Written by Nick Gee - Conjugation Expert
Part 1 of 2: Random conjugation of labels to antibody lysines—is the concern over the impact of CDR modification on antigen binding overstated?
The degree of labelling (DoL), defined as the average number of labels per antibody in a population of conjugate molecules, typically ranges from 1 to 6, depending on the type of label used. DoLs for larger labels, such as alkaline phosphatase, tend to be at the lower end of this range. By contrast, it is common to attach 4 to 6 small molecules, such as fluorescent dyes, to an antibody. What is the probability of targeting a lysine residue in the complementarity-determining regions (CDRs)?
A typical IgG molecule may contain approximately 100 lysine residues, some of which are located in the complementarity-determining regions. I analysed around 50 monoclonal antibody sequences, and the average lysine count was 5 per antigen binding site. For now, let us assume that all lysine residues are available for modification and exhibit equal reactivity.
In a population of antibodies with three labels, 73% of the molecules will possess two unmodified antigen-binding sites, while 97% will have at least one. By contrast, the labels will be found exclusively within the CDRs in only 0.07% of cases.
In reality, the situation is more complex because a conjugate with a DoL of 3 contains subpopulations of antibodies that carry 1, 2, 3, 4, or 5 labels. Even within the subpopulation of antibodies with 5 labels, 92% will have at least one unmodified antigen-binding site, and 58% will possess two.
The presence of 100 potential modification points means that numerous combinations (or positional isomers) can arise within each subpopulation. Specifically, there are 100, 4,950, 161,700, 3.9 million, and 75 million combinations for 1, 2, 3, 4, and 5 attached labels, respectively. Furthermore, all combinations can be found in 100 ug of conjugate; such is the magnitude of Avogadro’s number.
This analysis suggests that one should prioritise tightly controlling the DoL to ensure consistent proportions of the various positional isomers, rather than unduly worrying about the CDRs.
Obviously, many factors related to conjugates and their performance are not addressed in this type of analysis. In Part 2, I will review data on the lysine reactivity and the actual distribution of labels.
---------------------------------------------------------