A CE-MALDI interface based on the use of prestructured sample supports.

We have developed an off-line coupling of capillary electrophoresis (CE) to matrix-assisted laser desorption/ionization time-of-flight mass spectrometry(MALDI-TOF-MS) based on CE fraction collection onto prestructured MALDI sample supports. Analyte carryover and detection sensitivity were investigated using a standard peptide mixture. Low femtomole amounts were detected, and no noticeable carryover was discovered. The performance of the method was evaluated with a mixture of tryptic digests of proteins from a human fetal brain cDNA expression library. The total number of identified peptides was increased from 47 to 211 when the CE-MALDI interface was used compared to direct MALDI-MS analysis. Sequence coverage with CE-MALDI was in the 25-60% range for the different proteins, corresponding to an increase of 1.3-4.9 times relative to that obtained with MALDI-MS of the crude mixture. Fractionation of sample components also facilitated protein identification by MALDI postsource decay analysis. Our initial results suggest this CE-MALDI interface can be used for the analysis of complex peptide mixtures isolated from biological tissues.

Complex polar lipids of a hot spring cyanobacterial mat and its cultivated inhabitants.

The complex polar lipids of the hot spring cyanobacterial mat in the 50 to 55 degrees C region of Octopus Spring, Yellowstone National Park, and of thermophilic bacteria cultivated from this or similar habitats, were compared in an attempt to understand the microbial sources of the major lipid biomarkers in this community. Intact complex lipids were analyzed directly by fast atom bombardment mass spectrometry (FAB-MS), two-dimensional thin-layer chromatography (TLC), and combined TLC-FAB-MS. FAB-MS and TLC gave qualitatively similar results, suggesting that the mat contains major lipids most like those of the cyanobacterial isolate we studied, Synechococcus sp. strain Y-7c-s. These include monoglycosyl, diglycosyl, and sulfoquinosovyl diglycerides (MG, DG, and SQ, respectively) and phosphatidyl glycerol (PG). Though Chloroflexus aurantiacus also contains MG, DG, and PG, the fatty acid chain lengths of mat MGs, DGs, and PGs resemble more those of cyanobacterial than green nonsulfur bacterial lipids. FAB-MS spectra of the lipids of nonphototrophic bacterial isolates were distinctively different from those of the mat and phototrophic isolates. The lipids of these nonphototrophic isolates were not detected in the mat, but most could be detected when added to mat samples. The mat also contains major glycolipids and aminophospholipids of unknown structure and origin. FAB-MS and TLC did not always give quantitatively similar results. In particular, PG and SQ may give disproportionately high FAB-MS responses.

Phospholipid-hydroperoxide glutathione peroxidase. Genomic DNA, cDNA, and deduced amino acid sequence.

The complete amino acid sequence of the selenoprotein phospholipid-hydroperoxide glutathione peroxidase (PHGPX) from pig heart has been deduced from the corresponding genomic DNA, the cDNA covering the coding region, and by sequencing the N terminus of the protein. The maximum length of the peptide chain derived from the cDNA amounts to 170 amino acid residues. By protein sequencing the N-terminal residues methionine and cysteine of the deduced sequence were found to be cleaved. The molecular mass of 19,671 Da obtained by laser desorption mass spectroscopy, however, significantly exceeds the mean molecular mass of 19,257.09 calculated for the sequence 3-170 of PHGPX, thus indicating posttranscriptional modification. In contrast to glutathione peroxidase (GPX) the coding area of the PHGPX gene is composed of seven exons. Only the amino acid sequences encoded by the third and fifth exon are highly homologous to GPX sequences. The amino acid residues selenocysteine, tryptophan, and glutamine forming the catalytic site in bovine GPX are conserved in homologous positions of PHGPX, whereas the arginine residues presumed to bind GSH in GPX are not. Gaps in the PHGPX sequence correspond to subunit interaction sites of the tetrameric GPX. The data suggest an identical catalytic mechanism of the selenoperoxidases, a less stringent substrate specificity of PHGPX, and explain the monomeric nature of PHGPX. As in other selenoproteins, the selenocysteine residue of PHGPX is encoded by UGA. The 3'-untranslated region (UTR) of the PHGPX shows a limited consensus with that of GPX and 5'-deiodinase, where it was shown to be responsible for the decoding of UGA as selenocysteine. The 3'-UTR of PHGPX can form a stem/loop as in other mammalian selenoprotein genes. The 5'-UTR and the first intron of the PHGPX gene contain a variety of putative regulatory elements indicating hormonal control.

Structures of sialylated oligosaccharides of human erythropoietin expressed in recombinant BHK-21 cells.

The native structures of the Asn-linked oligosaccharides and the O-glycans at Ser126 of human erythropoietin expressed from recombinant BHK cells have been elucidated. Enzymatically released N-glycans were studied by methylation analyses, fast-atom-bombardment mass spectrometry as well as one- and two-dimensional 1H-NMR spectrometry at 600 MHz. Many (82.7%) were found to be tetraantennary N-acetyllactosamine-type (22.8% with one, 3.6% with two and 0.4% with three N-acetyllactosamine repeats) being tetrasialylated (41%), trisialylated (29.6%) and disialylated (12.2%). A few (9.7%; 4.1% 2,4-branched, 5.6%, 2,6-branched) of the chains were triantennary (5.4% trisialyl, 4.3% disialyl) and 4.6% were of the disialyl diantennary type. Almost all of the innermost GlcNAc residues were alpha 1-6 fucosylated and NeuAc was exclusively alpha 2-3 linked to Gal beta 1-4GlcNAc-R; 60% of the protein was found to be O-glycosylated at Ser126; structures were monosialylated (70%) or disialylated (30%) forms of the Gal beta 1-3GalNAc core type. Glycosylation patterns at individual Asn-Xaa-Thr/Ser sites were determined by analytical high-pH anion-exchange chromatography with pulsed amperometric detection. Only tetraantennary chains with 0-3 N-acetyllactosamine repeats were detected at Asn38 and Asn83, while almost all of the di- and triantennary oligosaccharides were attached to Asn24. Batch analysis of different preparations of recombinant erythropoietin revealed the high reproducibility of the production procedure. Structures containing terminal GalNAc-GlcNAc were detected in small amounts in a few batches.

