Antibodies: what makes us stronger

ABSTRACT

Neutralizing antibodies are the basis of almost all approved prophylactic vaccines and the foundation of effective protection from pathogens, including the recently emerging SARS Coronavirus 2 (SARS-CoV-2). However, the contribution of antibodies to protection and to the course of the disease during first-time exposure to a pathogen is unknown. We analyzed the antibodies and B cell responses in severe and mild COVID-19 patients. Despite our primary assumption that high antibody titers contribute to a mild disease, we found that severe COVID-19 illness, and not mild infection, correlates with strong anti-viral antibody and memory B cell responses. This phenomenon was also demonstrated for anti-Mycobacterium tuberculosis inhibiting antibodies that we recently isolated from an actively infected Tuberculosis-sick donor. This correlation between disease severity and antibody responses can be explained by the fact that high viral loads drive B cell stimulation and generation of high-affinity antibodies that will be protective upon future encounter with the particular pathogen.

KEYWORDS:

• B cells
• neutralizing antibodies
• SARS-CoV-2
• tuberculosis

Our immune response is comprised of two main arms; the first is the innate immunity arm, which functions as an immediate line of response to pathogens and its components include mainly activated lymphocytes and epithelial cells, as well as soluble factors, such as proteins of the complement pathway. The second arm of our immune system is the adaptive immunity, which is separated into cellular and humoral immunity, represented by T cells and B cells, respectively. B cell immunity is responsible for the generation of antibodies and it is an important component of effective protection from pathogens. When infection occurs, naïve B cells are stimulated, and through a unique process called “affinity maturation,” differentiate into memory B cells and antibody-secreting plasma cells.¹ In some infections, plasma cells secreting pathogen-specific antibodies are detected as early as on day 4 post infection, peaking around days 7 post infection.² The pathogen-specific antibodies can directly bind to the infectious agent or bacteria-specific toxin and sometimes, this binding event itself blocks the pathogen and subsequently prevents the infection or blocks the biological activity of the toxin. These types of antibodies are commonly termed “neutralizing antibodies,” and they have been defined as one of the most important factors in protection during infections and the basis of all approved vaccines.³ The importance of antibodies in protection from pathogenic infections is also demonstrated by the fact that B cell immunodeficient individuals who are unable to mount IgG responses have recurrent and increased susceptibility to infections.⁴

While the role of preexisting pathogen-specific antibodies in protection from infections has been set in stone, the role of pathogen-specific antibodies in protection during first-time encounter with the corresponding pathogen, is not yet clear. During first-time exposure to a pathogen, the B cell response is maturing in parallel with the developing infection. Therefore, one might ask what is the correlation between asymptomatic first-time infection and the production of highly effective and robust neutralizing antibodies, and is failure to produce such an effective antibody response during the first-time encounter with the pathogen result in greater morbidity? The emergence of the COVID-19 pandemic provided a unique opportunity to systematically address this question by studying the relationships between a new virus previously unknown to the immune system, a new disease, and the antibodies co-evolving against it.

Correlation between COVID-19 disease severity and serum IgG

Infection with the recently emerging SARS Coronavirus 2 (SARS-CoV-2) results in severe disease manifestations in approximately 10% of the virologically confirmed infections, leading to death in 1–4% of all confirmed infection.^5,6 Nonetheless, the vast majority (according to most reports between 60% and 75% of the infectees) demonstrate a mild disease that clears on its own, whereas unundefined portion (according to some reports between 20% and 40% of the infections^6,7) is asymptomatic. Given the wide range of possible clinical outcomes, and the fact that antibody responses against SARS-CoV-2 have been reported very early in the course of the pandemic,⁸ we hypothesized that mild asymptomatic infection correlates with high levels of antibodies that hold SARS-CoV-2 back, thus preventing viral spread and subsequent tissue damage. With this assumption in mind, our group initiated a search for highly potent anti-SARS-CoV-2 neutralizing antibodies in mild SARS-CoV-2 infectees that will serve as candidates for therapeutics, and provide insights for vaccine targets. However, analysis of serological and B cell responses in convalescent mildly and severely sick patients by our group,⁹ as well as by others^10–12 revealed a reversed picture, thus refuting our primary postulation. Our in-depth analyses of B cells showed that in the case of COVID-19, it is a severe disease that is associated with both higher titers and better antibody responses, which are directed against numerous nonoverlapping epitopes on the virus.⁹ In contrast to our initial assumption, mild infection resulted in significantly lower titers of inhibiting antibodies (p < .0001),⁹ as assayed by the ability of the serum to inhibit SARS-CoV-2 Spike receptor-binding domain (RBD) and ACE2 interaction. RBD is a major neutralizing component on the SARS-CoV-2 and the majority of neutralizing anti-SARS-CoV-2 antibodies are directed against this component.¹³ We also found greater B cell clonal expansion and higher frequencies of anti-SARS-CoV-2 Spike receptor-binding domain (RBD) memory B cells, in severe convalescent SARS-CoV-2 donors compared to asymptomatic SARS-CoV-2-infected individuals.⁹ The severe donors we analyzed developed multi-clonal neutralizing anti-SARS-CoV-2 antibodies directed against different sites on the SARS-CoV-2 Spike RBD.⁹ When combined, the neutralizing antibodies from one of the severe donors we analyzed, CoV02, were capable of suppressing the virus for several days in culture, more effectively than the individual effect of each single antibody, suggesting that the antibody response in this donor was not only neutralizing but also synergistic.⁹

