/

University of Southern California Dissertations and Theses

/

Association testing of polymorphisms in the vitamin D metabolism and signaling pathway with multiple sclerosis across multiple populatuions

(USC Thesis Other)

Association testing of polymorphisms in the vitamin D metabolism and signaling pathway with multiple sclerosis across multiple populatuions

PDF

Transcript (if available)

Content
ASSOCIATION TESTING OF POLYMORPHISMS IN THE VITAMIN D METABOLISM AND
SIGNALING PATHWAY WITH MULTIPLE SCLEROSIS ACROSS MULTIPLE POPULATIONS

by
Jenn Rong Chiam

A Thesis Presented to the
FACULTY OF THE USC GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
MASTER OF SCIENCE
(BIOSTATISTICS)

August 2012

Copyright 2012 Jenn Rong Chiam

ii

ACKNOWLEDGEMENTS
I would like to thank my advisor Dr. Talat Islam, for his guidance and support throughout
my research and writing process. I would like to also thank Dr. Thomas Mack and Dr. Jim
Gauderman for serving as my committee members and providing valuable suggestions.
This thesis would not have been possible without the collaboration with Dr. Jorge
Oksenberg and his team from the Department of Neurology, University of California,
San Francisco. Finally, I would like to thank my family and friends for all the love,
support and encouragement.

iii

TABLE OF CONTENTS
ACKNOWLEDGEMENTS ii

LIST OF TABLES iv

LIST OF FIGURES v

ABSTRACT vi

CHAPTER 1: INTRODUCTION 1
1.1 Background 1
1.2 Primary Specific Aim 2
1.3 Secondary Specific Aim 4

CHAPTER 2: METHODS AND MATERIALS 6
2.1 Study Populations 6
2.1.1 International Twins Study (ITS) 6
2.1.2 University of San Francisco Multiple Sclerosis Families Study (UCSF-MS) 7
2.1.3 60k Study 8
2.1.4 3L-GWAS Study 8
2.2 Polymorphism Selection and Genotyping 9
2.3 Covariates 11
2.4 Statistical Analyses 11
2.4.1 Meta Analysis 12
2.4.2 Adjustment for Multiple Genetic Testing 13
2.4.3 Effect Modification 15

CHAPTER 3: RESULTS 16
3.1 Association between Vitamin D polymorphisms and MS (Meta Analysis) 16
3.2 Effect Modification by HLA-DRB1*1501 18

CHAPTER 4: DISCUSSION 22

REFERENCES 30

iv

LIST OF TABLES
Table 1: Characteristics of study populations used 6

Table 2: Descriptive characteristics of individual 3L-GWAS cohorts 9

Table 3: Candidate Vitamin D metabolism and signaling pathway related genes 10

Table 4: Number of independent tests needed to correct for multiple testing 14

Table 5: Top 20 vitamin D related SNPs and its association with MS 17

Table 6: Effect modification of Vitamin D genes by HLA (PCA approach) 18

Table 7: Effect modification of Vitamin D polymorphisms by HLA 19

Table 8: Gene-gene interaction between HLA and rs17467825 by genotype 19

Table 9: Gene-gene interaction between HLA and rs7041 by genotype 20

Table 10: Gene-gene interaction between HLA and rs213212 by genotype 20

Table 11: Gene-gene interaction between HLA and rs6531 by genotype 20

Table 12: Gene-gene interaction between HLA and rs1704996 by genotype 21

Table 13: Functionality of SNPs associated with MS on chromosome 12q13-14 25

v

LISTS OF FIGURES
Figure 1: Vitamin D metabolism and signaling pathway 4

Figure 2: Linkage Disequilibrium of the CYP27B1 and adjacent gene regions 23

Figure 3: Linkage Disequilibrium of the RXRB and adjacent gene regions 27

vi

ABSTRACT
Vitamin D is an environmental factor known to protect against multiple sclerosis
(MS). Furthermore, it has also been shown to influence activities of immunological
genes such as the HLA-DRB1*1501, which have found to be associated with this
autoimmune disease. Genes controlling the metabolism and signaling functions of
vitamin D are also likely to be MS susceptibility regions.
In this thesis, we examined the association between nine genes of the vitamin D
pathway and MS. The functional relevance of significantly identified single nucleotide
polymorphisms (SNPs) as they contribute to MS was also explored. After conducting a
meta-analysis, we found rs928329 from CYP27B1 to be significantly associated with MS
(meta p < 0.0002) while all other vitamin D genes did not yield significant associations.
Significant interactions between RXRB and HLA-DRB1*1501 identified in this
thesis indicate the importance of complex gene-gene interaction in MS. Given the
importance of vitamin D as an immune-regulator and the important role of immunity in
MS, future investigation of interaction between vitamin D and other immune-related
genes will likely reveal important biological mechanisms that cause MS.

1

CHAPTER 1: INTRODUCTION
1.1 Background
Multiple Sclerosis (MS) is a chronic autoimmune disease in which the protective
myelin sheaths surrounding the nerve cells become damaged
1
. Even though the
etiological pathway of MS is poorly understood, previous studies have elucidated that
both environmental and genetic factors act independently and synergistically to cause
the disease
2
. In particular, vitamin D is one important MS factor shown to be not only a
protective environmental factor but also an important regulator of immunologic cell
known to increase MS susceptibility
2, 3
.
The geographical differences in MS prevalence led to the discovery that
ultraviolet radiation and vitamin D are two closely related environmental MS factors
4
.
Epidemiological studies of prenatal and childhood sun exposure
4-7
, as well as vitamin D
dietary supplementation
8, 9
contribute to the notion that increasing vitamin D levels
have a protective effect against MS. Furthermore, a prospective nested case-control
study of 444 subjects drawn from a cohort of 7 million US military personnel also found
that a 50 nmol/L increase in serum vitamin D reduced the risk of MS by 41%
10
.
In addition, vitamin D is a crucial immune-modulator of the immunological
pathway known to affect the activities of T regulatory cells
11, 12
. Cells of the immune
system express CYP27B1 and VDR, two vitamin D related genes, which allow for local
response to 1,25-dihydroxyvitamin D3 (1,25(OH)
2
D
3
), the biologically active vitamin D
metabolite
13
. Vitamin D deficiency and weakened VDR signals has been shown to

2

increase activities of T-helper 1 (Th1) cells while simultaneously decreasing regulatory T-
cells and T-helper 2 (TH2) cell activity
14
. These immunological changes may result in
autoimmune attacks against the myelin basic protein (MBP) in MS
15, 16
.
The genes controlling the metabolism and signaling functions of vitamin D are
also important MS susceptibility regions and have received attention in MS genetic
research in recent years
17-19
. By examining the association between the vitamin D genes
and MS, we can deepen our understanding of the role vitamin D plays in MS. This
approach also allows us to postulate if any particular vitamin D metabolite is especially
important in the disease mechanism. Several genome-wide association studies (GWAS)
have identified genome-wide significant associations between MS and two vitamin D
genes (CYP27B1 and CYP24A1)
20-22
despite non-significant findings from smaller case-
control studies
17-19, 23-25
. However, a distinction of whether these identified susceptibility
variants have causative effects or are just simply in linkage disequilibrium (LD) with yet-
to-be identified causal variants remains elusive. Therefore, a thorough understanding of
the functional properties of identified variants as they pertain to MS development is
needed.

