Blackwell Science, LtdOxford, UKBIJBiological Journal of the Linnean Society0024-4066The Linnean Society of London, 2003 78 Original Article SOIL FUNGI DISTRIBUTION IN ‘EVOLUTION CANYON’ II, ISRAELI. GRISHKAN ET AL.
Biological Journal of the Linnean Society, 2003, 78, 527–539. With 3 figures
Adaptive spatiotemporal distribution of soil microfungi in ‘Evolution Canyon’ II, Lower Nahal Keziv, western Upper Galilee, Israel ISABELLA GRISHKAN, EVIATAR NEVO*, SOLOMON P. WASSER and ALEX BEHARAV Institute of Evolution, University of Haifa, Mt Carmel, Haifa 31905, Israel Received 25 June 2002; accepted for publication 29 October 2002
We describe and interpret spatiotemporal micromycete community structure and adaptive complexes in contrasting xeric and mesic microclimates in the soils of ‘Evolution Canyon’ II, Lower Nahal Keziv, western Upper Galilee, Israel. A total of 192 species from 60 genera belonging to Zygomycota (nine species), Ascomycota (13 species), and mitosporic fungi (170 species) were isolated. The fungal communities on the south-facing, xeric, ‘African’ slope (AS) demonstrated significantly greater diversity than on the north-facing, mesic, ‘European’ slope (ES) and the valley bottom (VB). Seasonally, winter slope communities were less heterogeneous. Forest localities on the ES and the VB in all seasons and the shady localities on the AS in the winter were overwhelmingly dominated by mesophilic Penicillium species. The sunny locality on the AS was characterized by a dominance of melanin-containing micromycetes that was most pronounced in the summer and by high occurrence and abundance of thermotolerant and thermophilic Aspergillus and Fusarium species. Ascomycetes and zygomycetes were the minor components in all local mycobiota studied; sexual ascomycetes, being stress-selected fungi, were more than ten times more abundant in the soil of the AS than in that of the ES, with the peak of abundance in the sunny summer community. The results demonstrated a microscale adaptive spatiotemporal inter- and intraslope divergence in soil mycobiota structure. Microclimatic natural selection appears to be the major factor affecting soil fungus diversity patterns. © 2003 The Linnean Society of London. Biological Journal of the Linnean Society, 2003, 78, 527–539.
ADDITIONAL KEYWORDS: adaptive complexes – biodiversity characteristics – community structure – natural selection – spatiotemporal dynamics.
INTRODUCTION Ecology of soil microfungal communities is now a subject of growing mycological interest. Many studies have recently been conducted on different aspects of soil fungal biodiversity: on local geographical scales (e.g. Betucci, Rodrigez & Indarte, 1993; Keller & Bidochka, 1998; Persiani et al., 1998; McLean & Huhta, 2000), on a regional scale (e.g. Maggi & Persiani, 1992; Zak, 1992), on distribution of different taxonomic and functional micromycete groupings (Mouchacca, 1993; Christensen, Frisvad & Tuthill, 2000; Klich, 2002), and on the relation between taxonomic and functional biodiversity (Zak & Visser, 1996).
*Corresponding author. E-mail:
[email protected]
One of the most interesting aspects of biodiversity (mycodiversity) investigation is studying the effect of microscale environmental variability and stress on biodiversity patterns. Microsite ecological contrasts are excellent critical tests for evaluating biodiversity evolution across all organizational levels, from genes to biota. They may also reveal spatiotemporal evolutionary dynamics (Nevo, 2001). The Institute of Evolution (University of Haifa, Israel) established such a test, displaying sharp ecological contrasts at a microscale in a canyon located in Lower Nahal Oren, Mt. Carmel, called ‘Evolution Canyon’ I (EC I). In EC I, problems of biodiversity evolution have been intensively investigated on different groups of organisms across life (Nevo, 1995, 1997, 2001). Mycological studies in this microsite revealed soil mycobiota comprising 204 species (Ellanskaya et al., 1997; Grishkan et al., 2000). The results demon-
© 2003 The Linnean Society of London, Biological Journal of the Linnean Society, 2003, 78, 527–539
527
528
I. GRISHKAN ET AL.
strated a strong interslope impact of edaphic and climatic conditions (mainly soil moisture), both on the spatial distribution of micromycete density and on the pattern of microfungal species composition. A long-term biodiversity research program is also being conducted in a second ‘Evolution Canyon’ (EC II) located in Lower Nahal Keziv, western Upper Galilee, 38 km north-east of EC I (Finkel, Fragman & Nevo 2001; Finkel, Chikatunov & Nevo 2002). The main goal of the mycological investigation presented in this paper addresses the adaptive pattern of mycobiota structure in the soil at EC II. The characteristics of the micromycete communities were examined for spatiotemporal dynamics: species composition, contribution of major groupings to mycobiota structure, diversity level, and dominant groups of species. The effect of soil abiotic factors (moisture, pH, and organic matter content) on the above characteristics was also estimated. The main question concerned differences in soil microfungi diversity patterns on the opposite slopes separated only by 50 m at the bottom, but sharply divergent ecologically (climatically). We also
set out to determine whether the results for EC II would be similar to those found in EC I. If so, we would be able to show biodiversity and adaptive divergent parallelism on a microscale, reinforcing the hypothesis that natural selection is a major divergent evolutionary force of soil fungi diversity (e.g. Gams, 1992).
