- Email: [email protected]
Multiple sclerosis risk pathways differ in Caucasian and Chinese populations
Accepted Manuscript Multiple sclerosis risk pathways differ in Caucasian and Chinese populations
Guiyou Liu, Fang Zhang, Yang Hu, Yongshuai Jiang, Zhongying Gong, Shoufeng Liu, Xiuju Chen, Qinghua Jiang, Junwei Hao PII: DOI: Reference:
S0165-5728(16)30289-2 doi: 10.1016/j.jneuroim.2017.03.012 JNI 476548
To appear in:
Journal of Neuroimmunology
Received date: Revised date: Accepted date:
22 September 2016 5 March 2017 15 March 2017
Please cite this article as: Guiyou Liu, Fang Zhang, Yang Hu, Yongshuai Jiang, Zhongying Gong, Shoufeng Liu, Xiuju Chen, Qinghua Jiang, Junwei Hao , Multiple sclerosis risk pathways differ in Caucasian and Chinese populations. The address for the corresponding author was captured as affiliation for all authors. Please check if appropriate. Jni(2017), doi: 10.1016/j.jneuroim.2017.03.012
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 Multiple sclerosis risk pathways differ in Caucasian and Chinese populations
Guiyou Liu1, #, Fang Zhang2, #, Yang Hu1, Yongshuai Jiang3, Zhongying Gong4, Shoufeng Liu5, Xiuju
PT
Chen6, Qinghua Jiang1, *, Junwei Hao2, *
1
RI
School of Life Science and Technology, Harbin Institute of Technology, Harbin, China
2
SC
Department of Neurology and Tianjin Neurological Institute, Tianjin Medical University General
Hospital, Tianjin, China 3
NU
College of Bioinformatics Science and Technology, Harbin Medical University, Harbin, China
4
MA
Department of Neurology, Tianjin First Central Hospital, Tianjin, China
5
Department of Neurology, Tianjin HuanHu Hospital, Tianjin, China
6
These authors contributed equally to this work.
CE
#
PT E
D
Department of Neurology, Tianjin NanKai Hospital, Tianjin, China
AC
*Corresponding author: Qinghua Jiang School of Life Science and Technology, Harbin Institute of Technology, Harbin, Heilongjiang 150001, China
E-mail: [email protected] *Corresponding author: Junwei Hao Department of Neurology and Tianjin Neurological Institute, Tianjin Medical University General Hospital, Tianjin 300052, China, Tel/Fax: +86-22-6081-7429, Email: [email protected] 1
ACCEPTED MANUSCRIPT Abstract Large-scale genome-wide association study (GWAS) datasets provide strong support for investigations of the mechanisms underlying multiple sclerosis (MS) by using pathway analysis methods. In our recent study, we conducted a three-stage pathway analysis of GWAS and
PT
expression datasets. After identifying 15 shared MS pathways in separate MS GWAS datasets, we
RI
found that dysregulated MS genes were significantly enriched in 10 of the 15 MS risk pathways.
SC
Evidence showed that 17%-30% of genes are differentially expressed among individual ethnic populations. We then verified the potential disruption of genes in the 10 MS risk pathways cited
NU
above in Chinese MS patients. Here, we investigated potential up- and down-regulation of 42 MS
MA
genes in these 10 MS risk pathways using 132 Chinese MS patients and 76 healthy control subjects. We then identified 31 differentially expressed genes in Chinese MS patients compared to healthy
D
control subjects. Moreover, the expression patterns of 28 of these genes were consistent with those
PT E
obtained from Caucasian (European and American) MS patients, although 14 genes differed from the latter group’s. Our results provide clinically useful clues about the link between these risk genes
AC
CE
and MS susceptibility in the Chinese population.
Keywords: Multiple sclerosis; pathway analysis; genome-wide association study; Chinese population 2
ACCEPTED MANUSCRIPT 1 Introduction Multiple sclerosis (MS) is considered to be an inflammatory disease and is grouped amongst the most common diseases of the central nervous system (CNS) 1. As a means of detecting genetic risk variants, large-scale genome-wide association studies (GWAS) have been performed and
PT
provide strong support for identifying disease risk pathways 2-8. Baranzini et al. conducted a
RI
pathway analysis of two MS GWAS 3. Their identification of several significantly enriched pathways
SC
included such immune system pathways as cell adhesion, communication, and signaling as well as the neural pathways axon-guidance and synaptic potentiation 3. The International Multiple Sclerosis
NU
Genetics Consortium (IMSGC) similarly performed a pathway analysis of two large-scale MS
MA
GWAS datasets 9. Their report covered three significant gene ontology (GO) biological processes, including leukocyte activation, apoptosis, and positive regulation of macromolecule metabolic
D
process as well as three KEGG pathways consisting of the JAK-STAT signaling pathway, acute
PT E
myeloid leukemia, and T cell receptor signaling 9. We propose that the various MS GWAS datasets may share genetic pathways, although this
CE
possibility was not reported in previous studies 3, 9. Therefore, to test this hypothesis, we recently
AC
conducted a three-stage pathway analysis of GWAS and expression datasets 10. In stage 1, we analyzed pathways of two MS GWAS datasets. There, 15 shared pathways were found, most of which were related to the immune system and various diseases, environmental information processing, genetic information processing, and infectious diseases. In stage 2, we performed a candidate pathway analysis of the large-scale MS GWAS dataset from IMSGC to further verify the 15 shared pathways. We successfully replicated 14 of these 15 MS risk pathways. In stage 3, our pathway analysis was directed to the dysregulated MS gene list from seven human MS case-control 3
ACCEPTED MANUSCRIPT expression datasets derived from Caucasian (European and American) populations. In this stage, we found that dysregulated MS genes were significantly enriched in 10 of the 15 MS risk pathways identified in stages 1 and 2. Prior evidence showed that 17%-30% of genes are differentially expressed among individual
PT
ethnic populations 11. Here, to further verify the potential disruption of MS genes in Chinese MS
RI
patients, we performed real-time PCR to determine the potential up- and down-regulation of these
SC
MS genes in the MS risk pathways noted above using gene expression datasets from Chinese MS
NU
patients and healthy controls. 2 Materials and methods
MA
2.1 MS risk pathways and dysregulated genes
The MS risk pathways and dysregulated genes were collected from our recent study 10. Overall, we
D
performed a pathways analysis of three MS GWAS datasets including 334,923 single nucleotide
PT E
polymorphisms (SNPs) from 931 trios 12, 551,642 SNPs from 978 MS patients and 883 control subjects 13, and 441,547 SNPs from 9,772 MS patients and 17,376 controls of European descent 14,
CE
as well as 229 differentially expressed MS genes 15. Ten MS risk pathways as well as the
AC
corresponding dysregulated genes are described in Table 1. More detailed information appears in our recent study 10.
2.2 Study population and trial design in the Chinese population During open enrollment for the study, a total of 132 patients with relapsing-remitting MS in the acute stage of the disease, who also satisfied the McDonald Criteria of MS as revised in 2010, were recruited at Tianjin Medical University General Hospital, Tianjin HuanHu Hospital, Tianjin First Central Hospital, and Tianjin NanKai Hospital, Tianjin, China (Table 2). All patients manifested 4
ACCEPTED MANUSCRIPT multiple disseminations of the disease in space (i.e., involvement of multiple areas of the CNS) and time (i.e., ongoing disease activity over time). MS was also verified by the presence of oligoclonal bands in the cerebrospinal fluid (CSF) of all these patients. Exclusion criteria were the following: (1) presence of other diseases of the CNS in addition to MS, (2) presence of tumor(s) and systemic
PT
hematological diseases, (3) presence of recent infection, and (4) concomitant use of antineoplastic
RI
or immune-modulating therapies prior to blood sampling. We also recruited 76 age- and gender-
SC
matched healthy volunteers (control subjects) for the comparative study. This project was designed as a multicenter study. The trial protocol and supporting documentation were approved by the
MA
from all patients or legally acceptable surrogates.
