Exploring the epigenome to identify biological links between the urban environment and neurodegenerative disease: an evidence review

ABSTRACT

Global urbanisation has occurred in tandem with population ageing, having implications on human cognitive health. Urban environmental factors such as air pollution are known risk factors of cognitive impairment and neurodegenerative disease. However, due to the sparse evidence base, the biological pathways by which urban environmental factors operate are not well understood. The aim of this review is to explain how exploring the epigenome (i.e. chemical modifications to the genome which do not change the underlying gene sequence) can further our understanding of these biological pathways. The epigenome is influenced by environmental factors and has implications for cognitive impairment and neurodegenerative disease. Utilising complex epigenetic analytical techniques including epigenetic clocks, Mendelian randomization and multi-omic approaches, it is possible to identify environmental consequences on underlying biology. Through better understanding of how epigenetic modifications, which can be inherited or change dynamically in response to environmental exposures, impact cognitive outcomes, we can work to encourage the development of public health policies, as well as urban planning and design policies to reduce the burden of neurodegenerative disease and encourage healthier ageing in the older adult population.

KEYWORDS:

• Ageing
• neurodegenerative disease
• cognitive impairment
• epigenetic
• methylation
• urban environment

Introduction

Urbanisation and neurodegenerative disease

In 2021, 55% of the global population lived in urban areas (World Health Organisation, 2021), a substantial change from the 1960s when 34% of the global population was urban living (United Nations Population Division, 2019). Urban environmental factors (physical, natural and built aspects of towns and inner cities including environmental by-products) can be harmful to human health and contribute to disease outcomes (Flies et al. 2019, Fuller et al. 2022). However, there are urban environment factors which benefit environmental health and human health, such as provision of high-quality urban green and blue space, and high neighbourhood walkability. Supporting the health of a growing urban population requires a healthy urban environment.

In tandem, our global population is ageing, with 2.1 billion people predicted to be 60 years of age or older by 2050 (World Health Organisation, 2021). Cellular impairment and disease risk increases with age (Hui 2017, Dominguez et al. 2021, Liu 2022). Neurodegenerative diseases (e.g. Alzheimer’s disease, Parkinson’s disease) are a type of non-communicable disease of the central nervous system which come with a high likelihood of premature mortality from neurodegenerative progression (White et al. 2020). Due to the lack of regenerative capacity of the central nervous system (Cooke et al. 2022), strategies to identify and develop preventative mechanisms are required. In 2022, over 55 million older adults were living with dementia globally, projected to double to 139 million persons by 2050 (Nichols 2022). The longer life expectancy of populations coupled with increasing risk of disease and ill health requires strategies to help populations age well with minimal health complications. Furthermore, the economic burden of disease has also become increasingly evident. In the UK, the total cost of dementia care per year is 34.7 billion pounds (Wittenberg et al. 2020). This highlights a research necessity to improve the ageing experience of the population, including reducing neurodegenerative risk.

There are numerous identified risk factors for neurodegenerative diseases. Age is believed to be the largest contributor (Hou et al. 2019, Ju and Tam 2022) to neurodegenerative risk. The global ageing of the population is therefore a concern and highlights the urgency for preventive population-level interventions. Other known risk factors include sex and vascular disease (Mielke et al. 2014, Sumien et al. 2021) which can arise from atherosclerotic pathology. Socioeconomic risk factors include being from a lower income household (Wang et al. 2021) and having lower educational attainment (Takasugi et al. 2021). In addition, environmental exposures, including heavy metals, pesticides (Chin-Chan et al. 2015, Tang 2020) and air pollutants (Wu et al. 2022), are also known risk factors of neurodegenerative disease. Other urban environment factors such as noise (Ho et al. 2012) and light (Baranyi et al. 2022) pollution have been associated with higher risk of cognitive impairment. Conversely, urban green and blue space presence has been evidenced to be protective against cognitive impairment (Zhang et al. 2022b). However, the biological pathways by which urban environment factors influence cognitive impairment and neurodegenerative risk are not fully understood (Heusinkveld et al. 2016, Chandra et al. 2022, Meng et al. 2022).

An attractive risk factor for neurodegenerative disease is epigenetic modification (Millan 2014, Zhang et al. 2022a). The epigenome is the term used to describe chemical modifications to the genome which do not change the underlying gene sequence (Morgensztern et al. 2018). These modifications can alter the expression of genes, resulting in a biological cascade which could either contribute to disease risk, or protect from disease (Ho et al. 2012). The epigenome is modifiable by environmental exposures, with urban environment factors such as air pollution being associated with epigenetic modifications (Baranyi et al. 2022, Holliday et al. 2022). This modification may contribute to the biological risk of neurodegenerative disease and is thereby a potential mechanism by which environmental exposures influence disease risk (Jirtle and Skinner 2007, Wu et al. 2023). This makes multi-omic exposomic research on urban environment and neurodegenerative disease of interest, including investigation into the consequences of exposures at both the genetic and epigenetic level.

Exposomic research

Recently, the exposome – endogenous and exogenous factors to which the body is exposed in the lifetime (Kalia et al. 2022) – is being explored to identify its role in age-associated conditions such as neurodegenerative disease. It has been hypothesised that epigenetic modifications could be used as a proxy to explore the exposome (Colwell et al. 2023), as it is possible to analyse and visualise epigenetic modifications which are the result of environmental exposures. By investigating epigenetic modifications in urban environmental-neurodegenerative research, it is possible to explore biological pathways by which urban environment factors influence neurodegenerative disease induction and progression.

Focusing on exogenous environmental risk factors for disease is beneficial considering the ability to modify exposure to them and the potential impact at a population level. Implementation of urban health policies have the potential to reduce exposure to harmful pollutants whilst encourage exposure to urban green and blue spaces which enhance both environmental and human health (Wilson and Xiao 2023). These policies would be a potential therapeutic, reducing the risk of neurodegenerative disease (Besser 2021).

Aim of review

This review aims to summarise the evidence on the influence of key urban environment factors which are known to affect human cognitive function and highlight how they influence the biological risk of neurodegenerative disease.

The urban environment and neurodegenerative risk

The next section synthesises the evidence for environmental factors.

Air pollution

It is estimated that 91% of urban residents live within polluted air (World Health Organisation 2021), in an area not conforming to the global air quality guidelines provided by the WHO (World Health Organisation 2021). Vehicle emissions are considered the largest contributor to poor urban air quality (Kim et al. 2015). The most common type of air pollution included in biological research is particulate matter (PM). PM is composed of inhalable pollutants such as nitrates, sulfates, endotoxins, carbon and heavy metals (World Health Organisation 2013). PM is categorised by its size: PM₁₀ is composed of coarse material, PM_2.5 of fine and PM_0.1 of ultra-fine (Benabed and Boulbair 2022). The size indicates the level of penetration into the respiratory tract, as PM₁₀ will infiltrate the tracheal and bronchial mucosa (Atkinson et al. 2010) while PM_0.1 is so small it can behave similar to a gas, diffusing at the level of the alveoli into the systemic circulation (Valavanidis et al. 2008). This makes PM_0.1 a highly penetrative and potentially damaging aspect of air pollution.

