PAT-H-MS coupled with laser microdissection to study histone post-translational modifications in selected cell populations from pathology samples

Abstract

Background: Aberrations in histone post-translational modifications (hPTMs) have been linked with various pathologies, including cancer, and could not only represent useful biomarkers but also suggest possible targetable epigenetic mechanisms. We have recently developed an approach, termed pathology tissue analysis of histones by mass spectrometry (PAT-H-MS), that allows performing a comprehensive and quantitative analysis of histone PTMs from formalin-fixed paraffin-embedded pathology samples. Despite its great potential, the application of this technique is limited by tissue heterogeneity.

Methods: In this study, we further implemented the PAT-H-MS approach by coupling it with techniques aimed at reducing sample heterogeneity and selecting specific portions or cell populations within the samples, such as manual macrodissection and laser microdissection (LMD).

Results: When applied to the analysis of a small set of breast cancer samples, LMD-PAT-H-MS allowed detecting more marked changes between luminal A-like and triple negative patients as compared with the classical approach. These changes included not only the already known H3 K27me3 and K9me3 marks, but also H3 K36me1, which was found increased in triple negative samples and validated on a larger cohort of patients, and could represent a potential novel marker distinguishing breast cancer subtypes.

Conclusions: These results show the feasibility of applying techniques to reduce sample heterogeneity, including laser microdissection, to the PAT-H-MS protocol, providing new tools in clinical epigenetics and opening new avenues for the comprehensive analysis of histone post-translational modifications in selected cell populations.

Keywords: Epigenetic marker; Formalin-fixed paraffin embedded; Histone post-translational modifications; Laser microdissection; Mass spectrometry; PAT-H-MS; Proteomics.

Fig. 1

Quantitation of hPTMs by PAT-H-MS in H&E-stained samples. a Schematic representation of the procedures used to isolate histones from FFPE mouse spleen using PAT-H-MS on H&E-stained sections compared with the classical approach. b Percent relative abundances (%RA) profiles for H3 peptides obtained using the two preparation methods. c hPTM %RA correlation for the values shown in b. Pearson correlation coefficients (r) and p values are shown

Fig. 2

PAT-H-MS coupled with manual macrodissection. a H&E staining of mouse brains bearing glioblastoma xenografts (x). Scale bar = 2 mm. b Number of sections and histone octamer yields for the samples shown in a (samples 1–3) and d (sample 4). c Heatmap display and nonsupervised clustering of the log₂ of ratios for the indicated hPTMs for macrodissected mouse brains and xenografts from samples 1–3. L/H relative abundances ratios, obtained with the super-SILAC strategy (light channel: breast cancer biopsy, heavy channel: spike-in super-SILAC standard) normalized over the average value across the samples are shown (d) H&E staining of a mouse brain bearing a glioblastoma xenografts (x) of reduced size

Fig. 3

PAT-H-MS coupled with laser microdissection. a Schematic representation of the procedures used to isolate histones from FFPE mouse spleen using the classical PAT-H-MS approach (right) or LMD-PAT-H-MS (left). The appearance of extracted histones using the two procedures is shown for two representative breast cancer samples (LuA1 and TN1). b H&E staining for LuA1 and TN1 sections. The encircled areas show the microdissected tumor cells. Scale bar = 10 mm. c Number of sections and histone octamer yields for six breast cancer luminal A-like (LuA) or triple negative (TN) samples (including the samples LuA1 and TN1 (bold) show in a and b). d Elution profiles of H3-modified (27–40) forms for sample TN1, which was processed using LMD-PAT-H-MS from one section of starting material. Extracted ion chromatograms (XICs) for 14 differentially modified forms are displayed

Fig. 4

Analysis of luminal A-like and triple negative breast cancer samples by LMD-PAT-H-MS. a Heatmap display and nonsupervised clustering of the log₂ of ratios obtained for the indicated hPTMs for microdissected luminal A-like and triple negative breast cancer samples, using LMD-PAT-H-MS (left) or the classical PAT-H-MS approach (right). L/H relative abundances ratios obtained with the super-SILAC strategy (light channel: breast cancer biopsy, heavy channel: spike-in super-SILAC standard) normalized over the average value across the samples are shown. Modified peptides significantly different in the two subtypes are indicated by asterisks. The grey color indicates peptides that were not quantified in the heavy channel or in both heavy and light channels, for which a L/H ratio could not be calculated. Encircled grey squares indicate peptides that were not quantified only in the light channel (L/H ratio = 0). b Ratios obtained for the indicated peptides in the luminal A-like or triple negative breast cancer samples shown in a. c Ratios obtained for the K9me3 peptide in frozen luminal A-like or triple negative breast cancer samples compared with their corresponding normal breast tissue. Five samples were analyzed in the top panel and three in the bottom one. Samples in b and c were compared by t test. Error bars represent standard error from three to six patient samples. *p < 0.05, **p < 0.01

Fig. 5

Validation of the K36me1 increase in triple negative breast cancer samples by PAT-H-MS. a Heatmap display of the log₂ of ratios obtained for differentially modified forms of the H3 27–40 peptides in 10 luminal A-like and 10 triple negative breast cancer samples, using the classical PAT-H-MS approach. L/H relative abundances ratios obtained with the super-SILAC strategy (light channel: breast cancer biopsy, heavy channel: spike-in super-SILAC standard) normalized over the average value across the samples are shown. Modified peptides significantly different in the two subtypes are indicated by asterisks. b Ratios obtained for the indicated peptides in the luminal A-like or triple negative breast cancer samples shown in A. Samples were compared by t test. Error bars represent standard error from ten patient samples. **p < 0.01, ***p < 0.001