OUP user menu

Transcriptional regulation of plant cell wall degradation by filamentous fungi

Nina Aro, Tiina Pakula, Merja Penttilä
DOI: http://dx.doi.org/10.1016/j.femsre.2004.11.006 719-739 First published online: 1 September 2005

Abstract

Plant cell wall consists mainly of the large biopolymers cellulose, hemicellulose, lignin and pectin. These biopolymers are degraded by many microorganisms, in particular filamentous fungi, with the aid of extracellular enzymes. Filamentous fungi have a key role in degradation of the most abundant biopolymers found in nature, cellulose and hemicelluloses, and therefore are essential for the maintenance of the global carbon cycle. The production of plant cell wall degrading enzymes, cellulases, hemicellulases, ligninases and pectinases, is regulated mainly at the transcriptional level in filamentous fungi. The genes are induced in the presence of the polymers or molecules derived from the polymers and repressed under growth conditions where the production of these enzymes is not necessary, such as on glucose. The expression of the genes encoding the enzymes is regulated by various environmental and cellular factors, some of which are common while others are more unique to either a certain fungus or a class of enzymes. This review summarises our current knowledge on the transcriptional regulation, focusing on the recently characterized transcription factors that regulate genes coding for enzymes involved in the breakdown of plant cell wall biopolymers.

Keywords
  • Cellulase
  • Ligninolytic enzyme
  • Hemicellulase
  • Transcriptional regulation
  • Cre
  • ACEI
  • ACEII
  • PacC
  • Hap complex

1 Introduction

Plant cell wall consists mainly of polysaccharides such as cellulose, hemicelluloses and pectin, which together with proteins and lignin form a complex and rigid structure. It is estimated that approximately 4 × 109 tons of cellulose is formed annually, however, its amount on earth does not accumulate due to fungi and bacteria that efficiently degrade plant cell wall materials. These microorganisms play a key role in the recycling of carbon back into the ecosystem. Degradation of plant cell wall compounds is a complex process involving the synergistic action of a large number of extracellular enzymes. In filamentous fungi, these plant cell wall degrading enzymes, cellulases, hemicellulases, pectinases and ligninases provide the fungus with the means to obtain energy and nutrients from plant cell wall biopolymers. In addition, plant cell wall degrading enzymes are believed to contribute to the action of plant pathogenic fungi [1, 2]. Carbohydrate polymer degrading enzymes may also play a role in the ability of antagonistic fungi to attack their target fungi due to hydrolysis of the host cell wall [3].

Many of the biopolymer degrading enzymes have received considerable attention because of their potential applications in food, feed, textile, and pulp and paper industries [48, and references therein]. Several applications, such as bleaching of pulp by xylanolytic enzymes, clarification of juices with pectinases and biobleaching of textiles with cellulases are currently in use. The use of fungal lignocellulose degrading enzymes for total hydrolysis of plant biomass to sugars is under intensive study. Liberated sugars could serve as raw material in the bioproduction of chemicals and fuels by microbes. For instance, bio-ethanol production by yeasts and bacteria from hexose and pentose sugars derived from agricultural waste materials could replace part of the fossil fuel use with a more sustainable alternative.

Industrial production strains of Aspergilli and Trichoderma can produce extremely high amounts of extracellular enzymes, several tens of grams per litre [9, 10]. The existence of hypersecreting strains and strong promoters, such as cellulase promoters, make filamentous fungi potential hosts for heterologous protein production as well. These fungi are easy and inexpensive to grow in large bioreactors and they possess good secretion capacity capable of carrying out similar type of protein modifications as occurs in many higher eukaryotes. As an example, calf chymosin, used in the dairy industry is produced at commercial levels in Aspergillus niger [11].

The importance of fungi for the global carbon cycle, the significance of extracellular enzymes in the life of these organisms and the biotechnological importance of filamentous fungi and their enzymes have promoted an interest towards understanding the regulation of expression of the extracellular enzymes and especially the characterization of the transcription factors involved. The most studied filamentous fungi in respect to biopolymer degrading enzyme production are brown-rot fungi Trichoderma reesei (Hypocrea jecorina), A. niger and Penicillium and the ligninolytic white-rot fungus Phanerochaete. Full genomic data is now available for Phanerochaete chrysosporium (http://www.jgi.doe.gov/programs/whiterot.htm), Aspergillus nidulans (http://www-genome.wi.mit.edu/annotation/fungi/aspergillus/index.html) and T. reesei (http://shake.jgi-psf.org/trire1/trire1.home.html).

2 Plant cell wall polysaccharides and lignin

Plant cell wall consists of cellulose, hemicelluloses and pectin and the phenolic polymer lignin. Cellulose is the most abundant polysaccharide in nature and the major constituent of plant cell wall providing its rigidity. Cellulose consists of β-1,4 linked d-glucose units that form linear polymeric chains of about 8000–12 000 glucose units. In crystalline cellulose, these polymeric chains are packed together by hydrogen bonds to form highly insoluble structures. Hemicelluloses, the second most abundant polysaccharides in nature, have a heterogeneous composition of various sugar units. Hemicelluloses are usually classified according to the main sugar residues in the backbone of the polymer. The main hemicellulose, xylan, found in cereals and hardwood, and (galacto)glucomannan found in both softwood and hardwood, consists of β-1,4 linked d-xylose and β-1,4 linked d-mannose in the polymer backbone, respectively. The less abundant hemicellulose arabinan consist of α-1,5 linked l-arabinose units, and galactan of β-1,3 linked d-galactose units. The main chain sugars of hemicelluloses are modified by various side groups such as 4-O-methylglucuronic acid, arabinose, galactose, and acetyl, making hemicelluloses branched and variable in structure [12, 13]. Pectins are a family of complex polysaccharides containing a backbone of α-1,4 linked d-galacturonic acid. Pectins contain two different types of regions. In the “smooth” region of pectin d-galacturonic acid residues can be methylated or acetylated, whereas the “hairy” region consists of two different structures, d-xylose substituted galacturonan and rhamnogalacturonan to which long arabinan and galactan chains are linked via rhamnose residues (for a more detailed structure of pectin we refer to [14] and [15]). The cellulose wall is strengthened by lignin, a highly insoluble complex branched polymer of substituted phenylpropane units joined together by carbon–carbon and ether linkages forming an extensive cross-linked network within the cell wall.

3 Enzymes involved in the breakdown of plant cell wall polysaccharides

3.1 Cellulases

Complete hydrolysis of cellulose to glucose requires the combined actions of multiple enzymes with different substrate specificities. The cellobiohydrolases (EC 3.2.1.91; 1,4-β-d-glucan cellobiohydrolases) cleave cellobiose units from the ends of the polysaccharide chains. The endoglucanases (EC 3.2.1.4; 1,4-β-d-glucan-4-glucanohydrolase) cut the cellulose chains internally mainly from the amorphous regions, providing the cellobiohydrolases more ends to act upon [16]. And, finally, β-glucosidases (EC 3.2.1.21) hydrolyse cellobiose to glucose, which serves as an easily metabolisable carbon source for fungi [17]. A schematic view on the synergistic action of cellulotic enzymes is shown in Fig. 1(a).

Figure 1

Degradation of cellulose and hemicellulose. (a) A schematic view on cellulotic systems. The sites for the major activities of cellulolytic enzymes are shown. In addition a putative route for formation of sophorose by the transglucosylating activity of β-glucosidase is presented. (b) A schematic view on a hemicellulolytic system, degradation of arabinoxylan as an example. The enzymes active on the substrate are listed in the table.

