Purification and characterization of an extracellular

Purification and characterization of an extracellular thiol-containing serine proteinase from Thermomyces lartuginosusl

Biochem. Cell Biol. Downloaded from www.nrcresearchpress.com by Huazhong University of Science and Technology on 06/04/13 For personal use only.

SADIQHASNAIN,~ KHOSROW A D E L I AND , ~ ANDREW C.

STORER~

Institute for Biological Sciences, National Research Council of Canada, Ottawa, Ont., Canada KIA OR6 Received September 9, 1991

HASNAIN, S.. ADELI,K., and STORER, A. C. 1992. Purification and characterization of an extracellular thiol-containing serine proteinase from Thermomyces lanuginosus. Biochem. Cell Biol. 70: 117-122. An extracellular protease produced by the filamentous fungus Thermomyces lanuginosus has been purified and characterized. The results indicate that the enzyme, which we have called humicolin, is a thiol-containing serine protease with a molecular mass of 38 000 kilodaltons. Secretion of humicolin, which is glycosylated, is tightly regulated by protein substrates. Kinetic characterization has revealed that humicolin activity is highly dependent upon the deprotonation of a group with a pK, of 6.6 and that the enzyme has a specificity for phenylalanine in the P I position of the substrate. Key words: thiol-containing serine proteinase, characterization, kinetics, inhibitor, specificity.

HASNAIN, S., ADELI,K., et STORER, A. C. 1992. Purification and characterization of an extracellular thiol-containing serine proteinase from Thermomyces lanuginosus. Biochem. Cell Biol. 70 : 117-122. Nous avons purifit et caracttrist une prottase extracellulaire produite par un champignon filamenteux, Thermomyces lanuginosus. Les rtsultats montrent que l'enzyme, nommee humicoline, est une strine prottase renfermant du thiol et dont la masse moltculaire est de 38 000 kilodaltons. La stcrttion de l'humicoline, qui est glycosylte, est Ctroitement contr61te par les substrats prottiques. La caracttrisation cinttique rtvtle que I'activitt de I'humicoline est fortement dtpendante de la dtprotonation d'un groupe avec un pK, de 6,6 et que l'enzyme manifeste une sptcificitt pour la phtnylalanine a la position P, du substrat. Mots elks : strine prottinase renfermant du thiol, caracttrisation, cinttique, inhibiteur, sptcificitt. [Traduit par la rtdaction]

Introduction Shenolikar and Stevenson (1982) reported the purification and partial characterization of a thiol proteinase from the thermophilic fungus Thermomyces lanuginosus (Humicola lanuginosa).They followed the lead from Ong and Gaucher (1973), who first reported the presence of an extracellular, thermostable thiol proteinase in cultures of this organism. The enzyme that was purified almost to homogeneity (Shenolikar and Stevenson 1982) had an M, of 23 700 by gel filtration and sedimentation, and was shown to cleave at the C-terminal side of hydrophobic amino acid residues. Its substrate specificity was shown to be unique when compared with papain. Our interest in cysteine proteinases led us to investigate further the T. lanuginosus enzyme that we call humicolin. In our earlier attempts to purify humicolin we followed the published procedure of Shenolikar and Stevenson (1982). ABBREVIATIONS: M,, relative mass; CDR, cell debris remover; SDS-PAGE, sodium dodecyl sulfate - polyacrylamide gel electrophoresis; &ME, 0-mercaptoethanol; IEF, isoelectric focusing; NP-40, Nonidet P-40; TCA, trichloroacetic acid; ConA, concanavalin A; Cbz-Gly-pNP, N-benzyloxycarbonylglycinep-nitrophenyl ester; kDa, kilodalton(s); PMSF, phenylmethylsulfonyl fluoride; pCMB, p-chloromercuribenzoate; pCMPSA, p-chloromercuriphenylsulfonic acid; AMPSF, p-amidinophenylmethylsulfonyl fluoride; TPCK, N-tosyl-L-phenylalanyl chloromethyl ketone; DTNB, 5,s'-dithiobis(2-nitrobenzoic acid); E-64, (~-3-trans-carboxyoxiran-2-carbonyl)-~-leucyl-agmatin; NEM, N-ethylmaleimide; TLCK, N-tosyl-L-lysinechloromethyl ketone. 'NRCC 31 969. ' ~ u t h o rto whom all correspondence should be addressed. 'present address: Biochemistry Department, University of Windsor, Windsor, Ont.. Canada N9B 3P4. 4~resentaddress: Biotechnology Research Institute, National Research Council of Canada, Protein Engineering Section, 6100 Royalmount Avenue, Montrtal, Que., Canada H4P 2R2. Printed in Canada / Imprim&nu Canada

