SEARCH

SEARCH BY CITATION

Keywords:

  • Antibiotic resistance;
  • bacterial evolution;
  • microbial ecology;
  • subinhibitory concentration

Abstract

  1. Top of page
  2. Abstract
  3. Transparency Declaration
  4. References

Antibiotics are likely the most important compounds used for human therapy. Conversely, antibiotic resistance is a relevant medical problem. However, besides their relevance for human health, antibiotics and their resistance genes are important elements that can influence the structure of microbial populations. In this article, we discuss antibiotics and antibiotic resistance genes in non-clinical enviromnents.

Antibiotics (including synthetic antimicrobials) are probably the most successful therapeutic agents developed by humans. Although sulphonamides, the first widely used antibiotics, are synthetic, the large majority of these compounds have a natural origin. Furthermore, production of antibiotics is a frequent characteristic of environmental microorganisms. Conversely, antibiotic resistance is widely disseminated in non-clinical environments, and antibiotic producers contain a large number of antibiotic resistance genes in their genomes [1] that might eventually be transferred to pathogenic bacteria.

Antibiotic producers are able to outcompete, at least under laboratory growing conditions, other microorganisms [2]. Thus, a suitable ecological role for these molecules in nature could be the Darwinian struggle for life against competitors. As Waksman and Woodruff stated in 1940, the reason for searching for antibiotic producers in soil was the assumption that this type of organism must be present in the environment and that production of antimicrobials should have an ecological role in inhibiting the growth of other microorganisms: ‘The soil was searched for bacterial agents of infectious diseases, until the conclusion was reached that these do not survive long in the soil. It was suggested that the cause of the disappearance of these disease-producing organisms in the soil is to be looked for among the soil-inhabiting microbes, antagonistic to the pathogens and bringing about their rapid destruction in the soil’ [3].

Following the same reasoning, the presence of large numbers of antibiotic resistance genes in natural (non-clinical) environments strongly suggests that they must have an ecological role, either in endogenous detoxification in the case of producer organisms, or in protecting bacteria from the action of exogenous antimicrobials.

This ‘weapon/shield’ function for antibiotics and their resistance genes is useful for pathogenic bacteria growing in a treated host. This is an ecosystem with high concentrations of antibiotics, where the unique function of these molecules is to inhibit bacterial growth. Being resistant in an environment with a high load of antibiotics is a rewarding strategy because it allows the utilization of resources in the absence of competitors, which are being killed by the antimicrobial. Antibiotic resistance can be considered as a colonization factor in the presence of antibiotics [4].

Do antibiotics and antibiotic resistance genes play the same role in non-clinical environments? This may be true in some instances, but compelling evidence indicates they may have very different roles in mediating the interactions of microbial communities in natural environments. There are two important key concepts. First, the toxicity of a molecule is always concentration-dependent. Furthermore, multiple compounds are beneficial at low concentrations and toxic at high ones, an effect that has been called hormesis [5]. For example, iron is a cofactor in several biomolecules and is required for life. However, at high concentrations, iron is toxic. For this reason, all organisms have developed complex and exquisitely regulated mechanisms for acquiring iron and avoiding its toxic effects. Second, production of antibiotics (and, thus, their concentrations in natural, non-clinical environments) is highly regulated in the producer organisms (a situation that obviously does not occur during clinical treatment of infection). Similarly, the expression of antibiotic resistance genes is finely tuned in the organisms in which they evolve, whereas this regulation does not occur after horizontal gene transfer to a pathogenic bacterium.

