OUP user menu

The ectomycorrhizal community in natural Tuber borchii grounds

Mirco Iotti, Enrico Lancellotti, Ian Hall, Alessandra Zambonelli
DOI: http://dx.doi.org/10.1111/j.1574-6941.2010.00844.x 250-260 First published online: 1 May 2010


Although Tuber borchii is a commercially valuable truffle, its habitat has been virtually ignored. Here, we examine the ectomycorrhizal fungal communities in natural T. borchii grounds. Ectomycorrhizas under T. borchii ascomata and up to 1 m away were collected and morphologically assigned to pine or oak host plants. They were then morphotyped and molecular typed using internal transcribed spacer regions. Seventy ectomycorrhizal taxa were identified, many of which were rare. Tuber borchii dominated, forming 20% of ectomycorrhizas, with Thelephoraceae, Inocybaceae and Sebacinaceae being the other main species. Species composition was markedly affected by the host plant, although community structure and composition was also influenced by the location from which the soil cores were collected. Tuber dryophilum, an edible truffle, but without commercial value, shared the habitat with T. borchii. Its mycorrhizas were never found together with those of T. borchii. Tuber borchii was present on both oaks and pines, but was more abundant in soil cores where the roots of both hosts were present. It is suggested that the presence of young oaks contributed to the maintenance of T. borchii colonization on pines.

  • bianchetto
  • ITS-rDNA
  • morphotyping
  • mixed forest
  • ectomycorrhizal diversity
  • truffière


Truffles are ectomycorrhizal Ascomycetes belonging to genera in the Pezizales. They include the genus Tuber, which produce edible truffles of considerable economic and culinary importance (Hall et al., 2003; Mello et al., 2006). The most commercially important species are Tuber magnatum Pico and Tuber melanosporum Vittad., but other species such as Tuber aestivum Vittad. and Tuber borchii Vittad. are excellent truffles and are becoming increasingly popular in the marketplace (Hall et al., 2007). Tuber borchii (bianchetto) is a particularly important species in north-east Italy, where it is commonly used fresh in quality restaurants as well as in preserved products (Zambonelli et al., 2002).

Since its first cultivation in Italy in 2000 T. borchii has become a more popular species (Zambonelli et al., 2000). It has also been cultivated in New Zealand out of season to the Northern Hemisphere, where it commands a price similar to T. melanosporum (Hall, 2008). Although T. borchii, like the other edible truffles, is commonly found in calcareous soils with pH between 7 and 8, it is also found in acidic soils (Zambonelli et al., 2002; Gardin, 2005). Tuber borchii also has a low host specificity, forming ectomycorrhizas with a wide range of broad leaf trees including oaks, hazel, poplar, linden, chestnut and alder and coniferous species such as pine and cedar (Zambonelli et al., 2002; Hall et al., 2007). It is reported to have a wide distribution in Europe, being found from southern Finland to Sicily and from Ireland to Hungary and Poland (Hall et al., 2007). Because of its wide host range and adaptation to a wide range of soils and climates, it is likely that it will be possible to cultivate T. borchii in areas where other important species of truffle will not grow (Zambonelli et al., 2002). However, information on the ecology of T. borchii is lacking and limited to studies of the soils, vegetation and geology in northern and central Italy (Gardin, 2005), while underpinning research on the microbial environment has been virtually ignored.

Considerable research has been conducted in the past on ectomycorrhizal communities of both natural and cultivated truffle grounds using morphological methods that allow classification of ectomycorrhizal fungi in a habitat using morphotypes (Donnini & Bencivenga, 1995; Zambonelli et al., 2005; García-Falces & De Miguel Velasco, 2008; García-Montero et al., 2008). However, these methods provide only a partial picture of the ectomycorrhizal community because the morphological characteristics of ectomycorrhizas are influenced by age, host tree species and soil conditions. For this reason, it is now recognized that ectomycorrhizal morphotyping should be supported by molecular methods in order to provide a complete and precise picture of the ectomycorrhizal diversity and reveal previously misidentified or undetected species (Dahlberg, 2001; Selosse, 2001). The most popular locus for taxonomic affiliation of ectomycorrhizal mycobionts is the internal transcribed spacer (ITS) of the nuclear ribosomal repeat unit (Horton & Bruns, 2001; Martin, 2007). Although taxonomic reliability in public sequence databases is less than ideal (Bridge et al., 2003; Nilsson et al., 2006), the large number of ITS fungal accessions that have now been generated has increased the usefulness of this region for the purposes of comparison of fungal communities (Nilsson et al., 2008).