In fact, the correlation between severe disease caused by a high pathogenic burden and effective antibody neutralization is not limited to COVID-19. In HIV-1 infection, anti-HIV-1 broadly neutralizing antibodies were isolated from individuals with high viral loads.¹⁴ Several studies showed that viral stimulation is needed to drive the evolution of neutralizing responses, occurring in ‘waves’ that follow the preceding ‘waves’ of viremia in donors with chronic HIV-1 infection.^15,16 A unique subset of HIV-1-infected individuals, elite HIV-1 controllers, who are able to maintain low levels of viremia without the use of antiviral drugs, are the source of several of the most potent anti-HIV-1 antibodies.^17–19 However, viral control in these individuals was found to be largely due to specific alleles of the human leukocyte antigen (HLA) types, resulting in highly effective CD8 + T cells response.²⁰ In these cases, HIV-1 is kept under control mainly by the T cell arm of the immune system; however, the fact that the virus is still chronically present (even if viremia is low) is sufficient to stimulate and drive the B cell arm. This condition enables the generation of extremely broad cross-clade neutralizing antibodies that sometimes coexist with the viral variants.²¹

Inhibiting antibody responses in Mycobacterium tuberculosis actively infected donors

While it is becoming more and more appreciated, the role of antibodies in some bacterial infections, such as Mycobacterium tuberculosis (Mtb) is still controversial.^22,23 Health-care workers repeatedly exposed to sick patients exhibited polyclonal serum antibodies that when passively transferred to mice, protected them from the Mtb challenge,²⁴ indicating that anti-bacterial inhibiting antibodies are produced upon exposure. Antibodies against Mtb lipopolysaccharide residues were shown to provide a certain degree of protection,²⁵ however the main protein-based targets for neutralizing antibodies remain to be largely unknown. This presents one of the major challenges in the field of anti-bacterial antibodies – the vast complexity of the bacteria and the high abundance of proteins compared to viruses that usually encode only a small number of structural proteins. We recently used a recombinant phosphate transporter protein (PstS1) to isolate protective anti-Mtb monoclonal antibodies from a symptomatic, actively infected patient who was hospitalized due to severe tuberculosis disease and treated until recovery.²⁶ The two antibodies, p4-36 and p4-163, showed 50% protection both in vitro and in a mouse model, when administered to BALB/c mice prior to infection.²⁶ Interestingly, affinity maturation was crucial for the activity of both of these antibodies as reversion of their somatic hypermutations canceled their activity.²⁶ This emphasizes the fact that while the donor himself was sick, B cell responses and affinity maturation were ongoing, resulting in the production of high-affinity antibodies. Although such antibodies did not reduce Tuberculosis disease loads in the donor during the time of their generation, they might be protective during future bacterial encounters. Monoclonal antibodies, such as p4-36 and p4-163 can be evaluated as potential biological therapeutics to serve as an adjunctive treatment of antibiotic-resistant strains, as well as provide important leads for vaccine design.

The correlation between the severity of the disease during first-time exposure and the quality of the produced antibodies fits Friedrich Nietzsche’s aphorism, “what does not kill me makes me stronger” – in this case, meaning that the high pathogenic loads ultimately push the B cell responses forward to generate an effective germinal center reaction, and subsequently produce an effective and diverse polyclonal response. In light of this, in the pursuit for the next effective neutralizing antibody, not only individuals who had mild disease should be considered, but one should also consider looking into the antibodies developed by donors who recovered from a severe illness, whose antibodies might be very effective.

Vaccination makes perfect

While the lack of symptoms following first-timeinfection may not be antibody-mediated,B cells and antibodies are the main mediators of protectionduring second-timeinfections. The concept of vaccination is based on the notion that the B cell community needs to be trained on the most relevant targets while depositing long-lastingpolyclonal and effective B cell memory. For this reason, vaccination can be even more effective than natural infection by providing an improved and targeted immune response compared to immune response to infection. Accordingly, it has been shown that while the B cell response following seasonal influenza vaccination is highly effective in neutralizing the virus, a significant fraction of the antibodies produced following influenza infection are based on recall of less effective memory B cells.²⁷ Similarly, in the case of COVID-19, anti-SARS-CoV-2 antibody responses following vaccination were shown to be more robust and effective compared to those produced in response to an infection.²⁸ Similar phenomena, i.e., that antibody response to a vaccine is advantageous to the response to natural infection, were reported for HPV.²⁹ In conclusion, the revolution of vaccines enables us to train B cells by direct and sufficient antigenic stimulation, without pathogen-induced disease, and circumventing the need to “survive” the first encounter with the virus or bacteria, while still gaining effective protection.

Disclosure of potential conflicts of interest

No potential conflicts of interest were disclosed.

Acknowledgments

The author thanks Jonathan M. Gershoni for helpful discussions and Anna Roitburd-Berman for help with scientific editing. The author is supported by The Campbell Foundation for HIV-1/AIDS Research and by ISF grants #1422/18 and #3711/20.