Ectomycorrhizal communities above and below ground and truffle productivity in a Tuber aestivum orchard

Aim of study : The diversity of ectomycorrhizal fungal communities (EM) above (EMFb) and below (EMMt) ground associated with Quercus cerris L., Q. pubescens Willd., and Pinus nigra J.F.Arnold was analyzed. Area of study : A 20 year-old orchard that produces Tuber aestivum truffles, located a few kilometers from Chiusi della Verna (latitude 43° 41' 53"; longitude 11° 56' 9") in Tuscany (central Italy) was observed. Material and methods : This investigation combined analyses of EMFb, EMMt, T. aestivum productivity, different host trees, and statistical data on community ecology. Main results : The EM communities showed high species richness and differed slightly in relation to both the host tree and their location above or below ground, providing frequent findings of Tricholoma and Tomentella , respectively. Positive correlations were found between the number of truffles and host trees, and between the weight and number of truffles and EMFb. Research competition.


330
E. Salerni et al. / Forest Systems (2014) 23 (2): [329][330][331][332][333][334][335][336][337][338] composition is also, as yet, not fully understood, although several studies have demonstrated the existence of interspecif ic competition in ectomycorrhizal community development (Zambonelli et al., 2000;Lilleskov and Bruns, 2003;Koide et al., 2004;Kennedy and Bruns, 2005;Baciarelli-Falini et al., 2006;Kennedy et al., 2007). In this study we analyzed the diversity of ectomycorrhizal fungal communities above and below ground associated with Quercus cerris L., Q. pubescens Willd., and Pinus nigra J.F.Arnold. We aimed to achieve the following: (a) characterize the ectomycorrhizal communities through observations of the fruiting bodies (EMFb) and morphotypes on root tips (EMMt); (b) verify and quantify the presence and distribution of T. aestivum truffles; (c) analyze whether the fungal communities differed in relation to the host trees; (d) compare the fungal communities in relation to truffle production, in both number and weight; (e) identify any interactions between the communities above and below ground.

Description of the truffle orchard
The study was conducted in a 20 year-old orchard that produces Tuber aestivum truffles, located a few kilometers from Chiusi della Verna (latitude 43°41' 53''; longitude 11°56' 9'') in Tuscany (central Italy). The site has an area of 0.7 ha, an average altitude of 1000 m a.s.l., and a slight slope (5-7%). The climate is fresh to moderate, with a mean annual rainfall of 900-1000 mm and mean annual temperature of 9.5°C. The soil is constituted of limestones and marly limestones belonging to the Alberese formation. Rock outcrops are absent, while stones of small dimensions (2-75 mm) formed by unaltered calcareous elements are frequent on the surface; the pH ranges from 7.4 to 8.3 .
The truffle orchard was established in 1989 by planting 58 Quercus cerris, 85 Q. pubescens and 85 Pinus nigra, all originally inoculated with Tuber aestivum. Seedlings were placed with a distance of 5 m between the rows and 4 m within the rows. In 2006 various silvicultural treatments (mulching, adding organic matter and granular gravel) were performed with the aim of improving the habitat for T. aestivum fruiting.

Data collection and morphological analysis
In line with studies conducted in the Netherlands (for the methodology see Arnolds 1981; for the trophic groups see Arnolds et al., 1995), epigeous and hypogeous ectomycorrhizal fruiting bodies larger than 1mm (EMFb) were observed from September 2008 to December 2011. The abundance and frequency was considered in the whole orchard and in relation to the 3 types of host tree. Identification was performed with the usual morphological techniques and employing general analytic keys and monographs .
The exsiccata are conserved at the Herbarium Universitatis Senensis (Siena). The nomenclature of fungal species refers to the CABI Bioscience Database of Fungal Names available on the internet: www.indexfungorum.org/Names/Names.asp, which was updated at the end of August 2012.
Observations on the hypogeous fruiting bodies, were performed in the same period with the help of truffle hunters and trained dogs, to monitor the presence of T. aestivum by quantifying its numbers and weight.
To analyze the EM morphotypes present on root tips (EMMt), 84 soil cores of 30x6 cm were collected between September 2008 and December 2011. After removing the litter and organic horizon, cores were taken from the exact points in which EMFb were found. In order to have soil samples for every fungal species and every host tree, samples were taken for the same fungal species when fructifying under different trees. Each sample was individually soaked overnight in tap water and sieved to separate the root fragments and EMMt from the soil; they were then stored at 4°C and processed within the following 10 days. Morphotyping was performed using a stereomicroscope and a light microscope, with reference to the anatomo-morphological characteristics described by Agerer (1987Agerer ( -20082006) and the on-line EctoMycorrhizal Community DataBase (www.emyco.uniss.it).
Spearman's rank correlation coefficient was calculated to determine correlations between the number and weight of the T. aestivum fruiting bodies and host trees and the qualitative (number of species) and quantitative (number of fruiting bodies and of morphotypes) data for EMFb and EMMt. The P-value was set at the 5% signif icance level. Statistical analyses were performed using the STATISTICA 5.0 (StatSoft. Inc., Tulsa, Ok, USA) package of programmes.