A number of genes encoding for cellulose degrading enzymes have been isolated from the most studied cellulolytic fungus T. reesei. These include two cellobiohydrolase-encoding genes [1821], eight endoglucanase genes [2226] and seven genes encoding β-glucosidases [2629]. Cellulolytic activities have been characterised, and numerous cellulase-encoding genes have also been isolated from other fungi including, cellobiohydrolase-encoding genes from Penicillium janthinellum [30], Phanerochaete chrysosporium [31], Humicola grisea [32], Trichoderma viride [33], different Aspergilli species [34, 35], Agaricus bisporus [36] and Neurospora crassa [37]. In addition to the more classical cellulases, novel types of cellulases have recently been described, one of these being swollenin (SWOI) from T. reesei. SWOI has amino acid similarity to plant expansins that disrupt cellulosic fibres without possessing hydrolytic activity. It has been suggested that the action of swollenin would make the cellulose fibers more accessible for the cellulases to act upon [38]. The sequencing of expressed sequence tags (ESTs) and full genomes from cellulolytic and hemicellulolytic organisms will most likely lead to discovery of new genes encoding cellulolytic and hemicellulolytic enzymes, some of which may possess novel activities. For instance, the EST sequencing of T. reesei has led to isolation of a novel endoglucanase gene predicted to contain a glycosylphosphatidylinositol membrane anchor suggesting that this endoglucanase is membrane bound [26].

3.2 Hemicellulases

Although hemicelluloses are relatively complex polysaccharides, their enzymatic degradation is well understood. The hydrolysis of hemicelluloses occurs by the concerted action of endo-enzymes cleaving internally the main chain, exo-enzymes liberating monomeric sugars and ancillary enzymes cleaving the side chains of the polymers or oligosaccharides leading to the release of various mono- and disaccharides depending on the hemicellulose type. As an example, the breakdown of xylan involves at least endo-1,4-β-d-xylanases (EC 3.2.1.8) and β-xylosidases (EC 3.2.1.37) acting on the main sugar chain and depending on the type of xylan, the side-chain cleaving enzymes such as α-glucuronidase (EC 3.2.1.131) and acetyl xylan esterase (3.1.1.72). For a complete list of enzymes involved in degradation of hemicellulose and their mode of action we refer to the recent review by de Vries et al. [39]. A schematic view on degradation of arabinoxylan is shown in Fig. 1 (b) as an example of a hemicellulolytic system.

From the Aspergillus species nearly 20 genes encoding endoxylanases and β-xylosidases have been cloned. Xylanase genes have been cloned from a number of other fungi as well, e.g. Penicillium, A. bisporus, and the rice blast fungus Magnaporthe grisea [4044]. Numerous other genes encoding hemicellulose backbone and side chain cleaving enzymes have been characterized, most from the species of Aspergillus. Also, nearly 30 genes encoding for enzymes acting on pectin have been cloned from A. niger [5, 45]. From T. reesei altogether four xylanase genes, designated xyn1 to xyn4, and more than 10 other hemicellulase genes have been isolated [26, and references therein].

3.3 Ligninases

In the plant cell wall lignin forms a matrix that surrounds cellulose and hemicellulose. Degradation of lignin is a prerequisite for hydrolysis of cellulose and hemicelluloses and thus for the usage of the main carbon and energy sources by the fungi. However, due to hydrophobicity and the complex random structure lacking regular hydrolysable bonds, lignin is poorly degraded by many microorganisms capable of cellulose degradation. Basidiomycetous white rot fungi are capable of efficient depolymerization and mineralization of lignin. The enzymes degrading lignin are oxidative, non-specific and act via non-protein mediators in contrast to hydrolytic cellulases and hemicellulases. The main fungal ligninolytic enzymes are manganese peroxidases (MnP; EC 1.11.1.13), lignin peroxidases (LiP; EC 1.11.1.14) that catalyse a variety of oxidative reactions that are dependent on H2O2, and laccases (EC 1.10.3.1) that oxidize phenolic compounds and reduce molecular oxygen to water. In addition, extracellular hydrogen peroxide-generating enzymes, glyoxal oxidase and glucose-2-oxidase (EC 1.1.3.10), that generate peroxides essential for peroxidase function, play a role in the delignification process.

Since the cloning of the first lignin peroxidase encoding lip gene from P. chrysosporium by Tien and Tu [46], several other lip genes have been characterised from P. chrysosporium [47, 48], as well as from other white rot fungi such as T. vesicolor, P. radiata, Bjerkanderasp., and Pleurotus eryngii [4954]. In comparison to the lip genes, the number of genes characterised that encode manganese peroxidases is smaller [5557]. The ligninolytic genes are sometimes found in clusters in the fungal genomes as is the case for two lip and one manganese peroxidase gene,mnp, of T. versicolor [58].

4 Regulation of genes encoding extracellular enzymes

As production of extracellular enzymes in large quantities is an energy-consuming process, the carbon source-dependent regulation of the genes encoding plant cell wall degrading enzymes ensures that the enzymes are produced only under conditions in which the fungus needs to use plant polymers as an energy and carbon source. A synergistic action of the enzymes is needed for degradation of the plant cell wall components and thus production of many of the enzymes is coordinately regulated. There is considerable cross-talk in induction of expression of the genes encoding different classes of enzymes. For instance, the same compounds may provoke expression of both cellulases and hemicellulases, albeit to different extent [59]. Majority of the genes studied are repressed by the presence of easily metabolisable carbon sources, such as glucose, and induced in the presence of the substrate polymers. However, as the polymers are too large to enter the cells, the inducing molecules are expected to be small hydrolysis products of the polymers or their derivatives. Glucose repression takes place very efficiently and typically overrides induction of the genes.

4.1 Expression of cellulases

Regulation of cellulase genes has been analysed at the molecular level mainly in T. reesei, and the largest variety of compounds that provoke cellulase expression has been described for this fungus. The main cellulase genes have been shown to be regulated, to a large part, co-ordinately [60, 61]. It has been proposed that a low constitutive level of cellulase expression is responsible for the formation of an inducer from cellulose. El-Gogary et al. [62], reported the lack of cellulase induction in T. reesei upon growth on cellulose when cellulase action was blocked by addition of antibodies against the main cellulases and β-glucosidase in the culture medium prior to induction. However, the level of expression must be extremely low since cellulase transcripts were not detected from glucose-, sorbitol- or glycerol-grown mycelia in Northern analysis overloaded with RNA, even though it was estimated that for cbh1 up to 6000-fold lower levels than those seen in cellulose-induced cultivations could have been detected this way [61].

Sophorose (two β-1,2-linked glucose units) is the most potent soluble inducer in T. reesei and it has been for many years considered to be the natural inducer of cellulases in T. reesei. Sophorose is believed to be formed from cellobiose by the transglycosylating activity of β-glucosidase [63, 64]. In accordance with this, deletion of the bgl1 gene in T. reesei encoding the major extracellular β-glucosidase led to a delay in induction of cellulase gene transcription during growth on cellulose [60]. However, sophorose is more specific for only certain fungi including T. reesei, A. terreus and P. purpurogenum since sophorose has been shown not to induce cellulase expression in P. janthinellum, P. chrysosporium, A. nidulans and A. niger [35, 6567]. In T. reesei sophorose has been reported to be a poor substrate for β-glucosidase but to be readily transported into the mycelium by the β-linked disaccharide permease [68]. The addition of sophorose into resting cells was also reported to increase the β-linked disaccharide permease activity of the cells. This increase was proposed to be due to the enhanced transcription of the gene encoding the permease [68]. The results are, however, only tentative since the gene encoding the permease has not been cloned.

Growth in the presence of cellobiose (two β-1,4-linked glucose units), the end product of the action of cellobiohydrolases [61, 6971], has been shown to induce cellulase expression in many species of fungi, including for instance all the main cellulases of T. reesei [61], eg1 of Volvariella volvacea [72], egl2 of P. janthinellum [66], and eglA of A. nidulans [73]. However, the reports concerning the inducing effect of cellobiose have been somewhat controversial, most likely due to the various culture conditions and e.g. cellobiose concentrations used in the experiments. The β-glucosidase(s) can either cleave cellobiose, the end product of cellobiohydrolase action, into glucose, which may cause repression, or transglycosylate it into e.g. sophorose, which would lead to activation of enzyme gene transcription. Thus the outcome in cellobiose cultures depends on the balance of hydrolysis and transglycosylation, and the subsequent uptake of the generated sugars and the intracellular signals they provoke. During cellulose hydrolysis in nature, cellobiose and glucose may not accumulate to high enough levels to cause significant glucose repression.