Recently, studies of ectomycorrhizal fungal communities of natural white truffle (T. magnatum Pico) truffières in northern and central Italy using rDNA-ITS sequences showed that T. magnatum mycorrhizas were very rare adjacent to where its ascomata were found, whereas the mycorrhizas of other species of truffle such as Tuber rufum Pico and Tuber maculatum Vittad. were frequent (Murat et al., 2005; Bertini et al., 2006; Iotti & Zambonelli, 2006). Similarly, some species of Boletus edulis s.l. group (B. edulis Bull., Boletus pinophilus Pilát & Dermek and Boletus aereus Bull.) are also poorly represented adjacent to fruiting bodies. In contrast, Baestivalis aestivalis (Paulet) Fr. (sin. Breticulatus reticulatus Schaeff.) and Tricholoma matsutake (S. Ito & S. Imai) mycorrhizas are concentrated just below their fruiting bodies (Lian et al., 2006; Peintner et al., 2007). Our research presented here extends these studies to ectomycorrhizal fungal communities in natural T. borchii truffières in Ferrara province, the most productive area in Italy (Zambonelli et al., 2002). In particular, we focused on (1) studying the characteristics of the whole ectomycorrhizal fungal community; (2) comparing the ectomycorrhizal fungal communities in the soil cores collected below the ascomata and at 1 m distance; and (3) comparing the ectomycorrhizal fungal communities on pine and oak, which are the common hosts of T. borchii in the studied area.

Materials and methods

Area studied

The area studied is located inside the regional park ‘Delta del Po’ (44°50′N; 12°15′E) (http://www.parcodeltapo.it/er/Eindex.html) in the littoral area of Ferrara Province (Italy). Inside this park, the T. borchii fruiting areas are man-made pine woods (http://www.parcodeltapo.it/er/natura/ambienti/Epinete.html) that are no longer used for timber production and are slowly reverting to thermophilic evergreen oak woods (Quercion ilicis Br.-Bl. ex Molinier 1934 em. Riv.-Mart. 75) (Barkman et al., 1986) – the natural vegetation for the area (Corticelli et al., 2004). The experimental natural truffières were located in Volano and Mesola conifer woods established between 1933 and 1938, where Pinus pinaster and Pinus pinea are the main planted species, with Quercus ilex dominating the natural shrub layer. The two truffle grounds (Volano and Mesola) are located 10 km apart and are separated by agricultural land. Both the truffières have similar ecological characteristics and are representative of productive T. borchii areas. Truffles were harvested with trained dogs in an area of about 1 ha in each truffière. The ground vegetation is characterized by Phillyrea angustifolia, Asparagus acutifolius, Clematis flammula, Rubia peregrina, Ruscus aculeatus, Osyris alba and Rosa sempervirens (Corticelli et al., 2004). In these areas, the T. borchii host plants are P. pinaster, P. pinea and Q. ilex (Hall et al., 2007).

The mean annual air temperature is 13–15 °C, with a mean annual rainfall of 400–900 mm, with March and October being the wettest months. June and July are the hottest and driest months and January is the coldest month, with a mean daily temperature between 1 and 5 °C (http://www.provincia.fe.it/).

The study area has a sandy calcareous soil classified as Aquic Ustipsamments, mixed, mesic (USDA, 2003) or Calcaric Arenosols (Gleyic) (FAO, 1998) by the ERMES Agricoltura (1994).

Sampling techniques

Between January and April 2007, truffle ascomata were collected in the selected truffières using trained dogs and global positioning system (GPS) coordinates recorded for each. During the survey, 15 ascomata of T. borchii were found: seven in the Volano ground and eight in the Mesola ground. The ascomata were identified using morphological and molecular methods (Pegler et al., 1993; Bonuso et al., 2009), weighed and then deposited in the herbarium of the ‘Centro di Micologia’ in Bologna (CMI-UNIBO). Root samples were collected directly under where each T. borchii truffle was found (Tbo0) using a 6-cm-diameter soil corer 30 cm long. A second was taken randomly 1 m away (Tbo1) to verify whether mycorrhizae of T. borchii were also present near the fruiting points at a distance where most of the dominant ectomycorrhizal taxa usually show patchiness (Lilleskov et al., 2004). Root samples were stored overnight at 4 °C in plastic bags.

