- Email: [email protected]
Deep Sequencing of Small RNAs in Blood of Patients with Brain Arteriovenous Malformations
Accepted Manuscript Deep sequencing of small RNAs in blood of patients with brain arteriovenous malformations Yong Chen, Zhili Li, Yi Shi, Guangfu Huang, Longyi Chen, Haibin Tan, Zhenyu Wang, Cheng Yin, Junting Hu PII:
S1878-8750(18)30816-7
DOI:
10.1016/j.wneu.2018.04.097
Reference:
WNEU 7933
To appear in:
World Neurosurgery
Received Date: 24 November 2017 Revised Date:
11 April 2018
Accepted Date: 13 April 2018
Please cite this article as: Chen Y, Li Z, Shi Y, Huang G, Chen L, Tan H, Wang Z, Yin C, Hu J, Deep sequencing of small RNAs in blood of patients with brain arteriovenous malformations, World Neurosurgery (2018), doi: 10.1016/j.wneu.2018.04.097. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT Deep sequencing of small RNAs in blood of patients with brain arteriovenous malformations Short Title: Key miRNAs in brain arteriovenous malformations
Wang1, Cheng Yin1, Junting Hu1 1
Department of Neurosurgery, Sichuan Academy of Medical Sciences & Sichuan
SC
Provincial People’s Hospital
Key Laboratory of SiChuan Province in Human Disease Gene Study, Sichuan
M AN U
2
RI PT
Yong Chen1, Zhili Li1* Yi Shi2, Guangfu Huang1, Longyi Chen1, Haibin Tan1, Zhenyu
Academy of Medical Sciences & Sichuan Provincial People’s Hospital * Correspondence to Zhili Li,
Department of Neurosurgery, Sichuan Academy of Medical Sciences & Sichuan People’s
Hospital,
No.32
TE D
Provincial
Road,Chengdu,Sichuan, 610072, China Email: [email protected]
EP
Tel: 86-13808067618
AC C
Fax: 86-028-87394376
West
Second
Section
First
Ring
ACCEPTED MANUSCRIPT Abstract Background: The deregulation of circulating microRNAs (miRNAs) is always associated with development and progression of human diseases. We are aimed to
miRNA signature compared with healthy subjects.
RI PT
assess whether brain arteriovenous malformations (BAVMs) patients present a distinct
Methods: Three unruptured BAVMs patients and three normal controls were recruited
SC
as case and normal groups. Peripheral blood was collected and miRNA signature was
M AN U
obtained by next-generation sequencing, followed by comparative, functional, and network analyses. Further qRT-PCR was performed to validate the expression of specific miRNAs.
Results: Deep sequencing detected 246 differentially expressed miRNAs in blood
TE D
samples of BAVM patients compared with normal controls. For the top five miRNAs, 946 target genes were predicted and a BAVM-specific miRNA-target gene regulatory network was then constructed. Functional annotation suggested that 15 of the
EP
predicted miRNAs targeted genes were involved in VEGF signaling, in which three
AC C
critical miRNAs involved, including miR-7-5p, miR-199a-5p and miR-200b-3p. Conclusion: We explored the miRNA expression signature of BAVMs, which will provide an important foundation for future studies on the regulation of miRNAs involved in BAVMs. Keywords
brain arteriovenous malformations, miRNAs, VEGF signaling
ACCEPTED MANUSCRIPT Introduction Brain arteriovenous malformations (BAVMs) are major causes of intracranial hemorrhage or subarachnoid hemorrhage especially in young adults, leading to
RI PT
serious neurological symptoms and substantial mortality 1. BAVMs are brain vascular lesions, which consisting of abnormal tangles of vessels (nidus) or “bag of worms” and leading to direct communication of arteries and veins without an interposing
SC
capillary bed 2. Current medical treatments, such as surgical removal, endovascular
M AN U
treatment, and stereotactic radiosurgery are highly invasive and can lead to significant risks to nearby brain structures. Moreover, current medical treatments can’t prevent development or rupture of BAVMs 3-5.
Although it is classically believed that BAVMs exist at birth, growing evidences
TE D
support the postnatal growth of BAVMs 6. Knowledge from the previous studies showed that some SNPs in the inflammatory cascades and in the regulation of angiogenesis play a role in the development of hemorrhage of BAVMs 7,8
. Moreover, angiogenic factors and inflammatory cytokines are
EP
nonspecifically
9-12
. Even so, the mechanism of BAVM
AC C
observed to contribute to manifest BAVMs
formation is not yet well elucidated and it is difficult to diagnose the patients who present unruptured with other clinical symptoms (e.g., seizure, headache, focal neurologic deficit) or asymptomatic. Thus, identification of biomarkers that predict BAVM would help to improve the optimal patient management and represent a major advance in clinical care. MicroRNAs (miRNAs) are a group of non-coding small RNAs which play an
ACCEPTED MANUSCRIPT important regulatory role in various biological processes. The expression profiles of miRNAs in tissue and peripheral blood can reflect pathological changes in the body and brain. Peripheral blood continuously interacts with the tissues of the body and
which could be measured at earlier stages of disease 13.
RI PT
different disease states may trigger specific changes in blood cell gene expression
For many diseases, especially those which affect the vasculature, tissue
SC
specimens are usually only obtained after a patient has become symptomatic and
M AN U
treated. Therefore, in the present study, we collected peripheral blood samples from three patients with unruptured BAVMs and three controls. The miRNA expression profiles were evaluated by next-generation sequencing. Further prediction of target genes and functional enrichment of target genes suggested that VEGF signaling
TE D
pathway may be involved in both physiologic and pathologic angiogenesis of BAVMs. Our results suggested that three miRNAs (hsa-miR-7-5p, hsa-miR-199a-5p and
EP
hsa-miR-200b-3p) may serve as potential biomarkers for BAVMs.
AC C
Materials and Methods Patient Recruitment
In this study, three specimens from patients with BAVMs were recruited at the
Sichuan Academy of Medical Sciences & Sichuan Provincial People’s Hospital according to the standardized guidelines
14
. These patients were diagnosed by
angiography and were defined as unruptured BAVMs at presentation without evidence of new bleeding. Moreover, three specimens were recruited as the controls, which
ACCEPTED MANUSCRIPT underwent craniocerebral trauma with no history of cardiovascular or cerebrovascular disease and came to our hospital. All three patients (BAVM-3) with BAVM have no family history.
BAVM 1 was a 22 years old female with right dome cerebrovascular
RI PT
malformation (3×4cm) and epilepsy while without other neurological dysfunction; BAVM 2 was a 28 years old male with left occipital cerebral vascular malformation (3×3cm), headache and normal nerve function; BAVM 3 was a 26 years old male with
SC
left dome cerebrovascular malformation (4×4cm) and visual field defect. All the three
M AN U
control patients (Control1-3) have no family history neither basic diseases. Control 1 was a 16 years old male with right frontotemporal contusion; Control 2 was a 26 years old male with right dome subdural hematoma and Control 3 was a 11 years old male with left frontal brain contusion. None of these three controls has neurofunctional loss.
TE D
The Glasgow Coma Scale of (GCS) of all these three controls patients was 15. Except slight cranial trauma, there is no other pathologic factor affects the results in this study. After obtaining the informed consent, the blood specimens were drawn from each
. Blood samples of control patients were obtained after 3-6 hours of trauma which
AC C
15
EP
subject into PAXgene tubes (PreAnalytiX, Hilden, Germany) as previously described
reduce the influence of cranial trauma on results in this study. Our study complied with the declaration of Helsinki principles and was approved by the Ethics Committee of Sichuan Academy of Medical Sciences & Sichuan Provincial People’s Hospital.
RNA preparation and next-generation sequencing Total RNA from each sample was isolated using the Trizol reagent (Invitrogen,
ACCEPTED MANUSCRIPT CA, USA) according to the manufacturer’s protocols. The RNA quality and quantity was verified by NanoDrop 2000 (NanoDrop Technologies, Wilmington, DE, USA) and RNA integrity number (RIN) was measured by Agilent 2100 (Agilent, Santa
RI PT
Clara,CA, USA). Then, we constructed the small RNA (sRNA) library using TrueSeq small RNA library prep kit (Illumina San Diego CA, USA) according to the manufacturer's instruction. Further, Adapters were ligated to each end of the RNAs
SC
and subsequently reverse transcribed to create single-stranded cDNA and sequenced
miRNA sequencing data analysis
M AN U
on the Hiseq2500/Miseq platform with a read length of 50bps.
The RNA-sequencing data was used to discover the differentially expressed
TE D
miRNAs. Firstly, reads were quality checked, and low quality bases and the adapter sequences were then trimmed using the cutadapt. Secondly, the trimmed reads were mapped to the human reference genome GRCh37 using the popular alignment tool 16
. Finally, we used miRDeep2 (https://www.mdc-berlin.de/8551903/en/) to
EP
Bowtie
17
. Mature miRNA and miRNA precursors were download from
AC C
quantify miRNAs
miRBase. Moreover, we used the DEGseq package in R to determine differentially expressed miRNAs. The FDR<0.001 and |log2(Fold_change)|>2 were used as the cut-off criteria. To visualize the expression profiles that distinguish unruptured BAVM patients from healthy controls, the differentially expressed miRNAs identified from the groups were organized using the hierarchical clustering method.
