasc. Dis ., 23, 2607– 2613. Google Scholar CrossRef Search ADS PubMed  27 Lee J.E., Hong E.J., Nam H.Y., Kim J.W., Han B.G., Jeon J.P. ( 2011) MicroRNA signatures associated with immortalization of EBV-transformed lymphoblastoid cell lines and their clinical traits. Cell. Prolif ., 44, 59– 66. Google Scholar CrossRef Search ADS PubMed  28 Wheeler H.E., Dolan M.E. ( 2012) Lymphoblastoid cell lines in pharmacogenomic discovery and clinical translation. Pharmacogenomics , 13, 55– 70. Google Scholar CrossRef Search ADS PubMed  29 Ziu M., Fletcher L., Rana S., Jimenez D.F., Digicaylioglu M. ( 2011) Temporal differences in microRNA expression patterns in astrocytes and neurons after ischemic injury. PLoS One , 6, e14724. Google Scholar CrossRef Search ADS PubMed  30 Martinez B., Peplow P.V. ( 2016) Blood microRNAs as potential diagnostic and prognostic markers in cerebral ischemic injury. Neural Regen. Res ., 11, 0– 1378. Google Scholar CrossRef Search ADS   31 Koutsis G., Siasos G., Spengos K. ( 2013) The emerging role of microRNA in stroke. Curr. Top. Med. Chem ., 13, 1573– 1588. Google Scholar CrossRef Search ADS PubMed  32 Zhao H., Li G., Ma Q., Tao Z., Wang R., Fan Z., Feng Y., Ji X., Luo Y. ( 2017) MicroRNA-99a-5p in circulating immune cells as a potential biomarker for the early diagnosis of ischemic stroke. Brain Circ ., 3, 21– 28. 33 Griffiths-Jones S., Saini H.K., van Dongen S., Enright A.J. ( 2007) MiRbase: tools for microRNA genomics. Nucleic Acids Res ., 36, D154– D158. Google Scholar CrossRef Search ADS PubMed  34 Griffiths-Jones S., Grocock R.J., van Dongen S., Bateman A., Enright A.J. ( 2006) MiRbase: microRNA sequences, targets and gene nomenclature. Nucleic Acids Res ., 34, D140– D144. Google Scholar CrossRef Search ADS PubMed  35 Griffiths-Jones S. ( 2004) The microRNA Registry. Nucleic Acids Res ., 32, 109D– D111. Google Scholar CrossRef Search ADS   36 Kumar S., Vijayan M., Reddy P.H. ( 2017) MicroRNA-455-3p as a potential peripheral biomarker for Alzheimer's disease. Hum. Mol. Genet ., 26, 3808– 3822. Google Scholar CrossRef Search ADS PubMed  37 Livak K.J., Schmittgen T.D. ( 2001) Analysis of relative gene expression data using real-time quantitative PCR and the 2#DDCT method. Methods , 25, 402– 408. Google Scholar CrossRef Search ADS PubMed  38 Vannucci R.C., Vannucci S.J. ( 1997) A model of perinatal hypoxic-ischemic brain damage. Ann. N.Y. Acad. Sci ., 835, 234– 249. Google Scholar CrossRef Search ADS PubMed  39 Cengiz P., Kleman N., Uluc K., Kendigelen P., Hagemann T., Akture E., Messing A., Ferrazzano P., Sun D. ( 2011) Inhibition of Na+/H+ exchanger isoform 1 is neuroprotective in neonatal hypoxic ischemic brain injury. Antioxid. Redox. Signal ., 14, 1803– 1813. Google Scholar CrossRef Search ADS PubMed  40 Uluc K., Kendigelen P., Fidan E., Zhang L., Chanana V., Kintner D., Akture E., Song C., Ye K., Sun D., Ferrazzano P. et al.   ( 2013) TrkB receptor agonist 7, 8 dihydroxyflavone triggers profound gender- dependent neuroprotection in mice after perinatal hypoxia and ischemia. CNS Neurol. Disord. Drug. Targets , 12, 360– 370. Google Scholar CrossRef Search ADS PubMed  41 Cikla U., Chanana V., Kintner D.B., Udho E., Eickhoff J., Sun W., Marquez S., Covert L., Otles A., Shapiro R.A. et al.   ( 2016) Signaling is required for TrkB-mediated hippocampal neuroprotection in female neonatal mice after hypoxic ischemic encephalopathy (1, 2, 3). eNeuro , 3, 1– 14. Google Scholar CrossRef Search ADS   42 Chanana V., Tumturk A., Kintner D., Udho E., Ferrazzano P., Cengiz P. ( 2016) Sex differences in mouse hippocampal astrocytes after in-vitro ischemia. J. Vis. Exp ., 116, 1– 9. 43 Yu Z., Liu J., Guo S., Xing C., Fan X., Ning M., Yuan J.C., Lo E.H., Wang X. ( 2009) Neuroglobin-overexpression alters hypoxic response gene expression in primary neuron culture following oxygen glucose deprivation. Neuroscience , 162, 396– 403. Google Scholar CrossRef Search ADS PubMed  44 Zhu H., Wang Z., Xing Y., Gao Y., Ma T., Lou L., Lou J., Gao Y., Wang S., Wang Y. ( 2012) Baicalin reduces the permeability of the blood-brain barrier during hypoxia in vitro by increasing the expression of tight junction proteins in brain microvascular endothelial cells. J. Ethnopharmacol ., 141, 714– 720. Google Scholar CrossRef Search ADS PubMed  45 Guo F., Wang H., Li L., Zhou H., Wei H., Jin W., Wang Q., Xiong L. ( 2013) A novel domain of amino-Nogo-A protects HT22 cells exposed to oxygen glucose deprivation by inhibiting NADPH oxidase activity. Cell. Mol. Neurobiol ., 33, 443– 452. Google Scholar CrossRef Search ADS PubMed  46 Alluri H., Anasooya Shaji C., Davis M.L., Tharakan B. ( 2015) Oxygen-glucose deprivation and reoxygenation as an in vitro ischemia-reperfusion injury model for studying blood-brain barrier dysfunction. J. Vis. Exp ., 99, e52699. 47 Lewis B.P., Burge C.B., Bartel D.P. ( 2005) Conserved seed pairing, often flanked by adenosines, indicates that thousands of human genes are microRNA targets. Cell , 120, 15– 20. Google Scholar CrossRef Search ADS PubMed  48 Huang D.W., Sherman B.T., Lempicki R.A. ( 2009) Systematic and integrative analysis of large gene lists using DAVID Bioinformatics Resources. Nat. Protocol ., 4, 44– 57. Google Scholar CrossRef Search ADS   49 Tan L., Yu J.T., Liu Q.Y., Tan M.S., Zhang W., Hu N., Wang Y.L., Sun L., Jiang T., Tan L. ( 2014) Circulating miR-125b as a biomarker of Alzheimer's disease. J. Neurol. Sci ., 336, 52– 56. Google Scholar CrossRef Search ADS PubMed  © The Author(s) 2018. Published by Oxford University Press. All rights reserved. For permissions, please email: journals.permissions@oup.com This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices) http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Human Molecular Genetics Oxford University Press

Identification of novel circulatory microRNA signatures linked to patients with ischemic stroke

Loading next page...
