Comprehensive micro-scaled proteome and phosphoproteome characterization of archived retrospective cancer repositories.
Author(s): Friedrich C, Schallenberg S, Kirchner M, Ziehm M, Niquet S, Haji M, Beier C, Neudecker J, Klauschen F, Mertins P
Publication: Nat Commun, 2021, Vol. 12, Page 3576
PubMed ID: 34117251 PubMed Review Paper? No
Purpose of Paper
This paper investigated potential differences due to protein extraction protocol, input amount, the duration of heat-assisted de-crosslinking, and time in formalin on the proteome and phosphoproteome of formalin-fixed paraffin-embedded (FFPE) non-small cell lung cancer (NSCLC) specimens analyzed using a tandem mass tag (TMT) approach with LC-MS/MS.
Conclusion of Paper
Protein extraction using a sodium dodecyl sulfate (SDS) protocol was superior to the other protocols evaluated (SDC; direct trypsinization, DTR), resulting in the lowest mean coefficient of variation (CV) for protein intensities, better overall proteome coverage, better coverage of proteins relevant to NSCLC (such as napsin A, cytokeratins, and EGFR), and the highest reproducibility. NSCLC specimens fixed in formalin for 12-24 h or 72 h did not differ from one another in the number of proteins or the abundance of formalin-induced peptide modifications (dimethlyation, formylation, M oxidation, presence of a methylol group) identified by LC-MS/MS; and significant differences in abundance were limited to a lower abundance of two extracellular proteins in specimens fixed for 72 h compared to those fixed for 12-24 h. The amount of protein input used for TMT-labelling (20 µg, 200 µg) resulted in comparable numbers of identified proteins, although the number of identified phosphosites was lower in samples with 20 µg of protein input. The number of identifiable proteins and phosphosites were comparable among surgically resected and needle biopsy lung tumor specimens when the protein input amount was standardized (20 µg). Based on a cell culture experiment the authors concluded that heat-mediated protein de-crosslinking at 95°C for 120 min would maximize protein yield, although it was accompanied by a reduction in the number of phosphoproteins identified.
Studies
-
Study Purpose
The purpose of this study was to identify a preferred method of protein extraction for mass spectrometric analysis of the proteome and phosphoproteome of four archived FFPE adenocarcinoma specimens. Specimens were fixed in 4% buffered formalin-fixed for an unspecified amount of time and paraffin-embedded; additional information on processing and storage were not provided. Three protein extraction protocols were compared that primarily differed in the detergent used in the lysis buffer and subsequent purification measures. The detergents used were: SDS (sodium dodecyl sulfate in conjunction with paramagnetic beads), SDC detergent (sodium deoxycholate), DTR (direct trypsinization using RapiGest detergent). For each extraction protocol, one 10 µm-thick section per extraction protocol was deparaffinized in xylene, incubated in lysis buffer at 95°C for 2 h in a thermoshaker, sonicated, centrifuged, purified if required, and analyzed by LC-MS/MS for proteomic analysis using a Q Exactive Plus instrument.
Summary of Findings:
The SDS protocol was the superior protein extraction method resulting in the lowest mean CV for protein intensities (0.35) compared to DTR (0.77) and SDC (0.69); better overall proteome coverage given 2,917 proteins were identified in SDS samples compared to the 2,488 identified in DTR and the 958 proteins identified in SDC samples. SDS also had better coverage of proteins relevant to NSCLC (such as napsin A, cytokeratins, and EGFR), as well as the highest reproducibility.
