The program is concerned with proteomic research including the analysis, synthesis, protein expression, and structure-function relationships of peptide and protein hormones and their associated proteins. Development of mass spectrometric analysis for the identification of serine and threonine phosphorylation sites (P-sites), identification of P-sites in human angiotensin II (Ang II) type 1 receptor (hAT1), over-expression of protein corresponding to the carboxyl-terminal domain (residues 268-359) of hAT1, identification of an anti-HIV active site in human lysozyme, cloning and sequencing of pigeon egg white protein cDNAs to deduce their primary protein sequences, and characterization of expressed yeast ribonuclease H2 were current focuses as described below: (1) Protein phosphorylation on Ser, Thr and Tyr plays a major role in the regulation of eukaryotic cell function. Characterization of the P-sites is critical in order to understand how these modifications modulate the biological activities of proteins. Mass spectrometry (MS) is one of the most advanced methods for analyzing protein P-sites because of its sensitivity, accuracy, and selectivity. Tandem MS/MS identification of P-sites directly using phosphopeptides as precursors is complicated by the loss of phosphate moiety during low energy collision induced dissociation (CID) in most cases. To improve the ionization of phosphopeptides, methodologies based on beta?elimination and Michael addition to modify the P-Ser and P-Thr into CID stable derivatives have been developed. The same reactions can also be employed to liberate O-linked glycans from glycopeptides and used to identify the exact O-glycosylation sites by MS/MS. However, under commonly employed conditions, i.e., when using 0.2-0.5N NaOH for the modification, un-expected beta?elimination on un-modified Ser or Thr occurs. The additional modification site(s) may lead to an erroneous conclusion. The need of an optimal modification condition that could both obviate side reaction, and permit accurate assignment of the P-sites is obvious. A barium hydroxide catalyzed reaction to eliminate O-linked phosphate or glycans on Ser or Thr followed by addition of propanethiol or butanethiol as nucleophile was studied. An optimum condition that included 20 mM barium hydroxide, 30% n-propanol, and 0.5M alkanethiols incubated for 24 hours at 25 degree C was established for P-Ser and P-Thr peptides. The resulted n-propylthio and n-butylthio derivatives not only were stable during CID but also yielded at least seven times more positive mode ionization than the parent molecule when analyzed by matrix-assisted laser desorption ionization (MALDI) time-of-flight (TOF) MS. Furthermore, abundant y and b fragment ions were easily identifiable under general conditions for ESI tandem MS/MS and MALDI-TOF/TOF MS. This procedure was applied to determine accurately the P-sites in bovine milk beta-casein that contains monophospho-hexadecapeptide FQpSEEQQQTEDELQDK and tetraphospho-pentacosapeptide RELEELNVPGEIVEpSLpSpSpSEESITR. However, the carbohydrate moiety in either synthetic O-GalNAc-Ser or -Thr glycopeptide was only slightly eliminated under the same condition indicating that barium hydroxide could catalyze beta-elimination efficiently on O-phospho- but not on O-GalNAc-Ser and ?Thr. Thus this condition allows for distinction between O-phospho- and O-glycan sites. For the identification of P-sites in hAT1, tryptic fragments of solubilized proteins from Ang II stimulated human adrenal cortical carcinoma cells (H295R) were obtained. Phosphopeptides from the tryptic digest will be isolated by immobilized metal (Fe3+) affinity chromatography, and then submitted to the beta-elimination/propanethiol addition followed by nanoflow HPLC on an ion-trap MS and MS/MS as described above. (2) The cytoplasmic carboxyl-terminal sequence of G protein-coupled receptors plays a crucial role in receptor internalization, desensitization, phosphorylation and recycling. A partial segment of carboxyl-terminal cytoplasmic domain (residues 268-359) of hAT1 has been over-expressed in E. coli and purified, which will be phosphorylated and used as a bait to isolate AT1 associated proteins in H295R cells. (3) Our previous studies have indicated that lysozyme associated with the beta-core fraction of human chorionic gonadotropin is accountable for anti-HIV activity in the urine extracts of pregnant women. To define the structural and sequence requirements for anti-HIV activity, we have identified two peptides that consist of residues 98-115 of human lysozyme and its C-terminal 9 residue segment RAWVAWRNR (designated HL18 and HL9, respectively). Both HL18 and HL9 are potent inhibitors of HIV-1 infection and replication with EC50 of 50 to 55 nM, a level comparable to that of intact lysozyme. The importance of Arg and Trp at certain positions in the peptide was demonstrated by loss of the anti-HIV activity by scrambling or substitution of each or a combination at residues 107, 113 or 115, and 109, 112, respectively. Although HL9 exists as an alpha-helix and forms a pocket with its basic residues on the surface of the lysozyme molecule, it does not display muramidase activity because it is located outside of the muramidase catalytic site in human lysozyme. Monte Carlo peptide folding energy minimizing simulation modeling and CD studies indicate that HL9 sequence has high alpha-helical propensity. However, several mutants that also show high alpha-helical propensity were inactive, suggesting that alpha-helical conformation may be required but not sufficient for the activity. Amphiphilicity, basicity and charge distribution may contribute to the anti-HIV activity. Our findings indicate that HL18 and HL9 are active against HIV-1 and may lead to new strategies for the treatment of HIV-1 and other viral infections. (4) Previously, we purified and partially sequenced four components from pigeon egg white glycoproteins and identified them as ovomucoid, ovotransferrin and two ovalbumins (high and low molecular weight forms). The major components of the two albumins gave a mass of 48,551.6 and 53,129.7, respectively as revealed by MALDI-MS. Only the low molecular weight albumin has been found in most of avian, despite the fact that chickens contain multiple ovalbumin genes. Based on the amino acid sequences from fragments obtained from each of the four proteins, full-length cDNA clones encoding the four proteins were obtained by screening a pigeon oviduct library. The four cDNAs encode proteins of the following lengths of amino acid sequence with identity/similarity to the respective chicken counterparts as shown in parenthesis: ovalbumin-high (pigeon 388 vs. chicken 386 residues; 58.29/68.91% of chicken), ovalbumin-low (386 vs. 386; 66.58/75.65), ovomucoid (209 vs. 210; 70.53/60.38), ovotransferrin (706 vs. 705; 78.01/83.40). Aside from multiple forms of ovalbumins in pigeon, dissimilarity in protein sequences of egg white glycoproteins between pigeon and chicken is significant and may lead a way to understand the evolution of avian in general and that of related glycoproteins in particular. (5) Saccharomyces cerevisiae Rnh2Ap, a homologue of ribonuclease (RNase) H2 in bacteria was expressed in E. coli. The expressed protein was inactive as a single chain. To rule out that the inactivity was due to an error in expression or isolation of the target protein, N-terminal sequence analysis by Edman degradation method and tandem LC-MS/MS analyses of its tryptic fragments were carried out, and the expressed protein of interest was verified as Rnh2Ap. This finding led to the elucidation of the complex nature of RNase H2 of Saccharomyces cerevisiae. Two more proteins, namely Ydr279p and Ylr154p, together with Rnh2Ap, are necessary and sufficient for RNase H2 activity.
