MYCOSES
35, 343-348 (1992)
ACCEPTED:MAY4, 1992
Effect of reducing agents on proteolytic and keratinolytic activity of enzymes of Microsporum gypseum Die Wirkung von Reduktionsmitteln auf die proteolytische und keratinolytische Aktivitat der Enzyme von Microsporum gypseum
Key words. Microsporum gypseum, proteolytic enzymes, keratinolysis, reducing agents. Schliisselworter. Microsporum gypeum, proteolytische Enzyme, Keratinolyse, Reduktionsmittel.
Summary. The effect of sodium sulphite, cysteine, glutathione, mercaptoethanol and dithioerythritol (0.1-10 mmol I - ’ ) on the activity of proteases of Microsporum Qpseum was studied using azocasein, cross-linked bovine serum albumin and keratin as substrates. With the substrate without disulphide bonds (casein) no stimulation was found, and reducing agents inhibited proteolysis in most cases. With the remaining two substrates, all substances enhanced the activity of proteases probably through the cleavage of the substrate disulphide bonds. Sulphite was more effective than the four used thiols and enhanced the activity against serum albumin up to 3.2 times and against keratin up to 2.9 times. Using sulphitolysed sheep wool, keratinolytic activity increased after sulphitolysis of more than 20% of disulphide bonds. With the fully sulphitolysed wool the activity increased 43 times. The obtained results support the author’s hypothesis on keratin degradation by sulphite excretion prior to attack by fungal proteases. Stimulation of proteolysis and keratinolysis by cleavage of disulphide bonds is not specific for dermatophytic proteases because trypsin and pronase behaved similarly in the experiments.
sporum gypseum untersucht. Azokasein, vernetztes Rinderserumalbumin und Keratin dienten als Substrate. Bei dem Substrat ohne Disulfidbindungen (Kasein) wurde keine Stimulierung der Proteolyse durch Reduktionsmittel gefunden; diese wurde meistens gehemmt. Beim Serumalbumin und Keratin wurde die Aktivitat der Proteasen durch alle Reduktionsmittel erhoht, wahrscheinlich infolge der Spaltung von Disulfidbindungen des Substrats. Sulfit war wirksamer als die vier benutzten Thiole und erhohte die Aktivitat fur Serumalbumin bis 3.2 ma1 und fir Keratin bis 2.9 mal. Mit sulfitolysierter Schafwolle als Substrat wurde die keratinolytische Aktivitat nach einer Spaltung von mehr als 20% aller Disulfidbindungen gesteigert. Bei der vollig sulfitolysierten Wolle wurde die Aktivitat 43 ma1 erhoht. Die Ergebnisse stimmen mit der Hypothese des Autors uber den Keratinabbau durch Sulfitausscheidung vor dem Angriff der Pilzproteinasen iiberein. Die Stimulierung der Proteolyse und Keratinolyse durch Spaltung von Disulfidbindungen ist nicht fur die Dermatophyten-Proteinasen spezifisch, weil Trypsin und Pronase ein ahnliches Verhalten aufwiesen.
Es wurde die Wirkung von Natriumsulfit, Cystein, Glutathion, Merkaptoethanol und Dithioerythritol (0.1-10 mmol 1- I ) auf die Aktivitat der Proteinasen von Micro-
Introduction
Zusammenfassung.
~~
~~
~
~
Department of Biology, Medical Faculty, Palacky University, Olomouc, Czechoslovakia. Correspondence: Dr J. Kunert, Department of Biology, Medical Faculty, CS-77515 Olomouc, Czechoslovakia.
