Abundance and absence: Human-microbial co-evolution in the Anthropocene

Abstract

Human-microbe relations have undergone a profound shift over the past 100 years. The discovery of antibiotics, increasing levels of pollution, and urban and agricultural intensification have led to the proliferation and diversification of novel resistance genes and microorganisms. This abundance has unfolded against a backdrop of microbial absence that is the other side of the antimicrobial coin; reductions in the quantity and diversity of human-microbe interactions are now registering as epidemics of chronic non-communicable diseases in urban populations. Building from this paradoxical situation of ‘abundance’ and ‘absence’, this article reviews the molecular-genetic, macroscale-infrastructural, and community-ecological aspects of microbial evolution at a time when human actions are a critical force in shaping their directions.

Introduction

In her essay ‘antibiotic resistance and the biology of history’, Landecker (2016) writes that ‘bacterial life today is appearing as a specific instantiation of the biology of the Anthropocene: human efforts to control life’s productivity become the matter of the world’. (p. 23). Recalling Chakrabarty’s (2009) thesis that the Anthropocene spells the end of the distinction between human and natural history, Landecker contends that ‘antibiotic resistance’ too, ‘ruptures assumptions about divisions between human history and natural history’. (p. 22). According to her analysis, microbial genetic structures and sequences are beginning to read like an archive of the sequential anthropogenic stressors to which they have been subject over the past century, forming a physical registration of human activities within the microbial world (see also Landecker, 2019; Moreno and Gibbons, 2022: 9–10). This article engages with Landecker’s insights to review some of the characteristic molecular-genetic, macroscale-infrastructural, and community-ecological aspects of microbial evolution at a time when human actions are a critical force in shaping their directions.

The effects of antibiotics on microbial species move outward in both space and time from their localised application on pathogenic strains; although the emergence of antimicrobial resistance (AMR) over the past century might be the latest, most obvious and immediately concerning change in the microbial world, the genetic and social stage for AMR was set by the widespread use of early disinfectant compounds (like the quaternary ammonium compounds; Landecker, 2019) and the pollution of groundwater and soil with metals and micropollutants from heavy industry (Buelow et al., 2021). Beyond this, antibiotic usage and AMR has had more subtle effects in reducing microbial diversity and putting symbiotic human-microbe relations on impoverished ground. Prior to the industrial and antibiotic revolutions microorganisms have undergone a phased response to the anthropogenising of the earth over the past 10,000 years, including the use of fire to transform foods, and the widespread adoption of agriculture, each of which have exerted different constraints, opportunities, and selection pressures on microbial evolution (Gillings and Paulsen, 2014). ¹ Collectively, these changes highlight the responsiveness and adaptiveness of microbial evolution to anthropogenic change. Indeed, in the last century this evolution has taken on a new order, in which the evolvability itself, or the ability to evolve, of bacterial strains and communities has increased directly as a result of anthropogenic factors (Gillings and Stokes, 2012).

This article argues that at the molecular level, the microbiological Anthropocene is characterised by the emergence of novel genetic combinations and enzymatic activities targeted towards antibiotic and synthetic agents, as well as the propagation and recombination of these activities throughout microbial communities that are increasingly connected through novel pathways. At the macroscale this is reinforced by the proliferation of geographical, infrastructural, and biological hotspots in which bacterial gene evolution and exchange are catalysed (Cabello et al., 2016; Hsu et al., 2014). In other words, the micro- and macroscale potentialities of microbial evolution are synergising in the Anthropocene to create unintended and runaway evolution of novel (and chemically resistant) organisms.

At the community and ecological level, the Anthropocene is also foregrounding a precarious balance between the negative effects of microbial vitality on human wellbeing and those symbiotic relationships with microbes to whom we are indebted for our very biological existence (developmentally and physiologically). In a rapidly urbanising world we are now seeing reduced exposure to certain ‘old friends’ (Ege et al., 2011; Parajuli et al., 2018); microorganisms with which humans have co-evolved over deep time, which train and prime our immune systems, induce developmental pathways, and provide adaptive advantages to human life (Rook, 2013). The loss of these ancient relationships, which is registering in the emergence of chronic inflammatory diseases (McDade et al., 2010; Wilkins et al., 2019) as well as psychiatric and neurological illness (Rogers et al., 2016), are reinforced and underscored by dramatic alterations to the interlinked human and environmental microbiota; reduced biodiversity in urban environments is increasingly linked to dysbiosis in human bodies (Haahtela et al., 2013), which, like the story of antimicrobial resistance, affirms the interconnections (and lack of; Parajuli et al., 2018) between microorganisms living in different environments, their various multicellular hosts and vectors, and physical infrastructures.

Context, concepts and scope

Elaborating on these two themes, this review uses the concepts of ‘abundance’ and ‘absence’ to organise an overview of how microbes, microbial communities, and human-microbe relations are (co)evolving in the Anthropocene. On the one hand, focusing on bacterial genetic evolution in the Anthropocene, and particularly the mechanisms of AMR, ² offers a potentially provocative – although not redeeming – counterpoint to the narratives of overwhelming anthropogenic biospheric destruction and extinction with which we usually associate it (Giraud et al., 2019). As microbes come into contact with anthropogenic pollutants in complex combinations, they are evolving in all manner of unforeseen and unknowable ways, producing a diversity – an abundance ³ – of genetic structures and phenotypes that may be commensurate with, and potentially in excess of the manifold chemical species and selection pressures which we have subjected them to (Clark and Hird, 2013, 2018). But as Giraud et al. (2019) remind us, ‘the show of life’s resilience offered by AMR, however, does not represent hope that life can flourish within the chemical infrastructures that permeate and toxify our environments. Instead it is a direct and serious threat to the efficacy of antibiotics and their crucial place in modern medical practice’ (p. 363). If the Anthropocene is teeming with resilient and adaptive organisms that are thriving in its ‘emergent ecologies’ (Kirksey, 2015), then in the case of AMR it is this very vitality that disrupts and constrains the possibilities for our own. AMR is a threat to our current ways of life, and raises serious questions about how humans and microbes might continue to co-evolve in the coming decades.

