Given that: (1) the worldwide consumption of antimicrobial drugs (AMDs) used in food-producing animals will increase over the coming decades; (2) the prudent use of AMDs will not suffice to stem the rise in human antimicrobial resistance (AMR) of animal origin; (3) alternatives to AMD use are not available or not implementable, there is an urgent need to develop novel AMDs for food-producing animals. This is not for animal health reasons, but to break the link between human and animal resistomes. In this review we establish the feasibility of developing for veterinary medicine new AMDs, termed “green antibiotics,” having minimal ecological impact on the animal commensal and environmental microbiomes. We first explain why animal and human commensal microbiota comprise a “turnstile” exchange, between the human and animal resistomes. We then outline the ideal physico-chemical, pharmacokinetic, and pharmacodynamic properties of a veterinary green antibiotic and conclude that they can be developed through a rational screening of currently used AMD classes. The ideal drug will be hydrophilic, of relatively low potency, slow clearance and small volume of distribution. It should be eliminated principally by the kidney as inactive metabolite(s). For oral administration, bioavailability can be enhanced by developing lipophilic pro-drugs. For parenteral administration, slow-release formulations of existing eco-friendly AMDs with a short elimination half-life can be developed. These new eco-friendly veterinary AMDs can be developed from currently used drug classes to provide alternative agents to those currently used in veterinary medicine and mitigate animal contributions to the human AMR problem.
Keywords: innovative eco-friendly antimicrobials, veterinary medicine, public health, commensal microbiota, environmental hazard
he links between animals and humans, in respect of the emergence and spread of resistance, is a major global issue. This review proposes the development for veterinary medicine of new and innovative drugs for food-producing animals; this is not for animal health reasons but rather to mitigate the veterinary contribution to the human resistome.
Currently, food-producing animal medicine does not face the same critical situation as human medicine, because there are neither life-threatening infections of multiply drug resistant microorganisms causing sepsis nor chronic conditions in poultry, pigs, or cattle for which AMD therapy is mandatory. In addition, the prevalence of resistance for major veterinary pathogens that cannot be treated by any AMD is very limited. However, this review proposes a renewal of the veterinary armamentarium with drugs designed to break, for public health reasons, the link between human and veterinary medicine. We have termed, these innovative compounds “green antibiotics,” as they will have minimal (ideally no) ecological impact on the animal commensal microbiome and, more broadly, on the environmental resistome The rationale for and urgency of this proposal is that (i) the worldwide consumption of AMDs to treat or prevent health conditions in food-producing animals will ineluctably increase over the coming decades the so-called prudent use of AMDs will not stem the rise in AMR of animal pathogens and commensals and its subsequent impact on the human resistome.
New, eco-friendly, veterinary AMDs can readily be developed from currently used drug classes to provide credible alternative agents. The viability of this approach is enhanced by the fact that alternatives to AMD use are either not available or not implementable as reviewed by Cheng et al. (2014). The green antibiotic principle is in line with the so-called eco-evo concepts that consider AMR in the broad light of evolution and ecology, rather than that of narrow clinical practices relating to infections (Baquero et al., 2011). In this review, we first consider the animal and human commensal microbiota as comprising a “turnstile” for exchanges between the two resistomes, and then outline the ideal properties of a green antibiotic. Finally, the regulatory aspects that should be addressed to facilitate the promotion of green antibiotics will be discussed.
