Flue-cured tobacco intercropping with insectary floral plants improves rhizosphere soil microbial communities and chemical properties of flue-cured tobacco

Abstract

Background

Continuous cropping of the same crop leads to land degradation. This is also called the continuous-cropping obstacle. Currently, intercropping tobacco with other crops can serve as an effective strategy to alleviate continuous cropping obstacles.

Results

In this study, tobacco K326 and insectary floral plants were used as materials, and seven treatments of tobacco monoculture (CK), tobacco intercropped with Tagetes erecta, Vicia villosa, Fagopyrum esculentum, Lobularia maritima, Trifolium repens, and Argyranthemum frutescens respectively, were set up to study their effects on rhizosphere soil chemical properties and composition and structure of rhizosphere soil microbial community of tobacco. The 16 S rRNA gene and ITS amplicons were sequenced using Illumina high-throughput sequencing. tobacco/insectary floral plants intercropping can influence rhizosphere soil chemical properties, which also change rhizosphere microbial communities. The CK and treatment groups tobacco rhizosphere soil microorganisms had significantly different genera, such as tobacco intercropping with T. repens and A. frutescens significantly increased the number of Fusarium and intercropping T. erecta, V. villosa, L. maritima, T. repens, and A. frutescens significantly increased the number of Sphingomonas and unknown Gemmatimonadaceae. Additionally, intercropping T. erecta, V. villosa and L. maritima changed the rhizosphere fungal and bacteria community and composition of tobacco and the positive correlation between tobacco rhizosphere the genera of fungi and bacterial were greater than CK. The pathway of the carbohydrate metabolism, amino acid metabolism, and energy metabolism in rhizosphere bacteria were significantly decreased after continuous cropping. Fungal symbiotic trophic and saprophytic trophic were significantly increased after intercropping V. villosa, L. maritima and plant pathogen and animal pathogen were increased after intercropping T. repens and A. frutescens. Additionally, bacterial and fungal communities significantly correlated with soil chemical properties, respectively.

Conclusion

This study reveals that intercropping tobacco with insectary floral plants, particularly T. erecta, V. villosa, L. maritima and A. frutescens significantly affects soil chemical properties and alters rhizosphere microbial communities, increasing the abundance of certain microbial genera. Additionally, intercropping enhances pathways related to carbohydrate, amino acid, and energy metabolism in rhizosphere bacteria. These findings suggest that intercropping could provide a promising strategy to overcome challenges associated with continuous tobacco cropping by regulating the rhizosphere environment.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12866-024-03597-7.

Background

Continuous cropping is commonly employed in modern intensive agricultural production as a means to enhance crop yields [1, 2]. The practice of continually cultivating the same or similar crop in the same soil year after year enables the optimal utilization of local soil, climate, and other natural resources [2, 3], and this usually leads to severe continuous-cropping obstacle. Generally, the continuous-cropping obstacle induced by excessive cultivation is regarded as one of the primary types of land degradation [4]. This degradation is accompanied by a decline in soil quality, manifesting as loss of soil carbon, nutrient depletion, reduction in land fertility, decreased biodiversity, soil acidification, salinity, compaction, and alterations in soil microbial communities, thereby creating challenges for the implementation of continuous cropping [5–10]. To date, large tracts of cropland are imperiled by continuous-cropping obstacles, which manifest as stunted plant growth, diminished plant vigor, and reduced crop productivity [5, 9]. For example, long-term monoculture leads to a decline in the yields of maize, wheat, medicinal plants, and vegetable crops such as cucumbers [11–14]. It also results in a reduction of nitrogen, phosphorus, and potassium levels in the soil, causing an imbalance in nutrient uptake by plants [5, 8]. Additionally, it accelerates the accumulation of plant toxins, such as phenolic acids, which promotes the prevalence of plant diseases [5, 15].

Chinese tobacco is an economically important crop that faces difficulties when grown continuously [16]. Given the constraints of limited land resources and the absence of sustainable cropping systems, tobacco cultivation in most mountainous areas in southwest China relies heavily on a continuous cropping approach [10]. In continuous tobacco cropping systems, tobacco selectively absorbs certain soil nutrients, leading to declines in soil mineral nitrogen (NH₄⁺-N + NO₃⁻-N), Olsen-P (available phosphorus), organic carbon (SOC), total nitrogen (TN), and pH levels, resulting in an imbalance of soil nutrients [10, 17]. The exudates of the tobacco root contain specific bioactive substances, including secondary metabolites and aroma components, which have been demonstrated to facilitate the proliferation of harmful bacteria, such as Ralstonia solanacearum, while simultaneously reducing the population of beneficial bacteria, including Arthrobacter and Lysobacter [18, 19]. This distinctive trait of tobacco, particularly the presence of allelopathic substances that are primarily secondary metabolites, plays a pivotal role in the difficulties associated with continuous cropping. As identified by Ren et al. [20], these allelochemicals are responsible for autotoxic allelopathy, a phenomenon that poses significant challenges in agricultural practices. Therefore, continuous cropping obstacles may be caused by the deterioration of soil physicochemical properties, nutrient imbalance, allelopathy, and the imbalance of soil microorganisms [10, 21, 22]. Among these factors, soil microorganisms have been confirmed to be one of the important influences on tobacco growth [23]. In conclusion, to address soil degradation resulting from continuous monoculture and the loss of rhizosphere soil microbial communities, it is essential to explore innovative solutions in the context of climate change and biodiversity loss. This approach aims to enhance the sustainability and productivity of tobacco agriculture [24].

Soil microorganisms play a pivotal role in the functioning of the soil ecosystem, influencing processes such as nutrient cycling, organic matter turnover, soil structure maintenance, and toxin degradation [10, 25]. They enhance and maintain soil fertility, functioning as effective indicators of soil health [17, 26]. Furthermore, they are indispensable for the biological, chemical, and physical processes that significantly impact soil quality, fertility, and productivity [23]. The microbes in the rhizosphere affect the uptake and conversion of soil nutrients [27]. This implies that the richness and diversity of rhizosphere microorganisms play significant roles in the growth, development, and health of plants [28]. In parallel, plant root exudates can shape the rhizosphere microbiome and influence the composition of the microbial community [29]. Continuous cropping of a single species often leads to an imbalance in the soil microbial flora [30]. However, intercropping and crop rotation, which are traditional agricultural practices, have long been recognized as essential and sustainable approaches in various regional agroecosystems [12, 31, 32]. It has been demonstrated that reasonable intercropping and rotation systems can reduce the occurrence of tobacco diseases and enhance the planting efficiency. For example, maize and lily crop rotation with tobacco significantly changed the composition and structure of soil bacterial community and effectively reduced the occurrence of tobacco bacterial wilt [33], and the intercropping of tobacco and soybean could also reduce the occurrence of tobacco black shank [34]. In addition, other studies have also demonstrated that intercropping, compared to monocropping, can enhance crop diversity, thereby increasing the diversity of root secretions. This alteration in root secretions affects the composition and quantity of soil microorganisms, contributing to the restoration of soil ecosystems [35]. Thus, diversified crop cultivation has demonstrated advantages in promoting the abundance and diversity of soil macro- and microorganisms, as well as enhancing crop yields. So far, in China, the cultivation of insectary floral plant strips (IFPS) has become a common practice in agricultural ecosystem habitat management [36]. In comparison to conventional techniques such as intercropping, crop rotation, or other cultivation management methods, the implementation of insectary floral plant strips (IFPS) has the potential to not only enhance insect and vegetation diversity in agricultural ecosystems, but also to positively influence the structure of soil rhizosphere microbial communities and to prevent the occurrence of continuous cropping obstacles. Tschumi et al. [37] demonstrated that the planting of seven flowering plants from the families Asteraceae, Chenopodiaceae, Apiaceae, and Papaveraceae in winter wheat fields is an effective method for reducing the population of cereal leaf beetles. This suggests that flower strips can be an effective method for pest control, reducing pest levels below economic thresholds and providing a viable alternative for pest management. Meanwhile, Zhou et al. [38] also found that tobacco and sweet potato intercropping helps improve soil microbial structure and enzyme activity, thereby optimizing nutrient availability and microbial diversity in the rhizosphere soil. This improvement contributes to maintaining plant productivity and ensuring ecological sustainability. Furthermore, an investigation into the effects of intercropping Salvia miltiorrhiza with tobacco revealed that the intercropping of Salvia miltiorrhiza led to notable enhancements in the nutrient status of tobacco field soils, as well as in the composition and structure of soil microorganisms. This resulted in an enhancement of the quality of flue-cured tobacco, as well as an increase in the proportion of premium tobacco leaves and an improvement in planting efficiency. Ultimately, these integrated effects served to alleviate the obstacles caused by continuous tobacco cropping [39]. Currently, insectary floral plants strips are not only employed in orchards [40], vegetables [41], and grain crops [37], but their application practices and research reports on economic crops are relatively limited. Especially under intercropping systems, studies on changes in tobacco rhizosphere soil microbial communities and their relationships with soil properties are scarce.

