What can deliberately infecting healthy people tell us about infectious diseases? How is this useful for developing treatments, and how do we manage the risks?
- SARS-CoV-2, the virus that causes COVID-19, carries its genetic information (genome) in the form of a single strand of RNA containing around 30,000 letters of genetic code. The process by which viruses replicate their genetic material is prone to errors, leading to mutations in the genetic code.
- Most mutations are harmful to the virus and will not persist in the viral population. However, any mutation that provides a benefit to the virus may allow the new variant to out-compete other forms of the virus.
- New variants of the virus have been detected in many countries around the world, including in Europe, the UK, the Americas and Africa. Of these, the variants of most concern are those that have emerged in the UK, South Africa and Brazil because they include changes to the region of the viral genome that codes for the spike protein.
- Mutations in the spike protein are of particular interest because the interaction between this protein and a receptor on the cells that line human airways is the first step in the infection process. Any changes to the spike protein could affect the strength of its interaction with human cells and therefore how easily the virus can be transmitted.
- For the new UK variant, there is good evidence that it a) has a transmission advantage that allows it to out-compete other variants circulating in the UK, b) continued to spread (albeit at a slower rate) despite the imposition of national lockdown measures in November and c) spread very rapidly when those measures were eased in early December. Stricter measures than those imposed in November may be required to contain the new variant.
- The spike protein is also the target of virtually all the COVID-19 vaccines currently in development. Any changes to it might allow the virus to evade (natural or vaccine-induced) immunity.
- Lab tests suggest that one of the vaccines used in the UK (Pfizer/BioNTech) will protect against the new UK variant. Researchers are confident that all of the vaccines in use or under development will protect against the new UK variant because they target multiple sites on the viral genome. However, there is concern that some of the mutations found in the South Africa and Brazil (but not the UK) variants may compromise vaccine efficacy in some individuals.
- Initial studies suggested there was no evidence that any of the new variants caused a more severe form of COVID-19. However, recent evidence suggests that the UK variant may be associated with a slight increase in case fatality rates.
How new variants arise
The SARS-CoV-2 virus consists of a single strand of genetic material (RNA) surrounded by a membrane studded with structural proteins. The viral genome (the RNA strand) is one of the largest known RNA genomes and contains instructions for making:
- Four structural proteins, one of which is the spike (S) protein that the virus uses to gain entry to the human epithelial cells that line the airways.
- At least 16 non-structural proteins (nsp) that are associated with a range of functions including making copies of the viral components (proteins and RNA), assembling them into new viral particles and budding them of from the host cell.
- Six or seven other (accessory) proteins.
Other RNA viruses, such as influenza viruses and HIV, are associated with very high rates of mutation due to errors in the RNA replication process. These high mutation rates affect the way the diseases are managed. For example, flu vaccines have to be reformulated each flu season to reflect the predominant strains in circulation, and anti-HIV drugs are given as combination therapy to minimise the risk of drug-resistant mutations emerging. However, despite being an RNA virus, the mutation rate of SARS-CoV-2 is much lower than that observed in influenza viruses and HIV. This is because the SARS family of viruses have sophisticated and accurate machinery for replicating RNA, including having a built-in proof-reading function. It has been estimated that this machinery limits the mutation rate of SARS-CoV-2 to around two changes per genome per month.
Why mutations matter
Mutations arise randomly and most will either be silent (have no discernible impact) or be harmful (for instance by making the virus non-viable) and therefore not persist in the viral population. But mutations that occur in the parts of the genome that code for proteins can result in changes to the properties of those proteins. This is because proteins are built by assembling long chains of amino acids, with the sequence of the amino acids in the chain being determined by the sequence of letters in the genetic code.
Changing just one letter of the genetic code can cause a different amino acid to be incorporated into the chain, and this can have a profound effect on the way the resulting protein functions. In most cases these changes will be detrimental to the virus, but a very small proportion may provide some advantage to the virus and allow it to out-compete other variants. The following sections summarise some of the general trends in genetic variation in the SARS-CoV-2 genome and examine some of the new variants of most concern.
Tracking variations in the SARS-CoV-2 genome
The first full genome sequence (Wuhan-Hu-1) of the SARS-Cov-2 virus behind the outbreak of COVID-19 in Wuhan was published in February 2020. Since this time, GISAID has maintained a database of all published (complete) SARS-CoV-2 genomes in order to promote the rapid sharing of data. These data have been used by the COVID-19 Genomics UK consortium (COG UK) to power apps such as CoV-GLUE that allow researchers to track changes in the SARS-CoV-2 genome as it spreads around the world. CoV-GLUE can be used to track changes to the genome sequence, such as the insertion and deletion of letters of the genetic code, as well as mutations that lead to changes in the amino acid sequence of the proteins produced by the virus. Details of more than 32,400 different amino acid changes are held in the CoV-GLUE database.
