- Email: [email protected]
Long Noncoding RNAs and Their Regulatory Network: Potential Therapeutic Targets for Adult Moyamoya Disease
Accepted Manuscript Long non-coding RNAs and their regulatory network: Potential therapeutic targets for adult moyamoya disease Faliang Gao, M.D., Lanbing Yu, M.D., Dong Zhang, M.D., Yan Zhang, M.D., Rong Wang, M.D., Jizong Zhao, MD. PII:
S1878-8750(16)30355-2
DOI:
10.1016/j.wneu.2016.05.081
Reference:
WNEU 4142
To appear in:
World Neurosurgery
Received Date: 22 February 2016 Revised Date:
24 May 2016
Accepted Date: 25 May 2016
Please cite this article as: Gao F, Yu L, Zhang D, Zhang Y, Wang R, Zhao J, Long non-coding RNAs and their regulatory network: Potential therapeutic targets for adult moyamoya disease, World Neurosurgery (2016), doi: 10.1016/j.wneu.2016.05.081. 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
Title page: Title: Long non-coding RNAs and their regulatory network: Potential therapeutic targets for adult moyamoya disease 1,2,3,4
, Lanbing Yu, M.D.
1,2,3,4
, Dong Zhang, M.D.
1,2,3,4
,Yan Zhang, M.D.
1,2,3,4
,
RI PT
Faliang Gao, M.D.
Rong Wang, M.D. 1,2,3,4,Jizong Zhao, MD. 1,2,3,4 1
Department of Neurosurgery, Beijing Tiantan Hospital, Capital Medical University, Beijing, P. R.
SC
China; 2China National Clinical Research Center for Neurological Diseases, Beijing, P. R.
M AN U
China; 3Center of Stroke, Beijing Institute for Brain Disorders, Beijing, P. R. China; 4Beijing Key Laboratory of Translational Medicine for Cerebrovascular Diseases, Beijing, P. R. China;
Correspondence author: Ji-Zong Zhao MD, Department of Neurosurgery BeijingTiantan Hospital
Capital Medical University. NO.6 Tiantanxili, Dongcheng
TE D
District, Beijing, China, 100050. Fax: 0086-010-67096523 Tel: 0086-010-67096523. Email: [email protected] OR [email protected]
EP
Key words: moyamoya disease, long non-coding RNA, MAPK signaling pathway,
AC C
vascular disease, stroke
Abbreviations list:
MMD: moyamoya disease; lncRNA: long noncoding RNA MAPK signaling pathway: mitogen-activated protein kinase (MAPK) signaling pathway
ACCEPTED MANUSCRIPT IVH= intraventricular hemorrhage TIA = transient ischemic attack
AC C
EP
TE D
M AN U
SC
RI PT
mRS: modified Rankin scale
ACCEPTED MANUSCRIPT ABSTRACT Objective: To investigate long non-coding RNA (lncRNA) expression patterns in adult moyamoya disease (MMD) patients and explore their possible roles in the
RI PT
pathophysiology of MMD. Methods: A healthy control group (n = 10) and a MMD group (n = 15) were
evaluated. RNA was extracted from peripheral blood samples and hybridized to
SC
microarray to get lncRNAs expression profiles. Then predicted lncRNA target genes
M AN U
were identified, and bioinformatics analysis was performed to investigate their molecular functions. Results: In the MMD group, 3,649 lncRNAs exhibited over two-fold expression than their counterparts in the healthy control group; of these, 1,494 were up-regulated, while 2,155 were down-regulated. Principal component analysis and Hclust analysis
TE D
produced completely different clusters between the two groups. Gene Ontology and KEGG pathway enrichment analysis suggested that the differentially expressed lncRNAs regulate multiple signaling pathways that were related with inflammation
EP
and vascular disease, and mitogen-activated protein kinase (MAPK) signaling
AC C
pathway was the core regulatory pathway. Conclusions: Long noncoding RNA expression profiles were quite different between MMD and control groups. Multiple signaling pathways that were closely associated with immune response, vasculogenesis and smooth muscle contraction were indicated to participate in lncRNAs regulatory mechanism; Of these, MAPK signaling pathway, which has been well studied for the treatment of many other cardiovascular diseases, was the core of this regulatory-network. Our findings could help further understand
ACCEPTED MANUSCRIPT the pathophysiology of MMD and provide new potential therapeutic targets. Key words: moyamoya disease, long non-coding RNA, MAPK signaling pathway,
AC C
EP
TE D
M AN U
SC
RI PT
vascular disease, stroke
ACCEPTED MANUSCRIPT Introduction Moyamoya disease (MMD), a chronic, progressive cerebrovascular disease, is often characterized by stroke.1 The pathophysiology of MMD has been extensively studied;
which
has
been
a
major
breakthrough
for
RI PT
for example, it is confirmed that RNF213 is a susceptibility gene for the disease, understanding
of
MMD
pathophysiology.2,3 However, a deeper understanding of the pathophysiology of
SC
MMD is still needed. Long non-coding RNAs (lncRNAs) are defined as genomic
M AN U
transcripts longer than 200 nucleotides without protein coding function.It can exert regulatory roles at multiple levels and are generally considered to engage in almost all biological process, including X-chromosome inactivation, DNA methylation, as well as other transcriptional, post-transcriptional and epigenetic regulations.4,5 In recent
TE D
years, interest in lncRNAs has dramatically increased. Specific disease-related lncRNAs have been increasingly identified and a library of disease-specific lncRNAs has been established;6 Furthermore, some lncRNAs have been found to exert many
EP
important regulatory roles in vascular diseases.7 For example, MEG3 has a role in
AC C
angiogenesis during the brain ischemic process and inhibit angiogenesis as a tumor suppressor,8 MALAT1 is associated with endothelial cell function and vessel growth;9 ANRIL has significant involvement with atherosclerosis and it can modulate the expression of genes involved in cell proliferation, apoptosis and inflammatory response to impact in the risk of cardiovascular disease.10-13 Undoubtedly, the identification of specific lncRNAs has provided better understanding of the pathogeneses of the above-listed conditions and has identified novel therapeutic
ACCEPTED MANUSCRIPT targets to treat them. Whether lncRNAs play regulatory roles in MMD remains to be clarified. To date, it remains unknown whether MMD patients exhibit altered lncRNA expression profiles and, if so, whether these changes carry disease-related
RI PT
significance. Currently, techniques used to study long non-coding RNAs (lncRNAs) are initially based on bioinformatics analyses of microarrays (to identify potentially important
SC
lncRNAs and their regulatory mechanisms), followed by in vivo and in vitro
M AN U
experiments to verify functional mechanisms. In the present study, we tried to evaluate lncRNA expression changes in peripheral blood samples collected from MMD patients; and we further attempted to explore the possible regulatory mechanisms of lncRNAs in MMD pathophysiology and the core signaling pathway
TE D
involved. To these aims, lncRNAs microarray was used to get lncRNAs expression profiles in MMD and control groups; then differentially-expressed lncRNAs were identified and their target genes were predicted and input into the bioinformatics
EP
database to analyze their molecular functions and the potential pathways involved.
AC C
Finally, pathway-network modeling was built to explore the interactions among different pathways. To our knowledge, this is the first report, on moyamoya disease with lncRNA and its regulatory network studied. Materials and methods
Patient samples and RNA extraction The study population included 10 healthy volunteers and 15 MMD patients. All MMD patients underwent cerebral digital subtraction angiography (DSA), and met
ACCEPTED MANUSCRIPT moyamoya diagnostic criterias.14 The exclusion criteria involved the presence of secondary moyamoya phenomenon caused by atherosclerosis, meningitis, Down syndrome, hyperthyroidism, neurofibromatosis, leptospiral infection, or prior
RI PT
skull-base radiation therapy. We excluded patients with hypertension or diabetes, as well as patients taking certain oral drugs, to avoid potential interference to the experimental results. All patients were admitted into hospital at least 2 weeks after
SC
initial clinical symptoms were identified and all the blood samples were collected
M AN U
before the operation. The clinical information of MMD patients was indicated in Table 1. The control group of healthy adults was of similar age(36.1±6.95 vs. 34.6± 6.10, control group vs. MMD group, P>0.05) and similar sex ratio(female/male:5/5 vs. 7/8, control group vs. MMD group) with MMD group. All patients and healthy
TE D
subjects were of an ethnically homogeneous Han Chinese origin.
Then 15 MMD patients were divided into 2 subgroups based on their initial symptoms: hemorrhagic group (7 patients) , ischemic group (8 patients, including 2
EP
patients suffered hemorrhage after initial ischemic symptoms, named Ischemic-hemorrhagic, IH
AC C
1 and IH2 ).The 2 subgroups of MMD were also of similar age(35±5.88 vs. 34.3±
6.27,hemorrhagic
group
vs.
ischemic
group,
P>0.05)
and
similar
sex
ratio(female/male:3/4 vs. 4/4, hemorrhagic group vs. ischemic group). The study was approved by the hospital ethics committee and was performed after obtaining written consent from all the patients and volunteers. For whole blood RNA collection, 2.5ml blood were collected in PAX gene tubes (PreAnalytiX, Hombrechtikon, Switzerland). Total RNA was extracted and purified
ACCEPTED MANUSCRIPT using PAXgeneTM
Blood RNA Kit (Cat# 762174, QIAGEN, GmBH, Germany)
following the manufacturer’s instructions and checked for a RIN number to inspect RNA integration by an Agilent Bio analyzer 2100 (Agilent technologies, Santa Clara,
RI PT
CA, US). Microarray and data analysis
Human 4x180 K lncRNA and mRNA arrays manufactured by Agilent Technologies
SC
(SantaClara, CA) represented all long transcript (63431 lncRNAs and 39887 mRNAs)
M AN U
in the human genome following the manufacturer’s instructions. Labeled cRNA were purified by RNeasy mini kit (Cat#74106, QIAGEN, GmBH, Germany). Each Slide was hybridized with 1.65µg Cy3-labeled cRNA using Gene Expression Hybridization Kit (Cat#5188-5242, Agilent technologies, Santa Clara, CA, US) in Hybridization
TE D
Oven (Cat#G2545A, Agilent technologies, Santa Clara, CA, US), according to the manufacturer’s instructions. After 17 hours hybridization, slides were washed in staining dishes (Cat#121, Thermo Shandon, Waltham, MA, US) with Gene
EP
Expression Wash Buffer Kit(Cat#5188-5327, Agilent technologies, Santa Clara, CA,
AC C
US), followed the manufacturer s instructions. Slides were scanned by Agilent Microarray Scanner (Cat#G2565CA, Agilent technologies, Santa Clara, CA, US) with default settings, Dye channel: Green, Scan resolution=3µm, 20bit. Data were extracted with Feature Extraction software 10.7 (Agilent technologies, Santa Clara, CA, US). Raw data were normalized by Quantile algorithm, Gene Spring Software 12.6(Agilent technologies, Santa Clara, CA, US) Bioinformatics analysis
ACCEPTED MANUSCRIPT We used two independent algorithms to identify the targets of differentially expressed lncRNAs via cis- or trans-regulatory effects. The first algorithm searches for target genes acting in cis-lncRNAs and potential target genes were paired and visualized
RI PT
using UCSC (http://genome.ucsc.edu/) genome browser. The second algorithm is based on mRNA sequence complementarity and RNA duplex energy prediction. Then
predicted lncRNA target genes were input into the Database for Annotation,
SC
Visualization and Integrated Discovery (DAVID; http://david.abcc.ncifcrf.gov/),
the
KEGG
(Kyoto
M AN U
which utilized Gene Ontology (GO) to identify the molecular function. Next we used Encyclopedia
of
Genes
and
Genomes)
database
(http:www.genome.ad.jp/kegg/) to analyze the potential functions in the pathways. Pathway-network modeling was then performed to explore the interactions among
TE D
pathways. The pathway network was the interaction net of the significant pathways(p<0.05) of the differential expression genes using Fisher’s exact test , and was built according to the interaction among pathways of the KEGG database to find
AC C
Real time PCR
EP
the interaction among the significant pathways directly and systemically.