Insectary floral plants, also known as flowering plants or honey plants, are defined as plant species that can provide pollen, nectar, or extrafloral nectar for natural enemies, especially parasitic natural enemies [42]. The planting of a diverse array of flowering plants has been demonstrated to be a viable and impactful ecological agricultural strategy, as it provides rich nutrients that enhance the vitality of natural enemy insects [43, 44]. The attractiveness of flowering plants to natural enemies varies greatly [45–47]. When selecting flowering plants, several factors should be considered: (1) The candidate plants should have traits that are compatible with the target natural enemies and crop systems [48, 49]; (2) Flowering plants should bloom early in the crop cycle to attract natural enemies before pest infestations occur, especially for crops with short growing seasons, early-flowering plants should be chosen, and vice versa [46]; (3) The selected flowering plants should attract fewer pests compared to the crop itself [50]; (4) The cost of planting flowering plants, field management measures, and farmers’level of mastery of the technique should be considered [51]. In this study, the selected floral plants, Tagetes erecta (L.), Vicia villosa (Roth.), Fagopyrum esculentum (Moench), Lobularia maritima ((L.) Desv.), Trifolium repens (L.), and Argyranthemum frutescens ((L.) Sch.-Bip), are chosen due to their ability to flower during the entire growth cycle of tobacco. In addition to attracting natural enemies to control pests in tobacco fields, these plants also alter the types and quantities of soil microorganisms, thereby promoting the restoration of soil ecological balance [52–54].

In the current study, we hypothesize that intercropping different floral plants with tobacco can alleviate soil degradation in tobacco continuous cropping systems by improving rhizosphere soil properties and microbial communities, and reducing microbial nutrient limitations. Moreover, there are notable differences in rhizosphere soil microbial communities and nutrient limitations between monocropped tobacco fields and those intercropped with different floral strips. To test this, we studied the differences in the chemical properties and microbial communities of the rhizosphere soils of tobacco intercropped with different floral plants and those of tobacco under long-term monoculture.

Materials and methods

Study sites

We have selected an experimental field site in Wanqiao County, Dali City in the west of Yunnan Province, China (25°48’34"N, 100°9’0"E). Dali belongs to the northern subtropical plateau monsoon climate. The local farmland soil classification is dark-brown earths, with an altitude of 1974.31 m, annual average sunshine hours of 2253.9 h, and annual average temperature of 15.0 ℃; The annual average precipitation is 1065.7 mm, and the frost-free period is about 240 days. The altitude, temperature, frost free period and rainfall in this area can satisfy the cultivation and growth of N. tabacum. The fields at the site were continuously planted with tobacco plants of the cultivar K326 for the last 25 years. During the whole tobacco field period, film mulching cultivation in dry land is adopted. Conventional chemical control methods are mainly used for disease and pest control, supplemented by agricultural control methods such as topping and forking, and manual weeding. Water and fertilizer are irrigated by water diversion canal, applied fertilizer twice in the whole growth period of tobacco, and irrigated the tobacco roots with humic acid containing water-soluble fertilizer (Yunnan Shunfeng Erhai Environmental Protection Technology Co., Ltd.) at the volume ratio of water to fertilizer of 10:1 two weeks after transplanting and at the tobacco rosette stage. Insecticides (Imidacloprid, Abamectin), fungicides (Triazolone, Bordeaux mixture) and humic acid containing water-soluble fertilizer for tobacco used in tobacco field control are uniformly distributed and used under the guidance of Wanqiao Tobacco Work Station of Dali Tobacco Company, Dali City.

Experimental design

Experimental plot design

The seven treatments (including the control) were laid out in a randomized complete block design (Fig. 1). The seven treatments were (1) tobacco plot with tobacco border strip (CK); (2) tobacco plot with an Tagetes erecta (marigold, “Fengfu”series) border flower strip (T-T); (3) tobacco plot with an Vicia villosa border flower strip (T-V); (4) tobacco plot with an Fagopyrum esculentum (buckwheat) border flower strip (T-F); (5) tobacco plot with an Lobularia maritima (sweet alyssum, “Carpet of Snow”) border flower strip (T-L); (6) tobacco plot with an Tifolium repens (white clover) border flower strip (T-R); (7) tobacco plot with an Argyranthemum frutescens (marguerite, “Red Candy”) border flower strip (T-A). Three rows of each border flower strips are intercropping with the tobacco.

Fig. 1.

Spatial layout of the field experiment in Xinxiyi Village Wanqiao Town, Dali City, Yunnan Province, China. Note: CK: Flue-cured tobacco plots with flue-cured tobacco border strips; T-L: Flue-cured tobacco plots with Lobularia maritima (sweet alyssum)border strips; T-T: Flue-cured tobacco plots with Tagetes erecta (marigold) border strips; T-F: Flue-cured tobacco plots with Fagopyrum esculentum (buckwheat) border strips; T-R: Flue-cured tobacco plots with Trifolium repens (white clover) border strips; T-V: Flue-cured tobacco plots with Vicia villosa border strips; T-A: Flue-cured tobacco plots with Argyranthemum frutescens (marguerite) border strips (the same below). Sizes are indicated in X and Y axes (in meters). The red five-pointed star indicates the sampling point of flue-cured tobacco rhizosphere soil

Establishment of the tobacco crop and floral plants strips in experimental fields

Two planting patterns were established: tobacco intercropped with floral plants (polyculture) and tobacco alone with tobacco strips (monoculture). The whole test plot was 45.50 m wide and 79.30 m long, and the row was oriented from south to north. The tobacco monoculture (CK) as the control and tobacco intercropped with marigold, Vicia villosa, buckwheat, sweet al.yssum, white clover, and marguerite flower strips as the treatment. Each treatment was repeated three times, with a total of twenty-one plots. Each plot area was 81 m², and placed in a random arrangement on the field. Each polyculture plot contains 5 rows of tobacco plants with a row spacing of 120 cm, and a plant spacing of 50 cm between each plant, while each monoculture plot contains 8 rows of tobacco plants with the same row and plant spacing. Additionally, in a tobacco field with intercropping strips of flower plants, 3 rows of flower plants are interspersed on the east and west sides of the tobacco field. The rows of tobacco and floral plant strips are 20 to 30 cm apart. The spacing between plots is at least 5 m to prevent interaction, as shown in Fig. 1 [55].

All of floral insectary plants were transplanted in the tobacco fields. Buckwheat has a row spacing of 20 cm × 30 cm, Vicia villosa has a row spacing of 10 cm × 20 cm, marigold has a plant spacing of 45 cm × 50 cm when transplanted. Marguerite has a plant spacing of 10 cm × 10 cm, and sweet alyssum has an average of 8–10 transplanted seedlings per hole. White clover has a plant spacing of 10 cm × 25–30 cm when transplanted. The seedlings of Argyranthemum frutescens were purchased from Kunming Huaxianzi Gardening and Greening Engineering Co., Ltd., while the seedlings of other insectary floral plants were self-propagated.

In all tobacco plots, we planted K326 on May 8th in 2022. These tobacco seedlings were also planted at a rate of 9 plants/m². The tobacco was harvested in mid-September. Fagopyrum esculentum, Vicia villosa and Tifolium repens seeds were sown on April 23th, 2022, and the seedlings of them came out on May 1st, 2022. Tagetes erecta, Lobularia maritima and Argyranthemum frutescens were transplanted on April 30th, 2022. The weeds in flower strips were pulled out manually.

Collection and treatment of rhizosphere soil in tobacco field

At the mature stage of tobacco, collect the rhizosphere soil in the rows between insectary floral plants and tobacco. Shake off loose soil around tobacco plants intercropped with insectary floral plants (treatment group) and tobacco plants not intercropped with insectary floral plants (CK group), and the attached soil was brushed gently and collected [56]. A four-point sampling method was used to collect the soil from each plot, and finally mixed all soil samples from one plot and then divided it into three replications. Finally, the samples were transported to the laboratory in ice bags. A portion of the samples were air dried for soil physical and chemical analysis, while another portion was kept at 4℃ for soil microbial assessment.

