Variants of concern - the turning point

At the start of COG-UK, there was uncertainty over what the genomic data would reveal and how it could be used. Some of this uncertainty was rooted in the fact that coronaviruses typically evolve slowly and the same was expected with the SARS-CoV-2 virus. Dr Andrew Page, who headed up the Consortium's effort at the Quadram Institute, admits 'to be honest, we didn't really know if this stuff would be useful at all. We just assumed that everything would more or less look the same, there would be very little diversity. We weren't really sure it would actually work'. To start with, he says, he and his colleagues went in 'blind in the hope that we might be useful. Actually, we didn't even know what to do with the results, or how it would be useful locally' (Page transcript).

Similar sentiments were expressed by Dr Richard Myers at Public Health England [PHE}. He says, 'I think it would be fair to say there was a degree of uncertainty about what we'd be able to do with that data. Not so much around the fact that we would be able to generate it, but when SARS-CoV-2 came, when we started looking at genomes, coronavirus would be very homogeneous. So you've got 29,000 base pairs, but actually the amount of variation that we observed within those genomes was very little, certainly at the beginning' (Myers transcript).

While it was obvious at the start that the diversity of the virus would increase over time, it took months before it became clear how the genomic data could be used. The turning point came in December 2020 when the first variant of concern (VOC), Alpha, was found. According to Dr Cordelia Langford, from the Sanger Institute, that 'really brought the understanding and the knowledge of genomic surveillance and how powerful it is onto the front pages.' She argues that this 'reinforced the importance' of COG-UK's work and the way that it could be applied to inform government decisions about whether there should be 'local lockdowns' (Langford transcript).

The same point is also made by Dr Ian Johsnton also at the Sanger Institute. He describes the identification of the Alpha as a VOC as 'the lightbulb moment'. Importantly it 'enabled the world to see the value of genomic surveillance' [Johnston transcript]. Getting to this point, however, had taken many months of sequencing thousands of samples and trying to determine which mutations spotted in the different SARS-CoV-2 virus genomes really mattered in terms of transmissibility and severity of disease.

Mutations

Measuring 80 billionths of a metre in diameter, the genetic material of SARS-CoV-2 virus is encoded in a nucleic acid, which in the case of SARS-CoV-2 is (RNA). The RNA is simply a very long polymer made from 4 chemical sub-units, known as nucleotides, which are: Uracil (U), adenine (A), cytosine (C), and guanine (G). Like other viruses, the SARS-CoV-2 virus cannot reproduce on its own and can only replicate by invading a host cell (such as humans, other animals, plants or bacteria) and hijacking and customising its machinery to make copies of itself. Once inside a living cell, the virus makes use of an enzyme of the host cell, RNA polymerase, to replicate. But this process is prone to error because the RNA polymerase lacks proof-reading capacity found in the polymerase of DNA viruses, which means that occasionally one of the nucleotides get flipped during replication, resulting in mutations.

Overall, the mutation rate of the SARS-CoV-2 virus is half the rate of flu viruses and a quarter that of HIV. Nonetheless, as it infected more and more people it had greater scope to mutate. Like other coronaviruses, the SARS-CoV-2 virus has one of the longest genomes amongst all RNA viruses (Cao). Part of the problem with the virus is the fact that while its replication error is relatively low, the increasing pool of infected people and large reservoir of individuals potentially susceptible to infection increased made the likelihood of mutations much greater (Thomson). In many cases mutations do not matter and often weaken the virus leading to its disappearance. But in some cases, they can confer new characteristics which gives them the ability to spread faster or better able to evade the immune system (Makin; Gale).

Figure 11.1 Diagram showing how a SARS-CoV-2 virus particle replicates. Credit: Lara Marks. The virus particle enters the living cell by hooking itself on to angiotensin converting enzyme-2 (ACE2), a specific protein receptor found on the cell surface. Once inside the cell, the protein shell coating the RNA gets destroyed, releasing the genetic material into the cytoplasm ready for replication with the help of RNA polymerase. The replicated virus particle leaves the host cell after maturation to infect other cells. An infected cell can replicate and release hundreds to thousands of virus particles.

A lot depends on how the mutation in the genome affects the protein structure of the virus. Like other coronaviruses, the SARS-CoV-2 virus particle contains four structural proteins (figure 11.2). The first, the Spike protein (S) which is found on the surface of the virus and gives the virus a crown or sun corona appearance - hence the name - plays an important role in binding to the ACE2 receptors on host cells to gain entry into the cell. The second protein, the nucleocapsid (N), wrapped by envelope (E) and membrane (M) proteins, helps package the RNA and serves several roles in the life cycle of the virus, including facilitating viral RNA production and suppressing the innate immune responses of the host cells. Because they are important to enabling the virus to function, both the S and N proteins quickly became targets for the development of vaccines and therapeutics (Ye). They were also an important focus for understanding mutations.

Figure 11.2: Diagram of the four key structural proteins in a SARS-CoV-2 virus. Credit: Oxford Genomics Institute.

Very early on, a number of COG-UK participants began to have regular meetings to discuss variants of interest (VOIs). These are variants that display genetic changes predicted or known to affect the behaviour of the virus such as transmissibility, and which could reduce the effectiveness of diagnostics and therapeutics. One of those involved in these discussions was Professor Emma Thomson based at the Centre for Virus Research in Glasgow. For her these meetings 'was almost like a break in the day, because it was so interesting. We were just talking about the different bits of the spike protein that might be important for different functions and so on' (Thomson transcript).

D614G mutation

The first inkling that there might have been a significant change in the SARS-CoV-2 genome to cause a change in the behaviour of the virus surfaced in May 2020 (COG-UK May 2020). It was picked up by a group of American researchers led by Professor Bette Korber, a computational biologist at Los Alamos National Laboratory. They found it as a result of using a computational tool to hunt for changes in the gene that codes for the Spike protein in the thousands of SARS-CoV-2 viral genomes regularly deposited in GISAID, an online global database which also included data uploaded from COG-UK members (Vartabedian).

