Cellulase Induction in Trichoderma reesei by Cellulose Requires Its Own Basal Expression*

The induction of cellulases by cellulose, an insoluble polymer, in the filamentous fungus Trichoderma reesei is puzzling. We previously proposed a mechanism that is based on the presence of low levels of cellulase in the uninduced fungus; this basal cellulase activity would digest cellulose-releasing oligosaccharides that could enter the cell and trigger expression of cellulases. We now present experiments that lend further support to this model. We show here that transcripts of two mem- bers of the cellulase system, cbh1 and egl1 , are present in uninduced T. reesei cells. These transcripts are in- duced at least 1100-fold in the presence of cellulose. We also show that a construct containing the hygromycin B resistance-encoding gene driven by the cbh1 promoter confers hygromycin B resistance to T. reesei cells grown in the absence of cellulose. Moreover, cellulose-induced production of the cbh1 transcript was suppressed when antisense RNA against three members of the cellulase system was expressed in vivo . Experiments are pre- sented indicating that extracellular cellulase activity is the rate-limiting event in induction of synthesis of the cellulase transcripts by cellulose. The results reveal a critical requirement for basal expression of the cellulase system for induction of synthesis of its own transcripts by cellulose.

The induction of cellulases by cellulose, an insoluble polymer, in the filamentous fungus Trichoderma reesei is puzzling. We previously proposed a mechanism that is based on the presence of low levels of cellulase in the uninduced fungus; this basal cellulase activity would digest cellulose-releasing oligosaccharides that could enter the cell and trigger expression of cellulases. We now present experiments that lend further support to this model. We show here that transcripts of two members of the cellulase system, cbh1 and egl1, are present in uninduced T. reesei cells. These transcripts are induced at least 1100-fold in the presence of cellulose. We also show that a construct containing the hygromycin B resistance-encoding gene driven by the cbh1 promoter confers hygromycin B resistance to T. reesei cells grown in the absence of cellulose. Moreover, cellulose-induced production of the cbh1 transcript was suppressed when antisense RNA against three members of the cellulase system was expressed in vivo. Experiments are presented indicating that extracellular cellulase activity is the rate-limiting event in induction of synthesis of the cellulase transcripts by cellulose. The results reveal a critical requirement for basal expression of the cellulase system for induction of synthesis of its own transcripts by cellulose.
Cellulose, a ␤-(1,4)-linked glucose polymer, is a product of the utilization of solar energy and carbon dioxide by plants, through photosynthesis, a process which is estimated to produce 7.2 ϫ 10 10 tons of cellulosic biomass annually (1). The degradation and oxidation of cellulose to carbon dioxide, a process carried out mainly by microorganisms, is a key transformation step in the biological carbon cycle in nature (2). The filamentous fungus Trichoderma reesei is considered to be the most efficient producer of cellulases (3,4). Its cellulases are classified into two broad classes: cellobiohydrolases (CBH), 1 whose major activity involves the cleavage of cellobiose resi-dues consecutively from the ends of the cellulose chains (5), and endoglucanases (EG), whose major activity involves the cleavage of ␤-glycosidic bonds in the cellulose chain (6). The members of this system act synergistically and are necessary for the efficient hydrolysis of cellulose to soluble oligosaccharides (7,8). The genes encoding two cellobiohydrolases, cbh1 and cbh2, and two endoglucanases, egl1 and egl2, have been isolated and characterized (9 -15).
In T. reesei, cellulase transcripts are induced when the natural inducer cellulose is the only carbon source available (16). The natural cellulose is insoluble; hence, we developed an interest in studying how this insoluble polymer, which cannot traverse the cell membrane, would induce cellulase production. Previously published observations suggested that basal levels of cellulolytic enzymes may be necessary to generate soluble breakdown products of cellulose such as sophorose (2-O-␤-glucopyranosyl-D-glucose), a soluble disaccharide inducer of cellulase synthesis (17), which could pass through the cell wall and initiate the inductive process. Experimental evidence in vitro was reported supporting this mechanism. The potential soluble disaccharide sophorose was detected, as well as other glucose disaccharides, during growth of T. reesei on cellobiose (18), and after hydrolysis of cellulose with the T. reesei cellulase system (19). A membrane-bound ␤-glucosidase is believed to catalyze the formation of sophorose through its transglycosylation activity (20). Indirect evidence was also reported showing that the presence of antibodies to the major cellulase inhibited production of the cbh1 transcript with cellulose present. However, the antibodies did not affect transcription of the cbh1 transcript when sophorose was present (21).