NU
institutional review boards of each participating center. Informed consent was obtained at enrollment
2.3 Quantitative real-time PCR validation in MS patients and control subjects
D
Total RNA was extracted from peripheral blood mononuclear cells (PBMCs) by using Trizol®
PT E
reagent (Invitrogen) according to the manufacturer’s instructions. RNA quantity and quality were assessed using a Nanodrop ND-100 Spectrophotometer (Nanodrop Technologies, USA) and a 2100
CE
Bioanalyzer (Agilent RNA 6000 Nano Kit, Germany). For the reverse transcriptase (RT) reaction,
AC
SYBR Green RT reagents (Bio-Rad, USA) were used. The mRNA PCR results were quantified using the 2ΔΔct method against β-actin and GAPDH for normalization. The resulting data represent the means (±SD) of three experiments. 2.4 Statistical Analysis Genes expressed differentially in MS patients compared to healthy controls were analyzed by using Student’s t tests. Statistical significance was considered as P<0.05. All statistical data were analyzed by using SPSS 17.0 software (SPSS Inc., Chicago, IL, USA). 5
ACCEPTED MANUSCRIPT 3 Results 3.1 MS risk pathways and dysregulated genes Previously, we analyzed the large-scale MS GWAS and expression datasets in Caucasian populations 10. To summarize briefly here, that report highlighted 10 significantly dysregulated MS
PT
risk pathways (Table 1). Comparison of the corresponding dysregulated genes in those 10
RI
pathways yielded 42 unique genes. Subsequent performance of real-time PCR revealed the primers
SC
for all the 42 genes, which are presented here in Supplementary Table 1. 3.2 Gene expression in Chinese MS patients and control subjects
NU
Results from our in-depth analysis of these 42 genes are provided in Table 3. Interestingly, 31 of
MA
these 42 genes showed significant differential expression in Chinese MS patients compared to the matched control subjects (Figure 1). We then compared in more detail the expression of these 31
D
significant genes in Chinese MS patients with those from European and American MS patients and
PT E
found that the expression trends of 28 of the 31 genes from Chinese participants were consistent with the results obtained from European and American MS patients (Table 3). COL11A2 was the
CE
most elevated (5.09-fold higher expression), followed by CCL3 (4.0-fold higher expression), and
AC
PNN (3.91-fold higher expression). PIM1 was the least expressed gene (3.31-fold lower expression), followed by JAK1 (2.81-fold lower expression), and TGFB1 (2.76-fold lower expression). The trends of these genes tested in Chinese MS patients were consistent with results obtained from the Caucasian populations, with the exception of CCL3, which had a lower expression in Caucasians. 3.3 Comparison of gene expression in Chinese and Caucasian MS patients The expression patterns of 28 (67%) (COL11A2, PNN, CLNS1A, ITGA4, TNFAIP3, EDEM1, CD40, EIF3D, IL7R, PIM1, EIF1, CXCR4, JAK1, TGFB1, EIF4A1, TNFRSF1A, GAG2, NFKBIA, MYLK,
6
ACCEPTED MANUSCRIPT IL2RG, HSPA1L, MAP3K7, IRF9, HLA-DQB1, JUN, BCL2, PAK2, CCR5) of these 42 genes were consistent with those obtained from Caucasian (European and American) MS patients. However, 14 (33%) of the 42 genes were expressed inconsistently with regard to those obtained from Caucasian MS patients. Eleven of the latter group’s genes (NFKB1, ITGA6, HRAS, CDC42, KRAS, TNF, IL1B,
PT
HSPA1A, ADCY9, IL2RB, CXCR6) had no different expression in Chinese MS patients, and 3
RI
genes (CCL3, IL8, CXCL1) expressed an opposite trend.
SC
3.4 Top three significant pathway analyses
In our recent study 10, we found that the toxoplasmosis pathway was the most significant pathway
NU
among the three MS GWAS datasets and seven human MS case-control expression datasets
MA
derived from Caucasian populations. Ten genes (TGFB1, HSPA1L, JAK1, HLA-DQB1, NFKBIA, TNFRSF1A, BCL2, MAP3K7, CD40, CCR5) were aberrantly expressed in the toxoplasmosis
D
pathway of Chinese MS patients compared to 14 genes in Caucasian populations. However, the
PT E
expression tendency in Chinese MS patients correlated strongly with that of the Caucasian populations.
CE
The chemokine signaling pathway was the second most significant pathway in the GWAS and
AC
expression analyses. Seven genes (GNG2, CXCR4, NFKBIA, IL8, CXCL1, CCL3, CCR5) had aberrant expression of the chemokine signaling pathway in Chinese MS patients compared to 13 in Caucasian populations.
In the NOD-like receptor signaling pathway, TNFAIP3 and NFKBIA were highly expressed in both Chinese MS patients and Caucasian populations. However, the expression of IL8, CXCL1 and MAP3K7 was low in Chinese MS patients. 4 Discussion 7
ACCEPTED MANUSCRIPT Two former studies that attempted pathway and network-based analysis methods for investigating multiple MS GWAS datasets failed to identify any shared genetic pathway 3, 9. To extend that search, we recently performed a three-stage analysis of GWAS and expression datasets. As a result, both the GWAS and expression datasets provided outcomes that highlight the involvement of 10 MS risk
PT
KEGG pathways in Caucasian (European and American) MS patients.
RI
With our interest in differential gene expression among diverse ethnic populations, we
SC
investigated the potential up- and down-regulation of 42 MS genes in the MS risk pathways. To that end, we examined 132 Chinese MS patients and 76 healthy control subjects and successfully
NU
identified 31 genes expressed in Chinese MS group but not in their healthy counterparts. The
MA
mRNA PCR results were quantified using the 2ΔΔct method against β-actin and GAPDH for normalization. Compared to healthy control subjects, in Chinese MS patients 31 differentially
D
expressed genes were identified. We further found that the expression trends of 28 of those 31
PT E
genes (90%) were consistent with the results obtained from European and American MS patients but another 5 genes (16%) that were inconsistent with those of the Caucasians. Our findings are in
CE
agreement with previous studies showing that 17%-30% of genes are differentially expressed in
AC
separate ethnic populations 11. Lu et al. analyzed about 15,000 transcripts expressed in lymphoblastoid samples and identified about 4,700 (31.3%) transcripts that were differentially expressed between the HapMap CEU and Asian (CHB + JPT) populations 11. In the 10 MS risk KEGG pathways cited above, the toxoplasmosis pathway is the most significant signal. In this pathway, 14 genes showed different expression in Caucasian populations. Here, we identified 10 of these 14 genes with clearly aberrant expression in Chinese MS patients. Extensive evidence indicates that MS is a chronic demyelinating disease of the central nervous 8
ACCEPTED MANUSCRIPT system and that its occurrence is more prevalent in economically advanced nations than in the developing countries 16-17. The lesser prevalence of parasitic infections (which can activate immune responses and prevent or modulate damage to host antigens) in the more modernized areas is among the possible responsible factors for such a difference 18-19. Toxoplasmosis is not highly
PT
prevalent in MS patients; that is, 36% and 49%, respectively, of MS patients and their family
RI
members were documented as having toxoplasmosis 20. Stascheit et al. investigated the prevalence
SC
of Toxoplasma gondii IgG and IgM antibodies in 163 MS patients and 178 control subjects 21. The negative association between parasitic Toxoplasma gondii infection and the presence of MS they
21
.
MA
compared with healthy controls
NU
reported reflected significantly lower counts of Toxoplasma gondii IgG antibodies in MS patients
We further compared the second significant genetic pathway, chemokine signaling, in
D
Caucasian and Chinese populations. In our previous study, 13 genes of Caucasian populations
PT E
differed in chemokine activity from that of the Chinese. Here, we successfully replicated 7 of these 13 genes showing aberrant expression in Chinese MS patients. Growing evidence suggests that
CE
chemokines and chemokine receptors participate in the recruitment of macrophages and T
AC
lymphocytes into the CNS, and it has been considered the most critical mechanism in the pathogenesis of MS 22-23. In these pathways, two genes, CXCR4 and NFKBIA, are highly expressed in both Chinese and Caucasian MS patients. CXCR4 is a receptor of CXCL12. In turn, CXCL12–CXCR4 interactions initiate multiple signaling pathways that augment T cell co-stimulation, proliferation, cytokine production, migration, and survival 24. NFKBIA, another important target of the chemokine signaling pathway, is also a highly expressed participant. NFKBIA normally retains NFkB within the cytoplasms of unstimulated cells and thereby inhibits the inflammatory response 9
25-26
.
ACCEPTED MANUSCRIPT Therefore, the overexpression of NFKBIA may be a protective response in MS patients. The NOD-like receptor signaling pathway is the third significant signal of interest in this survey. NOD-like receptors are intracellular receptors capable of regulating both innate and adaptive immune responses 27. Upon recognition of their respective ligands, NOD-like receptors become
PT
activated, and that transformation results in the subsequent triggering of multiple pro-inflammatory
RI
molecular pathways; one such example was nuclear factor-kappa B (NF-κB). In our study, we found
SC
that TNFAIP3 is highly expressed in both Chinese MS patients and Caucasian populations. Inactivation of the TNFAIP3 gene, encoding the A20 protein, is associated with critical inflammatory
NU
diseases including MS, rheumatoid arthritis and Crohn’s disease28. All the foregoing findings
MA
suggest that TNFAIP3 and NOD-like receptors pathway may be the important targets of MS pathogenesis.
D
Despite the clarity of this outcome, our use of PBMCs for the real time PCR validation research
PT E
represented a limitation of the study. PBMCs were the basis for our real-time PCR validation research. PBMCs consist of a variety of cells in various abundance that may lead to differences in
CE
observed gene expression for the whole PBMC group. These cells were selected because they
AC
reflect the change in cellular composition of MS patients as compared to controls but are somewhat nonspecific. Additionally, 11 MS patients had calcium tablet treatments and 6 received hemp seed soft capsules for purging before blood sampling. In summary, our multiple pathway analyses of GWAS and expression datasets highlight shared and dysregulated MS risk pathways in Caucasian (European and American) populations 10. In the Chinese population, we reported consistent but also inconsistent results compared to those in European and American Caucasians. The expression patterns of 28 (67%) genes were consistent 10
ACCEPTED MANUSCRIPT with those obtained from all Caucasian MS patients; however, 14 (33%) genes were inconsistent with those obtained from the Caucasians. Of those 14 genes, 11 were no different in Chinese MS patients, and 3 genes (CCL3, IL8, CXCL1) had the opposite expression trend. The importance of these results is new knowledge that may serve to determine treatment options according to patients’
PT
genetic heritage.
RI
Conflict of interest
SC
The authors declare no conflict of interests. Acknowledgments
NU
This work was financially supported by the National Natural Science Foundation of China
MA
(81300945 to G. Y. L., and 81571600, 81322018, 81273287 and 81100887 to J. W. H.); the Youth Top-notch Talent Support Program; and the National Key Clinical Specialty Construction Project of
D
China.
1.
PT E
References
Damotte V, Guillot-Noel L, Patsopoulos NA, et al. A gene pathway analysis highlights the
2.
AC
32.
CE
role of cellular adhesion molecules in multiple sclerosis susceptibility. Genes Immun. 2014; 15: 126-
Liu G, Jiang Y, Wang P, et al. Cell adhesion molecules contribute to Alzheimer's disease:
multiple pathway analyses of two genome-wide association studies. J Neurochem. 2012; 120: 190-8. 3.