ONE WORLD, ONE HEALTH AND “ONE RESISTOME”
In adhering to the principle of One world, One health(i.e., acknowledging the interconnections between animal and human health and the environment) the priorities for veterinary AMD therapy are dictated by public health rather than animal health issues. The tonnage consumption of AMDs in veterinary medicine exceeds that of human medicine (World Health Organisation, 2012), and it is recognized that veterinary medicine contributes to the emergence and spread of AMR in humans. Available epidemiological methods alone are often insufficient to accurately describe the relationships between agricultural AMD use and resistance (Singer and Williams-Nguyen, 2014). Therefore, the veterinary contribution to human AMR remains uncertain, with opinion ranging from globally negligible (Phillips et al., 2004) or irrelevant [an example is resistance to fluoroquinolones in Escherichia coliand non-typhoidal Salmonellazoonosis (de Jong et al., 2012)], to one of a major concern (Collignon et al., 2013). Despite this division of opinion, it is clear that AMD uses in livestock play some role in the emergence, amplification, persistence and transfer of resistance determinants to all ecosystems (Marshall and Levy, 2011) and the main justification in promoting the concept of green antibiotics is to minimize the veterinary contribution to the enrichment of human and environmental resistomes.
WHAT TYPES OF ANTIMICROBIAL DRUG RESISTANCE DOES VETERINARY MEDICINE FACE AND WHICH RAISE PUBLIC HEALTH ISSUES?
Veterinary medicine faces AMR of three types: AMR for specific animal pathogens; AMR for zoonotic pathogens; and AMR of the commensal bacteria harbored by animals (Figure 1).
FIGURE 1. Classes of bacteria developing resistance under the influence of veterinary AMD usage and consequences for animal, human and public health.
The veterinary use of AMDs is inescapably linked to the risk of emergence of AMR in veterinary targeted pathogens, zoonotic food-borne pathogens and the animal commensal microbiota. AMR for the targeted pathogen raises animal health issues only, whereas AMR for zoonotic pathogens and bacteria in the commensal microbiota (principally in the digestive tract) raises public health issues. However, for zoonotic pathogens, the problem, although important, is solely an individual medical problem. For the animal commensal microbiota, on the other hand, the concern is ecological, collective and worldwide. It is this latter issue that drives the requirement for new, green AMDs in veterinary medicine.
Antimicrobial resistance of specific animal pathogens raises veterinary problems in terms of efficacy but, for two reasons, has no direct impact on human medicine: (i) these pathogens (resistant or not) are not in most instances zoonotic and (ii) more importantly, the size of these pathogenic microbiota is negligible, when compared to the size of the commensal microbiota that are collaterally exposed during AMD treatment (see later).
Antimicrobial resistance of foodborne zoonotic pathogens, causing enteritis and diarrhea, such as Salmonella (Salmonella typhimurium and S. enteritidis), Campylobacter jejuni (in poultry), C. coli (in pigs)and some strains of E. coli suchas E. coli O157:H7are of greater concern. This was the case for the in ovo administration of ceftiofur, an emblematic example of AMD misuse in hatcheries, with the aim of improving the sanitary status of chickens (Dutil et al., 2010). However, the emergence of AMR in foodborne pathogens is not the most relevant hazard of veterinary origin for human medicine. According to a recent EFSA report on human cases of salmonellosis and campylobacteriosis in the EU (European Food Safety Authority, 2014b), approximately 200 deaths were attributable annually to these zoonoses. This number is placed in perspective by the claimed 25,000 deaths attributed to AMR annually in the EU (World Health Organisation, 2015). Similar figures were reported in the USA (Centers for Disease Control and Prevention, 2013). Moreover, AMR per seis not responsible for these fatalities, as most zoonotic Salmonellaand Campylobacterof EU foodborne origin are susceptible to the drugs available to treat these infections (European Food Safety Authority, 2014a). Furthermore, most cases of salmonellosis and campylobacteriosis in humans are self-limiting, not requiring antimicrobial treatment. Moreover, outbreaks of salmonellosis, at least in the EU, are decreasing; a 30% decline has been reported over the past 5 years (European Food Safety Authority, 2014b). Therefore, it can be concluded that resistance to zoonotic pathogens is an individual person medical issue and not a global ecological and economic hazard for the future.