The principle of their purification relied on the specific binding of the proteinase to an organomercury-Sepharose column. We were unable to use this covalent chromatography method, in that we could not detect specific binding of the enzyme to the column. Consequently, we have developed an alternative method to purify the proteolytic activity found in culture filtrates of T. lanuginosus. In this paper we describe the purification and characterization of the extracellular protease of T. lanuginosus and present data which show that this enzyme is a thiol-containing serine proteinase and not a cysteine proteinase as previously thought. Other examples of thiol-containing serine proteinases previously reported in the literature include enzymes from Bacillus thuringiensis (Epremyan et al. 1980; Stepanov et al. 198l), Thermoactinomyces vulgaris (Stepanov et al. 1981; Meloun et al. 1985), Streptomyces rectus var. proteolyticus (Mizusawa and Yoshida 1973), and Tritirachium album Limber (Epremyan et al. 1980; Betzel et al. 1988). Materials and methods Thermomyces lanuginosus (ATCC strain 3078) was grown in potato dextrose agar slants, which were then used to inoculate 20-mL precultures (1 L of medium contained 20 g casein, 2 g NaCI, 1 g KCI, 1 g KH2P04,0.5 g MgS04.7H20, 0.01 g FeS04.7H20, 0.3 mg ZnS04.7H20, 1 g glucose, and 0.1 g yeast extract at pH 7.4). these precultures were grown for 2 days at 45°C. The cultures were checked for good sporulation by light microscopy and then used to inoculate 400 mL of the same medium in 4-L shake flasks. Cultures were shaken at 150 rpm and 45°C and assayed periodically for protease activity until maximal proteolytic activity was achieved. At this point (normally 60 h), the cultures were harvested and filtered through Whatman No. 1 filter paper. For larger preparations, 4-L cultures were grown in an 8-L fermenter with 1-2 L/min aeration. These cultures were inoculated with 800 mL of inoculum grown in two 4-L shake flasks. The culture filtrate was treated with cell debris remover (CDRcellulose. Whatman) and concentrated 20 times using a Pellicon concentrator fitted with a Millipore PTGC 10 000 NMWL mem-

BIOCHEM. CELL BIOL. VOL. 70,

1992


TABLE1. Purification of humicolin Purification step

Volume

Protein (mg)

Activity (Cbz-Gly-pNP assay) (u/~L)

