Barcellos, Lisa F.

Background: Rare genetic variants are emerging as important contributors to the heritability of multiple sclerosis (MS). Whether rare variants also contribute to pediatric-onset multiple sclerosis (POMS) is unknown. Objective: To test whether genes harboring rare variants associated with adult-onset MS risk ( PRF1, PRKRA, NLRP8 , and HDAC7 ) and 52 major histocompatibility complex (MHC) genes are associated with POMS. Methods: We analyzed DNA samples from 330 POMS cases and 306 controls from the US Network of Pediatric MS Centers and Kaiser Permanente Northern California for which Illumina ExomeChip genotypes were available. Using the gene-based method “SKAT-O,” we tested the association between candidate genes and POMS risk. Results: After correction for multiple comparisons, one adult-onset MS gene ( PRF1, p = 2.70 × 10 −3 ) and two MHC genes ( BRD2, p = 5.89 × 10 −5 and AGER, p = 7.96 × 10 −5 ) were significantly associated with POMS. Results suggest these are independent of HLA-DRB1*1501. Conclusion: Findings support a role for rare coding variants in POMS susceptibility. In particular, rare minor alleles within PRF1 were more common among individuals with POMS compared to controls while the opposite was true for rare variants within significant MHC genes, BRD2 and AGER . These genes would not have been identified by common variant studies, emphasizing the merits of investigating rare genetic variation in complex diseases.

Abstract Objective To identify features of the gut microbiome associated with multiple sclerosis activity over time. Methods We used 16S ribosomal RNA sequencing from stool of 55 recently diagnosed pediatric‐onset multiple sclerosis patients. Microbiome features included the abundance of individual microbes and networks identified from weighted genetic correlation network analyses. Prentice‐Williams‐Peterson Cox proportional hazards models estimated the associations between features and three disease activity outcomes: clinical relapses and both new/enlarging T2 lesions and new gadolinium‐enhancing lesions on brain MRI. Analyses were adjusted for age, sex, and disease‐modifying therapies. Results Participants were followed, on average, 2.1 years. Five microbes were nominally associated with all three disease activity outcomes after multiple testing correction. These included butyrate producers Odoribacter (relapse hazard ratio = 0.46, 95% confidence interval: 0.24, 0.88) and Butyricicoccus (relapse hazard ratio = 0.49, 95% confidence interval: 0.28, 0.88). Two networks of co‐occurring gut microbes were significantly associated with a higher hazard of both MRI outcomes (gadolinium‐enhancing lesion hazard ratios (95% confidence intervals) for Modules 32 and 33 were 1.29 (1.08, 1.54) and 1.42 (1.18, 1.71), respectively; T2 lesion hazard ratios (95% confidence intervals) for Modules 32 and 33 were 1.34 (1.15, 1.56) and 1.41 (1.21, 1.64), respectively). Metagenomic predictions of these networks demonstrated enrichment for amino acid biosynthesis pathways. Interpretation Both individual and networks of gut microbes were associated with longitudinal multiple sclerosis activity. Known functions and metagenomic predictions of these microbes suggest the important role of butyrate and amino acid biosynthesis pathways. This provides strong support for future development of personalized microbiome interventions to modify multiple sclerosis disease activity.