In this study, we present evidence for the requirement of basal levels of cellulases by showing: 1) the presence of the transcripts of two major members of the cellulase system, cbh1 and egl1, in uninduced T. reesei cells; 2) the activity of a reporter gene driven by the cbh1 promoter in the absence of cellulose; and 3) the inhibition of cbh1 expression under celulose but not under sophorose by expressing antisense RNA against three members of the cellulase system. We also show that increasing the concentration of added cellulase decreases the time necessary for the induction of synthesis of the cbh1 and egl1 transcripts. Moreover, extracellularly added purified cellulases also have a positive effect on cbh1 and egl1 expression, consistent with a model in which induction is dependent on the extracellular activity of cellulases. the cbh1 gene fused in the correct and the opposite orientation to the hygromycin B resistance gene (hph) were constructed as follows. A 2.2-kb EcoRI-BglI DNA fragment containing the 5Ј-flanking region of the cbh1 gene was treated with T4 DNA polymerase and ligated to a BamHI linker. The coding region for the hph gene was amplified using the plasmid pLG90 (22) in conjunction with PCR in which a BamHI site was included at the 5Ј ends of the primers. The 5Ј-flanking region of the cbh1 gene and the amplified hph coding sequence were digested with BamHI, ligated and then inserted into the BamHI site of a pBluescript containing a polyadenylation signal. A construct containing the 5Јflanking region, the hph gene and the polyadenylation signal placed in the correct orientation was isolated and designated pCBH1-Hph-2.2. The pCBH1R-Hph-2.2 was constructed by placing the 5Ј-flanking region in opposite orientation using the SacII sites located 10 bp 5Ј of the ATG.
The antisense plasmid, pJO51, was constructed as follows. Specific oligonucleotides, in which suitable restriction enzymes sites were placed at the 5Ј end, were used in conjunction with PCR to amplify the required cbh2 and egl2 sequences. The amplified cbh2 DNA fragment contained 399 bp (from base 1477 to 1876) from the fourth exon of the gene (14). The amplified egl2 DNA fragment contained 398 bp (from base 982 to 1380) from the second exon of the gene (11). A 587-bp KpnI-EcoRI DNA fragment (from base 86 to 673) from the first exon of the egl1 gene (10) was isolated and used in the construction of the plasmid pJO51. After digestion with the suitable restriction enzymes, the fragments were ligated and inserted in the reverse orientation between the TrpC promoter and the 3Ј terminator from Aspergillus nidulans (23,24).
RNA Isolation and Analyses-Total RNA was isolated as described (25). RNA (10 g) was separated by electrophoresis on a 1.2% agarose gel, after denaturation with glyoxal and dimethyl sulfoxide (26), and transferred to a Zeta-Probe membrane. Membranes were hybridized with a random primer [␣-32 P]dCTP radiolabeled EcoRI fragment (720 bp) or KpnI-PstI fragment (1050 bp) containing part of the coding region of the cbh1 or egl1 respectively (10,12).
Primers and Polymerase Chain Reaction-Each of the forward primers was a 21-mer oligonucleotide complementary to the mRNA; cb 1 and eg 1 primers correspond to DNA sequences 740 -760 and 885-905 of the second exon of cbh1 (12) and egl1 (10) genes, respectively. Each of the reverse primers was a 21-mer oligonucleotide of the same polarity as the mRNA; cb 2 and eg 2 primers correspond to DNA sequence 612-632 and 711-731 of the first exon of cbh1 (12) and egl1 (10) genes, respectively. cDNA was synthesized by mixing 1 g of total RNA, 1 M of forward primers (cb 1 for cellobiohydrolase I and eg 1 for endoglucanase I), 2.5 units/l of murine leukemia virus reverse transcriptase (Perkin-Elmer) in polymerase buffer (10 mM Tris-HCl, pH 8.3, 50 mM KCl, 1.5 mM MgCl 2 , 0.01% (w/v) gelatin) and 1.25 mM each dNTP. Samples were incubated 60 min at 42°C, and then second strand synthesis and 25 cycles of amplification were performed by addition of 1 M reverse primers (cb 2 for cellobiohydrolase I and eg 2 for endoglucanase I), 2.5 units of Taq DNA polymerase, and 20 Ci of [␣-32 P]dATP (3000 Ci/ mmol). Each cycle of PCR consisted of 1.5 min of denaturation at 94°C, 1.5 min of annealing at 55°C, and 3 min of polymerization at 72°C. Genomic amplification was performed as described above, except that 10 ng of genomic DNA clone were used in place of RNA, and the first strand synthesis was omitted.