Biospecimens
Preservative Types
- Formalin
Diagnoses:
- Neoplastic - Carcinoma
Platform:
Analyte Technology Platform Peptide LC-MS or LC-MS/MS Protein LC-MS or LC-MS/MS Pre-analytical Factors:
Classification Pre-analytical Factor Value(s) Analyte Extraction and Purification Protein solubilization SDS
SDC
DTR
-
Study Purpose
The purpose of this study was to determine if time in formalin can adversely affect LC-MS/MS analysis of the proteome and phosphoproteome of FFPE NSCLC specimens. A total of 30 archived FFPE NSCLC specimens were analyzed representing different diagnoses and fixation times: 12-24 h (10 adenocarcinoma specimens, 8 squamous cell carcinoma specimens), 72 h (6 adenocarcinoma specimens, 6 squamous cell carcinoma specimens). Additional information on processing and storage were not provided. One 10 µm-thick section per case was xylene deparaffinized and subjected to protein extraction using a SDS (sodium dodecyl sulfate) protocol that included removal of the detergent by acidification and centrifugation. The proteome was analyzed by LC-MS/MS using a Q Exactive Plus instrument.
Summary of Findings:
NSCLC specimens fixed in formalin for 12-24 h or 72 h did not differ from one another in the number of proteins identified by LC-MS/MS. Significant differences between specimens fixed for different durations were limited to two proteins (Clusterin and immunoglobulin heavy constant mu, IGHM), which although higher in specimens fixed for 12-24 h compared to those fixed for 72 h, were noted by the authors to be difficult to reliably quantify given they are extracellular proteins. There were also no differences in the abundance of formalin-induced modifications (peptides with dimethlyation, formylation, M oxidation or containing a methylol group) between specimens fixed for 12-24 h and those fixed for 72 h.
Biospecimens
Preservative Types
- Formalin
Diagnoses:
- Neoplastic - Carcinoma
Platform:
Analyte Technology Platform Protein LC-MS or LC-MS/MS Peptide LC-MS or LC-MS/MS Pre-analytical Factors:
Classification Pre-analytical Factor Value(s) Biospecimen Preservation Time in fixative 12 h
24 h
72 h
-
Study Purpose
The purpose of this study was to determine the optimal duration of a heat-mediated de-crosslinking step for LC-MS/MS analysis of the proteome and phosphoproteome using frozen and formalin-fixed HEK293 cultured cells. Cells were fresh frozen, immediately fixed in formalin, or fixed in formalin after a delay to fixation of 1 h at 4°C. Protein extraction included lysis in buffer containing SDS,, incubation at 95°C for 10, 30, 60, or 120 min, cleaning with SP3 (to remove SDS), and digestion. Proteomes were assessed by TMT-labelling by single shot LC-MS/MS analysis using a Q Exactive Plus instrument.
Summary of Findings:
The duration of heat-mediated protein de-crosslinking affected the stability of the proteome and phophoproteome in cultured HEK293 cells, although effects were specific to preservation method. The number of peptides identified in HEK293 cells that were fixed with formalin displayed a 20% increase over the lysis incubation timecourse, while frozen cells displayed a 20% reduction in the number of peptides identified following heat exposure for 120 min compared to 10 min. The shortest duration of lysis incubation (95°C for 10 min) resulted in the fewest number of phosphopeptides identified regardless of fixation method, increasing by ~50% after 60 min at 95°C. Phosphorlyation modifications were stable during lysis incubations of up to 2 h at 95°C. A TMT comparison between differentially fixed cells (frozen and formalin) revealed strong correlations between proteomes (r=0.96-0.99), and phosphoproteomes (r=0.7-0.9).
