Promoting Fc-Fc interactions between anti-capsular antibodies provides strong immune protection against Streptococcus pneumoniae
Curation statements for this article:-
Curated by eLife
eLife assessment
This paper will be of interest to immunologists and infectious disease experts, as it reports the investigation of a novel treatment of invasive pneumococcal diseases using complement-activating monoclonal antibodies. Using a combination of in vitro and in vivo methods, the authors demonstrate convincingly that the introduction of specific mutations in human monoclonal antibodies that target the surface of pneumococcus bacteria can result in enhanced complement activation after these antibodies bind to the bacterial surface.
This article has been Reviewed by the following groups
Listed in
- Evaluated articles (eLife)
- Microbiology and Infectious Disease (eLife)
Abstract
Streptococcus pneumoniae is the leading cause of community-acquired pneumonia and an important cause of childhood mortality. Despite the introduction of successful vaccines, the global spread of both non-vaccine serotypes and antibiotic-resistant strains reinforces the development of alternative therapies against this pathogen. One possible route is the development of monoclonal antibodies (mAbs) that induce killing of bacteria via the immune system. Here, we investigate whether mAbs can be used to induce killing of pneumococcal serotypes for which the current vaccines show unsuccessful protection. Our study demonstrates that when human mAbs against pneumococcal capsule polysaccharides (CPS) have a poor capacity to induce complement activation, a critical process for immune protection against pneumococci, their activity can be strongly improved by hexamerization-enhancing mutations. Our data indicate that anti-capsular antibodies may have a low capacity to form higher-order oligomers (IgG hexamers) that are needed to recruit complement component C1. Indeed, specific point mutations in the IgG-Fc domain that strengthen hexamerization strongly enhance C1 recruitment and downstream complement activation on encapsulated pneumococci. Specifically, hexamerization-enhancing mutations E430G or E345K in CPS6-IgG strongly potentiate complement activation on S. pneumoniae strains that express capsular serotype 6 (CPS6), and the highly invasive serotype 19A strain. Furthermore, these mutations improve complement activation via mAbs recognizing CPS3 and CPS8 strains. Importantly, hexamer-enhancing mutations enable mAbs to induce strong opsonophagocytic killing by human neutrophils. Finally, passive immunization with CPS6-IgG1-E345K protected mice from developing severe pneumonia. Altogether, this work provides an important proof of concept for future optimization of antibody therapies against encapsulated bacteria.
Article activity feed
-
-
Author Response
Reviewer #1 (Public Review):
I noticed 2 weaknesses, the first related to the killing assays: considering that WT IgG less efficiently promotes complement-mediated phagocytosis of bacteria, one would assume that the ingested bacteria (to be killed) would be lower in neutrophils exposed to this IgG, to begin with - which is not accounted for in the analyses shown.
We now included a better explanation of our opsonophagocytic killing assay.
A second weakness in my mind pertains to the in vivo experiment: the model used obviously requires a very high number of bacteria (the inoculum), somehow indicating that this specific bacterial strain does not lead to progressive infection (i.e. with replicating bacteria) but mice experience a severe acute inflammatory response followed by the rapid elimination of bacteria. This …
Author Response
Reviewer #1 (Public Review):
I noticed 2 weaknesses, the first related to the killing assays: considering that WT IgG less efficiently promotes complement-mediated phagocytosis of bacteria, one would assume that the ingested bacteria (to be killed) would be lower in neutrophils exposed to this IgG, to begin with - which is not accounted for in the analyses shown.
We now included a better explanation of our opsonophagocytic killing assay.
A second weakness in my mind pertains to the in vivo experiment: the model used obviously requires a very high number of bacteria (the inoculum), somehow indicating that this specific bacterial strain does not lead to progressive infection (i.e. with replicating bacteria) but mice experience a severe acute inflammatory response followed by the rapid elimination of bacteria. This explains the high mortality - and indicates that mice succumb to acute inflammation, rather than the progressive replication of bacteria. To conclusively prove the therapeutic value of those modified antibodies, a clinically more relevant S. pneumoniae model would be helpful.
The inoculum used in our mouse model was based on a dose finding study. Although the initial starting dose was 5x106 bacteria (based on previously published mouse infection models with S. pneumoniae serotype 6A), we needed a higher dose (1x108 bacteria) to reach 80-100% mortality. While we agree that the final dose was relatively high, this does not mean that capsule type 6 is not a clinically relevant strain. It is well known that clinically relevant serotypes in humans are not always invasive in mice (doi: 10.1128/iai.60.1.111-116.1992). This is the exact reason why we chose to perform in vivo experiments with serotype 6A, which is known to be more invasive in mice (while serotype 6B is more virulent in humans). Of course, while our in vivo data provide an important proof-of-concept for the capacity of hexamer-enhancing mutations to improve protection by anti-capsular antibodies, future studies are needed to verify the potential use of mAbs against other serotypes.