Noticing that one particular gene mutation appeared to have spread faster across the globe since February 2020 than previous lineages, Professor Korber's group reached out to Dr Thushan de Silva at the University of Sheffield because he and his team had deposited quite a lot of sequences early on. De Silva remembers 'they wanted to know if we had any sequence linked to clinical metadata, so to see whether patients had gone to ICU, etc, with this variant compared to others.' He points out that at that time the issue was 'quite controversial' because 'we all thought this virus didn't really change in that way' (de Silva transcript). Subsequently called D614G, the mutation caused an amino acid change from aspartate (D) to glycine (G) in the SARS-CoV-2 spike protein, which attaches to the ACE2 receptor on host cells and is critical for viral entry into the host cell' (Volz 2020; COG-UK May 2020).

Figure 11.3: Diagram showing D614 mutation change and its global spread. Credit: Fig 1, Korber. The mutation appears in the 614th amino acid of the protein's peptide chain.

When COG-UK first cited the mutation in its sixth report, it took pains to stress 'There are a number of mutations across the SARS-CoV-2 genome that are seen at a high frequency around the world, however it is unclear whether these mutations are under positive selection (i.e. have become established as they improve some aspect of viral performance) …The nature of a virus pandemic (rapidly growing, spatially expanding, hierarchically structured) means the simplest and most likely explanation for spatial and temporal changes in mutation frequency is expected to be random chance processes until proven otherwise.' It also pointed out that cellular models or computer models of changes that might indicate a possible increase in the infectivity or transmissibility of the virus 'may not translate into a meaningful advantage for the virus in the real world' (COG-UK May 2020).

The controversy over whether the mutation made the virus potentially more contagious and dangerous continued for some months (Vartabedian). In July 2020 Korber published a paper together with de Silva and other colleagues concluding that while the mutation appeared to make the virus more infectious the evidence did not suggest it was associated with greater disease severity. But they argued, 'These findings illuminate changes important for a mechanistic understanding of the virus and support continuing surveillance of Spike mutations to aid with development of immunological interventions' (Korber).

N439K mutation

Not long after D614G was discovered, another mutation - N439K - was observed in October 2020. Like D614G it affected a change in the spike protein. Spotted by Thomson and her Glasgow team, the mutation was especially concerning to them because they realised 'It was an escape mutation from natural immunity, as we didn't have vaccines around then – but we could see that the virus was going to be able to evade immunity readily'. They could also see that it 'increased the binding affinity between the spike and the ACE-2 receptor' which highlighted 'how plastic the receptor binding domain was more broadly'. Thomson recalls, 'We did an analysis looking at the receptor binding motif, and we could see that there were lots of changes all around the world in the global sequence data. We saw, also, for example, that E484 was very wobbly, and other sites as well. The single polymorphisms that we started to see early on really heralded the new variants, well before the Alpha variant was found' (Thomson transcript).

Describing it as a 'concerning moment', Thomson recalls that she and her group immediately put out the data on GISAID. This data immediately attracted the attention of researchers at the University of Washington and Vir Biotechnology, a company based in San Francisco developing monoclonal antibodies for the treatment of COVID-19. They had also noticed the mutation and were working on it in parallel to the Glasgow team. Based on this, they quickly began to have regular Zoom calls together to write a joint paper to describe the N439K mutation, which was published in March 2021 (Thomson transcript). In the paper they showed that the mutation conferred 'resistance against several neutralizing monoclonal antibodies, including one authorized for emergency use by the US Food and Drug Administration (FDA), and reduces the activity of some polyclonal sera from persons recovered from infection (Thomson).

Reporting their findings to SAGE, they highlighted that 'N439K was initially identified in a single lineage first detected in March 2020 and until recently was almost unique to Scotland where it infected more than 500 individuals.' Its transmission in Scotland had ceased with lockdown in the Spring of 2020 and had 'not been observed since the 20th June in South Lanarkshire.' But it had been identified 'in another fast-growing lineage that has been sampled between late June and mid-August (2020) in Romania, Norway, Switzerland, Ireland, Belgium, Germany and now in all parts of the UK.' They argued that the 'apparent sudden rise in August/September appears to be linked to relaxation of control measures, the degree of sampling in these countries and its recent emergence in the UK with a high sampling rate.' In the conclusion to the report, they noted 'While there is no evidence that this variant will affect the efficacy of vaccines currently in development, it does highlight the need to establish a systematic approach for monitoring the appearance and spread of all variants and prioritising mutations of interest for further characterisation, in particular when selective pressure from mass vaccination programmes begins' (COG-UK Oct 2020).

Figure 11.4: Map showing the distribution of the N439K mutant found around the world as of March 2021. Credit: Thomson.

For Dr Ana da Silva Felipe taking part in the research in Glasgow, this experience was 'really a critical point when we started to push our know-how and our capabilities in terms of putting together different tools to understand the virus.' Importantly, they improved their 'capabilities in terms of growing clinical isolates in vitro to be able to characterise these new mutations and doing cross competition assays to understand the effect of mutations on viral fitness'. From her perspective this was vital in preparing the CVR to be able to respond to other variants as they emerged (da Silva Felipe transcript).

Alpha - the turning point

One of the most important moments in COG-UK's history was when the Alpha variant was identified as a variant of concern (VOC) in December 2020. Being the first variant to be labelled in this way, Alpha marked a shift in the way variants were treated. Prior to the designation of VOC, variants 'considered to have concerning epidemiological, immunological or pathogenic properties' were labelled as ‘Variant Under Investigation (VUI) with a year, month and number'. This label signified the variant needed to undergo a risk assessment by an expert committee, following which it could be redesignated a VOC (PHE Dec 2020). The category VOC is applied to cases where the variant is 'known to spread more easily, cause more severe disease, escape the body’s immune response, change clinical presentation, or decrease effectiveness of known tools – such as public health measures, diagnostics, treatments and vaccine' (WHO Dec 2021).