To verify the microarray results, we randomly choose 12 differentially expressed lncRNAs and detect their expression change between two groups. Total cellular RNA was isolated from all samples using PAXgene Blood RNA Kit Qiagen#762174and then reversely transcribed using iScript cDNA synthesis kit (BIO-RAD, USA) in accordance with the manufacturer’s instructions. Glyceralde-hyde 3-phosphate dehydrogenase (GAPDH) mRNA was used as an internal control. The 12 lncRNAs
ACCEPTED MANUSCRIPT and the primers (5' to 3: CATCCATCACTGTCCTTGTCA
ENST00000416385-R
GGGGATAACTGTGGCACCTA
ENST00000454968-F
GCAATGTTGTGCTCCTGTTAGTCT
ENST00000454968-R
TATCATGTGCCTCAGCCTGTTT
ENST00000409898-F
CCACCAGCCTCTCCTTGACA
ENST00000409898-R
GGATACAGTTCATTAAGGTTGGAAAAG
ENST00000409054-F
GATGAGAGCTGGACTTCTGTGTGT
ENST00000409054-R
GCGTGGGCTTAACTGATGCT
ENST00000453213-F
TGGAGATGGAAGGAGGATGCT
ENST00000453213-R
GTGGACTCGCACTGTCTGACA
SC
M AN U
CGAGCTGTAAAAGCCAAAGG
NR_033908-R
CCTGGGCGATAAGAGTGAAA
TE D
NR_033908-F
N344580-F TACCACGTCACCACCAACAC
EP
N344580-R CCTGAGCAGTTTGCATGGTA GGTGGCCTGACTTTTAGCTG
AC C
ENST00000565401-F
RI PT
ENST00000416385-F:
ENST00000565401-R
GCGGTTTCTTCCAAAATCAA
ENST00000408564-F
TCAAGCAGAGGCCTAAAGGA
ENST00000408564-R
TTCCCTACTGAGGTCCCAGA
NR_027252-F
ATCCCAGGGTAAGGAATGAAAGTA
NR_027252-R
GAGTTATGGTCACCTTGGAATGG
NR_015395-F
CTAATTTGCCACCACCCTGT
ACCEPTED MANUSCRIPT NR_015395-R
AAGACCCAGATGCCGTTTTA
NR_024420-F
CTGCAACGAATCCCAAAAGT
NR_024420-R
ACCACTTTCCAGAGGCTGAA
RI PT
Statistical analysis Statistical analysis was performed using Student’s t-test to compare two variables of microarray data. Statistical significance was determined at P<0.05, the raw data were
SC
normalized using the quantile algorithm in GeneSpring software 12.6 (Agilent
M AN U
technologies, CA, US). The genes with significant differential expression (a fold change >2) between the two groups were filtered using Student’s t-test (p < 0.05). Results
Differentially expressed long non-coding RNAs in moyamoya disease
TE D
Using authoritative data sources, we assessed lncRNA expression patterns in the MMD and control groups. Differentially expressed lncRNAs were identified using fold-change filtering (fold change > 2, P < 0.05). We found that 3,649 lncRNAs were
EP
differentially expressed (fold change > 2, P < 0.05) between the MMD and control
AC C
groups. Among them, 1,494 lncRNAs were up-regulated, and 2,155 lncRNAs were down-regulated in the MMD group (fold change > 2) (Figure 1 A and D). We next performed principal component analysis (PCA) and Hclust analysis to better understand the population separations in all 25 samples. PCA scatterplot showed that two distinct clusters formed, and the MMD and control groups showed excellent separation. However, 2 MMD subgroups did not show clear separation (Figure 1 B). The Hclust results showed that all 25 samples were separated into two populations
ACCEPTED MANUSCRIPT (MMD and control groups), and no obvious separation was noticed between the 2 subgroups (Figure 1 C). Real-time PCR validation
RI PT
Validation of the microarray results was conducted using RT-PCR.12 lncRNA probes were randomly selected for PCR amplification by applying specific primers. The PCR
results for all of the evaluated lncRNA probes were concordant with the microarray
SC
results. This agreement confirmed the accuracy of the microarray results (Figure 2).
M AN U
Potential functional targets
Although lncRNAs can theoretically exert both cis- and trans-regulation of target genes, current studies suggested the superiority of cis-regulation. In our results of GO and KEGG enrichment analysis revealed no more than 10 trans-regulated target genes
TE D
in each enriched GO and KEGG term (data not shown). Therefore, only the enrichment analysis results of the cis-regulated target genes were presented in the following GO and KEEG enrichment analysis.
EP
Gene ontology analysis
AC C
To identify the potential functions of the differentially expressed lncRNAs, we next performed GO analysis. This analysis revealed that the cis-regulated lncRNA target genes were enriched in the terms Molecular function, Biological process and Cellular component. The top 10 GO terms are listed (Table 2). Pathway analysis and pathway network The signaling pathways potentially regulated by the distinctly expressed lncRNAs were further investigated by performing enrichment analysis of lncRNA target genes
ACCEPTED MANUSCRIPT using the KEGG database. Sixteen most significantly enriched signaling pathways were identified (Table 3, p≤0.01). Surprisingly, many of the identified signaling pathways have close associations with immune response, vasculogenesis and smooth
RI PT
muscle contraction and these processes are the basis of the onset of MMD. Subsequently, a pathway network was constructed using the significantly enriched
signaling pathways (P<0.05). During pathway network construction, we attempted to
SC
reveal the interactions between each signaling pathway in the pathway network and to
M AN U
identify the network’s core signaling pathway. As shown in Figure 3, several signaling pathways can interact in this network, and the MAPK signaling pathway is the core pathway of the network, indicating that the MAPK pathway plays a pivotal role in the identified regulatory network.
TE D
Analyses of lncRNA target genes and mRNA intersections
We at last evaluated the intersections of the lncRNA target genes and the differentially expressed mRNAs. The genes in this intersection were identified based on two
EP
factors: (1) their potential regulation by the identified differentially expressed
AC C
lncRNAs in the bioinformatics analysis, and (2) their differential mRNA expression patterns in the microarray analysis. The microarray results contained several lncRNAs with multiple potential target genes that are difficult to verify individually; therefore, the identification of intersections helped narrow the exploration range to some degree. Interestingly, the effects of the MAPK signaling pathway were also detected (Gene Ontology analysis showed that 7 target gene were predicted in negative regulation of MAP kinase activity, P=9.52E-05).
ACCEPTED MANUSCRIPT Discussion In recent years, an increasing number of studies have been conducted to evaluate associations between non-coding RNAs and various diseases. From these studies,
RI PT
special databases, such as “The LncRNA and Disease Database” has been established to identify non-coding RNAs that are associated with specific diseases. In this
database, it contains more than 1000 lncRNA-disease entries. In 2014, Dai et al firstly
SC
reported miRNA expression profile changes in the sera of patients with MMD.15
M AN U
lncRNAs have longer base chains than miRNAs and have many regulatory roles outside of post transcriptional regulation. Therefore, lncRNAs are generally considered to possess more extensive and complex regulatory roles than miRNAs.4-6 Unfortunately, the association between lncRNAs and MMD has not been previously
TE D
reported. Based on the substantial regulatory roles of lncRNAs in many vascular diseases,7 the present study aimed to identify alterations in lncRNA expression in
pathophysiology.
EP
patients with MMD and to explore the significance of these changes in MMD
AC C
For ethical issues, we could not obtain moyamoya vessels for analysis from either direct or indirect revascularization. Therefore, peripheral blood (primarily white blood cells) was chosen as our research material, consistent with most recent studies on MMD. Fifteen cases of MMD, including 7 hemorrhagic type and 8 ischemic type were included. The current study primarily focused on differences in the hemodynamics of these types. For example, anterior choroid artery dilation is an important risk factor for hemorrhagic MMD.16 No current study has suggested that
ACCEPTED MANUSCRIPT different molecular mechanisms underlie the two subtypes; thus, 15 cases were included to uncover possible clues regarding their molecular mechanisms. More importantly, a common question of many studies evaluating disease-related lncRNAs
RI PT
is whether differential lncRNA expression is the "cause" of a disease or the "result" of internal environmental changes due to the disease. Our PCA and Hclust results showed that the lncRNA profiles of the 2 subtypes did not form distinct clusters. It is
SC
well known that these 2 subtypes produced substantially different internal blood
M AN U
environments; Given the information that the 2 subgroups of MMD were of similar age and sex ratio and DSA Suzuki stage, as shown in table 1, the results indicated that lncRNA expression may not be influenced by the internal environment created by different MMD subgroups. The changes in lncRNA expression patterns identified in
TE D
the present study are more likely to be the cause rather than the result of MMD. As mentioned above, we used whole blood samples for this analysis; however, after the experimental procedures, the lncRNAs were mostly isolated from white blood
EP
cells. We know that lumen stenosis caused by hyperplasia of intimal smooth muscle
AC C
cells is an important cause of MMD, in which the leukocyte-mediated inflammatory response plays an important role.17,18 In fact, congenital factors and acquired vascular inflammatory response are currently believed to be the two most important aspects of MMD pathophysiology.19 Pathway analysis showed that MAPK, VEGF, and smooth muscle signaling pathways were identified as important signaling pathways which have close associations with immune response, vasculogenesis and smooth muscle contraction and these processes are the basis of the onset of MMD. The results
ACCEPTED MANUSCRIPT demonstrated that several signaling pathways might participate in regulating the onset and progression of MMD; however, too many signaling pathways were identified to provide a clear causal relationship. It is difficult to verify this number of signaling
RI PT
pathways individually or to perform intervention measures to inhibit the onset of MMD.
