( Springer-Verlag 1995

Appl Microbiol Biotechnol (1995) 44:106—111

OR I G I N A L P AP E R

Y. Morikawa · T. Ohashi · O. Mantani · H. Okada

Cellulase induction by lactose in Trichoderma reesei PC-3-7

Received: 12 December 1994/Received revision: 3 March 1995/Accepted: 27 March 1995

Abstract In an attempt to clarify the function of lactose in cellulase induction, experiments were carried out on cellulase formation by lactose along with other sugars in a resting cell system of ¹richoderma reesei PC-3-7, a hypercellulase-producing mutant. Although lactose alone induces little cellulase under the conditions used, a synergistic effect on cellulase formation was observed following the respective addition of sophorose, cellobiose or galactose to lactose. The lactose consumption was more rapid when these sugars were added than in their absence. Furthermore, following lactose addition 10 h after the beginning of cultivation in the presence of cellobiose, cellulase formation was initiated with only a little lag, and lactose consumption started immediately, being complete in 14 h. b-Galactosidase induction experiments suggested that the rapid consumption of lactose is possibly not dependent on lactose degradation by the enzyme. From these results, it is suggested that lactose may function as an inducer for cellulase formation if it is taken up in the mycelium of ¹. reesei PC-3-7, and that sophorose, cellobiose or galactose may induce a putative lactose permease.

Introduction Cellulolytic enzymes from a filamentous fungus, ¹richoderma reesei, have been the subject of intensive research because this organism secretes large amounts of enzymes having all the activities required for complete hydrolysis of crystalline cellulose (Kubicek 1992; Teeri et al. 1992). Consequently, the molecular structure, function and genetics of ¹. reesei cellulases have

Y. Morikawa ( ) · T. Ohashi · O. Mantani · H. Okada Department of Bioengineering, Nagaoka University of Technology, 1603-1 Kamitomioka, Nagaoka, Niigata 940-21, Japan. Fax: #87 258 468163

been fully elucidated, and their biosynthesis has also been studied in some detail. In ¹. reesei the biosynthesis of cellulases is induced by cellulose or other substances such as cellulose derivatives and is susceptible to repression by a readily metabolizable carbon source such as glucose, glycerol or fructose (Nisizawa et al. 1972; Merivuori et al. 1984). One of the most likely models for cellulase induction at present is that the presence of extracellular cellulose is recognized by the constitutive conidial cellulases, which release small amounts of cellooligosaccharides from the cellulose. These cellooligosaccharides may be taken up by the fungus and act as inducers for the further synthesis of cellulases in large amounts (Messner et al. 1991; Kubicek 1992; Seiboth et al. 1992). The identity of the natural inducer of cellulase biosynthesis, however, is still a matter of dispute. It has been reported that sophorose, cellobiono-dlactone, cellobiose (also cellooligosaccharides) and Lsorbose induce cellulase formation in ¹. reesei (Sternberg and Mandels 1979; Hrmova` et al. 1986; Kawamori et al. 1986a; Iyayi et al. 1989; Fritscher et al. 1990). Among them, sophorose has been shown to be a very powerful inducer, and may be the true inducer of cellulases. It should be noted, however, that although sophorose is an excellent inducer of cellulases in ¹. reesei and some bacteria, other highly cellulolytic fungi such as Phanerochaete chrysosporium and Aspergillus purpurogenum do not respond to sophorose (Teeri et al. 1992; Kurosawa et al. 1992). Furthermore, it was reported that cellobiose promoted cellulase formation when its hydrolysis by b-glucosidase was inhibited by nojirimycin or when a b-glucosidase-defective strain was used (Fritscher et al. 1990; Strauss and Kubicek 1990; Fowler and Brown 1992). This indicates that cellobiose may be a natural inducer of cellulases under cellulolytic conditions. Although ¹. reesei also produces cellulases when cultivated in the presence of lactose (Pourquie and Desmarquest 1989; Warzywoda et al. 1992; Chaudhuri

107

and Sahai 1993), it has long been suggested that lactose relieves the fungus from catabolite repression because it is only slowly hydrolysed by the fungus (Merivuori et al. 1984). However, Messner and Kubicek (1991) suggested that cellulase formation on lactose was most probably due to induction, on the basis of their experiment results, which showed that slow growth alone was insufficient to promote cellulase formation. Thus, the nature of ‘‘true inducer’’ is still unknown and it is also possible that several substances can act as inducers. In order to clarify the functions of these inducers, especially lactose, experiments were carried out on cellulase (endoglucanase) formation by lactose along with other inducers under the conditions in which growth was lacking (a resting cell system), using ¹. reesei PC-3-7, which is a hypercellulase-producing mutant that shows elevated cellulase induction by Lsorbose (Kawamori et al. 1986a, b). In the present paper, we conclude that lactose can act as an inducer in ¹. reesei PC-3-7 if it can be taken up at least partly by the fungus, and that sophorose, cellobiose and galactose would be able to induce a putative lactose permease.