Baranzini SE, Galwey NW, Wang J, et al. Pathway and network-based analysis of genome-
wide association studies in multiple sclerosis. Hum Mol Genet. 2009; 18: 2078-90. 4.
Jiang Q, Jin S, Jiang Y, et al. Alzheimer's Disease Variants with the Genome-Wide
Significance are Significantly Enriched in Immune Pathways and Active in Immune Cells. Mol 11
ACCEPTED MANUSCRIPT Neurobiol. 2016. 5.
Bao X, Liu G, Jiang Y, et al. Cell adhesion molecule pathway genes are regulated by cis-
regulatory SNPs and show significantly altered expression in Alzheimer's disease brains. Neurobiol Aging. 2015; 36: 2904 e1-7. Quan B, Qi X, Yu Z, et al. Pathway analysis of genome-wide association study and
PT
6.
RI
transcriptome data highlights new biological pathways in colorectal cancer. Mol Genet Genomics.
7.
SC
2015; 290: 603-10.
Liu G, Yao L, Liu J, et al. Cardiovascular disease contributes to Alzheimer's disease:
Liu G, Jiang Y, Chen X, et al. Measles contributes to rheumatoid arthritis: evidence from
MA
8.
NU
evidence from large-scale genome-wide association studies. Neurobiol Aging. 2014; 35: 786-92.
pathway and network analyses of genome-wide association studies. PLoS One. 2013; 8: e75951. Network-based multiple sclerosis pathway analysis with GWAS data from 15,000 cases and
D
9.
10.
PT E
30,000 controls. Am J Hum Genet. 2013; 92: 854-65. Liu G, Zhang F, Jiang Y, et al. Integrating genome-wide association studies and gene
Lu J and Clark AG. Impact of microRNA regulation on variation in human gene expression.
AC
11.
CE
expression data highlights dysregulated multiple sclerosis risk pathways. Mult Scler. 2016.
Genome Res. 2012; 22: 1243-54. 12.
Hafler DA, Compston A, Sawcer S, et al. Risk alleles for multiple sclerosis identified by a
genomewide study. N Engl J Med. 2007; 357: 851-62. 13.
Baranzini SE, Wang J, Gibson RA, et al. Genome-wide association analysis of susceptibility
and clinical phenotype in multiple sclerosis. Hum Mol Genet. 2009; 18: 767-78. 14.
Sawcer S, Hellenthal G, Pirinen M, et al. Genetic risk and a primary role for cell-mediated 12
ACCEPTED MANUSCRIPT immune mechanisms in multiple sclerosis. Nature. 2011; 476: 214-9. 15.
Kemppinen AK, Kaprio J, Palotie A and Saarela J. Systematic review of genome-wide
expression studies in multiple sclerosis. BMJ open. 2011; 1: e000053. 16.
Ahlgren C, Lycke J, Oden A and Andersen O. High risk of MS in Iranian immigrants in
La Flamme AC, Ruddenklau K and Backstrom BT. Schistosomiasis decreases central system
inflammation
and
alters the
progression
of
experimental
autoimmune
SC
nervous
RI
17.
PT
Gothenburg, Sweden. Multiple sclerosis (Houndmills, Basingstoke, England). 2010; 16: 1079-82.
encephalomyelitis. Infection and immunity. 2003; 71: 4996-5004.
Ascherio A and Munger KL. Epstein-barr virus infection and multiple sclerosis: a review.
NU
18.
MA
Journal of neuroimmune pharmacology : the official journal of the Society on NeuroImmune Pharmacology. 2010; 5: 271-7.
Koch-Henriksen N and Sorensen PS. The changing demographic pattern of multiple
D
19.
20.
PT E
sclerosis epidemiology. The Lancet Neurology. 2010; 9: 520-32. Pestehchian N, Etemadifarr M, Yousefi HA, Chiani M, Aslani N and Nasr Z. Frequency of
CE
Blood-tissue Parasitic Infections in Patients with Multiple Sclerosis, as Compared to their Family
21.
AC
Members. International journal of preventive medicine. 2014; 5: 1578-81. Stascheit F, Paul F, Harms L and Rosche B. Toxoplasma gondii seropositivity is negatively
associated with multiple sclerosis. J Neuroimmunol. 2015; 285: 119-24. 22.
Sellebjerg F, Bornsen L, Khademi M, et al. Increased cerebrospinal fluid concentrations of
the chemokine CXCL13 in active MS. Neurology. 2009; 73: 2003-10. 23.
Opdenakker G, Nelissen I and Van Damme J. Functional roles and therapeutic targeting of
gelatinase B and chemokines in multiple sclerosis. The Lancet Neurology. 2003; 2: 747-56. 13
ACCEPTED MANUSCRIPT 24.
Balabanian K, Brotin E, Biajoux V, et al. Proper desensitization of CXCR4 is required for
lymphocyte development and peripheral compartmentalization in mice. Blood. 2012; 119: 5722-30. 25.
Chen G and Goeddel DV. TNF-R1 signaling: a beautiful pathway. Science (New York, NY).
2002; 296: 1634-5. Herrington FD and Nibbs RJ. Regulation of the Adaptive Immune Response by the IkappaB
PT
26.
Kumar H, Kawai T and Akira S. Pathogen recognition by the innate immune system.
International reviews of immunology. 2011; 30: 16-34.
Wertz IE, Newton K, Seshasayee D, et al. Phosphorylation and linear ubiquitin direct A20
NU
28.
SC
27.
RI
Family Protein Bcl-3. Cells. 2016; 5.
AC
CE
PT E
D
MA
inhibition of inflammation. Nature. 2015; 528: 370-5.
14
ACCEPTED MANUSCRIPT Figure legends Figure 1．Validation of disrupted mRNA expression in Chinese MS patients and control subjects. Verification of potential disruption of mRNA expression in Chinese MS patients. Real-time PCR was performed to determine the potential up- and down-regulation of these MS genes (42
PT
unique genes in these 10 risk pathways) in Chinese MS patients compared to healthy control
RI
subjects. Interestingly, 31 of the 42 genes show significant differential expression in Chinese MS
AC
CE
PT E
D
MA
NU
SC
patients.
15
ACCEPTED MANUSCRIPT Table 1. 10 MS risk pathways and differentially expressed MS genes Pathway ID
Pathway Name
P value
Differentially expressed MS genes
hsa05145
Toxoplasmosis
1.42E-13
TNF TGFB1 HSPA1L JAK1 HLA-DQB1 ITGA6 NFKBIA NFKB1 TNFRSF1A BCL2 MAP3K7 CD40 HSPA1A CCR5
hsa04062
Chemokine signaling pathway
1.60E-10
GNG2 CXCR4 CDC42 ADCY9 NFKBIA IL8 NFKB1 CXCL1 KRAS CCL3 CXCR6 HRAS CCR5
hsa04621
NOD-like receptor signaling pathway
3.49E-09
TNF TNFAIP3 NFKBIA IL8 NFKB1 CXCL1 MAP3K7 IL1B
hsa04510
Focal adhesion
3.43E-06
JUN CDC42 COL11A2 ITGA4 ITGA6 PAK2 MYLK BCL2 HRAS
hsa04672
Intestinal immune network for IgA production
1.29E-05
TGFB1 CXCR4 CD40 HLA-DQB1 ITGA4
hsa05221
Acute myeloid leukemia
3.00E-04
KRAS PIM1 HRAS NFKB1
hsa04630
Jak-STAT signaling pathway
3.00E-04
PIM1 IL2RB JAK1 IL7R IL2RG IRF9
hsa04612
Antigen processing and presentation
1.00E-03
hsa03013
RNA transport
hsa04141
Protein processing in endoplasmic reticulum
D E 1.40E-03
T P E
1.16E-02
T P
I R
C S U
N A
M
TNF HSPA1L HLA-DQB1 HSPA1A EIF4A1 PNN EIF1 EIF3D CLNS1A HSPA1L BCL2 EDEM1 HSPA1A
C C
A
16
ACCEPTED MANUSCRIPT Table 2. Main Characteristics of selected MS cases and healthy control subjects in a Chinese population Control (n=70)
MS (n=132)
P value
Gender, M/F
27/43
53/79
0.88
Age (years), mean±SD
41±15
38±13
0.62
Age at onset (years), median (range)
-
31 (17-59)
-
Annual relapse rate, median (range)
-
0.4 (0.03-3.5)
-
Proportion OCBs positive/tested (%)
-
63/112 (56)
-
Proportion brain MRI abnormalities (%)
-
132/132 (100)
-
Proportion spinal MRI abnormalities (%)
-
89/132 (67)
-
EDSS score, median (range)
-
3 (1-9)
-
Poor neurological outcome (%)
-
29/132 (22)
-
NU
SC
RI
PT
Characteristics
MS, multiple sclerosis; OCBs, oligoclonal bands; MRI, magnetic resonance imaging; EDSS, Expanded Disability Status
AC
CE
PT E
D
MA
Scale.
17
ACCEPTED MANUSCRIPT Table 3. Differential mRNA expression profile of Chinese and European or American MS patients Expression in Chinese MS patients
Expression in European/American MS patients
Gene
Toxoplasmosis
Expression level
P value
Trend
Trend
Original studies
TNF
0.92±0.27
P=0.63
-
↓
Achiron et al. ; Bomprezzi et al.
TGFB1
2.76±0.93
P<0.001
↓
↓
Achiron et al. 16; Satoh et al. 18-19
HSPA1L
0.47±0.27
P<0.001
↓
↓
JAK1
2.81±2.04
P<0.001
↓
↓
HLA-DQB1
0.46±0.04
P<0.001
↓
↓
ITGA6
0.86±0.3
P=0.411
-
NFKBIA
2.49±0.48
P<0.001
↑
NFKB1
0.84±0.32
P=0.405
-
TNFRSF1A
2.72±1.22
P<0.001
BCL2
1.97±1.10
MAP3K7
2.36±1.31
T P
I R
17
17
Bomprezzi et al. ; Satoh et al.