Morphological and molecular analysis of ectomycorrhizas

The roots were removed from the soil cores by careful washing over a 2 mm mesh sieve, washed in sterile water and then examined under a dissecting stereomicroscope (× 20). The mycorrhizas were assigned to pine or oak host plant on the basis of the morphological features of the colonized roots (colour, diameter and branching pattern) (Peterson et al., 2004). To assign each mycorrhiza to morphotypes, anatomo-morphological characteristics described by Agerer (1995) were used.

The anatomical structures of the mantles and of the external elements (hyphae, rhizomorphs and cystidia) (Agerer, 2006) of each morphotype were examined under an Eclipse TE 2000-E microscope (× 1000) (Nikon). The degree of ectomycorrhizal infection was measured by counting the number of living mycorrhizal tips of each morphotype in all the root samples and expressing the results as a percentage of infected out of the total number of tips examined.

Four to five infected tips, representative of each morphotype, were selected for molecular analyses, whereas the remaining mycorrhizas were stored in FAA (formaldehyde, 70% ethanol, acetic acid − 5 : 90 : 5) at 5 °C as a reference. The selected tips were vortexed in a microcentrifuge tube for 30 s and then spun for 2 min at 17 089 g to remove soil particles from the symbiotic tissues. These selected tips were then transferred into another 1.5-mL tube containing 500 μL of sterile water and stored at −80 °C pending further molecular characterization.

Molecular identification of the mycorrhizas was performed by applying direct PCR techniques as described by Iotti & Zambonelli (2006), and so avoiding DNA isolation. ectomycorrhizal manipulation for molecular analyses was carried out in Petri dishes containing sterile-distilled water under a dissecting microscope (× 20). The mantles of each morphotype were first cleaned of surrounding hyphae to prevent PCR contamination with rhizoplane fungi; a very small portion, approximately 0.01–0.02 mm2 wide, was removed from each mantle cap, and transferred directly to a PCR tube containing 20 mL of sterile water.

To amplify the ITS1-5.8S–ITS2 region of the ribosomal DNA, PCRs were conducted using a T gradient Thermal Cycler (Biometra, Göttingen, Germany) in 50-μL volume reactions using the primers pair ITS1F and ITS4 (White et al., 1990; Gardes & Bruns, 1993) at a concentration of 300 nM each, 10 mM Tris-HCl (pH 8.3), 50 mM KCl, 1.5 mM MgCl2, 200 μM for each dNTP and 1.5 U of TaKaRa Taq DNA polymerase (Takara, Otsu, Japan). Twenty to 40 μg of bovine serum albumin was added to compensate for PCR inhibitors. The cycling parameters were as follows: 6 min of initial denaturation at 95 °C; 30 cycles of 94 °C for 30 s, 55 °C for 30 s and 72 °C for 1 min; and a final extension step of 72 °C for 10 min. Amplified ITS fragments were electrophoresed in a 1% agarose gel and visualized by staining with ethidium bromide in a GeneGenius Imaging System (SynGene, Cambridge, UK). Images were elaborated using gene-tools analysis software (SynGene).

The amplified products were first purified using the Gene Clean II kit (BIO 101, Vista, CA) and then sequenced using both the primers ITS1F and ITS4. Sequence reaction was performed using the Abi Prism 3700 DNA Analyzer (Applied Biosystem, Foster City, CA) with Big Dye Terminator v3.1 chemistry. The ITS sequences obtained were compared with those present in GenBank database (http://www.ncbi.nlm.nih.gov/BLAST/) and Unite database (Kõljalg et al., 2005) using the blastn search (Altschul et al., 1997). Sequences were regarded as belonging to operational taxonomic units (OTUs) following the criteria cited in Landeweert (2003): sequence similarity of ≥99%, identification to the species level; sequence similarity of 95–99%, identification to the genus level; and sequence similarity of ≤95%, identification to the family or the ordinal level. It is difficult to establish a priori the intraspecific variability within a fungal taxon, but adopting a cut-off value of 99% for OTU identification to the species level can be a satisfactory compromise to separate intra- from interspecific variability particularly as most fungal species have an intraspecific variability of ITS1-5.8S–ITS2 region ranging from 0% to 1% (Nilsson et al., 2008).