The disaccharide lactose (1,4-O-β-d-galactopyranosyl-d-glucose) is currently the only soluble and economically feasible carbon source that allows cellulase gene expression in T. reesei. Therefore lactose has been used as a carbon source in the production of homologous enzymes and heterologous proteins under the cbh1 promoter in T. reesei [74]. The expression provoking effect of lactose is still not understood. Lactose itself is not a component of plant cell wall polymers. Lactose is cleaved by extracellular β-galactosidase into glucose and galactose. Recent studies with T. reesei strains deleted for the galactose utilisation pathway genes, gal1 or gal7, encoding galactokinase and uridyltransferase, respectively, suggest that intracellular concentration of galactose-1-phosphate may be involved in lactose signalling [75, 76]. However, since growth on galactose also normally leads to the formation of galactose-1-phosphate, and galactose in the culture medium does not induce cellulase expression, the formation of galactose-1-phosphate in the intracellular galactose utilisation pathway cannot alone explain cellulase expression on lactose [59, 76]. It was recently shown that galactose could be metabolised through another route, which produces galactitol as an intermediate and involves l-arabinitol-4-dehydrogenase encoded by the lad1 gene of T. reesei. The lactose effect seems to be mediated at least partially through an other intracellular mechanism than sophorose induction since genomic deletion of the gene, gal1, affected only lactose-mediated induction [76]. On the other hand, in the basidiomycetous fungi A. bisporus, Volvariella volvacea, and Irpex lacteus addition of lactose into cellulose induced cultures has been shown to repress cellulase transcription [72, 7779].

In addition, the presence of various other oligosaccharides in the T. reesei cultures is known to induce cellulase expression. These include δ-cellobiono-1,5-lactone, laminaribiose, gentiobiose, xylobiose and the monosaccharide l-sorbose [9, 59, 80, 81]. Also l-arabitol and different xylans have been shown to induce expression of cbh1 in T. reesei [59].

Expression of a large majority of the cellulase genes that have been studied in filamentous fungi is repressed during growth on glucose. Glucose repression overrides induction since addition of glucose to induced cultures results in repression of the cellulase gene expression [61, 62]. This repression by glucose has been shown to occur very efficiently [61]. However, cellulase transcripts have been detected in cultures of T. reesei where glucose had already been consumed [61]. Simple incubation in media lacking any carbon source did not induce cellulase transcription suggesting that an induction mechanism is involved in the observed derepression of cellulase gene expression in carbon depleted conditions. It is possible that an inducing sugar is derived from carbohydrates released from the fungal cell wall or formed e.g. by transglycosylation from the glucose that had been present in the culture [61, 70]. A similar derepression phenomenon appears to occur for a majority of cellulases and hemicellulases (see e.g. [59]).

4.2 Expression of hemicellulases

In respect to the inducing sugars, hemicellulase expression has been studied mostly in the species of Aspergilli, and T. reesei. Generally the presence of the hemicelluloses xylan, arabinan and mannan, as well as some compounds derived from these larger polymers, is known to provoke high hemicellulase production. As hemicelluloses, like cellulose, are too large to enter the cell, the smaller molecules derived from those are likely to play a role in activation of the genes. However, the possibility that the inducing signal would be transduced from the extracellular space or that the larger polymers outside the cell would be involved in induction cannot be ruled out. The genes encoding for the xylanolytic enzymes in Aspergillus species are induced by the monosaccharide d-xylose. Furthermore, xylobiose and the heterodisaccaride d-glucose-β-1,2-d-xylose induce expression of xylanolytic genes [67]. d-xylose induces expression of various other enzyme genes in addition to those encoding xylanases in A. niger, including genes encoding the side chain cleaving enzymes α-glucuronidase (aguA), acetylxylan esterase (axeA) and feruloyl esterase (faeA) [82, 83]. Production of many of the T. reesei hemicellulases is induced during growth in the presence of cellulose, xylan, sophorose, xylobiose, and l-arabitol. Xylobiose has been shown to induce T. reesei genes involved in the hydrolysis of xylan, the xylanase genes xyn1 and xyn2 and β-xylosidase gene bxl1, as well as the side chain cleaving enzyme genes, α-galactosidase genes encoded by agl1 and agl2 and α-glucuronidase encoding gene glr1 [59]. Xylose at low concentrations, however, has been shown to induce only one of the two main xylanase genes in T. reesei (xyn1) whereas the expression of the other xylanase gene (xyn2) has been shown to be partially constitutive and inducible by xylobiose and sophorose [84, 85]. In other studies no xylanase expression was detected when high xylose concentrations were used [59]. Some of the controversies in these results can be explained by the different concentrations of the carbon source or the inducer in the culture medium at the time of the analysis. Xylose at very low concentrations can act as an inducer and at higher concentrations as a repressing carbon source, as has been demonstrated in A. niger [86].

Arabinose and l-arabitol induce expression of a set of genes encoding for enzymes involved in the degradation of arabinoxylan, e.g., genes encoding arabinofuranosidases (afbA and afbB) and arabinoxylan arabinofuranohydrolase (axhA) in A. niger [8789]. In T. reeseil-arabitol induces the expression of xyn1, xyn2, bxl1, agl1, agl2, and the gene abf1 encoding α-l-arabinofuranosidase that cleaves arabinose from arabino(glucurono)xylans [59]. Studies done with A. niger suggest that l-arabitol serves as the intracellular inducer of the enzymes involved in the hydrolysis of arabinan containing hemicelluloses. A mutant strain of A. nidulans that accumulates high intracellular levels of l-arabitol produces higher amounts of α-arabinofuranosidase and endoarabinases than the wild type strain [87]. A specific transcriptional activator for arabinases of A. niger was first proposed by Flipphi et al. [88], based on the finding that additional copies of the abfA or abfB genes lead to a decrease in expression of the other abf genes. Subsequently, de Groot et al. [90] recently reported the isolation of two ara mutants of A. niger that are defective for induction of the genes encoding the arabinolytic enzymes but not for the induction of the genes encoding enzymes of the xylanolytic system.

d-galacturonic acid, polygalacturonate, and sugar beet pectin have been shown to induce expression of nearly all the genes encoding for enzymes involved in the degradation of pectin and its side chains in A. niger [39, 45]. It was suggested that all genes encoding pectinolytic enzymes would be under a general pectinolytic regulatory system responding to d-galacturonic acid or a metabolite derived from it [45, 82].

In addition to the compounds mentioned here many other mono- and disaccharides have been shown to induce expression of a set of hemicellulases. In T. reesei growth on galactose induces only the expression of the α-galactosidase encoding genes (agl1 and agl2) and the acetyl xylan esterase encoding gene (axe1) [59]. In A. niger galactose and mannose are involved in the induction of α-galactosidases and β-mannosidases [91].

In addition, the expression of xylanase genes has also been shown to be dependent on the ambient pH of the culture medium. The pH dependent regulation of xylanase expression will be discussed in Section 5.6.

4.3 Expression of ligninases

Regulation of the genes encoding ligninolytic enzymes has been studied at the mRNA level mainly in Phanerochaete chrysosporium. Studies on the expression of the manganese peroxidase (mnp) and lignin peroxidase (lip or LIG) encoding genes are complicated by the fact that most ligninolytic fungi possess several closely related but differentially regulated lip and mnp genes. As an example, P. chrysosporium has at least ten lip genes designated from lip A to lip J [47] and three mnp genes, mnp1 to mnp3 [56, 92, 93]. Due to the number of homologous lip, mnp and laccase genes often found in ligninolytic fungi the study of laccase gene expression by traditional methods such as Northern blot analysis is troublesome due to the difficulties in the selection of specific probes. Reverse transcription coupled to the PCR technique (RT-PCR) has recently been used to quantitatively study the expression of ligninolytic genes under different environmental conditions. However, the reliability of this technique as a quantitative method is controversial [94].