1.2 Primary Specific Aim
This thesis reports the results of an association study of polymorphisms in the
vitamin D gene pathway with MS across multiple populations using a meta-analysis. The
primary aim of this thesis is to identify significant single-nucleotide polymorphisms

3

(SNPs) associated with MS, and to understand their linkage disequilibrium (LD) structure
and to generate hypotheses on the plausible biological mechanisms by which these
SNPs confer disease risk based on online databases describing gene function and
regulation.
The genes analyzed were selected based on their role in metabolizing and
regulating the various vitamin D substrates after vitamin D
3
is produced in the skin when
exposed to UVR (Figure 1). Vitamin D can also be obtained from dietary sources,
although it usually only accounts for a small proportion of the initial vitamin D entering
the body. Vitamin D
3
is biologically inert and requires sequential hydroxylation in order
to be metabolized into its active 1,25(OH)
2
D
3
form
26
. Vitamin D binding protein (DBP)
binds and transports the gut- and skin-produced vitamin D3 to the liver, where vitamin
D-25-hydrxylase enzymes (CYP27A1 and CYP2R1) hydroxylates vitamin D3 into 25-
hydroxyvitamin D3 (25(OH)D
3
). Once formed, this major circulating form of vitamin D is
again transported by DBP to the kidneys to be further hydroxylated into 1,25-
dihydroxyvitamin D3 (1,25(OH)
2
D
3
), by 1-hydroxylase (CYP27B1)
14
. Whenever blood
plasma 1,25(OH)
2
D
3
are too high, 24-OHase (CYP24A1) catabolizes 25-(OH)D
3
and
1,25(OH)
2
D
3
into inactive 24,25(OH)
2
D
3
and 1,24,25(OH)
2
D
3
respectively. This ensure
that the biologically active 1,25(OH)
2
D
3
is maintained within a narrow optimal range
27
.
Once transported to target tissues, 1,25(OH)
2
D
3
binds to the Vitamin D Receptor (VDR)
to form a complex that dimerizes with nuclear receptor retinoic X receptors (RXR -
RXRA, RXRB, and RXRG). This heterodimer complex, 1,25(OH)
2
D
3
/VDR/RXR
,
then binds

4

to the vitamin D receptor element (VDRE) of the target tissue to exert the effect of
vitamin D
28
.
Figure 1: Vitamin D metabolism and signaling pathway
Pathway genes are printed in bold letters. 1,25(OH)
2
D
3
in target tissue is represented by .

Identifying significant SNPs from of these vitamin D related genes that may be
associated with MS and (exploring its functional importance) will undoubtedly reveal
plausible defects in biological mechanisms of vitamin D regulation and receptivity in
cells that leads to MS.

1.3 Secondary Specific Aim
The HLA-DRB1*1501 haplotype is another important genetic risk factor that has
been consistently found to be associated with MS. The vitamin D responsive element
(VDRE) in the promoter region of HLA DRB1*1501 was recently found to be completely

5

conserved
29
. No mutation of this VDRE sequence was found among 322 HLA-DRB1*15
homozygous individuals regardless of MS status, suggesting that there exists some
evolutionary pressures that allow for the preservation of this response element. This
VDRE has been functionally shown to impact higher vitamin D sensitivity to HLA
DRB1*1501, providing strong biological reasons to test if HLA interacts with vitamin D
genes to cause MS. Hence, the secondary aim of this thesis will be to explore the
possible effect modification of the MS-vitamin D relationship by HLA.

6

CHAPTER 2: METHODS AND MATERIALS
2.1 Study Populations
To study the role of the polymorphisms from the nine vitamin D pathway genes,
a meta-analysis approach combining results from four study populations were used. The
four distinct populations used in this thesis consists of: (i) 564 subjects from the
International Twins Study (ITS)
30
(ii) 444 subjects from the University of California, San
Francisco MS families study (UCSF-MS)
31
, (iii) 2722 participants across three clinical sites
across the United States and United Kingdom in a GWAS
32
which we shall call “60k”,
and lastly (iv) 9844 participants from three large international genome-wide association
studies across Europe and USA (3L-GWAS)
33
. Table 1 provides a description of the
characteristics of all populations used in this study.
Table 1: Characteristics of study populations used
Population ITS UCSF-MS 60k 3L-GWAS
Study Design Case-relative
control
Gender-
matched case-
sibling control
General
population case-
control
General
population
case-control
Female: Male Ratio 1.66 3.0 1.75 3.1
Cases (N) 250 222 1343 2624
Controls (N) 314 222 1379 7220
tagSNPs (N) 190 190 155 183

2.1.1 International Twins Study (ITS)
Between 1980 and 1992, non-Hispanic Caucasian twins with chronic disease
(including MS) were recruited to form a volunteer twin registry, through newspaper
advertisements placed throughout North America
30
. Ascertainment of the original
cohort of MS subjects was set up to identify twin pairs in whom at least one member

7

had physician-diagnosed MS. Blood samples were later collected from all contactable
twins as well as that from either a non-twin sibling or relative control. All participants
received informed consent approved by the University of Southern California. Overall,
DNA samples were collected from total of 749 eligible cases and controls (328 cases and
421 controls). However, approximately an equal proportion of cases and controls (24%
and 25% respectively) had a poor genotyping call rate of less than 90%, resulting in
these individuals being excluded. A total of 250 cases and 314 controls were retained for
the analyses.
2.1.2 University of San Francisco Multiple Sclerosis Families Study (UCSF-MS)
The UCSF-MS study was a research effort employing rigorous ascertainment
criteria to identify single- and multiple-case MS families from the MS registry at UCSF, or
by physician referral, collaborative arrangements with other MS centers in the USA, and
advertisement in newsletters of MS societies
31
. These families, of European ancestry,
were required to have an affected proband with either two living parents or at least one
unaffected sibling. All study participants gave informed consent as approved by the
appropriate Institutional Review Boards (IRB).
A brief telephone interview was used to determine eligibility for the study and to
collect detailed family history information. Unaffected siblings were interviewed and
examined by one of the study neurologists before their blood sample was taken. A total
of 222 cases and an equal number of sibling controls from only single-case MS families
were genotyped for our study.