MATERIAL AND METHODS SITE
DESCRIPTION
EC II (Fig. 1) is a Plio-Pleistocene canyon (Geological Map of Israel, 1998) in Lower Nahal Keziv, western Upper Galilee (33∞02¢ N; 35∞11¢ E). The soil type is classified as colluvial-alluvial at the bottom and terrarossa on slopes. The climate is mediterranean, with warm, dry summers (mean temperature of the hottest months, July–August, is 22–27∞C) and humid winters (mean temperature of the coldest months, January– February, is 6–12∞C). The average annual rainfall of the vicinity is about 700 mm (Atlas of Israel, 1985).
Figure 1. ‘Evolution Canyon’ II, Lower Nahal Keziv, western Upper Galilee. A cross-section with numbers of sampling stations. Note the plant formation on opposite slopes: the dense forest on the cool, mesic north-facing ‘European’ slope (ES), and the garrigue on the warm, xeric, south-facing ‘African’ slope (AS). © 2003 The Linnean Society of London, Biological Journal of the Linnean Society, 2003, 78, 527–539
SOIL FUNGI DISTRIBUTION IN ‘EVOLUTION CANYON’ II, ISRAEL EC II consists of two opposite slopes – south-facing ‘African’ slope (AS) and north-facing ‘European’ slope (ES) – that incline 20–40∞ and 30–40∞, respectively. These slopes are separated by 50 m at the bottom and 350 m at the top, but they display substantial physical and biotic contrasts because of the difference in solar radiation received, which in such canyons may be as much as 600% more intense on the ‘African’ slopes than on ‘European’ slopes (Nevo, 1997). The plant communities vary remarkably between the slopes (Finkel et al., 2001). The number of vascular plant species on the AS (205 species) is substantially higher than on the ES (54 species). The AS changes from Calicotome villosa and Salvia fruticosa garrigue at the bottom to a dry, Mediterranean, savannoid, open park forest of Ceratonia siliqua–Pistacia lentiscus association at the top, with plant cover increasing by nearly 70% downslope. By contrast, the ES is 100% covered by homogenous dense forest of Acer obtusifolium and Laurus nobilis. A dense Quercus calliprinos community covers the summer-dry riverbed in the valley bottom (VB). The study site presents a natural area which has developed for a long time without disturbance by human activity.
SAMPLING EC II has seven sampling stations, three on each slope at altitudes of approximately 140, 170, and 200 m above sea level (a.s.l.; numbers 3, 2, 1, and 5, 6, 7 on AS and ES, respectively), and one at the valley bottom at 110 m a.s.l. (no. 4). The soil samples were collected four times during 1999–2000: in the autumn (November), winter (February), spring (April), and summer (August). The seasonal soil sample series included 24 samples. On the AS, soil was collected in a sunny, open habitat and in a shady habitat under shrubs and
529
grasses (two samples from each habitat in each of three stations); on the ES and in the VB, three samples were taken from each station (shady habitats only, because these sites are 100% covered by trees). Altogether, 96 soil samples were examined. Samples were collected from the upper soil layer (1–5 cm deep) and placed in sterile paper bags. At each microlocality, 8–10 subsamples from a plot of 25 cm2 were combined into a single pooled sample (30–40 g). The samples were stored before processing (1–5 days) in a dry, cool place.
SOIL
MOISTURE,
pH,
AND ORGANIC MATTER CONTENTS
These soil parameters were measured in the Testing Laboratory of the Technion Research and Development Foundation (Israel). The methods used are shown in Table 1.
ISOLATION
OF MICROMYCETES
Micromycetes were isolated using the soil dilution plate method (Davet & Rouxel, 2000). Soil samples of 10 g each were initially diluted. Three culture media with different carbon and nitrogen sources were used: malt extract agar, Czapek’s agar (Sigma), and carboxymethylcellulose agar (10 g carboxy-methylcellulose; 1 g K2HPO4; 0.5 g KCL; 0.5 g MgSO4.7H2O; 0.01 g Fe2SO4.7H2O; 0.5 g yeast extract; 15 g agar). Streptomycine (Spectrum, ST135) was added to each medium (100 mg/mL) to suppress bacterial growth. 1 mL of sample suspension from the dilution 1 : 1000 (soil : sterile water) was mixed with the agar medium at a temperature near 40 ∞C in Petri dishes of 90 mm diameter. The plates were incubated at 25 ∞C in darkness for 10–15 days (three plates for each medium).
