Observations on how antibodies evolve
A recent study on anti-SARS-CoV-2 antibodies retraces the steps they take during affinity maturation.
Antibodies evolve differently from most proteins. Instead of maintaining functions and/or interactions across different organisms, antibodies evolve specificity to targets of interest within organisms, in a Darwinian process called somatic hypermutation. During this process, they collect point mutations that increase their affinity to their target, with the end result being a slightly modified antibody sequence that binds with nanomolar or picomolar affinity and (usually) very high specificity.
Kirby et al recently examined the evolutionary trajectories taken by fourteen high-affinity antibodies to see what the overall effect of each mutation was, relative to the germline precursor. Their results are quite interesting and are summarized as follows:
- The mutations obtained during affinity maturation can be obtained in any order, arguing that epistasis is not a problem.
- Most mutations (47 out of 51) are the result of single-nucleotide changes, suggesting that the DNA sequence of the germline introduces a hard constraint on the accessible evolutionary space.
- Germline precursors to high-affinity antibodies already bind with nanomolar affinity, and most mutations obtained during affinity maturation don’t actually improve on this
- More evolutionary “steps” are required for germline antibodies to evolve to broadly neutralizing antibodies that can bind to many different variants of SARS-CoV-2
Incidentally, this is the first time I used an LLM to find a paper - I asked Gemini to find a citation for the claim that germline antibodies can already have nanomolar binding affinity to their targets, and it replied with this paper.