8

2.1.3 60k Study
MS cases were recruited from three MS clinics of University of California, San
Francisco; Harvard/MIT Board Institute; and Cambridge University as part of MS genetic
study
32, 33
. Cases and controls were frequency-matched on age and gender. For the US
cases, unrelated hospital controls with no familial history of chronic inflammatory
diseases were selected from each respective clinic, whereas controls for the British
cases were selected from existing 1958 Birth Cohort Study. All participants were self-
reported non-Hispanic whites and gave informed consent. The study was approved by
local IRBs prior to enrollment in the study. Blood samples were collected from all
participants at the respective clinic. Overall, there were 1343 cases and 1379 controls in
the 60k study.
2.1.4 3L-GWAS Study
Participants from three genome-wide association studies
33, 34
were combined in
a large international MS genetic consortium (IMSGC) to identify potentially important
genetic loci. A total of 9844 self-reported Non-Hispanic subjects from six distinct cohorts
across Europe and USA were obtained (Table 2). All cases met the McDonald criteria at
enrollment. The cases were mostly recruited by neurological associations or by
neurologists at MS specialist clinical centers in their respective location. Controls
(hospital or friend-referral) were primarily unrelated individuals with no family history
of chronic inflammatory diseases. Each participating site was responsible for collecting
blood samples from eligible participants who had given informed consent prior to

9

enrollment in the study. The overall female-to-male ratio in the 3L-GWAS population
was 3.1.
Table 2: Descriptive characteristics of individual 3L-GWAS cohorts

Cohort
IMSGC
UK
IMSGC
US
GeneMSA
NL
GeneMSA
CH
GeneMSA
US

BWH
Cases 453 342 253 230 486 860
Controls 2950 1679 208 232 431 1720
Sex (F:M) ratio 3.0 3.2 2.9 2.8 3.1 3.2
Platform Affy
500k
Affy
500k
Illumina
550
Illumina
500
Illumina
500
Affy 6.0
IMSGC, International MS Genetics Consortium; UK, United Kingdom; US, United States; NL,
Netherlands; CH, Switzerland; BWH, Brigham and Women’s Hospital, Boston, MA;
Reprinted by permission from Macmillan Publishers Ltd: [Nature Genetics] (De Jager PL et al:
Meta-analysis of genome scans and replication identify CD6, IRF8, and TNFRSF1A as new
multiple sclerosis susceptibility loci), copyright © 2009

2.2 Polymorphism Selection and Genotyping
For the ITS and UCSF-MS studies, a SNP selection program was developed to
select a minimal set of surrogate SNPs (tagSNPs) that could efficiently capture genetic
information across the targeted gene (20kb upstream of transcription to 10kb
downstream of each gene). The program utilized linkage disequilibrium (LD) information
based on either the HapMap samples of European ancestry (CEU) data or user-defined
populations in order to group SNPs that satisfied pre-specified pair-wise LD thresholds in
the same bin. The algorithm also exploited additional information associated with each
individual SNP, such as the minor allele frequency (MAF ≥ 0.05), the type of SNP
(intronic or exonic) and any known functionality. Fine-tuning the initial selection
conditions ensured that there was a balanced spacing between the SNPs throughout the

10

targeted chromosome. Furthermore, all SNPs with previous known functionality were
also included unconditionally while SNPs with a low probability of being genotyped
successfully were excluded. After applying this method, a total of 196 SNPs from the
nine pathway genes of interest were identified (Table 3); DNA genotyping for the ITS
and UCSF-MS data was conducted using the Illumina platform.
Table 3: Candidate Vitamin D metabolism and signaling pathway related genes
Gene
Name
Chromosomal
Location

NM_Number
Number
of Exons
HapMap
tagSNPs
tagSNP Average
Spacing
GC 4q12-q13 NM_000583 13 15 5575.00
CYP27A1 2q33-qter NM_000784 9 11 9044.29
CYP2R1 11p15.2 NM_024514 11 12 4017.64
CYP27B1 12q13.3-q14 NM_000785 9 8 5714.29
CYP24A1 20q13.2-q13.3 NM_000782 12 42 1203.24
VDR 12q12-q14 NM_000376 11 41 2460.32
RXRA 9q34 NM_002957 10 27 5143.64
RXRB 6p21.3 NM_021976 10 11 3076.92
RXRG 1q22-q23 NM_006917 10 29 2554.45

Of the 196 SNPs identified above, a total of 149 and 183 SNPs were available
from the 60k and 3L-GWAS studies, respectively. For the 60k study, a total of 48,767
custom SNPs across the genome were genotyped using the Illumina Infinium 60K Bead-
Chip assay
33
. All SNPs genotyped in the ITS, UCSF-MS and 60k studies, are coded as 0, 1,
2 copies of the minor allele. In the 3L-GWAS study, as different genotyping platforms
were used across the six distinct cohorts (Table 2), imputation of missing SNPs were
conducted using the MACH imputation algorithm
35
based on phased data from HapMap
CEU samples. The case-control status of all subjects was ignored during imputation to
avoid introducing artifacts between cases and controls. Fractional genotypic scores
generated from the imputation algorithm were directly analyzed without conversion

11

into categorical genotypes so as to minimize variance inflation. The distribution of
fractional genotype scores was tri-modal with the peaks at 0, 1 and 2. Overall, the
combined 3L-GWAS dataset produced a common panel of 2.56 million SNPs.

2.3 Covariates
Gender was the only covariate collected across all study population. Age of
diagnosis for cases were collected for both the ITS and UCSF-MS samples. Sun exposure
information and HLA information were available from participants in the ITS population,
including place of birth. The SNP rs3135388 was the tagging SNP for the HLA-DRB1*1501
haplotype genotyped in the ITS population.

2.4 Statistical Analyses
Given that there were four distinct study populations, we could have adopted
either 1) a discovery-replication approach using the ITS population as the discovery set
or 2) a meta-analysis approach combining results from all study populations. With
insufficient power in the ITS discovery set, the discovery-replication approach was not
optimal for identifying important vitamin D related SNPs associated with MS. Hence the
more powerful meta-analysis method was implemented, thereby allowing the sample
sizes of the respective populations to be appropriately taken into account.

