Soil characteristics
The concentrations of soil TC (F = 8.71, P = 0.01), TN (F = 6.21, P = 0.02), C/N (F = 12.55, P = 0.00), AN (F = 13.34, P = 0.00), TP (F = 6.00, P = 0.02), AP (F = 20.52, P = 0.00), and AK (F = 6.59, P = 0.02) differed significantly among PKSZ, RPL, and PSC, however, the soil pH existed no significant difference (F = 2.13, P = 0.18) (Table 1). RPL hold the highest contents of TC, TN, TP, and AP with 5.71 g/kg, 0.74 g/kg, 2.89 g/kg, and 20.22 mg/kg, respectively, while, PSC hold the highest AK content with 108.98 mg/kg (Table 1). Soil pH, TC, TN, and TP in unrestored area were 8.17, 1.30g/kg, 0.36g/kg, and 1.03g/kg (Deng et al., 2020a).
Table 1
Differences in soil characteristics among PKSZ, RPL, and PSC. PKSZ: Pinus koraiensis Sieb. et Zucc., RPL: Robinia pseudoacacia L., PSC: Populus simonii Carr. Average value ± standard error (n = 12). TC: total carbon, TN: total nitrogen, C/N: C to N ration, AN: available nitrogen, TP: total phosphorus, AP: available phosphorus, AK: available potassium. Different capital letters in same row indicated the significant difference at 0.01 level, and different lowercase letters in same row indicated the significant difference at 0.05 level.
|
PKSZ
|
RPL
|
PSC
|
F
|
P
|
pH
|
7.18aA
|
6.84aA
|
7.00aA
|
2.13
|
0.18
|
TC (g/kg)
|
3.23bA
|
5.71aA
|
5.48aA
|
8.71
|
0.01
|
TN (g/kg)
|
0.49bA
|
0.74aA
|
0.64abA
|
6.21
|
0.02
|
C/N
|
6.64bB
|
7.66abAB
|
8.64aA
|
12.55
|
0.00
|
AN (mg/kg)
|
25.65bB
|
38.92aAB
|
46.43aA
|
13.34
|
0.00
|
TP (g/kg)
|
2.26bA
|
2.89aA
|
2.38abA
|
6.00
|
0.02
|
AP (mg/kg)
|
14.89bB
|
20.22aA
|
13.30bB
|
20.52
|
0.00
|
AK (mg/kg)
|
69.88bA
|
94.75abA
|
108.98aA
|
6.59
|
0.02
|
Sequencing and OTUs
A total of 779, 651 fungal sequences (reads) were obtained by high-throughput amplification sequencing. After filtering, denoising, merging, removing chimera, and nonsingleton, 659, 453 (54, 955 per sample) high quality sequences remained in the dataset, which were clustered into 1,172 amplicon sequence variants (ASV) (Fig. 1). As the sequencing depth increased, the rarefaction curves for evaluating the observed_species per sample universally tended to be saturation, demonstrating that the number of sequences was sufficient (Fig. S1). The number of ASVs of PKSZ, RPL, and PSC was 479, 697, and 356, respectively, and the number of ASVs shared by PKSZ, RPL, and PSC was 97 (Fig. 1).
Ectomycorrhizal fungal diversity
Ectomycorrhizal fungal Chao1 index (F = 17.98, P < 0.01), Pielou_e index (F = 9.44, P < 0.01), Shannon index (F = 15.54, P < 0.01), and Observed_species (F = 27.37, P < 0.01) differed obviously among PKSZ, RPL, and PSC (Fig. 2). In addition, the mean Chao1 index, Pielou_e index, Shannon index, and Observed_species were even greater in RPL than PKSZ and PSC with 331.96, 0.72, 5.99, and 314.35, severally (Fig. 2). Neither Goods_coverage (F = 0.08, P = 0.92) nor Simpson index (F = 4.20, P = 0.05) of ECM fungi in roots of PKSZ, RPL, and PSC differed distinctly (Fig. 2). Ectomycorrhizal fungal Chao1 index (r = 0.59, p < 0.05) and Observed_species (r = 0.60, p < 0.05) increased with the increase of TP (Table 2). Chao1 index (r = 0.76, p < 0.01), Observed_species (r = 0.78, p < 0.01), Pielou_e index (r = 0.64, p < 0.05), and Shannon index (r = 0.70, p < 0.05) of ECM fungal communities inhabiting the roots increased with the increase of soil AP (Table 2).
