Stability of ADCs in sub-cellular fractions
04th Jun 2024
Measurement of warhead release of common ADC linkers conjugated to antibodies after incubation in lysosomes and S9 fraction
Antibodies are highly targeted biomolecules which are able to recognise proteins within biological systems. Mechanisms exist within the cell membrane which transport the antibodies into vesicles which are internalised into lysosomes containing high enzyme concentrations . Antibody drug conjugates (ADCs) utilise the targeting properties of antibodies to deliver highly cytotoxic molecules directly to sites of interest. By tuning the chemistry of the linkers between the protein and drug portion, their release can be controlled to occur within these lysosomes, minimising the amount of off-target toxic effects. Lysosomal and liver extracts can be utilised to determine how linkers and toxins will behave under these conditions to ensure a complete and expected release. Mass spectrometry represents an ideal tool for this work due to the complex matrix effects of these subcellular fractions.
Here we outline an analysis strategy to determine the release of three ADCs consisting of three toxin-linkers (Vedotin (Val-Cit), Deruxtecan (Gly-Gly-Phe-Gly), and Mafodotin (uncleavable)) with the same monoclonal antibody scaffold in human liver S9 fraction and human lysosome. The samples were incubated in human liver lysosomes, or S9 fraction (Xenotech), at an ADC concentration of 1.3 mg/ml, for time periods up to one week at 37°C. Subcellular fractions were also bolstered with catabolic buffer and NADPH RapidStart (Xenotech) to ensure metabolic activity. The samples were heat inactivated after incubation for five minutes at 95°C before removal of the protein via solvent crash. All data was collected on a SCIEX X500B mass spectrometer with an EXION HPLC. The instrument was operated in IDA mode to confirm species via fragmentation. The chromatography was performed with a Kinetix C18 Column (1.7μm, 50×2.1mm) on a five minute gradient of water and acetonitrile both with 0.1% formic acid.
Figure 1: XIC of MMAE and Dxd in human liver lysosomal extract. MMAE (718 m/z) and Dxd (494 m/z) were monitored over a seven day period and saw an increase over time as the peptide linkers are cleaved releasing the toxin (t=blue 0, red 60, green 480, fuchsia 1440, purple 10080 (min))
In trials, all the ADCs were stable at up to seven days in catabolic buffers lacking the sub-cellular fractions. Figure 1 shows the XIC of MMAE and Dxd over 24 hours in lysosomal extract. After removal of the proteinatious portion of the cell fraction, the resulting chromatogram is still complicated and could not be resolved without mass spectrometry. Cleavage progresses steadily with around 50% cleavage after four hours for deruxtecan and 70% for vedotin and reaching a plateau after three days incubation.
Figure 2: Appearance of cysteine-MMAF adduct (top), MS (blue) and isotopic matching (red). Cysteine adduct of MMAF appears slowly as antibody is degraded (t=blue 0, red 60, green 480, fuchsia 1440, purple 4320, yellow 10080 (min))
The conjugate with mafodotin has no peptide cleavable site and therefore we did not see a released toxin. However, we did see the appearance of a peak at 1046 m/z which shares fragments in common with the control of mafodotin. This indicates that this peak is likely due to the cysteine adduct of mafodotin which has shown to be the released form of the molecule after proteolytic cleavage of the antibody . The levels of cysteine-mc-mafodotin increased more slowly than those of the cleaved.
Speak to a member of the team.
Related Content
Poster Presentation
A standardised LC-MS workflow for rapid DAR determination of thiol-conjugated ADCs across a variety of toxin linkers
Poster Presentation
SDE-100 a stochastic cysteine linked Vedotin ADC: Assessing comparability of extent and site of conjugation using multiple orthogonal analytical approaches
Poster Presentation