There is research debate surrounding the efficiency with which pollutants can bypass and disrupt the blood–brain barrier to directly impact the brain. Multiple studies have provided evidence that the blood–brain barrier, a highly selective protective barrier to the brain parenchyma, is disrupted through pollutant exposure (Oppenheim et al. 2013, Marin et al. 2018). Nanosized PM deposits have been discovered in the frontal cortex of human brains (Maher et al. 2016). A study by Maher et al., specifically analysed the presence of magnetite in the brain (Maher et al. 2016), which has been linked to Alzheimer’s disease (Pankhurst et al. 2008). Barrier disruption, allowing pollutant infiltration into the brain, could increase the risk of an intercranial inflammatory response and the generation of reactive oxygen species. The nasal olfactory tract is a further proposed inhalation mechanism by which pollutants target the brain. Olfactory deficit and dysfunction are common early signs of neurodegenerative disease (Lafaille-Magnan et al. 2017), including Alzheimer’s disease (Lafaille-Magnan et al. 2017) and Parkinson’s disease (Morley et al. 2018). Further validation of these findings is required to confirm or disprove these theories and identify clear pathways by which air pollutants access and impact the brain.

The impact of air pollutants on human health has been well documented in the literature, commonly focusing on PM_2.5. In a meta-analysis by Yu et al., investigating aspects of air pollution and risk of cognitive impairment, there was a significant positive correlation between PM_2.5 and the risk of cognitive impairment in population-based studies (Yu et al. 2020). However, there was moderate observed heterogeneity (I² 68.5%) in the 10 studies included in this meta-analysis. Visually, the funnel plot indicated publication bias; however, the Egger analysis suggested no significant evidence of this bias. In a recent systematic review by Peters et al. (2019), 13 population-based and cohort studies were analysed with an overall conclusion that greater exposure to air pollutants may be associated with increased risk of cognitive decline and dementia. PM_2.5 was the most commonly analysed pollutant in included studies. Included within this review was a study by Chen et al. (2017a), which assessed the risk of dementia against exposure to numerous air pollutants, finding varying results. Chen et al. found an increased risk of dementia associated with PM_2.5 (HR 1.04 95%CI 1.03-1.05). However, no association was found between dementia incidence and ozone. In a further study by Chen et al. (2017b), the residential proximity to a roadway was assessed against the risk of dementia. Using proximity to the road can be beneficial in environmental studies as it can relate to both traffic-related air pollution and traffic-related noise pollution. This study found that living within 50 m of a roadway increased the risk of dementia (HR1.07 95%CI 1.06-1.08).

Although the literature concerning air pollution is largely PM_2.5 focused; other pollutants such as NO₂ have been explored. NO₂ exposure has been associated with an increased risk of Parkinson’s disease (Jo et al. 2021) and in the beforementioned Chen et al. (2017a) study NO₂ exposure was associated with an increased risk of dementia (HR 1.1 95%CI 1.08-1.12). Studies have also found significant associations between NO₂ and amyloid deposition (Alemany et al. 2021), one of the most common pathological marks of Alzheimer’s disease (Duran-Aniotz et al. 2021). Alemany et al. also found associations between PM_2.5 and amyloid deposition, as well as PM_2.5 and PM₁₀ with cerebrospinal fluid neurofilament light, a neurofilament biomarker of neurodegeneration. An interesting study by Lin et al. compared air pollution components NO₂, PM₁₀, S0₂, ozone, and carbon monoxide between patients with Alzheimer’s disease in two Taiwanese cities (Lin et al. 2022). Patients with Alzheimer’s disease living in the city with higher pollutant exposure were shown to be experiencing faster cognitive deterioration. This would suggest air pollutants may also influence the rate of neurodegenerative progression.

Overall, these studies reflect how air pollution is evidenced to impact the risk of cognitive impairment and neurodegenerative disease and is considered a risk factor for dementia (Livingston et al. 2020). However, there is no universal agreement on how pollutants disrupt specific biological pathways to contribute to disease pathology. Genotoxicity from oxidative stress and inflammatory induction (Calderón-Garcidueñas et al. 2020, Dumax-Vorzet et al. 2015, Lee et al. 2023, Pandics et al. 2023) are commonly agreed general mechanisms which need further experimental exploration.

Water pollution

A further concern for urban populations is usage or consumption of contaminated water supplies. Contaminated water supplies contain toxic levels of heavy metals from contaminated soil (Ahmad et al. 2021) or street run-off containing traffic-related pollutants (Müller et al. 2020). High usage of plastics has resulted in high levels of bisphenol A being present in our global water supply (Thompson et al. 2009). Bisphenol A is a xenoestrogen which despite having a low affinity with oestrogen receptors, has been implicated in cognitive decline from long-term exposure and consumption (Suresh et al. 2022). This is suspected to be due to the role of oestrogen receptors in cognitive functions such as verbal memory (Hara et al. 2015) being disrupted. Recently, there has been emerging evidence on the impact of water quality on the brain (Bondy and Campbell 2017) and cognitive impairment risk (Liu et al. 2017, Pan et al. 2022a). Bondy and Campbell (2017), specifically analyse the neurotoxicity of metals such as lead and copper, two elements abundant in brake wear emissions which can infiltrate water supplies via urban street run-off (Grigoratos and Martini 2015).

Early life exposure to lead is suspected to be involved in genomic imprinting which impacts the expression of Alzheimer’s related genes in later life, contributing to neurodegenerative risk (Lahiri and Maloney 2010). A case–control study investigating blood lead levels found that levels were higher in those with Alzheimer’s disease (OR 1.05 95%CI 1.01-1.09) in comparison to controls (Fathabadi et al. 2018). This study did not specifically investigate sources of lead exposure but did acknowledge that accumulative lead exposure over the life course does seem to associate with higher neurodegenerative risk. The pathogenic nature of lead in terms of Alzheimer’s disease likely involves neuroinflammation (Adekomi 2017) and encouraging amyloid deposition (Zhou et al. 2018). The study did not consider co-exposure of elements on the impact of Alzheimer’s which would be desirable. In a study by Pan et al. (2022b), a sample of the China Health and Retirement Longitudinal Study (CHARLS) was used to investigate the relationship between water quality and cognitive function in the older adult Chinese population. Higher water quality was associated with higher scores in mental status (relating to orientation and numerical ability), episodic memory and global cognition. This finding is consistent with a Portuguese study examining the association between heavy metals referred to as potentially toxic elements (PTE) and cognitive function (Cabral Pinto et al. 2018). Higher urinary PTE levels were associated with cognitive impairment. The groundwater of the study area was examined for PTE which showed elevated levels of aluminium, iron, manganese and zinc. These elements are main contributors to both water and soil pollution.