Table 2
The relationships between soil characteristics and ectomycorrhizal fungal diversity (n = 12). *P<0.05, ** P<0.01. TC: total carbon, TN: total nitrogen, C/N: C to N ration, AN: available nitrogen, TP: total phosphorus, AP: available phosphorus, AK: available potassium.
|
PH
|
TC
|
TN
|
C/N
|
AN
|
TP
|
AP
|
AK
|
Chao1 index
|
-0.31
|
0.27
|
0.44
|
-0.16
|
-0.03
|
0.59*
|
0.76**
|
-0.10
|
Goods_coverage
|
-0.03
|
0.09
|
0.16
|
-0.07
|
-0.03
|
-0.10
|
0.03
|
0.07
|
Observed_species
|
-0.32
|
0.29
|
0.48
|
-0.17
|
-0.04
|
0.60*
|
0.78**
|
-0.10
|
Pielou_e index
|
-0.08
|
0.24
|
0.46
|
-0.30
|
-0.25
|
0.20
|
0.64*
|
-0.32
|
Shannon index
|
-0.14
|
0.25
|
0.48
|
-0.29
|
-0.20
|
0.31
|
0.70*
|
-0.27
|
Simpson index
|
0.02
|
0.16
|
0.35
|
-0.30
|
-0.35
|
0.01
|
0.44
|
-0.49
|
Ectomycorrhizal fungal community composition
A total of 12 phyla, 38 classes, 395 genus, and 575 species were identified in our study. The dominant fungal groups were Ascomycota, Basidiomycota, and Mortierellomycota at the phylum level, accounting for 99.43% (Fig. 3A). Agaricomycetes, Pezizomycetes, Sordariomycetes, Eurotiomycetes, Dothideomycetes, Mortierellomycetes, Leotiomycetes, and Tremellomycetes were the dominant fungal groups at the class level (Fig. S2A). At the genus level, the fungal groups with the average relative abundance more than 3.5% were Hebeloma, Geopora, Sebacina, Tomentella, Penicillium, Fusarium, Metarhizium, Mortierella, Pulvinula, and Clavulina (Fig. 3B). At the species level, the fungal groups with the average relative abundance more than 2.0% were Hebeloma_mesophaeum, Geopora_arenicola, Clavulina_cinerea, Cenococcum_geophilum, Tomentella_ellisii, Gibberella_baccata, and Mortierella_alpina (Fig. S2B).
The NMDS (stress = 0.062) demonstrated that ectomycorrhizal fungal community composition clearly differed among PKSZ, RPL, and PSC especially along NMDS1 (Fig. 5). The biomarkers in RPL were Ascomycota (74.72%), Glomeromycota (0.94%), Mortierellomycota (10.64%), Eurotiomycetes (18.04%), Leotiomycetes (2.66%), Sordariomycetes (42.34%), Tremellomycetes (3.52%), Penicillium (10.71%), Fusarium (16.11%), Metarhizium (6.36%), Mortierella (10.64%), Gibberella (7.01%), and Didymella (3.34%) (Fig. 4). The biomarkers in PKSZ were Pezizomycetes (15.68%), Geopora (22.55%), Suillus (5.52%), Pulvinula (12.71%), and Cenococcum (9.25%) (Fig. 4). The biomarkers in PSC were Basidiomycota (74.67%), Agaricomycetes (25.925), Laccaria (4.41%), Hebeloma (19.29%), Inocybe (3.47%), Sebacina (20.31%), Tomentella (9.70%), and Clavulina (10.26%), and Tuber (9.45%) (Fig. 4).
