Secondary Logo

Share this article on:

Autism Spectrum Disorder: Interaction of Air Pollution with the MET Receptor Tyrosine Kinase Gene

Volk, Heather E.a,b,c; Kerin, Taraa; Lurmann, Fredd; Hertz-Picciotto, Irvae; McConnell, Roba; Campbell, Daniel B.c,f,g

doi: 10.1097/EDE.0000000000000030
Air Pollution

Background: Independent studies report association of autism spectrum disorder with air pollution exposure and a functional promoter variant (rs1858830) in the MET receptor tyrosine kinase (MET) gene. Toxicological data find altered brain Met expression in mice after prenatal exposure to a model air pollutant. Our objective was to investigate whether air pollution exposure and MET rs1858830 genotype interact to alter the risk of autism spectrum disorder.

Methods: We studied 252 cases of autism spectrum disorder and 156 typically developing controls from the Childhood Autism Risk from Genetics and the Environment Study. Air pollution exposure was assigned for local traffic-related sources and regional sources (particulate matter, nitrogen dioxide, and ozone). MET genotype was determined by direct resequencing.

Results: Subjects with both MET rs1858830 CC genotype and high air pollutant exposures were at increased risk of autism spectrum disorder compared with subjects who had both the CG/GG genotypes and lower air pollutant exposures. There was evidence of multiplicative interaction between NO2 and MET CC genotype (P= 0.03).

Conclusions: MET rs1858830 CC genotype and air pollutant exposure may interact to increase the risk of autism spectrum disorder.

Supplemental Digital Content is available in the text.

From the aDepartment of Preventive Medicine, Keck School of Medicine, University of Southern California, Los Angeles, CA; bDepartment of Pediatrics, Children’s Hospital Los Angeles, University of Southern California, Los Angeles, CA; cZilkha Neurogenetic Institute, Keck School of Medicine, University of Southern California, Los Angeles, CA; dSonoma Technology, Inc., Petaluma, CA; eDepartment of Public Health Sciences, University of California, Davis, Davis, CA; fDepartment of Psychiatry and the Behavioral Sciences, Keck School of Medicine, University of Southern California, Los Angeles, CA; and gCenter for Genomic Psychiatry, Keck School of Medicine, University of Southern California, Los Angeles, CA.

F.L. is an employee of Sonoma Technology Inc., Petaluma, CA. R.M.C. received support from an air quality violations settlement agreement between the South Coast Air Quality Management District, a California state regulatory agency, and BP. H.E.V. received travel funds from Autism Speaks to present a paper at an academic conference. The other authors declare no competing financial interests.

Supported by NIEHS; ES019002, ES013578, ES007048, ES11269, ES015359, ES016535, and ES011627; EPA Star-R; 823392, 833292; and MIND Institute matching funds and pilot grant program.

Supplemental digital content is available through direct URL citations in the HTML and PDF versions of this article ( This content is not peer-reviewed or copy-edited; it is the sole responsibility of the author.

Correspondence: Heather E. Volk, University of Southern California, 2001 N. Soto Street, MC 9237, Los Angeles, CA 90089. E-mail:

Autism and autism spectrum disorders are complex neurodevelopmental disorders characterized by deficits in social interaction, communication, and behavioral flexibility. The complex phenotypic presentation of these disorders suggests that multiple genetic and environmental factors contribute to risk, and gene-environment interactions are widely believed to underlie autism spectrum disorders. Few studies have addressed joint risk from specific genetic susceptibility in combination with a specific environmental exposure or class of exposures.1 In previous independent studies, we have identified (1) increased risk of autism spectrum disorder among children exposed to high levels of local near-roadway traffic-related air pollution and regional particulate matter near the time of birth2,3; (2) increased risk of autism spectrum disorder among children with the C allele of the MET gene promoter variant rs1858830,4,5 which is associated with decreased expression of MET protein in brain6 and immune system7; and (3) decreased MET protein expression in brain and altered behavior in offspring of mouse dams exposed during pregnancy to the polycyclic aromatic hydrocarbon benzo(a)pyrene (a component of traffic-related air pollution and particulate matter).8 Based on these independent autism spectrum disorder associations and the biological link between benzo(a)pyrene and MET, we hypothesized that a gene-environment interaction contributes to the risk of autism spectrum disorder.

In children, as in animals, prenatal polycyclic aromatic hydrocarbon exposure has been associated with intelligence (IQ) deficits at 5 years of age and with increased anxiety, depression, and inattention at 6 to 7 years of age.8–10 In this study, we investigated the relationship of air pollution exposure and genotype at the MET rs1858830 locus with autism spectrum disorder.

