Immune Enhancement: The Dark Side of Antibodies and Implications for COVID-19 Vaccine Development
With any coronavirus, literature has shown that the potential for complications and enhanced disease due to pre-existing antibodies exists independent of vaccine technology. This must be considered carefully when designing vaccine candidates or treatment approaches. The best way forward to safely investigate the longevity of protective immunity, cross-reactivity between coronavirus strains, and the potential for antibody-dependent disease enhancement is perhaps the design of sizeable human challenge studies, using less virulent coronaviruses or attenuated strains of SARS-CoV-2. The latter would be more specific due to genetic and immunogenic similarity and would be able to provide higher confidence in the results.
Since the genome sequence of the SARS-CoV-2 virus was published in January 2020, researchers worldwide have been developing a vaccine to limit the severe impact on public health the pandemic continues to cause. As of April 2020, well over a hundred candidates are being investigated, many of which are at exploratory or preclinical stages. Some have recently moved into human clinical trials, and individuals have argued for the accelerated development and fast-tracking of vaccine testing and approval 1. Most SARS-CoV-2 vaccines are designed to target the highly immunogenic spike protein, which is located on the external membrane surface of the virus and enables virus attachment and entry into host cells. But, despite being under intense stabilizing selective pressure, the spike protein sequence is also one of the most variable regions on the SARS-CoV-2 genome 2. Vaccines designed to target epitopes on the spike protein may thus lead to inefficient antibody binding in new variants of the circulating SARS-CoV-2 virus and result in viral antigenic evolution towards escaping antibody recognition 3.
Besides potentially being non- or only partially protective and possibly leading to viral evolution, potential risks are associated with vaccines that necessitate extreme caution when vaccine development efforts are to be fast-tracked and approval accelerated. A fairly common phenomenon documented for many other viral diseases is immune enhancement, in which the patient shows augmented symptoms and enhanced disease severity due to pre-existing antibodies. The phenomenon has been observed for antibodies derived from previous infection with the same or closely related viruses or from vaccination.
What is Immune Enhancement?
Immune enhancement is an antibody-mediated exacerbation of viral illness, which usually has two aspects to it: Antibody-dependent enhancement (ADE) describes the process of antibody-mediated viral entry into cells of the immune system expressing an Fc receptor. Antibodies contain two binding sites, a variable Fab portion that recognizes the foreign antigens and an invariable Fc portion. The latter functions in presenting the immunogen to cells of the immune system for neutralization. In ADE, the antibody will recognize and bind the virus particle with its Fab portion and subsequently attach to the Fc-receptor. Fc-receptor binding then enables the entry of antibody-virus complex into such immune cells, including myeloid-derived cells such as monocytes and macrophages. Instead of being neutralized, the virus persists in immune cells, causing an aberrant downstream cascade of immune effects and may replicate further. This antibody-induced immunopathology is characterized mainly by exacerbated severity of inflammatory symptoms through antibody-mediated effector functions. Here, Fc-receptor activated cells initiate excessive complement activation and antibody-dependent cellular cytotoxicity, potentially leading to severe disease in otherwise fairly innocuous respiratory viruses 4.
In the past, ADE was documented specifically for antibodies with a sub-neutralizing or non-neutralizing effect on virus particles, usually due to an incomplete fit of antibody and epitope. It has now also been observed in fully neutralizing antibodies via an alternative mechanism.
Researchers found initial evidence for ADE in feline infectious peritonitis virus (FIPV). Here, coronavirus entry into macrophages is facilitated, increasing viral progeny, and upregulating the production of pro-inflammatory tumor necrosis factor. The resulting immunopathology leads to augmented apoptosis (cell death) in leukocytes 5.
Although only few clinical associations have been reported for ADE and diseases, there are some significant instances. Previous infection with one serotype of dengue virus (DENV), for example, causes a subsequent secondary infection with another distinct serotype (heterologous serotype) to be much more severe. ADE in DENV alters antiviral mechanisms to suppress type 1 interferon (IFN) production, increasing DENV replication 6.