 
/lp/ou_press/identification-of-novel-circulatory-microrna-signatures-linked-to-ZXVDT2rNAa
Publisher
Oxford University Press
Copyright
© The Author(s) 2018. Published by Oxford University Press. All rights reserved. For permissions, please email: journals.permissions@oup.com
ISSN
0964-6906
eISSN
1460-2083
D.O.I.
10.1093/hmg/ddy136
Publisher site
See Article on Publisher Site

Abstract

Abstract MicroRNAs (miRNAs) are involved in growth, development, and occurrence and progression of many diseases. MiRNA-mediated post-transcriptional regulation is poorly understood in vascular biology and pathology. The purpose of this is to determine circulatory miRNAs as early detectable peripheral biomarkers in patients with ischemic stroke (IS). MiRNAs expression levels were measured in IS serum samples and healthy controls using Illumina deep sequencing analysis and identified differentially expressed miRNAs. Differentially expressed miRNAs were further validated using SYBR-green-based quantitative real-time PCR (qRT-PCR) assay in postmortem IS brains, lymphoblastoid IS cell lines, oxygen and glucose deprivation/reoxygenation -treated human and mouse neuroblastoma cells, and mouse models of hypoxia and ischemia (HI)-induced stroke. A total of 4656 miRNAs were differentially expressed in IS serum samples relative to healthy controls. Out of 4656 miRNAs, 272 were found to be significantly deregulated in IS patients. Interestingly, we found several novel and previously unreported miRNAs in IS patients relative to healthy controls. Further analyses revealed that some candidate miRNAs and its target genes were involved in the regulation of the stroke. To the best of our knowledge, this is the first study identified potential novel candidate miRNAs in IS serum samples from the residents of rural West Texas. MiRNAs identified in this study could potentially be used as a biomarker and the development of novel therapeutic approaches for stroke. Further studies are necessary to better understand miRNAs-regulated stroke cellular changes. Introduction Stroke is a common neurological disease with diverse etiologies that occurs when the blood supply to the brain is interrupted, resulting in a shortage of oxygen and nutrients to brain tissue. Due to multifactorial nature, stroke may be classified as a syndrome, not as a single disease. Stroke is the second leading cause of death globally and third leading cause of disability-adjusted life years worldwide (1,2). An estimated 7.2 million Americans ≥20 years of age self-report having had a stroke and approximately 795 000 strokes occur in the United States every year. On an average, every 40 s, someone in the United States has a stroke, and on an average, every 4 min, someone dies of stroke. Prevalence of stroke in the United States increases with age in both men and women (3). Ischemic stroke (IS) is described as a lack of blood supply and oxygen availability to an area of the brain due to narrowed or blocked arteries leading to or within the brain and the most predominant type of stroke accounting for approximately 87% of stroke cases (4). Stroke doubles the risk for dementia (post-stroke dementia), and approximately 30% of stroke patients go on to develop cognitive dysfunction within 3 years (5,6). Biomarkers might be useful in identifying different diseases, such as stroke, cancer, diabetes and disease severity (7,8). Identification of biomarkers could inform researchers in their attempts to develop early detectable peripheral biomarkers and could contribute to a better understanding of the etiologies and mechanisms underlying particular diseases, such as stroke. Recent molecular biology discoveries have revealed that microRNAs (miRNAs) can detect changes in the bodily organs, including brain that may lead to IS. MiRNAs are important post-transcriptional regulators that connect with multiple target messenger RNAs coordinately regulating target genes. MiRNAs have also been found to be important regulators of leukocyte gene expression in acute IS cases (9). Many studies showed that miRNAs altered after central nervous system injury moderate processes that stimulate neuronal death with inflammation, apoptosis and oxidative stress (10,11). Furthermore, miRNAs can act as sensitive biomarkers of secondary brain damage. Studies also suggested that peripheral blood miRNAs and their profiles could be developed as diagnostic and prognostic biomarkers of IS, as well as serving as innovative targets in the treatment of this disease (12). Clinical approaches accessible for the diagnosis and prognosis of stroke were restricted to radiological imaging, which was with limited availability and higher cost. Diagnosis of early stage of stroke and its development could be improved through the finding of new biomarkers. MiRNA-mediated post-transcriptional regulation is poorly understood in vascular biology and pathology. Currently, there are no drugs/agents and peripheral biomarkers available that can delay and/or detect IS in humans. Hence, identification of blood-based early detectable miRNAs could contribute to a better understanding of the etiologies and mechanisms underlying IS. In the present study, we sought to determine miRNAs as early detectable biomarkers in serum samples from IS patients relative to healthy controls. We used miRNA deep sequencing method and validated differentially expressed miRNAs using quantitative real-time PCR (qRT-PCR). Further, we validated the trend of selected miRNAs using postmortem IS brains, lymphoblastoid IS cell lines, oxygen and glucose deprivation/reoxygenation (OGD/R)-treated human (SH-SY5Y) and mouse neuroblastoma (N2a) cells and hypoxia and ischemia (HI)-induced stroke mouse model. Results Differentially expressed miRNA profile by deep sequencing Illumina deep sequencing analysis of serum samples provided a total of 484 651 777 raw RNA reads. Among these, 341 678 616 (70.5%) were mapped to miRNAs, and 39 890 853 reads were mapped to mRNA and 24 723 087 reads were mapped to other RNAs (RFam: rRNA, tRNA, snRNA, snoRNA and others) (Supplementary Material, Fig. S1). Based on the size distribution of all known miRNAs, 15–32 nucleotide (nt) reads were selected as ‘mappable reads’ for further analysis (Supplementary Material, Fig. S2). Of these reads, 87.6% of the small RNAs were 17–22 nt in size, which is typical miRNA sizes produced by RNA Dicer-digested products. The mappable reads sequences were subjected to advance bioinformatics analysis and to simplify the data from sequencing, all identical sequence reads were grouped and then assigned a unique sequence tag (Supplementary Material, Table S1). Our miRNA sequencing analysis revealed/detected a total of 4656 miRNAs in serum samples of IS patients versus healthy controls. Among them, 272 miRNAs were differentially deregulated (FC ±2, P ≤ 0.05) in IS patients, compared with healthy controls. Interestingly, 173 miRNAs were significantly upregulated, while 76 were found to be significantly down-regulated in IS patients. Hierarchical clustering performed with differentially expressed miRNAs, revealed that miRNA