| The DNA-binding domains of E. coli uracil-DNA glycosylase (Ung) and human replication protein A (hRPA) were studied using a general protocol developed in our laboratory for probing protein-DNA interactions. The procedure involves purification and mass spectrometric analysis of the nucleopeptide-products of a tryptically digested UV-crosslinked protein-nucleic acid complex.; In the case of Ung x dT20 nucleoprotein complex, three nucleopeptide isomers having the same peptide backbone (T18 peptide) but with dinucleotides attached to different aminoacids were separated. The tandem mass spectra from the isomers provided new structural information about Ung binding to DNA. Specifically, His187, Ser189, and His 194 from T18 nucleopeptide were putatively identified as sites that photocrosslink to dT20.; Photochemical crosslinking of hRPA to oligonucleotide dT30 produced two covalent hRPA70 x dT30 complexes involving one of the protein's subunits (hRPA70). Three crosslinked tryptic peptides were isolated using the same protocol as used for Ung and MALDI-TOF and nanoLC-ESI-MS/MS analyses revealed the identity of these peptides as T43, T 28/29, and a truncated *T24/25 (without the last 5 aminoacids from the C-terminal). Additional experiments showed that at least one amino acid from the sequence 383-VSDF-386 (located in T43), at least one residue from 235-ATAFNE-240 (*T24/25), and at least one residue from F269/T270 (T28/29) is involved in crosslinking. Aromatic residues contained in these peptides (F238, F269 and F386), which can form base stacking interactions with the DNA, are the residues most likely to be involved in crosslinking. These observations are in good agreement with previously published data regarding the single stranded-DNA binding site of hRPA obtained from crystal structure and from site-directed mutagenesis experiments. |