12

2.4.1 Meta Analysis
For each individual study, logistic regression models were used to test the
association between the SNPs and MS, where an additive model was assumed to assess
the effect of having increasing number of the minor allele of each particular variant.
Both the ITS and 60k models were adjusted for gender, whereas conditional logistic
regression models were used for the UCSF-MS study as cases and controls were
specifically matched on gender. Summary statistics for the unconditional logistic
regression of the 3L-GWAS model (odds ratios and p-value) were provided by the IMSGC
researchers. The 3L-GWAS analyses were adjusted for gender and centre (Table 2). We
then undertook a fixed effect meta-analysis approach
36
to combine the association
results from all four different study populations, taking into account both the strength
and direction of association of each individual SNP from all populations, where
available. This approach allowed for sufficient power to detect genetic variations with
moderate or small associations with MS after adjusting for multiple testing.
The odds ratio (OR), p-value (and corresponding z-scores) as well as the sample
size associated with a specific allele in each population were incorporated in the meta-
analysis to obtain a single meta p-value. The meta-analysis test statistic was calculated
as a linear combination of the individual population z-scores with weights proportional
to the square root of the corresponding sample size. This statistic also took into account
whether the direction of OR (in the ITS, USCF-MS or 60K study) was in agreement with
that of the 3L-GWAS population, which had the largest sample size. Any OR in the

13

opposite direction from that of the 3L-GWAS association resulted in a negative weight
given to the corresponding z-score for that population being compared. Under the null
hypothesis, the statistics follows a standard normal distribution and the final meta p-
values were obtained using normal approximations. As 3L-GWAS was used as the
reference population, meta p-values were calculated only for 183 SNPs that were
available from the 3L-GWAS population.
2.4.2 Adjustment for Multiple Genetic Testing
In testing the association of the vitamin D SNPs with MS, multiple hypotheses
were tested simultaneously. To adjust the significance level using the Bonferroni
method such that the individual hypotheses can be directly compared, all tests must be
assumed to be independent of each other. In the context of association testing
involving SNPs, high linkage disequilibrium (LD) exists between SNPs closely spaced on
the same gene. As such, the total number of hypotheses conducted does not necessarily
equate to the number of truly independent tests needed. However, if the number of
independent tests can be correctly inferred, the Bonferroni method can then be used to
rapidly adjust for multiple testing.
To address our concern regarding multiple testing, we used the SimpleM
approach established by Gao et al
37
. The number of independent tests for each
particular gene was determined via Principle Components (PC). Taking into account the
pair-wise correlations between the N SNPs of a specified gene, we used the PC approach
to parse out the total variation of the joint distribution of these SNPs into N

14

independent principle components (eigenvectors). The N corresponding eigenvalues
(λ’s) represent progressively decreasing amounts of variance explained by each
eigenvector. For each gene-specific PC analysis, we required that the first x (x ≤ N)
eigenvalues explain at least 99.5% of the variation that exists between the genic SNPs.
To determine the total number of truly independent tests conducted, denoted as M
eff
,
all x’s from each gene-specific PC analysis were summed together. At this stage, we can
then appropriately apply the simple Bonferroni method by dividing the family-wise type
I error by the total number of truly independent tests.
Based on SimpleM approach, even though we conducted a total of 183
association tests, there were only 139 truly independent tests (Table 4). Applying the
Bonferroni correction, the effective significance cut-off used for each analysis was set to
3.6x10
-4
.
Table 4: Number of independent tests needed to correct for multiple testing

Gene

Chromosome
No. of SNPs analyzed
in Meta Analysis
No. of PC explaining 99.5% of total
variance in SimpleM
RXRG 1 29 21
CYP27A1 2 7 6
GC 4 14 13
RXRB 6 11 10
RXRA 9 27 22
CYP2R1 11 12 8
VDR 12 38 27
CYP27B1 12 7 6
CYP24 20 38 26
Total SNPs:183 M
eff
=139

15

2.4.3 Effect Modification
We also examined possible effect modification of the association between the
vitamin D polymorphisms and MS by another important MS risk factor – HLA
DRB1*1501 haplotype. The analyses were restricted to individuals in the ITS population
as the covariate was available only in this population. For the interaction models, we
first focused on a gene-centric approach using Principle Component Analysis (PCA),
before moving forward with specific vitamin D SNP-HLA interactions. This PCA approach
reduces the dimensionality of the data while incorporating information from across all
SNPS in the same gene, ensuring we do not miss any important gene-HLA interactions
using a significance cut-off of 0.05. In this gene-centric approach, PCA was conducted to
determine the number of PCs needed to explain at least 80% of total genic variance.
Likelihood ratio tests were then used to compare base models containing only HLA and
PC terms as explanatory variables to the full model which also contained appropriate
product terms between HLA and the PC terms. All vitamin D genes that were
significantly modified by HLA were then further analyzed in individual vitamin D SNP-
HLA interaction models. Test of interaction between the relevant vitamin D related
polymorphisms and HLA were conducted using conditional logistic regression with
appropriate product terms. All interaction models were adjusted for gender as it is an
important MS factor that also confounds the relationship between HLA and MS.

16

CHAPTER 3: RESULTS
3.1 Association between Vitamin D polymorphisms and MS (Meta Analysis)
The meta-analysis combining the MS risk association from all four studies yielded
one significant finding. After adjusting for multiple testing, rs923829 (meta p < 0.0002)
from CYP27B1 was significantly associated with MS (Table 5). However, several
polymorphisms from RXRB (rs2854028, meta p < 0.004; rs9277936, meta p < 0.005;
rs213212, meta p < 0.01) were also found to be marginally significant. We present the
top 20 SNPs from the meta analysis and associated polymorphism-related information
(such as chromosomal location and minor allele frequency) in Table 5.

17

18

3.2 Effect Modification by HLA-DRB1*1501
The tagging SNP for the HLA-DRB1*1501 haplotype, rs3135388, was significantly
associated with MS after adjustment for gender only in the ITS sample (OR=2.28,
p<0.0001). Using the PCA approach to identify important HLA-vitamin D gene
interactions, we found that the association between MS and RXRB was significantly
modified by HLA (p = 0.05). Marginally significant vitamin D genes included GC, CYP27B1
and RXRB (Table 6).
Table 6: Effect modification of vitamin D genes by HLA (PCA approach)
Gene Interaction P-value
GC 0.10
‡

CYP27A1 0.18
CYP2R1 0.29
CYP27B1 0.07
‡

CYP24A1 0.64
VDR 0.64
RXRA 0.20
RXRB 0.05
†

RXRG 0.92
†: Significant interaction (p ≤ 0.05)
‡: Marginally significant interaction (0.10 ≤ p < 0.05)

We then focused on the individual SNP-HLA interactions for GC, CYP27B1 and
RXRB genes. Using a likelihood ratio test to test for possible interactions between the
vitamin D polymorphisms and rs3135388, significant HLA interactions were found for GC
SNPs: rs17467825 and rs7041, and RXRB SNPs: rs212312, rs6531 and rs1704996 (Table
7). For example, the odds of developing MS for a HLA-DRB1*1501 positive subjective
increases by 94% for each additional minor allele of rs17467825 after adjustment for