18-19
↑
C S U
↑
Kemppinen et al. ; Satoh et al.
↓
Achiron et al. ; Arthur et al. ; Mandel et al.
↓
↓
Kemppinen et al. ; Satoh et al.
P<0.001
↑
↑
Arthur et al. ; Bomprezzi et al.
P<0.001
↓
↓
Kemppinen et al. ; Satoh et al.
3.43±2.33
P<0.001
↑
↑
Arthur et al. ; Kemppinen et al.
1.21±0.25
P=0.088
-
↓
Bomprezzi et al. 17; Gandhi et al. 21; Mandel et al. 22; Satoh et al. 18-19
CCR5
1.9±0.63
P<0.001
↓
↓
Kemppinen et al. ; Satoh et al.
GNG2
2.53±1.17
P<0.001
↓
↓
Gandhi et al. ; Satoh et al. [16]
CXCR4
2.88±1.8
P<0.001
↑
↑
Arthur et al. ; Kemppinen et al.
CD40 HSPA1A
Chemokine signaling pathway
16
PT
E C
C A
D E
N A
M
18
20
Arthur et al. ; Satoh et al.
18-19
Kemppinen et al. 15; Gandhi et al. 21 17
Bomprezzi et al. ; Mandel et al. 15
16
22
18-19
20
15
20
15
18-19
17
18-19
20
15
15
18-19
21
20
22
15
ACCEPTED MANUSCRIPT 21
0.89±0.28
P=0.739
-
↑
Gandhi et al. ; Satoh et al.
ADCY9
1.28±1.79
P=0.943
-
↓
Arthur et al. ; Kemppinen et al.
NFKBIA
2.49±0.48
P<0.001
↑
↑
Kemppinen et al. ; Satoh et al.
IL8
2.25±0.97
P<0.001
↓
↑
Kemppinen et al. 15; Mandel et al. 22
P=0.405
-
↓
20
15
0.84±0.32 NFKB1
NOD-like receptor signaling pathway
18-19
CDC42
CXCL1
1.78±0.58
P<0.001
↓
↑
KRAS
0.90±0.39
P=0.633
-
↓
CCL3
4.00±2.02
P=0.004
↑
CXCR6
1.77±0.99
P=0.774
-
HRAS
0.88±0.49
P=0.174
-
CCR5
1.9±0.63
P<0.001
TNF
0.92±0.27
TNFAIP3
3.55±2.14
18-19
T P
I R
16
20
Achiron et al. [7]; Arthur et al. ; Mandel et al.
C S U
N A ↓
15
20
Arthur et al. ; Satoh et al.
22
18-19
Gandhi et al. 21; Satoh et al. 18-19 Arthur et al. 20; Bomprezzi et al. 17; Mandel et al. 22
↓
Arthur et al. 20; Kemppinen et al. 15
↓
Arthur et al. 20; Satoh et al. [16]
↓
↓
Kemppinen et al. ; Satoh et al.
P=0.63
-
↓
Achiron et al. ; Bomprezzi et al.
P<0.001
↑
↑
Arthur et al. 20; Kemppinen et al. 15; Satoh et al. 18-19
2.49±0.48
P<0.001
↑
↑
Kemppinen et al. ; Satoh et al.
2.25±0.97
P<0.001
↓
↑
Kemppinen et al. 15; Mandel et al. [13]
NFKB1
0.84±0.32
P=0.405
-
↓
Achiron et al. ; Arthur et al. ; Mandel et al.
CXCL1
1.78±0.58
P<0.001
↓
↑
Arthur et al. ; Satoh et al.
MAP3K7
2.36±1.31
P<0.001
↓
↓
Kemppinen et al. ; Satoh et al.
NFKBIA IL8
PT
E C
C A
D E
M
19
15
16
15
16
18-19
17
18-19
20
20
15
18-19
18-19
22
ACCEPTED MANUSCRIPT Focal adhesion
Intestinal immune network for
15
0.93±0.45
P=0.907
-
↓
Achiron et al. ; Kemppinen et al. ; Mandel et al.
JUN
0.62±0.41
P<0.001
↓
↓
Bomprezzi et al. ; Satoh et al.
CDC42
0.89±0.28
P=0.739
-
↑
Gandhi et al. 21; Satoh et al. 18-19
COL11A2
5.09±1.26
P<0.001
↑
↑
Arthur et al. 20; Ramanathan et al. 23
ITGA4
3.69±2.4
P<0.001
↑
↑
Achiron et al. ; Arthur et al.
ITGA6
0.86±0.3
P=0.411
-
↑
Bomprezzi et al. 17; Mandel et al. 22
PAK2
0.56±0.3
P<0.001
↓
↓
MYLK
0.45±0.25
P<0.001
↓
↓
BCL2
1.97±1.10
P<0.001
↑
HRAS
0.88±0.49
P=0.174
-
TGFB1
2.76±0.93
P<0.001
↓
CXCR4
2.88±1.8
P<0.001
↑
CD40
3.43±2.33
P<0.001
↑
HLA-DQB1
0.46±0.04
P<0.001
↓
↓
Kemppinen et al. 15; Gandhi et al. 21
3.67±2.22
P<0.001
↑
↑
Achiron et al. ; Arthur et al.
0.90±0.39
P=0.633
-
↓
Gandhi et al. 21; Satoh et al. 18-19
PIM1
0.42±0.29
P=0.035
↓
↓
Gandhi et al. 21; Satoh et al. 18-19
HRAS
0.88±0.49
P=0.174
-
↓
Arthur et al. 20; Satoh et al. 18-19
NFKB1
0.84±0.32
P=0.405
-
↓
Achiron et al. ; Arthur et al. ; Mandel et al.
ITGA4 KRAS
E C
C A
D E
PT
IgA production
Acute myeloid leukemia
16
IL1B
17
M
I R
C S U
N A ↑
T P
16
21
18-19
Gandhi et al. ; Satoh et al.
20
18-19
Arthur et al. 20; Gandhi et al. 21 20
Arthur et al. ; Bomprezzi et al.
↓
Arthur et al. 20; Satoh et al. 18-19
↓
Achiron et al. 16; Satoh et al. 18-19
↑
Arthur et al. ; Kemppinen et al.
20
15
Kemppinen et al. ; Satoh et al.
20
17
16
16
15
18-19
20
20
22
22
ACCEPTED MANUSCRIPT Jak-STAT signaling pathway
Antigen processing and presentation
RNA transport
PIM1
0.42±0.29
P=0.035
↓
↓
Gandhi et al. 21; Satoh et al. 18-19
IL2RB
0.46±0.31
P=0.460
-
↓
Arthur et al. ; Satoh et al.
20
18-19
JAK1
2.81±2.04
P<0.001
↓
↓
Arthur et al. ; Satoh et al.
20
18-19
IL7R
3.32±1.93
P<0.001
↑
↑
IL2RG
0.50±0.27
P<0.001
↓
↓
Arthur et al. 20; Satoh et al. 18-19
IRF9
0.52±0.28
P<0.001
↓
↓
Arthur et al. 20; Satoh et al. 18-19
TNF
0.92±0.27
P=0.63
-
↓
HSPA1L
0.47±0.27
P<0.001
↓
↓
HLA-DQB1
0.46±0.04
P<0.001
↓
HSPA1A
1.21±0.25
P=0.088
-
EIF4A1
2.72±1.68
P<0.001
↑
PNN
3.91±2.18
P<0.001
EIF1
3.28±2.01
EIF3D
3.36±2.22
T P
C S U
N A ↓
I R
16
Achiron et al. , Bomprezzi et al. 17
Bomprezzi et al. ; Satoh et al.
17
18-19
Kemppinen et al. 15; Gandhi et al. 21
↓
Bomprezzi et al. 17; Gandhi et al. 21; Mandel et al. 22; Satoh et al. 18-19
↑
Arthur et al. 20; Kemppinen et al. 15; Gandhi et al. 21
↑
↑
Arthur et al. 20; Mandel et al. 22
P<0.001
↑
↑
Arthur et al. 20; Kemppinen et al. 15;
P<0.001
↑
↑
Arthur et al. 20; Gandhi et al. 21
3.74±1.85
P<0.001
↑
↑
Arthur et al. 20; Gandhi et al. 21
2.42±0.91
P<0.001
↓
↓
Bomprezzi et al. ; Satoh et al.
BCL2
1.97±1.10
P<0.001
↑
↑
Arthur et al. ; Bomprezzi et al.
EDEM1
3.44±2.12
P<0.001
↑
↑
Arthur et al. 20; Kemppinen et al. 15
HSPA1A
1.21±0.25
P=0.088
-
↓
Bomprezzi et al. 17; Gandhi et al. 21; Mandel et al. 22; Satoh et al. 18-19
CLNS1A Protein processing in endoplasmic reticulum
Bomprezzi et al. [17]; Ramanathan et al. [14]
HSPA1L
PT
E C
C A
D E
M
21
17
20
18-19
17
MA
NU
SC
RI
PT
ACCEPTED MANUSCRIPT
AC
CE
PT E
D
Fig. 1
22
AC
CE
PT E
D
MA
NU
SC
RI
PT
ACCEPTED MANUSCRIPT
Graphical Abstract
23
ACCEPTED MANUSCRIPT Highlights 1． We investigated potential up- and downregulation of 42 MS genes in 10 MS risk pathways.
AC
CE
PT E
D
MA
NU
SC
RI
PT
2． 31 differentially expressed genes in Chinese MS patients compared to healthy control subjects were identified. 3． the expression patterns of 28 of these genes were consistent with those obtained from Caucasian MS patients.