Initially Alpha was designated B.1.1.7 under the Pango lineage naming system and renamed Alpha on 31 May 2021 when the World Health Organization announced a new naming system for COVID-19 variants of interest and concern based on the Greek alphabet in order to reduce stigma and misinformation in the naming of variants (WHO May 2021). A key characteristic of the Alpha variant was it displayed a cluster of mutations including a deletion of six nucleotides in the Spike protein gene, which resulted in the loss of two amino acids (?69-70) at positions 69 and 70. It also showed a mutation labelled N501Y in the spike protein, also found in a variant in a new and concerning variant in South Africa. Because of the important role the Spike protein plays in the ability of the SARS-CoV-2 virus to enter host cells, such changes raised questions about whether this made the virus more transmissible and flagged up the need for closer observation and investigation (COG-UK March 2023).

Alpha's profound impact is captured by Amy Gaskin based at Public Health Wales. As she puts it, 'I think the main moment that sticks out for me is the emergence of the Alpha variant, which at the time was known as the Kent variant. This was a time where efforts at COG-UK had a really direct impact on UK-wide policy and on our immediate daily lives. Winter in the NHS is always busy, to say the least, and I was expecting us to be busy leading up to Christmas, everyone was fatigued, and the fact that Wales in autumn of 2020 had effectively had fluctuating local lockdowns, including a weird firebreak in October, but then COG-UK discovered the emergence of the Alpha variant and then that took off in the UK and led to Christmas plans being cancelled or changed around at the very last minute' (Gaskin transcript).

Among those first to spot the new variant was Nicholas Ellaby, a bioinformatician at Public Health England (PHE). Helping to evaluate different lineages in terms of their transmissibility and virulence, Ellaby says 'I remember identifying the first emergence of the Kent lineage, Alpha, and being quite perplexed by it.' When they checked samples that were closely related, known as cluster detecting analysis, Ellaby recalls 'we started seeing lots of cases in the south-east. And they all sort of appeared very quickly, and we were a bit confused by it.' It just so happened that Andrew Rambaut had also seen the same thing so when Ellaby mentioned it, he recalls Rambaut said, '“Yup, this is really weird.” I don't think I quite appreciated how much of a dramatic shift that lineage's emergence was at the time.' On discussing it further they realised it 'raised a very concerning signal' (Ellaby transcript). Dr Meera Chand, one of the national COVID-19 incident directors at PHE, also remembers 'We immediately knew we had found something very concerning. Normally when you’re looking at samples you would expect to see lots of small clusters made up of multiple strains that are all slightly different. But when we looked at Kent, we saw about 50% of the samples were extremely similar, forming one massive cluster' (Myers, Chand).

Dr Nabil-Fareed Alikhan, who was analysing bioinformatics data at the Quadram Institute, also recalls being initially unclear about the implications of the new variant. He recalls, 'The first week or so we're not quite sure, is it going to do something serious? Is it just another one of these [lineages] 'come and go' chaps? But then after it started to ramp up, it's like 'no, this is something new and different' and we've got to change the way we think about it in a matter of weeks and get as much data as possible' (Alikhan transcript). Part of the difficulty stemmed from the fact that there were so many different variants emerging at that point (figure 11.5) it was difficult to get a handle on what was really happening.

Figure 11.5: Data from COG-UK Mutation Explorer based on all UK genomes uploaded into the CLIMB database between October 2020 and June 2021. This included both community and hospital samples collected from people with COVID-19. The data shows the detection of multiple variants (pink) prior to Alpha (green) which began to be picked up in early December 2020. Credit: COG-UK March 2023.

In early December, PHE assembled a team of experts to study the characteristics of the new variant. Coming from a range of disciplines, including virology, epidemiology, modelling and genomics, the team was drawn from across PHE and academic institutions. According to Chand, Alpha really marked the first time experts from different disciplines from both academic and PHE came together 'to evaluate a variant at pace and understand how it behaves' (Myers, Chand). Some of the work that they carried out is outlined in a Technical Briefing Report that the PHE subsequently put out on 21 December 2020. The report indicates that the variant was designated a VUI on 1 December 2020 (VUI-202012/01), and by 8 December 2020 had become the focus of intense investigation. This work led B.1.1.7 being re-designated a VOC on 18 December 2020 (VOC-202012/01) (PHE Dec 2020).

Figure 11.6: Tweet put out by Jeffrey Barrett announcing a post about the new variant put up on Virological on 19 Dec 2020 and intense work COG-UK team was putting in to understand its implications. At the time Barrett was the Director of the COVID-19 Genomics Initiative at the Wellcome Sanger Institute.

Figure 11.7: Data about new B.1.1.7 variant released on virological, 19 Dec 2020. Click here for the full post.

The PHE Technical Briefing report summarised an analysis of routinely available genomic data together with an epidemiological investigation to understand an increasing incidence of COVID-19 cases. It looked at 962 genomes in a cluster from Kent, of which ‘data was available for 915 individuals; most specimen dates were in November (828/915) followed by October (79/915), with a small number of cases in September (4/915)’ (PHE Dec 2020).

The epidemiological investigation was also undertaken with the help of data collected from the national community testing system which was using PCR tests to detect positive COVID-19 cases. What helped in this process was soon after the new variant was spotted, researchers at the Sanger Institute noticed that some of its PCR tests, designed to look for a number of gene targets in the virus sample, failed to detect one of its spike genes because it had mutated. What was puzzling was that this phenomenon seemed to only happen with certain PCR tests and the test was still detecting other target genes, which indicated the virus was in the sample. For this reason, Alikhan explains, 'Initially it took a few days for everyone to put two and two together that there's a problem with the tests' (Alikhan transcript). The same phenomenon was also picked up at the Lighthouse labs which experienced problem detecting the S-gene in their samples. Chand remembers, 'Initially they thought there was an issue with the test. It wasn’t working, and the rate at which it wasn’t working was increasing very steeply' (Myers, Chand).

The discovery of the single gene target failure (SGTF) proved quite fortuitous. Importantly, it provided a short cut to determine the presence of Alpha with a diagnostic test without needing to complete the genome sequence which took longer to process. From Chand's perspective, the observation of the SGTF provided a huge step forward in the case of Lighthouse testing. She points out that 'Whole genome sequencing takes time, but lighthouse results are a real time way of detecting what’s happening. A huge proportion of samples go through them, so we were able to give good, real time monitoring of big data' (Myers, Chand). Some idea of how useful it proved can be seen from the figure 11.8 below from the PHE’s Technical Report. The figure shows that the presence of B.1.1.7 (Alpha) and other variants containing the deletion ?69-70 only began to become noticeable in PCR tests carried out at Milton Keynes Lighthouse Lab from early November 2020. A similar pattern was also observed in tests carried out at other Lighthouse labs in Alderney Park and Glasgow (PHE Sept 2021).

