• The genetic potential of ectomycorrhizal fungi to produce N-acetylhexosaminidases was investigated here. N-acetylhexosaminidases are enzymes that cleave monosaccharides from oligomers of N-acetylhexosamines and play an important role in the degradation of chitin.
• Degenerate PCR-primers were designed against genes coding for N-acetylhexosaminidases in basidiomycetes. PCR was performed with DNA templates extracted from sporocarps of 26 ectomycorrhizal fungal species and two saprotrophs.
• PCR-products were obtained from 18 species representing 12 genera distributed throughout the basidiomycete phylogeny. Sequencing confirmed that the products were homologous with N-acetylhexosaminidase genes from plants, animals and other fungi. Some species yielded two PCR-products representing isoenzymes.
• Chitin constitutes a potentially important nitrogen source in soil. Our results demonstrate that a wide range of ectomycorrhizal fungi have the genetic potential to produce N-acetylhexosaminidases, and the expression of this potential would enable them to exploit polymers of amino sugars as a source of nitrogen for themselves and their host plants.
Boreal forest ecosystems with acidic soils and a low availability of mineral nitrogen are generally dominated by plants associated with ectomycorrhizal fungi (Read & Perez-Moreno, 2003). Many ectomycorrhizal fungi can take up small organic molecules, such as amino acids, and are able to supply their host plants with nitrogen from these sources (Finlay et al., 1992; Näsholm et al., 1998). A major fraction of the organic nitrogen in forest soils is, however, incorporated into polymeric macromolecules that are too large for direct uptake. Most of the potential nitrogen sources in acidic temperate forests thus require degradation by extracellular enzymes into smaller organic molecules before microorganisms may utilize them (Leake & Read, 1997; Read & Perez-Moreno, 2003). This article presents a study of basidiomycete genes coding for N-acetylhexosaminidases (EC 220.127.116.11), which are enzymes that cleave monosaccharides from oligomers of amino sugars.
Chitin is a polymer composed of β-1,4 linked N-acetylglucosamine units. It is produced by fungi as a component of their cell walls and is also found in the shells and cuticles of invertebrates (Gooday, 1990). Chitin is one of the most abundant organic polymers in nature. A major fraction of the global pool is found in marine environments, but chitin is also abundant in terrestrial environments. In the organic top-layers of a boreal forest podzol, Bååth & Söderström (1979) determined that up to 20% of the total soil nitrogen might be incorporated into dead and alive fungal mycelium. Dead or vacuolated hyphae, mainly consisting of chitin-containing cell walls, make up a major fraction (> 90%) of the mycelium in these soils (Söderström, 1979). Commonly, 4–10% of the total nitrogen in soil is found in polymers of amino sugars (Kelley & Stevenson, 1996; Johnsson et al., 1999; Greenfield, 2001) and from heathland soil figures up to 30% have been reported (Kerley & Read, 1997). A major fraction of soil nitrogen is incorporated into recalcitrant polyphenolic complexes (Kelley & Stevenson, 1996). However, nitrogen in amino sugars and amino acids is more readily available to soil organisms compared with the bulk of the organic nitrogen in soil (Johnsson et al., 1999). Polymers of amino sugars, such as chitin, thus constitute a potentially important source of nitrogen for soil organisms.
The ability of ectomycorrhizal fungi to utilise chitin in axenic culture has been tested only for four species within the Boletoid clade, with some displaying increased growth when chitin was present as the only nitrogen source (Leake & Read, 1990; Hodge et al., 1995). In the growth media, Hodge et al. (1995) also detected the activity of chitinolytic enzymes by using fluorogenic substrates. N-acetylhexosaminidase activity has also been found to be associated with the fungal mantle surrounding certain ectomycorrhizal beech roots (Hodge et al., 1996). Dighton et al. (1987) found increased decomposition of chitin in microcosms with Pinus contorta Dougl. seedlings colonized by the ectomycorrhizal fungus Suillus bovinus (L.Fr.) Roussel compared to systems with non-colonized seedlings, although the increase was not statistically significant. It therefore appears that at least some ectomycorrhizal fungi are able to break down chitin into compounds that could serve as a nitrogen source for themselves and their host plants.
