Two-Dimensional LC-MS Deconvolution to Achieve Deep Understanding and Accurate Quantification of Biological Drugs

Complete characterization of large and complex drug candidates remains a challenging task. Scientists are asked to report not only the main components but also traces of truncated or modified versions of the drug candidate. Intact mass analysis is an analytical assay for the identification and relative quantification of large molecules in a sample.

Acquired LC-MS spectra are deconvoluted to condense the individual LC-MS peaks into a neutral mass signal. In order to achieve sufficient mass accuracy, recombinant proteins like intact IgG-type antibodies of about 150 kDa in size are cut by specific enzymes into fragments. For example, the IdeS enzyme cuts the heavy chain of an antibody at the hinge region just below the disulfide bridges that hold the two light-heavy chain pairs together. This treatment produces a fragment that still has the ability to bind the antigen but is smaller in size, called the F(ab’)2 fragment (Figure 1). Further reduction of the inter-chain disulfide bonds results in a further cleavage into the light chain as well as the remainder of the heavy chain (Fd’ fragment), both with a size of about 25 kDa, a size for which current mass spectrometers have sufficient resolution for an accurate mass determination.??

A straightforward approach is then to detect a rough elution time range for each fragment and to deconvolve an averaged LC-MS spectrum, as shown in Figure 2 for the Fd’ fragment. We can clearly identify the Fd’ fragment with intact intra-chain disulfide bonds as well as an oxidized version of it. While such an approach results in the identification of the major compounds, a lot of information about lower abundant species is missed. A total ion chromatogram of the corresponding elution range (Figure 3) shows a more complex structure, not just two major components as indicated by the deconvoluted spectrum.

Scientists usually have to split the elution time range further, deconvolve each subrange, and judge if a further splitting of the elution time subranges is necessary (Figure 4). Lower abundant versions of the Fd’ fragment may include partial reduction of the intra-chain disulfide bonds as well as other amino acid modifications. Such modifications only result in a marginal change in the overall hydrophobicity of the Fd’ fragment, making it a challenge to separate them chromatographically. This causes a problem if accurate quantification is required since overlapping peaks result in only partial integration of the total available molecule amount. Furthermore, the high number of deconvoluted spectra makes it difficult to gain a complete understanding of the sample since the relationship between the signals is lost. Finally, such a manual approach is extremely time-consuming and may even introduce significant bias.

To address this issue, Genedata Expressionist? for Mass Spectrometry features a two-dimensional deconvolution algorithm, which allows for the identification and accurate quantification of even trace amounts of large molecules in one single plot (Figure 5). It is quite obvious that the two major signals belong to the Fd’ fragment with intact inter-chain disulfide bonds plus the oxidized version of it (boxes I and XI in Figure 5). Even better, more modified versions of each main signal are visible. A signal at +162 Da from the main signals (Figure 5, black lines) indicates glycation of an amino acid. Detection and quantification of the extent of the glycation is a critical parameter in drug development, since glycation may alter the protein function. Another low abundant signal at -17 Da from the main signals (Figure 5, orange lines) is likely to be a succinimide formation from an asparagine amino acid. Furthermore, we can spot the partially reduced and the fully reduced versions of the disulfide bridges (Figure 5, red lines), possibly an effect from the sample preparation procedure. We can also identify several ion adducts, like a sodium adduct (Figure 5, blue lines). Although these signals might also be detectable in manually generated deconvoluted mass spectra, it is much more difficult to identify their relationship to each other.

Besides providing an easy way to gain a complete understanding of the sample content, 2D-deconvolution in Genedata Expressionist also provides other advantages for quantification (Figure 6).

Each signal on the two-dimensional plot can be integrated and used for relative quantification (Figure 5, grey boxes), bypassing the problem of overlapping peaks in a total-ion chromatogram. Even low abundant signals (on the sub-percent level) can be quantified and associated to specific modifications of the Fd’ fragment.

Summary

Genedata Expressionist for Mass Spectrometry uses a two-dimensional?deconvolution approach for accurate?identification and quantification of drug?candidate fragments and modifications. This fully automatable approach overcomes the time-intensive and non-reproducible?selection of elution time ranges. Furthermore, even co-eluting molecules can be?identified and accurately quantified.?

Learn more: