Abstract

Arbuscular mycorrhizal (AM) fungi in a chronosequence of 5–42-year-old Caragana korshinskii plantations in the semi-arid Loess Plateau region of northwestern China were investigated. AM fungi colonization, spore diversity and PCR-denatured gradient gel electrophoresis-based AM fungal SSU rRNA gene sequences were analyzed. AM fungi colonization [measured as the percent of root length (%RLC), vesicular (%VC) and arbuscular (%AC) colonization] and spore density were significantly correlated with sampling month, but not with plant age, except for %RLC. The percent of vesicular colonization was negatively correlated with soil total nitrogen and organic carbon, and spore density was negatively correlated with soil moisture and available phosphorus. Ten distinguishable AM fungal spore morphotypes, nine Glomus and one Scutellospora species, were found. Nine AM fungal Glomus phylotypes were identified by sequencing, but at each sampling time only four to six AM fungal phylotypes were detected. The AM fungal community was significantly seasonal, whereas the AM fungal species richness did not increase with plantation age. A significant change in AM fungal colonization and community composition over an annual cycle was observed in this study, and our results suggest that the changes of AM are the product of the interaction between host phenology, soil characteristics and habitat. Understanding these interactions is essential if habitat restoration is to be effective.

Introduction

Arbuscular mycorrhizal (AM) fungi (phylum Glomeromycota, Schüßleret al., 2001) can form mutualistic relationships with the roots of most plant families (Smith & Smith, 1997). It is generally accepted that AM fungi can promote plant nutrient uptake, especially phosphate (Smithet al., 2003), reduce the damage of root pathogens (Borowicz, 2001) and protect plants against several types of stress (Al-Karaki, 1998; Arocaet al., 2007). Additionally, as abundant and widespread organisms in terrestrial ecosystems, AM fungi can influence ecosystem processes through many direct and indirect pathways (Rillig, 2004). Although AM fungi have such physiological benefits and ecological importance, the actual number of AM fungal species is still unknown. Investigations that have estimated the biodiversity and distribution of AM fungi in various ecosystems showed that communities of AM fungi occurring in different ecosystems have different species composition (Öpiket al., 2006), and the distributions of AM fungi may be products of environment, interspecific competition and regional spatial dynamics (Lekberget al., 2007).

The composition of AM fungal communities may play an important role in the maintenance of plant biodiversity (van der Heijdenet al., 1998); conversely, plant community can also affect AM fungal diversity and community composition (Johnsonet al., 2003, 2005). Moreover, some evidence indicates that the number of AM fungi and plant species are positively correlated and might have similar responses to environmental variations (e.g. Beveret al., 2001; Hartet al., 2001). We hypothesize that the composition of an AM fungal community will change in response to the changes in plant community composition, for example during succession. Successional patterns in species of ectomycorrhizal fungi are common in nature (Smith & Read, 1997), but in the AM fungi, a general pattern has been hard to identify (c.f. Benjaminet al., 1989; Johnsonet al., 1991). Therefore, under field conditions, AM fungal community succession is probably complex; it may not only depend on the plant and field succession but also on the pools of AM fungal species in the region and the dispersal ability of AM fungi from other regions. To understand the relationship between AM fungi and plant community succession, we chose a chronosequence of plantations in the semi-arid loess and gully region of northwestern China to study the AM fungal succession in monoculture plantations.

The Loess Plateau region of northwestern China, about 45 million hectares in size, is a region where the natural vegetation has been seriously degraded by a 1000 years of human overexploitation. Since the middle of the 20th century, many vegetation restoration programs have been implemented in this region. Caragana korshinskii Kom, a native leguminous shrub with extensive root systems and high stress tolerance, is the species most widely planted in these programs for stabilization of eroded land and improving vegetation coverage. The potential that AM fungi have to influence the growth and performance of this important plant species requires an understanding of AM fungal community development if habitat restoration is to be effective.

We predicted that the AM fungal species diversity in C. korshinskii roots would increase with increasing plantation age, and that AM fungal colonization, spore density and community composition would change over season. To test this hypothesis, we used microscopy and PCR-denatured gradient gel electrophoresis (DGGE) techniques to investigate the AM fungal patterns in a chronosequence of C. korshinskii plantations at three sampling times to address the following three questions: (1) what is the AM fungal diversity in the C. korshinskii plantations? (2) What are the seasonal dynamics of AM fungal colonization, spore density and AM fungal community composition? (3) Does AM fungal diversity increase with plantation age?

Materials and methods

Sampling sites and procedures

This experiment was carried out in Yuzhong county of Gansu Province in northwestern China (N36°02′, E104°24′; 2400 m above sea level). Long-term climate data were available from the nearby Loess Plateau Experimental Station of Lanzhou University. This region is located in a medium temperate semi-arid climate with an annual mean air temperature of 6.2 °C, annual mean precipitation of 300 mm and plant vegetation belonging to the typical desert grassland region of western Loess Plateau of China, where the soil type is yellow-cultivated loessial soil. We chose four C. korshinskii plantations (protected by the County Forestry Administration) that have been established for 5, 13, 20 and 42 years, and the distance between each site is no more than 10 km. Around the plantations, the extensive loess plateau is highly eroded, and most of the region has little vegetation cover. Caragana korshinskii individuals were planted using contour planting (terracing) from the top to the bottom of the gully. Vegetation coverage, plant species coverage and species richness were estimated by Sutherland's method (1996). Vegetation coverage of each site was remarkably different and the plant species richness tended to increase with plantation age; more details of the sampling sites are presented inTable 1.

