Conceived and designed the experiments: ZH DY JZ ZL. Performed the experiments: ZH CA. Analyzed the data: ZH DY KT. Wrote the paper: ZH YD AMM LE MFB JZ ZL.
The authors have declared that no competing interests exist.
Traumatic brain injury (TBI) is a leading cause of injury-related death and disability worldwide. Effective treatment for TBI is limited and many TBI patients suffer from neuropsychiatric sequelae. The molecular and cellular mechanisms underlying the neuronal damage and impairment of mental abilities following TBI are largely unknown. Here we used the next generation sequencing platform to delineate miRNA transcriptome changes in the hippocampus at 24 hours and 7 days following TBI in the rat controlled cortical impact injury (CCI) model, and developed a bioinformatic analysis to identify cellular activities that are regulated by miRNAs differentially expressed in the CCI brains. The results of our study indicate that distinct sets of miRNAs are regulated at different post-traumatic times, and suggest that multiple miRNA species cooperatively regulate cellular pathways for the pathological changes and management of brain injury. The distinctive miRNAs expression profiles at different post-CCI times may be used as molecular signatures to assess TBI progression. In addition to known pathophysiological changes, our study identifies many other cellular pathways that are subjected to modification by differentially expressed miRNAs in TBI brains. These pathways can potentially be targeted for development of novel TBI treatment.
Traumatic brain injury (TBI) is an insult to the brain from the application of external physical force that leads to temporary or permanent structural and functional impairment of the brain. TBI is a leading cause of injury-related death and disability
MicroRNAs (miRNAs) are a class of small non-coding RNAs which repress gene expression at the post-transcriptional level by translational inhibition and/or promoting mRNA degradation through binding of the miRNA seed regions (usually nucleotides 2–8 at the 5′ end) to their binding sites in the 3′ untranslated region (3′ UTR) of the target mRNAs
Animal models are important tools for studying TBI. The lateral fluid percussion injury (FPI) model and the controlled cortical impact injury (CCI) model are two widely used animal models of focal TBI. FPI is delivered through a pressurized pulse of saline solution created by a piston
Here, by employing the next generation sequencing platform, we analyze the miRNA transcriptomes in the rat hippocampus at 24 hours and 7 days after TBI in the CCI model. Deep sequencing provides superb sensitivity, quantifiability and throughput over microarray and quantitative RT-PCR assays for miRNAs. We show that distinct miRNAs are up- or down-regulated at the two post-CCI time points. To decipher the biological implication of altered miRNA expression, we design a miRNA-gene-GO enrichment analysis to identify biological functions that are regulated by the target genes of changed miRNAs. We find that different gene ontology (GO) terms are targeted by the miRNAs altered in expression at the early and late phase of TBI. Intriguingly, the biological processes associated with these GO terms are consistent with the characteristic post-CCI pathological, physiological and structural changes. These findings indicate that miRNA expression profile changes are a distinctive molecular signature of post-TBI brains, and regulation of gene expression by miRNAs might be an important component of the cellular programs used by the injured brain to modulate gene expression for management and repair of injury-related cell damages.
To interrogate the effect of TBI on miRNA expression, we used Illumina sequencing to analyze CCI-induced miRNA transcriptome changes. Male rats aged 60 days were sham operated or subjected to CCI (3–4 rats for each condition), and euthanized for collection of brain tissues at 24 hours and 7 days after CCI. RNA was extracted from the dorsal hippocampus, which corresponds to the human posterior hippocampus and is involved in learning and memory
The normalized tag counts of individual miRNAs in each library were used to assess their relative abundance. The fold change and p-values of miRNA expression changes detected by deep sequencing were illustrated in
The hippocampus on the operated side of the brain was removed for miRNA expression profiling by deep sequencing at 24 hours (A, B) and 7 days (C, D) after CCI. Each circle represents a miRNA species. (A, C) The fold change (on log2 scale) of each miRNA species in CCI rats versus sham controls is plotted against its normalized read number (on log2 scale). (B, D) The p-value (on log10 scale) of each miRNA species is plotted against its fold change (on log2 scale). Blue and green dots are miRNAs that have p-values less than 0.05. (E, F) The fold change (on log2 scale) of miRNAs analyzed by qRT-PCR. Data are presented as mean ± s.e.m. Student’s t test was used for statistical analysis. * p<0.05; ** p<0.01.
Heat map of mean signal intensities following normalization for all miRNAs detected in the CCI and sham control hippocampi by deep sequencing.
