Enhanced activity and altered specificity of phospholipase A2 by deletion of a surface loop.

Protein engineering and x-ray crystallography have been used to study the role of a surface loop that is present in pancreatic phospholipases but is absent in snake venom phospholipases. Removal of residues 62 to 66 from porcine pancreatic phospholipase A2 does not change the binding constant for micelles significantly, but it improves catalytic activity up to 16 times on micellar (zwitterionic) lecithin substrates. In contrast, the decrease in activity on negatively charged substrates is greater than fourfold. A crystallographic study of the mutant enzyme shows that the region of the deletion has a well-defined structure that differs from the structure of the wild-type enzyme. No structural changes in the active site of the enzyme were detected.

Crystallization and preliminary X-ray study of a recombinant cutinase from Fusarium solani pisi.

Recombinant cutinase from Fusarium solani pisi is expressed and excreted with very high yields in Escherichia coli cultures. Cutinase was crystallized at 20 degrees C using the vapour diffusion technique, with polyethylene glycol 6000 as precipitant. Best crystals were obtained at pH 7.0 with polyethylene glycol 6000 as precipitant. Best crystals were obtained at pH 7.0 with polyethylene glycol at 15 to 20%. They are monoclinic, with space group P2(1) and cell dimensions a = 35.1 A, b = 67.4 A, c = 37.05 A and beta = 94.0 degrees; they diffract beyond 1.5 A resolution. The asymmetric unit contains one molecule of 22,000 Da (Vm = 1.98 A3/Da; 38% water).

Fusarium solani cutinase is a lipolytic enzyme with a catalytic serine accessible to solvent.

Lipases belong to a class of esterases whose activity on triglycerides is greatly enhanced at lipid-water interfaces. This phenomenon, called interfacial activation, has a structural explanation: a hydrophobic lid, which at rest covers the catalytic site, is displaced on substrate or inhibitor binding and probably interacts with the lipid matrix. Fusarium solani pisi cutinase belongs to a group of homologous enzymes of relative molecular mass 22-25K (ref. 7) capable of degrading cutin, the insoluble lipid-polyester matrix covering the surface of plants, and hydrolysing triglycerides. Cutinases differ from classical lipases in that they do not exhibit interfacial activation; they are active on soluble as well as on emulsified triglycerides. Cutinases therefore establish a bridge between esterases and lipases. We report here the three-dimensional structure of a recombinant cutinase from F. solani pisi, expressed in Escherichia coli. Cutinase is an alpha-beta protein; the active site is composed of the triad Ser 120, His 188 and Asp 175. Unlike other lipases, the catalytic serine is not buried under surface loops, but is accessible to solvent. This could explain why cutinase does not display interfacial activation.

Possible active site of the sweet-tasting protein thaumatin.

Epitopes on thaumatin and monellin were studied using the PEPSCAN-technology. The antibodies used were raised against thaumatin. Only antibodies that, in an ELISA, both recognized thaumatin and monellin were used in the PEPSCAN-analyses. On thaumatin two major overlapping epitopes were identified. On monellin no epitopes could be identified. The identified epitope region on thaumatin shares structural features with various peptide and protein sweeteners. It contains an aspartame-like site which is formed by Asp21 and Phe80, tips of the two extruding loops KGDAALDAGGR19-29 and CKRFGRPP77-84, which are spatially positioned next to each other. Furthermore, sub-sequences of the KGDAALDAGGR19-29 loop are similar to peptide-sweeteners such as L-Asp-D-Ala-L-Ala-methyl ester and L-Asp-D-Ala-Gly-methyl ester. Since the aspartame-like Asp21-Phe80 site and the peptide-sweetener-like sequences are also not present in non-sweet thaumatin-like proteins it is postulated that the KGDAALDAGGR19-29- and CKRFGRPP77-84 loop contain important sweet-taste determinants. This region has previously not been implicated as a sweet-taste determinant of thaumatin.

Molecular cloning of pldA, the structural gene for outer membrane phospholipase of E. coli K12.