Synthesis and characterization of the hydrophilic C-terminal domain of the human immunodeficiency virus type 1-encoded virus protein U (Vpu).

The HIV-1-specific virus protein U (Vpu) is an amphipathic membrane-associated phosphoprotein that probably possesses an N-terminal hydrophobic membrane anchor and a hydrophilic moiety. Because of an as yet undefined cytotoxic effect and the low concentration of Vpu in host cells, it has not been possible to obtain sufficient quantities of this protein for biochemical and structural investigations. We describe the synthesis, in two forms, of 50-residue peptides representing the polar cytoplasmic domain of Vpu: pVpu, comprising the wild-type Vpu sequence of HIV-1, strain HTLVIIIB, from position Ile32 to Leu81, and a mutant, pVpum2/6, with replacement of Ser52-56 with Asn. This mutation affects the two phosphorylation sites of Vpu for casein kinase-2 (CK-2), the enzyme that phosphorylates Vpu in HIV-1-infected cells. The peptides were purified to homogeneity and characterized by N-terminal sequencing, mass spectrometry and SDS-PAGE. Furthermore, both peptides were immunologically characterized by Western blot and ELISA using Vpu-specific monoclonal and polyclonal antibodies. Recombinant human CK-2 caused in vitro phosphorylation of pVpu, but had no effect on the mutant pVpum2/6. Investigation of the peptides by circular dichroism showed that addition of trifluoroethanol stabilized the alpha-helical secondary structure. Preliminary 1H NMR spectroscopic data indicated that, in the presence of trifluoroethanol, both peptides in solution have stable secondary structures with considerable alpha-helical content.

Class II tubulin, the major brain beta tubulin isotype is polyglutamylated on glutamic acid residue 435.

Protein sequencing shows that porcine brain tubulin retains the N-terminal sequences of alpha and beta tubulin after a mild treatment with subtilisin. C-terminal peptides released by subtilisin were purified and characterized by automated Edman degradation and mass spectrometry. We confirm the polyglutamylation of alpha tubulin on glutamic acid residue 445 reported by others and show in addition that class II beta tubulin, the major beta tubulin isotype of adult brain, is also polyglutamylated. The substitution is restricted to glutamic acid residue 435. Thus all major tubulin isotypes of adult brain are subjected to polyglutamylation.

Proteinase-catalyzed conversion of a substance P-precursor peptide.

The protease-catalyzed conversion of peptides and proteins produced by recombinant DNA technology is a promising method for large-scale production of peptides including those with non-proteinogenic structural elements. As a model system we have investigated the proteinase-catalyzed modification of a chemically synthesized substance P-precursor. In the precursor peptide the residues of substance P(1-8) were flanked by tripeptide linkers on both sides. In the first step the C-terminal tripeptide amide was replaced by the authentic C-terminal tripeptide amide of substance P via alpha-chymotrypsin-catalyzed transpeptidation. The enzyme simultaneously attacks two peptide bonds in the precursor molecule leading to the formation of several side-products. The desired peptide was obtained with 25% yield. In the second step the other tripeptide linker was selectively and almost quantitatively removed from the N-terminus of the precursor via trypsin-catalyzed hydrolysis. This study demonstrates that substance P can be obtained from an engineered protein by proteinase-catalyzed processing.

Fast atom bombardment mass spectrometry as a rapid means of screening mixtures of ether-linked polar lipids from extremely halophilic archaebacteria for the presence of novel chemical structures.

Fast atom bombardment mass spectrometry was used to analyse intact polar ether lipids present at microgram levels in crude lipid mixtures extracted from Halobacterium halobium, Natronococcus occultus and Halobacterium marismortui. Negative-ion spectra showed the intact deprotonated lipid molecules and in some instances their sodium salts. The simplicity of the mass spectra permits the rapid screening of polar lipid mixtures for the presence of novel lipids. Additional structural information of ions with selected masses was obtained after collisionally induced decomposition.

Detoxification of the macrolide toxin brefeldin A by Bacillus subtilis.

The macrolide toxin brefeldin A is a determinant of Alternaria leaf blight disease in safflower, which causes severe economic losses worldwide. Soilborne bacteria, classified as Bacillus subtilis spp., were isolated and shown to readily metabolize brefeldin A in laboratory culture to one major product. This product was identified by high resolution 2D 1H NMR and FAB mass spectroscopies as the acid resulting from hydrolysis of the macrolide ring in brefeldin A . In contrast to brefeldin A, the acid completely lacked phytotoxic activity in the standard leaf bioassay. Detoxification of brefeldin A by the lactonase activity from Bacillus subtilis may be exploited in the future to introduce resistance to Alternaria leaf blight in safflower.