The hazards associated with AMR at the level of the animal’s commensal microbiota, i.e., organisms of the GIT and possibly of the skin are potentially much more serious ecologically. This is because their biomasses greatly exceed those of the specific or zoonotic pathogens harbored by the same treated animals (Figure 2). It is likely that the amplification of pre-existing or emerging genes of resistance displays some proportionality with both the size and genetic richness of each category of microbiota, whether pathogenic or not, harbored by treated animals. The commensal microbiota bacteria are not pathogenic but they ineluctably harbor, even before any antimicrobial treatment, a range of genes of resistance (the so-called resistome). The use of veterinary AMDs can promote the selection and amplification of the GIT resistant genes, which may then be transmitted directly (principally by the food chain) or indirectly (via the environment) to man. After gaining access to the human GIT microbiota, these “Trojan horse” bacteria may transmit their resistance genes to the human commensal bacteria by horizontal transfer; these genes of resistance may then be acquired by human-specific pathogens (Angulo et al., 2004) or by opportunistic bacteria such as Enterococcusspp. responsible for nosocomial infections.
FIGURE 2. Bacterial load exposed to AMDs during and after treatment, and the duration of exposure.
One important category of bovine respiratory disease is pasteurellosis. Roof (2011)estimated the pathogen load for the entire lung to be 2 × 107–2 × 108colony forming units (CFU) for Pasteurella multocidaand 9 × 106–9 × 108CFU for Mannheimia haemolytica. Therefore, the estimated pathogen biomass in the lung of an infected cow does not exceed a few mg in toto, whereas the bacterial mass in the animal’s commensal GIT microbiota amounts to several kg. The duration of exposure of the target pathogen can be nil (prophylactic use) or very short (with rapid pathogen eradication during metaphylaxis). In contrast, the duration of intestinal microbiota exposure will never be less than the duration of treatment plus the delay in fully clearing the AMD, together with any newly formed resistant bacteria from the intestinal microbiota, i.e., several weeks (Hansen et al., 2002). The GIT microbiota are continually eliminated at a high rate into the environment, often into an aqueous matrix, thereby allowing further dissemination of the excreted and potentially resistant bacteria. Furthermore, this process will favor horizontal exchanges of resistance factors between organisms within this vast ecosystem. In consequence, this pathway of bacterial elimination (together with their genes) via the excreta is by far the largest connection between animal and human resistomes.
To assess this collateral risk of AMD use in veterinary medicine, AMR in commensal bacteria is being monitored in indicator organisms of food-producing animals, E. colifor the Gram negative microbiota and Enterococcus (Enterococcus faeciumand E. faecalis) for the Gram-positive commensal intestinal microbiota (European Food Safety Authority, 2013). For commensal E.coli, the link between the quantities of the different classes of AMDs administered to food-producing animals in EU and the prevalence of resistance in isolates from cattle, pigs, and poultry was demonstrated (Chantziaras et al., 2014). These epidemiological data drive the proposal that both human and veterinary medicine will benefit from the development of green AMDs by limiting the impact of AMD treatment on the animal commensal resistome and thence on that of humans.
SHORTCOMINGS OF THE PARADIGM OF PRUDENT USE OF AMDs IN VETERINARY MEDICINE
Many recommendations have been made on the prudent use of AMDs in livestock, to mitigate the emergence of AMR by promoting their sustainable use in animals. The most effective decisions, a priori, are those that enforce drastic reductions in the overall consumption of AMDs. However, the reduction of such use has, in some cases, produced unexpected results. For tetracycline resistance in fecal coliforms isolated from swine, the decrease was less than 50%, after the use of all classes of AMD had been discontinued for 126 months (Langlois et al., 1983). In the USA, the decision in 2005 to ban enrofloxacin for metaphylactic use in poultry was not followed by the expected decrease in AMR for Campylobacter. Indeed, by 2010 the prevalence of ciprofloxacin-resistant C. jejuniremained >20% for poultry and human clinical isolates (Food and Drug Administration, 2012).