Filtrate CDR-cellulose and concentration PhenylSepharose

1585

-

3.45

236

533

64.5

63.3

Cultivation time (hours) FIG. 1. The effect of nutrients on growth of Thermomyces lanuginosus. Protease activity in culture filtrates was determined spectrophotometricallyusing the substrate Cbz-Gly-pNP. One unit of activity is the amount of enzyme that gave a change of 1 absorbance unit in 1 min. 0,2% casein and 0.1% glucose; A, 1% casein and 1.0% glucose; , 0.5% casein and 1.5% glucose; , 2.0% glucose; A, 2.0% lactose. brane cassette. The concentrate was again clarified using CDR and applied to a 2.5 x 20 cm phenyl-Sepharose column preequilibrated with 0.25 M NaCl - 25 mM sodium phosphate - 1 mM EDTA (pH 8.5). The column was washed with one bed volume (80 mL) of the preequilibrated buffer and then with three bed volumes of a wash buffer containing 0.1 M sodium phosphate and 5 mM EDTA (pH 7.0), followed by one bed volume of the wash buffer containing 30% ethylene glycol. The protease activity was then eluted upon increasing the ethylene glycol content of the buffer to 50%. Samples containing proteolytic activity were treated with mercuric chloride prior to preparation for SDS-PAGE, which was performed using a modification to the method of Laemmli (1970). The modification involved the removal of 8-ME from the sample buffer to prevent autodigestion. Gels were stained with 0.25% Coomassie brilliant blue in 7% acetic acid - 40% methanol and destained in 7.5% acetic acid - 5% methanol. IEF gels consisted of 5% acrylamide and 0.2% pH 3.5-10 ampholine (Bio-Rad Laboratories). Samples were solubilized in a sample buffer containing 9.5 M urea, 2% NP-40, 5% 8-ME, and 2% ampholines. Gels were prefocused at 200-400 V for 1 h and focused at 400 V for 16 h. The gels were stained and destained as above. Staining for proteolytic activity was performed on IEF gels according to the method of Ward (1975). Immediately after electrophoresis, gels were washed with distilled water for 15 min and then incubated with 8% cytochrome c in 6 mL of 8 M urea and 6 mL of 33 rnM 2-amino-2-methyl-1,3-propanediol for 1 h at 37°C. The gels were then removed and incubated for 1 h with 4 M urea 20 mM 2-amino-2-methyl-1,3-propanediol. Finally the gels were incubated with 12.5% TCA for 30 min.

Specific A,,,/ activity Total r n ~ ( / 2 8 0 ) activity 13.9

0.248

5488

15.6

6.9

2.26

3681

35.0

0.25

137.8

2258

Three milligrams of humicolin were dissolved in ConA buffer (10 mM Tris-HC1 (pH 7.3), 300 mM NaCI, 1 mM MgCl,, 1 mM CaCl,, and 1 mM MnClJ and applied to a 5-mL column of ConA-Sepharose packed in the same buffer. After application of the sample, the column was extensively washed with the ConA buffer and the eluate was monitored for protein and protease activity. The bound fraction was eluted from the column by 0.8 M methyl 8-D-mannopyranoside in ConA buffer. The amino acid composition, after hydrolysis for 20 h at 110°C in 6 M HCI, was determined using a Durrum D-500 analyzer. The tryptophan content was determined after hydrolysis in 4 M methane sulfonic acid containing 0.2% 3-(2-aminoethyl)indole, whereas the cysteine was determined as cysteic acid after treatment of the protein with performic acid. The molar absorptivity of the enzyme at 280 nm was calculated to be 36 730 using the amino acid composition and the extinction coefficients of the relevant amino acids. This value was used to calculate enzyme concentration for kinetic experiments. Two enzyme assays were routinely used: a discontinuous caseinolytic assay at pH 8.0 (Ong and Gaucher 1973), in which the extent of casein digestion was determined from the Azsoof the acid-soluble fraction, and also a continuous spectrophotometric assay using Cbz-Gly-pNP as the substrate, at 2S°C and pH 8.0 (Shenolikar and Stevenson 1982). Kinetic measurements were carried out spectrophotometrically under the conditions given in the text. Tests were conducted to determine the ideal monitoring wavelength for each of the different substrates at the various pH values used. Below pH 6, absorbance was monitored at 270 nm and at pH 6 and above it was monitored at 400 nm. Difference extinction coefficients were determined from the measured optical densities after allowing selected reaction mixtures to go to completion. An evaluation of the error type indicated that the most appropriate weighting system for data analysis was one in which the errors are proportional to the measured velocity. Data were fitted by the nonlinear regression data analysis program Enzfitter, of Leatherbarrow, supplied by Elsevier-Biosoft. Kinetic constants for Cbz-Gly-pNP at different pH values were calculated from kinetic experiments by varying the concentration of the substrate from 10 to 150 pM.