Back to Top | Article Outline


Description of Sample

The Childhood Autism Risks From Genetics and the Environment Study is a population-based, case-control study of preschool children from California. Participants were born in California and lived with at least one English- or Spanish-speaking biological parent in one of the study catchment areas related to specific regional centers in California. Subjects were 24 to 60 months of age at the time of recruitment; additional details on study design are provided elsewhere.11 For this analysis, cases met criteria for autism or autism spectrum disorder based on the Autism Diagnostic Observation Schedules and the Autism Diagnostic Interview-Revised. Typically developing controls were children who received a score <15 on the Social Communication Questionnaire and also showed no evidence of other types of developmental delay (composite scores >70 on Mullen Scales of Early Learning and Vineland Adaptive Behavior Scales). We assigned air pollution exposure to 669 study participants based on their residential histories and available exposure databases (as described below).3 For 63% of participants, parents agreed to give blood and consented to share biospecimens with researchers outside the original study team. This analysis includes 251 cases with a confirmed diagnosis of autism or autism spectrum disorder and 156 controls with typical development.

In parental interviews, we collected data on demographic characteristics, medical conditions, and environmental exposures, including residential history.11 Residential histories recorded dates and address locations where the mother lived, beginning at conception through the most recent place of residence, as well as any other place of residence where the child lived. These dates and addresses were used to develop air pollution exposure metrics.3 Prenatal and birth addresses were used to develop a weighted average of pollution exposure. In this analysis, we focus on air pollution exposure during the prenatal period.

Back to Top | Article Outline

Air Pollution Exposure Assignment

We assigned modeled estimates of traffic-related air pollution exposure to study participants using the CALINE4 line-source air quality dispersion model.12 Included in the model is information on roadway geometry, link-based traffic volumes, period-specific meteorological conditions (wind speed and direction, atmospheric stability, and mixing heights), and vehicle emission rates.3 CALINE4 pollutant concentration estimates are indicators of the traffic-related air pollutant mixture rather than of a specific pollutant. We estimated residential exposure derived from freeways, non-freeways, and all roads located within 5 km from home.

We also used regional air quality data to assign exposure for particulate matter <2.5 and <10 microns in diameter (PM2.5 and PM10), nitrogen dioxide, and ozone using data from the US Environmental Protection Agency Air Quality System ( supplemented for Southern California by the University of Southern California’s Children’s Health Study data for 1997–2009.3 When no Federal Reference/Equivalent Method data for particulate matter were available for a given monitoring station in the Air Quality System, Children’s Health Study continuous particulate matter data were used. The monthly air quality data from monitoring stations located within 50 km of each residence were used for spatial interpolation of ambient concentrations. The spatial interpolations were based on inverse distance-squared weighting of data from up to four closest stations located within 50 km of each participant residence; however, if one or more stations were located within 5 km from residence, then only data from the stations within 5 km were used for the interpolation.

Back to Top | Article Outline

Genotyping Methods

Blood was collected from participants as a part of the study protocol, with genomic DNA extracted from peripheral blood leukocytes using standard methods (Gentra Puregene Blood Kit; Qiagen, Germantown, MD). Because the rs1858830 single nucleotide polymorphism (SNP) falls within a highly GC-rich region, indirect genotyping methods fail when using genomic DNA. A 652-bp fragment containing the rs1858830 SNP was amplified from 15-ng genomic DNA with primers 5'-GATTTCCCTCTGGGTGGTG-3' (forward) and 5'-CAAGCCCCATTCTAGTTTCG-3' (reverse). Polymerase chain reaction (PCR) analysis was performed with the KOD Xtreme Hot Start Polymerase Kit (EMD Millipore, Billerica, MA), which is designed to amplify regions with high GC content. Cycling conditions were as follows: 95°C for 5 minutes followed by 35 cycles of 95°C for 30 seconds, 68°C for 30 seconds, and 72°C for 1 minute. Specific amplification of the 652-bp product was confirmed by agarose gel electrophoresis. Each PCR product was subjected to direct resequencing using an ABI 3730xl using Big Dye Terminator chemistry (Life Technologies, Grand Island, NY). Genotype at the MET rs1858830 locus was determined from the sequencing result using Sequencher software (Gene Codes, Ann Arbor, MI).

Back to Top | Article Outline

Statistical Analysis

Logistic regression models were used to estimate odds ratios (ORs) and 95% confidence intervals (CIs) for air pollution exposure and MET genotype. We examined each pollutant separately, categorizing children as “high exposure” if the pregnancy average exposure for traffic-related air pollution, PM2.5 or PM10, nitrogen dioxide, or ozone was in the top 25% of the exposure distribution. Participants in the other 75% served as “low exposure” in our analyses. These categorizations are consistent with findings identified in our previous work.3 We also explored more and less extreme exposure cut points (eTable 1, Because previous research demonstrated an increased risk of autism spectrum disorder owing to overtransmission of the C allele and because functional studies suggest the MET CC genotype is associated with decreased MET expression, we compared the CC genotype to the CG and GG genotypes in our analyses.5 Analyses were adjusted for potential confounders, including child’s sex and ethnicity, maximum education level in the home, maternal age, home ownership, and prenatal smoking.

