Skip to content

Science Museum Group

The body’s protective immune system holds the key to diagnosing, treating and vaccinating against coronavirus. Science Director, Roger Highfield, focuses on antibodies.

What is the immune system?

This is the body’s system for protecting itself.

Part of it is ‘innate’, dealing with general threats, and part ‘adaptive’, where the body learns to counter a specific infection, for instance, a virus or bacterium.

Our adaptive immune response is the reason why, once we have had measles (or the MMR vaccine), we can shrug off a measles infection at the next encounter with the measles virus.

Salivette used in measels prevalence surveys by the Communicable Diseases Surveillance Centre (PHLS).
Salivette used in measels prevalence surveys by the Communicable Diseases Surveillance Centre (PHLS), complete with postal packaging, 1996. Black, matt background.

Why are antibodies important?

When the human body is exposed to an invader, such as a coronavirus, the immune ramps up production of white blood cells – called B lymphocytes – that produce antibodies that bind to viral proteins that the virus needs to infect cells.

A second part of the immune response relies on another type of white blood cell, T lymphocytes, which can kill infected cells.

Antibodies, which are also called immunoglobulins, stop the spread of infection between cells while T-cells stop the virus once it is inside a cell.

What is an antibody?

Human antibodies are Y-shaped proteins, measuring around 10 billionths of a metre (10 nanometres), that are highly specific to combat whichever infection a person has previously encountered.

They work by recognising and sticking to specific proteins, in this case, proteins on the surface of the COVID-19 virus.

At the molecular level, the Y shaped proteins come in four parts, two that form the ‘trunk’ and tips, and two more that form the tips alone.

The structure of tips of the Y varies depending on the invader or virus, binding to them in a highly-specific way.

The stem and rest of the molecule falls into five major classes –  IgM, IgG, IgA, IgD, and IgE – which reflect where they are made in the body and if the Y-shaped molecules clump together.

Wellcozyme HIV 1+2' used to test for antibodies to HIV in blood samples, England, 1993.
`Wellcozyme HIV 1+2′ enzyme immunoassay kits, one of two, used to test for antibodies to HIV in blood samples, with 480 wells, without chemicals, from Murex Diagnostics Ltd, England, 1993.

What antibodies are we most interested in?

Some antibodies are better than others at providing protection from COVID-19. These ‘neutralising antibodies’ usually block the virus from attaching to a cell.

When are antibodies made?

Antibodies first become detectable by the most sensitive tests when the patient develops symptoms: the median (middle) incubation time from infection to symptoms is five days.

However, more typically, antibodies can begin to be detected by tests at 7 days earliest and reliably from around two weeks.

If you are infected with SARS-CoV-2, do you develop immunity?

Surprisingly little is known about how our immune system responds to COVID-19, how immunity develops, and for how long. Filling in these missing details is vital, not least because efforts to develop an effective COVID-19 vaccine depends on understanding how to create a strong and enduring immune response.

Encouragingly, one study by Linqi Zhang of Tsinghua University, Beijing, found antibodies with potent anti-SARS-CoV-2 neutralization activity that correlates with their ability to bind to the region of the virus spike protein that docks with a human protein (called angiotensin converting enzyme 2, better known by the acronym ACE2). This is the crucial first step that allows the virus to infect a human cell. These antibodies were specific to SARS-CoV-2, not to the other coronaviruses SARS-CoV or MERS.

Another study that explored neutralizing antibodies by George Gao of the Chinese Academy of Sciences, Beijing, and colleagues, isolated two antibodies from a recovering COVID-19 patient, CA1 and CB6. The latter CB6 antibody inhibited SARS-CoV-2 infection in rhesus monkeys by interfering with the way the virus binds to cells.

Can we use naturally made antibodies as treatments?

Antibodies are in plasma, the yellowish liquid which remains when red and white blood cells have been removed from blood. This is the basis of an old-fashioned treatment dating back to the 1918 pandemic that can be effective, in which a seriously-ill COVID-19 patient is given plasma collected from a recovered patient.

