High-throughput proteomic studies of archival formalin-fixed paraffin-embedded (FFPE) tissues have the potential to be a powerful tool for examining the clinical course of disease. However, advances in FFPE tissue-based proteomics have been hampered by inefficient methods to extract proteins from archival tissue and by an incomplete knowledge of formaldehyde-induced modifications in proteins. To help address these problems, we have developed a procedure for the formation of 'tissue surrogates' to model FFPE tissues. Cytoplasmic proteins, such as lysozyme or ribonuclease A, at concentrations approaching the protein content in whole cells, are fixed with 10% formalin to form gelatin-like plugs. These plugs have sufficient physical integrity to be processed through graded alcohols, xylene, and embedded in paraffin according to standard histological procedures. In this study, we used tissue surrogates formed from one or two proteins to evaluate extraction protocols for their ability to quantitatively extract proteins from the surrogates. Optimal protein extraction was obtained using a combination of heat, a detergent, and a protein denaturant. The addition of a reducing agent did not improve protein recovery; however, recovery varied significantly with pH. Protein extraction of >80% was observed for pH 4 buffers containing 2% (w/v) sodium dodecyl sulfate (SDS) when heated at 100 degrees C for 20 min, followed by incubation at 60 degrees C for 2 h. SDS-polyacrylamide gel electrophoresis of the extracted proteins revealed that the surrogate extracts contained a mixture of monomeric and multimeric proteins, regardless of the extraction protocol employed. Additionally, protein extracts from surrogates containing carbonic anhydrase:lysozyme (1:2 mol/mol) had disproportionate percentages of lysozyme, indicating that selective protein extraction in complex multiprotein systems may be a concern in proteomic studies of FFPE tissues.