24
Guiyou Liu, Fang Zhang, Yang Hu, Yongshuai Jiang, Zhongying Gong, Shoufeng Liu, Xiuju Chen, Qinghua Jiang, Junwei Hao PII: DOI: Reference:
S0165-5728(16)30289-2 doi: 10.1016/j.jneuroim.2017.03.012 JNI 476548
To appear in:
Journal of Neuroimmunology
Received date: Revised date: Accepted date:
22 September 2016 5 March 2017 15 March 2017
Please cite this article as: Guiyou Liu, Fang Zhang, Yang Hu, Yongshuai Jiang, Zhongying Gong, Shoufeng Liu, Xiuju Chen, Qinghua Jiang, Junwei Hao , Multiple sclerosis risk pathways differ in Caucasian and Chinese populations. The address for the corresponding author was captured as affiliation for all authors. Please check if appropriate. Jni(2017), doi: 10.1016/j.jneuroim.2017.03.012
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 Multiple sclerosis risk pathways differ in Caucasian and Chinese populations
Guiyou Liu1, #, Fang Zhang2, #, Yang Hu1, Yongshuai Jiang3, Zhongying Gong4, Shoufeng Liu5, Xiuju
PT
Chen6, Qinghua Jiang1, *, Junwei Hao2, *
1
RI
School of Life Science and Technology, Harbin Institute of Technology, Harbin, China
2
SC
Department of Neurology and Tianjin Neurological Institute, Tianjin Medical University General
Hospital, Tianjin, China 3
NU
College of Bioinformatics Science and Technology, Harbin Medical University, Harbin, China
4
MA
Department of Neurology, Tianjin First Central Hospital, Tianjin, China
5
Department of Neurology, Tianjin HuanHu Hospital, Tianjin, China
6
These authors contributed equally to this work.
CE
#
PT E
D
Department of Neurology, Tianjin NanKai Hospital, Tianjin, China
AC
*Corresponding author: Qinghua Jiang School of Life Science and Technology, Harbin Institute of Technology, Harbin, Heilongjiang 150001, China
E-mail: [email protected] *Corresponding author: Junwei Hao Department of Neurology and Tianjin Neurological Institute, Tianjin Medical University General Hospital, Tianjin 300052, China, Tel/Fax: +86-22-6081-7429, Email: [email protected] 1
ACCEPTED MANUSCRIPT Abstract Large-scale genome-wide association study (GWAS) datasets provide strong support for investigations of the mechanisms underlying multiple sclerosis (MS) by using pathway analysis methods. In our recent study, we conducted a three-stage pathway analysis of GWAS and
PT
expression datasets. After identifying 15 shared MS pathways in separate MS GWAS datasets, we
RI
found that dysregulated MS genes were significantly enriched in 10 of the 15 MS risk pathways.
SC
Evidence showed that 17%-30% of genes are differentially expressed among individual ethnic populations. We then verified the potential disruption of genes in the 10 MS risk pathways cited
NU
above in Chinese MS patients. Here, we investigated potential up- and down-regulation of 42 MS
MA
genes in these 10 MS risk pathways using 132 Chinese MS patients and 76 healthy control subjects. We then identified 31 differentially expressed genes in Chinese MS patients compared to healthy
D
control subjects. Moreover, the expression patterns of 28 of these genes were consistent with those
PT E
obtained from Caucasian (European and American) MS patients, although 14 genes differed from the latter group’s. Our results provide clinically useful clues about the link between these risk genes
AC
CE
and MS susceptibility in the Chinese population.
Keywords: Multiple sclerosis; pathway analysis; genome-wide association study; Chinese population 2
ACCEPTED MANUSCRIPT 1 Introduction Multiple sclerosis (MS) is considered to be an inflammatory disease and is grouped amongst the most common diseases of the central nervous system (CNS) 1. As a means of detecting genetic risk variants, large-scale genome-wide association studies (GWAS) have been performed and
PT
provide strong support for identifying disease risk pathways 2-8. Baranzini et al. conducted a
RI
pathway analysis of two MS GWAS 3. Their identification of several significantly enriched pathways
SC
included such immune system pathways as cell adhesion, communication, and signaling as well as the neural pathways axon-guidance and synaptic potentiation 3. The International Multiple Sclerosis
NU
Genetics Consortium (IMSGC) similarly performed a pathway analysis of two large-scale MS
MA
GWAS datasets 9. Their report covered three significant gene ontology (GO) biological processes, including leukocyte activation, apoptosis, and positive regulation of macromolecule metabolic
D
process as well as three KEGG pathways consisting of the JAK-STAT signaling pathway, acute
PT E
myeloid leukemia, and T cell receptor signaling 9. We propose that the various MS GWAS datasets may share genetic pathways, although this
CE
possibility was not reported in previous studies 3, 9. Therefore, to test this hypothesis, we recently
AC
conducted a three-stage pathway analysis of GWAS and expression datasets 10. In stage 1, we analyzed pathways of two MS GWAS datasets. There, 15 shared pathways were found, most of which were related to the immune system and various diseases, environmental information processing, genetic information processing, and infectious diseases. In stage 2, we performed a candidate pathway analysis of the large-scale MS GWAS dataset from IMSGC to further verify the 15 shared pathways. We successfully replicated 14 of these 15 MS risk pathways. In stage 3, our pathway analysis was directed to the dysregulated MS gene list from seven human MS case-control 3
ACCEPTED MANUSCRIPT expression datasets derived from Caucasian (European and American) populations. In this stage, we found that dysregulated MS genes were significantly enriched in 10 of the 15 MS risk pathways identified in stages 1 and 2. Prior evidence showed that 17%-30% of genes are differentially expressed among individual
PT
ethnic populations 11. Here, to further verify the potential disruption of MS genes in Chinese MS
RI
patients, we performed real-time PCR to determine the potential up- and down-regulation of these
SC
MS genes in the MS risk pathways noted above using gene expression datasets from Chinese MS
NU
patients and healthy controls. 2 Materials and methods
MA
2.1 MS risk pathways and dysregulated genes
The MS risk pathways and dysregulated genes were collected from our recent study 10. Overall, we
D
performed a pathways analysis of three MS GWAS datasets including 334,923 single nucleotide
PT E
polymorphisms (SNPs) from 931 trios 12, 551,642 SNPs from 978 MS patients and 883 control subjects 13, and 441,547 SNPs from 9,772 MS patients and 17,376 controls of European descent 14,
CE
as well as 229 differentially expressed MS genes 15. Ten MS risk pathways as well as the
AC
corresponding dysregulated genes are described in Table 1. More detailed information appears in our recent study 10.
2.2 Study population and trial design in the Chinese population During open enrollment for the study, a total of 132 patients with relapsing-remitting MS in the acute stage of the disease, who also satisfied the McDonald Criteria of MS as revised in 2010, were recruited at Tianjin Medical University General Hospital, Tianjin HuanHu Hospital, Tianjin First Central Hospital, and Tianjin NanKai Hospital, Tianjin, China (Table 2). All patients manifested 4
ACCEPTED MANUSCRIPT multiple disseminations of the disease in space (i.e., involvement of multiple areas of the CNS) and time (i.e., ongoing disease activity over time). MS was also verified by the presence of oligoclonal bands in the cerebrospinal fluid (CSF) of all these patients. Exclusion criteria were the following: (1) presence of other diseases of the CNS in addition to MS, (2) presence of tumor(s) and systemic
PT
hematological diseases, (3) presence of recent infection, and (4) concomitant use of antineoplastic
RI
or immune-modulating therapies prior to blood sampling. We also recruited 76 age- and gender-
SC
matched healthy volunteers (control subjects) for the comparative study. This project was designed as a multicenter study. The trial protocol and supporting documentation were approved by the
MA
from all patients or legally acceptable surrogates.
NU
institutional review boards of each participating center. Informed consent was obtained at enrollment
2.3 Quantitative real-time PCR validation in MS patients and control subjects
D
Total RNA was extracted from peripheral blood mononuclear cells (PBMCs) by using Trizol®
PT E
reagent (Invitrogen) according to the manufacturer’s instructions. RNA quantity and quality were assessed using a Nanodrop ND-100 Spectrophotometer (Nanodrop Technologies, USA) and a 2100
CE
Bioanalyzer (Agilent RNA 6000 Nano Kit, Germany). For the reverse transcriptase (RT) reaction,
AC
SYBR Green RT reagents (Bio-Rad, USA) were used. The mRNA PCR results were quantified using the 2ΔΔct method against β-actin and GAPDH for normalization. The resulting data represent the means (±SD) of three experiments. 2.4 Statistical Analysis Genes expressed differentially in MS patients compared to healthy controls were analyzed by using Student’s t tests. Statistical significance was considered as P<0.05. All statistical data were analyzed by using SPSS 17.0 software (SPSS Inc., Chicago, IL, USA). 5
ACCEPTED MANUSCRIPT 3 Results 3.1 MS risk pathways and dysregulated genes Previously, we analyzed the large-scale MS GWAS and expression datasets in Caucasian populations 10. To summarize briefly here, that report highlighted 10 significantly dysregulated MS
PT
risk pathways (Table 1). Comparison of the corresponding dysregulated genes in those 10
RI
pathways yielded 42 unique genes. Subsequent performance of real-time PCR revealed the primers
SC
for all the 42 genes, which are presented here in Supplementary Table 1. 3.2 Gene expression in Chinese MS patients and control subjects
NU
Results from our in-depth analysis of these 42 genes are provided in Table 3. Interestingly, 31 of
MA
these 42 genes showed significant differential expression in Chinese MS patients compared to the matched control subjects (Figure 1). We then compared in more detail the expression of these 31
D
significant genes in Chinese MS patients with those from European and American MS patients and
PT E
found that the expression trends of 28 of the 31 genes from Chinese participants were consistent with the results obtained from European and American MS patients (Table 3). COL11A2 was the
CE
most elevated (5.09-fold higher expression), followed by CCL3 (4.0-fold higher expression), and
AC
PNN (3.91-fold higher expression). PIM1 was the least expressed gene (3.31-fold lower expression), followed by JAK1 (2.81-fold lower expression), and TGFB1 (2.76-fold lower expression). The trends of these genes tested in Chinese MS patients were consistent with results obtained from the Caucasian populations, with the exception of CCL3, which had a lower expression in Caucasians. 3.3 Comparison of gene expression in Chinese and Caucasian MS patients The expression patterns of 28 (67%) (COL11A2, PNN, CLNS1A, ITGA4, TNFAIP3, EDEM1, CD40, EIF3D, IL7R, PIM1, EIF1, CXCR4, JAK1, TGFB1, EIF4A1, TNFRSF1A, GAG2, NFKBIA, MYLK,
6
ACCEPTED MANUSCRIPT IL2RG, HSPA1L, MAP3K7, IRF9, HLA-DQB1, JUN, BCL2, PAK2, CCR5) of these 42 genes were consistent with those obtained from Caucasian (European and American) MS patients. However, 14 (33%) of the 42 genes were expressed inconsistently with regard to those obtained from Caucasian MS patients. Eleven of the latter group’s genes (NFKB1, ITGA6, HRAS, CDC42, KRAS, TNF, IL1B,
PT
HSPA1A, ADCY9, IL2RB, CXCR6) had no different expression in Chinese MS patients, and 3
RI
genes (CCL3, IL8, CXCL1) expressed an opposite trend.
