Construction and properties of an intracellular serine protease mutant of Bacillus subtilis

The first structure in a family of peptidase inhibitors reveals an unusual Ig-like fold

We report the crystal structure solution of the Intracellular Protease Inhibitor (IPI) protein from Bacillus subtilis, which has been reported to be an inhibitor of the intracellular subtilisin Isp1 from the same organism. The structure of IPI is a variant of the all-beta, immunoglobulin (Ig) fold. It is possible that IPI is important for protein-protein interactions, of which inhibition of Isp1 is one. The intracellular nature of ISP is questioned, because an alternative ATG codon in the ipi gene would produce a protein with an N-terminal extension containing a signal peptide. It is possible that alternative initiation exists, producing either an intracellular inhibitor or a secreted form that may be associated with the cell surface.  Homologues of the IPI protein from other species are multi-domain proteins, containing signal peptides and domains also associated with the bacterial cell-surface. The cysteine peptidase inhibitors chagasin and amoebiasin also have Ig-like folds, but their topology differs significantly from that of IPI, and they share no recent common ancestor. A model of IPI docked to Isp1 shows similarities to other subtilisin:inhibitor complexes, particularly where the inhibitor interacts with the peptidase active site.

Identification of the degradome of Isp-1, a major intracellular serine protease of Bacillus subtilis, by two-dimensional gel electrophoresis and matrix- assisted laser desorption/ionization-time of flight analysis

Intracellular serine protease-1 (Isp-1) is a major intracellular serine protease of Bacillus subtilis, whose functions still remain largely unknown. Furthermore, physiological substrates are yet to be determined. To identify Isp-1 substrates, we digested extract obtained from an Isp-1 deficient Bacillus mutant with purified Isp-1 and examined eliminated or decreased spots by two-dimensional gel and matrix-assisted laser desorption/ionization-time of flight analyses. Proteins degraded by Isp-1, termed the Isp-1 degradome, are involved in a variety of cellular functions such as DNA packing, genetic competence, and protein secretion. From the degradome we selected ClpC and EF-Tu as putative Isp-1 substrates and studied their in vitro degradation. ClpC and EF-Tu contain putative cleavage sites for Isp-1. N-terminal sequencing of in vitro proteolytic fragments of ClpC and EF-Tu revealed that these sites are indeed recognized and cleaved by Isp-1. Moreover, the cellular levels of ClpC and EF-Tu were dramatically reduced at the late stationary phase, where the expression level of Isp-1 was greatly increased. These results suggest that the regulated proteolysis of ClpC by Isp-1 plays an important role in the stationary phase adaptive response. This degradomic approach could provide a powerful tool for finding physiological substrates of many proteolytic enzymes whose functions remain to be determined.

The IspA protease's involvement in the regulation of the sporulation process of Bacillus thuringiensis is revealed by proteomic analysis

We have observed that the process of sporulation of the ispA-deficient mutant was delayed under phase-contrast microscopy. The protein profiles of the ispA-deficient mutant have been analyzed using two-dimensional gel electrophoresis. The results of a proteomic analysis using MALDI-TOF MS indicated that a sporulation-associated protein, pro- [Formula: see text], was upregulated, while two other sporulation-associated proteins, SpoVD and SpoVR, were downregulated in the ispA-deficient mutant. It has been known that pro- [Formula: see text] is a precursor of [Formula: see text] and is required for gene expression related to the late stage of sporulation. Moreover, SpoVD and SpoVR are known to be involved in the formation of the spore cortex. Based on these observations, we propose that the delay in the sporulation process observed in the ispA-deficient mutant may be due to a failure of [Formula: see text] to signal sporulation. This phenomenon may be further enhanced by insufficient amount of SpoVD and SpoVR for cortex formation. In this study, we have revealed for the first time a possible pathway for the regulation of sporulation-associated proteins via IspA.