Soil pollution

Air pollution impacts not just the quality of water but also the quality of and composition of the soil. Airborne pollutants are suspended at the soil level, resulting in their penetration and infiltration into the soil. Street run-off can further relocate air and traffic-related pollutants from the urban streets to surrounding soil. The pollutant levels in the soil can also directly impact water quality, as groundwater streams become infiltrated with pollutants. Human exposure to soil pollutants can be through inhalation, ingestion or through dermal (skin) contact (Marini et al. 2021, Lupolt et al. 2022). A systematic review by Killin et al. (2016) investigated the association between environmental exposures including soil and water pollutants and dementia risk. The outcomes of studies were conflicting, showcasing the necessity for further research. In Shen et al., a study included in this review, higher levels of aluminium were associated with reduced risk of dementia (comparing highest to lowest exposure RR 0.267 95%CI 0.265-0.268) (Shen et al. 2014). This is conflicting to a meta-analysis (2016) of studies (n = 8) which suggested that individuals chronically exposed to aluminium were 71% more likely to develop Alzheimer’s disease (Wang et al. 2016). No publication bias was observed amongst these eight studies and there was low heterogeneity. A more recent (2021) meta-analysis (n = 23) confirmed this association of high aluminium exposure and cognitive impairment specifically with declines in global cognition scores, working memory and processing speed (Bagepally et al. 2021). In this analysis, however, heterogeneity and publication bias were significant. Aluminium is a neurotoxicant which is implicated in the promotion of amyloid deposition (Zhao et al. 2014). This metal has been shown to alter the expression of genes involved in the formation of amyloid and neurofibrillary tangles, a further pathological mark of Alzheimer’s disease (Duran-Aniotz et al. 2021) as well as decreasing expression of neurotrophins (Kawahara and Kato-Negishi 2011, He et al. 2021). Heavy metals are generally believed to contribute to cognitive impairment and neurodegenerative risk through inducing oxidative stress (Poprac et al. 2017) and altering gene expression (Lahiri et al. 2009, Bakulski et al. 2020).

Noise pollution

With much industrial activity, road traffic and high usage of airports in large cities, the urban area is subject to noise pollution (de Paiva Vianna et al. 2015). In a study by Chen et al. (2017b), traffic-related noise pollution was associated with increased incidence of dementia. The relationship was strongest in urban residents, especially those within major cities and this may be mediated by traffic-related air pollution. This paper by Chen et al. was included in a systematic review and meta-analysis which aimed to investigate the relationship between chronic noise exposure and the risk of dementia (Meng et al. 2022). The overall finding from this review was a significant positive association between the two variables, chronic noise exposure was seen to increase the risk of dementia (>57 decibels. RR 1.47 95%CI 1.21-1.78). However, heterogeneity was observed between the nine included studies in the meta-analysis (I² = 78.7%) and publication bias analysis was absent. The low number of included studies (quantitative synthesis = 9) within the review by Meng et al. (2022) highlights the limited study with regards to noise pollution and neurodegenerative risk. Despite the suspected relationship, there are few human studies in this area. Animal experimentation has suggested that chronic high-level noise exposure promotes tau hyperphosphorylation and amyloid beta overproduction (Su et al. 2018). Chronic noise exposure is also associated with sleep disturbance (Le et al. 2017) which may increase the risk of dementia. Noise has been evidence to evoke a neuronal stress response in which oxidative stress is a downstream consequence (Daiber et al. 2020), a large contributor to neurodegenerative pathophysiology. A study in rats by Cui et al. (2012) suggests that noise induced tau phosphorylation may be partially reversible in the absence of chronic exposure, a beneficial insight for public health personnel who can encourage the incorporation of noise prevention into urban policy.

Light pollution

Light pollution, in particular outdoor light at night (OLAN), is a further issue in urban areas, as artificial light is used extensively in cities after sunset. These high levels of light can disrupt circadian rhythms, which may have numerous consequences as hormone levels, cognitive function, and several other homeostatic mechanisms, are modulated by the circadian clock (Xu et al. 2021). Compromised homeostasis makes the body vulnerable to disease induction (Escobar et al. 2011) from further exogenous or endogenous stressors. The circadian rhythm disruption itself can also contribute to multiple neurological issues such as sleep disorders (contributed to by noise pollution), depression and anxiety (Walker et al. 2020), which could serve as a predeterminant to future cognitive impairment and neurodegenerative disease. Sleep disorders and ageing are both associated with decreased pineal melatonin production (Hardeland 2012, Chen et al. 2022). Decreased melatonin production at night is associated with neurodegenerative diseases including Alzheimer’s (Nous et al. 2021), likely impacting neurodegenerative risk due to its neuroprotective properties (Alghamdi 2018). Pineal melatonin has been described as anti-amyloidogenic (Shukla et al. 2017), and has been evidenced to encourage neuroplasticity (Stefanova et al. 2015) and reduce oxidative stress in Alzheimer’s disease (Shukla et al. 2017). There is limited literature on the effect of OLAN on cognitive impairment and neurodegenerative disease risk. A recent study of veterans in China found that long-term exposure to OLAN was associated with increased risk of mild cognitive impairment (OR 1.44 95%CI 1.36-1.52), considered a transitional state before dementia (Chen et al. 2022). A further study in Italy found that high exposure to OLAN was associated with an increased risk of late-onset dementia (OR 3.5 95%CI 0.32-38.87) (Mazzoleni et al. 2023).

Urban green and blue space

Lack of good-quality natural environments including green and blue spaces can be an issue in urban areas. Urban green space includes public parks, gardens, greenways, and other types of open and recreational spaces that are present in urban areas. Similarly, blue space describes natural areas of water within the urban area, such as a river. These features serve to positively impact the health of urban populations through their perceived benefits, for example, through encouragement of physical and social activity (Beyer et al. 2014, Besser et al. 2021). Physical activity is associated with reduced risk of dementia, including Alzheimer’s disease (Iso-Markku et al. 2022), with one of the suggested reasons for this being that it reduces the risk of future cardiometabolic disorders associated with Alzheimer’s disease risk (Tai et al. 2022). Social activity can encourage neural stimulation, increasing cognitive reserve which is considered protective of Alzheimer’s disease (Liu et al. 2019). There are also more specific neurological benefits of being within the natural landscape, such as decreased sympathetic nerve activity (Mueller et al. 2022). A recent cohort study has found associations between having a ‘healthy lifestyle’ (including engaging in physical and social activity and cognitive stimulation) with reduced risk of memory decline in both APOE4 carriers and non-carriers (Jia et al. 2023). This study had a 10-year follow-up period, allowing memory loss to be examined over a substantial time period; however, lifestyle behaviour measurements were self-assessed, potentially decreasing reliability of study results.

Urban green and blue spaces have been suggested to decrease risk of cognitive impairment highlighted by the results of intervention studies such as Gidlow et al. (2016b). In this study, walking within green and blue spaces resulted in improvements in cognitive function when compared to walking in an urban environment. Cognitive improvements were most significant after walking in blue space. Studies have also been carried out to investigate residential proximity to these green and blue spaces and cognitive measures, with these studies suggesting a reduction in the risk of cognitive impairment when urban green and blue spaces are present close to residence (Sylvers et al. 2022, Rodriguez-Loureiro et al. 2022). However, due to high levels of within-study and between-study variations, results from studies in this area have provided conflicting evidence. Wu et al. investigated the association between residential distance to an urban park and risk of dementia finding that living further from a park was associated with lower odds of dementia (OR 0.64 95%CI 0.42-0.95) (Wu et al. 2020). There is also a smaller amount of literature concerning blue space and its relationship with cognitive measures (White et al. 2020). Considering the potential for urban green and blue space to reduce the risk of cognitive impairment, issues can arise in urban areas if these spaces are of limited presence or are not seen as safe areas for people to spend time in, due to associations with crime. The accessibility of urban green and blue spaces is also important if they are to be used by all individuals in the population despite their age or disability status. In the creation of healthy cities, urban green and blue space development and maintenance will be integral for improving environmental health through mitigating temperature rises (Hasan et al. 2022) and pollution levels (Junior et al. 2022) and will also have impacts on population health including enhancing the walkability of the city (Zuniga-Teran et al. 2019, Zhang et al. 2020).