Enzyme activity measurement Cellulase activity from the supernatant was examined as carboxymethylcellulase activity measured by the method described by Morikawa et al. (1985). b-Galactosidase activity was measured in a similar manner to that of b-glucosidase (Wood and Bhat 1988) using o-nitrophenyl b-D-galactopyranoside as a substrate. The enzyme activity in mycelium was also determined as follows: mycelium was suspended in a tartrate buffer (50 mM, pH 4.0), sonicated with glass beads and decanted, and the crude suspension thus obtained was used for the measurement. A unit of enzyme activities represents 1 lmol product released/min.

Analysis of induced enzymes by SDS-PAGE and isoelectric focusing Sodium dodecyl sulphate/polyacrylamide gel electrophoresis (SDSPAGE) was performed in the system of Laemmli (1970). The supernatant from the resting cell system was concentrated about ten-fold and analysed by SDS-PAGE on 12.5% (w/v) polyacrylamide gel with a low-molecular-mass protein standards kit (Daiichi Pure Chemicals, Japan). The gel was stained with Coomassie brilliant blue. The isoelectrofocusing was carried out in a Resolmax apparatus (Atto Inc., Japan) using carrier ampholytes (pH range 3.5—9.5) as described by the manufacturer.

Determination of sugar consumption

Materials and methods Microorganism The strain used throughout this study was ¹. reesei PC-3-7, which is a hypercellulase-producing mutant having an enhanced ability to respond to cellulase induction by L-sorbose (Kawamori et al. 1986a). This strain was obtained from Kyowa Hakko Kogyo Co. Ltd. and maintained on a potato/dextrose/agar (PDA) slant.

Culture conditions The spores of ¹. reesei PC-3-7, obtained from a PDA plate culture, were inoculated into 300-ml conical flasks containing 50 ml basal medium described by Kawamori et al. (1986a) with 0.3% glucose as a carbon source, and incubated for 28 h at 28°C on a rotary shaker at 220 rpm. The weight of mycelium collected with filter-paper was measured after drying at 90°C for 20 h.

Resting cell system for cellulase induction The 28-h old mycelium obtained in the manner described above was collected on glass filters, washed twice and suspended in saline. For the cellulase induction studies, the modified method by Sternberg and Mandels (1979) was used. The washed mycelium was mixed with the basal medium containing various concentrations of inducers and 50 mM tartrate buffer pH 4.0, giving a final mycelium dry weight of about 2.0 mg/ml. The basal medium lacked a nitrogen source and thus did not support growth. The induction experiment was carried out in 25-ml test-tubes containing 5-ml medium by incubating them at 28 °C for an appropriate period on a reciprocal shaker at 120 strokes/min. After the mycelium had been removed by centrifugation the supernatant was assayed for cellulase activity and inducer consumption.

The concentration of sugars remaining in the supernatant from the resting cell system was determined with a Shimadzu high-pressure liquid chromatograph LC-9A (Shimadzu, Japan) equipped with a Shodex Ionpak KC811 column and Shodex RI SE-61 (Showa Denko, Japan), using 0.1% H PO as an eluent. 3 4

Results Synergistic cellulase formation by lactose with several sugars ¹. reesei PC-3-7 produces cellulases with yields two or three times as high as ¹. reesei QM9414 does when cultivated with lactose as a carbon source (Y. Morikawa et al., unpublished data). In a resting cell system lacking cell growth, however, this strain required about 30 h before a certain amount of endoglucanase was formed and more than 80% of the added lactose remained in the medium for 24 h, as shown in Fig. 1. On the other hand, it took only several hours for endoglucanase induction and 8 h for complete sugar consumption in the case of sophorose. From these results, it is not clear whether cellulase formation by lactose is due to catabolite derepression or induction. In order to clarify the function of lactose in cellulase formation, we investigated the synergistic effect of lactose in the presence of sophorose on cellulase formation. When a low concentration of sophorose (10 lg/ml) was added to the resting cell system containing lactose, which alone could produce little endoglucanase under these conditions (for 24 h), further stimulation of