Abstract Abstract 4048 Poster Board III-983 Introduction Iron deficiency is the most common nutritional disorder in the world with an estimated two billion affected persons. Although commonly considered environmental in origin, the existence of multiple genetic disorders of iron metabolism in man, rodents and other vertebrates suggest a genetic contribution to iron deficiency. Methods: The Hemochromatosis and Iron Overload Screening (HEIRS) Study is a multi-center, multi-ethnic study in which transferrin saturation (TS), serum ferritin (SF), and HFE mutations were determined in 101,168 adults. To identify genomic locations associated with iron deficiency, we performed a genome-wide association study (GWAS) using DNA collected from white HEIRS Study participants who had SF ≤ 12 μg/L (cases) and an equal number of white controls (SF > 100 μg/L in men, SF > 50 μg/L in women) frequency-matched to cases by sex and geographic location. Men aged ≥ 25 y and women ≥ 50 y were included in both groups. Tissue body iron, an index of iron deficiency, was estimated from serum transferrin receptor (sTfR) and SF. Genotyping was performed with the Illumina HumanCNV370K Beadchip platform. Quality control filters excluded single nucleotide polymorphisms (SNPs) or samples with > 5% missing genotypes, SNPs showing heterozygosity or Hardy-Weinberg deviations (P<10−7), and SNPs with minor allele frequency < 0.02. Population admixture/structure was assessed using principal component analysis. Regression analysis was used to examine the association between outcomes (case-control status, tissue body iron, serum ferritin, transferrin receptor, serum iron, total iron-binding capacity [TIBC], and unsaturated iron-binding capacity [UIBC]) and each SNP genotype variable; covariates included age, sex, and geographic location. Replication for 56 SNPs was conducted in a population attending primary care clinics at a Veterans Affairs (VA) medical center using the iPlex platform. Eligibility within the VA replication population was restricted to age and self-reported white ethnicity as for the HEIRS subset from a total of 2559 people (138 women). VA participants with SF ≤ 20 μg/L were classified as iron-deficient cases and frequency matched 1:2 with controls (men with SF > 100 μg/L and women with SF > 50 μg/L). Results The GWAS genomic control parameter was not significantly different from 1.0. There were 392 cases (96 men) and 390 controls (96 men) in the HEIRS subset GWAS with average age (SD) of 59 (10) y and 61 (11) y, respectively. Geometric mean SF (minimum, maximum), and mean (SD) for sTfR and tissue body iron in the HEIRS subset were 7.5 (1.2, 12) μg/L, 6.4 (3.77) mg/kg and -2.0 (2.50) for cases and 141 (51, 881) μg/L, 3.0 (0.98) mg/kg and 10.8 (2.5) for controls. After quality control tests, GWAS analysis included genotype data for 331,060 SNPs in 734 individuals (364 cases, 370 controls). For the VA replication population there were 67 male and 11 female cases, and 136 male and 27 female controls for whom DNA was successfully prepared; the average age (SD) was 68 (12) y for cases and 65 (11) y for controls. Regression analysis identified seven SNPs within four independent regions that replicated associations found in the GWAS (GWAS P<1×10-4 and VA P<0.05).The SNP rs6735681 on chromosome 2p24 was associated with serum iron (GWAS P<3.9×10-5, VA P=0.038). Three SNPs on chromosome 2p14 (rs6750096, rs2698541 and rs2698530) significantly influenced both TIBC and UIBC (GWAS P<2.9×10-5, VA P< 0.04 for all). Two SNPs in the TF gene region on chromosome 3q22 also showed significant effects on TIBC and UIBC (GWAS P<4.7×10-6, VA P<0.03 for all). The SNP rs10512064 on chromosome 9q21 was associated with serum ferritin concentration and tissue body iron (GWAS P<2.5×10-5, VA P<0.05 for both). Conclusion: From these GWAS and replication studies, we have identified three new genetic loci and one known iron gene, TF, associated with iron phenotype variability. These results point to specific loci as targets for gene identification and TF polymorphisms as determinants of iron metabolism, which in turn may play a role in regulation of body iron status. Disclosures: No relevant conflicts of interest to declare.

