J. Biochem. 116, 1156-1163 (1994)

Three Novel Subtilisin- Inhibitors from Streptomyces: Primary Structures and Inhibitory Properties

Mahito Terabe,*Shuichi Kojima,*Seiichi Taguchi,tHaruo Momose,tand Kin-ichiro Miura*,1 *Institute for Biomolecular Science , GakushuinUniversity, 1-5-1, Mejiro, Toshima-ku, Tokyo 171; and tDepartment of BiologicalScience and Technology, Science University of Tokyo, Noda , Chiba278 Receivedfor publication, July 20, 1994

Three novel proteinaceous inhibitors, which had been identified as "Streptomyces subtilisin inhibitor-like (SIL) proteins" and exhibited trypsin inhibition in addition to strong inhibition toward subtilisin BPN', were purified from the culture broth of three Streptomyces strains: SIL10 from S. thermotolerans, SIL13 from S. galbus, and SIL14 from S. azureus. Their primary structures were determined by sequence analysis of intact SIL inhibitors and obtained by enzymatic digestions of S-pyridylethylated SIL inhibitors. These inhibitors were composed of about 110 amino acids and existed as dimer proteins. The reactive site was identified as Lys-Gln for all three inhibitors by sequence analysis of their modified forms in which the reactive-site bond was specifically cleaved by subtilisin BPN' under acidic conditions. Thus, their inhibition toward trypsin and subtilisin BPN' was due to the presence of a Lys residue at the P1 site. Inhibitor constants toward subtilisin BPN' and trypsin were also determined. These inhibitors showed relatively high sequence homology to other SSI-family inhibitors possessing a Lys residue at the P1 site, with amino acid replacements on their molecular surface.

Key words: inhibitory property, primary structure, inhibitor, Streptomyces subtilisin inhibitor-like inhibitor, Streptomyces subtilisin inhibitor.

Streptomyces subtilisin inhibitor (SSI) is a well character ized proteinaceous inhibitor of (1). It was isolated MATERIALSAND METHODS from the culture broth of Streptomyces albogriseolus S 3253 (2). To clarify whether such inhibitors are produced Materials -Proteolytic , lysyl , by specific strains of Streptomyces or widely distributed in Staphylococcus aureus V8 protease, arginylendopeptidase, , we have tested various species of streptomycetes subtilisin BPN', and trypsin were obtained from Wako and found that homologous proteins of SSI are widely Pure Chemicals, Boehringer Mannheim, Takara Shuzo, present in various Streptomyces strains (3-5). Termed Nagase Biochemicals, and Sigma, respectively. All other "SSI -like (SIL) proteins," they exhibited strong inhibitory chemicals were of reagent grade for biochemical research. activity toward subtilisin BPN'. Some also inhibited tryp Three Streptomyces strains (S. thermotolerans 4519, S. sin or . To obtain data on the structure galbus 4222, and S. azureus 4217) were generously pro function relationship of these SSI-family inhibitors, we vided by Dr. T. Nakase, the Institute of Physical and have purified SIL inhibitors, determined their primary Chemical Research, Riken, Wako. structures, and measured their inhibitory activities (3 7), Purification of SIL Inhibitors-Three Streptomyces in addition to mutational analysis of SSI by protein engi strains were cultured in 500 ml of SSI medium (3) at 30`C neering techniques (8-17). Recently we have determined for 7 days. Purification of SIL10 from S. thermotolerans, the primary structures of SIL1-4, all of which inhibited SIL13 from S. galbus, and SIL14 from S. azureus was both subtilisin BPN' and trypsin, and discussed the differ carried out essentially by the procedures described previ ences in their inhibitory strength on the basis of these ously (18). structures (6, 7). Estimation of Relative Molecular Mass-SDS-poly In this paper, we purified novel SIL inhibitors from three acrylamide gel electrophoresis (SDS-PAGE) to estimate strains of Streptomyces and determined their primary the relative molecular mass in the denatured state was structures and inhibitor constants toward proteases to performed by the method of Laemmli (19), and analytical obtain further information on the naturally occurring gel-filtration to estimate that in the native state was carried inhibitors belonging to the SSI family. out by the procedures described previously (7, 18). Amino Acid Composition and Sequence Analyses Amino acid composition was analyzed after hydrolysis in ' To whom correspondence should be addressed . 5-7N HCl containing 0.1% phenol at 110'C for 24h. Abbreviations: SIL, Streptomyces subtilisin inhibitor-like; SSI, Cysteine residues in the intact SIL inhibitors or digested Streptomyces subtilisin inhibitor. peptides were S-carboxymethylated (20) or S-pyridyl

1156 J. Biochem . Subtilisin-Trypsin Inhibitors from Streptomyces 11.57 ethylated (18), respectively. Automated sequence analysis sequences of about 30 residues were determined. They by Edman degradation was carried out using an Applied were also subjected to enzymatic digestions after reduction Biosystems model 476A protein sequencer. and S-pyridylethylation of cystine residues. Enzymatic Digestions-Portions of 4 nmol of S-pyridyl (i) Amino acid sequence of SIL10: Since amino acid ethylated SIL inhibitors were digested with lysyl endopep composition analysis indicated that SIL10 had four Lys tidase or S. aureus V8 protease at 37'C for 16 h with an residues, S-pyridylethylated SIL10 was digested with /substrate ratio of 1 : 50 (mol/mol) in 100 pl of 2 , and five peaks were detected on M urea/0.1M Tris-HCI (pH 8.6) or 2 M urea/0.1M reverse-phase HPLC [Fig. 2S(a) ]. Amino acid sequences of NH4HCO3 (pH 7.8), respectively. The digestion of 2 nmol these peptides up to their carboxy termini were deter of peptides by arginylendopeptidase was carried out in 2 M mined, as shown in Fig. IA. The connectivity of each urea/50 mM Tris-HC1 (pH 8.0). Digested peptides were peptide obtained by lysyl endopeptidase digestion was separated using reverse-phase HPLC on C18 column (L investigated by sequence analysis of the peptides derived column ODS, 0.46 X 15 cm) in 0.1% trifluoroacetic acid with from S. aureus V8 protease digests of S-pyridylethylated a gradient of acetonitrile at a flow rate of 0.8 ml/min. SIL10 [Fig. 2S(b) ], and was concluded to be in the order of Peptides were designated by a serial number prefixed by a L3-L2-L5-L1-L4 in the SIL10 molecule, as shown in Fig. letter representing the type of digestion: L, lysyl endopep IA. The revealed carboxy-terminal sequence was in good tidase; V, S. aureus V8 protease; A, arginylendopeptidase. agreement with the result of carboxypeptidase A treat Carboxypeptidase A Treatment-SIL inhibitors (2 nmol) ment: 3.8 nmol of Phe, 1.8 nmol of Ser, and 1.1 nmol of Val were treated with 20 pmol of carboxypeptidase A in 100 pl from 2 nmol of SIL10. of 20 mM Tris-HC1(pH 8.0) at 37'C for 3 h, then applied to (ii) Amino acid sequence of SIL13: S-Pyridylethylated a Hitachi model L-8500 amino acid analyzer to analyze the SIL13 was digested with S. aureus V8 protease, and four released amino acids. peaks were detected, as shown in Fig. 3S(a). The V-4 Specific Digestion at the Reactive Site by Subtilisin peptide was an amino-terminal fragment, and the V-1 BPN'-A mixture of subtilisin BPN' and 100 p g of SIL peptide was thought to follow it, since its amino acid inhibitors was incubated in 600 pl of 10 mM Tris-HCl (pH sequence up to the 18th residue was identical to that of 8.0) at 37°C for 10 min to form an enzyme/inhibitor intact SIL13 from position 17 to position 34. The V-2 complex. The molar ratios of SIL inhibitors to subtilisin peptide was considered to be a carboxy-terminal fragment BPN' were 1.40 for SIL10 and SIL14, and 1.20 for SIL13. because of its lack of a Glu residue and the agreement with Then an equal volume of ice-cold 0.4 M glycine-HCI (pH the result of carboxypeptidase A treatment: 3.9 nmol of 2.5) was added, and the mixture was allowed to stand at 4'C Phe, 1.9 nmol of Ala, and 1.6 nmol of Val from 2 nmol of for 1 h, followed by precipitation of proteins with a final SIL13. The amino acid sequence of V-3 peptide was concentration of 15% trichloroacetic acid to obtain a determined up to the 43rd residue. The remaining sequence modified form of SIL inhibitor in which the reactive-site in the carboxy-terminal region of the V-3 peptide was determined by sequence analysis of the peptide obtained by peptide bond was specifically digested. Determination of Inhibitor Constants-Inhibitor con lysyl endopeptidase (see below). Thus, SIL13 was deduced to be composed of four peptides obtained by S. aureus V8 stants (K,) toward subtilisin BPN' and trypsin were deter mined using synthetic substrates by the procedures de protease digestion in the order of V4-Vl-V3-V2. To confirm the connectivity of these peptides and to scribed previously (8-10). determine the remaining sequence in the V-3 peptide, S-pyridylethylated SIL13 was digested with lysyl endo RESULTS peptidase, and four peaks were detected [Fig. 3S(b)]. Purification and Characterization of SIL Inhibitors Amino acid sequence analysis of L-1 peptide confirmed the Production of SIL10 from S. thermotolerans and SIL13 connectivity of the V-3 and V-2 peptides and revealed the from S. galbus was reported previously (4). In this study, undetermined sequence in the carboxy-terminal region of the V-3 peptide. Sequence analysis of a long L-2 peptide we also report the production of SIL14 from S. azureus. All suggested the overlapping of the V-1 and V-3 peptides, and these SIL inhibitors have been isolated as proteinaceous thus the L-2 peptide with two Arg residues was digested inhibitors that inhibit both subtilisin BPN' and typsin. The with arginylendopeptidase. Two peaks were detected on inhibitors were purified from the culture medium essen reverse-phase HPLC [Fig. 3S(c)]. Peak A-1 was found to tially by the procedures described previously (18): ion be composed of two peptides, since two amino acids were exchange chromatography on DEAE-cellulose, hydrophobic identified at each cycle. One possible sequence correspond chromatography on Phenyl-Sepharose, and gel-filtration on ed to the amino-terminal sequence of the V-3 peptide that Sephacryl S-200. Yield from 500 ml of culture was about had been determined. The other sequence in.the peak A-1 14.5 mg for SIL10, 4.6 mg for SIL13, and 33 mg for SIL14. confirmed the connectivity of the V-1 and V-3 peptides, as The relative molecular mass of each SIL inhibitor in the shown in Fig. 1B. denatured state was estimated to be about 10,000 by (iii) Amino acid sequence of SIL14: S-Pyridylethylated SDS-PAGE [Fig. 1S(A) ] , and that in the native state to be SIL14 was digested with lysyl endopeptidase, and three about 20,000 by gel-filtration [Fig. 1S(B) ], suggesting that peaks were detected on reverse-phase HPLC [Fig. 4S(a) ]. all these SIL inhibitors existed as dimers, like other The L-3 peptide was considered to be a carboxy-terminal members of the SSI family (1, 6, 7). The amino acid fragment, because it did not have a Lys residue and the compositions of the SIL inhibitors are shown in Table IS. determined sequence was consistent with the result of Determination of Complete Amino Acid Sequences of carboxypeptidase A treatment: 3.8 nmol of Phe, 1.9 nmol SIL Inhibitors-Purified SIL inhibitors were initially of Ala, and 1.8 nmol of Val from 2 nmol of SIL14. The L-1 applied to the automated protein sequencer, and the