ACCEPTED MANUSCRIPT Prediction of potential targets of differentially expressed miRNAs The miRNAs can recognize and regulate the transcripts of its target genes at the post-transcriptional level. A bioinformatic study was conducted to predict the target
RI PT
genes regulated by differentially expressed miRNAs. We herein used six different bioinformatic algorithms, including RNA22 v2 (https://cm.jefferson.edu/rna22v2.0/) ,
miRanda-mirSVR
19
(http://www.microrna.org/) 20
(http://mirdb.org/miRDB/)
,
SC
18
(http://pictar.mdc-berlin.de/)
22
M AN U
(http://www.umm.uni-heidelberg.de/apps/zmf/mirwalk/index.html)
,
miRDB
miRWalk
21
,
PICTAR2
and Targetscan6.2 (http://www.targetscan.org/) 23, and
the target genes were obtained. Moreover, we identified the validated targets of differentially
expressed
miRNAs
using
the
miRTarBase
TE D
(http://mirtarbase.mbc.nctu.edu.tw/), which has collected 51,460 miRNA–target interactions from 18 species having experimental evidence
24
. Furthermore, the
miRNA-target pairs were subjected to construct the specific miRNA-targets network,
which
EP
interaction
was
then
visualized
using
Cytoscape
AC C
(http://www.cytoscape.org/) 25.
Functional interpretation of target genes To provide a functional interpretation of the target genes related to BAVM, we
performed functional annotation of target genes using the Gostats package in R. The listed target genes can be mapped to associated Gene Ontology (GO) terms, including biological process, cellular component and molecular function 26. Moreover,
ACCEPTED MANUSCRIPT the listed target genes can be assigned to associated Kyoto Encyclopedia of Genes and Genomes (KEGG) terms
27
. The graphical output was then produced by ggplot2
RI PT
package in R.
qRT-PCR validation
The peripheral blood was obtained from three patients with BAVMs and other
SC
five individuals were used as normal samples. Total RNA was extracted according to
M AN U
the method described above. Then cDNA was synthesized and qRT-PCR was performed with SYBR Green I Master (Roche, China) on the BIO-RAD IQ5 real-time PCR platform. Relative gene expression was analyzed using 2-∆∆Ct method. The human U6 and GAPDH gene were used as endogenous controls for miRNA and RNA
TE D
expression, respectively. Primers used in qRT-PCR validation were listed in Table 1. All reactions were analyzed triplicates. One-way ANOVA was used to calculate the significant differences of miRNAs and genes between BAVM and controls. *
AC C
EP
indicated p-value <0.05.
Results
Differential expression profiles of BAVM patients vs. controls The primary analysis compared blood miRNA expression profiles between three
BAVM patients and three healthy controls to identify miRNAs showing consistent expression differences. As was shown in Table 2, a total of 10619255, 10407754, 10515006 clean reads were obtained for case 1, case 2 and case 3, respectively. While
ACCEPTED MANUSCRIPT there were 11842374, 11829993 and 11828494 clean reads obtained for normal 1, normal 2 and normal 3, respectively. Above 91% of clean reads can be mapped to human reference genome. we identified 246 differentially
RI PT
With FDR<0.001 and |log2(Fold_change) |>2
expressed miRNAs in blood samples of BAVM patients compared with normal controls. All of the differentially expressed miRNAs were up-regulated, suggesting
SC
the important roles of up-regulated miRNAs in BAVM. Hierarchical cluster analysis
M AN U
of the top 50 differentially expressed miRNAs showed good segregation of three BAVM patients and three normal controls into two main branches (Figure 1). BAVM patients tended to have a pattern of miRNAs up-regulated (red color), whereas normal controls had a greater proportion of the same miRNAs down-regulated (green color).
TE D
Based on the read counts of miRNAs in each sample, we identified the differentially expressed miRNAs in peripheral blood between BAVMs and controls. The top ten differentially expressed miRNA in peripheral blood between BAVMs and controls
EP
(hsa-miR-7-5p, hsa-miR-629-5p, hsa-miR-199a-5p, hsa-let-7b-3p, hsa-miR-200b-3p, hsa-miR-148b-5p,
hsa-miR-1976,
hsa-miR-1284
and
AC C
hsa-miR-6747-3p,
hsa-miR-4433b-5p) were listed in Table 3. The read counts of these 10 miRNAs in each sample were listed in Supplemental Table S1.
Prediction of miRNA target genes It is well-known that discovering posttranscriptional gene silencing by miRNAs has led to an explosion of new hypotheses in human disease. To understand the
ACCEPTED MANUSCRIPT function of the top five differentially expressed miRNAs in BAVM, we predicted the target genes of the miRNAs using RNA22 v2 miRanda-mirSVR, miRDB, miRWalk, PICTAR2 and Targetscan 6.2, and counted miRNAs that were predicted. As a result,
RI PT
946 genes were found to be targeted by the top five differentially expressed miRNAs. Totally, we obtained 985 miRNA-target gene pairs. Based on the miRNA-target gene pairs, a BAVM-specific miRNA-target gene regulatory network was established
SC
(Figure 2). In the network, the top three miRNAs which regulated the most target
M AN U
genes were hsa-miR-7-5p, hsa-miR-199a-5p and hsa-miR-200b-3p.
Functional enrichment analysis of miRNA targets
GO enrichment analysis revealed that the miRNA targets were significantly
TE D
enriched in GO terms of cellular response to growth factor stimulus, response to growth factor and enzyme linked receptor protein signaling pathway for biological process (Figure 3A). For cellular component, the miRNA targets were significantly
EP
enriched in GO terms of intracellular part, intracellular and membrane-bounded
AC C
organelle (Figure 3B). For molecular function, the miRNA targets were significantly enriched in GO terms of protein binding, enzyme binding and organic cyclic compound binding (Figure 3C). KEGG pathway enrichment analysis identified 43 pathways that could be
involved in regulation of BAVM. We listed the top 25 significantly enriched pathways in Figure 4. Among these pathways, VEGF signaling pathway is crucial for both physiologic and pathologic angiogenesis. The binding of VEGF to VEGFR-2 leads to
ACCEPTED MANUSCRIPT a cascade of different signaling pathways, resulting in the up-regulation of genes involved in mediating the proliferation and migration of endothelial cells and promoting their survival and vascular permeability. The miRNA target genes in VEGF
RI PT
signaling pathway were further showed in Figure 5. The VEGF signaling pathway is involved in 15 of the genes, which are targeted by the hsa-miR-7-5p, hsa-miR-199a-5p and hsa-miR-200b-3p (Figure 5). Moreover, Pathway of focal
SC
adhesion plays essential roles in the considerable crosstalk between adhesion- and
M AN U
growth factor-mediated signalling.
Validation of miRNA and target expression
The peripheral blood of patients with BAVMs and normal samples were used to
TE D
verify the expression patterns of miRNAs and targets, including five miRNAs (miR-7-5p, miR-629-5p, miR-199a-5p, let-7b-3p, miR-200b-3p) and two target genes (VEGFA and KDR). We have performed the quality control data for U6 miRNA.
EP
Firstly, we calculated the mean 2-CT. The fold change of U6 miRNA in the internal
AC C
control in the BAVM samples compared to the control samples was: 5.59966E-05/6.22226E-05=0.89994 (p-value=0.10689) which suggested that human U6 gene was a good internal control gene in this study28. The results indicated that the expressions of miR-199a-5p, let-7b-3p and miR-200b-3p were consistent with next-generation sequencing. Moreover, VEGFA and KDR were significantly up-regulated in BAVMs compared with normal samples. (Figure 6)
ACCEPTED MANUSCRIPT Discussion To the best of our knowledge, this is the first study to present the global miRNA expression profiling in peripheral blood of BAVMs. In the current study,
RI PT
next-generation sequencing was used to identify the miRNA expression profiling in a paired design comparing BAVMs and normal controls, which showed a total of 246 differentially expressed miRNAs, which were all up-regulated in bloods of BAVMs
SC
and revealed that the aberrantly expressed miRNAs may aberrantly regulate target
M AN U
genes. Based on the qRT-PCR validation, three miRNAs (has-let-7b-3p, has-miR-200b-3p and has-miR-199a-5p) were up-regulated in BAVM which were consistent with our RNA-sequencing results generally and increased credibility of our RNA-sequencing results. However, the expression of let-7b-3p and miR-200b-3p in
TE D
BAVM in qRT-PCR validation were inconsistent with the results in RNA-sequencing which needs further research.