19

gender (OR= 2.62; 95% CI= 1.26 – 5.42). The top 10 HLA-vitamin D polymorphism
interaction based on the genic PCA approach are summarized in Table 7.
Table 7: Effect modification of vitamin D polymorphisms by HLA

Gene

SNP
Interaction
OR†
Interaction
P

N

Ran k‡
GC rs17467825 2.62 0.01 528 1
rs7041 2.07 0.03 527 3
rs221999 1.81 0.11 527 7
RXRB rs213212 2.31 0.02 528 2
rs6531 2.15 0.03 523 4
rs1704996 0.19 0.04 525 5
rs2744505 2.13 0.06 522 6
rs213208 1.55 0.13 528 10
CYP27B1 rs1048691 1.94 0.11 526 8
CYP27B1 rs2270777 1.59 0.12 526 9
†: Adjusted for sex
‡: Rank is based upon interaction P-value (in order of decreasing significance)

We further explored the interactions for all HLA-interactions where the p-value
was less than 0.05: GC SNPs: rs17467825 and rs7041 and RXRB SNPs: rs212312, rs6531
and rs1704996 by across the specific genotypes (Table 7-11). We set the reference
group to be that of those individuals who were HLA-DRB1*1501 negative and had 2
major alleles of the respective vitamin D related SNP.
Table 8: Gene-gene interaction between HLA and rs17467825 by genotype
GC SNP:
rs17467825
HLA tagging SNP:
rs3135388

OR†
AA N REF
AG N 0.38 (0.20-0.76)
GG N 0.21 (0.05-0.92)
AA Y 1.22 (0.67-2.24)
AG Y 1.64 (0.83-3.25)
GG Y 0.69 (0.15-3.12)

P for trend: 0.34
†: Adjusted for sex

20

Table 9: Gene-gene interaction between HLA and rs7041 by genotype
GC SNP:
rs7041
HLA tagging SNP:
rs3135388

OR†
TT N REF
TG N 0.42 (0.22-0.79)
GG N 0.26 (0.09-0.76)
TT Y 1.08 (0.53-2.21)
TG Y 1.06 (0.51-2.21)
GG Y 1.10 (0.36-3.30)

P for trend: 0.12
†: Adjusted for sex

Table 10: Gene-gene interaction between HLA and rs213212 by genotype
RXRB SNP:
rs213212
HLA tagging SNP:
rs3135388

OR†
TT N REF
TG N 0.66 (0.36-1.20)
GG N 0.40 (0.07-2.28)
TT Y 1.49 (0.81-2.75)
TG Y 1.72 (0.88-3.37)
GG Y 3.44 (1.28-9.23)

P for trend: 0.01
†: Adjusted for sex

Table 11: Gene-gene interaction between HLA and rs6531 by genotype
RXRB SNP:
rs6531
HLA tagging SNP:
rs3135388

OR†
TT N REF
TC N 0.63 (0.34-1.17)
CC N 0.45 (0.13-1.63)
TT Y 1.42 (0.77-2.60)
TC Y 1.57 (0.80-3.07)
CC Y 1.99 (0.87-4.57)

P for trend: 0.04
†: Adjusted for sex

21

Table 12: Gene-gene interaction between HLA and rs1704996 by genotype
RXRB SNP:
rs1704996
HLA tagging SNP:
rs3135388

OR†
TT N REF
TG N 2.03 (0.66-6.25)
TT Y 2.50 (1.50-4.18)
TG Y 1.03 (0.36-2.97)

P for trend: 0.67
†: Adjusted for sex

When considering both the HLA and GC genotypes simultaneously, no pattern
was observed; however an interesting trend was observed for the interaction between
HLA and RXRB. Being HLA-DRB1*1501 positive increased disease risk, regardless of
genotype for RXRB SNPs rs213212 and rs6531 (Table 9-10). In particular, a three-fold
increased odds ratio (OR = 3.44, 95% CI = 1.28 – 9.23) of MS was observed for those who
were both HLA-positive with two minor rs213212 G allele compared to those who were
HLA-negative with two major rs213212 T allele. A similar trend was observed when
considering both HLA-DRB1 and rs6531. Given that the homozygous variant was not
observed for rs1704996, no distinct pattern with increasing number of minor allele
across the HLA haplotype could be observed.

22

CHAPTER 4: DISCUSSION
Previous genetic studies in MS have identified SNPs from two vitamin D pathway
genes, CYP27B1 and CYP24A1, to be associated with MS. In this thesis, we found a
positive finding for CYP27B1, but not for CYP24A1 (rs2248359; meta p=0.04). Other
vitamin D gene regions were not found to be associated with MS, consistent with that of
published literature. After genotyping and analyzing the set of tagSNPs distributed
across the nine vitamin D related genes in a meta-analysis, rs923829, a SNP in close
approximation to CYP27B1, was found to be statistically associated with MS (meta p <
0.0002). Marginally significant SNPs include rs2854028 and rs9277936 from RXRB (meta
p < 0.004 and p < 0.005 respectively. Overall, these findings lend further support to the
involvement of vitamin D and its related genes in MS.
Within chromosome 12q13-14, there exists an MS susceptibility region
consisting of 17 genes (including CYP27B1) that are in tight linkage disequilibrium (LD)
38
. The SNP rs703842, which lies 1.76 kb upstream in the promoter region of CYP27B1,
has been found to be protective against MS in a GWAS study conducted by the
Australian and New Zealand MS Genetic Consortium (ANZgene)
17
(OR=0.81, p=5.4 x 10
-
11
). Other CYP27B1 SNPs located on the same LD block also appear to protect against MS
(rs4646536, rs10877012 and rs10877015, OR range: 0.87-0.88, p = 3.5 × 10
−3
)
39, 40
. It
should be noted that the significant CYP27B1 association identified in this thesis was
largely driven by the 3L-GWAS study, which had the largest sample size. A recently
published genome-wide meta-analysis combining this 3L-GWAS population together

23

with that of the ANZgene has replicated the significant finding for rs703842
41
, which is
at high LD with rs923829 (Figure 2). Future studies aiming to replicate the CYP27B1
finding in MS requires the use of independent populations.
Figure 2: Linkage Disequilibrium of the CYP27B1 and adjacent gene regions

24

Nonetheless, it remains uncertain whether any of the previously identified SNPs
from this chromosome 12q13-14 region is actually the causal SNP. We present the gene
location and known functionality of previously identified SNPs from chromosome
12q13-14 that were found to be associated with MS in Table 8. Functionality studies
may elucidate whether any of these SNPs is the truly casual SNP. It is equally important
that synonymous SNPs are not excluded from functionality studies; recent experimental
work has uncovered possible mechanisms by which synonymous SNPs can alter the
dynamics of the DNA/mRNA or expression levels of proteins
42
. As a result, synonymous
SNPs such as rs923829 may be involved in disease pathogenesis even if no peptide
change in the resulting protein occurs. Given its importance in the bio-activation of
vitamin D, CYP27B1 remains a strong biological susceptibility gene from this region. Fine
mapping of the chromosome 12q13-14 region may help to identify the actual disease-
causing SNP, as the SNPs identified thus far may simply be in LD with the truly casual
SNP. In addition, DNA methylation and chromatin modification studies involving
CYP27B1 (and other important genetic regions) may provide a more detail
understanding of the genetic influences in MS pathogenesis. This is especially true
because the identified SNPs tend to lie in non-coding regions (Table 8), such as the
promoter region of CYP27B1, implying a putative role in the regulation of gene
expression through epigenetic mechanisms.