On the other hand, radical changes to the environments in which humans live and work over the past century, as well as changes in medicine and industrialisation – what may be heuristically signalled under the rubrics of ‘modernisation’ and ‘urbanisation’ – are registering as reduced – absent – interactions between humans and environmental microorganisms, reductions in the diversity of human gut microbiota, and of the microbiota colonising the increasingly urban environments ⁴ in which we live. But neither is this loss of microorganisms a promising situation; dysbiosis of our own microbial ecosystems (our ‘microbiota’) register as chronic disease and sickness, whilst reductions in the diversity of environmental microbiota impact ecosystem resilience and adaptiveness (Cavicchioli et al., 2019). As a distorted mirror image to the concept of abundance, this microbial absence points to the depths and complexities of extinction in the Anthropocene, an extinction that recedes from view and that registers in complex and drawn out relations. As Bird Rose (2017) writes ‘extinction cascades involve failing connectivities. . .we are looking at worlds of loss that are much greater than the species extinction numbers suggest.’ (p. G52).

This particular framing is also intended to draw out the paradoxical and contradictory nature of microbial life, our relationship with it, and how it has been represented in discourse over the past century. This is what Sariola and Gilbert (2020) refer to as ‘the powerful ambivalence of microbes in the Anthropocene’. Which is epitomised by the fact that, as they put it: ‘in our failed attempts to eliminate microbes from our lives, the Anthropocene has created the conditions for recombinant viruses and antibiotic-resistant bacteria’, whilst at the same time forcing us to ‘recognise that microbes are part of our very being’. (2) This ‘paradoxical’ situation unfolds through the interaction between different material-discursive practices towards microbial lives, and the ontological vitality of microbial life itself, as it moves to constantly respond to and disrupt these practices. Thus, as we move away from antibiotic worldviews and an ‘antiseptic-consciousness’ (Ironstone, 2019) towards probiotic mentalities (Greenhough et al., 2018; Paxson, 2008; Paxson and Helmreich, 2014), we are simultaneously seeing the emergence of novel pathogens and viruses that respond to disinfectants (McLeod et al., 2020) and whose emergence is historically linked to antibiotic practises in the past. In other words, changing views on human-microbe relations such as probiosis unfold incompletely, alongside and amidst other antibiotic practices in various mixtures, a picture that becomes even more complex when it is read through the uneven socio-economic conditions that constrain and enable different relations with microbes within and between different parts of the world (Nading, 2016). Human-microbe relations are intensely situated in different epistemologies and materialities (Hinchliffe, 2021).

Thus, whilst in classical biological definitions ‘co-evolution’ refers to the reciprocal genetic evolution of species existing in ecological relations (Thompson, 2010), within the Anthropocene, in which humans are the ‘greatest evolutionary force on the planet’ (Palumbi, 2001), genetic co-evolution of interdependent species exists alongside and is entangled with wider patterns of human-environment co-evolution (Ross and Richerson, 2014; Søgaard Jørgensen et al., 2020). Here, eco-evolutionary dynamics emerge from feedback between cultural, technological and social evolution on the one hand (Creanza et al., 2017), and biospheric systems on the other, composed of earth systems, ecosystems, communities, and individuals. As Søgaard Jørgensen et al. (2020) put it, although ‘genetic evolution is now shaped by human actions, ranging from the level of the gene to that of the biosphere. . .cultural change not only is a driver of, but is itself shaped by ongoing environmental change’.

Antimicrobial resistance is a key example of this complex cultural-ecosystemic-genetic co-evolution. Widespread antibiotic usage was the result of human technological, social, and cultural evolution that directly impacted the genetic evolution of microorganisms. As microbial genetic evolution occurs at a radically faster rate than human genetic evolution – hours to days versus decades to centuries – the effects of antibiotic use on human genetic evolution is obscured by more rapid changes in cultural and technological evolution, such as the development of new antibiotic classes (Martínez and Baquero, 2002). At the same time, the consequences of antibiotic use are paramount for human biology at more readily perceived levels: In the Anthropocene, where human biology has become increasingly permeable to a contaminated world (Meloni et al., 2022), antibiotics use is linked to epigenetic changes, microbiota dysbiosis, and complex non-communicable diseases (Austvol et al., 2020; see below). The co-evolutionary dynamics between humans and microorganisms, therefore, operate at distinct spatial-temporal levels and as complex interactions between various biological, cultural, and technological strata.

In a related way there is not one Anthropocene, but many complex, uneven, and geographically situated processes that interact to change the earth system in different ways, and that is why this review is limited in its scope. My focus is primarily on urban systems and processes, which I am using as a proxy for the Anthropocene, and their effects on microbial life. These processes can be roughly summarised as antibiotic use, pollution, agricultural intensification, dietary changes, and sanitation as well as their networked interactions through logistical and infrastructural reorganisations. Around half of all people currently live in urban settings, and this is expected to increase to 70% by 2050 (World Bank, 2020). In urban centres anthropogenic forces and selection pressures are concentrated, mixed, and unfold side by side, particularly those processes that impact microbial evolution. The studies focused on in this review are drawn from the primary scientific literature on microbial evolution, as well as their conceptual interpretation in review articles, opinion pieces, and hypotheses. The first section focuses on microbial genetics and their interactions with anthropogenic stressors and pollutants both at the molecular and macroscales. The second section is more concerned with (lost) human-microbe relations and the microbial communities – the microbiota – that live within human bodies and their environments. Whilst a focus on genetic evolution in the Anthropocene moves between spatial scales to link microbiological processes to anthropogenic changes in land use, the microbiota takes us outward from the increasingly porous boundary of human body to the biota of our built environments and anthropogenic ecologies.

Abundance; microbial genetic evolution and diversification in the Anthropocene

Properties of microbial genetic evolution

In a recent paper Meloni et al. (2022) discuss biological plasticity in the Anthropocene, and how changes in earth systems are increasingly registered in the biological substratum of living organisms. Although their focus is on epigenetic mechanisms as a form of biological plasticity that exhibit rapid, reversible, flexible, and sometimes heritable responses to environmental stressors, a focus on bacteria gives a view of the genetic structure itself – including the sequence of base pairs that comprise it – as flexible and responsive over short time periods, as well as mobile, and conjugative. ⁵ The bacterial world is awash with a form of genetic flexibility, exchange, and recombination that unsettles species (and individual) borders and genetic identity (Gilbert et al., 2012).