Ectomycorrhizal fruiting body (EMFb) communities
In the truffle orchard studied, 48 fungal species were identified (Table 1), of which 44 were epigeous species and 4 hypogeous: 15 genera belonged to the Basidiomycota and only 1 to the Ascomycota. The genus Inocybe had the highest number of species (14), followed by Cortinarius with 11, Hygrophorus, Lactarius and Tricholoma with 3 species each, Boletus, Hebeloma and Tuber with 2 species, and others with a single species. Regarding the average abundance in the whole area studied, Tricholoma terreum produced the highest number of fruiting bodies, with an average of 4.114, followed by Laccaria laccata (3.052), while Tuber aestivum, found 49 times, was the most frequent. These fungal species were also the most abundant when the three different host trees were observed separately: the most abundant species in the area with Quercus cerris was Laccaria laccata (11.948), in the Pinus nigra area Tricholoma terreum (6.882) and, among the few species found in the Q. pubescens area, Tuber aestivum had the highest productivity and frequency (Table 1). In the Q. cerris area the most frequent species, found exclusively here, was Hebeloma sinapizans, which was observed in association with 41 of the 58 total trees. The area with 85 Pinus trees was strongly preferred by Chroogomphus rutilus, with 31 findings. Here one of the most frequent and abundant species was Suillus granulatus, known to always be associated with pines. Surprisingly, a few fruiting bodies were also collected in the neighboring oak areas in this study.
The three indices (Table 2) used in this study provided an important tool to summarize the overall diversity of the fruiting body communities. They were characterized by a low diversity and a good evenness of the samples, although these indices varied significantly among the different host trees. In particular, the area planted with Q. pubescens showed the lowest values for species richness and Shannon's diversity index (11 and 0.048 respectively).
Rank-abundance curves revealed that the Quercus pubescens community showed a different trend in relation to the other two communities, and that only a few species of EMFb dominated in relation to each host species (Fig. 1).
With regard to their identity, 69 were Basidiomycota and only 5 Ascomycota. The most common family was Cortinariaceae with 14 different ectomycorrhizal morphotypes, 6 of which were identified at genus level (Cortinarius) ( Table 3), followed by Inocybaceae with 12 different EMMt (all of the genus Inocybe) and Thelephoraceae with 11 (7 of which belonged to the genus Tomentella). The genus Inocybe, followed by Tomentella, also had the highest number of different EMMt. Regarding relative abundance, Tricholoma sp. 2 had the largest number of colonized root tips (6.429), followed by Hebeloma sp. 2 (3.060), Inocybe sp. 8 (3.048), Cenococcum geophilum (2.631) and Tuber aestivum (2.512). The last two species were also among the most frequent. In the area with Q. cerris, Tricholoma sp. 2 was again the most abundant (13.769), whereas in the Pinus nigra and Q. pubescens areas the most abundant species were Hebeloma sp. 2, with 4.571 colonized root tips, and Tomentella sp. 5, with 5.550, respectively. The most frequent EMMt in the orchard studied as a whole and in the Pinus area was Cenococcum geophilum (14), while under Q. cerris it was T. aestivum. The third area showed only single findings. T. aestivum was found in only 9 out of the total of 84 soil cores, with a considerable variation in relative abundance between individual samples (from 0.07 to 0.64).
Among the 32 EMMt exclusive to Quercus cerris the most abundant were Inocybe sp. 8 and Tricholoma sp.; The values of the Shannon and Pielou indices (0.214 and 0.629, respectively) indicated that the EMMt communities were also characterized by a low diversity Ectomycorrhizal communities and truffle productivity 333   and an even distribution of the samples ( Table 2). As described above for the EMFb, the fungal community below Quercus pubescens showed lower values for richness and Shannon's diversity index (17 and 0.047 respectively).

Tuber aestivum
In total 232 truffles were collected, weighing 5485 grams. The largest number of truffles was counted in the area with Pinus nigra (154), while in each of the 2 oak areas less than a quarter of this number and weight was collected ( Table 2).
The Kruskal-Wallis test (α = 0.05) was performed on the host trees to compare the number of truffles and total weight of Tuber aestivum with the species richness and abundance of EMFb and EMMt communities. Signif icant differences were found in all cases (p < 0.001).
Spearman's rank correlation coeff icient showed significant correlations (p < 0.05) between the number and weight of T. aestivum fruiting bodies and the number of EMFb species and their abundance, but not the number of species or abundance of the EMMt community. The host trees showed a statistically significant correlation (p < 0.05) with the number of truffles, but not their weight (Table 4).

Discussion
This is the first study to date in which fungal communities above and below ground, analyzed through observations of fruiting bodies (EMFb) and morphotypes (EMMt), have been compared with the production of T. aestivum, expressed in both the number and weight of fruiting bodies. Ectomycorrhizal communities on root tips in natural and cultivated truffières have been amply investigated (Donnini and Bencivenga, 1995;Donnini et al., 1999;Murat et al., 2005;Baciarelli Falini et al., 2006;Pruett et al., 2008;Águeda et al., 2010;González-Armada et al., 2010;Iotti et al., 2010;Benucci et al., 2011;Garcia-Barreda and Reyna, 2012;Leonardi et al., 2013). However, according to Tóth and Barta (2010) very few studies have simul-Ectomycorrhizal communities and truffle productivity 335 taneously analyzed fungal species both above and below ground in relation to varying environmental factors or with the aim of determining their suitability as indicators of environmental change. In particular, there is limited information available on the ecology of ectomycorrhizae that combines both communities in truffle environments (Donnini et al., 2008;Salerni et al., 2011). Our result confirmed that fungal communities differed significantly in relation to the host trees. Quercus pubescens showed lower levels of richness than Q. cerris and Pinus nigra for both fruiting bodies (EMFb) and root tips (EMMt). The Shannon index values obtained in this study were rather low, which may be due to the non-natural character of the truffières. In fact, this index is much higher in natural environments: Buée et al. (2011) reported a Shannon index of 4.84 for epigeous saprotrophic and ectomycorrhizal fungi studied in a temperate deciduous forest in France, while Oria- de-Rueda et al. (2010) found an index of 1.2-1.3 for fungal communities observed in natural Quercus faginea and Q. pyrenaica forests. The Shannon index for the EMMt communities was closer to agreement with, but still lower than, other studies (Pruett et al., 2008;Benucci et al., 2011;Garcia-Barreda and Reyna, 2012). In contrast Leonardi et al. (2013) found relatively high values for the ectomycorrhizal communities present in four natural Tuber magnatum truffle grounds. More host plants are generally present in natural truffle grounds. This was confirmed by Belfiori et al. (2012), who studied fungal communities in natural and cultivated T. melano-sporum sites, also providing evidence that fungal communities can diverge under identical environmental conditions.
The genera Inocybe, Cortinarius, Hygrophorus, Lactarius and Tricholoma are fairly common ectomycorrhizal fruiting bodies in various plant communities (Oria- de-Rueda et al., 2010;Buée et al., 2011;Hernández-Rodriguez et al., 2013) and in environments in which truffles are produced (Donnini et al., 2008;Salerni et al., 2011). Similarly, Cortinariaceae and Thelephoraceae are fairly common morphotypes on root tips in various communities (Kõljalg et al., 2000;Glen et al., 2002;Selosse et al., 2002;Urban et al., 2003) and in environments characterized by truffle production (Murat et al., 2005;Baciarelli Falini et al., 2006;Pruett et al., 2008;Iotti et al., 2010;Benucci et al., 2011;Belfiori et al., 2012;Leonardi et al., 2013). Our data confirmed that the genus Tomentella is widespread and an important component in orchards with Tuber spp. (Murat et al., 2005;Pruett et al., 2008;Águeda et al., 2010;Iotti et al., 2010;Benucci et al., 2011;Belfiori et al., 2012;Leonardi et al., 2013). In agreement with Donnini et al. (2008), who reported the frequent fruiting of Tricholoma species in truff ières, Tricholoma terreum was also found to be more abundant than Tuber aestivum in the present study. Note that Tricholoma sp., together with Inocybe sp. 8, were the most abundant EMMt growing exclusively on the roots of Quercus cerris. A separate mention is merited by Suillus species, which is always associated with pine trees. In the present study S. granulatus fruiting bodies and morphotypes were also observed under the nearby oak plantation. This could be related to the root system of P. nigra, which often extends well beyond the projection of its foliage.
Positive correlations were found between the number of T. aestivum fruiting bodies and the host species. This is in agreement with Garcìa-Montero et al. (2007), who found a significant variability in T. melanosporum truffle production in f ive types of woods in Spain. Nevertheless, Benucci et al. (2011) suggested that T. aestivum production was not affected by differences between the ectomycorrhizal communities associated with hazel and hornbeam in a 24-yearold orchard. This was also conf irmed by our data, which showed no correlation between the production of truffles and the EMMt community below ground. Finally, positive correlations were identified between the weight and number of T. aestivum fruiting bodies and qualitative and quantitative data for EMFb above 336 E. Salerni et al. / Forest Systems (2014) 23 (2): 329-338 ground. This result confirms the hypothesis supported by Donnini et al. (2008) that mycorrhizal fungi and truffle production are not in competition. It is known that ectomycorrhizal fungi provide mineral nutrition to dominant trees, but differ in their enzymatic activities (Courty et al., 2005). Moreover, they deliver species-specif ic benef its to their host plants (van der Heijden and Kuyper, 2003), which render their biodiversity of high importance to plant nutrition (Tedersoo et al., 2006). Ectomycorrhizal fungi are highly diverse in most ecosystems, even though the mechanisms behind this diversity remain unclear. Their different preference for soil conditions and host plants seems to play a key role (Bruns, 1995). Koide et al. (2004), on the other hand, argue that signif icant associations between species occur when resources are not a limiting factor. In this case competition would not be prejudicial.