Mechanism of inactivation of enveloped viruses by the biosurfactant surfactin from Bacillus subtilis.

The antiviral activity of surfactin, a cyclic lipopeptide antibiotic and biosurfactant produced by Bacillus subtilis, was determined for a broad spectrum of different viruses, Semliki Forest virus (SFV), herpes simplex virus (HSV-1, HSV-2), suid herpes virus (SHV-1), vesicular stomatitis virus (VSV), simian immunodeficiency virus (SIV), feline calicivirus (FCV), murine encephalomyocarditis virus (EMCV). In vitro experiments showed biphasic virus inactivation kinetics for enveloped viruses during treatment. Inactivation of enveloped viruses, especially herpes- and retroviruses, was much more efficient than that of non-enveloped viruses. For those viruses susceptible to its action, surfactin was active at 25 microM in medium containing 5% fetal calf serum (FCS). Concentrations up to 80 microM of surfactin led to a titre reduction of >4.4 log10 CCID50/ml for HSV-1 in 15 min and for SIV and VSV in 60 min. The inactivation rate increased linearly with the incubation temperature by a factor 2.4/10 degrees C and logarithmically with the concentration. Serum components, probably proteins and/or lipids, influence the effective surfactin concentration. A disruption of the viral lipid membrane and partially of the capsid was observed by electron microscopy. These findings suggest that the antiviral action, postulated also in other investigations, seems to be due to a physicochemical interaction of the membrane-active surfactant with the virus lipid membrane. Surfactin may be useful for application in virus safety enhancement of biotechnological and pharmaceutical products.

Analysis of surfactin synthetase subunits in srfA mutants of Bacillus subtilis OKB105.

The srfA operon of Bacillus subtilis functions in the biosynthesis of the lipopeptide antibiotic surfactin. On the basis of nucleotide sequence and genetic analysis, it is believed to encode three enzymes (E1A, E1B, and E2) that catalyze the incorporation of the surfactin substrate amino acids. Insertion, deletion, and amino acid substitution mutations of srfA were analyzed for subunit composition and activity as determined by assays of both amino acid-dependent ATP-PPi exchange and aminoacyl thioester formation. Insertion mutations in srfAA (encoding E1A, the subunit that incorporates Glu, Leu, and D-Leu) eliminated production and activity of all three enzymes. Deletions within srfAA and extending from srfAA to srfAB (encoding E1B, which incorporates Val, Asp, and D-Leu) abolished the activity and production of all three enzymes. Insertions between srfAA and srfAB and within srfAB eliminate the production and activity of E1B and E2. An insertion mutation in srfAC (encoding E2, which incorporates Leu) abolished the activity of E2 only. Mutations of the active serine in the putative 4'-phosphopantetheine-binding motif of the second and third domains of E1A eliminated thioester formation and severely reduced the ATP-PPi exchange activity of the two domains. However, the same mutation in the first domain of E1B had little effect on Val-dependent ATP-PPi exchange activity but abolished thioester formation. These results indicate that the coding assignments of the srfA genes are srfAA (E1A), srfAB (E1B), and srfAC (E2).

Structural and functional organization of the surfactin synthetase multienzyme system.

By gel filtration of a crude extract of Bacillus subtilis ATCC 21332 and OKB 105, the multienzyme system that forms the lipoheptapeptide surfactin was separated into three enzyme fractions, E1, E2, and E3. E1, which appeared near the exclusion limit of the column, activates all amino acid components of surfactin as aminoacyladenylates and thioesters according to the thioester mechanism. In addition, a leucine-activating enzyme (E2) and an acyltransferase (E3) were detected that show molecular masses of approximately 160 and 40 kDa, respectively. The surfactin synthetase multienzyme system was reconstituted by complementation of all three enzyme fractions, yielding high rates of lipopeptide formation. E1 is composed of two multifunctional polypeptides (E1A and E1B) with molecular masses of 460 and 435 kDa, respectively, that can be separated by high-resolution anion-exchange chromatography on Pharmacia Mono Q. E1A binds L-Glu and L-Leu in a molar ratio of 1:2, whereas E1B incorporates L-Val, L-Asp, and L-Leu in a molar ratio of 1:1:1. The hydroxy fatty acid moiety is contributed by the acyltransferase accepting the hydroxy fatty acid coenzyme A thioester as substrate. The transfer of the hydroxy fatty acid to E1A and the formation of the hydroxyacyl-L-glutamate intermediate are the initiation steps in the biosynthesis of surfactin. The amino acid-activating enzyme components E1A, E1B, and E2 have been highly purified and partially characterized.