Among the factors explaining the limited efficiency of a ban or of voluntary restriction of AMD use, the most challenging, is of ecological origin. When a wild-susceptible bacterial population has been replaced in the environment by an antimicrobial-resistant population, having no or low fitness costs associated with the mechanism of resistance, the emergent resistant population can become very stable in its ecosystem (Andersson and Hughes, 2010). Indeed, it was shown for Campylobacterin the USA that some mutations conveying resistance to ciprofloxacin might even provide a fitness advantage (Luo et al., 2005; Zhang et al., 2006). The general conclusion from bans and moratoriums is that retrospective measures will be less effective than tackling the factors leading to AMR emergence and dissemination in the first instance.
When the use of AMDs in animals is justified by welfare and economic considerations, veterinary prescribers have been encouraged to follow guidelines to ensure their so-called “prudent” use. Unfortunately, many recommendations have simply been transposed from human to veterinary medicine, without recognition that they may be inefficient and even counterproductive in respect of public health (Figure 3). As indicated above, the microbiota of public health interest are the animal GIT microbiota and ultimately the microbiota in the environment, rather than the target pathogen. For example, there is no certainty that the priority given to using older AMD classes, such as tetracyclines, qualified by EMA as category 1 (lower risk) drugs (European Medicine Agency, 2014) will be less detrimental to the human and environmental resistomes than, for example, a third generation cephalosporin, specifically developed to have minimal impact on the GIT microbiota and which is rapidly degraded in the environment. Indeed, the poor oral bioavailability of tetracyclines in food-producing animals is a factor leading to extensive animal GIT and environmental exposures. Resistance to tetracyclines is commonly associated with multi-drug resistant bacteria, able to co-select genes conferring resistance to highly critical AMDs for man (Herrick et al., 2014), despite the fact that these AMDs are not marketed or legally restricted in use for food-producing animals.
FIGURE 3. Priority and stewardship for human and veterinary medicine and the paradigm of prudent use of AMDs.
The paradigm of prudent use of AMDs in animals can be insufficient and even counter-productive. This is because such recommendations fail to recognize that the main sources of the resistance determinants, which are amplified by veterinary AMD usage, are derived not from the pathogenic microbiota but from the commensal GIT microbiota. An appropriate stewardship regarding the target pathogen (the priority for human medicine) can actually increase the public health issues when directly extrapolated to veterinary medicine. For example, one recommendation, for the prudent use of AMDs in human and animal medicine, fully justified from both animal and human health perspectives, is the possible increase of dosage regimens for older drugs to comply with current PK/PD concepts. However, this may be detrimental from the perspective public health. A second example of a questionable recommendation is the compulsory recourse to Antimicrobial Susceptibility Testing for AMDs classified as critical for human use, when, in most instances, there are no specific corresponding veterinary breakpoints. Another issue is the marketing of inexpensive generic products in veterinary medicine. Although these have important uses in disease control, there is the possibility that they might be used clinically when (more costly) hygienic, husbandry and disease containment options would be more appropriate [for details see (Toutain and Bousquet-Melou, 2013)]. An a priorisound recommendation is to give preference to local rather than systemic AMD administration, as in the treatment of clinical mastitis or at drying off in dairy cattle. In fact this may be counterproductive as it does not allow for the fact that the waste and unsaleable milk (containing a higher residual amount of AMD than that associated with systemic treatment), is commonly used to feed calves and may be responsible for the emergence of resistance in their GIT microbiota (Brunton et al., 2012; Duse et al., 2013).