Immunocytochemical localization of a calmodulinlike protein in Bacillus subtilis cells

To determine possible functions of the calmodulinlike protein of Bacillus subtilis, the time course of its expression during sporulation and its cellular localization were studied. The protein was expressed in a constitutive manner from the end of logarithmic growth through 8 h of sporulation as determined by antibody cross-reactivity immunoblots and enzyme-linked immunosorbent assays (ELISAs). In partially purified extracts, the immunopositive protein comigrated upon electrophoresis with a protein which selectively bound [(45)Ca]CaCl(2), ruthenium red, and Stains-all. Previous studies showed increased extractability of the calmodulinlike protein from B. subtilis cells when urea and 2-mercaptoethanol were used in breakage buffers, implying that the protein might be partially associated with the membrane fraction. This was confirmed by demonstrating that isolated membrane vesicles of B. subtilis also gave positive immunological tests with Western blotting and ELISAs. To more precisely locate the protein in cells, thin sections of late-log-phase cells, sporulating cells, and free spores were reacted first with bovine brain anticalmodulin specific antibodies and then with gold-conjugated secondary antibodies; the thin sections were examined by transmission electron microscopy. The calmodulinlike protein was found almost exclusively associated with the cell envelope of these fixed, sectioned cells. A possible function of the calmodulinlike protein in sensing calcium ions or regulating calcium ion transport is suggested.

Preparation of general proteinase substrates using 3,5-dinitrosalicylaldehyde

To search for new proteinases in Bacillus subtilis we have developed a general method for synthesizing chromogenic proteinase substrates using 3,5-dinitrosalicylaldehyde (DNSA). Hammersten casein and soluble protein from extracts from B. subtilis cells were labeled with DNSA in the presence of NaBH4. After dialysis (pH 7.8), the resultant 3,5-dinitro-2-hydroxybenzyl-casein (DNHB-casein) and DNHB-bacterial cell protein solutions were a light orange color. A model compound, N-benzyl-3,5-dinitro-2-hydroxybenzylamine was synthesized and estimated to have a molar absorption coefficient of 14,100 M-1 cm-1 at 366 nm at pH 8, which was used to calculate dye loading on casein. Chromogenic substrates prepared in this way should retain positive charges on lysine residues. DNHB-casein and DNHB-bacterial cell protein were incubated with varying concentrations of subtilisin BPN' for varying times, precipitated with trichloroacetic acid and centrifuged. The acid-soluble supernatant fractions were made basic with NaOH and absorbances were measured at 366 nm, the absorption maximum. Color production was proportional to subtilisin concentration and times of incubation; under the assay conditions used, the limit of detection of subtilisin was about 100 ng. Five proteinase activities were detected in soluble extracts of B. subtilis using DNHB-labeled proteins as substrates.

Characterization of the gene encoding an intracellular proteinase inhibitor of Bacillus subtilis and its role in regulation of the major intracellular proteinase

The gene (ipi) for an intracellular proteinase inhibitor (BsuPI) from Bacillus subtilis was cloned and found to encode a polypeptide consisting of 119 amino acids with no cysteine residues. The deduced amino acid sequence contained the N-terminal amino acid sequence of the inhibitor, which was chemically determined previously, and showed no significant homology to any other proteinase inhibitors. Analysis of the transcription initiation site and mRNA showed that the ipi gene formed an operon with an upstream open reading frame with an unknown function. The transcriptional control of ipi gene expression was demonstrated by Northern (RNA) blot analysis, and the time course of transcriptional enhancement roughly corresponded to the results observed at the protein level. Strains in which the ipi gene was disrupted or in which BsuPI was overexpressed constitutively sporulated normally. Analysis of the time course of production of the intracellular proteinase and proteinase inhibitor in these strains suggested that BsuPI directly regulated the major intracellular proteinase (ISP-1) activity in vivo.