Coordinated synthesis of leghemoglobin and root protein R18 in yellow lupin.

Uninfected roots of yellow lupin contain an abundant 18 kDa protein (referred to as R18), absent in the mature nodules. Some properties of this polypeptide are apparently similar to those of lupin leghemoglobins. However, the lack of any immuno crossreaction between R18 and leghemoglobin and differences in N-terminal amino acid sequences indicate that these proteins are coded by different genes. The decrease in the content of R18 protein in developing nodule is associated with the increased synthesis of leghemoglobin. This implies coordination of both events.

Cross-linked amino acids in the protein pair S13-S19 and sequence analysis of protein S13 of Bacillus stearothermophilus ribosomes.

The 30S ribosomal subunits from Bacillus stearothermophilus were cross-linked under native conditions with the bifunctional reagent diepoxybutane. The dominant protein-protein cross-link in the 30S ribosomal subunit between proteins S13 and S19 [Brockmöller, J., & Kamp, R.M. (1986) Biol. Chem. Hoppe-Seyler 367, 925-935] was isolated on a preparative scale. The presence of a single cross-link site between cysteine-83 of protein S13 and histidine-68 of protein S19 was established by microsequence analysis of isolated cross-linked peptides. This cross-link site was further confirmed by different analytical methods including fast atom bombardment mass spectrometry of the cross-linked peptide. The cross-linking site is located in the highly conserved C-terminal regions of proteins S13 and S19. In addition, the complete amino acid sequence of protein S13 from B. stearothermophilus is determined. Sequence comparison with the homologous Escherichia coli protein S13 revealed 58% identical amino acid residues.

Primary structure of the archaebacterial Methanococcus vannielii ribosomal protein L12. Amino acid sequence determination, oligonucleotide hybridization, and sequencing of the gene.

The primary structure of ribosomal protein L12 from Methanococcus vannielii has been determined by direct amino acid sequence analysis with automated liquid phase Edman degradation of the entire protein and manual 4-N,N'-dimethylaminoazobenzene-4'-isothiocyanate/phenylisothiocyanate sequencing of fragments obtained by enzymatic digestion and by partial acid hydrolysis. The knowledge of the amino acid sequences of these various fragments allowed the synthesis of two oligonucleotide probes complementary to the 5'- and the 3'-end of the gene, and they were used for hybridization with digested M. vannielii chromosomal DNA. Both oligonucleotide probes gave similar and clear hybridization signals. The plasmid pMvaX1 containing the entire gene of protein L12 was obtained. The nucleotide sequence complemented the partial amino acid sequence, and it is in full agreement with the protein sequence and the amino acid analysis. Comparison of secondary structural elements and hydrophobicity plots of the M. vannielii protein L12 with the known L12 sequences derived from other archaebacterial and eukaryotic sources show strong homologies among these sequences. They contain an exceptional highly conserved hydrophilic sequence area in the C-terminal part of the proteins. In comparison with eubacterial L12 proteins, the conservation is reduced to single amino acid residues. However, the eubacterial L12 proteins have hydrophilic regions similar to those of L12 from M. vannielii. These regions are predicted to be located at the surface of the proteins, as has been proven to be the case in crystallized Escherichia coli L12 protein. It is possible that the strongly conserved hydrophilic sequence regions form part of the factor-binding domain.

Nucleotide sequence of the colicin B activity gene cba: consensus pentapeptide among TonB-dependent colicins and receptors.