Sequences were deposited in GenBank database with the accession numbers FJ210727FJ210782 (‘Supporting Information, Table S2’).

Statistical analysis

Statistical analyses were carried out only on 14 paired samples at Tbo0 and Tbo1 (seven for each truffle ground), with one sample excluded from the analysis because no roots were found in the paired Tbo1 sample.

The relative OTU abundances and frequencies were calculated in soil samples collected in the T. borchii fruiting point (Tbo0) and in the soil samples collected at 1 m (Tbo1). The relative abundance of fungal individual OTUs was calculated in percent of the fungal community per sample as follows: Abri=(xi/n) × 100 (Abri is the relative abundance of the OTU i in the sample, xi is the number of colonized tips of the OTU i in the sample and n is the total number of colonized tips in the sample).

The dominance–diversity curves were constructed by ranking the abundance values of the OTUs from the highest to the lowest (Magurran, 1988).

Frequencies (a measure of the distribution of a species in the area) were calculated as the percentage of occurrence in soil samples. A value close to 100% represents a wide diffusion, whereas a low value indicates the presence of the species in just one or a few samples.

Spearman's test was used to investigate whether there was a possible correlation between the weight of the T. borchii ascomata and the relative abundances of T. borchii colonized tips in the sample collected directly under the ascomata.

We used three diversity measures to characterize the ectomycorrhizal fungal community: richness, Pielou's index and the Shannon–Wiener index (Magurran, 1988), where richness (R) is the total number of distinct OTUs detected; Pielou's index measures how evenly the individuals are distributed among the OTUs (a measure of the evenness of species, where J=1 if all species occur at the same proportion); and the Shannon–Wiener index (H) is a measure that takes into account the taxa richness and their relative abundance (the index is increased either by having additional unique species or by having a greater species evenness).

The diversity measures were compared using Friedman's (1940) nonparametric test. The same test was used to compare the number of root tips found in Tbo0 and Tbo1 and T. borchii relative abundances on oak and pine.

The similarity values between the communities present in Tbo0 and Tbo1 samples and of the pine and the oak were calculated using the Bray–Curtis dissimilarity index. The Bray–Curtis index is one of the most used indices to quantify the compositional dissimilarity between two different communities in ecological studies (Faith et al., 1987). It is bound between 0 and 1, where 0 indicates that the two communities share all the species and 1 indicates that the two communities do not share any species. The Bray–Curtis dissimilarity index was used to create a species dissimilarity matrix, which was analysed using Adonis and Mantel tests and the Constrained Analysis of Principal Coordinates (CAP) (Anderson & Willis, 2003).

The Adonis test was used on the species dissimilarity matrix to evaluate the effect of T. borchii collection site, host plant and sampling distance from the T. borchii ascomata collection points (Tbo0 and Tbo1) on the composition of the fungal community.

The Mantel test was used to compare species dissimilarity matrix and linear distance matrix between sampling points calculated using the GPS coordinates to determine whether autocorrelation among sampling points could influence community structure.

CAP was used to graphically represent trends within the ectomycorrhizal fungal community composition in relation to sampling position (Tbo0 and Tbo1) and to the relative abundance of ectomycorrhizal tips of pine or oak in each sample. For this analysis, the data of the abundance in the species dissimilarity matrix were subjected to the Hellinger transformation (Legendre & Gallagher, 2001).

Statistical analyses were performed using R 2.9 software (R Development Core Team, 2009) with the ‘vegan’ package v. 11.1-4 (Oksanen, 2009).


General description of the fungal community

The 15 ascomata of T. borchii collected in the two truffle grounds ranged between 2 and 10 g. No correlation was found between the size of the T. borchii ascomata and the occurrence of its mycorrhizas (test of Spearman ρ=0.028, P=0.935). The ITS sequences obtained from of all ascomata showed that they belong to the ITS clade 1 of T. borchii described by Bonuso (2009), which is the most common in Italy. Some Tuber dryophilum Tul. ascomata were also found, but never in the same points where T. borchii ascomata were found.

