Nanopore sequencing for disease surveillance in low-resource settings — updates from London Calling 2023

Nanopore sequencing for real-time genomic surveillance of Plasmodium falciparum

William Hamilton
Wellcome Sanger Institute, UK, and Cambridge University Department of Medicine, UK

Dr Lucas Amenga-Etego
The West African Centre for Cell Biology of Infectious Pathogens (WACCBIP), University of Ghana

William Hamilton, from the Wellcome Sanger Institute, UK, presented the results of his research collaboration with Dr Lucas Amenga-Etego from the West African Centre for Cell Biology and Infectious Pathogens, using nanopore sequencing to explore resistance of malaria parasites to antimalarial drugs¹. Malaria, caused by Plasmodium falciparum, is a eukaryotic parasite that is transmitted to humans via Anopheles spp. mosquitoes, and can cause serious disease and death. William described how there were over 600,000 deaths attributed to malaria in 2021, with the greatest burden occurring in sub-Saharan Africa, where 95% of global malaria deaths occurred that year; with 76% of recorded deaths occurring in young children.

There are drugs available for treatment of malaria, but in some places, notably southeast Asia, the parasite has become resistant to multiple antimalarial drugs, including to certain forms of the current front-line treatment, ACT (artemisinin-based combination therapy). Worryingly, parasites with partial resistance to artemisinins have been reported in East Africa in recent years. William commented that ‘The WHO has highlighted this as a major issue’ and emphasised the importance of surveillance in order to monitor for the emergence and spread of resistance. However, checking for the presence of resistance in malaria parasites is difficult, as traditionally they need to be isolated from blood samples and tested in a laboratory — by contrast, genomics allows rapid access to resistance information without the need for lab culture. In addition to drug treatments, William explained that the first malaria vaccine (RTS,S) has recently been implemented in Ghana, Malawi, and Kenya, with plans for further rollout. The vaccine targets an antigen within the parasite called circumsporozoite protein (CSP); monitoring any changes in the diversity of the vaccine antigen target gene that occur as the vaccine is rolled out is one of the goals of the project. William explained that the further goals of the project were to perform nanopore sequencing in Ghana for malaria genomic surveillance, to describe the prevalence of drug resistance markers and also to provide training and capacity-building for nanopore sequencing.

‘Absolutely everything from sample collection, to processing, to sequencing, to bioinformatics analysis and final interpretation, all needed to be done in Ghana.... and we wanted to build a platform that was sustainable, that had a pretty simple and straightforward workflow that could be implemented in endemic settings’

The sequencing workflow consisted of DNA extraction followed by multiplex PCR and library preparation with the Native Barcoding Kit, and sequencing using a MinION with R10 MinION Flow Cells. Real-time basecalling was performed in super accuracy (SUP) mode with the Guppy basecaller and subsequent bioinformatics analysis involved the mapping of reads to reference sequences and variant calling with Medaka. Before testing the workflow at the sample source, the team validated their assay using mocked-up isolates with parasite gDNA spiked into human clinical research samples. They found that even from low-volume blood research samples and at very low infection levels, the resistance genotype was correctly called with nanopore sequencing. William remarked that ‘we know what the genotype should be for all of these lab isolates because they are very well described, and we found complete agreement using R10 sequencing chemistry’. Although the data used in this study were from venous blood clinical research samples, William commented that they had tested the workflow on dried blood spots, which consist of small volumes of blood, and the assay worked very well.

The workflow was then tested at two sites in Ghana: Accra, the capital city, and Navrongo, a rural town in the north. During the study period, the team collected 142 venous blood research samples and analysed 109 using the nanopore sequencing workflow.

William went on to summarise the drug resistance results, remarking that they found very high coverage using their amplicon sequencing assay, with over 1000x coverage for each amplicon. Reassuringly, they found no parasites that were resistant to the current frontline antimalarial, artemisinin. However, there was a high frequency of resistance to sulfadoxine and pyrimethamine, which, William explained, are used together prophylactically to prevent malaria during pregnancy. Although ‘high-level’ resistance to the combination, which reduces the protective effect in pregnancy, was not found, the results highlight the need for ongoing surveillance. One particularly interesting result was the discovery that chloroquine resistance in malaria parasites, which was common in Ghana a decade ago, was not observed, and the majority of parasites sequenced were chloroquine susceptible. William explained that this may be due to the fact that chloroquine has not been used to treat malaria in Ghana for many years, so the resistance genotype has reverted.