Vol. 116, No. 5, 1994 1158 M . Terabe et al.

Fig. 1. Complete amino acid sequences and sequencing strategy of SIL10 (A), SIL13 (13), and SIL14 (C). Long arrows show the amino acid sequences identified by sequencer and dashed lines indicate the remaining regions. Amino acids released by carboxypeptidase A are indicated by CPase A. Peptide nomenclature is described in "MATERIALS AND METHODS ."

J. Biochem. Subtilisin-Trypsin Inhibitors from Streptomyces 1159

TABLE I. Inhibitor constants for subtilisin BPN' and trypsin of SIL10, SIL13, SIL14, and other subtilisin-trypsin inhibitors in the SSI family (6, 7).

Fig. 2. Reverse-phase HPLC of subtilisin-treated SIL13 under acidic conditions. SIL13 was treated by subtilisin BPN' at 4'C for 1 h under acidic conditions, and proteins were precipitated with dicated that peak A corresponded to the modified form. trichloroacetic acid. The precipitate was dissolved in 0.1% trifluoro Thus, peak A was subjected to sequence analysis, and a acetic acid and subjected to HPLC on a C18 column (L-column ODS) . sequence Gln-Tyr-Ala-Pro-Val-Val, which corresponded to positions 70 to 75 of SIL13, was obtained in addition to that from an amino terminus. This indicated that the reactive peptide, a long amino-terminal fragment, was further site of SIL13 was Lys69-G1n70. The reactive sites of SIL10 digested with arginylendopeptidase. Three peaks were and SIL14 were similarly identified as Lys-Gln. observed [Fig. 4S(b)]. The A-3 peptide was an amino Inhibitory Activities of SIL Inhibitors-The inhibitory terminal fragment, and the A-1 peptide was found to follow activities of SIL10, SIL13, and SIL14 toward subtilisin it, since its amino-terminal sequence was identical to that BPN' and trypsin were measured using synthetic sub of the intact SIL14 from position 24. These results suggest strates, and the inhibitor constants (K,) were determined ed that the three peptides obtained by arginylendopep from the inhibition kinetics. The K; values obtained are tidase digestion were arranged in the order of A3-Al-A2 in summarized in Table I with those of other subtilisin the L-1 peptide. trypsin inhibitors of the SSI family (6, 7). The three SIL To determine the overlapping of these peptides, S inhibitors inhibited subtilisin BPN' strongly with K; values pyridylethylated SIL14 was digested with S. aureus V8 of about 10-11 M, which were close to those of other protease. Peak V-1 in Fig. 4S(c) was found to contain two members of the SSI family possessing a Lys or Arg residue peptides, one of which confirmed the connectivity of the L-2 at the P1 site (6, 7). They inhibited trypsin with K; values and L-3 peptides, although it was cleaved artificially at the of about 10-s-10-1° M, showing weaker inhibition than Gly-Ser peptide bond. Amino acid sequence analysis of the toward subtilisin BPN', as in the case of other subtilisin V-2 peptide confirmed the connectivity of both the A-1 and trypsin inhibitors of the SSI family bearing an amino acid A-2 peptides and the A-2 and L-2 peptides, as shown in Fig. residue with a bulky side chain such as Met or Trp at the P4 1C. Thus, five peptides obtained by the lysyl endopeptidase site (6). Of the three SIL inhibitors in the present study, and subsequent arginylendopeptidase digestions were SIL13 was less effective than SIL10 and SIL14 as a trypsin concluded to be arranged in the order of A3-A1-A2-L2-L3 inhibitor. in the SIL14 molecule. (iv) Full sequences of SIL inhibitors: The full amino acid DISCUSSION sequences of SIL10, SIL13, and SIL14 are shown in Fig. 1 along with the sequencing strategy employed. Amino acid In this paper, we have determined the complete amino acid compositions of peptides subjected to sequence analysis are sequences of three novel protease inhibitors from Strep shown in Tables IIS-IVS. The determined sequences were tomyces which exhibited inhibitory activities toward both in good agreement with the results of amino acid composi subtilisin BPN' and trypsin, and measured their inhibitory tion analysis, as shown in Table IS. SIL10 and SIL14 were activities. Figure 3 compares the determined primary composed of 107 amino acids, and SIL13 was composed of structures of SIL10, SIL13, and SIL14 with those of nine 109 amino acids. The difference in length was due to an other SSI-family inhibitors whose complete amino acid extension of SIL13 of only two amino acids in the amino sequences have been determined, and their sequence terminal region. similarity is summarized in Table II. Their dimeric forms Identification of the Reactive Site of SIL Inhibitors-To and the absence of insertions or deletions in the sequence identify their reactive sites, the SIL inhibitors were com alignment suggest that the overall tertiary and quaternary plexed with subtilisin BPN' in the neutral-pH buffer, then structures of the inhibitors in this study were not dramati incubated under acidic conditions at 4-C to obtain the cally different from those of SSI. From Table II, SIL10 is modified forms of SIL inhibitors in which the reactive-site found to be most closely related to SIL14 in the SSI-family peptide bond was specifically cleaved. Proteins were pre inhibitors with relatively high similarity (91 amino acids cipitated with trichloroacetic acid, then subjected to re identical), and vice versa. SIL13 is most closely related to verse-phase HPLC to isolate the modified inhibitors. A plasminostreptin, though with lower similarity than that typical HPLC pattern for SIL13 is shown in Fig. 2. Peak A between SIL10 and SIL14. In the previous study, sub produced two bands on SDS-PAGE, whereas peak B grouping of the SSI-family inhibitors based on the sequence similarity corresponded to that based on the PI site residue produced one band, and their amino acid compositions were essentially identical (data not shown). These results in (6). In several cases, however, the sequence similarity

