Genetic Susceptibility to Traffic Related Pollutants

Student thesis: Doctoral ThesisDoctor of Philosophy


A strong correlation exists between acute and chronic exposure to traffic derived pollutants and poor respiratory health. Specifically, diesel exhaust (DE) components such as NO2 and fine particles (PM2.5) have been related to impaired lung growth and increased respiratory and allergic symptoms in children and adults living near busy roads. On this basis, implementation of strategies to reduce diesel emissions and improve air quality should provide a measureable improvement in the respiratory health of populations resident in high traffic areas. The introduction of the London Low Emission Zone (LEZ), the largest of its kind in the world, covering an area of 2,644 km2 and a resident population of more than 8 million, provided a unique opportunity to examine this, as well to quantify the impact of DE emissions on the respiratory health of London's population.
London’s Low Emission Zone was introduced as part of the Mayor of London’s Air Quality Strategy, with the aim of improving public health through targeted reductions in tail pipe emissions from the most polluting vehicles entering the city. The objective of decreasing PM10 concentrations was to be achieved by restricting the entry of the oldest and most polluting diesel vehicles (heavy goods vehicles (HGVs), buses and coaches, larger vans and minibuses) into Greater London by providing incentives to operators to upgrade their fleets to lower emission vehicles. The Low Emission Zone was enacted as a phased tightening of emission standards for each vehicle class, with the first phase coming into force at the beginning of February 2008. This applied to HGVs greater than 12 tonnes and restricted entry to the zone for those vehicles not meeting the Euro III emissions standard for PM10. Phase 2 followed in July 2008 widening restrictions to include HGVs between 3.5 and 12 tonnes, buses and coaches. Vehicles failing to meet these emissions standards within the zone were initially charged £200 (£100 for vans an minibuses) per day, with enforcement achieve using cameras to identify the registration numbers of vehicles and the Driver and Vehicle Licensing Agency (DVLA) database to identify a vehicle’s emissions standard. In its initial configuration phase 3, restricting access to heavier LGVs and mini-buses not meeting Euro III PM standard was planned for October 2008.
In our initial four year study design we planned to examine the respiratory health of cross sectional panels of 8-9 year old school children living within the zone from November 2008; encompassing the first two years post phases 1 and 2, and two years post phase 3 (see Figure s1). As the subject recruitment and health assessments began in November 2008, this afforded us the opportunity of addressing the impact of the third phase of the LEZ, by comparing lung function and respiratory symptoms in the two years before and after phase 3. We also planned to examine year-on-year changes related to projected reductions in vehicle emissions as newer cleaner vehicles entered the fleet, independent of the LEZ, and the increased period the children had lived within the zone, from 11-15 months (11.5-13.9% of lifetime) in year 1, to 44-60 months (45.8-55.6%) in year 4. During annual school visits, children were asked to perform spirometry and provide a urine sample for the assessment of exposure (metals, reflective of defined traffic sources) and response biomarkers (oxidative damage). In addition, the parents/guardians of the children completed a questionnaire on respiratory / allergic symptoms and the children provided DNA samples to investigate genetic susceptibility to the detrimental effects of air pollution, focusing on a panel of antioxidant and xenobiotic genes, as well as a genetic marker associated with the onset of childhood asthma.
In May 2008 Boris Johnson was elected the new Mayor of London, with a manifesto commitment to review ongoing traffic management schemes within the city, including the LEZ, and on the 2nd February 2009 he announced intention to cancel the third phase of the LEZ, subject to the outcome of a public consultation later in the year. This political decision therefore robbed us of the original intervention we were planning to address in our original design. Following a further consultation, the scheme was finally fully implemented and expanded on the 3rd of January 2012 (LEZ phase 3 and 4), with Euro III emission standards for minibuses and vans and a further tightening of emission standards (Euro IV) on Lorries over 12 tonnes, between 3.5-12 tonnes, as well as buses and coaches. In light of this development we obtained additional funding to examine furthers panels of school children in Nov 2012 - March 2013 and Nov 2013 - March 2014, extending our study to six years, allowing a formal assessment of the three years pre and two years post LEZ phase 3 and 4, with year 4 straddling the periods of phase 3 and 4 implementation (Figure s2). Children at the conclusion of the study in March 2014, who have been resident within the LEZ since birth will have lived within the zone for 68.8-83.3% of their lives.
The data presented in this thesis is therefore based upon the first three years of the study and therefore constitutes a baseline analysis of the relationship between air pollution in London and our key respiratory endpoints prior to the formal evaluation of Phase 3 and 4 in 2014/15.