1

Description of the sampling sites of Caragana korshinskii plantations

Sites (years) Slope gradient Slope orientation Vegetation coverage Dominance of plant species Species richness 
45° c. 15% Leymus secalinus (c. 70%) 17 
   Stipa bungeana (c. 10%)  
   C. korshinskii (c. 1%)  
13 35° c. 30% C. korshinskii(c. 60%) 17 
   S. bungeana (c. 10%)  
   Artemisia capillaries (c. 5%)  
20 30–35° E15°N c. 50% C. korshinskii (c. 50%) 28 
   S. bungeana (c. 10%)  
42 6° E30°N c. 85% C. korshinskii (c. 80%) 18 
   S. bungeana (c. 5%)  
Sites (years) Slope gradient Slope orientation Vegetation coverage Dominance of plant species Species richness 
45° c. 15% Leymus secalinus (c. 70%) 17 
   Stipa bungeana (c. 10%)  
   C. korshinskii (c. 1%)  
13 35° c. 30% C. korshinskii(c. 60%) 17 
   S. bungeana (c. 10%)  
   Artemisia capillaries (c. 5%)  
20 30–35° E15°N c. 50% C. korshinskii (c. 50%) 28 
   S. bungeana (c. 10%)  
42 6° E30°N c. 85% C. korshinskii (c. 80%) 18 
   S. bungeana (c. 5%)  
*

E, east; N, north; E15°N represents east by north 15°.

Dominance of plant species was estimated by the percent of coverage of each plant species.

All samples were collected in April, July and October 2006. At each site, three layers were randomly selected from the top to the bottom of the gully (the same layers were selected at each sampling time). In each layer, fine roots and rhizosphere soils (depth of 15–45 cm) were randomly sampled from each of three C. korshinskii individuals by tracing three laterals from the taproot. The sampled roots were mixed, washed carefully with tap water and stored in 40% ethanol at 4 °C until processed. Soil samples were stored at 4 °C in sealed bags and transported to the laboratory for spore isolation and soil analysis. Soil moisture content was measured gravimetrically on each sampling date in the Loess Plateau Experimental Station of Lanzhou University.

AM fungal DNA extraction from roots

From each sample, about 30 root fragments, 1 cm in length, were randomly taken and washed four times with sterilized distilled water. Root fragments were dried at room temperature, pulverized with liquid nitrogen and about 0.2 g of ground root was placed into 2-mL reaction tubes. Eight microliters of β-mercaptoethanol, 8 μL protease K (10%; w/v) and 1000 μL CTAB buffer [2% (w/v) cetylammoniumbromide, 20 mM EDTA, 100 mM Tris-HCl, 1.4 M NaCl] were added and shaken (225 r.p.m., 37 °C for 15 min). Two hundred microliters of sodium dodecyl sulfate (SDS) (20%; w/v) was then added to the reaction tubes and incubated at 65 °C for 60 min. Thereafter, samples were centrifuged (3300 g) for 10 min, the upper phase was extracted with an equal volume of chloroform : isoamyl alcohol (24 : 1) and mixed before a 10-min centrifugation at 6000 g. The upper phase was collected, precipitated with 750 μL of isopropanol at −20 °C for at least 60 min and centrifuged for 30 min (16 000 g). The supernatant was removed and the pellet was washed with 200 μL 4 °C 70% ethanol, centrifuged for 5 min at 16 000 g, dried and resuspended in 30 μL TE (10 mM Tris-HCl, 1 mM EDTA, pH 8.0).

Roots staining and AM fungal colonization assessment

Roots were randomly selected from the root samples, cut into c. 1-cm segments, clarified in 10% KOH (w/v) at 121 °C for 40 min, acidified in 2% HCl (v/v) for 60 min and stained in 0.05% trypan blue (w/v) in lactoglycerol (lactic acid, glycerol and water, 1 : 1 : 1) at 90 °C for 15 min. About 60 root fragments were randomly selected from a sample, mounted in glycerin on microscope slides with coverslips and examined under a compound microscope. Percent root length colonization (%RLC), arbuscular colonization (%AC) and vesicular colonization (%VC) were quantified using the magnified intersection method (McGonigleet al., 1990).

AM fungal spore isolation and identification

AM fungal spores were separated from each air-dried soil sample by wet-sieving (a pair of sieves: 750- and 38-μm mesh) and sucrose centrifugation (Brundrettet al., 1994). Spores were clustered according to spore morphology and color using a dissecting microscope and the spore densities were counted. For each spore morphotype, permanent slides were mounted with PVLG (polyvinyl alcohol, lactic acid, glycerin) and PVLG+Melzer's reagent (1 : 1, v/v) and then classified using current taxonomic criteria and information published by INVAM (http://invam.caf.wvu.edu).

Nested PCR and DGGE analysis

The PCR-DGGE approach has proven to be a rapid and cost-effective strategy for describing the intraradical AM fungal community (Kowalchuket al., 2002; Öpiket al., 2003; Santoset al., 2006). Although advanced molecular methods make it feasible for analysis of AM fungal communities, the extent of AM fungal root colonization cannot be estimated reliably using these methods. The extracted DNA from roots was subjected to nested PCR. The first DNA amplification was performed with the universal fungal primers GeoA2 and Geo11 to amplify a c. 1.8-kb fragment of the SSU rRNA gene (Schwarzott & Schüßler, 2001). PCR was carried out in a 25 μL reaction volume with 2 μL template (c. 10 ng μL−1) and 0.5 μM of each primer using the Pfu PCR mastermix system (Tiangen, Beijing, China) according to the manufacturer's instructions, and using the following cycling conditions: 94 °C for 2 min; 30 × (94 °C for 30 s; 63 °C for 1 min and 72 °C for 2.5 min); and 72 °C for 10 min. PCR products were examined on a 1.5% (w/v) agarose gel with ethidium bromide staining to confirm product integrity. The product of the first amplification was diluted 1 : 100 and 2 μL of this dilution used as a template for the second PCR. The second amplification used the specific primers AM1 (Helgasonet al., 1998) and GC-NS31 (Kowalchuket al., 2002), and identical reaction conditions were used with the following program: 94 °C for 2 min; 30 × (94 °C for 30 s; 67 °C for 1 min and 72 °C for 1 min); and 72 °C for 10 min. The nested PCR products were then examined on an agarose gel as described above.