Considering that miRNAs work as small guide molecules in RNA silencing by
EP
degrading their mRNA target and/or by silencing translation 29, we suggested the gene
AC C
expression profiling may tend to be down-regulated in bloods of BAVMs. Interestingly, a previous study suggested that unruptured BAVMs patients tended to have a greater proportion of genes down-regulated
30
. Our results seemed to support
that the differential miRNA expression in BAVMs may be responsible for both physiologic and pathologic angiogenesis.Functional annotation of the target genes regulated by differentially expressed miRNAs supported the role for MAPK, Wnt and VEGF signaling in BAVMs. Previous study reported that the MAPK pathway has
ACCEPTED MANUSCRIPT been implicated in cerebral cavernous malformation
31
. Moreover, Wnt signaling
influences vascular development 32. In particular, angiogenesis plays a key role in the formation and evolution of BAVMs. The VEGF acts as a vascular endothelial
RI PT
cell-specific mitogen and is a major regulator of angiogenesis, which is expressed abundantly during embryonic development but is absent in most normal adult vascular beds, including the cerebral vasculature
33
. The VEGF family plays a key
SC
role in vasculogenesis and angiogenesis, and an experiment on animal indicated that
M AN U
VEGF pathway genes were dysregulated in endothelial cells derived from BAVMs, contributing to aberrant angiogenic characteristics and the pathogenesis of vascular malformations
34
. VEGF signaling in angiogenesis was involved in pathogenesis of
cerebral cavernous malformations and hereditary hemorrhagic telangiectasia
35
. Our
TE D
data strongly supported that differentially expressed miRNAs in peripheral blood of BAVMs may target the VEGF pathway, contributing to the BAVM phenotype. The VEGF signaling pathway is involved in 15 of the genes, which are targeted
EP
by the hsa-miR-7-5p, hsa-miR-199a-5p and hsa-miR-200b-3p. Prior study showed
AC C
that miR-7-5p was significant distinct in the peripheral blood of patients with glioblastoma pathologies 36. Another study also showed a significant down-regulation of miR-7-5p in glioblastoma microvasculature comparing with normal brain capillaries. Moreover, overexpression of miR-7-5p significantly inhibited human umbilical vein endothelial cell proliferation, which suggested that miR-7-5p might be an angiogenesis inhibitor of microvascular proliferation in glioblastoma
37
. Our
present study showed that miR-7-5p was significantly up-regulated in peripheral
ACCEPTED MANUSCRIPT blood of BAVMs and its target genes involved in VEGF signaling pathway, suggesting its crucial role of miR-7-5p in the pathogenesis of BAVMs. Chan et al provided the first evidence that induction of miR-199a-5p in human
RI PT
dermal microvascular endothelial cells blocked angiogenic response of endothelial cells in Matrigel culture, whereas miR-199a-5p-deprived cells exhibited enhanced angiogenesis in vitro
38
. Moreover, Zhang et al determined that miR-199a-5p acts as
SC
an endogenous inhibitor of SIRT1/eNOS to strengthen angiogenesis and involved in
39
M AN U
the pathological angiogenesis induced by human cytomegalovirus (HCMV) infection . In the present study, we demonstrated the up-regulation of miR-199a-5p in
peripheral blood of BAVMs and targeted VEGF pathway, which may provide a potential miRNA-based mechanism in the pathological angiogenesis of BAVMs.
TE D
Through microarray analysis, Shi et al found that miR-200b-3p was overexpressed in aortic tissue from degenerative aortic stenosis (AS) patients
40
. Our
present investigation provided novel evidence that up-regulation of miR-200b-3p in
EP
peripheral blood may be involved in the BAVMs. As a limitation of our study, the
AC C
VEGF pathway affected by miRNAs and which cause BAVMs remain elusive. Further studies are required to functionally characterize the role of specific candidate miRNAs.
Besides, our result identified a number of novel dysregulated miRNAs in in
peripheral blood of BAVMs with unknown function, such as miR-629-5p, let-7b-3p, miR-6747-3p,
hsa-miR-148b-5p,
hsa-miR-1976,
hsa-miR-1284
and
hsa-miR-4433b-5p. Further elucidation of the miRNA expression signature may
ACCEPTED MANUSCRIPT provide a better understanding of the mechanisms of BAVMs
41
. Taken together, our
study identified a miRNA expression signature in peripheral blood of patients with BAVMs, and the targeted pathway of VEGF signaling may have a crucial role in
RI PT
formation of BAVMs. We also suggested that three miRNAs (miR-7-5p, miR-199a-5p and miR-200b-3p) may serve as the potential diagnostic markers and anti-angiogenic therapies for BAVMs.
SC
However, age and gender differences between BAVM and controls and small
M AN U
sample size were limitations of this study. Further validation analysis with larger sample size and biological test were needed to confirm our finding. Moreover, follow-up study comparing miRNA signature between unruptured and rupture-prone BAVMs would be of great importance and would be performed in our following
Acknowledgement
TE D
research.
This study was supported by The study of molecular mechanisms of brain
EP
arteriovenous vascular malformation (No. 140072).
AC C
Competing interests
The authors declare that they have no competing interests.
References 1.
Komiyama
M.
Pathogenesis
of
brain
arteriovenous
malformations.
Neurologia
medico-chirurgica 2016;56:317-25. 2.
Yashar P, Amar AP, Giannotta SL, et al. Cerebral arteriovenous malformations: Issues of the interplay between stereotactic radiosurgery and endovascular surgical therapy. World neurosurgery 2011;75:638-47.
3.
van Beijnum J, van der Worp HB, Buis DR, et al. Treatment of brain arteriovenous
ACCEPTED MANUSCRIPT malformations: A systematic review and meta-analysis. Jama 2011;306:2011-9. 4.
Mohr JP, Parides MK, Stapf C, et al. Medical management with or without interventional therapy for unruptured brain arteriovenous malformations (aruba): A multicentre, non-blinded, randomised trial. Lancet 2014;383:614-21.
5.
Haw CS, terBrugge K, Willinsky R, Tomlinson G. Complications of embolization of arteriovenous malformations of the brain. Journal of neurosurgery 2006;104:226-32.
6.
Mullan S, Mojtahedi S, Johnson DL, Macdonald RL. Embryological basis of some aspects of
RI PT
cerebral vascular fistulas and malformations. Journal of neurosurgery 1996;85:1-8. 7.
Pawlikowska L, Poon KY, Achrol AS, et al. Apolipoprotein evarepsilon2 is associated with new hemorrhage risk in brain arteriovenous malformations. Neurosurgery 2006;58:838-43.
8.
Pawlikowska L, Tran MN, Achrol AS, et al. Polymorphisms in genes involved in inflammatory and angiogenic pathways and the risk of hemorrhagic presentation of brain
9.
SC
arteriovenous malformations. Stroke; a journal of cerebral circulation 2004;35:2294-300.
Chen Y, Zhu W, Bollen AW, et al. Evidence of inflammatory cell involvement in brain arteriovenous malformations. Neurosurgery 2008;62:1340-9; discussion 9-50.
10.
Chen Y, Pawlikowska L, Yao JS, et al. Interleukin-6 involvement in brain arteriovenous
11.
M AN U
malformations. Annals of neurology 2006;59:72-80.
Xu B, Wu YQ, Huey M, et al. Vascular endothelial growth factor induces abnormal microvasculature in the endoglin heterozygous mouse brain. Journal of cerebral blood flow and metabolism : official journal of the International Society of Cerebral Blood Flow and Metabolism 2004;24:237-44.
12.
Hao Q, Zhu Y, Su H, et al. Vegf induces more severe cerebrovascular dysplasia in endoglin than in alk1 mice. Translational stroke research 2010;1:197-201.
Mohr S, Liew CC. The peripheral-blood transcriptome: New insights into disease and risk
TE D
13.
assessment. Trends in molecular medicine 2007;13:422-32. 14.
Joint Writing Group of the Technology Assessment Committee American Society of I, Therapeutic N, Joint Section on Cerebrovascular Neurosurgery a Section of the American Association of Neurological S, et al. Reporting terminology for brain arteriovenous
EP
malformation clinical and radiographic features for use in clinical trials. Stroke; a journal of cerebral circulation 2001;32:1430-42. 15.
Xu H, Tang Y, Liu DZ, et al. Gene expression in peripheral blood differs after cardioembolic
AC C
compared with large-vessel atherosclerotic stroke: Biomarkers for the etiology of ischemic stroke. Journal of cerebral blood flow and metabolism : official journal of the International
Society of Cerebral Blood Flow and Metabolism 2008;28:1320-8.
16.
Langmead B, Trapnell C, Pop M, Salzberg SL. Ultrafast and memory-efficient alignment of short DNA sequences to the human genome. Genome biology 2009;10:R25.
17.
Friedlander MR, Mackowiak SD, Li N, Chen W, Rajewsky N. Mirdeep2 accurately identifies
known and hundreds of novel microrna genes in seven animal clades. Nucleic acids research 2012;40:37-52. 18.
Miranda KC, Huynh T, Tay Y, et al. A pattern-based method for the identification of microrna binding sites and their corresponding heteroduplexes. Cell 2006;126:1203-17.
19.
Betel D, Koppal A, Agius P, Sander C, Leslie C. Comprehensive modeling of microrna targets predicts functional non-conserved and non-canonical sites. Genome biology 2010;11:R90.