Abstract Abstract 2105 The existence of multiple inherited disorders of iron metabolism in man, rodents and other vertebrates suggests genetic contributions to iron deficiency. We hypothesized that common variants in genes involved in iron metabolism may modulate susceptibility or resistance to the development of iron deficiency in humans. To examine the association between single nucleotide polymorphisms (SNPs) in key genes involved in iron metabolism pathways, we previously performed a genome-wide association study using DNA collected from white men aged ≥25 y and women ≥50 y in the Hemochromatosis and Iron Overload Screening (HEIRS) Study with serum ferritin (SF) ≤12 μg/L (cases) and controls (SF >100 μg/L in men, SF >50 μg/L in women). We now report on a multiethnic follow-up association study of HEIRS participants. Candidate SNPs were identified from our GWAS and the scientific literature. Population samples of whites, African Americans, Hispanics, and Asians from the U.S. and Canada were analyzed separately for association between SNPs and case-control status and each of seven quantitative outcomes including serum iron, total iron-binding capacity (TIBC), unsaturated iron-binding capacity (UIBC), transferrin saturation, SF, serum transferrin receptor, and body iron. There were 1084 white (357 cases, 727 controls), 153 Asian (51 cases, 102 controls), 221 African American (77 cases, 144 controls) and 233 of 239 Hispanic individuals (79 cases, 160 controls) that passed quality control. For the African-American and Hispanic samples, ancestry proportions were estimated based on genotypes of ancestry informative markers. Regression analysis was used to examine the association between case-control status and quantitative serum iron measures and 1134, 1115, 1113 and 1134 SNP genotypes in the white, African-American, Hispanic, and Asian population samples, respectively. Model predictors included age, sex, the estimated ancestry proportion (for African American and Hispanic only), genotype, and measured covariates that showed nominally significant associations with the outcome. Three chromosomal regions showed evidence of association across multiple populations, including SNPs in the TF gene on chromosome 3q22, the TMPRSS6 gene on chromosome 22q12, and loci on chromosome 18q21. SNP rs1421312 in TMPRSS6 was associated with serum iron in whites (p=4.7×10−7) and was replicated in African Americans (p=0.0012).Twenty SNPs in the TF gene region were significantly associated with TIBC in the white sample (p<4.4×10−5); six SNPs were replicated in other ethnicities (p< 0.01). SNP rs10904850 in the CUBN gene on 10p13 was significantly associated with serum iron in the African-American sample (P=1.0×10−5). Mutations in the TMPRSS6 gene have been implicated in iron-refractory iron deficiency anemia through linkage studies. We found a novel SNP in TMPRSS6 that was associated with serum iron in whites and replicated in African Americans, suggesting a role for this SNP in increasing the risk of iron deficiency in affected persons. Our results confirm known associations with iron measures and give evidence of their role in different ethnic groups, a unique aspect of this study, suggesting origins in a common founder. Disclosures: No relevant conflicts of interest to declare.

Abstract It is well-recognized that Helicobacter pylori infection causes gastrointestinal blood loss and iron deficiency. H. pylori gastritis has also been implicated in iron deficiency refractory to oral iron replacement therapy. Although H. pylori infection is common in multiple populations, less is known about the extent to which H. pylori contributes to iron deficiency in different racial/ethnic groups. To address this question, we tested the serum collected from men aged ≥ 25 y and women ≥ 50 y in the Hemochromatosis and Iron Overload Screening (HEIRS) Study to identify those with iron deficiency (defined as serum ferritin ≤ 12 mg/L [cases]) and iron-replete controls (serum ferritin > 100 mg/L in men, serum ferritin > 50 mg/L in women). All samples were tested for H. pylori IgG antibodies and for antibodies to CagA (cytotoxin-associated gene-bearing strain). We tested 571 cases and 1142 controls. Participants were considered to have evidence of H. pylori infection if they had positive results for both antibodies or for either antibody alone. Participants who did not have a result for either test were excluded. We examined the association between the presence of H. pylori in cases and controls among Caucasians, Asian/Pacific Islanders, African Americans, and Hispanics. Among Caucasians, evidence of H. pylori infection was comparable in cases and controls; 170 of 735 controls (23.1%) and 97 of 363 cases (26.7%), respectively, were positive. Similar results were found in African Americans (56.3% of 144 controls; 55.8% of 77 cases) and Hispanics (66.7% of 159 controls; 59.5% of 79 cases). However, in Asian/Pacific Islanders, the prevalence of H. pylori was higher among cases (62.8% of 51) than controls (44.1% of 102), p=0.029 by the likelihood ratio Chi-Square test. The relative risk (RR) of iron deficiency among those with H. pylori-positivity was significantly greater than in those without (RR, 1.66; 95% CI [1.04, 2.66]). A comparison of subgroups by positivity for IgG or CagA showed that the relative risk of iron deficiency was increased among those with IgG positivity, with or without Cag A positivity (RR, 1.80 and 1.90, respectively), but not among those with CagA positivity alone (RR, 1.00). H. pylori infection is a frequent contributor to iron deficiency but is also common in the general population. Our results indicate that among HEIRS Study Asian/Pacific Islander participants, H. pylori infection is more prevalent in cases with iron deficiency than in controls. In contrast, there was no difference in the prevalence of H. pylori infection between cases and controls among Caucasians, African Americans, and Hispanics. These results have implications for the investigation of iron deficiency in Asians and Pacific Islanders. Disclosures No relevant conflicts of interest to declare.