Intracellular serine protease-4, a new intracellular serine protease activity from Bacillus subtilis

A previously undiscovered intracellular serine protease activity, which we have called intracellular serine protease-4, was identified in extracts of stationary Bacillus subtilis cells, purified 260 fold from the cytoplasmic fraction, and characterized. The new protease was stable and active in the absence of Ca2+ ions and hydrolyzed azocasein and the chromogenic substrate carbobenzoxy-carbonyl-alanyl-alanyl-leucyl-p-nitroanilide, but not azocollagen or a variety of other chromogenic substrates. The protease was strongly inhibited by phenylmethylsulfonylfluoride, chymostatin and antipain, but not by chelators, sulfhydryl-reactive agents or trypsin inhibitors. Its activity was stimulated by Ca2+ ions and gramicidin S; its pH and temperature optima were 9.0 and 37 degrees C, respectively. Although intracellular serine protease-4 was immunochemically distinct from intracellular serine protease-1, it was absent from a mutant in which the gene encoding the latter was disrupted.

Energy and calcium ion dependence of proteolysis during sporulation of Bacillus subtilis cells

Bacterial cells degrade intracellular proteins at elevated rates during starvation and can selectively degrade proteins by energy-dependent processes. Sporulating bacteria can degrade protein with apparent first-order rate constants of over 0.20 h-1. We have shown, with an optimized [14C]leucine-labeling and chasing procedure, in a chemically defined sporulation medium, that intracellular protein degradation in sporulating cells of Bacillus subtilis 168 (trpC2) is apparently energy dependent. Sodium arsenate, sodium azide, carbonyl cyanide m-chlorophenylhydrozone, and N,N'-dicyclohexylcarbodiimide, at levels which did not induce appreciable lysis (less than or equal to 10%) over 10-h periods of sporulation, inhibited intracellular proteolysis by 13 to 93%. Exponentially growing cells acquired arsenate resistance. In contrast to earlier reports, we found that chloramphenicol (100 micrograms/ml) strongly inhibited proteolysis (68%) even when added 6 h into the sporulation process. Restricting the calcium ion concentration (less than 2 microM) in the medium had no effect on rates or extent of vegetative growth, strongly inhibited sporulation (98%), and inhibited rates of proteolysis by 60% or more. Inhibitors of energy metabolism, at the same levels which inhibited proteolysis, did not affect the rate or degree of uptake of Ca2+ by cells, which suggested that the Ca2+ and metabolic energy requirements of proteolysis were independent. Restricting the Ca2+ concentration in the medium reduced by threefold the specific activity in cells of the major intracellular serine proteinase after 12 h of sporulation. Finally, cells of a mutant of B. subtilis bearing an insertionally inactivated gene for the Ca2(+)-dependent intracellular proteinase-1 degraded protein in chemically defined sporulation medium at a rate indistinguishable from that of the wild-type cells for periods of 8 h.

Intracellular serine protease 1 of Bacillus subtilis is formed in vivo as an unprocessed, active protease in stationary cells

Western immunoblots and assays of Bacillus subtilis extracts showed that intracellular serine protease 1 is produced in a form larger than previously reported, appears not to have undergone N-terminal processing, and is active in the presence or absence of calcium. No evidence for an inactive precursor form of the protease was found.

Control of intracellular serine protease expression in Bacillus subtilis

Expression of the major intracellular serine protease (ISP-1) gene of Bacillus subtilis was studied by using a translational fusion plasmid in which the isp promoter region was fused to the lacZ gene. beta-Galactosidase activity, used to measure transcription from the isp promoter, was produced immediately after the end of exponential growth, whereas intracellular protease activity was not detected until 4 h later. These results are consistent with a previous suggestion that ISP-1 initially accumulates in the cell in an enzymatically inactive form. ISP-1 activity was detected in all of the sporulation-deficient strains examined, and the amount of protease activity always corresponded to the amount of beta-galactosidase activity. These results indicate that the activation of ISP-1 is not dependent on a sporulation-specific gene product. Expression of ISP-1 is regulated by a number of mutations known to affect the expression of extracellular enzymes. In sacU(h) and sacQ(h) mutants, the expression of ISP-1 was 10-fold higher than in the wild-type strain. In catA, hpr, and scoC strains, expression of ISP was stimulated two- to threefold, whereas in sacU mutants the expression of ISP-1 was reduced to less than 10% of the wild-type level. The temporal expression and activation of ISP-1 was not affected by any of these mutations. This is the first evidence that the expression of a native intracellular protein is affected by these hyperproduction mutations.