Bacterial evolution is not limited to chance mutations in genes, inheritance of these mutations by offspring, and natural selection of the corresponding phenotypes. Bacteria modulate their own rate of DNA mutation following exposure to environmental stressors and are able to transfer genetic material between themselves in a process referred to as ‘lateral gene transfer’ (in addition to the vertical transmission represented by clonal/asexual reproduction). In lateral gene transfer, ⁶ genes and other DNA elements are spread from one microbial cell to another (transduction and conjugation), or from the environment into a cell (transformation), a process that can occur across diverse taxa (species, genera, and even kingdoms) (Ochman et al., 2000). Bacterial genomes contain different kinds of mobile elements that allow DNA to move between chromosomal locations within a single cell, and between bacterial cells. Bacteria can fuse together – or ‘conjugate’, allowing the dissemination of mobile genetic elements between them. Plasmids refer to extrachromosomal circular DNA molecules that contain elements enabling their own replication as well as genes encoding the machinery that promotes bacterial conjugation and therefore their own dissemination between different cells. Other regions of DNA, known as integrative conjugative elements undergo excision and reintegration into new sites, into mobile vectors like plasmids, and into new organisms through conjugative transfer (Wozniak and Waldor, 2010). Bacterial DNA is hypermobile, lively, and flexible, moving between chromosomes, plasmids, and bacterial species, encoding its own replication and dissemination.

These mechanisms of microbial genetic fluidity have largely been studied and elucidated with reference to the growing issue of antimicrobial and antibiotic ⁷ resistance (AMR). In this article I use the mechanisms of AMR to provide a case study of how human forces have influenced microbial evolution. But as we will see, the cellular and genetic processes driving AMR intersect with a wide variety of other anthropogenic pollutants and their processes, pointing to complex and emergent properties in Anthropocene bacteria that may have implications for microbial evolution far beyond the province of antibiotic resistance. Soon after the introduction of antibiotics certain bacterial strains developed resistance to these agents and, although ‘it was assumed that new antimicrobials would be discovered and/or developed at a rate faster than bacteria evolved resistance’, (LaPara and Burch, 2011: 241–242) the discovery of new antimicrobials has slowed significantly since the 1960s. It is important to note that although antibiotics are not strictly anthropogenic, in that their original structures (or analogues) were found within the microbial world, the current rates at which they are synthesised and used, coupled with their discharge into all manner of environments, far outstrips their usage by microorganisms. Significant increases in the use of antibiotics since the 1950s also coincides with the Great Acceleration (Steffen et al., 2015), and is intimately tied into the demographic intensities of that period, the clinical/medical pressures of an expanding population, and the agricultural intensification required to feed it. In this sense antibiotic use can be seen as an enabling condition for the Anthropocene on the one hand, whilst on the other antibiotic pollution and the corresponding emergence of resistance genes can be seen as an informal proxy or signature of the Anthropocene (Zhu et al., 2018).

The origin of many of the antibiotic resistant determinants detected in pathogenic bacteria is from environmental microorganisms, and their spread to and proliferation in the commensal and pathogenic bacteria of the human microbiome follows a story that links the molecular logic of flexible and transmissible DNA elements with the industrialisation of human society (Aminov, 2009). In order to understand this story we are led to massive levels of industrial pollution (primarily of metals), the reorganisation of human societies into tightly linked and shortened food chains concentrated in dense urban centres, and the radical mobilisation of diverse disinfectant, antimicrobial, and antibiotic compounds through them. The rise of antibiotic resistance has occurred rapidly over the past 100 years (and primarily in the last 50 years) giving the image of an instantaneous revolt by the microbial world at our attempts to control it, but looking at the fine-grained structure of the microbial DNA sequences implicated in resistance, and as revealed through deep-sequencing metagenomic technologies, the sequential effects of, and responses to, anthropogenic factors by these organisms can be finely resolved.

Microbial genetic fluidity in action: The class I integron

The evolution of the ‘clinical’ ⁸ class I integron provides a powerful illustration of the effects of anthropogenic selection on the structure, function, and dissemination of microbial genetic elements, as well as the sequential accumulation of resistance in response to specific pressures and events. Integrons are sections of microbial chromosomal DNA that are able to capture other genes called ‘cassettes’ through recognition of specific sequences in these target genes. In this way, integrons may accumulate up to hundreds of gene cassettes that can be combined and recombined in various combinations and permutations, and the protein products of which may confer selective advantages to their hosts. Whilst integrons may have some limited autonomous mobility, it was the insertion of the class I integron into the Tn402 transposon (Kholodii et al., 1995) that allowed its propagation throughout the microbial world, an event that ‘happened in a single cell perhaps as recently as 100 years ago’ (Ghaly et al., 2017; 8). Transposons are DNA elements that encode their own mobility between different sections of DNA, and the Tn402 transposon is particularly successful at targeting its own insertion into plasmids, thus ensuring its spread between bacterial cells. In this way, the class I integron has, over the last century, been integrated into numerous mobile DNA elements that can encode their own ability to transfer between cells. Over the intervening time period, and during its travels through various microbial taxa and phyla, it has accumulated numerous genes for resistance to various antibiotics (Ghaly et al., 2017). Within a highly modified landscape of hard selection pressures the class I integron ensures its own survival, evolution, diversity, and transmission via co-opting catabolic resistance genes that confer selective advantage to the hosts to which it spreads. Although it is likely the class I integron originated within an environmental microbe, it is now detected in 40-70% of Gram-negative pathogens that infect humans and livestock, as well as in Gram-positive microorganisms and plant pathogens.

But how and why did this particular integron become so successful? The class I integron was ‘amongst the earliest elements to move from the environmental resistome into human ecosystems’ ⁹ (Gillings, 2017; 11) and this is linked to the ‘increasing use of hard selective agents’ (Gillings, 2016: 27) during the Anthropocene. The use of disinfectants, and increasing pollution of groundwater, freshwater, and sediments with heavy metals falling out from industrialisation at the turn of the 1900s increased the selective advantages of resistant genes that, before such pressures, would have been more costly to maintain. It was in this way that numerous other resistance genes were fixed in bacterial populations. For instance, although mercury resistance genes predate anthropogenic influence, these elements became fixed in the human microbiome at the upstart of industrial activity ‘as early as the mid-1800s’ (Gillings, 2016: 27). In a similar way to the success of disinfectant resistance genes, the success of the mercury resistance gene was driven by historical patterns of metal pollution that predate the antibiotic era, a point reinforced by the observation of plasmids that contain metal resistance genes but lack antibiotic resistant determinants (Datta and Hughes, 1983; see also Poulain et al., 2015). Whilst in earlier periods the advantage conferred by newly acquired mobile DNA elements were often relative owing to continuous fluctuations in selection pressures, and sometimes costly, as new proteins had to integrate into preexisting metabolic networks, the spread of resistant genes are linked to the formation of standalone phenotypes that allow organisms to flourish in environments that are polluted with novel pressures, such as metals, disinfectants and antibiotics (Gillings, 2016).