A total of 13 134 colonized root tips were present in the 29 soil cores. Each was examined and assigned to 70 different morphotypes using morpho-anatomical features and 56 OTUs were identified using sequence data from the rDNA-ITS regions (Table 1 and Tables S1, S2). Fourteen morphotypes were considered to be different fungal taxa on the basis of their anatomical features, but were not identified on a molecular basis because the mycorrhizas were too old and/or molecular analyses failed (Agerer, 2006). A close agreement was found between morphological characteristics and molecular data for each OTU and no direct sequencing reaction failed due to multispecies assemblage on the same mantle fragment used as a PCR target. The direct PCR approach minimized the risk of amplification of aspecific amplicon from ectomycorrizas as reported recently by Iotti & Zambonelli (2006).

View this table:

ECM diversity in the total Tuber borchii community, in fruiting (Tbo0) and nonfruiting zones (Tbo1), and associated at Quercus and Pinus

Number of samples2915142424
Number of root tips13 1348931420358837251
Specie richness7050394041
Shannon index3.4332.5293.2203.1253.155
Pielou index0.8080.6460.8710.8470.849
Bray–Curtis dissimilarity index0.6550.580

The values of the Shannon and Pielou indexes (3.433 and 0.808, respectively) indicated that the community was characterized by a high diversity and an even spread of the OTUs (Table 1). The dominance–diversity curve shows that T. borchii was dominant (Fig. 1) and only a few OTUs belonging to Thelephoraceae and Sebacinaceae have a value of relative abundance higher than 3%. Tuber borchii was also present in the majority of the samples (18 soil samples); only 13 taxa were found in three or more samples, while most of the ectomycorrhizal fungi (30 OTUs and 14 unknown taxa) were only found in a single sample (Fig. 1 and Table S1). Some taxa dominated in only one soil sample, for example: Thelephoraceae (three OTUs), Suillus granulatus and Tomentella lapidum. Only three OTUs of thelephoroid fungi and Inocybe rufuloides were relatively abundant and frequent.


Dominance–diversity curves in the studied community. The bars show the incidence of different OTUs.

The ectomycorrhizal community was dominated by basidiomycota (47 OTUs and five unknown taxa) that colonized 70.5% of root tips and ascomycota (nine OTUs, 26.7%). These were primarily Tuber spp. (two OTUs), which were well represented (21.7%) (Fig. 2). The most abundant basidiomycete belonged to Thelephoraceae (42.4%, 20 OTUs) and Sebacinaceae (9.7%, six OTUs), while Inocyb-aceae represented 6.3% of colonized tips (eight OTUs).


Abundance of the fungal families in the whole community in fruiting (Tbo0) and nonfruiting points (Tbo1) and under both Quercus and Pinus. The abundances of rare families are indicated in the boxes at the top of the columns.

Twenty-eight and eight mycorrhizal tips were amplified and sequenced for T. borchii and T. dryophylum, respectively. The diversity of the ITS regions of these two truffle species was >10%; moreover, T. dryophylum had an insertion of 84 bp in the ITS1 region, whereas the intraspecific variability was found to be very low: 0% for T. borchii (0/481 nt) and 0.4% for T. dryophylum (2/564 sites). The morphology of the mycorrhizas was similar, except for the mantle anatomy (Table S2 and Fig. S1). No ITS type belonging to clade 2 of T. borchii (Bonuso et al., 2009) was sequenced from the analysed samples, which was also not found as ascomata in the area during this survey.

The Adonis test showed no statistically significant differences in fungal community composition between the paired samples (P=0.057). The Adonis test also showed no differences in OTU composition between pine and oak in Tbo0 (P=0.477) and Tbo1 (P=0.7688) and between the two different sampling points in pine (P=0.2858) and oak (P=0.3116).

The Mantel test showed a low autocorrelation between the samples influencing the community structure (r=0.2417, P=0.027).