108

Fig. 1 Induction of endoglucanase by lactose in ¹. reesei PC-3-7. Endoglucanase activities (——) and residual sugars (---) in the supernatant from the resting cell system at appropriate times were measured as described in Materials and methods. j h Lactose (1 mg/ml) as an inducer; m n sophorose (250 lg/ml) as an inducer

Fig. 3 Effect of several sugars on endoglucanase induction and lactose consumption in the presence of lactose. Each sugar (500 lg/ml) was added to the system including lactose (1 mg/ml). Endoglucanase activities (») and residual lactose concentrations (h) in the supernatant were assayed after 24 h of incubation. 1 lactose alone; 2—7 cellobiose, glucose, maltose, fructose, galactose and sorbitol respectively

that is, 1.0, 1.0 and 0.5 mg/ml were used for lactose, cellobiose and galactose respectively. When different concentrations of sophorose up to 250 lg/ml were added along with 1 mg/ml lactose, a marked synergistic effect on cellulase formation was observed at a lower sophorose concentration (5—10 lg/ml) and the endoglucanase activity was almost the same as the maximum level obtained with sophorose (250 lg/ml) alone. These concentrations were used in further studies. The course of cellulase formation

Fig. 2 Synergistic effect of sophorose and lactose on endoglucanase induction. Sophorose (10 lg/ml) and lactose were added as inducers in the resting cell system. Endoglucanase activities in the supernatant from the system were assayed after 24 h of incubation. (---) Endoglucanase activity level without lactose addition (10 lg/ml sophorose alone)

cellulase formation was observed depending on the lactose concentration (Fig. 2). We therefore investigated synergistic effects of several sugars other than sophorose with lactose on cellulase formation. Although these sugars alone do not induce cellulase under the conditions used in this study, cellobiose and galactose promoted cellulase formation, while glucose, fructose, maltose and sorbitol did not (Fig. 3). It was also noted that lactose consumption was more rapid following the addition of cellobiose and galactose than in their absence. The concentrations of lactose and other sugars required for endoglucanase formation were first optimized;

Typical courses of cellulase formation and consumption of lactose with and without other sugars in ¹. reesei PC-3-7 are shown in Fig. 4. With respect to cellulase formation, addition of sophorose and cellobiose produced similar phenomena; that is, endoglucanase activities reached about the maximum level (1.1—1.6 U/mg mycelium) obtained in this system in 24 h. On the other hand, in the case of galactose addition the enzyme activity appeared slowly, and after 24 h only half the activity seen following sophorose and cellobiose addition was obtained. Addition of the three sugars significantly promoted lactose consumption. When sophorose was added, lactose consumption was initiated within 2 h and lactose did not remain in the medium after 12 h, while a time lag of several hours was observed when cellobiose and galactose were added separately and the lactose consumption terminated within 24 h. In this system cellobiose and galactose consumptions were almost not affected by the presence or absence of lactose (data not shown); these sugars were consumed completely in about 8 h.

109

Fig. 5 Course of endoglucanase induction and lactose consumption by subsequent addition of lactose in the presence of cellobiose. Lactose (1 mg/ml) was added after 10 h in the presence of cellobiose (1 mg/ml). d Endoglucanase; s lactose consumption

10 h after the beginning of cultivation, when cellobiose consumption was complete. As shown in Fig. 5, lactose consumption started immediately after its addition and was complete in 24 h. There was only a brief lag in endoglucanase formation and the activity rose to over 1.0 U/mg mycelium 14 h after the addition. b-Galactosidase induction with various sugars

Fig. 4A—C Course of endoglucanase induction and lactose consumption by sophorose (A), cellobiose (B) and galactose (C) in the presence of lactose (1 mg/ml). —— Endoglucanase activities, --lactose consumption. d s Lactose and other sugars, j h lactose alone, m sophorose alone (10 lg/ml)

It was suggested that the rapid consumption of lactose depends either on the increasing rate of lactose uptake (inductive formation of a putative lactose permease) or on the rate of lactose degradation by bgalactosidase, or on both. Furthermore, the stimulation of endoglucanase formation was supposed to be due to the lactose taken up in the mycelium, but not to sophorose or cellobiose, because the same phenomenon was also observed with galactose, which did not induce cellulases in ¹. reesei at all.

Second, the effect of various sugars on induction of b-galactosidase activity was investigated (Table 1). While a trace of the enzyme activity was found in the mycelium constitutively, only galactose led to an increase in activity both in the mycelium and extracellularly. From these results (Fig. 5 and Table 1) it is suggested that the three sugars used may induce a putative lactose permease and that lactose may function as an inducer for cellulase formation when it is taken up into the mycelium of ¹. reesei PC-3-7.