Vol. H I F, No. 5. 1994 1160 M . Terabe et al.

Fig. 3. Sequence comparison of protease inhibitors from vari common to more than eight of the twelve aligned inhibitors are ous Streptomyces species. SIL10 from S. thermotolerans, SIL14 shaded. The "f,-strand," "a,-helix," and so on are the regions of from S. azureus, SIL13 from S. galbus, plasminostreptin (PSN) from secondary structures in the SSI. The symbols "c" and "+" under the S. antifibrinolyticus, STI2 from S. longisporus, SIL4 from S. laven sequence are the exposed and hydrophobic-core-forming residues, dulae, SIL2 from S. parvulus, SIL3 from S. coelicolor, SLPI from S. respectively. Residue numbering is that for SSI. The arrow indicates lividans, SSI from S. albogriseolus, API-2c' from S. griseoincarnatus, the reactive site of inhibitors. and SILL from S. cacaoi are aligned. Conserved residues that are

TABLE II. Sequence similarity among the SSI-family inhibitors. Comparison was carried out for the amino acid sequence from Tyr7 of SSI, since each inhibitor has a different amino-terminal extension. The numbers of identical residues are shown. PSN, plasminostreptin.

between a Lys and an Arg-possessing inhibitor at the PI always be suitable. In the future, the establishment of a site is higher than that between two Lys-possessing in phylogenic tree based on statistical calculation will clarify hibitors. For example, SIL14 (P1, Lys) is more closely this point. related to SLPI (P1, Arg) than to SIL4 (P1, Lys). As the Sequence alignment in Fig. 3 indicates that many resi number of known SSI-family inhibitors increases, it is dues in the /3-sheet and hydrophobic core regions are well possible that subgrouping based on the P1 site may not conserved in the molecules of SIL10, SIL13, and SIL14 , as