In general, the ligninolytic gene expression is triggered by the depletion of nutrient nitrogen, carbon or sulphur and the expression of most of the ligninolytic genes is regarded as a stress response to nutrient depletion. In P. chrysosporium expression of the mnp gene is activated by the depletion of nutrient nitrogen in the culture [48, 95]. Also the presence of Mn(II) in the culture medium activates mnp expression [48, 9698]. The presence of putative metal response elements has been reported in the manganese peroxidase promoters [48, 96, 98]. The expression of mnp is also transiently induced by a heat shock in nitrogen limited cultures and sequences matching to the consensus heat shock element (CN2GAAN2 TTCN2G) are found in the promoters of mnp genes [92, 93]. Furthermore, addition of H2O2 and other chemical stress compounds into the cultures also induce mnp expression in P. chrysosporium [99]. It appears that the regulation of the mnp genes occurs at least at two different levels since the induction caused by the different stress factors occurs only in the nitrogen limited cultures and the level of induction is in most cases dependent on the presence of Mn(II). Similarly to the mnp genes the lip genes of P. chrysosporium are also differentially regulated according to various factors, e.g. as a response to carbon and nitrogen starvation [100103].

Expression of laccase encoding genes is not dependent on nutrient depletion. Laccases are expressed constitutively in many basidiomycetous fungi [104106]. This constitutive low expression is often enhanced by inducers. Increase in laccase transcription in response to aromatic compounds such as 2,5-xylidine has been reported for the laccase genes from e.g. T. versicolor [107], Trametes villosa [108] and A. bisporus [106]. It has been postulated that the constitutively expressed laccases are the first enzymes degrading lignin and that the possible degradation products released from lignin by the laccases act as inducers that further increase laccase expression and induce the expression of other ligninolytic genes [105].

5 Transcription factors regulating plant cell wall degrading enzymes

Regulatory elements found in the promoters of genes encoding plant cell wall degrading enzymes include binding sites for the carbon catabolite repressor CRE, CCAAT element, and binding sites for transcriptional activators or factors modulating expression. Regulation of these genes has been shown to be subject to many regulatory pathways. For some of the well-known pathways, like carbon catabolite repression, the DNA-binding factor, the consensus binding sequences and functional sites in the promoters, have been characterised. For some other pathways, e.g. those leading to high level of expression, only fragmentary data are available at the moment. In the following chapters we summarise the data on the different regulatory pathways known to regulate cellulase and hemicellulase genes in filamentous fungi. To the best of our knowledge, transcription factors regulating genes encoding for enzymes involved in degradation of lignin have not yet been identified.

5.1 XlnR

The first transcriptional activator controlling the expression of genes encoding xylanolytic and cellulolytic enzymes in filamentous fungi was isolated from A. niger by cDNA library complementation of a pyrG genotype in a strain expressing the double-selectable marker pyrG under the xynA promoter and being deficient for xynA promoter activation [109]. XlnR has a zinc binuclear cluster DNA-binding domain, and it was shown to bind in vitro to the 5′-GGCTAATAA sequence in the xylanase xlnA promoter. The binding site was proposed to contain a core of 5′-GGCTAR that is found in most hemicellulase and cellulase gene promoters of A. niger [35, 82, 109, 110]. The expression of two endoxylanases (xlnB and xlnC), a β-xylosidase (xlnD), two endoglucanases (eglA and eglB), two cellobiohydrolases (cbhA and cbhB) and several genes encoding the side chain cleaving hemicellulases are co-ordinately regulated by this xylanolytic activator [35, 109, 110]. xlnR loss of function mutation leads to reduced or complete loss of induction of the regulated genes by xylose or xylan. Accordingly, the expression of some of the genes controlled by XlnR was increased by xlnR overexpression [35, 110]. The fact that both hemicellulase- and cellulase-encoding genes are regulated by XlnR indicates that the transcriptional activation mechanism in Aspergillus is at least partially shared between cellulases and hemicellulases.

The role of XlnR in A. niger seems not to be restricted to regulation of genes encoding secreted polysaccharide degrading enzymes but also to an intracellular enzyme involved in d-xylose catabolism. d-xylose reductase, xyrA, which catalyses the first step of d-xylose catabolism, the reduction of xylose to xylitol, was shown to be regulated by XlnR [111]. This finding demonstrates for the first time a link between regulation of extracellular enzymes and intracellular catabolic enzymes in filamentous fungi. A similar situation where the same transcriptional regulator controls the expression of genes encoding the intracellular and extracellular enzymes needed for the utilisation of a particular sugar oligomer has been characterised for e.g. S. cerevisiae. GAL4 regulates the expression of the secreted α-galactosidase and several genes of the galactose utilisation pathway [112].

The gene encoding XlnR has also been isolated from A. oryzae where the corresponding protein AoXlnR was demonstrated to control the expression of the xylanase-encoding genes xynF1 and xynG2 [113]. In addition, the databases contain sequences of putative XlnR homologues from A. kawachii (AB064658), A. nidulans (AJ272537) and T. reesei (AF479644). Furthermore, based on promoter analysis the regulation of the genes encoding ferulic acid esterase, faeB, of Penicillium funiculosum and xylanases of P. purpurogenum have been suggested to involve the Penicillium homologue of the Aspergillus transcriptional activator XlnR [114, 115].

5.2 ACEII

The ace2 gene was isolated in a yeast expression screening designed for isolation of factors binding to and activating the main cellulase promoter cbh1 of T. reesei. ace2 encodes a zinc binuclear cluster DNA-binding protein with no clear amino acid similarity to sequences in the databases. ACEII has been shown to bind in vitro to 5′GGSTAA sequences in the promoters of cbh1 and xyn2 of T. reesei [116, 117]. Deletion of the ace2 gene from the T. reesei genome resulted in the reduction of expression of all the main cellulase genes, cbh1, cbh2, egl1, egl2, and the xylanase gene xyn2, when the genes were induced upon growth on cellulose as the sole carbon. This indicates that ACEII acts as an activator but that in addition to ACEII, other factors also play a role in the cellulose-derived induction of these genes. Induction by sophorose was not affected by the ace2 deletion [116], suggesting that in T. reesei induction derived from sophorose and cellulose may use at least partially different mechanisms. Two types of mutations have been introduced into the ACEII binding site within the xylanase-activating element (XAE) in the xyn2 promoter of T. reesei. The mutations resulted either in 50% reduction in xyn2 expression or to a complete loss of basal and induced expression upon growth on xylobiose [117], suggesting that also in xylobiose induced cultures another factor in addition to ACEII regulates xyn2 expression and that this factor also acts through the XAE sequence in the xyn2 promoter.

Furthermore, it still remains to be studied to what extent the XlnR homologue XYRI of T. reesei regulates the expression of cellulases and xylanases. The lack of an ace2 homologue in A. nidulans, and most likely in A. niger as well, implies that at least partly different transcriptional regulators are used by these fungi to achieve the induced expression levels. So far ace2 gene has been isolated only from T. reesei and somewhat unexpectedly the genome sequences of N. crassa, A. nidulans and Magnaporthe grisea appear not to contain a gene homologous to ace2.

