Antibody-dependent enhancement


Antibody-dependent enhancement, sometimes less precisely called immune enhancement or disease enhancement, is a phenomenon in which binding of a virus to suboptimal antibodies enhances its entry into host cells, and sometimes also its replication. This phenomenon—which leads to both increased infectivity and virulence—has been observed with mosquito-borne flaviviruses such as Dengue virus, Yellow fever virus and Zika virus, with HIV, and with coronaviruses.
There are various hypotheses on how ADE happens and it is likely that more than one mechanism exists. In one such pathway, some cells of the immune system lack the usual receptors on their surfaces that the virus uses to gain entry, but they have Fc receptors that bind to one end of antibodies. The virus binds to the antigen-binding site at the other end, and in this way gains entry to and infects the immune cell. Dengue virus can use this mechanism to infect human macrophages, if there was a preceding infection with a different strain of the virus, causing a normally mild viral infection to become life-threatening.
ADE can hamper vaccine development, as a vaccine may cause the production of antibodies which, via ADE, worsen the disease the vaccine is designed to protect against. This is an important issue to be studied during the Phase 3 of vaccine development against COVID-19. Vaccine candidates for Dengue virus, and feline infectious peritonitis virus had to be stopped because they elicited ADE.

In coronavirus infection

Alphacoronavirus

The feline infectious peritonitis virus is an alphacoronavirus that is a very common pathogen in both domestic and wild cats. It is believed that FIPV can cause antibody-dependent enhancement. Thus, vaccination against FIPV surprisingly increases the disease seriousness. Moreover, it was demonstrated that infection of macrophages by FIPV in vitro can be triggered by non-neutralising monoclonal antibodies targeting the Spike -protein, and this phenomenon can also occur with diluted neutralising antibodies. ADE explains why half of cats develop peritonitis after being passively immunized with antivirus antibodies and being challenged with the same FIPV serotype. In several countries an attenuated virus vaccine is available in a form of nasal drops; however, its application is still considered controversial by many experts, both in terms of safety and efficacy. The mechanism of antibody-dependent enhancement for FIPV is illustrated in the video.

Betacoronavirus

The neutralization ability of an antibody on a virion is dependent on concentration and the strength of interaction between antibody and antigen. High-affinity antibodies can cause virus neutralization by recognizing specific viral epitopes. However, pathogen-specific antibodies can promote a phenomenon known as antibody-dependent enhancement, which can be induced when the strength of antibody-antigen interaction is below the certain threshold.
There are multiple examples of ADE triggered by betacoronaviruses.

ADE related vaccine development problems

The development of immunopathology upon exposure has been a major challenge for coronavirus vaccine development and may similarly impact SARS-CoV-2 vaccine research. Non-human primates vaccinated with modified vaccinia Ankara virus encoding full-length SARS-CoV spike glycoprotein and challenged with the SARS-CoV virus had lower viral loads but suffered from acute lung injury due to ADE. ADE has been observed in both severe acute respiratory syndrome and Middle East respiratory syndrome animal models allowing the respective viruses to enter cells expressing Fc?R including myeloid lineage cells.
ADE of acute lung injury has been documented in animal models of both SARS and MERS. Rabbits intranasally infected with MERS-CoV developed a pulmonary infection characterized by viremia and perivascular inflammation of the lung, and an antibody response that lacked neutralizing antibodies. The rabbits developed more severe lung disease on re-exposure to MERS-CoV, and developed neutralizing antibodies after reinfection. In SARS, mice vaccinated with four types of vaccines against SARS-CoV, as well as those infected with SARS-CoV itself, developed neutralizing antibodies. Mice were then challenged with live SARS-CoV, upon which all developed immunopathologic-type lung disease, although none had detectable virus two days after challenge and were protected compared to control.
It was shown that antibodies triggered by a SARS-CoV vaccine candidate, represented by recombinant, full-length SARS-CoV Spike-protein trimers, promote infection of immune cell lines. Fluorescence microscopy and real-time quantitative reverse transcriptase polymerase chain reaction were used for demonstration that anti-Spike immune serum increases infection of human monocyte-derived macrophages by SARS-CoV. Antibody-mediated infection was dependent on binding of immune complexes to cell surface FcγRII. Intact cytoplasmic signaling domains FcγRII were important for sustaining ADE of SARS-CoV infection. The authors of the study concluded that primary human immune cells where susceptible to infection by SARS-CoV in the presence of anti-Spike antibodies. The SARS-CoV-1 virus can enter macrophages via an antibody-mediated pathway and can replicate in these cells.

Mechanism of ADE for coronavirus infections

ADE in coronavirus infection can be caused by conformation changes of spike -protein or/and high mutation rate of the gene that encodes spike protein. A thorough analysis of amino acid variability in SARS-CoV-2 virus proteins, that included the S-protein, revealed that least conservative amino acids are in most exposed fragments of S-protein including receptor binding domain. Therefore, antigenic drift is a most likely cause of amino-acids variability in this protein and ADE. This drift can occur in a form of changes of both types of antigenic epitopes, including conformational and linear.

