Pathogens, Vol. 11, Pages 1437: SARS-CoV-2 Antibody Effectiveness Is Influenced by Non-Epitope Mutation/Binding-Induced Denaturation of the Epitope 3D Architecture

1. IntroductionProtein structures are determined by amino acid sequences and are considered denatured when their biological, chemical, and physical properties are altered due to distortions in their 3D native structures [1]. Distortion of protein structures can be caused by several factors including chemical denaturants, pH, force, pressure, temperature, and mutation [2]. Mutations in severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) have produced Alpha, Beta, Gamma, Delta, and Omicron, also known as variants of concern (VOC) since they exhibit increased capabilities for immune evasion, transmissibility, and are highly pathogenic [3]. Omicron is the predominant VOC with several subvariants including the newly reported BA.4 and BA.5 poised to supplant BA.1 and BA.2 [4]. The spike protein present on the surface of SARS-CoV-2 comprises the N-terminal S1, consisting of the N-terminal domain (NTD) and a receptor-binding domain (RBD), and the C-terminal S2 domain. While the S2 domain is responsible for viral fusion leading to host cell infection, the S1 domain through the RBD attaches the SARS-CoV-2 virus to angiotensin-converting enzyme 2 (ACE2) receptors on host cells to initiate infection [5]. Hence, COVID-19 vaccines and therapeutic antibodies are designed to target the spike protein, precisely the RBD [5,6]. Donor/acceptor atoms on original RBD and paratope/ACE2 amino-acid sidechains (AASCs) facilitate the formation of hydrogen bonds [7]. Hydrogen bonds are essential for both stabilizing the RBD-paratope/ACE2 complex and enabling antibodies/ACE2 to function effectively [8]. Replacing native amino acids with alien amino acids could result in the disruption of hydrogen bonds and abrogation of the RBD-paratope/ACE2 interaction consequently. Epitope amino acid substitutions in the RBD promote immune evasion and viral transmission [9,10]. We hypothesized that viral escape can also occur when epitope AASCs are spatially repositioned since this may disrupt hydrogen bonds. Epitope AASC repositioning can be facilitated by non-epitope mutations. However, information on the effect of non-epitope mutation-induced RBD AASC repositioning on RBD-antibody and RBD-ACE2 function remains elusive. This will require deciphering the structural basis for the increased immune evasion and strengthened ACE2 engagement by constantly emerging SARS-CoV-2 VOC. Allosteric modulation of epitope 3D structures driven by non-epitope RBD amino acid substitutions can influence antibody function [11]. However, the mechanism remains uncharacterized.

We examined several Protein Data Bank (PDB) deposited 3D structures of the original, Beta, Delta, Kappa (a variant of interest), BA1, and BA2 RBD proteins in complex with either antibodies or ACE2.