Proteolytic enzymes of keratinolytic fungi, although often termed ‘keratinases’, are incapable of dissolving by themselves the native hard keratin. When using cut human hair or wool as substrate, the loss of weight was not measurable or did not exceed the expected content of nonkeratins (5-10%) in those tissues [l-41. In guinea
344
J. KUNERT
pig hair, a purified 'keratinase' of Trichophyton mentagrophytes caused a 30% loss in weight [ l , 51. However, mainly medulla was dissolved which is not a true keratin [6]. Therefore the hypothesis was postulated that a complete hydrolysis of keratin could be achieved only after its denaturation by cleavage of a part of the disulphide bridges representing the main source of the extraordinary resistance of this scleroprotein. It is well known that dermatophytes growing on media with free or combined cystine metabolize this substance intensively and utilize it not only as a source of sulphur but also as a source of carbon and nitrogen [7-111. Excess sulphur is excreted back to the medium in the oxidized form as sulphate and/or sulphite. Sulphite reacts in a neutral to alkaline solution with cystine cleaving it to cysteine and S-sulphocysteine according to the equation
+
CYS-SS-CYS HSO, +Cys-SH
+ CYS-SSO,
(sulphitolysis, see Milligan & Swan [12]). Since this reaction takes place also with cystine combined in proteins including keratin, the hypothesis was presented that keratin is denatured prior to the attack by proteases by excretion of sulphite and by sulphitolysis of its disulphide bonds [9]. This hypothesis was supported by experiments revealing the presence of expected products of sulphitolysis (peptides containing S-sulphocysteine) in the medium of cultures cultivated on keratin [2, 13, 141 and also in the attacked substrate itself [15, 161. The presumption on enhanced availability of the denatured keratin to fungal proteases was based mainly on current knowledge about the activation of proteolysis by reduction of disulphide bonds of the substrate (see, e.g., Ref. 17) and on the experiments of Everett et al. [18] dealing with the degradation of sheep wool by non-specific proteases in the presence of reducing agents. The effect of dermatophyte proteases on keratin and other proteins in the presence of sulphite or other reducing agents has not yet been studied in detail and it is therefore the subject of the present paper.
Materials and methods As a preparation of proteolytic enzymes, dialysed filtrates of cultures of the dermatophyte Microsporum gypseum (strain MG 155=ATCC 14544) on keratin was used. The cultures in 100 ml conical flasks contained 0.2 g of cut child's hair (sterilized by autoclaving at 115 "C for 20 min), 10 ml of 1 g 1-' KH,PO,, p H 6.5, and about 50,000 fungal spores as inoculum. After 15 days of
stationary cultivation at 28 "C the cultivation fluid was collected by filtration, set to pH 6, dialysed against distilled water and kept in small amounts at - 20 "C. For determination of proteolytic activity on azocasein (synthesized according to Langner et al. [19]), the incubation mixture contained 1 ml of 2% substrate solution, 0.2 ml of enzyme preparation and 1.8 ml of 0.1 moll- Tris-HC1 buffer, pH 7.2. After 60 min at 40"C, 0.3 ml of 40% trichloracetic acid was added and the precipitate filtered off after a further 30 min. The activity was determined as absorbance at 440 nm, compared with the control containing the enzymes denatured by heat ( 5 min, 100°C). Another substrate was bovine serum albumin (BSA) cross-linked by epichlorohydrine and dyed with Remazol Brilliant Blue, used in the commercial kit 'S-test Protease' (Slovakofarma, Hlohovec, Czechoslovakia). Thirty milligrams of this chromolytic substrate was put into a 3 ml tube to which 0.5 ml of enzyme preparation and 2.3 ml of Tris-HC1 buffer (pH 8.0) were added. The plugged tube was placed on a roller keeping the substrate in suspension. After 2 h at 23 & 1"C, the substrate was filtered off and the activity determined as an increment of absorbance at 585 nm compared with the control. Keratinolytic activity was determined essentially after Yu et al. [l] on cut defatted hairs of albino guinea pigs. A 3 ml tube contained 50 mg substrate, 1.8 ml enzyme preparation, 1 ml 0.2 moll-' borate buffer (pH 9.0) and 50 pl toluene. Incubation was the same as in the method described above, but lasted for 6 h. Absorbance was measured at 280 nm. Similar experiments were performed with cut sheep wool (see below). Besides fungal proteases, crystalline bovine trypsin (Spofa, Prague, Czechoslovakia) and pronase (Boehringer, Mannheim, Germany) were used. Trypsin was applied at a final concentration of 2 pg ml I - ' with dyed BSA and both enzymes at a concentration of 10 pg ml 1-' with keratin as substrate. As reducing agents L-cysteine (Loba-Chemie, Vienna, Austria), reduced glutathione, 2-mercaptoethanol (Koch & Light, Colnbrook, U K ) , dithioerythritol (Serva, Heidelberg, Germany) and sodium sulphite p.a. (Lachema, Brno, Czechoslovakia) were used. These compounds were added to the incubation mixture as solutions in buffer to reach the final concentrations of 0.1, 0.3, 1.0, 3.0, 10 (and 30 and 100, if necessary) mmol 1-'. Activity of proteases was always measured against a blank containing the same concentration of reducing agent. All the used solutions were oxygen-protected by flushing through pure nitro-
'
mycoses 35, 343-348 (1992)
REDUCING AGENTS AND
gen, and the incubation mixtures were kept under N,. Keratin substrate denatured by sulphitolysis was prepared from sheep wool (Australian Merino, cystine content 9.4%) according to Wolfram & Bruggemans [20]. Washed and cut wool was processed by 1 mol I - ' Na,SO, and 0.5 mol 1-' sodium tetrathionate (pH 6.5, 35 "C, 5 g wool per 1 litre). After various time intervals (30 min up to 24 h), wool displayed varying degrees of cystine conversion to S-sulphocysteine up to complete sulphitolysis. The content of S-sulphocysteine in the sulphitolysed wool was determined by our method based on its oxidation by performic acid to cysteic acid and inorganic sulphate [12]. Between 10 and 100 mg of thoroughly washed and dried substrate was suspended in 10ml of 5% performic acid. After 16 h of continuous movement the liberated sulphate was determined turbidimetrically with BaCl, [2 13. Performic acid was prepared as the mixture 30% H,O,-formic acid p.a. ( l / l O , vol/vol) which was kept for 1 h at 23 f 1°C. For calculation of the degree of cleavage of disulphide bonds (conversion to S-sulphocysteine) the cystine content of the used wool was derived from elementary sulphur analysis.
Results Table 1 illustrates that none of the reducing substances stimulated hydrolysis of azocasein by proteases of the dermatophyte. By contrast, inhibitory effects were recorded, being weaker with sodium sulphite and stronger with glutathione, mercaptoethanol and cysteine. With dithioerythritol and with cysteine at a concentration of 10 mmol 1-', reducing agents bleached the substrate colour so that proteolytic activity could not be measured precisely (results not shown). The reaction of insoluble chromolytic substrate (dyed BSA) was quite different (Table 2). Sodium sulphite stimulated proteolysis even at a concentration of 0.1 mmol 1-I, and at 10 mmol 1-' the ~~
Table 1. Proteolytic activity of enzymes from Mzcrosfiorum gypseum on azocasein in the presence of reducing agents (data are given as
yo proteolytic activity) Concentration of reducing agent (mmol I - ' )
Sulphite Cysteine Mercaptoethanol Glutathione
0
0.1
0.3
1.0
3.0
10.0
100 100
98 95 92 82
102 92 87 83
103 86 77 79
94 58 61 71
74 37 47
100 100
mycoses 35, 343-348 (1992)
FUNGAL KERATINOLYSIS
345