Contribution of soil properties to ectomycorrhizal fungal community composition
For the ectomycorrhizal fungal community at the phylum level, all the eight soil characteristics explained 99.9% of the variance, with axis 1 explaining 82.40% of the variance and axis 2 explaining 16.60% (Fig. 6A). For the fungal community at the genus level, all the eight soil characteristics explained 65.0% of the variance (Fig. 6B), with axis 1 explaining 40.10% of the variance and axis 2 explaining 24.90% (Fig. 6B). Notably, the concentration of AP in soil was positively correlated with Mortierellomycota (r = 0.68, p < 0.05) and Glomeromycota (r = 0.59, p < 0.05). Aphelidiomycota was significantly positive correlation with the concentration of soil TN (r = 0.59, p < 0.05), TP (r = 0.77, p < 0.01), and AP (r = 0.90, p < 0.01). While, Mortierellomycota (r = -0.73, p < 0.01), Glomeromycota (r = 0.65, p < 0.05), and Aphelidiomycota (r = -0.75, p < 0.01) decreased with the increase of soil pH (Table 3).
Table 3
The relationship between soil basic characteristics and ectomycorrhizal fungal community composition. (n = 12). *P<0.05, ** P<0.01. TC: total carbon, TN: total nitrogen, C/N: C to N ration, AN: available nitrogen, TP: total phosphorus, AP: available phosphorus, AK: available potassium.
|
pH
|
TC
|
TN
|
C/N
|
AN
|
TP
|
AP
|
AK
|
Ascomycota
|
-0.17
|
-0.12
|
0.01
|
-0.52
|
-0.40
|
-0.11
|
0.22
|
-0.58*
|
Basidiomycota
|
0.47
|
-0.07
|
-0.24
|
0.34
|
0.29
|
-0.15
|
-0.47
|
0.52
|
Mortierellomycota
|
-0.73**
|
0.11
|
0.29
|
-0.28
|
-0.35
|
0.48
|
0.69*
|
-0.53
|
Glomeromycota
|
-0.65*
|
0.29
|
0.39
|
-0.06
|
-0.13
|
0.30
|
0.59*
|
-0.34
|
Aphelidiomycota
|
-0.75**
|
0.49
|
0.59*
|
0.15
|
0.27
|
0.77**
|
0.90**
|
0.03
|
|
pH
|
TC
|
TN
|
C/N
|
AN
|
TP
|
AP
|
AK
|
Hebeloma
|
0.55
|
-0.78**
|
-0.81**
|
-0.46
|
-0.33
|
-0.28
|
-0.37
|
-0.05
|
Geopora
|
0.75**
|
-0.65*
|
-0.69*
|
-0.43
|
-0.41
|
-0.69*
|
-0.64*
|
-0.21
|
Sebacina
|
0.51
|
-0.27
|
-0.40
|
0.16
|
0.04
|
-0.41
|
-0.76**
|
0.14
|
Tomentella
|
0.79**
|
-0.17
|
-0.29
|
0.05
|
0.00
|
-0.74**
|
-0.74**
|
0.25
|
Penicillium
|
-0.57
|
0.18
|
0.38
|
-0.36
|
-0.14
|
0.13
|
0.43
|
-0.28
|
Fusarium
|
-0.68*
|
0.29
|
0.43
|
-0.22
|
-0.17
|
0.42
|
0.61*
|
-0.51
|
Metarhizium
|
-0.61*
|
-0.24
|
-0.06
|
-0.48
|
-0.58*
|
0.55
|
0.77**
|
-0.62*
|
Mortierella
|
-0.73**
|
0.11
|
0.29
|
-0.28
|
-0.35
|
0.48
|
0.70*
|
-0.53
|
Pulvinula
|
0.57
|
-0.61*
|
-0.63*
|
-0.64*
|
-0.68*
|
-0.67*
|
-0.51
|
-0.47
|
Clavulina
|
0.63*
|
0.28
|
0.07
|
0.44
|
0.46
|
-0.53
|
-0.60*
|
0.70*
|
At the genus level, soil fungal community composition was driven by soil properties (Table 3). The relation abundance of Hebeloma (r = -0.78, p < 0.01; r = -0.81, p < 0.01) and Geopora (r = -0.65, p < 0.05; r = -0.69, p < 0.05) declined with the increase of TC and TN. There were high correlation coefficients present between AP and Geopora (r = -0.64, p < 0.05), Sebacina (r = -0.76, p < 0.01), Tomentella (r = -0.74, p < 0.01), Fusarium (r = 0.61, p < 0.05), Metarhizium (r = 0.77, p < 0.01), Mortierella (r = 0.70, p < 0.05), Clavulina (r = -0.60, p < 0.05) (Table 3).