The ADE effect is also documented for pre-existing antibodies resulting from a vaccination as opposed to a previous infection in other diseases, such as respiratory syncytial virus (RSV) and atypical measles. Here, vaccination with inactivated virions predisposed persons to more severe illness. The presence of immune sera can also trigger ADE in MERS-CoV, SARS-CoV, Zika virus, and HIV 4,7. These observations are a reason for concern due to the close genetic relatedness of SARS-CoV-2 to MERS-CoV and SARS-CoV.
Enhancement mechanisms of viral infections
Coronaviruses SARS-CoV, MERS-CoV, as well as SARS-CoV-2 use an external spike (S) protein for attachment, fusion with the cell membrane and entry into host cells, employing a structurally similar receptor-binding domain (RBD). SARS-CoV and SARS-CoV-2 bind to the ACE2 receptor (angiotensin-converting enzyme 2) found on lung alveolar epithelial cells and enterocytes of the intestine. Cell entry under normal conditions proceeds via pH-dependent proteolytic activation of spike and endocytosis. The process depends on the binding of RBD of the viral spike protein to the host cell receptor leading to its cleavage by host protease enzymes. After part of the spike protein is removed, the remaining portion undergoes a conformational change to be able to fuse the host and viral membranes for entry 8.
Coronaviruses, however, have evolved a secondary cell infection pathway that relies on the presence of antiviral immune serum, or host antibodies, and is independent of pH and cellular protease enzymes. They effectively exploit the host’s immune reaction by infecting immune cells 5,9. This way, the virus can gain entry into an entirely new cell type that does not carry the ACE2 receptor. This different mechanism is relevant as many other instances of ADE in other viruses only improve the infectivity to already susceptible cell types 5.
The alternative entry pathway depends on the presence of Fc or complement receptors on immune cells, particularly the FcγRII (CD32) receptor found on human leukocytes 5. When antibodies attach to the viral epitope on the spike protein, the ADE2 receptor-dependent entry to epithelial cells is blocked. Instead, the antibody-immune complex now functionally mimics the viral receptor on epithelial cells. Thus, it facilitates access to immune cells, including macrophages, B cells, and monocytes 7,8. IgM antibody subclasses, in particular, have been implicated for ADE of several viruses, including DENV, Ebola, and HIV 8,9. Furthermore, ADE can be caused independently of antibody origin by pre-existing antibodies raised through vaccination, infection, or passive maternal immunity 10.
Nonetheless, different vaccine candidates for SARS-CoV, for instance, despite being equally effective to block ADE2-dependent viral entry, have been shown to differ in their effectiveness in causing ADE 5. Through the synthesis of antibodies targeting different antigenic regions on the spike protein of the SARS virus, in vitro studies have identified that particular protein sequences on the spike protein differ in their immunogenicity as well as their capacity to elicit ADE. Thus, an epitope sequence-dependent (ESD) mechanism for enhanced infectivity of SARS-CoV both in vitro and in non-human primates was identified 9,11.
Since antibodies raised in a natural infection may target different epitopes on viral proteins, they may represent a mixture of neutralizing and enhancing species, partially counteracting each other 11. Initially, researchers assumed that only antibodies that could not completely neutralize or block entry of virions caused ADE due to their sub-optimal function and poor fit on viral epitope targets. It has been shown, however, that even antibodies that completely block entry via ACE2 receptors can facilitate access through the alternative Fc-mediated pathway 8.
Other factors that determine whether a particular immune serum or vaccine strategy will cause neutralization or enhancement are complex. In some models, neutralization requires more than one antibody to bind to an individual epitope to neutralize the virus completely, protection thus depending on the quantitative ratio of antibodies versus virions 10. Alternatively, strongly neutralizing antibodies for important epitopes may inhibit infection at lower concentrations, whereas weakly neutralizing antibodies binding less critical epitopes might be required at much higher concentrations to afford protection 11. Hence, the difference between neutralization and enhancement of infection depends on multiple variables, including antibody class and type, relative concentrations or titers, viral epitope variant, as well as other immune system components 10.