Table 1 b-Galactosidase induction with various sugars. b-Galactosidase activities (mU/mg mycelium) in the supernatant (extracellular) and in mycelium (intracellular) from the resting cell system were assayed after 24 h of incubation while adding each sugar as an inducer Inducer

Cellulase formation and lactose consumption by late addition of lactose To test the above assumption, two experiments were carried out. First, cellulase formation was observed in the initial presence of cellobiose with lactose addition

None Lactose Sophorose Cellobiose Galactose

b-Galactosidase (mU/mg mycelium) Extracellular

Intracellular

0 1 2 2 27

2 2 4 3 11

110

Discussion Carbon source control in cellulase formation in ¹. reesei is very complicated because of permeation, degradation and transconversion of the carbon source used, and induction and repression (direct for cellulase formation, and also indirect for glycosylation and the secretion pathway, for example). Because of this, cellulase formation on lactose has been ambiguous for a long time. From the results of this study, it is suggested that lactose acts as an inducer of cellulase formation in ¹. reesei PC-3-7 when taken up rapidly into the mycelium without its degradation. However, it is also possible that the cellulase formation is a result of an escape of the fungus from catabolite repression. Such an assumption is, however, contradicted by the fact that the addition of galactose along with lactose led to cellulase formation despite the fact that galactose is not associated with cellulase induction but related to catabolite repression. Furthermore, although the rate of lactose consumption was not more rapid in the presence of sugars than that of glucose, the consumption was complete within 12 h when sugars such as sophorose were added to the resting cell system. Such results suggested that the cellulase formation on lactose was most probably due to induction and not derepression caused by the slow uptake of lactose. When proteins in supernatants, following 24 h induction by lactose in the presence of three sugars and by sophorose (250 lg/ml) alone, were analysed by means of SDS-PAGE and isoelectric focusing, no difference between their patterns could be detected (data not shown), suggesting that lactose induced primarily the same cellulase species as those induced by sophorose. When lactose was added within 10 h after galactose addition, however, no cellulase formation was observed in spite of the immediate consumption of lactose (data not shown), unlike the results following cellobiose addition. Time-course experiments of b-galactosidase induction by galactose showed little activity of the enzyme after 6 h but considerable activity in mycelium following lactose addition (after 10 h; data not shown). This suggests that in this case lactose was not present to perform cellulase induction because the lactose taken up was possibly degraded largely by the enzyme induced. On the other hand, the simultaneous addition of galactose and lactose at the beginning may result in cellulase formation by lactose taken up at the initial stage (up to 10 h). As it appears to take several hours or more for the induction of cellulase formation by lactose (the results shown in Fig. 4) the possibility that cellobiose (and also sophorose) could play some role in cellulase induction cannot be excluded since the time lag in cellulase formation following galactose addition was longer than that when either cellobiose or sophorose were added

(Fig. 4C), and lactose addition after the complete consumption of cellobiose caused cellulase induction almost instantly (Fig. 5). However, this needs to be tested further. This report also suggests that sophorose, cellobiose and galactose induce a putative lactose permease. Although galactose is well known to induce lactose permease (Jung et al. 1994), we show here for the first time that sophorose and cellobiose might also induce a putative lactose permease. It is particularly emphasized that sophorose induces the enzyme immediately at a low concentration (10 lg/ml) compared to galactose, with which there is a lag of several hours before the induction. This putative lactose permease may not be identical with the ‘‘b-linked diglucoside permease’’ reported by Kubicek et al. (1993), since according to their report the permease that is induced by sophorose is specific for cellobiose, sophorose, laminaribiose and gentiobiose, but not for lactose. Although these roles for sophorose are of interest in connection with cellulase formation under physiological conditions in ¹. reesei, further experiments are needed to obtain direct evidence of the presence of lactose permease and to characterize its major properties. Cellulase biosynthesis by the filamentous fungus ¹. reesei is still a subject of considerable interest in view of its commercial potential. In this regard, the results obtained from this study suggest that when a small amount of cellulose or cellobiose is used in addition to lactose as a carbon source, the cellulase production titre may be enhanced and the fermentation period may be shortened by using ¹. reesei PC-3-7.