All DGGE analyses were performed using the Dcode Universal Mutation Detection System (Bio-Rad, Hercules, CA) using the method described by Kowalchuket al. (2002) modified as follows: 6% (w/v) polyacrylamide gel (37.5 : 1 acrylamide : bis-acrylamide, 1 × TAE buffer, 1.5 mm thick, 16 cm × 16 cm) containing a linear denaturing gradient from 20% to 35%. Electrophoresis was run for 6 h at 160 V in 1 × TAE buffer at a constant temperature of 60 °C. Gels were stained using silver staining (Sanguinettiet al., 1994), and the gel images were captured digitally using a scanner (Canon, Japan).

Cloning, sequencing and sequence analysis

Dominant bands were excised from the gels and washed twice with sterilized distilled water. The excised piece was mashed and incubated in 30 μL ddH2O at 4 °C overnight. One microliter gel-eluted DNA was reamplified with primers AM1/NS31 (Helgasonet al., 1998) as the second PCR described above, except that the annealing temperature was 60 °C. DNA fragments of expected length (c. 550 bp) were purified using the Gel and PCR Clean-up System (Promega, Madison, WI) according to the manufacturer's instructions. Purified PCR products were digested with the restriction enzymes HinfI and AluI (Takara, Japan) to confirm that bands with the same DGGE gel mobility contained the same sequence. All bands in the same mobility group had the same restriction fragment length polymorphism (RFLP), although some bands with different mobilities shared RFLPs. One PCR product from each RFLP/mobility group was selected randomly for ligation into pGEM-T (Promega, Madison, WI) and cloned into Escherichia coli DH5α according to the manufacturer's recommended protocol.

The presence of inserts of the expected size was confirmed by PCR using the primers AM1/NS31 (PCR conditions as described above) and restriction analysis with the enzyme EcoRI (Takara). Reconfirmed clones were extracted using the TIANprep Mini Plasmid Kit (Tiangen, Beijing, China) and sequenced by Generay Biotech Company (Shanghai, China) using the M13 F/R primer. Sequences were edited and assembled using the contigexpress program of Vector NTI Suite 6.0 (InforMax, MD). All sequences were compared with public databases using the blast (Altschulet al., 1997), and were screened for possible chimeric origin using the online chimera detection program (http://rdp8.cme.msu.edu/html/analyses.html). The sequences obtained in this study, the 14 most closely related sequences, some other AM fungal sequences and two outgroup taxa [Endogone pisiformis (X58724), Mortierella polycephala (X89436)] were used for phylogenetic analysis. Sequences were edited manually to remove ambiguous nucleotides and gaps using bioedit software, and then aligned using the program clustal x, and the neighbor-joining tree was constructed using the mega version 3.1 with the Kimura two-parameter model.

Soil properties analysis

Soil organic carbon and total nitrogen concentrations were analyzed using the CHNS-analyser system (Elementar Vario EL, Elementar Analysensysteme GmbH, Hanau, Germany) with the burning method at 450 and 1250 °C, respectively. Soil available phosphorus was extracted using the Olsen method, the total phosphorus content was determined after digesting with perchloric acid and the phosphorus concentrations were measured colorimetrically (Nanjing Agricultural University, 1996).

Statistical analysis

Statistical analyses were performed using the spss 13.0 (SPSS Inc., IL). Before analysis, percentage variables (including data of %RLC, %AC, %VC, total nitrogen, organic carbon and soil moisture) were arcsine square root transformed and other variables (including data of spore density, total phosphorus and available phosphorus) were ln transformed. A principal component analysis (PCA) was carried out using the soil characters in order to generate fewer compound variables (principal components) that characterize the soil types. The relationships between these compound variables and AM fungal colonization variables were then tested using a Pearson correlation. The effects of site and month on percent of AM fungal colonization and spore density were analyzed using one-way and two-way anova. AM fungal community composition (based on phylotypes) was calculated on the basis of the fungal sequence groups' presence or absence in a root system (Öpiket al., 2003); all data of AM fungal communities species composition were ordinated separately using the PCA, implemented in canoco version 4.51.

Results

Ordination of edaphic characters

The ordination was successful in explaining the variation in soil characteristics among samples: 44.7% and 25.5% of the variation was explained by the first (PC1) and the second (PC2) principal component, respectively (Fig. 1). The loadings of the axes show that PC1 reflects the total nitrogen, organic carbon and total phosphorus status of the soil and PC2 reflects the available phosphorus and soil moisture. Only vesicle colonization was significantly negatively correlated with PC1, suggesting that as soil nitrogen and organic carbon increase, vesicle colonization decreases. Spore density was significantly negatively correlated with PC2, suggesting that as the soils dry and available phosphorus declines, sporulation increases (Table 2).

1

PCA of soil characteristics (including soil total nitrogen, organic carbon, carbon/nitrogen, total phosphorus, available phosphorus and soil moisture) marked by sampling site and time. The points represent the mean of PC1 and PC2 scores of each site in each month. Bars represent the SEs. Black closed symbols, gray closed symbols and open symbols represent the April, July and October samples, respectively.

1

PCA of soil characteristics (including soil total nitrogen, organic carbon, carbon/nitrogen, total phosphorus, available phosphorus and soil moisture) marked by sampling site and time. The points represent the mean of PC1 and PC2 scores of each site in each month. Bars represent the SEs. Black closed symbols, gray closed symbols and open symbols represent the April, July and October samples, respectively.

2

Correlations between percent of AM fungal colonization, spores density with month, plant age and two PCA extracted components of soil properties (PC1 and PC2)

 Month Plant age PC1 PC2 
% RLC −0.564 −0.552 +0.174 +0.22 
% AC +0.472 −0.044 +0.094 −0.135 
% VC +0.483 +0.294 −0.423 −0.055 
Spore density +0.423 +0.136 +0.299 −0.54 
 Month Plant age PC1 PC2 
% RLC −0.564 −0.552 +0.174 +0.22 
% AC +0.472 −0.044 +0.094 −0.135 
% VC +0.483 +0.294 −0.423 −0.055 
Spore density +0.423 +0.136 +0.299 −0.54 

Data are Pearson's correlation coefficients.

*

P≤0.01;

**

P≤0.001.

%RLC, percent of root length colonization; %AC, percent of arbuscular colonization; %VC, percent of vesicular colonization.