4. DiscussionProtein binding and neutralization profiles are highly dependent on their native 3D structures which are determined by amino acid sequences. While it is known that epitope amino acid substitutions influence RBD-antibody/ACE2 interactions, the role of epitope AASC repositioning induced by non-epitope mutations in promoting immune escape and ACE2 binding simultaneously remains unexplored. We show that non-epitope mutations, including truncations, and induced-fit can reposition RBD AASCs leading to the weakening, complete disruption, or establishment of new hydrogen bonds with paratope/ACE2 AASCs. Discordant epitope-paratope/ACE2 interactions facilitated by structural variations might explain inconsistencies in the effectiveness of the same antibodies/ACE2 against SARS-CoV-2 VOC and when antibodies are used in cocktails relative to individual use [10,11,14,16,25,26]. Additionally, truncation-induced structural differences might also explain discordant effectiveness reported for Fabs relative to their parent immunoglobulins [18,19,20,27] which traditionally has been explained by the difference in the number of antigen-binding sites (antibody valency) [28,29]. However, this unproven hypothesis does not explain why the same Fab binds and neutralizes some but not other subvariants of the same virus [18,19,20,27]. Another study explored the role of antibody valency in virus neutralization by comparing F(ab’)2 and Fab variants to their parent immunoglobulin. They found that while the F(ab’)2 exhibited reduced activity, the Fab completely lost activity. These researchers repeated the same experiment but this time using a different antibody and found that the Fab did not lose its binding and neutralization activity relative to the parent immunoglobulin. Therefore, they fairly concluded that antibody valency does not explain the loss of Fab activity [30]. Furthermore, if antibody valency determined antibody binding and neutralization characteristics, then decavalent immunoglobulin m would be more effective than tetravalent immunoglobulin-A which should in turn be more effective than bivalent immunoglobulin g possessing identical variable regions. However, this is not always the case given that findings are mixed [31,32], just like with Fabs and parent immunoglobulins. Based on our structural findings, we propose that whether truncated proteins/antibodies will conserve their binding and neutralization profiles will largely depend on variable region amino acid sequences, the potential for variable region AASCs to undergo induced-fit required to optimize binding, circumvent clashes with epitope AASCs or glycans [27], how severe epitope/paratope AASCs are repositioned and whether these denatured AASCs will result in clashes or in the disruption of critical paratope-epitope interactions necessary for antibodies to properly engage their respective antigens. Interestingly, b12 Fab and its parent immunoglobulin g inhibited HIV using different mechanisms of action [19] indicating that epitope AASC repositioning might not only influence antibody potency but the mechanism of inhibition potentially. Vaccines are designed to induce antibodies with spatially positioned acceptor/donor atoms on paratope AASCs to specifically interact with corresponding donor/acceptor atoms on RBD AASCs. Since vaccine-induced antibodies have fixed structures, epitope amino acid substitutions or mutation-induced AASC repositioning in the RBD of VOC could influence the effectiveness of vaccine-elicited antibodies. The spike protein from the original SARS-CoV-2 is still used in current COVID-19 vaccines despite the emergence of VOC with different amino acid sequences and RBD 3D structures. Changes in RDB native 3D structures might explain the consistent reduction in COVID-19 vaccine-induced antibody effectiveness against continuously emerging VOC [17], underscoring the urgent need for designing and developing multi-variant COVID-19 vaccines. Variant-specific COVID-19 vaccines might not be ideal given the high SARS-CoV-2 mutation rate. The FDA recently authorized the use of updated bivalent COVID-19 vaccines containing spike proteins from the original and BA.4/5 SARS-CoV-2 variants [33], consistent with our reporting and proposals. Waning immunity is believed to cause breakthrough infections while booster COVID-19 vaccine doses have been proposed as one way of increasing the level of waning neutralizing antibodies [34,35,36]. However, the fact that the same serum from COVID-19 vaccinated/infected individuals can be highly potent against the original or structurally similar SARS-CoV-2 D614G variant but less potent against predominant Omicron subvariants [11,37] challenges the “waning-immunity” concept. Vaccine/infection-induced antibodies are present in the serum but many of them have been rendered obsolete by structural alterations in RBD epitopes of Omicron subvariants. Therefore, booster doses from using first-generation COVID-19 vaccine formulations will likely increase the quantity and not the effectiveness of elicited antibodies since most antibodies will most likely have donor/acceptor atoms on paratope amino acids for which complementary acceptor/donor atoms on RBD amino acids are inexistent or no longer accessible because they are structurally repositioned in SARS-CoV-2 VOC including Omicron BA.2, BA.4, and BA.5. It is true that waning immunity or immune status modifies vaccine effectiveness, but they do not play a major role in driving SARS-CoV-2 breakthrough infections. Mutation-induced structural alterations in SARS-CoV-2 VOC are responsible for escaping vaccine/infection-induced immunity instead [34,38,39,40]. To prevent and control the rapid emergence, acquisition, and spread of SARS-CoV-2 VOC including the newly identified BA.4/5 subvariants, the focus must be to improve the quality and not increase the quantity of mismatched COVID-19 vaccine-induced antibodies. As of August 31, 2022, the FDA no longer recommends the use of monovalent (first-generation) but bivalent (containing both first-generation and BA.4/BA.5) COVID-19 vaccines as boosters [33], congruent with our observations and suggestions.Furthermore, AASC repositioning has challenged the notion that amino acid conservation directly translates into maintained binding, and by extension neutralization activity. We have shown that spatially conserved donor/acceptor atoms directly translated into conserved binding and neutralization activity irrespective of the presenting amino acid. Hence, antibodies must be engineered to target conserved RBD donor/acceptor atoms to protect against current and future SARS-CoV-2 VOC. This should be extended to designing universal COVID-19 vaccines. We also showed that AASC repositioning must occur before immune-pressured epitope amino acids are eventually substituted and/or new contacts with ACE2 are established. Identifying immune-pressured AASCs and monitoring trends of donor/acceptor atoms on RBD and ACE2 AASCs will enable modeling of their immune-evasion and transmission capabilities which is essential for pandemic preparedness. Prospective mapping of escape mutations has been explored [41] concordant with our findings/proposals. AASC repositioning driven by induced-fit requires that various antibody combinations are assessed to come up with the most effective therapeutic antibody cocktails [22,23]. Our findings can be extended to the design of antibody-based diagnostic/detection kits. Upon binding, the capture antibody might induce structural changes in the protein which may influence optimal binding by the detection antibody. Our study has some limitations. We did not analyze all interactions key for binding such as salt bridges. However, our findings using hydrogen bonds can be extrapolated to salt bridges and potentially other interactions influenced by distances between complementary donor/acceptor atoms on paratope and RBD epitope AASCs. Interactions modeled by structural alignment should be interpreted with caution since binding can involve induced-fit. Nonetheless, findings obtained by structural alignment were corroborated by complete 3D structures. We did not include 3D structures for all Omicron subvariants as these were not deposited in the Protein Data Bank at the time of mining. A study published recently explored BA.2.12.1, BA.4, and BA.5 Omicron subvariants and demonstrated that these subvariants exhibited inconsistent neutralization profiles consistent with our findings [11]. Some comparisons were performed using structures that were resolved from different methods and, therefore, it is possible that differences observed using these structures might be artifacts. Nonetheless, our findings are supported by functional studies from the literature.

留言 (0)

沒有登入
gif