20.
Wang X. Mirdb: A microrna target prediction and functional annotation database with a wiki
ACCEPTED MANUSCRIPT interface. Rna 2008;14:1012-7. 21.
Dweep H, Sticht C, Pandey P, Gretz N. Mirwalk--database: Prediction of possible mirna binding sites by "walking" the genes of three genomes. Journal of biomedical informatics 2011;44:839-47.
22.
Krek A, Grun D, Poy MN, et al. Combinatorial microrna target predictions. Nature genetics 2005;37:495-500.
23.
Garcia DM, Baek D, Shin C, et al. Weak seed-pairing stability and high target-site abundance
RI PT
decrease the proficiency of lsy-6 and other micrornas. Nature structural & molecular biology 2011;18:1139-46. 24.
Hsu SD, Tseng YT, Shrestha S, et al. Mirtarbase update 2014: An information resource for experimentally validated mirna-target interactions. Nucleic acids research 2014;42:D78-85.
25.
Smoot ME, Ono K, Ruscheinski J, Wang PL, Ideker T. Cytoscape 2.8: New features for data
26.
SC
integration and network visualization. Bioinformatics 2011;27:431-2.
Ashburner M, Ball CA, Blake JA, et al. Gene ontology: Tool for the unification of biology. The gene ontology consortium. Nature genetics 2000;25:25-9.
Kanehisa M, Goto S. Kegg: Kyoto encyclopedia of genes and genomes. Nucleic acids research 2000;28:27-30.
28.
Schmittgen TD, Livak KJ. Analyzing real-time pcr data by the comparative c(t) method. Nat Protoc 2008;3:1101-8.
29.
M AN U
27.
Oliveto S, Mancino M, Manfrini N, Biffo S. Role of micrornas in translation regulation and cancer. World journal of biological chemistry 2017;8:45-56.
30.
Weinsheimer SM, Xu H, Achrol AS, et al. Gene expression profiling of blood in brain arteriovenous malformation patients. Translational stroke research 2011;2:575-87. Dejana E, Tournier-Lasserve E, Weinstein BM. The control of vascular integrity by
TE D
31.
endothelial cell junctions: Molecular basis and pathological implications. Developmental cell 2009;16:209-21. 32.
Franco CA, Liebner S, Gerhardt H. Vascular morphogenesis: A wnt for every vessel? Current opinion in genetics & development 2009;19:476-83. Nauck M, Karakiulakis G, Perruchoud AP, Papakonstantinou E, Roth M. Corticosteroids
EP
33.
inhibit the expression of the vascular endothelial growth factor gene in human vascular smooth muscle cells. European journal of pharmacology 1998;341:309-15. Jabbour MN, Elder JB, Samuelson CG, et al. Aberrant angiogenic characteristics of human
AC C
34.
brain arteriovenous malformation endothelial cells. Neurosurgery 2009;64:139-46; discussion
46-8.
35.
Kar S, Baisantry A, Nabavi A, Bertalanffy H. Role of delta-notch signaling in cerebral
cavernous malformations. Neurosurgical review 2016;39:581-9.
36.
Dong L, Li Y, Han C, et al. Mirna microarray reveals specific expression in the peripheral blood of glioblastoma patients. International journal of oncology 2014;45:746-56.
37.
Liu Z, Liu Y, Li L, et al. Mir-7-5p is frequently downregulated in glioblastoma microvasculature and inhibits vascular endothelial cell proliferation by targeting raf1. Tumour biology : the journal of the International Society for Oncodevelopmental Biology and Medicine 2014;35:10177-84.
38.
Chan YC, Roy S, Huang Y, Khanna S, Sen CK. The microrna mir-199a-5p down-regulation switches on wound angiogenesis by derepressing the v-ets erythroblastosis virus e26 oncogene
ACCEPTED MANUSCRIPT homolog 1-matrix metalloproteinase-1 pathway. The Journal of biological chemistry 2012;287:41032-43. 39.
Zhang S, Liu L, Wang R, et al. Mir-199a-5p promotes migration and tube formation of human cytomegalovirus-infected endothelial cells through downregulation of sirt1 and enos. Archives of virology 2013;158:2443-52.
40.
Shi J, Liu H, Wang H, Kong X. Microrna expression signature in degenerative aortic stenosis. BioMed research international 2016;2016:4682172. Xiao J, Liang D, Zhang Y, et al. Microrna expression signature in atrial fibrillation with mitral
AC C
EP
TE D
M AN U
SC
stenosis. Physiological genomics 2011;43:655-64.
RI PT
41.
ACCEPTED MANUSCRIPT Figure legends Figure 1 Heat map of top 50 differentially expressed miRNAs in BAVMs. Figure 2 The regulatory network between top five differentially expressed miRNAs
RI PT
and target genes in BAVMs. Red color represents the up-regulated miRNAs; Blue color represents target genes.
Figure 3 The significantly enriched GO terms of target genes regulated by the top five
M AN U
Cellular component; (C) Molecular functions.
SC
differentially expressed miRNAs in BAVMs. (A) Biological process; (B)
Figure 4 The significantly enriched KEGG pathways of target genes regulated by the top five differentially expressed miRNAs in BAVMs.
Figure 5 miRNAs and target genes involved in VEGF signaling pathway. (A) Target
TE D
genes regulated by the top five differentially expressed miRNAs in VEGF signaling; (B) miRNA-target pairs.
Figure 6 qRT-PCR validation of differentially expressed miRNAs and target genes in
EP
BAVMs. One-way ANOVA was used to calculate the significant differences
AC C
of miRNAs and genes between BAVM and controls. * indicated p-value <0.05.
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
Table 1 Primers used in qRT-PCR validation Gene Primers GAPDH Forward: 5′GGAGCGAGATCCCTCCAAAAT 3′ Reverse: 5′GGCTGTTGTCATACTTCTCATGG 3′ VEGFA Forward: 5′CTGTCTTGGGTGCATTGGAGC 3′ Reverse: 5′AGGGTCTCGATTGGATGGCAG 3′ KDR Forward: 5′GTGATCGGAAATGACACTGGAG 3′ Reverse: 5′CATGTTGGTCACTAACAGAAGCA 3′ hsa-miR-7-5p 5′UGGAAGACUAGUGAUUUUGUUGU 3′ hsa-miR-629-5p 5′UGGGUUUACGUUGGGAGAACU 3′ hsa-miR-199a-5p 5′CCCAGUGUUCAGACUACCUGUUC 3′ hsa-let-7b-3p 5′CTATACAACCTACTACTGCCTTCCC 3′ hsa-miR-200b-3p 5′UAAUACUGCCUGGUAAUGAUGA 3′
ACCEPTED MANUSCRIPT Table 2 Summary of sequencing reads in BAVM patients and controls
BAVM 1
Sex/age
Total reads
Male/26
mapped
9750342
868913
BAVM 2
Female/18 10407754
9604484
803270
BAVM 3
Female/36 10515006
9790850
724156
Male/16
11842374 11751866
90508
Control 2
Male/26
11829993 11737060
92933
Control 3
Male/11
11828494 11729217
M AN U
TE D
(%)
91.8
8.2
92.3
7.7
93.1
6.9
99.2
0.8
99.2
0.8
99.2
0.8
SC
Control 1
EP
(%)
unmapped
10619255
AC C
Unmapped
RI PT
Sample
Mapped
99277
ACCEPTED MANUSCRIPT Table 3 Top ten differentially expressed miRNAs in BAVM Value (BAVM) Value (Control) log2 (fold change)
FDR
hsa-miR-7-5p
6960
0
NA
0
hsa-miR-629-5p
4002
0
NA
0
hsa-miR-199a-5p
3497
0
NA
hsa-let-7b-3p
1444
0
NA
hsa-miR-200b-3p
1380
0
NA
hsa-miR-6747-3p
1156
0
hsa-miR-148b-5p
1035
hsa-miR-1976
793
hsa-miR-1284
773
SC
M AN U
NA
0
0
0
0
0
NA
0
0
NA
0
0
NA
0
0
NA
0
TE D
hsa-miR-4433b-5p 622
RI PT
Gene name
Value in BAVM and Control samples respectively represented the sum of read count of miRNAs in BAVM and Control samples based on RNA-sequencing results. FDR,
EP
false discovery rate was obtained by using DEGseq package, based on the default
AC C
parameter settings of DEGseq package, FDR=0 indicated that FDR was less than 1E-16.
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT 1. We firstly obtained the global miRNA expression profile in peripheral blood of BAVMs. 2. MAPK, Wnt and VEGF signaling were involve in the process of BAVMs.
RI PT
3. miR-7-5p, miR-199a-5p and miR-200b-3p may be potential diagnostic markers of
AC C
EP
TE D
M AN U
SC
BAVMs.