Specific to the case of the clinical class I integron is its early association with the qacE gene that codes for an efflux pump protein. This protein can pump cationic ions out of the cell, and was co-opted to extrude quaternary ammonium ions, amongst the earliest disinfectants in widespread use. It is suggested that the use of such disinfectants ‘selected for an event that transferred a chromosomal class I integron, with an attendant qacE gene cassette, from a betaproteobacterial chromosome onto a transposon in the Tn402 family’. (Ghaly et al., 2017; Gillings, 2016: 28. See also, Gillings et al., 2009) Betaproteobacteria intersect with the human food chain through their colonisation of plant roots, and it is likely that this was the source of its transmission to the human microbiota; this transposon is now ubiquitous in the commensal microbes of the human gut, where it has captured other resistance genes for disinfectants and antimicrobials.

It was the genetic and phenotypic properties of the class I integron that allowed it to spread so prolifically throughout the microbial world. These factors – the ability to move between chromosomal locations, onto plasmids, and to sample catabolic and resistance genes, ‘combined synergistically’ (Gillings, 2016: 11) both with each and with the rapid modification of the environment by human use of metals, disinfectants, and antibiotics to ensure the spread, diversification, and evolution of this genetic element. Earlier resistance genes provided the selective advantage – and the physical genomic context – for the acquisition of further resistant genes as novel antibiotics came into widespread usage. The class I integron became a vector for the dissemination of antimicrobial resistance genes; as it moved between bacteria it sampled and incorporated various resistance encoding elements.

Synergies between different chemical pollutants driving microbial evolution

Michael Gillings has argued that the proliferation of the class I integron and its transmission between microorganisms is so strongly linked to human activity that the presence of this genetic element in a particular environmental context can be regarded as a proxy for anthropogenic activity in that region; class I integron levels concentrate around human settlements and taper off the further one samples away from these locations (although they have been found on all continents, including Antarctica, Gillings, 2017; Koczura et al., 2016; Power et al., 2016). In a similar way, Gillings et al. (2018) have suggested the integron itself represents a novel type of pollutant that they term ‘xenogenetic’ pollutants. The critical difference between this genetic pollutant and other forms of anthropogenic pollutants that are primarily chemical in nature is that although the latter have intrinsic half-lives, the former has the ability to replicate and ensure its own persistence in the environment. They draw an analogy between these ‘invasive elements’ and more traditionally recognised ‘invasive species’ or ‘neobiota’ that are also a characteristic component of an Anthropocene biosphere (Williams et al., 2015). Moreover, the selection pressures giving rise to the recombination events driving the reticulated complex structures of the integron could not have emerged before ‘the advent of the Anthropocene’. (Zhu et al., 2018: 1494) Perhaps more importantly, the ‘survival and dissemination’ of xenogenetic elements ‘is promoted’ by other types of pollution ¹⁰ stemming from agricultural, industrial, and urban intensification (Gillings et al., 2018: 976). Indeed, there is strong ‘potential for unanticipated synergies’ (p. 976) between more traditionally defined chemical pollution and the generation and propagation of xenogenetics.

The kinds of negative synergies that Gillings et al. recognise as driving the diversification and propagation of the class I integron might be characteristic of a more general phenomenon of microbial genetic adaptation in the Anthropocene. It is now widely recognised that antimicrobial resistance isn’t driven by antimicrobials alone, but a slew of anthropogenic pollutants are able to promote resistance through emergent and contingent relationships with antimicrobial resistance genes and proteins. As Rodgers et al. (2019) state ‘elevated antibiotic resistance is evident in environments with high levels of [general] anthropogenic stress’, which, as they argue ‘challenges the perception that AMR occurs solely as a consequence of antibiotics.’ (p. 61). Indeed, there is strong evidence from a number of studies, including early studies on AMR, that heavy metal exposure increases rates of antibiotic resistance (Foster, 1983; Stepanauskas et al., 2005; Summers et al., 1993). Certain trace metals are essential for microbial function, and microbes have specific physiological pathways for utilising and metabolising particular metal ions, but in larger concentrations, or in situations of chronic exposure, metals exert a selective pressure on microbial communities and resistance to antibiotics can emerge within these metal-stressed communities through various mechanisms.

On the one hand, ‘cross-resistance’ describes a situation in which the gene product for metal resistance also confers resistance to antibiotic compounds. Proteins which extrude metal ions from the bacterial cell, for instance, can be co-opted for pumping out antibiotics. More generally there are crossovers between the molecular mechanisms that produce antibiotic and metal resistance, including reduction in membrane permeability, chemical alteration of the molecule, mutation of its cellular target, and sequestration of the molecule in question (Baker-Austin et al., 2006). On the other hand, through the formation of the complex genetic elements discussed above, the resistance genes for a large number of anthropogenic pollutants have become physically linked in tight clusters and their expression regulated by the same promoter (for instance the promoter for the integrase I element). In this way the lateral transmission of metal resistance, which proliferates and propagates in sites of metal pollution, simultaneously transfers genes for antibiotic resistance. Cross-resistance is an emergent property that arises from the conjoining of many resistance elements together; the tight linkage between genomically connected resistance genes means that exposure to just one of the stressors will result in the selection of the other resistant genes, too.

Synthetic organic pollutants, such as polychlorinated biphenyls and polycyclic aromatic hydrocarbons – which are emerging contaminants and have been suggested as chemostratigraphic markers for the Anthropocene (Gałuszka et al., 2020) – might also increase antibiotic resistance by various mechanisms, including selecting for those microbial taxa that constitutively produce more antimicrobial compounds and which therefore have intrinsic defence mechanisms to them (Gorovtsov et al., 2018). The kinds of negative synergies between different environmental pollutants and bacterial resistance are mediated through various mechanisms that converge on microbial stress and their metabolic responses to it. The SOS response, for instance, is a general response to environmental stressors and increases the rate of microbial DNA mutation (McKenzie et al., 2000), as well as the lateral transfer of genes which might confer adaptive advantages in times of exposure to novel stressors (Beaber et al., 2004).