Figure 11.8: Figure 2 presented in PHE Dec 2020. The deletion of ?69-70 was not only found in the Alpha variant. Just before Alpha was flagged up as a VOC, in November 2020, the same deletion was also observed by Danish investigators in SARS-CoV-2 genome sequences obtained from infected minks and humans living on farms (Lassaunière).

Figure 11.9: Tweets put out by Dr Jeffrey Barrett, 29 December 2020 indicating the way in which the new variant could be detected through SGTF in Thermofisher TaqPath diagnostics tests. The same phenomenon was picked up by French researchers, who also proposed using it as a method to screen for VOC-202012/01 in early January 2021 (Bal). Delta, a variant of concern, first identified in October 2020, did not have the SGTF, which again provided a shortcut to distinguishing it from Alpha. By contrast, Omicron behaved like Alpha (Alikhan transcript; ECDC).

A subsequent investigation published in early January 2021 by PHE together with a team of academics from COG-UK confirmed that the B.1.1.7 'lineage was detected in November 2020 and likely originated in September 2020 in the South East region of England. As of 20 December 2020, the regions in England with the largest numbers of confirmed cases of the variant are London, the South East, and the East of England'. The collaborators highlighted that over a third of the PCR-positive community COVID cases for November and December showed SGTF which allowed them to 'use SGTF frequency as a proxy for VOC frequency, and thus estimate VOC and non-VOC incidence trends by region over that time period.’ They continued ‘We see a very clear visual association between SGTF frequency and epidemic growth in nearly all areas' (Volz 2021).

Figure 11.10: Expansion and growth of B.1.1.7 lineage as presented in figure 1, Volz 2021. A) The number of UK Lower Tier Authorities reporting at least one samples VOC genome. B) Empirical (solid) and estimated (dash) frequency of true positive rate-adjusted SGTF in three regions of England. C) Empirical (points) and estimated (line) frequency (log odds) of VOC inferred from genomic data by epidemiological week. D) Empirical (points) and estimated (line) frequency (log odds) of SGTF based on the same data as B.

Dr Alessandro Carabelli, tasked with running COG-UK's Mutational Analysis working group from October 2020, indicates that many expected new variants to surface. But the emergence of Alpha was a surprise because they 'didn't know the extent, or how it was spreading, what were the characteristics of this variant, if it was more lethal, and how it was more transmissible' (Carabelli transcript). What was particularly worrying was the highly unusual number of virus changes found in the Alpha variant. As the authors of the post released on Virological, on 19 December 2020, noted 'The accrual of 14 lineage-specific amino acid replacements prior to its detection is, to date, unprecedented in the global virus genomic data for the COVID-19 pandemic. Most branches in the global phylogenetic tree of SARS-CoV-2 show no more than a few mutations and mutations accumulate at a relatively consistent rate over time. Estimates suggest that circulating SARS-CoV-2 lineages accumulate nucleotide mutations at a rate of about 1-2 mutations per month' (Rambaut).

Figure 11.11: Tweet of media interview with Susan Hopkins, Strategic Response Director, COVID-19, PHE, put out by Sophy Ridge. Hopkins indicated that the new variant was first brought to the attention of the UK Government on 11 Dec 2020, three days following PHE and its collaborators beginning a detailed analysis, and the government was warned it was significantly more transmissible than the original Wuhan strain on 18 Dec 2020. By this point, Chand says PHE and its academic collaborators had established that 'the rate of spread was very rapid and was much bigger than just Kent. It was in London, Essex and other parts of the South East and expanding across the rest of the country' (Myers, Chand). According to the UK parliamentary committee investigating COVID-19, 'The eventual knowledge of this new variant and its heightened transmissibility explained what had been observed earlier: that North Kent and neighbouring areas were experiencing unaccountably high and persistent levels of covid infections during the late autumn. For example, on 30 November 2020, the rate of confirmed COVID-19 cases in Swale, in North Kent, was 568 per 100,000 population - over three times as high as the average UK rate of 154 per 100,000' (UK Parliament Committee).

Figure 11.12: Tweet put out by Dr Anthony Underwood, 20 December 2020. His tweet illustrates some of the pressure Alpha caused at the time. In his interview he points out 'It was initially described as the Kent variant. But it became blatantly clear, using the analysis and the map, that it was not just Kent, it was kind of everywhere.' Involved in routinely updating the visualisation tool for COVID-19, he could 'see clearly that locking down wasn't necessarily going to have much effect because it had already spread.' It was for this reason that he wrote on his personal Twitter account 'the horse has already bolted' (Underwood transcript).

For many, the appearance of the Alpha variant was what Thomson describes as 'quite a shock' because it represented a 'stepwise change with all of the mutations that were present' (Thomson transcript). Professor Sharon Peacock also comments 'Alpha was the first learning ground for us, where people realised that variants can emerge that are very genetically different. That came [out of] left of field. Nobody really expected a variant that was so genetically diverged. Nor did they realise that a variant could arise with so many different mutations' (Peacock transcript).

The experience with Alpha helped galvanise PHE to put 'protocols and systems' in place to help it 'evaluate and ensure' that it had 'surveillance in the correct sort of areas, and the appropriate sort of analyses to interpret that surveillance, and quite a rapid turnaround system' (Ellaby transcript). As part of this process, the Department of Health and Social Care developed a dashboard which allowed it to see how the data was changing so that surge testing could be triggered where needed. Originally this process had been dependent on manual reports sent in daily from COG-UK centres which entailed a lot of work on their part (Ariani transcript).

The speed with which Alpha was detected and identified as concerning meant many within COG-UK had little knowledge of it before Matt Hancock, the Minister of Health, announced it in the UK House of Commons on 14 December 2020 (Gallagher). Following this, on 19 December 2020, Boris Johnson, the UK Prime Minister, issued a 'Stay at Home' alert and imposed tougher restrictions for large parts of the South East and London, which introduced tighter travel and social contact restrictions (GOV.UK).