AC C
EP
TE D
M AN U
SC
Abbreviations AS, aortic stenosis BAVMs, brain arteriovenous malformations GO, Gene Ontology HCMV, human cytomegalovirus KEGG, Genes and Genomes miRNAs, microRNAs sRNA, small RNA
RI PT
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT Conflict of interest
AC C
EP
TE D
M AN U
SC
RI PT
The authors declare that they have no competing interests.
S1878-8750(18)30816-7
DOI:
10.1016/j.wneu.2018.04.097
Reference:
WNEU 7933
To appear in:
World Neurosurgery
Received Date: 24 November 2017 Revised Date:
11 April 2018
Accepted Date: 13 April 2018
Please cite this article as: Chen Y, Li Z, Shi Y, Huang G, Chen L, Tan H, Wang Z, Yin C, Hu J, Deep sequencing of small RNAs in blood of patients with brain arteriovenous malformations, World Neurosurgery (2018), doi: 10.1016/j.wneu.2018.04.097. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT Deep sequencing of small RNAs in blood of patients with brain arteriovenous malformations Short Title: Key miRNAs in brain arteriovenous malformations
Wang1, Cheng Yin1, Junting Hu1 1
Department of Neurosurgery, Sichuan Academy of Medical Sciences & Sichuan
SC
Provincial People’s Hospital
Key Laboratory of SiChuan Province in Human Disease Gene Study, Sichuan
M AN U
2
RI PT
Yong Chen1, Zhili Li1* Yi Shi2, Guangfu Huang1, Longyi Chen1, Haibin Tan1, Zhenyu
Academy of Medical Sciences & Sichuan Provincial People’s Hospital * Correspondence to Zhili Li,
Department of Neurosurgery, Sichuan Academy of Medical Sciences & Sichuan People’s
Hospital,
No.32
TE D
Provincial
Road,Chengdu,Sichuan, 610072, China Email: [email protected]
EP
Tel: 86-13808067618
AC C
Fax: 86-028-87394376
West
Second
Section
First
Ring
ACCEPTED MANUSCRIPT Abstract Background: The deregulation of circulating microRNAs (miRNAs) is always associated with development and progression of human diseases. We are aimed to
miRNA signature compared with healthy subjects.
RI PT
assess whether brain arteriovenous malformations (BAVMs) patients present a distinct
Methods: Three unruptured BAVMs patients and three normal controls were recruited
SC
as case and normal groups. Peripheral blood was collected and miRNA signature was
M AN U
obtained by next-generation sequencing, followed by comparative, functional, and network analyses. Further qRT-PCR was performed to validate the expression of specific miRNAs.
Results: Deep sequencing detected 246 differentially expressed miRNAs in blood
TE D
samples of BAVM patients compared with normal controls. For the top five miRNAs, 946 target genes were predicted and a BAVM-specific miRNA-target gene regulatory network was then constructed. Functional annotation suggested that 15 of the
EP
predicted miRNAs targeted genes were involved in VEGF signaling, in which three
AC C
critical miRNAs involved, including miR-7-5p, miR-199a-5p and miR-200b-3p. Conclusion: We explored the miRNA expression signature of BAVMs, which will provide an important foundation for future studies on the regulation of miRNAs involved in BAVMs. Keywords
brain arteriovenous malformations, miRNAs, VEGF signaling
ACCEPTED MANUSCRIPT Introduction Brain arteriovenous malformations (BAVMs) are major causes of intracranial hemorrhage or subarachnoid hemorrhage especially in young adults, leading to
RI PT
serious neurological symptoms and substantial mortality 1. BAVMs are brain vascular lesions, which consisting of abnormal tangles of vessels (nidus) or “bag of worms” and leading to direct communication of arteries and veins without an interposing
SC
capillary bed 2. Current medical treatments, such as surgical removal, endovascular
M AN U
treatment, and stereotactic radiosurgery are highly invasive and can lead to significant risks to nearby brain structures. Moreover, current medical treatments can’t prevent development or rupture of BAVMs 3-5.
Although it is classically believed that BAVMs exist at birth, growing evidences
TE D
support the postnatal growth of BAVMs 6. Knowledge from the previous studies showed that some SNPs in the inflammatory cascades and in the regulation of angiogenesis play a role in the development of hemorrhage of BAVMs 7,8
. Moreover, angiogenic factors and inflammatory cytokines are
EP
nonspecifically
9-12
. Even so, the mechanism of BAVM
AC C
observed to contribute to manifest BAVMs
formation is not yet well elucidated and it is difficult to diagnose the patients who present unruptured with other clinical symptoms (e.g., seizure, headache, focal neurologic deficit) or asymptomatic. Thus, identification of biomarkers that predict BAVM would help to improve the optimal patient management and represent a major advance in clinical care. MicroRNAs (miRNAs) are a group of non-coding small RNAs which play an
ACCEPTED MANUSCRIPT important regulatory role in various biological processes. The expression profiles of miRNAs in tissue and peripheral blood can reflect pathological changes in the body and brain. Peripheral blood continuously interacts with the tissues of the body and
which could be measured at earlier stages of disease 13.
RI PT
different disease states may trigger specific changes in blood cell gene expression
For many diseases, especially those which affect the vasculature, tissue
SC
specimens are usually only obtained after a patient has become symptomatic and
M AN U
treated. Therefore, in the present study, we collected peripheral blood samples from three patients with unruptured BAVMs and three controls. The miRNA expression profiles were evaluated by next-generation sequencing. Further prediction of target genes and functional enrichment of target genes suggested that VEGF signaling
TE D
pathway may be involved in both physiologic and pathologic angiogenesis of BAVMs. Our results suggested that three miRNAs (hsa-miR-7-5p, hsa-miR-199a-5p and
EP
hsa-miR-200b-3p) may serve as potential biomarkers for BAVMs.
AC C
Materials and Methods Patient Recruitment
In this study, three specimens from patients with BAVMs were recruited at the
Sichuan Academy of Medical Sciences & Sichuan Provincial People’s Hospital according to the standardized guidelines
14
. These patients were diagnosed by
angiography and were defined as unruptured BAVMs at presentation without evidence of new bleeding. Moreover, three specimens were recruited as the controls, which
ACCEPTED MANUSCRIPT underwent craniocerebral trauma with no history of cardiovascular or cerebrovascular disease and came to our hospital. All three patients (BAVM-3) with BAVM have no family history.
BAVM 1 was a 22 years old female with right dome cerebrovascular
RI PT
malformation (3×4cm) and epilepsy while without other neurological dysfunction; BAVM 2 was a 28 years old male with left occipital cerebral vascular malformation (3×3cm), headache and normal nerve function; BAVM 3 was a 26 years old male with
SC
left dome cerebrovascular malformation (4×4cm) and visual field defect. All the three
M AN U
control patients (Control1-3) have no family history neither basic diseases. Control 1 was a 16 years old male with right frontotemporal contusion; Control 2 was a 26 years old male with right dome subdural hematoma and Control 3 was a 11 years old male with left frontal brain contusion. None of these three controls has neurofunctional loss.
TE D
The Glasgow Coma Scale of (GCS) of all these three controls patients was 15. Except slight cranial trauma, there is no other pathologic factor affects the results in this study. After obtaining the informed consent, the blood specimens were drawn from each
. Blood samples of control patients were obtained after 3-6 hours of trauma which
AC C
15
EP
subject into PAXgene tubes (PreAnalytiX, Hilden, Germany) as previously described
reduce the influence of cranial trauma on results in this study. Our study complied with the declaration of Helsinki principles and was approved by the Ethics Committee of Sichuan Academy of Medical Sciences & Sichuan Provincial People’s Hospital.
RNA preparation and next-generation sequencing Total RNA from each sample was isolated using the Trizol reagent (Invitrogen,
ACCEPTED MANUSCRIPT CA, USA) according to the manufacturer’s protocols. The RNA quality and quantity was verified by NanoDrop 2000 (NanoDrop Technologies, Wilmington, DE, USA) and RNA integrity number (RIN) was measured by Agilent 2100 (Agilent, Santa
RI PT
Clara,CA, USA). Then, we constructed the small RNA (sRNA) library using TrueSeq small RNA library prep kit (Illumina San Diego CA, USA) according to the manufacturer's instruction. Further, Adapters were ligated to each end of the RNAs
SC
and subsequently reverse transcribed to create single-stranded cDNA and sequenced
miRNA sequencing data analysis
M AN U
on the Hiseq2500/Miseq platform with a read length of 50bps.
The RNA-sequencing data was used to discover the differentially expressed
TE D
miRNAs. Firstly, reads were quality checked, and low quality bases and the adapter sequences were then trimmed using the cutadapt. Secondly, the trimmed reads were mapped to the human reference genome GRCh37 using the popular alignment tool 16
. Finally, we used miRDeep2 (https://www.mdc-berlin.de/8551903/en/) to
EP
Bowtie
17
. Mature miRNA and miRNA precursors were download from
AC C
quantify miRNAs
miRBase. Moreover, we used the DEGseq package in R to determine differentially expressed miRNAs. The FDR<0.001 and |log2(Fold_change)|>2 were used as the cut-off criteria. To visualize the expression profiles that distinguish unruptured BAVM patients from healthy controls, the differentially expressed miRNAs identified from the groups were organized using the hierarchical clustering method.