Enzymes involved in chitin degradation have been extensively studied in mycoparasitic strains of ascomycetes, particularly within the genus Trichoderma, which are frequently used as biocontrol-agents of fungal plant pathogens (Lorito, 1998). Chitin degradation is initiated by endochitinases (EC 18.104.22.168) that hydrolyse bindings between N-acetylglucosamine residues at random locations within the chitin macromolecule. Degradation by endochitinases disrupts the structural integrity of chitin and produces oligomers of N-acetylglucosamine of varying length. N-acetylhexosaminidases then further degrade the oligomers, releasing monosaccharides from the non-reducing ends. A third type of enzyme, chitobiosidase, cleaves chitobiose, which is the disaccharide of N-acetylglucosamine, from the end of chitin chains (Lorito, 1998). Chitobiose, released by chitobiosidases, may possibly be taken up intact by fungi (Gooday, 1990), or split by N-acetylhexosaminidases into N-acetylglucosamine. The nomenclature used to describe enzymes involved in chitin degradation is not consistent. N-acetylhexosaminidases and chitobiosidases have both been termed exochitinases and N-acetylhexosaminidases have commonly been termed chitobiases. N-acetylhexosaminidases involved in chitin degradation are usually termed N-acetylglucosaminidases, but since these enzymes also have been described to act against substrates containing galactosamines (Cannon et al., 1994), the more general term is more suitable, until detailed information on substrate specificity is obtained.
Endochitinases play a variety of roles in organisms. In addition to be involved in degradation of chitin for nutritional purposes, endochitinases play important roles in fungal morphogenesis and are also involved in recognition and antagonism against fungi by both fungi and plants (Sahai & Manocha, 1993). In contrast to endochitinases, N-acetylhexosaminidases alone are unable to degrade intact chitin (Tronsmo & Harman, 1993). Their function outside organisms should therefore be restricted to the processing of degradation products, produced by other chitinolytic enzymes, into smaller assimilable compounds.
The aim of the present study was to investigate the incidence of N-acetylhexosaminidase-encoding genes in ectomycorrhizal fungi. Degenerate PCR primers specific for genes coding for N-acetylhexosaminidases were designed based on DNA-sequences from Trichoderma harzianum (Draborg et al., 1995) in combination with genome sequence data from fully sequenced basidiomycetes. The primers were then used with DNA templates from identified ectomycorrhizal sporocarps representing a range of phylogenetically widely separated taxa. PCR amplification products were sequenced and the predicted amino acid sequences were compared to published database sequences.
Materials and Methods
The amino acid sequence of a N-acetylhexosaminidase (exc1) from T. harzianum (NCBI accession number AAB47060; Draborg et al. 1995) was used to find homologies within the translated full genome sequences of the saprotrophic basidiomycetes Phanerochaete chrysosporium strain ‘RP 78’ (DOE Joint Genome Institute, Walnut Creek, CA, USA) and Coprinus cinerea strain ‘Okayama 7’ (Fungal Genome Initiative, USA) using tBLASTn (Altschul et al., 1997). Three homologous genes (expectancy < 1e-80) were identified in P. chrysosporium, and two homologues (expectancy < 1e-46) were found in C. cinerea. These five genes together with an N-acetylhexosaminidase cDNA sequence from the wood rotting basidiomycete Hypholoma fasciculare (NCBI accession number AY615427) were compared, and conserved sites were identified. Two primer sites were selected and two different degenerate primers were designed for each site. Primers 1a and 1b coded for different amino acid sequences and primers 2a and 2b coded for the same amino acid sequences but with 2a more specific than 2b (Table 1).
Table 1. PCR primers developed for amplification of the central parts of genes coding for N-acetylhexosaminidases in basidiomycetes
Amino acids coded for
In total, 28 fungal species were investigated; 26 ectomycorrhizal and two saprotrophic species (Table 2). Nomenclature follows Hansen & Knudsen (1992, 1997). Sporocarps were collected from the area around Uppsala, central Sweden. For the Piloderma species, Paxillus involutus and H. fasciculare, pure culture material was used. Extraction was performed from dried sporocarp material or fresh culture material, which was homogenised in buffer containing 3% (w/v) CTAB, 2.5 m NaCl, 0.15 m Tris and 2 mm EDTA and incubated at 65°C for 1 h. After centrifugation, the supernatant was extracted once with chloroform, after which nucleic acids were precipitated through the addition of 1.5 volumes of isopropanol. The pellet was washed with 70% ethanol and re-suspended in water.
Table 2. Strains of ectomycorrhizal and saprotrophic (indicated by *) fungi that were screened for the presence of genes coding for N-acetylhexosaminidases
PCR was performed in a GeneAmp PCRsystem 2700 thermal cycler (Applied Biosystems, Foster City, CA, USA) using all four combinations of primers. Initial denaturation at 95°C for 5 min was followed by 35 amplification cycles of denaturation at 95°C for 30 s, annealing at 55°C for 30 s and extension at 72°C for 30 s. The thermal cycling was ended by a final extension step at 72°C for 7 min. The PCR reaction solutions contained 10 mm Tris-HCl, pH 8.3, 50 mm KCl, 1.85 mm MgCl2, 0.2 mm dNTP, 1 µm of each degenerate primer, and 25 µml−1 RedTaq polymerase (Sigma-Aldrich Ltd, St. Louis, MO, USA). Template concentrations were optimized separately for each sample.
