Mitogen‐activated protein kinase (MAP kinase) is a 42 kd serine/threonine protein kinase whose enzymatic activity requires phosphorylation of both tyrosyl and threonyl residues. As a step in elucidating the mechanism(s) for activation of this enzyme, we have determined the sites of regulatory phosphorylation. Following proteolytic digestion of 32P‐labeled pp42/MAP kinase with trypsin, only a single phosphopeptide was detected by two‐dimensional peptide mapping, and this peptide contained both phosphotyrosine and phosphothreonine. The amino acid sequence of the peptide, including the phosphorylation sites, was determined using a combination of Fourier transform mass spectrometry and collision‐activated dissociation tandem mass spectrometry with electrospray ionization. The sequence for the pp42/MAP kinase tryptic phosphopeptide is similar (but not identical) to a sequence present in the ERK1‐ and KSS1‐encoded kinases. The two phosphorylation sites are separated by only a single residue. The regulation of activity by dual phosphorylations at closely spaced threonyl and tyrosyl residues has a functional correlate in p34cdc2, and may be characteristic of a family of protein kinases regulating cell cycle transitions.

A new ion desorption method is described that utilizes a primary beam of massive, multiply charged cluster ions to generate secondary ions of peptides in a glycerol matrix. The massive cluster ion beam is generated via electrohydrodynamic emission using a 1.5 M solution of ammonium acetate in 30% aqueous glycerol. Negative ion spectra of peptides obtained using this technique show greatly decreased relative intensities for fragment ions and 'chemical noise' background when compared to spectra obtained using a xenon atom primary beam. The near absence of fragments derived from radiation damage to the sample solution is attributed to the impact of primary particles with energies less than 1 eV/nucleon.

A shock wave model is proposed to explain certain features of recently reported spectra obtained by massive duster impact (MCI) mass spectrometry. It is suggested that clusters that impact glycerol matrices with energies/nucleon in the range 0.01 eV/u < E/N < 1.0 eV/u provide an extremely soft method for sputtering intact biomolecules, Compared to the high energy/nucleon characteristic of atomic or molecular ion primary beams (typically < 50 eV/u), massive cluster primary beams possess much lower energies/nucleon, which are insufficient to cause appreciable ionization and radiation damage of matrix material. Moreover, fragmentation products of parent molecular ions are effectively lower. With these benefits, MCI spectra show lower chemical noise background and enhanced signalto-noise ratios. Rankine-Hugoniot analysis of the shock conditions is used to arrive at an estimate of the heat retained in the collision-affected matrix volume after bombardment by a characteristic cluster. For a cluster collision resulting in a 26.8 GPa shock pressure, by analogy with water data, rapid heating of the shocked volume to 1000 °C or more is plausible. In a beam consisting of clusters distributed in size and charge, an estimate is made for the range of cluster sizes over which hyrodynamic shock wave theory applies.

We have compiled evidence that nonmuscle isoforms of both myosin heavy chain (NM MHC) and myosin regulatory light chain (NM LC20) are present in fully differentiated smooth muscles (SM). In swine carotid media sodium dodecyl sulfate-gel electrophoresis separated three MHC bands. The upper two bands were identified by immunoblotting as SM-specific isoforms. The lowest MHC band amounted to 14 +/- 2% of the total MHC and was electrophoretically and antigenically similar to platelet MHC. Two-dimensional gel electrophoresis of swine carotid media extracts resolved multiple LC20 species, including phosphorylated and "satellite" forms. Mass spectrometric analysis of tryptic peptides from blots of these gels demonstrated two LC20 isoforms. The measured peptide masses correspond with two published cDNA sequences proposed to represent SM and NM LC20 isoforms. These sequences readily explain the electrophoretic behavior of the isoforms. The minor isoform's abundance (16 +/- 3%, corresponding to NM MHC), antigenic properties, and pattern of expression in tissue culture all confirm that this is a NM LC20 isoform. The localization and functional significance of NM myosin in smooth muscle is unknown.

The amino acid sequence of mouse liver NAD(P)H:quinone acceptor oxidoreductase (EC 1.6.99.2) has been determined by tandem mass spectrometry and deduced from the nucleotide sequence of the cDNA encoding for the enzyme. The electrospray mass spectral analyses revealed, as previously reported (Prochaska HJ, Talalay P, 1986, J Biol Chem 261: [1372][1373][1374][1375][1376][1377][1378], that the 2 forms -the hydrophilic and hydrophobic forms-of the mouse liver quinone reductase have the same molecular weight. No amino acid sequence differences were found by tandem mass spectral analyses of tryptic peptides of the 2 forms. Moreover, the amino-termini of the mouse enzymes are acetylated as determined by tandem mass spectrometry. Further, only 1 cDNA species encoding for the quinone reductase was found. These results suggest that the 2 forms of the mouse quinone reductase have the same primary sequences, and that any difference between the 2 forms may be attributed to a labile posttranslational modification.Analysis of the mouse quinone reductase cDNA revealed that the enzyme is 273 amino acids long and has a sequence homologous to those of rat and human quinone reductases. In this study, the mouse quinone reductase cDNA was also ligated into a prokaryotic expression plasmid pKK233.2, and the constructed plasmid was used to transform Escherichia coli strain JM109. The E. coli-expressed mouse quinone reductase was purified and characterized. Although mouse quinone reductase has an amino acid sequence similar to those of the rat and human enzymes, the mouse enzyme has a higher NAD(P)H-menadione reductase activity and is less sensitive to flavones and dicoumarol, 2 known inhibitors of the enzyme. The results would indicate that the regions in mouse quinone reductase that contain amino acids different from the rat and human enzymes are critical for the binding of menadione, flavones, and dicoumarol.