SC
3.4 Top three significant pathway analyses
In our recent study 10, we found that the toxoplasmosis pathway was the most significant pathway
NU
among the three MS GWAS datasets and seven human MS case-control expression datasets
MA
derived from Caucasian populations. Ten genes (TGFB1, HSPA1L, JAK1, HLA-DQB1, NFKBIA, TNFRSF1A, BCL2, MAP3K7, CD40, CCR5) were aberrantly expressed in the toxoplasmosis
D
pathway of Chinese MS patients compared to 14 genes in Caucasian populations. However, the
PT E
expression tendency in Chinese MS patients correlated strongly with that of the Caucasian populations.
CE
The chemokine signaling pathway was the second most significant pathway in the GWAS and
AC
expression analyses. Seven genes (GNG2, CXCR4, NFKBIA, IL8, CXCL1, CCL3, CCR5) had aberrant expression of the chemokine signaling pathway in Chinese MS patients compared to 13 in Caucasian populations.
In the NOD-like receptor signaling pathway, TNFAIP3 and NFKBIA were highly expressed in both Chinese MS patients and Caucasian populations. However, the expression of IL8, CXCL1 and MAP3K7 was low in Chinese MS patients. 4 Discussion 7
ACCEPTED MANUSCRIPT Two former studies that attempted pathway and network-based analysis methods for investigating multiple MS GWAS datasets failed to identify any shared genetic pathway 3, 9. To extend that search, we recently performed a three-stage analysis of GWAS and expression datasets. As a result, both the GWAS and expression datasets provided outcomes that highlight the involvement of 10 MS risk
PT
KEGG pathways in Caucasian (European and American) MS patients.
RI
With our interest in differential gene expression among diverse ethnic populations, we
SC
investigated the potential up- and down-regulation of 42 MS genes in the MS risk pathways. To that end, we examined 132 Chinese MS patients and 76 healthy control subjects and successfully
NU
identified 31 genes expressed in Chinese MS group but not in their healthy counterparts. The
MA
mRNA PCR results were quantified using the 2ΔΔct method against β-actin and GAPDH for normalization. Compared to healthy control subjects, in Chinese MS patients 31 differentially
D
expressed genes were identified. We further found that the expression trends of 28 of those 31
PT E
genes (90%) were consistent with the results obtained from European and American MS patients but another 5 genes (16%) that were inconsistent with those of the Caucasians. Our findings are in
CE
agreement with previous studies showing that 17%-30% of genes are differentially expressed in
AC
separate ethnic populations 11. Lu et al. analyzed about 15,000 transcripts expressed in lymphoblastoid samples and identified about 4,700 (31.3%) transcripts that were differentially expressed between the HapMap CEU and Asian (CHB + JPT) populations 11. In the 10 MS risk KEGG pathways cited above, the toxoplasmosis pathway is the most significant signal. In this pathway, 14 genes showed different expression in Caucasian populations. Here, we identified 10 of these 14 genes with clearly aberrant expression in Chinese MS patients. Extensive evidence indicates that MS is a chronic demyelinating disease of the central nervous 8
ACCEPTED MANUSCRIPT system and that its occurrence is more prevalent in economically advanced nations than in the developing countries 16-17. The lesser prevalence of parasitic infections (which can activate immune responses and prevent or modulate damage to host antigens) in the more modernized areas is among the possible responsible factors for such a difference 18-19. Toxoplasmosis is not highly
PT
prevalent in MS patients; that is, 36% and 49%, respectively, of MS patients and their family
RI
members were documented as having toxoplasmosis 20. Stascheit et al. investigated the prevalence
SC
of Toxoplasma gondii IgG and IgM antibodies in 163 MS patients and 178 control subjects 21. The negative association between parasitic Toxoplasma gondii infection and the presence of MS they
21
.
MA
compared with healthy controls
NU
reported reflected significantly lower counts of Toxoplasma gondii IgG antibodies in MS patients
We further compared the second significant genetic pathway, chemokine signaling, in
D
Caucasian and Chinese populations. In our previous study, 13 genes of Caucasian populations
PT E
differed in chemokine activity from that of the Chinese. Here, we successfully replicated 7 of these 13 genes showing aberrant expression in Chinese MS patients. Growing evidence suggests that
CE
chemokines and chemokine receptors participate in the recruitment of macrophages and T
AC
lymphocytes into the CNS, and it has been considered the most critical mechanism in the pathogenesis of MS 22-23. In these pathways, two genes, CXCR4 and NFKBIA, are highly expressed in both Chinese and Caucasian MS patients. CXCR4 is a receptor of CXCL12. In turn, CXCL12–CXCR4 interactions initiate multiple signaling pathways that augment T cell co-stimulation, proliferation, cytokine production, migration, and survival 24. NFKBIA, another important target of the chemokine signaling pathway, is also a highly expressed participant. NFKBIA normally retains NFkB within the cytoplasms of unstimulated cells and thereby inhibits the inflammatory response 9
25-26
.
ACCEPTED MANUSCRIPT Therefore, the overexpression of NFKBIA may be a protective response in MS patients. The NOD-like receptor signaling pathway is the third significant signal of interest in this survey. NOD-like receptors are intracellular receptors capable of regulating both innate and adaptive immune responses 27. Upon recognition of their respective ligands, NOD-like receptors become
PT
activated, and that transformation results in the subsequent triggering of multiple pro-inflammatory
RI
molecular pathways; one such example was nuclear factor-kappa B (NF-κB). In our study, we found
SC
that TNFAIP3 is highly expressed in both Chinese MS patients and Caucasian populations. Inactivation of the TNFAIP3 gene, encoding the A20 protein, is associated with critical inflammatory
NU
diseases including MS, rheumatoid arthritis and Crohn’s disease28. All the foregoing findings
MA
suggest that TNFAIP3 and NOD-like receptors pathway may be the important targets of MS pathogenesis.
D
Despite the clarity of this outcome, our use of PBMCs for the real time PCR validation research
PT E
represented a limitation of the study. PBMCs were the basis for our real-time PCR validation research. PBMCs consist of a variety of cells in various abundance that may lead to differences in
CE
observed gene expression for the whole PBMC group. These cells were selected because they
AC
reflect the change in cellular composition of MS patients as compared to controls but are somewhat nonspecific. Additionally, 11 MS patients had calcium tablet treatments and 6 received hemp seed soft capsules for purging before blood sampling. In summary, our multiple pathway analyses of GWAS and expression datasets highlight shared and dysregulated MS risk pathways in Caucasian (European and American) populations 10. In the Chinese population, we reported consistent but also inconsistent results compared to those in European and American Caucasians. The expression patterns of 28 (67%) genes were consistent 10
ACCEPTED MANUSCRIPT with those obtained from all Caucasian MS patients; however, 14 (33%) genes were inconsistent with those obtained from the Caucasians. Of those 14 genes, 11 were no different in Chinese MS patients, and 3 genes (CCL3, IL8, CXCL1) had the opposite expression trend. The importance of these results is new knowledge that may serve to determine treatment options according to patients’
PT
genetic heritage.
RI
Conflict of interest
SC
The authors declare no conflict of interests. Acknowledgments
NU
This work was financially supported by the National Natural Science Foundation of China
MA
(81300945 to G. Y. L., and 81571600, 81322018, 81273287 and 81100887 to J. W. H.); the Youth Top-notch Talent Support Program; and the National Key Clinical Specialty Construction Project of
D
China.
1.
PT E
References
Damotte V, Guillot-Noel L, Patsopoulos NA, et al. A gene pathway analysis highlights the
2.
AC
32.
CE
role of cellular adhesion molecules in multiple sclerosis susceptibility. Genes Immun. 2014; 15: 126-
Liu G, Jiang Y, Wang P, et al. Cell adhesion molecules contribute to Alzheimer's disease:
multiple pathway analyses of two genome-wide association studies. J Neurochem. 2012; 120: 190-8. 3.
Baranzini SE, Galwey NW, Wang J, et al. Pathway and network-based analysis of genome-
wide association studies in multiple sclerosis. Hum Mol Genet. 2009; 18: 2078-90. 4.