The concentrations of antibiotics in natural environments is frequently too low to play their proposed functional role as inhibitors of competitors. Antibiotics are frequently produced at stationary growth phase, not during active growth when the production of a weapon to inhibit competitors would be a rewarding ecological strategy. As antibiotics are usually present at low concentrations in natural environments, studies on the effect of antibiotics at subinhibitory concentrations may give insights into the role that these molecules play in natural ecosystems [6]. These studies have demonstrated that each antibiotic produces a rather specific transcription pattern in global bacterial gene expression [7]. Functional analyses have also demonstrated that antibiotics can trigger expression of phenotypes, such as biofilm production or cytotoxicity, that can give an ecological advantage to bacteria growing under low antibiotic concentrations [7,8]. From these studies, it has been suggested that antibiotics may be signalling molecules mediating intercellular communication in natural environments [8]. The more classic role of ‘weapon’, with production of high concentrations of antibiotic, might occur in some local micro-environments in the rhizosphere.

If antibiotics are not always weapons, it is possible that the natural function of some antibiotic resistance genes is also other than to encode shields. Bacteria acquire resistance by a few mechanisms (Fig. 1), some of them as the consequence of mutations [9], and some due to horizontal gene transfer [10]. The most frequent antibiotic resistance genes acquired by horizontal gene transfer are antibiotic-inactivating enzymes and efflux pumps. It was earlier stated that aminoglycoside-inactivating enzymes could have a metabolic role, in addition to a function in resistance, because ‘phosphorylated streptomycin might be important as a metabolic precursor’ [11]. More recent work has shown that some aminoglycoside-modifying enzymes may be involved in acetylation of peptidoglycan [12]. The similar structure of the enzymatic substrate and the antibiotic would allow these enzymes to confer resistance even though they evolved to play a completely different role. The investigation of sequenced microbial genomes indicates that this could be a frequent situation, because all bacteria contain in their genome antibiotic resistance genes, whether the bacteria produce antibiotics or not. As these genes are present in all members of each species, they have not been acquired recently by horizontal gene transfer as the consequence of antibiotic pressure from antibiotics used by man and probably have functional roles different to resistance.

image

Figure 1.  Mechanisms of antibiotic resistance: (A) A schematic representation of a Gram-negative bacterium with two membranes. To inhibit microbial growth, an antibiotic needs to cross the cellular envelopes (1 and 2), on some occasions to be modified (3) and to reach its target (4) at concentrations high enough for its effective binding. (B) Mutations in all the elements involved in this pathway can produce resistance because the membrane transporters suffer structural changes (a) or are not expressed (c), because there is a modification (e), or lack of expression of, the enzyme that activates the pre-antibiotic or because the target presents structural modifications (f) that impede the antibiotic’s binding. Antibiotic-inactivating enzymes (b) and efflux pumps (d), either present in the bacterial genomes or acquired by horizontal gene transfer, also contribute greatly to antibiotic resistance.

Download figure to PowerPoint

Multidrug resistance efflux pumps constitute another category of antibiotic resistance genes for which a functional role other than resistance has been proposed. Common substrates of efflux pumps are the quinolones, and it has been demonstrated that environmental bacteria isolated before the discovery of these synthetic drugs were capable of extruding quinolones [13]. Notably, some multidrug pumps from Pseudomonas aeruginosa can extrude signal molecules involved in quorum sensing in this bacterial species, including one natural quinolone-like compound. Overexpression of those pumps increases resistance to several antibiotics and simultaneously leads to a reduction in the quorum sensing response and virulence [14–16]. It is thus likely that these efflux pumps might be involved in the bacterial response to intercellular signalling.

In summary, we propose that antibiotics may be signalling molecules (and not only weapons) in natural environments. Conversely, antibiotic resistance genes could have metabolic roles (and act not only as shields) that may include biosynthesis of macromolecules, maintenance of homeostasis, and signal trafficking.

Transparency Declaration

  1. Top of page
  2. Abstract
  3. Transparency Declaration
  4. References

This work was supported by grants BIO2005-04278, LSHM-CT-2005-518152 and LSHM-CT-2005-018705. A. Fajardo is the recipient of a fellowship from the Spanish ‘Ministerio de Educacion y Ciencia’. All authors declare no conflicts of interests.

References

  1. Top of page
  2. Abstract
  3. Transparency Declaration
  4. References