Therefore, we constructed a pathway network with the significantly enriched
SC
signaling pathways (P < 0.05) to identify the core signaling pathway within the
M AN U
network. The results demonstrated that the MAPK signaling pathway is the core of this network. Many studies have demonstrated the important roles of the MAPK signaling pathway in vascular inflammation and other vascular physiological and pathophysiological processes, such as renal,20-22 pulmony23-26 and cardiovascular
TE D
stenosis and remodeling,27 and many specific MAPK inhibitors have been studied to prevent the MAPK-induced biological process.22,28,29 As written in the Guidelines for Diagnosis and Treatment of Moyamoya Disease: “Degeneration of the smooth muscle
EP
cells in the media and the resultant death of the vascular smooth muscle cells cause
AC C
thinning of the media” and “These changes noted in the terminal portion of the internal carotid arteries suggest the possibility of similar occurrence in the systemic arteries”,14 it indicates a similar pathology in MMD and systemic stenosis. Moreover, it was also noted in the guidelines that “some transcription and growth factors, such as bFGF are implicated in the migration and proliferation of vascular smooth muscle cells”,14 where studies have demonstrated that in these process activation of the MAPK signaling pathway is of vital importance. In fact, the MAPK signaling
ACCEPTED MANUSCRIPT pathway is the major mediator for these cytokines to induce the proliferation of vascular smooth muscle cells.30-32 To date, a definite correlation between a specific signaling pathway and MMD
RI PT
pathophysiology has not been reported, and the potential therapeutic targets remain unknown. lncRNAs have multiple action modes and a quite complicated regulatory network, the identification of multiple distinctly expressed lncRNAs is challenging, as
SC
is verifying the numerous associated signaling pathways. Due to the important role of
M AN U
the MAPK signaling pathway in the process of systemic artery stenosis22,26,27 and atherosclerosis,33-37 focusing on this signaling pathway narrowed the scope of our study. Given the fact that MAPK signaling pathway inhibitors had been successfully used to treat atherosclerosis in animal experiments,35,37 our results provide leads to
therapeutic targets.
TE D
further understand the pathophysiology of MMD and may help identify new available
In our study, a single-channel, dual-labeled DNA microarray that contains probes for
EP
all currently known lncRNAs and mRNAs was employed. Thus, the expression
AC C
profiles of all known lncRNAs and mRNAs were simultaneously obtained. Differentially expressed lncRNAs were screened, and their target mRNAs were predicted. These mRNAs were then compared with the identified differentially expressed mRNAs to reveal intersections between the two. Following this, we performed GO of this intersection. Interestingly, we also noted the involvement of the MAPK signaling pathway. Because lncRNAs function at multiple levels in addition to regulating mRNA, the lncRNAs involved in the identified intersections would
ACCEPTED MANUSCRIPT constitute only a small portion of those of significance. Nevertheless, analysis of such intersections may enable further narrowing of the scope of future explorations. Cerebral Revascularization (direct or indirect) has always been the primary method to
RI PT
treat Moyamoya disease (MMD). Even with the improvement of surgical techniques and plans, and more scientific standards for the case control, the failure rate of this
operation is still 20-30%, combined with 10-20% severe operative complications.38
SC
Until now, there is no specific treatment to prevent MMD progression. With a better
M AN U
understanding of MMD pathophysiology, non-surgical approaches targeting MMD pathophysiology may be available to stop or slow the progression of this disease39. The time of considering lncRNA just as "transcriptional noise" has passed. And in this entirely new field, the researches on lncRNA indicated the strong biological
TE D
regulatory effects in almost all the biological processes. Therefore, the lncRNA research in disease researches is of great importance with lncRNA itself and the targeted regulatory network as the new hope for the cure of relevant diseases.
EP
Our study still has limitations, and there remains future work to be done. First,
AC C
although this is the first lncRNA study of moyamoya disease (MMD), because of the limited sample size, specific lncRNAs and core members of the MAPK signalling pathway could not be verified. The number of genes involved is large, which increases the difficulty for the next research. The identification of certain specific lncRNAs or a smaller panel of lncRNAs should be the focus of future large-sample studies. Second, our results were based on bioinformatics analyses and require further confirmation by in vitro and in vivo experiments. We note that aberrant angiogenesis
ACCEPTED MANUSCRIPT is an active angiogenetic process that causes stenosis through the proliferation of vascular progenitor cells (including both endothelial and smooth muscle progenitor cells) 39. By targeting specific lncRNAs and MAPK signalling molecules in these
RI PT
cells, future experiments could determine whether the abnormal proliferation process could be blocked. Finally, it would also be helpful to clarify the relationship between
lncRNAs and other key molecules (for example, miRNAs and RNF213) and their
SC
participation in MMD pathophysiology. In this study, we have taken the first step
M AN U
towards investigating the role of lncRNAs in MMD, but future work is needed to investigate its practical application in treating this disease. Conclusion
Long noncoding RNA expression profiles are quite different in MMD group and
TE D
produced completely different clusters. Multiple signaling pathways that are closely associated with immune response, vasculogenesis and smooth muscle contraction were indicated to participate in lncRNAs regulatory mechanism, of these,
EP
mitogen-activated protein kinase (MAPK) signaling pathway was found to have a
AC C
core role in this pathway-network. lncRNAs and MAPK signaling pathway should be further studied as new therapeutic targets in future. Conflict of interest
The authors report no conflict of interest concerning the materials or methods used in this study or the findings specified in this paper. Study Funding Supported by the National Natural Science Foundation of China (81371292) and the
ACCEPTED MANUSCRIPT “13th Five-Year Plan” National Science and Technology supporting plan (2015BAI09B04), Acknowledge
RI PT
The authors thank Xiaona Zhang and Weiping Yu for the data collection and analysis. The authors also thank the research nurses, Donghong Zhao, Xi Ren, Na Shi and
References 1.
Suzuki J, Takaku A. Cerebrovascular moyamoya disease: disease showing abnormal net-like
vessels in base of brain. Arch Neurol 1969;20:288-299.
Cecchi AC, Guo D, Ren Z, et al. RNF213 rare variants in an ethnically diverse population with
M AN U
2.
SC
Xiaorong Sun for their assistance in the clinical collection of blood samples
Moyamoya disease. Stroke 2014;45:3200-3207. 3.
Mineharu Y, Takagi Y, Takahashi JC, et al. Rapid progression of unilateral moyamoya disease in a
patient with a family history and an RNF213 risk variant. Cerebrovasc Dis 2013;36:155-157. 4.
Mathieu E-L, Belhocine M, Dao L, Puthier D, Spicuglia S. [Functions of lncRNA in development and
diseases]. Med Sci M/S 2013;30:790-796.
Esteller M. Non-coding RNAs in human disease. Nat Rev Genet 2011;12:861-874.
6.
Taft RJ, Pang KC, Mercer TR, Dinger M, Mattick JS. Non-coding RNAs: regulators of disease. J
Pathol 2010;220:126-139. 7.
TE D
5.
Schonrock N, Harvey RP, Mattick JS. Long noncoding RNAs in cardiac development and
pathophysiology. Circ Res 2012;111:1349-1362. 8.
Gordon FE, Nutt CL, Cheunsuchon P, et al. Increased expression of angiogenic genes in the brains
of mouse meg3-null embryos. Endocrinology 2010;151:2443-2452. Michalik KM, You X, Manavski Y, et al. Long noncoding RNA MALAT1 regulates endothelial cell
EP
9.
function and vessel growth. Circ Res 2014;114:1389-1397. 10. Holdt LM, Beutner F, Scholz M, et al. ANRIL expression is associated with atherosclerosis risk at
AC C
chromosome 9p21. Arterioscler Thromb Vasc Biol 2010;30:620-627. 11. Congrains A, Kamide K, Oguro R, et al. Genetic variants at the 9p21 locus contribute to atherosclerosis through modulation of ANRIL and CDKN2A/B. Atherosclerosis 2012;220:449-455. 12. Congrains A, Kamide K, Katsuya T, et al. CVD-associated non-coding RNA, ANRIL, modulates expression of atherogenic pathways in VSMC. Biochem Biophys Res Commun 2012;419:612-616. 13. Holdt LM, Hoffmann S, Sass K, et al. Alu elements in ANRIL non-coding RNA at chromosome 9p21 modulate atherogenic cell functions through trans-regulation of gene networks. PLOS Genet 2013;9:e1003588. 14. Hashimoto N, Tominaga T, Miyamoto S, et al. Guidelines for diagnosis and treatment of moyamoya disease (spontaneous occlusion of the circle of Willis). Neurol Med Chir 2012;52:245-266. 15. Dai D, Lu Q, Huang Q, et al. Serum miRNA signature in Moyamoya disease. PLoS One 2014;9:e102382.
ACCEPTED MANUSCRIPT 16. Ryan RW, Chowdhary A, Britz GW. Hemorrhage and risk of further hemorrhagic strokes following cerebral revascularization in Moyamoya disease: a review of the literature. Surg Neurol Int 2012;3:72. 17. Takekawa
Y,
Umezawa
T,
Ueno
Y,
Sawada
T,
Kobayashi
M.
Pathological
and
immunohistochemical findings of an autopsy case of adult moyamoya disease. Neuropathology 2004;24:236-242. 18. Gosalakkal J. Moyamoya disease: a review. Neurol India 2002;50:6. 19. Weinberg DG, Arnaout OM, Rahme RJ, Aoun SG, Batjer HH, Bendok BR. Moyamoya disease: a
RI PT
review of histopathology, biochemistry, and genetics. Neurosurg Focus 2011;30:E20.