Potential linkage between pathophysiology of COVID-19 and ADE

The pathophysiology of SARS and COVID-19 diseases may be associated with ADE. The authors of the study believe that ADE is a key step in the progression of disease from its mild to severe form. Onset of ADE, due to antigenic drift, can explain the observed sudden immune dysregulation, including apoptosis of immune cells, which promotes the development of T-cell lymphopenia and an inflammatory cascade with the lung accumulation of macrophages and neutrophils, as well as a cytokine storm. ADE goes along with reduction of Th1 cytokines IL2, TNF-α and IFN-γ and increase of Th2 cytokines IL-10, IL-6, PGE-2 and INF-α, as well as with inhibition of STAT pathway.
Perhaps ADE is the reason why the course of SARS and COVID-19 is more severe for older people compared to younger people. It is likely that in older people the production of antibodies is slower and by the time the antibodies are developed in the titer that is sufficient to neutralize the virus, the virus changes its antigenic determinants. In this case, immuno-dominant neutralizing antibodies might start forming unstable complexes with the new form of the virus and start to infect monocytes/macrophages causing ADE. This process can trigger generalized infection of immune cells in multiple organs and cytokine storm.
It is interesting that in mice similar phenomenon of developing more severe disease in old compared to young animals exists. In contrast to old mice, in young mice, despite detectable viral replication оf SARS-CoV-1 in the lungs upon infection, clinical signs of the disease do not develop.
Also, an ongoing question in the COVID-19 pandemic is whether—and if so, to what extent—COVID-19 receives ADE from prior infection with other coronaviruses.

In influenza infection

Prior receipt of 2008–09 TIV was associated with an increased risk of medically attended pH1N1 illness during the spring-summer 2009 in Canada. The occurrence of bias or confounding cannot be ruled out. Further experimental and epidemiological assessment is warranted. Possible biological mechanisms and immunoepidemiologic implications are considered.
Natural infection and the attenuated vaccine induce antibodies that enhance the update of the homologous virus and H1N1 virus isolated several years later, demonstrating that a primary influenza A virus infection results in the induction of infection enhancing antibodies.
ADE was suspected in infections with influenza A virus subtype H7N9, but knowledge is limited.

In dengue virus infection

The most widely known example of ADE occurs in the setting of infection with dengue virus, a single-stranded positive-polarity RNA virus of the family Flaviviridae. It causes a disease of varying severity in humans, from dengue fever, which is usually self-limited, to dengue hemorrhagic fever and dengue shock syndrome, either of which may be life-threatening. It is estimated that as many as 390 million individuals are infected with dengue virus annually.
The phenomenon of ADE may be observed when a person who has previously been infected with one serotype of the dengue virus becomes infected months or years later with a different serotype. In such cases, the clinical course of the disease is more severe, and these people have higher viremia compared with those in whom ADE has not occurred. This explains the observation that while primary infections cause mostly minor disease in children, secondary infection is more likely to be associated with dengue hemorrhagic fever and/or dengue shock syndrome in both children and adults.
There are four antigenically different serotypes of dengue virus. In 2013 a fifth serotype was reported. Infection with dengue virus induces the production of neutralizing homotypic immunoglobulin G antibodies which provide lifelong immunity against the infecting serotype. Infection with dengue virus also produces some degree of cross-protective immunity against the other three serotypes. Neutralizing heterotypic IgG antibodies are responsible for this cross-protective immunity, which typically persists for a period of several months to a few years. These heterotypic antibody titers decrease over long time periods. While heterotypic IgG antibody titers decrease, homotypic IgG antibody titers increase over long time periods. This could be due to the preferential survival of long-lived memory B cells producing homotypic antibodies.
In addition to inducing neutralizing heterotypic antibodies, infection with the dengue virus can also induce heterotypic antibodies that neutralize the virus only partially or not at all. The production of such cross-reactive but non-neutralizing antibodies could be the reason for more severe secondary infections. It is thought that by binding to but not neutralizing the virus, these antibodies cause it to behave as a "trojan horse", where it is delivered into the wrong compartment of dendritic cells that have ingested the virus for destruction. Once inside the white blood cell, the virus replicates undetected, eventually generating very high virus titers which cause severe disease.
A study conducted by Modhiran et al. attempted to explain how non-neutralizing antibodies down-regulate the immune response in the host cell through the Toll-like receptor signaling pathway. Toll-like receptors are known to recognize extra- and intracellular viral particles and to be a major basis of the cytokines production. In vitro experiments showed that the inflammatory cytokines and type 1 interferon production were reduced when the ADE-dengue virus complex bound to the Fc receptor of THP-1 cells. This can be explained by both a decrease of Toll-like receptor production and a modification of its signaling pathway. On one hand, an unknown protein induced by the stimulated Fc receptor reduces the Toll-like receptor transcription and translation, which reduces the capacity of the cell to detect viral proteins. On the other hand, many proteins involved in the Toll-like receptor signaling pathway are down-regulated, which led to a decrease of the cytokine production. Two of them, TRIF and TRAF6, are respectively down-regulated by 2 proteins SARM and TANK up-regulated by the stimulated Fc receptors.
To illustrate the phenomenon of ADE, consider the following example: an epidemic of dengue fever occurred in Cuba, lasting from 1977 to 1979. The infecting serotype was dengue virus-1. This epidemic was followed by two more outbreaks of dengue fever—one in 1981 and one in 1997; dengue virus-2 was the infecting serotype in both of these later epidemics. 205 cases of dengue hemorrhagic fever and dengue shock syndrome occurred during the 1997 outbreak, all in people older than 15 years. All but three of these cases were demonstrated to have been previously infected by the dengue virus-1 serotype during the epidemic of 1977–1979. Furthermore, people who had been infected with dengue virus-1 during the 1977-79 outbreak and secondarily infected with dengue virus-2 in 1997 had a 3-4 fold increased probability of developing severe disease than those secondarily infected with dengue virus-2 in 1981. This scenario can be explained by the presence of neutralizing heterotypic IgG antibodies in sufficient titers in 1981, the titers of which had decreased by 1997 to the point where they no longer provided significant cross-protective immunity.