A third aspect, which should be addressed in the discussion, unless tested and not shown, is how anti-pneumococcal IgM antibodies compare to hexamerized IgGs. Is there any advantage, or do they perform similarly with regards to complement activation?
We have now generated and tested IgM against CPS6 (Figure 2g). Although anti-CPS6 IgM can induce complement-dependent phagocytosis to some extent, but IgM was less potent than IgG variants with hexamer-enhancing mutations. This suggests that complement activation via pre-assembled IgM oligomers was less effective than via IgG hexamers that are formed after target binding.
These new data are now included in the revised manuscript as figure 2g, supplemental figure 9 and commented in results section lines 172-179.
Reviewer #2 (Public Review):
The results are intriguing, and one consideration is whether enhancing complement activation is beneficial or harmful for a therapeutic antibody. Based on these results is there the possibility of a natural selection against strong levels of complement activation?
We appreciate the positive feedback to our presented work. Indeed, it is believed there is a natural selection against these mutations to avoid uncontrolled complement activation by naturally occurring IgGs in solution. It is important to realize that formation of IgG hexamers is a surface-dependent process. When IgG molecules bind to surface-bound antigens (via Fab), they can organize into higher-ordered hexamers via Fc-Fc interactions. The specific point mutations used in this paper increase hexamer formation after antigen binding on the cell surface. However, at high concentrations of IgG (as those occurring in our blood (>10 mg/ml), IgG hexamers might be formed independent of target binding (van Kampen et al Journal of Pharmaceutical Sciences Volume 111, Issue 6, June 2022, Pages 1587-1598). If naturally occurring IgGs would have hexamer-enhancing mutations, IgG hexamers could be formed in solution resulting in massive complement activation and depletion of the complement system.
The study clearly shows that the introduction of the hexamerisation mutations affects the ability of the antibodies to bind and activate complement. The studies in Fig 2 examining the role of Fc are particularly elegant. One issue is that it is surprising that the WT IgG1 and IgG3 monoclonals have a minimal capacity to fix and activate complement, despite IgG1/3 to other antigens being efficient isotypes at fixing complement. In the absence of data showing whether IgG1/3 from normal human sera against capsule fixes complement then it is difficult to contextualise these results or to assess if other changes, such as in glycosylation, contribute to the results presented. Related to this, there is reasonable evidence that antibodies induced to capsules can be protective yet the data in Fig 5 suggests that without the mutations then the monoclonals are not effective at all for 6B and only effective at the highest concentration for 19A.
As mentioned in Essential revision 3 our data with S. aureus antibodies demonstrate that this is not a consequence of how these mAbs are produced or differences in their Fc glycosylation profile. We agree with the fact that there are reasonable evidence that antibodies induced to capsules can be protective. However, not all vaccine serotypes are able to induce a strong immune protection. Serotype 6B, for instance, which is covered by current vaccines, is found to be poorly immunogenic (manuscript lines 101-103). For further studies, it would be really interesting to find out what makes this difference between mAbs and, specifically in our case between anti-CPS antibodies.
The adoptive transfer experiments demonstrate that the antibodies can moderate bacteraemia. The mechanism of this is not explored and the importance of hexamerisation and complement activation not demonstrated, especially as it is not clear if human antibodies and mouse complement are a productive combination in this context.
We have now included additional phagocytosis assays with mouse sera (supplemental figure 15) that demonstrate that human antibodies and mouse complement are a productive combination.
-
eLife assessment
This paper will be of interest to immunologists and infectious disease experts, as it reports the investigation of a novel treatment of invasive pneumococcal diseases using complement-activating monoclonal antibodies. Using a combination of in vitro and in vivo methods, the authors demonstrate convincingly that the introduction of specific mutations in human monoclonal antibodies that target the surface of pneumococcus bacteria can result in enhanced complement activation after these antibodies bind to the bacterial surface.
-
Reviewer #1 (Public Review):
Aguinagalde et al. investigated alternative treatment options for invasive pneumococcal diseases and considered the use of monoclonal antibodies to promote the killing of bacteria. For this to happen, complement deposition and activation are required to occur on the bacterial surface. The authors discovered that hexamerization of IgG strongly augments the recruitment of C1, and specific mutation of the antibodies can support this process and enable efficient phagocytosis and killing of bacteria by neutrophils. Further, sets of in vivo studies support the idea that passive immunization with these modified antibodies improves survival from pneumococcal pneumonia in mice.