The chemical properties were assessed based on previous studies [56, 57]. The content of total nitrogen (TN) in rhizosphere soil (tightly attached to the tobacco root) of tobacco field was determined by sulfuric acid-accelerant digestion method. The content of total phosphorus (TP) in rhizosphere soil was determined using NaOH fusion and molybdenum-antimony anti-spectrophotometry method .The content of total potassium (TK) in rhizosphere soil was determined by NaOH alkali melting and flame spectrophotometry. A pH meter (FE28, METTLER-TOLEDO, USA) was utilized to measure the soil pH in a soil-water suspension (1:2.5, air-dried soil/distilled water to eliminate CO2) following 30 min of shaking. The contents of available N (AN), available P (AP), available K (AK), and soil organic matter (SOM) were determined using the alkaline hydrolysis diffusion method [58], sodium bicarbonate extraction followed by Mo-Sb colorimetry [59], ammonium acetate extraction with flame photometry using a flame spectrophotometer (FP6450, XinYi Instrument, CN) [60], and the potassium dichromate titrimetric method [61] (Wang et al., 2014), respectively. A conductometer (Bante902P, Bante, CN) was utilized to measure the rhizosphere soil electrical conductivity (EC).

DNA extraction and MiSeq sequencing

DNA extraction and MiSeq sequencing refer to the study of Wang et al. [56]. Rhizosphere microorganism genomic DNA extraction, along with 16 S rRNA gene and ITS gene sequencing, was conducted with the assistance of Majorbio Co. Ltd., Shanghai. The microbial DNA from the rhizosphere soil was isolated using the Qiagen E.Z.N.A.^® Soil DNA Kit (Omega Bio-Tek, USA). The 16 S rRNA gene and ITS rDNA were amplified from the extracted DNA using bacterial primers, 338 F/806R [62], and fungal primers, ITS1F/ITS2R, respectively [63]. The PCR product was analyzed using 2% agarose gel electrophoresis, purified with the AxyPrep DNA Gel Extraction Kit (AXYGEN, Union, CA), and quantified using QuantiFluor™-ST (Promega, Wisconsin, USA). The combined amplicon library underwent sequencing on the Illumina MiSeq platform, utilizing the TruSeqTM DNA Sample Prep Kit (Illumina, USA) and adhering to the manufacturer’s guidelines. The Flash (version 1.2.11) and Trimmomatic programs were utilized to filter and trim the raw sequences, which included quality trimming, chimera detection, and removal. The Silva database (Release 138) was employed for aligning the sequences of 16 S rRNA gene data [64], while the Unite database (version 8.0) was used for aligning ITS gene data [65]. The Silva and Unite databases were utilized to align the operational taxonomic units (OTUs), thereby acquiring species taxonomic information for each OTU. The OTU table underwent manual filtering, specifically eliminating chloroplast sequences from all samples. The fungal and bacterial sequencing data are uploaded into the Sequence Read Archive (SRA) of NCBI (http://www.ncbi.nlm.nih.gov/sra) and can be accessed through Accession number PRJNA1058964 (Accession number: SRR27378722, SRR27378733, SRR27378738—SRR27378754, SRR27378756, SRR27378757; SRR27385172, SRR27385177—SRR27385193, SRR27385195, SRR27385196, SRR27385220).

Fungal and bacterial functional analyses

The software PICRUSt2 (Version 2.2.0) and FUNGuild (Version 1.0) were employed for the functional prediction of 16 S and ITS amplicon sequencing outcomes, respectively. The metagenomes were predicted from 16 S data using Phylogenetic Investigation of Communities by Reconstruction of Unobserved States (PICRUSt2) [66]. The functional genes were identified from the Kyoto Encyclopedia of Kyoto Genes and Genomes (KEGG) database [67]. FUNGuild was used to obtain fungal functional guild data [68].

Statistical analyses

The operational taxonomic units (OTUs) with a 97% similarity threshold were grouped using Usearch (version 11.0) in accordance with the methodology outlined by Edgar [69]. Mothur (vision 1.30.2) software was used to calculate Alpha diversity index and sequencing depth Coverage of soil rhizosphere microbial communities in tobacco fields. The Qiime (vision 1.9.1) software for Beta diversity distance matrix is calculated [70], then principal coordinate analysis (PCoA) was performed for the samples using R software (version 3.3.1), and statistical analysis of ANOSIM inter-group and intra-group differences was performed. Based on the proportion of microorganisms in each treatment sample at phylum and genus level, R language was used to draw the relative abundance map of each microorganism in each rhizosphere soil sample, and the differences of microbial relative abundance between the peri-species treatment and non-peri-species treatment were analyzed respectively at phylum and genus level. Species with significant differences in sample classification were identified using linear discriminant analysis (LDA ≥ 3.5). The nonparametric Kruskal-Wallis (KW) sum-rank test and Wilcoxon rank-sum test were used to analyze the LDA and the nonparametric Kruskal-Wallis (KW) sum-rank test and Welch’s were used to analyze the test for the significance of differences between groups of samples. Using Nephi (vision 0.10.1) software, a co-occurrence network of tobacco rhizosphere soils microbial community structure correlation at the genus level was constructed [71]. Redundancy Analysis (RDA) was used and the significance of all chemical factors was evaluated using Monte Carlo permutations (permu = 999). The selection of environmental factors was determined by the functions envfit (permu = 999) and vif.cca, where all environmental factors with vif > 20 were excluded from further analysis [72]. The vif values of AN (vif = 33.70) and pH (vif = 30.53) were higher than 20 and removed. The RDA analyses were performed in R (version 3.3.1).

The one-way ANOVA (DMRT) and least significant difference (LSD) method was used to test the difference significance of the chemical properties of tobacco rhizosphere soil and Alpha diversity index of fungi and bacteria in rhizosphere soil, respectively. The Tukey’s multiple tests was used to test the difference significance of the relative abundance of trophic modes and mainly guilds assigned by FUNGuild for fungal communities. Differences were considered to be statistically significant if P < 0.05.

Results

Chemical properties of N. tabacum rhizosphere soil under the tobacco/insectary floral plants intercropping patterns

The chemical properties of collected tobacco rhizosphere soil from different sample were measured, the results of which are shown in Table 1. The changes of soil physical and chemical properties were different among different samples. The contents of TN and AP in T-A tobacco rhizosphere soil samples were the highest, 5.68 g•kg⁻¹ and 209.87 g•kg⁻¹, respectively, which were significantly increased by 23.21% and 95.05% compared with those in control (CK) samples (P<0.05). However, the pH of T-A soil sample was the lowest as well as was significantly different from that of CK (P<0.05). The AN, AK, SOM contents and pH value of T-L tobacco rhizosphere soil samples were the highest, significantly increased by 16.34%, 127.74%, 6.25% and 3.54% compared with CK, respectively(P<0.05), while the EC value was the lowest, 15.49% lower than CK. In T-R sample, EC was the largest, which was significantly increased by 190.30% compared with CK(P<0.05). The content of TK in CK was the largest and significantly different from that in other samples (P<0.05).

Table 1.

Effects of floral plants on chemical properties in N. tabacum rhizosphere

Sample	TN (g•kg -1)	TP (g•kg -1)	TK (g•kg -1)	AN (mg•kg -1)	AP (mg•kg -1)	Sample	AK (mg•kg -1)	SOM (g•kg -1)	EC (μs/cm)	pH
CK	4.61 ± 0.04d	1.45 ± 0.04abc	18.33±0.23a	361.33±2.33f	107.60±2.91de	CK	188.67 ± 5.24ef	65.53 ± 0.17bc	391.67 ± 10.84d	6.78 ± 0.04c
T-T	4.28 ± 0.05e	1.50 ± 0.04ab	17.37±0.07bc	367.33±4.41ef	119.37±1.28cd	T-T	201.67 ± 3.38e	62.58 ± 0.98c	731.00 ± 26.08b	6.78 ± 0.04c
T-V	4.66 ± 0.03d	1.37 ± 0.04c	17.53±0.19bc	373.33±3.18e	101.47±0.23e	T-V	153.33 ± 1.76f	62.57 ± 1.25c	677.67 ± 46.24b	6.82 ± 0.02bc
T-F	4.99 ± 0.04b	1.46 ± 0.02abc	17.47±0.15bc	394.00±1.00c	113.60±0.06d	T-F	311.33 ± 3.93c	66.78 ± 1.13ab	382.67 ± 7.22d	6.92 ± 0.01ab
T-L	5.08 ± 0.02b	1.40 ± 0.05bc	17.07±0.27c	420.00±1.53a	136.67±1.57b	T-L	429.67 ± 5.24a	69.56 ± 0.91a	331.00 ± 15.00d	7.02 ± 0.06a
T-R	4.88 ± 0.05c	1.51 ± 0.04ab	17.80±0.06ab	384.67±1.45d	130.20±8.47bc	T-R	246.33 ± 4.37d	62.52 ± 1.11c	1137.00 ± 24.02a	6.41 ± 0.04d
T-A	5.68 ± 0.01a	1.54 ± 0.02a	17.63±0.27bc	407.00±1.15b	209.87±5.71a	T-A	361.67 ± 34.07b	69.31 ± 1.36a	565.00 ± 11.59c	6.10 ± 0.05e
F	F(6,12)=172.504	F(6,12)=2.691	F(6,12)=3.782	F(6,12)=73.967	F(6,12)=99.071	F	F(6,12)=60.461	F(6,12)=9.040	F(6,12)=127.703	F(6,12)=63.325
P	P = 0.0001	P = 0.0682	P = 0.0238	P = 0.0001	P = 0.0001	P	P = 0.0001	P = 0.0007	P = 0.0001	P = 0.0001