Comparison between the T. borchii productive and nonproductive soil cores

In the 15 soil Tbo0 samples and 14 Tbo1 soil samples, 8931 and 4203 mycorrhizal root tips were found, respectively. No statistically significant differences were found between root tips number in Tbo0 and Tbo1 by Friedman's test (P=0.096). Fifty taxa were found in Tbo0 and 39 in Tbo1. The fungal community at Tbo0 had a lower biodiversity as measured by the Shannon index and a lower evenness as indicated by Pielou's index (Table 1), and by the dominance–diversity curves (Fig. 3a). However, Friedman's test only showed significant differences in evenness revealed through Pielou's index (P=0.033) and no statistically significant differences in OTU richness and diversity between Tbo0 and Tbo1.


Comparison between the dominance–diversity curves of ectomycorrhizal fungal communities in relation to the sample collecting point (Tbo0, Tbo1) and (a) the Quercus (Oak) and (b) Pinus (Pine) host plants.

The Bray–Curtis (0.655) index indicated differences in the ectomycorrhizal OTU composition between Tbo0 and Tbo1 communities. The Adonis test also showed significant differences between the fungal community composition at Tbo0 and Tbo1. Nineteen taxa were present in both the Tbo0 and Tbo1 ectomycorrhizal fungal communities. Four of six taxa of Sebacinaceae and five of eight taxa of Inocybaceae spp. were found in both. Tuber borchii was present in both fungal communities, but significantly more abundant at Tbo0 (P=0.005).

Thirty-one and 20 taxa, respectively, were present only at Tbo0 (e.g. Cortinarius sp., Sebacinaceae sp. 1, Tomentella ferruninea, Rizopogon sp.) and at Tbo1 (e.g. T. dryophilum and Tomentella sublilacina). The ectomycorrhizal family composition in Tbo0 and Tbo1 was different (Fig. 2) with Basidiomycetes, in particular Sebacinaceae, Russulaceae and Inocybaceae, more abundant at Tbo1, whereas Tuberaceae and Thelephoraceae were more abundant at Tbo0. Cortinariaceae, Entolomataceae, Rhizopogonaceae and Suillaceae were represented only at Tbo0 and Atheliaceae and Tricholomataceae and Amanitaceae only at Tbo1.

Distribution of the ectomycorrhizal fungi in different host plants

A similar number of root tips were found on the oaks and pines and Shannon and Pielou indexes (Table 1), and dominance–diversity curves (Fig. 3b) showed that species richness, diversity and evenness were similar on both host species. However, differences were found in the OTU composition of the two communities using the Adonis test (P=0.023). Only 18 taxa (26%) were common on both oak and pine, which included T. dryophilum and T. borchi, with the comparison between the relative abundance of T. borchii on oak and pine nonsignificant (P=0.102). Some taxa such as Rhizopogon sp., S. granulatus, I. rufuloides and T. lapidum were found only on pine and Tomentella lillacinogrisea only on oak. The most abundant taxa on pine belonged to Thelephoraceae, whereas Sebacinaceae were common on oak, with Sebacina sp. being the most abundant. Five families of Agaricales (Cortinariaceae, Inocybaceae, Entolomataceae, Tricholomataceae and Amanitaceae) were represented on oak and only two (Cortinariaceae, Inocybaceae) on pine (Fig. 2).


Direct gradient ordination of ectomycorrhizal communities using CAP explained 11.8% of the total variability, with a P=0.008. The position of soil core collection (Tbo0 and Tbo1) explains 6.53% (P=0.030), whereas the host plant explains 5.27% of the total variability (P=0.025). The different position in the ordination plot of the samples collected at Tbo0 and at Tbo1 shows differences in OTU composition of the communities (Fig. 4a). In Fig. 4b, T. borchii is located in the upper part of the ordination plot and characterizes the fungal community at Tbo0. In contrast, the fungal community at Tbo1 is characterized by fungal taxa such as T. dryophilum and T. sublilacina located in the lower part of the ordination plot.


Ordination of samples (a) (○, Tbo0, △, Tbo1) and species (b) obtained by CAP in relation to the collection point and host plant.


