Disease index and incidence and plant weights.
Figure. 1 shows the phenotypic differences in leaves of P. ginseng in 9 dpi among treatment groups: CK, A, AS, and S. Significant differences were observed in the severity of A. panax infections under Si treatment (Fig. 1). As shown in Figure 1, no effect of Si on biomass was observed compared to the CK group. The differences between CK plants and group S plants were not obvious, however, group AS plants were obviously healthier than group A plants (Fig. 1). The first symptom of leaf spots appeared soon (3 days) after post inoculation (dpi), followed by stunting and blight within a few days. As shown in Table 1, Si treatment significantly reduced the disease incidence and disease index of ginseng black spot.
There was no significant difference in dry weight among non-inoculated (pathogen) plants: CK plants (1.12 ± 0.81 g) and group S plants (1.23 ± 0.59 g). However, the plant dry weight was significantly reduced in group A plants. Apparently, Si treatment resulted in significantly heavier plants (Table 2a). After 9 days post-treatment, the fresh weight of group AS plants was 15% higher than that of group A plants (Table 2b).
Soil properties and plant growth responses.
Soil properties are presented in Table 3. A one-way ANOVA showed that the treatments significantly recovered the soil property parameters from disease treatment (P < 0.05). (Table 3). The pH value of the CK soil samples was ~7.39. Compared with the CK group, the soil pH, NO3--N, and NH4+-N were significantly reduced in Group A (P < 0.05). In contrast, the ratio of available P and available K (P < 0.05) were significantly increased. Furthermore, AS significantly increased the soil pH, and NO3--N and NH4+-N contents (P < 0.05), and significantly reduced the ratio of available P and available K (P < 0.05) compared to the A treatment, i.e., without Si (P < 0.05). No significant differences in the above-mentioned nutrients, except available Si, was detected between group S and CK.
Analysis of bacterial composition and diversity of soil bacterial community structure based on 16S rRNA gene sequencing.
Bacteria-targeted regions were completely amplified by PCR and fully sequenced for all soil samples. The raw sequence libraries were screened to remove reads that originated from sequencing noise or putative chimeric sequences. Using the 12 soil samples from the different treatments, a total of 815,609 valid 16S rDNA sequences were obtained by filtering and processing according to a 97% similarity. Variation of a single soil sample ranged from 56,510 to 76,384 sequences, and the above sequences were retained for further analysis.
The effective sequence number and OTU number of each group of samples did not significantly differ between the treatment groups and the CK group, as shown in Table 4. The sequencing coverage of samples ranged from 98.5% to 98.6%. After sample diversity (alpha diversity) analysis, the indexes reflecting the abundance and diversity of microbial communities were calculated, and the results of all treatments were analyzed using a one-way ANOVA (Table 4). The coverage index of the sample library was more than 98.5%, which indicated that the sequencing results represented the real situation of the bacterial population in the sample. The microflora richness index (Chao1, ACE) and biodiversity index (Shannon, Simpson) of the samples revealed that the diversity of the bacterial populations in the soil samples was relatively high (Table 4). Further analysis revealed that at a 97% similarity level, the Shannon index and Simpson index of soil bacteria of each treatment group were not significantly different from those in the CK group.
Analysis of soil bacterial community structure.
According to the abundance of bacterial OTU types in the 12 soil samples, a non-metric multidimensional scale (NMDS) diversity analysis was conducted to determine the differences in the bacterial compositions of the different samples and treatments (Fig. 2). The NMDS results were evaluated using the UniFrac distances to estimate the phylogenetic relatedness among the bacterial communities (Fig. 3a, Fig. 3c). The soil bacterial communities were found to be totally distinct between groups A and S, i.e., when treated with A. panax or Si (NMDS). Among treatment groups, the soil bacterial composition of group CK was most similar to that of group AS, i.e., had the highest phylogenetic relatedness, and the group AS bacterial flora could be independently distinguished from that in the infected soil (group A). However, the composition of bacterial flora in group S differed from that of the other treatments. In summary, Si application significantly regulated the changes in bacterial flora (back to the composition of CK) that were induced by inoculation of ginseng black spot (group AS).
Cluster analysis of soil bacterial community structure.
Based on a Beta diversity analysis, a distance matrix was obtained for the 12 soil samples, and a hierarchical clustering analysis was conducted using the unweighted group average method (UPGMA) (Fig. 3b, Fig. 3d). The soil samples of groups S and AS were classified as one branch, and those of groups CK and AS group were classified as one branch. The results were consistent with those of the NMDS analysis, which fully demonstrated that the soils inoculated with Si + ginseng black spot (AS group) were significantly recovered compared with the soils inoculated with only ginseng black spot (S group). PLS-DA analysis showed that the microbial composition of soil in the AS group was significantly altered following Si treatment. The results suggested similarities between group CK and AS, but not with nor among the other two groups. In summary, Si was again shown to have alleviated the changes in soil bacteria caused by ginseng black spot (Fig. 4).
Heat map analysis of the soil bacterial community structure.
A heat map of the bacterial community structure among different samples (Fig. 5) revealed the relative abundances of the various bacterial groups (at phylum and genus levels) and that significant differences were observed among different groups of samples. The results showed that, at the phylum level, Proteobacteria, Nitrospirae, Actinobacteria, and Bacteroidetes were the four main groups (Fig. 5). The relative abundances (represented by the color depth in Fig. 5) of Sandaracinus, Polycyclovorans, Hirschia, Bdellovibrio, Haliangium, and Nitrospira were significantly higher in group CK than those of group A (P < 0.05). In addition, the relative abundances of Sandaracinus, Polycyclovorans, Hirschia, Haliangium, and Nitrospira in group AS were significantly higher than those in A (P < 0.05). The results showed that Si application significantly regulated the structural impact of soil microorganisms caused by ginseng black spot inoculation.
Factors influencing the quantity and composition of soil bacteria.
Correlation analysis showed (Table 5) that most of the other dominant bacterial groups had significant correlations with soil chemical properties, except Arenimonas, H16, and RB41, which showed no correlations with all chemical indicators. Haliangium and available K were significantly negatively correlated; Phenylobacterium (phenyl coli) was very significantly negatively correlated with pH and Gemmatimonas (bacillus); Nitrospira (nitrification spirillum) was negatively correlated with NO3--N; Mesorhizobium (rhizobia) was very significantly positively correlated with NO3--N; Gemmatimonas (bacillus), Nitrospira (nitrobacteria), and available Si were significantly negatively correlated; Lactobacillus (lactobacillus), Mesorhizobium (rhizobium), and available Si were significantly positively correlated. Haliangium was significantly positively correlated with pH.