Note: 1. Data in the Table 1 are mean ± SE. Different small letters in the same column indicate significant difference at P ＜ 0.05 level by one-way ANOVA (DMRT). 2. T-T, T-V, T-F, T-L, T-R and T-A represent the treatment of planting Tagetes erecta (marigold), Vicia villosa, Fagopyrum esculentum (buckwheat), Lobularia maritima (sweet alyssum), Trifolium repens (white clover) and Argyranthemum frutescens (marguerite) intercropping with tobacco, respectively (the same below). 3. CK means not planting other floral plants except tobacco (the same below)

The diversity and composition of tobacco rhizosphere soil microbial under the tobacco/insectary floral plants intercropping patterns

In order to assess the effects of intercropping different floral plants in tobacco fields on soil microbial communities, fungal primer ITS1F-ITS2R and bacterial primer 338 F-806R were used to sequence rhizosphere soil samples of different treatments with high throughput. The total number of sequences was 1,059,096 and 796,307, respectively. Among them, the effective sequence of each sample under ITS1F-ITS2R primer was 1,032,589, and the effective sequence of each sample under 338 F-806R primer was 406,186. Based on the 97% similarity level, 2734 and 5748 OTUs were obtained by bioinformation statistical analysis, respectively. The results of the rarefaction curve of soil fungi and bacteria show that (Fig. 2A, B) at the OTU classification level of 97% similarity, the rarefaction curve of Sobs index of each sample gradually tends to be flat, that is, the amount of sequencing data is sufficient to reflect the species diversity of soil fungi and bacteria communities.

Fig. 2.

Fungal (A) and bacterial (B) rarefaction curves in tobacco rhizosphere from seven soil samples

The obtained OTUs were subjected to comparative analyses in order to identify the diversity and richness of microorganisms in tobacco rhizosphere soil caused by various insectary floral plants (Table 2). In terms of fungal community richness (Chao and ACE indices), the rhizosphere soil fungal community in T-V samples had the highest richness, but there was no significant difference from CK. T-A sample has the lowest richness of fungal community in tobacco rhizosphere soil, and its richness is significantly lower than CK. For the bacterial community richness of rhizosphere soil in tobacco field, except that the T-R and T-A samples were significantly different from the CK and T-L samples, respectively, the other samples have no significant difference compared with CK. In terms of community diversity (Shannon and Simpson indices), there was no significant difference in rhizosphere soil fungal community diversity. But the Shannon indices of soil rhizosphere bacterial community in T-V and T-A samples were significantly lower than CK; the Simpson indices of soil rhizosphere bacterial community in T-A sample was significantly higher than that of CK.

Table 2.

Effects of function plants on α-diversity index of fungi and bacteria in N. tabacum rhizosphere

Treatments	Fungal
	Chao1 indices	ACE indices		Shannon indices	Simpson indices	Coverage
CK	1171.68 ± 58.42ab	1233.58 ± 95.5ab		4.46 ± 0.18ab	0.032 ± 0.008a	0.9947
T-T	1110.75 ± 38.57ab	1121.74 ± 46.46ab		4.49 ± 0.15ab	0.032 ± 0.006a	0.9952
T-V	1226.45 ± 61.09a	1252.91 ± 75.89a		4.23 ± 0.35ab	0.054 ± 0.028a	0.9945
T-F	1161.28 ± 60.98ab	1160.98 ± 53.22ab		4.64 ± 0.07a	0.023 ± 0.002a	0.9946
T-L	1079.38 ± 32.00b	1087.32 ± 35.87b		4.34 ± 0.26ab	0.049 ± 0.021a	0.9956
T-R	1175.26 ± 26.22ab	1186.27 ± 19.00ab		4.41 ± 0.04ab	0.030 ± 0.002a	0.9949
T-A	898.25 ± 56.79c	914.14 ± 63.34c		4.00 ± 0.01b	0.043 ± 0.002a	0.9958
F	F (6,12) = 6.802	F (6,12) = 5.193		F (6,12) = 1.300	F (6,12) = 0.757
P	P = 0.0025	P = 0.0075		P = 0.3283	P = 0.6163
Treatments	Bacterial
	Chao1 indices	ACE indices	Shannon indices		Simpson indices	Coverage
CK	3548.42 ± 34.07b	3586.88 ± 16.40b	6.68 ± 0.03a		0.004 ± 0.0003bc	0.9510
T-T	3754.69 ± 112.33ab	3822.89 ± 133.22ab	6.62 ± 0.07ab		0.004 ± 0.0004bc	0.9501
T-V	3688.30 ± 81.83ab	3976.92 ± 214.25ab	6.51 ± 0.07bc		0.005 ± 0.0005b	0.9451
T-F	3663.45 ± 68.69ab	3766.03 ± 65.58ab	6.70 ± 0.05a		0.004 ± 0.0003c	0.9472
T-L	3459.68 ± 24.26b	3504.48 ± 7.62b	6.57 ± 0.01ab		0.004 ± 0.0001bc	0.9520
T-R	3860.49 ± 24.95a	3925.04 ± 23.03ab	6.64 ± 0.02ab		0.004 ± 0.0001bc	0.9476
T-A	3741.88 ± 189.77ab	4212.43 ± 355.78a	6.41 ± 0.07c		0.006 ± 0.0005a	0.9497
F	F (6,12) = 1.824	F (6,12) = 2.172	F (6,12) = 3.973		F (6,12) = 6.316
P	P = 0.1768	P = 0.1191	P = 0.0201		P = 0.0034

Note: Data in the Table 2 are mean ± SE. Different small letters in the same column indicate significant difference at P＜0.05 level by LSD test

In a complement approach, we analyzed beta diversity of the rhizosphere soil microbial community of intercropping different floral plants in the tobacco field in order to explore potential community shifts (Fig. 3). Based on fungi and bacteria OUT level, PCA with the weighted UniFrac distance were carried out. The PCA score plot of fungi indicated that the T-T, T-A and T-R groups were closely related, and grouped to the left of the graph along PC2, which represented 15.07% of the total variation, whereas CK, T-V, T-F and T-L groups were separated from T-T, T-A and T-R along PC1, which accounted for 24.81% of the total variation (Fig. 3A). The PCA score plot of bacteria indicated that the T-A and T-R groups were closely related, and grouped to the left of the graph along PC1, which represented 26.46% of the total variation, whereas CK, T-T, T-V, T-F and T-L groups were separated from T-A and T-R along PC2, which accounted for 14.18% of the total variation (Fig. 3B). Overall, the two PCA axes explained 39.88% and 40.64% of variation between the fungal and bacterial different communities, respectively. By ANOSIM analysis, significant effects of different floral plants intercropped in tobacco field on rhizosphere soil fungal (Fig. 3C) and bacterial (Fig. 3D) community structure (β-diversity) were observed (R = 0.8735, P = 0.001 and R = 0.7511, P = 0.001 respectively).

Fig. 3.