20. Park KM, Chen A, Bonventre JV. Prevention of kidney ischemia/reperfusion-induced functional injury and JNK, p38, and MAPK kinase activation by remote ischemic pretreatment. J Biol Chem 2001;276:11870-11876.
21. Chen H, Brahmbhatt S, Gupta A, Sharma AC. Duration of streptozotocin-induced diabetes
differentially affects p38-mitogen-activated protein kinase (MAPK) phosphorylation in renal and
SC
vascular dysfunction. Cardiovasc Diabetol 2005;4:3-3.
22. Wang D, Warner GM, Yin P, et al. Inhibition of p38 MAPK attenuates renal atrophy and fibrosis in a murine renal artery stenosis model. Am J Physiol Renal Physiol 2013;304:F938-F947.
M AN U
23. Hashimoto S, Matsumoto K, Gon Y, et al. p38 mitogen-activated protein kinase regulates IL-8 expression in human pulmonary vascular endothelial cells. Eur Respir J 1999;13:1357-1364. 24. Morrell NW, Upton PD, Kotecha S, et al. Angiotensin II activates MAPK and stimulates growth of human pulmonary artery smooth muscle via AT1 receptors. Am J Physiol Lung Cell Mol Physiol 1999;277:L440-L448.
25. Nick JA, Young SK, Brown KK, et al. Role of p38 mitogen-activated protein kinase in a murine model of pulmonary inflammation. J Immunol 2000;164:2151-2159.
TE D
26. Urashima T, Zhao M, Wagner RA, et al. Molecular and physiologic characterization of RV remodeling in a murine model of pulmonary stenosis. Am J Physiol Heart Circ Physiol 2008;295:H1351-H1368.
27. Nordmeyer J, Eder S, Mahmoodzadeh S. Upregulation of myocardial estrogen receptors in human aortic stenosis. Circulation 2004;110:3270-3275.
EP
28. Öllinger R, Bilban M, Erat A. Bilirubin: A natural inhibitor of vascular smooth muscle cell proliferation. Circulation 2005;112:1030-1039. 29. Underwood DC, Osborn RR, Bochnowicz S, et al. SB 239063, a p38 MAPK inhibitor, reduces
AC C
neutrophilia, inflammatory cytokines, MMP-9, and fibrosis in lung. Am J Physiol Lung Cell Mol Physiol 2000;279:L895-L902.
30. Cui Y, Sun Y, Lin H, et al. Platelet-derived growth factor-BB induces matrix metalloproteinase-2 expression and rat vascular smooth muscle cell migration via ROCK and ERK/p38 MAPK pathways. Mol Cell Biochem 2014;393:255-263. 31. Suwanabol PA, Seedial SM, Shi X, et al. Transforming growth factor-β increases vascular smooth muscle
cell
proliferation
through
the
Smad3
and
extracellular
signal-regulated
kinase
mitogen-activated protein kinases pathways. J Vasc Surg 2012;56:446-454. 32. Martin-Garrido A, Williams HC, Lee M, et al. Transforming growth factor β inhibits platelet derived growth factor-induced vascular smooth muscle cell proliferation via Akt-independent, Smad-mediated cyclin D1 downregulation. PLoS One 2013;8:e79657.
ACCEPTED MANUSCRIPT 33. Hu Y, Sun B, Liu K, et al. Icariin attenuates high-cholesterol diet induced atherosclerosis in rats by inhibition of inflammatory response and p38 MAPK signaling pathway. Inflammation Epub 2015 Aug 26. 34. Tiwari RL, Singh V, Barthwal MK. Macrophages: an elusive yet emerging therapeutic target of atherosclerosis. Med Res Rev 2008;28:483-544. 35. Rahaman SO, Lennon DJ, Febbraio M, Podrez EA, Hazen SL, Silverstein R. A CD36-dependent signaling cascade is necessary for macrophage foam cell formation. Cell Metab 2006;4:211-221.
RI PT
36. Ricci R, Sumara G, Sumara I. Requirement of JNK2 for scavenger receptor A-mediated foam cell formation in atherogenesis. Science 2004;306:1558-1561.
37. Zhao M, Liu Y, Wang X, New L, Han J, Brunk UT. Activation of the p38 MAP kinase pathway is required for foam cell formation from macrophages exposed to oxidized LDL. APMIS 2002;110:458-468.
38. Kim D, Huh P, Kim H, et al. Surgical treatment of moyamoya disease in adults: combined direct
SC
and indirect vs. indirect bypass surgery. Neurol Med Chir 2012;52:333-338.
39. Bang OY, Fujimura M and Kim S-K. The pathophysiology of moyamoya disease: an update. J stroke.
M AN U
2016; 18(1):12-20.
Figure legends
TE D
Figure 1.lncRNA expression profile in MMD and control group.A.Scatter-plot comparison of lncRNAs expression levels between MMD and control group. The red dots of the scatter plot in the upper left corner represented up-regulated lncRNAs (fold
EP
change >2,P<0.05), and the green dots in low right corner represented down-regulated
AC C
lncRNAs (fold change <0.5, P<0.05) B. Principal Component Analysis (PCA) of lncRNAs in MMD and control group. The PCA scatter plot showed that the two groups (MMD and control) is the best separation and formed two distinct clusters. However, the 2 subgroups of MMD were not able to separated.Black :control group.Red:Hemorrhagic
group.Green
and.Blue:Ischemic
group.C.
HClust
of
lncRNAs among 25 samples. The results showed 25 samples were separated into two populations (MMD and control group), no obvious separation was noticed in MMD
ACCEPTED MANUSCRIPT group between MMD subgroups. D. Heat map showed the differentially expressed genes between two groups. Red color indicates over expressed genes, whereas green color indicates the opposite. The color scale bar is shown.
RI PT
Figure 2. Real time PCR was performed for validation of microarray results. Using specific primers, all the randomly selected 12 lncRNAs PCR results were consistent with microarray results.
SC
Figure. 3 Pathway network of differentially-expressed lncRNAs target genes.
M AN U
Cycle nodes represent pathways, the arrow between two nodes represents an interaction target between pathways. The more edges of a pathway, the more
AC C
EP
TE D
pathways connecting to it, and the more central role it has within the network.
ACCEPTED MANUSCRIPT Table 1. Clinical characteristics of 15 MMD patients No.
age
sex
initial clinical findings
Suzuki stage
mRS score
2
hemorrhagic group 19
female
IVH
rt2 lt2
2
32
female
IVH
rt4 lt4
1
3
33
male
IVH
rt2 lt2
2
4
47
male
IVH
rt5 lt5
1
5
44
male
IVH
6
28
male
IVH
7
41
female
female
2
21
male
3
43
male
4
20
male
5
21
female
6
42
7
SC rt3 lt4
2
rt3 lt3
TIA
2
2
rt3 lt3
2
infarction
rt5 lt5
2
infarction
rt4 lt4
1
TIA
rt3 lt3
3
female
infarction
rt2 lt4
2
20
male
infarction
rt4 lt4
1
46
female
infarction
rt3 lt2
1
TE D
TIA
AC C
8
3
EP
29
rt3 lt3
M AN U SAH
ischemic group 1
RI PT
1
†:IVH= intraventricular hemorrhage; TIA = transient ischemic attack; lt = left; rt = right;mRs= modified Rakkin score
ACCEPTED MANUSCRIPT Table 2 Enriched Go terms for potential functional targets (cis-effect, top 10,) GO
term
type
Target Gene
P value
in GO Molecular function
G-protein coupled receptor signaling pathway
Biological process
69
7.20 E-05
olfactory receptor activity
Molecular function
27
8.51 E-05
detection of chemical stimulus involved in sensory
Biological process
27
1.01 E-04
Cellular component
40
9.58 E-04
Biological process
23
9.99 E-04
Molecular function
925
1.14 E-03
Biological process
8
1.87 E-03
Molecular function
36
1.88 E-03
Biological process
22
2.09 E-03
cell projection
proteasome-mediated ubiquitin-dependent protein catabolic process
TE D
protein binding sterol metabolic process
EP
RNA polymerase II core promoter proximal region sequence
SC
M AN U
perception of smell
AC C
cellular response to lipopolysaccharide
47
3.40 E-05
RI PT
G-protein coupled receptor activity
Table 3 Enriched significant pathways for potential functional targets (cis-effect, P≤0.01) ACCEPTED MANUSCRIPT KEGG name
Target gene in this pathway
All gene
P value
in this pathway
50
207
9.46 E-05
Insulin signaling pathway - Homo sapiens (human)
34
141
1.07 E-03
Neurotrophin signaling pathway - Homo sapiens
30
31
Non-small cell lung cancer - Homo sapiens (human)
16
M AN U
Dopaminergic synapse - Homo sapiens (human)
120
1.13 E-03
131
2.28 E-03
SC
(human)
RI PT
Focal adhesion - Homo sapiens (human)
56
3.63 E-03
Olfactory transduction - Homo sapiens (human)
33
405
4.39 E-03
MAPK signaling pathway - Homo sapiens (human)
51
257
5.84 E-03
Glioma - Homo sapiens (human)
17
65
7.08 E-03
16
60
7.19 E-03
Oxytocin signaling pathway - Homo sapiens (human)
34
159
7.30 E-03
Platelet activation - Homo sapiens (human)
29
131
7.78 E-03
ErbB signaling pathway - Homo sapiens (human)
21
87
8.08 E-03
Vascular smooth muscle contraction - Homo sapiens
27
121
8.93 E-03
43
215
9.30 E-03
FoxO signaling pathway - Homo sapiens (human)
29
133
9.54 E-03
NOD-like receptor signaling pathway - Homo
15
57
1.01 E-02
AC C
EP
TE D
mTOR signaling pathway - Homo sapiens (human)
(human)
Regulation of actin cytoskeleton - Homo sapiens (human)
sapiens (human)
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 Highlights
1. lncRNAs expression were quite different in adult moyamoya patients. 2. Multiple signaling pathways that were closely associated with immune response,
lncRNAs regulatory mechanism.
RI PT
vasculogenesis and smooth muscle contraction were predicted involved in
3. Pathway network analysis showed that the MAPK signaling pathway, which has
SC
been well studied for the treatment of many other cardiovascular diseases, was the core of lncRNAs regulatory-network. Our findings could help further understand
AC C
EP
TE D
M AN U
the pathogenesis of MMD and provide new potential therapeutic targets.