J. Biochem. Subtilisin-Trypsin Inhibitors from Streptomyces 1161 in the case of SIL2, SIL3, SLPI, and so on. This is because chem. 57, 522-524 of their significant contribution to dimeric structure forma 5. Taguchi, S., Kikuchi, H., Suzuki, M., Kojima, S., Terabe, M., tion and high stability. The amino acid residues that had Miura, K., Nakase, T., and Momose, H. (1993) Streptomyces subtilisin inhibitor-like proteins are distributed widely in strep been thought, and have recently been shown, to be required tomycetes. Appl. Environ. Microbiol. 59, 4338-4341 for the inhibitory action of SSI are completely conserved. 6. Taguchi, S., Kojima, S., Terabe, M., Miura, K., and Momose, H. For instance, Cys71-Cys1O1 near the reactive site (12) and (1994) Comparative studies on the primary structures and Asn99 in the secondary contact region are required for inhibitory properties of subtilisin-trypsin inhibitors from Strep providing rigidity around the reactive site of SSI. The tomyces. Eur. J. Biochem. 220, 911-918 Arg29 interacting with the carboxy terminus (17), Trp86 7. Kojima, S., Terabe, M., Taguchi, S., Momose, H., and Miura, K. (1994) Primary structure and inhibitory properties of a pro between the /3-sheet and a,-helix (11) and Arg9O on the teinase inhibitor produced by Streptomyces cacaoi. Biochim. dimer interface (Kojima et al., in preparation) have also Biophys. Acta 1207, 120-125 been shown to be needed for the inhibitory action of SSI. In 8. Kojima, S., Obata, S., Kumagai, I., and Miura, K. (1990) other words, the completely conserved residues in all Alteration of the specificity of the Streptomyces subtilisin in members of the inhibitor family are likely to be absolutely hibitor by gene engineering. Bio/Technology 8, 449-452 necessary for their function as protease inhibitors. Deter 9. Kojima, S., Kumagai, I., and Miura, K. (1990) Effect on inhibi tory activity of mutation at reaction site P4 of the Streptomyces mination of the primary structures of many other SIL subtilisin inhibitor, SSI. Protein Eng. 3, 527-530 inhibitors and mutational analysis of SSI to investigate the 10. Kojima, S., Nishiyama, Y., Kumagai, I., and Miura, K. (1991) structural requirements for its inhibitory action will clarify Inhibition of subtilisin BPN' by reaction site P1 mutants of this point. In contrast, amino acid replacements are ob Streptomyces subtilisin inhibitor. J. Biochem. 109, 377-382 served on the molecular surface of the SIL inhibitors, 11. Tamura, A., Kanaori, K., Kojima, S., Kumagai, I., Miura, K., and Akasaka, K. (1991) Mechanisms of temporary inhibition in possibly due to their lesser importance, as in the case of Streptomyces subtilisin inhibitor induced by an amino acid other SSI-familv inhibitors (6, 7, 18). substitution, tryptophan 86 replaced by histidine. Biochemistry SIL10, SIL13, and SIL14 inhibited subtilisin BPN' 30,5275-5286 strongly with K; values of about 10-" M, like other 12. Kojima, S., Kumagai, 1., and Miura, K. (1993) Requirement for members of the SSI-family possessing a Lys residue at the a disulfide bridge near the reactive site of protease inhibitor SSI Pi site (6). These three inhibitors also exhibited inhibitory (Streptomyces subtilisin inhibitor) for its inhibitory action. J. Mol. Biol. 230, 395-399 activity toward trypsin, although this was