Interestingly, allelic polymorphisms in Fc receptors in individuals also affect disease severity of SARS-CoV. As different IgG subclasses bind to different Fc receptor types, persons with a specific isoform of receptor able to bind several IgG subclasses instead of just one suffered worse pathology, presumably through the increased incidence of ADE 9.
Evidence for ADE in coronaviruses
During the 2003 SARS epidemic, a nonintuitive relationship between the quantities of antibodies and disease outcomes in patients was observed. Patients who developed a robust antibody response earlier during the infection developed more severe symptoms than patients who only produced antibodies later 4. It was then discovered that despite primarily being a respiratory pathogen, immune and hematopoietic cells not expressing the ACE2 receptor used for viral entry were also infected (Jaume et al., 2011). Antibody-dependent enhancement as a mechanism of potential disease exacerbation was identified in 2005 and implicated in the unusually high mortality rate of people in China 12.
Findings were confirmed through in vitro studies, in which SARS virus infection of macrophage cells was reliant on the presence of IgG immune serum targeting the SARS spike protein in a concentration-dependent manner 10,13,14. The relationship between antisera and SARS disease enhancement is complex, however, and has prevented the development of a safe and effective vaccine. Animal studies showed the immunization with a recombinant full-length spike protein to afford protection, but to cause viral infection of human immune cells in vitro, while double-inactivated vaccine enhanced pulmonary inflammation in mice 5,9. In viral vector vaccines, immunization targeted at different epitopes could induce disparate effects. IgG antibodies were generated to both spike protein and nucleocapsid protein, yet upon re-challenge, only mice immunized with anti-nucleocapsid vaccines developed excessive inflammatory response and augmented lung pathology 9. There is no evidence to suggest that virions gaining entry into immune cells via the ACE2-independent pathway can replicate productively and spread further, instead of terminating in what is called abortive infection. Despite this, immunopathological responses result from immune cells’ invasion, promoting inflammation and tissue damage 9. The evidence thus suggests that alternative cellular pathways play a role in SARS pathogenicity and can significantly affect disease outcomes.
The clinical presentation of infections with SARS-CoV-2, primarily a respiratory pathogen, initially puzzled researchers. Yet, there are many similarities between the current pandemic and the SARS outbreak in 2003, indicating an essential role for ADE. While the presence of SARS-CoV-2-specific IgM and IgG antibodies is used as a diagnostic tool, associations between high antibody titers and poor disease outcome were discovered 15. While IgM levels appear independent of illness severity, early appearance of IgG and overall higher neutralizing antibody titers are associated with more severe cases similar to what was observed in SARS-CoV 3,14. A correlation also exists between the advanced age of patients, higher IgM and IgG levels, and worse outcomes 9. Notably, the most accurate predictor of disease prognosis is lymphocyte count 16.
The general pattern of relatively mild disease occurring in younger people can also be explained in terms of ADE. As with SARS-CoV, cases in children are generally less severe or asymptomatic, and complications are rare. This relationship has been widely documented in other, less pathogenic coronaviruses such as 229E, NL63, OC43 7. In children under six years of age, anti-CoV IgG antibodies are absent but increase in prevalence soon after due to the widespread exposure to common CoVs and subsequent seroconversion. Older people have a broader repertoire of anti-CoV antibodies produced by long-lived plasma cells. Thus, cross-reactions might occur between IgGs raised during previous infections with commonly circulating strains of CoVs and SARS-CoV-2 in this population group 7,17.
Prior exposure to SARS-CoV or other bat-derived coronaviruses was also suggested in regions experiencing especially severe disease progression, such as Hubei province in China, potentially indicating the presence of pre-existing antibodies against similar antigenic epitopes 12. Hubei province is considered the potential introduction site of SARS-CoV to humans, and two specific epitopes on the SARS-CoV spike protein exhibit 72.7 and 100 % similarity to the corresponding epitopes on the SARS-CoV-2 spike. Even more antigenic similarity was revealed for certain bat CoV strains, making serological interactions likely in potential successive infections 12.