Abstract Little is known about the interaction between constitutive and somatic genetic variation in carcinogenesis. Genome-wide association studies (GWAS) of childhood acute lymphoblastic leukemia (ALL) have identified 6 genetic associations, including the single nucleotide polymorphism (SNP) rs3731217, located between the CDKN2A and CDKN2B genes. Somatic deletions of this chromosome 9p21.3 region are common in cancer, and occur in ~30% of childhood ALL. We replicated the rs3731217 association with childhood ALL in Hispanics (P = 0.021) via a GWAS of 323 childhood ALL cases and 454 controls, using the Illumina OmniExpress SNP array. SNP imputation analysis across chromosome 9p21.3 revealed a stronger association at SNP rs3731249 (P = 6.4x10-4; OR = 2.77), a low-frequency missense variant in exon 2 (p.Ala148Thr) of CDKN2A (MAFcontrols = 2.1%). These associations were replicated in an additional 378 ALL cases and 536 controls using Taqman assays, where the missense variant again had a greater magnitude of effect (Prs3731249 = 6.5x10-3; OR rs3731249 = 2.00 versus Prs3731217 = 0.029; OR rs3731217= 1.38). We assessed the effect of rs3731249 on risk of other cancers with known chr9p21.3 GWAS associations, but observed no association between this variant and either adult glioblastoma or melanoma risk, suggesting an ALL-specific role for rs3731249. We hypothesized that cases heterozygous for rs3731249 may show allelic imbalance, with preferential loss of the wildtype (WT) allele when hemizygous chr9p21.3 deletions occur in leukemic cells. To test this, we developed a novel method termed “SMART-ddPCR” (Somatic Mutation Allelic Ratio Test using Droplet Digital PCR). Concentration of risk and WT alleles was measured by ddPCR using the rs3731249 Taqman assay in diagnostic bone marrow (i.e. tumor) DNA from 35 known heterozygotes. Allelic imbalance due to somatic CDKN2A alterations was identified in 17 cases, of which 14 had a higher risk:WT allele ratio (P = 0.006, 1-sided binomial significance test), thus demonstrating preferential loss of WT rs3731249 and retention of the missense allele. Subclonal heterogeneity in CDKN2A copy number was also evident, suggesting that CDKN2A alterations are secondary events in leukemogenesis. Preferential allelic imbalance towards the rs3731249 risk allele in ALL tumor DNA provides strong evidence of an important role for this missense variant in leukemogenesis. We show, for the first time, a direct relationship between constitutive and somatic genetic variants underlying ALL development, suggesting that such constitutive variants can provide an additional substrate for selection during tumor evolution. Disclosures No relevant conflicts of interest to declare.