AM fungal colonization

AM fungal colonization in the roots of C. korshinskii was intense and the mean %RLC varied from 80.3% to 96% in all samples. Statistical analysis of the %RLC showed a significant effect of sampling site (F=19.026, P<0.001) and of sampling time (F=7.954, P<0.01); furthermore, site and time had a significant interaction on %RLC (F=6.096, P<0.01). Combining all months or all sites together, the 5-year site (mean=94.3%) and April (mean=92.4%) had the highest %RLC, respectively (Fig. 2a). Time of sampling (month) was significantly correlated with %RLC (Table 2), and there was an obviously declining trend in all sites from spring to autumn. In each month, %RLC appeared to decline during the C. korshinskii succession, with the exception of the 13-year site.

2

Percent of AM fungal colonization and spore density of different samples from four sites in each month (bars represent SDs). (a) %RLC, percent of root length colonization; (b) %AC, percent of arbuscular colonization; (c) %VC, percent of vesicular colonization; (d) spore density. Significant differences between columns of the same cluster (plantation age class) were determined using Fisher's Least Significant Difference at the 5% level after one-way anova and indicated by dissimilar letters above the bars.

2

Percent of AM fungal colonization and spore density of different samples from four sites in each month (bars represent SDs). (a) %RLC, percent of root length colonization; (b) %AC, percent of arbuscular colonization; (c) %VC, percent of vesicular colonization; (d) spore density. Significant differences between columns of the same cluster (plantation age class) were determined using Fisher's Least Significant Difference at the 5% level after one-way anova and indicated by dissimilar letters above the bars.

Arbuscules were occasionally present in the observed root segments. The average of %AC was 5.96%, with the highest colonization (9.39%) in the 5-year site of October and the lowest (3.43%) in the 13-year site of April (Fig. 2b). The %AC that had a similar seasonal trend in each site had a significant correlation with sampling time (Table 2). Seasonal variation in %VC was also present in all sites, whereas there were no distinctly seasonal differences in the 5- and 13-year sites (Fig. 2c). In overall analysis, the mean of %VC was 8.78%, the highest value of each site was present in July and the 42-year site had the highest total %VC (9.47%, 3 months combined).

Spore density and spore identification

AM fungal spore densities in rhizosphere soil of C. korshinskii ranged from 3.3 to 30 per 10 g soil (mean=12.7 per 10 g soil). Spore density was correlated significantly with soil moisture and soil phosphorus, demonstrated by the ordination of the soil variables (Table 2). The spore density had a significantly seasonal variation in each site, with the highest and the lowest spore densities presented in October and July, respectively (Fig. 2d). Spore densities of different sites showed no significant difference in each sampling time, except for October (data not shown). Although not significant, the 5-year site (mean=14.2 per 10 g soil) had a higher spore density than others in three sampling times together.

AM fungal spore morphotypes of different samples were similar. Glomus and Scutellospora were the only two genera observed. A total of 10 distinguishable AM fungal spore morphotypes were found in all soil samples. Seven spore morphotypes could clearly be identified, six grouped in the genus Glomus and one in Scutellospora. The spores of the Glomus species resembled Glomus intraradices, Glomus etunicatum, Glomus geospororum, Glomus constrictum, Glomus coronatum and Glomus versiforme, and the Scutellospora species was Scutellospora calospora. The remaining three spore morphotypes could not be fully identified, but all of them belonged to the genus Glomus. Accurate assessment of the number of each morphotype was difficult, but classification was straightforward at the genus level. The proportional distribution of Glomus and Scutellospora varied in each sampling site, but the S. calospora species were more abundant in the 20- and 42-year sites (Fig. 3).

3

Proportional distribution of the genus of Glomus and Scutellospora spores in different sites of each month.

3

Proportional distribution of the genus of Glomus and Scutellospora spores in different sites of each month.

PCR-DGGE and phylogenetic analysis

The expected DNA fragments were amplified via the first and second PCR from all samples (36 in total), which produced c. 1.8-kb and 590-bp DNA fragments, respectively. One of the secondary PCRs of the 42-year samples in July yielded very little product, resulting in failure of the DGGE; therefore, this sample was excluded from further analysis. DGGE profiles varied among samples from different months, but within each month, patterns derived from different sites were similar (Fig. 4). The signal intensities of DNA bands were variable; some bands showed a very strong signal (e.g. Ap8, Ju2, Oc2) while some were weak (e.g. Ap6, Ju4, and Oc5).

4

DGGE profiles of partial SSU rRNA gene sequences of AM fungi colonizing roots of Caragana korshinskii. Ap1–Ap10, Ju1–Ju6 and Oc1–Oc10 represent bands from samples of April, July and October, respectively. Glo-A–Glo-I represent the sequence groups to which the bands belong.

4

DGGE profiles of partial SSU rRNA gene sequences of AM fungi colonizing roots of Caragana korshinskii. Ap1–Ap10, Ju1–Ju6 and Oc1–Oc10 represent bands from samples of April, July and October, respectively. Glo-A–Glo-I represent the sequence groups to which the bands belong.

Six to ten DNA fragments with distinct mobility were detected from each sample. From each distinct band position (26 in total), prominent bands (158 in total) were excised from the gels, reamplified and screened using RFLP typing as described inMaterials and methods. A total of 26 partial SSU rRNA gene sequences were thus obtained and analyzed using blast to reveal similarities with published sequences. The results of blast showed that 23 sequences had high homology to members of Glomeromycota, while two sequences were more related to ascomycetes and one sequence (Fig. 4, Oc7) appeared to be chimeric. Phylogenetic analysis using our sequences and those of the closest blast hits revealed nine sequence groups (Fig. 5), of which all were grouped in the putative genus of Glomus. Three sequence groups Glo-A, Glo-B and Glo-E showed high similarity to sequences of G. intraradices (bootstrap value 62%), G. fasciculatum (67%) and G. caledonium (100%), respectively; the remaining six groups were related to root-derived, but taxonomically unknown sequence types: Glo-C (related to the sequence of GenBank accession nos. AY330278, bootstrap value 55%), Glo-D (AJ563895, 94%), Glo-F (AJ854089, 76%), Glo-G (AJ418868, 99%), Glo-H (EF177590, 75%) and Glo-I (AY129582, 99%). All sequences of clade Glo-C held over 98% identity to the sequence of AY330278, whereas these sequences were distinctly diverged and our sequences clustered into a monophyletic clade. Sequence type Oc9 distantly diverged from most sequences obtained from this study and showed the closest relationship (96% identity) with the sequence type of Glo45 from the roots of Luehea seedings in a tropical forest (Husbandet al., 2002b). Additionally, the non-AM fungal sequence cluster containing sequences of Ap9 and Ap10 was also presented in the tree, confirming that the NS31/AM1 PCR does not exclude all non-AM fungal sequences.