5.3 The CCAAT binding complex

CCAAT sequences are found in the 5′ regions of approximately 30% of eukaryotic genes. A conserved multimeric protein complex recognises and binds to this sequence. The first CCAAT-binding complex described, designated as the Hap complex, consisting of Hap2, Hap3, Hap4 and Hap5 proteins, was identified from S. cerevisiae [118120]. Since then, homologues of the Hap protein complex, excluding Hap4, have been found from various organisms [121]. In filamentous fungi, the cloned CCAAT binding complexes include the Hap2/3/5 from T. reesei [122], AnCF from A nidulans [123, 124], and a Hap5 homologue AAB1 from N. crassa [125]. Also the genome and EST sequences of various other fungi contain genes encoding proteins with clear amino acid sequence identity/similarity with those encoded by the hap genes. In each Hap2/3/5 subunit, the segments needed for the formation of the protein–DNA complex and subunit association are conserved between the different species, suggesting conservation in the mode of action. In S. cerevisiae, the Hap complex is required in various functions, e.g. for the positive regulation of many respiratory genes involved in oxidative phosphorylation in response to growth on non-fermentable carbon sources [126]. In A. nidulans, the deletion of any of the genes encoding the AnCF subunits results in slowly growing and weakly conidiating phenotypes [123, 124]. Mutation of the CCAAT motif in promoters or deletions of the subunits generally reduce the expression level of the gene studied, the reduction occurring either at the basal expression level or in the response to specific stimulatory signal, indicating that the CCAAT motif co-operates with other specific elements to induce transcription. The AnCF complex has subsequently been shown to have a role in the formation of an open chromatin structure in the promoter region of the amdS and fmdS genes of A. nidulans, indicating that CCAAT binding complexes are necessary for the full transcriptional activation of certain promoters [127].

CCAAT sequences have been found in the promoters of many cellulase and hemicellulase genes as well as in the promoters of the ligninolytic genes mnp1 and mnp2 of P. chrysosporium and mnp1 of Ceriporiopsis subvermispora [128, 129]. In the T. reesei cbh2 promoter a region designated CAE consists of a CCAAT box, and a GTAATA motif suggested to bind ACEII. Mutation of either the CCAAT element or the GTAATA motif resulted in reduction of cbh2 transcription while simultaneous mutation of both led to a complete loss of cbh2 transcription [130]. In vivo footprinting analyses of the CAE region in the cbh2 promoter showed that both motifs are protected by bound proteins in inducing and repressing conditions suggesting that the activity of the CCAAT binding complex and the GTAATA binding protein is regulated by other means than their mere binding to the promoter. A somewhat opposite result was obtained when the CCAAT sequence within the XAE in the T. reesei xyn2 promoter was mutated, resulting in a slight increase in xyn2 transcription in glycerol and xylan grown cultures [117]. A protein complex has also been shown to bind to fragments of the xylanase xyn1 promoter of T. reesei containing the CCAAT motif suggesting a role for the Hap complex in the regulation of xyn1 expression as well.

5.4 ACEI

The ace1 gene was isolated in a similar screen as ace2 designed for the isolation of transcription factors binding to and activating the T. reesei cbh1 promoter in S. cerevisiae. ACEI contains three Cys2His2-type zinc fingers and it was shown to bind in vitro to eight sites in the cbh1 promoter, all of which contain the core 5′AGGCA followed by a sequence rich in A and T [131]. 5′AGGCA sequences are found in nearly all of the sequenced T. reesei cellulase promoters but their functionality has not yet been studied. Although ace1 activated the cbh1 promoter in yeast, its deletion in T. reesei led to increased expression of all the main cellulase and xylanase genes studied in cellulose and sophorose induced cultures, indicating that ACEI has a negative effect on the induced expression of these genes. T. reesei strains deleted for ace1 expressed cellulases at 2–30-fold higher level than the host strain in the inducing conditions [132].

The discrepancy between the effects of ace1 expression in the yeast system and in T. reesei could be explained by the fact that ACEI expressed from the yeast library plasmid was truncated and lacked 242 amino acids from the N-terminus. This N-terminus is conserved between the putative ACEI proteins of A. nidulans (AF202995), Talaromyces emersonii (AAL69549) and N. crassa (genome sequence), suggesting its functional importance. The truncated protein may have functioned as an activator in yeast while the full-length protein is a repressor in T. reesei. Alternatively, it is possible that ACEI is capable to act as an activator or a repressor depending on the context.

In addition to resulting in an increase in cellulase and xylanase expression, deletion of ace1 led to clear reduction in growth when sorbitol was used as the sole carbon source. No difference in growth between the ace1 deleted strain and the host strain was seen when glucose, glycerol or fructose were used as carbon sources [132]. The reason for this growth defect on sorbitol is not known. Furthermore, the A. nidulans gene encoding a protein showing highest overall amino acid similarity with ACEI, stzA (AF202995), is deposited into the database as a gene encoding a protein that alleviates sensitivity to salt and DNA damaging agents. Together these findings suggest that ACEI may have a more general regulatory role in addition to being only a repressor for cellulase and xylanase expression.

5.5 Glucose repression – Cre proteins

The presence of the easily metabolisable and energetically favourable carbon source glucose in the microbial surroundings results in the repression of various genes needed for the use of other alternative carbon sources. The mechanism controlling this preferential use of substrates such as glucose (or fructose) over other alternative carbon sources is called glucose repression or carbon catabolite repression (CCR). In many ascomycetous fungi, glucose repression is mediated by the Cys2His2 type transcription factor CreA/CREI that has been cloned from numerous filamentous fungi. CRE mediated regulation has been studied in most detail for the alcohol gene cluster in A. nidulans [133]. The role of the glucose phosphorylating enzymes hexokinase and glukokinase in mediating CCR in A. nidulans and the connections to CreA were recently addressed by Flipphi et al. [134].

Numerous cellulase, hemicellulase and pectinase genes have been shown to be regulated by Cre proteins in T. reesei and Aspergillus species by using cre mutant strains in the studies [59, 61, 89, 135139]. In general, mutations of the cre gene lead to (partial) derepression of enzyme gene expression on glucose. For instance, a hypercellulolytic industrial mutant strain (Rut-C30) of T. reesei, in which the cellulase and most of the hemicellulase genes are expressed to some extent during growth on glucose, was shown to contain a truncated cre1 gene. Transformation of a full-length cre1 gene into this strain restored glucose repression of the studied cellulase and hemicellulase genes, indicating that glucose repression of cellulase and most of the hemicellulase genes is mediated by CREI in T. reesei [59, 141]. Other studies have also addressed glucose repression of the T. reesei cellulase and hemicellulase genes [59, 84, 85, 141145].

To the best of our knowledge, cre genes have not been cloned from any of the basidiomycetous white rot fungi and the full genome of P. chrysosporium does not contain any clear homologue for the A. nidulans CreA either. However, the genes encoding for the ligninolytic enzymes are in general not expressed during growth on glucose [102]. Recently glucose repression and the presence of sequences fitting to the consensus binding site of Cre in the promoter of the cellobiose dehydrogenase (cdh) gene involved in lignin degradation was reported for the basidiomycete T. versicolor [146].

The binding consensus motif for A. nidulans CreA was determined to be 5′-SYGGRG and the binding was shown to be context dependent [147149]. In vivo functionality of the CRE binding sites have been shown for the cbh1 and xyn1 promoters of T. reesei and the xlnA promoter of A. nidulans where mutations in the binding sequences led to expression of these genes in the presence of glucose [84, 140, 142]. In the xlnA promoter of A. tubingensis the removal of four putative CreA binding sites led to increased expression of this gene indicating the functionality of at least one of these binding sites [150]. In all functional CREI/CreA binding sites characterized so far two closely spaced 5′SYGGRG sequences are present, and it has been suggested that direct CREI/CreA repression would occur only through the double binding sites. However, in vitro binding of CREI/CreA into DNA fragments containing only single site has been shown [133, 143, 144, 151, and references therein]. In the T. reesei cbh2 promoter a single CREI binding site has been suggested to be essential for strict nucleosome positioning in both inducing and repressing conditions [152].

Phosphorylation of a serine in a conserved short stretch within an acidic region of T. reesei CREI has been demonstrated to regulate its DNA binding. Phosphorylation of this serine may involve a casein kinase II [153] which are known from various other organisms to play a role in the regulation of a large number of transcription factors [154]. No casein kinase encoding genes have yet been cloned from filamentous fungi. However, the genome of N. crassa and the ESTs sequences of T. reesei and A. nidulans contain sequences that have amino acid similarity with S. cerevisiae casein kinase subunits. Furthermore, it has been shown that in A. nidulans CreA–DNA complex was formed in vivo more efficiently under repressing than derepressing conditions [155].