References Chaudhuri BK, Sahai V (1993) Production of cellulase enzyme from lactose in batch and continuous cultures by a partially constitutive strain of ¹richoderma reesei. Enzyme Microb Technol 15:513—518 Fritscher C, Messner R, Kubicek CP (1990) Cellobiose metabolism and cellobiohydrolase I biosynthesis by ¹richoderma reesei. Exp Mycol 14:405—415 Fowler T, Brown RD Jr (1992) The bgl1 gene encoding extracellular b-glucosidase from ¹richoderma reesei is required for rapid induction of the cellulase complex. Mol Microbiol 6:3225—3235 Hrmova` M, Biely P, Vrsanska M (1986) Specificity of cellulase and b-xylanase induction in ¹richoderma reesei QM9414. Arch Microbiol 144:307—311 Iyayi CB, Bruchmann E-E, Kubicek CP (1989) Induction of cellulase formation in ¹richoderma reesei by cellobiono-1,5-lactone. Arch Microbiol 151:326—330 Jung K, Jung H, Kaback HR (1994) Dynamics of lactose permease of Escherichia coli determined by site-directed fluorescence labeling. Biochemistry 33:3980—3985 Kawamori M, Morikawa Y, Takasawa S (1986a) Induction and production of cellulases by L-sorbose in ¹richoderma reesei. Appl Microbiol Biotechnol 24:449—453 Kawamori M, Morikawa Y, Ado Y, Takasawa S (1986b) Production of cellulase from alkali-treated bagasse in ¹richoderma reesei. Appl Microbiol Biotechnol 24:454—458

111 Kubicek CP (1992) The cellulase proteins of ¹richoderma reesei: structure, multiplicity, mode of action and regulation of formation. Adv Biochem Eng Biotechnol 45:1—27 Kubicek CP, Messner R, Gruber F, Mandels M, Kubicek-Pranz EM (1993) Triggering of cellulase biosynthesis by cellulose in ¹richoderma reesei. J Biol Chem 268:19364—19368 Kurosawa T, Yachi M, Suto M, Kamagata Y, Takao S, Tomita F (1992) Induction of cellulase by gentiobiose and its sulfurcontaining analog in Penicilium purpurogenum. Appl Environ Microbiol 58:106—110 Laemmli UK (1970) Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227:680—685 Merivuori H, Siegler KM, Sands JA, Montenecourt BS (1984) Regulation of cellulase biosynthesis and secretion in fungi. Biochem Soc Trans 13:411—414 Messner R, Kubicek CP (1991) Carbon source control of cellobiohydrolase I and II formation by ¹richoderma reesei. Appl Environ Microbiol 57:630—635 Messner R, Kubicek-Pranz EM, Gsur A, Kubicek CP (1991) Cellobiohydrolase II is the main conidial-bound cellulase in ¹richoderma reesei and other ¹richoderma strains. Arch Microbiol 155:601—606 Morikawa Y, Kawamori M, Ado Y, Shinsha Y, Oda F, Takasawa S (1985) Improvement of cellulase production in ¹richoderma reesei. Agric Biol Chem 49:1869—1871

Nisizawa T, Suzuki H, Nisizawa K (1972) Catabolite repression of cellulase formation in ¹richoderma viride. J Biochem (Tokyo) 71:999—1007 Pourquie J, Desmarquest J-P (1989) Scale up of cellulases production by ¹richoderma reesei. In: Coughlan MP (ed) Enzyme systems for lignocellulose degradation. Elsevier, London, pp 283—292 Seiboth B, Messner R, Gruber F, Kubicek CP (1992) Disruption of the ¹richoderma reesei cbh2 gene coding for cellobiohydrolase II leads to a delay in the triggering of cellulase formation by cellulose. J Gen Microbiol 138:1259—1264 Sternberg D, Mandels GR (1979) Induction of cellulolytic enzymes in ¹richoderma reesei by sophorose. J Bacteriol 139:761—769 Strauss J, Kubicek CP (1990) b-Glucosidase and cellulase formation by a ¹richoderma reesei mutant defective in constitutive bglucosidase formation. J Gen Microbiol 136:1321—1326 Teeri TT, Penttila M, Kera¨nen S, Nevalainen H, Knowles JKC (1992) Structure, function, and genetics of cellulases. In: Finkelstein DB, Ball C (eds) Biotechnology of filamentous fungi: technology and products. Butterworth-Heinemann, Boston, pp 417—445 Warzywoda M, Larbre E, Pourquie J (1992) Production and characterization of cellulolytic enzymes from ¹richoderma reesei grown on various carbon sources. Bioresource Technol 39:125—130 Wood TM, Bhat KM (1988) Methods for measuring cellulase activities. Methods Enzymol 160:87—112