5

Neighbor-joining phylogenetic tree inferred from partial SSU rRNA gene sequences of all identified AM fungi in Caragana korshinskii roots and referenced sequences. Bootstrap value (1000 replicates) >55% are shown. Sequence groups (Glo-A, etc.) identify distinct clusters of sequences with similarity >98% (except Glo-I, 96%). All sequences have been submitted in the GenBank database under the accession numbers EU350045EU350069.

5

Neighbor-joining phylogenetic tree inferred from partial SSU rRNA gene sequences of all identified AM fungi in Caragana korshinskii roots and referenced sequences. Bootstrap value (1000 replicates) >55% are shown. Sequence groups (Glo-A, etc.) identify distinct clusters of sequences with similarity >98% (except Glo-I, 96%). All sequences have been submitted in the GenBank database under the accession numbers EU350045EU350069.

AM fungal community

The AM fungal phylotypes colonizing C. korshinskii roots showed little difference among four sites at the same sampling time, whereas dramatic differences occurred in different months (Table 3). In April, the AM fungal population was heavily dominated by the Glo-E type, and thereafter, this type was not detected; Glo-C was a subdominant type in April, was absent in July and occasionally reoccurred in October. By contrast, many types including Glo-A, Glo-F, Glo-G and Glo-I, which were absent in C. korshinskii roots in April, showed various frequencies in July and October. As a whole, the frequency of Glo-B and Glo-H types increased with sampling time at a moderate and a low frequency, respectively.

3

AM fungal phylotypes detected from samples of different sites in each month

Months Sites (years) AM fungal phylotypes 
Glo-A Glo-B Glo-C Glo-D Glo-E Glo-F Glo-G Glo-H Glo-I 
April  ▪ ▪ ▪ ▪   ▪  
13  ▪ ▪  ▪   ▪  
20  ▪ ▪  ▪   ▪  
42  ▪ ▪ ▪ ▪   ▪  
July ▪ ▪    ▪  ▪  
13 ▪ ▪    ▪  ▪  
20 ▪ ▪    ▪  ▪  
42 ▪ ▪    ▪  ▪  
October ▪ ▪ ▪    ▪ ▪ ▪ 
13 ▪ ▪     ▪ ▪ ▪ 
20 ▪ ▪ ▪    ▪ ▪ ▪ 
42 ▪ ▪     ▪ ▪ ▪ 
Months Sites (years) AM fungal phylotypes 
Glo-A Glo-B Glo-C Glo-D Glo-E Glo-F Glo-G Glo-H Glo-I 
April  ▪ ▪ ▪ ▪   ▪  
13  ▪ ▪  ▪   ▪  
20  ▪ ▪  ▪   ▪  
42  ▪ ▪ ▪ ▪   ▪  
July ▪ ▪    ▪  ▪  
13 ▪ ▪    ▪  ▪  
20 ▪ ▪    ▪  ▪  
42 ▪ ▪    ▪  ▪  
October ▪ ▪ ▪    ▪ ▪ ▪ 
13 ▪ ▪     ▪ ▪ ▪ 
20 ▪ ▪ ▪    ▪ ▪ ▪ 
42 ▪ ▪     ▪ ▪ ▪ 

▪, presence.

PCA of the AM fungal phylotype compositions showed that samples from each sampling month were clustered together (Fig. 6), and indicated a significant difference in the AM fungal population colonizing C. korshinskii roots presented through the season. However, the total variation among sites was much smaller and suggests that the species composition of AM the fungal community in roots of C. korshinskii was unrelated to plantation age.

6

PCA pattern of AM fungal community composition marked by sampling time. Scatterplots (35 in total) represent AM fungal communities described in terms of phylotypes.

6

PCA pattern of AM fungal community composition marked by sampling time. Scatterplots (35 in total) represent AM fungal communities described in terms of phylotypes.

Discussion

Change of AM colonization and spore density

The AM fungal colonization and spore density were significantly correlated with the sampling month (Table 2), suggesting a seasonal pattern of AM fungal colonization and spore density in all successional stages of C. korshinskii. The highest and lowest %RLC were found at the beginning (April) and the end (October) of the growing season, a result that is consistent with Bohreret al. (2004), who observed a similar seasonal trend in a wetland ecosystem. There was a seasonal increase in %AC from April to October. In Arum-types, arbuscules are generally considered as the primary site for phosphate exchange between the plant and the fungus (Smith & Smith, 1997), and the variation of %AC corresponds to the dynamics of plant demand for phosphate (Sanders & Fitter, 1992).

Many studies have shown that seasonal variations in AM fungal colonization are related to host plant phenological events (Lugoet al., 2003; Bohreret al., 2004), and our analysis shows that the changes in %RLC and %AC are not significantly correlated with edaphic characters (Table 2). Thus, we suggest that the seasonality of %RLC and %AC is a response to the plant physiological demands and changes during plant growth.

Although plant phenology influences AM fungal colonization and the whole AM fungal lifecycle, soil properties also have a significant effect on fungal biology. The correlation analysis indicated that %VC and spore density were significantly negatively correlated with PC1 and PC2, respectively (Table 2), and suggests that the variation in the soil environment may result in variation of AM colonization and spore density. Spores are important survival organs for AM fungi, and increased production as the soil dries is entirely consistent with this (Andersonet al., 1984). Vesicles are considered to be an important storage structure: decreased vesicle colonization with increased soil nitrogen and organic carbon suggests a storage response of AM fungi to nitrogen and carbon deposition, and this result is consistent with a recent study in which AM fungal storage structures declined with simulated nitrogen deposition (van Diepenet al., 2007).