Enzymes Purification-T. reesei was induced for 6 days with 2% Avicel. The culture filtrate was concentrated and chromatographed on a Sephacryl S-200 column, and the fraction containing the excluded cellulolytic activity was used as crude cellulase. The major members of the cellulolytic system composed of CBHI, CBHII, EGI, and EGII were purified and assayed as described previously (21).
Inoculum, Culture Conditions, and Transformation-Maintenance of T. reesei (QM9414) cultures, inocula preparation, culture medium, and washing of Avicel were carried out as described previously (21). Mycelia from germinated spores were centrifuged, washed twice with 100 mM potassium phosphate buffer (pH 6.0), and incubated on a rotary shaker for 2 h. Mycelia (2 mg, dry weight) were suspended in 5 ml of culture medium and cellulose, sophorose, or cellulase was added to the reaction mixture as indicated. All cultures were incubated on a rotary shaker (200 rpm) at 28°C for the indicated time.
Preparation and transformation of protoplasts were carried out as described (27,28). After transformation, the protoplasts were plated on minimal medium in agarose overlay. The minimal medium used was as described by Pentillä et al. (28) except that glucose was substituted with 0.4% glycerol.

Transcription of cbh1 and egl1 in Uninduced Cells-The
transcripts of two members of the cellulase system, cbh1 and egl1, could not be detected by Northern analysis in total RNA isolated from uninduced T. reesei cells (Fig. 1). Moreover, the physical heterogeneity of substrates and the complexity of the cellulase system's multiple enzymes, synergism, and glucanases with overlapping specificity (6), present obstacles to assaying the expression of a single member of the cellulase system, specially at low activity levels. To detect low levels of transcription of cellulases in uninduced T. reesei cells, we assayed for the presence of cbh1 and egl1 transcripts using reverse transcription-PCR. The strategy of the amplification process is presented in Fig. 2A. The position of the primer and the expected amplified product using RNA or DNA as a template is also shown. cDNAs were synthesized from total RNA extracted from Avicel-induced and glycerol-grown (uninduced) T. reesei cells and then amplified by PCR. This procedure produced 82-and 125-bp segments of the cbh1 and egl1 transcripts, respectively ( Fig. 2B; refer to details in Fig. 2A). However, if T. reesei genomic DNA was used as a template instead of mRNA, the presence of introns in cbh1 and egl1 genes, resulted in fragments of 149 and 196 bp, respectively (Fig. 2C). In addition, digestion of amplified products by appropriate restriction enzymes produced fragments of the expected sizes ( Fig. 2C; see also Fig. 2A). This excluded any possible artifactual amplification of contaminating DNA and confirmed that the amplified DNA fragments corresponded to spliced cbh1 and egl1 transcripts. Quantitation of both transcripts from uninduced relative to induced T. reesei cells was performed using 5Ј 32 P-labeled reverse primers (cb 2 for cbh1 and eg 2 for egl1) and comparing the extent of amplification achieved after different numbers of PCR cycles as described by Chelly et al. (29). The levels of cbh1 and egl1 transcripts were, respectively, 1156-and 1148-fold lower in uninduced cells relative to induced cells. It is worth mentioning that similar results were obtained when total RNA was isolated from uninduced T. reesei cells grown in a culture medium in which peptone was omitted.
