AMF species diversity
We identified 77 total AMF OTUs in our 27 soil samples, which were separated into 9 groups. Dilution curves generated for these 9 groups were flat, indicating that sequencing depth was sufficient and that additional sequencing depth would have revealed only a small number of additional species (Supplementary Fig. S1).
We next conducted taxonomic analyses of these representative OTUs, leading us to determine that these fungi were associated with 1 Class, 4 Orders, 4 Families, 4 Genera, and 20 Species, with additional unidentified species having been detected at various taxonomic levels. No significant differences in Alpha diversity, Sobs, Shannon, Chao1, or Simpson index values were observed among samples as a function of soil depth, whereas these index values did vary significantly as a function of slope position. Phylogenetic diversity (PD) was unrelated to soil depth or slope position. Greater than 99.98% coverage was achieved in this sequencing analysis, confirming that these data met with the targeted sequencing depth requirements (Table 1).
Table 1. Rhizosphere AMF diversity indices.
|
depth
|
sobs
|
shannon
|
simpson
|
chao
|
pd
|
coverage
|
sw
|
0-20cm
|
20.67±6.03
|
2.09±0.35
|
0.21±0.09
|
21.00±5.57
|
1.43±0.52
|
99.99%
|
|
20-40cm
|
28.00±1.73
|
2.27±0.15
|
0.15±0.02
|
28.87±1.86
|
1.97±0.33
|
99.98%
|
|
40-60cm
|
23.67±6.66
|
2.04±0.33
|
0.23±0.08
|
25.00±5.68
|
1.78±0.50
|
99.98%
|
|
mean
|
24.11±5.58B
|
2.13±0.27B
|
0.20±0.07B
|
24.96±5.31B
|
1.72±0.46
|
99.99%
|
zw
|
0-20cm
|
16.00±7.00
|
1.51±0.26
|
0.34±0.08
|
116.33±7.02
|
1.45±0.45
|
99.99%
|
|
20-40cm
|
17.00±5.57
|
1.51±0.22
|
0.34±0.15
|
17.50±6.06
|
1.56±0.74
|
99.99%
|
|
40-60cm
|
20.00±2.00
|
1.75±0.45
|
0.25±0.11
|
21.00±2.65
|
2.14±0.47
|
99.98%
|
|
mean
|
15.67±4.92C
|
1.60±0.31C
|
0.31±0.11A
|
18.28±5.26C
|
1.72±0.59
|
99.99%
|
xw
|
0-20cm
|
30.00±6.25
|
2.41±0.16
|
0.12±0.04
|
35.33±13.65
|
1.45±0.22
|
99.98%
|
|
20-40cm
|
29.33±1.15
|
2.35±0.05
|
0.15±0.03
|
29.67±1.53
|
1.53±0.05
|
99.99%
|
|
40-60cm
|
31.33±1.15
|
2.48±0.09
|
0.13±0.03
|
33.33±4.16
|
1.62±0.21
|
99.99%
|
|
mean
|
30.22±3.35A
|
2.41±0.11A
|
0.13±0.03C
|
32.75±7.60A
|
1.54±0.17
|
99.99%
|
Lowercase letters are significantly difference among three soil depth,capital letters are significantly difference among three plant species (P < 0.05). SW, ZW, and XW respectively correspond to the top, middle, and bottom slope positions.
Intermediate junctions were used to identify OTUs common to all samples as well as OTUs unique to specific samples. These analyses revealed that there were 12 core OTUs in our samples, with two of these OTUs belonging to the Diversispora genus and all others belonging to the Glomus genus (Fig. 1).
AMF community composition and distribution
Clear differences in AMF community composition in different rhizosphere soil samples were observed in this study. Glomus species accounted for 83.46% of total AMF species, while Diversispora accounted for 14.73%, Ambispora for 1.21%, and Paraglomus species were only detected in a few samples (Fig. 2a).
In our constructed species relationship diagrams, Glomus distributions in individual samples ranged from 7.7% - 13%. Of these Glomus species, 31%, 28%, and 39% were detected in the soil samples from the upper, middle, and lower slope positions, respectively, while 31%, 28%, and 39% of Glomus species were detected in samples collected at respective soil depths of 0-20 cm, 20-40 cm, and 40-60 cm. Diversispora species were primarily distributed in the upper and middle slope positions, with 52%, 47.8%, and 0.2% of these species being found in the upper, middle, and lower slope positions, respectively. In addition, 30%, 31%, and 39% of Diversispora species were detected in samples collected at 0-20 cm, 20-40 cm, and 40-60 cm depths, respectively (Fig. 2b). Glomus and Diversispora species were present at all soil depths and slope positions, with Glomus species composing the largest proportion of all samples. Ambispora species were only detected in soil samples collected at depths of 0-20 cm and 40-60 cm from the middle slope position (Supplementary Fig. S2)
A cluster analysis conducted according to Unweighted UniFrac distance values revealed that samples were separable into three primary categories based upon environmental variables. Clusters of samples collected at the same slope site indicated that AMF community composition was more similar at a given slope site, whereas this composition differed substantially among slope sites (Fig. 3).
A Bray-Curtis PCoA analysis indicated that there was no difference in AMF community composition at different soil depths, whereas these communities did differ significantly as a function of sample slope position. While there were no differences in AMF community composition in the middle or upper slope positions, the composition of these samples did differ significantly with respect to the composition of AMF communities in soil samples collected from lower slope positions (R=0.332, P =0.001) (Fig. 4a).
LEFse analyses were also used to identify biomarkers associated with AMF diversity in rhizosphere soil. At different soil depths, only Glomus species were significantly enriched at the species and OTU levels. The most abundant biomarkers were detected in soil samples collected at a depth of 0-20 cm, with decreasing levels of these biomarkers as soil depth increased (Fig. 5b). With respect to slope position, biomarker abundance was highest in the lower slope position. However, Glomus species were significantly enriched in samples collected from the lower slope position, whereas Diversispora species were significantly enriched in samples collected from the upper slope position (Fig. 5a).
Relationships between soil AMF diversity and soil properties
Significant differences in pH and TP were observed between soil samples as a function of soil depth (Table 2; P<0.05), but these differences may be a function of different slope positions, as soil from the same slope position did not differ as a function of depth. When comparing soil samples collected from different slope positions, there were significant differences in OM, TN, TP, TK, AN, NN, TDS, pH, and SM values (P<0.05).
Table 2. Mean values of nonbiological factors in soils of different depths.
|
depth
|
OM(g/kg)
|
TN(g/kg)
|
TP(g/kg)
|
TK(g/kg)
|
NN(mg/kg)
|
AN(mg/kg)
|
AP(mg/kg)
|
AK(mg/kg)