In the first experimental chapter (Chapter 3) I evaluated the associations between traffic-related air pollutants and respiratory/allergic symptoms within our cross-sectional children's cohort. Information on respiratory/allergic symptoms was obtained using a parent-completed questionnaire and linked to modelled annual air pollutant concentrations based on the residential address of each child, using a multivariable mixed effects logistic regression analysis. Exposure to traffic-related air pollutants was associated with current rhinitis (NOx [OR 1.01, 95% CI 1.00-1.02], NO2 [1.03, 1.00-1.06], PM10 [1.16, 1.04-1.28] and PM2.5 [1.38, 1.08-1.78], all per g/m3), but not with other respiratory/allergic symptoms. Furthermore, over the first three years of the operation of London's LEZ I did not observe evidence of reduced ambient air pollution levels, or year-on-year changes in the prevalence of respiratory/allergic symptoms. I found no evidence that these associations were modified by polymorphisms in gasdermin B, located at the chromosome 17q12, associated with the risk of childhood asthma. These data confirm previously reported associations between traffic-related air pollutant exposures and symptoms of current rhinitis. Importantly, whilst the data is largely confirmatory, this remains one of the few studies that has addressed respiratory symptoms in urban children over the period of rapid dieselization within Europe.
In Chapter 4 I report evidence of reduced lung volumes (FVC - Forced Vital Capacity) in children living within the study area. This negative association was small and most strongly associated with modeled annual NOx concentrations, at the residential address level. A straightforward method to discriminate between acute versus chronic pollutant effects was developed for the study. Acute exposures were assessed by scaling annual mean concentrations according to a ‘Nowcast’ factor calculated for each pollutant for the period immediately prior to the health assessment. This factor was defined as the ratio between concentrations measured by a local subset of London Air Quality Network monitoring sites in the prior period, and the annual mean measured by the same sites. Using this approach I was able to dissect out whether basal lung function was related to short or long term exposures. In the absence of relationships between FEV1 and FVC with 24 hour and 7 day average exposures, the association between FVC and annual pollutant exposures was interpreted as reflecting evidence of impaired lung growth. In this initial analysis I found no evidence that polymorphisms in the commonly studied glutathione S transferases (GSTM1 and GSTP1) and NADPH quinone oxidoreducatse (NQO1) genes modified the association between lung function and pollutant exposure. In a secondary analysis I examined whether polymorphisms in Cytochrome P450, family 1, subfamily A, polypeptide 1 (CYP1A1) and the Aryl hydrocarbon receptor (AhR) might modify the association between pollutant exposures and lung function, based on their role in the xenobiotic metabolism of Polycyclic aromatic hydrocarbons (PAHs). This is the first time polymorphisms in these genes have been investigated in the context of air pollution – lung function interactions. I found evidence that two SNPs in CYP1A1 influenced acute lung function responses to air pollution, one being protective and the other detrimental. Again all the significant relationships were with FVC and not FEV1, in contrast to much of the pre-existing literature. No noticeable improvement in respiratory health was observed in the children studied over the first three years of the LEZ.
In the final results chapter (5) I explored the relationship between pollutant exposures in the same cohort of children with urinary biomarkers of oxidative stress: 8-oxo-2'-deoxyguanosine (8-oxodG, as a marker of DNA damage) and 8-isoprostane (lipid oxidation). In addition, I also explored whether the determination of a panel of urinary metals would be informative as a biomarker of traffic exposure. These data demonstrated a positive relationship between urinary 8-oxodG concentrations and short term, acute particulate exposures (both PM10 and PM2.5). These responses appeared to reverse with longer term exposures (both 7 day and annual), i.e. high pollutant exposures were associated with lower urinary 8-oxodG concentrations, suggesting some form of biologic adaption to the pollutant stress. I therefore examined a panel of SNPs in genes, either directly related to oxidative stress (GCLM, Glutamate-cysteine ligase regulatory subunit and SOD3, extracellular superoxide dismutase), or the coordination of the antioxidant response (Nrf2, Nuclear factor (erythroid-derived 2)-like 2). Whilst polymorphisms in SOD3 did not modify the relationship between urinary markers of oxidative stress and the pollutant exposures, there was evidence that SNPs in Nrf2 appeared to blunt the sub-chronic exposure adaption. The work on the use of urinary metals as biomarkers did not prove fruitful. Therefore overall this series of studies have demonstrated long term effects of primary pollutants on indices of lung function and systemic oxidative stress.
Date of Award2014
Original languageEnglish
Awarding Institution
  • King's College London
SupervisorIan Mudway (Supervisor) & Frank Kelly (Supervisor)

Cite this