(a) Pulmonary oxidative stress and inflammation in humans

See Table 4.17.

Table 4.17

Oxidative stress and inflammation biomarkers in exhaled breath condensate from humans exposed to air pollution.

Studies on oxidative stress in humans who have been exposed to air pollution have mainly applied biomarkers of oxidatively damaged lipids or inflammation markers in exhaled breath or bronchoalveolar lavage fluid (BALF). The publications are grouped into controlled exposure, panel, and cross-sectional studies.

Among the controlled exposure studies are some that have obtained direct measurements of pulmonary inflammation by analysis of BALF cell counts. The Working Group has included studies on bronchial instillation of PM in humans for the purpose of bridging observations in animal models on the same exposure, although it recognizes that extrapolation to real-life human exposures is challenging. Ghio & Devlin (2001) reported results from a study in which young and healthy people were exposed by bronchial instillation to aqueous extracts of PM₁₀ collected before, during, and after a steel mill strike in Utah Valley, Utah, USA (Ghio & Devlin, 2001). The particles that were collected before and after the strike were associated with higher levels of neutrophils, pro-inflammatory cytokines (interleukin-1 beta [IL-1β], tumour necrosis factor [TNF], and IL-8), and protein (a marker of epithelial damage) in the BALF at 24 hours after the instillation of 500 µg of particle extract into the lungs of non-smoking volunteers. Extracts from periods when the steel mill was operating had high concentrations of metals, and the extracts generated ROS in acellular conditions, which was diminished by addition of DFO (a metal chelator) or dimethylthiourea (an antioxidant) (Ghio & Devlin, 2001). Schaumann et al. (2004) instilled into the lungs of 12 healthy volunteers PM_2.5 (100 µg, corresponding to 24 hours of inhalation of 100 µg/m³) from two locations in Germany, characterized as being an area with mining and smelter industry and a non-polluted area. The instillation increased the total number of cells in BALF, whereas there was no difference in differential cell counts of neutrophils, lymphocytes, and monocytes. Nevertheless, it was only instillation of particles from a polluted area that was associated with increased concentrations of some pro-inflammatory cytokines in BALF (IL-6 and TNF, but not IL-1 and IL-8) and increased ex vivo ROS production in zymosan-stimulated BAL cells. There were unaltered levels of markers of cell damage in the BALF (protein, albumin, and lactate dehydrogenase activity), and the glutathione concentration was unaltered (Schaumann et al., 2004). A study on inhalation exposure to concentrated outdoor particles (23–311 µg/m³) from Chapel Hill, North Carolina, USA, for 2 hours with moderate exercise (exercise 15 minutes; rest 15 minutes) and analysis of pulmonary inflammation 18 hours after cessation of the exposure showed a mild increase in neutrophils in BALF, whereas there were unaltered levels of IL-6, IL-8, and prostaglandin E2 (PGE₂) (Ghio et al., 2000). Another study in 19 healthy non-smoking volunteers from Chapel Hill on outdoor UFP (40 848–205 648 UFP/cm³; 1–50 µg/m³) showed a moderate increase in the concentration of IL-8 in BALF, whereas there were unaltered levels of IL-6 and PGE₂ (Samet et al., 2009).

A non-invasive way to study the effect of air pollution is by analysis of markers of inflammation and oxidative stress in exhaled air. As this is a relatively easy method, it has been used in epidemiological studies as well as studies on controlled exposures to air pollution. It was shown that healthy young subjects had elevated levels of malondialdehyde (MDA) in exhaled breath condensate (EBC) after exercise at a location with high traffic-generated UFP (252 290 UFP/cm³) compared with the same type of exercise at a location with less traffic (7382 particles/cm³) (Rundell et al., 2008). This study also showed lower concentrations of exhaled NO and nitrate, which was hypothesized to be due to the formation of peroxynitrite (Rundell et al., 2008). Another study of controlled exposure to elderly men with stable coronary heart disease showed that inhalation of CAPs (190 µg/m³) increased the concentration of 8-isoprostanes in EBC at 6 hours and 24 hours, whereas there was no effect on 3-nitrotyrosine levels (Mills et al., 2008). Exposure to air in a road tunnel in Stockholm, Sweden (80 µg/m³ of PM_2.5 for 2 hours), had no effect on exhaled NO in people with asthma, and there were inconsistent associations between exposure and levels of IL-10, IL-1β, IL-6, IL-8, and TNF in nasal lavage fluid (Larsson et al., 2010). Strak et al. (2012) studied subjects who were exposed to outdoor air at a traffic site or an urban site (5 hours, with intermittent exercise); they showed associations between the particle number concentration and fractional exhaled NO (FeNO).

