Periodic Reporting for period 2 - SNUGly (Systems-level novel understanding of anti-glycan immunity)
Reporting period: 2022-01-01 to 2023-06-30
Intestinal bacteria have an enormous influence on health, both in the form of major pathogens and as major constituents of the microbiota. We have demonstrated that high-affinity intestinal antibodies offer huge potential to protect from infection and to manipulate microbiota composition. However, designing good oral vaccines to induce these antibody responses is a compound problem. Most protective intestinal antibodies target bacterial glycans: long, repetitive sugar polymers that coat the surface of all bacterial cells. As these are typically amorphous hydrophilic structures the way that these interact with antibodies is fundamentally different to a rigid protein or small molecule. It is also currently unclear how these glycans or glycolipids are presented to B cells during germinal center responses. Starting from the well-established concept that whole-cell inactivated oral vaccines can induce high-affinity intestinal antibody responses against some glycan structures, we will dissect the critical determinants of antigen sampling, presentation and affinity maturation that are necessary for vaccine success. By combining advanced biophysical methods (e.g. atomic force spectroscopy), fluorescence and electron microscopy, and synthetic biology, we aim to:
1) Generate a "toolbox" for glycan-binding antibody research: Recombinant antibodies, BCR knock-in mice and defined glycan antigens as purified molecules, on whole bacteria or on virus-like particle will be developed for model antigens: Salmonella Typhimurium O-antigen and the E.coli K100 capsule.
2) Determine the quantitative relationship between glycan antigen sampling into gut-associated lymphoid tissues and particle size, glycan flexibility/structure, digestion resistance and natural IgA binding.
3) Determine how biophysical properties of glycan antigens affect B cell antigen uptake, T cell help and antibody affinity maturation.
4) Combine models of antigen sampling efficiency and anti-glycan antibody affinity maturation to generate a systems-level model of mucosal vaccine efficacy.
This will uncover the fundamental principles governing the induction of high-affinity anti-glycan sIgA, driving urgently required progress in mucosal vaccine design.
1) Generate a "toolbox" for glycan-binding antibody research: Recombinant antibodies, BCR knock-in mice and defined glycan antigens as purified molecules, on whole bacteria or on virus-like particle will be developed for model antigens: Salmonella Typhimurium O-antigen and the E.coli K100 capsule.
2) Determine the quantitative relationship between glycan antigen sampling into gut-associated lymphoid tissues and particle size, glycan flexibility/structure, digestion resistance and natural IgA binding.
3) Determine how biophysical properties of glycan antigens affect B cell antigen uptake, T cell help and antibody affinity maturation.
4) Combine models of antigen sampling efficiency and anti-glycan antibody affinity maturation to generate a systems-level model of mucosal vaccine efficacy.
This will uncover the fundamental principles governing the induction of high-affinity anti-glycan sIgA, driving urgently required progress in mucosal vaccine design.
During the first half of this project we have established a wide range of tools and have begun to probe the biophysics of antibody-glycan binding in detail. We have produced two recombinant antibodies with specificity for the Salmonella Typhimurium O-antigen backbone and could demonstrate using classical techniques (surface plasmon resonance) that these two antibodies have very different binding properties when expressed as IgG isotypes (µM versus nM) and sensitivies to alterations in glycan chain length. These recombinant antibodies have been designed and produced, together with our main external collaborator Prof. Beth Stadtmueller, Illinois, USA, as murine IgG2, dimeric IgA, secretory IgA and FAb fragements. Salmonella O-glycans with different lengths (between 1 and 200 repeats) and chemistries (O-acetylated, glucosylated). Combined with a newly developed bifunctional crosslinker, we can now generate O-antigen glycoconjugates with different properties, giving us a third class of vaccine structures, on top of whole-cell inactivated vaccines and outer membrane-vesicle-based vaccines to compare in oral vaccination studies.
In order to generate quantitative data on the relationship between glycan structure/presentation and immunogenicity we need to quantify the properties of our vaccines. Techniques to achieve this have been developed and include:
1) A novel UPLC-based method based on fluorescent labeling of O-antigen core sugars has been developed that allows quantification of labeled O-antigen abundance at different lengths.
2) Impedance flow cytometry that quantifies total biomass and particle density in vaccine preps.
3) Biophysical analysis, including dynamic light scattering and AFM measurements to determine parameters such a stiffness/elastic modulus and shell-thickness of our vaccine structures.
4) Analysis of resistance to digestion, which is a critical parameter for successful oral vaccination, based on incubation in mouse and/or pig stomach/small intestine/fecal content followed by quantification of antigen abundance and structurel
Oral vaccinations with well-defined vaccine preps are currently in progress and are expected to yield quantitative data within the next months.