Many aspects of the symptomatic picture described for severe infections with SARS-CoV-2 further indicate the incidence of ADE, such as the invasion of endothelial and immune cells lacking ACE2 receptors but expressing specific Fc receptors17. Monocytes and macrophages express the CD32a receptor (implicated explicitly in the antibody-mediated viral entry) and release pro-inflammatory cytokines upon infection. High levels of cytokines, chemokines, and immune dysregulation are frequently observed in severe or critical cases 3. Further consistent with immune-mediated injury are the findings of tissue damage in heart, liver and kidneys, pulmonary infiltration of neutrophils and macrophages, as well as spleen and lymph node necrosis in fatal cases 3,12,15. Underlying conditions that have been linked to especially poor outcomes, such as cardiovascular disease and diabetes mellitus, are known to coincide with increased cellular expression of CD32a receptors on monocytes and macrophages 18.
Thus, ADE as a significant pathomechanism is probable, especially if potentially activated by neutralizing as well as non-neutralizing antibodies targeted at epitopes of varying similarity to SARS-CoV-2 18.
Implications for vaccine development and treatment approaches
Scientists and policymakers in different countries have been proposing various public health strategies in light of the SARS-CoV-2 pandemic, many of which rely on the discovery of an effective vaccine. Much of the assumptions these solutions are based upon, however, are not grounded in robust science. Unknown, for instance, is if an infection with SARS-CoV-2 will lead to immunity and protection against subsequent disease, or how long this protection would last. Already, signs of multiple infections with different forms of SARS-CoV-2 in a single individual have been documented 2.
Human challenge studies with other common, less pathogenic CoVs have shown that protection may last for only 1 to 2 years, and little is known about the amount of cross-reactivity between different strains 14. For instance, antibodies to the nucleocapsid protein of alpha-coronaviruses protect against subsequent infections within the alpha-coronavirus group, with the same observed for beta-coronaviruses. However, no protection is afforded between the different groups. Additionally, in some coronaviruses studied, pre-existing high neutralizing antibody titers pre-disposed individuals to worse cold symptoms upon viral challenge, again indicating the occurrence of ADE 14.
In SARS-CoV and MERS-CoV, spike protein epitope sequence-dependent disease enhancement has been observed11, and ADE has been linked to the cytokine storm syndrome in the most severe cases of MERS, SARS-CoV, and SARS-CoV-2 7. Moreover, clinical trials for DENV and RSV vaccines were terminated due to ADE in the past.
Overall, the evidence is ambiguous, however, and points to many remaining unknown factors rather than likely conclusions. Many of the concerns raised for the potential of ADE originate from in vitro studies or animal models in experimental settings, which can be extrapolated to clinical implications only with little certainty.
Other reports have shown sequential infection with human coronaviruses without indications of disease enhancement, for instance, and some SARS-CoV and MERS-CoV vaccine candidates have proceeded to early-phase clinical trials. There are also animal studies in which vaccines targeted at CoV spike proteins showed strong immunogenicity and provided protection upon challenge 4.
Lower SARS and COVID-19 disease severity in children may also result from other factors besides the lack of CoV antibody prevalence, such as macrophage and memory T cell immaturity found in infants and adolescents, respectively, and a smaller memory B cell diversity 7.
With this uncertainty, many other factors that can potentially influence whether an antibody becomes enhancing or protective have not yet been investigated. These include the adjuvant used, the route of immunization, or vaccine protein glycosylation 4,7,9 in addition to age at vaccination and prior exposure to CoVs. The effects of antibody-specific variables such as concentration, specificity, antigen target, affinity, and isotype are to be elucidated 9,18. Other critical gaps in the knowledge concern antibody kinetics in subclinical infections. These affect the validity of antibody assays used in diagnostics, since antibody titers, whether neutralizing or not, are not necessarily linked to lasting immunity 14,19.
The best way forward to safely investigate antibody kinetics, cross-reactivity, and the potential for ADE is perhaps the design of sizeable human challenge studies, using less virulent CoVs 4 or attenuated strains of SARS-CoV-2. The latter would be more specific due to spike protein similarity and would be able to provide higher confidence in the results. With any coronavirus, literature has shown that the potential for ADE exists independent of vaccine technology and must be considered carefully when designing vaccine candidates or treatment approaches.