The Promoter of the cbh1 Gene Drives Basal Expression of a Heterologous Gene Placed under Its Control-We used a heterologous gene fusion to demonstrate that the promoter of the major member of the cellulase system, cbh1, is functionally active in the absence of cellulose, and that this low basal promoter activity is sufficient to endow a new phenotype on transformed T. reesei cells. Series of plasmids were constructed in which the Escherichia coli gene encoding hygromycin B phosphotransferase (22)  T reesei cells were grown on 0.8% glycerol (uninduced) or induced with 1% cellulose for 20 h as described under "Experimental Procedures." Total RNA was isolated and aliquots containing 10 g of RNA were denatured, fractionated electrophoretically in a 1.5% agarose gel, transferred to Hybond-N membrane, and hybridized with labeled cbh1 or egl1 probes as indicated (for details see "Experimental Procedures"). mycin B phosphotransferase, catalyzes the phosphorylation and subsequent inactivation of the antibiotic hygromycin B, a potent inhibitor of protein synthesis in both pro-and eukaryotes (30). The resistance to hygromycin B of transformed T. reesei cells is shown in Fig. 3. The results demonstrate that the cbh1 promoter fused to the hygromycin B phosphotransferase gene in the correct orientation (plasmid pCBH1-Hph-2.2) conferred resistance to the antibiotic in noninduced cells. No transformants were obtained in control experiments in which the cbh1 promoter was fused in the opposite orientation (plasmid pCBH1R-Hph-2.2) or with promoterless construct (pHph). In six independent transformations, an average of 500 transformants/g of the plasmid pCBH1-Hph-2.2 was usually obtained (Fig. 3). It is important to note that in T. reesei transformants harboring the plasmid pCBH1-Hph-2.2, the transcript of hygromycin B phosphotransferase was found to be induced with cellulose or sophorose in a manner resembling that of the cbh1 gene (data not shown). These results indicate that the promoter of the cbh1 gene has basal transcription activity in the absence of an inducer and that low levels of mRNA do indeed give rise to hygromycin B phosphotransferase activity and subsequent resistance of transformed T. reesei cells to the antibiotic.
Expression of Antisense RNA against cbh2, egl1, and egl2 mRNA Leads to Suppression of Cellulose-induced Expression of the cbh1 Transcript-We designed an antisense strategy to establish convincingly that the inductive mechanism of the cellulase mRNA by cellulose in T. reesei requires basal expression of the cellulase transcripts. The cellulase system of T. reesei is made up of hydrolases catalyzing the cleavage of ␤-(1,4)-glycosidic bonds in the cellulase chain. The members of this system include at least two cellobiohydrolases, CBHI and CBHII, and two major endoglucanases, EGI and EGII, that act synergistically (7,8) in the hydrolysis of cellulose to oligosaccharides. Therefore, we decided to examine the effect that the expression of an antisense RNA against CBHII, EGI, and EGII mRNAs would have on the induction of cbh1. We reasoned that if the basal activity of cbh2, egl1, and egl2 was necessary for induction by cellulose, the expression of antisense transcripts against those three cellulases should inhibit the induction of cbh1 by cellulose but not by sophorose. A 399-bp DNA sequence from the fourth exon of the cbh2 gene, a 587-bp DNA sequence from the first exon of the egl1 gene and a 398-bp DNA sequence from the second exon of the egl2 gene were ligated together and then inserted in reverse orientation between the TrpC promoter and the 3Ј terminator (23, 24) from A. nidulans (Fig. 4,  egl1 and cbh1 genes). a, sequence and position of primers in the target mRNA region; b, as above, except that amplification was performed using genomic DNA clone as the template. Sizes of the expected amplified fragments are indicated below the sequences. Arrowheads indicate the exon limits; introns are indicated by lowercase letters. B, amplification products using mRNA as a template. After the single strand cDNA synthesis from total RNA using cb 1 or eg 1 as forward primers, amplification was performed with reverse primers, cb 2 for cellobiohydrolase I and eg 2 for endoglucanase I in the presence of [␣-32 P]dATP. Aliquots were taken after set numbers of cycles as indicated, electrophoresed in a 15% acrylamide gel, dried, and exposed for 90 min. C, ethidium bromide stained PCR product for cbh1 and egl1 using genomic DNA as a template after electrophoresis in a 15% acrylamide gel. Lanes 1 and 2 are specific fragments amplified from genomic DNA or total RNA respectively; lanes 3 and 5 are as in lane 1, and lanes 4 and 6 are as lane 2 after restriction by MboII or BstNI as indicated.
