Comparison of two FFPE preparation methods using label-free shotgun proteomics: Application to tissues of diverticulitis patients
Graphical abstract
Introduction
Formalin-fixed paraffin-embedded (FFPE) tissues stored in hospital biobanks represent a valuable resource for retrospective analysis. Hence, larger populations can be studied, increasing the possibility of identifying significant and specific potential biomarkers. FFPE tissues are advantageous in many ways when compared to fresh frozen specimens; they can be stored for very long periods of time and are easier to obtain as they are collected for systematic routine diagnosis. However, the formalin fixation of tissue involves the creation of inter- and intra-molecular crosslinks [1], [2]. Therefore, one of the challenges for proteomic studies consists of extracting the highest number of different proteins present in the sample and obtaining confident protein identifications. Many protocols exist for “unlocking” FFPE crosslinks induced by the formalin fixation of tissue. These involve protein solubilization and digestion prior to analysis using various techniques including proteomics [3], [4], [1], [5], [6], [7], [8]. These techniques were recently reviewed [1], [9], [10]. Among these methods, the citric acid antigen retrieval (CAAR) method, which was initially developed for imaging by mass spectrometry (IMS), was adapted using laser capture microdissection (LCM) before shotgun proteomics [4], [11]. The advantage of the CAAR method is that it is performed on slices mounted on glass slides allowing conservation of the 2D spatial resolution for IMS. Applying an AR strategy using a tissue section on a slide for FFPE preparation before shotgun proteomics is rarely described in the literature [11].
Some papers commented on the stability and quality of FFPE material over time, the variability of the fixation protocols, the capacity to identify the same range of proteins as with fresh frozen samples and the possibility for studying posttranslational modifications such as glycosylation and phosphorylation [7], [12], [13], [14], [15]. However, the variability of these “unlocking” protocols on the quality of results, mostly regarding the quantitative reproducibility or repeatability, was seldom commented on.
In the context of the discovery of disease biomarkers, various studies addressed FFPE specimens, and some used differential analysis by label-free shotgun proteomics [16], [17], [14]. One of the instrument systems enabling a differential label-free proteomics study of complex samples with high sensitivity, high specificity and a linearity over 4 orders of magnitude in terms of the dynamic range of the protein concentration is the 2D-nano UPLC (ultra-performance liquid chromatography) Q-Tof Synapt HDMS™ G2 system (Waters Corporation, Milford, USA). This data-independent MSE offers some gains over traditional data-dependent analysis, including good measurement reproducibility, identification of low-abundance peptides, faster throughput due to the simultaneous fragmentation of multiple peptides and direct relative quantification [18].
Translational research by shotgun proteomics for the discovery of biomarkers might involve the use of pooled sample extracts [19], [20], [21]. Hence, working on pools somehow results in a smoothing out of characteristics with high biological variability linked to demographic differences of the patients. The biological variability can also be driven by the clinical staging of diseased samples, the selection of the specimens, the surfaces of cells and tissues treated (with or without micro- or macrodissection), the sample treatments before the FFPE process and differences in the storage of the samples. All of these parameters have to be carefully considered and controlled when composing balanced pools to avoid bias in the selection of samples. Tissue biobanks involve the collection and storage of residual material used first for the diagnosis of patients. Hence, some pathological stages or grades are, by nature, rather tiny specimens (such as some colorectal adenoma polyps). Therefore, limited quantities of materials are available for translational research.
These limitations and precautions drove us to perform the preliminary study presented in this paper comparing the respective performances (efficiencies and repeatabilities) of two FFPE preparation protocols using differential label-free quantitative analysis to select the best option for application to a clinical study. Hence, a few slices of a FFPE specimen of the same colorectal cancer patient were treated by the following two methods in triplicate: the commercial FFPE–FASP (filter-aided sample preparation) kit and an on slice antigen retrieval-derived protocol (“On Slice AR”). Finally, to evaluate its reproducibility, the FFPE–FASP method was applied on FFPE tissue samples of diverticulitis patients for comparison of the proteins obtained within tissue areas with inflammation to those obtained from within matched healthy zones.
Section snippets
Materials and methods
Fig. 1A shows the workflow of the experimental strategy steps followed in our work for the two FFPE preparation protocols that were tested and compared.
FFPE–FASP and On Slice AR protocol considerations
The differential label-free analysis in this work took place prior to a translational retrospective cancer research project being performed on FFPE materials. Our aim was to compare the efficiencies and more importantly the repeatabilities of the two methods of FFPE preparation. The methods tested were: a commercial solution “FFPE–FASP” and a method developed in-house on FFPE sections mounted on glass slides using an antigen retrieval-derived protocol “On Slice AR” [4]. The comparison of the
Number of proteins identified and quantified
The “simulated 1D-UPLC” separation was chosen to limit the running time. These conditions enabled approximately one-third of the identifications observed on the same sample analyzed with a classical 2D-UPLC separation with five elution steps taking 120 min per separation (data not shown). The number and quantity (μg) of proteins identified and common to the three UPLC–MSE runs performed per process as well as those present in at least one run are provided in Table 2. The results of the total
Total protein recoveries and protein identification
Table 3 summarizes the results concerning the total quantities of proteins recovered and the proteomics results obtained for the diverticulitis study.