ACCEPTED MANUSCRIPT MMD: moyamoya disease; lncRNA: long noncoding RNA MAPK signaling pathway: mitogen-activated protein kinase (MAPK) signaling
RI PT
pathway IVH: intraventricular hemorrhage TIA : transient ischemic attack
AC C
EP
TE D
M AN U
SC
mRs: modified Rankin scale
S1878-8750(16)30355-2
DOI:
10.1016/j.wneu.2016.05.081
Reference:
WNEU 4142
To appear in:
World Neurosurgery
Received Date: 22 February 2016 Revised Date:
24 May 2016
Accepted Date: 25 May 2016
Please cite this article as: Gao F, Yu L, Zhang D, Zhang Y, Wang R, Zhao J, Long non-coding RNAs and their regulatory network: Potential therapeutic targets for adult moyamoya disease, World Neurosurgery (2016), doi: 10.1016/j.wneu.2016.05.081. 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
Title page: Title: Long non-coding RNAs and their regulatory network: Potential therapeutic targets for adult moyamoya disease 1,2,3,4
, Lanbing Yu, M.D.
1,2,3,4
, Dong Zhang, M.D.
1,2,3,4
,Yan Zhang, M.D.
1,2,3,4
,
RI PT
Faliang Gao, M.D.
Rong Wang, M.D. 1,2,3,4,Jizong Zhao, MD. 1,2,3,4 1
Department of Neurosurgery, Beijing Tiantan Hospital, Capital Medical University, Beijing, P. R.
SC
China; 2China National Clinical Research Center for Neurological Diseases, Beijing, P. R.
M AN U
China; 3Center of Stroke, Beijing Institute for Brain Disorders, Beijing, P. R. China; 4Beijing Key Laboratory of Translational Medicine for Cerebrovascular Diseases, Beijing, P. R. China;
Correspondence author: Ji-Zong Zhao MD, Department of Neurosurgery BeijingTiantan Hospital
Capital Medical University. NO.6 Tiantanxili, Dongcheng
TE D
District, Beijing, China, 100050. Fax: 0086-010-67096523 Tel: 0086-010-67096523. Email: [email protected] OR [email protected]
EP
Key words: moyamoya disease, long non-coding RNA, MAPK signaling pathway,
AC C
vascular disease, stroke
Abbreviations list:
MMD: moyamoya disease; lncRNA: long noncoding RNA MAPK signaling pathway: mitogen-activated protein kinase (MAPK) signaling pathway
ACCEPTED MANUSCRIPT IVH= intraventricular hemorrhage TIA = transient ischemic attack
AC C
EP
TE D
M AN U
SC
RI PT
mRS: modified Rankin scale
ACCEPTED MANUSCRIPT ABSTRACT Objective: To investigate long non-coding RNA (lncRNA) expression patterns in adult moyamoya disease (MMD) patients and explore their possible roles in the
RI PT
pathophysiology of MMD. Methods: A healthy control group (n = 10) and a MMD group (n = 15) were
evaluated. RNA was extracted from peripheral blood samples and hybridized to
SC
microarray to get lncRNAs expression profiles. Then predicted lncRNA target genes
M AN U
were identified, and bioinformatics analysis was performed to investigate their molecular functions. Results: In the MMD group, 3,649 lncRNAs exhibited over two-fold expression than their counterparts in the healthy control group; of these, 1,494 were up-regulated, while 2,155 were down-regulated. Principal component analysis and Hclust analysis
TE D
produced completely different clusters between the two groups. Gene Ontology and KEGG pathway enrichment analysis suggested that the differentially expressed lncRNAs regulate multiple signaling pathways that were related with inflammation
EP
and vascular disease, and mitogen-activated protein kinase (MAPK) signaling
AC C
pathway was the core regulatory pathway. Conclusions: Long noncoding RNA expression profiles were quite different between MMD and control groups. Multiple signaling pathways that were closely associated with immune response, vasculogenesis and smooth muscle contraction were indicated to participate in lncRNAs regulatory mechanism; Of these, MAPK signaling pathway, which has been well studied for the treatment of many other cardiovascular diseases, was the core of this regulatory-network. Our findings could help further understand
ACCEPTED MANUSCRIPT the pathophysiology of MMD and provide new potential therapeutic targets. Key words: moyamoya disease, long non-coding RNA, MAPK signaling pathway,
AC C
EP
TE D
M AN U
SC
RI PT
vascular disease, stroke
ACCEPTED MANUSCRIPT Introduction Moyamoya disease (MMD), a chronic, progressive cerebrovascular disease, is often characterized by stroke.1 The pathophysiology of MMD has been extensively studied;
which
has
been
a
major
breakthrough
for
RI PT
for example, it is confirmed that RNF213 is a susceptibility gene for the disease, understanding
of
MMD
pathophysiology.2,3 However, a deeper understanding of the pathophysiology of
SC
MMD is still needed. Long non-coding RNAs (lncRNAs) are defined as genomic
M AN U
transcripts longer than 200 nucleotides without protein coding function.It can exert regulatory roles at multiple levels and are generally considered to engage in almost all biological process, including X-chromosome inactivation, DNA methylation, as well as other transcriptional, post-transcriptional and epigenetic regulations.4,5 In recent
TE D
years, interest in lncRNAs has dramatically increased. Specific disease-related lncRNAs have been increasingly identified and a library of disease-specific lncRNAs has been established;6 Furthermore, some lncRNAs have been found to exert many
EP
important regulatory roles in vascular diseases.7 For example, MEG3 has a role in
AC C
angiogenesis during the brain ischemic process and inhibit angiogenesis as a tumor suppressor,8 MALAT1 is associated with endothelial cell function and vessel growth;9 ANRIL has significant involvement with atherosclerosis and it can modulate the expression of genes involved in cell proliferation, apoptosis and inflammatory response to impact in the risk of cardiovascular disease.10-13 Undoubtedly, the identification of specific lncRNAs has provided better understanding of the pathogeneses of the above-listed conditions and has identified novel therapeutic
ACCEPTED MANUSCRIPT targets to treat them. Whether lncRNAs play regulatory roles in MMD remains to be clarified. To date, it remains unknown whether MMD patients exhibit altered lncRNA expression profiles and, if so, whether these changes carry disease-related
RI PT
significance. Currently, techniques used to study long non-coding RNAs (lncRNAs) are initially based on bioinformatics analyses of microarrays (to identify potentially important
SC
lncRNAs and their regulatory mechanisms), followed by in vivo and in vitro
M AN U
experiments to verify functional mechanisms. In the present study, we tried to evaluate lncRNA expression changes in peripheral blood samples collected from MMD patients; and we further attempted to explore the possible regulatory mechanisms of lncRNAs in MMD pathophysiology and the core signaling pathway
TE D
involved. To these aims, lncRNAs microarray was used to get lncRNAs expression profiles in MMD and control groups; then differentially-expressed lncRNAs were identified and their target genes were predicted and input into the bioinformatics
EP
database to analyze their molecular functions and the potential pathways involved.
AC C
Finally, pathway-network modeling was built to explore the interactions among different pathways. To our knowledge, this is the first report, on moyamoya disease with lncRNA and its regulatory network studied. Materials and methods
Patient samples and RNA extraction The study population included 10 healthy volunteers and 15 MMD patients. All MMD patients underwent cerebral digital subtraction angiography (DSA), and met
ACCEPTED MANUSCRIPT moyamoya diagnostic criterias.14 The exclusion criteria involved the presence of secondary moyamoya phenomenon caused by atherosclerosis, meningitis, Down syndrome, hyperthyroidism, neurofibromatosis, leptospiral infection, or prior
RI PT
skull-base radiation therapy. We excluded patients with hypertension or diabetes, as well as patients taking certain oral drugs, to avoid potential interference to the experimental results. All patients were admitted into hospital at least 2 weeks after
SC
initial clinical symptoms were identified and all the blood samples were collected
M AN U
before the operation. The clinical information of MMD patients was indicated in Table 1. The control group of healthy adults was of similar age(36.1±6.95 vs. 34.6± 6.10, control group vs. MMD group, P>0.05) and similar sex ratio(female/male:5/5 vs. 7/8, control group vs. MMD group) with MMD group. All patients and healthy
TE D
subjects were of an ethnically homogeneous Han Chinese origin.
Then 15 MMD patients were divided into 2 subgroups based on their initial symptoms: hemorrhagic group (7 patients) , ischemic group (8 patients, including 2
EP
patients suffered hemorrhage after initial ischemic symptoms, named Ischemic-hemorrhagic, IH
AC C
1 and IH2 ).The 2 subgroups of MMD were also of similar age(35±5.88 vs. 34.3±
6.27,hemorrhagic
group
vs.
ischemic
group,
P>0.05)
and
similar
sex
ratio(female/male:3/4 vs. 4/4, hemorrhagic group vs. ischemic group). The study was approved by the hospital ethics committee and was performed after obtaining written consent from all the patients and volunteers. For whole blood RNA collection, 2.5ml blood were collected in PAX gene tubes (PreAnalytiX, Hombrechtikon, Switzerland). Total RNA was extracted and purified
ACCEPTED MANUSCRIPT using PAXgeneTM
Blood RNA Kit (Cat# 762174, QIAGEN, GmBH, Germany)
following the manufacturer’s instructions and checked for a RIN number to inspect RNA integration by an Agilent Bio analyzer 2100 (Agilent technologies, Santa Clara,
RI PT
CA, US). Microarray and data analysis
Human 4x180 K lncRNA and mRNA arrays manufactured by Agilent Technologies
SC
(SantaClara, CA) represented all long transcript (63431 lncRNAs and 39887 mRNAs)
M AN U
in the human genome following the manufacturer’s instructions. Labeled cRNA were purified by RNeasy mini kit (Cat#74106, QIAGEN, GmBH, Germany). Each Slide was hybridized with 1.65µg Cy3-labeled cRNA using Gene Expression Hybridization Kit (Cat#5188-5242, Agilent technologies, Santa Clara, CA, US) in Hybridization
TE D
Oven (Cat#G2545A, Agilent technologies, Santa Clara, CA, US), according to the manufacturer’s instructions. After 17 hours hybridization, slides were washed in staining dishes (Cat#121, Thermo Shandon, Waltham, MA, US) with Gene
EP
Expression Wash Buffer Kit(Cat#5188-5327, Agilent technologies, Santa Clara, CA,
AC C
US), followed the manufacturer s instructions. Slides were scanned by Agilent Microarray Scanner (Cat#G2565CA, Agilent technologies, Santa Clara, CA, US) with default settings, Dye channel: Green, Scan resolution=3µm, 20bit. Data were extracted with Feature Extraction software 10.7 (Agilent technologies, Santa Clara, CA, US). Raw data were normalized by Quantile algorithm, Gene Spring Software 12.6(Agilent technologies, Santa Clara, CA, US) Bioinformatics analysis
ACCEPTED MANUSCRIPT We used two independent algorithms to identify the targets of differentially expressed lncRNAs via cis- or trans-regulatory effects. The first algorithm searches for target genes acting in cis-lncRNAs and potential target genes were paired and visualized
RI PT
using UCSC (http://genome.ucsc.edu/) genome browser. The second algorithm is based on mRNA sequence complementarity and RNA duplex energy prediction. Then
predicted lncRNA target genes were input into the Database for Annotation,
SC
Visualization and Integrated Discovery (DAVID; http://david.abcc.ncifcrf.gov/),
the
KEGG
(Kyoto
M AN U
which utilized Gene Ontology (GO) to identify the molecular function. Next we used Encyclopedia
of
Genes
and
Genomes)
database
(http:www.genome.ad.jp/kegg/) to analyze the potential functions in the pathways. Pathway-network modeling was then performed to explore the interactions among
TE D
pathways. The pathway network was the interaction net of the significant pathways(p<0.05) of the differential expression genes using Fisher’s exact test , and was built according to the interaction among pathways of the KEGG database to find
AC C
Real time PCR
EP
the interaction among the significant pathways directly and systemically.