activity increased 3.2 times. Higher concentrations were less effective and at 100 mmol 1-' (about 12.6 mg ml-I) the sulphite was strongly inhibitory. With other substances the stimulation was weaker and reached its maximum at lower concentrations. Higher concentrations also displayed an inhibitory effect which was maximal with cysteine. Stimulation of proteolysis was detectable only when proteases and reducing agents acted simultaneously. The substrate pre-exposed to the effect of reducing agents (10 rnmol 1- I , 2 h, 23 "C) and washed with nitrogen-saturated water did not show any greater availability to proteolytic enzymes of the fungus. With keratin (guinea pig hair) as substrate, the effects of reducing agents were similar, except there was a smaller tendency to inhibition at higher concentrations (Table 3), Sodium sulphite enhanced protease activity maximally at a concentration of 30 mmol 1-' (3.78 mglml), namely about 2.9 times. Even at a concentration of 100 mmol 1- ', the activity increased by about 70%. Cysteine, mercaptoethanol and glutathione were, in this order, less stimulatory. They were most effective at concentrations of 0.3 to 1 mmol I-'. The weakest was the effect of dithioerythritol where 1 mmol 1-' was already inhibitory. Table 4 summarizes the activities of various enzymes on various substrates in the presence of sulphite. The effect of trypsin on cross-linked BSA was significantly activated by sulphite. At lower concentrations the activation was weaker, at higher concentrations rather stronger than in fungal proteases. With keratin as substrate trypsin was less stimulated than M . gypseum proteases, but this activity was also enhanced up to the concentration of 10 mmol 1-' Na,SO,. The activity of pronase was much less influenced. Table 5 illustrates the effect of dermatophyte proteases on the sulphitolysed wool. In the samples with 8 and 17% cystine converted to S-sulphocysteine, no increase in susceptibility to proteolysis was demonstrated. The activity was enhanced 3.7 times in the sample with reduction of 32% cystine, and after cleavage of 66% and 97% cystine it was enhanced 29 times and 43 times, respectively. Microscopical examination of the two latter samples revealed that proteases caused maceration of the cuticle and cortex to individual cells and their gradual degradation and dissolution. This is supported by a gravimetrically measurable substrate weight loss of 47% and 82%, respectively. The fully sulphitolysed wool was greyish and lost its lustre. In alkaline buffers it rapidly swelled but was not macerated and did not show any loss in weight.
346
J. KUNERT
Table 2. Proteolytic activity of enzymes from Microsporum gvpseum on cross-linked serum albumin in the presence of reducing agents (data are given as yo proteolytic activity) Concentration of reducing agents (mmol 1-
Sulphite Mercaptoethanol Glu tat hione Dithioerythritol Cysteine
I)
0
0.1
0.3
1 .O
3.0
10.0
30.0
100.0
100 100 100 100 100
126 117 114 108 116
189 157 129 116 107
227 173 136 105 50
299 128 126 78 16
322 72 75 47 12
262 -
52 -
Table 3. Keratinolytic activity of enzymes from Minosporum gvpseum on keratin (guinea pig hair) in the presence of reducing agents (data are given as yo keratinolytic activity) Concentration of reducing agents (mmol 1-')
Sulphite Cysteine Mercaptoethanol Dithioerythritol Glutathione
0
0.1
0.3
1.o
3.0
10.0
30.0
100.0
100 100 100 100 100
110
160 174 148 145 129
184 172 153 99 138
216 85
263 28 82 96
293 -
172
111 115
122 113
111
79 124
-
-
Table4. Activity of various proteolytic enzymes on various substrates in the presence of sulphite (data are given as proteolytic activity) Enzyme
M.gvpseum enzymes Trypsin 10 pg rno1-l Pronase 10 pg mol-l
Substrate
Cross-linked BSA Keratin Cross-linked BSA Keratin Cross-linked BSA Keratin
Yo
Sulphite concentration (mmol I-')
0
0.1
0.3
1 .O
3.0
10.0
30.0
100.0
100
126
189
227
299
322
262
52
100 100
110 95
160 116
184 183
216 26 1
263 328
293 29 1
172 134
100 ND
103 ND
110 ND
148 ND
191 ND
236 ND
204 ND
ND ND
100
102
107
114
136
142
135
ND
ND, not determined.