Walkability and cyclability

Neighbourhood walkability, the ability to walk to desired locations without a car, is an important characteristic of the urban area. The walkability of the urban area will depend partly on urban green spaces, as well other features such as proximity to and density of roads, the number of pedestrian walkways and crossings (Knapskog et al. 2019). Limited neighbourhood walkability can lead to issues in terms of physical inactivity of urban residents and increased usage of private transport, exacerbating the issue of traffic-related air pollution in urban areas. Neighbourhood walkability has been associated with improved global cognitive scores and attention psychomotor speed at baseline in a recent study (Timmermans et al. 2023), especially in individuals with vascular cognitive impairment. However, these associations were not found to be significant in the subsequent 2-year follow-up period. Another study investigating access to amenities through neighbourhood walkability found no significant association between walkability and cognition (Katayama et al. 2020). This study did find other interesting significant associations, as study participants engaging in a more active lifestyle were less likely to have mild or global cognitive impairment. Neighbourhood walkability was hypothesised to be protective of this association.

A high degree of cyclability in a city is beneficial in numerous ways including traffic-related emissions (Scorza and Fortunato 2021) and noise accumulation. Cycling also impacts an individual’s physical activity which is one of the modifiable risk factors for cognitive impairment and neurodegenerative disease (Iso-Markku et al. 2022). A study by Wittfield et al. (2020) found that increased cardiorespiratory fitness (achieved through walking, cycling and running) was associated with higher grey matter volumes in specified brain regions. Grey matter atrophy is a common occurrence in neurodegenerative diseases such as Alzheimer’s due to neuronal loss (Wu et al. 2021). Cycling may be an activity which could be encouraged to protect from brain atrophy. However, Williams et al. found that cortical volumes were dependent on a further factor, mean diffusivity, when assessing if higher cortical volume was associated with higher or lower longitudinal memory performance (Williams et al. 2023). Further study is therefore required with a more complex assessment of brain volumes to identify if physical activity through cycling for example can protect from brain atrophy. Current issues in cities resulting in poor cyclability include absent or inadequate cycling infrastructure (Iwińska et al. 2018, Caicedo et al. 2021). Addressing these issues will work to improve the health of cities and potentially protect residents of the city from future cognitive impairment and neurodegenerative disease.

As urban environment factors and associated health behaviours, including physical and social activity discussed above, have been associated with cognitive impairment and neurodegenerative disease in the literature, it is of interest to explore the potential biological mediators of these associations. The epigenome is a modifiable and measurable mechanism which could provide insight to the biological pathways implicated in how urban environment factors of interest impact the risk of cognitive impairment and neurodegenerative disease.

The urban environment and epigenetic influence

The epigenome helps regulate gene expression to enable required cellular functions and to help people respond to environmental stimuli (Gibney and Nolan 2010). The epigenome is modified throughout the human lifespan to regulate development (Skinner 2011) and ageing (Wang et al. 2022). These modifications are commonly reversible and induced by environmental factors (Kanherkar et al. 2014). Epigenetic modifications include DNA methylation (DNAm), histone modification, and non-coding RNAs. DNAm commonly occurs when a methyl group is transferred to a cytosine of a CpG site of DNA (Zeng et al. 2022), while histones are structural proteins with roles in nucleosome remodelling and chromatin structure (Millán-Zambrano et al. 2022). Non-coding RNAs such as microRNAs and long-ncRNAs can also alter the epigenome through post translational gene regulation and influence the level of DNAm and histone modification (Panni et al. 2020).

The epigenome is modifiable through exogenous influences, including an individual’s surrounding environment. Urban environment factors have been evidenced to modify the epigenome (Skinner 2011, Wang et al. 2022), and epigenetic modifications are known to be cognitively influential and be implicated in neurodegenerative disease (Hwang et al. 2017, Zhang et al. 2022a). It is of interest to identify if a biological link exists between urban environment factors and neurodegenerative risk, via epigenetic modification. There is a small amount of research dedicated to the investigation of this area. Focusing research on this link is beneficial considering the potential for the epigenome to be modified and epigenetic modifications to potentially be reversed (Gibney and Nolan 2010). It is hoped that the dynamic nature of the epigenome (Cazaly et al. 2019) will allow us to alter exposure to influential environmental factors, to reverse epigenetic modifications which increase risk of disease or alternatively, encourage protective modifications to alleviate neurodegenerative risk. To enable beneficial alterations of environmental exposures, further depth of knowledge is required with regards to the consequences of the urban environment on the epigenome. Developing health recommendations backed up by this additional biological research would potentially encourage policymakers to take action and create interventions to enhance environmental and population health, thereby reducing the risk of neurodegenerative disease.

Identifying epigenetic modifications which occur as a result of environmental exposures, and exploring how these modifications are involved in the causation of neurodegenerative disease, requires complex analytical techniques. Epigenetic analysis mainly involves epigenetic clock analysis and epigenome-wide association studies (EWAS), additionally utilising Mendelian randomization (MR) as a technique to assess causation. Harnessing these techniques will be greatly beneficial in future research, with the aim of finding biological links between urban environment factors and disease. These techniques aid the identification and visualisation of epigenetic markers which may be implicated in causal pathways and are crucial to begin developing further biological pathway knowledge.

Epigenetic analytical techniques

Current literature has identified epigenetic markers associated with urban environment factors, helping explain the underlying biological pathways by which these environment factors influence disease risk. The majority of research in this area has focused on DNAm, with a particular focus on epigenetic clocks and EWAS.

Epigenetic clocks

Epigenetic clocks are often used to capture accelerated biological ageing, predicting risk of morbidity and premature mortality (Hannum et al. 2013, Levine et al. 2015, 2018, Horvath and Raj 2018, Bell et al. 2019, Noroozi et al. 2021). Epigenetic clocks use patterns of CpG methylation, where a methyl group is placed in the genomic region of a cytosine followed by a guanine nucleotide, to predict biological age (Li et al. 2022). When biological age exceeds that of an individual’s chronological age, this indicates biological age acceleration has occurred (Jain et al. 2022). Biological ageing is linked to cellular senescence (Yang and Sen 2018), a process in which cells exit the cell cycle and no longer have proliferative functions (Di Micco et al. 2021). High numbers of senescent cells can disrupt homeostasis and leave the body vulnerable to disease induction, a potential pathway as to how epigenetic modification influences disease risk (Crouch et al. 2022).