It is not possible from these data to distinguish between a direct response of the AM fungi to the environmental cue or an indirect response due to decreased carbon supply from the plant hosts. Nevertheless, both the status of the host (physiology and phenology) and the edaphic variation can influence the AM fungal colonization and spore density. The shift of AM fungal community composition (seeFig. 6) may also result in such a seasonal dynamics due to variation in life-history characters of the different AM fungal species (Beveret al., 2001).

AM fungal diversity

At three time points, a total of nine phylotypes all belonging to the genus Glomus were detected in C. korshinskii roots. At each sampling time, however, only four to six AM fungal phylotypes were detected. Many previous studies have used the AM1/NS31 primer pair to estimate AM fungal diversity (e.g. Helgasonet al., 1998, 1999, 2002; Daniellet al., 2001; Husbandet al., 2002a, b; Vandenkoornhuyseet al., 2002), and the diversity estimate here is markedly lower. The AM1/NS31 primer pair is known to sample only a subset of the Glomeromycota, and it is possible that the RFLP screen before cloning missed some sequence variants, and so it cannot be ruled out that the diversity is an underestimate. However, inspection of rarefied species accumulation curves showed that species accumulation was asymptotic (data not shown), suggesting that the majority of sequence types that could be identified were found. The genetic diversity is similar to that found in agricultural fields using a similar methodology, and so the molecular data obtained in this study are likely to reflect a genuinely low diversity of AM fungi colonizing C. korshinskii roots (Daniellet al., 2001).

AM fungal spores collected from rhizosphere soils of C. korshinskii were classified into 10 morphotypes, and all but one of the morphotypes belong to the genus Glomus. Although a dominant morphotype in spores, S. calospora was not detected in the roots, indicating that AM fungal community structure based on spores may not reflect those in roots. Taken as a whole, both morphological and molecular data suggest that AM fungal diversity in C. korshinskii plantations was relatively low. In addition, molecular analysis strongly indicated that an extremely low AM fungal diversity presented in C. korshinskii roots in any given time point.

Low AM fungal diversity appears to be characteristic of habitats that experience periods of high environmental stress, for example semi-arid mediterranean scrub (Ferrolet al., 2004), arable crops (Daniellet al., 2001) and in this study. It is difficult to separate this from the effect that plant diversity has on AM fungal community as such habitats are often also species poor. As we might predict, in tropical forests which have high plant diversity and members of AM fungal types potentially equating to a fifth of the world's described AM fungal species (Husbandet al., 2002b), the average AM fungal taxa per plant species is 18.2 (Öpiket al., 2006). Recently, however, the AM fungal community with the highest diversity reported to date was found in a boreal forest (Öpiket al., 2008). In the arid habitat studies here, the effect of plant community diversity could play a part, but it is clear that an abiotic environment plays a significant role as a determinant of community structure.

Dynamics of AM fungal community

Our results reveal a significant change in the AM fungal community through season. At each sampling time, there was a specific AM fungal community composition: some dominant AM fungal types were replaced by others and some disappeared (e.g. Glo-E type). This pattern is consistent with the results of previous studies in different habitats. Merryweather & Fitter (1998) found significant seasonal shifts in root colonization in a temperate woodland species, and the pattern was confirmed in a subsequent season by DNA sequencing (Helgasonet al., 1999). Temporal shifts have also been identified in grasslands (Vandenkoornhuyseet al., 2002; Santos-Gonzálezet al., 2007). It has been shown that different AM fungi vary in their response to changes in an abiotic environment (Beveret al., 2001), and so the dominant AM fungal species is likely to change as the environment changes over time (Husbandet al., 2002b); furthermore, such temporal changes of AM fungal community may be related to host phenology (Merryweather & Fitter, 1998; Eomet al., 2000), resulting in changes in the AM fungi colonizing the roots at different stages of plant growth. Seasonal changes of AM fungal community are likely to be an interaction between the abiotic environment and the host phenology.

In contrast to the significant seasonal change, surprisingly, we observed few or no changes in the species composition of the AM fungal communities across the C. korshinskii plantation chronosequence at the same time point (Table 3), suggesting that the AM fungal population colonizing C. korshinskii roots was stable regardless of the change in plant age. This result is in contrast to our first hypothesis, but it is consistent with an earlier study by Benjaminet al. (1989), who showed that AM fungal species compositions of little bluestem (Schizachyrium scoparium) showed no change in an Illinois prairie-forest chronosequence. The most likely explanation for our observations is that the abiotic environment and lack of host plants select for a particular group of AM fungi. Further study of the AM fungal community in the alternative hosts and estimates of migration into the site are required to confirm this. The extensive loess plateau around the C. korshinskii plantations is sufficiently large that migration of AM fungi into the region may be low. As a result, we detected no evidence of AM community succession.

Implications for plant and AM fungal community restoration

It has been suggested that the rehabilitation of plant and AM fungal communities is slow in semi-arid ecosystems (Roldánet al., 1997). AM fungi and plant communities potentially show parallel succession (Beveret al., 2001; Hartet al., 2001), and so we predict that the recovery of AM fungal community should interact closely with the rehabilitation in the plant community. It is not clear from our findings whether the recovery of the AM fungal community is slow or whether the maximum diversity has already been reached, because the diversity of AM fungi did not increase with the age of the C. korshinskii plantations.