To verify the microarray results, we randomly choose 12 differentially expressed lncRNAs and detect their expression change between two groups. Total cellular RNA was isolated from all samples using PAXgene Blood RNA Kit Qiagen#762174and then reversely transcribed using iScript cDNA synthesis kit (BIO-RAD, USA) in accordance with the manufacturer’s instructions. Glyceralde-hyde 3-phosphate dehydrogenase (GAPDH) mRNA was used as an internal control. The 12 lncRNAs
ACCEPTED MANUSCRIPT and the primers (5' to 3: CATCCATCACTGTCCTTGTCA
ENST00000416385-R
GGGGATAACTGTGGCACCTA
ENST00000454968-F
GCAATGTTGTGCTCCTGTTAGTCT
ENST00000454968-R
TATCATGTGCCTCAGCCTGTTT
ENST00000409898-F
CCACCAGCCTCTCCTTGACA
ENST00000409898-R
GGATACAGTTCATTAAGGTTGGAAAAG
ENST00000409054-F
GATGAGAGCTGGACTTCTGTGTGT
ENST00000409054-R
GCGTGGGCTTAACTGATGCT
ENST00000453213-F
TGGAGATGGAAGGAGGATGCT
ENST00000453213-R
GTGGACTCGCACTGTCTGACA
SC
M AN U
CGAGCTGTAAAAGCCAAAGG
NR_033908-R
CCTGGGCGATAAGAGTGAAA
TE D
NR_033908-F
N344580-F TACCACGTCACCACCAACAC
EP
N344580-R CCTGAGCAGTTTGCATGGTA GGTGGCCTGACTTTTAGCTG
AC C
ENST00000565401-F
RI PT
ENST00000416385-F:
ENST00000565401-R
GCGGTTTCTTCCAAAATCAA
ENST00000408564-F
TCAAGCAGAGGCCTAAAGGA
ENST00000408564-R
TTCCCTACTGAGGTCCCAGA
NR_027252-F
ATCCCAGGGTAAGGAATGAAAGTA
NR_027252-R
GAGTTATGGTCACCTTGGAATGG
NR_015395-F
CTAATTTGCCACCACCCTGT
ACCEPTED MANUSCRIPT NR_015395-R
AAGACCCAGATGCCGTTTTA
NR_024420-F
CTGCAACGAATCCCAAAAGT
NR_024420-R
ACCACTTTCCAGAGGCTGAA
RI PT
Statistical analysis Statistical analysis was performed using Student’s t-test to compare two variables of microarray data. Statistical significance was determined at P<0.05, the raw data were
SC
normalized using the quantile algorithm in GeneSpring software 12.6 (Agilent
M AN U
technologies, CA, US). The genes with significant differential expression (a fold change >2) between the two groups were filtered using Student’s t-test (p < 0.05). Results
Differentially expressed long non-coding RNAs in moyamoya disease
TE D
Using authoritative data sources, we assessed lncRNA expression patterns in the MMD and control groups. Differentially expressed lncRNAs were identified using fold-change filtering (fold change > 2, P < 0.05). We found that 3,649 lncRNAs were
EP
differentially expressed (fold change > 2, P < 0.05) between the MMD and control
AC C
groups. Among them, 1,494 lncRNAs were up-regulated, and 2,155 lncRNAs were down-regulated in the MMD group (fold change > 2) (Figure 1 A and D). We next performed principal component analysis (PCA) and Hclust analysis to better understand the population separations in all 25 samples. PCA scatterplot showed that two distinct clusters formed, and the MMD and control groups showed excellent separation. However, 2 MMD subgroups did not show clear separation (Figure 1 B). The Hclust results showed that all 25 samples were separated into two populations
ACCEPTED MANUSCRIPT (MMD and control groups), and no obvious separation was noticed between the 2 subgroups (Figure 1 C). Real-time PCR validation
RI PT
Validation of the microarray results was conducted using RT-PCR.12 lncRNA probes were randomly selected for PCR amplification by applying specific primers. The PCR
results for all of the evaluated lncRNA probes were concordant with the microarray
SC
results. This agreement confirmed the accuracy of the microarray results (Figure 2).
M AN U
Potential functional targets
Although lncRNAs can theoretically exert both cis- and trans-regulation of target genes, current studies suggested the superiority of cis-regulation. In our results of GO and KEGG enrichment analysis revealed no more than 10 trans-regulated target genes
TE D
in each enriched GO and KEGG term (data not shown). Therefore, only the enrichment analysis results of the cis-regulated target genes were presented in the following GO and KEEG enrichment analysis.
EP
Gene ontology analysis
AC C
To identify the potential functions of the differentially expressed lncRNAs, we next performed GO analysis. This analysis revealed that the cis-regulated lncRNA target genes were enriched in the terms Molecular function, Biological process and Cellular component. The top 10 GO terms are listed (Table 2). Pathway analysis and pathway network The signaling pathways potentially regulated by the distinctly expressed lncRNAs were further investigated by performing enrichment analysis of lncRNA target genes
ACCEPTED MANUSCRIPT using the KEGG database. Sixteen most significantly enriched signaling pathways were identified (Table 3, p≤0.01). Surprisingly, many of the identified signaling pathways have close associations with immune response, vasculogenesis and smooth
RI PT
muscle contraction and these processes are the basis of the onset of MMD. Subsequently, a pathway network was constructed using the significantly enriched
signaling pathways (P<0.05). During pathway network construction, we attempted to
SC
reveal the interactions between each signaling pathway in the pathway network and to
M AN U
identify the network’s core signaling pathway. As shown in Figure 3, several signaling pathways can interact in this network, and the MAPK signaling pathway is the core pathway of the network, indicating that the MAPK pathway plays a pivotal role in the identified regulatory network.
TE D
Analyses of lncRNA target genes and mRNA intersections
We at last evaluated the intersections of the lncRNA target genes and the differentially expressed mRNAs. The genes in this intersection were identified based on two
EP
factors: (1) their potential regulation by the identified differentially expressed
AC C
lncRNAs in the bioinformatics analysis, and (2) their differential mRNA expression patterns in the microarray analysis. The microarray results contained several lncRNAs with multiple potential target genes that are difficult to verify individually; therefore, the identification of intersections helped narrow the exploration range to some degree. Interestingly, the effects of the MAPK signaling pathway were also detected (Gene Ontology analysis showed that 7 target gene were predicted in negative regulation of MAP kinase activity, P=9.52E-05).
ACCEPTED MANUSCRIPT Discussion In recent years, an increasing number of studies have been conducted to evaluate associations between non-coding RNAs and various diseases. From these studies,
RI PT
special databases, such as “The LncRNA and Disease Database” has been established to identify non-coding RNAs that are associated with specific diseases. In this
database, it contains more than 1000 lncRNA-disease entries. In 2014, Dai et al firstly
SC
reported miRNA expression profile changes in the sera of patients with MMD.15
M AN U
lncRNAs have longer base chains than miRNAs and have many regulatory roles outside of post transcriptional regulation. Therefore, lncRNAs are generally considered to possess more extensive and complex regulatory roles than miRNAs.4-6 Unfortunately, the association between lncRNAs and MMD has not been previously
TE D
reported. Based on the substantial regulatory roles of lncRNAs in many vascular diseases,7 the present study aimed to identify alterations in lncRNA expression in
pathophysiology.
EP
patients with MMD and to explore the significance of these changes in MMD
AC C
For ethical issues, we could not obtain moyamoya vessels for analysis from either direct or indirect revascularization. Therefore, peripheral blood (primarily white blood cells) was chosen as our research material, consistent with most recent studies on MMD. Fifteen cases of MMD, including 7 hemorrhagic type and 8 ischemic type were included. The current study primarily focused on differences in the hemodynamics of these types. For example, anterior choroid artery dilation is an important risk factor for hemorrhagic MMD.16 No current study has suggested that
ACCEPTED MANUSCRIPT different molecular mechanisms underlie the two subtypes; thus, 15 cases were included to uncover possible clues regarding their molecular mechanisms. More importantly, a common question of many studies evaluating disease-related lncRNAs
RI PT
is whether differential lncRNA expression is the "cause" of a disease or the "result" of internal environmental changes due to the disease. Our PCA and Hclust results showed that the lncRNA profiles of the 2 subtypes did not form distinct clusters. It is
SC
well known that these 2 subtypes produced substantially different internal blood
M AN U
environments; Given the information that the 2 subgroups of MMD were of similar age and sex ratio and DSA Suzuki stage, as shown in table 1, the results indicated that lncRNA expression may not be influenced by the internal environment created by different MMD subgroups. The changes in lncRNA expression patterns identified in
TE D
the present study are more likely to be the cause rather than the result of MMD. As mentioned above, we used whole blood samples for this analysis; however, after the experimental procedures, the lncRNAs were mostly isolated from white blood
EP
cells. We know that lumen stenosis caused by hyperplasia of intimal smooth muscle
AC C
cells is an important cause of MMD, in which the leukocyte-mediated inflammatory response plays an important role.17,18 In fact, congenital factors and acquired vascular inflammatory response are currently believed to be the two most important aspects of MMD pathophysiology.19 Pathway analysis showed that MAPK, VEGF, and smooth muscle signaling pathways were identified as important signaling pathways which have close associations with immune response, vasculogenesis and smooth muscle contraction and these processes are the basis of the onset of MMD. The results
ACCEPTED MANUSCRIPT demonstrated that several signaling pathways might participate in regulating the onset and progression of MMD; however, too many signaling pathways were identified to provide a clear causal relationship. It is difficult to verify this number of signaling
RI PT
pathways individually or to perform intervention measures to inhibit the onset of MMD.