Pressure intensifies

In some cases, COG-UK participants had already begun to see a rise in the number of samples prior to the identification of Alpha. For example, in Portsmouth both Angela Beckett and Dr Sharon Glaysher were struck by the increasing numbers they began receiving, which included both local samples and ones coming in from other COG-UK centres and Lighthouse labs. Not yet knowing about Alpha, Glaysher says they first assumed it was the result of 'predicted winter pressures' (Glaysher transcript; Robson and Beckett transcript).

Dr Sarah Buchan, who was aiding the COG-UK effort from Bournemouth University also saw a rise in the number of positive samples. They were so high, she says, 'At one point we were questioning whether our platform was working, I mean we were saying that we've got so many positives, is this right? And we ran the negative controls again, and went, no, the negatives are negative; they're just all positive. There were so many positives that it was taking me hours to type them up for COG' (Buchan transcript). A similar trend was also picked up by the team sequencing samples at the Quadram Institute. In late November they noticed that something odd seemed to be happening in a small village outside Norwich, which appeared to have a very high infection rate. This showed up in both samples collected from community testing sites as well as from patients admitted to hospital (Meadows transcript; Alikhan transcript).

Alpha massively increased the pressure on COG-UK participants. In part, Catherine Ludden explains, this was because it was the first major VOC to surface and 'a lot more people became very interested in getting genomics done quicker.' According to her, 'The pressure from various different organisations got heavy quite quickly'. This had a knock-on effect on sequencing operations because of the increased volume of samples now needing to be sequenced at speed (Ludden and Blane transcript). Dr Leigh Jackson, who contributed to COG-UK's effort from the University of Exeter, made the same observation. He comments there was a real sense of 'urgency' in terms of 'how quickly we needed to scale up even further because of Alpha and get the data to the ministers and to the policymakers so they could make informed decisions. There was huge pressure coming from delivery boards and things like that, that we needed to get quicker and faster and do more' (Jackson and Thomas-McEwen transcript).

What also contributed to the strain was the timing. Critically, the news about Alpha broke in the week before Christmas when many centres had agreed for its staff to take well-deserved annual leave with the blessing of COG-UK. Just how much people needed this break is brought home by the interview with da Silva Felipe. At that point, she remembers, many of her team in Glasgow were 'desperate for a break' and was adamant that nothing should stop them getting a holiday. But, as she says, 'then suddenly Alpha happened and we were like, “Okay, we cannot stop for Christmas, we have to do something about this”'. Based on this, she did 'a careful assessment of how the teams were feeling and whether they could actually cope with additional work. Interestingly, there always seemed to be someone that still had the energy somewhere to help push things forward' (da Silva Felipe transcript).

Many people elsewhere in COG-UK also gave up their Christmas in December 2020 to help sequence samples. That included Glaysher who, for a week over Christmas, did all the sequencing in Portsmouth single-handed. Her memory is that 'During the Christmas period I would set up the first 96 samples to be extracted early in the morning. Then come back to our department and do the most recent RT-PCR [Reverse transcription-polymerase chain reaction test] samples before carrying on to do the library prep in the afternoon and early evening. Moving through that pathway I would then come in the next day and then repeat it all over again.' Not having any people around during the Christmas period, Glaysher recalls, 'I would pop on some music or an audiobook in the background just to keep myself going, so it wasn't pure silence or the hum of machines' (Glaysher transcript). Commenting on Glaysher's sequencing, Beckett says 'she really smashed it, she did some long hours that week.' According to Beckett, Glaysher 'managed to process more samples herself as one person than I think we'd managed in a week before (Robson and Beckett transcript).

Figure 11.13: Photograph of Dr Sharon Glaysher preparing samples in the laboratory. Credit: Bethany Lavin Photography. Glaysher developed a fascination for science at an early age. One of her strongest memories is jumping out of bed every morning to see how her crystals had grown after her parents gave her a crystal kit one Christmas. She recalls 'I would measure them and record their progress on a homemade chart. It ignited my passion for knowledge about the world around me and how everything worked.' After completing her undergraduate and master's degree at Portsmouth University, Glaysher did a PhD in Molecular Biology and Cancer Sciences while based at one of the local hospitals in Portsmouth. She did her doctorate part-time over six years while working in medical laboratories in Portsmouth Hospitals University NHS Trust. While she found working and doing a PhD at the same time challenging, at the same time she found it rewarding when she began to see her research published. Her doctoral research focused on signal transduction pathways and resistance mechanisms to chemotherapy. When the pandemic started, Glaysher was managing the research lab at Queen Alexandra Hospital in Portsmouth helping to support a number of research projects and clinical trials. This was where she and Beckett worked together to get the sequencing up and running for the COG-UK effort (Glaysher transcript; COG-UK Jan 2023).

Figure 11.14: Photograph of Angela Beckett with a model of the SARS-CoV-2 virus. Credit: Bethany Lavin Photography. The first person in her family to go to university, Beckett developed an interest in biology when she embarked on her A Levels. After completing a BSc degree in Biology at the University of Portsmouth, Beckett, spent some time travelling around South America and working in hospitality before going on to do an MSc in Biomedical Science at Kingston University, London. Following her Master's, she helped develop vital diagnostic assays and specialised vaccines at Porton Down Science Park in Wiltshire. After spending five years in this position, she was ready for a new challenge and moved to the Centre for Enzyme Research at Portsmouth University in March 2020 to work on plastic recycling research. Due to start the day when lockdown officially began, she quickly shifted to helping set up the South Coast sequencing hub for COG-UK which she did alongside Glaysher (COG-UK Jan 2023).

The week after Christmas, Beckett joined Glaysher in the laboratory to assist with the sequencing of Alpha samples. She highlights that 'Alpha was quite a daunting time in a lot of ways because at this point we did have a really good capacity for testing samples, and yet for the first time, we couldn't test all of the samples from Portsmouth hospital. We couldn't feasibly do it because the samples were coming in so fast we weren't able to process all of them. We used to go once a week to the biomedical lab freezers to collect the nasopharyngeal swabs to be processed and we used to have a few bags a week, and at some point we had freezers full. Maybe harrowing isn't the right word, but it was very sobering because you realise that each one of those people is somebody that's infected in your city, your hometown, your hospital, the place you're working. So, it was a lot during that time' (Robson and Beckett transcript).