Colicin B formed by Escherichia coli kills sensitive bacteria by dissipating the membrane potential through channel formation. The nucleotide sequence of the structural gene (cba) which encodes colicin B and of the upstream region was determined. A polypeptide consisting of 511 amino acids was deduced from the open reading frame. The active colicin had a molecular weight of 54,742. The carboxy-terminal amino acid sequence showed striking homology to the corresponding channel-forming region of colicin A. Of 216 amino acids, 57% were identical and an additional 19% were homologous. In this part 66% of the nucleotides were identical in the colicin A and B genes. This region contained a sequence of 48 hydrophobic amino acids. Sequence homology to the other channel-forming colicins, E1 and I, was less pronounced. A homologous pentapeptide was detected in colicins B, M, and I whose uptake required TonB protein function. The same consensus sequence was found in all outer membrane proteins involved in the TonB-dependent uptake of iron siderophores and of vitamin B12. Upstream of cba a sequence comprising 294 nucleotides was identical to the sequence upstream of the structural gene of colicin E1, with the exception of 43 single-nucleotide replacements, additions, or deletions. Apparently, the region upstream of colicins B and E1 and the channel-forming sequences of colicins A and B have a common origin.

Primary structure of colicin M, an inhibitor of murein biosynthesis.

The DNA sequence of the colicin M activity gene cma was determined. A polypeptide consisting of 271 amino acids was deduced from the nucleotide sequence. The amino acid sequence agreed with the peptide sequences determined from the isolated colicin. The molecular weight of active colicin M was 29,453. The primary translation product was not processed. In the domain required for uptake into cells, colicin M contained the pentapeptide Glu-Thr-Leu-Thr-Val. A similar sequence was found in all colicins which are taken up by a TonB-dependent mechanism and in outer membrane receptor proteins which are constituents of TonB-dependent transport systems. The structure of colicin M in the carboxy-terminal activity domain had no resemblance to the pore-forming colicins or colicins with endonuclease activity. Instead, the activity domain contained a sequence which exhibited homology to the sequence around the serine residue in the active site of penicillin-binding proteins of Escherichia coli. The colicin M activity gene was regulated from an SOS box upstream of the adjacent colicin B activity gene on the natural plasmid pColBM-Cl139.

Desalting and concentration of proteins in dilute solution using reversed-phase high-performance liquid chromatography.

A rapid method for desalting and concentrating dilute protein solutions using short reversed-phase columns (3-4 cm) has been described. The recovery of proteins is usually 90-100%. The method is simple and rapid and allows the desalting and concentration of protein samples simultaneously. A wide variety of proteins in the range up to 80 kDa can be desalted in microgram to milligram amounts, and volumes up to 1 liter can be concentrated to a few milliliters by a single injection.

Isolation and identification of two protein-protein crosslinks within the two subunits of Bacillus stearothermophilus ribosomes.

Bacillus stearothermophilus 30S and 50S ribosomal subunits were isolated and crosslinked with diepoxybutane. The crosslinked proteins were extracted with LiCl or with 67% acetic acid and purified by a combination of different high performance liquid chromatography techniques. The protein fractions were analysed by two-dimensional and diagonal polyacrylamide gel electrophoresis and by immunological methods. Two crosslinked protein pairs, one from the large and one from the small subunit, consisting of proteins L23-L29 and S13-S19 respectively, were isolated in milligram amounts for determination of the crosslinked amino acids.

Two-dimensional polyacrylamide gel electrophoresis of ribosomal proteins in the nanogram range.

A two-dimensional gel electrophoresis system to identify and check the purity of ribosomal proteins from different organisms with nanogram quantities is described. This procedure combines the method of Geyl et al. for the separation of ribosomal proteins of Escherichia coli, and the microscale electrophoresis system for proteins described by Neuhoff and Poehling, with several modifications. The first gel dimension is carried out in capillaries and the second in the form of slab gels, both are run in newly designed chambers suitable for 10-20 samples. This electrophoresis system enables a reduction of the running time from 2 days to 2 hours and an increase in sensitivity, with Coomassie blue staining, from 3-5 micrograms for the normal 100 X 100 mm gels to 50-100 ng. The resolution of all ribosomal proteins on the micro-gel (30 X 38 X 0.5 mm) is similar to the separation on the mini-gel of 100 X 100 X 3 mm as described by Geyl et al.