While amino acid changes have been reported for virtually all of the proteins found in the SARS-CoV-2 virus, the main focus to date has been on those that affect the spike protein. This is because changes to the spike protein have the potential to increase the transmissibility of the virus, modify the range of cells it can infect, evade being neutralised by the monoclonal antibodies being trialled as COVID-19 therapy, and evade natural- or vaccine-induced immune responses (because it is the spike protein that is the target of such responses). COG-UK suggests that around 4,000 different mutations in the spike protein have been reported to date. The following sections describe the main new SARS-CoV-2 variants that contain one or more changes to the spike protein.
The D614G variant
The first mutation in the spike protein to be tracked worldwide was the D614G variant. In this variant the original amino acid aspartame (D) found in the 614th position of the spike protein in the Wuhan-Hu-1 variant has been replaced by the amino acid glycine (G). This substitution was first seen in sequences from patients in China and Germany in January 2020. While the original D variant was the predominant virus seen until early March 2020, cases of the new G variant have risen sharply since then and D614G has become the predominant variant.
The rapid rise of the D614G variant prompted speculation that the mutation confers some benefit that allows it to out-compete the original D variant. This triggered a wide range of studies to investigate whether the D614G variant was more efficient in infecting cells, more easily transmissible or caused a more severe version of COVID-19 than the original D variant. Laboratory studies used pseudo-viruses (model viruses that have been genetically modified to include the spike protein fragment from a D or G variant SARS-CoV-2 virus) to infect cultured human cell lines or hamsters. Such studies showed that the D614G variant was more infective in the human cell lines than the original D variant. The hamster studies showed that the D614G variant produced higher viral loads (amount of detectable viral RNA) in the upper respiratory tract but not in the lung compared with the D variant. These findings are consistent with the idea that the D614G variant is more transmissible than the D variant, but does not cause a more serious version of COVID-19.
A new variant in the UK
A new UK variant (called B.1.1.7 or VOC 202012/01) was detected in November 2020. It is thought to have arisen in South East England in September 2020 and is characterised by 17 mutations, eight of which are found in the spike protein. The most significant of these are thought to be:
- N501Y, a mutation in the part of the spike protein that directly interacts with the human receptor protein. In animal models it has been shown to increase receptor binding and cell infectivity. Changes at the N501 site have been linked with evasion of the therapeutic monoclonal antibody LY-CoV16 which is currently in clinical trials as a potential treatment for COVID-19.
- P618H, a mutation in a part of the S gene that is thought to be important in gaining entry to the epithelial cells that line the human respiratory tract.
- Deletion of two letters of the spike protein code. This is significant because the deletion falls inside one of three sites checked by a commonly used diagnostic test for SARS-CoV-2 (see below).
Three UK diagnostic labs use the TaqPath PCR test for SARS-CoV-2 testing. This test looks for sequences at three sites of the viral genome (S, N and ORF1ab) and records a positive test result if two or more of these sequences are detected. Until November 2020, the vast majority of positive results recorded by these labs tested positive at all three sites. From November 2020 onwards an increasing proportion of positive results from these labs recorded a negative result for the S gene site with positive results for the other two sites. This profile is known as S gene target failure (SGTF). The proportion of SGTF results increased sharply after the national lockdown measures were eased in December 2020.
Subsequent studies suggest that, in South East England, the new UK variant accounted for 98% of the SGTF profiles and SGTF is now widely used a proxy to track the spread of the new variant. While the new variant has been detected in all regions of England, it is highest in the South East of England (where SGTF accounts for 85% of positive COVID-19 tests), London and the East of England, and lowest in Yorkshire and the Humber (15%). The new UK variant appears to have a significant transmission advantage over other variants circulating in the UK, and is slightly more prevalent among 0–19 year olds and less prevalent in 60–79 year olds than has been seen to date.
A new variant in South Africa
A rapid rise in SARS-CoV-2 cases in the Mandela Bay area of the East Cape province of South Africa in mid-October 2020 has been linked to a new variant of the virus. It is characterised by eight mutations in the spike protein, including three mutations (N501Y, E484K and K417N) in or around the part of the spike protein directly involved in binding human cells. Two of these (N501Y and E484K) are thought to enhance the binding of the spike protein with a receptor on the outer surface of human epithelial airway cells. Lab studies have implicated the E484K mutation and substitutions at the K417 site with evasion of a range monoclonal antibodies.
In one study, the E484K mutation (in combination with other mutations) enabled evasion of multiple neutralising antibodies from the blood of several patients recovering from COVID-19. The rapid spread of the new variant within the East Cape province and into the West Cape province suggests that it may have a transmission advantage over the variants it is displacing. However, there is no evidence for this to date. Although the South Africa variant shares the N501Y mutation with the UK variant, whole genome data suggest that they emerged independently.