Synergies at the macroscale driving microbial evolution

Although ‘antimicrobial resistance is far better understood at the small scale than at the large scale’ there is growing knowledge on the ecology of antimicrobial resistance and how resistance is ‘spread over geographic distances’ (LaPara and Burch, 2012: 242). Research into the macroscale dynamics of antimicrobial resistances is furnished with the concepts of ‘reservoirs’ for antibiotic resistance bacteria and their associated (mobile) genetic elements (the geographical and ecological mapping of the ‘resistome’), as well as the idea of hubs, where bacteria and pollutants meet and mix, and the correspondent interfaces – inlets and outlets ¹¹ – between these sites. One thing that emerges from this growing research is the recognition that the kinds of negative synergy characterising antimicrobial evolution at the molecular level are mirrored, catalysed and reinforced by phenomena at the macro-scale. In the Anthropocene, this reinforcement is driven by the generation and proliferation of sites in which numerous different pollutants, genetic elements, and microorganisms mingle, mix, hybridise, and recombine. Certain locations concentrate these elements; urban centres, for instance, shed large amounts of various forms of pollution into waste streams, including a host of pharmaceuticals and antimicrobial agents, as well as a distinctive microbiota.

Accordingly, the water cycle, and particularly its routes through urban systems and aquaculture is an emerging locus for the dissemination of antibiotic resistance genes and bacteria, as well as the pollutants that select for them. Hydrosystems are ‘mixing places’ for chemical pollutants and microbial organisms, places where bacteria and pollutants meet, mingle, and transfer antibiotic resistance genes (Almakki et al., 2019: 64). The process of urbanisation has altered the water cycle significantly, for instance by abstracting ground and surface water for drinking and agricultural purposes. The built structures of urban centres, with their non-porous and hydrophobic surfaces, provide ready-made pathways for the flow of water which may accumulate pollutants and toxins as it flows over these surfaces (for instance dust from roads, metal from roofing, etc) before being discharged into rivers. In a similar way the large surface areas of hydrological infrastructure provides ‘relatively new ecological niches for colonization by microorganisms’ into biofilms, large consortia of physically associated microorganisms that are metabolically coordinated and which share genetic material freely (Manaia et al., 2016; McLellan et al., 2015: 141). In urban centres these hydrological pathways tend to converge on wastewater treatment plants, which collect the effluent of modern life, and are ‘unique interfaces between human society and the environment’ (Karkman et al., 2018: 220) where a multitude of compounds – pharmaceuticals, cosmetics, metals, etc. – hybridise to form novel opportunities for bacterial evolution and genetic adaptation. In this way, sewage treatment plants and wastewater act as a ‘reactor for assembling complex DNA elements’ as well as the ‘dissemination of these newly formed assemblages to diverse species’ (Gillings and Stokes, 2012: 350).

If the water cycle, and wastewater treatment sites in particular, represent the outputs of human societies, then how do the inputs of society affect microbial genetic transmission and evolution? Agricultural intensification and the production of animals for consumption is suggested as a critical point for the incubation and downstream dissemination of antimicrobial resistance. Antibiotics are widely used in agriculture for growth promotion and as a prophylactic practice, and animals bred for human consumption, such as ruminants, harbour diverse antibiotic resistance genes in their guts (Noyes et al., 2016). The human commensal microbiome too is a reservoir for potential antimicrobial resistance genes (Sommer et al., 2010). The organisation of these organisms into highly shortened food chains focused on humans, in addition to creating novel ecologies (like monocultures) and ‘unusual encounters between humans and nonhumans’ (like wet markets) (Aronsson and Holm, 2022: 25) therefore begins to link together environmental, commensal, and pathogenic bacterial species through these various hosts and heightens the transmission of novel genetic elements between them, ¹² and of antibiotic resistance genes from environmental to human ecologies. The animal gut has been conceived as a ‘melting pot’ (Shterzer and Mizrahi, 2015) for lateral gene transfer, suggesting that the human food chain might be considered a superhighway for hypermobile DNA elements, and a reactor for lateral gene transfer.

In addition to the geographical, infrastructural and biological hotspots where flows of bacteria, genetic materials, and pollutants are concentrated, the Anthropocene is seeing the emergence of new mobile hotspots for bacterial evolution and genetic exchange. Plastic pollution is temporally confined to the modern era, with globalised production of plastic accelerating from the 1950s onwards (Geyer et al., 2017). Indeed, the accumulation of plastic into stratal deposits has been suggested as one potential marker for stratigraphic characterisation of the Anthropocene’s onset (Zalasiewicz et al., 2016). Discarded plastics, as well as their weathered products such as microplastics and nanoplastics in aquatic and terrestrial ecosystems serve as colonisation sites for diverse microbial taxa, which form biofilms on these debris in tightly linked metabolic communities that are taxonomically and functionally distinct from the surrounding media (Dussud et al., 2018; Puglisi et al., 2019). Recent studies have suggested that plastic-associated biofilms have increased propensity for gene transfer (Arias-Andres et al., 2019), as well as the suggestion that plastisphere communities serve as reservoirs for antibiotic and metal resistance genes (Yang et al., 2019). As Vos (2020) suggests, because plastic adsorbs contaminants from the surrounding environment, including metals, pesticides, and antimicrobials, plastic pollution could ‘potentially facilitate resistance evolution’ (p. 4). Moreover, marine plastic is highly mobile, and increases the chances that indigineous and allochthonous microorganisms may come into contact and potentially exchange genetic material. Plastics are also a major component of waste streams that end up in sewage treatment plants, and the microplastic component of activated sludge in these treatment plants enriches pathogenic microorganisms and antibiotic resistance genes (Pham et al., 2021). Plastics in terrestrial soils are also an emerging ecology, and the soil plastisphere is enriched in pathogenic microorganisms as well as antibiotic resistance genes (Zhu et al., 2022).