Another person for whom Alpha had a profound impact was Peijun Zhang, a research technician who helped manage the COG-UK work at the University of Sheffield. Just before Christmas her group in Sheffield identified the first case with the variant in the South Yorkshire region. One particularly strong memory that sticks out for her was when she uploaded what she thought would be the last sequence she completed during an evening shift before her Christmas break. Once she had uploaded the sequence, she recalls that she notified Matt Parker, the bioinformatician, on Slack with words to the effect 'saying “Okay, that's it, I will see you in two weeks time” thinking that 'it was the last run before Christmas'. Much to her surprise, he replied 'he wasn't so sure it would be.' Not long after this she remembers 'we received emails from our head of department, and from COG-UK as well, asking if we were willing to come back during the holiday so we can keep on monitoring the spread. Of course, we were paid overtime. Before this job, I always worked in university, but I never experienced such a chaotic time, and Christmas holiday, I just took it for granted having two weeks off, but not for this one' (Zhang transcript).

While many stepped up to work over Christmas, for some this effort was tinged with guilt. Some of the feelings are expressed by Alikhan. He recalls Christmas was 'really bad because firstly we had to tell our families we're working over Christmas…And then we had to break the bad news to them that Christmas is going to be cancelled and we're not going to be allowed to travel because of this COVID thing' (Alikhan transcript).

The double-edged sword

In some ways the identification of Alpha was a double-edged sword. On the one hand it demonstrated the importance of COG-UK's work. Crucially, its effort helped to both identify Alpha and track its spread across the UK (Gaskin transcript; Bashton transcript; de Silva transcript; Chapman transcript). As Peacock pointed out in an interview with the BBC, the high level of genomic surveillance undertaken in the UK through COG-UK meant that 'if you're going to find something anywhere, you're going to find it probably here first' (Schraer). The same point is made by Dr Emma Hodcroft, a molecular epidemiologist based at the University of Bern in Switzerland and part of Nextstrain, an open-source initiative that provides tools for visualising the genetics behind the spread of viral outbreaks. She argues, 'the fact that Alpha appeared in the UK where they had enough sequences to actually characterise this and figure out that it was a new variant was also important. I mean, if it had appeared in Eastern Europe, we would have probably spent months scrambling to try and figure out what was going on, instead of realising so quickly that this really was a change in the virus and it really was making a difference. That completely changed the response to Alpha and the impact that it had in the rest of Europe' (Hodcroft transcript).

On the other hand, the news of Alpha shone an unprecedented spotlight on COG-UK. Just how much media attention went up is captured by the interview with Georgie McManus who was part of COG-UK's communication team at this point. She recalls 'To put it into numbers, before the Alpha variant was announced I think we had about 2,000 followers on Twitter, and over the space of a week it went up by about 10,000. So, it was just a crazy amount of attention all of a sudden within such a short space of time. I am actually talking about Twitter followers' [McManus transcript]. A lot of the interest stemmed from the role COG-UK's data played in persuading the government to curb travel and mingling over Christmas and introduce a third lockdown from 6 January 2020 (UK Parliament Committee).

Not surprisingly, the government's response made some COG-UK participants feel like they were responsible for cancelling Christmas. Gaskin sums up some of what they had experienced. For her, what made the situation with Alpha feel worse was because its discovery 'coincided with a lot of people in the UK, obviously not everybody' planning to celebrate Christmas 'as an escape from COVID' which meant 'that when that was taken away at such short notice, it resulted in a feeling of powerlessness, almost like the last 10 months of sacrifice and social isolation, financial troubles, was all for nothing' (Gaskin transcript).

The feelings COG-UK participants experienced at the start of the Alpha wave were not helped by media headlines. One headline printed in the Daily Mail that Gaskin remembers, read 'How Britain shot itself in the foot being the best in the world at sequencing viruses and spotting the super infectious mutation that left the UK a pariah' (Daily Mail). She recalls, 'I felt like saying, 'Sorry, we're too good at sequencing. What do you want us to do?' But, as she points out, 'it would have been 10 times worse to have not discovered the variant at all, or discovered it but not told anybody about it. When you've got this knowledge you have to share it, because otherwise January after that Christmas would have been so much worse because we wouldn't understand why the spread was being driven so quickly?' (Gaskin transcript).

Figure 11.15 Tweet put out by O'Toole on 24 December reporting the development of Grinch. Lineage B.1.1.7 refers to the Alpha variant. B.1.351 refers to the Beta variant which was detected in South Africa around the time that news broke about Alpha. Like Alpha, the Beta variant also carried a mutation in its spike protein and appeared to spread rapidly. It was first identified in a South African sample collected in October 2020 (O'Toole).

Figure 11.16: Photograph of Verity Hill. Credit: Bethany Lavin Photography. Not really knowing what she wanted to do as a career when she was younger, Hill's interest in science was first kindled when, at the age of 16, she read an article on the microbiome in The New Scientist. She first grasped the importance of science applications when she learnt about the outbreak of the Ebola virus in Africa while studying Biological Sciences as an undergraduate at the University of Oxford. After Oxford, Hill completed a master's degree in the Control of Infectious Diseases at the London School of Hygiene and Tropical Medicine and then did a doctorate at the University of Edinburgh. She is now a postdoctoral researcher at Yale School of Public Health in the United States (COG-UK Jan 2023).

What added to the pressure was the fact that the variant had first been identified in the UK. For example, Dr Verity Hill, who was developing analytical tools for COG-UK, recalls, 'Watching that spread from day to day, it was like the start of the pandemic again.' Keen to try and help the situation, she and O'Toole quickly designed a software tool to download and summarise GISAID data to track the spread of Alpha around the world. Hill remembers it took them about two days, between 22 and 23 December 2020, to get it working as a functional tool and then finessed it over the Christmas period. According to Hill, they called the tool 'Grinch' because Alpha 'stole Christmas for almost everyone in the UK'. The name stands for global report investigating novel coronavirus haplotypes. It had the capacity to track Alpha and other new variants of concern (Hill transcript).