Therefore, we constructed a pathway network with the significantly enriched
SC
signaling pathways (P < 0.05) to identify the core signaling pathway within the
M AN U
network. The results demonstrated that the MAPK signaling pathway is the core of this network. Many studies have demonstrated the important roles of the MAPK signaling pathway in vascular inflammation and other vascular physiological and pathophysiological processes, such as renal,20-22 pulmony23-26 and cardiovascular
TE D
stenosis and remodeling,27 and many specific MAPK inhibitors have been studied to prevent the MAPK-induced biological process.22,28,29 As written in the Guidelines for Diagnosis and Treatment of Moyamoya Disease: “Degeneration of the smooth muscle
EP
cells in the media and the resultant death of the vascular smooth muscle cells cause
AC C
thinning of the media” and “These changes noted in the terminal portion of the internal carotid arteries suggest the possibility of similar occurrence in the systemic arteries”,14 it indicates a similar pathology in MMD and systemic stenosis. Moreover, it was also noted in the guidelines that “some transcription and growth factors, such as bFGF are implicated in the migration and proliferation of vascular smooth muscle cells”,14 where studies have demonstrated that in these process activation of the MAPK signaling pathway is of vital importance. In fact, the MAPK signaling
ACCEPTED MANUSCRIPT pathway is the major mediator for these cytokines to induce the proliferation of vascular smooth muscle cells.30-32 To date, a definite correlation between a specific signaling pathway and MMD
RI PT
pathophysiology has not been reported, and the potential therapeutic targets remain unknown. lncRNAs have multiple action modes and a quite complicated regulatory network, the identification of multiple distinctly expressed lncRNAs is challenging, as
SC
is verifying the numerous associated signaling pathways. Due to the important role of
M AN U
the MAPK signaling pathway in the process of systemic artery stenosis22,26,27 and atherosclerosis,33-37 focusing on this signaling pathway narrowed the scope of our study. Given the fact that MAPK signaling pathway inhibitors had been successfully used to treat atherosclerosis in animal experiments,35,37 our results provide leads to
therapeutic targets.
TE D
further understand the pathophysiology of MMD and may help identify new available
In our study, a single-channel, dual-labeled DNA microarray that contains probes for
EP
all currently known lncRNAs and mRNAs was employed. Thus, the expression
AC C
profiles of all known lncRNAs and mRNAs were simultaneously obtained. Differentially expressed lncRNAs were screened, and their target mRNAs were predicted. These mRNAs were then compared with the identified differentially expressed mRNAs to reveal intersections between the two. Following this, we performed GO of this intersection. Interestingly, we also noted the involvement of the MAPK signaling pathway. Because lncRNAs function at multiple levels in addition to regulating mRNA, the lncRNAs involved in the identified intersections would
ACCEPTED MANUSCRIPT constitute only a small portion of those of significance. Nevertheless, analysis of such intersections may enable further narrowing of the scope of future explorations. Cerebral Revascularization (direct or indirect) has always been the primary method to
RI PT
treat Moyamoya disease (MMD). Even with the improvement of surgical techniques and plans, and more scientific standards for the case control, the failure rate of this
operation is still 20-30%, combined with 10-20% severe operative complications.38
SC
Until now, there is no specific treatment to prevent MMD progression. With a better
M AN U
understanding of MMD pathophysiology, non-surgical approaches targeting MMD pathophysiology may be available to stop or slow the progression of this disease39. The time of considering lncRNA just as "transcriptional noise" has passed. And in this entirely new field, the researches on lncRNA indicated the strong biological
TE D
regulatory effects in almost all the biological processes. Therefore, the lncRNA research in disease researches is of great importance with lncRNA itself and the targeted regulatory network as the new hope for the cure of relevant diseases.
EP
Our study still has limitations, and there remains future work to be done. First,
AC C
although this is the first lncRNA study of moyamoya disease (MMD), because of the limited sample size, specific lncRNAs and core members of the MAPK signalling pathway could not be verified. The number of genes involved is large, which increases the difficulty for the next research. The identification of certain specific lncRNAs or a smaller panel of lncRNAs should be the focus of future large-sample studies. Second, our results were based on bioinformatics analyses and require further confirmation by in vitro and in vivo experiments. We note that aberrant angiogenesis
ACCEPTED MANUSCRIPT is an active angiogenetic process that causes stenosis through the proliferation of vascular progenitor cells (including both endothelial and smooth muscle progenitor cells) 39. By targeting specific lncRNAs and MAPK signalling molecules in these
RI PT
cells, future experiments could determine whether the abnormal proliferation process could be blocked. Finally, it would also be helpful to clarify the relationship between
lncRNAs and other key molecules (for example, miRNAs and RNF213) and their
SC
participation in MMD pathophysiology. In this study, we have taken the first step
M AN U
towards investigating the role of lncRNAs in MMD, but future work is needed to investigate its practical application in treating this disease. Conclusion
Long noncoding RNA expression profiles are quite different in MMD group and
TE D
produced completely different clusters. Multiple signaling pathways that are closely associated with immune response, vasculogenesis and smooth muscle contraction were indicated to participate in lncRNAs regulatory mechanism, of these,
EP
mitogen-activated protein kinase (MAPK) signaling pathway was found to have a
AC C
core role in this pathway-network. lncRNAs and MAPK signaling pathway should be further studied as new therapeutic targets in future. Conflict of interest
The authors report no conflict of interest concerning the materials or methods used in this study or the findings specified in this paper. Study Funding Supported by the National Natural Science Foundation of China (81371292) and the
ACCEPTED MANUSCRIPT “13th Five-Year Plan” National Science and Technology supporting plan (2015BAI09B04), Acknowledge
RI PT
The authors thank Xiaona Zhang and Weiping Yu for the data collection and analysis. The authors also thank the research nurses, Donghong Zhao, Xi Ren, Na Shi and
References 1.
Suzuki J, Takaku A. Cerebrovascular moyamoya disease: disease showing abnormal net-like
vessels in base of brain. Arch Neurol 1969;20:288-299.
Cecchi AC, Guo D, Ren Z, et al. RNF213 rare variants in an ethnically diverse population with
M AN U
2.
SC
Xiaorong Sun for their assistance in the clinical collection of blood samples
Moyamoya disease. Stroke 2014;45:3200-3207. 3.
Mineharu Y, Takagi Y, Takahashi JC, et al. Rapid progression of unilateral moyamoya disease in a
patient with a family history and an RNF213 risk variant. Cerebrovasc Dis 2013;36:155-157. 4.
Mathieu E-L, Belhocine M, Dao L, Puthier D, Spicuglia S. [Functions of lncRNA in development and
diseases]. Med Sci M/S 2013;30:790-796.
Esteller M. Non-coding RNAs in human disease. Nat Rev Genet 2011;12:861-874.
6.
Taft RJ, Pang KC, Mercer TR, Dinger M, Mattick JS. Non-coding RNAs: regulators of disease. J
Pathol 2010;220:126-139. 7.
TE D
5.
Schonrock N, Harvey RP, Mattick JS. Long noncoding RNAs in cardiac development and
pathophysiology. Circ Res 2012;111:1349-1362. 8.
Gordon FE, Nutt CL, Cheunsuchon P, et al. Increased expression of angiogenic genes in the brains
of mouse meg3-null embryos. Endocrinology 2010;151:2443-2452. Michalik KM, You X, Manavski Y, et al. Long noncoding RNA MALAT1 regulates endothelial cell
EP
9.
function and vessel growth. Circ Res 2014;114:1389-1397. 10. Holdt LM, Beutner F, Scholz M, et al. ANRIL expression is associated with atherosclerosis risk at
AC C
chromosome 9p21. Arterioscler Thromb Vasc Biol 2010;30:620-627. 11. Congrains A, Kamide K, Oguro R, et al. Genetic variants at the 9p21 locus contribute to atherosclerosis through modulation of ANRIL and CDKN2A/B. Atherosclerosis 2012;220:449-455. 12. Congrains A, Kamide K, Katsuya T, et al. CVD-associated non-coding RNA, ANRIL, modulates expression of atherogenic pathways in VSMC. Biochem Biophys Res Commun 2012;419:612-616. 13. Holdt LM, Hoffmann S, Sass K, et al. Alu elements in ANRIL non-coding RNA at chromosome 9p21 modulate atherogenic cell functions through trans-regulation of gene networks. PLOS Genet 2013;9:e1003588. 14. Hashimoto N, Tominaga T, Miyamoto S, et al. Guidelines for diagnosis and treatment of moyamoya disease (spontaneous occlusion of the circle of Willis). Neurol Med Chir 2012;52:245-266. 15. Dai D, Lu Q, Huang Q, et al. Serum miRNA signature in Moyamoya disease. PLoS One 2014;9:e102382.
ACCEPTED MANUSCRIPT 16. Ryan RW, Chowdhary A, Britz GW. Hemorrhage and risk of further hemorrhagic strokes following cerebral revascularization in Moyamoya disease: a review of the literature. Surg Neurol Int 2012;3:72. 17. Takekawa
Y,
Umezawa
T,
Ueno
Y,
Sawada
T,
Kobayashi
M.
Pathological
and
immunohistochemical findings of an autopsy case of adult moyamoya disease. Neuropathology 2004;24:236-242. 18. Gosalakkal J. Moyamoya disease: a review. Neurol India 2002;50:6. 19. Weinberg DG, Arnaout OM, Rahme RJ, Aoun SG, Batjer HH, Bendok BR. Moyamoya disease: a
RI PT
review of histopathology, biochemistry, and genetics. Neurosurg Focus 2011;30:E20.
20. Park KM, Chen A, Bonventre JV. Prevention of kidney ischemia/reperfusion-induced functional injury and JNK, p38, and MAPK kinase activation by remote ischemic pretreatment. J Biol Chem 2001;276:11870-11876.
