Amin Ziaei1, Amy M Lavery2, Xiaorong Ma Shao3, Cameron Adams3, T Charles Casper4, John Rose4, Meghan Candee4, Bianca Weinstock-Guttman5, Greg Aaen6, Yolanda Harris7, Jennifer Graves8, Leslie Benson9, Mark Gorman9, Mary Rensel10, Soe Mar11, Tim Lotze12, Benjamin Greenberg13, Tanuja Chitnis14, Janace Hart1, Amy T Waldman2, Lisa F Barcellos3, Emmanuelle Waubant1. 1. University of California, San Francisco, San Francisco, CA, USA. 2. Division of Child Neurology, Children's Hospital of Philadelphia, Philadelphia, PA, USA. 3. Genetic Epidemiology and Genomics Laboratory, Divisions of Epidemiology and Biostatistics, School of Public Health, University of California, Berkeley, Berkeley, CA, USA. 4. The University of Utah, Salt Lake City, UT, USA. 5. Buffalo General Hospital, State University of New York at Buffalo, Buffalo, NY, USA. 6. Loma Linda University Children's Hospital, Loma Linda, CA, USA. 7. The University of Alabama, Tuscaloosa, AL, USA. 8. University of California, San Diego, San Diego, CA, USA. 9. Pediatric Multiple Sclerosis and Related Disorders Program, Boston Children's Hospital, Boston, MA, USA. 10. Cleveland Clinic, Cleveland, OH, USA. 11. Washington University in St. Louis, St. Louis, MO, USA. 12. Texas Children's Hospital, Houston, TX, USA. 13. University of Texas Southwestern Medical Center, Dallas, TX, USA. 14. Brigham and Women's Hospital and Harvard Medical School, Boston, MA, USA.
Abstract
BACKGROUND: We previously reported a relationship between air pollutants and increased risk of pediatric-onset multiple sclerosis (POMS). Ozone is an air pollutant that may play a role in multiple sclerosis (MS) pathoetiology. CD86 is the only non-HLA gene associated with POMS for which expression on antigen-presenting cells (APCs) is changed in response to ozone exposure. OBJECTIVES: To examine the association between county-level ozone and POMS, and the interactions between ozone pollution, CD86, and HLA-DRB1*15, the strongest genetic variant associated with POMS. METHODS: Cases and controls were enrolled in the Environmental and Genetic Risk Factors for Pediatric MS study of the US Network of Pediatric MS Centers. County-level-modeled ozone data were acquired from the CDC's Environmental Tracking Network. Participants were assigned ozone values based on county of residence. Values were categorized into tertiles based on healthy controls. The association between ozone tertiles and having MS was assessed by logistic regression. Interactions between tertiles of ozone level and the GG genotype of the rs928264 (G/A) single nucleotide polymorphism (SNP) within CD86, and the presence of DRB1*15:01 (DRB1*15) on odds of POMS were evaluated. Models were adjusted for age, sex, genetic ancestry, and mother's education. Additive interaction was estimated using relative excess risk due to interaction (RERI) and attributable proportions (APs) of disease were calculated. RESULTS: A total of 334 POMS cases and 565 controls contributed to the analyses. County-level ozone was associated with increased odds of POMS (odds ratio 2.47, 95% confidence interval (CI): 1.69-3.59 and 1.95, 95% CI: 1.32-2.88 for the upper two tertiles, respectively, compared with the lowest tertile). There was a significant additive interaction between high ozone tertiles and presence of DRB1*15, with a RERI of 2.21 (95% CI: 0.83-3.59) and an AP of 0.56 (95% CI: 0.33-0.79). Additive interaction between high ozone tertiles and the CD86 GG genotype was present, with a RERI of 1.60 (95% CI: 0.14-3.06) and an AP of 0.37 (95% CI: 0.001-0.75) compared to the lowest ozone tertile. AP results indicated that approximately half of the POMS risk in subjects can be attributed to the possible interaction between higher county-level ozone carrying either DRB1*15 or the CD86 GG genotype. CONCLUSIONS: In addition to the association between high county-level ozone and POMS, we report evidence for additive interactions between higher county-level ozone and DRB1*15 and the CD86 GG genotype. Identifying gene-environment interactions may provide mechanistic insight of biological processes at play in MS susceptibility. Our work suggests a possible role of APCs for county-level ozone-induced POMS risk.