Current findings utilising epigenetic clocks

Studies have found that accelerated biological age is associated with cognitive impairment (Levine et al. 2015), Parkinson’s disease age of onset (Tang et al. 2022) and dementia (Wu et al. 2021). However, many additional studies have failed to find significant associations between accelerated biological age and neurodegenerative disease. A detailed review concerning studies investigating biological age acceleration and dementia has been carried out by Zhou et al. (2022). Studies utilising epigenetic clocks have associated urban environment factors with biological ageing. A recent study (Yannatos et al. 2022) focused on identifying racial differences in DNA methylation age with socioeconomic status and air pollution exposure in a large US cohort. Black participants were exposed to greater levels of air pollution than White participants, with lower socioeconomic status and education level partially explaining the accelerated biological age observed in Black participants for Dunedin Pace of Ageing; however, it was hypothesised that air pollution, specifically PM_2.5 exposure, was also a contributor. PM_2.5 components have been associated with biological ageing in other studies (White et al. 2019, Wang et al. 2020, Gao et al. 2022), suggesting this is one of the mechanisms which mediate air pollution related ill health and potentially neurodegenerative risk.

Important considerations regarding epigenetic clocks

Multiple generations and types of epigenetic clocks have been developed. First-generation clocks Hannum and Horvath were designed to be predictors of chronological age (Hannum et al. 2013, Horvath 2013) however, second-generation clocks PhenoAge and GrimAge were trained with the inclusion of more clinical-based markers such as C-reactive protein (Lu et al. 2019). Second-generation clocks are considered better predictors of mortality and morbidity (Noroozi et al. 2021, McCrory et al. 2021) Third-generation clocks consider longitudinal and cross-species data (Belsky et al. 2020, Lu et al. 2023). There is substantial transcriptional overlap for some clocks, for example in the dorsolateral prefrontal cortex with Horvath1, Horvath2, Levine, Hannum, and Lin (Liu et al. 2020). Considering differences between clock generations and individual clock formations, clocks demonstrate different associations with underlying biological pathways. Utilising appropriate clocks in future urban environmental-neurogenerative disease research will be important. Focusing analyses using second- and third-generation clocks may be more reliable in the prediction of neurodegenerative risk. Further exploration of what type of underlying pathological pathways each clock may best associate with will be additionally helpful.

Epigenome-wide association studies

EWAS primarily analyse DNAm levels across the genome with single-site resolution (Campagna et al. 2021), taking either a cross-sectional or longitudinal approach. EWAS can identify epigenetic markers, epigenetic modifications which are associated with a statistically significant higher risk of disease. Alternatively, EWAS can uncover markers of resilience, epigenetic modifications associated with statistically significant lower risk of disease. These resilience markers are of considerable interest as they indicate a protective mechanism which could be explored in preventative policy of disease risk (Rakyan et al. 2011).

Current findings in epigenome-wide association studies

Neurodegenerative disease focused EWAS have located differentially methylated loci in cortex derived DNA (Dashtipour et al. 2017, Li et al. 2020, Smith et al. 2021) which serve as markers for the diseases. EWAS have also been carried out with a specified focus on urban environment factors. Long term exposure to air pollution, including PM and NO₂, has been associated with differentially methylated genes and regions (Lee et al. 2019, Chi et al. 2022). Chi et al., for example, found that PM_2.5 exposure was associated with cg05926640 (false discovery rate (FDR) <0.05) in monocytes (Chi et al. 2022), with this CpG site being commonly associated with the TOMM20 protein-coding gene (Translocase of Outer Mitochondrial Membrane 20). The TOM20 protein synthesised from TOMM20 transcription has been implicated in Parkinson’s disease through alteration of mitochondrial protein import (Li et al. 2021). Noise exposure related to different modes of transport and air pollution was investigated in a study by Eze et al. (2020), who found that PM_2.5 increased methylation of a FARS2 gene region (Phenylalanyl-TRNA Synthetase 2). The FARS2 gene has been implicated in methylation-related changes in Multiple Sclerosis lesions (Schirmer et al. 2019, Kular and Jagodic 2020). Some EWAS complete additional pathway-based enrichment analysis which gives an indicator of functional biological pathways which may be influenced by modifications. In Lee et al., pathway analysis highlighted enrichment in inflammatory and cardiovascular pathways from PM_2.5 and NO₂ exposure (Lee et al. 2019).

Other urban factors which have been examined through EWAS include greenness. Residential greenness measured using the normalised difference vegetation index and enhanced vegetation index was associated with differential methylation in a study by Xu et al. (2021). A significant association was found for the CNP gene promoter (2‘,3’-Cyclic Nucleotide 3’ Phosphodiesterase), a gene previously considered to play a role in Multiple Sclerosis (Parade et al. 2017, Kao et al. 2020). A separate study (Alfano et al. 2023) analysing maternal exposure to green space and the methylation levels in umbilical cord blood found associations with green space and the methylation of genes including HTR2A (5-Hydroxytryptamine Receptor 2A), implicated in major depressive disorder (Parade et al. 2017, Kao et al. 2020).

Issues to overcome to maximise epigenome-wide association study usage in urban environmental-neurodegenerative research

Authors of EWAS concerning urban environment factors are in universal agreement that more studies are required in this field, including multiple validation of findings and further exploration of associations and biological pathways (Mostafavi et al. 2018, Chi et al. 2022, Freydenzon et al. 2022). EWAS results are most convincing when the studies are performed robustly using adequately powered cohorts and including replication of top-ranked findings. However, EWAS projects are expensive and, as they are relatively newly employed in this field, truly independent datasets that explore the same variables are rarely available for replication. Advancing the available data to perform EWAS will be highly beneficial to the field, enabling the identification and validation of epigenetic markers which can be further used in causative analysis. EWAS focused on exploring the neurodegenerative disease outcome as a consequence of urban environmental exposures are required.

To minimize false positives from future large-scale EWAS, appropriate adjustment for multiple testing is essential, ideally followed by validation or replication from independent cohorts. Both single site resolution and combined estimators/composite biomarkers should be considered for the phenotype and environmental exposures.

Harnessing Mendelian randomization to explore causation in the pathway to disease

While a person’s genetic risk of disease is largely fixed at conception, due to the modifiable nature of epigenetic changes, their epigenetic risk score alters across the life course (Relton and Smith 2015). Challenges exist in determining if epigenetic changes are a cause or consequence of disease; however, strategies have been developed to tease this out through functional studies or MR (Rakyan et al. 2011, Paul and Beck 2014).

MR has been adopted as an epidemiological research technique, effectively acting as in silico randomised control trials (RCTs). RCTs are a gold standard research technique (Cartwright 2009) but can be unethical and impractical for the requirements of epidemiological study. Observational studies on the other hand have the limitation of being unable to measure all possible confounders. MR overcomes this problem through the use of genetic variants (single nucleotide polymorphisms) as proxies (instrumental variables) for specific exposures. Because genetic variants are defined at conception, they are independent of confounders, allowing the investigation of causation.

For analysis concerning epigenetic modifications such as DNAm, a methylation site cannot be used directly as an instrumental variable in the MR process, as they are modified throughout the life course and therefore have potential to be influenced by confounders. Genetic variants significantly associated with certain DNAm changes, which occur due to specific environmental exposures, are therefore used as proxies to determine the causal effect of these risk-associated DNAm changes on disease outcomes (Smith and Ebrahim 2003, Thanassoulis and O’Donnell 2009) (Figure 1). For a variant to be used as an instrumental variable within MR, it is assumed that the variant only has a relationship with the disease outcome through its relationship with the risk factor (Didelez and Sheehan 2007). MR can provide strong statistical evidence for causal relationships, which can then be explored in biological pathway research to identify how urban environment factors are influencing the epigenome to contribute to neurodegenerative risk.