A and B). The TrpC promoter was found to be functional in T. reesei (16). This antisense construct, pJO51, was transformed into QM9414 T. reesei cells, and five stable transformants were isolated. Analyses of DNA from those transformants showed that the plasmid pJO51 was integrated in the fungal genome (data not shown). The effect of cellulose and sophorose on the expression of the cbh1 transcript were analyzed and the results of one of those transformants, QMJO51, and the original QM9414 cells are presented in Fig. 4, C and D. The   4. Effect of expression of antisense RNA against cbh2, egl1, and egl2 mRNAs on cellulose-and sophorose-induced expression  of the cbh1 transcript. A, restriction maps of the genes coding for CBHII, EGI, and EGII. Boxes represent exon and lines introns; shaded parts within the fourth exon of the cbh2 gene and the second exon of egl2 gene were amplified using the primer presented on the top and the bottom of each of the restriction map. A restriction site was added at the 5Ј end of each primer (bold letters), as indicated, to facilitate the construction of the plasmid pJO51. The shaded part within the first exon of the egl1 gene was isolated using the indicated restriction sites. These fragments were ligated and inserted in the opposite orientation between the TrpC promoter and terminator as presented in B and named pJO51. In T. reesei cells, QM9414 was transformed with pJO51, and a stable transformant, QMJO51, was isolated in which the plasmid pJO51 was integrated within the fungal genome. QM9414 and QMJO51 were induced with 1% cellulose (C) or 3 mM sophorose (D). Aliquots were removed at the indicated time, and total RNA was isolated and analyzed by Northern blot with labeled cbh1 probes as indicated. The transcript of actin (act) of T. reesei (33) was analyzed, and is included as a control. RNA isolated from QMJO51 and QM9414 cells exposed to cellulose for 17, 20, and 24 h, or sophorose for 4 and 6 h. Using cellulose as an inducer, the results revealed reduction of the expression of the cbh1 transcript by at least 85% in cells carrying the antisense construct QMJO51, compared with the original QM9414 (Fig. 4C). However, expression of the antisense RNA did not repress the synthesis of the cbh1 transcript if the soluble disaccharide sophorose was used as an inducer (Fig. 4D).
Effect of the Addition of Exogenous Cellulase on the Induction of cbh1 and egl1 Transcripts-Studies on the nature of the physiological inducer of the cellulases by growing T. reesei on cellobiose have indicated that sophorose might be one of the naturally occurring inducers of the cellulase system (18). Therefore we examined the kinetics of induction of the cbh1 and egl1 transcript by the soluble disaccharide sophorose and compared it with that of cellulose. The results show that the transcripts could not be detected up to 14 h after induction with cellulose, while 4 h were needed for induction with sophorose (Fig. 5A). Similar results were obtained with the egl1 transcript (data not shown). This result is expected if the formation of sophorose is dependent on the initial hydrolysis of cellulose by the basal cellulase activity and suggest that the concentration of the extracellular cellulase system could be the rate-limiting step in the induction of the cellulase transcript with cellulose. Therefore, we examined whether the induction of the cbh1 and egl1 transcripts by cellulose could be detected earlier in the presence of added cellulase. To this end, a crude cellulase fraction was prepared as described under "Experimental Procedures." The induction of cbh1 and egl1 transcripts by cellu-lose was detected earlier in the presence of the crude cellulase fraction, and the response was also dependent on the amount of added cellulase (Fig. 5B, compare also lanes 6 and 7). The enzymes of the cellulolytic system, composed of CBHI, CBHII, EGI, and EGII were purified to homogeneity from the crude cellulase fraction as described under "Experimental Procedures." All of them individually accelerated the induction of cbh1 and egl1 transcripts (Fig. 5C). In the above-mentioned experiments, actin transcript presented no alteration (data not shown). A summary of these kinetic studies is presented in Fig.  5D. The time required for induction of the cbh1 and egl1 transcripts using cellulose, cellulose ϩ cellulase, and sophorose is 14, 10, and 4 h, respectively. These results implicate extracellular cellulase activity in the generation of soluble(s) inducer(s), such as sophorose, from cellulose and are consistent with the proposed mechanism for cellulase induction by cellulose. DISCUSSION In T. reesei, cellulase is an inducible enzyme system that has drawn considerable interest