The Beijing Olympics in 2008 has formed the basis for studies on the association between improvements in outdoor air quality and biomarkers of oxidative stress and inflammation. A study in 125 healthy adults showed that the EBC content of NO, nitrates, nitrites, MDA, and 8-isoprostanes (using an ELISA method) was lower in young and healthy subjects during the Olympics compared with periods before and after the Olympics (Gong et al., 2013; Huang et al., 2012).

Panel studies have typically focused on subjects with lung or cardiovascular diseases. It was shown that there was a positive association between personal PM_2.5 exposure and exhaled NO in children (aged 9–13 years) with asthma from Seattle, Washington, USA, who did not take corticosteroid medication (Koenig et al., 2003, 2005; Mar et al., 2005). Another panel study in Ontario, Canada, showed no association between air pollution exposure and FeNO; there was a positive association between levels of air pollution exposure components (PM_2.5, NO₂, and SO₂) and thiobarbituric acid reactive substances (TBARS; an oxidative stress marker) in EBC, which was not a particularly reliable assay for detection of lipid peroxidation products (Liu et al., 2009a). Positive associations between levels of PM_2.5 and ozone, based on stationary monitoring data, and MDA levels in EBC were observed in children with asthma in Mexico City (Romieu et al., 2008). A study in elderly subjects with asthma or COPD in Seattle, Washington, USA, showed an association between outdoor concentrations of PM_2.5 and FeNO, whereas there was no association between FeNO and personal exposure to PM_2.5 (Jansen et al., 2005). In addition, elderly subjects with coronary artery disease in the Los Angeles basin, California, USA had a positive association between exposure markers (PM_2.5 and ozone) and FeNO (Delfino et al., 2010a). Studies of healthy children in Steubenville, Ohio, USA, or Mexico City also indicated positive associations between air pollution exposure and FeNO (Adamkiewicz et al., 2004; Barraza-Villarreal et al., 2008), whereas there was no association between levels of hydrogen peroxide (H₂O₂) in EBC and air pollution exposure among students in Christchurch, New Zealand (Epton et al., 2008).

A cross-sectional study of subjects with lung diseases (asthma or COPD) in four European cities showed an association between levels of coarse particles at a central monitoring station and levels of NO_x in EBC, whereas there was no association with personal exposure to PM_2.5, PM₁₀, or coarse fraction as measured either near or inside the homes (Manney et al., 2012). Another cross-sectional study on children in the Netherlands showed a positive association between PM₁₀ exposure and exhaled NO; children from an urban area had higher nasal lavage levels of IL-8 and NO_x compared with children from a suburban area (Steerenberg et al., 2001).

(b) Systemic effects of inflammation in humans

The studies on systemic inflammation have mainly centred on markers of cardiovascular diseases, including acute-phase proteins (fibrinogen and C-reactive protein [CRP]), platelets, von Willebrand factor, haematocrit, whole blood viscosity, and leukocyte counts (Delfino et al., 2005). The measurement of CRP especially has been popular because it is used clinically, it can increase by more than 3 orders of magnitude during an acute-phase response, and it has a relatively short half-life in plasma (~19 hours). A recent review of the association between air pollution levels and CRP levels in humans encompassed a total of 44 publications, stratified into cross-sectional, panel, and randomized cross-over trials (Li et al., 2012). The most important conclusions from that survey were that there was an association between air pollution exposure and elevated levels of CRP in children in cross-sectional studies, whereas there were inconsistent results in adults, which might be related to the inclusion of subjects with prescribed statins or anti-inflammatory drugs. It was also noted that the randomized cross-over trials mainly showed no association between air pollution exposure and CRP levels in plasma, which could be because these studies had few subjects and relatively short exposure duration (Li et al., 2012). One of the studies on controlled exposure used relatively high concentrations of CAPs (190 µg/m³ for 2 hours) and found no change in serum levels of CRP and total leukocyte counts, although there was a transient and marginal increase in the number of monocytes in blood (Mills et al., 2008). In addition, two studies on indoor air filtration for 24–48 hours with relatively low exposure gradients of traffic-generated emissions in Copenhagen, Denmark, showed no effect on levels of CRP, IL-6, TNF, and fibrinogen (Bräuner et al., 2008a, b). A 7-day intervention period with air filtration in homes of a wood smoke-affected area in British Columbia, Canada, found an association between the indoor concentration of fine particles and CRP levels, whereas there was no association with IL-6 levels (Allen et al., 2011).