Jiang Q, Jin S, Jiang Y, et al. Alzheimer's Disease Variants with the Genome-Wide
Significance are Significantly Enriched in Immune Pathways and Active in Immune Cells. Mol 11
ACCEPTED MANUSCRIPT Neurobiol. 2016. 5.
Bao X, Liu G, Jiang Y, et al. Cell adhesion molecule pathway genes are regulated by cis-
regulatory SNPs and show significantly altered expression in Alzheimer's disease brains. Neurobiol Aging. 2015; 36: 2904 e1-7. Quan B, Qi X, Yu Z, et al. Pathway analysis of genome-wide association study and
PT
6.
RI
transcriptome data highlights new biological pathways in colorectal cancer. Mol Genet Genomics.
7.
SC
2015; 290: 603-10.
Liu G, Yao L, Liu J, et al. Cardiovascular disease contributes to Alzheimer's disease:
Liu G, Jiang Y, Chen X, et al. Measles contributes to rheumatoid arthritis: evidence from
MA
8.
NU
evidence from large-scale genome-wide association studies. Neurobiol Aging. 2014; 35: 786-92.
pathway and network analyses of genome-wide association studies. PLoS One. 2013; 8: e75951. Network-based multiple sclerosis pathway analysis with GWAS data from 15,000 cases and
D
9.
10.
PT E
30,000 controls. Am J Hum Genet. 2013; 92: 854-65. Liu G, Zhang F, Jiang Y, et al. Integrating genome-wide association studies and gene
Lu J and Clark AG. Impact of microRNA regulation on variation in human gene expression.
AC
11.
CE
expression data highlights dysregulated multiple sclerosis risk pathways. Mult Scler. 2016.
Genome Res. 2012; 22: 1243-54. 12.
Hafler DA, Compston A, Sawcer S, et al. Risk alleles for multiple sclerosis identified by a
genomewide study. N Engl J Med. 2007; 357: 851-62. 13.
Baranzini SE, Wang J, Gibson RA, et al. Genome-wide association analysis of susceptibility
and clinical phenotype in multiple sclerosis. Hum Mol Genet. 2009; 18: 767-78. 14.
Sawcer S, Hellenthal G, Pirinen M, et al. Genetic risk and a primary role for cell-mediated 12
ACCEPTED MANUSCRIPT immune mechanisms in multiple sclerosis. Nature. 2011; 476: 214-9. 15.
Kemppinen AK, Kaprio J, Palotie A and Saarela J. Systematic review of genome-wide
expression studies in multiple sclerosis. BMJ open. 2011; 1: e000053. 16.
Ahlgren C, Lycke J, Oden A and Andersen O. High risk of MS in Iranian immigrants in
La Flamme AC, Ruddenklau K and Backstrom BT. Schistosomiasis decreases central system
inflammation
and
alters the
progression
of
experimental
autoimmune
SC
nervous
RI
17.
PT
Gothenburg, Sweden. Multiple sclerosis (Houndmills, Basingstoke, England). 2010; 16: 1079-82.
encephalomyelitis. Infection and immunity. 2003; 71: 4996-5004.
Ascherio A and Munger KL. Epstein-barr virus infection and multiple sclerosis: a review.
NU
18.
MA
Journal of neuroimmune pharmacology : the official journal of the Society on NeuroImmune Pharmacology. 2010; 5: 271-7.
Koch-Henriksen N and Sorensen PS. The changing demographic pattern of multiple
D
19.
20.
PT E
sclerosis epidemiology. The Lancet Neurology. 2010; 9: 520-32. Pestehchian N, Etemadifarr M, Yousefi HA, Chiani M, Aslani N and Nasr Z. Frequency of
CE
Blood-tissue Parasitic Infections in Patients with Multiple Sclerosis, as Compared to their Family
21.
AC
Members. International journal of preventive medicine. 2014; 5: 1578-81. Stascheit F, Paul F, Harms L and Rosche B. Toxoplasma gondii seropositivity is negatively
associated with multiple sclerosis. J Neuroimmunol. 2015; 285: 119-24. 22.
Sellebjerg F, Bornsen L, Khademi M, et al. Increased cerebrospinal fluid concentrations of
the chemokine CXCL13 in active MS. Neurology. 2009; 73: 2003-10. 23.
Opdenakker G, Nelissen I and Van Damme J. Functional roles and therapeutic targeting of
gelatinase B and chemokines in multiple sclerosis. The Lancet Neurology. 2003; 2: 747-56. 13
ACCEPTED MANUSCRIPT 24.
Balabanian K, Brotin E, Biajoux V, et al. Proper desensitization of CXCR4 is required for
lymphocyte development and peripheral compartmentalization in mice. Blood. 2012; 119: 5722-30. 25.
Chen G and Goeddel DV. TNF-R1 signaling: a beautiful pathway. Science (New York, NY).
2002; 296: 1634-5. Herrington FD and Nibbs RJ. Regulation of the Adaptive Immune Response by the IkappaB
PT
26.
Kumar H, Kawai T and Akira S. Pathogen recognition by the innate immune system.
International reviews of immunology. 2011; 30: 16-34.
Wertz IE, Newton K, Seshasayee D, et al. Phosphorylation and linear ubiquitin direct A20
NU
28.
SC
27.
RI
Family Protein Bcl-3. Cells. 2016; 5.
AC
CE
PT E
D
MA
inhibition of inflammation. Nature. 2015; 528: 370-5.
14
ACCEPTED MANUSCRIPT Figure legends Figure 1．Validation of disrupted mRNA expression in Chinese MS patients and control subjects. Verification of potential disruption of mRNA expression in Chinese MS patients. Real-time PCR was performed to determine the potential up- and down-regulation of these MS genes (42
PT
unique genes in these 10 risk pathways) in Chinese MS patients compared to healthy control
RI
subjects. Interestingly, 31 of the 42 genes show significant differential expression in Chinese MS
AC
CE
PT E
D
MA
NU
SC
patients.
15
ACCEPTED MANUSCRIPT Table 1. 10 MS risk pathways and differentially expressed MS genes Pathway ID
Pathway Name
P value
Differentially expressed MS genes
hsa05145
Toxoplasmosis
1.42E-13
TNF TGFB1 HSPA1L JAK1 HLA-DQB1 ITGA6 NFKBIA NFKB1 TNFRSF1A BCL2 MAP3K7 CD40 HSPA1A CCR5
hsa04062
Chemokine signaling pathway
1.60E-10
GNG2 CXCR4 CDC42 ADCY9 NFKBIA IL8 NFKB1 CXCL1 KRAS CCL3 CXCR6 HRAS CCR5
hsa04621
NOD-like receptor signaling pathway
3.49E-09
TNF TNFAIP3 NFKBIA IL8 NFKB1 CXCL1 MAP3K7 IL1B
hsa04510
Focal adhesion
3.43E-06
JUN CDC42 COL11A2 ITGA4 ITGA6 PAK2 MYLK BCL2 HRAS
hsa04672
Intestinal immune network for IgA production
1.29E-05
TGFB1 CXCR4 CD40 HLA-DQB1 ITGA4
hsa05221
Acute myeloid leukemia
3.00E-04
KRAS PIM1 HRAS NFKB1
hsa04630
Jak-STAT signaling pathway
3.00E-04
PIM1 IL2RB JAK1 IL7R IL2RG IRF9
hsa04612
Antigen processing and presentation
1.00E-03
hsa03013
RNA transport
hsa04141
Protein processing in endoplasmic reticulum
D E 1.40E-03
T P E
1.16E-02
T P
I R
C S U
N A
M
TNF HSPA1L HLA-DQB1 HSPA1A EIF4A1 PNN EIF1 EIF3D CLNS1A HSPA1L BCL2 EDEM1 HSPA1A
C C
A
16
ACCEPTED MANUSCRIPT Table 2. Main Characteristics of selected MS cases and healthy control subjects in a Chinese population Control (n=70)
MS (n=132)
P value
Gender, M/F
27/43
53/79
0.88
Age (years), mean±SD
41±15
38±13
0.62
Age at onset (years), median (range)
-
31 (17-59)
-
Annual relapse rate, median (range)
-
0.4 (0.03-3.5)
-
Proportion OCBs positive/tested (%)
-
63/112 (56)
-
Proportion brain MRI abnormalities (%)
-
132/132 (100)
-
Proportion spinal MRI abnormalities (%)
-
89/132 (67)
-
EDSS score, median (range)
-
3 (1-9)
-
Poor neurological outcome (%)
-
29/132 (22)
-
NU
SC
RI
PT
Characteristics
MS, multiple sclerosis; OCBs, oligoclonal bands; MRI, magnetic resonance imaging; EDSS, Expanded Disability Status
AC
CE
PT E
D
MA
Scale.
17
ACCEPTED MANUSCRIPT Table 3. Differential mRNA expression profile of Chinese and European or American MS patients Expression in Chinese MS patients
Expression in European/American MS patients
Gene
Toxoplasmosis
Expression level
P value
Trend
Trend
Original studies
TNF
0.92±0.27
P=0.63
-
↓
Achiron et al. ; Bomprezzi et al.
TGFB1
2.76±0.93
P<0.001
↓
↓
Achiron et al. 16; Satoh et al. 18-19
HSPA1L
0.47±0.27
P<0.001
↓
↓
JAK1
2.81±2.04
P<0.001
↓
↓
HLA-DQB1
0.46±0.04
P<0.001
↓
↓
ITGA6
0.86±0.3
P=0.411
-
NFKBIA
2.49±0.48
P<0.001
↑
NFKB1
0.84±0.32
P=0.405
-
TNFRSF1A
2.72±1.22
P<0.001
BCL2
1.97±1.10
MAP3K7
2.36±1.31
T P
I R
17
17
Bomprezzi et al. ; Satoh et al.