BACKGROUND: We previously reported a relationship between air pollutants and increased risk of pediatric-onset multiple sclerosis (POMS). Ozone is an air pollutant that may play a role in multiple sclerosis (MS) pathoetiology. CD86 is the only non-HLA gene associated with POMS for which expression on antigen-presenting cells (APCs) is changed in response to ozone exposure. OBJECTIVES: To examine the association between county-level ozone and POMS, and the interactions between ozone pollution, CD86, and HLA-DRB1*15, the strongest genetic variant associated with POMS. METHODS: Cases and controls were enrolled in the Environmental and Genetic Risk Factors for Pediatric MS study of the US Network of Pediatric MS Centers. County-level-modeled ozone data were acquired from the CDC's Environmental Tracking Network. Participants were assigned ozone values based on county of residence. Values were categorized into tertiles based on healthy controls. The association between ozone tertiles and having MS was assessed by logistic regression. Interactions between tertiles of ozone level and the GG genotype of the rs928264 (G/A) single nucleotide polymorphism (SNP) within CD86, and the presence of DRB1*15:01 (DRB1*15) on odds of POMS were evaluated. Models were adjusted for age, sex, genetic ancestry, and mother's education. Additive interaction was estimated using relative excess risk due to interaction (RERI) and attributable proportions (APs) of disease were calculated. RESULTS: A total of 334 POMS cases and 565 controls contributed to the analyses. County-level ozone was associated with increased odds of POMS (odds ratio 2.47, 95% confidence interval (CI): 1.69-3.59 and 1.95, 95% CI: 1.32-2.88 for the upper two tertiles, respectively, compared with the lowest tertile). There was a significant additive interaction between high ozone tertiles and presence of DRB1*15, with a RERI of 2.21 (95% CI: 0.83-3.59) and an AP of 0.56 (95% CI: 0.33-0.79). Additive interaction between high ozone tertiles and the CD86 GG genotype was present, with a RERI of 1.60 (95% CI: 0.14-3.06) and an AP of 0.37 (95% CI: 0.001-0.75) compared to the lowest ozone tertile. AP results indicated that approximately half of the POMS risk in subjects can be attributed to the possible interaction between higher county-level ozone carrying either DRB1*15 or the CD86 GG genotype. CONCLUSIONS: In addition to the association between high county-level ozone and POMS, we report evidence for additive interactions between higher county-level ozone and DRB1*15 and the CD86 GG genotype. Identifying gene-environment interactions may provide mechanistic insight of biological processes at play in MS susceptibility. Our work suggests a possible role of APCs for county-level ozone-induced POMS risk.
Authors: Christen L Mumaw; Shannon Levesque; Constance McGraw; Sarah Robertson; Selita Lucas; Jillian E Stafflinger; Matthew J Campen; Pamela Hall; Jeffrey P Norenberg; Tamara Anderson; Amie K Lund; Jacob D McDonald; Andrew K Ottens; Michelle L Block Journal: FASEB J Date: 2016-02-10 Impact factor: 5.191
Authors: John W Hollingsworth; Meghan E Free; Zhuowei Li; Laura Novack Andrews; Hideki Nakano; Donald N Cook Journal: J Allergy Clin Immunol Date: 2010-04-15 Impact factor: 10.793
Authors: Emmanuelle Waubant; Robyn Lucas; Ellen Mowry; Jennifer Graves; Tomas Olsson; Lars Alfredsson; Annette Langer-Gould Journal: Ann Clin Transl Neurol Date: 2019-08-07 Impact factor: 4.511
Authors: Amy M Lavery; Emmanuelle Waubant; T Charles Casper; Shelly Roalstad; Meghan Candee; John Rose; Anita Belman; Bianca Weinstock-Guttman; Greg Aaen; Jan-Mendelt Tillema; Moses Rodriguez; Jayne Ness; Yolanda Harris; Jennifer Graves; Lauren Krupp; Leigh Charvet; Leslie Benson; Mark Gorman; Manikum Moodley; Mary Rensel; Manu Goyal; Soe Mar; Tanuja Chitnis; Teri Schreiner; Tim Lotze; Benjamin Greenberg; Ilana Kahn; Jennifer Rubin; Amy T Waldman Journal: Ann Clin Transl Neurol Date: 2018-09-27 Impact factor: 4.511