Learning curve

According to Dr Sonia Goncalves, who was managing COG-UK operations at the Sanger Institute, Alpha marked a major shift towards focusing more directly on variants within COG-UK. Prior to this she says 'it was about looking at the importation of where the positive samples are coming from, which countries they are coming from, or specific outbreaks in hospitals or care homes. But then, the Kent [Alpha] variant was a big game changer, because this was a completely different scenario.' She remembers this was the moment when it became clear that they could actually see 'the virus evolving in real time' and the impact of a 'particular variant taking over' (Goncalves transcript).

Alpha was not just a huge learning curve for COG-UK, but also for government bodies. It demonstrated to them the need for continuous genomic data. Just how much of an impact it had in government circles is borne out by the fact that the Department of Health and Social Care became much more directly engaged with the Sanger Institute than previously and put it on the equivalent footing as testing centres in terms of supporting it with consumable supplies (Johnston transcript; Ariani transcript).

Subsequent variants of concern

The Alpha variant provided a major training ground for the next VOCs that came after it (Peacock transcript). It helped create an analytical framework for talking about new variants. But each novel variant raised new issues because, as Ellaby, points out 'none of the variants are exactly the same' (Ellaby transcript). In the words of da Silva Felipe 'After each variant appeared we realised we were not yet fully prepared and needed more tools and to get better at growing and studying the virus'. From her perspective each variant provided a learning opportunity to get better at understanding the virus. She comments, 'Every time there is a new variant we already have tools from the previous one to enable us to respond a bit better and to continue to build on them' (da Silva Felipe transcript).

Figure 11.17: Tweet put out by Professor Dr Moritz Gerstung, a researcher based at EMBL’s European Bioinformatics Institute (EMBL-EBI), highlighting the unpredictable nature of the SARS-CoV-2 virus.

Figure 11.18: Graphic distribution of cases and different lineages within the UK from September 2020 to June 2021. The image was compiled based on the analysis of genomic surveillance data collected across England by researchers from the Sanger Institute together with collaborators at EMBL-EBI. The team characterised the growth rates and geographical spread of 71 lineages to get an insight into how newly emerging variants changed over the course of the early part of the COVID-19 pandemic. Credit: Figure C, Vöhringer.

Figure 11.19: Tweet put out by Hodcroft, 21 June 2022 with a phylogenetic tree from Nextstrain showing the Alpha variant and subsequent VOCs. To visualise what is happening with the SARS-CoV-2 virus, Nextrain downloads and analyses publicly available data from websites like GISAID and GenBank, which includes data uploaded from COG-UK. Hodcroft indicates 'the UK makes up a very important chunk of our open source data.' She is especially appreciative of the fact 'that COG-UK makes that data available and is okay with us sharing all of the details'. According to her, COG-UK really stands out because of its willingness to share data which other organisations are less willing to do (Hodcroft transcript).

In October 2021, Barrett commented the genomic surveillance data collected through COG-UK provided a ‘totally new way of watching an outbreak unfold, which has taught us how a new infectious agent spreads and evolves’. Critically the emergence of new variants led both him and other researchers to believe that the COVID-19 crisis that ‘gripped the UK between September 2020 and June 2021’ should be ‘thought of as a series of overlapping epidemics, rather than a single event’ (Sanger Institute Oct 2021).

Mutation Explorer Dashboard

Over the course of the pandemic, several bioinformatic tools were developed for visualising and analysing genomic data from the SARS-CoV-2 virus to support public health measures. Most of these tools focused on the epidemiological aspects of the virus but provided little in the way of how the mutations impacted the immune response, which had ramifications for both vaccine and therapeutic developments. To address this gap, COG-UK's Mutation Analysis Group, led by Carabelli, created a publicly accessible web-based resource called 'Mutation Explorer'. The group comprised clinicians, evolutionary biologists, epidemiologists, microbiologists, virologists, immunologists, structural analysts, bioinformaticians and web-developers. Designed to complement dashboards and bioinformatic tools developed elsewhere, COG-UK-ME aimed to combine UK SARS-CoV-2 genome sequence data, drawn from CLIMB-COVID, with information collated from the literature on the 'impact of mutations on various functional aspects of the virus' (Wright; Carabelli transcript).

Figure 11.20: Photograph of Dr Alessandro Maria Carabelli. Born in Italy, Carabelli joined COG-UK's effort in October 2020 after five years as a postdoctoral researcher investigating new biomaterials at the University of Nottingham. He first developed a passion in science as a result of one of his primary school teachers. Prior to starting his PhD, he taught chemistry and biology to school students in Italy for seven years. His inspiration for teaching came from his grandfather who used to say to him "If you are not able to make a child understand something it means you haven't understood it yet". For Carabelli, COG-UK provided a golden opportunity to help make a difference to the pandemic which first made an impression on him when he started seeing the suffering it caused early on in his home country. (Carabelli; Carabelli transcript).

The initial coding for the dashboard was undertaken by Derek Wright, a bioinformatician based at Glasgow Centre for Virus Research at the University of Glasgow. He did this by drawing on GLUE (Genes Linked by Underlying Evolution), a software system for tracking amino acid sequence variation in viruses developed before COVID-19 by Joshua Singer, another CVR bioinformatician (Singer). At first, the dashboard focused on taking data from GISAID to show mutations as they accumulated, which worked for a while but subsequently needed to be updated to cope with the increasing volume of data. At this point another bioinformatician in Glasgow, Richard Orton, stepped in to help (Robertson transcript).

Bioinformaticians were not the only ones involved in the development of COG-UK-ME. So too were several academics who were tasked with gathering information from the literature to upload into COG-UK-ME. Carabelli points out that this was a significant challenge because 'there were papers and preprints coming out on a daily basis, and it was pretty hard to keep up and remain informed'. Another significant hurdle was making sure that the visual information COG-UK-ME provided was presented in a way that the message was clear (Carabelli transcript). Designed to be used by researchers, public health agencies and drug and vaccine developers, COG-UK-ME's interface allows 'users to track mutations that are a potential threat based on a phenotypic impact on virus biology or by conferring resistance to the human immune response, including that boosted by vaccines or antiviral drugs' (Wright).