Application of high-performance liquid chromatographic techniques to the separation of ribosomal proteins of different organisms.

The ribosomal proteins from Escherichia coli, Bacillus stearothermophilus and Methanococcus vannielii were separated by size-exclusion, ion-exchange and reversed-phase high-performance liquid chromatography (HPLC), employing new column materials, different gradient systems, and preparative columns, respectively. The purity of the isolated proteins was analysed by one- and two-dimensional gel electrophoresis and by direct micro-sequencing. The separation of ribosomal proteins could be improved by employing propanol gradients in combination with Vydac reversed-phase columns. From the E. coli ribosome, fifteen S and twenty-three L proteins were isolated in sequencer purity by this method. In addition, ion-exchange HPLC was proven to be useful for isolating ribosomal proteins under native conditions: six S proteins and sixteen L proteins from E. coli could be purified. Some of these proteins were not isolated by the reversed-phase procedures, e.g. proteins L9, L14 and L21.

Purification of Escherichia coli 50 S ribosomal proteins by high performance liquid chromatography.

The 50 S subunit proteins from the Escherichia coli ribosome were purified by size-exclusion, ion-exchange or reversed phase high performance liquid chromatography (HPLC) avoiding any precipitation or desalting procedures during isolation. Best resolution of this complex protein mixture was achieved by reversed phase chromatography on supports with short alkyl chains and C18 hydrocarbon-bonded phases; 23 out of the 32 proteins from the 50 S subunit were purified as shown by two-dimensional gel electrophoresis, amino acid analysis and direct micro-sequencing. Protein recoveries varied between 25 and 84% as determined by amino acid analysis. Ribosomal proteins of other organisms can be separated under similar conditions.

Purification of Escherichia coli 30S ribosomal proteins by high performance liquid chromatography.

High performance liquid chromatography was applied to the separation of proteins derived from the Escherichia coli 30S ribosomal subunit. Several methods of separating this protein mixture has been tested: size-exclusion chromatography on hydrophilic phases; ion exchange and reversed phase chromatography (on C2 to C18 hydrocarbon-bonded supports). Various elution systems were examined in order to obtain pure proteins suitable for micro-sequence analysis. The resolution and yields of the proteins varied considerably, depending on the type of support and gradient system used. The best results were achieved with uniformly globular-shaped supports of large pore size, and by combining high performance size exclusion with rechromatography on reversed phase columns. Purification conditions for the individual proteins are listed. The methods employed avoid any precipitation step and allow easy identification of the proteins by one or two-dimensional gel electrophoresis, amino-acid analysis or direct manual or automatic micro-sequencing. Since the isolation time is much reduced compared with conventional purification procedures, the proteins obtained by the techniques described here are well suited for topographical and immunological studies or reconstitution assays. Ribosomal proteins of other organisms can be separated under similar conditions.

Direct micro-sequence analysis of peptides from Escherichia coli ribosomal proteins S11, L9 and L29 after separation by reversed phase chromatography.

Tryptic peptides of the ribosomal proteins S11, L9 and L29 were separated by reversed phase chromatography under conditions which enabled direct micro-sequencing with the 4-(dimethylamino)azobenzene-4'-isothiocyanate/phenylisothiocyanate double coupling method [Chang, Brauer, Wittmann-Liebold (1978) FEBS Lett. 93, 205-214]. The peptides were separated on a RP-18 column employing volatile buffers at pH 2.0, 4.1 and 7.8. Depending on the different chromatographic behaviour of the peptide mixture, the elution gradient was optimised for each hydrolysate using 20 micrograms of the hydrolysed protein. Preparative separations were made with 150-250 micrograms. At least 80% of the peptides could be isolated by these techniques and used for direct micro-sequencing without further purification or desalting. The results show that the high-performance liquid chromatographic method employed allows easy isolation and sequencing with minute amounts of peptides.