In HIV-1 virus infection

ADE of infection has also been reported in HIV. Like dengue virus, non-neutralizing level of antibodies have been found to enhance the viral infection through interactions of the complement system and receptors. The increase in infection has been reported to be over 350 fold which is comparable to ADE in other viruses like dengue virus. ADE in HIV can be complement-mediated or Fc receptor-mediated. Complements in the presence of HIV-1 positive sera have been found to enhance the infection of MT-2 T-cell line. The Fc-receptor mediated enhancement was reported when HIV infection was enhanced by sera from HIV-1 positive guinea pig enhanced the infection of peripheral blood mononuclear cells without the presence of any complements. Complement component receptors CR2, CR3 and CR4 have been found to mediate this Complement-mediated enhancement of infection. The infection of HIV-1 leads to activation of complements. Fragments of these complements can assist viruses with infection by facilitating viral interactions with host cells that express complement receptors. The deposition of complement on the virus brings the gp120 protein close to CD4 molecules on the surface of the cells, thus leading to facilitated viral entry. Viruses pre-exposed to non-neutralizing complement system have also been found to enhance infections in interdigitating dendritic cells. Opsonized viruses have not only shown enhanced entry but also favorable signaling cascades for HIV replication in interdigitating dendritic cells.
HIV-1 has also shown enhancement of infection in HT-29 cells when the viruses were pre-opsonized with complements C3 and C9 in seminal fluid. This enhanced rate of infection was almost 2 times greater than infection of HT-29 cells with virus alone. Subramanian et al., reported that almost 72% of serum samples out of 39 HIV positive individuals contained complements that were known to enhance the infection. They also suggested that the presence of neutralizing antibody or antibody-dependent cellular cytotoxicity-mediating antibodies in the serum contains infection-enhancing antibodies. The balance between the neutralizing antibodies and infection-enhancing antibodies changes as the disease progresses. During advanced stages of the disease the proportion of infection-enhancing antibodies are generally higher than neutralizing antibodies. Increase in viral protein synthesis and RNA production have been reported to occur during the complement-mediated enhancement of infection. Cells that are challenged with non-neutralizing levels of complements have been found have accelerated release of reverse transcriptase and the viral progeny. The interaction of anti-HIV antibodies with non-neutralizing complement exposed viruses also aid in binding of the virus and the erythrocytes which can lead to more efficient delivery of viruses to the immune-compromised organs.
ADE in HIV has raised questions about the risk of infections to volunteers who have taken sub-neutralizing levels of vaccine just like any other viruses that exhibit ADE. Gilbert et al., in 2005 reported that there was no ADE of infection when they used rgp120 vaccine in phase 1 and 2 trials. It has been emphasized that much research needs to be done in the field of the immune response to HIV-1, information from these studies can be used to produce a more effective vaccine.

Mechanism

There are several possibilities to explain the phenomenon:
  1. A viral surface protein studded with antibodies against a virus of one serotype binds to a similar virus with a different serotype. The binding is meant to neutralize the virus surface protein from attaching to the cell, but the virus-antibody complex also binds to the Fc-region antibody receptor on the cell membrane. This brings the virus into close proximity to the virus-specific receptor, and the cell internalizes the virus through the normal infection route.
  2. A virus surface protein may be attached to antibodies of a different serotype, activating the classical pathway of the complement system. The complement cascade system instead binds C1Q complex attached to the virus surface protein via the antibodies, which in turn bind C1q receptor found on cells, bringing the virus and the cell close enough for a specific virus receptor to bind the virus, beginning infection. This mechanism has not been shown specifically for dengue virus infection, but may occur with Ebola virus infection in vitro.
  3. When an antibody to a virus is present for a different serotype, it is unable to neutralize the virus, which is then ingested into the cell as a sub-neutralized virus particle. These viruses are phagocytosed as antigen-antibody complexes, and degraded by macrophages. Upon ingestion the antibodies no longer even sub-neutralize the body due to the denaturing condition at the step for acidification of phagosome before fusion with lysosome. The virus becomes active and begins its proliferation within the cell.