Mining The Human "Igome"

 Antibody discovery has traditionally relied on isolating B-cells, however this approach overlooks the true defenders circulating in plasma – the antibodies that actually neutralize infection. This gap is particularly pressing in malaria, where protection against severe disease often hinges on antibodies targeting Plasmodium falciparum erythrocyte membrane protein 1 (PfEMP1), the parasite adhesin responsible for trapping infected red blood cells in small vessels.   

In a recent study published in PNAS, researchers combined antigen-specific enrichment with mass spectrometry and B-cell transcript sequencing to capture and reconstruct antibodies directly from human plasma. Working with samples from a nine-year-old Tanzanian child with naturally acquired immunity, they identified a clonal antibody family that potently inhibits EPCR binding across diverse PfEMP1 variants. Structural analysis confirmed that the antibodies target conserved residues at the interaction site, overlapping with – but distinct from – previously described broadly protective antibodies.  

The results highlight both the feasibility and the potential of plasma-based antibody sequencing to illuminate naturally acquired immune responses in malaria – and beyond. To find out how the team pieced their findings together, and the broader implications for infectious disease research, we spoke with coauthors Thierry Le Bihan, Thomas Lavstsen, and Louise Turner. 

Could you explain how you were able to sequence a naturally acquired antibody directly from plasma? 

Thierry: The breakthrough really came down to solving what I like to call a "mixed puzzle problem." Imagine you've got 10 to 50 different jigsaw sets – say, different castles against slightly different skies – and someone’s mixed all the pieces together before asking you to reconstruct each set. That's essentially what you're doing when sequencing polyclonal antibodies from plasma – highly similar, yet distinct sequences.  

Traditional antibody discovery has always relied on isolating individual B-cells. However, this approach has a fundamental limitation: it doesn't directly reveal what's circulating in the blood – the molecules responsible for neutralizing pathogens. Most circulating antibodies are produced by long-lived plasma cells in the bone marrow, rather than the circulating B-cells we typically study. There’s often a disconnect between what we find in B-cell repertoires and that which provides immunity.  

Our approach combines mass spectrometry analysis of antigen-purified plasma IgG with sequencing of the B-cell receptor. We're essentially mining the human “Igome" – the set of antibodies circulating in humans that are, for the most part, already proven functional through real immune responses.  

Louise: As Thierry says, we first began talks with Rapid Novor’s team on how to identify antibodies from plasma of malaria-exposed individuals quite a while ago. This has been a goal of ours since 2015, when we first found we could affinity purify really interesting antibody pools from the plasma of Tanzanian children who had experienced malaria. In brief, individuals exposed to malaria develop immunity to severe disease through antibodies targeting the parasite’s virulence proteins, known as PfEMP1. These proteins enable infected red blood cells to adhere to blood vessel walls, allowing the parasite to avoid circulation – and thus destruction – in the spleen. Through evolution, PfEMP1 proteins have diversified to escape antibody attack, and many had predicted the immune system to be incapable of developing broadly reactive antibodies against PfEMP1. 

Once we realized we could physically purify and directly observe IgG that was effective not only against the recombinant PfEMP1 protein used for purification, but also against other PfEMP1 variants, we became eager to determine the molecular makeup of these antibodies and understand how they work. 

What motivated your team to focus on antibodies against PfEMP1 and its binding to the EPCR receptor?  

Thierry: The whole pathology of severe malaria effectively boils down to a deadly game of “biological hide-and-seek.” Malaria parasites have evolved a mechanism to make infected red blood cells sticky by displaying PfEMP1 proteins on their surface. These proteins bind to receptors like EPCR on the walls of small blood vessels, causing infected cells to stick to the walls, rather than circulate normally. It's this "sequestration" that blocks blood flow to vital organs (e.g. the brain) and is the cause of major issues – not the parasite itself, but rather the traffic jam it creates in your microvasculature. What makes EPCR particularly concerning is that parasites using EPCR-binding PfEMP1 variants are consistently associated with severe malaria cases across Africa and Asia. Therefore, blocking this molecular “handshake” between PfEMP1 and EPCR offers the potential to prevent severe disease altogether. 

Here's where things get interesting from an immunological perspective: people living in malaria-endemic areas naturally develop antibodies that block the PfEMP1-EPCR interaction, and this correlation is notable for its protective effect against severe malaria.

The technical challenge that made this perfect for our approach is that PfEMP1 proteins are incredibly diverse, as this is the parasite's immune evasion strategy. However, here's the catch: the EPCR-binding site must remain relatively conserved to maintain its function; in essence, if you need the same key to work, you can't change the lock. This represents an area of weakness, susceptible to target from (broadly) protective antibodies – and these are exactly the kinds of antibodies we can now directly identify from human plasma.  

