Response of ectomycorrhizal communities to past Roman occupation in an oak forest

Soil Biology & Biochemistry 41 (2009) 2206–2213

Contents lists available at ScienceDirect

Soil Biology & Biochemistry journal homepage: www.elsevier.com/locate/soilbio

Response of ectomycorrhizal communities to past Roman occupation in an oak forest Abdala G. Diedhiou a, *, Jean-Luc Dupouey b, Marc Bue´e a, Etienne Dambrine c, Laure Lau¨t d, Jean Garbaye a a

UMR INRA-UHP 1136 Interactions Arbres-Microorganismes, Centre INRA de Nancy, 54280 Champenoux, France UMR INRA-UHP Ecologie et Ecophysiologie forestie`res, Centre INRA de Nancy, 54280 Champenoux, France c Unite´ 1138 Bioge´ochimie des Ecosyste`mes Forestiers, Centre INRA de Nancy, 54280 Champenoux, France d Universite´ de Paris 1-Panthe´on-Sorbonne, UFR 03, Institut d’art et d’arche´ologie, 3, rue Michelet, 75006 Paris, France b

a r t i c l e i n f o

a b s t r a c t

Article history: Received 25 March 2009 Received in revised form 23 June 2009 Accepted 3 August 2009 Available online 8 August 2009

The impact of past Roman occupation on the composition of ectomycorrhizal (ECM) communities was analysed in 12 Roman settlements in an oak forest in Central France. At each Roman settlement, soils and ECM roots were sampled from two plots (600 m2 each), one plot close to the remains of the buildings (<100 m), supposed to be impacted by ancient Roman agriculture, and the second plot 250–500 m away from the remains of the buildings, supposed to be less intensively influenced by previous cultivation. Soils were analysed and ECM fungal taxa were identified by morphotyping and sequencing the rDNA ITS region. The soil properties were significantly affected by the past Roman occupation, in terms of nutrient availability, especially for P, N and Mg. The enhancement of soil nutrient levels by past Roman land-use had significantly modified alpha diversity and species composition of ECM communities. Among the 67 determined ECM morphotypes, 40 were shared by the occupied and non-occupied plots, 17 were found only in the occupied plots and 10 only in the non-occupied plots. Six morphotypes were significantly more frequent near the antique remnants. Our study showed, for the first time, that ectomycorrhizal communities are impacted by previous Roman land-use, even after nearly two thousand years of forest state. Crown Copyright Ó 2009 Published by Elsevier Ltd. All rights reserved.

Keywords: Community composition Ectomycorrhiza Land-use history Morphotype Oak forest Roman period

1. Introduction The strong impact of ancient agricultural land-use, dating from the last few centuries, on the plant species composition, ecological cycles and productivity of present day forests has been repeatedly evidenced in Europe and North America (Peterken, 1996; Hermy et al., 1999; Foster, 2002). Indeed, the complex modifications of structural and chemical properties of soils induced by past farming and manure inputs have impacted the forests which have replaced croplands (Koerner et al., 1997, 1999; Compton et al., 1998; Jussy et al., 2002). More recently, the impact of much older occupations, dating from the Roman period, on the present soil properties and biodiversity of forests has been revealed (Dupouey et al., 2002; Dambrine et al., 2007). Traces of Roman occupations have been discovered in exceptionally large areas of present forests in France (Peltre and Bruant, 1991; Dambrine et al., 2007). In the Tronçais forest of Central France, more than 100 Roman settlements have been found (Bertrand, 1996; Lau¨t, 2001). Dupouey et al. (2002) have

* Corresponding author. Tel.: þ33 0 3 83 39 40 79; fax: þ33 0 3 83 39 40 69. E-mail address: [email protected] (A.G. Diedhiou).

shown that 200 years of farming during Roman times had induced gradients in soil nutrient availability and biodiversity that are still measurable in present forests almost 2000 years later. Up to now, phanerogamic communities only have been investigated, although other biota could also have been impacted. In forest ecosystems, fungi are associated with the roots of trees, forming mixed symbiotic organs called ectomycorrhizas (ECMs), which perform the uptake of water and nutrients for the trees. The genetic and functional diversities of ECMs are considered key to forest soil ecosystem functioning through decomposition and mineralization of organic matter (Smith and Read, 1997). Numerous reports indicate that fertilization influences the composition and function of ECM communities. For instance, increased N deposition reduces the production and diversity of ECM sporocarps, and modifies the species composition of ECM fungi colonizing roots belowground (Taylor et al., 2000; Peter et al., 2001; Lilleskov et al., 2001, 2002; Avis et al., 2003). Moreover, Taylor and Read (1996) noted that, in the areas with high N deposition, the ECM fungi which can readily use organic N were replaced in areas with high N deposition by those which rely largely or solely upon inorganic N sources. Despite the increasingly evidences of shifts in the species composition of ECM communities following increasing N

0038-0717/$ – see front matter Crown Copyright Ó 2009 Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.soilbio.2009.08.005