ACCEPTED MANUSCRIPT Prediction of potential targets of differentially expressed miRNAs The miRNAs can recognize and regulate the transcripts of its target genes at the post-transcriptional level. A bioinformatic study was conducted to predict the target
RI PT
genes regulated by differentially expressed miRNAs. We herein used six different bioinformatic algorithms, including RNA22 v2 (https://cm.jefferson.edu/rna22v2.0/) ,
miRanda-mirSVR
19
(http://www.microrna.org/) 20
(http://mirdb.org/miRDB/)
,
SC
18
(http://pictar.mdc-berlin.de/)
22
M AN U
(http://www.umm.uni-heidelberg.de/apps/zmf/mirwalk/index.html)
,
miRDB
miRWalk
21
,
PICTAR2
and Targetscan6.2 (http://www.targetscan.org/) 23, and
the target genes were obtained. Moreover, we identified the validated targets of differentially
expressed
miRNAs
using
the
miRTarBase
TE D
(http://mirtarbase.mbc.nctu.edu.tw/), which has collected 51,460 miRNA–target interactions from 18 species having experimental evidence
24
. Furthermore, the
miRNA-target pairs were subjected to construct the specific miRNA-targets network,
which
EP
interaction
was
then
visualized
using
Cytoscape
AC C
(http://www.cytoscape.org/) 25.
Functional interpretation of target genes To provide a functional interpretation of the target genes related to BAVM, we
performed functional annotation of target genes using the Gostats package in R. The listed target genes can be mapped to associated Gene Ontology (GO) terms, including biological process, cellular component and molecular function 26. Moreover,
ACCEPTED MANUSCRIPT the listed target genes can be assigned to associated Kyoto Encyclopedia of Genes and Genomes (KEGG) terms
27
. The graphical output was then produced by ggplot2
RI PT
package in R.
qRT-PCR validation
The peripheral blood was obtained from three patients with BAVMs and other
SC
five individuals were used as normal samples. Total RNA was extracted according to
M AN U
the method described above. Then cDNA was synthesized and qRT-PCR was performed with SYBR Green I Master (Roche, China) on the BIO-RAD IQ5 real-time PCR platform. Relative gene expression was analyzed using 2-∆∆Ct method. The human U6 and GAPDH gene were used as endogenous controls for miRNA and RNA
TE D
expression, respectively. Primers used in qRT-PCR validation were listed in Table 1. All reactions were analyzed triplicates. One-way ANOVA was used to calculate the significant differences of miRNAs and genes between BAVM and controls. *
AC C
EP
indicated p-value <0.05.
Results
Differential expression profiles of BAVM patients vs. controls The primary analysis compared blood miRNA expression profiles between three
BAVM patients and three healthy controls to identify miRNAs showing consistent expression differences. As was shown in Table 2, a total of 10619255, 10407754, 10515006 clean reads were obtained for case 1, case 2 and case 3, respectively. While
ACCEPTED MANUSCRIPT there were 11842374, 11829993 and 11828494 clean reads obtained for normal 1, normal 2 and normal 3, respectively. Above 91% of clean reads can be mapped to human reference genome. we identified 246 differentially
RI PT
With FDR<0.001 and |log2(Fold_change) |>2
expressed miRNAs in blood samples of BAVM patients compared with normal controls. All of the differentially expressed miRNAs were up-regulated, suggesting
SC
the important roles of up-regulated miRNAs in BAVM. Hierarchical cluster analysis
M AN U
of the top 50 differentially expressed miRNAs showed good segregation of three BAVM patients and three normal controls into two main branches (Figure 1). BAVM patients tended to have a pattern of miRNAs up-regulated (red color), whereas normal controls had a greater proportion of the same miRNAs down-regulated (green color).
TE D
Based on the read counts of miRNAs in each sample, we identified the differentially expressed miRNAs in peripheral blood between BAVMs and controls. The top ten differentially expressed miRNA in peripheral blood between BAVMs and controls
EP
(hsa-miR-7-5p, hsa-miR-629-5p, hsa-miR-199a-5p, hsa-let-7b-3p, hsa-miR-200b-3p, hsa-miR-148b-5p,
hsa-miR-1976,
hsa-miR-1284
and
AC C
hsa-miR-6747-3p,
hsa-miR-4433b-5p) were listed in Table 3. The read counts of these 10 miRNAs in each sample were listed in Supplemental Table S1.
Prediction of miRNA target genes It is well-known that discovering posttranscriptional gene silencing by miRNAs has led to an explosion of new hypotheses in human disease. To understand the
ACCEPTED MANUSCRIPT function of the top five differentially expressed miRNAs in BAVM, we predicted the target genes of the miRNAs using RNA22 v2 miRanda-mirSVR, miRDB, miRWalk, PICTAR2 and Targetscan 6.2, and counted miRNAs that were predicted. As a result,
RI PT
946 genes were found to be targeted by the top five differentially expressed miRNAs. Totally, we obtained 985 miRNA-target gene pairs. Based on the miRNA-target gene pairs, a BAVM-specific miRNA-target gene regulatory network was established
SC
(Figure 2). In the network, the top three miRNAs which regulated the most target
M AN U
genes were hsa-miR-7-5p, hsa-miR-199a-5p and hsa-miR-200b-3p.
Functional enrichment analysis of miRNA targets
GO enrichment analysis revealed that the miRNA targets were significantly
TE D
enriched in GO terms of cellular response to growth factor stimulus, response to growth factor and enzyme linked receptor protein signaling pathway for biological process (Figure 3A). For cellular component, the miRNA targets were significantly
EP
enriched in GO terms of intracellular part, intracellular and membrane-bounded
AC C
organelle (Figure 3B). For molecular function, the miRNA targets were significantly enriched in GO terms of protein binding, enzyme binding and organic cyclic compound binding (Figure 3C). KEGG pathway enrichment analysis identified 43 pathways that could be
involved in regulation of BAVM. We listed the top 25 significantly enriched pathways in Figure 4. Among these pathways, VEGF signaling pathway is crucial for both physiologic and pathologic angiogenesis. The binding of VEGF to VEGFR-2 leads to
ACCEPTED MANUSCRIPT a cascade of different signaling pathways, resulting in the up-regulation of genes involved in mediating the proliferation and migration of endothelial cells and promoting their survival and vascular permeability. The miRNA target genes in VEGF
RI PT
signaling pathway were further showed in Figure 5. The VEGF signaling pathway is involved in 15 of the genes, which are targeted by the hsa-miR-7-5p, hsa-miR-199a-5p and hsa-miR-200b-3p (Figure 5). Moreover, Pathway of focal
SC
adhesion plays essential roles in the considerable crosstalk between adhesion- and
M AN U
growth factor-mediated signalling.
Validation of miRNA and target expression
The peripheral blood of patients with BAVMs and normal samples were used to
TE D
verify the expression patterns of miRNAs and targets, including five miRNAs (miR-7-5p, miR-629-5p, miR-199a-5p, let-7b-3p, miR-200b-3p) and two target genes (VEGFA and KDR). We have performed the quality control data for U6 miRNA.
EP
Firstly, we calculated the mean 2-CT. The fold change of U6 miRNA in the internal
AC C
control in the BAVM samples compared to the control samples was: 5.59966E-05/6.22226E-05=0.89994 (p-value=0.10689) which suggested that human U6 gene was a good internal control gene in this study28. The results indicated that the expressions of miR-199a-5p, let-7b-3p and miR-200b-3p were consistent with next-generation sequencing. Moreover, VEGFA and KDR were significantly up-regulated in BAVMs compared with normal samples. (Figure 6)
ACCEPTED MANUSCRIPT Discussion To the best of our knowledge, this is the first study to present the global miRNA expression profiling in peripheral blood of BAVMs. In the current study,
RI PT
next-generation sequencing was used to identify the miRNA expression profiling in a paired design comparing BAVMs and normal controls, which showed a total of 246 differentially expressed miRNAs, which were all up-regulated in bloods of BAVMs
SC
and revealed that the aberrantly expressed miRNAs may aberrantly regulate target
M AN U
genes. Based on the qRT-PCR validation, three miRNAs (has-let-7b-3p, has-miR-200b-3p and has-miR-199a-5p) were up-regulated in BAVM which were consistent with our RNA-sequencing results generally and increased credibility of our RNA-sequencing results. However, the expression of let-7b-3p and miR-200b-3p in
TE D
BAVM in qRT-PCR validation were inconsistent with the results in RNA-sequencing which needs further research.