Abstract Abstract 2095 Celiac disease (CD) is an increasingly recognized disorder in Caucasian populations of European origin. Little is known about its prevalence in non-Caucasians. While CD is thought to be a cause of iron deficiency anemia, the extent to which CD contributes to iron deficiency in Caucasians and especially non-Caucasians is unknown. Iron deficiency is one of the most common nutritional deficiencies in both the developed world and in developing countries. Thus, it is important to know the prevalence of CD in different populations and in iron-deficient and iron-replete persons. To answer these questions, we tested serum collected from Caucasian and non-Caucasian men aged ≥ 25 y and women ≥ 50 y in the Hemochromatosis and Iron Overload Screening (HEIRS) Study. In the HEIRS Study, 101,168 participants were screened with serum biochemical tests of iron status, which not only detected participants with iron overload but also identified a unique multiethnic population of participants with iron deficiency. We hypothesized that CD is more common in those with iron deficiency than in iron-replete individuals. We also examined the difference in frequency of CD between Caucasians and non-Caucasians with iron deficiency. Cases with iron deficiency (defined as serum ferritin ≤ 12 μg/L [cases]) and iron-replete controls (serum ferritin > 100 μg/L in men, serum ferritin > 50 μg/L in women) were further evaluated by using a sequential serological testing scheme whereby all samples were tested with an ELISA for human recombinant tissue transglutaminase IgA antibodies, and positive values were confirmed by endomysial antibodies. Those positive for both tests were considered double positives and were presumed to have untreated CD. Those positive for tissue transglutaminase IgA antibodies but not for endomysial antibodies were termed single positives and were considered indeterminate and excluded from analysis. We further hypothesized that low serum ferritin and an appropriate human leukocyte antigen (HLA) variant, may be predictive of CD. We used SNP-based analysis to determine HLA genotype and DQ8, DQ2.2 or DQ4, and DQ2.5 risk variants. Fisher's exact test was used to examine the association between iron−deficient case and control status and the presence of celiac disease. Log-binomial regression was applied to estimate the odds of celiac disease in iron-deficient cases relative to that of controls. A total of 1713 subjects were tested for tissue transglutaminase IgA antibodies. Participants included 1100 Caucasians, 221 African Americans, 153 Asians, and 239 Hispanics. Fourteen of 571 iron-deficient cases were positive for the double serology compared to just one of 1142 controls. All of the seropositive tests were seen in Caucasians and none in the non-Caucasians. Excluding data from participants with indeterminate serological screen results (6 Caucasians, 4 non-Caucasians), we found that CD occurred in 14 of 567 cases (2.5%) and in only 1 of 1136 (0.1%) controls (Fisher's exact test, p=1.92 10−6). CD was more common in Caucasian cases (14 of 363, 4%) than in non-Caucasian cases (0 of 204, p=0.003). Only one Caucasian control and none of the non-Caucasian controls had CD. The odds of CD in individuals with iron deficiency was 28 (3.7, 212.8) times that in controls. Thirteen of 14 iron-deficient cases with CD carried the DQ2.5 variant of the HLA genotype. Presence of DQ2.5 was highly significantly associated with celiac disease status in iron-deficient cases compared to controls (Fisher's exact test, p = 7.183 × 10−8). The results indicate that CD is a small but significant contributor to iron deficiency in Caucasians. CD is very rare in other races, even among individuals with features frequently seen in CD, such as iron deficiency. CD is also rare in Caucasians who are iron replete. CD testing should be considered for adult Caucasian males or post-menopausal females having iron deficiency without evidence of other etiologies such as gastrointestinal blood loss. Patients identified with CD may benefit from appropriate treatment. Disclosures: No relevant conflicts of interest to declare.