PCR amplification products were cloned into Escherichia coli using TOPO TA cloning® and One-Shot® TOP10 chemically competent bacteria (Invitrogen, Carlsbad, CA, USA) according to the manufacturer's protocols. PCR products from reactions with the more specific primer 2a were used only when no products were obtained from PCR reactions with primer 2b. For each cloned PCR product, 16 clones were selected, and the cloned fragments were PCR amplified using the M13 Forward (−20) and M13 Reverse primers and the same PCR conditions as earlier but with primer concentrations reduced to 0.25 µm and bacterial cells as templates. PCR products from the clones were digested, using the restriction enzymes CfoI or TaqI according to the manufacturers’ recommendations. Cleaved PCR products were separated on 1.8% agarose (D1, Laboratorios Conda, SA, Madrid, Spain) gels, stained with ethidium bromide and visualized under UV light, after which the clones were sorted into groups according to their restriction patterns. From each restriction pattern group, the PCR products from three clones were selected for sequencing (if a restriction pattern group contained less than three clones, all were sequenced). PCR products were purified using the QIAquick PCR purification kit (QUIAGEN GmbH, Hilden, Germany). Sequencing was performed on an ABI PRISM™ 310 Genetic Analyzer (Applied Biosystems) using BigDye™ Terminator and the M13 Forward (−20) primer according to the manufacturer's protocols.
Sequences were aligned according to their predicted amino acid sequences using the program ‘Dialign 2’ (Morgenstern et al., 1996). Introns were identified by comparison with a cDNA sequence from H. fasciculare (NCBI accession number AY615427). After removal of introns and primers, similarity between protein sequences was analysed using the program ‘paup* 4.0b10’ (Swofford, 2002). A distance tree was created using neighbour joining, and the support of branches was tested by bootstrap analysis with 1000 replicates.
Results and Discussion
PCR products, visible as distinct bands on agarose gels, were obtained from 19 of the 28 tested taxa. From Polyporus melanopus, a PCR product was obtained with primer 1b, whereas all other products came from PCR reactions with primer 1a. In some cases, PCR products were obtained with primer 2a, but not with primer 2b. Often products were obtained with both 2a and 2b. In all cases where products were obtained from primer 2b, products were also obtained from primer 2a.
Sequencing of the cloned products and comparison with sequence databases (NCBI) showed that most of the obtained sequences were homologous with identified genes coding for N-acetylhexosaminidases from ascomycetes, animals, plants and bacteria, as indicated by BLASTx expectancy values of e-7 to e-17 for the best matches. However, some of the obtained sequences showed no homology with N-acetylhexosaminidase genes, indicating that the primers were not completely specific, at least not under the PCR conditions used in this study. These sequences either showed homology to genes with other functions without any apparent pattern, or did not match any database sequences.
PCR products homologous to N-acetylhexosaminidase genes were found in 18 taxa of which 16 were ectomycorrhizal (Table 2). In 6 species, two different sequences corresponding to N-acetylhexosaminidase genes were identified (Table 2). The finding of isoenzymes is consistent with the presence of two isoforms of N-acetylhexosaminidases in T. harzianum (Draborg et al., 1995) and Trichoderma virens (Kim et al., 2002) as well as with the presence of two and three N-acetylhexosaminidase genes within the genomes of P. chrysosporium and C. cinerea, respectively.
No PCR products were obtained for any of the four tested species within the genera Lactarius, neither for species within the cantharelloid clade or from Hygrophorus persoonii, Gomphus clavatus, Pisolithus arhizus and Thelephora terrestris. Absence of PCR products should not, however, be interpreted as absence of N-acetylhexosaminidase genes. Although PCR products were obtained for a wide range of different basidiomycetes, mutations in the primer sites may have occurred during the evolution of some groups of fungi. Similarly, the fact that most of the investigated species yielded only one single N-acetylhexosaminidase sequence does not preclude the presence of several other isoforms.
The amplified fragments ranged from 429 to 618 nucleotides in length. Aided by the cDNA sequence from H. fasciculare, it was possible to identify 5 different introns within the amplified fragments. Two of the introns were present in almost all investigated sequences, two were present in about half of the sequences, whereas one of the introns was found only in Piloderma byssinum. After removal of introns, 372 base pair long fragments remained corresponding to 124 amino acids. One of the sequences obtained from Tricholoma focale was only 118 amino acids long, due to a deletion of six amino acids, and in the sequence obtained from Piloderma fallax, a 23 amino acid fragment was missing compared to sequences from other species.