25

Table 13: Functionality of SNPs associated with MS on chromosome 12q13-14
SNP found to be associated with MS Gene Function
Rs923829 METTL21B Coding synonymous
Rs703842 CYP27B1 Promoter
Rs4646536 CYP27B1 Intron
Rs10877012 CYP27B1 Promoter
Rs10877015 METTL21B Intron

Previous case-control studies investigating the association between VDR and MS
have yielded inconsistent results. Initial association studies investigating VDR in MS
among a population of Japanese MS patients and controls, positively identified
rs1544410 (Bsm1) to be associated with MS (OR (95%CI) = 2.44 (1.20 -5.00)).
Subsequent association studies among Caucasian populations, however, yielded
negative findings
18, 23, 43
. A recent meta-analysis combining 11 case-controls involving
VDR polymorphisms found that none of the frequently investigated VDR SNPs were
associated with MS (Apa1: OR(95% CI) = 0.90 (0.69-1.17); Bsm-1: OR(95% CI) = 1.09
(0.48-2.46)
44
. Furthermore, while several GWAS have identified vitamin D related genes
such as CYP27B1 and CYP24A1 to be associated with MS, no genome-wide significant
association has been identified for VDR. Similarly, we did not find any association for
VDR with Ms in this thesis. All 38 VDR polymorphisms investigated were not found to be
associated with MS, including Apa-1 (3L-GWAS OR: 0.99; meta p = 0.84) and Bsm-1(3L-
GWAS OR: 0.99; meta-p = 0.65). Given the recent identification of increased VDR
binding sites on many immune-related MS-associated genes
45
, one possible route in
which VDR contributes to MS is likely to involve complex interaction with the immune
system.

26

Using a PCA approach to analyze HLA interactions at the genic level, we
identified significant vitamin D gene-HLA interaction for RXRB. Further investigation at
the individual polymorphism level revealed a complex relationship between some RXRB
SNPs, HLA-DRB1 and MS. There appears to be an increasing MS risk among HLA-positive
individuals with increasing number of minor allele for both rs213212 and rs6531.
However, the similar trend of increasing MS risk among HLA-positive individuals with
increasing number of minor allele in the interaction for both rs213212 and rs6531 may
be attributed to high LD structure within RXRB and adjacent region (Figure 3).
Polymorphisms which were genotyped in this thesis from the neighboring RXRB region
includes the ring finger protein gene (RING1), Estradiol 17-beta-dehydrogenase 8
(HSD17B8), zinc transporter (SLC39A7) and Collagen, type XI, alpha 2 gene (COL11A2).
Out of the aforementioned genes, SLC39A7 may also be important in MS base on the
recent findings that serum zinc levels were significantly lower among a group of Iranian
MS patients compared to healthy subjects
46
. However, it remains to be determined how
SLC39A7 may interact with HLA-DRB1 in MS etiology. Given the highly conserved VDRE
in the promoter of HLA-DRB1 results in increased vitamin D sensitivity
29
, RXRB remains
the most likely candidate for which a meaningful biological interaction with HLA, via the
1,25(OH)
2
D
3
-VDR-RXR complex, may occur. It remains to be determined whether the
coding synonymous rs6531 polymorphism within RXRB is the truly causal variant that
can influence HLA to cause MS. Nevertheless, future gene expression studies and
functional studies involving rs6531 as well as others SNPs in this region will be valuable

27

for uncovering the biological link between RXRB and HLA in MS. The HLA-rs213212
association within MS identified in this thesis may be attributable to the LD pattern of
this SNP with respect to the true underlying variant as the functional properties of SNP
rs213212, which is located 20k upstream from RXRB within the RING1 gene, is unknown.
Figure 3: Linkage Disequilibrium of the RXRB and adjacent gene regions

Strengths of this thesis include the combination of four distinct study
populations in a meta-analysis, which enables us to conduct a powerful study. This
ensured that significantly identified associations detected represent plausible biological

28

mechanisms involving vitamin D in MS development. In addition, we have provided for
the first time a candidate gene approach involving the metabolism and signaling
pathway of vitamin D in MS etiology. In this approach a minimal but sufficient set of
polymorphisms across nine genes that directly exert influence on the function and levels
of the various vitamin D metabolites within the body, were genotyped. However, given
that these candidate genes are essentially part of the vitamin D metabolism and
signaling pathway, integrating pathway analysis with results from published GWAS could
further identify putative disease mechanisms involving vitamin D genes.
The limitations of the thesis include no appropriate adjustments for ancestry-
related variables. All participants in all four populations were assumed to be of
European ancestry based on self-reported ethnicity data. Insignificant associations for
some vitamin D related genes may be a result of self-reporting error and possible
genetic variations that are likely to exist across diverse Caucasians populations. Future
studies investigating the association between vitamin D genes and MS among
populations of European ancestry may to adjust for such ancestry-related variable. Poor
genotyping sample for the ITS population where 25% of the original sample had a
genotyping call rate of less than 90%, resulted in decreased statistical power to detect
any significant effect signal in this population. In addition, the directional inconsistencies
of odds ratios across the individual populations in the meta-analysis made it difficult to
determine an overall magnitude of the effects of the vitamin D SNPs on MS. However,
given the random nature of the directional inconsistency across all population, there

29

may truly be no association between most vitamin D SNPs with MS. Careful quality
control of all genetic data and the use of next generation high-resolution sequencing to
map the gene regions would help eliminate such limitations in the future.
As researchers increasingly identify novel genetic regions individually exhibiting
at least moderate risk on MS, it is clear that MS is a complex disease involving
interactive forces between many environmental and genetic factor. The results from
this thesis confirm that some vitamin D genes are associated with MS status. Vitamin D
genes are likely to influence the disease mechanism process, and in light of the recent
focus on the immune system in MS
47, 48
, the contribution of vitamin D and its associated
genes in MS is also likely to involve complex interactions with MS-associated immune-
related genes
49
. Future research involving larger high-resolution GWAS, transcriptomic
research together with the integration of pathway analysis incorporating relevant
biological knowledge will undoubtedly improve the understanding of the MS
mechanism. Hopefully these endeavors will elucidate the interactive dynamics between
vitamin D genes, other immune-related MS susceptibility genes and environmental
factors as they pertain to MS.

