Barriers to the spread of resistance

Despite identifying abundant genes capable of conferring antibiotic resistance in soil microorganisms, a study finds that few are shared by human pathogens and that there is little transfer of the genes within the soil communities. See Letter p.612

Antibiotic resistance is complicating the treatment of many bacterial infections, leading to increasing mortality and health-care costs across the globe. Pathogens frequently acquire resistance to antibiotics from other sources, and soil-dwelling bacteria are considered an important reservoir of resistance genes. In this issue, Forsberg et al.1 (page 612) identify nearly 3,000 genes conferring antibiotic resistance from the soil, and find that only a minute fraction of these genes is shared with human pathogens. Furthermore, their data suggest that it is the mobilization of and selection for such genes, rather than their supply, that limits their transfer among soil bacteria and with other bacteria, including human pathogens.

The genome sequencing of thousands of bacterial pathogens has shown that antibiotic-resistance genes are often acquired by pathogens from other sources through horizontal gene transfer2. This process allows even distantly related bacteria to transfer genes through the action of viruses (bacteriophages) or conjugative plasmids (small DNA molecules separate from chromosomal DNA that have the machinery to facilitate transfer between bacteria), or through uptake of free-floating DNA in the environment. All of these mechanisms are known to contribute to the emergence of antibiotic resistance in human pathogens, yet the origin of most clinically relevant resistance genes remains elusive. Accordingly, research efforts are directed at addressing this question through the elucidation of antibiotic-resistance genes harboured by different microbial communities3.

Several soil bacteria can produce antibiotics, and these bacteria also have the necessary genes to confer immunity against the toxins they produce. Accordingly, it has been proposed2 that these antibiotic producers could be an origin of antibiotic-resistance genes. Subsequent research has also pointed to other soil microbes that might serve as sources of resistance genes, including bacteria that can subsist on antibiotics4. Yet, in spite of substantial research in this area, only a few studies have demonstrated a link between resistance genes in the soil and resistance genes in human pathogens5.

Forsberg and colleagues deployed functional metagenomics to study soil microbial communities and characterize genes that confer antibiotic resistance on a non-resistant strain of the bacterium Escherichia coli. They identified nearly 3,000 resistance-conferring genes, which is a number comparable to all currently known antibiotic-resistance genes6. Thus, this study shows clearly that an extraordinary diversity of antibiotic-resistance genes exists in nature, as suggested by previous analysis of soil microbial communities7.

An earlier study5 from the group presenting the current paper reported the first case of the transfer of several drug-resistance genes between innocuous soil bacteria and human pathogens, highlighting that transfer of genes between such bacteria is possible. But that study included bacterial-enrichment steps that prevent quantification of the extent of such transfer. In the present study, the authors used a method that did not require enrichment and which allowed them to quantify the extent of antibiotic-resistance genes that are shared between soil bacteria and previously characterized bacteria.

They found that only around 0.1% of the identified resistance genes from soil are highly similar (greater than 99% nucleotide identity) to previously detected resistance genes, indicating that there is only limited overlap between the resistance genes of soil bacteria and other bacteria, including those that cause infections in humans (Fig. 1). Although this low overlap does not exclude the possibility of soil bacteria acting as an origin of antibiotic-resistance genes that cause clinical problems, it does demonstrate that only a minute fraction of resistance genes from soil bacteria have been transferred to human pathogens.

Figure 1: Limited transfer.

Forsberg and colleagues' metagenomic analysis of soil microorganisms1 identified around 3,000 genes capable of conferring antibiotic resistance, but found that less than 0.1% of these genes have been identified as associated with antibiotic resistance in bacteria isolated from human patients. They also show that fewer resistance genes in the soil bacteria are flanked by mobilization elements than those in pathogens, suggesting that there is limited transfer of these genes within the soil community and from the soil to other bacteria.

Forsberg et al. also investigated whether the limited overlap might result from limited transfer of antibiotic-resistance genes within the soil microbial community. If this is the case, specific resistance genes should be stably associated with specific phylogenetic divisions. The authors show that this is correct and conclude that the resistance-gene pool of different soil communities is closely linked to the phylogenetic architecture of those communities. The authors were not able to resolve the phylogenetic architecture beyond the phylum level, and so horizontal transfer of genes within a specific phylum cannot entirely be ruled out. However, they show that soil bacteria, in contrast to human pathogens, have a much lower number of mobilization elements flanking their resistance genes, which supports their hypothesis of limited transfer of resistance genes between soil bacteria. These results are consistent with the hypothesis that there is limited selection for antibiotic resistance within the soil microbiota compared to the selection for antibiotic resistance in human pathogens.

These findings fuel the ongoing question of what is the function of antibiotic-resistance genes in their natural hosts. For instance, the MFS transporter proteins identified by the authors as conferring resistance to a wide range of antibiotic classes may not actually function as antibiotic-resistance proteins in their hosts, but rather in different processes, such as the transport of other small molecules that may be more abundant than antibiotics in the soil. Similarly, the identified β-lactamase enzymes might serve as cell-wall remodelling enzymes in their natural hosts. The apparent paucity of mobilization elements flanking these genes would suggest that selection for and transfer of resistance functions in the soil is not as strong as in other environments.

Irrespective of the function of these genes in their natural hosts, Forsberg and colleagues' study demonstrates that the soil microbiota harbours an extraordinary diversity of genes that have the potential to confer antibiotic resistance in human pathogens such as E. coli. Their findings also suggest that it may not be the availability of genes encoding proteins capable of conferring antibiotic resistance that limits the spread of resistance, but rather the mobilization and transfer of these genes. Functional metagenomic studies of soils that have been exposed to inhibitory concentrations of antibiotics should be performed to test whether this increases the extent of resistance-gene mobilization.


  1. 1

    Forsberg, K. J. et al. Nature 509, 612–616 (2014).

    ADS  CAS  Article  Google Scholar 

  2. 2

    Davies, J. & Davies, D. Microbiol. Mol. Biol. Rev. 74, 417–433 (2010).

    CAS  Article  Google Scholar 

  3. 3

    Allen, H. K. et al. Nature Rev. Microbiol. 8, 251–259 (2010).

    CAS  Article  Google Scholar 

  4. 4

    Dantas, G., Sommer, M. O. A., Oluwasegun, R. D. & Church, G. M. Science 320, 100–103 (2008).

    ADS  CAS  Article  Google Scholar 

  5. 5

    Forsberg, K. J. et al. Science 337, 1107–1111 (2012).

    ADS  CAS  Article  Google Scholar 

  6. 6

    McArthur, A. G. et al. Antimicrob. Agents Chemother. 57, 3348–3357 (2013).

    CAS  Article  Google Scholar 

  7. 7

    Riesenfeld, C. S., Goodman, R. M. & Handelsman, J. Environ. Microbiol. 6, 981–989 (2004).

    CAS  Article  Google Scholar 

Download references

Author information



Corresponding author

Correspondence to Morten O. A. Sommer.

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Sommer, M. Barriers to the spread of resistance. Nature 509, 567–568 (2014). https://doi.org/10.1038/nature13342

Download citation

Further reading


By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.


Quick links

Nature Briefing

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing