Natural production of Tuber aestivum in central Spain: Pinus spp. versus Quercus spp. brûles

Aim of study: Tuber aestivum is the most widespread edible truffle, with increasing commercial interest. This species can produce carpophores with conifer hosts, in contrast with the inability of Pinus spp. to induce fruiting in other truffle species such as Tuber melanosporum. Therefore the objective is to compare the characteristics and carpophore production of T. aestivum brûlés associated with Pinus spp. versus Quercus spp. Area of study: We studied the natural habitats of T. aestivum in the Alto Tajo Nature Reserve in central Spain. Material and methods: During 5 years, we monitored the production of carpophores and brûlé size of 145 T. aestivum brûlés associated with Pinus nigra subsp. salzmanni and P. sylvestris and Quercus ilex subsp. ballota and Q. faginea hosts. Statistical treatment was performed using the Statistica Program v. 6. Main results: The size of brûlés associated with Pinus was significantly smaller than that of brûlés associated with Quercus. However, carpophore production per brûlé, and especially for brûlés of similar size, was greater when the host plant was a pine. After accounting for brûlé size, the production of brûlés associated with Pinus spp. was 2.23 (95% CI, between 1.35 and 3.69) and 1.61 (95% CI, between 1.02 and 2.54) times greater than the production of brûlés associated with Quercus faginea and Q. ilex subsp. ballota, respectively. Research highlights: The considerable ability of Pinus nigra subsp. salzmanni and P. sylvestris to form effective brûlés and to produce carpophores of Tuber aestivum in natural conditions was clearly demonstrated, and suggest that those species can be of use in the culture of T. aestivum.


Introduction
Tuber aestivum Vittad. (summer truffle) is the most cosmopolitan edible truffle. Tuber eastivum has a widespread distribution from Sweden to Morocco, from UK to Ukraine in Europe, but it can also be found as far as China. However, despite increasing commercial interest, the wide ecological and geographical amplitude of T. aestivum has hindered the ecological knowledge of this valuable species (Montechi and Sarasini, 2000;Arredondo-Ruiz et al., 2014).
The production points of T. aestivum, known as burns or brûlés, are characterized by clearings in the vegetation, a result of the phytotoxic capacity of the Tuber mycelia (Plattner and Hall, 1995;Hall et al., 2003;Pomarico et al., 2007). Tuber aestivum dominates the ectomycorrhizal populations within their brûlés, excluding other mycorrhizal species. There also may be a "host effect", where plants can allocate resources selectively to preferred and ideal ectomycorrhizal associates. All host species present varying morphological characteristics that condition mycorrhization under particular environment (Benucci et al., 2011).
The association between European pines and highly prized truffles is also well-known. However, there is little information estimating the productivity of edible black truffles in European pine forests. In Central Spain, García-Montero et al. (2007) monitored the production of 433 Tuber melanosporum Vittad. (black truffle) brûlés for 7 years, in mixed forests of pines and oaks. They concluded that Pinus nigra Arnold subsp. salzmannii (Dunal) Franco and P. sylvestris L. hinder the production of Tuber melanosporum carpophores and, therefore, pine species are not recommended for black truffle culture.
In those mixed forests of Central Spain, there was also a significant presence of Tuber aestivum brûlés, competing against T. melanosporum brûlés. García-Montero et al. (2008) compared the soil properties of brûlés from those two truffle species, both associated with oaks [Quercus faginea Lam. and Q. ilex L. subsp. ballota (Desf.) Samp.]. They concluded that active carbonate was a major factor in the fruiting and aggressiveness of Tuber melanosporum versus T. aestivum. However, no evidence of oak host species effect was revealed (single species stands of Quercus faginea, Q. ilex subsp. ballota, and mixed-species stands) in the production of carpophores of either Tuber melanosporum or T. aestivum .
Despite current interest in the harvest and culture of Tuber aestivum, there is little information about its ecology and, in particular, the effect of host species in the production of valuable carpophores. Thus, the objective of this study is to compare the characteristics and carpophore production of T. aestivum brûlés associated with Pinus spp. versus Quercus spp. in natural forests of central Spain. The results of this study could provide significant information to evaluate the suitability of Pinus nigra subsp. salzmannii and P. sylvestris for Tuber aestivum cultivation.
The study area is dominated by Jurassic and Cretaceous limestone and dolomites. Soils are lithic and rendzic leptosols with marked surface stoniness and high peaks. Elevation ranged from 1,000 to 1,450 m.a.s.l. The dominant aspect is southern. Mean annual temperature is 10.8 , and mean monthly minimum and maximum temperature is 3.3 and 18.5 , respectively. Mean annual precipitation is 650 mm (Table 1).

Data collection and analysis
Tuber aestivum has been harvested regularly in the area since the early 2000s. Carpophores are harvested with the help of trained dogs. This study was based on the collaboration with 4 truffle collectors who shared their knowledge and collection data, including the location of several harvesting points. During many field trips undertaken between 2007 and 2012 with these collectors, 145 of the collectors' preferred brûlés were selected, based on their regular production.
In each brûlé, we determined the host plant, size (m 2 ) of the brûlé and carpophore production. The maximum annual production of Tuber aestivum carpophores was calculated as the greatest quantity of fresh carpophores, in grams, picked during a harvest season, over a minimum period of 5 years. The maximum annual production (instead of the average yield) was chosen to standardize and compare the optimum yields of brûlés (in the most favorable year for each site) located in different areas (García-Montero et al., 2007). Harvest data were supplied by the 4 collectors who harvested those 145 brûlés regularly. The reliability of the information provided by the collectors was confirmed independently in systematic field trips made with the collectors at the time of harvesting throughout the period between 2007 and 2012. In 2012, at the end of monitoring period, we measured the longest and shortest axis and estimated brûle area using an ellipse model. In brûles with an irregular shape, we used a combination of several small ellipses. The exact geographic coordinates of these brûlés were not provided, to maintain the confidentiality of the T. aestivum locations.
We compared the size and Tuber aestivum carpophore production of the brûlés associated with different host plants species using analysis of variance. We used analysis of covariance to compare truffle production among hosts while controlling for brûlé size. The response variables were log transformed. Therefore, following Ramsey and Schaffer (1997, p: 207), we interpreted the results as multiplicative effects on the group medians. The family-wise error rate was controlled with Tukey-Kramer's procedure. All analyses were done in R v. 3.0.2 (R Core Team 2013).

Results
The average area of the 145 brûlés was 40.01 m 2 (SE, 3.66 m 2 ), but there was a wide range of sizes, from 1 to 314 m 2 . The highest yielding brûlé produced 7.00 kg of carpophores in one season, but the mean maximum production for all 145 brûlés was 0.976 kg per brûlé and season (SE, 0.095 kg yr -1 ). Table 2 summarizes the natural production and brûlé size data depending on the host plant of Tuber aestivum.

Is there a difference in truffle production among hosts?
There is convincing evidence that the maximum production of Tuber aestivum carpophore per brûlé  There is no significant difference in production between the two oak hosts, or between Q. ilex subsp. ballota and Pinus spp. (Fig. 1a).

Is there a difference in the size of the brûlé among hosts?
There is convincing evidence that the area of Tuber aestivum brûlés differs among hosts (F 2,144 = 14.13; p < 0.0001). The estimated median brûlé area for Quercus faginea and Q. ilex subsp. ballota hosts is 178.8% (95% CI from 73.4 to 348.1%) and 142.1% (95% CI from 55.2 to 277.6%) greater than the area of Pinus spp. host, respectively. There is no significant difference in brûlé size between the two oak hosts (Fig. 1b).

Is there a difference in production among hosts, after accounting for the size of the brûlé?
Because the size of the brûlé was different among hosts, and the production was greater but the area much smaller when the host was a pine versus an oak, we compare truffle production after accounting for brûlé size. The effect of both host and brûlé size on truffle production was significant, so truffle production depends on host (p = 0.001) and increases with the area of the brûlé (p = 0.001). There was no evidence that the relationship between production and brûlé size was different among hosts (p-value for the interaction, 0.13). Holding brûlé size constant, the production of Pinus spp. is estimated to be 123.3% (95% CI from 35.1 to 269.0%) and 60.6% (95% CI from 1.7 to 153.6%) greater than the production in Quercus faginea and Q. ilex subsp. ballota, respectively. There is suggestive but inconclusive evidence of a difference in production between the two oak species (p = 0.077) (Fig. 2).

Discussion
Mediterranean pine forests are rich in mushrooms and truffles. In many regions, harvesting edible mycorrhizal species generate greater economic benefits than any other forest product (Oria, 1989(Oria, , 1991. Even though species from the genus Tuber are one of the most valuable and sought after, and pine forests one of the most common forest types, there is very little information about truffle production in association with Pinus spp. (García-Montero et al., 2007).
It is remarkable the significantly smaller area of the brûles associated with Pinus spp. versus those associated with Quercus spp., despite the greater production of the former (Table 2; Fig. 1). As a result, the truffle production of brûles of the same size was much greater in pines (2.23 and 1.61 times greater) that than the production in oaks. García-Montero et al. (2009 hypothesized that the ecological significance of the brûlé is to modify the soil environment, increasing active carbonate and exchangeable calcium and/or soil pH, to stress the host plant with nutritional deficiencies by lime induced chlorosis. The plant would respond to the stress by increasing the number of root tips and, therefore, favoring Tuber spp. mycorrhization. This hypothesis and our results suggest that Pinus nigra subsp. salzmannii and P. sylvestris could be more sensitive to the stress caused by the Tuber aestivum brûlé than Quercus spp. Thus, to accomplish the same impact on the host plant, the fungus would have to invest fewer resources, leading to smaller brûlés and greater truffle production per unit of brûlé area.
Our results indicate that Pinus nigra subsp. salzmannii and P. sylvestris may have an important role in Tuber aestivum cultivation because, at least in natural habitats and under certain ecological conditions, both species produced high yields of T. aestivum carpophores. However, new studies are necessary to evaluate whether Pinus nigra subsp. salzmannii and P. sylvestris differ in their production carpophores and/or their ability to form brûles of Tuber aestivum, because, in the Iberian Peninsula, Pinus sylvestris seems indifferent to soil type while Pinus nigra subsp. salzmannii favors marl-limestone soils (Amaral Franco, 1986), as does Tuber aestivum. This fact suggests that P. nigra subsp. salzmannii may better tolerate increases in soil pH and/or exchangeable calcium that occurs inside Tuber brûlés, as proposed by García-Montero et al. (2009. In summary, we found that Pinus nigra subsp. salzmannii and P. sylvestris have considerable ability and facility of to form effective brûlés and to produce carpophores of Tuber aestivum. Although both Pinus are very little-known species in truffle culture, the results support their use under the appropriate climatic and soil conditions required for their growth and the development of T. aestivum. Moreover, we can hypothesize that other Pinus spp. could have a highly relevant role in the cultivation of several Tuber spp. In this regard, it has been observed that Pinus nigra subsp. salzmannii produces carpophores of Tuber aestivum (= uncinatum), T. mesentericum Vittad., T. borchii Vittad. and, exceptionally, T. magnatum Pico; Pinus pinea L. produces Tuber borchii, T. aestivum, T. magnatum and T. indicum; Pinus halepensis Mill. produces Tuber borchii; Pinus tabulaeformis Carr. var. yunnanensis (Franch.) Shaw produces Tuber indicum, T. pseudoexcavatum and T. sinense Tao and Liu; and Pinus armandii Franch. produces Tuber indicum and T. sinense (Pacioni 1987;Chevalier and Frochot, 1997;Granetti et al., 2005;Montecchi and Sarasini, 2000;Zambonelli et al. 2000;Riousset et al., 2001;García-Montero et al., 2010). However, little is known about the actual capacity of those Pinus species for producing significant amounts of truffles with commercial interest.