3.3 Analysis of amplicon sequencing results
3.3.1 Sequencing data statistics and OTU analysis
The original high-throughput sequences of rhizosphere soil genomic DNA of groups F and L were spliced to obtain spliced sequences, and the effective sequences were obtained after filtering the chimeras (Table 6). After continuous cropping of S. miltiorrhiza, the number of raw tags and effective tags decreased, while those of bacterial and fungal OTUs increased by 3.45% and 3.78%, respectively. Subsequently, cluster analysis was performed on the representative OTU sequences obtained from groups F and L, and the sequences were subjected to species annotation at six different classification levels of phylum, class, order, family, genus, and species. As shown in Table 7, following continuous cropping, the OTU number of the soil bacterial community decreased by 0.3%, 1.67%, and 2.48% at the order, family, and genus levels, respectively, while that of the fungal community increased by 15.38%, 5.45%, and 1.67% at the phylum, order, and family levels, respectively. In addition, the fungi:bacteria ratio in continuous cropping soil showed an increasing trend, when compared with that in non-continuous cropping soil.
Table 6
Sequencing data statistics of bacteria and fungi in rhizosphere soil of continuous and non-continuous cropping of Salvia miltiorrhiza Note: Original tags = tag sequences obtained by splicing; effective label = the label sequence finally used for subsequent analysis after filtering the chimera; base = the base of the final valid data; effectiveness (%) = the percentage of the number of valid labels to the number of original PEs; oTUs = the number of operational taxonomic units.
index | Bacteria | Fungi |
F group | L group | F group | L group |
Raw Tags | 83256 | 82089 | 75428 | 74409 |
Effective Tags | 77993 | 76756 | 74600 | 73511 |
Base(nt) | 19741183 | 19424898 | 17312674 | 16595521 |
Effective (%) | 92.52 | 92.32 | 89.00 | 86.85 |
OTUs | 3942 | 4078 | 980 | 1017 |
Table 7
OTU statistics of bacterial and fungal communities in rhizosphere soil of continuous and non-continuous cropping of Salvia miltiorrhiza
Microbial type | Treatment | Phylum | Class | Order | Family | Genus | Species |
Bacteria | F group | 66 | 150 | 307 | 420 | 644 | 288 |
L group | 70 | 154 | 306 | 413 | 628 | 278 |
Fungi | F group | 13 | 48 | 110 | 235 | 381 | 524 |
L group | 15 | 50 | 116 | 242 | 385 | 528 |
3.3.2 Differences in the influences of α-diversity index
The PD-whole-tree of the soil bacterial and fungal communities in groups F and L varied by 5.89% and 4.32%, respectively, indicating that the diversity of rhizosphere microbial communities significantly changed after continuous cropping (Table 8). After continuous cropping, the Shannon index decreased by 9%, whereas the Chao1 index increased by 1.80%, implying that the rhizosphere soil bacterial diversity decreased after continuous cropping. In contrast, the Shannon and Simpson indices of the rhizosphere soil fungal community increased by 8.6% and 3.6%, respectively, after continuous cropping, when compared with those noted after non-continuous cropping, suggesting that the diversity of the soil fungal community increased and the species distribution was more uniform after continuous cropping.
Table 8
Effects of continuous cropping on the alfa diversity index of bacteria and fungi in rhizosphere soil of Salvia miltiorrhiza
Microbial type | Treatment | Shannon | Simpson | Chao1 | Goods- coverage | PD-whole-tree |
Bacteria | F group | 9.94 | 0.997 | 4298.82 | 0.9904 | 241.0228 |
L group | 9.85 | 0.993 | 4376.41 | 0.9912 | 255.2186 |
Fungi | F group | 5.12 | 0.896 | 1237.13 | 0.9936 | 507.4566 |
L group | 5.56 | 0.932 | 1254.23 | 0.9940 | 485.54 |
Note: Shannon: The total number of categories in the sample and their proportions. The higher the community diversity, the more uniform the species distribution, and the larger the Shannon index. Chao1: Estimation of the total number of species in community samples. Coverage of goods: Coverage. The higher the coverage of the sort, the larger the index. PD whole tree: the genetic relationship of species in the community. Simpson: Diversity and uniformity of species distribution in a community. The analysis used the Simpson diversity index (1-D). The better the species evenness, the greater the Simpson index. |
3.3.3 Differences in the effects of β-diversity index
The bacterial composition significantly varied between continuous and non-continuous cropping soils. Principal coordinate analysis showed the differences in the soil bacterial communities between groups F and L (Fig. 2A). The first and second principal component axes (contribution: 22.67% and 12.71%, respectively) distinguished the soil bacterial communities between continuous cropping and non-continuous cropping groups. In addition, the first and second principal component axes (contribution: 44.01% and 28.54%, respectively) also exhibited differences in the soil fungal community composition between groups F and L (Fig. 2B).
3.3.4 Analysis of the rhizosphere soil bacterial community structure after continuous cropping
The soil bacterial community OTUs derived from the non-continuous cropping and continuous cropping groups were analyzed (Fig. 3A). The Venn diagram showed that the number of OTUs shared by the two groups was 5747. The number of unique OTUs in the non-continuous cropping and continuous cropping groups were 1099 and 1297, respectively. With the increase in the duration of continuous cropping, some specific bacterial strains disappeared and some new strains were enriched, indicating change in the structure and composition of the soil bacterial community. There was no significant difference in the relative abundance of each dominant phylum, class, and order of the soil bacterial community in groups F and L. The dominant phyla in groups F and L were Firmicutes (17.2% and 18.1%, respectively), Proteobacteria (24.1% and 24.3%, respectively), Acinetobacter (7.2% and 7.5%, respectively), Actinobacteria (4.8% and 4.4%, respectively), and Bacteroidetes (4.9% and 4.8%, respectively) (Fig. 3B). The dominant classes in groups F and L were Clostridium (10.2% and 8.0%, respectively), Bacillus (6.3% and 9.2%, respectively), γ-Proteobacteria (13.8% and 14.2%, respectively), α-Proteobacteria (10.3% and 10.1%, respectively), and Bacteroidia (4.8% and 4.8%, respectively) (Fig. 4A). The dominant orders in groups F and L were Leptospirales (5.3% and 3.3%, respectively), Oscillospirales (3.1% and 1.8%, respectively), Lactobacillus (2.2% and 2.3%, respectively), and Rhizopus (4.2% and 4.2%, respectively) (Fig. 4B).
Simper (Similarity percentage) was used to break down the Bray-Curtis difference index to quantify the contribution of each species to the difference between the continuous cropping and non-continuous cropping groups. At the family level (Fig. 5A), the relative abundance of Lachnospiracea and Nitrosopharacea, both of which exhibit soil nitrogen cycle function, decreased by 38.32% and 22.75%, respectively, in the rhizosphere soil of the continuous cropping group. Furthermore, the relative abundance of Erysipelotrichaceae and Ruminococcaceae decreased by 42.49% and 40.82%, respectively, while those of Staphylococcaceae, Peptostreptococcaceae, Lactobacillaceae, and Acidithiobacillaceae decreased at varying degrees in the rhizosphere soil of the continuous cropping group. At the genus level (Fig. 5B), the relative abundance of the dominant genera of the rhizosphere soil bacterial community in groups F and L showed significant variation after continuous cropping. In particular, the relative abundances of Staphylococcus (0.3% and 2.2%, respectively), Romboutsia (0.6% and 2.1%, respectively), Acidithiobacillus (0.006% and 1.1%, respectively), Leptospirillum (0.0006% and 1.0%, respectively), Allobaculum (0.9% and 0.5%, respectively), Blautia (1.3% and 0.8%, respectively), Lactobacillus (0.6% and 0.8%, respectively), Pseudomonas (1.4% and 1.0%, respectively), Ruminococcus (0.9% and 0.4%, respectively), and Bacillus (0.8% and 1.6%, respectively) significantly differed between groups F and L.
3.3.5 Analysis of the rhizosphere soil fungal community structure after continuous cropping
The soil fungal community OTUs (Fig. 6A) derived from the non-continuous cropping and continuous cropping groups were analyzed. The Venn diagram showed that the number of OTUs shared between groups F and L was 1356. The number of unique OTUs in the non-continuous cropping and continuous cropping groups were 968 and 1007, respectively. Furthermore, significant differences in the soil fungal community at the phylum level were observed between the groups F and L (Fig. 6B), and the relative abundances of the top 10 dominant phyla, including Mortiellomycota, Chytridillomycota, Zoopagomycota, Aphelidiomycota, Mucoromycota, Glomeromycota, and Blastocladiomycota, increased by 127.5%, 325.0%, 32.4%, 253.1%, 265.7%, 14.0%, and 1230.8%, respectively, while those of Ascomycota, Basidiomycota, and Rozellomycota decreased by 19.42%, 90.6%, and 30.9%, respectively, after continuous cropping.