To validate the results of deep sequencing with an independent assay, we used quantitative reverse transcription polymerase chain reaction (qRT-PCR) to measure the expression change of miR-124, miR-135, miR-153 and miR-222. As shown in
By using deep-sequencing, we found that 13 miRNAs were down-regulated and 8 miRNAs were up-regulated at 24 hours after CCI (p<0.05,
The cellular and physiological effects of altered miRNA expression are mainly mediated by their regulation of target mRNAs. Many miRNAs are predicted to have hundreds of mRNA targets, and each mRNA usually has binding sites for multiple miRNA species. While the repressive effect of an individual miRNA species on its target expression can be moderate, the combinatorial control of a gene by multiple miRNA species that bind to the same mRNA target, and the concurrent action of one miRNA species on several target genes belonging to the same signaling pathway are expected to regulate gene expression and biological functions in a more robust and fine-tuned manner
To test the possibility that multiple miRNA species function cooperatively in TBI, we developed a miRNA-gene-GO enrichment analysis to computationally examine the potential effects of miRNA expression changes in CCI. The effect size of miRNA expression change on its target gene expression and cellular activities varies greatly among miRNAs, and is determined by the fold change of the miRNAs. Because the absolute fold changes of miRNAs positively correlate with their p-values (
The miRNA-gene-GO enrichment analysis was conducted in three steps. Because the rat genome is annotated less well than the mouse genome
(A) Genes and GO terms enriched by down-regulated miRNAs. (B) Genes and GO terms enriched by up-regulated miRNAs.
(A) Genes and GO terms in the cellular component ontology enriched by down-regulated miRNAs. (B) Genes and GO terms in the cellular component ontology enriched by up-regulated miRNAs. (C) Genes and GO terms in the molecular function ontology enriched by down-regulated miRNAs. (D) Genes and GO terms in the molecular function ontology enriched by up-regulated miRNAs.
(A) Genes and GO terms enriched by down-regulated miRNAs. (B) Genes and GO terms enriched by up-regulated miRNAs.
(A) Genes and GO terms in the cellular component ontology enriched by down-regulated miRNAs. (B) Genes and GO terms in the cellular component ontology enriched by up-regulated miRNAs. (C) Genes and GO terms in the molecular function ontology enriched by down-regulated miRNAs. (D) Genes and GO terms in the molecular function ontology enriched by up-regulated miRNAs.
Some identified GO terms have common cellular functions. For example, the GO-terms “mRNA export from nucleus”, “nuclear export” and “nucleus localization” are all related to nuclear transport. In
The GO enrichment and clustering analysis reveals that: 1. at the early times after brain injury, miRNAs might be involved in the inhibition of white blood cell apoptosis to augment the inflammatory response, the dampening of transcription and facilitation of protein folding and catabolism to reduce stress-induced denaturation and aggregation of intracellular proteins, and the promotion of aerobic respiration to exert protective actions; 2. At the later post-CCI times, miRNAs preferentially regulate gene expression, intracellular trafficking and metabolism, cytoskeleton and cell adhesion to allow cell structure remodeling and synaptogenesis for repair of neuronal circuitry. These findings suggest that miRNAs are involved in both the early post-CCI pathological changes and stress management, and the later brain repair mechanisms.
In this study, we examine the effect of CCI injury on the miRNA transcriptome in the hippocampus, and identify cellular functions and biological processes that are potentially regulated by miRNAs in the injured hippocampus. Our findings reveal that distinct sets of miRNAs are regulated at different post-CCI time points. To enquire the biological pathways targeted by the differentially-expressed miRNAs in a high-throughput manner and obtain a comprehensive view on the functional significance of miRNAs in TBI, we developed the miRNA-gene-GO enrichment analysis to functionally annotate miRNA expression changes. Our bioinformatic analysis shows that miRNAs target to multiple GO terms in a post-CCI-time-specific way. These findings indicate that miRNAs act in concert to regulate a broad range of cellular functions in the brain at different stages of TBI.
The fact that both up- and down-regulation of miRNAs are detected in the injured hippocampus indicates that the miRNA expression change is not caused by global changes of miRNA biogenesis or turnover. The observations that more miRNAs are altered at 24 hours than at 7 days post-CCI and that ∼75% of miRNAs changed at 24 hours restore their expression levels at 7 days suggest that the mechanisms to alter miRNA expression at various post-CCI times are different. In addition to the number, the identities of miRNAs altered at the two post-CCI time points that we examined are also different, thus each post-CCI stage has a unique miRNA expression profile, which can be potentially used as molecular markers for TBI progress. Redell and colleagues have used the microarray assay to examine miRNA expression changes at 3 hour and 24 hour after CCI
miRNA expression changes may be the outcome of injury-related pathological changes in the brain, such as cell death and neuroinflammation. If this is the case, we predict a uniform change direction for all or the majority of miRNA species, but only 6% of miRNAs detected in the hippocampus show altered expression, and both increases and decreases are observed, suggesting that miRNAs do not passively respond to injury. It is likely that modulating miRNA expression is an active approach taken by cells to cope with injury.