Results Variations in the nature and concentration of the culture nutrients have a profound effect on the production of protease activity. Figure 1 shows the effects of various concentrations of casein on protease production. In the presence of 0.5-2% casein, protease activity appeared and peaked at about 45-70 h. The levels of protease activity decreased afterwards. This drop was more significant in the cultures with a low initial concentration of casein. The 0.5% casein culture showed a 40% drop in protease activity only 21 h after reaching its maximal activity. No protease activity was observed when the organism was grown on glucose or lactose in the absence of casein (see Fig. 1). These obsemations suggest a tight regulation of protease expression and (or) secretion induced by external protein substrate.

119

HASNAIN ET AL-


a

IEF + (+-

FIG. 2. Determination of humicolin purity using SDS-PAGE. Lane 1, molecular mass markers. Lanes 2 and 8, culture filtrate. Lane 3, cell extract. Lanes 4, 5, 6,and 7,purified humicolin. The gel was stained with Coomassie blue. Mass units are in kilodaltons.

A simple two-step purification method was developed. Clarification of the culture filtrate with CDR-cellulose resulted in a 10-fold purification. Elution from the phenylSepharose column resulted in a high specific activity with a 555-fold purification and a yield of 41% (Table 1). SDSPAGE under nonreducing conditions revealed that humicolin was a minor component in the crude filtrate (lanes 2 and 8 of Fig. 2) and that the purification procedure resulted in a homogeneous protein with a molecular mass of about 38 000 kDa (lanes 4-7). When &ME was present in the sample buffer, humicolin readily autodigested on exposure to SDS and heat (data not shown). Purified humicolin was analyzed by two-dimensional electrophoresis, SDS-PAGE, and IEF to reveal any charge heterogeneity (Fig. 3). A single major component was observed along with a trace contaminant. The major polypeptide had an approximate isoelectric point of 5.0. An in situ assay of a parallel IEF lane showed proteolytic activity (Fig. 3a). All of the proteolytic activity in a humicolin sample applied to a ConA Sepharose column bound to the column. This indicates that the humicolin has a significant amount of glycosylation. The presence of 2-glucosamine has been determined by amino acid analysis. The amino acid composition of humicolin is compared with other serine proteases in Table 2. No major difference is apparent when comparing humicolin with the other examples in the table. Of the proteases shown, humicolin and proteinase K contain five residues of cysteine, while the other enzymes contain one or none, with the exception of thermoycolin which contains two. Table 3 contains the results of inhibition experiments. The inhibitors were mixed with the enzyme and aliquots were removed and assayed by the continuous spectrophotometric method at specific time intervals. The most effective

FIG. 3. Two-dimensional gel electrophoresis of humicolin. (a) IEF gel

electrophoresis was performed in the first dimension

with an ampholine pH gradient from 3.5 to 10. The top gel was

used for activity staining, while the bottom gel was stained with Coomassie blue. (b) SDS-PAGE was performed in a slab gel in the second dimension and stained with Coomassie blue. inhibitors, H ~ and ~ PMSF, + resulted in a complete loss of activity in less than 3 h. Significant inhibition after 24 h was also observed with pCMB, pCMPSA, AMPSF, and TPCK. Common thiol reagents such as DTNB did not react with this thiol group, and significant inhibition was only observed ~ organomercuric + compounds such as pCMB with H ~ and and pCMPSA. The specific cysteine proteinase inhibitor E-64 (Rich 1986) did not affect activity. Peptide aldehydes are potent inhibitors of both serine and cysteine proteases, forming covalent hemiacetal and hemithioacetal adducts with the active site serine and cysteine residues, respectively. Table 4 contains the results of an inhibition study of humicolin using four peptidyl aldehyde inhibitors and pepstatin, a peptidyl acid protease inhibitor. The results clearly show that peptidyl aldehydes are also potent inhibitors of humicolin. From the results in Table 4 it is possible to infer information regarding the subsite specificity of the enzyme. For example, a comparison of chymostatin and antipain indicate that an aromatic side chain is preferred in the P I position. A comparison of antipain and leupeptin indicates that the preferences of the S3 and S4 subsites are for basic (arginine) and aromatic (phenylalanine) residues, respectively. This S3 and S4 subsite specificity must account to a large extent for the inhibition of humicolin by antipain, which has an unfavourable basic residue at P I (see above and below). The invariability of the kc,, values given in Table 5 for the various substituted benzylglycine substrates indicates