expression patterns were able to classify individuals according to their disease status. Among these miRNAs, we chose 16 miRNAs (Supplementary Material, Table S2) that were differentially expressed between the IS patients and healthy controls and where number of reads 10 in either IS patients or healthy controls were detected and at least a meaningful ±2-fold change between the group was identified (Fig. 1). Figure 1. View largeDownload slide Heat map of the hierarchical cluster analysis of differentially expressed miRNAs between IS patients and healthy controls detected by deep Sequencing. The color indicates the log 2-fold change from high (red) to low (green), as indicated by the color key. Figure 1. View largeDownload slide Heat map of the hierarchical cluster analysis of differentially expressed miRNAs between IS patients and healthy controls detected by deep Sequencing. The color indicates the log 2-fold change from high (red) to low (green), as indicated by the color key. Validation of candidate miRNAs in serum samples by real-time RT-PCR We validated 16 miRNAs using real-time RT-PCR analysis in same RNA samples that were used for deep sequencing analysis. A few known and several novel and previously unreported miRNAs were found in IS serum samples. Of the 16 miRNAs differentially expressed between IS patients and healthy controls in the discovery cohort, the validation studies found that four miRNAs [PC-3p-57664 (P = 0.01), PC-5p-12969 (P = 0.04), hsa-miR-122-5p (P = 0.01) and hsa-miR-211-5p (P = 0.03)] were significantly upregulated in IS patients compared with healthy controls. Whereas four miRNAs [hsa-miR-22–3p (P = 0.01), PC-3p-32463 (P = 0.0001), hsa-miR-30d-5p (P = 0.0009) and hsa-miR-23a-3p (P = 0.03)] were significantly down-regulated in the same comparison (Fig. 2A). Figure 2. View largeDownload slide (A) Validation of candidate miRNAs in serum samples by qRT-PCR. Significantly deregulated miRNA expression in IS versus the healthy controls. The y-axis depicts lnΔCq. P-values were determined by Mann-Whitney test. (B) Validation of serum miRNAs using postmortem IS brains by qRT-PCR. Box plots of lnΔCq values of significant serum miRNAs in IS brains compared to healthy control brains. Figure 2. View largeDownload slide (A) Validation of candidate miRNAs in serum samples by qRT-PCR. Significantly deregulated miRNA expression in IS versus the healthy controls. The y-axis depicts lnΔCq. P-values were determined by Mann-Whitney test. (B) Validation of serum miRNAs using postmortem IS brains by qRT-PCR. Box plots of lnΔCq values of significant serum miRNAs in IS brains compared to healthy control brains. Validation of serum miRNAs using postmortem IS brains We analyzed the expression of above-selected 16 miRNAs in the postmortem IS brains (n = 10) and control brains (n = 10) by real-time RT-PCR (Supplementary Material, Table S3). Analysis showed that four miRNAs [PC-3p-57664 (P = 0.04), PC-5p-12969 (P = 0.006), hsa-miR-122-5p (P < 0.0001) and hsa-miR-211-5p (P < 0.0001)] were consistently upregulated and three miRNAs [PC-3p-32463 (P = 0.01), hsa-miR-30d-5p (P = 0.01) and hsa-miR-23a-3p (P = 0.03)] were significantly down-regulated in the IS brains compared with control brains (Fig. 2B). The expression of PC-3p-57664, PC-5p-12969, hsa-miR-122–5p and hsa-miR-211-5p were the most significantly upregulated in both the IS serum and postmortem IS brains, suggesting that these upregulated miRNAs are relevant to IS—in terms of early detection and disease progression. Validation of serum miRNAs using lymphoblastoid IS cell lines To further validate the miRNA sequencing data, expression of 16 miRNAs were measured in lymphoblastoid cell line (LCL) strains by real-time RT-PCR (Supplementary Material, Table S4). Seven miRNAs [mmu-mir-6240-p5 (P = 0.007), ggo-miR-139 (P = 0.002), hsa-mir-760 (P = 0.001), PC-3p-57664 (P = 0.0009), PC-5p-12969 (P = 0.02), hsa-miR-122-5p (P = 0.03) and hsa-miR-211-5p (P < 0.0001)] were upregulated in IS LCL compared with control LCL strains. PC-3p-32463 (P = 0.01) and hsa-miR-30d-5p (P = 0.01) were significantly down-regulated in the IS LCL strains (Fig. 2C). These results further confirmed the significant response of these four miRNAs (PC-3p-57664, PC-5p-12969, hsa-miR-122-5p and hsa-miR-211-5p) in IS pathogenesis. Candidate miRNAs expression in OGD-treated cells Human neuroblastoma cells (SH-SY5Y) To determine the involvement of candidate miRNA expression in hypoxic-ischemic-induced neuronal death, OGD-stimulated human neuroblastoma cells was monitored. We selected 16 miRNAs for analysis based on a number of factors including, their expression levels in IS serum, postmortem IS brains and IS LCL strains. PC-5p-211 (P = 0.006), ggo-miR-139 (P = 0.001), hsa-mir-760 (P = 0.02), hsa-miR-96 (P = 0.0007), hsa-miR-99a-5p (P = 0.0004), PC-3p-57664 (P = 0.01), PC-5p-12969 (P = 0.01), hsa-miR-122–5p (P = 0.0006), hsa-miR-211–5p (P = 0.001) were increased significantly in human neuroblastoma cells following OGD/R exposure compared with control cells (Fig. 3A). Mmu-miR-5124a (P = 0.03), PC-3p-32463 (P = 0.0003) were significantly down-regulated in the OGD-treated cells. Figure 3. View largeDownload slide (A) MicroRNAs expression in OGD/R-treated human neuroblastoma cells (SH-SY5Y) by qRT-PCR. Data are presented as the mean ± SD of three independent experiments. (B) MicroRNAs expression in OGD/R-treated mouse neuroblastoma cells (N2a) by qRT-PCR. Figure 3. View largeDownload slide (A) MicroRNAs expression in OGD/R-treated human neuroblastoma cells (SH-SY5Y) by qRT-PCR. Data are presented as the mean ± SD of three independent experiments. (B) MicroRNAs expression in OGD/R-treated mouse neuroblastoma cells (N2a) by qRT-PCR. Mouse neuroblastoma (N2a) cells We further evaluated the miRNA expression profiles of the OGD/R-activated N2a cells. To test the hypothesis, we selected 12 miRNAs. In these, nine miRNAs that exhibited significantly altered expression levels between the hypoxic and normoxic conditions. PC-3p-57664 (P = 0.04), PC-5p-12969 (P = 0.005), mmu-miR-122-5p (P = 0.002) and mmu-miR-211-5p (P = 0.04) were upregulated significantly in mouse neuroblastoma cells following OGD exposure compared with normoxia-treated cells (Fig. 3B). Mmu-miR-5124a (P = 0.01) and PC-3p-32463 (P = 0.002) were significantly down-regulated in the OGD-treated cells. Differential expression of miRNAs in the brain of hypoxia and ischemia (HI)-induced neonatal mice To verify the accuracy of miRNA sequencing results, we selected 11 miRNAs for further validation in brains of HI-induced mouse models. We studied four different brain regions, including hippocampus, striatum, cerebral cortex and cerebellum from HI-induced neonatal and control, naive mice. Out of 11 miRNAs, the following five miRNAs [mmu-miR-211-5p (P = 0.003), PC-5p-211 (P = 0.02), PC-3p-57664 (P = 0.02), PC-5p-12969 (P = 0.0005) and mmu-miR-122-5p (P = 0.002)] were significantly upregulated in the hippocampus of HI mice when compared with that of naïve control mice (Fig. 4). Details of 11 miRNA expressions in other brain regions are given in Supplementary