weaker than that 13. Kojima, S., Furukubo, S., Kumagai, I., and Miura, K. (1993) toward subtilisin BPN'. This may be due to structural Effects of deletion in the flexible loop of the protease inhibitor SSI differences in substrate- between subtilisin (Streptomyces subtilisin inhibitor) on interactions with proteases. BPN' and trypsin, as discussed in the previous papers (7, 9, Protein Eng. 6, 297-303 18). Trypsin has only the S1-S3 subsite, whereas subtilisin 14. Masuda-Momma, K., Hatanaka, T., Inouye, K., Kanaori, K., BPN' has a hydrophobic S4 site with a large pocket. The Tamura, A., Akasaka, K., Kojima, S., Kumagai, I., Miura, K., and Tonomura, B. (1993) Interaction of subtilisin BPN' and presence of a bulky side chain at the P4 site of an inhibitor recombinant Streptomyces subtilisin inhibitors with substituted is considered to cause steric hindrance in the interaction P, site residues. J. Biochem. 114, 553-559 with trypsin. In contrast, the ability of the S4 pocket of 15. Kumazaki, T., Kajiwara, K., Kojima, S., Miura, K., and Ishii, S. subtilisin BPN' to accommodate a bulky side chain at the P4 (1993) Interaction of Streptomyces subtilisin inhibitor (SSI) with site of an inhibitor seems to result in the strong interaction Streptomyces griseus metallo-endopeptidase II (SGMP II). J. of the three SIL inhibitors with subtilisin BPN'. Biochem. 114, 570-575 16. Masuda-Momma, K., Shimakawa, Y., Inouye, K., Hiromi, K., Kojima, S., Kumagai, I., Miura, K., and Tonomura, B. (1993) This work was supported in part by a Grant-in-Aid for Scientific Identification of amino acid residues responsible for the changes Research from the Ministry of Education, Science and Culture of of absorption and fluorescence spectra on the binding of subtilisin Japan. BPN' and Streptomyces subtilisin inhibitor. J. Biochem. 114, 906-911 REFERENCES 17. Kojima, S., Fujimura, K., Kumagai, I., and Miura, K. (1994) Contribution of salt bridge in the protease inhibitor SSI (Strepto 1. Hiromi, K., Akasaka, K., Mitsui, Y., Tonomura, B., and Murao, myces subtilisin inhibitor) to its inhibitory action. FEBS Lett. S. (eds.) (1985) Protein Protease Inhibitor-The Case of Strep 337,195-199 tomyces Subtilisin Inhibitor (SSI), Elsevier, Amsterdam 18. Ueda, Y., Kojima, S., Tsumoto, K., Takeda, S., Miura, K., and 2. Sato, S. and Murao, S. (1973) Isolation and crystallization of Kumagai, I. (1992) A protease inhibitor produced by Strepto microbial alkaline protease inhibitor, S-SI. Agric. Biol. Chem. myces lividans 66 exhibits inhibitory activities toward both 37,1067-1074 subtilisin BPN' and trypsin. J. Biochem. 112, 204-211 3. Taguchi, S., Kojima, S., Kumagai, I., Ogawara, H., Miura, K., 19. Laemmli, U.K. (1970) Cleavage of structural proteins during the and Momose, H. (1992) Isolation and partial characterization of assembly of the head of bacteriophage T4. Nature 227, 680-685 SSI-like protease inhibitors from Streptomyces. FEMS Microbiol. 20. Crestfield, A.M., Moore, S., and Stein, W.H. (1963) The prepa Lett. 99, 293-298 ration and enzymatic hydrolysis of reduced and S-carboxymeth 4. Taguchi, S., Kikuchi, H., Kojima, S., Kumagai, I., Nakase, T., ylated proteins. J. Biol. Chem. 238, 622-627 Miura, K., and Momose, H. (1993) High frequency of SSI-like protease inhibitors among Streptomyces. Biosci. Biotech. Bio