What is the consequence of miRNA expression changes? A miRNA usually has numerous mRNA targets and the effect of altering a single miRNA species on its targets may be moderate. The robustness of miRNA mediated gene expression regulation can be enhanced through cooperative action of several types of miRNAs that bind to the same target gene or to several genes serving for the same cellular process. We designed a miRNA-gene-GO enrichment analysis to test if distinct miRNAs changed in the same direction would also coordinately regulate cellular functions. miRNAs changed in the same direction have similar effects on their target expression, hence are put in the same group for the test. The results of our analysis show that many differentially expressed miRNAs are related to common GO terms. Both 24 hours and 7 days post-CCI correspond to the secondary damage stage of TBI, which occurs at minutes to days and even months after the initial trauma as the results of delayed histological, biochemical, metabolic and cellular changes
The fact that distinct GO terms are identified for the two post-CCI time points is intriguing. At 24 hours post-CCI, miRNAs are modulated to inhibit leukocyte cell death for the inflammatory response, prevent stress-induced protein denaturation and aggregation, and protect cells from energy exhaustion. At the later recovery-oriented time, cells switch to miRNAs that bolster a permissive cellular environment for cell repair and structural remodeling. Hence, miRNA mediated gene expression regulation is an integral component of the cellular strategies to adapt to, compensate for and repair brain damages from traumatic brain injury. Future studies to understand the specific functions of each differentially expressed miRNA and the significance of the targeted cellular pathways for TBI may contribute to the amelioration of long-term cognitive dysfunction as a consequence of brain injury.
CCI injury was performed in male rats (Sprague Dawley, 200 g) as previously described
Dorsal hippocampi of rats were removed and homogenized with a Polytron homogenizer (Kinematica) in the lysis buffer provided by the mirVana miRNA Isolation Kit (Ambion). The lysate was then extracted with acid-phenol:chloroform, added ethanol to bring up the sample to 25% ethanol and fractionated to isolate the large and small RNAs by using the glass-fiber filter of mirVana miRNA Isolation Kit. miRNAs were purified from the small RNA fraction by using denaturing polyacrylamide gel electrophoresis (15%) followed by recovering the 17∼27-nt-long RNA fragments which were enriched for miRNAs. RNA integrity number (RIN) of the extracted RNA samples was measured to assess RNA quality by using the Agilent 2100 Bioanalyzer. Deep-sequencing libraries were constructed by using the Small RNA Sample Prep kit (Illumina). Briefly, 3′ adaptors and 5′ adaptors were ligated to purified miRNAs sequentially. The adaptor tagged miRNAs were reverse-transcribed and PCR amplified. The PCR products were purified by polyacrylamide gel electrophoresis (8%) and sequenced by using an Illumina Genome Analyzer II.
Sequence data analysis was performed as described previously
The following equation was used for the hypergeometric statistical test in the gene enrichment analysis for each test group (e.g. 24 h positive).
The enriched GO terms that annotate the over-represented genes were identified by using the conditional hypergeometric statistical test [R package GOStats,
miRNAs were extracted by using the mirVana miRNA Isolation Kit (Ambion). The TaqMan miRNA assay kit (Applied Biosystems) was used for qRT-PCR of miRNAs. Briefly, miRNAs were transcribed into cDNAs and amplified by PCR with the Taqman primers specific for miRNAs of interest by using the Applied Biosystems 7900 Fast Real-Time PCR System according to the manufacture’s manual. U6 was used as the endogenous control gene to normalize input amounts.
This study was carried out in strict accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health. The protocol was approved by Institutional Animal Care and Use Committee (IACUC) of Uniformed Services University of the Health Sciences (Permit Number: APG 09-748). All surgery was performed under isoflurane anesthesia, and all efforts were made to minimize suffering.
(TIF)
(DOCX)
(DOCX)
(DOCX)
We thank Dr. Yanqing Yang (NHLBI/NIH) for help with data analysis and Dr. Heinz Arnheiter (NINDS/NIH) for fruitful discussions and critical reading of the manuscript.