Compared to the results obtained with the CRC sample, the material recovered after deparaffinization was better with diverticulitis tissue samples than with the CRC tissue. This is probably due to differences in the cell tissue and the macrodissections. The total protein and peptide recoveries obtained using FFPE–FASP on these sample types were at
Conclusions
In an ideal situation, both methods of extraction could be combined to pool and analyze the extracted peptides. However, in a reasonable experimental plan involving dozens of sample specimens, this remains rather technically challenging. Overall, the potential advantage of the new On Slice AR method described in this work was based on the fact that the “unlocking” and digestion steps were performed directly on slices placed on the glass slides and not in microtubes. Our hypothesis was that this
Transparency document
Acknowledgments
We thank the Biothèque universitaire de Liège, BUL, CHU de Liège for the selection and preparation of the sample specimens.
We also thank the GIGA Proteomic Facility, ULg involved in the shotgun proteomic analyses.
This research was supported by the Belgian Cancer Plan, Belgian Ministry of Health, Belgium, the “Centre Anti-Cancereux”, CHU de Liège, and the “Fédération Wallonie Bruxelles”. This work was possible due to the contributions of MSD.
NS is a FNRS logistic collaborator.
We thank Stephanie
References (34)
- et al.
Internal calibrants allow high accuracy peptide matching between MALDI imaging MS and LC–MS/MS
J Proteomics
(2012) - et al.
Comparison of ultrafiltration units for proteomic and N-glycoproteomic analysis by the filter-aided sample preparation method
Anal Biochem
(2011) - et al.
Comparative proteomic analysis of cysteine oxidation in colorectal cancer patients
Mol Cells
(2013) - et al.
Absolute quantification of proteins by LCMSE: a virtue of parallel MS acquisition
Mol Cell Proteomics
(2006) - et al.
Analysis and quantification of diagnostic serum markers and protein signatures for Gaucher disease
Mol Cell Proteomics
(2007) - et al.
Comparability of differential proteomics data generated from paired archival fresh-frozen and formalin-fixed samples by GeLC–MS/MS and spectral counting
J Proteomics
(2012) - et al.
Proteomics of colorectal cancer: overview of discovery studies and identification of commonly identified cancer-associated proteins and candidate CRC serum markers
J Proteomics
(2010) - et al.
Advancing formaldehyde cross-linking towards quantitative proteomic applications
Anal Bioanal Chem
(2012) - et al.
Toward deciphering proteomes of formalin-fixed paraffin-embedded (FFPE) tissues
Proteomics
(2012) - et al.
Extraction and analysis of diagnostically useful proteins from formalin-fixed, paraffin-embedded tissue sections
J Histochem Cytochem
(1998)
Citric acid antigen retrieval (CAAR) for tryptic peptide imaging directly on archived formalin-fixed paraffin-embedded tissue
J Proteome Res
Elevated pressure improves the extraction and identification of proteins recovered from formalin-fixed, paraffin-embedded tissue surrogates
PLoS One
Novel prognostic protein markers of resectable pancreatic cancer identified by coupled shotgun and targeted proteomics using formalin-fixed paraffin-embedded tissues
Int J Cancer
Evaluation of the suitability of archival Bouin-fixed paraffin-embedded tissue specimens to proteomic investigation
Electrophoresis
Proteomic identification of immunoproteasome accumulation in formalin-fixed rodent spinal cords with experimental autoimmune encephalomyelitis
J Proteome Res
Analysis of the formalin-fixed paraffin-embedded tissue proteome: pitfalls, challenges, and future prospectives
Amino Acids
Toward improving the proteomic analysis of formalin-fixed, paraffin-embedded tissue
Expert Rev Proteomics
Cited by (17)
A laser microdissection-based workflow for FFPE tissue microproteomics: Important considerations for small sample processing
2016, MethodsCitation Excerpt :Although it is generally considered that the temperature is more important than the antigen retrieval solution itself, pH adjustment can lead to more or less efficient immunohistochemical staining [8]. Despite the fact that detergent solutions such as sodium dodecyl sulfate (SDS) are commonly used for tissue proteomics, it is generally applicable for the lysis of large amounts of tissues [9]. For studies requiring a carefull preservation of the tissue such as for matrix-assisted laser desorption/ionization mass spectrometry imaging (MALDI MSI), IHC solutions are more adequate.
Quantitative Proteomic Analysis Using Formalin-Fixed, Paraffin-Embedded (FFPE) Human Cardiac Tissue
2021, Methods in Molecular Biology