Figure 1. Directed acyclic graph to explain how causative investigation through Mendelian randomization could be used in urban environment-neurodegenerative research.

Note: Instrumental variables Z (single nucleotide polymorphism) are used as proxies for risk factor X (a DNA methylation site which is associated with an environmental exposure, for example PM_2.5) which is hypothesised to have a causal relationship with the disease outcome Y (a neurodegenerative disease such as Alzheimer’s disease). The ZX and ZY relationships will be used to ultimately calculate the extent to which X causes Y. U represents the unknown confounding factors and C represents known confounding factors in the XY relationship (such as sex, age, weight and physical activity). The instrumental variables Z are required to 1) have a relevant relationship with X, 2) only have a relationship with the outcome Y through X and 3) not be associated with U/C.

A statistically significant result from this Mendelian randomization analysis would indicate that a genetic variant associated with a specific DNA methylation change (a consequence of PM_2.5 exposure), is causative in neurodegenerative disease. Providing statistical evidence of this relationship not only consolidates a causative link, but also provides an encouraging base for biological pathway research to explore how PM_2.5 exposure is modifying the epigenome and how this is impacting on neurodegenerative risk and pathophysiology.

Multiple variations on the traditional MR approach have been developed. Bidirectional MR evaluates the direction of causation (Zheng et al. 2017) through performing MR in both the forward and reverse direction of the acyclic graph. Two-sample MR has been developed which allows the risk factor and outcome instruments to be taken from two different sample populations (Pierce and Burgess 2013) in both GWAS and EWAS meta-consortia (Burgess et al. 2015, Walker et al. 2019). A recent study aimed to test the causal effect of alcohol-related DNAm on colon cancer risk (Zhou et al. 2022). This study used a two-sample MR approach by selecting genetic instruments of predisposition to alcohol drinking from previously published GWAS and assessing the relationship of these variants with colon cancer risk using separate previously published GWAS.

One of the difficulties faced by researchers using MR is acquiring appropriate genetic variants (used as instrumental variable Z) for CpG sites considering methylation at CpG sites is altered by both underlying genetics and the environment (Hannon et al. 2018). To enable causal links to be assessed, it is important that genetic variants account for enough variability at CpG sites of interest and have an F statistic > 10, considered acceptable instrument strength (Burgess et al. 2011).

For urban environmental-neurodegenerative research, MR is a useful technique as the downstream consequences (epigenetic modifications) of urban environment factors can be assessed. Assessing causative relationships between methylation changes and neurodegenerative disease will enable the identification of environmental exposures which can be modified.

Future research

There is a growing need for identifying urban environmental–neurodegenerative relationships. Recent analyses have independently associated epigenetic modifications and urban environment factors with neurodegenerative diseases. This review highlights the need to integrate both these variables in analysis. Through identifying relationships between urban environmental exposures and epigenetic modifications, it will then be possible to expand on biological pathways, which may have cognitive consequences. This will enable opportunities to collaborate with those in environmental health policymaking to alleviate molecular risk of disease.

Using epigenetic analytical techniques to analyse causal relationships

Combining EWAS and MR enables causative relationships to be identified between urban environment factors and neurodegenerative diseases. Identifying these relationships is an important starting point. In other fields, MR studies have made use of genetic variants to investigate disease causation related to epigenetics (Dekkers et al. 2016). It is rare that urban environment exposures and their influence on neurodegenerative diseases are explored in this way. A multi-omic approach, exploring genetic and epigenetic variation, as well as harnessing methods such as MR, will be greatly beneficial in beginning to understand the complex interactions to help translate research into preventative policy. It would be beneficial for future work to identify epigenetic markers through EWAS analysis associated with important urban environment factors and also consider co-exposure. These epigenetic markers enable MR analyses to explore complex causal associations which can be interpreted with appropriate caution.

Epigenetic clocks are useful in the prediction of biological age which has been associated with neurodegenerative risk (Gonzales et al. 2022). Formulating clocks traditionally utilises blood samples; however, recently, the Cortical clock (Shireby et al. 2020) has been developed using human post-mortem cortical tissue as opposed to the traditional systemic blood samples. The Cortical clock identified 347 DNAm sites which can predict age in the human cortex (Shireby et al. 2020) and may have higher sensitivity to specific brain pathological epigenetic changes (Grodstein et al. 2021, Gonzales et al. 2022). Further validation and replication of the Cortical clocks sensitivity could make it a prominent feature of biological and specifically cortical ageing. Expanding work on both cortical and traditional blood made epigenetic clocks will aid in identifying underlying epigenetic mechanisms of ageing in the brain, providing useful data in which urban environment methylation-related changes can be linked.

Disentangling complex interactions

It is important that future research does not solely focus on more commonly researched urban environment factors. For example, there is a larger volume of research including EWAS centred around air pollution, commonly PM_2.5, while the literature on noise and light pollution is sparse. Moreover, from our understanding, there are very limited published EWAS concerning noise or light pollution using human samples. This limits the knowledge within the field on the potential impact of these exposures at the epigenetic level. The inclusion of these lesser researched factors will aid in filling in the bigger picture as to how the whole urban environment impacts disease risk. The inclusion of more recently implicated and novel urban environmental pollutants, including microplastics (Meng et al. 2020) and particulates as a consequence of wildfires (Li et al. 2023), would further bridge the knowledge gap. Additionally, current research commonly investigates a singular factor, such as PM or chronic noise exposure, and the impact on neurodegenerative disease. These studies are important in building an understanding of how these individual factors influence disease risk. However, urban environmental exposure is multifactorial, which can impact both the epigenome and neurodegenerative risk simultaneously. Despite the added complexity, it would be beneficial for future research to begin the investigation of co-exposures.

Additionally, social factors such as socioeconomic status and education can confound and mediate relationships between urban environment factors and disease (Clifford et al. 2016, Peters et al. 2019, Hajat et al. 2021). Social factors have also been associated with similar epigenetic mechanisms to urban environment factors such as accelerated biological ageing (Fiorito et al. 2017). Researchers should consider this overlap when interpreting their results and always consider the relationship between environmental and social factors.

Disentangling the interactions as to how environmental exposures interact with each other and other social factors to influence the modification of the epigenome, and ultimately impact neurodegenerative disease susceptibly, will be a complex future goal in the field.

Biological pathways

Biological pathways tract the body’s cellular progression from exposure to disease states. They often involve complex molecular mechanisms including epigenetic modifications (McGowan and Roth 2015, Alameda et al. 2022). Few papers at present have focused on exploring the biological pathways which explain the link between the urban environment and neurodegenerative disease through epigenetics. This lack of research has left multiple unanswered questions and hypotheses. Through reviewing the literature, a summary of the proposed biological pathways can be seen in Figures 2 and 3.

Figure 2. The biological pathways involving epigenetic mechanisms which link urban environment pollutants and neurodegenerative disease.