This is the first report where ectomycorrhizal fungal communities in natural T. borchii truffières have been surveyed and quantified using both morphological and molecular techniques and adds to recent work by Murat (2005) and Bertini (2006) on T. magnatum. Using both molecular and morphological techniques, rather than just the latter (Donnini & Bencivenga, 1995; García-Falces & De Miguel Velasco, 2008), it was possible to assess the potential importance of all ectomycorrhizal species adjacent to fruiting bodies.

Despite their similar morphologies, T. borchii behaved very differently from T. magnatum. In the latter, <2% of the root tips supported fruiting (Murat et al., 2005; Bertini et al., 2006; Donnini, 2006), while T. borchii dominated the ectomycorrhizal population – a feature also observed by Hall (2008). In doing so, it was more akin to T. melanosporum (Baciarelli Falini & Granetti, 1998; Napoli et al., 2010) and T. matsutake (Lian et al., 2006), which also dominate where they are found to be fruiting. This different behaviour among T. borchii and T. magnatum is perhaps not surprising because the two species are phylogenetically not close (Jeandroz et al., 2008), but even so, this disparate behaviour needs to be elucidated.

Tuber borchii ectomycorrhizal community was dominated by just a few common species and large numbers of rare species – a feature common to other ectomycorrhizal communities (Taylor, 2002). The lack of other codominant ectomycorrhizal fungal species, which are usually patchy at a scale of <3 m (Lilleskov et al., 2004), might explain why the ectomycorrhizal community showed a low, nonsignificant, spatial autocorrelation (as shown by the Mantel test) either among samples collected at different fruiting sites and between the samples collected at Tbo0 and Tbo1. The higher incidence of T. borchii mycorrhizas in Tbo0 demonstrates a reduction in evenness due to decreased number of individual of other species in comparison with Tbo1. In contrast to T. melanosporum brùlès (the burnt area around the productive tree characterized by scanty vegetation) (Napoli et al., 2010), T. borchii did not reduce the ectomycorrhizal richness, i.e., there was no inhibitory effect on ectomycorrhizal biodiversity.

Thelephoraceae are generally the most common ectomycorrhizal fungal taxa in ectomycorrhizal communities with Inocybaceae, Sebacinaceae and Pezizales, in particular, Pyro-nemataceae, also well represented (Kõljalg et al., 2000; Horton & Bruns, 2001). In contrast, Russulaceae, which are very common in many ectomycorrhizal fungal communities (Horton & Bruns, 2001), were rare. Similarly, Cenococcum geophylum was also absent, even though it is the most widespread ectomycorrhizal fungal species (Lobuglio et al., 1996). Surprisingly, these species are also missing from T. magnatum truffières (Murat et al., 2005). The possibilities that might account for this are the high pH and high levels of plant available calcium present in both T. borchii and T. magnatum soils, and most other soils where truffles are found (Hall et al., 2007) and that are known to influence ectomycorrhizal communities (Bakker et al., 2000).

Surprisingly, only five species of the dominant family of Telephoraceae were present in the fruiting (Tbo0) and nonfruiting points (Tbo1). In contrast, the other species of this family are only present in either Tbo0 or Tbo1. It is well known that other soil microorganisms, including other ectomycorrhizal fungi, influence ectomycorrhizal communities through competition (Jumpponen & Egerton-Warburton, 2005), while Hall (2003) hypothesized that the presence of certainly ectomycorrhizal fungal species may positively interact with other ectomycorrhizal fungi. It might be that within the Telephoraceae, there are species that vary in their behaviour when associated with T. borchii, with some having no impact on it while others might stimulate or inhibit its development.

The Sebacinaceae and Inocybaceae were found in both fruiting (Tbo0) and nonfruiting points (Tbo1), but their distribution was strongly influenced by the host plant. Sebacina spp. were predominantly associated with oak at our study site, a feature that has been reported in the past with oak and other Fagaceae (Glen et al., 2002; Richard et al., 2004a, b; Ishida et al., 2006). Interestingly, Sebacinaceae spp. are also common in T. magnatum truffières (Murat et al., 2005; Iotti & Zambonelli, 2006). In contrast, research by Iotti (2005) showed that I. rufuloides was only associated with Pinus, which is confirmed by the present study, and that it is a potential competitor for T. borchii. However, the present study now indicates that I. rufuloides does not compete with T. borchii, but instead its presence in sandy soils along the Mediterranean coasts simply indicates that the ecological conditions suit T. borchii (Kuyper, 1986; Zambonelli et al., 2002).