BIOCHEM. CELL BIOL. VOL. 70, 1992

TABLE2. Comparison of the amino acid composition of humicolin with other microbial serine proteases


-

Residues/molecule

Amino acid

Humicolin*

BTH

5 3 39 19 24 19 12 37 38 23 19 15 13 8 4 20 8 9

1 1 33 21 27 24 12 32 37 24 14 12 15 3 5 11 5 5

CY~ Met Asx Glx Ser Thr Pro GlY Ala Val Ile Leu TYr Phe T~P LY~ Arg His

BCE 1

1 33 20 32 26 11 34 35 21 14 12 14 2 6 12 2 5

THA

STR

BAM

TMN

K

1 1 33 22 24 16 16 30 38 20 14 8 16 4 4 11 4 6

1 1 35 22 30 14 12 34 44 23 14 9 15 3 8 10 5 4

0 5 28 13 37 15 14 33 37 30 13 15 10 3 3 11 2 6

2 2 39 16 33 21 10 57 45 24 20 14 8 3 6 4 14 7

5 5 31 21 36 12 9 34 33 10 10 13 17 6 2 8 12 4

Nore: BTH, Bacillus thuringiensis; BCE, Bacillus cereus; THA, Thermoactinomyces vulgar&; STR. Strepromyces rectus; BAM, subtilisin BPN; TMN, thermomycolin; K, proteinase K. *M, = 31 858 (on basis of amino acid composition).

TABLE3. Inhibition of humicolin by common protease inhibitors qo activity remaining

Inhibitor

After 3 h

After 24 h

Iodoacetarnide Iodoacetic acid pCMB pCMPSA NEM PMSF AMPSF DTNB TLCK TPCK E-64 H ~ ~ +

-

92 94 9.3 0 74 0 60 89 110 48 87 0

0

-

0

NOTE: The final concentration of inhibitors was 0.1 mM except pCMB, PMSF, and TPCK. For these 20 pL of a saturated solution of inhibitor was used after filtering. Enzyme solution (I0 pL, 0.5 mg/mL) was added to inhibitor solution that was made to a final volume of 110 pL with 50 mM phosphate - 1 mM EDTA (pH 7.0).For the reaction with H~'', EDTA was not used. Activity was measured at pH 7.0 in 50 mM phosphate using the continuous spectrophotometricassay described in the text. For pCMB, inhibitor and enzyme were mixed at pH 8.0 in 50 mM phosphate.

(with these substrates) that, with humicolin as with other serine proteases, deacylation is rate limiting. The pH profiles of kcat/Kmand kcatobtained using Cbz-Gly-pNP with humicolin are shown in Figs. 4 and 5, respectively. The solid line represents the best fit of the data to an equation defining a single ionization process (Alberty and Massey 1954). For both kcat/Kmand kcat, a pKa of approximately 6.6 was obtained, and as with other serine proteases, this pKa probably reflects the ionization of an active-site histidine side

TABLE 4. Inhibition of humicolin by peptide derivatives Inhibitors ChymostatinT ~nti~ain~ Elastinal Leupeptin Pepstatin

I50

(MI*

< 0.1 x

*Concentration of inhibitor that results in 50% inhibition of the enzyme under assay conditions [I] > > [EF For these inhibitors, owing to a combination of assay sensitivity and very tight binding of inhibitor, the condition [I] > > [El could not be satisfied.

chain. The decrease in kcat at high pH values was reproducible; however, instability of the substrate at high pH prevented further characterization of the decrease. In Table 6, it is clear from the specificity constant (kcat/Km) values that phenylalanine is the preferred amino acid in the P I position of the substrate, whereas the basic amino acid lysine is the least preferred. This result confirms the finding already described above for the S1 subsite specificity obtained from the peptidylaldehyde inhibitor study. From Table 6 it is also clear that humicolin exhibits a broad specificity in its S1 subsite. The order of preference is aromatic amino acids (Phe,Tyr) > non-&branched hydrophobic amino acids (Leu) > small side chains (Ala,Gly) > other amino acids. Discussion The thermophilic fungus Thermomyces lanuginosus, upon induction by protein in the medium, secretes a protease into its extracellular environment. The protease, which we have called humicolin, is secreted about 45-70 h after initiation of the culture and after reaching a maximum its activity

HASNAIN ET

AL.