Material, Figure S3A–C. Figure 4. View largeDownload slide Quantitative RT-PCR analysis of miRNAs in hippocampus region of stroke hypoxia ischemia model. Fold change was calculated by 2-ΔΔCT method. Significant difference among groups were calculated by paired t-test with two-tailed P < 0.05 is considered significant. Figure 4. View largeDownload slide Quantitative RT-PCR analysis of miRNAs in hippocampus region of stroke hypoxia ischemia model. Fold change was calculated by 2-ΔΔCT method. Significant difference among groups were calculated by paired t-test with two-tailed P < 0.05 is considered significant. Receiver operating characteristics (ROC) curve analysis Expression of four miRNAs (PC-3p-57664, PC-5p-12969, hsa-miR-122-5p and hsa-miR-211–5p) consistently upregulated throughout the validation analysis. Therefore, we evaluated the diagnostic value of these four miRNAs by plotting ROC curve in IS serum, postmortem IS brains, LCL strains and HI stroke mouse models. The curves were plotted based on the ΔCt value of candidate miRNAs expression in different sources. Upon analysis, PC-3p-57664 (AUROC = 0.76; 95% CI: 0.571–0.953; P = 0.01), PC-5p-12969 (AUROC = 0.80; 95% CI: 0.6053–0.996; P = 0.006), hsa-miR-122–5p (AUROC = 0.72; 95% CI: 0.569–0.874; P = 0.03) and hsa-miR-211–5p (AUROC = 0.72; 95% CI: 0.533–0.919; P = 0.04) showed significant area under curve in IS serum samples compared with the healthy controls. The same trend was observed in postmortem IS brains, IS LCL strains as well in the HI stroke mice (Fig. 5). Thus, ROC analysis confirmed that the profile of the four serum miRNAs could be a simple, specific and non-invasive molecular biomarker for diagnosing IS. Figure 5. View largeDownload slide ROC curve analysis of serum miRNAs as diagnostic biomarkers differentiating IS patients from healthy controls. (A) Serum, (B) postmortem IS brains and (C) IS lymphoblastoid IS cell lines. (D) HI stroke mouse model (hippocampus). Figure 5. View largeDownload slide ROC curve analysis of serum miRNAs as diagnostic biomarkers differentiating IS patients from healthy controls. (A) Serum, (B) postmortem IS brains and (C) IS lymphoblastoid IS cell lines. (D) HI stroke mouse model (hippocampus). Discussion The overall objective of our study was to identify early detectable peripheral biomarkers for IS in the residents of rural West Texas. MiRNAs have been identified as circulating biomarkers in several diseases, including IS (13–15). Our Illumina deep sequencing and further validation analysis revealed that 16 circulating miRNAs that distinguishes between IS patients and healthy controls. Of the 16 miRNAs differentially expressed between IS patients and healthy controls, 12 miRNAs are previously reported in stroke and other diseases and four miRNAs are novel and unreported miRNAs. As of now, no study is reported to validate the expression of these miRNAs in post-mortem IS brains, IS LCL strains, OGD/R-treated human and mouse neuroblastoma cells and HI stroke mice. For the first time, we identified the novel potential candidate miRNAs (PC-3p-57664, PC-5p-12969) in IS serum samples from the residents of rural West Texas and verified their expression in mice as well. MiRNAs are cell specific, interestingly these novel candidates were consistently upregulated in all stroke sources and showed a strong correlation with stroke pathology. Hence, these miRNAs might provide the possibilities of unique biomarker candidate for stroke. Chen and Zhang identified the variant rs2507800 in the 3′-untranslated region of angiopoietin-1 that might reduce the risk of stroke by interfering with hsa-miR-211 binding site (16). Interestingly in our current study, hsa-miR-211-5p was upregulated in IS patients. Hsa-miR-122 was identified to be related to human stroke based on the Human MicroRNA Disease Database (17). Another study investigated miRNA expression profile and found that miR-122 was down-regulated (18). Jickling et al. identified that miR-122 were decreased in acute IS patients compared with controls (9). Hsa-miR-23a and hsa-miR-22 were significantly down-regulated in stroke patients (19). MiR-23a levels differed in male and female ischemic brains, providing evidence for sex-specific miRNA expression in stroke (20). Hypertension is a well-established risk factor for stroke. Several studies showed that miRNAs were known to impact the state of hypertension directly or indirectly. In another research, miR-30d was down-regulated known to be involved in hypertension (21). Interestingly, in our study also miR-30d was down-regulated, which clearly meant there was a link between stroke and hypertension. A study by Long et al. identified that circulating miR-30a was markedly down-regulated in all patients with IS until 24 weeks (22). Postmortem human brain tissue was being used for quantifying cellular and molecular markers of neural courses with the area of improved understanding the variations in the brain caused by neurological diseases (23). However, the miRNA expression levels and molecular characterizations were not investigated using postmortem IS brains. To the best of our knowledge, our study was first to validate the miRNAs using postmortem IS brain specimens. MiRNAs PC-3p-57664, PC-5p-12969, hsa-miR-122-5p and hsa-miR-211-5p were consistently upregulated and PC-3p-32463, PC-5p-211, ggo-miR-139, hsa-miR-30d-5p, mmu-mir-6240-p5 and hsa-miR-23a-3p were significantly down-regulated in the IS brains. A recent study reported that a decrease of brain miRNA-122 level was deleterious and could be considered as an early marker of stroke in the stroke-prone spontaneously hypertensive rat (24). Elevating miR-122 improves stroke outcomes and this occurred via down-regulating miR-122 target genes in blood leukocytes (25). Down-regulation of miRNA-30a improves ischemic injury through enhancing beclin 1-mediated autophagy in N2a cells and cultured cortical neurons after OGD, and mouse brain with MCAO-induced IS (26). LCLs are the biological resources that have been used in various research fields related to human genetics, pharmacogenomics and immunology (27,28). LCLs have the potential to disclose at least a subset of brain-related miRNAs implicated in stroke. Hypoxia induces time-dependent alteration of the expression levels of miRNAs suggesting their involvement in the cellular response to ischemic injury (29). In this study, we performed miRNA expression in IS LCL strains and OGD/R on human and mouse neuroblastoma cells to mimic ischemia in vitro. To our knowledge, this is the first study to examine the roles of miRNA expression variations in IS LCLs. MiRNAs have essential roles in brain function, including neurogenesis, neural development and cellular responses leading to changes in synaptic plasticity. They are also implicated in neurodegeneration and neurological disorders, in responses to HI, and in ischemic tolerance induced by ischemic pre-conditioning (30). Expression levels of few miRNAs could be differently modulated in both in vivo and in vitro experimental models (25,31). We assessed the expressions of 11 miRNAs using