In my work, I mainly look at antigen-bound antibodies and this means a lot of analysing interfaces. Specifically, I spend a lot of my time examining the contributions of complementarity-determining regions (CDRs) to antigen binding, but what about antibodies where the framework (FW) region also contributes to binding? Such structures do exist, and these interactions are rarely trivial. As such, a recent preprint I came across where the authors examined the DE loops of antibodies was a great motivator to broaden my horizons!
Antibodies achieve their high specificity and affinity through use of the so-called CDRs. Six loops (CDRH1-3 on the heavy chain and CDRL1-3 on the light chain) of highly varied length and sequence are found on each antibody and these contribute the majority of interface residues. These loops extend from the FW region, the sequence of which is more conserved. However, next to CDR1 and CDR2 there exists another loop. This loop joins the D and E strands of the immunoglobulin fold, and is often referred to as the DE loop. Traditionally, this loop has not been considered as being a CDR and is often regarded as part of the FW region. This paper examines the DE loop as another CDR (CDRL4 and CDRH4), and makes conclusions about the impact of the loop on traditional CDRs. It also shows that in certain disease states, the DE loop may contribute significantly to antigen binding and as such might be an important factor to consider when looking for effective antibodies.
As part of the analysis, the authors retrieved all antibody structures from the PDB, and filtered these to remove structures where models have been poorly built. From these they also extracted DE loops which were then clustered using the DBSCAN algorithm. The authors also calculated buried surface areas for the antibodies and sequence variability in DE loops.
CDRL4 loops tend to mostly have six residues, with some loops having eight. Upon clustering of length-matched loops, the authors were able to identify four L4-6 clusters and show that these clusters are mainly separated by chain identity. 2 of the clusters included only κ chains, while one included a mixture of κ and λ chains. The last structure included only λ chains. Sequence analyses of these clusters show that the differences mainly come from a small number of positions within the loop. Markedly, a single residue being either Glycine or another residue makes all the difference between the two most populated clusters. For L4-8 loops, the authors were only able to find one cluster with very little sequence variation. H4-8 loops partition into two clusters with very high sequence variation. One of these clusters, H4-8-1 includes most of the structures, and is very similar to the L4-8 conformation, but with substantially different sequences.
Next the authors show that CDRL4 interacts in some way with CDRL1 and CDRL2. For instance, certain CDRL1 canonical forms are shown to prefer different CDRL4 conformations, and that in these cases CDRL1 is stabilised by hydrogen bonds made with CDRL4, suggesting that considering DE loop conformation may be of value during rational antibody design.
The team also examined insertions in L4 and H4. Surprisingly, in all PDB entries with an insertion in DE loops apart from one, the antibody is broadly neutralising against HIV-1. Further, when there is an insertion in either of these loops, the loop tends to be more involved in the binding interface than most of the traditional CDRs, highlighting the ability of the loop to provide at least part of the paratope. The paper shows that CDRL4 loops with insertions tend to stabilise CDRL1 loops through hydrogen bonding. Lastly, using Aleks’ database (OAS), the team analysed antibody sequences from HIV patients and compared them to healthy sequences, showing that there are substantially more insertions in sequences from studies on broadly neutralising anti-HIV-1 antibodies.
In summary, the paper argues that DE loops have many of the characteristics of traditional CDR loops and can significantly contribute to antibody-antigen interactions. The authors highlight the importance of the loop when searching for effective antibodies and in rational antibody engineering.