The pldA gene of Escherichia coli K12, which is involved in the synthesis of an outer membrane (OM) phospholipase, has been cloned using a cosmid cloning system. For detection of the cloned gene a newly developed, in vivo phospholipase assay was used. Subsequent cloning of the pldA gene was performed into the multicopy plasmid vectors pBR322 and pACYC184. The gene was localised on these hybrid plasmids by the analysis of in vitro-constructed deletion plasmids and mutant plasmids generated by transposon gamma delta-insertions. Analysis of plasmid-encoded proteins in a minicell system showed that the pldA gene product is a polypeptide with apparent molecular weight of 29,000. This apparent molecular weight changes from 29,000 to 26,000 when the denaturing temperature is changed from 95 degrees C to 37 degrees C. These data are in agreement with those on purified OM phospholipase (Nishijima et al. 1977), and therefore strongly suggest that pldA is the structural gene for this phospholipase. From the minicell experiments the direction of transcription of pldA could be established relative to the metE gene, which is also cloned on the same hybrid plasmids. Strains carrying the pldA gene on these high copy vectors do not appear to be affected by the product with respect to cell growth in any way. However they do harbour increased amounts of 29 K protein in cell envelope fractions, indicating that gene expression and product translocation to the OM are proportional to the increased gene copy number. We therefore conclude that phospholipase enzymatic activity is strictly regulated at the protein level.

Engineering cysteine mutants to obtain crystallographic phases with a cutinase from Fusarium solani pisi.

Cutinases are extracellular enzymes involved in the disruption of cutine, an insoluble polyester which covers the surface of plants. They belong to a class of serine esterases that are able to hydrolyse fatty acid esters and emulsified triglycerides as efficiently as lipases, but without displaying interfacial activation. Classical crystallographic methods for obtaining heavy-atom derivatives failed, so the cutinase structure has been solved exclusively by the multiple isomorphous replacement method using four Hg derivatives obtained from mutants S4C, S92C, S120C and S129C. Two of these derivatives behaved as expected: (i) the cys mutant of the catalytic Ser S120C, located at the surface of the active site pocket, leads to a good derivative; and (ii) the Hg atom of the derivative obtained with the S92C mutant is completely accessible to the solvent and occupies two alternative positions--consequently a poor derivative results. In contrast, two mutants show an unexpected behaviour: (i) the Hg atom in the S129C mutant was completely buried 10 A below the protein surface and yielded the best derivative; and (ii) a poor quality derivative was obtained with the S4C mutant. Cys 4 belongs to the disordered propeptide 1-16. The Cys 4 bound Hg atom is located in front of the Asp58 side chain, but neither Cys4 nor parts of the propeptide are clearly visible in the electron density maps of the derivative structure.

Overproduction and high-yield purification of phospholipase A from the outer membrane of Escherichia coli K-12.

The cloned gene for the outer-membrane-bound phospholipase A from Escherichia coli was placed under control of the strong PL promoter of phage lambda. Induction of PL resulted in a 250-fold overexpression up to about 2% total cellular protein. This overproduced enzyme was indistinguishable from the wild-type enzyme. A homogeneous phospholiphase A preparation was obtained in high yield from overproducing bacteria, using the zwitterionic detergent C12-Sulfobetaine and anion-exchange chromatography. Detergent gradients were found to exert great influence on the elution characteristics. Considerations for the choice of optimal detergent gradients are discussed. The purified enzyme migrated as a single 29-kDa band in SDS/polyacrylamide gels, and required Ca(II) for activity. Maximum activity was displayed by enzyme samples taken from solutions with detergent concentrations near the critical micelle concentration. However, upon switching from high to optimal detergent concentration, maximum activity was restored in several hours, probably reflecting a slow conformational transition of the protein. Because the final pure protein was found to hydrolyze phospholipids in the intact erythrocyte membrane, a densely packed bilayer, we assume that this protein is in its biological native state.

Secretion of biologically active porcine prophospholipase A2 by Saccharomyces cerevisiae. Use of the prepro sequence of the alpha-mating factor.