The obtained protein sequences showed homology to a 72–73 kDa protein in Trichoderma spp. (Draborg et al., 1995; Peterbauer et al., 1996; Kim et al., 2002). When the basidiomycete sequences were compared to the Trichoderma sequences using BLASTp (Altschul et al., 1997), expectancy values of between e-21 and e-13 were obtained and the sequences had 33% to 46% identical amino acids. The exception was the shorter sequence of P. fallax, which had only 25% amino acids identical to Trichoderma sequences and a BLAST expectancy value of e-6. The presence of other differently sized N-acetylhexosaminidases has been demonstrated in Trichoderma spp. (Ulhoa & Peberdy, 1991; Haran et al., 1995). However, the 72 kDa enzyme seems to be responsible for a major fraction of the extracellular N-acetylhexosaminidase activity produced in liquid culture, as a mutant with the gene disrupted produced only low activity (Brunner et al., 2003).
In a tree showing sequence similarity between the different PCR products (Fig. 1), sequences from species within the same genus usually clustered together, with amino acid sequences being > 79% identical to each other. However, sequences from within single species did not always cluster together; P. chrysosporium, C. cinerea, Amanita crocea and two Tricholoma species each yielded two or three sequences that were 53–74% similar to each other. On the other hand, duplicate genes from P. byssinum and Suillus flavidus were similar within the species (81% similarity). The two sequences obtained from Amanita virosa were highly similar to each other (98% similarity) and may represent two alleles of the same gene. The resemblance between sequences from Paxillus involutus and Suillus spp. is consistent with both genera being placed within the boletoid clade (Hibbet et al., 2000). It thus appears that gene duplications have occurred at different stages during basidiomycete evolution; early, as exemplified by the Tricholoma species, and later, as exemplified by P. byssinum and the boletoid clade.
An interesting observation is that the sequences from P. byssinum and P. fallax are only 46–49% similar to each other. One explanation for this dissimilarity could be that the P. fallax gene has degenerated and is not functional, as suggested by the deletion of amino acid that are conserved in other species. In addition to the diverging gene, P. fallax may, however, contain genes that are more similar to the P. byssinum gene but that remain undetected in the present study.
The presence of N-acetylhexosaminidase genes in basidiomycetous fungi representing several different phylogenetic clades (Table 2) indicates that the genetic capacity for extracellular degradation of amino sugar oligomers is widespread within this group of fungi. Hibbet et al. (2000) underlined that the ectomycorrhizal habit is evolutionary unstable and has evolved on several independent occasions and that several present saprotrophic basidiomycetes seem to have evolved from ectomycorrhizal ancestors. Members of the two functional groups, ectomycorrhizal and saprotrophic fungi, may therefore be similar with respect to their overall physiology. This notion is well illustrated by the presence of genes for enzymes degrading complex organic compounds in both groups. Further studies are, however, necessary to reveal under what conditions, if ever, the N-acetylhexosaminidase genes in ectomycorrhizal fungi are expressed and if the products are functional enzymes. Other genes coding for extracellular enzymes involved in the degradation of complex substrates that have so far been identified in ectomycorrhizal fungi are: a manganese peroxidase in Tylospora fibrillosa (Chambers et al., 1999), two different proteases in Amanita muscaria (Nehls et al. 2001) and laccases in a range of different taxa (Chen et al., 2003).
In Trichoderma atroviride, N-acetylhexosaminidases are produced in response to carbohydrate starvation. They are also produced in response to a shortage of nitrogen, even when glucose is provided in excess, but only in the presence of hexosamine oligomers in the medium (Donzelli & Harman, 2001). Högberg et al. (1999) showed that a wide range of studied ectomycorrhizal fungi depended mainly on current photoassimilates for their supply of carbohydrates. Ectomycorrhizal fungi would therefore benefit from the production of N-acetylhexosaminidases mainly in improved nitrogen nutrition for the fungi and their host plants. In order to exploit the significant nitrogen pool incorporated into soil chitin, fungi would, in addition to N-acetylhexosaminidases, also have to produce endochitinases that can degrade chitin into oligosaccharides, which the N-acetylhexosaminidases degrade further to N-acetylglucosamine. In acidic forest soils nitrogen mineralization may often be limited, due to the high availability of carbohydrates in the litter, low availability of nitrogen, and efficient conservation of nitrogen within saprotrophic biomass (Lindahl et al., 2002). In these types of ecosystems it is likely that N-acetylglucosamine and amino acids may replace ammonium and nitrate as the principal sources of nitrogen for ectomycorrhizal plants (Read & Perez-Moreno, 2003).
Katarina Ihrmark, Malin Elfstrand, Åke Olsson, Olov Pettersson and Gregory Heller are gratefully acknowledged for their good advice and help in the laboratory, and Ursula Eberhardt for help with the sequence analysis. Financial support from FORMAS (The Swedish Research Council for Environment, Agricultural Sciences and Spatial Planning) is also gratefully acknowledged.