A.G. Diedhiou et al. / Soil Biology & Biochemistry 41 (2009) 2206–2213

deposition during the last 20th century (Wallenda and Kottke, 1998; Cairney and Meharg, 1999; Lilleskov and Bruns, 2001), nothing is known about the influence of past agricultural use of soils on the ECM community composition when the formerly farmed sites have been re-colonized by forest during later historical periods. Yet, past changes in land-use are a major factor currently affecting the availability of nutrients in forest ecosystems all over the world. Mycorrhizae play a central role in phosphorous supply and demand of plants (Koide, 1991). In a meta-analysis of mycorrhizal responses to nitrogen and phosphorus manipulation in field studies, Treseder (2004) showed that the decrease of mycorrizal abundance following phosphorus addition was stronger and more consistent across studies than the decrease following nitrogen addition. Moreover, phosphorus is the soil element whose concentration shows the most increase in the vicinity of ancient human settlements (Craddock et al., 1985). Traces of human occupation, even dating back as far as the Neolithic, are routinely mapped by soil phosphorus analysis. Thus, the hypothesis that mycorrhizae communities could be, in the long run, affected by ancient land use seems natural. The aim of this study was to determine if the past Roman occupation of presently forested areas has durably modified the ECM community composition. For this purpose, we set up and sampled a network of paired sites, previously disturbed or undisturbed during the Roman period. 2. Materials and methods 2.1. Studied area and plot selection The state forest of Tronçais is located in Central France (46 38’28.6000 N, 2 42’58.9100 E). The history of the Tronçais forest has been marked by a diversity of successive types of human occupation and management, as described in many historical documents. Recently (Bertrand, 1996; Lau¨t, 2001), 108 Roman settlements have been found by surface surveys of stones, tiles and ceramics, and have been dated from the 1st to the 4th century AD according to standard archaeological references. Each settlement is defined by the ruins of one or a few Roman buildings. Historical documents point to a very ancient forest recolonisation. From the 17th to the 19th century, the forest of Tronçais has been intensively exploited for wood and grazed, before being used for making charcoal. It is nowadays managed as a pure oak high forest and is the source of high-quality oak wood used for making wine casks. It covers 10,600 ha of a homogenous plateau with a mean elevation of 250 m. The canopy is dominated by sessile oak (Quercus petraea > 80%) mixed with beech (Fagus sylvatica). Hornbeam (Carpinus betulus) is frequent in the understory. The soils are mostly sandy acidic inceptisols developed from standstone or alluvial sand deposits with various degrees of hydromorphy. The humus form is mull to moder (see Dambrine et al., 2007). Twelve Roman settlements (P1 to P12) were selected, evenly distributed across the forest. The main criteria of selection were the homogeneity of topography and stand characteristics, both in terms of species composition (mature sessile oak dominant) and structure (regular high forest) within the site. At each site, we made our observations in two plots (600 m2 each), one plot close to the remains of the central building (<100 m), supposed to be impacted by ancient Roman agriculture, and the second plot at 250–500 m away from the central building, supposed to be out of the influence or at least less influenced by previous cultivation. The age of each plot was obtained from ancient forest management records, which provided the date of stand regeneration. Stands were between 60 and 160 year-old. The two plots at a given site always were the same age.

2207

2.2. Soil and root sampling For each of the 12 sites, at both the centre and the outer plots, five cylindrical soil cores (4 cm diameter, 20 cm deep) were collected within an area of less than 50 m2 and more than 1 m from tree trunks, during the year 2006. The soil of each plot was air-dried and sieved (<2 mm) to discard roots and stones, and analysed for pH (H2O), total C, N and P contents, available P according to the methods of Olsen (Olsen et al., 1954) and Duchaufour (Duchaufour and Bonneau, 1959), and exchangeable cations (Ca, Mg, K, Na) at pH 7 (ammonium acetate method) at the INRA Central Soil Analysis Laboratory in Arras, France. For ECM analysis, sampling was performed twice in the year in each of the 24 selected plots (12 from centres and 12 from outer zones). Soil cores were extracted following the same protocol as for soil analyses. The soil cores were separately wrapped with polythene film and kept into isotherm boxes until arrival at the laboratory within 24 h. The soil cores were kept at 4  C from 1 to 2 days after sampling. For each plot, the 5 cores were pooled in order to perform soil and ECM analysis. The roots were gently separated from the surrounding soil, washed and observed in water using a stereomicroscope. ECM tips were classified into morphotypes (MTs) based on distinctive macroscopic and microscopic features: branching, colour and texture of the mantle, presence or absence of external hyphae, mycelial strands or rhizomorphs, and sclerotia linked to ectomycorrhizae. In all samples, three ECM tips of each encountered MT were frozen in liquid nitrogen and stored at – 20  C for later molecular identification. The relative abundance (in percent) of dominant MTs in each plot was determined from a subsample of 150 ECM tips randomly taken from the root sample. The frequency of each MT was calculated as the number of times it was encountered among the 24 samples (12 sites  2 dates) from the centre and outer plots. 2.3. Molecular analyses Total DNA was isolated from the three frozen ECM tips of each morphotype in each sample using the DNeasy Plant mini-kit following the manufacturer’s recommendations (Qiagen SA, Courtaboeuf, France). The internal transcribed spacer (ITS) region of the fungal nuclear ribosomal DNA was amplified with the fungus-specific primer pair ITS1f - ITS4 (White et al., 1990). All amplifications were performed as in Diedhiou et al. (2004). Before sequencing, the amplified DNA was purified on a 96-well MultiScreen-PCR plate system following the manufacturer’s recommendations (Millipore SA, Molsheim, France). The ITS region was directly sequenced using ITS1/ITS4 or ITS1/ ITS2 primers with CEQ DTCS-Quick Start Kit on 8-capillar sequencer CEQ 2000XL (Beckman, Fullerton, CA, USA). Forward and reverse DNA sequences of each MT were aligned to produce a consensus DNA sequence. In order to determine the taxonomical belonging of each MT forming fungus, DNA sequences were compared to the GenBank (http://www.ncbi.nlm.nih.gov/) and UNITE (http://unite.ut.ee/index. php) databases using the BLASTn algorithm (Altschul et al., 1990). 2.4. Statistical analysis Since the study plots were paired, one element of each pair being in the centre of the site, the other one away from the zone of heavy disturbance, we tested the differences in soil characteristics and MT richness between the two ancient land uses by paired t-tests. In addition to individual soil variables analysis, a principal component analysis of the ten measured (see above) and two calculated (C/N ratio and sum of exchangeable cations Ca, Mg, K, Na, noted SEC) soil characteristics was conduced and differences in position along the first axis were tested. In order to simultaneously

2208

A.G. Diedhiou et al. / Soil Biology & Biochemistry 41 (2009) 2206–2213

test site and ancient land-use effects, a combined model of analysis of variance was fitted: Yij ¼ a þ ALUi þ SITEj þ errij Where Yij is the value of variable of interest (soil characteristics, number of taxa.) at site j, ALUi is the effect of ancient land use i, either disturbed or undisturbed, SITEj is the effect of the jth site (12 sites) and err is the residual error term. In case of repeated measurements along time, the variable of interest was averaged over the different dates before analysis. Normality of the distribution of residuals was checked by visual inspection. The diversity of ECM communities was assessed through different indices. We computed the number of taxa per plot (alpha diversity, also called richness). We calculated the Shannon-Wiener diversity and equitability indices, which weight each species by its relative abundance. The composition of the ECM community was investigated with multivariate analysis. We analysed the table crossing all the identified MTs with all the samples analysed (12 sites  2 plots  2 dates). The aim of this multivariate analysis was to identify the main gradients responsible for the differentiation of morphotype communities. We used a non-metric multidimensional scaling (NMDS) approach because it has been shown to be superior to classical methods such as factor analysis or correspondence analysis for retrieving such gradients in community composition (Fasham, 1977; Minchin, 1987). The Sorensen coefficient (Sorensen, 1948) was preferred for the calculation of similarity between pairs of plots because it provides double weight to the presence of a species in both plots. This coefficient is based on presence-absence of the morphotypes. Seven axes were retained after inspection of the stress value as a function of the number of axes. The morphotypes significantly associated to a given multivariate axis issued from the multidimensional scaling were identified by logistic regression of the presenceabsence of all MTs on the axis coordinates and subsequent X2 test of the significance of this axis coordinates in the logistic regression equation. In addition, morphotype frequencies were compared between centre and outer plots by an exact Fisher test. All analyses were performed with SAS 9.0 (procedure TTEST for t-tests, GLM for analysis of variance, MDS for multidimensional scaling, LOGISTIC for fitting logistic regression, FREQ for exact Fisher test).

Table 1 Soil properties according to Roman ancient land-use. Mean values and standard errors of the samples from centre and outer plots, followed by the effect of occupation (þ positive,  negative) and associated paired t-tests (*: P < 0.05; **: P < 0.01; ns: not significant). All comparisons were made over 12 pairs of plots. Centre plots Total carbon (g/kg) Total nitrogen (g/kg) Carbon/Nitrogen (C/N) Phosphorus (g/kg, Olsen) Phosphorus (g/kg, Duchaufour) Total phosphorus (g/100 g) Calcium (Ca, g/kg) Magnesium (Mg, g/kg) Potassium (K, g/kg) Sodium (Na, g/kg) Sum of exchangeable cations (cmol/kg) pH (H2O)

Outer plots

Effect of past occupation

C 16.1  N 1.01  C/N 15.9  Pols 0.012  Pduch 0.114 

1.2 0.06 0.7 0.002 0.017

13.9 0.84 16.5 0.008 0.072

    

0.5 0.02 0.4 0.001 0.006

þ þ – þ þ

ns * ns * *

Ptot Ca Mg K Na SEC

0.005 0.041 0.003 0.009 0.0002 0.25

0.069 0.062 0.017 0.069 0.003 0.64

     

0.003 0.006 0.001 0.005 0.0003 0.06

þ þ þ þ þ þ

** ns * ns * ns

þ

ns

pH

0.086 0.140 0.025 0.087 0.004 1.14

     

4.96  0.07

4.84  0.07

Thelephorales (9%), and ‘‘others’’ grouping Atheliales, Cantharellales and Sebacinales (11%). The frequency and abundance of each MT are in Table 2. Cenococcum geophilum (MT09) was the most abundant and most frequent MT; it represented 19.5% of MT total relative abundance and was found in 45 samples out of 48. The following most abundant ECM taxa were Lactarius quietus (MT01) and Russula nigricans (MT07), with 10.5% and 7.2% of total relative abundance respectively. Their frequencies were 23/48 and 32/48 respectively. There is a highly significant and positive correlation between the frequency (presence/absence) and the relative abundance (r ¼ 0.89, P < 0.001, n ¼ 67) of MTs (Fig. 2).

3.3. Effect of past Roman occupation on ECM community composition

3. Results 3.1. Soil properties The analyses of soils from the centre and outer plots of the Roman sites showed clear differences of fertility (Table 1). A general trend towards a higher nutrient content in centre plots appeared (Fig. 1). Position of plots along the first axis of principal component analysis significantly differed between centre and outer plots (P < 0.01). The most representative element of this difference was phosphorus, which concentration was significantly higher in the centre plots. The contents of the exchangeable cations Mg and Na were significantly higher in the occupied areas, but those of K and Ca were not significantly different (Table 1). A significant increase in nitrogen was also observed. However, no significant difference was recorded for total carbon or the carbon to nitrogen ratio. 3.2. General traits of the ECM communities The 7200 root tips observed in the 48 samples (24 plots and two sampling dates) fell into 67 morphotypes considered as due to different putative ECM species (Table 2). From the 67 MTs, 53 were successfully identified at the genus or species level after ITS sequencing and BLASTn search (Table 2). Six fungal taxonomical groups were defined from the identified MTs by BLASTn search: Ascomycota (34% of total relative abundance of identified ECM taxa), Russulales (23%), Agaricales (10%), Boletales (13%),

From the 67 morphotypes, 40 were shared by the centre and outer plots, 17 were found only in the centre plots and 10 only in the outer plots (Table 2). The ECM alpha diversity (total number of MTs per plot) was significantly (P < 0.05) higher in the centre plots (N ¼ 10.3) than in the outer plots (N ¼ 9.0). Significant and larger variations in the number of ECM were also observed between sites, with a difference of 5 MTs between the poorest and the richest sites. Shannon-Wiener diversity and equitability indices did not differ between ancient land uses. The analysis of the ECM community composition of the 48 samples from the 24 plots (2 sampling dates per plot) using multidimensional scaling discriminated the centre plots and outer plots, on the fourth axis (Fig. 3). The ancient land-use meaning of the fourth axis was confirmed by the significant correlations with soil variables: total phosphorus content (r ¼ 0.59, P < 0.01) or phosphorus measured with the Olsen’s method (r ¼ 0.53, P < 0.01), pH (r ¼ 0.54, P < 0.01), Mg (r ¼ 0.56, P < 0.01), N (r ¼ 0.54, P < 0.01). Correlation with soil phosphorus content measured with the Duchaufour’s method was not significant. The first axis of the same NMDS showed a significant spatial organisation of ECM communities, separating between eastern and western sites in the whole Tronçais forest. The correlation between the longitude of sites and position along this first axis was 0.50 (P < 0.05). This geographical segregation between ECM communities was not linked to any soil chemical property. The second axis of NMDS was correlated with the date of sampling in the year and the third one was linked to inter-site variations.

A.G. Diedhiou et al. / Soil Biology & Biochemistry 41 (2009) 2206–2213

2209

Table 2 Q. petraea ectomycorrhizae sampled from the 12 studied Roman sites in the Tronçais forest. Identification was based on morphotyping and sequencing of the internal transcribed spacer (ITS) rDNA followed by BLASTn search in the NCBI and UNITE databases. Morphotype (MT), and best BLASTn full-length ITS match (taxonomical group)

GenBank accession number

Frequency in the centre plots

Frequency in the outer plots

Abundance in the centre plots (%)

Abundance in the outer plots (%)

MT01, Lactarius quietus (Russulales) MT02, Genea spp. (Ascomycota) MT03, Genea hispidula (Ascomycota) MT04, Unidentified taxon MT05, Tomentella sp. (Thelephorales) MT06, Laccaria amethystina (Agaricales) MT07, Russula nigricans (Russulales) MT08, Sebacinaceae sp. (others) MT09, Cenococcum geophilum (Ascomycota) MT10, Cortinarius subsertipes (Agaricales) MT11, Russula sp. (Russulales) MT12, Oidiodendron citrinum (Ascomycota) MT13, Xerocomus pruinatus (Boletales) MT14, Thelephoraceae sp. (Thelephorales) MT15, Cortinarius sertipes (Agaricales) MT16, Cortinarius saturninus (Agaricales) MT17, Helotiales sp. (Ascomycota) MT18, Cortinarius subsertipes (Agaricales) MT19, Unidentified taxon MT20, Xerocomus communis (Boletales) MT21, Unidentified taxon MT22, Xerocomus chrysenteron (Boletales) MT23, Boletus edulis (Boletales) MT24, Thelephoraceae sp. (Thelephorales) MT25, Sebacinaceae sp. (others) MT26, Unidentified taxon MT27, Tomentella botryoides (Thelephorales) MT28, Russula atropurpurea (Russulales) MT29, Russula sp. (Russulales) MT30, Cantharellaceae sp. (others) MT31, Lactarius chrysorheus (Russulales) MT32, X. pruinatus (Boletales) MT33, Xerocomus dryophilus (Boletales) MT34, Leptodontidium sp. (Ascomycota) MT35, Byssocorticium atrovirens (others) MT36, Thelephoraceae sp. (Thelephorales) MT37, Tomentellopsis zygodesmoides (Thelephorales) MT38, R. atropurpurea (Russulales) MT39, Scleroderma verrucosum (Boletales) MT40, Unidentified taxon MT41, Unidentified taxon MT42, Helotiales sp. (Ascomycota) MT43, Unidentified taxon MT44, Entolomataceae sp. (Agaricales) MT45, Oidiodendron maius (Ascomycota) MT46, Unidentified taxon MT47, Dermocybe olivaceopicta (Agaricales) MT48, Tomentella badia (Thelephorales) MT49, Inocybe maculata (Agaricales) MT50, Tarzetta cf. Cupularis (Ascomycota) MT51, Tomentella subtestacea (Thelephorales) MT52, Russula fellea (Russulales) MT53, Hydnotrya tulasnei (Ascomycota) MT54, Leccinum pseudoscabrum (Boletales) MT55, Unidentified taxon MT56, Sebacinaceae sp. (others) MT57, Russula ochroleuca (Russulales) MT58, R. atropurpurea (Russulales) MT59, Unidentified taxon MT60, Unidentified taxon MT61, Russula rosea (Russulales) MT62, Amanita rubescens (Agaricales) MT63, Tricholoma sulphureum (Agaricales) MT64, Tomentella lateritia (Russulales) MT65, Unidentified taxon MT66, Unidentified taxon MT67, Inocybe aurea (Agaricales)

FM995571 nda FM995589 nd FM995562 FM995554 FM995572 FM995582 FM995597 FM995585 FM995579 FM995592 FM995547 FM995561 FM995588 FM995586 FM995594 FM995587 nd FM995552 nd FM995550 FM995551 FM995566 FM995580 nd FM995563 FM995573 FM995599 FM995598 FM995570 FM995548 FM995549 FM995596 FM995583 FM995567 FM995569 FM995575 FM995553 nd nd FM995595 nd FM995558 FM995593 nd FM995557 FM995568 FM995556 FM995590 FM995565 FM995576 FM995591 FM995584 nd FM995581 FM995577 FM995574 nd nd FM995578 FM995560 FM995559 FM995564 nd nd FM995555

14 14 4 1 9 3 17 5 23 2 5 8 2 4 9 3 2 1 2 0 3 1 7 8 16 1 4 1 0 1 3 0 13 5 9 1 1 0 6 1 1 1 1 2 2 5 4 1 1 1 1 1 1 2 0 0 4 1 4 1 1 0 0 1 0 0 1

9 11 2 2 1 4 15 4 22 3 10 10 6 1 5 1 1 0 4 2 2 0 6 5 16 0 0 4 3 0 3 1 9 8 6 2 0 3 0 1 5 0 2 0 0 3 6 0 0 0 0 1 0 2 1 1 1 1 2 2 0 3 2 1 1 1 0

13 3.3 1.1 0.14 6.2 1.7 8.0 1.25 19 0.56 1.6 0.44 0.33 0.44 3.9 0.08 0.89 0.50 0.28 0 1.7 0.22 0.86 2.5 6.1 0.75 1.0 0.03 0 0.03 1.3 0 4.7 0.42 1.8 0.03 0.33 0 2.1 0.56 0.58 0.19 0.47 2.4 0.72 4.1 0.17 0.03 0.28 0.03 0.47 0.03 0.14 1.75 0 0 0.58 0.17 1.5 0.03 0.03 0 0 0.03 0 0 0.03

8.0 5.3 1.4 0.44 0.03 3.5 6.4 0.47 20 0.53 3.1 4.0 3.6 0.22 3.6 0.08 0.47 0 0.94 0.67 0.50 0 1.4 2.25 7.8 0 0 0.56 0.36 0 1.2 0.03 6.1 1.3 1.25 0.78 0 0.42 0 1.1 1.4 0 0.03 0 0 4.7 0.36 0 0 0 0 0.03 0 0.56 0.44 0.11 0.14 0.61 0.81 0.56 0 0.06 0.58 0.89 0.92 0.03 0

a

nd, not determined.

2210

A.G. Diedhiou et al. / Soil Biology & Biochemistry 41 (2009) 2206–2213

Fig. 1. Principal component analysis of the twelve measured and calculated soil characteristics. Projection of variables (left) and plots (right) on the plane of axes 1 and 2. Open circles are outer plots and black circles centre plots.

A comparison of the frequency of each MT according to ancient land use revealed that some MTs occurred preferentially either in the centre or in the outer plots (Table 3). At taxonomical group level, only Thelephorales were markedly associated with ancient settlements (encountered 29 times in the centre plots, against only 9 in the outer plots, P < 0.001). The occurrence of thirteen MTs was significantly related to the position of sites along dimension 4 according to a logistic regression (Table 3). Scleroderma verrucosum (MT39), Tomentella botryoides (MT27), Tomentella sp. (MT05), Cortinarius sertipes (MT15), L. quietus (MT01), Byssocorticium atrovirens (MT35) and Thelephoraceae sp. (MT24) were stronly linked with the centre

plots along dimension 4. Among those, S. verrucosum (MT39) and T. botryoides (MT27) are likely to be specific of the centre plots, because they were never found in the outer plots. In contrast, Russula atropurpurea (MT58), Genea hispidula (MT03), Xerocomus pruinatus (MT13), Boletus edulis (MT23), unidentified taxon (MT46) and unidentified taxon (MT11) were strongly linked with the outer zones; however, they were not specific, occurring even in the centre plots. R. nigricans (MT07), L. quietus (MT01), Genea spp. (MT02), C. geophilum (MT09) and C. sertipes (MT15) were linked with soils with high carbon content and situated in the SW of the forest. In contrast, Thelephoraceae sp. (MT24), Dermocybe olivaceopicta

Fig. 2. Relation between local abundance and global frequency of ECM morphotypes (MTs) found in the 48 samples.

A.G. Diedhiou et al. / Soil Biology & Biochemistry 41 (2009) 2206–2213

2211

Fig. 3. NMDS analysis of the ECM communities sampled from the 12 Roman sites. Fourth axis of the analysis as a function of soil total phosphorus content. Open circles represent ECM morphotypes of the outer plots and black circles represent those of the centre plots.

(MT47), Leptodontidium sp. (MT34) and Oidiodendron citrinum (MT12) were linked with low carbon contents and sites in the NE. 4. Discussion The soil properties were significantly affected by the past Roman occupation, reflected as increased nutrient availability, especially for P, N, Mg and Na. The enhancement of soil nutrient levels may be the result of ancient agricultural practices. Indeed, during the Roman period, according to Roman agronomic treatises, club wheat (Triticum compactum Host) was commonly cultivated, accompanied by barley (Hordeum vulgare L.), oats (Avena sativa L.), and rye (Secale cereale L.). Nitrogen-fixing legumes (Pisum sativum L., Lens esculenta, and Vicia sp.) were also intensively cultivated. These different crops were grown in ploughed fields which were regularly fertilized using ashes, animal manure or green manure. Moreover, beside crop production, animal rearing was also practised as attested by animal macroremains (Dupouey et al., 2002). In the Tronçais forest, pollens of cereals were identified in two wet depression in the close vicinity of the buildings, confirming previous cultivation (Dambrine et al., 2007). The enhancement of soil nutrient levels by past Roman occupation has impacted on ECM communities. Morphotypes displayed higher richness in the occupied compared to the non-occupied plots, and the ECM community composition was different. This shift may be, for a part, related to resource availability, which may contribute to the outcome of different ecological niches (Baier et al., 2006; Bue´e et al., 2007; Tedersoo et al., 2008). Thus, some ECM species such as S. verrucosum (MT39) and T. botryoides (MT27) are likely to be specific of richest soils as attested by their exclusive occurrences in the centre plots. S. verrucosum was previously described to inhabit the richest soils (Laessoe and Conte, 1996). We therefore confirm that S. verrucosum is a good bio-indicator of rich soils. T. botryoides is included in the Thelophorales group, which occurred significantly and preferentially in the centre (enriched plots) compared to the outer plots. Fransson et al. (2000), (2001) also noted that Thelephorales group showed an increased abundance in the fertilized plots. ECM species belonging to other

taxonomical groups preferentially occurred either in the centre plots (i.e. L. quietus), or in the outer plots (i.e. R. atropurpurea). C. geophilum, the most frequent and abundant ECM taxon was not affected by high soil nutrient levels. The lack of response of C. geophilum to enhanced soil nutrient levels in our work is consistent with other works by Jonsson et al. (2000) or by Avis (2005). Our results thus suggest that ECM fungal species respond differently to the availability of resources, and disturbance (Bergemann and Miller, 2002; Lilleskov and Bruns, 2003). Interestingly, the same distribution patterns were observed for frequencies and relative abundances of ECM fungal species. These different responses displayed by ECM fungal taxa to enhanced nutrient levels are likely to contribute to the resilience of forest ecosystems exposed to environmental changes; having diverse responses of ECM fungal taxa in the same ECM community might allow trees to buffer changing environmental conditions over time. Strikingly, the higher soil nutrient levels in the occupied zones did not reduce ECM richness, in contrast to other studies of ECM community composition following N deposition or fertilization (Kåre´n, 1997; Peter et al., 2001; Lilleskov et al., 2002), or litter addition (Cullings et al., 2003). However, these other works were made essentially in conifer forests. Taylor et al. (2000) also found a negative relationship between ECM MT richness and soil inorganic N in spruce stands, while they noted a weaker positive relationship between these variables in beech stands. This emphasizes the need for more comparison of conifer and deciduous forests for understanding the effect of shifting resources on ECM biodiversity in the context of environmental changes (Lilleskov and Bruns, 2001). The NMDS analysis of all samples from the 12 studied sites discriminated the MTs from the centre plots and those from the outer plots, except for sites P2 and P10 of which the ECM communities from the occupied plots displayed a similar composition to that of the ECM communities from the outer plots. These two sites are those where the overall impact of the Roman occupation has not been great and thus the human impact is not still visible in ECM composition. In the case of site P10, this may be due

2212

A.G. Diedhiou et al. / Soil Biology & Biochemistry 41 (2009) 2206–2213

Table 3 Ectomycorrhizal (ECM) fungi showing biased occurrence between the centre plots and outer plots, and following geographic location and soil carbon content. Preferential occurrence Frequency according to ancient land use

Associated test (P value) Fisher exact test

ECM taxa Scleroderma verrucosum (MT39) Tomentella sp. (MT05) Tomentella botryoides (MT27) Russula sp. (MT29) Russula atropurpurea (MT38) Amanita rubescens (MT62)

Centre plots Centre plots Centre plots Outer plots Outer plots Outer plots

0.0024 0.0026 0.0151 0.0368 0.0368 0.0368

Taxonomical group Thelephorales

Centre plots

0.0009

NMDS dimension 4 Scleroderma verrucosum (MT39) Tomentella sp. (MT05) Byssocorticium atrovirens (MT35) Tomentella botryoides (MT27) Thelephoraceae sp. (MT24) Cortinarius sertipes (MT15) Lactarius quietus (MT01) Russula atropurpurea (MT58) Unidentified taxon (MT46) Genea hispidula (MT03) Boletus edulis (MT23) Xerocomus pruinatus (MT13) Unidentified taxon (MT11)

Centre plots Centre plots Centre plots Centre plots Centre plots Centre plots Centre plots Outer plots Outer plots Outer plots Outer plots Outer plots Outer plots

X2 test <0.0001 0.0033 0.0186 0.0023 0.0153 0.0016 0.0191 0.0168 0.0087 0.0173 0.0142 0.0372 0.0471

NMDS dimension 2 Russula nigricans (MT07) Lactarius quietus (MT01) Genea spp. (MT02) Tomentella sp. (MT05) Helotiales sp. (MT17) Cenococcum geophilum (MT09) Xerocomus pruinatus (MT13) Cortinarius sertipes (MT15) Thelephoraceae sp. (MT24) Xerocomus dryophilus (MT33) Leptodontidium sp. (MT34) Unidentified taxon (MT46) Dermocybe olivaceopicta (MT47) Unidentified taxon (MT43) Oidiodendron citrinum (MT12)

SW & high carbon SW & high carbon SW & high carbon SW & high carbon SW & high carbon SW & high carbon SW & high carbon SW & high carbon NE & low carbon NE & low carbon NE & low carbon NE & low carbon NE & low carbon NE & low carbon NE & low carbon

<0.0001 0.0006 0.0007 0.0034 0.0077 0.0208 0.0372 0.0382 0.0002 0.0028 0.0049 0.0061 0.0069 0.0178 0.0322

communities in comparison with undisturbed areas. This longterm human impact is important to take into account when analyzing the determinants of present ECM species distribution in forests. Since significant effects on ECM taxa were observed after almost 2000 years of afforestation, they should be even larger for shorter periods of times. This suggests that the huge areas which were re-colonized by temperate forests on lands left over by agriculture during the last two centuries should display very different ECM communities than those in ancient forests. Our study raises other fascinating questions: do these ECM communities function differently, and do they benefit plant growth? Acknowledgements

to a lower intensity of past land use (Dupouey et al., 2002): we observed lower levels of soil nutrients in the centre plot of this site compared to those of the other sites (data not shown). In the case of site P2, archaeological investigations based on the remaining edifices, density and spatial distribution of walls, abundance of potteries and tiles revealed that it was a cultural site with a temple instead of an agricultural settlement (Dambrine et al., 2007). After nearly two thousand years of abandonment, the effects of ancient land use are still visible in soil characteristics. The causes for such a long memory of forest soils remain largely unknown. Considering the potential nutrient losses through historical periods, the induced soil fertility should progressively disappear, and therefore should not be detectable nowadays. However, microbial communities may play an important role for the long-term persistence of the initial enhancement of soil fertility. Indeed, changes in both the activity and the species composition of soil microbial communities likely influenced the nutrient cycling (Paterson et al., 1997; Olsson and Wallander, 1998; Conn and Dighton, 2000). Once initiated by ancient land use, changes in soil microbial communities may be more long-lasting and therefore may contribute to the maintenance of the enhancement of soil nutrient levels. Our study showed for the first time that Roman agricultural activities significantly changed the species composition of ECM

We thank Bruno Jube´ra and the French National Forest Service (ONF) for practical and technical support during fieldwork. We acknowledge Aurore Coince for help in the enzymatic activities measurements. We thank the anonymous referees for their valuable comments on this paper. This project was financially supported by the French GIP-Ecofor program ‘‘Typologie des Stations’’. References Altschul, S.F., Gish, W., Miller, W., Myers, E.W., Lipman, D.J., 1990. Basic local alignment search tool. Journal of Molecular Biology 215, 403–410. Avis, P.G., 2005. The response of ectomycorrhizal fungal inoculum to long-term increases in nitrogen supply. Mycologia 97, 329–337. Avis, P.G., McLaughlin, D.J., Dentinger, B.C., Reich, P.B., 2003. Long-term increase in nitrogen supply alters above and below ground ectomycorrhizal communities and increases the dominance of Russula spp. in a temperate oak savanna. New Phytologist 160, 239–253. Baier, R., Ingenhaag, J., Blaschke, H., Go¨ttlein, A., Agerer, R., 2006. Vertical distribution of an ectomycorrhizal community in upper soil horizons of a young Norway spruce (Picea abies [L] Karst.) stand of the Bavarian Limestone Alps. Mycorrhiza 16, 197–206. Bergemann, S.E., Miller, S.L., 2002. Size, distribution, and persistence of genets in local populations of the late-stage ectomycorrhizal basidiomycete, Russula brevipes. New Phytologist 156, 313–320. Bertrand, E., 1996. Les sites antiques. La lettre du Pays de Tronçais 2, 1–48. Bue´e, M., Courty, P.E., Mignot, D., Garbaye, J., 2007. Soil niche effect on species diversity and catabolic activities in an ectomycorrhizal fungal community. Soil Biology and Biochemistry 39, 1947–1955. Cairney, J.W.G., Meharg, A.A., 1999. Influences of anthropogenic pollution on mycorrhizal fungal communities. Environmental Pollution 106, 169–182. Compton, J.E., Boone, R.D., Motzkin, G., Foster, D.R., 1998. Soil carbon and nitrogen in a pine-oak sand plain in central Massachusetts: role of vegetation and land-use history. Oecologia 116, 536–542. Conn, C., Dighton, J., 2000. Litter quality influences on decomposition, ectomycorrhizal community structure and mycorrhizal root surface acid phosphatase activity. Soil Biology and Biochemistry 32, 489–496. Craddock, P.T., Pryor, F., Hugues, M.J., 1985. The application of phosphate analysis to the location and interpretation of archaeological sites. Archeological Journal 142, 361–376. Cullings, K.W., New, M.H., Makhija, S., Parker, V.T., 2003. Effects of litter addition on ectomycorrhizal associates of a lodgepole pine (Pinus contorta) stand in Yellowstone National Park. Applied and Environmental Microbiology 69, 3772–3776. Dambrine, E., Dupouey, J.L., Lau¨t, L., Humbert, L., Thinon, M., Beaufils, T., Richard, H., 2007. Biodiversity patterns in present French forests related to former Roman agriculture. Ecology 88, 1430–1439. Diedhiou, A.G., Baˆ, A.M., Sylla, S.N., Dreyfus, B., Neyra, M., Ndoye, I., 2004. The early-stage ectomycorrhizal Thelephoroid fungal sp. is competitive and effective on Afzelia africana Sm. in nursery conditions in senegal. Mycorrhiza 14, 313–322. Duchaufour, P., Bonneau, M., 1959. Une me´thode nouvelle de dosage du phosphore assimilable dans les sols forestiers. Bulletin de l’Association Française d’e´tude du sol 4, 193–198. Dupouey, J.L., Dambrine, E., Laffite, J.D., Moares, C., 2002. Irreversible impact of past land use on forest soils and biodiversity. Ecology 83, 2978–2984. Fasham, M.J.R., 1977. A comparison of nonmetric multidimensional scaling, principal components and reciprocal averaging for the ordination of simulated coenoclines and coenoplanes. Ecology 58, 551–561. Foster, D.R., 2002. Insights from historical geography to ecology and conservation: lessons from the New England landscape. Journal of Biogeography 29, 1269–1275. Fransson, P.M.A., Taylor, A.F.S., Finlay, R.D., 2000. Effects of continuous optimal fertilisation upon belowground ectomycorrhizal community structure in a Norway spruce forest. Tree Physiology 20, 599–606.

A.G. Diedhiou et al. / Soil Biology & Biochemistry 41 (2009) 2206–2213 Fransson, P.M.A., Taylor, A.F.S., Finlay, R.D., 2001. Elevated atmospheric CO2 alters root symbiont community structure in forest trees. New Phytologist 152, 431–442. Hermy, M., Honnay, O., Firbank, L., Grashof-Bokdam, C., Lawesson, J., 1999. An ecological comparison between ancient and other forest plant species of Europe, and the implications for forest conservation. Biological Conservation 91, 9–22. Jonsson, L., Dahlberg, A., Brandrud, T.E., 2000. Spatiotemporal distribution of an ectomycorrhizal community in an oligotrophic Swedish Picea abies forest subjected to experimental nitrogen addition: above- and below-ground views. Forest Ecology and Management 132, 143–156. Jussy, J.H., Koerner, W., Dambrine, E., Dupouey, J.L., Benoit, M., 2002. Influence of former agricultural land use on net nitrate production in forest soils. European Journal of Soil Science 53, 367–374. Kåre´n, O., 1997. Effects of Air Pollution and Forest Regeneration Methods on the Community Structure of Ectomycorrhizal Fungi. PhD thesis, Department of Forest Mycology and Pathology, Swedish University of Agricultural Sciences, Uppsala, Sweden. Koerner, W., Dupouey, J.L., Dambrine, E., Benoıˆt, M., 1997. Influence of past land use on the vegetation and soils of present day forest in the Vosges mountains, France. Journal of Ecology 85, 351–358. Koerner, W., Dambrine, E., Dupouey, J.L., Benoit, M., 1999. d15N of forest soil and understorey vegetation reflect the former agricultural land use. Oecologia 121, 421–425. Koide, R.T., 1991. Nutrient supply, nutrient demand and plant response to mycorrhizal infection. New Phytologist 117, 365–386. Laessoe, T., Conte, A.D., 1996. L’Encyclope´die des champignons. Ed. Bordas, Paris, pp. 256. Lau¨t, L., 2001. L’occupation du sol dans l’Antiquite´ en foreˆt de Tronçais. Rapport a` la Direction Re´gionale de l’Arche´ologie et de la Culture de la Re´gion Centre, Paris, France. Prospection, Inventaire, Prospection the´matique annuelle, Sondages, Document Final de Synthe`se. Lilleskov, E.A., Bruns, T.D., 2001. Nitrogen and ectomycorrhizal fungal communities: what we know, what we need to know. New Phytologist 149, 154–158. Lilleskov, E.A., Bruns, T.D., 2003. Root colonization dynamics of two ectomycorrhizal fungi of contrasting life history strategies are mediated by addition of organic nutrient patches. New Phytologist 159, 141–151. Lilleskov, E.A., Fahey, T.J., Lovett, G.M., 2001. Ectomycorrhizal fungal aboveground community change over an atmospheric nitrogen deposition gradient. Ecological Applications 11, 397–410. Lilleskov, E.A., Fahey, T.J., Horton, T.R., Lovett, G.M., 2002. Belowground ectomycorrhizal fungal community change over a nitrogen deposition gradient in Alaska. Ecology 83, 104–115. Minchin, P.R., 1987. An evaluation of the relative robustness of techniques for ecological ordination. Vegetation 69, 89–107.

2213

Olsen, S.R., Cole, C.V., Watanable, F.S., Dean, L.A., 1954. Estimation of Available Phosphorus in Soils by Extraction with Sodium Bicarbonate. U.S. Dept. of. Agriculture, Washington DC. Cir. No 939. Olsson, P.A., Wallander, H., 1998. Interactions between ectomycorrhizal fungi and the bacterial community in soils amended with various primary minerals. FEMS Microbiology Ecology 27, 195–205. Paterson, E., Hall, J.M., Rattray, E.A.S., Griffiths, B.S., Ritz, K., Killham, K., 1997. Effect of elevated CO2 on rhizosphere carbon flow and soil microbial processes. Global Change Biology 3, 363–377. Peltre, J., Bruant, P., 1991. Terroirs fossiles me´connus en Lorraine. Le Pays Lorrain 72, 51–55. Peter, M., Ayer, F., Egli, S., 2001. Nitrogen addition in a Norway spruce stand altered macromycete sporocarp production and belowground ectomycorrhizal species composition. New Phytologist 149, 311–325. Peterken, G.F., 1996. Natural Woodland - Ecology and Conservation in Northern Temperate Regions. Cambridge University Press, Cambridge, UK, pp. 522. Smith, S., Read, J., 1997. Mycorrhizal Symbiosis, second ed. Clarendon Press, Oxford. Sorensen, T., 1948. A method of establishing groups of equal amplitude in plant sociology based on similarity of species content and its application to analyses of the vegetation on Danish commons. Kongelige Danske Videnskabernes Selskab Biologiske Skrifter 5, 1–34. Taylor, A.F.S., Martin, F., Read, D.J., 2000. Fungal diversity in ectomycorrhizal communities of Norway spruce (Picea abies [L.] Karst.) and beech (Fagus sylvatica L.) along north–south transects in Europe. In: Schulze, E.-D. (Ed.), Carbon and Nitrogen Cycling in European Forest Ecosystems. Ecological Studies, 142, pp. 343–365. Taylor, A.F.S., Read, D.J., 1996. A European North–south survey of ectomycorrhizal populations on spruce. Mycorrhizas in integrated systems from genes to plant development. In: Azcon-Aguilar, C., Barea, J.M. (Eds.), Proceedings of the Forth European Symposium on Mycorrhizas. Luxembourg: Office for Official Publications of the European Communities, Luxembourg, pp. 144–147. Tedersoo, L., Suvi, T., Jairus, T., Ko˜ljalg, U., 2008. Forest microsite effects on community composition of ectomycorrhizal fungi on seedlings of Picea abies and Betula pendula. Environmental Microbiology 10, 1189–1201. Treseder, K.K., 2004. A meta-analysis of mycorrhizal responses to nitrogen, phosphorus, and atmospheric CO2 in field studies. New Phytologist 164, 347–355. Wallenda, T., Kottke, I., 1998. Nitrogen deposition and ectomycorrhizas. New Phytologist 139, 169–187. White, T.J., Bruns, T.D., Lee, S.B., Taylor, J.W., 1990. Amplification and direct sequencing of fungal ribosomal RNA genes for phylogenetics. In: Innis, M.A., Gelfand, D.H., Sninsky, J.J., White, T.J. (Eds.), PCR Protocols: a Guide to Methods and Applications. Academic Press, London, pp. 315–322.