30

REFERENCES
13. Adams JS, Hewison M. Unexpected actions of vitamin D: new perspectives on the
regulation of innate and adaptive immunity. Nat Clin Pract Endocrinol Metab. 2008 Feb;4(2):80-
90.

3. Ascherio A, Munger K, Simon K. Vitamin D and multiple sclerosis. Lancet Neurology.
2010 JUN 2010:599-612.

20. Bahlo M, Booth D, Broadley S, et al. Genome-wide association study identifies new
multiple sclerosis susceptibility loci on chromosomes 12 and 20. Nature Genetics. 2009 JUL
2009:824-U84.

21. Baranzini S, Wang J, Gibson R, et al. Genome-wide association analysis of susceptibility
and clinical phenotype in multiple sclerosis. Human Molecular Genetics. 2009 FEB 15 2009:767-
78.

31. Barcellos LF, Oksenberg JR, Green AJ, et al. Genetic basis for clinical expression in
multiple sclerosis. Brain. 2002 Jan;125(Pt 1):150-8.

45. Berlanga-Taylor AJ, Disanto G, Ebers GC, Ramagopalan SV. Vitamin D-gene interactions
in multiple sclerosis. J Neurol Sci. 2011 Dec;311(1-2):32-6.

28. Bikle D. Nonclassic actions of vitamin D. J Clin Endocrinol Metab. 2009 Jan;94(1):26-34.

33. Briggs FB, Bartlett SE, Goldstein BA, et al. Evidence for CRHR1 in multiple sclerosis using
supervised machine learning and meta-analysis in 12,566 individuals. Hum Mol Genet. 2010
Nov;19(21):4286-95.

49. Correale J, Ysrraelit MC, Gaitán MI. Immunomodulatory effects of Vitamin D in multiple
sclerosis. Brain. 2009 May;132(Pt 5):1146-60.
34. De Jager PL, Jia X, Wang J, et al. Meta-analysis of genome scans and replication identify
CD6, IRF8 and TNFRSF1A as new multiple sclerosis susceptibility loci. Nat Genet. 2009
Jul;41(7):776-82.

48. Disanto G, Sandve GK, Berlanga-Taylor AJ, et al. Genomic regions associated with
multiple sclerosis are active in B cells. PLoS One. 2012;7(3):e32281.

9. Esparza M, Sasaki S, Kesteloot H. Nutrition, latitude, and multiple sclerosis mortality - An
ecologic study. American Journal of Epidemiology. 1995 OCT 1 1995:733-7.

38. Gandhi KS, McKay FC, Cox M, et al. The multiple sclerosis whole blood mRNA
transcriptome and genetic associations indicate dysregulation of specific T cell pathways in
pathogenesis. Hum Mol Genet. 2010 Jun;19(11):2134-43.

31

37. Gao X, Starmer J, Martin ER. A multiple testing correction method for genetic
association studies using correlated single nucleotide polymorphisms. Genet Epidemiol. 2008
May;32(4):361-9.

46. Ghazavi A, Kianbakht S, Ghasami K, Mosayebi G. High copper and low zinc serum levels
in Iranian patients with multiple sclerosis: a case control study. Clin Lab. 2012;58(1-2):161-4.

12. Gorman S, Kuritzky LA, Judge MA, et al. Topically applied 1,25-dihydroxyvitamin D3
enhances the suppressive activity of CD4+CD25+ cells in the draining lymph nodes. J Immunol.
2007 Nov;179(9):6273-83.

16. Hedegaard CJ, Krakauer M, Bendtzen K, Lund H, Sellebjerg F, Nielsen CH. T helper cell
type 1 (Th1), Th2 and Th17 responses to myelin basic protein and disease activity in multiple
sclerosis. Immunology. 2008 Oct;125(2):161-9.

11. Hewison M, Burke F, Evans KN, et al. Extra-renal 25-hydroxyvitamin D3-1alpha-
hydroxylase in human health and disease. J Steroid Biochem Mol Biol. 2007 Mar;103(3-5):316-
21.

14. Hewison M. Vitamin D and innate and adaptive immunity. Vitam Horm. 2011;86:23-62.

44. Huang J, Xie ZF. Polymorphisms in the vitamin D receptor gene and multiple sclerosis
risk: a meta-analysis of case-control studies. J Neurol Sci. 2012 Feb;313(1-2):79-85.

42. Hunt R, Sauna ZE, Ambudkar SV, Gottesman MM, Kimchi-Sarfaty C. Silent (synonymous)
SNPs: should we care about them? Methods Mol Biol. 2009;578:23-39.

32. (IMSGC) IMSGC. Comprehensive follow-up of the first genome-wide association study of
multiple sclerosis identifies KIF21B and TMEM39A as susceptibility loci. Hum Mol Genet. 2010
Mar;19(5):953-62.

6. Islam T, Gauderman WJ, Cozen W, Mack TM. Childhood sun exposure influences risk of
multiple sclerosis in monozygotic twins. Neurology. 2007 Jul;69(4):381-8.

26. Jones G, Strugnell SA, DeLuca HF. Current understanding of the molecular actions of
vitamin D. Physiol Rev. 1998 Oct;78(4):1193-231.

8. Kampman MT, Brustad M. Vitamin D: a candidate for the environmental effect in
multiple sclerosis - observations from Norway. Neuroepidemiology. 2008;30(3):140-6.

4. Kurtzke J. Multiple sclerosis in time and space - geographic clues to cause. Journal of
Neurovirology. 2000 MAY 2000:S134-S40.

32

2. Lincoln J, Cook S. An overview of gene-epigenetic-environmental contributions to MS
causation. Journal of the Neurological Sciences. 2009 NOV 15 2009:54-7.

15. Lutton JD, Winston R, Rodman TC. Multiple sclerosis: etiological mechanisms and future
directions. Exp Biol Med (Maywood). 2004 Jan;229(1):12-20.

35. MACH algorithm. Available from:
http://www.sph.umich.edu/csg/abecasis/MACH/download/.

30. Mack TM, Deapen D, Hamilton AS. Representativeness of a roster of volunteer North
American twins with chronic disease. Twin Res. 2000 Mar;3(1):33-42.

10. Munger KL, Levin LI, Hollis BW, Howard NS, Ascherio A. Serum 25-hydroxyvitamin D
levels and risk of multiple sclerosis. JAMA. 2006 Dec;296(23):2832-8.