Table 1. Selected soil parameters for the different sites in ‘Evolution Canyon’ II, Lower Nahal Keziv, western Upper Galilee. Means with the same letter are not significantly different using the Duncan Multiple Range Test (SAS Institute, 1996) at the 5% level Moisture content, %a (N = 6, 9, 3 for AS, ES and VB, respectively) pH Localities
Autumn
AS, sunny AS, shady ES VB
15.9 ± 4.4 17.8 ± 5.3 21.4 ± 4.9 24.3 ± 2.8
a a a a
Winter
Spring
40.8 ± 4.8 b 46.6 ± 8.9 b 61.2 ± 11.4 a 59.6 ± 10.1 a
8.9 ± 1.7 12.5 ± 4.0 34.5 ± 7.7 43.2 ± 7.3
b
(N = 3)
Summer c c b a
5.5 ± 1.2 8.9 ± 2.2 16.2 ± 3.1 18.2 ± 4.6
b b a a
8.09 ± 0.20 7.92 ± 0.25 7.78 ± 0.15 7.36 ± 0.14
a
a a a b
= (c - g)100/g, where c is weight of wet soil, and g is weight of dry soil (dried at 105 ∞C for 24 h) spring samples, in water paste: 1 (water) : 5 (soil); c spring samples, Walkley and Black Method (Walkley & Black, 1934). b
© 2003 The Linnean Society of London, Biological Journal of the Linnean Society, 2003, 78, 527–539
Organic matter,%c (N = 3) 4.65 ± 2.84 9.17 ± 3.72 12.15 ± 3.15 15.52 ± 5.64
c bc ab a
530
I. GRISHKAN ET AL.
TAXONOMIC
IDENTIFICATION
After incubation, the emerging fungal colonies were transferred to malt extract agar, Czapek’s agar (the Penicillium and Aspergillus isolates), and oatmeal agar (Sigma, the Fusarium isolates) for purification and further taxonomic identification. Taxonomic identification was based on morphological characteristics of fungal isolates. Because species concepts, taxonomic base, and nomenclature strategies are subject to constant changes in soil mycology, we tried to follow modern approaches in identification of isolated fungi, especially of such dominant genera as Penicillium, Aspergillus and Fusarium. In our study, the following basic identification keys were used: Kiffer & Morelet (2000) for genera of mitosporic fungi; Domsch, Gams & Anderson (1993), Samson et al. (1996), Watanabe (1994) for miscellaneous micromycetes; Ellis (1971, 1976) for melanin-containing micromycetes; Gams (1971) for Acremonium; Rifai (1969) for Trichoderma; Sutton (1980) for Phoma and other coelomycetous fungi; Nelson, Toussoun & Marasas (1983), O’Donnell, Gigelnik & Nirenberg (1998) for Fusarium; Pitt (1979; 1994), Frisvad & Filtenborg (1989), Pitt & Cruickshank (1989), Stolk et al. (1989), Peterson (2000a) for Penicillium; Raper & Fennell (1965), Samson (1979, 1994), Peterson (2000b) for Aspergillus. All names of the Penicillium and Aspergillus species are cited according to Pitt, Samson & Frisvad (2000).
DATA
ANALYSES
For all species in each sampling period, frequency of occurrence was calculated as percent of samples in which a particular species was observed. Analysis of biodiversity was based on the evenness index (J), which was defined as J = H/Hmax, where H is the Shannon–Wiener index of heterogeneity calculated as H = –Spi ln(pi) (pi is the proportion of species i in a sampling plot), Hmax is the maximum value of H for the number, S, of species present (Hmax = ln S) (Krebs, 1989). Independence of species richness is considered a main requirement for evenness indices (Smith & Wilson, 1996). The J index in our study responded to this criterion (by the Pearson correlation test). To analyse spatiotemporal variations in the micromycete community structure, five major groupings were chosen: Penicillium spp., Aspergillus spp., Ascomycota (species with sexual stage appearing on the plates), Zygomycota, and melanin-containing microfungi. The contribution of each group to mycobiota structure was estimated as their relative abundance (percentage) in the Shannon index value. This characteristic was chosen instead of direct density because it is logarithmic, thus preventing an over-estimation of heavily sporulating species.
To evaluate similarity between local micromycete communities, the following similarity indices were used: (i) the Jaccard coefficient = a/(a + b + c), where a is number of species common to both localities, b is number of species found only in soil of the first locality, and c is number of species found only in soil of the second locality; (ii) the percentage similarity = S minimum (p1i, p2i), where p1i is percentage of species i in soil of the first locality, and p2i is percentage of species i in soil of the second locality (Krebs, 1989); (iii) the Sorensen–Czekanowski coefficient = 2a/(b + c), modified by Mirchink (1988), where a is S minimal frequency of occurrence of common species, b is S frequency of occurrence of all species in soil of the first locality, and c is S frequency of occurrence of all species in soil of the second locality. These three indices characterize similarity of communities from different sides: (i) species presence/absence (ii) species relative abundances, and (iii) species frequencies of occurrence. To analyse the slope community organization, complexes of typical micromycete species were determined based on spatial and temporal frequencies of occurrence (frequency) for each species (Mirchink, 1988): 100% temporal and ≥75% spatial frequency for dominant species; ≥75% temporal and ≥50% spatial frequency for frequent species; ≥50% temporal and ≥30% spatial frequency for rare species. Statistical analysis was conducted using Statistica (StatSoft, 1996). We used the non-parametric Wilcoxon matched pair test and c2 test to compare data from different localities on biodiversity characteristics and relative abundance of micromycete groupings. The correlation of these data with soil abiotic parameters was estimated by the Pearson correlation test. A log-linear model was employed to compare proportion of ascomycetes displaying only a sexual state in different localities and seasons.
RESULTS SOIL
ABIOTIC CHARACTERISTICS
Soil moisture displayed both spatial and seasonal variations. The spatial variation was more pronounced in the spring and the seasonal variation was more pronounced in the AS soil (Table 1). The EC II soil is alkaline, with maximal and minimal pH in the AS sunny and the VB localities, respectively. Organic matter content was highest in the VB soil and lowest in the soil of the AS sunny locality.
SPECIES
COMPOSITION
The soil mycobiota of EC II comprised 192 species, from Zygomycota (nine species), Ascomycota (13 species) and mitosporic fungi (170 species). The species
© 2003 The Linnean Society of London, Biological Journal of the Linnean Society, 2003, 78, 527–539
531
SOIL FUNGI DISTRIBUTION IN ‘EVOLUTION CANYON’ II, ISRAEL belonged to 60 genera, the most prominent being Penicillium (47 species), Aspergillus (27), Acremonium (11), Phoma (eight), Trichoderma and Fusarium (six each). RICHNESS AND EVENNESS
We isolated 149, 108, and 78 species from the AS, ES and VB, respectively, with 107 from AS sunny and 117 from AS shady. Twenty-five species (13%) were common to all of these four localities, 44 (23%) to AS, ES, and VB, and 61 (32%) to AS and ES. Among isolated strains, 3.8% were sterile (6.4%, 5.8%, 2.2% and 0.4% from AS sunny, AS shady, ES, and VB, respectively). Species richness was significantly higher on the AS 2 = 21.7, P < 0.001). Although we than on the ES (c(1) examined an unequal number of samples from the AS (48) and ES (36), it should not significantly influence interslope difference in species richness because the number of species in each habitat of the AS, isolated from 24 samples, was the same (sunny) or even higher (shady) than on the ES. A comparison of the studied micro-environments exposed the intra- and interslope differences in biodiversity level (Table 2). The AS sunny localities were characterized by the highest values of evenness. This was observed in each season (except winter) as well as across the year (Wilcoxon test, P < 0.05). By contrast, the shady ES microfungal community was the least heterogeneous and even. In seasonal terms, the slope communities had the highest species richness in the spring and the lowest in the autumn. In the winter, both slopes had the lowest equitability.
STRUCTURE
OF LOCAL MICROMYCETE COMMUNITIES
Spatial pattern The composition of local micromycete communities demonstrated prominent spatial (both inter- and
60
31
22
50
% of Shannon index
SPECIES
intraslope) differences. As Figure 2 shows, the soil of AS sunny habitats is characterized by dominance of melanin-containing micromycetes (36% of species number, 30.5% relative abundance). The contributions of Aspergillus and Penicillium species were
40 27 41 30
20
16
36 17 23
13
14
18
10
10 5
7 5
3
7 7
AS, sunny
AS, shady
ES, shady
1
7
0 VB, shady
Figure 2. Spatial dynamics of relative abundance of main micromycete groupings in the local communities of ‘Evolution Canyon’ II. The numbers above the bars represent species number in each grouping. Wilcoxon test (N = 12 in each case): (1) AS, sunny vs. shady: Penicillium spp. (Pen), P < 0.01; Aspergillus spp. (A) NS; Melanincontaining micromycetes (MC), P < 0.05; Ascomycota (ASC), NS; Zygomycota (Z). (2) AS sunny vs. ES: Pen, P < 0.01; A, P