PCoA of fungal (A) and bacterial (B) community in tobacco rhizosphere soil; ANOSIM of fungal (C) and bacterial (D) community in tobacco rhizosphere soil

The results of high-throughput sequencing showed that 15 phyla, 53 classes, 488 genera, and 794 species of fungi were present in all rhizosphere soil samples. The 10 phylum with the highest abundance were selected to generate a column stack diagram of relative species abundance (Fig. 4A). The overall fungal species in the samples were similar, but the proportion of species richness varied at different taxonomic levels. Ascomycota, Mortierellomycota and Basidiomycota were the three main phyla. The relative abundances of three phylum were the highest in the T-V treatment, reaching 96.69%, while they reached 95.44%, 94.83%, 94.88%, 93.45%, 95.43% and 96.32% in the CK, T-T, T-F, T-R, T-L and T-A treatments, respectively. The results showed that intercropped different floral plants in tobacco field changed the structure of the rhizosphere soil fungal flora. The relative abundance of the top10 species at the genus level is shown in Fig. 4B. The main genera were: Mortierella, Gibellulopsis, Gibberella, Thielavia, Saitozyma, Aspergillus, Retroconis, unclassified_k_Fungi, Fusarium and Pseudeurotium. These accounted for 46.49% of all fungi in all treatments, among which T-A accounted for the highest proportion (54.47%); CK, T-V, T-F, T-R and T-L treatment, accounted for 44.10%, 49.56%, 44.57%, 50.48% and 41.44%, respectively; and T-T accounted for the lowest proportion (40.59%). These results indicate that intercropped different floral plants changed the dominant genera of fungi.

Fig. 4.

Relative abundance of soil microbial phylum (fungi (A), bacteria (C)) and genus (fungi (B), bacteria (D)) in the rhizosphere of tobacco for different treatments (top 10). Note：“unclassified”in the figure means a scientific name that is not classified in taxonomy;“norank” indicates a scientific name that does not have that level in the classification

The result shows that 46 phyla, 147 classes, 1002 genera, and 2120 species of bacteria were present in all rhizosphere soil samples. The 10 phylum with the highest abundance were selected to generate a column stack diagram of relative species abundance (Fig. 4C). Actinobacteriota (26.53%), Proteobacteria (24.10%), Firmicutes (15.68%), Chloroflexi (11.41%) and Acidobacteriota (8.92%) were the dominant bacteria in the soil (with relative abundance > 5%), accounting for 86.64% of all bacteria. The relative abundance of Actinobacteriota was highest in the T-R treatment, whereas it decreased by 5.60% in the CK treatment. Similarly, the relative abundance of Proteobacteria peaked in the T-V treatment, but declined by 23.68% in CK. Firmicutes showed the greatest relative abundance in the T-L treatment, with a corresponding decrease of 23.37% in CK. Chloroflexi reached its maximum relative abundance in the T-T treatment, but decreased by 6.10% in CK. Acidobacteriota exhibited its highest relative abundance in the T-F treatment, with a decrease of 15.54% in CK. The relative abundances of the top 10 species at the genus level are shown in Fig. 4D. Bacillus, Defluviicoccus, norank_f__norank_o__Gaiellales, norank_f__norank_o__norank_c__KD4-96, norank_f__norank_o__Vicinamibacterales, Sphingomonas, norank_f__67 − 14, Gaiella, Phycicoccus and Arthrobacter were the main genera. The top 10 bacteria in relative abundance accounted for 28.37% of all bacteria, among which T-A accounted for the highest proportion (32.61%), CK, T-T, T-V, T-L and T-R accounted for 27.35%, 27.16%, 29.06%, 26.89%, and 29.83%, respectively, and T-F accounted for the lowest proportion (25.50%). Overall, intercropping different floral plants changed the structure of the soil microbial community and had a significant impact on the dominant microbial community.

LDA revealed the most characteristic genera of fungi and bacteria in the rhizosphere microbiota of N. tabacum under the tobacco/insectary floral plants intercropping patterns

To identify the featured genus associated with the different floral plants intercropping with tobacco we used LDA on microbial abundance profiles at the genus level. The LEfSE analysis (LDA>3.5) results identified 32 fungal genera with proportions of abundance in the different samples of 0.58% ~ 16.85%. Analysis of the different groups of fungi in the rhizosphere soil of each treatment showed that (Fig. 5A), Monocillium, Saitozyma, Apiotrichum, Leptodiscella were most featured genera in CK (P<0.05); Cladosporiun, Plectosphaerella, Alternaria, Epicoccum were most featured genera in T-T (P<0.05);Mortierella was most featured genera in T-V; Pseudeurotium, Westerdykella, Minimedusa, unclassified_f__Microascaceae, unclassified_f__Nectriaceae and Geotrichum were most featured genera in T-F (P<0.05); unclassified_f__Didymellaceae, Talaromyces, unclassified_c__Dothideomycetes, Naganishia, Cystofilobasidium, Stachybotrys were most featured genera in T-R (P<0.05); Kernia, Trichocladium, Cercophora, Cladorrhinum were most featured genera in T-L (P<0.05); Aspergillus, Retroconis, Fusarium, Penicillium, Papulaspora, Chaetomium, Pseudopithomyces were most featured genera in T-A (P<0.05).

Fig. 5.

Relative abundance of fungal (A) and bacterial (B) community Genus in different treatments tobacco rhizosphere samples. Note：“unclassified”in the figure means a scientific name that is not classified in taxonomy;“norank” indicates a scientific name that does not have that level in the classification

The LEfSe analysis (LDA > 3.5) results showed that 15 genera of fungi exhibited different abundances, relative abundances accounting for 2.5% ~ 16.72% of the total genera. Figure 5B shows the different groups of rhizosphere soil bacteria in each treatment. The results show that, norank_f__norank_o__Subgroup_17 was most featured genera in CK (P<0.05); Meiothermus, norank_f__norank_o__Saccharimonadales, norank_f__JG30-KF-CM45, Brevibacillus were most featured genera in T-T (P<0.05); norank_f__67 − 14 and Pseudolabrys were most featured genera in T-V (P<0.05); norank_f__Methyloligellaceae; Turicibacter, and norank_f__Caldilineaceae were most featured genera in T-L (P<0.05); Defluviicoccus, Sphingomonas, norank_f__norank_o__Gaiellales, norank_f__Gemmatimonadaceae and Streptomyces were most featured genera in T-A (P<0.05).

Tobacco/insectary floral plants intercropping changed tobacco rhizosphere soil microbial community and composition

To further investigate the effects of different floral plants on rhizosphere soil microorganisms of tobacco, we performed significant difference analysis on the composition of CK and treatment group tobacco rhizosphere microbial communities at the genera level. With respect to soil fungal composition, T-V, T-F, T-R and T-L significantly increased the genera of Mortierella, T-T, T-F, T-R, T-L and T-A significantly increased the genera of Aspergillus, T-A significantly increased the genera of Retrocomis, T-R and T-A significantly increased the genera of Fusarium, but CK significantly increased the genera of Saitozyma (Fig. 6A). Besides, under the tobacco/insectary floral plants intercropping patterns changed tobacco rhizosphere soil fungal community (Fig. 6C). CK and treatment groups were mainly clustered into 6 distinct modules in co-occurrence networks, which we examined to decipher module-trait relationships (Fig. 6C, Table S1). The ratios of positive correlations to negative correlations were increased in T-T, T-V, T-F and T-L compared to CK, especially, in T-V (71.62%) and T-L (70.97%), the positive correlation between tobacco rhizosphere the genera of fungi were greater than CK (57.74). (Table S1).

Fig. 6.

Tobacco/insectary floral plants intercropping induced changed tobacco rhizosphere soil microbial community and composition. Histogram showing significantly different fungal genera (A) and bacterial genera (B). C Tobacco rhizosphere soil bacterial community. D Tobacco rhizosphere soil fungi community Note: In co-occurrence network of rhizosphere soils, the node represents different genera in tobacco rhizosphere soil microorganisms, and the edges represent the correlation between different genera. Red edges indicate positive, green edges negative connections. ***P < 0.001; **P < 0.01; *P < 0.05

In the bacterial level, all treatment groups significantly increased the genera of Defluviicoccus, norank_f_67 − 14 and norank_f_Gemmatimonadaceae. However, for norank_f_norank_o_Gaiellales and Sphingomonas, all treatments except T-F were significantly higher than CK (Fig. 6B). Besides, under the tobacco/insectary floral plants intercropping patterns also changed tobacco rhizosphere soil bacterial community (Fig. 6D). The network was also assigned to 6 modules with CK and treatment groups. Compared with CK, the proportion of positive correlation and negative correlation increased in other treatments except T-R (Fig. 6D, Table S1). Especially, in T-L (57.07%) and T-T (58.83%), the positive correlation between tobacco rhizosphere the genera of bacterial were greater than CK (50.74). (Table S1).

