BALTIMORE – Members of the Poliovirus Sequencing Consortium, an international collaboration between academic and public health labs aiming to develop next-generation sequencing-based poliovirus detection protocols, have developed a method that can directly detect poliovirus using nested PCR and nanopore sequencing.
Named Direct Detection by Nanopore Sequencing (DDNS), the approach, originally published in 2020 but since improved, can detect and sequence poliovirus from stool and environmental samples, eliminating the need for time-consuming cell culture, which is necessary for the current standard poliovirus detection workflow.
Although poliovirus has been largely eradicated across the world, the virus is still endemic in a small number of countries, said Alex Shaw, a researcher at Imperial College London who helped develop the DDNS method and is part of the Poliovirus Sequencing Consortium. It can also crop up elsewhere, for example in New York last month, where an unvaccinated adult was diagnosed with polio.
According to Shaw, the standard procedures currently adopted by the Global Polio Laboratory Network (GPLN), a group of polio laboratories across the world accredited by the World Health Organization (WHO) to detect and surveil poliovirus, are culture based. With this method, stool or environmental samples, such as sewage, are cultivated in a transformed cell line that is susceptible to poliovirus.
By gauging the sample's cytotoxic effects, indicated by the amount of cell death in the tissue culture, researchers can detect the presence of poliovirus. This is followed by a qPCR step to determine the serotype of the virus. After that, researchers will carry out Sanger sequencing for certain poliovirus samples to retrieve genetic information pertaining to the VP1 capsid region.
One of the major drawbacks of the current culture-based method, Shaw pointed out, is that it's time-consuming. He said the initial culturing step can take about three to seven days, and if Sanger sequencing is involved, the total turnaround time can add up to several weeks.
Adding to that, because there are only a small number of labs within the GPLN, many samples will need to be shipped internationally for analysis, presenting another bottleneck for the rapid detection of the virus.
"Ideally, rather than getting a sample and amplifying the virus, you would receive the sample and then immediately lyse it to kill the virus and then do a direct detection method," he added.
To improve on this, the Poliovirus Sequencing Consortium, which is funded by the Bill and Melinda Gates Foundation, Wellcome Trust, and UK Medical Research Council, aims to come up with NGS-based methods that can bypass the need for culturing and achieve direct poliovirus detection.
The consortium currently consists of about 40 researchers from institutions across the globe, including the National Institute for Biological Standards and Control (NIBSC) UK, Imperial College London, the University of Edinburgh, the University of Nebraska Medical Center, the Institut National de la Recherche Biomédicale of the Democratic Republic of the Congo (INRB), and the National Institutes of Health Islamabad, Pakistan.
According to Shaw, the DDNS method developed by the consortium is built upon a nested PCR protocol that was previously published by a group of Japanese researchers. The workflow starts with RNA extraction from stool or environmental samples. After that, the extracted RNA undergoes a two-step nested PCR reaction, which first amplifies the entire enterovirus capsid region through RT-PCR, followed by a second PCR step with poliovirus-specific barcoded primers that target the VP1 region of the viral genome.
Shaw said the team has designed a set of 96 barcoded primers for the VP1 PCR step, enabling the protocol to multiplex up to 96 samples.
After the nested PCR, VP1 amplicons are sequenced using the Oxford Nanopore Technologies platforms. While the sequencing can be done with any Oxford Nanopore devices, Shaw said the consortium has mainly used the MinIon and the GridIon so far.
Post-sequencing, raw reads are demultiplexed based on sample barcodes and mapped to a poliovirus reference database using the open-source RAMPART software that was previously developed by Andrew Rambaut's group, another collaborating team for the consortium, at the University of Edinburgh. A custom module named realtime-polio is further deployed to generate consensus sequences and identifies mutations in comparison to reference strains.
Shaw said the entire DDNS workflow, from sample to sequence, can be executed in about two days, significantly truncating the detection turnaround time for poliovirus compared with the traditional method. The reagent and sequencing kit cost is about $16 per sample, he added.
Compared with other sequencing modalities, Shaw pointed out a few advantages of nanopore sequencing for detecting poliovirus. For one, he said, the technology's ability to sequence the entire VP1 region, which is crucial for vaccine responses, in a single read is "very handy." Especially for environmental samples, he explained, short-read sequencing might have trouble getting the correct consensus sequence of the sample, given that there are multiple highly related viruses, which is typically the case with sewage.
In addition, Shaw said the relatively low upfront investment and overhead costs for nanopore sequencing make the technology more accessible to many labs, bringing sequencing capabilities to countries that previously did not have them. "You can set up a nanopore sequencing facility with a MinIon very cheaply," he said. "It doesn't [require] quite the same upfront investment that other technologies would."
To benchmark the performance of DDNS, the consortium compared the method with the standard culture-based approach through prospective testing at its collaborating site INRB, a GPLN lab in Congo.
Presenting preliminary data at a webinar hosted by Oxford Nanopore in June, Shaw said the more than 2,300 prospective samples tested at INRB using DDNS or cell culture in parallel showed that the two methods have comparable sensitivity and specificity.
Specifically, he pointed out that of the 27 poliovirus-positive samples detected and sequenced using both DDNs and cell culture paired with Sanger sequencing, the median time between stool collection and the generation of a sequence was 36 days for cell culture versus 14 days for DDNS. In addition, he showed that DDNS was able to confirm three VDPV2 outbreaks during the study period 23 days earlier than the culture-based method.
Despite the promising results, Shaw also noted some challenges that the team needs to overcome in order to further drive the adoption of DDNS, especially in low-resource countries. One of these is the supply of reagents. "Getting the kits out to these countries can be very challenging," he said. "There is definitely a need to improve the shipping of the reagents and kits, so we can actually get them to the labs where they are needed."
Another obstacle, he said, is the lack of mature sequencing IT infrastructure in some of these countries. "We are hoping to work with the labs that we train trying to develop abilities in this area," he said, "so they can process their own data, store it, share it, and publish it."
Moving forward, Shaw said the consortium will continue to optimize the DDNS method and its analysis pipeline to better cope with environmental samples harboring multiple closely related virus species. As Oxford Nanopore continues to roll out new sequencing chemistries, Shaw said the team will make updates on the method's protocols, adding that it has observed improved accuracy for the raw reads with the evolution of Oxford Nanopore's reagents and software.
Moreover, Shaw said, the consortium will work to train more labs across the globe on how to use DDNS in order to further drive adoption of the method.
"The hope in the long run would be that [DDNS] gets adopted by the GPLN, so that labs take this on themselves," he said.