B Cell Immunology

Malaria immune responses and vaccine development 

The malaria parasite Plasmodium falciparum (Pf) with its complex life cycle has developed sophisticated immune evasion strategies to escape the humoral and cellular immune responses of the human host, similar to the complex interplay between cancer cells and anti-tumor responses. In this project, we characterize the human B and T cell responses to natural Pf infection and malaria vaccine candidates at single-cell level to understand the selection and differentiation processes of cells that target these antigens and to identify protective epitopes. By tracking the clonal evolution of the response and through the functional characterization of antigen-receptors, we develop and test ideas for the design of improved malaria vaccines in animal models. Subsequent deep molecular and cellular analyses using diverse single-cell based high-throughput methods allow us to identify parameters that critically define the quality of the vaccine response and associate with protection. The specific focus lies on understanding how to overcome immunodominance, to provide optimal T cell help, and to improve the duration of vaccine responses. Work under this project is funded by the Gates Foundation, a Wellcome Trust Discovery Award (BigFc; https://wellcome.org/grant-funding/people-and-projects/grants-awarded/breadth-igg-fc-antibody-mediated-protective), and the European Union Horizon Europe Research and Innovation Program (CAPTIVATE; https://www.captivate-malaria.eu).

The impact of pre-existing humoral immunity on B cell responses

Through the formation of immune complexes with antigen, serum antibodies modulate vaccine responses positively by promoting B cell activation and antigen-deposition in germinal centers and negatively by blocking access to epitopes. In this project, we aim to determine how pre-existing antibodies impact the cellular selection and differentiation processes of booster B cell vaccine responses in space and time. In mouse immunization experiments with adjuvanted SARS-CoV-2 spike protein as model, we track the dynamic, clonal evolution, and epitope-specificity of the cellular response by combining multi-parameter spectral flow cytometry, single-cell based Ig gene repertoire analyses and paired functional antibody assessments with spatial mass-spectrometry-based proteomics approaches. The work will provide fundamental insights into how soluble antibodies modulate vaccine responses to help guide the timing of homologous boost immunizations. 

The pathological role of anti-tumor responses in paraneoplastic neurological syndromes 

Paraneoplastic neurological syndromes (PNS) are characterized by severe motoric or mental disfunction caused by autoimmune neural cell damage. The ectopic expression of neural antigens in tumor cells likely drives the development of autoreactive B and T cells that cross the blood-brain barrier. In this project, we characterize the tumor-infiltrating and central nervous system B and T cell response in patients with PNS to determine their role in the anti-cancer response and in brain pathology. We are specifically interested in understanding the cellular origin and antigen specificity of the brain-infiltrating immune cells for the development of targeted prophylactic or therapeutic intervention strategies. The project is funded by the German Research Association as part of the Research Unit SYNABS (Synaptic pathology in autoimmune encephalitis).

AI-driven protein design has enabled the generation of targeted protein binders, highly-stable scaffolds for epitopes or functional protein motifs and the development of proteins with new-to-nature properties.

The protein design team led by Dr. Michael Jendrusch is developing cutting-edge, AI-driven tools for protein design and applying them to research questions in immunology and beyond.

The team’s activities range from routine protein design for research applications, to pushing the boundaries of both computational and experimental methods in protein design by co-developing the next generation of protein generative models in tandem with high-throughput experimental methods for protein characterization.

Protein design pipelines @ DKFZ

The team is developing and maintaining a growing number of automated pipelines for common design tasks (binder design, epitope design, protein stabilization, nanoparticle design), based on efficient and powerful protein foundation models. In the spirit of open science, we are providing our models (https://github.com/mjendrusch/salad) and pipelines (https://github.com/mjendrusch/flexcraft) to the community.

If you want to apply protein structure prediction and design in your research, please do not hesitate to get in touch – the protein design team can provide support for both routine design projects and collaborations which involve more complex co-design of pipelines and experimental approaches.