AbstractObjectiveOnset of multiple sclerosis (MS) occurs in childhood for approximately 5% of cases (pediatricMS, or ped‐MS). Epigenetic influences are strongly implicated inMSpathogenesis in adults, including the contribution from microRNAs (miRNAs), small noncodingRNAs that affect gene expression by binding target genemRNAs. Few studies have specifically examined miRNAs in ped‐MS, but individuals developingMSat an early age may carry a relatively high burden of genetic risk factors, and miRNAdysregulation may therefore play a larger role in the development of ped‐MSthan in adult‐onsetMS. This study aimed to look for evidence of miRNAinvolvement in ped‐MSpathogenesis.MethodsGWASresults from 486 ped‐MScases and 1362 controls from the U.S. PediatricMSNetwork and Kaiser Permanente Northern California membership were investigated for miRNA‐specific signals. First, enrichment of miRNA‐target gene network signals was evaluated usingMIGWASsoftware. Second,SNPs in miRNAgenes and in target gene binding sites (miR−SNPs) were tested for association with ped‐MS, and pathway analysis was performed on associated target genes.ResultsMIGWASanalysis showed that miRNA‐target gene signals were enriched inGWAS(P = 0.038) and identified 39 candidate biomarker miRNA‐target gene pairs, including immune and neuronal signaling genes. The miR‐SNPanalysis implicated dysregulation of miRNAbinding to target genes in five pathways, mainly involved in immune signaling.InterpretationEvidence fromGWASsuggests that miRNAs play a role in ped‐MSpathogenesis by affecting immune signaling and other pathways. Candidate biomarker miRNA‐target gene pairs should be further studied for diagnostic, prognostic, and/or therapeutic utility.

Objective: The aim of this study was to determine whether a vitamin D genetic risk score (vitDGRS) is associated with 25-hydroxyvitamin D (25(OH)D) level and multiple sclerosis (MS) relapses in children. Methods: DNA samples were typed for single nucleotide polymorphisms (SNPs) from four genes previously identified to be associated with 25(OH)D levels. SNPs with strong associations with 25(OH)D after multiple comparison correction were used to create a genetic risk score (vitDGRS). Cox regression models tested associations of vitDGRS with relapse hazard. Results: Two independent SNPs within or near GC and NADSYN1/DHCR7 genes were strongly associated with 25(OH)D levels in the discovery cohort ( n = 182) after Bonferroni correction. The vitDGRS of these SNPs explained 4.5% of the variance of 25(OH)D level after adjustment for genetic ancestry. Having the highest versus lowest vitDGRS was associated with 11 ng/mL lower 25(OH)D level (95% confidence interval (CI) = −17.5, −4.5, p = 0.001) in the discovery cohort. Adjusting for ancestry, sex, disease-modifying therapy (DMT), and HLA-DRB1*15 carrier status, the highest versus lowest vitDGRS was associated with 2.6-fold (95% CI = 1.37, 5.03, p = 0.004) and 2.0-fold (95% CI = 0.75, 5.20, p = 0.16) higher relapse hazard in the discovery and replication cohorts, respectively. Conclusion: The vitDGRS identifies children at greater risk of relapse. These findings support a causal role for vitamin D in MS course.

AbstractBackgroundThere is limited information about the potential associations of multiple sclerosis (MS) and commonly used household chemicals.MethodsWe performed a case‐control study of exposures to common household chemicals during childhood in children with MS and healthy pediatric controls. Exposures to household products were collected from a comprehensive questionnaire (http://www.usnpmsc.org/Documents/EnvironmentalAssessment.pdf) completed by parents at the time of enrollment in the study. Cases included children diagnosed with MS or clinically isolated syndrome with at least two silent T2 bright lesions on MRI, recruited within 4 years of disease onset from 16 pediatric MS clinics in the USA. Multivariate analyses using logistic regression were adjusted for possible confounders including age, sex, race, ethnicity, mother's highest level of education, and urban versus rural living.ResultsQuestionnaire responses to household chemicals were available for 312 eligible cases (median age 15.7 years, 63% girls) and 490 healthy controls (median age 15.0, 57% girls). Exposure to rodenticides (odds ratio [OR] 2.10, 95% confidence interval [CI] 1.35–3.26, P ≤ 0.001), weed control agents (OR 1.99, 95% CI 1.36–2.92, P ≤ 0.001) and products for plant/tree disease control (OR 2.72, 95% CI 1.54–4.82, P ≤ 0.001) anytime during childhood were associated with an increased risk for pediatric‐onset MS in adjusted and multiple comparisons analyses.ConclusionsOur findings suggest that exposure to specific household chemicals during early childhood is associated with the risk of developing pediatric‐onset MS. Future studies are needed to elucidate a causal relationship and the exact agents involved.