47. Oksenberg JR, Hauser SL. Decoding multiple sclerosis. Ann Neurol. 2011 Dec;70(6):A5-7.

19. Orton S, Ramagopalan S, Para A, et al. Vitamin D metabolic pathway genes and risk of
multiple sclerosis in Canadians. Journal of the Neurological Sciences. 2011.

24. Partridge JM, Weatherby SJ, Woolmore JA, et al. Susceptibility and outcome in MS:
associations with polymorphisms in pigmentation-related genes. Neurology. 2004
Jun;62(12):2323-5.

41. Patsopoulos NA, Esposito F, Reischl J, et al. Genome-wide meta-analysis identifies novel
multiple sclerosis susceptibility loci. Ann Neurol. 2011 Dec;70(6):897-912.

1. Pender MP, Greer JM. Immunology of multiple sclerosis. Curr Allergy Asthma Rep. 2007
Jul;7(4):285-92.

29. Ramagopalan S, Maugeri N, Handunnetthi L, et al. Expression of the Multiple Sclerosis-
Associated MHC Class II Allele HLA-DRB1*1501 Is Regulated by Vitamin D. Plos Genetics. 2009
FEB 2009:-.

40. Ramagopalan SV, Dyment DA, Cader MZ, et al. Rare variants in the CYP27B1 gene are
associated with multiple sclerosis. Ann Neurol. 2011 Dec;70(6):881-6.

22. Sawcer S, Hellenthal G, Pirinen M, et al. Genetic risk and a primary role for cell-mediated
immune mechanisms in multiple sclerosis. Nature. 2011 Aug;476(7359):214-9.

18. Simon K, Munger K, Yang X, Ascherio A. Polymorphisms in vitamin D metabolism related
genes and risk of multiple sclerosis. Multiple Sclerosis. 2010 FEB 2010:133-8.

33

25. Smolders J, Peelen E, Thewissen M, et al. The relevance of vitamin D receptor gene
polymorphisms for vitamin D research in multiple sclerosis. Autoimmunity Reviews. 2009 JUN
2009:621-6.

43. Smolders J, Damoiseaux J, Menheere P, Tervaert J, Hupperts R. Association Study on
Two Vitamin D Receptor Gene Polymorphisms and Vitamin D Metabolites in Multiple Sclerosis.
Contemporary Challenges in Autoimmunity. 2009 2009:515-20.

7. Staples J, Ponsonby A, Lim L. Low maternal exposure to ultraviolet radiation in
pregnancy, month of birth, and risk of multiple sclerosis in offspring: longitudinal analysis.
British Medical Journal. 2010 APR 29 2010:-.

17. Steckley J, Dyment D, Sadovnick A, et al. Genetic analysis of vitamin D related genes in
Canadian multiple sclerosis patients. Neurology. 2000 FEB 8 2000:729-32.

39. Sundqvist E, Baarnhielm M, Alfredsson L, Hillert J, Olsson T, Kockum I. Confirmation of
association between multiple sclerosis and CYP27B1. European Journal of Human Genetics. 2010
DEC 2010:1349-52.

36. Torgerson DG, Ampleford EJ, Chiu GY, et al. Meta-analysis of genome-wide association
studies of asthma in ethnically diverse North American populations. Nat Genet. 2011 Jul.

27. van Etten E, Stoffels K, Gysemans C, Mathieu C, Overbergh L. Regulation of vitamin D
homeostasis: implications for the immune system. Nutr Rev. 2008 Oct;66(10 Suppl 2):S125-34.

5. Willer C, Dyment D, Sadovnick A, et al. Timing of birth and risk of multiple sclerosis:
population based study. British Medical Journal. 2005 JAN 15 2005:120-3.

23. Yeo T, Maranian M, Singlehurst S, Gray J, Compston A, Sawcer S. Four single nucleotide
polymorphisms from the vitamin D receptor gene in UK multiple sclerosis. Journal of Neurology.
2004 JUN 2004:753-4.

Abstract (if available)

Abstract Vitamin D is an environmental factor known to protect against multiple sclerosis (MS). Furthermore, it has also been shown to influence activities of immunological genes such as the HLA-DRB1*1501, which have found to be associated with this autoimmune disease. Genes controlling the metabolism and signaling functions of vitamin D are also likely to be MS susceptibility regions. ❧ In this thesis, we examined the association between nine genes of the vitamin D pathway and MS. The functional relevance of significantly identified single nucleotide polymorphisms (SNPs) as they contribute to MS was also explored. After conducting a meta-analysis, we found rs928329 from CYP27B1 to be significantly associated with MS (meta p < 0.0002) while all other vitamin D genes did not yield significant associations. ❧ Significant interactions between RXRB and HLA-DRB1*1501 identified in this thesis indicate the importance of complex gene-gene interaction in MS. Given the importance of vitamin D as an immune-regulator and the important role of immunity in MS, future investigation of interaction between vitamin D and other immune-related genes will likely reveal important biological mechanisms that cause MS.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

A genome wide association study of multiple sclerosis (MS) in Hispanics

PDF

Environmental risk factors of Multiple Sclerosis: a twin study

PDF

The association between sun exposure and multiple sclerosis

PDF

Hispanic ethnicity in multiple sclerosis

PDF

Association of subclinical atherosclerosis with plasma B-vitamin, cysteine, homocysteine, and cysteinyl glycine in a cardiovascular disease-free population

PDF

Carcinogen metabolism genes, meat intake, and colorectal cancer risk

PDF

Targeting Ryk-signaling to attenuate multiple sclerosis by promoting oligodendrocyte differentiation and myelination

PDF

Variation in CRY2 and MTNR1B have independent effects on insulin secretion in Mexican Americans

PDF

Two-step testing approaches for detecting quantitative trait gene-environment interactions in a genome-wide association study

PDF

Combination of quantile integral linear model with two-step method to improve the power of genome-wide interaction scans

PDF

The effect of vitamin D supplementation on the progression of carotid intima-media thickness and arterial stiffness in elderly African American women: Results of a randomized placebo-controlled trial

PDF

The role of genetic ancestry in estimation of the risk of age-related degeneration (AMD) in the Los Angeles Latino population

PDF

The interplay between tobacco exposure and polygenic risk score for growth on birthweight and childhood acute lymphoblastic leukemia

PDF

The association of asthma and asthma-related medications on subclinical atherosclerosis: a cross-sectional analysis of four randomized clinical trials

PDF

Targeted metabolic signatures and diet in associations with obesity and insulin resistance in young adults

PDF

Association between common taxa in gut microbiota and obesity as well as the influence of diet on these taxa

PDF

Evaluating the effects of testing framework and annotation updates on gene ontology enrichment analysis

PDF

Examining the relationship between bioavailable vitamin D and IVF outcomes

PDF

Polygenic analyses of complex traits in complex populations

PDF

The effects of heat and air pollution on mental-health related mortality