Competition between Tuber spp. and other ectomycorrhizal fungi is a major problem in truffle cultivation particularly during the first few years after planting when the introduced Tuber sp. can be replaced by native ectomycorrhizal fungi (Hall et al., 2007). For example, T. aestivum and Tuber brumale Vittad. often replace T. melanosporum in French, Italian and Spanish truffières (Donnini et al., 2001; Sourzat et al., 2001; García-Montero et al., 2008). Soil conditions and particularly active carbonate seem to be a major factor in the fruiting and aggressiveness of T. melanosporum over the other less precious Tuber species (García-Montero et al., 2008). Similarly, in a closely planted T. borchii infected P. pinea experiment (Zambonelli et al., 2000; Iotti et al., 2006), T. borchii initially dominated the ectomycorrhizal community, but colonization of the root system and fruiting declined and tended to move in the periphery of the plantation as the canopy closed and a litter layer developed. This, together with its ability to infect plant by spores and its early fruit-body production (Zambonelli et al., 2000), led to the conclusion that T. borchii was an early-stage fungus (Mason et al., 1983; Deacon & Fleming, 1992).

In the soil samples where the roots of both host plants were in equal proportions, we did not detect differences between the ectomycorrhizal communities in pine and oak. Only in the samples where pine dominated did pine-specific fungi such as Rhizopogon sp. and S. granulatus appear (Bruns et al., 2002). Despite this, T. borchii was present in old pinewoods. A likely explanation is that T. borchii, being an early-stage ectomycorrhizal fungus (Deacon & Fleming, 1992; Zambonelli et al., 2000), is colonizing young oaks establishing in forest gaps formed when the older pines die during the succession to thermophilic Q. ilex forest (Barkman et al., 1986; Corticelli et al., 2004). It might be suggested that T. borchii is not an early-stage fungus because declines in production have not been detected in commercial bianchetto truffières in Italy or New Zealand. However, in truffières, low planting densities are used and contaminating ectomycorrhizal fungi are unlikely to be present, and so ‘normal’ successions, at least in the short term, will not occur (Bencivenga & Urbani, 1996; Ceccucci, 2005; Hall, 2008).

Tuber dryophilum is a common, morphologically similar, poorly flavoured, degrading contaminant in batches of T. borchii (Zambonelli et al., 2002; Gioacchini et al., 2005; Hall et al., 2007). The presence of its fruit bodies and mycorrhizas adjacent to, but not mixed with, T. borchii and the differences in the micro niches that the two species occupy are therefore of considerable interest to those interested in competition between ectomycorrhizal fungi as well as those involved in the commercialization of T. borchii (Zambonelli et al., 2002). Fortunately, the anatomical differences in mantel structure, with typical roundish-epidermoid cells in T. borchii and the coarse net of irregularly shaped and thick hyphae in T. dryophilum, can be used to exclude T. dryophilum during the quality control of T. borchii-infected plants. This is in spite of some variability in mantle anatomical characteristics of T. borchii genotypes (Giomaro et al., 2000). However, the interaction between T. borchii and T. dryophilum in the field, including ecological studies on the micro environmental variables that favour one species or the other, warrants further investigation.

Supporting Information

Additional Supporting Information may be found in the online version of this article:

Table S1. Relative abundance (Abb) and absolute incidence (Freq) of ECM fungal species.

Table S2. Description of the most important morphological and anatomical characteristics of 70 ECM morphotype encountered in T. borchii truffiéres.

Fig S1. Mantle of Tuber borchii (a) and Tuber dryophilum (b) mycorrhizas. Bar = 10 μm.


The work was supported by Amministrazione Provinciale di Ferrara and the Italian ‘Ministero dell'Ambiente e della Tutela del Territorio e del Mare’ as a part of the target Project on: ‘Tuber: studio del ruolo e funzione del tartufo come nuovo modello per la salvaguardia dell'ambiente e della biodiversità’. The authors would like to thank the ‘Corpo Forestale’ of Comacchio for providing very useful information and assistance in the field.


  • Editor: Philippe Lemanceau


View Abstract