The United States has launched a national effort to roll out these blood-based therapies to more than 16,000 people while, in the UK, a trial has started using plasma collected by NHS Blood and Transplant.

‘The results are deliberately kept blinded to us as investigators, so that we don’t influence the trial in any way. However, we have been monitoring safety and we have no safety concerns at this stage,’ said Anthony Gordon, UK Chief Investigator of the REMAP-CAP trial.

Based on earlier research using convalescent plasma to treat Ebola, it is critical to select donors properly – that is, to only use plasma from recovered people who show a strong antibody response, with lots of neutralising antibodies – commented Richard Tedder, Visiting Professor in Medical Virology, Imperial College London.

Tedder has also contributed to A Message from Corona, an ebook by his daughter to help children cope with COVID-19.

Dried human plasma bottle with foil covered cap.
Dried human plasma by Lister Institute, Britain, 1971 from the North East Thames Regional Transfusion Centre.

How do we test for antibodies?

There are four basic ways to show if antibodies have been made in response to a COVID-19 infection.

Known as ELISA tests (enzyme-linked immunosorbent assay), they all depend on detecting the reaction between a virus protein (known as the antigen) and the antibody.

For example, one uses a thin layer of proteins from the virus that are known to provoke an immune reaction in a tray of plastic wells. If antibodies are present in a person’s blood sample, they will attach themselves to the protein in the wells. A chemical reaction – and colour change – signals the presence of the antibodies.

What can an antibody test tell us?

If a person has antibodies, it might be because of a recent and active SARS-CoV-2 infection, or because they had an infection in the past (which may or may not have been accompanied by symptoms).

These tests can reveal the degree of immunity that is present in a population which, for example, is the aim of the Swiss ‘Corona Immunitas’ programme.

Between April and October 2020, through long-term monitoring of antibody development, studies will determine the spread of coronavirus in various regions, populations, and occupational groups.

The antibody test complements direct tests for the virus which work best during the first days of infection. Using swabs from the nose and throat, locations in the body where the virus is most likely to be found, tiny quantities of viral genetic material are multiplied to detectable levels using a technique called PCR, or polymerase chain reaction.

Prototype polymerase chain reaction (PCR) machine
‘Baby Blue’ – a prototype polymerase chain reaction (PCR), c 1986. This machine automates the process of making large quantities of DNA from a tiny starting sample.

What about the new antibody tests used by the Government?

The government has announced the start of a major new national antibody testing programme, with plans to provide antibody tests to NHS and care staff in England from the end of May.

Public Health England has evaluated commercial diagnostic kits that detect antibodies made against the ‘N protein’ of the virus, the ‘nucleocapsid protein’ that helps package up the virus’s genetic code and plays a role in its manufacture.

One test, made by Abbott, was highly specific (100 per cent, so it correctly identifies all those who have been infected) and its sensitivity was 93 per cent, so it correctly identifies 93 per cent of people who have not been infected.

A second, developed by Roche, was just under 100 per cent specific and its overall sensitivity was around 84 per cent, though the manufacturer reported 100 per cent.

However, antibody tests aimed at internal viral proteins – the N protein in this case  – are not ideal targets for an antibody test and are at greater risk of false positives, according to Tedder.

Along with Myra McClure, Professor of Retrovirology at Imperial, Tedder is developing a sophisticated test for neutralising antibodies that bind to the ‘receptor binding domain’, the part of the coronavirus spike that sticks to human cells.

‘That target, RBD, would be the target of choice for an antibody test,’ said McClure.

There is another problem with antibody testing that arises when the prevalence of an infection in a population is low (it is likely less than 10 per cent in the UK), when the total number of people who receive false positives can match or even exceed the number receiving true positives (find out for yourself using this web tool).

For example, in a population where the prevalence is 5%, a test with 90% sensitivity and 95% specificity will yield a positive predictive value of 49%. In other words, less than half of those testing positive will truly have antibodies.

