How chemicals and heavy metals contribute to antimicrobial resistance

Antimicrobial resistance (AMR) in the clinical setting has been getting increasing coverage in the media in the past few years. However, emerging research shows that there is more than just the overuse of antibiotics which leads to AMR. The use of biocides and heavy metals contribute to the global problem^[1],
[2]
of AMR and only targeting antibiotic use will not solve it.

In 2011, Dame Sally Davies, chief medical officer for England, highlighted the challenges of AMR in her office’s annual report^[3]
, and single-handedly renewed the dialogue in the UK on AMR mitigation — a debate that began more than three decades earlier^[4],
[5]
. In recognition of the fact that there are no silver bullets to solving the AMR challenge, the World Health Organization (WHO), in collaboration with the Food and Agriculture Organization and World Organisation for Animal Health, pushed forward the agenda for all members of the United Nations to draft a national antimicrobial resistance action plan. Of the 26 national action plans that currently exist (as of the end of 2016), most elaborate on seven core mitigation measures. These are: surveillance around infections, drug-resistant infections and drug use; prevention (e.g. immunisation, hygiene); better antimicrobial stewardship; education and professional development; drug development; rapid diagnostics; and scientific research^[6],
[7]
.

The importance of co-resistance and cross-resistance cannot be understated

The WHO^[8]
, European Commission^[9]
and UK antimicrobial action plans^[10]
, among all 26 of the existing AMR action plans^[10]
, as well as the UK Review on Antimicrobial Resistance Final Report^[11]
, operate under the fundamental premise that the primary driver of AMR is antibiotics.

However, a growing body of evidence points to a much wider breadth of chemicals that are responsible for selecting and maintaining elevated levels of resistance genes, including biocides and heavy metals^[12]
.

Trace concentrations of antibiotic are often sufficient to select for or retain the corresponding resistance gene

Cross-resistance and co-resistance

Biocides are used in all parts of society, from home and hospital to farms and industry. They are found in personal-care products (e.g. toothpaste), household cleaners, wipes and detergents, as well as many hospital hygiene-related products, and are widely used throughout the manufacturing and farming (animal husbandry and food) industries. The presence of biocides in all areas of our built and natural environment contributes to the selection of resistance genes that can directly or indirectly select for antimicrobial resistance genes (ARGs). These biocide resistance genes can be the same genes as antibiotic resistance genes (i.e. cross-resistance), or they can be co-located with one or more resistance genes on mobile genetic elements (e.g. plasmids), a phenomenon termed co-resistance^[13]
.

The importance of co-resistance and cross-resistance cannot be understated because both phenomena facilitate the selection of resistance genes for chemicals that do not need to be constantly present. As a result, micro-organisms in the environment can regularly be found carrying multi-drug resistance despite the fact most of the chemicals that are thought to drive the selection of such genes are not present^[14],
[15]
. Or so we once thought.

Emerging evidence from the literature suggests that trace concentrations of antibiotic are often sufficient to select for or retain the corresponding resistance gene — often at concentrations that are at least 10 times lower than the minimum inhibitory concentration and sometimes over 100 times lower^{[16],
[17],
[18],
[19],
[20]}
.

Determination of the minimum concentration of each antibiotic and biocide that can select for resistance is an active area of research, but is still in its infancy. Early evidence suggests a traditional ‘break point’ as used in the clinical setting might not be easily derived or even feasible, since the ‘minimum selective concentration,’ as it is called, is not only sensitive to each antibiotic but to the presence of co-selective agents (i.e., antibiotic and metal) as well as the location of the resistance gene (i.e. chromosome or plasmid)^[16]
. Owing to the ubiquity of pollutants in sewage and farms, the complexity of microbial communities^[21]
, and the additional complexity of the mobile genome that microbes contain^[22]
, an approximate ‘minimum selection concentration’ might be both pragmatic and the only available way forward for estimating environmental targets for biocide and antibiotic mitigation^[20]
.

