Exploring resistomes in microbial communities

How diverse are the antibiotic resistance genes in our microbial communities? Do human resistomes differ among nations, or according to lifestyle, diet, age, medicament use or health status? How does the resistome in humans compare to those in animals and in the environment? Are they interrelated? These and other questions regarding the human resistome have been propelled by the urgent need to better understand antibiotic resistance. The first studies addressing the diversity of resistance genes in human microbial communities relied mostly on culture, in which microbes are grown in the laboratory using selective media. Culture-dependent techniques enable direct testing of antibiotic resistance. However, it involves the isolation of pure cultures and the mapping of resistance to a panel of antibiotics, which requires several growth steps.

These limit the scalability required for the study of the complex microbial communities found in humans. Also, it does not account for bacteria in microbial communities for which conditions for growth in the laboratory remain unknown. With advancements in sequencing technologies and computational methods, it has become possible to achieve the larger-scale requirements for resistome studies of complex microbial communities. One of the pioneering studies used the whole metagenome shotgun sequence to compare human gut resistomes from individuals who live in China, Spain and Denmark. The study showed that Chinese individuals harboured a higher abundance of antibiotic resistance genes than those from Spain and Denmark.

It also showed that some of the same genes in samples from China had different sequence signatures compared to the other two countries. And in the samples analysed, resistance genes accounted for 0.7, almost 1% of the total gut genes. From 2013, when this was first published, studies on the resistomes of human microbial communities have increased exponentially, and benefited from the improvement in analytical tools and reduction in sequence costs. How does whole metagenome shotgun sequencing work? Whole metagenome shotgun provides sequence information for the entire genome content in the sampled microbial community. The workflow can be divided into four main steps. First step, the experiment is designed and samples are collected, followed by DNA isolation, library preparation, and sequencing.

The second step is computational preprocessing. In this step, quality filtering is performed, including, for instance, removal of adaptors, and in the case of human samples, removal of sequence corresponding to human DNA. And the third step is analysis of the clean reads. In read mapping, which is the most common approach, the clean reads are aligned to taxonomic antibiotic resistance gene reference databases. This provides coverage, and that information of the represented genes in the samples. And finally, the fourth and last step is the downstream analysis. In this step, the date is explored using statistics and visual tools. This approach has, however, some limitations. For instance, sequence costs are still high, despite a dropping in prices.

It’s also demanding in terms of computational resources, and there are challenges in identifying novel resistance genes. Also, the downstream step in the analytical process is usually best addressed by multidisciplinary teams, but these usually require computational skills that most of the team is not familiar with. For this, new resources have been developed for downstream analysis exploration by researchers and clinicians with little or no bioinformatics training. I hope that this introduction to the tools for exploring human resistomes will help you in understanding the principles already being used in groundbreaking studies in the field. These are starting to provide an overall picture on the emergence and dissemination of antibiotic resistance genes in humans and across species and environmental boundaries.

As the technology advances and with the continued reduction in sequencing costs, it’s expected that such knowledge will help individualise antibiotic therapies and increase its applicability in surveillance. These are exciting times.