What about home antibody tests?

The government is also working in partnership with the private sector to develop a ‘finger-prick’ type test, similar to a diabetes test, which will be suitable for use at home. The Government is not offering them at present.

Richard Tedder commented that finger-prick style tests would probably have to be checked with a laboratory test to be confident in the results. Ideally, he added, a home test would not rely on using blood but a swab, taken from the mouth, that could be reliably tested in a central laboratory.

Throat swab used to diagnose diphtheria
A swab with cotton wool on a metal rod, in a glass tube, in wooden case, supplied by Kent County Council, early 20th century, for diphtheria diagnosis.

How long will antibodies persist?

Neutralizing antibodies are typically only produced in the body for a limited time after initial infection, past research with coronaviruses shows these neutralizing antibodies last one or two years.

But we still don’t know how long immunity to SARS-CoV-2 will endure.

However, while one study found that patients who recently recovered from COVID-19 produce virus-specific antibodies and T cells, the responses of different patients are not all the same, and understanding what immune responses are the most effective is crucial for developing vaccines.

The study by Chen Dong of Tsinghua University and colleagues assessed the levels of immunoglobulin M (IgM) antibodies, which are the first to appear in response to an infection, as well as immunoglobulin G (IgG) antibodies, which are the most common type found in blood circulation.

Compared to healthy controls, both newly discharged and follow-up patients showed higher levels of IgM and IgG antibodies that bind to the SARS-CoV-2 nucleocapsid protein, mentioned earlier. In addition, the receptor-binding domain, RBD, which binds to receptors on host cells during the process of viral entry.

The latter, which is also the best marker for infection, induced both antibody and T cell responses. “Our results suggest that S-RBD is a promising target for SARS-CoV-2 vaccines,” says co-senior author Fang Chen of Chui Yang Liu Hospital’ affiliated to Tsinghua University. “But our findings need further confirmation in a large cohort of COVID-19 patients.”

Other studies have also shown that infected people have T cells that target the virus which is encouraging regarding long term immunity and could also help researchers develop better vaccines.

Can studies of the immune response to the virus help develop vaccines?

Very much so. Vaccines work by priming the body’s immune system to recognise and remember a virus so that the body can quickly respond when under attack.

Vaccines train the body to produce its own antibodies by introducing a carefully chosen part of the SARS-CoV-2 virus – typically a molecule from its outer surface, such as its spike protein, or a weakened or inert version of the entire virus.

Various studies are underway into detailed features of the immune response to the COVID-19 virus, notably what aspects that allow people to contain the virus without showing symptoms, or those that are linked with serious symptoms, for instance by Ajit Lalvani, Chair of Infectious Diseases at Imperial College London, and by Jörg Goldhahn at the ETH, Zurich.

Antibodies from infected individuals can be used to identify targets for vaccine development, as shown by this study by the ETH team.

Large, well-designed studies of the immune response of people will reveal ‘correlates of protection’, features of the immune response that allow people to successfully contain the virus.

Finding the correlates of protective immunity ‘will guide rational design and evaluation of novel vaccines as well as the development of new treatments that could enhance beneficial immune responses in patients with symptoms to hasten recovery,’ said Lalvani.

How are antibody studies helping vaccine development?

Preliminary studies in monkeys now suggest that that humans may develop immunity after being vaccinated or after recovering from COVID-19.

“We demonstrate in rhesus macaques that prototype vaccines protected against SARS-CoV-2 infection and that SARS-CoV-2 infection protected against re-exposure,” said Dan Barouch, co-author and the director of the Center for Virology and Vaccine Research at the Beth Israel Deaconess Medical Center in Boston.

Six prototype vaccines were tested, based on DNA instructions for the virus spike protein — when injected, they program the body to make the protein and train the immune system to recognise the virus. The vaccines prevented 25 rhesus macaques from becoming infected by the novel coronavirus.