Table 5. Activity of various proteolytic enzymes on sulphitolysed wool (data are given as relative proteolytic activity) Degree of sulphitolysis*
Proteases of M.gypseum Trypsin 10 pg ml-' Pronase 10pgml-'
0
8
17
32
66
97
100
88
94
370
2930
4310
100
-
-
290
2610
4130
100
-
-
570
2190
3940
*Percentage of half-cystine converted to S-sulphocysteine.
mycoses 35, 343-348 (1992)
REDUCING AGENTS AND FUNGAL
Sulphitolysis of disulphide bonds also stimulated the hydrolysis of wool by trypsin and pronase. When these enzymes were used at concentrations degrading the native wool similarly as fungal enzymes, stimulation of proteolysis in sulphitolysed samples was also similar (Table 5).
Discussion The obtained results confirmed that proteolytic enzymes of M. gypseum are more active in the presence of reducing agents, of sulphite in particular. This effect depends on the content of cystine (disulphide bonds) in the substrate. In a protein without cystine (azocasein), inhibitory effects prevailed. This is probably due to the reduction of disulphide bonds in the enzyme molecules themselves. During hydrolysis of substrates without disulphide bonds, inhibition in the presence of reducing agents has been described in several papers. Ziegler [22] reported that gelatin hydrolysis by proteases of Trichophyton mentagrophytes and Microsporum canis was inhibited by 26-55% with 10 mmol 1- I cysteine or thioglycollate. The cleavage of benzoyl-arginine ethyl ester by T. rubrum enzymes was inhibited in the presence of 5 mmol I-' cysteine by l8%, and of 5 mmol 1-' dithiothreitol by 66% [23]. Hydrolysis of leucyl-p-nitroanilide by aminopeptidases of T. ajelloi was strongly inhibited by as little as 0.1 mmol 1- I mercaptoethanol or dithiothreitol [24]. Roberts & Doetsch [25] reported that with a purified proteinase of Microsporum sp., millimolar concentrations of sodium sulphite and thioglycollate were inhibitory, while cysteine and mercaptoethanol markedly enhanced the activity on casein. Chattaway et al. [26] performed the experiments with enzymes of T. rubrum and 1. uerrucosum on casein and found that 20-40 mmol 1-' cysteine had either inhibitory (at pH 5-6) or stimulating (at p H 8-11) effects. Elastase of M. fuluum was not markedly affected by the presence of 50mmol I - ' cysteine [27]. An increased activity in the presence of reducing agents on substrates without disulphide bonds might indicate the presence of thiol proteinases having cysteine at the active site. However, production of proteases of this class has not yet been demonstrated in dermatophytes. Bovine serum albumin is a protein with a compact globular structure stabilized by 17 disulphide bonds and resistant to proteolysis and reduction [28]. In our experiments, the protein was further insolublized by cross-linking. With its insolubility and high cystine content it served mycoses 35, 343-348 (1992)
KERATINOLYSIS
347
as a model of keratin proteins. The results obtained with cross-linked BSA and native keratin were indeed almost identical (Tables 2, 3 and 4) and can be explained as a stimulation of proteolysis by cleavage of disulphide bonds of the substrate. Moreover, both reducible substrates evidently consumed a significant part of reducing agents during incubation so that the inhibitory effect on the enzymes was diminished. With BSA it was found that in the absence of proteases, no significant reduction of the protein occurred so that proteolysis and reduction conditioned one another. In our experiments, inorganic sulphite exhibited higher stimulating and lower inhibitory effects on the enzymes under study than four thiols. However, it cannot be concluded yet that dermatophyte proteases are adapted to sulphitolysis during keratin degradation because similar results were obtained with trypsin. Everett et al. [I81 reported a tenfold increase in the activity of trypsin against cut wool in the presence of 30 mmol 1sodium metabisulphite, while dithionite, thioglycollate and thiosulphate were less effective. The data comparable with our results are still sporadical in the literature. Page & Stock [29] studied a protease from M . gypseum macroconidia and recorded a 