The CreI of Sclerotinia sclerotiorum was demonstrated to be localised in the nucleus during growth on glucose and transported into the cytoplasm when glucose was removed from the culture medium, implying that the activity of the CreI is controlled by its nuclear translocation [156]. Furthermore, mutation of the corresponding serine in the S. sclerotiorum CreI that has been shown to be phosphorylated in the T. reesei CREI, abolished its repressor activity but not its nuclear localisation. The authors suggest the involvement of a cyclic AMP-dependent protein kinase in the phosphorylation of S. sclerotiorum CreI [157]. The sequence does not, however, fully fit into the consensus recognition site of AMP-dependent protein kinases and therefore the mechanism of CRE/Cre phosphorylation awaits further confirmation.

It appears that Cre/CRE proteins have roles beyond acting as repressors in conditions where glucose (or fructose) is present, and that their effects on cellulase and hemicellulase gene expression can be mediated in a more complex manner than by mere binding to the enzyme gene promoters. Firstly, it has been noticed that the steady-state levels of cre transcripts can be higher in inducing conditions than on glucose in T. reesei, A. nidulans, Gibberella fujikuroi and Botrytis cinerea [141, 155, 158], and that the cre transcript is negatively autoregulated as a response to repressing carbon sources in A. nidulans [155]. Secondly, genomic deletion of the cre genes or their mutagenesis leads to a higher expression of enzyme genes even in inducing conditions, suggesting that Cre/CRE imposes a repressing effect also in these conditions and not only in the presence of glucose. In A. niger expression of arabinofuranosidase encoding genes, abfA and abfB and the endo-arabinase gene abnA is up to 10 times higher in the creA mutants than in the wild type when the fungus is grown in inducing conditions, having arabinose or arabitol as carbon sources [135, 136]. Similarly, the expression of the A. nidulans xylanase genes xlnA and xlnB are induced to a higher level in a creAd 30 mutant than in the wild type strain when xylanase expression is induced upon transfer to xylose-containing medium [134, 140, 159]. Expression levels of the hemicellulase genes xlnB, xnlD, faeA and aguA of A. niger are all higher in the creA mutant than in the wild-type strain during growth in inducing conditions on various xylose concentrations [86]. Furthermore, Orejas et al. [159] have shown that the A. nidulans xlnB gene promoter is repressed by glucose and that its expression is higher in the creAd30 mutant than in the wild type. However, none of the four putative binding sites for CreA in the promoter were found to be functional in vivo. Results like these have promoted suggestions that balancing mechanisms between repression and activation exist, and that Cre would enforce CCR also by an indirect mechanism repressing the expression of the gene encoding the transcriptional activator such as XlnR [86, 134, 140, 159]. This type of a double-lock mechanism mediated by CreA has been shown for the well established transcriptional regulation of the A. nidulans alcA gene involved in ethanol utilisation [133, 149].

In addition, the presence of a CCR-mediating carbon source such as glucose may provoke the repressing effect already at the sugar permease level. CreA may control the expression of permease genes responsible for the uptake of glucose [160, 161] or the inducing compounds [140, 159]. Furthermore, recent results on the studies of glucose repression of cbh2 of T. reesei and xlnA of A. nidulans suggest that repression of these enzyme genes in the presence of glucose may be caused by inducer exclusion, i.e. glucose inhibiting the transport of an inducer into the mycelium and thereby causing repression and/or preventing induction [134, 152]. Flipphi et al. [134] have shown that expression of the A. nidulans xlnA gene is lowered when excess of glucose is added to xylose cultures. This reduction in expression occurs also in the creAd 30 mutant, indicating that creA does not play a role in repression of xlnA expression, but that the effect could be caused by permease level competition for the sugars or alterations in the permease protein activity. Fructose did not have such an effect. In accordance with this, higher amounts of xylose together with glucose (50 mM of both xylose and glucose) allowed induction of xlnA in the cre mutant (in the wild type cre strain repression occurred). As a whole, the various mechanisms how glucose (and CCR) mediates its control over cellulase and hemicellulase expression requires and merits further investigations.

In addition to creA two other genes, creB and creC were isolated in the genetic screening designed for isolation of regulatory proteins affecting CCR in A. nidulans [162]. CreC protein contains five WD40 repeat motifs and a proline rich region, which in other proteins have been shown to be involved in protein–protein interactions [163]. However, no biochemical role for CreC has been established yet. The creB gene encodes a deubiquinating enzyme [164]. Deubiquinating enzymes are generally cysteine proteases that cleave ubiquitin. The most common role for ubiquitin is to target proteins for degradation by the proteasome. In addition, ubiquitinylation has been proposed to play a role in macromolecular assembly and function. Only recently activation domains of certain transcription factors have been demonstrated to serve as direct ubiquitylation targets, and modulation of activation domains by the ubiquitylation level has been hypothesized to provide an important mechanism for the regulation of gene transcription [165]. However, no direct role for deubiquinating enzymes in transcription regulation has been demonstrated. The protein products CreB and CreC were found to exist in a complex where CreC is proposed to stabilise CreB and a model was proposed by Lockington et al., where the CreB/CreC complex acts directly on CreA and affects the stability of CreA [166]. This model, however, still lacks adequate biochemical proof.

5.6 PacC

In nature, the enzymatic breakdown of plant cell wall polymers can occur in different surrounding pH. Thus, a mechanism is needed to ensure that only enzymes functional in a certain pH range are produced with respect to the ambient pH of the culture habitat. The regulatory mechanism controlling pH-dependent transcriptional regulation has been analysed in detail in A. nidulans and a major role has been demonstrated for the zinc finger transcription factor PacC [167, 168], which acts as an activator for alkaline-expressed genes and prevents expression of acid-expressed genes. PacC contains a DNA-binding domain with three Cys2His2 zinc fingers and it binds to promoter sites containing the core hexanucleotide sequence 5′GCCARG [167169]. In response to a signal transduced by the pal genes, PacC is proteolytically processed to its functional form [170173]. This proteolytic cleavage removes? 420 C-terminal amino acids that contain the negative-acting C-terminal domain [173, 174]. Conformation of the full-length PacC alternates between a protease-resistant and protease-sensitive form, the latter being cleaved to the active form in alkaline conditions. The genes regulated by PacC or its homologues include enzymes involved in the synthesis of biologically active metabolites such as penicillin [167, 175], and proteases [176]. Also xylanases [177] and an endopolygalacturonase involved in pectin degradation have been shown to be under PacC control [178]. Studies with alkaline- and acidic-mimicking mutants of PacC showed that two genes, xlnA and xlnB encoding xylanases in A nidulans, are also regulated by the PacC. XlnB, encoding a minor xylanase, is expressed under conditions of acidic ambient pH, and xlnA encoding a major xylanase at alkaline ambient pH. Using the same mutants, Gielkens et al. [136] also established the pH-dependent regulation of the abfB gene encoding α-l-arabinofuranosidase, an enzyme releasing l-arabitol from hemicelluloses. In S. sclerotiorum mutants with loss-of-function alleles of the pac1 gene, the expression of the endopolygalacturonase gene pg1 that is important for pathogenesis, was increased when the fungus was shifted to higher ambient pH [179].