As shown in Fig. 7A, the dominant fungal classes in groups F and L were Sordariomycetes (37.2% and 38.7%, respectively), Mortierellomycetes (9.1% and 20.7%, respectively), Agaricomycetes (5.8% and 4.4%, respectively), Orbiliomycetes (10.4% and 6.1%, respectively), Dothideomycetes (10.2% and 4.0%, respectively), Eurotiomycetes (7.4% and 4.9%, respectively), Leotiomycetes (2.3% and 1.2%, respectively), and Tremellomycetes (1.7% and 2.6%, respectively). Furthermore, the dominant fungal orders in groups F and L were Pyrrophyta (23.7% and 16.6%, respectively), Mortierela (9.1% and 20.7%, respectively), Thesephorales (4.6% and 0.08%, respectively), Orbiliales (10.4% and 6.1%, respectively), Microscales (3.0% and 9.2%, respectively), Capnodiales (8.0% and 1.1%, respectively), Eurotiales (8.0% and 1.1%, respectively), Pleospores (7.0% and 4.6%, respectively), and Sordariales (2.0% and 2.7%, respectively) (Fig. 7B). The cluster heatmap of species abundance (Fig. 7C) revealed significant differences in species richness and composition at the family level between groups F and L. The proportion of Coniochaetacaeae and Coniochaetacaeae in the rhizosphere fungal community of continuous cropping S. miltiorrhiza increased, while the proportion of Coniochaetacaeae and Coniochaetacaeae decreased.
With the increase in the duration of continuous cropping, the structure and composition of the soil fungal community changed, and some common soil-borne pathogenic fungi, such as Leptosphaeria turcica, some Cladosporium spp., some subspecies of Alternaria, Fusarium solani, and other subspecies of Fusarium, were enriched. Based on LEfSe (LDA Effect Size) analysis, the relationship between continuous cropping and non-continuous cropping soils was further explored. The results revealed that the abundance of the microbial species in the S. miltiorrhiza soil was significantly different before and after continuous cropping. A total of 30 biomarkers with LDA score > 4 were enriched in the two groups (Fig. 8A), with the abundance of Mortierella, Lophotrichus, Conocybe, Aspergillus, Arthrobotrys, and Cladosporium being statistically different in the two groups (Fig. 8B). Mortierella polycephala, Lophotrichus sp., Mortierella stylospora, and Conocybe pubescens were the biomarkers in the continuous cropping soil, while Mortierella alpina, Arthrobotrys amerospora, and Cladosporium sp. were the biomarkers in the non-continuous cropping soil.
3.3.6 Function prediction analysis
Based on the FunGuild tool, the corresponding categories of microbial ecological functions were annotated, and the statistical results are shown in Fig. 9A and B. In the soil bacterial community, plant and animal pathological nutrition functions were relatively predominant. The abundance of saprophytic bacteria was relatively low and presented only slight variation between groups F and L. These results suggested that the function of bacteria in the rhizosphere soil of S. miltiorrhiza before and after continuous cropping was similar. The abundance of internal parasitic nutritional fungi in the soil fungal community decreased, while that of saprophytic nutritional fungi significantly increased after continuous cropping. Subsequently, principal component analysis was performed on the abundance statistics results based on the database functional annotations (Fig. 9C). The soil fungal functional abundance showed good separation. The first and second principal component axes (contribution: 24.3% and 18.76%, respectively) distinguished the soil fungal composition between groups F and L. In contrast, the soil bacterial functional abundance between the two groups did not exhibit good separation (Fig. 9D). Therefore, it can be concluded that continuous cropping changed the function of the rhizosphere soil fungi, and that long-term continuous cropping caused the enrichment of some pathogenic fungi that gradually became the dominant groups.
3.3.7 Correlation analysis between soil microbial species abundance and S. miltiorrhiza growth index
The S. miltiorrhiza root morphology and root bioaccumulation were significantly positively correlated with the relative abundances of Lecythophora, Aspergillus, Alternaria, Eurotium, and other fungi, and the bacterium Pseudomonas, and were negatively correlated with Rhizophlycti and Papulaspora. The contents of chlorophyll a, chlorophyll b, and carotenoids in S. miltiorrhiza were significantly positively correlated with the relative abundances of Eurotium, Macroventuria, Gibberella, Geomyces, Aspergillus, Cladosporium, Arthrobotrys, Solirubrobacter, Clostridium innocuum, Lysobacter, and Gaiella, but significantly negatively correlated with the relative abundances of fungi such as Papulaspora, Fusicolla, Chaetomium, Solicoccozyma, and Mortierella, and some bacteria such as Ohtaekwangia, Streptomyces, Bacillus, and Leptospirillum (Fig. 10A and 10B). In general, the microbial species abundance and community structure were significantly correlated with the root morphological accumulation and photosynthetic intensity of S. miltiorrhiza.