Back to Top | Article Outline


Genotyped subjects were similar to nongenotyped subjects in autism spectrum disorder status and air pollution exposure (eTable 1, Genotyped subjects were less likely to have a mother who smoked during pregnancy and less likely to have high nitrogen dioxide exposure compared with nongenotyped subjects. MET rs1858830 genotype frequencies did not vary across cases and controls (χ2 = 1.40, 2df). We did not find an increased risk of autism spectrum disorder for the MET CC genotype compared with CG/GG genotypes (crude OR = 0.9 [95% CI = 0.6–1.4]). Autism spectrum disorder was associated with exposure to the top quartile of traffic-related air pollution (1.7 [1.0–2.7]), PM10 (2.5 [1.6–4.3]), PM2.5 (1.9 [1.2–3.1]), and nitrogen dioxide (1.7 [1.1–2.7]).

We then parameterized our model based on both MET genotype and air pollution exposure. Synergistic effects were observed between MET CC genotype and local traffic-related air pollution, regional PM10, and regional nitrogen dioxide exposure; adjusted ORs were, respectively, 2.9 (1.0–10.6), 3.2 (1.3–9.1), and 3.6 (1.3–13), comparing the high-risk genotype and highly exposed children to those with low exposure and without the risk genotype (Table). There was evidence of multiplicative interaction between NO2 and MET CC genotype (P = 0.03) and between local traffic-related air pollution and MET CC genotype (P = 0.09). Analyses exploring alternative cut points found the persistence of joint effects of traffic-related air pollution and MET CC genotype using either lower or higher cut points for defining high exposure (eTable 2, Joint effects of MET CC genotype with PM10 or nitrogen dioxide are additionally present at higher cut points.



Back to Top | Article Outline


Examination of joint pollution and gene effects suggests that subjects with both the MET rs1858830 CC genotype and high air pollutant exposure were at increased risk of autism spectrum disorder compared with subjects who had both the CG/GG genotypes and lower air pollutant exposure. Given that the MET CC genotype had no impact on the 75% of the population with lower air pollutant exposures, these data suggest a gene-environment interaction for autism spectrum disorder based on MET genotype and air pollution exposure. These results require independent replication and a more detailed understanding of the underlying biology. However, these data add to the literature supporting a role for gene-environment interactions in autism spectrum disorder etiology. They also point to the contribution of common alleles for which gene-only analyses show inconsistent evidence of a link to autism spectrum disorder.

Back to Top | Article Outline


We thank CHARGE families for participating in this research and CHARGE study staff for their time and effort.

Back to Top | Article Outline


1. Schmidt RJ, Hansen RL, Hartiala J, et al. Prenatal vitamins, one-carbon metabolism gene variants, and risk for autism. Epidemiology. 2011;22:476–485
2. Volk HE, Hertz-Picciotto I, Delwiche L, Lurmann F, McConnell R. Residential Proximity to Freeways and Autism in the CHARGE Study. Environ Health Perspect. 2011;119:873–877
3. Volk H, Lurmann F, Penfold B, Hertz-Picciotto I, McConnell R. Traffic Related Air Pollution, Particulate Matter, and Autism Risk. JAMA Psychiatry. 2012;70:71–77
4. Campbell DB, Li C, Sutcliffe JS, Persico AM, Levitt P. Genetic evidence implicating multiple genes in the MET receptor tyrosine kinase pathway in autism spectrum disorder. Autism Res. 2008;1:159–168
5. Campbell DB, Sutcliffe JS, Ebert PJ, et al. A genetic variant that disrupts MET transcription is associated with autism. Proc Natl Acad Sci U S A. 2006;103:16834–16839
6. Campbell DB, D’Oronzio R, Garbett K, et al. Disruption of cerebral cortex MET signaling in autism spectrum disorder. Ann Neurol. 2007;62:243–250
7. Heuer L, Braunschweig D, Ashwood P, Van de Water J, Campbell DB. Association of a MET genetic variant with autism-associated maternal autoantibodies to fetal brain proteins and cytokine expression. Transl Psychiatry. 2011;1:e48
8. Sheng L, Ding X, Ferguson M, et al. Prenatal polycyclic aromatic hydrocarbon exposure leads to behavioral deficits and downregulation of receptor tyrosine kinase, MET. Toxicol Sci. 2010;118:625–634
9. Perera FP, Li Z, Whyatt R, et al. Prenatal airborne polycyclic aromatic hydrocarbon exposure and child IQ at age 5 years. Pediatrics. 2009;124:e195–e202
10. Perera FP, Wang S, Vishnevetsky J, et al. Polycyclic aromatic hydrocarbons-aromatic DNA adducts in cord blood and behavior scores in New York city children. Environ Health Persp. 2011;119:1176–1181
11. Hertz-Picciotto I, Croen LA, Hansen R, Jones CR, van de Water J, Pessah IN. The CHARGE study: an epidemiologic investigation of genetic and environmental factors contributing to autism. Environ Health Perspect. 2006;114:1119–1125
12. Bensen PE. A review of the development and application of the CA-LINE3 and 4 models. Atmos Environ. 1992;26B:379–390

Supplemental Digital Content

Back to Top | Article Outline
© 2014 by Lippincott Williams & Wilkins, Inc