Correlation analysis of tobacco rhizosphere soil microbial community and environment factors under the tobacco/insectary floral plants intercropping patterns

R vegan package was used for redundancy analysis (RDA) of fungal and bacterial communities at OTU level. The results of redundancy analysis show that, the change of soil chemical properties plays an important role in shaping the structure and composition of microbial community. In rhizosphere soil samples of tobacco fields with different insectary floral plant treatments, the changes of TP (r² = 0.5314, P = 0.001), AP (r² = 0.6619, P = 0.001), AK (r² = 0.3181, P = 0.018) and TN (r² = 0.3272, P = 0.024) had significant effects on the structure and composition of fungal community; however, the influence of TK (r² = 0.1421, P = 0.259), SOM (r² = 0.2360, P = 0.088) and EC (r² = 0.1455, P = 0.2680) changes is relatively small (Fig. 7A). The changes of soil AP (r² = 0.7652, P = 0.001),TN (r² = 0.5560, P = 0.004), TP (r² = 0.3473, P = 0.023) and EC (r² = 0.3408, P = 0.026) had significant effects on the structure and composition of bacterial communities; the change of AK (r² = 0.2494, P = 0.077), SOM (r² = 0.0891, P = 0.456)and TK (r² = 0.0308, P = 0.751) have relatively little effect on bacteria (Fig. 7B).

Fig. 7.

RDA analysis of fungal (A) and bacterial (B) communities and chemical characteristics of soil. The red arrows indicate the chemical properties of the soil. The position and length of the arrows represent the direction and intensity of the influence of the chemical indexes on the microbial community, respectively

Tobacco soil rhizosphere microbial community functional prediction under the tobacco/insectary floral plants intercropping patterns

To study the effect of different floral plants intercropping in tobacco field on the soil fungal function, we used FUNGuild to predict the trophic and floral mode of the fungal communities with different treatments. The results showed that of the 2734 fungal OTUs identified, 1065 were allocated to different trophic models by FUNGuild, accounting for 38.88% of all OTUs. Among these allocated OTUs, 63 (5.92%) were symbiotropic, 117 (10.99%) were pathotrophic, 531 (49.86%) were saprotroph, 64 (6.01%) were saprotroph-symbiotroph, and 17(1.60%) were pathotroph-symbiotroph, 186 (17.46%) were pathotroph-saprotroph-symbiotroph, 86 (8.08%) were pathotroph-saprotroph, and 1 (0.09%) was saprotroph-pathotroph-symbiotroph.

The trophic modes of fungi in rhizosphere soil of tobacco field under different floral plant intercropping treatments were classified and analyzed. The results showed that (Fig. 8A), In T-V, the relative proportion of symbiotic trophic fungi reached 0.96%, which was significantly higher than that of T-F, T-L and T-A (P<0.05), but had no significant difference with other treatments and CK. The proportion of saprophytic trophic fungi was the highest in T-L (42.20%) and the lowest in T-R (29.06%), but there was no significant difference among all treatments (P>0.05). The proportion of T-R in pathogenic trophic fungi was the highest (18.68%), and the proportion of T-F was the lowest (10.94%), and there was no significant difference among all treatments (P>0.05). For saprophytic trophic fungi, the dominant functional groups were soil saprophytic fungi, dung saprophytic fungi and undefined soil saprophytic fungi (Fig. 8C), among which there was no significant difference between dung saprophytic fungi and undefined soil saprophytic fungi among all treatments. However, the proportion of soil saprophytic fungi in CK was the largest (7.98%), which was significantly higher than that of T-F, T-L, T-R and T-A (P< 0.05), while there was no significant difference between them and T-V and T-T. As shown in Fig. 8D, the symbiotrophic mode groups primarily consisted of arbuscular mycorrhizal fungus (AMF), endophyte, and ectomycorrhizal fungus (ECM). However, there was no significant difference in the relative proportion of ECM among all treatments (P > 0.05). For AMF, the highest proportion was 62.98% in T-R, which was significantly increased by 130.02% compared with CK (P<0.05). As for endophytic fungi, the proportion of T-V was the highest (85.42%), which was significantly increased by 280.57% compared with T-T (P<0.05). The pathotrophic mode was used in the detection of uncultivated and different floral plants rhizosphere soil samples. The ratios of plant pathogens in T-L, T-R, T-A samples were significantly higher than those observed in the CK, while animal pathogens in T-L, T-R, T-A samples were significantly lower than that observed in the control samples (Fig. 7B).

Fig. 8.

The relative abundance of trophic modes (A) and mainly Guilds (B-D) assigned by FUNGuild for fungal communities. Bars with different letters are significantly different by Tukey’s multiple test (P<0.05). Error bars represent +/- standard error of the mean. AMF: Arbuscular mycorrhizal fungus, ECM: Ectomycorrhizal fungus

Based on PICRUSt2 software, the functional analysis of bacteria was carried out to obtain the functional prediction information of different samples of bacteria. The results were compared using the KEGG (kyoto encyclopedia of genes and genomes) database. The samples were related to six biological metabolic pathways: metabolism, genetic information processing, environmental information processing, cellular processes, human diseases and organismal systems. Metabolism, genetic information processing, and environmental information processing are the main components, accounting for 78.37%, 6.64%, and 5.43% respectively. The analysis of predicted gene functions at the second level identified 46 sub-functions. After removing human diseases and peculiar classifications, functional differences analysis was conducted using the remaining 26 sub-functions. (Fig. 9). The main functions of microorganisms include membrane transport (3.12%), amino acid metabolism (8.51%), carbohydrate metabolism (9.82%), replication and repair (2.51%), energy metabolism (4.66%) and metabolism of cofactors and vitamins (4.36%). In the treatment of T-A, T-L, T-F and T-T, there were significant differences in the abundance of the three-core resource metabolic pathways: carbohydrate metabolism, amino acid metabolism, and energy metabolism, compared to CK.

Fig. 9.

The heatmap displays predicted bacterial community functional categories (KEGG level 2) based on PICRUSt2 analysis. Rows represent 46 KEGG orthologous functions, and columns represent 7 samples. The intensity of colors in the heatmap represents the abundance of functional genes. The legend for color interpretation can be found in the online version of this article

Discussion

Tobacco/insectary floral plants intercropping changes tobacco rhizosphere soil nutrients

The practice of intercropping, a sustainable agricultural method, has been the subject of extensive study with regard to its potential to enhance soil fertility and crop productivity [16, 38]. One of the primary benefits of intercropping is the utilization of plant interactions to enhance soil physicochemical properties and enzymatic activities, thereby promoting crop growth and development [38]. In the present study, nine indicators (TN, TK, AN, AP, AK, SOM, EC, and pH) were observed with regard to their influence on soil nutrients in the rhizosphere of tobacco plants intercropped with different flowering plants. The practice of continuous cropping over an extended period of time frequently results in the depletion of soil nutrients [3, 13]. The soil chemical properties tested were found to be comparable between the monoculture and continuous cropping systems. However, the differences in total phosphorus (TP) between the various treatments were not statistically significant (Table 1). Nitrogen in the soil represents the primary source for plant uptake and utilization, and is also the principal limiting factor for high-yielding crops [57]. Meanwhile, soil organic matter (SOM) is of great importance for soil ecology, assisting in water conservation and providing essential nutrients such as nitrogen, phosphorus, sulfur, and micronutrients to plants and microorganisms. This is of great consequence for the maintenance of soil fertility and crop productivity [10]. Here, our study demonstrated that the T-A and T-L treatments substantially enhanced the TN, AN, and SOM levels in the rhizosphere soil of tobacco. These findings suggest that intercropping tobacco with L. maritima and A. frutescens, respectively, can enhance the soil fertility of the tobacco rhizosphere. Consequently, the continuous cropping obstacle may result in a reduction of SOM content and soil fertility. The notable elevation in these three soil variables can be ascribed to the intercropping impact of L. maritima and A. frutescens with tobacco, as observed under the T-A and T-L treatments. The intercropping of tobacco with L. maritima and A. frutescens, respectively, has been demonstrated to reinforce the ecological interactions within the soil microenvironment, with the potential to promote nutrient cycling and the transfer of nitrogen between plants, thereby enhancing the efficiency of nitrogen absorption in the rhizosphere of tobacco [57]. Additionally, intercropping may facilitate the decomposition of plant and animal residues, thereby increasing the organic matter content in the soil [73]. Meanwhile intercropping may modify the structure of the rhizosphere microbial community, elevate nitrogen content in the soil, and stimulate the secretion of root exudates (including carbohydrates, alcohols, and phenolics), ultimately resulting in an increase in beneficial bacteria (such as Bacillus, Fig. 4D, T-A and T-L) in the rhizosphere [53, 74]. At present, the analysis of the effects of T-A and T-L treatments remains speculative. To gain a full understanding of the impact of intercropping L. maritima and A. frutescens with tobacco on the nitrogen content in the rhizosphere soil and the organic matter content of the soil, further in-depth research is required in related fields.

A significant increase in soil pH and salt ion concentrations will result in a deterioration of the environment necessary for plant survival and growth, as well as nutrient deficiencies in the soil, which will greatly hinder plant growth, development and yield [57]. Intercropping does indeed change soil pH, with the degree of change primarily influenced by nitrogen levels [57]. Increased nitrogen is an important factor in soil acidification [75]. In our study, we found that the pH of tobacco rhizosphere soil under T-A treatment was significantly lower than that in monoculture tobacco fields, which may be closely related to the contents of TN and AT in the tobacco rhizosphere soil. Previous research has shown that higher plant diversity can lead to increases in soil pH, organic carbon, and nitrogen levels, while also resulting in a decrease in EC [76–78]. Similar results were observed in our study: The tobacco/sweet al.yssum intercropping system increased the soil AN, pH and SOM, and decreased soil EC (Table 1). However, T-R treatment significantly increased the EC of the tobacco rhizosphere soil. This can be attributed to the following two aspects: (1) Under the intercropping pattern of legumes, legumes release protons, organic acids, and acid phosphatases into the rhizosphere to enhance soil phosphorus forms, thereby promoting the growth of intercropped plants and their uptake and utilization of soil phosphorus, which may lead to an increase in soil electrical conductivity [79–81]; (2) Changes in soil electrical conductivity are associated with changes in soil microbial community structure. Intercropped legumes may indirectly affect soil electrical conductivity by altering the structure of the soil microbial community in the rhizosphere [82]. These results suggest that the effect of intercropping on crop rhizosphere soil metabolic processes is actually more complex. In intercropping systems, a more in-depth analysis of rhizosphere soil microbial communities and investigation of the specific effects of different crop root exudates on soil fertility may be potential research directions and development priorities for the future.

Phosphorus (P) and potassium (K) are two important nutrients necessary for plant growth and development. Although total phosphorus and total potassium levels in soil are relatively high, the bioavailability of these elements in soil is often inadequate because they tend to form insoluble complexes with other elements in the soil [83, 84]. The presence of these insoluble complexes limits the uptake and utilization of these essential nutrient elements by plants, thereby affecting plant growth and yield. Phosphate solubilizing microorganisms (PSMs) and potassium solubilizing microorganisms (KSMs) can convert insoluble phosphorus and potassium in the soil into plant-available forms by producing organic acids, lowering soil pH, acidolysis, chelation and exchange reactions [84, 85]. These microorganisms not only increase the bioavailability of phosphorus and potassium in the soil, but also further promote plant growth and development by facilitating the uptake and use of other nutrients by plants and stimulating the production of plant hormones [86, 87]. In intercropping systems, root exudations from different crops may change the soil environment, which affects the availability of total available phosphorus and quick available potassium. Therefore, in this study, in the T-A treatment, the TP and AP contents in the rhizosphere soil of tobacco reached their highest values. This phenomenon indicates that the root exudates of A. frutescens, or the phosphate solubilizing microorganisms (PSMs) induced by them, may effectively promote the solubilization of insoluble phosphorus in the soil and significantly increase the content of available phosphorus in the rhizosphere soil, which in turn leads to an increase in the total phosphorus content in the rhizosphere soil. In addition, the decrease in rhizosphere soil pH under this treatment may also be closely related to this phenomenon. At the same time, different intercrop combinations may affect the availability of total and available potassium in the soil by altering the structure of soil microbial communities and the composition of root exudates. For example, in the T-L treatment, KSMs could release available potassium by decomposing organic matter, resulting in the highest levels of available potassium in this treatment. However, the TK content is highest in the monoculture tobacco fields, indicating that potassium may not have been fully released in the monoculture system.

In the soil rhizosphere microenvironment, the interactions among soil properties, soil microorganisms, and plant roots form an integrated system of mutual coordination and dependence. In intercropping systems, the close interactions and cooperation among these three elements collectively facilitate the transmission of signals within the soil, promote the cycling and utilization of nutrients, and enhance the adaptability of plants to environmental changes. Soil physicochemical characteristics play a crucial role in determining soil fertility [88]. In two planting systems, the chemical properties of rhizosphere soil are correlated among various indicators, except for two indicators TK and EC (Supplementary Figure S1, P < 0.05). In T-F, T-L, T-R and T-A treatment (Table 1), TN of the tobacco rhizosphere soil is positively correlated with AN, AP, and SOC indicators, indicating that intercropping with buckwheat, sweet al.yssum, white clover and marguerite can increase the content of TN, AN, AP, and SOC in rhizosphere soil. This result is similar to the findings of the study of Jin et al. [16]. In our research, the AK levels in the T-F, T-L, T-R and T-A treatment exhibit a positive correlation with AN and SOC as well as were higher under tobacco-flower intercropping system compared to the continuous cropping (Supplementary Figure S1; Table 1, P < 0.05). Additionally, in this study, we employed RDA and conducted Spearman’s correlation analyses to illustrate the connections between soil chemical properties and microbial communities (Fig. 7). The dominant bacterial taxa that experienced negative regulation due to intercropping with different insectary floral plants were negatively correlated with soil pH, total nitrogen (TN), available nitrogen (AN), available potassium (AK), soil organic matter (SOM), and total phosphorus (TP) (Figure S3B, C). On the other hand, the dominant fungal taxa only showed negative correlations with soil pH, TN, and TP (Figure S3A, B). The results of the current study are in line with previous findings demonstrating the importance of soil pH in shaping the construction of microbial communities [3, 89]. It is noteworthy that certain Sphingomonas also exhibit culturability as endophytic and rhizospheric bacteria, establishing interaction patterns with plants [90]. Here, we found that Sphingomonas was significantly positively correlated with AP and negatively correlated with rhizosphere soil pH value (Fig. 3D), which was consistent with the previous findings [91, 92]. Yao et al. [93] discovered that as the soil pH decreased, there was an increase in the relative abundance of Acidobacteria and our findings are consistent with this result. In additional, in this study, it was observed that the relative abundance of Fusarium varied significantly with changes in soil pH (Figure S3B). This suggests that intercropping with buckwheat and alyssum in tobacco fields can reduce the abundance of Fusarium in the soil, thereby decreasing the occurrence of tobacco root rot caused by continuous cropping. This finding may provide a practical basis for mitigating the issues caused by continuous tobacco cropping. However, further long-term research is needed to determine the feasibility of this intercropping measure.

Tobacco/insectary floral plants intercropping has the potential to enhance sustainable tobacco cultivation by antagonizing harmful microorganisms

In addition to the influence of soil’s chemical properties on soil fertility, soil microbes, as the primary drivers of the biogeochemical cycles of elements in soil, also directly or indirectly affect soil fertility, thereby influencing plant growth [94, 95]. It has been observed that there are variations in soil microbial communities between monoculture and diversified cropping systems [96]. In this study, the principal coordinate analysis (PCoA) demonstrated a distinct separation of the rhizosphere soil fungal and bacterial communities in the continuous tobacco cropping system when intercropped with marigold, white clover and marguerite (Fig. 3). This observation was further supported by hierarchical clustering (Figure S2).

