By Teresa Street, Senior Research Scientist
My name is Teresa Street and I’m a senior postdoctoral scientist in the Modernising Medical Microbiology (MMM) research group, part of the Nuffield Department of Medicine at the University of Oxford, UK. MMM aims to transform how we analyse and treat infections so we can improve patient care. For the last few years, we’ve worked on studies that try to develop better ways of detecting bacterial infections and predicting antibiotic resistance using DNA sequencing.
Our latest study focused on detecting gonorrhoea in patient samples. Gonorrhoea is a common sexually transmitted infection, caused by the bacteria Neisseria gonorrhoeae. It is usually successfully treated with a combination of two antibiotics. Increasingly, there are more cases arising where these two antibiotics are no longer effective at treating gonorrhoea, as the bacteria has developed resistance to them. This resistance has been detected globally, making gonorrhoea a significant public health risk.
To tackle this problem, we need to be able to identify and treat infections quickly: early detection and faster treatment will help control the spread of antibiotic-resistant strains. Gonorrhoea is usually detected by collecting urine or a swab sample from patients and then growing any bacteria contained within the samples in a laboratory. If bacteria do grow, further tests can be done to identify which antibiotics will successfully kill them. Scientists also do molecular tests (PCR), which involve trying to detect DNA from the gonorrhoea bacteria. It can take a while to get the results back from all of these tests as often bacteria can take a few days to grow. A single test that could be done much faster, and which would identify both whether a gonorrhoea infection is present and which antibiotics would treat it best would allow the correct treatment for each patient to be started sooner. This, in turn, would reduce the onward transmission of gonorrhoea. This is particularly important in those cases where the bacteria is resistant to multiple antibiotics.
Recently, a molecular method called Metagenomic Sequencing, or mNGS, has shown potential as a new diagnostic test. It utilises next generation sequencing (NGS) technologies to identify DNA from bacteria directly from patient samples, without needing to grow the bacteria in a laboratory first. The bacterial DNA can be identified by comparing it to a database of many known bacterial sequences, and this helps to identify the bacteria causing an infection. If we can extract enough bacterial DNA from a sample, not only can mNGS work out which type of bacteria is causing the infection, but it can also identify specific parts of its DNA that we know lead to antibiotic resistance (AMR). mNGS can also be faster than current tests, often detecting the cause of an infection and which treatments will work within a few hours. In this way we have the potential for a single test that both identifies the cause of infection and gives us information about which antibiotics will (or won’t) treat it much faster than current methods.
In a previous study we tested the ability of mNGS to detect N. gonorrhoeae directly from urine samples. We were able to detect gonorrhoea and in some cases also see some AMR determinants. mNGS can, however, sometimes be hampered by high levels of host contamination: DNA extracted from a clinical sample will contain DNA from the patient as well as from any bacteria. This limits the detection of bacterial DNA in our sequence data and makes it more difficult to identify AMR determinants. We observed this in our previous work, and so our latest study tested a method to enrich for gonorrhoea before sequencing.
We used a technique called Target Enrichment to capture any gonorrhoea DNA in our extracts before sequencing. This involves designing probes – short sequences of RNA that are complementary to specific regions of the target DNA. In this case our target DNA was the N. gonorrhoeae genome and we also focussed on known AMR determinants, including probes that match to these known sequences. By mixing the probes with DNA extracted from patient samples they will selectively hybridize, or bind, to their complementary gonorrhoea target sequences. Subsequent steps remove the unbound DNA (which we hoped would be human and any other bacterial DNA), increasing the relative abundance of the gonorrhoea DNA compared to non-target DNA in the sample. In this way we should be able to enrich for gonorrhoea over the human DNA.
We tested this enrichment method for gonorrhoea-positive urine and urethral swab samples. Our results demonstrated a substantial improvement in the proportion of DNA sequences classified as N. gonorrhoeae in comparison to the same sample without enrichment. This enhanced genome coverage enabled detection of AMR determinants in chromosomal genes that are known to confer resistance, and we were able to predict the resistance seen by the laboratory to certain antibiotics in our samples. We also tested the feasibility of multiplexing, where multiple samples were pooled, enriched and sequenced simultaneously, to improve efficiency and reduce the costs associated with enrichment and sequencing. We obtained enough genome coverage to detect AMR determinants in these samples, too.
We hope our results have shown the usefulness of enrichment for detecting gonorrhoea directly from patient samples without needing to culture it in the laboratory first. We were even able to detect an AMR determinant in a sample which did not grow in the laboratory and so didn’t have a recommendation for which antibiotic would be best to kill it. We think this really highlights the utility of mNGS for looking at infections where bacteria are difficult to grow in a laboratory.
Original article posted in:
https://microbiologysociety.org/blog/using-genomics-to-identify-antibiotic-resistant-gonorrhoea.html