a HI-induced in post-natal day nine (P9) C57BL/6J mice. Hippocampal region of the HI-induced neonatal mouse brain showed the most consistent differential expression of miRNA compared with other regions. A recent global expression of miRNAs in a P10 rat model of cerebral HI found that miR-30d-5p was one of the most deregulated miRNAs in neonatal brains in response to HI. Collectively, these results indicated that miR-30d-5p modulated survival programs of neural cell by regulating autophagy and apoptosis (32). It is possible that miRNAs identified in this study may have implications for both consequences and risk factors of stroke. In this study, we investigated serum samples from IS patients and we strongly feel that differentially expressed miRNAs are consequences of disease process, and these differentially miRNAs can also be used for the development of novel therapeutic targets for IS. However, further studies are necessary to better understand miRNAs-regulated stroke not only for risk but also for consequences of stroke. Based on our findings, we can only say that observed miRNAs in IS patients are different from healthy controls. In summary, miRNA sequencing analysis of IS serum samples showed significant deregulation of 16 miRNAs. Among 16 miRNAs, four miRNAs (PC-3p-57664, PC-5p-12969, hsa/mmu-miR-122–5p and hsa/mmu-miR-211–5p) were almost consistently upregulated in human IS serum samples, human post-mortem IS brain specimens, human lymphoblastoid IS cell lines, OGD/R-treated human and mouse neuroblastoma cells and HI stroke mouse models. ROC curve analysis in serum and postmortem brain also confirmed their diagnostic potential for stroke. Further, GO and KEGG pathway analysis showed the regulation of many stroke-related genes and pathways by these miRNAs. Based on intense analysis, we conclude that circulatory levels of PC-3p-57664, PC-5p-12969, miR-122-5p, miR-211-5p might be the potential biomarkers for the diagnosis of IS. Materials and Methods Enrollment of study samples For this, 34 IS patients (13 males, 21 females: mean age of 62.88 ± 11.94 years) and 11 healthy controls (5 males, 6 females: mean age: 62.63 ± 6.6 years) were used as the study group. Sera samples were collected from patients and healthy controls under Facing Rural Obstacles to healthcare Now Through Intervention, Education & Research (FRONTIER) project based at Garrison Institute on Aging (GIA), Texas Tech University Health Sciences Center. The Institutional Review Board (IRB) protocol was approved for Project FRONTIER (IRB#L06–028). All the bio-specimens were stored at the GIA. Information on demographic characteristics, medical history, biochemical profile and established risk factors were recorded by using a standardized questionnaire (Supplementary Material, Table S5). RNA extraction, small RNA library construction RNA was isolated from 1.5 ml of serum using Plasma/Serum RNA purification Midi Kit as per manufacturer’s instructions (Cat. No: 56100; Norgen Biotek Corp., Thorold, ON, Canada). All RNA samples were processed and analyzed by LC Sciences (Houston, TX). The quality and quantity of the RNA samples were tested using an Agilent 2100 Bioanalyzer (Agilent). A small RNA library was generated using the Illumina Truseq™ Small RNA Preparation kit according to Illumina’s TruSeq™ Small RNA Sample Preparation Guide [(15004197 C), Illumina Inc., Part # 1004239 Rev. A, 2008; Cat. No. RS-930-1012, Part No. 15004197 Rev. B, January 2011]. Primary screening by deep sequencing and data analysis The purified cDNA library was used for cluster generation on Illumina’s Cluster Station and then sequenced on Illumina GAIIx following vendor’s instruction for running the instrument. Raw sequencing reads (40 nts) were obtained using Illumina’s Sequencing Control Studio software version 2.8 (SCS v2.8) following real-time sequencing image analysis and base-calling by Illumina’s Real-Time Analysis version 1.8.70 (RTA v1.8.70). The extracted sequencing reads were stored and then a proprietary pipeline script, ACGT101-miR v4.2 (LC Sciences), was used for sequencing data analysis. After the raw sequence reads, or sequenced sequences (sequ seqs) were extracted from image data, a series of digital filters (LC Sciences) were employed to remove various un-mappable sequencing reads. Impurity sequences due to sample preparation, sequencing chemistry and processes, and the optical digital resolution of the sequencer detector were also removed. Remaining sequ seqs with lengths between 15 and 32 bases were grouped by families (unique seqs), and were used to map with the reference database files. Various ‘mappings’ were performed on unique seqs against pre-miRNA (mir) and mature miRNA (miR) sequences listed in the latest release of miRbase (v21.0; ftp://mirbase.org/pub/mirbase/CURRENT/; specific species: hsa; selected species: ggo, ppa, ptr, ppy, ssy, age, lla, sla, pbi, mml, mne, lca, cgr, mmu, rno, cfa, ocu, efu, aja, eca, mdo, sha, meu, oan, bta, chi, oar, tch, ssc) (33–35) or genome based on the public releases of appropriate species (V37.1; ftp://ftp.ncbi.nih.gov/genomes/H sapiens/). Mappings were also done on mirs of interest against genome sequence. Mappable unique seqs were mapped to other defined databases, such as mRNA, RFam and Repbase (V37.1; ftp://ftp.ncbi.nih.gov/genomes/H sapiens/RNA/). Methods and criteria used for various mappings were documented in the ACGT-101 User’s Manual. Sequences were mapped against reported miRNA, species’ genomes, and other RNA databases (e.g. RFam, repase, mRNA) and were classified as follows (Supplementary Material, Fig. S4): Mappable reads mapped to selected mirs in miRbase Mirs mapped to species specific genome (Homo sapiens) Mirs are of specific species (Homo sapiens) (group1a) Mirs are of selected species (Mammalia) (group1b) Reads mapped to other locations too and Reads mapped only to the same locations in the genome as that of mirs (group 1c) Mirs un-mapped to species specific genome Reads mapped/un-mapped to species specific genome Extended sequences potentially form hairpins (group 2a) Extended sequences potentially cannot form hairpins (group 2b) Reads mapped to miRs of selected species (group 3a) Reads unmapped to miRs of selected species (group 3b) Mappable reads un-mapped to selected mirs in miRbase Reads un-mapped to mRNA, Rfam and repbase Reads mapped to species specific genome Extended sequences potentially form hairpins (group 4a) Extended sequences potentially cannot form hairpins (group 4b) Reads un-mapped to species specific genome (no hit) Reads mapped to mRNA, Rfam, or repbase (others) Validation of differently expressed serum miRNAs using quantitative real-time RT-PCR To support the data obtained from the deep sequencing results, qRT-PCR analysis was performed to validate further. One µG of total RNA was reverse transcribed using miRNA First-Strand cDNA synthesis kit (Agilent Technologies Inc., CA), following manufacturer’s instructions. Resulting cDNAs were diluted with 20 μl of RNase-free water and stored at −80°C for further analysis (36). Primers for 16 miRNAs were synthesized