18-19
↑
C S U
↑
Kemppinen et al. ; Satoh et al.
↓
Achiron et al. ; Arthur et al. ; Mandel et al.
↓
↓
Kemppinen et al. ; Satoh et al.
P<0.001
↑
↑
Arthur et al. ; Bomprezzi et al.
P<0.001
↓
↓
Kemppinen et al. ; Satoh et al.
3.43±2.33
P<0.001
↑
↑
Arthur et al. ; Kemppinen et al.
1.21±0.25
P=0.088
-
↓
Bomprezzi et al. 17; Gandhi et al. 21; Mandel et al. 22; Satoh et al. 18-19
CCR5
1.9±0.63
P<0.001
↓
↓
Kemppinen et al. ; Satoh et al.
GNG2
2.53±1.17
P<0.001
↓
↓
Gandhi et al. ; Satoh et al. [16]
CXCR4
2.88±1.8
P<0.001
↑
↑
Arthur et al. ; Kemppinen et al.
CD40 HSPA1A
Chemokine signaling pathway
16
PT
E C
C A
D E
N A
M
18
20
Arthur et al. ; Satoh et al.
18-19
Kemppinen et al. 15; Gandhi et al. 21 17
Bomprezzi et al. ; Mandel et al. 15
16
22
18-19
20
15
20
15
18-19
17
18-19
20
15
15
18-19
21
20
22
15
ACCEPTED MANUSCRIPT 21
0.89±0.28
P=0.739
-
↑
Gandhi et al. ; Satoh et al.
ADCY9
1.28±1.79
P=0.943
-
↓
Arthur et al. ; Kemppinen et al.
NFKBIA
2.49±0.48
P<0.001
↑
↑
Kemppinen et al. ; Satoh et al.
IL8
2.25±0.97
P<0.001
↓
↑
Kemppinen et al. 15; Mandel et al. 22
P=0.405
-
↓
20
15
0.84±0.32 NFKB1
NOD-like receptor signaling pathway
18-19
CDC42
CXCL1
1.78±0.58
P<0.001
↓
↑
KRAS
0.90±0.39
P=0.633
-
↓
CCL3
4.00±2.02
P=0.004
↑
CXCR6
1.77±0.99
P=0.774
-
HRAS
0.88±0.49
P=0.174
-
CCR5
1.9±0.63
P<0.001
TNF
0.92±0.27
TNFAIP3
3.55±2.14
18-19
T P
I R
16
20
Achiron et al. [7]; Arthur et al. ; Mandel et al.
C S U
N A ↓
15
20
Arthur et al. ; Satoh et al.
22
18-19
Gandhi et al. 21; Satoh et al. 18-19 Arthur et al. 20; Bomprezzi et al. 17; Mandel et al. 22
↓
Arthur et al. 20; Kemppinen et al. 15
↓
Arthur et al. 20; Satoh et al. [16]
↓
↓
Kemppinen et al. ; Satoh et al.
P=0.63
-
↓
Achiron et al. ; Bomprezzi et al.
P<0.001
↑
↑
Arthur et al. 20; Kemppinen et al. 15; Satoh et al. 18-19
2.49±0.48
P<0.001
↑
↑
Kemppinen et al. ; Satoh et al.
2.25±0.97
P<0.001
↓
↑
Kemppinen et al. 15; Mandel et al. [13]
NFKB1
0.84±0.32
P=0.405
-
↓
Achiron et al. ; Arthur et al. ; Mandel et al.
CXCL1
1.78±0.58
P<0.001
↓
↑
Arthur et al. ; Satoh et al.
MAP3K7
2.36±1.31
P<0.001
↓
↓
Kemppinen et al. ; Satoh et al.
NFKBIA IL8
PT
E C
C A
D E
M
19
15
16
15
16
18-19
17
18-19
20
20
15
18-19
18-19
22
ACCEPTED MANUSCRIPT Focal adhesion
Intestinal immune network for
15
0.93±0.45
P=0.907
-
↓
Achiron et al. ; Kemppinen et al. ; Mandel et al.
JUN
0.62±0.41
P<0.001
↓
↓
Bomprezzi et al. ; Satoh et al.
CDC42
0.89±0.28
P=0.739
-
↑
Gandhi et al. 21; Satoh et al. 18-19
COL11A2
5.09±1.26
P<0.001
↑
↑
Arthur et al. 20; Ramanathan et al. 23
ITGA4
3.69±2.4
P<0.001
↑
↑
Achiron et al. ; Arthur et al.
ITGA6
0.86±0.3
P=0.411
-
↑
Bomprezzi et al. 17; Mandel et al. 22
PAK2
0.56±0.3
P<0.001
↓
↓
MYLK
0.45±0.25
P<0.001
↓
↓
BCL2
1.97±1.10
P<0.001
↑
HRAS
0.88±0.49
P=0.174
-
TGFB1
2.76±0.93
P<0.001
↓
CXCR4
2.88±1.8
P<0.001
↑
CD40
3.43±2.33
P<0.001
↑
HLA-DQB1
0.46±0.04
P<0.001
↓
↓
Kemppinen et al. 15; Gandhi et al. 21
3.67±2.22
P<0.001
↑
↑
Achiron et al. ; Arthur et al.
0.90±0.39
P=0.633
-
↓
Gandhi et al. 21; Satoh et al. 18-19
PIM1
0.42±0.29
P=0.035
↓
↓
Gandhi et al. 21; Satoh et al. 18-19
HRAS
0.88±0.49
P=0.174
-
↓
Arthur et al. 20; Satoh et al. 18-19
NFKB1
0.84±0.32
P=0.405
-
↓
Achiron et al. ; Arthur et al. ; Mandel et al.
ITGA4 KRAS
E C
C A
D E
PT
IgA production
Acute myeloid leukemia
16
IL1B
17
M
I R
C S U
N A ↑
T P
16
21
18-19
Gandhi et al. ; Satoh et al.
20
18-19
Arthur et al. 20; Gandhi et al. 21 20
Arthur et al. ; Bomprezzi et al.
↓
Arthur et al. 20; Satoh et al. 18-19
↓
Achiron et al. 16; Satoh et al. 18-19
↑
Arthur et al. ; Kemppinen et al.
20
15
Kemppinen et al. ; Satoh et al.
20
17
16
16
15
18-19
20
20
22
22
ACCEPTED MANUSCRIPT Jak-STAT signaling pathway
Antigen processing and presentation
RNA transport
PIM1
0.42±0.29
P=0.035
↓
↓
Gandhi et al. 21; Satoh et al. 18-19
IL2RB
0.46±0.31
P=0.460
-
↓
Arthur et al. ; Satoh et al.
20
18-19
JAK1
2.81±2.04
P<0.001
↓
↓
Arthur et al. ; Satoh et al.
20
18-19
IL7R
3.32±1.93
P<0.001
↑
↑
IL2RG
0.50±0.27
P<0.001
↓
↓
Arthur et al. 20; Satoh et al. 18-19
IRF9
0.52±0.28
P<0.001
↓
↓
Arthur et al. 20; Satoh et al. 18-19
TNF
0.92±0.27
P=0.63
-
↓
HSPA1L
0.47±0.27
P<0.001
↓
↓
HLA-DQB1
0.46±0.04
P<0.001
↓
HSPA1A
1.21±0.25
P=0.088
-
EIF4A1
2.72±1.68
P<0.001
↑
PNN
3.91±2.18
P<0.001
EIF1
3.28±2.01
EIF3D
3.36±2.22
T P
C S U
N A ↓
I R
16
Achiron et al. , Bomprezzi et al. 17
Bomprezzi et al. ; Satoh et al.
17
18-19
Kemppinen et al. 15; Gandhi et al. 21
↓
Bomprezzi et al. 17; Gandhi et al. 21; Mandel et al. 22; Satoh et al. 18-19
↑
Arthur et al. 20; Kemppinen et al. 15; Gandhi et al. 21
↑
↑
Arthur et al. 20; Mandel et al. 22
P<0.001
↑
↑
Arthur et al. 20; Kemppinen et al. 15;
P<0.001
↑
↑
Arthur et al. 20; Gandhi et al. 21
3.74±1.85
P<0.001
↑
↑
Arthur et al. 20; Gandhi et al. 21
2.42±0.91
P<0.001
↓
↓
Bomprezzi et al. ; Satoh et al.
BCL2
1.97±1.10
P<0.001
↑
↑
Arthur et al. ; Bomprezzi et al.
EDEM1
3.44±2.12
P<0.001
↑
↑
Arthur et al. 20; Kemppinen et al. 15
HSPA1A
1.21±0.25
P=0.088
-
↓
Bomprezzi et al. 17; Gandhi et al. 21; Mandel et al. 22; Satoh et al. 18-19
CLNS1A Protein processing in endoplasmic reticulum
Bomprezzi et al. [17]; Ramanathan et al. [14]
HSPA1L
PT
E C
C A
D E
M
21
17
20
18-19
17
MA
NU
SC
RI
PT
ACCEPTED MANUSCRIPT
AC
CE
PT E
D
Fig. 1
22
AC
CE
PT E
D
MA
NU
SC
RI
PT
ACCEPTED MANUSCRIPT
Graphical Abstract
23
ACCEPTED MANUSCRIPT Highlights 1． We investigated potential up- and downregulation of 42 MS genes in 10 MS risk pathways.
AC
CE
PT E
D
MA
NU
SC
RI
PT
2． 31 differentially expressed genes in Chinese MS patients compared to healthy control subjects were identified. 3． the expression patterns of 28 of these genes were consistent with those obtained from Caucasian MS patients.
24