THE DISCOVERY AND IMPLEMENTATION OF ALTERNATIVES TO AMDs
Superficially attractive alternatives to AMDs include vaccines, antibodies to specific pathogens, immunomodulatory agents, bacteriophages, antimicrobial peptides and pro-, pre-, or symbiotic products. An example is the marked reduction of AMD consumption in Norway, following the marketing of an efficacious vaccine for the prevention of furunculosis (Midtlyng et al., 2011). Similarly, the use of a vaccine to prevent diarrhea due to Lawsonia intracellularisin pigs led to the reduction of AMD consumption in Danish pigs (DANMAP, 2014). However, for biological, technical, economic, medical and regulatory reasons, vaccines (like many putative alternatives to antibiotics) may be difficult to develop in veterinary medicine [reviewed by (Cheng et al., 2014)]. Moreover, some AMD alternatives can have negative consequences for public health, including the unexpected promotion of AMR. For example, food supplementation with trace elements, such as Zn and Cu, proposed as alternatives to AMDs to control colibacillosis in pigs (Fairbrother et al., 2005; Hojberg et al., 2005) increased the proportion of multi-resistant E. coli in vivoin the enteral microbiome of pigs (Bednorz et al., 2013) and also increased resistance to methicillin in staphylococci (reviewed by Yazdankhah et al., 2014).
WHICH ANIMAL BACTERIAL ECOSYSTEMS PROMOTE AMR OF HUMAN HEALTH RELEVANCE?
The animal bacterial ecosystems possibly exposed to AMDs during treatment, and able to promote AMR of human health relevance, must be identified. In this review, three types of microbial ecosystems are considered, based on the two main hazard factors for the emergence and spread of resistance. The factors are their biomass and the link of each system with the environment: (i) large, open bacterial ecosystems, such as the GIT and skin microbiota; (ii) small, open ecosystems, such as the respiratory tract; and (iii) small, closed ecosystems, such as the udder in cattle. The healthy udder is a closed system with no resident flora and is unable to foster a significant source of AMR during systemic AMD administration. The lung is an open system with no relevant resident microbiota and the bacterial biomass in the lung exposed to AMD during a lung infection is very small, not exceeding a few mg (Figure 2). In contrast, considering the estimated total numbers of prokaryotes in the GIT of some domestic species (Whitman et al., 1998) and assuming an average weight of 1 pg per prokaryote cell, the bacterial biomasses in the digestive tract of a typical pig, cow, chicken, and man are approximately 500, 3,000, 0.2, and 70 g, respectively. Thus GIT biomass is several orders of magnitude greater than that of the target lung pathogens. In addition, the GIT microbiota contain a large genetic diversity, including many resistance genes that can be amplified, and they have a long residence time in the GIT, favoring exchanges of resistance genes. There is, moreover, a regular large scale release of this bacterial population into the environment, thereby potentially impacting on the emergence and/or selection of AMR.
Sludge and manure are waste products exposed to AMDs or their active metabolites, not only during antimicrobial treatments but also long after its completion; the bacterial biomass exposed to AMDs is expressed in tons, not mg as for the target pathogen or kg for the commensal microbiota (Figure 3). Indeed, in cattle feces production rates are 12 kg/day for calves, 26 kg/day for beef and 62 kg/day for milk cows. In pigs, the daily production of manure is 1 to 4 kg and for egg-laying poultry it is approximately 100 g (Hofmann, 2008). In consequence, the risk, when treating a pulmonary infection in domestic species, is not due to AMD exposure of the targeted pathogens but to the unwanted exposure of the intestinal microbiota to the administered drug and beyond, to the persistency of its biological effects in soil and water bacterial populations (van den Bogaard and Stobberingh, 2000; O’Brien, 2002).
WHY VETERINARY ANTIMICROBIAL TREATMENTS ARE ABLE TO ALTER THE RESISTOME OF THE ANIMAL GIT MICROBIOTA
In food-producing animals, the most convenient route of AMD administration is orally (Figure 4); this allows collective treatment at the herd or flock level for prophylaxis or metaphylaxis. Metaphylaxis also termed control in US correspond to the collective treatment of all animal of a group when only a given percentage of subjects of this group display the first signs of infection while prophylaxis is the administration of antibiotic when only a risk factor is present (weaning in piglet, transportation in calves, drying off in dairy cattle…).