In order to investigate the biophysical properties of our antibodies, we have modified atomic force microscope cantilevers with a Spy-catcher-FAb fragment fusion protein derived from either a strong-binding or weak-binding Salmonella O-antigen-specific antibody. These cantilevers were used to measure force-distance relationships of binding/unbinding to the surface of Salmonella with different surface properties. The same antibodies were studied in IgG form for binding to purified O-antigen in classical surface-plasmon resonance (SPR) studies. Interestingly, monoclonal antibody STA5 showed 1000-fold lower affinities of binding to purified O-antigen as a (dimeric-binding) IgG, than the clone STA121, but both antibodies showed similar force-distance curves in atomic force microscopy analysis of FAb-fragment (monomeric) binding. Combined with structural data currently under analysis and our observations of the specificity of STA5 and STA121 for single-repeat and polymerized glycans, the most plausible explanation for this is that the high off-rate of STA121 is SPR is driven by avidity. We are currently extending our observations to test this hypothesis directly. As STA5 would also usually encounter highly clustered glycan molecules on the bacterial surface, it is likely that avidity, rather than affinity is also a major driver of bacteria-antibody interactions. The consequences of multi-valent interactions for the strength of crosslinking between bacteria of dimeric IgA will be tested and modeled in the next months. These observations are also soon to be extended to bacterial capsular polysaccharides, starting with the MenA capsule from Neisseria meningitidis.
In order to generate quantitative data on the relationship between glycan structure/presentation and immunogenicity we need to quantify the properties of our vaccines. Techniques to achieve this have been developed and include:
1) A novel UPLC-based method based on fluorescent labeling of O-antigen core sugars has been developed that allows quantification of labeled O-antigen abundance at different lengths.
2) Impedance flow cytometry that quantifies total biomass and particle density in vaccine preps.
3) Biophysical analysis, including dynamic light scattering and AFM measurements to determine parameters such a stiffness/elastic modulus and shell-thickness of our vaccine structures.
4) Analysis of resistance to digestion, which is a critical parameter for successful oral vaccination, based on incubation in mouse and/or pig stomach/small intestine/fecal content followed by quantification of antigen abundance and structurel
Oral vaccinations with well-defined vaccine preps are currently in progress and are expected to yield quantitative data within the next months.
In order to investigate the biophysical properties of our antibodies, we have modified atomic force microscope cantilevers with a Spy-catcher-FAb fragment fusion protein derived from either a strong-binding or weak-binding Salmonella O-antigen-specific antibody. These cantilevers were used to measure force-distance relationships of binding/unbinding to the surface of Salmonella with different surface properties. The same antibodies were studied in IgG form for binding to purified O-antigen in classical surface-plasmon resonance (SPR) studies. Interestingly, monoclonal antibody STA5 showed 1000-fold lower affinities of binding to purified O-antigen as a (dimeric-binding) IgG, than the clone STA121, but both antibodies showed similar force-distance curves in atomic force microscopy analysis of FAb-fragment (monomeric) binding. Combined with structural data currently under analysis and our observations of the specificity of STA5 and STA121 for single-repeat and polymerized glycans, the most plausible explanation for this is that the high off-rate of STA121 is SPR is driven by avidity. We are currently extending our observations to test this hypothesis directly. As STA5 would also usually encounter highly clustered glycan molecules on the bacterial surface, it is likely that avidity, rather than affinity is also a major driver of bacteria-antibody interactions. The consequences of multi-valent interactions for the strength of crosslinking between bacteria of dimeric IgA will be tested and modeled in the next months. These observations are also soon to be extended to bacterial capsular polysaccharides, starting with the MenA capsule from Neisseria meningitidis.
By combining these tools to examine antibody-glycan interactions across scales, we will dissect how protective mucosal immunity against bacterial glycans is generated, and will identify the determinants for its success. Already in the first half of this project we have developed novel glycan analysis techniques, new glycoconjugation technqiues and advanced data on the physical interactions between antibodies and highly flexible glycan polymers.
Having established the tools and quantitative analysis of our vaccines, the second half of this project will focus on addressing the implications of chemical and biophysical properties of our vaccines on each stage of oral vaccine uptake and presentation. This will be achieved by looking an vaccine-induced antibody responses and using surrogate in vitro systems with anti-IgD linked to lipid bilayers via polymers with differing length and flexibility. BCR knock-in mice will also continue to be generated and studied with an aim to understanding how BCRs are generated with the potential to bind bacterial glycans in mouse and man.
Having established the tools and quantitative analysis of our vaccines, the second half of this project will focus on addressing the implications of chemical and biophysical properties of our vaccines on each stage of oral vaccine uptake and presentation. This will be achieved by looking an vaccine-induced antibody responses and using surrogate in vitro systems with anti-IgD linked to lipid bilayers via polymers with differing length and flexibility. BCR knock-in mice will also continue to be generated and studied with an aim to understanding how BCRs are generated with the potential to bind bacterial glycans in mouse and man.