|
TDS(g/kg)
|
PH
|
SM(%)
|
sw
|
0 -20cm
|
23.47±3.75
|
1.99±0.34
|
0.76±0.09a
|
11.82±2.94
|
3.71±1.05
|
5.59±0.62
|
3.05±0.62
|
239.71±99.13
|
5.70±3.29
|
8.07±0.14b
|
7.44±1.73
|
20-40cm
|
23.12±1.88
|
2.00±0.23
|
0.77±0.09a
|
11.86±2.95
|
3.37±0.61
|
5.58±0.99
|
4.32±2.06
|
217.33±66.7
|
5.45±3.15
|
7.95±0.15b
|
8.00±1.24
|
40-60cm
|
21.92±1.42
|
2.03±0.07
|
0.76±0.07a
|
12.05±2.42
|
3.72±1.10
|
5.06±1.24
|
3.11±1.05
|
234.01±82.87
|
5.55±3.20
|
7.98±0.15b
|
6.5±1.65
|
mean
|
22.84±2.32B
|
2.00±0.21B
|
0.76±0.07A
|
11.91±2.41B
|
3.60±0.84B
|
5.41±0.89AB
|
3.49±1.34
|
230.35±73.40
|
6.13±4.83A
|
8.00±0.14B
|
7.31±1.50A
|
zw
|
0 -20cm
|
24.13±2.58
|
2.14±0.21
|
0.53±0.03b
|
15.08±1.79
|
3.62±2.36
|
4.40±0.61
|
2.43±0.68
|
200.11±19.55
|
5.64±3.26
|
8.04±0.16b
|
5.38±2.7
|
20-40cm
|
23.52±3.73
|
2.16±0.18
|
0.54±0.02b
|
15.60±1.6
|
3.68±1.74
|
4.80±0.35
|
2.17±0.56
|
253.91±101.89
|
5.86±3.38
|
8.03±0.11b
|
5.54±9.80
|
40-60cm
|
25.40±1.11
|
2.30±0.02
|
0.55±0.01b
|
16.48±0.45
|
4.46±1.64
|
4.22±0.32
|
2.35±0.50
|
251.63±28.73
|
4.43±2.55
|
8.19±0.10b
|
5.20±3.80
|
mean
|
24.35±2.48AB
|
2.20±0.16A
|
0.53±0.21C
|
15.72±1.37A
|
3.92±1.73AB
|
4.47±0.46B
|
2.31±0.52
|
235.22±59.93
|
5.20±4.67AB
|
8.09±0.13B
|
5.37±0.56B
|
xw
|
0 -20cm
|
27.24±1.79
|
2.16±0.13
|
0.70±0.02a
|
15.06±1.47
|
5.56±2.25
|
5.56±1.24
|
3.09±1.65
|
314.8±257.37
|
0.28±0.16
|
8.58±0.10a
|
7.76±1.66
|
20-40cm
|
26.50±2.04
|
2.16±0.10
|
0.70±0.01a
|
15.04±1.57
|
5.08±2.17
|
5.74±0.47
|
3.57±2.84
|
359.52±284.25
|
0.45±0.26
|
8.62±0.11a
|
7.21±3.08
|
40-60cm
|
26.02±2.99
|
2.18±0.13
|
0.70±0.05a
|
15.35±1.6
|
5.59±3.20
|
6.33±0.78
|
3.17±1.38
|
419.81±342.26
|
0.45±0.26
|
8.56±0.23a
|
6.73±1.81
|
|
mean
|
26.59±2.09A
|
2.17±0.10A
|
0.70±0.03B
|
15.15±1.35A
|
5.41±2.25A
|
5.87±0.84A
|
3.28±1.80
|
364.71±261.01
|
2.08±0.49B
|
8.59±0.14A
|
7.23±2.02A
|
Lowercase letters are significantly difference among three soil depth, capital letters are significantly difference among three plant species (P < 0.05). SW, ZW, and XW respectively correspond to the top, middle, and bottom slope positions.
Spearman correlation analyses revealed that TP, AN, and soil pH were significantly positively correlated with Shannon index values (P<0.05), whereas TP, TDS, pH, AE, and Chao 1 values were positively correlated with one another (P<0.05) (Supplementary Table S1).
CCA analysis revealed that OM (R2=0.3939, P=0.002), TN (R2=0.2701, P=0.021), TK (R2=0.3209, P=0.012), NN (R2=0.3003, P=0.012), AN (R2=0.2692, P=0.03), AK (R2=0.3803, P=0.001), TDS (R2=0.5328, P=0.001), pH (R2=0.4834, P=0.001), SM (R2=0.4108, P=0.004), and AE (R2=0.335, P=0.011) all had a significant impact on AMF community composition, explaining 24.07% of the overall variability in this composition. In addition, AE was positively correlated with TDS and depth, whereas it was negatively correlated with other environmental factors (Fig. 4b).
The top 20 most abundant OTUs were identified and used to assess relationships between species and environmental factors. Among these OTUs were OTU33, which corresponded to an Ambispora species and OTU61, OTU12, and OTU24 which corresponded to Diversispora species, with all other top OTUs corresponding to Glomus species. OM, NN, AN, TP, AE, TDS, and pH were all significantly correlated with three or more OTUs (Fig. 6).