Biospecimens
Preservative Types
- Frozen
- Formalin
Diagnoses:
- Normal
Platform:
Analyte Technology Platform Peptide LC-MS or LC-MS/MS Protein LC-MS or LC-MS/MS Pre-analytical Factors:
Classification Pre-analytical Factor Value(s) Analyte Extraction and Purification Incubation duration/condition 10 min in SDS buffer at 95°C
30 min in SDS buffer at 95°C
60 min in SDS buffer at 95°C
120 min in SDS buffer at 95°C
-
Study Purpose
The purpose of this study was to determine if a smaller quantity of protein used as input impacts deep proteome and phosphoproteome analysis of FFPE specimens by LC-MS/MS. Eight archived FFPE surgically resected lung tumor specimens (4 adenocarcinoma specimens, 4 squamous cell carcinoma specimens) were used to compared 200 µg and 20 µg protein input quantities; eight additional archived needle biopsy FFPE specimens (4 adenocarcinoma specimens, 4 squamous cell carcinoma specimens) were used to verify results obtained with “biopsy equivalent” specimens (20 µg protein input). Six 10 µm-thick FFPE sections from surgically resected specimens were used for protein extraction, while three 10 µm-thick FFPE sections from needle biopsy specimens were used for protein extraction. All specimens had been fixed in a buffered formalin solution containing 3.7% formaldehyde for an unspecified amount of time prior to paraffin-embedding; additional information on processing and storage were not provided. Protein was extracted from all specimens using a SDS protocol (sodium dodecyl sulfate) protocol that included removal of the detergent by acidification and centrifugation. Protein input quantities for TMT labeling included 200 µg or 20 µg protein. For deep proteome and phosphoproteome analyses, samples underwent TMT labeling and fractionation prior to analysis by LC-MS/MS with a boosting sample using a Q Exactive Plus instrument.
Summary of Findings:
Deep proteome analysis of samples from surgically resected NSCLC specimens with a protein input amount of 200 ug resulted in identification of 14,133 proteins while 8,798 proteins were identified in microscaled (20 µg protein input) samples; 5,966 proteins overlapped between the two protein input quantities. Deep phosphoproteome analysis of samples with an input amount of 200 µg protein resulted in 14,000 identified phosphosites, while 8,800 phosphosites were identified in samples with an input amount of 20 µg protein, with an overlap of 3,519 phosphosites between the two input amounts. Based on analysis of two replicates, the 200 µg protein samples also had a higher percentage of phosphosites that were reproducibly identified (85.7%) than the 20 µg protein sample (63.5%). The hydrophobicity index was similar among both 200 µg and 20 µg protein samples. When the number of FFPE sections (from surgically resected tumors) used for protein extraction differed (1 FFPE section vs. 6 FFPE sections) but the input amount remained the same (20 µg), similar numbers of identified proteins and phosphosites were generated by deep global proteome and phosphoproteome analyses. Average reporter ion intensities were also strongly correlated among samples with a 20 µg protein input that differed in the number of FFPE sections used for extraction (0.93 and 0.96 for ADC and SCC, respectively). To verify that 20 µg protein is an amount that can be feasibly extracted from a tumor biopsy, the authors extracted protein from three 10 µm-thick FFPE sections of needle biopsies and obtained a mean of 79 µg of protein from each biopsy FFPE (4 adenocarcinomas, 4 squamous cell carcinomas). When 20 µg protein from needle biopsies was used as the input for TMT labeling, 6,800 proteins were quantified with no missing values. For the phosphoproteome, 90% of the extracted, labelled (TMT) sample was used and 5,200 phosphopeptides were quantified. Similar numbers of proteins were quantified in TMT experiments with needle biopsies and a biopsy equivalent amount of protein (20 µg) (7,137 versus 6,792, respectively), with an overlap between proteins of 78%, despite specimens originating from different patients. Conversely, fewer phosphosites were quantified in biopsy samples than the microscaled amount (20 µg) of protein derived from surgically resected FFPE specimens (5,243 versus 8,798).
Biospecimens
Preservative Types
- Formalin
Diagnoses:
- Neoplastic - Carcinoma
Platform:
Analyte Technology Platform Peptide LC-MS or LC-MS/MS Protein LC-MS or LC-MS/MS Pre-analytical Factors:
Classification Pre-analytical Factor Value(s) Biospecimen Acquisition Method of tissue acquisition Biopsy
Surgical resection
LC-MS or LC-MS/MS Specific Template/input amount 200 µg protein
20 µg protein