Simplified methodology has been developed for the direct N-terminal amino acid microsequencing of human liver and hepatoma-derived polypeptides using micropreparative immobilized pH gradient two-dimensional polyacrylamide gel electrophoresis (IPG 2D-PAGE). Utilization of IPG gel strips in the first dimension permitted protein loading concentrations of 0.5-2.0 mg with negligible diminution of polypeptide resolution. Following 2D separation and electrotransfer to PVDF membranes, nearly 100 well resolved Ponceau S stained polypeptides were readily visualized on respective blotted membranes, from which 32 adult liver S-9 and 72 HepG2 nuclear cytosolic polypeptides were subjected to N-terminal microsequencing. Twenty normal adult liver and 54 HepG2 polypeptides yielded N-terminal sequence information, of which 17 and 19 polypeptides, respectively, exhibited high sequence homology to previously identified proteins. The initial yields of the proteins sequenced ranged from 2-14 pmols and yielded sequences of 14-26 amino acid residues. Many of the adult liver and HepG2 proteins contained inferred leader sequences since the first sequenced residue was several (20-30) residues from the methionine initiation site predicted by the cDNA. Comparison of approximately 1000 silver stained whole cell lysate and purified nuclear proteins between normal adult liver, two nontransformed cell lines [Chang (adult) and WRL-68 (embryonic)], and four human hepatoma- derived cell lines (HepG2, Huh-7, FOCUS, and SK-Hep) revealed significant qualitative and quantitative differences in polypeptide expression. Chang and WRL-68 liver cells, whose 2D polypeptide patterns were almost completely superimposable, most resembled normal liver, while marked differences in polypeptide expression were observed between normal liver and each of the hepatoma-derived cell lines (HepG2, FOCUS, Huh-7 and SK-Hep). Preparative IPG 2D-PAGE in combination with protein microsequencing provides a convenient one-step procedure to rapidly obtain partial amino acid sequence information for nearly 100 individual polypeptides directly from a single 2D-PAGE gel. Comparison of partial amino acid sequences with existing protein and nucleic acid sequence databases provide a rapid and convenient method of protein identification in the absence of specific antibody preparations as well as to suggest possible biological function(s) for as yet unidentified proteins.