His team observed several kinds of antibodies in the animals’ blood, including neutralizing antibodies that latch onto the virus to prevent it from infecting cells, and various kinds of white blood cells called T cells, which rouse the immune system and kill infected cells.

Monkeys in the groups that received vaccines containing the virus’s entire spike protein fared better than their peers who received parts of the spike protein.

In a related experiment, nine macaques that tested positive for the virus were protected from re-infection when they encountered it a second time.

The bottom line of this animal study is that natural protective immunity to this virus seems to exist and this ‘increases our optimism that the development of a vaccine is likely going to be possible,” said Barouch. Barouch is working with Johnson & Johnson on a coronavirus vaccine that uses a specially modified virus, a type of adenovirus, that he developed.

How is COVID-19 vaccine research going?

Worldwide, more than 100 COVID-19 vaccines are under development according to the Coalition for Epidemic Preparedness Innovations (CEPI). They are listed here by the World Health Organisation.

Even if this effort pays off, and that is being optimistic, the earliest we might see a vaccine is around 18 months.

An adenovirus – another kind of virus that causes colds – had been genetically altered by the company CanSino to make the spike protein of the coronavirus and has been tested in Wuhan, China, the epicentre of the outbreak by Wei Chen of the Beijing Institute of Biotechnology, and colleagues.

The 108 participants received low, middle, and high doses of the vaccine, and aside from pain at the injection site, fever and fatigue, side effects were mild.

They observed how neutralising antibodies increased significantly at day 14, along with a peak in T-cells, and conclude this vaccine ‘warrants further investigation,’ though some are underwhelmed.

Meanwhile, tests on macaques of a Sinovac Biotech vaccine, based on inactivated COVID-19 virus, produced neutralizing antibodies and partial or complete protection.

Early animal test results have been reported of an Oxford University vaccine –  a chimpanzee adenovirus modified to make the COVID-19 spike protein –  which has been given to 1,000 adult volunteers and is about to be tested on 10,000 people, including older adults and children.

Six rhesus macaque monkeys were given the vaccine and three – the ‘control group’ – were given the adenovirus with a ‘reporter gene’. However, the study by Sarah Gilbert and her team has not yet been peer reviewed – a key part of the scientific process, when experts identify weaknesses in its assumptions, methods, and conclusions.  Some are unconvinced by the results.

Gilbert said the study showed that a theoretical harmful side-effect of the vaccine – ‘immune enhanced disease’ – did not occur.

‘A couple of months ago everyone was very worried about possible enhanced disease but now there have been several non-human primate studies published with no evidence of that problem, the focus has shifted to look at how protective the vaccines are.’

Mike Francis, a member of the UK vaccine network, added: “Whilst their vaccine did not appear to prevent infection of the macaques, it did reduce the severity of disease.

Indeed, whilst some vaccines can elicit “sterilising” immunity, others do simply act by modifying the disease outcome. Such a vaccine could still be very useful as against COVID-19 in vulnerable individuals.’

On 18 May, US biotech firm Moderna said its COVID-19 vaccine triggered an immune response in people and protected mice from lung infections with SARS-CoV-2.

The vaccine tricks the body into producing viral protein by using mRNA, or messenger RNA, the molecule that puts DNA instructions into action in cells.  However, scientists criticised the lack of hard data.

Much more research is needed to work out the duration of protection and to optimise COVID-19 vaccines because the current studies are small and, of course, when it comes to animal research, macaques do not perfectly match how the human immune system behaves.

Can we predict which vaccines might work?

A computer model is under development to predict how a cell is infected by a virus, from when it first invades to manufacturing mature virus to when fragments of viral proteins (broken down by the host cell, and called epitopes) eventually appear on the cell’s surface. This triggers the immune response.

The work by Microsoft and Peter Coveney’s team at CompBioMed, University College London has focused on HIV but, he said, as we understand more details of underlying mechanisms, computer modeling can help vaccine development as part of a broader effort to create ‘digital twins’ of people.

