Epigenetic toxicity measured by high-throughput mass-spectrometry based histone analysis

References, associated documents and other information

Lauschke, V.M., I. Barragan, and M. Ingelman-Sundberg, Pharmacoepigenetics and Toxicoepigenetics: Novel Mechanistic Insights and Therapeutic Opportunities. Annual review of pharmacology and toxicology, 2018. 58: p. 161-185

Pfister, S.X. and A. Ashworth, Marked for death: targeting epigenetic changes in cancer. Nature reviews. Drug discovery, 2017. 16(4): p. 241-263

hSWATH: Unlocking SWATH's Full Potential for an Untargeted Histone Perspective. De Clerck L, Willems S, Noberini R, Restellini C, Van Puyvelde B, Daled S, Bonaldi T, Deforce D, Dhaenens M. J Proteome Res. 2019 Nov 1;18(11):3840-3849. doi: 10.1021/acs.jproteome.9b00214. Epub 2019 Aug 29

Untargeted histone profiling during naive conversion uncovers conserved modification markers between mouse and human. Laura De Clerck, Jasin Taelman, Mina Popovic, Sander Willems, Margot Van der Jeught, Björn Heindryckx, Petra De Sutter, Hendrik Marks, Dieter Deforce, Maarten Dhaenens. Scientific Reports, accepted

Monitoring the DNA methylation status or sequencing miRNA are increasingly scalable to assay large sample batches for application as toxicoepigenetic assays. They yield rich datasets that allow broad biological interpretation. Histone modification analysis on the other hand, has classically relied on site specific antibody-based techniques such as western blotting, immunofluorescence and ChIP(seq). These approaches are confined by i) the limited number of targets that can be studied in a single experiment, ii) antibody cross-reactivity, iii) a lack of combinatorial information on hPTM co-occurrence and iv) epitope occlusion. In recent years, mass spectrometry (MS)-based approaches like multiple Reaction Monitoring (MRM) and Parallel Reaction Monitoring (PRM) have become viable alternatives to these classic techniques for analysing hPTM. However, they still require the user to make a limited selection of specific targets of interest and do not allow for retrospective analysis of novel targets because only the initial targets are acquired. In fact, in fundamental research, data-dependent acquisition (DDA) is the MS technique of choice to study histones in an untargeted way, as it does not rely on prior hypotheses. It has become a promising complementary option allowing discovery-driven histone analysis.

This technique has already proven its merit in many areas, such as e.g. identification and quantification of differential proteins of whole cell lysates and monitoring of phosphorylations over time in phospoproteomics. However, histone extracts are rather unique when compared to traditional samples, as they generally contain just five small basic proteins (11 to 21 kDa) and a few co-extracted proteins and contaminants. However, more than twenty different types of hPTMs present on the histone protein backbone structure have already been described. Theoretically, this allows for these five proteins to have 7⋅10¹⁷ different proteoforms with different hPTM combinations. Taken the high amount of lysines and arginines of histone peptides into account, this translates into 5⋅10⁷ different histone peptidoforms with ArgC-like specificity. This vast amount of highly similar peptides leads to several unconventional situations, including more co-elution of isobaric peptidoforms during acquisition. Consequently, this results in a higher amount of chimeric MS/MS spectra and ambiguous spectrum annotations, leading to problematic identification. At the same time, the total histone protein abundance can be considered constant from a quantitative perspective, as it is proportional to the amount of DNA present in a cell. It is in fact the hPTMs which induce changes in peptide abundance, thereby creating an intrinsic peptide-centric setting for histone analysis. Correctly quantifying even the smallest of these differences is of crucial importance, as small changes at specific regions in the genome can have large implications in terms of gene (in)activation. Unfortunately, the quantification of histone peptidoforms in DDA has to be performed at MS1 level and is therefore very susceptible to interference of the many co-eluting isobaric peptides. Recently, sequential window acquisition of all theoretical fragment ion spectra (SWATH) emerged as an intermediate MS technique between targeted approaches such as MRM and PRM on the one hand and untargeted approaches such as DDA on the other. This technique has great potential to address many of the issues mentioned above. SWATH is a methodology that combines data-independent acquisition (DIA) with peptide spectral library matching, originally developed for MRM-like, i.e. transition-centric, quantification of large sets of proteins across multiple samples. This implies that compared to DDA, it provides the indispensable specificity extracted from the ion traces to distinguish and thus quantify the many isobaric and (near-)coeluting histone peptides, while the acquisition itself remains untargeted, as opposed to MRM and PRM. Following 3 years of optimization, we have recently published a manuscript unlocking SWATH’s full potential to get an untargeted perspective on histones (hSWATH).

In this study we show that hSWATH provides very accurate and untargeted measurement of the histone code. We illustrate the power of this approach on a benchmark experiment, wherein we deacetylated commercial histones in a time-lapse manner using HDAC1. Finally, we performed a proof-of-principle toxicoepigenetic screening of breast cancer cell lines which are either sensitive or resistant to Panobinostat, a pharmacoepigenetic epidrug. Because SWATH is a data-independent acquisition methodology, it comprises all formative peptide ions in a sample. This makes it (i) particularly fit for generating a (pharmacoepigenetic) compound library for toxicoepigenetic interpretation, while (ii) allowing retrospective targeting of already acquired data with novel targets. Taken together, these features make hSWATH the perfect stepping stone to develop a high throughput hPTM assay that can address the urgent need for a toxicoepigenetic assay that can move the field of pharmacoepigenetics forward.