Note: This graph summarises the connections and interactions between biological pathways elucidated in the literature. Air, water and soil pollutants have been evidenced to modify epigenetic machinery and contribute to oxidative stress/inflammatory induction in the body. Air pollution can exacerbate the pollution levels of urban soils and water as discussed in this review, contributing to this. Noise and light pollution have been suggested to contribute to sleep disruption which collaterally has been suggested to also impact on the regulation of epigenetic machinery and inflammatory induction. Noise pollution has also been suggested to directly contribute to oxidative stress/inflammatory induction.

Oxidative stress has been evidenced to alter the activity of epigenetic machinery. Epigenetic machinery such as DNA methyltransferases modify the epigenome. Epigenetic modification over the lifetime has been evidenced to contribute to both cellular senescence and the alteration of the expression of genes implicated in neurodegenerative diseases, potentially contributing to the underlying molecular risk of cognitive impairment or neurodegenerative disease.

Current research has provided suggestions and some evidence for the discussed connections and interactions; however, research has not specifically investigated whole biological pathways providing concrete evidence for epigenetic mechanisms linking urban environment pollutants and neurodegenerative disease.

Figure 3. Resilience pathways involving epigenetic mechanisms which link urban environment factors and associated behaviours with neurodegenerative disease.

Note: This graph summarises the connections and interactions between urban environment factors and associated behaviours, and details how they potentially contribute to biological resilience through epigenetic mechanisms. Urban green and blue space, as discussed in this review, can contribute to walkability and cyclability, two aspects of active travel. Urban green and blue space and active travel have associations with numerous health impacting behaviours and factors, including decreasing traffic-related air pollution and increasing physical and social activity. Increased physical and social activity have been suggested to improve both mental health and well-being, and contribute to lower stress. Lower stress has further been evidenced to improve both mental health and well-being. Urban green and blue spaces are areas which have been suggested to directly impact on mental health and well-being through a variety of mechanisms, including reducing rumination or stress.

Air pollution can influence epigenetic machinery, as discussed in this review, and it has also been suggested that stress levels, as well as the presence of mental health conditions, may influence this machinery. Epigenetic machinery can modify the epigenome, altering the expression of genes. It is hypothesised that through changing health behaviours, such as lowering traffic-related pollution exposure, reducing stress and improving mental health, may reduce the risk of neurodegenerative disease through providing biological resilience. Alternatively, as epigenetic modifications are believed to be reversible, and urban green and blue space, active travel and other health associated behaviours discussed have been evidenced to improve health outcomes including cognitive outcomes, this may aid in reversing epigenetic modifications contributing to pathology. Reversing epigenetic modifications may both slow the biological ageing process and alter the expression of genes traditionally implicated in neurodegenerative disease to decrease the risk of these diseases.

Current research is majorly yet to explore the potential of biological resilience relating to changing urban environment exposures to reduce the risk of and prevent neurodegenerative disease.

Pathways involving urban environment pollutants

Urban pollutants can influence the activity of the enzymes involved in epigenetic modification (epigenetic machinery) including DNA methyltransferase (DNMT). PM has been evidenced to decrease DNMT expression, facilitating hypomethylation in skin keratinocytes (Ryu et al. 2019). PM has also been associated with hypomethylation via DNMT3B down regulation in blood (Wang et al. 2020). Histone post-translational modifying enzymes have been associated with air pollution exposure in post-mortem brain tissue of urban residents with pathological hallmarks of Alzheimer’s disease. Global expression of H3K9me2 and H3K9me3 (involved in histone methylation) was decreased in those with pathological hallmarks of the disease (neurofibrillary tangles and amyloid deposits) when compared to the controls (Calderón-Garcidueñas et al. 2020). Heavy metals implicated as soil and water pollutants have also been evidenced to alter DNTM expression (Ryu et al. 2015). For example, Cadmium has been shown to facilitate DNA hypomethylation and hypomethylation through altering the activity of DNMT (Takiguchi et al. 2003, Ryu et al. 2015, Athanasopoulos et al. 2016).

The number of studies investigating the influence of environment factors on epigenetic machinery is not extensive and still leaves questions unanswered. It is not clear how or why pollutants target these specific enzymes, nor how many enzymes pollutants interact with, subsequently altering their activity. The influence of urban environment factors beyond air pollutants and heavy metal constitutes of soil and water pollution remains largely unexplored in this context, highlighting an interesting area of future research. An additional research gap in the field is how dysregulation of epigenetic machinery enzymes, due to urban environmental exposures, disrupts gene expression and regulation, and the potential implications on cognitive impairment and neurodegenerative diseases.

Oxidative stress is an additional mechanism which has been evidenced to influence cognitive impairment and neurodegenerative risk which has been summarised in a review by Kandlur et al. (2020). Various pollutants, including air (Hahad et al. 2020, Murata et al. 2022), noise (Daiber et al. 2020), water and soil (Münzel et al. 2022) pollutants, are involved in inflammatory induction resulting in oxidative stress. Indeed, the potential for oxidative stress to itself induce these epigenetic changes is briefly discussed by Kandlur et al. (2020). Oxidative stress has been evidenced to influence DNA methylation and is suspected to alter the activity of epigenetic machinery (Guillaumet-Adkins et al. 2017); however, the literature is lacking on the mechanism(s) by which it specifically contributes to neurodegenerative risk. Further exploration of this pathway will provide much needed insight. Furthermore, oxidative stress has also been evidenced to be associated with both cellular senescence (Holmström and Finkel 2014) and biological ageing (Warraich et al. 2020); processes which are related with progressive epigenetic modification.

Current evidence suggests that epigenetic modifications influence the risk of cognitive impairment and neurodegenerative disease through either accelerated biological ageing (Levine et al. 2015, Wu et al. 2021, Tang et al. 2022), which can be linked to cellular senescence, or through gene expression changes (Berson et al. 2018, Sharma et al. 2020) having downstream molecular consequences which contribute to pathophysiology. Research surrounding which epigenetic modifications are consistently involved in neurodegenerative pathophysiology and specific outlines of molecular cascades affected by these modifications for example, is currently not comprehensive enough for sound understanding (Landgrave-Gómez et al. 2015, Delgado-Morales et al. 2017).

Other proposed biological pathways may involve circadian and sleep disruption, a consequence of noise and light pollution (Halperin 2014, Cho et al. 2015, Patel 2019). The epigenome has the ability to influence circadian cycles (Hudec et al. 2020), while sleep deprivation is suggested to be epigenetically modifying and have implications on epigenetic ageing, summarised in a review by Gaine et al. (2018). This review provides several examples of the implications of sleep deprivation on neurodegenerative diseases, including decreased histone acetylation observed in a Drosophila model of Alzheimer’s disease (Pirooznia et al. 2012, Gaine et al. 2018). Additionally, sleep deprivation has been evidenced to contribute to oxidative stress and inflammatory induction (Villafuerte et al. 2015, Kempf et al. 2019, Daiber et al. 2020). Current evidence has not been able to provide a comprehensive understanding of the impact of noise and light pollution on the human epigenome. Studying the impact of sleep deprivation using human epigenetic profiles provides an opportunity to deepen our understanding of this potential pathway.