Considering that miRNAs work as small guide molecules in RNA silencing by
EP
degrading their mRNA target and/or by silencing translation 29, we suggested the gene
AC C
expression profiling may tend to be down-regulated in bloods of BAVMs. Interestingly, a previous study suggested that unruptured BAVMs patients tended to have a greater proportion of genes down-regulated
30
. Our results seemed to support
that the differential miRNA expression in BAVMs may be responsible for both physiologic and pathologic angiogenesis.Functional annotation of the target genes regulated by differentially expressed miRNAs supported the role for MAPK, Wnt and VEGF signaling in BAVMs. Previous study reported that the MAPK pathway has
ACCEPTED MANUSCRIPT been implicated in cerebral cavernous malformation
31
. Moreover, Wnt signaling
influences vascular development 32. In particular, angiogenesis plays a key role in the formation and evolution of BAVMs. The VEGF acts as a vascular endothelial
RI PT
cell-specific mitogen and is a major regulator of angiogenesis, which is expressed abundantly during embryonic development but is absent in most normal adult vascular beds, including the cerebral vasculature
33
. The VEGF family plays a key
SC
role in vasculogenesis and angiogenesis, and an experiment on animal indicated that
M AN U
VEGF pathway genes were dysregulated in endothelial cells derived from BAVMs, contributing to aberrant angiogenic characteristics and the pathogenesis of vascular malformations
34
. VEGF signaling in angiogenesis was involved in pathogenesis of
cerebral cavernous malformations and hereditary hemorrhagic telangiectasia
35
. Our
TE D
data strongly supported that differentially expressed miRNAs in peripheral blood of BAVMs may target the VEGF pathway, contributing to the BAVM phenotype. The VEGF signaling pathway is involved in 15 of the genes, which are targeted
EP
by the hsa-miR-7-5p, hsa-miR-199a-5p and hsa-miR-200b-3p. Prior study showed
AC C
that miR-7-5p was significant distinct in the peripheral blood of patients with glioblastoma pathologies 36. Another study also showed a significant down-regulation of miR-7-5p in glioblastoma microvasculature comparing with normal brain capillaries. Moreover, overexpression of miR-7-5p significantly inhibited human umbilical vein endothelial cell proliferation, which suggested that miR-7-5p might be an angiogenesis inhibitor of microvascular proliferation in glioblastoma
37
. Our
present study showed that miR-7-5p was significantly up-regulated in peripheral
ACCEPTED MANUSCRIPT blood of BAVMs and its target genes involved in VEGF signaling pathway, suggesting its crucial role of miR-7-5p in the pathogenesis of BAVMs. Chan et al provided the first evidence that induction of miR-199a-5p in human
RI PT
dermal microvascular endothelial cells blocked angiogenic response of endothelial cells in Matrigel culture, whereas miR-199a-5p-deprived cells exhibited enhanced angiogenesis in vitro
38
. Moreover, Zhang et al determined that miR-199a-5p acts as
SC
an endogenous inhibitor of SIRT1/eNOS to strengthen angiogenesis and involved in
39
M AN U
the pathological angiogenesis induced by human cytomegalovirus (HCMV) infection . In the present study, we demonstrated the up-regulation of miR-199a-5p in
peripheral blood of BAVMs and targeted VEGF pathway, which may provide a potential miRNA-based mechanism in the pathological angiogenesis of BAVMs.
TE D
Through microarray analysis, Shi et al found that miR-200b-3p was overexpressed in aortic tissue from degenerative aortic stenosis (AS) patients
40
. Our
present investigation provided novel evidence that up-regulation of miR-200b-3p in
EP
peripheral blood may be involved in the BAVMs. As a limitation of our study, the
AC C
VEGF pathway affected by miRNAs and which cause BAVMs remain elusive. Further studies are required to functionally characterize the role of specific candidate miRNAs.
Besides, our result identified a number of novel dysregulated miRNAs in in
peripheral blood of BAVMs with unknown function, such as miR-629-5p, let-7b-3p, miR-6747-3p,
hsa-miR-148b-5p,
hsa-miR-1976,
hsa-miR-1284
and
hsa-miR-4433b-5p. Further elucidation of the miRNA expression signature may
ACCEPTED MANUSCRIPT provide a better understanding of the mechanisms of BAVMs
41
. Taken together, our
study identified a miRNA expression signature in peripheral blood of patients with BAVMs, and the targeted pathway of VEGF signaling may have a crucial role in
RI PT
formation of BAVMs. We also suggested that three miRNAs (miR-7-5p, miR-199a-5p and miR-200b-3p) may serve as the potential diagnostic markers and anti-angiogenic therapies for BAVMs.
SC
However, age and gender differences between BAVM and controls and small
M AN U
sample size were limitations of this study. Further validation analysis with larger sample size and biological test were needed to confirm our finding. Moreover, follow-up study comparing miRNA signature between unruptured and rupture-prone BAVMs would be of great importance and would be performed in our following
Acknowledgement
TE D
research.
This study was supported by The study of molecular mechanisms of brain
EP
arteriovenous vascular malformation (No. 140072).
AC C
Competing interests
The authors declare that they have no competing interests.
References 1.
Komiyama
M.
Pathogenesis
of
brain
arteriovenous
malformations.
Neurologia
medico-chirurgica 2016;56:317-25. 2.
Yashar P, Amar AP, Giannotta SL, et al. Cerebral arteriovenous malformations: Issues of the interplay between stereotactic radiosurgery and endovascular surgical therapy. World neurosurgery 2011;75:638-47.
3.
van Beijnum J, van der Worp HB, Buis DR, et al. Treatment of brain arteriovenous
ACCEPTED MANUSCRIPT malformations: A systematic review and meta-analysis. Jama 2011;306:2011-9. 4.
Mohr JP, Parides MK, Stapf C, et al. Medical management with or without interventional therapy for unruptured brain arteriovenous malformations (aruba): A multicentre, non-blinded, randomised trial. Lancet 2014;383:614-21.
5.
Haw CS, terBrugge K, Willinsky R, Tomlinson G. Complications of embolization of arteriovenous malformations of the brain. Journal of neurosurgery 2006;104:226-32.
6.
Mullan S, Mojtahedi S, Johnson DL, Macdonald RL. Embryological basis of some aspects of
RI PT
cerebral vascular fistulas and malformations. Journal of neurosurgery 1996;85:1-8. 7.
Pawlikowska L, Poon KY, Achrol AS, et al. Apolipoprotein evarepsilon2 is associated with new hemorrhage risk in brain arteriovenous malformations. Neurosurgery 2006;58:838-43.
8.
Pawlikowska L, Tran MN, Achrol AS, et al. Polymorphisms in genes involved in inflammatory and angiogenic pathways and the risk of hemorrhagic presentation of brain
9.
SC
arteriovenous malformations. Stroke; a journal of cerebral circulation 2004;35:2294-300.
Chen Y, Zhu W, Bollen AW, et al. Evidence of inflammatory cell involvement in brain arteriovenous malformations. Neurosurgery 2008;62:1340-9; discussion 9-50.
10.
Chen Y, Pawlikowska L, Yao JS, et al. Interleukin-6 involvement in brain arteriovenous
11.
M AN U
malformations. Annals of neurology 2006;59:72-80.
Xu B, Wu YQ, Huey M, et al. Vascular endothelial growth factor induces abnormal microvasculature in the endoglin heterozygous mouse brain. Journal of cerebral blood flow and metabolism : official journal of the International Society of Cerebral Blood Flow and Metabolism 2004;24:237-44.
12.
Hao Q, Zhu Y, Su H, et al. Vegf induces more severe cerebrovascular dysplasia in endoglin than in alk1 mice. Translational stroke research 2010;1:197-201.
Mohr S, Liew CC. The peripheral-blood transcriptome: New insights into disease and risk
TE D
13.
assessment. Trends in molecular medicine 2007;13:422-32. 14.
Joint Writing Group of the Technology Assessment Committee American Society of I, Therapeutic N, Joint Section on Cerebrovascular Neurosurgery a Section of the American Association of Neurological S, et al. Reporting terminology for brain arteriovenous
EP
malformation clinical and radiographic features for use in clinical trials. Stroke; a journal of cerebral circulation 2001;32:1430-42. 15.
Xu H, Tang Y, Liu DZ, et al. Gene expression in peripheral blood differs after cardioembolic
AC C
compared with large-vessel atherosclerotic stroke: Biomarkers for the etiology of ischemic stroke. Journal of cerebral blood flow and metabolism : official journal of the International
Society of Cerebral Blood Flow and Metabolism 2008;28:1320-8.
16.
Langmead B, Trapnell C, Pop M, Salzberg SL. Ultrafast and memory-efficient alignment of short DNA sequences to the human genome. Genome biology 2009;10:R25.
17.
Friedlander MR, Mackowiak SD, Li N, Chen W, Rajewsky N. Mirdeep2 accurately identifies
known and hundreds of novel microrna genes in seven animal clades. Nucleic acids research 2012;40:37-52. 18.
Miranda KC, Huynh T, Tay Y, et al. A pattern-based method for the identification of microrna binding sites and their corresponding heteroduplexes. Cell 2006;126:1203-17.
19.
Betel D, Koppal A, Agius P, Sander C, Leslie C. Comprehensive modeling of microrna targets predicts functional non-conserved and non-canonical sites. Genome biology 2010;11:R90.
20.
Wang X. Mirdb: A microrna target prediction and functional annotation database with a wiki
ACCEPTED MANUSCRIPT interface. Rna 2008;14:1012-7. 21.