The first iteration of COG-UK-ME was launched in February 2021 and was then continually updated. In March 2022 the dashboard had 'around 5,000 users per month, with approximately 30 per cent from the UK, 20 percent from the USA, and the remainder from other international locations'. The dashboard provides a number of tables and visualisations of VOIs and VOCs identified by the UK Health Security Agency. It also displays the spike protein structure, 'showing the position of the VOC-defining mutations' and percentage of variants per region (Wright).

Figure 11.21: Images from COG-UK-ME. Credit: Figure 1, Wright (A) provides a frequency plot showing the number of SARS-CoV-2 sequences per week for VOCs Alpha, Delta, Delta-AY4.2, Omicron BA.1 and BA.2, and 'other' pre-VOC variants in the UK. (B) illustrates the Spike protein structure showing the locations of Delta- and Omicron-specific spike mutations.

References

Bal, A, Destras, G, Gaymuard, A, et al (5 Jan 2021) 'Two-step strategy for the identification of SARS-CoV-2 variant of concern 202012/01 and other variants with spike deletion H69–V70, France, August to December 2020'.Back

Cao, C, Cai, Z, Xiao, X, et al (24 June 2021) 'The architecture of the SARS-CoV-2 RNA genome inside virion', Nature Communications, 12/3917.Back

Carabelli, A (25 Jan 2020) 'From teacher to SARS-CoV-2 mutation hunter', Peacock Blog.Back

COG-UK (14 May 2020) Report #6.Back

COG-UK (15 Oct 2020) SAGE Report #12.Back

COG-UK Women (Jan 2023) Snapshots of Women in COG: Scientific excellence during the COVID-19 pandemic.Back

COG-UK (17 March 2023) 'A look back to the detection of the first variant of concern, the Alpha variant'.Back

Daily Mail 22 Dec 2020) 'How Britain shot itself in the foot'.Back

ECDC, WHO (2 Aug 2022) Methods for the detection and characterisation of SARS-CoV-2 variants - second update.Back

Gale, J (31 Aug 2021) 'Alpha, Delta and More: Why virus variants are causing alarm', ($) Bloomberg.Back

Gallagher, J (14 Dec 2020) 'New variant' of coronavirus identified in England', BBC News.Back

GOV.UK Press release (19 Dec 2020) 'Prime Minister announces Tier 4: 'Stay At Home' Alert Level in response to new COVID variant'.Back

Korber, B, Fischer, WM, Granakaran, S et al (20 Aug 2020) 'Tracking changes in SARS-CoV-2 Spike: Evidence that D614G increases infectivity of the COVID-19 virus', Cell, 182/3, 812-27, E19.Back

Lassaunière, R, Fonager, J, Rasmussen, M et al (n.d.), 'Working Paper: SARS-CoV-2 spike mutations arising in Danish mink and their spread to 5 humans'.Back

Makin, S (5 Feb 2020) 'How coronaviruses cause infection—from colds to deadly pneumonia', Scientific American.Back

Myers, R, Chand, M (22 April 2021) 'Disease Detectives – Finding the UK’s dominant COVID-19 variant and the race to understand its impact', UKHSA Blog.Back

O'Toole, Á, Hill, V, Pybus, O et al (Feb 2021) 'Tracking the international spread of SARS-CoV-2 lineages B.1.1.7 and B.1.351/501Y-V2', Virological.Back

PHE (21 Dec 2020) Investigation of novel SARS-CoV-2 variant: Variant of Concern 202012/01: Technical Briefing.Back

PHE (17 Sept 2021) Investigation of novel SARS-CoV-2 variant: Variant of Concern 202012/01: Technical briefing 2.Back

Rambaut, A, Loman, N, Pybus, O et al (Dec 2020) 'Preliminary genomic characterisation of an emergent SARS-CoV-2 lineage in the UK defined by a novel set of spike mutations', Virological.Back

Sanger Institute (14 Oct 2021) 'The inside story of England COVID pandemic described in new study', Blog.Back

Schraer, R (22 Dec 2020) 'Covid: New variant found 'due to hard work of UK scientists', BBC News.Back

Singer, JB, Thomson, EC, McLauchlan, J et al (18 Dec 2018) 'GLUE: a flexible software system for virus sequence data', BMC Bioinformatics, 19/1, 532.Back

Thomson, E, Rosen, LE, Shepherd, JG, et al (4 March 2021) 'Circulating SARS-CoV-2 spike N439K variants maintain fitness while evading antibody-mediated immunity', Cell, 184/5, 1171-87, e20.Back

UK Parliament Committee (12 Oct 2021), 'Coronavirus: lessons learned to date: Lockdowns and social distancing'.Back

Vartabedian, B (2 July 2020) 'The coronavirus has changed since it left Wuhan. Is it more infectious?', Los Angeles Times.Back

Vöhringer, HS, Sanderson, T, Sinnott, M, et al (14 Oct 2021) 'Genomic reconstruction of the SARS-CoV-2 epidemic in England', Nature, 600, 506-11.Back

Volz, E (25 June 2020) 'Updated analysis of SARS-CoV-2 spike protein variant D614G in the UK: evaluating evidence for effects on transmission and pathogenicity', COG-UK Report #9.Back

Volz, E, Mishra, S, Chand, et al (4 Jan 2021)'Transmission of SARS-CoV-2 Lineage B.1.1.7 in England: Insights from linking epidemiological and genetic data', medRxiv.Back

WHO (31 May 2021)'WHO announces simple, easy-to-say labels for SARS-CoV-2 Variants of Interest and Concern'.Back

WHO (4 Dec 2021) ‘Coronavirus disease (COVID-19): Variants of SARS-COV-2'.Back

Wright, DW, Harvey, WT, Hughes, J, et al (18 March 2022) 'Tracking SARS-CoV-2 mutations and variants through the COG-UK-Mutation Explorer'.Back

Ye, Q, Lu, S, Corbett, KD (26 July 2021) 'Structural basis for SARS-CoV-2 nucleocapsid protein recognition by single-domain antibodies', Frontiers in Immunology, 12.Back