The production of CRP is regulated in response to elevated levels of IL-6, IL-1, and TNF. A study indicated that children in Mexico City, compared with children in a low-pollution city, had a systemic pro-inflammatory state as determined by elevated plasma/serum levels of TNF, IL-1β, PGE₂, and CRP (Calderón-Garcidueñas et al., 2008). A very large study of subjects in Lausanne, Switzerland, with 6183 adult participants showed associations between short-term exposures to PM₁₀ and elevated levels of IL-1β, IL-6, and TNF, whereas there was no effect on CRP levels (Tsai et al., 2012). Other studies of people with coronary artery disease also found associations between exposure to small particles (PM_0.25) and levels of CRP, IL-6, and TNF in plasma (Delfino et al., 2008, 2009, 2010b). A study in Singapore showed that subjects had elevated serum levels of TNF, IL-1β, and IL-6 during a period with haze, with high outdoor air pollution concentrations (PM₁₀, 125 µg/m³), compared with a period afterwards with a low air pollution level (PM₁₀, 14 µg/m³) (van Eeden et al., 2001). However, there are also studies showing inconsistent associations between air pollution exposure and levels of CRP, IL-6, IL-8, serum amyloid A, and fibrinogen (Huttunen et al., 2012; Rückerl et al., 2006, 2007; Strak et al., 2013; Wu et al., 2012) or no association with levels of CRP, IL-6, IL-10, TNF, and fibrinogen (Liu et al., 2007, 2009b; Zuurbier et al., 2011).

Studies on oxidative stress biomarkers in biomonitoring include oxidation products of lipids, proteins, and DNA. Several studies of these products from areas in the Czech Republic have shown positive associations between air pollution exposure and oxidative stress markers (Bagryantseva et al., 2010; Rossner et al., 2007, 2011c, 2013b).

(c) Pulmonary inflammation and ROS production in experimental systems

(a) Cardiorespiratory effects

The cardiovascular and respiratory effects of outdoor air pollution have been examined in many studies worldwide using diverse research designs and are summarized in numerous reviews and regulatory documents (e.g. Brook, 2008; Brook et al., 2010; EPA, 2006; Hoek et al., 2013; Lai et al., 2013; WHO, 2006). The effects of airborne PM have been most extensively studied. Exposure to PM is linked with increases in all-cause, cardiovascular, and respiratory mortality, as well as with other cardiovascular and respiratory effects, including hospitalization for acute respiratory events, decreased lung function, ischaemic events, stroke, arrhythmia, and reduced heart rate variability (Brook, 2008; Hoek et al., 2013; Pieters et al., 2012; Samet & Krewski, 2007; WHO, 2006). Positive associations have also been observed between exposure to NO_x and mortality from all causes and ischaemic health disease (Hoek et al., 2013; Mustafic et al., 2012). However, because NO₂ is closely correlated with other air pollutants from traffic-related sources, it is difficult to determine whether the effects are due specifically to NO₂, to other air pollutants, or to the complex mixture of pollutants (WHO, 2006).

(b) Reproductive effects

Prenatal exposure to air pollution has been hypothesized to affect the unborn child through several mechanisms, including oxidative stress, inflammatory processes, endocrine disruption, and germ-cell changes (Schwartz, 2004; Slama et al., 2008). Research on this topic is still inconclusive, but there is some evidence that prenatal exposure to outdoor air pollution increases the risk of preterm delivery, fetal growth deficit, and cardiac birth deficit (Wigle et al., 2008). A meta-analysis based on 62 studies estimated that exposure to PM, NO₂, and CO during pregnancy was associated with low birth weight and that exposure to PM and CO was associated with preterm birth (Stieb et al., 2012). Another systematic review of air pollution exposures and birth outcomes reported similar conclusions regarding low birth weight and preterm birth, as well as associations of small-for-gestational-age births with prenatal exposure to PM and of preterm birth with exposure to SO₂ (Shah & Balkhair, 2011). Transcriptomics allows for more mechanistic and holistic studies by analysing several genes that are upregulated or downregulated in exposed populations, for example as performed in the well-characterized Czech populations, including gene profiling of newborns (Šrám et al., 2013). These studies are explorative.