commercially (Integrated DNA Technologies, Inc., Iowa) (Supplementary Material, Table S2). U6, one of the uniformly expressed small RNAs, was used as the internal control for real-time RT-PCR. Briefly, 1 μl of miRNA-specific forward primer (10 μM), 1 μL of a universal reverse primer (3.125 μM) (Agilent Technologies Inc.), 10 μl of 2× SYBR® Green PCR master mix (Applied Biosystems, NY) and 1 μl of cDNA were mixed. To this mixture RNase-free water was added up to a 20 μl of final volume. The reactions were amplified for 5 min at 95°C, followed by 40 cycles of 95°C for 10 s, 60°C for 15 s and 72°C for 25 s at 7900HT Fast Real Time PCR System (Applied Biosystems). All reactions were performed in triplicate, and the controls (no template and no RT) were included for each gene, and the data were expressed as the mean ±SD. Fold change of each miRNAs were calculated as described previously (37). The threshold cycle (CT) values were automatically determined by the instrument, and the fold change of each miRNAs were calculated using the following equation: the formula (2-ΔΔCt), where ΔCt was calculated by subtracting Ct of U6snRNA from the Ct of particular miRNAs target, and ΔΔCt value was obtained by subtracting ΔCt of particular miRNAs target in the controls from the ΔCt of miRNAs target in the IS. Postmortem brains from stroke patients and controls In the present study, 20 postmortem brain samples were investigated which consisted of 10 IS [5 males and 5 females: Age ranged from 57–91 years (78.3 ± 11.89) and postmortem interval (PMI) varied from 4–23.9 h (average 16.48 h)] and 10 normal control subjects [5 males and 5 females: age ranged from 67 to 91 years (76.9 ± 8.62) and PMI varied from 11.8 to 25 h (average 17.99 h)] were obtained from the Human Brain and Spinal Fluid Resource Center (Los Angeles, CA) and Harvard Brain Tissue Resource Center (HBTRC) through NIH NeuroBiobank. These brain banks were responsible for obtaining subject consent and the unidentifiable coding of subject information. The study protocol was approved by the Institute Ethical Committee at TTUHSC (IBC protocol number: 14013). Lymphoblastoid cell lines Epstein-Barr Virus (EBV) transformed LCLs from 20 IS patients (10 males, 10 females: mean age of 66.9 ± 6.13 years) and 10 unrelated healthy subjects (5 males, 5 females: mean age of 62.8 ± 4.91 years) were obtained from the Coriell Cell Repository. These samples had been collected and anonymized by National Institute of Neurological Disorders and Stroke (NINDS), and all subjects had provided written consent for their experimental use. LCLs were cultured in Roswell Park Memorial Institute Medium 1640 (RPMI 1640 medium) with 2 mM l-glutamine (Gibco, Carlsbad, CA, #11875) supplemented with 15% heat-inactivated fetal bovine serum (FBS). Induction of neonatal hypoxia and ischemia HI was induced in post-natal day nine (P9) C57BL/6J mice. The pups were anesthetized with isoflurane (Butler Schein Animal Health Supply, Reno, NV) (5% for induction, 2–3% for maintenance) in 30% oxygen mixed with nitrous oxide. The body temperature of the pups were maintained at 36°C using a heated surgical table (Molecular Imaging Products, Bend, OR). Under a surgical microscope (Nikon SMZ-800 Zoom Stereo, Nikon, Melville, NY), a midline skin incision was made and the trachea was visualized through the muscle overlying it. The left common carotid artery was freed from the left common jugular vein and left vagus nerve by blunt dissection, electrically cauterized and cut. The incision was injected with 0.5% bupivacaine and closed with a single 6.0 silk suture. Animals were returned to their dams and monitored continuously for a 2 h recovery period. To induce unilateral ischemic injury, the animals were placed in a hypoxia chamber (BioSpherix Ltd, Redfield, NY) equilibrated with 10% O2 and 90% N2 at 36°C for 50 min. This is a well-characterized model of neonatal HI and results in reproducible brain injury ipsilateral (IL) to the electrocauterized left common carotid artery (38–41). Naïve mice were not exposed to HI or any surgical intervention. Mice were decapitated for collection of fresh tissue of the regions of striatum, hippocampus, cortex and cerebellum on day 3 (P12) post-HI (42). After extraction, the fresh tissues were kept in −20°C for 1 day and in −80°C for long-term storage. All procedures on animals were carried out in adherence with NIH Guide for the Care and use of Laboratory Animals and approved by the Institutional Animal Care and Use Committee at the University of Wisconsin-Madison. Cell cultures The human neuroblastoma cell line SH-SY-5Y and mouse neuroblastoma cell line N2a were grown in Dulbecco’s modified Eagle’s medium/F12 (1/1) (DMEM/F12, Gibco Life Technologies, Carlsbad, CA, #11320), minimum essential media respectively containing 10% FBS, penicillin (100 U/ml) and streptomycin (100 µG/ml). The cultures were maintained at 37°C in 95% air, 5% CO2 in a humidified incubator. After the cells were seeded, they were allowed to grow for 24–48 h or until 80% confluence in 6-well plates then they were used for experimentation. Oxygen and glucose deprivation/reoxygenation protocol OGD/R is a well-established in vitro model to study the pathology and pharmacology of ischemic damage, OGD/R was achieved using methods published earlier (43–46). Culture media were replaced with deoxygenated, no glucose DMEM (Gibco, Carlsbad, CA, #11966) and placed in hypoxia chamber (Biospherix, ProOx model 110) with 95% N2 and 5% CO2 for 4 h at 37°C to represent OGD condition. After the appropriate time, the culture plates were removed from the hypoxia chamber. Prior to re-oxygenation, cells were washed in PBS and the medium was replaced with complete culture medium then placed in a humidified incubator at 37°C for 20–24 h to represent reoxygenation. MiRNAs extraction from tissues and cell pellets and qRT-PCR Total RNA was isolated from the 80 mg of tissues and cell pellets using the TriZol RT reagent (Ambion, USA) as per manufacturer’s instructions. MiRNAs extraction and cDNA synthesis were followed as described earlier (36). The quality and quantity of the RNA were analyzed by NanoDrop analysis. The value of absorbance of each RNA sample (A260/A280) was 1.8–2.0. cDNA was synthesized from 1 μG of RNA using miRNA First-Strand cDNA synthesis kit (Agilent Technologies Inc.). Target gene prediction and enrichment analysis MiRNAs of interest were selected based on the statistical significance, fold change difference and biological rationales. Target gene prediction was performed by using TargetScan 6.0 (www.targetscan.org). For human, the program searches was run to match the miRNA seed regions and orthologous at 3' UTRs of human genes (47). Gene function enrichment and biological pathway analysis were performed using DAVID online tool suite (https://david.ncifcrf.gov/) (48). Validated and predicted target genes were uploaded at Gene List Manager according the tool instructions. Gene function enrichment and pathway analysis results were obtained and given in Supplementary Material, Figure S5A–C. Statistical Analysis Data were presented