ECD top-down analysis of antibodies presented by Professor Albert Heck at Proteomic Forum 2022
One of the first and traditionally the most common application of electron capture dissociation (ECD) is protein top-down analysis (and nowadays also so-called middle-down analysis) by mass spectrometry. What does this mean? Typically in proteomics experiments a single protein or a mixture of proteins is enzymatically digested into smaller peptides. For a number of reasons, trypsin is very popular for this. This creates a set of peptides with an average length of 10-12 Da and all ending either with a Lys or Arg residue on the C-terminal end (with the exception of the actual protein C-terminus). This set of peptides is then chromatographically separated and the peptides are fragmented by collision-induced dissociation (CID). The resulting fragment spectra are matched against databases to identify the corresponding peptide. It is important to stress here that this process does not identify proteins but only peptides! The proteins sequence results from the database against which the peptide-spectrum matches (PSM) are statistically matched. Sequences not in the database or modifications not considered can not be identified by this process. As it starts from small fragments, this approach is called “bottom-up“.
To better illustrate this, look at the picture below. You see a huge pile of building bricks. This represents the complex peptide mixture after digest. Now answer the following questions:
- How many different individual brick sets are in this pile? And how many of each?
- Which parts are overlapping between the brick sets?
- Are there any parts missing?
Difficult to say? Yes. But this is exactly the challenge of bottom-up proteomics. When you start from the complete sets of bricks, those questions are easy to answer. That’s the top-down approach.
While this technique has been extremely successful in the past years and the curent instrumentation is to a large extent optimized for this, there are also some major limitations with the bottom-up approach:
- The sequence arrangement of the peptides in the protein is solely based on previous knowledge and sequences in the database
- De-novo sequencing is not possible as it only can match against known or at least highly similar sequences
- The analysis of peptide mixtures is almost never comprehensive. Typical sequence coverage of a protein is maximum 70-80%. Full sequence coverage can only be achieved by the combination of several enzymes with alternative cleavage sites
- Some protein parts such as transmembrane domains contain hardly any cleavage site. These areas are usually difficult to analyze by bottom-up analysis
The counterpart is to skip the enzymatic cleavage entirely (“top-down“) or to use cleavage reactions which create only a few, much bigger fragments (such e.g. cyanogen bromide cleavage or IdeS [Genovis “FabRICATOR”]) in a “middle-down” analysis. In particular IdeS became very popular for monoclonal antibody characterization as it produces in a highly specific way 3 different antibody fragments of ~25 kDa, a weight ideally suited for top-down analysis. However, these much bigger fragments cannot be effectively fragmented by CID. But when ECD is used, the fragmentation of these highly charged bigger fragments happens very effectively allowing to deduce the protein sequence directly from the intact protein or a large peptide:
ECD has so far been successfully applied to molecules even in excess of 100 kDa. Yet the typical application is to characterize small to mid-size proteins in the range between 10 (histones) and 150 kDa (monoclonal antobodies) where the latter are at least separated into light (25 kDa) and heavy chain (50 kDa) by top-down mass spectrometry .
The classical by top-down workflow in the mass spectrometric biologics characterization is usually like this:
- determination of intact mass
- confirmation of termini (ragged ends or processing of termini such as pyroglutamate formation, Lys- removal, signal peptide removal, methionine formylation etc.)
- if further discrepancies between expected and observed mass remain, the entire protein sequence needs to be checked
- an additional task is to check e.g. for the conservation of the structure and the proper connectivity of disulfide bridges. This task could be addressed by using technologies which are structure sensitive such as IMS in general and collision-induced unfolding in particular. The latter creates a “melting curve” for the protein structure which can be compared e.g. for different batches or wild-type versus mutant
An important aspect of the top-down analysis is the spectra interpretation and sequence annotation. e-MSion’s ExdProcess viewer software has been especially developed to support this task and provides a starightforward spectra and sequence matching.