Skip to main content

Comparing the proteome of snap frozen, RNAlater preserved, and formalin-fixed paraffin-embedded human tissue samples

Bennike TB, Kastaniegaard K, Padurariu S, Gaihedec M, Birkelund S, Andersena V, Stensballe A

Publication: EuPA Open Proteomics (2016) • Vol 10 • Page 9-18

Purpose

The paper compared proteomes and post-translational modifications (PTMs) of case-matched colon biopsy specimens preserved by snap freezing, immersion in RNAlater, and formalin fixation and paraffin embedding.  Proteomic effects associated with a 30 min delay to formalin fixation were also investigated.

Conclusion

Biopsies preserved by snap freezing in liquid nitrogen and RNAlater had a comparable number of identifiable proteins with similar mean molecular weights and isoelectric points of proteins, as well as similar ratios of modified peptides. Case-matched formalin fixed, paraffin embedded (FFPE) specimens yielded a significantly lower number of identifiable proteins (P<0.05) than snap-frozen or RNAlater preserved specimens.  While 98.7% of the proteins quantified in FFPE specimens were also present in specimens preserved by snap-freezing or with RNAlater, 23 proteins were unique to FFPE and these proteins had a higher mean molecular weight and higher isoelectric points. The number of identifiable proteins and the percentage of modified peptides did not differ between FFPE specimens fixed immediately and those subjected to a 30 min delay at room temperature.

Studies (1)

Purpose

The study compared the proteomes as well as post-translational modifications (PTMs) of case-matched, differentially preserved specimens. Twelve colon mucosal biopsies (1–2 mm3) from 2 healthy patients were either snap frozen with liquid nitrogen and stored at -80°C, preserved in RNAlater at room temperature for 24 h and then placed in -80ºC storage, or stabilized in 4% formaldehyde immediately (iFFPE) or after a 30 min delay at ambient temperature (sFFPE) followed by paraffin embedding a week later. Specimens from all four preservation methods were stored for one month prior to homogenization. The homogenized samples were incubated at 95°C (snap frozen and RNAlater preserved specimens for 10 min and formalin-fixed specimens for 60 min for de-crosslinking), measured for total protein concentration, and then digested using a modified FASP tryptic protein digestion protocol. Protein identification and PTMs were performed by LC–MS analysis using 5 mg total peptide per specimen.

Summary of Findings

The mean number of proteins identified from the biopsies preserved by snap freezing and RNAlater were comparable (3840 and 3718, respectively) and nearly identical with an overlap of 3274 proteins (96.8%); however, mean numbers were lower in specimens preserved by iFFPE and sFFPE (3384 and 3328, respectively; P < 0.05), although the difference between iFFPE and sFFPE was not statistically significant (P > 0.05). Principal component analysis revealed that 5.9% (202) of proteins identified in both snap frozen and RNAlater-stabilized samples were not recovered in biopsies preserved by iFFPE and only 23 proteins were unique to iFFPE specimens. Mean molecular weight and isoelectric point of proteins were comparable for all preservation methods; however, the 23 unique proteins found in iFFPE biopsies were of higher mean molecular weight but with an increased variance. Analysis of peptide modifications revealed lower ratios of modified peptides in snap frozen and RNAlater preserved specimens (44.5% and 44.0%, respectively) than iFFPE and sFFPE specimens (49.3% and 50.1%, respectively). As expected, iFFPE and sFFPE specimens demonstrated increased ratios of formylated N-terminals (6.4% and 7.3%, respectively) compared to snap frozen specimens (5.1%) and RNAlater specimens (2.8%) as well as formylated lysines (3.7% and 3.3% versus 0.4% and 0.2%), but ratios less than 10% suggest de-crosslinking was effective. iFFPE and sFFPE specimens also revealed an increase in methylated, deamidated, and alkylated peptides. Analysis of subcellular location of proteins revealed minor biases in enrichment for ER and nucleus proteins in iFFPE biopsies compared to snap frozen and RNAlater-preserved biopsies. Scatterplot analysis of all quantifiable proteins revealed good correlation for all preservation methods with the lowest correlation between snap frozen and sFFPE biopsies (r=0.943) and the highest correlations between the snap frozen and RNAlater biopsies (r=0.987) and iFFPE and sFFPE biopsies (r=0.983).

  • Biospecimens

    TissueColorectal
  • Preservation Types

    FormalinFrozenRNAlater
  • Diagnoses

    Normal
  • Platforms

    PeptideLC-ESI-MS/MS
    ProteinLC-ESI-MS/MS
  • Pre-analytical Factors

    ClassificationPre-Analytical FactorValue(s)
    Biospecimen AcquisitionCold ischemia time
    0 min
    30 min
    Biospecimen PreservationTemperature of fixation/preservation
    Fixation in 4% formalin delayed by 30 min
    Immediate fixation in 4% formalin
    RNA-later
    Snap frozen