Metals

Metal resistance genes function in much the same way as biocide and antibiotic resistance genes and have even more potential sources, such as industry effluent, traffic-related emissions (e.g., tailpipe, tires), nanoparticles (e.g. food, clothing, gene therapy, drug delivery, and water treatment, and personal care products), textiles, mining, food additives in humans and animals, fertilisers, and pesticides^[23]
. Resistance genes that afford metal resistance often fall within the class of resistance mechanisms called ‘efflux pumps’^[24]
. Notably, efflux pumps also provide resistance to a wide range of antibiotics^[25]
and biocides^[26],
[27]
. When viewed in a holistic manner, it is clear that our lives and our environment are constantly exposed to three chemical drivers of resistance genes: antibiotics, metals and biocides^[28],
[29]
.

A key knowledge gap is determining the relative contribution that antibiotics, biocides and metals play in selecting for AMR

AMR in sewage

A fourth major input, antibiotic resistance genes (ARGs), might also play a deceptively important role in this complex story. Every month, in England, approximately 83 tonnes of active antibiotic are prescribed by doctors, the majority (70+%) of which are consumed by patients in the community^[30]
. In every case, the consumer of the antibiotic will be inadvertently selecting for antibiotic resistance in their (gut) microbiome^[31]
, which will be excreted in their faeces. Hence, every city receiving municipal wastewater will, arguably, have a constant stream of antibiotic resistant bacteria entering the wastewater treatment plant. Wastewater treatment plants are not designed to remove ARGs. Moreover, many ARGs have been shown to increase in prevalence while passing through wastewater treatment plants (relative to the size of the microbial population^[32]
). ARGs are also abundant in the sewage sludge that is ultimately deposited on agricultural land to improve soil fertility^[12]
. It is for this reason that there have been increasing calls for policy makers to take notice of these omissions in our AMR action plans because they have the potential to limit greatly our shared goal of reducing the overall prevalence of drug-resistant infections in humans and animals^[12]
,[33]
.

Knowledge gaps

A key knowledge gap is determining the relative contribution that antibiotics, biocides and metals play in selecting for AMR. It is possible that future research might reveal that antibiotics are a relatively trivial driver of AMR in the environment relative to metals and biocides. This might prove to be true owing to their ubiquity in sewage^{[34]
[35],
[36],
[37]}
, animal manure^[38]
, and industrial waste, often found orders of magnitude higher in concentration than antibiotics^[12]
,[39]
). However, it is not necessarily true that an increased concentration of AMR-driving chemicals (antibiotics, biocides and metals) in the environment will lead to a proportional increase in the prevalence of AMR because there are a number of factors that influence selection, most notably, bioavailability. Much like the pharmaceutical industry’s use of prodrugs to improve a drug’s absorption into the blood stream, AMR-driving chemicals are not necessarily all ‘accessible’ to a microbe in the environment, and as such, they will have limited, if any, impact on AMR selection. The factors that influence selection are at the centre of the environmental microbiological research on AMR. Unfortunately, few factors have been ruled out because everything seems to matter some of the time.

The scale and extent to which the global research community is struggling to catch up to where we need to be to inform policy and mitigation is overwhelming. However, there are some recent developments that offer reasons to be optimistic. The UK Research Councils have recently come together to begin to fund some research to fill these gaps in our knowledge^[40]
, with the aim to tackle both the issues highlighted by chief medical officer Davies, as well as the more neglected areas of research into AMR in the environment^[41]
. Moreover, research funding opportunities have emerged through the UK Newton Fund, alliances between the UK and China^[42]
, and UK and India^[43]
, to tackle what will likely include an environmental component of the AMR challenge. Notably, China and India are the two major manufacturing centres for antimicrobials in the world and both struggle with significant AMR within the air, water and soil^[44],
[45]
.

Co-ordinated effort required

The global challenge of AMR and the ubiquitous nature of resistance-driving chemicals is a daunting challenge for humanity. A co-ordinated global research effort is needed to ensure the efficacy of our existing antimicrobials into the future and inform policy and cost-effective mitigation strategies. It is the hope that these knowledge gaps can be filled and mitigation measures can be put in place before the ‘antibiotic apocalypse’ foreshadowed by Davies^[46]
.