21. Chen H, Brahmbhatt S, Gupta A, Sharma AC. Duration of streptozotocin-induced diabetes
differentially affects p38-mitogen-activated protein kinase (MAPK) phosphorylation in renal and
SC
vascular dysfunction. Cardiovasc Diabetol 2005;4:3-3.
22. Wang D, Warner GM, Yin P, et al. Inhibition of p38 MAPK attenuates renal atrophy and fibrosis in a murine renal artery stenosis model. Am J Physiol Renal Physiol 2013;304:F938-F947.
M AN U
23. Hashimoto S, Matsumoto K, Gon Y, et al. p38 mitogen-activated protein kinase regulates IL-8 expression in human pulmonary vascular endothelial cells. Eur Respir J 1999;13:1357-1364. 24. Morrell NW, Upton PD, Kotecha S, et al. Angiotensin II activates MAPK and stimulates growth of human pulmonary artery smooth muscle via AT1 receptors. Am J Physiol Lung Cell Mol Physiol 1999;277:L440-L448.
25. Nick JA, Young SK, Brown KK, et al. Role of p38 mitogen-activated protein kinase in a murine model of pulmonary inflammation. J Immunol 2000;164:2151-2159.
TE D
26. Urashima T, Zhao M, Wagner RA, et al. Molecular and physiologic characterization of RV remodeling in a murine model of pulmonary stenosis. Am J Physiol Heart Circ Physiol 2008;295:H1351-H1368.
27. Nordmeyer J, Eder S, Mahmoodzadeh S. Upregulation of myocardial estrogen receptors in human aortic stenosis. Circulation 2004;110:3270-3275.
EP
28. Öllinger R, Bilban M, Erat A. Bilirubin: A natural inhibitor of vascular smooth muscle cell proliferation. Circulation 2005;112:1030-1039. 29. Underwood DC, Osborn RR, Bochnowicz S, et al. SB 239063, a p38 MAPK inhibitor, reduces
AC C
neutrophilia, inflammatory cytokines, MMP-9, and fibrosis in lung. Am J Physiol Lung Cell Mol Physiol 2000;279:L895-L902.
30. Cui Y, Sun Y, Lin H, et al. Platelet-derived growth factor-BB induces matrix metalloproteinase-2 expression and rat vascular smooth muscle cell migration via ROCK and ERK/p38 MAPK pathways. Mol Cell Biochem 2014;393:255-263. 31. Suwanabol PA, Seedial SM, Shi X, et al. Transforming growth factor-β increases vascular smooth muscle
cell
proliferation
through
the
Smad3
and
extracellular
signal-regulated
kinase
mitogen-activated protein kinases pathways. J Vasc Surg 2012;56:446-454. 32. Martin-Garrido A, Williams HC, Lee M, et al. Transforming growth factor β inhibits platelet derived growth factor-induced vascular smooth muscle cell proliferation via Akt-independent, Smad-mediated cyclin D1 downregulation. PLoS One 2013;8:e79657.
ACCEPTED MANUSCRIPT 33. Hu Y, Sun B, Liu K, et al. Icariin attenuates high-cholesterol diet induced atherosclerosis in rats by inhibition of inflammatory response and p38 MAPK signaling pathway. Inflammation Epub 2015 Aug 26. 34. Tiwari RL, Singh V, Barthwal MK. Macrophages: an elusive yet emerging therapeutic target of atherosclerosis. Med Res Rev 2008;28:483-544. 35. Rahaman SO, Lennon DJ, Febbraio M, Podrez EA, Hazen SL, Silverstein R. A CD36-dependent signaling cascade is necessary for macrophage foam cell formation. Cell Metab 2006;4:211-221.
RI PT
36. Ricci R, Sumara G, Sumara I. Requirement of JNK2 for scavenger receptor A-mediated foam cell formation in atherogenesis. Science 2004;306:1558-1561.
37. Zhao M, Liu Y, Wang X, New L, Han J, Brunk UT. Activation of the p38 MAP kinase pathway is required for foam cell formation from macrophages exposed to oxidized LDL. APMIS 2002;110:458-468.
38. Kim D, Huh P, Kim H, et al. Surgical treatment of moyamoya disease in adults: combined direct
SC
and indirect vs. indirect bypass surgery. Neurol Med Chir 2012;52:333-338.
39. Bang OY, Fujimura M and Kim S-K. The pathophysiology of moyamoya disease: an update. J stroke.
M AN U
2016; 18(1):12-20.
Figure legends
TE D
Figure 1.lncRNA expression profile in MMD and control group.A.Scatter-plot comparison of lncRNAs expression levels between MMD and control group. The red dots of the scatter plot in the upper left corner represented up-regulated lncRNAs (fold
EP
change >2,P<0.05), and the green dots in low right corner represented down-regulated
AC C
lncRNAs (fold change <0.5, P<0.05) B. Principal Component Analysis (PCA) of lncRNAs in MMD and control group. The PCA scatter plot showed that the two groups (MMD and control) is the best separation and formed two distinct clusters. However, the 2 subgroups of MMD were not able to separated.Black :control group.Red:Hemorrhagic
group.Green
and.Blue:Ischemic
group.C.
HClust
of
lncRNAs among 25 samples. The results showed 25 samples were separated into two populations (MMD and control group), no obvious separation was noticed in MMD
ACCEPTED MANUSCRIPT group between MMD subgroups. D. Heat map showed the differentially expressed genes between two groups. Red color indicates over expressed genes, whereas green color indicates the opposite. The color scale bar is shown.
RI PT
Figure 2. Real time PCR was performed for validation of microarray results. Using specific primers, all the randomly selected 12 lncRNAs PCR results were consistent with microarray results.
SC
Figure. 3 Pathway network of differentially-expressed lncRNAs target genes.
M AN U
Cycle nodes represent pathways, the arrow between two nodes represents an interaction target between pathways. The more edges of a pathway, the more
AC C
EP
TE D
pathways connecting to it, and the more central role it has within the network.
ACCEPTED MANUSCRIPT Table 1. Clinical characteristics of 15 MMD patients No.
age
sex
initial clinical findings
Suzuki stage
mRS score
2
hemorrhagic group 19
female
IVH
rt2 lt2
2
32
female
IVH
rt4 lt4
1
3
33
male
IVH
rt2 lt2
2
4
47
male
IVH
rt5 lt5
1
5
44
male
IVH
6
28
male
IVH
7
41
female
female
2
21
male
3
43
male
4
20
male
5
21
female
6
42
7
SC rt3 lt4
2
rt3 lt3
TIA
2
2
rt3 lt3
2
infarction
rt5 lt5
2
infarction
rt4 lt4
1
TIA
rt3 lt3
3
female
infarction
rt2 lt4
2
20
male
infarction
rt4 lt4
1
46
female
infarction
rt3 lt2
1
TE D
TIA
AC C
8
3
EP
29
rt3 lt3
M AN U SAH
ischemic group 1
RI PT
1
†:IVH= intraventricular hemorrhage; TIA = transient ischemic attack; lt = left; rt = right;mRs= modified Rakkin score
ACCEPTED MANUSCRIPT Table 2 Enriched Go terms for potential functional targets (cis-effect, top 10,) GO
term
type
Target Gene
P value
in GO Molecular function
G-protein coupled receptor signaling pathway
Biological process
69
7.20 E-05
olfactory receptor activity
Molecular function
27
8.51 E-05
detection of chemical stimulus involved in sensory
Biological process
27
1.01 E-04
Cellular component
40
9.58 E-04
Biological process
23
9.99 E-04
Molecular function
925
1.14 E-03
Biological process
8
1.87 E-03
Molecular function
36
1.88 E-03
Biological process
22
2.09 E-03
cell projection
proteasome-mediated ubiquitin-dependent protein catabolic process
TE D
protein binding sterol metabolic process
EP
RNA polymerase II core promoter proximal region sequence
SC
M AN U
perception of smell
AC C
cellular response to lipopolysaccharide
47
3.40 E-05
RI PT
G-protein coupled receptor activity
Table 3 Enriched significant pathways for potential functional targets (cis-effect, P≤0.01) ACCEPTED MANUSCRIPT KEGG name
Target gene in this pathway
All gene
P value
in this pathway
50
207
9.46 E-05
Insulin signaling pathway - Homo sapiens (human)
34
141
1.07 E-03
Neurotrophin signaling pathway - Homo sapiens
30
31
Non-small cell lung cancer - Homo sapiens (human)
16
M AN U
Dopaminergic synapse - Homo sapiens (human)
120
1.13 E-03
131
2.28 E-03
SC
(human)
RI PT
Focal adhesion - Homo sapiens (human)
56
3.63 E-03
Olfactory transduction - Homo sapiens (human)
33
405
4.39 E-03
MAPK signaling pathway - Homo sapiens (human)
51
257
5.84 E-03
Glioma - Homo sapiens (human)
17
65
7.08 E-03
16
60
7.19 E-03
Oxytocin signaling pathway - Homo sapiens (human)
34
159
7.30 E-03
Platelet activation - Homo sapiens (human)
29
131
7.78 E-03
ErbB signaling pathway - Homo sapiens (human)
21
87
8.08 E-03
Vascular smooth muscle contraction - Homo sapiens
27
121
8.93 E-03
43
215
9.30 E-03
FoxO signaling pathway - Homo sapiens (human)
29
133
9.54 E-03
NOD-like receptor signaling pathway - Homo
15
57
1.01 E-02
AC C
EP
TE D
mTOR signaling pathway - Homo sapiens (human)
(human)
Regulation of actin cytoskeleton - Homo sapiens (human)
sapiens (human)
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 Highlights
1. lncRNAs expression were quite different in adult moyamoya patients. 2. Multiple signaling pathways that were closely associated with immune response,
lncRNAs regulatory mechanism.
RI PT
vasculogenesis and smooth muscle contraction were predicted involved in
3. Pathway network analysis showed that the MAPK signaling pathway, which has
SC
been well studied for the treatment of many other cardiovascular diseases, was the core of lncRNAs regulatory-network. Our findings could help further understand
AC C
EP
TE D
M AN U
the pathogenesis of MMD and provide new potential therapeutic targets.
ACCEPTED MANUSCRIPT MMD: moyamoya disease; lncRNA: long noncoding RNA MAPK signaling pathway: mitogen-activated protein kinase (MAPK) signaling
RI PT
pathway IVH: intraventricular hemorrhage TIA : transient ischemic attack
AC C
EP
TE D
M AN U
SC
mRs: modified Rankin scale