129- up to 790-fold increase in 'keratinase' activity in the presence of millimolar concentrations of sulphite, cysteine and dithiothreitol. However, their substrates were obtained from hairs or feathers by reduction, extraction and fractionation so that they were rather denaturated. Hose & Evans [30] also reported a stimulation of keratinolytic activity by sulphite but more details were not given. As in the experiments of Wolfram & Bruggemans [20] with human hairs, wool disulphide bonds were almost completely .cleaved and cystine converted to S-sulphocysteine by the sulphitetetrathionate mixture. According to the above authors, tetrathionate as an oxidizing agent reoxidizes the originating cysteine to cystine which is again sulphitolysed. However, a direct reaction of tetrathionate with cysteine giving rise to S-sulphocysteine has been described [31] and it is highly probable that this reaction takes place in the used mixture. A measurable enhancement of wool hydrolysis by dermatophyte proteases can be expected, according to our results, after removal of about 20% of disulphide bonds. Similar limits for decreased resistance to proteolysis were found in soluble high-sulphur proteins from wool [32]. In wool with conversion of over 30% 'half-cystine' to S-sulphocysteine, the proteolysis increased rap-
'
348
J. KUNERT
idly and substrate dissolution was microscopically demonstrable and gravimetrically measurable. There are only a few published data on the effect of dermatophyte proteases on denatured keratin. Chattaway et al. [26] described an enhanced proteolysis of human hairs and nails after reduction with thioglycollate, while Stahl et al. [33] in their experiments with wool under similar conditions found no increased activity of proteases of M . sypseum or of trypsin. The discrepancies might be due to the fact that the reduced substrates are highly unstable and readily re-oxidize in contact with the air. Under in vivo conditions, sulphitolysis occurs simultaneously with proteolysis in lytic holes surrounding the mycelium. It is difficult to determine in viuo sulphite concentration, protease activity, pH, etc. However, histochemical reactions revealed the presence of a large amount of S-sulpho groups at the sites of hard keratin degradation [9,161. Therefore it can be supposed that, at least in a thin layer on the walls of lytic channels, sulphitolysis is intense enough to allow a complete hydrolysis of the substrate.
References 1 Yu, R. J., Ragot, J. & Blank, F. (1972) Keratinases: hydrolysis of keratinous substrates by three enzymes of Trichophyton mentagrophytes. Experientia 28, 1512-1 5 13. 2 Kunert, J. (1976) Keratin decomposition by dermatophytes. 11. Presence of S-sulfocysteine and cysteic acid Allg. Mikrobiol. in soluble decomposition products. 16, 97-105. 3 Takiuchi, I., Sei, Y., Takagi, H. & Negi, M. (1984) Partial characterization of the extracellular kera tinase from Microsporum canis. Sabouraudia 22, 2 19-224. 4 Tsuboi, R., KO, I-J., Takamori, K. & Ogawa, H. (1989) Isolation of a keratinolytic proteinase from Trichophyton mentagrophytes with enzymatic activity at acidic pH. Infect. Immun. 57, 3479-3483. 5 Yu, R. J., Harmon, S. R. & Blank, F. (1969) Hair digestion by a keratinase of Trichophyton mentagrophytes. 3. Invest. Dermatol. 53, 166-171. 6 Marshall, R. C., Orwin D. F. G. & Gillespie, J. M. (1991) Structure and biochemistry of mammalian hard keratin. Electron Micro. Rev. 4, 47-84. 7 Ziegler, H. & Reichmann, G. (1968) Uber den 'schwefelstoffwechsel von Microsporum canis. Mykosen 11, 903-907. 8 Ziegler, H., Bohme, H. & Reichmann, G. (1969) Stoffwechselphysiologische Untersuchungen iiber den Abbau von Proteinen durch Microsporum gypseurn und M. canis. Dermalol. Mschr. 155, 835-856. 9 Kunert, J. (1972) Keratin decomposition by dermatophytes: evidence of the sulphitolysis of the protein. Experientia 28, 1025-1026. 10 Kunert, J. (1975) Formation of sulphate, sulphite and S-sulphocysteine by the fungus Microsporum gypseum during growth on cystine. Folia Microbiol. 20, 142-151. 1 1 Kunert, J. (1988) Utilization ofcystine by dermatophytes on glucose-peptone media. Folia Microbiol. 33, 188- 197,