V,,? : l6, No. 5, 1994 1162 M. Terabe et al.

Supplemental Materials

Table IS. Amino acid compositions of SIL10. SIL13 and SIL14. Table IVS. Amino acid compositions of SIL14-derived peptides subjected to automated Edman degradations.

aValues in parentheses are taken from the sequence data . bS-Pyridylethyleysteine . dNot determined.

aObtained after 24 h hydrolysis by 4 M methanesulfonic acid .

Table IIS. Amino acid compositions of SIL10-derived peptides subjected to automated Edman degradationa.

aValues in parentheses are taken from the sequence data. bS-Pyridylethylcysteine. cNot determined.

Table IIIS. Amino acid compositions of SIL13-derived peptides subjected to automated Edman degradationa.

Fig. 1S. SDS-polyacrylamide gel electrophoresis (A) and analytical gel-filtration (B) of SIL10, SIL13 and SIL14. (A) Purified SIL inhibitors were electrophoresed in SDS polyacrylamide gel using a gel concentration of 18.8%. Lane 1, standard proteins: cytochrome c (Mr 12,500) and bovine pancreatic (Mr 6,500); lane 2, SIL10; lane 3, SIL13; lane 4, SIL14. (B) SIL inhibitors or relative molecular mass markers were applied to a Superdex 75 column (1 x 35 cm) on an FPLC system. Elution was performed with 0.1 M Tris-HC1/0.5 H NaCl (pH 7.5) at a flow rate of 0.7 ml/min. Relative molecular mass markers were bovine serum albumin (BSA, Mr 89,000), soybean aValues in parentheses are taken from the sequence data. trypsin inhibitor (STI, Mr 20,100) and bovine pancreatic trypsin bS-Pyridylethylcysteine. cNot determined. inhibitor (BPTI, Mr 6,500).

J. Biochem. 6ubtitisin-Trypsin Inhibitors from Streptomyces 1163

Fig, 2S. HPLC of enzymatic digests of S-pyridylethylated SILIO obtained with lysyl endopeptidase (a) or S. aureus V8 protease (b).

Fig. 4S. HPLC of enzymatic digests of SIL14 and its fragment. The digests of S-pyridylethylated SIL14 obtained with lysyi endopeptidaae (a) or S. aureus V8 protease (c). (b) The digests of the L-1 peptide in (a) with arginylendopeptidase.

Fig. 3S. HPLC of enzymatic digests of SIL13 and its fragment. The digests of S-pyridylethylated SIL13 obtained with S, aureus VS protease (a) or lysyl endopeptidase (b). (c) The digests of the L-2 peptide in (b) with arginylendopeptidase.

t, ~ 16, No. 5, 1994