5.7 AreA

In filamentous fungi the regulation of nitrogen catabolic enzymes during the use of secondary nitrogen sources such as nitrate, nitrite and purines involves GATA factors that are a class of eukaryotic transcriptional activators or repressors characterised by a highly conserved DNA binding motif comprising a Cys(4) zinc finger followed by a basic domain. This motif mediates binding to DNA elements that have the core sequence 5′GATA. The major global positive-acting regulators, involved in nitrogen metabolism, are the GATA factors AreA and Nit-2 of A. nidulans and N. crassa, respectively, that activate transcription of many structural genes encoding enzymes for nitrogen source catabolism under nitrogen limiting conditions [180182]. In most cases the activation of nitrogen catabolic genes requires two distinct positive factors, AreA/Nit-2 that activates the nitrogen catabolic genes in the absence of a preferred nitrogen source such as l-glutamine (nitrogen derepression), and a pathway specific regulator, e.g. NirA or Nit-4, that mediates specific induction by nitrate in A. nidulans and N. crassa, respectively [183, 184]. The regulation of nitrogen catabolic genes is partly mediated by chromatin modelling since AreA of A. nidulans has been shown to be involved in chromatin remodelling in vivo of the niiAniaD promoter in A. nidulans [185]. Recently, A. nidulans areA mutants were used to study the role of AreA in the production of cellulases. The amount of total secreted cellulase activity was elevated in a strain containing a constitutively activating areA allele whereas the cellulase amount was reduced in an areA loss-of-function-mutant in cellulose induced cultures where ammonium was used as a nitrogen source. In accordance, mRNA ofcbhA was detectable only in the strain containing the constitutively activating allele of areA [186]. The results of the role of AreA in cellulase regulation, however, are still very preliminary since only one time point was studied in Northern analysis and only crude enzymatic assays were used to estimate the level of cellulase expression. The findings, nevertheless, suggest a role for AreA in the expression cellulases in A. nidulans, and cross-talk between the carbon and nitrogen utilisation pathways in filamentous fungi.

6 Repression of genes encoding extracellular enzymes in response to secretion stress

Impaired protein folding and transport in the secretory pathway induces various stress responses in the eukaryotic cells. Accumulation of unfolded proteins in the endoplasmic reticulum (ER) activates the unfolded protein response (UPR) pathway, which leads to induction of a broad set of genes required for enhancement of protein folding and their further transport as well as for clearance of the cells from the unfolded proteins. The mechanism of the UPR induction was first characterised in the yeast Saccharomyces cerevisiae and, currently the stress response is well-known also in higher eukaryotes (for reviews see [187190]. In S. cerevisiae the response is mediated by a transmembrane kinase/endonuclease Ire1p that is required for generation of the actively translated form of the transcript encoding Hac1p, the positive regulator of the UPR target genes. The counterparts of HACI and IREI have also been characterised in the filamentous fungus Trichoderma reesei [191, 192]. In mammalian cells, the UPR response is more complex involving several different transmembrane sensors, IRE1α and β, ATF6, and the kinase PERK. The mammalian ER stress response involves also translational control, inhibition of general protein translation as well as enhancement of synthesis of specific proteins under the stress conditions. The translational control is mediated by phosphorylation of the translation initiation factor eIF2α by the PERK kinase [187].

In filamentous fungi in particular, the protein production in vast quantities sets a requirement for an efficient system to eliminate deleterious effects of limitation in the secretory capacity. Recently, a novel type of regulatory mechanism has been found to down-regulate the genes encoding extracellular proteins in response to secretion stress in filamentous fungi. Treatment of T. reesei cultures with chemical agents shown to inhibit protein folding or transport, such as DTT or brefeldin A, resulted in rapid decrease in the steady-state mRNA levels of genes encoding secreted proteins [193]. The genes analysed and found to be subjected to this feedback control system included cbh1 and cbh2 (encoding cellobiohydrolases I and II), egl1 and egl2 (encoding endoglucanases I and II), xyn1 (encoding xylanase I) and hfb2 (encoding hydrofobin II). Interestingly, Northern analysis of a group of genes encoding intracellular proteins showed no down-regulation under these conditions. Among these constitutively expressed genes was bgl2 encoding an intracellular β-glucosidase. The expression pattern of bgl2 gene on different carbon sources resembles that of the cellulase genes [194]. The result suggests that the down-regulation phenomenon is characteristic to the genes encoding secreted proteins under secretion stress conditions. Expression of E. coli lacZ under cbh1 promoter sequences of different length was used as a reporter system to demonstrate that the cbh1 promoter sequences are involved in mediating the RESS response, and that the response is thus controlled at the transcriptional level. The RESS phenomenon is most likely independent of the cre1 mediated carbon catabolite repression system as the down-regulation of the transcripts took place also in strains deficient in a cre1 mutant strain.

A similar repression mechanism has been shown to be activated in A. niger. In DTT treated cultures of A. niger, the transcript levels of the genes encoding secreted glucoamylase and aspergillopepsin were down-regulated, whereas those encoding intracellular γ-actin and glyceraldehyde 3′phosphate dehydrogenase were not. Similarly, low transcript levels of the glucoamylase and aspergillopepsin genes were detected also in an A. niger strain expressing an antisense construct of pdiA and producing thus reduced amount of the foldase PDI [195].

Typically, the RESS phenomenon is observed under conditions that activate the UPR pathway [193]. Currently, it is not known whether the factors mediating UPR are directly involved in triggering RESS or whether the phenomena are independent. However, it is unlikely that the transcription factor HACI would be involved, as no consensus UPR element has been found in the cbh1 promoter known to be under RESS control. In addition, down-regulation of the transcripts has been shown to occur under conditions where the hac1 activation is weak or negligible, e.g. in A. niger expressing the pdiA antisense construct or in T. reesei cultures treated with the ionophore A23187 or tunicamycin. However, it is possible that a minor amount of the regulating factor is sufficient for the response and/or components, which are more upstream in the signalling pathway, are involved. It is also possible that the ER stress response signalling is more diverse in filamentous fungi compared to the UPR system in the yeast S. cerevisiae, and additional not yet discovered components are involved. As no homologues of PERK kinase have been found in lower eukaryotes, it is unlikely that general inhibition of translation would be involved in the fungal stress response under conditions where protein folding or transport becomes limiting. Activation of RESS provides the fungal cells an alternative way to efficiently reduce the protein load in the secretory pathway and thus to alleviate secretion stress.

7 Conclusions and future perspectives

The capacity of filamentous fungi to efficiently degrade plant polymers is an important feature that provides an interesting topic for studies concerning for instance microbial ecology and the basic mechanisms of nutritional regulation, and also provides a number of possibilities for industrial applications. Cellulose is chemically simple consisting of only glucose but in particular in its crystalline form very rigid and one of the most resistant materials found in nature. Hemicelluloses, on the other hand, are chemically highly variable and their hydrolysis leads to the formation of a variety of pentose and hexose sugars and acids, while lignin is a structurally undefined phenolic polymer. In order to understand this highly complex system, one needs to take into account the large variety of the polymer structures, the corresponding enzymatic diversity, the various oligomeric and monomeric sugars, acids and phenolic compounds created through the enzyme action, and subsequently the intracellular regulatory signals and metabolic responses these compounds provoke in the fungus. Furthermore, different fungal species ranging from anaerobic rumen fungi to saprophytic species in soil and plant pathogens that are highly specialised for particular plant host species, have each evolved to produce their particular set of enzymes that suit the fungal habitat. It is a challenge to study a system where the natural substrate, the carbon source, is very complex and often undefined, and furthermore changes in its structure with the culturing time due to enzymatic hydrolysis, in principle turning gradually from an inducing carbon source to a repressing one. On the other hand, studies carried out with polymeric model substrates or small soluble pure compounds can rarely mimic the regulatory challenges the fungi are facing to survive and compete in the nature.

The main aim of this research field has over the years been to gain understanding on the enzymatic diversity. Thus it is not surprising that much the information that has accumulated on regulation of enzyme production is quite sporadic and diverse as well. We still lack comprehensive understanding of the mechanism of induction by even one inducing compound, and still lack formal verification for many of the hypothesis set forth in the literature [199, 200]. It is only recently that the genes for the metabolic pathways for the utilisation of the pentose sugars xylose and arabinose have been characterised from eukaryotic microbes [111, 196198]. We are still far from understanding the nature of the intracellular inducers, and the possible signalling pathways and cofactors that mediate the induction to the enzyme gene promoters.