William went on to discuss the results of the vaccine antigen (csp) genotyping. The csp gene has a repetitive region which is difficult to sequence using short-read technologies, but William observed that when sequencing the isolates, he achieved ‘100% base-perfect sequence at the consensus level, including going right the way through that highly repetitive region’ and how this is ‘another good example of where nanopore would give an edge over short-read technologies to access these kind of complex antigens’. When assessing mutations along the C-terminal domain of csp – an important region for T-cell epitopes included in the RTS,S vaccine – the SNP variations in this region had good concordance with those discovered using short-read sequencing, with no statistically significant differences between the technologies.

William ended his talk by explaining that as part of the project, training in the use of nanopore sequencing was provided in Navrongo, from extraction through to analysis. This provides capacity strengthening and the potential for cross-application to other pathogens. The portable MinION device made it possible to rapidly implement the end-to-end sequencing workflow in an endemic setting. William concluded that this ‘brings data generation, and data processing and analysis and interpretation right to the data users, to the places where this disease is having its greatest impact— which, in my view, is the pathway to maximising the impact of genomics, also to make the genomics field more equitable rather than having samples shipped out of countries into the Global North’.

Pathogen and species identification using a mobile suitcase laboratory

Ahmed Abd El Wahed
University of Leipzig, Germany

Ahmed Abd El Wahed began his talk by introducing the ‘mobile suitcase lab’ which allows extraction, library preparation, and nanopore sequencing on a MinION Mk1C to be performed in low-resource areas using equipment and consumables contained in a single suitcase-sized container. Ahmed explained that this provides benefits in terms of sample preservation and reduced need for sample storage and transportation. This is especially useful in low-resource settings and the mobile suitcase laboratory has been used in many countries for a variety of reasons, including to identify pathogens from different sample types and for microbiome analysis. His colleagues went on to explain further.

Ahmed’s colleague, Rea Kobialka from Leipzig University in Germany, explained that in partnership with one German partner and seven sub-Saharan partners, they are aiming to identify antimicrobial resistance (AMR) in poultry, humans, and livestock. The suitcase kit allows them to sequence samples at the point of need. Arianna Ceruti, also from Leipzig University, explained that their extraction approach uses a magnetic bead-based system to purify the samples and that the extraction is performed all in one tube and takes about 20 minutes; so far, the group has tested the method on tissue samples for Leishmania spp. and on mosquito samples. Arianna explained that the nucleic acids obtained were 'purified enough to directly sequence within your device and offer really accurate results’ and emphasised the rapid time to result. Rea added that using the Rapid Barcoding Kit for multiplexing allows many samples to be processed together, providing further time savings as well as reducing the use of laboratory equipment.

Arianna stated that ’data analysis is the most crucial step in the whole sequencing process’ and that they needed to be able to perform data analysis directly in the field, where there may be no access to an internet connection; nanopore workflows enable offline analysis. One of the projects of the group requires sequencing of mosquitoes in the field, so they developed three different analysis pipelines — for mosquito species identification, detection of pathogens in mosquito samples, and identification of the species that comprise the blood meals. The software is user-friendly and Arianna remarked that even non-bioinformaticians can analyse the data in the field.

Kamal Eltom and Sanaa Idris from the University of Khartoum in Sudan went on to explain that the suitcase kit is being used in Sudan for whole-genome sequencing of Mycobacterium avium subsp. paratuberculosis from different livestock species, such as cattle, sheep, goats and camels, as well as from humans. Wisal Elmagzoub, also at the University of Khartoum, is investigating the faecal microbiome in samples from humans and cattle. Wisal explained that when using the MinION Mk1C device for metagenomic analysis, EPI2ME analysis workflows provided real-time results, and classification up to species level.

‘This technology suits our infrastructure in the Sudan and we are successful in doing whole-genome sequencing of some bacteria and also studying the microbiome of animals and humans’ — Kamal Eltom, University of Khartoum, Sudan.