1. Thanh Le, T. et al. The COVID-19 vaccine development landscape. Nat. Rev. Drug Discov. 19, 305–306 (2020).
2. Korber, B. et al. Spike mutation pipeline reveals the emergence of a more transmissible form of SARS-CoV-2. bioRxiv 2020.04.29.069054 (2020). doi:10.1101/2020.04.29.069054
3. Ricke, D. & Malone, R. W. Medical Countermeasures Analysis of 2019-nCoV and Vaccine Risks for Antibody-Dependent Enhancement (ADE). SSRN Electron. J. (2020). doi:10.2139/ssrn.3546070
4. de Alwis, R., Chen, S., Gan, E. S. & Ooi, E. E. Impact of immune enhancement on Covid-19 polyclonal hyperimmune globulin therapy and vaccine development. EBioMedicine 55, (2020).
5. Jaume, M. et al. Anti-Severe Acute Respiratory Syndrome Coronavirus Spike Antibodies Trigger Infection of Human Immune Cells via a pH- and Cysteine Protease-Independent Fc R Pathway. J. Virol. 85, 10582–10597 (2011).
6. Ubol, S. & Halstead, S. B. How innate immune mechanisms contribute to antibody-enhanced viral infections. Clin. Vaccine Immunol. 17, 1829–1835 (2010).
7. Wang, J. & Zand, M. S. The Potential for Antibody-Dependent Enhancement of SARS-CoV-2 Infection: Translational Implications for Vaccine Development. J. Clin. Transl. Sci. 1–11 (2020). doi:10.1017/cts.2020.39
8. Wan, Y. et al. Molecular Mechanism for Antibody-Dependent Enhancement of Coronavirus Entry. J. Virol. 94, 1–15 (2019).
9. Iwasaki, A. & Yang, Y. The potential danger of suboptimal antibody responses in COVID-19. Nat. Rev. Immunol. 1–3 (2020). doi:10.1038/s41577-020-0321-6
10. Smatti, M. K., Thani, A. A. Al & Yassine, H. M. Viral-Induced Enhanced Disease Illness. Front. Microbiol. 9, (2018).
11. Wang, Q. et al. Immunodominant SARS Coronavirus Epitopes in Humans Elicited both Enhancing and Neutralizing Effects on Infection in Non-human Primates. ACS Infect. Dis. 2, (2016).
12. Tetro, J. A. Is COVID-19 receiving ADE from other coronaviruses? Microbes Infect. 22, 72–73 (2020).
13. Yip, M. S. et al. Antibody-dependent infection of human macrophages by severe acute respiratory syndrome coronavirus. Virol. J. 11, 1–11 (2014).
14. Huang, A. T. et al. A systematic review of antibody mediated immunity to coronaviruses: antibody kinetics, correlates of protection, and association of antibody responses with severity of disease. medRxiv 2020.04.14.20065771 (2020). doi:10.1101/2020.04.14.20065771
15. Cao, X. COVID-19: immunopathology and its implications for therapy. Nat. Rev. Immunol. (2020).
16. Tan, L. et al. Validation of reported risk factors for disease classification and prognosis in COVID-19: a descriptive and retrospective study. medRxiv 2020.04.05.20053769 (2020). doi:10.1101/2020.04.05.20053769
17. Peron, J. & Nakaya, H. Susceptibility of the Elderly to SARS-CoV-2 Infection: ACE-2 Overexpression, Shedding and Antibody-dependent Enhancement (ADE). Preprints (2020).
18. Kadkhoda, K. COVID-19: an Immunopathological View. mSphere 5, 19–22 (2020).
19. Hotez, P. J., Corry, D. B. & Bottazzi, M. E. COVID-19 vaccine design: the Janus face of immune enhancement. Nat. Rev. Immunol. 1–2 (2020). doi:10.1038/s41577-020-0323-4