During the recent years genes encoding for the plant cell wall degrading enzymes have shown to be subject to regulation by various different factors and some of the cis-acting promoter elements have been characterised. Fig. 2 shows the most well studied inducing compounds and the positive and negative acting regulatory factors that are known. XlnR, ACEI and ACEII are factors that were specifically identified due to their roles in regulating the plant polymer degrading enzymes. Of the more general factors, the role of the until now best characterised factor, the glucose repressor protein Cre, is well established, while establishing the roles of the HAP complex, the pH regulator PacC and the nitrogen regulators AreA/Nit-2 and NirA/Nit-4 in regulation of these genes still need further experimentation. The results until now obtained suggest that we still miss some key regulatory factors even for the most highly expressed genes such as those encoding cellulases and xylanases.

Figure 2

Schematic representation of the different fungal trans-acting factors and regulatory responses affecting cellulase and xylanase expression. The carbon catabolite repressor CRE, activators XlnR and ACEII, the repressor ACEI, the CCAAT binding Hap2/3/5 complex, pH regulator PacC, and the nitrogen regulator AreA are shown. Enzyme load in the secretory pathway can activate a negative feed-back signal that down-regulates enzyme gene transcription (RESS). Outside the fungal hyphae are listed some known and possible environmental factors affecting cellulase and xylanase expression. The gene expression activating and repressing effects of the sugars depends on their concentration.

It appears that there is significant amount of overlap between the regulatory mechanisms, e.g. the same regulatory protein, such as XlnR and ACEII, regulates expression of cellulases as well as xylanases and some substrates induce expression of a variety of extracellular enzymes. This reflects the co-occurrence of the polymers in plant material in the nature. On the other hand, it appears that for instance ACEII mediates induction on cellulose-containing medium but not on sophorose. It is also clear that specific regulation for a certain class of enzymes occurs as well, and that fungi have a means to fine tune expression of the enzymes and their production levels according to a particular substrate and for instance the side chain sugar composition. Genome sequencing has revealed that there can easily be more than 100 genes in a fungus encoding extracellular hydrolytic enzymes, which complicates regulatory studies. A further complication is, for instance, that several genes encoding highly homologous ligninolytic enzymes exist in white rot fungi, and these appear to be differentially regulated.

Despite – and because of – the complexity, the fungal regulatory mechanisms controlling plant biopolymer utilisation are well worth of study. This subject provides many special features of basic and applied interest. Firstly, the utilisation of plant polymers is a special but one of the important means of microbes to acquire nutrients in the nature. The regulation of the production and secretion of the extracellular enzymes on one hand, and the metabolic pathways for the utilisation of the hydrolysis products on the other, need to be coordinated. The released sugars include both hexose (e.g. glucose, galactose) and pentose (xylose, arabinose) sugars that enter to the cell metabolism through different metabolic pathways, e.g. glycolysis and the pentose phosphate pathway. It is an interesting fundamental question how the cell balances the utilisation of these sugars, and how strong impact glucose repression has on the utilisation of the other sugars, and on repression of enzyme production. The hydrolysis of the polymers provides a natural example where the oligosaccharide and monomer concentration vary and thus provides an interesting system to study the balancing of enzyme production, polymer hydrolysis and sugar utilisation by the fungus, as well the impact of the sugar concentrations on their inducing/repressing effects. Furthermore, the novel finding that a load in the secretory pathway has a negative feed-back on the expression of genes encoding secreted enzymes (RESS) provides yet a completely new aspect towards understanding the interrelationships of different cellular processes that are important for fungi that obtain their carbon and energy from living on plant biomass in nature.

Fungal molecular and cell biology techniques have developed rapidly over the last 10–15 years. Although the techniques are still more tedious and time consuming to perform with filamentous fungi than with many other microbes, in principle fungi are now amenable to any experimental approach. The number of fully sequenced fungal genomes will increase rapidly according to the current plans (see e.g. http://www.broad.mit.edu/annotation). This will provide highly valuable information for comparative genomics, for the unravelling of the regulatory pathways, and for the studies concerning quantitative and temporal cellular events that occur as a response to the various plant polymers and their derivatives in the fungal environment.

Acknowledgements

We thank Pauliina Lankinen (University of Helsinki) for useful discussions and the financing from the research programme “VTT Industrial Biotechnology” (Academy of Finland; Finnish Centre of Excellence programme, 2000–2005, Project no. 64330).

References

  1. [1].
  2. [2].
  3. [3].
  4. [4].
  5. [5].
  6. [6].
  7. [7].
  8. [8].
  9. [9].
  10. [10].
  11. [11].
  12. [12].
  13. [13].
  14. [14].
  15. [15].
  16. [16].
  17. [17].
  18. [18].
  19. [19].
  20. [20].
  21. [21].
  22. [22].
  23. [23].
  24. [24].
  25. [25].
  26. [26].
  27. [27].
  28. [28].
  29. [29].
  30. [30].
  31. [31].
  32. [32].
  33. [33].
  34. [34].
  35. [35].
  36. [36].
  37. [37].
  38. [38].
  39. [39].
  40. [40].
  41. [41].
  42. [42].
  43. [43].
  44. [44].
  45. [45].
  46. [46].
  47. [47].
  48. [48].
  49. [49].
  50. [50].
  51. [51].
  52. [52].
  53. [53].
  54. [54].
  55. [55].
  56. [56].
  57. [57].
  58. [58].
  59. [59].
  60. [60].
  61. [61].
  62. [62].
  63. [63].
  64. [64].
  65. [65].
  66. [66].
  67. [67].
  68. [68].
  69. [69].
  70. [70].
  71. [71].
  72. [72].
  73. [73].
  74. [74].
  75. [75].
  76. [76].
  77. [77].
  78. [78].
  79. [79].
  80. [80].
  81. [81].
  82. [82].
  83. [83].
  84. [84].
  85. [85].
  86. [86].
  87. [87].
  88. [88].
  89. [89].
  90. [90].
  91. [91].
  92. [92].
  93. [93].
  94. [94].
  95. [95].
  96. [96].
  97. [97].
  98. [98].
  99. [99].
  100. [100].
  101. [101].
  102. [102].
  103. [103].
  104. [104].
  105. [105].
  106. [106].
  107. [107].
  108. [108].
  109. [109].
  110. [110].
  111. [111].
  112. [112].
  113. [113].
  114. [114].
  115. [115].
  116. [116].
  117. [117].
  118. [118].
  119. [119].
  120. [120].
  121. [121].
  122. [122].
  123. [123].
  124. [124].
  125. [125].
  126. [126].
  127. [127].
  128. [128].
  129. [129].
  130. [130].
  131. [131].
  132. [132].
  133. [133].
  134. [134].
  135. [135].
  136. [136].
  137. [137].
  138. [138].
  139. [139].
  140. [140].
  141. [141].
  142. [142].
  143. [143].
  144. [144].
  145. [145].
  146. [146].
  147. [147].
  148. [148].
  149. [149].
  150. [150].
  151. [151].
  152. [152].
  153. [153].
  154. [154].
  155. [155].
  156. [156].
  157. [157].
  158. [158].
  159. [159].
  160. [160].
  161. [161].
  162. [162].
  163. [163].
  164. [164].
  165. [165].
  166. [166].
  167. [167].
  168. [168].
  169. [169].
  170. [170].
  171. [171].
  172. [172].
  173. [173].
  174. [174].
  175. [175].
  176. [176].
  177. [177].
  178. [178].
  179. [179].
  180. [180].
  181. [181].
  182. [182].
  183. [183].
  184. [184].
  185. [185].
  186. [186].
  187. [187].
  188. [188].
  189. [189].
  190. [190].
  191. [191].
  192. [192].
  193. [193].
  194. [194].
  195. [195].
  196. [196].
  197. [197].
  198. [198].
  199. [199].
  200. [200].
View Abstract