Resilience pathways

Urban environment factors also have the capacity to operate in biological pathways in which epigenetic reversal and biological resilience may decrease the risk of cognitive impairment and neurodegenerative disease. Walkability and cyclability contribute to active travel (Fraser and Lock 2011, Kim et al. 2020), thereby increasing the physical activity levels of individuals. Literature has also directly associated urban green and blue space with increased physical activity levels (Richardson et al. 2013, Besser 2021). Physical activity has been shown to impact various aspects of health, namely cardiometabolic health (Rao et al. 2022). With atherosclerotic complications being a risk factor for neurodegenerative diseases, especially vascular dementia (Huang et al. 2021), improving heart and circulatory health has a potentially direct impact on reducing neurodegenerative risk.

Physical activity has also been evidenced to modify the epigenome (Plaza-Diaz et al. 2022). Physical activity has been found to influence the expression of multiple genes which are involved in neurodegenerative risk and pathophysiology (Panes et al. 2022), including PGC-1a (Peroxisome Proliferator-activated Receptor-gamma Coactivator 1 alpha) which is a key regulatory gene in mitochondrial biogenesis (Plaza-Diaz et al. 2022). PGC-1a hypomethylation occurs post-exercise (Rasmussen et al. 2014, Maasar et al. 2021), upregulating transcription. PGC-1a has been implicated in multiple neurodegenerative diseases and its upregulation is considered neuroprotective in neurodegenerative disease (Panes et al. 2022). In Alzheimer’s disease for example, PGC-1a downregulation is a critical risk factor for early onset of disease (Terada et al. 2020, Dong et al. 2020). PGC-1a hypomethylation through physical activity therefore suggests a resilience mechanism. This mechanism likely involves synaptic and neural protection from oxidative stress, although a confirmed pathway is yet to be identified.

Urban green and blue space have been associated with other health enhancing factors including promoting social activity (Maas et al. 2009, Jennings and Bamkole 2019), reducing stress (Roe et al. 2013, Gidlow et al. 2016a, Slawsky et al. 2022), improving mental health and well-being (Gascon et al. 2015, Hannah et al. 2015, Hystad et al. 2019), and mitigating the effects of traffic-related pollution (Gidlöf-Gunnarsson and Öhrström 2007, Dadvand et al. 2012). Health enhancing factors are interconnected, for example, social activity aids in stress reduction (Umberson and Karas Montez 2010); with active travel and physical activity also aiding their promotion (Carek et al. 2011, Cheng et al. 2019, Zukowska et al. 2022). Similar to physical activity, it is suggested that health enhancing factors may contribute to epigenetic modifications through altering epigenetic machinery. Stress (Hing et al. 2018), social and physical activity (Fernandes et al. 2017, Światowy et al. 2021), mental health conditions (Crawford et al. 2018, Wiegand et al. 2021) and pollutant exposure (Gondalia et al. 2019, Chi et al. 2022) have all been associated with differential methylation profiles. Chronic unpredictable stress in animal models for example, has been evidenced to alter DNMT3a activity in the pre-frontal cortex. Wei et al. study found that stressed rats displayed overexpression of DNMT3a, exhibited differential methylation profiles and cognitive changes including impaired memory (Wei et al. 2021). As some health enhancing behaviours can act similarly to discussed urban environmental pollutants, altering epigenetic machinery (Hunter et al. 2019, Rey et al. 2019), there is potential for promotion or reversal of epigenetic modifications from different exposure levels of these behaviours. Increasing physical activity or reducing stress may promote modifications related to a resilience pathway, decreasing the risk of cognitive impairment or neurodegenerative disease. Altering exposure from low to higher physical activity levels may also reverse epigenetic modifications which contribute to cognitive impairment of neurodegenerative risk. Epigenetic reversal has the additional potential benefit of slowing the effect of biological age acceleration, potentially providing a further resilience pathway.

Current research is able to elucidate biological pathways; however, it is not possible to confirm full pathways from environmental exposure to the disease outcome. Few studies have focused on investigating epigenetic modifications as the consequence of urban environmental exposures and the impact of this molecular change on cognitive impairment and neurodegenerative pathophysiology. Overall, current research has not provided a comprehensive understanding of biological resilience and epigenetic reversal in the context of cognitive impairment and neurodegenerative disease. Further work is required to explore how resilience can potentially be promoted and the extent to which epigenetic modifications can be reversed through altering environmental exposures.

An integral future direction in the field is to not only identify epigenetic markers, including those which indicate resilience mechanisms, but also exploring their influence on biological pathways. Multi-omic approaches will be especially important to explore this research area, considering individual urban environment factors likely contribute to multiple molecular risk pathways (Wu et al. 2023). Few studies have used multi-omic approaches, incorporating multiple types of environmental data to understand their interaction and subsequent impact on molecular biology. Maitre et al. recently used multi-omic approaches to explore the early life exposome (Maitre et al. 2022) for example. Utilising this multi-omic approach will create a more comprehensive understanding of whole complex pathways and system interconnections. Translating this molecular knowledge into both urban and public policy interventions has future potention for therapeutic protection from neurodegenerative diseases.

Concluding remarks

Urban environmental exposures have been associated with cognitive impairment and neurodegenerative disease. The biological pathways by which urban environment factors contribute to neurodegenerative disease have been summarised in this review. However, there are many hypothesised pathways that have yet to be tested and confirmed, and many unanswered questions remain. Further molecular analysis is required to tease out exactly how urban environment factors alter underlying molecular cascades to result in the disease outcome. Epigenetics is a research field which can aid the development of biological pathways considering that epigenetic modification contributes to neurodegenerative risk, and urban environment factors can modify the epigenome. Through utilising epigenetic analytical techniques discussed in this review it will be possible to expand research of the environmental consequences on human cognition and neurodegenerative status. It is hoped that developing knowledge of molecular risk from environmental exposures will aid in the creation and development of public and urban policy to prevent cognitive impairment and neurodegenerative disease.

Author contributions

SG: conceptualization, interpretation, writing (original draft, review & editing) and visualization. CH: interpretation, writing (review & editing) and visualization. BM: conceptualization, interpretation, writing (review & editing) and visualization. AJM: conceptualization, interpretation, writing (review & editing) and visualization. RH: conceptualization, interpretation, writing (review & editing) and visualization. All authors have read and approved the manuscript for submission.

Disclosure statement

No potential conflict of interest was reported by the authors.

Additional information

Funding

This work as supported by the Economic and Social Research Council and Innovate UK under the Healthy Ageing Challenge Social, Behavioural and Design Research Programme fund [ES/V016075/1]. CH is supported by funding from NI Health and Social Care Research and Development Division R&D STL/5569/19 and UKRI Medical Research Council [MC_PC_20026]. SG is supported by a PhD studentship from the NI Department for the Economy.

Notes on contributors

Sophie Glover

The SPACE (Supportive Environments for Physical and Social Activity, Healthy Ageing and Cognitive Health) research project at Queen’s University Belfast aims to assess the impact of the urban environment on brain health, namely relationships between the urban environment and mild cognitive impairment. The project has an interest in complex systems research to advance our universal understanding of how physical urban pollutants, physical and social activity, and green environments can impact our brain health as we age. The project has an interest in engaging with the public and policy makers to increase understanding on the topic and ensure policy takes health implications into consideration when creating environments.