The cDNA coding for porcine pancreatic prophospholipase A2 (proPLA) has been cloned and expressed in Saccharomyces cerevisiae. Expression and secretion of proPLA could only be obtained after fusing the proPLA to the prepro sequence of the yeast alpha-mating factor. Upon secretion, the fusion protein was cleaved by the KEX2 protease yielding a 140-amino-acid zymogen-like form of the phospholipase A2. This protein was purified in high yield by ion-exchange chromatography. Limited proteolysis with trypsin cleaved the 'zymogen' to yield active phospholipase A2, which was indistinguishable from the authentic porcine pancreatic enzyme. These results show that a protein with a disulphide bridge content as high as 7 per 124 amino acid residues can be correctly processed by the yeast secretory apparatus.

Probing olfactory receptors with sequence-specific antibodies.

Molecular cloning has revealed the structure of several putative odorant receptors. Chemically synthesized peptides, that correspond to a predicted extracellular domain of the encoded proteins, were employed to generate receptor-specific antibodies. Immunohistological approaches as well as Western-blot analysis confirmed the specificity of the antipeptide sera. Furthermore, deglycosylation experiments explained the observed discrepancy between the molecular mass of odorant receptors, as determined by SDS/PAGE and Western-blot analysis of ciliary proteins (M(r) 50,000), and the predicted protein size based on the deduced primary structure from cloned receptor genes (M(r) 30,000-35,000). Receptor proteins become phosphorylated upon odorant stimulation of olfactory cilia preparations; this was demonstrated by immunoprecipitation experiments employing the sequence-directed, receptor-specific antibodies. Functional assays revealed that the receptor-specific antibodies significantly attenuate second messenger signalling elicited by inositol 1,4,5-trisphosphate-inducing odorants, whereas activation of the cAMP cascade by appropriate odorants was not affected. These observation indicate that the sequence-specific antibodies not only recognize odorant receptors, but also discriminate between receptor subtypes coupling to different second-messenger pathways.

The pro- and mature forms of the E. coli K-12 outer membrane phospholipase A are identical.

The nucleotide sequence of the pldA gene, coding for the outer membrane (OM) phospholipase A of Escherichia coli K-12, and flanking sequences, was determined. Data were obtained from sequences of overlapping deletions which had been generated in vitro from both ends of the gene, using DNase I in the presence of Mn2+ and Bal31 nuclease. The deduced amino acid sequence of the pldA gene product is the first primary sequence of a membrane-bound phospholipase. The complete PldA protein contains 260 amino acids, which include a putative signal sequence, and has a calculated mol. wt. of 29 946 similar to that of the purified protein. Furthermore we found the N terminus of the purified protein to be blocked and the overall amino acid composition to be consistent with the one deduced from the complete pldA gene. Analysis of proteins synthesized in minicells with a pldA coding plasmid in the presence of 8% ethanol did not reveal any new bands on polyacrylamide gels, whereas the control beta-lactamase clearly showed its unprocessed form under the same conditions. These data are consistent with the empirical prediction from the primary sequence, that the PldA protein lacks any signal peptidase 'target' site. We therefore conclude that the PldA protein is exported to the OM without proteolytic removal of the signal peptide.

Expression of porcine pancreatic phospholipase A2. Generation of active enzyme by sequence-specific cleavage of a hybrid protein from Escherichia coli.

The cDNA coding for the porcine pancreatic prophospholipase A2 (proPLA) has been cloned and expressed in E. coli. Expression of proPLA could only be obtained in the form of intracellular aggregates after fusing the 15 kDa proPLA to a large (greater than or equal to 45 kDa) bacterial peptide. The fusion protein was readily purified from cell lysates, and specifically cleaved. Cleavage of the fusion protein was achieved with either hydroxylamine (at Asn/Gly sequences in the denatured protein), or trypsin (between the pro- and the mature PLA in the renatured fusion protein). The former method releases a proPLA-like enzyme, while the latter directly yields PLA. Renaturation of the fusion protein was made possible by the use of a recently reported new S-sulphonation method. The released (pro)PLA was purified (yields of 2-3 mg/ltr of culture medium), and showed identical properties compared to native (pro)PLA.