Thus, to the micrological and genetic mechanisms of hybridisation, recombination, integration, and conjugation, we might add the macroscale dynamics of effluent discharge, shedding, waste mixing and ecological selection (and not least antibiotic overuse) in driving and sculpting the emergence of novel bacterial genetic structures in the Anthropocene. The processes of industrialisation, agricultural intensification, and urbanisation are characterised by the creation of new links – and the destruction of old ones, for the circulation of bacteria and genetic elements, together with the selection pressures that drive their evolution. Elements that were once separated over vast geographical (and geological and temporal ¹³ ) distances are brought together in new relations. New sites emerge, whilst old ones are reconfigured and reorganised to streamline the flow of mobile genetic elements throughout the bacterial kingdom. These kinds of phenomena also force us to fundamentally reconceive how we think about spatiality and geographical relations; the transmission of microorganisms between bodies and sites has been conceived of as typifying a topological form of space, as opposed to a Euclidean geometry, where foldings and other forms of translation bring geographically distant points into close contact (Braun, 2008; Dixon and Jones, 2015; Hinchliffe et al., 2013). In a globalised world microorganisms are able to use our infrastructures to their own ends, and often ‘greater efficacy than we ourselves manage’ (Clark and Hird, 2018: 246).

Absence: Lost interactions between humans and environmental microbiota in the Anthropocene

Hygiene hypothesis, Old Friends

If the unintended consequences of antibiotic usage include the proliferation of resistant organisms and their associated mosaic gene structures, then perhaps a backdrop to these emphatic processes has been a more generalised reduction in the diversity of the microbial species to which humans are frequently exposed. Just as the presence of resistant strains is leading to sickness and death, the absence of other microorganisms is increasingly being associated with an epidemic of non-communicable illnesses in urban populations. In 1989 it was first suggested that exposure to infections during early life was associated with reduced incidence of allergic rhinitis and eczema in later life (Strachan, 1989). Strachan’s original article analysed the incidence of hay fever and eczema in a cohort of children to reveal that incidence of these illnesses were inversely proportional to the number of siblings. This phenomenon was referred to as the hygiene hypothesis, where hygiene is ‘understood not as a problem of individual behavioural hygiene but rather of the broad environmental infectious burden’ (Bach, 2018: 105) represented by the ‘vertical or horizontal transmission of infections’. (Ege, 2017: S349). The ‘post industrial epidemic’ (Emanuel, 1988) of hay fever was suggested to be prevented by viral and microbial infection in the early years of life, and to increase in incidence with smaller family size due to decreased cross infection between siblings, as well as general improvements in sanitation and personal hygiene. The generalised conditions of modernity in the Global North – sanitation systems, antibiotic (over)usage, as well as smaller family sizes and decreasing exposure to animals are producing hyper-sanitised spaces in which microbial exposure is significantly reduced.

The hygiene hypothesis (which is now suggested to be an outdated term because of its connotations with personal hygiene rather than systematic changes in the exposure to microbial diversity in the developed world – Bloomfield et al., 2016) has subsequently given way to the ‘The Old Friends’ hypothesis. This model of human-microbe interaction proposes that in industrialised contexts humans have ‘progressive loss of contact with organisms with which we co-evolved and that play a crucial role in setting up [physiological] regulatory pathways’ (Rook et al., 2014a: 47). Such organisms include environmental microbes, human pathogens such as helminths, and a host of commensal and parasitic symbionts ‘with which [humans] were in daily contact’ and which co-existed ‘in the environments in which humans evolved and lived until recently’ (Rook et al., 2014a: 47). Throughout evolution the human organism had to learn to tolerate a certain degree of coexistence with microbial organisms because contact was a fact of biological existence. Humans and these microorganisms co-evolved, such that human development and physiology became intimately entangled with the presence and function of certain microbial others. The Old Friends hypothesis focuses specifically upon the role of these relationships in ‘training’, modulating, and priming the immune system and how the loss of them is leading to a dramatic increase in immunological dysregulation in urbanised and urbanising contexts; ‘Contact with the immunoregulatory “Old Friends” diminishes rapidly when industrialization occurs, and individuals start to inhabit a plastic and concrete environment, to consume washed food and chlorine treated water, and to minimize their contact with mud, animals and faeces’ (Rook et al., 2014a: 48).

This is registering primarily in dysregulated immune responses against three different sites: (1) human self tissue – causing autoimmune diseases like multiple sclerosis and type I diabetes, (2) harmless allergens – causing atopic disease, like eczema and asthma and the (3) the human gut microbiota itself, which is causing inflammatory bowel diseases like Crohn’s disease and colitis (Liddicoat et al., 2016). Defective immunoregulation is resulting in overactive immune systems and low-grade and consistent inflammation that predisposes to other forms of chronic illness, including cardiac problems and psychiatric illness (Rook et al., 2014b). ¹⁴ In such individuals, stressors and insults that result in inflammation are exacerbated as inflammation cannot be switched off. Individuals live in a state of chronic inflammation, responding excessively to innocuous stimuli and, as such, showing predisposition to a range of chronic diseases (Rook et al., 2014b).

The mechanisms linking microbial exposure to prevention of autoimmune disease are related to the ‘selective activation of different environmental [microbial] agents on specific immunoregulatory cell types.’ (Bach, 2018: 113). Certain forms of endotoxin as produced by gram-negative bacteria, for instance, are able to activate subsets of anti-inflammatory immune cells, or to desensitise signalling in pro-inflammatory and autoimmune cells. Exposure to endotoxin – a metabolite of gram-negative bacteria – has, in this sense, been linked to the protective effect of the farming environment on the development of asthma (Schuijs et al., 2015).

The Old Friends model is found at the intersection of two phenomena that describe the co-evolution between humans and microbes. On the one hand, humans purposefully constrain and limit microbial flourishing through sanitation, antibiotics, etc. whilst on the other, microbial communities rapidly adapt to these environmental fluctuations, often directly undermining them (for instance through the evolution of AMR as described above). Indeed, the adaptive resilience of microbes often has maladaptive consequences for humans, painting the picture of a less than even co-evolution, and one in which microbial adaptation is more rapid and refined than human physiology (and technology) can keep up with. Changes in exposure to microorganisms and their metabolic products then registers in the biology of humans, as alterations in immunoregulation leading to consistent low-grade inflammation that predisposes to a number of chronic and non-communicable diseases. Indeed, If the Anthropocene is seeing an ‘altered balance between man and microbiome’ then this is because, relative to microbial evolution, human ‘adaptation. . .lags behind in a world of fast and profound environmental changes’ (Ege, 2017: S351).