More than 1.6 million deaths occur globally every year due to diarrhoeal diseases, disproportionately affecting developing countries such as Bangladesh, where these diseases represent a major public health challenge due to inadequate access to healthcare and safe drinking water. Early diagnostics are essential to avoid mortality and morbidity from diarrhoeal diseases but many healthcare facilities in Bangladesh do not have access to the resource- and time-demanding laboratory tests traditionally used to characterise the pathogens that cause diarrhoea. Disease sufferers are often treated with antibiotics without the causative pathogen having been identified, contributing to the burden of AMR in these countries. At the International Centre for Diarrhoeal Disease Research in Bangladesh, researchers are using the mobile suitcase laboratory to test the future potential of nanopore sequencing for rapidly identifying pathogens in faecal specimens.

Antimicrobial resistance wastewater epidemiology in low-resource communities

William Strike
University of Kentucky, USA

William Strike explained how his work using nanopore sequencing for wastewater-based epidemiology began by tracking  SARS-CoV-2 signals from wastewater on the University of Kentucky, USA, campus during the pandemic. Using this information, he was able to track infections back to specific university buildings, so that SARS-CoV-2 testing could be performed to identify infected individuals and prevent outbreaks. In order to collect samples from dormitories across Kentucky, wastewater samples had to be collected and shipped, but this resulted in increased time to result while samples were in transit, and problems retaining sample viability due to temperature-related degradation.

William introduced the solution to these problems — outfitting a transit van as a mobile BSL2 laboratory where the sample preparation, PCR, and nanopore sequencing could be performed in situ. This resulted in a faster time to result from days to within hours of sample receipt, due to a drastic reduction in sample degradation and the ability of nanopore sequencing to provide real-time data. The extraction method performed in the van utilised an exclusion-based sample preparation (ESP) method, in which silica-coated paramagnetic beads capture and concentrate nucleic acids in a reusable plate; the resulting prepped sample was more stable at room temperature than unprocessed samples. This extraction method reduced the use of plastic consumables by 40% and the preparation time by 70%, compared to traditional extraction techniques. William emphasised that these benefits were particularly important in low-resource settings, as well as the further benefits of a minimal reliance on electricity, pipetting, and prior experience.

The next target for the team was to characterise AMR in the United States. William noted that in the Appalachian region of the USA, there is a problem with over-use of antibiotics. This was a particular problem after the COVID-19 outbreak, with up to 70% of sufferers in the USA being prescribed antibiotics between 2019 and 2020, despite only 4% being thought to have a bacterial infection. Low-resource communities were disproportionately affected due to poor access to healthcare.

William and his team had access to an AMR enrichment panel which utilised short-read technology to identify the presence of AMR genes in wastewater. While it performed well in a pilot study, the cost was too high to be feasible for the wastewater project and William explained that the need to bring the cost down to a point that is conducive to the research environment in Eastern Kentucky compelled them to use Oxford Nanopore sequencing. After sample extraction, library preparation was performed using the Native Barcoding Kit and sequenced with Flongle and MinION Flow Cells, using the R9 and R10 chemistries. As Eastern Kentucky is prone to flooding and tornadoes, William is particularly interested in characterising AMR in Vibrio species, which can be a public health issue following natural disasters. He is currently collecting preliminary data on the sulfonamide resistance gene (sul1) commonly associated with Vibrio, using nanopore sequencing and EPI2ME analysis workflows. This will allow him to rapidly characterise newly emerging pathogens or resistant bacteria. William emphasised the importance of having access to offline analysis options during such projects, as they may be sequencing at locations where there is no access to consistent internet coverage.

William went on to describe a global initiative he is involved with, in partnership with the University of Arizona: the Kentucky-Arizona Pandemic Prediction Organizing Workshop (KAPPOW). The goals of this project are to set up sequencing centres in areas of the world where they are needed the most, for example, in areas that are at higher risk of future pandemics or natural disasters. William remarked that ‘a lot of the issues that we’ve overcome in Eastern Kentucky, with figuring out an on-site extraction process that’s simple and cheap and low form-factor to ship out to these places, is going to come in handy’. As part of this project, William is actively participating in a partnership with researchers in Uganda to set up sequencing capability there, including delivering training in nanopore sequencing to researchers and performing a pilot study. He explains that one of the most important aspects of these partnerships is to empower partners in low-resource communities to perform the data analyses on site themselves, so that the information is kept within the community.