Considering vaccine-escape serotypes, the (sometimes) suboptimal vaccine response together with the increased occurrence of strains resistant …
Reviewer #1 (Public Review):
Aguinagalde et al. investigated alternative treatment options for invasive pneumococcal diseases and considered the use of monoclonal antibodies to promote the killing of bacteria. For this to happen, complement deposition and activation are required to occur on the bacterial surface. The authors discovered that hexamerization of IgG strongly augments the recruitment of C1, and specific mutation of the antibodies can support this process and enable efficient phagocytosis and killing of bacteria by neutrophils. Further, sets of in vivo studies support the idea that passive immunization with these modified antibodies improves survival from pneumococcal pneumonia in mice.
Considering vaccine-escape serotypes, the (sometimes) suboptimal vaccine response together with the increased occurrence of strains resistant to antibiotics, the search for alternative treatments is highly warranted, and this study is an excellent example supporting the use of therapeutic antibodies targeting the capsule.
This study is generally very well performed and very well written. The authors conclusively show the importance of mAb hexamerization to augment complement deposition and activation on the surface of pneumococci, which promotes subsequent phagocytosis by neutrophils. Further experiments prove the importance of hexamerization of mAbs and the importance of this complement in the uptake of bacteria by neutrophils. Subsequent in vivo studies showed the therapeutic usefulness of modified antibodies in preventing mortality and bacteremia in female mice. All these data provide strong evidence for the claims of the authors.
I noticed 2 weaknesses, the first related to the killing assays: considering that WT IgG less efficiently promotes complement-mediated phagocytosis of bacteria, one would assume that the ingested bacteria (to be killed) would be lower in neutrophils exposed to this IgG, to begin with - which is not accounted for in the analyses shown.
A second weakness in my mind pertains to the in vivo experiment: the model used obviously requires a very high number of bacteria (the inoculum), somehow indicating that this specific bacterial strain does not lead to progressive infection (i.e. with replicating bacteria) but mice experience a severe acute inflammatory response followed by the rapid elimination of bacteria. This explains the high mortality - and indicates that mice succumb to acute inflammation, rather than the progressive replication of bacteria. To conclusively prove the therapeutic value of those modified antibodies, a clinically more relevant S. pneumoniae model would be helpful.
A third aspect, which should be addressed in the discussion, unless tested and not shown, is how anti-pneumococcal IgM antibodies compare to hexamerized IgGs. Is there any advantage, or do they perform similarly with regards to complement activation?
-
Reviewer #2 (Public Review):
In this clearly presented study, the authors are assessing the impact of introducing hexamerisation-associated mutations into human monoclonal antibodies that target the capsule of pneumococcus. The impact of these mutations is assessed in in vitro systems using human sera or neutrophils. The second series of studies use mouse models involving the adoptive transfer of antibody and the subsequent challenge of mice.
The major strengths of the study are that the authors are addressing an important area of unmet need, both in terms of alternative treatments for bacterial infections and also in how antibodies function against bacterial pathogens. This is a neglected area, particularly in the context of understanding how antibodies function after binding to bacterial capsules. The results are intriguing, and one …
Reviewer #2 (Public Review):
In this clearly presented study, the authors are assessing the impact of introducing hexamerisation-associated mutations into human monoclonal antibodies that target the capsule of pneumococcus. The impact of these mutations is assessed in in vitro systems using human sera or neutrophils. The second series of studies use mouse models involving the adoptive transfer of antibody and the subsequent challenge of mice.
The major strengths of the study are that the authors are addressing an important area of unmet need, both in terms of alternative treatments for bacterial infections and also in how antibodies function against bacterial pathogens. This is a neglected area, particularly in the context of understanding how antibodies function after binding to bacterial capsules. The results are intriguing, and one consideration is whether enhancing complement activation is beneficial or harmful for a therapeutic antibody. Based on these results is there the possibility of a natural selection against strong levels of complement activation?
The study clearly shows that the introduction of the hexamerisation mutations affects the ability of the antibodies to bind and activate complement. The studies in Fig 2 examining the role of Fc are particularly elegant. One issue is that it is surprising that the WT IgG1 and IgG3 monoclonals have a minimal capacity to fix and activate complement, despite IgG1/3 to other antigens being efficient isotypes at fixing complement. In the absence of data showing whether IgG1/3 from normal human sera against capsule fixes complement then it is difficult to contextualise these results or to assess if other changes, such as in glycosylation, contribute to the results presented. Related to this, there is reasonable evidence that antibodies induced to capsules can be protective yet the data in Fig 5 suggests that without the mutations then the monoclonals are not effective at all for 6B and only effective at the highest concentration for 19A.
The adoptive transfer experiments demonstrate that the antibodies can moderate bacteraemia. The mechanism of this is not explored and the importance of hexamerisation and complement activation not demonstrated, especially as it is not clear if human antibodies and mouse complement are a productive combination in this context.
-
-