Dweep H, Sticht C, Pandey P, Gretz N. Mirwalk--database: Prediction of possible mirna binding sites by "walking" the genes of three genomes. Journal of biomedical informatics 2011;44:839-47.
22.
Krek A, Grun D, Poy MN, et al. Combinatorial microrna target predictions. Nature genetics 2005;37:495-500.
23.
Garcia DM, Baek D, Shin C, et al. Weak seed-pairing stability and high target-site abundance
RI PT
decrease the proficiency of lsy-6 and other micrornas. Nature structural & molecular biology 2011;18:1139-46. 24.
Hsu SD, Tseng YT, Shrestha S, et al. Mirtarbase update 2014: An information resource for experimentally validated mirna-target interactions. Nucleic acids research 2014;42:D78-85.
25.
Smoot ME, Ono K, Ruscheinski J, Wang PL, Ideker T. Cytoscape 2.8: New features for data
26.
SC
integration and network visualization. Bioinformatics 2011;27:431-2.
Ashburner M, Ball CA, Blake JA, et al. Gene ontology: Tool for the unification of biology. The gene ontology consortium. Nature genetics 2000;25:25-9.
Kanehisa M, Goto S. Kegg: Kyoto encyclopedia of genes and genomes. Nucleic acids research 2000;28:27-30.
28.
Schmittgen TD, Livak KJ. Analyzing real-time pcr data by the comparative c(t) method. Nat Protoc 2008;3:1101-8.
29.
M AN U
27.
Oliveto S, Mancino M, Manfrini N, Biffo S. Role of micrornas in translation regulation and cancer. World journal of biological chemistry 2017;8:45-56.
30.
Weinsheimer SM, Xu H, Achrol AS, et al. Gene expression profiling of blood in brain arteriovenous malformation patients. Translational stroke research 2011;2:575-87. Dejana E, Tournier-Lasserve E, Weinstein BM. The control of vascular integrity by
TE D
31.
endothelial cell junctions: Molecular basis and pathological implications. Developmental cell 2009;16:209-21. 32.
Franco CA, Liebner S, Gerhardt H. Vascular morphogenesis: A wnt for every vessel? Current opinion in genetics & development 2009;19:476-83. Nauck M, Karakiulakis G, Perruchoud AP, Papakonstantinou E, Roth M. Corticosteroids
EP
33.
inhibit the expression of the vascular endothelial growth factor gene in human vascular smooth muscle cells. European journal of pharmacology 1998;341:309-15. Jabbour MN, Elder JB, Samuelson CG, et al. Aberrant angiogenic characteristics of human
AC C
34.
brain arteriovenous malformation endothelial cells. Neurosurgery 2009;64:139-46; discussion
46-8.
35.
Kar S, Baisantry A, Nabavi A, Bertalanffy H. Role of delta-notch signaling in cerebral
cavernous malformations. Neurosurgical review 2016;39:581-9.
36.
Dong L, Li Y, Han C, et al. Mirna microarray reveals specific expression in the peripheral blood of glioblastoma patients. International journal of oncology 2014;45:746-56.
37.
Liu Z, Liu Y, Li L, et al. Mir-7-5p is frequently downregulated in glioblastoma microvasculature and inhibits vascular endothelial cell proliferation by targeting raf1. Tumour biology : the journal of the International Society for Oncodevelopmental Biology and Medicine 2014;35:10177-84.
38.
Chan YC, Roy S, Huang Y, Khanna S, Sen CK. The microrna mir-199a-5p down-regulation switches on wound angiogenesis by derepressing the v-ets erythroblastosis virus e26 oncogene
ACCEPTED MANUSCRIPT homolog 1-matrix metalloproteinase-1 pathway. The Journal of biological chemistry 2012;287:41032-43. 39.
Zhang S, Liu L, Wang R, et al. Mir-199a-5p promotes migration and tube formation of human cytomegalovirus-infected endothelial cells through downregulation of sirt1 and enos. Archives of virology 2013;158:2443-52.
40.
Shi J, Liu H, Wang H, Kong X. Microrna expression signature in degenerative aortic stenosis. BioMed research international 2016;2016:4682172. Xiao J, Liang D, Zhang Y, et al. Microrna expression signature in atrial fibrillation with mitral
AC C
EP
TE D
M AN U
SC
stenosis. Physiological genomics 2011;43:655-64.
RI PT
41.
ACCEPTED MANUSCRIPT Figure legends Figure 1 Heat map of top 50 differentially expressed miRNAs in BAVMs. Figure 2 The regulatory network between top five differentially expressed miRNAs
RI PT
and target genes in BAVMs. Red color represents the up-regulated miRNAs; Blue color represents target genes.
Figure 3 The significantly enriched GO terms of target genes regulated by the top five
M AN U
Cellular component; (C) Molecular functions.
SC
differentially expressed miRNAs in BAVMs. (A) Biological process; (B)
Figure 4 The significantly enriched KEGG pathways of target genes regulated by the top five differentially expressed miRNAs in BAVMs.
Figure 5 miRNAs and target genes involved in VEGF signaling pathway. (A) Target
TE D
genes regulated by the top five differentially expressed miRNAs in VEGF signaling; (B) miRNA-target pairs.
Figure 6 qRT-PCR validation of differentially expressed miRNAs and target genes in
EP
BAVMs. One-way ANOVA was used to calculate the significant differences
AC C
of miRNAs and genes between BAVM and controls. * indicated p-value <0.05.
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
Table 1 Primers used in qRT-PCR validation Gene Primers GAPDH Forward: 5′GGAGCGAGATCCCTCCAAAAT 3′ Reverse: 5′GGCTGTTGTCATACTTCTCATGG 3′ VEGFA Forward: 5′CTGTCTTGGGTGCATTGGAGC 3′ Reverse: 5′AGGGTCTCGATTGGATGGCAG 3′ KDR Forward: 5′GTGATCGGAAATGACACTGGAG 3′ Reverse: 5′CATGTTGGTCACTAACAGAAGCA 3′ hsa-miR-7-5p 5′UGGAAGACUAGUGAUUUUGUUGU 3′ hsa-miR-629-5p 5′UGGGUUUACGUUGGGAGAACU 3′ hsa-miR-199a-5p 5′CCCAGUGUUCAGACUACCUGUUC 3′ hsa-let-7b-3p 5′CTATACAACCTACTACTGCCTTCCC 3′ hsa-miR-200b-3p 5′UAAUACUGCCUGGUAAUGAUGA 3′
ACCEPTED MANUSCRIPT Table 2 Summary of sequencing reads in BAVM patients and controls
BAVM 1
Sex/age
Total reads
Male/26
mapped
9750342
868913
BAVM 2
Female/18 10407754
9604484
803270
BAVM 3
Female/36 10515006
9790850
724156
Male/16
11842374 11751866
90508
Control 2
Male/26
11829993 11737060
92933
Control 3
Male/11
11828494 11729217
M AN U
TE D
(%)
91.8
8.2
92.3
7.7
93.1
6.9
99.2
0.8
99.2
0.8
99.2
0.8
SC
Control 1
EP
(%)
unmapped
10619255
AC C
Unmapped
RI PT
Sample
Mapped
99277
ACCEPTED MANUSCRIPT Table 3 Top ten differentially expressed miRNAs in BAVM Value (BAVM) Value (Control) log2 (fold change)
FDR
hsa-miR-7-5p
6960
0
NA
0
hsa-miR-629-5p
4002
0
NA
0
hsa-miR-199a-5p
3497
0
NA
hsa-let-7b-3p
1444
0
NA
hsa-miR-200b-3p
1380
0
NA
hsa-miR-6747-3p
1156
0
hsa-miR-148b-5p
1035
hsa-miR-1976
793
hsa-miR-1284
773
SC
M AN U
NA
0
0
0
0
0
NA
0
0
NA
0
0
NA
0
0
NA
0
TE D
hsa-miR-4433b-5p 622
RI PT
Gene name
Value in BAVM and Control samples respectively represented the sum of read count of miRNAs in BAVM and Control samples based on RNA-sequencing results. FDR,
EP
false discovery rate was obtained by using DEGseq package, based on the default
AC C
parameter settings of DEGseq package, FDR=0 indicated that FDR was less than 1E-16.
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT 1. We firstly obtained the global miRNA expression profile in peripheral blood of BAVMs. 2. MAPK, Wnt and VEGF signaling were involve in the process of BAVMs.
RI PT
3. miR-7-5p, miR-199a-5p and miR-200b-3p may be potential diagnostic markers of
AC C
EP
TE D
M AN U
SC
BAVMs.
AC C
EP
TE D
M AN U
SC
Abbreviations AS, aortic stenosis BAVMs, brain arteriovenous malformations GO, Gene Ontology HCMV, human cytomegalovirus KEGG, Genes and Genomes miRNAs, microRNAs sRNA, small RNA
RI PT
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT Conflict of interest
AC C
EP
TE D
M AN U
SC
RI PT
The authors declare that they have no competing interests.