Most published proteomics studies of bulk plant tissues use a procedure in which proteins are precipitated with trichloroacetic acid (TCA) and acetone (TCA-A), but few attempts have been made to contrast this approach in a systematic way with alternative methods against a spectrum of tissues. To address this, TCA-A was compared with another acetone-based protocol (TCA-B) or a phenol (Phe)-based method, targeting a range of tomato tissues and three species of fruits that contain high levels of contaminating compounds: banana, avocado and orange. The Phe method gave a higher protein yield and typically greater resolution and spot intensity, particularly with extracts from tissues containing high levels of soluble polysaccharides. The methods also generated remarkably different two-dimensional gel electrophoresis (2-DE) protein spot patterns. Peptide mass fingerprinting was used to identify polypeptides that were common to multiple extracts or uniquely present in one extract type. While no clear pattern emerged to explain the basis for the differential protein extraction, it was noted that the Phe method showed enhanced extraction of glycoproteins. These results suggest that the Phe protocol is highly effective with more recalcitrant tissues and that a combination of TCA-A and Phe methods provides enhanced 2-DE based proteomic analyses of most plant tissues.