If a test shows someone is immune, should they get an ‘immunity passport’?

Testing populations to provide snapshots of their infection history and immunity as the pandemic progresses could help address critical public health questions, such as when to relax stay-at-home orders or school closures, said Juliet Bryant of the  Laboratory of Emerging Pathogens, Fondation Mérieux, Lyon, France.

Bryant has weighed up the pros and cons  with a large international team led by Daniel Leung of the University of Utah School of Medicine.

But the tests would, in fact, need to be nearly perfect to be used for ‘immune passports’ under current conditions.  That’s because the ability of a test to accurately assess immune status depends critically upon how common infections are within a given community.

As mentioned earlier, if, for example, only 5% of the population has been infected because transmission has not taken off within an area, and COVID-19 cases are still relatively rare, then a test with 96% specificity and 90% sensitivity could yield misleading results.

As a result of this, not all people who test positive would be true positives.

Bryant said it is all about the mathematics of diagnostics, and how tests perform in situations where diseases are rare. Indeed, researchers are saying that specificity is the essential thing to look at under these conditions.

Specificity needs to be nearly perfect (100%) in situations of low prevalence, to ensure that people who test positive are true positives.

‘The real worry is the false positives,’ said Bryant. ’Because those would be the people who think they are immune (and who might get a ‘passport’) but in fact, they haven’t been exposed to the virus and are actually susceptible.  i.e. They shouldn’t feel ‘safe’ about resuming normal life.’

‘Even if we had a perfect antibody test, we still don’t know how that translates to protection,’ added Daniel Leung.

‘What the detection of SARS-CoV-2 antibody mean for an individual’s protection and immunity (how this may vary across diverse populations, how it is influenced by earlier infections, for instance with other coronaviruses, and most notably, how long it lasts) is still poorly understood.

This transmission electron microscope image shows SARS-CoV-2, the virus that causes COVID-19, emerging from the surface of cells
This transmission electron microscope image shows SARS-CoV-2, the virus that causes COVID-19, emerging from the surface of cells.
The image was captured and colorized at NIAID’s Rocky Mountain Laboratories (RML) in Hamilton, Montana.

To use these antibody tests as the basis of “immune passports” that certify an individual’s immunity is dangerously premature, and could even be harmful, as it could provide a false sense of security against reinfection.’

Are there ethical issues with immunity passports?

Yes. Ten reasons why immunity passports are a bad idea were listed in the journal Nature by Natalie Kofler of Harvard Medical School and Francoise Baylis of Dalhousie University: ‘Four huge practical problems and six ethical objections add up to one very bad idea.’

The ethical issues are: monitoring erodes privacy; marginalized groups will face more scrutiny; the wealthy and powerful are more likely to obtain a test than the poor and vulnerable; labeling people on the basis of their COVID-19 status would divide society (into the immunoprivileged and the immunodeprived); and it might drive new forms of discrimination.

Discrimination may occur as immune certification could easily be expanded to include other forms of personal health data, such as mental-health records and genetic-test results.

“Some proponents of immunity passports argue for personal freedom and that any such program could be designed to avoid inequity and discrimination,” said Kofler. “But data show that structural inequities are contributing to disproportionate health impacts of COVID-19 on communities of colour in the U.S. and the U.K.

It would be naive to think that an immunity passport program wouldn’t further disenfranchise those communities”

The practical problems are: COVID-19 immunity is a mystery; serological tests are unreliable; the volume of testing needed is unfeasible and the people who could possibly benefit make up no more than 2–3% of the global population who are estimated to have recovered from the virus.

They are also concerned that passports could create perverse incentives for healthy, non-immune individuals to wilfully seek out infection — putting themselves and others at risk.

What can studying antibodies tell us about treatments?

Neutralizing antibodies could themselves be used for prevention or as a treatment, an approach explored by an international team led by Davide Corti at Vir Biotechnology and David Veesler at the University of Washington.