In summary, the present study is the first investigation of the mycorrhizal status of plantations in the semi-arid loess and gully region of northwestern China. Our results indicate that the AM fungal community occurring in the roots of C. korshinskii varied significantly through season, both in AM fungal colonization and in community composition. The AM fungal diversity in C. korshinskii roots was rather limited and the species richness of AM fungal community did not increase with C. korshinskii plantation age, suggesting that the environment may be a limiting factor to development of the AM fungal community in these plantations. However, evaluation of the mycorrhizal status is only a first step in restoration of the degraded ecosystems using AM technology (Azcón-Aguilaret al., 2003). Isolation of the suitable AM fungal types and screening for a combination of functionally complementary species may be crucial for using AM technology in this region. Moreover, further research is needed to screen the appropriate phytomicrobial complexes, including stress-tolerant herbaceous plants and beneficial plant-associated microbial consortia.

Acknowledgements

We are grateful for financial support provided by National Natural Science Foundation of China (30570270, 30870438, J0630966), the Scientific Research Foundation for Returning Overseas Chinese Scholars, MOE of China, New Century Excellent Talents in University (NCET-07-0390), Natural Science Foundation of Gansu (3ZS051-A25-057) and Research Foundation for Middle-aged and Young Scientists of Northwest University for Nationalities, China (X2007-003, XBMU-2007-BD-36).

References

Al-Karaki
GN
(
1998
)
Benefit, cost and water-user efficiency of arbuscular mycorrhizal durum wheat grown under drought stress
.
Mycorrhiza
 
8
:
41
45
.
Altschul
SF
Madden
TL
Schäffer
AA
Zhang
J
Zhang
Z
Miller
W
Lipman
DJ
(
1997
)
Gapped BLAST and PSI-BLAST: a new generation of protein database search programs
.
Nucleic Acids Res
 
25
:
3389
3402
.
Anderson
RC
Liberta
AE
Dickman
LA
(
1984
)
Interaction of vascular plants and vesicular-arbuscular mycorrhizal fungi across a soil moisture-nutrient gradient
.
Oecologia
 
64
:
111
117
.
Aroca
R
Porcel
R
Ruiz-Lozano
JM
(
2007
)
How does arbuscular mycorrhizal symbiosis regelate root hydraulic properties and plasma membrane aquaporins in Phaseolus vulgaris under drought, cold or salinity stresses?
New Phytol
 
173
:
808
816
.
Azcón-Aguilar
C
Palenzuela
J
Roldán
A
Bautista
S
Vallejo
R
Barea
JMM
(
2003
)
Analysis of the mycorrhizal potential in the rhizosphere of representative plant species from desertification-threatened Mediterranean shrublands
.
Appl Soil Ecol
 
22
:
29
37
.
Benjamin
PK
Anderson
RC
Liberta
AE
(
1989
)
Vesicular-arbuscular mycorrhizal ecology of litter bluestem across a prairie-forest gradient
.
Can J Bot
 
67
:
2678
2685
.
Bever
JD
Schultz
PA
Pringle
A
Morton
JB
(
2001
)
Arbuscular mycorrhizal fungi: more diverse than meets the eye, and the ecological tale of why
.
BioSci
 
51
:
923
931
.
Bohrer
KE
Friese
CF
Amon
JP
(
2004
)
Seasonal dynamics of arbuscular mycorrhizal fungi in differing wetland habitats
.
Mycorrhiza
 
14
:
329
337
.
Borowicz
VA
(
2001
)
Do arbuscular mycorrhizal fungi alter plant-pathogen relations?
Ecology
 
82
:
3057
3068
.
Brundrett
M
Melville
L
Peterson
L
(
1994
)
Practical Methods in Mycorrhiza Research
 .
Mycologue Publications, University of Guelph
,
Canada
.
Daniell
TJ
Husband
R
Fitter
AH
Young
JPW
(
2001
)
Molecular diversity of arbuscular mycorrhizal fungi colonising arable crops
.
FEMS Microbiol Ecol
 
36
:
203
209
.
Eom
AH
Hartnett
DC
Wilson
GWT
(
2000
)
Host plant species effects on arbuscular mycorrhizal fungal communities in tallgrass prairie
.
Oecologia
 
122
:
435
444
.
Ferrol
N
Calvente
R
Cano
C
Barea
JM
Azcón-Aguilar
C
(
2004
)
Analysing arbuscular mycorrhizal fungal diversity in shrub-associated resource islands from a desertification-threatened semiarid Mediterranean ecosystem
.
Appl Soil Ecol
 
25
:
123
133
.
Hart
MM
Reader
RJ
Klironomos
JN
(
2001
)
Life-history strategies of arbuscular mycorrhizal fungi in relation to their successional dynamics
.
Mycologia
 
93
:
1186
1194
.
Helgason
T
Daniell
TJ
Husband
R
Fitter
AH
Young
JPW
(
1998
)
Ploughing up the wood-wide web?
Nature
 
384
:
431
.
Helgason
T
Fitter
AH
Young
JPW
(
1999
)
Molecular diversity of arbuscular mycorrhizal fungi colonising Hyacinthoides non-scripta (bluebell) in a seminatural woodland
.
Mol Ecol
 
8
:
659
666
.
Helgason
T
Merryweather
JW
Denison
J
Wilson
P
Young
JPW
Fitter
AH
(
2002
)
Selectivity and functional diversity in arbuscular mycorrhizas of co-occurring fungi and plants from a temperate deciduous woodland
.
J Ecol
 
90
:
371
384
.
Husband
R
Herre
EA
Young
JPW
(
2002a
)
Temporal variation in the arbuscular mycorrhizal communities colonizing seedlings in a tropical forest
.
FEMS Microbiol Ecol
 
42
:
131
136
.
Husband
R
Herre
EA
Turner
SL
Gallery
R
Young
JPW
(
2002b
)
Molecular diversity of arbuscular mycorrhizal fungi and patterns of host association over time and space in a tropical forest
.
Mol Ecol
 
11
:
2669
2678
.
Johnson
D
Vandenkoornhuyse
PJ
Leake
JR
Gilbert
L
Booth
RE
Grime
JP
Young
JPW
Read
DJ
(
2003
)
Plant communities affect arbuscular mycorrhizal fungal diversity and community composition in grassland microcosms
.
New Phytol
 
161
:
503
515
.
Johnson
D
IJdo
M
Genney
DR
Anderson
IC
Alexander
IJ
(
2005
)
How do plants regulate the function, community structure, and diversity of mycorrhizal fungi?
J Exp Bot
 