regarding the mechanism by which the insoluble inducer cellulose triggers synthesis of the cellulase transcripts. The proposed mechanism of cellulase induction is that the fungus produces basal levels of cellulase and that the activity of these extracellular enzymes on cellulose produces a soluble inducer, which can enter the cell and effect induction (18,31). In support of this mechanism, we have previously shown that antibodies against CBHI, CBHII, EGI, and EGII blocked the expression of cbh1 gene in the presence of cellulose but not the soluble inducer sophorose (21). However, these constitutive levels of cellulases and their direct role in The results presented here show that the mRNAs of the major members of the cellulase system, cbh1 and egl1, are transcribed under uninduced conditions, and that induction with cellulose results in at least 1100-fold increase of both transcripts. To examine further the basal activity of the promoter of the cbh1 gene, we constructed a chimeric vector in which the gene encoding hygromycin B phosphotransferase (22) was placed under the control of the 5Ј-flanking DNA sequence of the cbh1 gene. Indeed, under uninduced conditions, resistance to the antibiotic hygromycin B was observed with T. reesei cells transformed with this construct and grown on medium lacking cellulose. These sets of experiments indicate that the promoters of cbh1 and egl1 are transcriptionally active under uninduced conditions, and that the basal activity of the cbh1 promoter is sufficient to drive expression of a heterologous gene such as the hygromycin B resistance-encoding gene.
We also used an antisense RNA strategy to present direct and in vivo evidence for the requirement of the basal expression of the cellulase in induction of the cellulase transcripts by cellulose. The major cellulases from T. reesei, CBHI, CBHII, EGI, and EGII, have glucanase activities with overlapping specificity (6). Therefore, we decided to express an antisense RNA composed of parts of the coding sequences against cbh2, egl1, and egl2 mRNAs in T. reesei and examine the expression of the cbh1 transcript. The results show that the expression of this antisense RNA produced marked effects on the induction of the cbh1 transcript using cellulose but not sophorose as an inducer. The reduction of the expression of the cbh1 transcript by cellulose was between 80 and 90% in experiments with three independent transformants.
In this study we also present evidence indicating that the initial hydrolysis of cellulose is the rate-limiting step in the inductive process. This conclusion is based on the fact that the addition of the cellulase system or its purified enzyme members to a culture of T. reesei, in the presence of cellulose, resulted in earlier detection of the cbh1 and egl1 transcripts. That the inducer is produced from the cell wall of T. reesei cells, and not from cellulose, or that cellulase itself acts as inducer are ruled out, since the addition of cellulase in the absence of cellulose does not affect expression of cellulase transcripts (data not shown). The time required for induction of cbh1 and egl1 transcripts using cellulose, cellulose ϩ cellulase, or sophorose is 14, 10, and 4 h, respectively. This result supports our hypothesis that oligosaccharide(s) is(are) formed in vivo from cellulose by the activity of a low, constitutive, and extracellular cellulase activity. The relatively slow induction by sophorose could be explained by the fact that the inductive process is protein synthesis-dependent (data not shown). In addition, it has recently been shown that a sophorose-inducible ␤-diglucoside permease is involved in the induction of the cellulase system in T. reesei (32).
In summary, our results demonstrate that in T. reesei (i) the transcripts of two members of the cellulase system, cbh1 and egl1, are expressed at a low level in the absence of added inducer and are induced at least 1100-fold by cellulose; (ii) the promoter of the major member of the cellulase system, cbh1, is able to drive basal expression of a reporter gene placed under its control; (iii) expression of antisense RNA against the cbh2, egl1, and egl2 mRNAs resulted in inhibition of induction of the cbh1 transcript by cellulose but not sophorose; and (iv) the kinetic of expression of the cbh1 and egl1 transcripts by cellulose is influenced by the addition of cellulase to the culture medium.
Taken together, these results support a model that requires the presence of basal activity of cellulases for the induction of their own transcripts by cellulose. The identification of genes involved in this process and their mode of action will be critical to completely understand how the induction occurs at the molecular level.