Antibodies derived from SARS survivors could potently block entry of SARS-CoV-2 and other closely related coronaviruses into host cells. This is according to their research on antibodies from the B cells of a person who was infected with the SARs coronavirus in 2003, which drew on data gathered at Berkeley National Laboratory’s Advanced Light Source.

Their previous research on the SARS and MERS-causing coronaviruses revealed that some neutralizing antibodies produced in response to those diseases were also effective against closely related coronaviruses. So, they suspected that several might inhibit SARS-CoV-2, related to SARS-CoV.

They found eight antibodies that can bind to the SARS-CoV-2 spike that enables it to invade and take over human cells.  Further tests narrowed the field to reveal one SARS-CoV antibody, called S309, that successfully neutralizes SARS-CoV-2, and is already scheduled for clinical trials.

The virus responsible for the 2002-3 outbreak of severe acute respiratory syndrome (SARS), known as SARS-CoV, shares approximately 80% of its genetic sequence with SARS-CoV-2, which causes COVID-19.

Can we expect antibodies to other coronaviruses to help fight COVID-19?

Patients infected with either severe acute respiratory syndrome coronavirus (SARS-CoV) or SARS-CoV-2 can produce antibodies that bind to the other coronavirus, according to a study by Chris Mok of the University of Hong Kong and colleagues.

However, the cross-reactive antibodies are not cross-protective, at least in cell-culture experiments, so it remains unclear whether such antibodies offer cross protection in the human body. The findings suggest that more research is needed to identify parts of the virus that are critical for inducing a cross-protective immune response.

“There are related viruses still circulating in bats, and it is unclear whether any of these may also threaten human health in the future,” says co-senior study author Malik Peiris of the University of Hong Kong. “As such, whether infection by one of these viruses cross-protects against another is an important question.”

Any more antibodies that show promise as treatments?

Since the start of the COVID-19 outbreak, there has been a global quest to find useful antibodies.

A llama antibody that binds to the spike protein of SARS-CoV-2 was discovered by Xavier Saelens and Nico Callewaert at the Ghent University and VIB in Belgium, in collaboration with the lab of Jason McLellan at the University of Texas, Austin.

When llamas’ immune systems detect foreign invaders such as bacteria and viruses, these animals (and other camelids such as alpacas) produce two types of antibodies: one that is similar to human antibodies and another about half the size.

From these smaller ones, one can isolate so-called single-domain antibodies or nanobodies, that can be made at large scale, nebulized and potentially used in an inhaler, so they can be delivered to the site of the COVID-19 infection

For this study, the scientists were helped by Winter, a 4-year-old Belgian llama that had been injected with spike proteins from SARS-CoV-1 and MERS-CoV. They isolated a small antibody derived from llamas that also binds to the SARS-CoV-2 spike protein, preventing the virus from infecting human cells.

The team has engineered a new antibody that shows promise for treating SARS-CoV-2 by linking it with a part of a human antibody, thereby making it more potent (the llama was named after the season, not the UK scientist Greg Winter, who pioneered humanized antibodies, said Saelens).

When it comes to patients, ‘We are planning for a clinical trial later in 2020 or early in 2021,’ said Saelens.

Black Llama
Winter, a 4-year-old Belgian llama that had been injected with spike proteins from SARS-CoV-1 and MERS-CoV. Credit: Tim Coppens.

How can I find out more?

The latest picture of how far the pandemic has spread can be seen on the Johns Hopkins Coronavirus Resource Center or Robert Koch-Institute.

You can check the number of UK COVID-19 lab-confirmed cases and deaths along with figures from the Office of National Statistics.

There is more information in my earlier blog posts (including in German by focusTerra, ETH Zürich, with additional information on Switzerland), from the UKRI, the EU, US Centers for Disease Control, WHO, on this COVID-19 portal and Our World in Data.

The Science Museum Group is collecting objects and ephemera to document this health emergency for future generations.