New variants in Brazil
Manaus, the largest city in the Amazonas state in North Brazil has been badly hit by the COVID-19 pandemic and has seen a surge in COVID-19 cases since mid-December 2020. A recent study obtained a small number (37 samples giving rise to 31 whole genome sequences) of samples from people testing positive for SARS-CoV-2 virus in the city to investigate the variants of SARS-CoV-2 circulating there. It identified a new variant in 42% (13/31) of the samples.
This new variant contained multiple mutations, including three in the spike protein: E484K, N501Y and K417T. The E484K mutation was well established in previous variants found in Brazil and had also been identified as one of the mutations implicated in the first confirmed case of re-infection in Brazil. This was confirmed by the Brazilian Ministry of Health in December 2020 in a medical doctor who became infected with one variant of SARS-CoV-2 in June 2020, recovered and then contracted a different variant in October 2020. The second (re-infection) variant contained the E484K mutation. The fact that there were only 116 days between the two infections has led to speculation that the E484K mutation may have evaded any neutralising antibodies persisting from the first infection.
Significance of the new variants
The three new variants discussed above involve three different SARS-CoV-2 lineages that appear to have emerged independently of each other. Each has emerged in the context of a rapid upsurge in cases. The following headings cover the possible implications that the emergence of these new variants may have on the future course of the epidemic.
The S gene target failure feature of the new UK variant allows researchers to distinguish it from other disease-causing variants of the virus. This has allowed researchers to go back through the diagnostic data and track the rate of increase in prevalence of the new variant over time. Such studies suggest that the new UK variant has a significant transmission advantage compared with other disease-causing variants. If this is related to the N501Y mutation, then other variants with this mutation (South Africa and Brazil) may also have a transmission advantage, but that this has yet to be established.
A recent study showed that both the N501 and Y501 versions of the SARS-CoV-2 virus elicited similar (positive) antibody responses in blood from patients who had received the Pfizer/BioNTech vaccine. This suggests that the N501Y mutation does not allow the virus to escape immune responses stimulated by this vaccine. AstraZeneca, the manufacturer of another vaccine being used in the UK, has suggested that its vaccine will also be effective against the new variant and is conducting trials to establish this.
Recent research has tested the efficacy of the Moderna vaccine in viral models containing the mutations found in the UK and South Africa variants. It found that the UK variant showed full neutralisation, and the South Africa variant show reduced, but still significant, neutralisation by antibodies in the blood of people or higher primates that had received the Moderna vaccine. Further research is needed to assess vaccine efficacy against the Brazil variant.
Delaying the second vaccine dose
Both vaccines used in the UK require two doses before maximum immunity is achieved. Some commentators have suggested that the UK Government’s decision to extend the period between the first and second dose by nine weeks will increase the probability of new variants emerging that can evade vaccine-induced immune responses. The New and Emerging Virus Threats Advisory Group (NERVTAG) considered that the probability of this happening was unquantifiable but likely small, and that this had to be weighed against the established benefits of increasing the rate of vaccine-induced protection.
Severity of COVID-19 disease
Public Health England (PHE) conducted a study comparing the mortality rates among 1,769 patients with the new UK variant of SARS-CoV-2 against the same number of cases involving other variants matched for age, sex and area of residence. Preliminary analysis of a subset of these cases found no significant differences in mortality rates between the two groups. However, evidence considered by NERVTAG suggests there is a slight increase in case fatality rates among people infected with the new variant compared with those infected by other variants.
The new evidence includes unpublished papers from researchers at London School of Hygiene and Tropical Medicine, Imperial College London and Exeter University as well as a reanalysis involving all of the matched cases in the PHE study. The new studies suggest that the increase in case fatality rates may be between 1.3% to 1.9% in new variant cases compared with other variant cases.
The PHE study found no significant differences in re-infection rates between patients in the new variant group (two reinfections) and those in the comparator group (three reinfections).
Efficacy of the lockdown measures
It has been noted that the national lockdown measures imposed in November reversed the rise in cases of the non-new variant virus but merely slowed (not reversed) the rise in new variant cases. Stricter lockdown measures than those imposed in November may be needed to bring the new variant under control.
UKRI has announced £2.5 million funding for research into the new UK variant, including its severity and transmissibility and the efficacy of vaccines and treatments.
How do our bodies defend against Covid-19? Read how immune responses differ across people, variants, reinfection, vaccination, and current immunisation strategies.
Research studies involving thousands of people have allowed scientists to test which drugs are effective at treating COVID-19. Several drug therapies are now available to treat people who are in hospital with COVID-19, or to prevent infections in vulnerable people becoming more serious. This briefing explains which drugs are available, the groups of people in which they are used and how they work. It also outlines the importance of monitoring the emergence of new variants and drug resistance.