as means ± SD for other variables. The qRT-PCR validation analysis was based on the 2-ΔΔCT value of genes in each sample from IS and healthy controls. The Cq values of miRNAs were displayed as LnΔCq for statistical analysis (49). P-value was calculated, based on the paired and unpaired t-tests for analyzing two groups. MiRNAs levels between the probable IS and healthy controls were analyzed using a two-sided nonparametric Mann-Whitney test. Sensitivity and specificity of measured variable for IS biomarker were examined using a ROC curve analysis under a nonparametric approach. P-value < 0.05 was considered to be statistically significant. All analyses were performed by GraphPad Prism (version 6.0; GraphPad Software, La Zolla, CA). Supplementary Material Supplementary Material is available at HMG online. Acknowledgements We sincerely thank Ms. Annette Boles, Ms. Kathy Hudson and all field coordinators for their support in collecting data and blood samples for this study. We also thank all the participants of Project FRONTIER for providing blood specimens. Post-mortem brain tissues were obtained from the Human Brain and Spinal Fluid Resource Center, Los Angeles and Harvard Brain Tissue Resource Center (HHSN-271–2013-00030C). We are thankful to the Coriell Institute of Medical Research for providing IS and control LCLs. Conflict of Interest statement. A provisional patent is in progress with the scientific content of our paper. Funding The research presented in this article was supported by the National Institute of Health grants AG042178, AG47812 and NS105473, and the Garrison Family Foundation and CH Foundation (PHR), NIH-NINDS K08 NS088563 (PC) and NIH P30 HD03352 (Waisman Center). References 1 Lozano R., Naghavi M., Foreman K., Lim S., Shibuya K., Aboyans V., Abraham J., Adair T., Aggarwal R., Ahn S.Y. et al.   ( 2012) Global and regional mortality from 235 causes of death for 20 age groups in 1990 and 2010: a systematic analysis for the Global Burden of Disease Study 2010. Lancet , 380, 2095– 2128. Google Scholar CrossRef Search ADS PubMed  2 Murray C.J., Vos T., Lozano R., Naghavi M., Flaxman A.D., Michaud C., Ezzati M., Shibuya K., Salomon J.A., Abdalla S. et al.   ( 2012) Disability-adjusted life years (DALYs) for 291 diseases and injuries in 21 regions, 1990–2010: a systematic analysis for the Global Burden of Disease Study 2010. Lancet , 380, 2197– 2223. Google Scholar CrossRef Search ADS PubMed  3 Benjamin E.J., Blaha M.J., Chiuve S.E., Cushman M., Das S.R., Deo R., de Ferranti S.D., Floyd J., Fornage M., Gillespie C. et al.   ( 2017) American Heart Association Statistics Committee and Stroke Statistics Subcommittee. Heart disease and stroke statistics-2017 update: a report from the American Heart Association. Circulation , 135, e146– e603. Google Scholar CrossRef Search ADS PubMed  4 Rosamond W., Flegal K., Furie K., Go A., Greenlund K., Haase N., Hailpern S.M., Ho M., Howard V., Kissela B. et al.   ( 2008) American Heart Association Statistics Committee and Stroke Statistics Subcommittee. Heart disease and stroke statistics–2008 update: a report from the American Heart Association Statistics Committee and Stroke Statistics Subcommittee. Circulation , 117, e25– e146. Google Scholar CrossRef Search ADS PubMed  5 Pendlebury S.T., Rothwell P.M. ( 2009) Prevalence, incidence, and factors associated with pre-stroke and post-stroke dementia: a systematic review and meta-analysis. Lancet Neurol ., 8, 1006– 1018. Google Scholar CrossRef Search ADS PubMed  6 Allan L.M., Rowan E.N., Firbank M.J., Thomas A.J., Parry S.W., Polvikoski T.M., O'Brien J.T., Kalaria R.N. ( 2011) Long term incidence of dementia, predictors of mortality and pathological diagnosis in older stroke survivors. Brain , 134, 3716– 3727. Google Scholar CrossRef Search ADS PubMed  7 Schwarzenbach H., Nishida N., Calin G.A., Pantel K. ( 2014) Clinical relevance of circulating cell-free microRNAs in cancer. Nat. Rev. Clin. Oncol ., 11, 145– 156. Google Scholar CrossRef Search ADS PubMed  8 Jung H.J., Suh Y. ( 2014) Circulating miRNAs in aging and age related disease. J. Genet. Genom ., 41, 465– 472. Google Scholar CrossRef Search ADS   9 Jickling G.C., Ander B.P., Zhan X., Noblett D., Stamova B., Liu D. ( 2014) MicroRNA expression in peripheral blood cells following acute ischemic stroke and their predicted gene targets. PLoS One , 9, e99283. Google Scholar CrossRef Search ADS PubMed  10 Vijayan M., Reddy P.H. ( 2016) Peripheral biomarkers of stroke: focus on circulatory microRNAs. Biochim. Biophys. Acta , 1862, 1984– 1993. Google Scholar CrossRef Search ADS PubMed  11 Nieto-Diaz M., Esteban F.J., Reigada D., Muñoz-Galdeano T., Yunta M., Caballero-López M., Navarro-Ruiz R., Del Águila A., Maza R.M. ( 2014) MicroRNA dysregulation in spinal cord injury: causes, consequences and therapeutics. Front. Cell. Neurosci ., 8, 53. Google Scholar CrossRef Search ADS PubMed  12 Ouyang Y.B., Stary C.M., Yang G.Y., Giffard R. ( 2013) MicroRNAs: innovative targets for cerebral ischemia and stroke. Curr. Drug. Targets , 14, 90. Google Scholar CrossRef Search ADS PubMed  13 Lorenzen J.M., Thum T. ( 2012) Circulating and urinary microRNAs in kidney disease. Clin. J. Am. Soc. Nephrol ., 7, 1528– 1533. Google Scholar CrossRef Search ADS PubMed  14 Fichtlscherer S., Zeiher A.M., Dimmeler S. ( 2011) Circulating microRNAs: biomarkers or mediators of cardiovascular diseases? Arterioscler. Thromb. Vasc. Biol ., 31, 2383– 2390. Google Scholar CrossRef Search ADS PubMed  15 Laterza O.F., Lim L., Garrett-Engele P.W., Vlasakova K., Muniappa N., Tanaka W.K., Johnson J.M., Sina J.F., Fare T.L., Sistare F.D., Glaab W.E. ( 2009) Plasma MicroRNAs as sensitive and specific biomarkers of tissue injury. Clin. Chem ., 55, 1977– 1983. Google Scholar CrossRef Search ADS PubMed  16 Cheng Y., Zhang C. ( 2010) MicroRNA-21 in cardiovascular disease. J. Cardiovasc. Transl. Res ., 3, 251– 255. Google Scholar CrossRef Search ADS PubMed  17 Yuan Y., Kang R., Yu Y., Liu J., Zhang Y., Shen C., Wang J., Wu P., Shen C., Wang Z. ( 2016) Crosstalk between miRNAs and their regulated genes network in stroke. Sci. Rep ., 6, 20429. Google Scholar CrossRef Search ADS PubMed  18 He W., Chen S., Chen X., Li S., Chen W. ( 2016) Bioinformatic analysis of potential microRNAs in ischemic stroke. J. Stroke Cerebrovasc. Dis ., 25, 1753– 1759. Google Scholar CrossRef Search ADS PubMed  19 Sepramaniam S., Tan J.R., Tan K.S., DeSilva D.A., Tavintharan S., Woon F.P., Wang C.W., Yong F.L., Karolina D.S., Kaur P. et al.   ( 2014) Circulating microRNAs as biomarkers of acute stroke. Int. J. Mol. Sci ., 15, 1418– 1432. Google Scholar CrossRef Search ADS PubMed  20 Siegel C., Li J., Liu F., Benashski S.E., McCullough L.D. ( 2011) miR-23a regulation of X-linked inhibitor of apoptosis (XIAP) contributes to sex differences in the response to cerebral ischemia. Proc. Natl. Acad. Sci. U.S.A ., 108, 11662– 11667. Google Scholar CrossRef Search ADS PubMed  21 Da Costa Martins P.A., De Windt L.J. ( 2012) MicroRNAs in control of cardiac hypertrophy. Cardiovasc. Res ., 93, 563– 572. Google Scholar CrossRef Search ADS PubMed  22 Long G., Wang F., Li H., Yin Z., Sandip C., Lou Y., Wang Y., Chen C., Wang D.W. ( 2013) Circulating miR-30a, miR-126 and let-7b as biomarker for ischemic stroke in humans. BMC Neurol ., 13, 178. Google Scholar CrossRef Search ADS PubMed  23 Schmitt A., Parlapani E., Bauer M., Heinsen H., Falkai P. ( 2008) Is brain banking of psychiatric cases valuable for neurobiological research? Clinics (Sao Paulo) , 63, 255– 266. Google Scholar CrossRef Search ADS PubMed  24 Stanzione R., Bianchi F., Cotugno M., Marchitti S., Forte M., Busceti C., Ryskalin L., Fornai F., Volpe M., Rubattu S. ( 2017) A decrease of brain microRNA-122 level is an early marker of cerebrovascular disease in the stroke-prone spontaneously hypertensive rat. Oxid. Med. Cell. Longev ., 2017, 1. Google Scholar CrossRef Search ADS   25 Liu da Z., Jickling G.C., Ander B.P., Hull H., Zhan X., Cox C., Shroff N., Dykstra-Aiello C., Stamova B., Sharp F.R. ( 2016) Elevating microRNA-122 in blood improves outcomes after temporary middle cerebral artery occlusion in rats. J. Cereb. Blood Flow Metab ., 36, 1374– 1383. Google Scholar CrossRef Search ADS PubMed  26 Wang W., Sun G., Zhang L., Shi L., Zeng Y. ( 2014) Circulating microRNAs as novel potential biomarkers for early diagnosis of acute stroke in humans. J. Stroke Cerebrovasc. Dis ., 23, 2607– 2613. Google Scholar CrossRef Search ADS PubMed  27 Lee J.E., Hong E.J., Nam H.Y., Kim J.W., Han B.G., Jeon J.P. ( 2011) MicroRNA signatures associated with immortalization of EBV-transformed lymphoblastoid cell lines and their clinical traits. Cell. Prolif ., 44, 59– 66. Google Scholar CrossRef Search ADS PubMed  28 Wheeler H.E., Dolan M.E. ( 2012) Lymphoblastoid cell lines in pharmacogenomic discovery and clinical translation. Pharmacogenomics , 13, 55– 70. Google Scholar CrossRef Search ADS PubMed  29 Ziu M., Fletcher L., Rana S., Jimenez D.F., Digicaylioglu M. ( 2011) Temporal differences in microRNA expression patterns in astrocytes and neurons after ischemic injury. PLoS One , 6, e14724. Google Scholar CrossRef Search ADS PubMed  30 Martinez B., Peplow P.V. ( 2016) Blood microRNAs as potential diagnostic and prognostic markers in cerebral ischemic injury. Neural Regen. Res ., 11, 0– 1378. Google Scholar CrossRef Search ADS   31 Koutsis G., Siasos G., Spengos K. ( 2013) The emerging role of microRNA in stroke. Curr. Top. Med. Chem ., 13, 1573– 1588. Google Scholar CrossRef Search ADS PubMed  32 Zhao H., Li G., Ma Q., Tao Z., Wang R., Fan Z., Feng Y., Ji X., Luo Y. ( 2017) MicroRNA-99a-5p in circulating immune cells as a potential biomarker for the early diagnosis of ischemic stroke. Brain Circ ., 3, 21– 28. 33 Griffiths-Jones S., Saini H.K., van Dongen S., Enright A.J. ( 2007) MiRbase: tools for microRNA genomics. Nucleic Acids Res ., 36, D154– D158. Google Scholar CrossRef Search ADS PubMed  34 Griffiths-Jones S., Grocock R.J., van Dongen S., Bateman A., Enright A.J. ( 2006) MiRbase: microRNA sequences, targets and gene nomenclature. Nucleic Acids Res ., 34, D140– D144. Google Scholar CrossRef Search ADS PubMed  35 Griffiths-Jones S. ( 2004) The microRNA Registry. Nucleic Acids Res ., 32, 109D– D111. Google Scholar CrossRef Search ADS   36 Kumar S., Vijayan M., Reddy P.H. ( 2017) MicroRNA-455-3p as a potential peripheral biomarker for Alzheimer's disease. Hum. Mol. Genet ., 26, 3808– 3822. Google Scholar CrossRef Search ADS PubMed  37 Livak K.J., Schmittgen T.D. ( 2001) Analysis of relative gene expression data using real-time quantitative PCR and the 2#DDCT method. Methods , 25, 402– 408. Google Scholar CrossRef Search ADS PubMed  38 Vannucci R.C., Vannucci S.J. ( 1997) A model of perinatal hypoxic-ischemic brain damage. Ann. N.Y. Acad. Sci ., 835, 234– 249. Google Scholar CrossRef Search ADS PubMed  39 Cengiz P., Kleman N., Uluc K., Kendigelen P., Hagemann T., Akture E., Messing A., Ferrazzano P., Sun D. ( 2011) Inhibition of Na+/H+ exchanger isoform 1 is neuroprotective in neonatal hypoxic ischemic brain injury. Antioxid. Redox. Signal ., 14, 1803– 1813. Google Scholar CrossRef Search ADS PubMed  40 Uluc K., Kendigelen P., Fidan E., Zhang L., Chanana V., Kintner D., Akture E., Song C., Ye K., Sun D., Ferrazzano P. et al.   ( 2013) TrkB receptor agonist 7, 8 dihydroxyflavone triggers profound gender- dependent neuroprotection in mice after perinatal hypoxia and ischemia. CNS Neurol. Disord. Drug. Targets , 12, 360– 370. Google Scholar CrossRef Search ADS PubMed  41 Cikla U., Chanana V., Kintner D.B., Udho E., Eickhoff J., Sun W., Marquez S., Covert L., Otles A., Shapiro R.A. et al.   ( 2016) Signaling is required for TrkB-mediated hippocampal neuroprotection in female neonatal mice after hypoxic ischemic encephalopathy (1, 2, 3). eNeuro , 3, 1– 14. Google Scholar CrossRef Search ADS   42 Chanana V., Tumturk A., Kintner D., Udho E., Ferrazzano P., Cengiz P. ( 2016) Sex differences in mouse hippocampal astrocytes after in-vitro ischemia. J. Vis. Exp ., 116, 1– 9. 43 Yu Z., Liu J., Guo S., Xing C., Fan X., Ning M., Yuan J.C., Lo E.H., Wang X. ( 2009) Neuroglobin-overexpression alters hypoxic response gene expression in primary neuron culture following oxygen glucose deprivation. Neuroscience , 162, 396– 403. Google Scholar CrossRef Search ADS PubMed  44 Zhu H., Wang Z., Xing Y., Gao Y., Ma T., Lou L., Lou J., Gao Y., Wang S., Wang Y. ( 2012) Baicalin reduces the permeability of the blood-brain barrier during hypoxia in vitro by increasing the expression of tight junction proteins in brain microvascular endothelial cells. J. Ethnopharmacol ., 141, 714– 720. Google Scholar CrossRef Search ADS PubMed  45 Guo F., Wang H., Li L., Zhou H., Wei H., Jin W., Wang Q., Xiong L. ( 2013) A novel domain of amino-Nogo-A protects HT22 cells exposed to oxygen glucose deprivation by inhibiting NADPH oxidase activity. Cell. Mol. Neurobiol ., 33, 443– 452. Google Scholar CrossRef Search ADS PubMed  46 Alluri H., Anasooya Shaji C., Davis M.L., Tharakan B. ( 2015) Oxygen-glucose deprivation and reoxygenation as an in vitro ischemia-reperfusion injury model for studying blood-brain barrier dysfunction. J. Vis. Exp ., 99, e52699. 47 Lewis B.P., Burge C.B., Bartel D.P. ( 2005) Conserved seed pairing, often flanked by adenosines, indicates that thousands of human genes are microRNA targets. Cell , 120, 15– 20. Google Scholar CrossRef Search ADS PubMed  48 Huang D.W., Sherman B.T., Lempicki R.A. ( 2009) Systematic and integrative analysis of large gene lists using DAVID Bioinformatics Resources. Nat. Protocol ., 4, 44– 57. Google Scholar CrossRef Search ADS   49 Tan L., Yu J.T., Liu Q.Y., Tan M.S., Zhang W., Hu N., Wang Y.L., Sun L., Jiang T., Tan L. ( 2014) Circulating miR-125b as a biomarker of Alzheimer's disease. J. Neurol. Sci ., 336, 52– 56. Google Scholar CrossRef Search ADS PubMed  © The Author(s) 2018. Published by Oxford University Press. All rights reserved. For permissions, please email: journals.permissions@oup.com This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices)

Journal

Human Molecular GeneticsOxford University Press

Published: Apr 25, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off