56
:
1751
1760
.
Johnson
NC
Zak
DR
Tilman
D
Pfleger
FL
(
1991
)
Dynamics of vesicular-arbuscular mycorrhizae during old field succession
.
Oecologia
 
86
:
349
358
.
Kowalchuk
GA
De Souza
FA
Vanveen
JA
(
2002
)
Community analysis of arbuscular mycorrhizal fungi associated with Ammophila arenaria in Dutch coastal sand dunes
.
Mol Ecol
 
11
:
571
581
.
Lekberg
Y
Koide
RT
Rohr
JR
Aldrich-Wolfe
L
Morton
JB
(
2007
)
Role of niche restrictions and dispersal in the composition of arbuscular mycorrhizal fungal communities
.
J Ecol
 
95
:
95
105
.
Lugo
MA
González-Maza
ME
Cabello
MN
(
2003
)
Arbuscular mycorrhizal fungi in a mountain grassland II: seasonal variation of colonization studied, along with its relation to grazing and metabolic host type
.
Mycologia
 
95
:
407
415
.
McGonigle
TP
Miller
MH
Evans
DG
Fairchild
GL
Swan
JA
(
1990
)
A new method which gives an objective measure of colonization of roots by vesicular-arbuscular mycorrhizal fungi
.
New Phytol
 
115
:
495
501
.
Merryweather
J
Fitter
A
(
1998
)
The arbuscular mycorrhizal fungi of Hyacinthoides non-scripta II. Seasonal and spatial patterns of fungal populations
.
New Phytol
 
138
:
131
142
.
Nanjing Agricultural University
(
1996
)
Chemical Analysis Methods in Soil and Agriculture
 .
China Agricultural Press
,
Beijing
.
Öpik
M
Moora
M
Liira
J
Kõljalg
U
Zobel
M
Sen
R
(
2003
)
Divergent arbuscular mycorrhizal communities colonize roots of Pulsatilla spp. in boreal Scots pine forest and grassland soils
.
New Phytol
 
160
:
581
593
.
Öpik
M
Moora
M
Liira
J
Zobel
M
(
2006
)
Composition of root-colonizing arbuscular mycorrhizal fungal communities in different ecosystems around the globe
.
J Ecol
 
94
:
778
790
.
Öpik
M
Moora
M
Zobel
M
Saks
U
Wheatley
R
Wright
F
Daniell
T
(
2008
)
High diversity of arbuscular mycorrhizal fungi in a boreal herb-rich coniferous forest
.
New Phytol
 
179
:
867
876
.
Rillig
MC
(
2004
)
Arbuscular mycorrhizae and terrestrial ecosystem processes
.
Ecol Lett
 
7
:
740
754
.
Roldán
A
Garcia
C
Albaladejo
J
(
1997
)
AM fungal abundance and activity in a chronosequence of abandoned fields in a semiarid Mediterranean site
.
Arid Soil Res Rehabil
 
11
:
211
220
.
Sanders
IR
Fitter
AH
(
1992
)
The ecology and functioning of vesicular-arbuscular mycorrhizas in co-existing grassland species II. Nutrient uptake and growth of vesicular-arbuscular mycorrhizal plants in a semi-natural grassland
.
New Phytol
 
120
:
525
533
.
Sanguinetti
CJ
Dias Neto
E
Simpson
AJ
(
1994
)
Rapid silver staining and recovery of PCR products separated on polyacrylamide gels
.
Biotechniques
 
17
:
914
921
.
Santos
JC
Finlay
RD
Tehler
A
(
2006
)
Molecular analysis of arbuscular mycorrhizal fungi colonising a semi-natural grassland along a fertilization gradient
.
New Phytol
 
172
:
159
168
.
Santos-González
JC
Finlay
RD
Tehler
A
(
2007
)
Seasonal dynamics of arbuscular mycorrhizal fungal communities in roots in a semi-natural grassland
.
Appl Environ Microbiol
 
73
:
5613
5623
.
Schüßler
A
Schwarzott
D
Walker
C
(
2001
)
A new fungal phylum, the Glomeromycota: phylogeny and evolution
.
Mycol Res
 
105
:
1413
1421
.
Schwarzott
D
Schüßler
A
(
2001
)
A simple and reliable method for SSU rRNA gene DNA extraction, amplification, and cloning from single AM fungal spores
.
Mycorrhiza
 
10
:
203
207
.
Smith
FA
Smith
SE
(
1997
)
Structural diversity in (vesicular)-arbuscular mycorrhizal symbioses
.
New Phytol
 
137
:
373
388
.
Smith
SE
Read
DJ
(
1997
)
Mycorrhizal Symbiosis
 .
Academic Press
,
London
.
Smith
SE
Smith
FA
Jakobsen
I
(
2003
)
Mycorrhizal fungi can dominate phosphate supply to plants irrespective of growth responses
.
Plant Physiol
 
133
:
16
20
.
Sutherland
WJ
(
1996
)
Ecological Census Techniques: A Handbook
 .
Cambridge University Press
,
Cambridge
.
Vandenkoornhuyse
P
Husband
R
Daniell
TJ
Watson
IJ
Duck
JM
Fitter
AH
Young
JPW
(
2002
)
Arbuscular mycorrhizal community composition associated with two plant species in a grassland ecosystem
.
Mol Ecol
 
11
:
1555
1564
.
Van Der Heijden
MGA
Klironomos
JN
Ursic
M
Moutoglis
P
Streitwolf-Engel
R
Boller
T
Wiemken
A
Sanders
IR
(
1998
)
Mycorrhizal fungal diversity determines plant biodiversity, ecosystem variability and productivity
.
Nature
 
396
:
69
72
.
Van Diepen
LTA
Lilleskov
EA
Pregitzer
KS
Miller
RM
(
2007
)
Decline of arbuscular mycorrhizal fungi in northern hardwood forests exposed to chronic nitrogen additions
.
New Phytol
 
176
:
175
183
.

Author notes

Editor: Christoph Tebbe