Gut microbiota and dysbiosis

The old friends and hygiene hypotheses stake out a central role for human-microbe co-existence and proximity in shaping human health and disease; microbes are viewed as having constitutive roles in human health, and the boundary of the human organism can no longer be so easily delineated. In line with this, the direct effect of pathogenic organisms on a bounded human organism has given way to a view of microbial ecology as community-centric and relational, unfolding through ‘a shift in focus from individual pathogens to an ecological approach that considers the community as a whole’ (Lozupone et al., 2012: 220). In the past couple of decades, the human microbiota, which refers to the collection of microbes that live in and on various bodily hotspots – the oral and nasal cavities, the skin, the intestines, the urino-genital tracts – has become an increasingly recognised factor in human health and disease. The intestinal microbiota contains an immense quantity of cells - >10¹², roughly equal to the number of human cells in the body; (Bäckhed et al., 2005) – and genetic material – potentially outnumbering genetic information 100:1 (Gilbert et al., 2018).

The functions of these organisms support human health by contributing essential metabolic, immunoregulatory, and developmental functions. Just as each of us are indelibly linked to the trillions of bacteria that live out their days in our intestines and on our bodies, these communities link us to the environments outside of our skins; a significant amount of the microbiota is acquired from the environment during and after birth, and remains in dynamic contact with environmental biota throughout life. Our microbiota is dynamic, and is affected by the places we live and the food we eat (David et al., 2014; Voreades et al., 2014) forming an interface between human physiology and the environment, interpreting and responding to signals from each.

The microbiome interacts evolutionarily, developmentally, anatomically, immunologically, neurologically and endocrinologically with ‘human’ physiology (Gilbert et al., 2012) and to such an extent that ‘microbe-encoded metabolic pathways’ of the human microbiota ‘need to be regarded as part of human physiology’ that ‘provide crucial signals that drive development of all organ system including the brain.’ (Rook, 2020: 522) Changes to the human microbiota follow increasingly radical changes to human environments over the past 5 million years (Gillings et al., 2015; see also Davenport et al., 2017). Such changes predate the antibiotic era by millions of years, and have, similar to the changes reviewed above, ‘corresponded with the key transitions that define the Anthropocene’ (p. 847) including the widespread use of fire for transforming food matter and the rise of agriculture that altered the composition of the microbial ecologies to which humans were exposed (see Gillings and Paulsen, 2014; Gillings et al., 2015 for a review of these earlier changes).

Focus on the loss of early life infection and microbial exposure driving dysregulated immune function is thus being complemented by research into ‘dysbiosis’ of microbial communities living in and on the human body. Dysbiosis refers to an imbalance in the microbial flora resident in the gut microbiota, often represented by reductions in microbial diversity (Kriss et al., 2018) and is associated with numerous diseases. Although the causality of dysbiosis in specific diseases can be hard to establish, the phenomenon itself points to an underlying shift in the relations between humans and microorganisms. On the one hand, dysbiosis can arise from negative impacts on the formation and development of the microbiota, whilst on the other it can follow abrupt stressors such as antibiotic consumption.

In addition to the shifts mentioned above, the complex and intertwined processes of modernisation and urbanisation with their associated increased antibiotic use, Caesarean section, water sanitation, processed food, industrial pollution and built environments (Kembel et al., 2012; Parajuli et al., 2018) have wrought the most radical alterations to our internal ecologies. These changes are more tightly coupled to changes in the environment than differences in genetics (Rothschild et al., 2018), and humans living in rural and cosmopolitan societies show significant differences in their microbiota compositions, with rural and non-industrial populations more closely resembling those of ancient people (Wibowo et al., 2021). Numerous extant populations have also served as reference points for understanding the effects of urbanisation and industrialisation (‘Westernisation’) on the human gut microbiome, including the Hadza hunter gatherers of Tanzania (Rampelli et al., 2015; Schnorr et al., 2014; Smits et al., 2017), rural Papua New Guineans of the Asaro and Sasusi societies (Martínez et al., 2015), children from Burkina Faso (De Filippo et al., 2010), Amazonias of Venezuela, and rural Malawians (Yatsunenko et al., 2012). The findings demonstrate increased microbiome diversity in non-agricultural/non-industrialised communities, vastly different microbial abundance profiles, and the presence of microbial taxa that are undetectable in Western urbanised populations, a finding that is reinforced by a meta-analysis of metagenomic comparisons between the gut microbiomes of industrialised/urban and ‘pre-agricultural’ populations (Mancabelli et al., 2017). Non-industralised populations also exhibit Lower inter-individual variation in gut microbiota composition than Westernised counterparts that likely stem from less differentiated diets and reduced barriers to microbial dispersion (Martínez et al., 2015). These divergences are caused by key differences in subsistence patterns, and specifically the increased consumption of fibre in hunter-gatherer populations that leads to the enrichment of microbial species that can extract nutrition from plant foods (Schnorr et al., 2014). Moreover, in the Hadza population, patterns of gut microbial abundance are dynamically variable and exhibit seasonal cycling, and it is precisely these dynamic lineages that may be lost with progressive urbanisation and modernisation (Smits et al., 2017).

Notwithstanding the ethical and conceptual limitations of these studies – the quest for a ‘microbiomic purity’ projected onto ‘uncontaminated’ populations (Benezra, 2020) – the evidence suggests large scale temporal alterations in the gut microbiota corresponding to key features of the Anthropocene. As Tito et al. (2012) claim, ‘the modern cosmopolitan lifestyle [has] resulted in a dramatic change to the human gut microbiome’. The precise details of dysbiosis are complex and potentially singular; it has been suggested that ‘whilst healthy microbiomes are all alike in composition, microbiomes in dysbiosis are all unlike one another and exhibit no uniform pattern of dissimilarity’. (Fackelmann and Sommer, 2019: 198; see Zaneveld et al., 2017). Although a core microbiota may exist this is probably reflected in the functional annotation of the genetic and metabolic pathways it encodes, rather than its taxonomic composition (Berg et al., 2020). One emerging pattern, however, is a reduced diversity of microbial taxa in modern urban industrialised human gut microbiota (Blaser and Falkow, 2009; Yatsunenko et al., 2012) and the loss of what Sonnenburg and Sonnenburg (2019) refer to as ‘VANISH taxa’: taxa that are ‘volatile and/or associated negatively with industrialized societies of humans’. As noted above, human cultures who have had little to no contact with the wider world, such as the Amerindian Yanomami and who ‘maintain hunter gatherer lifestyles have unprecedented diversity of both microbial species and gene functions’ in their gut microbiota (Clemente et al., 2015; Gillings et al., 2015: 848). Metagenomic sequencing of fossilised human microbiota demonstrates that early humans also had significantly higher levels of microbiota diversity than modern day humans living in urban settings (Adler et al., 2013).