(c) Immunotoxic effects

Associations between exposures to outdoor air pollution and biomarkers of immunotoxicity associated with inflammatory responses, for example cytokines, have been reported in several studies.

(d) Endocrine effects

Statistically significant associations between long-term exposure to traffic-related air pollution at the residence and diabetes mortality were reported in a Danish follow-up study of 52 061 cohort participants (Raaschou-Nielsen et al., 2013a). An association of type 2 diabetes in women with traffic-related air pollution measured by NO₂ has also been reported (Brook et al., 2008). Urinary excretion of 17-ketosteroids and 17-hydroxycorticosteroids, markers of adrenal cortex functions, was reported to be significantly lower in male children living in polluted areas compared with children living in clean areas (Watanabe, 2000), but the paper lacks important details about the study population.

(a) DNA damage and chromosomal aberrations in human sperm

See Table 4.18.

Several studies in the Czech Republic have evaluated the association of outdoor air pollution exposure with chromatin damage in humans by SCSA. A preliminary study showed an increased percentage of sperm with abnormal chromatin structure (denatured DNA susceptibility) (expressed as COMPα_t, cells outside the main population of cells) at periods of high air pollution in Teplice (Selevan et al., 2000). High COMPα_t (> 30) has been associated with infertility and spontaneous abortion (Evenson et al., 1999). A subsequent 2-year follow-up study confirmed that there was a significant association between exposure to periods of air pollution and the increased DNA damage in human sperm; however, there was no association between aneuploidy in sperm and outdoor air pollution (Rubes et al., 2005). Rubes et al. also showed that men with the GSTM1 null genotype exhibited increased susceptibility to sperm DNA damage associated with exposure (Rubes et al., 2007). Furthermore, significantly higher sperm DNA fragmentation index (DFI) values were observed in the winter compared with in the spring for police officers working outdoors in Prague (Rubes et al., 2010). A study in male traffic police in Prague showed that they had higher frequencies of aneuploidy in their sperm when sampled in January, when the PM₁₀ concentrations were high, compared with in March, when the PM₁₀ concentrations were low (Rubes et al., 1996; Srám et al., 1999).

(b) Mutations in male germ cells and the germline in animals

Studies of sentinel wildlife and laboratory rodents exposed to outdoor air (summarized in Supplemental Table S24, available online) have provided evidence for elevated germline mutation induced by air pollution (Somers & Cooper, 2009; Somers, 2011). Increased rates of germline mutations at minisatellite loci (tandem repetitive DNA loci) were seen in herring gulls (Larus argentatus) collected from industrial areas with high levels of air pollution, and minisatellite mutation rates decreased with increasing distance from the industrial coking oven and urbanization site (Yauk & Quinn, 1996; Yauk et al., 2000). When laboratory mice were exposed to outdoor air in a polluted industrial area near steel mills, a significant 1.5–2.0-fold elevation in heritable mutation frequency in the offspring was observed, primarily through ESTR mutation events in the paternal germline (Somers et al., 2002). This heritable mutation frequency was significantly reduced when mice were exposed in situ to air from a polluted area treated by a HEPA filtration system that removed 99.97% of particles 0.3 µm in diameter (Somers et al., 2004). [The Working Group noted that airborne PM was an important factor in induction of germline ESTR mutations.] Taken together, the series of studies indicate that exposure to outdoor air pollution could cause damage to male gametes, as shown in observed germline mutations and DNA damage in sperm of mice, although their contributions to fertility and reproduction are still unknown.

(c) Sperm abnormality in animals

In a sperm morphology study of mice treated with extracts of air samples collected in Shanghai, China, an increased frequency of germ-cell deformations was observed for most sites in the winter (Mao et al., 1993). A subsequent animal experimental study in Taiyuan, China, showed that intraperitoneal treatment of male Kunming mice with particle extracts from the residential area downwind of a coal combustion power plant induced sperm abnormality, CAs of spermatogonia and primary spermatocytes, and MN in spermatids (Sun et al., 1995). Elevated frequencies of head and tail deformities in the sperm were observed in feral mice living in an area with air highly polluted by automobile traffic in Rome, Italy (Ieradi et al., 1996).

In summary, a series of studies in humans, wildlife, and experimental animals indicate that outdoor air pollution might cause heritable mutations, sperm abnormalities, and germ-cell DNA damage. The evidence for heritable mutation derives from studies examining gulls as well as inbred and outbred mice.