From our perspective, this was an ideal test case because we knew there should be functional, broadly reactive antibodies circulating in people who'd survived malaria exposure. We just needed the right tools to find them.  

Thomas: As explained above, EPCR-binding PfEMP1 has been associated with severe outcomes of P. falciparum infection in numerous studies across Africa and Asia. The PfEMP1 protein family is ancient – it’s also found in malaria parasites infecting great apes – meaning these proteins have had millions of years to evolve, becoming one of the most diverse protein families known. However, to retain receptor binding, the structure and key chemical characteristics of EPCR-binding domains must be maintained. This is the parasite's weak point – and we’re eager to understand more about how antibodies target it. 

Was there a “eureka” moment during development? 

Thierry: I wouldn’t say it was “eureka,” more a sigh of relief! The breakthrough came when we finally started detecting and assembling CDR3 regions – both heavy and light chains. These are like the real IgG fingerprints, unique to each antibody.  

When working with a mixture that’s too complex or a poorly purified IgG fraction (it's difficult to differentiate between the two when lacking feedback tools), the results are often skewed. The constant regions – the antibody Fc domain and framework regions – are conserved, and thus overrepresented in the MS data. Meanwhile, the complementarity-determining regions – CDR3 in particular – are highly variable and prone to mutation, resulting in their underrepresentation. CDR3 is where all the action happens – the “business end” of the antibody that contacts the antigen. But as every CDR3 is unique, with multiple somatic mutations from affinity maturation, they often became lost amidst the noise of our early experiments.  

The moment we started consistently detecting and sequencing CDR3 regions was a huge relief. We weren't just obtaining generic conserved antibody sequences; rather, we were acquiring the actual, functional signatures that make each antibody unique. It was as if, finally, we’d found the correct puzzle pieces to the castle tower, amidst endless segments of blue sky.  

That feeling of relief changed our confidence in the method. We then knew we were seeing the antibodies that were doing the real immunological work. 

Louise: First, we needed to purify IgG with broad reactivity to EPCR-binding PfEMP1. While we had successfully done this before using pooled plasma from many children, this time we needed enough material for MS-seq analysis from a single child. It was no easy task, but one we managed through close collaboration with our long-time colleagues in Tanzania. Because we had previously studied the acquisition of PfEMP1 immunity under varying malaria transmission intensities, we were able to identify a few samples that offered the best prospects. From these, we successfully isolated sufficient IgG using a single recombinant PfEMP1 protein. 

Our next big moment came when we screened the first recombinant monoclonal antibodies that Thierry and his team had proposed as candidates for broadly reactive PfEMP1 antibodies. Although the MS analysis was performed on plasma IgG purified for reactivity to PfEMP1, the sample was still expected to contain many distinct IgG species. We tested 12 recombinant monoclonal antibodies (mAbs), using different combinations of the heavy and light chain protein sequences identified by Thierry. We were very excited to see one of these bindings across 13 of 19 different PfEMP1 proteins, covering the extreme sequence diversity of the PfEMP1 family. 

What was the greatest technical or analytical challenge you faced? 

Thierry: One of the greatest challenges was simply handling the minuscule amount of material we had to work with. Ideally, we prefer to work with about 1mg of purified antibody – but we had nowhere near that available. Asking for additional materials from our source (a 9-year-old child in Tanzania) was neither realistic nor ethical. This forced us to rethink our experimental design.  

Normally, we'd run 7 to 10 different digests with various peptide chemistry modifications, such as altering cysteine into a "lysine analogue" for trypsin digestion, or adding positive charges at the C-terminal end for non-specific digestions with chymotrypsin and pepsin. All of this provides us with confidence in peptide overlap and redundancy.  

As we didn’t have that luxury here, we had to strip our approach down to the essentials – just a few proteases and a minimal experimental design. Every decision had to count; we knew we only had enough material for essentially one LC-MS run per digest. It goes without saying that this was not the time for a column to clog or an ESI spray to become unstable! 

How could this work impact infectious disease research more broadly? 

Thierry: We believe our approach could change how we study naturally acquired immunity across infectious diseases. Previous understanding of the B-cell capabilities has been limited, though this represents just a fraction of the full picture. We're further constrained when studying humans, as our access is mostly to peripheral blood mononuclear cells (PBMCs), which only represent roughly 2 percent of the total B-cell population. The real action, however, is with antibodies circulating in the blood, mostly secreted by long-lived plasma cells in the bone marrow – not the circulating B-cells we can easily sample. 