The urban gut microbiome, then, is most commonly associated with the depletion of certain microbial taxa, an imbalance in the ratios between them, and the loss of specific individual taxa that are involved in functions related to the biotransformation of lost dietary components, such as fibre (see Tyakht et al., 2014 for a review on some of the lesser abundant species and their co-metabolic relations). Diet is one of the most significant forces shaping the composition of the microbiota, and the Western diet which is ‘high in saturated fats, red meat and carbohydrates’ whilst being low in ‘fresh fruits, vegetables, whole grains, seafood and poultry’ has induced significant changes in the function and structure of the gut microbiota of humans living in urban contexts. Indeed, the ‘Western diet is associated with an irreversible gut microbiota dysbiosis.’ (Zuo et al., 2018: 443).

The urban context is also associated with larger differences between individual’s microbial communities (beta-diversity), but these communities are themselves less diverse (alpha-diversity). This observation might be explained by the fact that in urban contexts there are numerous barriers to the dispersal of microbial symbionts between hosts and between the environment and hosts. At the same time, people in urban centres also tend to eat highly individualised diets – which often contain large amounts of sugar and low amounts of fibre – thus generating a pattern of individualised selection pressures (Martínez et al., 2015). In addition to the dispersal hypothesis, Gillings et al. (2015) have proposed a model to explain the genesis of increased beta-diversity of the microbiome in modern societies that explains increased inter-individual microbial diversity through the ‘interplay between evolutionary and ecological factors’ (p. 849) including ‘interruption to colonisation opportunities. [and] a depauperate pool of microbial species’ to sample from, as well as a landscape of various selection pressures. According to these models, ‘each individual in modern society is a microbial island’ separated from contact with other islands by stringent barriers such as water sanitation, Caesarean-sections, decreased family size, and reduced breast feeding. The contours of this cut off and isolated microbiome is sculpted through the hard selection pressures of antibiotics, a ‘series of bottlenecks’ which each ‘takes the microbial assemblages of individuals down increasingly divergent paths.’ (p. 849).

Beyond the role of sanitation and antibiosis in sculpting and isolating the human microbiome, diverse chemical contaminants and pollutants also modify its structure and function (Jin et al., 2017). Pollutants interact with the microbiome, and the microbiome interacts with pollutants –microbes can aid in the detoxification of harmful compounds (Koppel et al., 2017), but also catalyse the bioactivation and toxification of others, for instance transforming polycyclic aromatic hydrocarbons into oestrogenic molecules that disrupt the function of the endocrine system (Van de Wiele et al., 2005). Bacterial communities mediate the toxicity of environmental pollution in human and nonhuman bodies, acting as an interpreter of and responder to a dramatically altered chemosphere. On the other hand, a diverse and resilient microbiota acts as a ‘buffer’ (Fackelmann and Sommer, 2019: 180) against communicable and non communicable illness alike. This raises important questions about how the human (gut) microbiome will continue to evolve in a landscape that is saturated with complex mixtures of chemical pollutants. Will we see an increasing dysbiosis with negative consequences for host health, or a new co-evolution of resilience between human bodies and their resident microbiota? We can imagine this latter option as perhaps represented by the incorporation into the microbiota of more species who efficiently detoxify and degrade pollutants, just as the guts of insects are beginning to harbour microorganisms that degrade microplastics (Jang and Kikuchi, 2020; see also Koppel et al., 2017).

Conclusions

A trope that has become familiar in this story of microbial life in the Anthropocene – as for the ontology and epistemology of biology in the Anthropocene in general (Mansfield, 2018) – is the unceasing crossing of boundaries and spatial scales; antibiotic resistance fluxes through infrastructures, networks, and bodies; human bodies lose their boundaries in a wider ecology; and the micro- and macroscales are repeatedly folded into one another. At the molecular level, the chemical landscape of the Anthropocene is increasing the very evolvability of microbial life, leading to a panoply of novel genetic structures which ‘promote their own diversification’ (Garriss et al., 2009). At the macroscale, demographic intensifications and their associated infrastructures concentrate, disperse, and mobilise these structures. Somewhat paradoxically, the diversity of other forms of microbial symbionts, and particularly those implicated in the development of host immune systems, is falling.

The changing microbiota of the Anthropocene also point beyond the boundaries of the human organism in another way – ‘at least a portion of the human microbiota is in exchange with microbes from the surrounding environment’ (Liddicoat et al., 2016: 1024) which suggests that changes to the human gut microbiota in the Anthropocene are likely to be following changes in the microbiomes of their wider environments. Microbes in urban systems are at the nexus of organism, ecosystem, and planetary health; they provide critical physiological, ecological, and biogeochemical functions in these nested systems (Cavicchioli et al., 2019; King, 2014). The urban microbiota is generally less diverse than those of surrounding rural areas (Flies et al., 2020). Initiatives focused on improving the diversity of urban microbiomes have been established, with the goal in mind that such improvements will have positive impacts for human health (Flies et al., 2018; Mills et al., 2017, 2019).

As Sariola and Gilbert (2020) argue, we need to rethink our relations with microorganisms. ‘Partnering’ with microbes (Bradshaw, 2022a) instead of waging war with them recognises a basic biological reality of being human; that we are not separate and sovereign individuals but intensely ecological beings. The conditions of the Anthropocene have ‘explicitated’ (Latour, 2010) this biological reality, whilst simultaneously threatening its integrity. Indeed, our biological reality is not fixed but responds to our relationship to it; individuals in modern societies are formed in the image of an antibiotic epistemology: as isolated islands of reduced microbial diversity. If the Anthropocene is seeing a radical disconnect between human social, technical, and biological evolution on the one hand, and microbial biology on the other, reconnecting to this world is an ongoing task that likewise cuts across these spheres. The plurality of practices that underscore the emergence and intensification of antimicrobial resistance and dysbiosis belie singular and universal approaches for their remediation. To this end, experimentation with fermentation (Hendy et al., 2021), rewilding (Lorimer, 2016, 2017), and probiotic practices grounded in an ethics of the more-than-human (Cañada et al., 2022) at a diversity of scales and contexts might hold promise. Beyond this, fostering generalised microbiological literacy in society (Timmis et al., 2019), and natural-social science collaboration (Barendse, 2021; Bradshaw, 2022b) might help us to reconnect to the microbial world.