Overlap of Formalin-Fixed Paraffin-Embedded and Fresh-Frozen Matched Tissues for Proteomics and Phosphoproteomics

Abstract

Many liquid chromatography-mass spectrometry (LC-MS) studies have compared formalin-fixed paraffin-embedded (FFPE) tissues with matched fresh-frozen (FF) tissues to examine the effect of preservation techniques on the proteome; however, few studies have included the phosphoproteome. A high degree of overlap and correlation between the two preservation techniques would demonstrate the importance of FFPE tissues as a valuable biomedical resource. Our aim was to quantitatively compare the proteome and phosphoproteome of matched FFPE and FF tissues using data-independent acquisition LC-MS. Four organs from three rats were cut in half to produce matched FFPE and FF tissue pairs. Excellent overlaps of 85-97% for the proteome and 82-98% for the phosphoproteome were observed, depending on the organ type, between the two preservation techniques. Most of the unique identifications were found in FF with less than 0.3% being unique to FFPE tissues. Strong agreement between FFPE and FF matched tissue pairs was observed with Pearson correlation coefficients of 0.93-0.97 and 0.79-0.87 for the proteome and phosphoproteome, respectively. Digestion efficiency was slightly higher in FFPE (92-94%) than in FF tissues (86-89%), and a search of a data subset for formaldehyde induced chemical modifications revealed that only 0.05% of precursors were unique to FFPE tissues. This suggests that with quality sample preparation methods it is not necessary to include formaldehyde induced chemical modifications when analyzing FFPE tissues. We attribute the lower number of identifications in FFPE tissues to inaccurate peptide quantitation, which resulted in a lower MS peptide load and tryptic peptide enrichment load. Our results demonstrate that both proteomic and phosphoproteomic analyses of FFPE and FF tissues are highly comparable and highlight the suitability of FFPE tissues for both proteomic and phosphoproteomic analysis.

Figure 1

Schematic representation of the experimental design. Top panel: Four rat organs (brain, kidney, liver, and testes) were collected from three rats and halved using a scalpel. One half organ was processed into FFPE tissue blocks while the other half of the same organ was frozen. For FFPE tissues, eight sequential 30 μm sections were cut from each of the 12 FFPE blocks. For FF tissues, eight tissue punches were taken for each of the 12 FF tissues. Middle panel: The 192 tissues (96 FFPE sections and 96 FF punches) were organized into 12 batches for proteomic sample preparation. After quantitation, the peptides from replicate sections/punches of the same FFPE block/FF tissue were pooled. Bottom panel: Triplicate enrichments of 200 μg tryptic peptide were enriched for phosphopeptides on MagResSyn zirconium and titanium IMAC paramagnetic beads for each FFPE and FF pooled tissue. A quarter of the first enrichment and the entire second enrichment was analyzed by MS. Created with BioRender.com.

Figure 2

Comparison of the proteomes of rat organs preserved by FFPE vs FF in the rat brain, kidney, liver, and testis. (A) The number of protein groups quantified between organs and preservation technique. (B) Ratio of quantified precursor intensity divided by the number of precursors between organ and tissue preservation technique. (C) Hierarchical clustering of proteomes, grouping first by organ, then preservation technique, then by individual rats. (D) Overlap of protein groups quantified in FFPE vs FF tissues in each of the four organs.

Figure 3

Pearson correlation coefficients of normalized rat organ proteomes (left) and phosphoproteomes (right) with log 2 transformation. Proteomes were acquired using 200 ng of peptide loads, and correlations were calculated at the protein group level. Phosphoproteomes were acquired using enrichments of 50 μg of tryptic peptide, and phosphoprecursor values were normalized before replicate enrichments were averaged.

Figure 4

Comparison of the phosphoproteomes of FFPE and FF between rat brain, kidney, liver, and testis. (A) Phosphoprecursors quantified between organ and preservation technique. (B) Ratio of phosphoprecursor intensity divided by the number of phosphoprecursors between organ and preservation technique. (C) Hierarchical clustering of phosphoproteomes, grouping first by organ, then preservation technique, and then individual rats. (D) Overlap of phosphoprecursors quantified in FFPE and FF tissues in each of the four organs.