121


TABLE5. Kinetic constants obtained for phenyl benzoylglycine ester substrates substituted on the phenyl ring, at 22OC in 50 mM phosphate buffer (pH 7.0) with 5 mM EDTA and 2% CH,CN

Substituent

Concn. range of substrate bM)

kca,/Km (mM 'es-')

kcar (s-)

TABLE6. Kinetic constants obtained for the humicolin-catalysed hydrolysis of Cbz-(amino acid)-pNP esters at 22°C in 50 mM phosphate buffer (pH 7.0), with 5 mM EDTA and 2% CH,CN. The reactions were monitored at 400 nm

Amino acid

Concn. range of substrate bM)

~cN/K~ (mM- ' - s -'I

FIG. 4. A pH profile for k,,/Km obtained for humicolin with Cbz-Gly-pNP as substrate. In the pH range of 5.2-5.6, the buffer consisted of 50 mM sodium acetate, 5 mM EDTA, and 200 mM NaCl. In the pH range of 5.8-8, the buffer consisted of 50 mM sodium phosphate, 5 mM EDTA, and 200 mM NaCl. The final concentration of acetonitrile in the reaction buffer was 2%.

Phe TY~ Leu Ala G~Y Asn T~P Val LY~

decreases slowly with time (Fig. 1). Secretion of proteolytic activity is tightly regulated by protein substrates, since no protease activity appears in the medium in the absence of casein and activity disappears once the protein substrate is utilized. Humicolin was purified to homogeneity in high yield by a simple two-step method including hydrophobic affinity chromatography with phenyl8epharose. Its molecular mass based on SDS-PAGE and amino acid composition is 38 000 and 32 000 kDa, respectively. While part of the discrepancy is probably due to the inherent error in molecular mass estimation by SDS-PAGE, glycosylation may account for the major difference between the two methods. As well, molecular mass estimation by SDS-PAGE may be even less reliable if not all disulfides are reduced. In our hands, the incorporation of &ME in the sample buffer led to rapid autodigestion of the enzyme. Only when the enzyme was inhibited by mercury and then denatured with SDScontaining sample buffer without 0-ME were we successful in visualizing a single band suggesting homogeneity. Humicolin was shown to bind to a ConA column and was eluted only with high concentrations of mannopyranoside (0.8 M), suggesting that the enzyme is glycosylated. As well, amino acid analysis revealed at least one residue of 2-glucosamine, further suggesting at least one N-linked glycosylation site. The amino acid composition of humicolin shows similarity to the free thiol containing class of subtilisin-like proteases

FIG. 5. A pH profile for k,, obtained for humicolin with CbzGly-pNP as substrate. Experimental details are given in the text and in Fig. 4.

(Jany et al. 1986), with the exception of a few prominent differences (Table 2). Most of the enzymes in this class examined to date contain one cysteine residue, with the exception of humicolin and proteinase K. Five residues of half cystine were determined for humicolin by amino acid analysis. In proteinase K, there are two disulfide bridges and a free cysteine positioned below the active site in the structure determined by X-ray crystallography (Betzel et al. 1988). An interesting feature of this class of enzyme is that the free thiol is essential for activity. Whether or not the thiol is involved in the catalytic mechanism is unclear. Betzel et al. (1988) show in their X-ray structure of proteinase K that the free cysteine is positioned beneath the active site. This in conjunction with inhibition data led them to suggest a possible catalytic function for free cysteine.


122

BIOCHEM. CELL BIOL. VOL.

Humicolin is similar to the other enzymes in its class in that it is inhibited by some reagents which react with free thiols, such as ~ g ~ pCMB, ' , or pCMPSA, but is not inhibited by alkylating agents such as iodoacetic acid, iodoacetamide, or DTNB. This suggests that the free cysteine is at least partially buried. Humicolin is also inhibited by serine protease specific PMSF and not by E-64, which is specific for true cysteine proteases. Finally, the dependence of kc, on the ionization of a group on the enzyme with a pK, of about 6.6, probably a histidine, is also consistent with a serine protease like mechanism. It is possible that this property of humicolin of inhibition by mercuric compounds may have led to the mistaken classification of this enzyme as a true thiol protease (Shenolikar and Stevenson 1982; Ong and Gaucher 1973). The other possibility is that the enzyme previously reported is different from the one that we identified and isolated. However, we were not able to detect any true thiol protease activity in the culture medium in any of our preparations. Furthermore, recent progress in the protein sequence analysis of humicolin (data not shown) in our laboratory has confirmed that this enzyme is a member of the subtilisin family of serine proteinases. This work will be published as soon as the full sequence is obtained. Enzyme specificity studies, as determined by reaction with inhibitors and synthetic substrates, show that while a broad range of side chains can be tolerated aromatic side chains are preferred in the P, position. In the P3 and P4 positions, arginine and phenylalanine are preferred, respectively. Acknowledgements The authors acknowledge the technical assistance of E. Armstrong and T. Hirama in obtaining the kinetic results, and M. Yaguchi for performing the amino acid analysis.

70, 1992

Alberty, R.A., and Massey, V. 1954. On the interpretation of the pH variation of the maximum initial velocity of an enzymecatalyzed reaction. Biochim. Biophys. Acta, 13: 347-353. Betzel, C., Bellemann, M., Pal, G.P., et al. 1988. X-ray and modelbuilding studies on the specificity of the active site of proteinase K. Proteins: Struct. Funct. Genet. 4: 157-164. Epremyan, A.S., Chestukhina, G.G., Azizbekyan, R.R., et al. 1980. Extracellular serine proteinase of Bacillus thuringiensis. Biokhimiya (Moscow), 46: 920-929. Jany, K.-D., Lederer, G., and Mayer, B. 1986. Amino acid sequence of proteinase K from the mold Tritirachium album Limber. FEBS Lett. 199: 139-144. Laemmli, U.K. 1970. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature (London), 227: 680-685. Meloun, B., Baudys, M., Kostka, V., et al. 1985. Complete primary structure of thermitase from Thermoactinomyces vulgaris and its structural features related to subtilisin-typeproteinases. FEBS Lett. 183: 195-200. Mizusawa, K., and Yoshida, F. 1973. Thermophilic Streptomyces alkaline proteinase: the role of a sulphydryl group and the conformational stability. J. Biol. Chem. 248: 4417-4423. Ong, P.S., and Gaucher, G.M. 1973. Protease production by thermophilic fungi. Can. J. Microbiol. 19: 129-133. Rich. D.L. 1986. Inhibitors of cysteine proteinases. In Proteinase inhibitors. Edited by A.J. Barrett and G. Salvesen. Elsevier, Amsterdam. pp. 153-178. Shenolikar, S., and Stevenson, K.J. 1982. Purification and partial characterization of a thiol proteinase from the thermophilic fungus Humicola lanuginosa. Biochem. J. 205: 147-152. Stepanov, V.M., Chestukhina, G.G., Rudenskaya, G.N., et al. 1981. A new family of microbial serine proteinase? Structural similarities of Bacillus thuringiensis and Thermoactinomyces vulgaris extracellular proteinases. Biochem. Biophys. Res. Commun. 100: 1680-1687. Ward, C.W. 1975. Aminopeptidases in webbing clothes moth larvae. Properties and specificities of enzymes of highest electrophoretic mobility. Austr. J. Biol. Sci. 28: 439-445.