The broader implications are significant. First, we're essentially gaining access to antibodies that have already passed the ultimate test: they've been selected and refined in real human immune responses over months or years of pathogen exposure. These aren't laboratory curiosities; they're battle-tested molecules proven against real infections, and so this could be transformative for vaccine development. Instead of guessing which epitopes might be important, we can directly identify the targets being hit by naturally protective antibodies. We're seeing this with our malaria work – the antibodies we've identified target conserved regions that the parasite can't easily mutate away from, as they’re essential for its function. This is exactly the kind of information you want when designing a vaccine.  

The method isn't limited to malaria either – any pathogen for which people develop naturally acquired immunity could be a target. Take tuberculosis, for example, where some individuals naturally control the infection, or hepatitis B, where a few can clear the virus. We could potentially identify the protective antibody responses in these people and, from that, reverse-engineer vaccine approaches. That said, I may be getting ahead of myself – there's still a lot of work to do and plenty of challenges ahead. 

There's also a broader shift in how we think about antibody discovery. Instead of starting from scratch with synthetic libraries or hoping to find the right B-cells, we're mining the human "Igome" – all those functional antibodies already circulating in people who've survived infections. We're all walking goldmines of immunological information, if you know how to extract it.  

The technical advances here extend beyond infectious diseases as well. This approach could potentially be effective in any situation where antigen-specific antibodies are present in circulation, such as autoimmune diseases, cancer, or allergies. We're democratizing access to the human antibody repertoire in a way that wasn't previously possible.  

Thomas: Understanding how our immune system can tackle the extreme diversity of P. falciparum PfEMP1 proteins is central to developing preventive vaccines and treatments against malaria. Antibodies provide a powerful tool to define the key epitopes that should be targeted in vaccine design. With recent advances in AI-driven protein engineering, we now have the opportunity not only to design improved vaccine antigens to capture these conserved epitopes but also to develop the antibodies themselves as potential prophylactic or therapeutic interventions. 

More generally, being able to tap into the functional antibody reservoir through MS analysis of antigen-bound plasma antibodies may enable us to acquire a more complete picture of how antibodies cooperate to fight foreign intruders. Recent work on other malaria antigens and cancer research has shown that interactions between antibodies can occur after the foreign protein is targeted. Such antibody-antibody interactions may lead to synergistic effects of multiple antibody binding, and could be important for efficient immunity against difficult pathogens such as malaria or other viral pathogens or cancer. 

What are the next steps for your team? 

Thierry: I should be upfront here: we're a for-profit company, so while we love to publish and contribute to science, we also need to generate revenue. Any method development, as well as exciting, has to be robust and commercially viable. We constantly balance providing client services with conducting R&D, as it’s important we stay competitive. Currently, we offer polyclonal antibody sequencing services for a continually expanding range of species, including humans, rabbits, alpacas, goats, sheep, chickens and dogs. We're slowly working out the details of mouse models too, which, while technically challenging, is crucial for pharmaceutical applications.  

On the human side, we're exploring several autoimmune conditions. The challenge here is that some patients have very low-titer, disease-specific antibodies circulating, which makes them technically demanding for our approach. Nonetheless, it's something worth tackling. This is clinically relevant work that could lead to a better understanding of disease mechanisms. We're also developing new separation technologies, as well as tackling more challenging antigens. One thing I've learned is that sample preparation and enrichment steps are critical. You can have the best mass spectrometer in the world, but if your antibody purification isn't clean, you're fighting an uphill battle.  

I'm fortunate to work with a very creative team of technicians, scientists, and bioinformaticians. This work spans multiple fields, and so we continually learn from one another. The bioinformatics team constantly encourages us to produce cleaner data and better separation, as it’s their job to make sense of it all. Technicians often come up with practical solutions that we scientists tend to overcomplicate – all towards a goal of making this technology accessible enough that it becomes a standard tool for immunologists, rather than a specialized technique.  

Thomas: We’ve just scratched the surface here. Everyone living in malaria-endemic regions will need to develop immunity to these PfEMP1 proteins to become immune. But we need to understand whether there’s a common path to immunity against severe malaria – whether people generate the same types of antibodies that bind the same epitope, and whether some individuals produce truly “magical monoclonals” capable of targeting all PfEMP1 variants, as we believe some do. 

Thierry Le Bihan is Vice-President of Research and Development at Rapid Novor Inc., Canada

Thomas Lavstsen is a Professor at the Department of Immunology and Microbiology, University of Copenhagen, Denmark

Louise Turner is a Senior Scientist at the Department of Immunology and Microbiology, University of Copenhagen, Denmark

Newsletters

Receive the latest analytical science news, personalities, education, and career development – weekly to your inbox.

I have read and understand the Privacy Notice*

Subscribe