The decrease in fungal diversity can elevate the risks and mortality rates attributed to plant diseases. On the contrary, an increase in soil fungal diversity has the potential to mitigate plant pathogenesis [13]. Typically, soil microbial biomass, functionality, and variety were found to be diminished in monoculture systems compared to diversified cropping systems [97, 98]. Here, we discovered that marigold, Vicia villosa, buckwheat and sweet al.yssum intercropping with tobacco increased the soil fungal community’s diversity. Hence, continuous tobacco cultivation may lead to the weakening of rhizosphere fungal resistance to the environment, which was not conducive to the sustainable development of soil. In this study, we discovered that Ascomycota, Basidiomycota, and Mortierellomycota were the prevailing phyla. Ascomycota are soil fungi that function as important decomposers in natural ecosystems, and their presence is affected by plant species and cropping systems [3, 99]. Intercropping marigold, sweet al.yssum and marguerite increased abundance of Ascomycota [99–101]. Mortierellomycota is a unique phylum predominantly composed of soil-dwelling saprotrophs [102, 103]. Continuous cropping decreased abundance of Mortierellomycota, which could solubilize mineral phosphorus in the soil and enhance soil nutrient concentrations through the synthesis and secretion of oxalic acid [104, 105]. However, this is inconsistent with our findings (Fig. 4A; Table 1), where in the T-A and T-V treatments, there is a negative correlation between the relative abundance of Mortierellomycota and AP in the soil. It was suggested that this disputed issue may be partially explained by the rapid absorption of soluble phosphorus by V. villosa, leading to the lowest AP content in the T-V treatment, as it promotes root nodule formation and nitrogen-fixing ability in leguminous plants [106]. Continuous cropping led to a higher presence of Basidiomycota, which could be linked to their involvement in the breakdown of plant litter [107]. The continuous cropping soil accumulated a greater amount of plant litter, and the active organic matter supplied diverse carbon substrates for the proliferation of Basidiomycetes [108]. At the genus level, Mortierella were the dominant genus in all treatments (Fig. 4B). However, as the years of continuous cropping increased, there was a continuous decline in the relative abundance of Mortierella. Prior research has revealed that Mortierella is capable of producing antagonistic arachidonic acid, which stimulates the production of phytoalexins and suppresses plant diseases [13]. Hence, Mortierella contributes to preserving the micro-ecological balance by suppressing soilborne pathogens [100]. The study revealed that Mortierella was most featured and difference genera in the control treatment in the rhizospheric soil of continuously planted tobacco (Fig. 4B). This indicates that the continuous cropping of tobacco has resulted in an imbalance in the fungal community in the rhizospheric soil, requiring substantial enrichment of Mortierella to control soilborne pathogens. Moreover, in T-T, T-A, T-L and T-A the abundance of potential pathogens was also increasing (Figs. 4B and 5A). Fusarium, a globally distributed soil-borne fungal pathogen, is present in multiple crop varieties [109, 110], leading to the rotting of plants, stems, flowers, and spikes. Fusarium-induced wilting and root rot are common soil-borne diseases affecting tobacco crops, posing a significant threat to tobacco production [99]. However, not all Fusarium species are harmful to crops. Ahmed et al. [111] found that some endophytic strains of the genus Fusarium are considered a treasure trove of natural products with novel structures for treating life-altering diseases. In addition, some studies have indicated that Fusarium fungi not only have positive effects on plant growth through secondary metabolites, but can also be used as biocontrol agents. First, the secondary metabolites of Fusarium fungi have a wide range of biological activities, including antibacterial, antiviral and antifungal effects [111, 112]. These secondary metabolites not only have an inhibitory effect on plant pathogens, but can also promote plant growth and improve plant resistance to pathogens [113].

Different intercropping combinations changed the bacterial community structure of tobacco rhizosphere soil (Fig. 4C, D). Some studies have explored the soil microbial bacterial community structure in continuous cropping systems [114, 115]. Numerous bacterial taxa have demonstrated their beneficial effects on crops by providing disease protection, enhancing plant growth, and triggering systemic pathogen resistance [116–118]. Collectively referred to as plant growth promoting bacteria (PGPB), these beneficial bacteria are a promising biotechnological approach for sustainable agriculture [119]. Rodriguez et al. [120] found that Actinobacteria, are oligotrophic microorganisms typically found in nutrient-poor environments. Taha et al. [121] reported that Proteobacteria have the ability to break down organic matter into oligosaccharides through the synthesis of enzymes (such as cellulases, chitinases, xylanases, and amylases), which in turn serve as carbon sources for other microorganisms (e.g., Acidobacteria). Species of the Firmicutes phylum have the capacity to break down hemicellulose and cellulose through the secretion of diverse enzymes (including xylanases, β-xylosidases, α-glucuronidases, and acetyl xylan esterases) and can also suppress plant pathogens by generating antibiotics (such as surfactin, iturin, fengycin, etc.) [122, 123]. Will et al. [124] found that Chloroflexi increased in nutrient-poor soil layers and was negatively correlated with soil nitrate concentration. Additionally, Chloroflexi has the ability to degrade organic residues [27]. Acidobacteria, crucial players in ecosystems, are highly abundant in soils and can effectively interact with other components to establish a stable ecological system [125, 126]. Here, in our study, we found that in T-T, T-V, T-L and T-R treatments, Proteobacteria, Firmicutes, Chloroflexi and Acidobacteriota are more abundant. These results indicated that plant growth promoting bacteria (PGPB) could directly or indirectly promote the growth of tobacco through different mechanisms. However, continuous cropping led to a higher presence of Bacteroidota, which could be linked to disease suppression [127]. Thus, in continuously cropped tobacco fields, the bacterial communities that exert antagonistic effects against harmful pathogens are maybe uniquely distinctive. At the genus level, Bacillus bacteria were the most prevalent (Fig. 4D), and they have been discovered to be advantageous for plant growth [128]. Previous research has shown that Bacillus casei strains can successfully colonize the rhizosphere of white mustard and alter the plant-associated microbial community [129]. Our study revealed that in T-A treatment, the structure of tobacco rhizobacterial communities, as assessed by 16 S amplicon data, was significantly altered compared to monoculture tobacco fields. Meanwhile, according to the results of LEfSe analysis, the genera that are most characteristic and show significant differences in the rhizosphere soil of tobacco under the T-T, T-V, T-R, and T-A treatments (Streptomyces, Sphingomonas, and Gemmatimonadaceae) not only promote crop growth but also help improve the composition of the soil microbial community. (Fig. 5B). For instance, Streptomyces show promise as biocontrol agents due to their ability to effectively combat a wide range of plant pathogens [130]. Sphingomonas have been shown to promote plant growth even in challenging environments with various stress conditions [131]. In addition, many Sphingomonas have properties that enhance plant growth and stress tolerance. These properties are attributed to their ability to fix nitrogen, solubilize phosphate, and produce plant growth hormones [132, 133]. Gemmatimonas have been shown to play a role in regulating phosphate metabolism and contributing to the stability of soil microbial communities [132]. While continuous cropping led to a notable decrease in their relative abundance, potentially impacting the growth and yield of tobacco. Recent studies have shown that PGPB do not promote plant growth in isolation; rather, they form PGPB consortia to alter the rhizosphere microbial community and interact synergistically with other beneficial microorganisms to enhance plant growth [119]. Therefore, in this study, the PGPB consortium may play a critical role in inhibiting harmful pathogens in continuous tobacco production systems.

Microbial community changes in continuous cropping systems are a complex and multifaceted problem. Currently, there are two views on microbiome changes in continuous cropping. Some researchers believe that in continuous cropping, host plants select for compatible taxa, thereby reducing microbiome diversity but increasing plant adaptability [134]. Other researchers have suggested that certain members of the microbial community show a reversal trend in continuous cropping, with their abundance or diversity increasing as continuous cropping progresses [9]. In this study, the α-diversity of fungi and bacteria in the rhizosphere soil of tobacco and bacterial community structure at the phylum and genus levels showed little variation between monoculture and intercropped tobacco fields (Table 2; Fig. 4). With regards to fungi, one unanticipated finding was that Ascomycota was highly abundant in T-T and T-A compared with that in CK groups. Mortierellomycota was highly abundant in T-V compared with that in CK groups. Mortierella was highly abundant in T-V, T-F, T-R, T-L compared with that in CK groups. Aspergillus was highly abundant in T-A compared with that in CK groups. In addition, based on Fig. 6 (A), which shows the most significant fungal genera, we found that Fusarium was significantly higher in continuous cropping tobacco fields compared to the T-T, T-V, T-F, and T-L treatments. These differences observed in different intercropping systems support previous reports that the abundance and structure of fungal communities are strongly influenced by continuous cropping [135, 136]. For bacteria, although the differences in abundance and diversity between continuous and intercropping systems are relatively small, these bacteria are largely classified as putative or potential plant growth promoting bacteria (PGPB). The interactions between these groups and the host plants may influence plant adaptability. Our study found that PGPB taxa are also present in the rhizosphere soil of continuously cultivated tobacco, suggesting that tobacco may select compatible microbial groups during continuous cultivation, thereby reducing microbial diversity and increasing its adaptability [137, 138].

Tobacco/insectary floral plants increase the levels of beneficial fungi in the tobacco rhizosphere and improve the metabolic function of rhizosphere bacteria in tobacco

The FUNGuild prediction results for tobacco rhizosphere fungi aligned with the research on Miscanthus rhizosphere functional groups conducted by Chen et al. [94]. The results mainly encompassed three trophic modes: pathotrophic, saprotrophic, and symbiotrophic (Fig. 8). The T-L and T-F samples were dominated by saprotroph, similar to the CK samples, although the proportions were different. LEfSe analysis of the composition fungal communities in the tobacco rhizosphere revealed that intercropping with various insectary floral plants could concomitantly enrich and enhance the proportions of Monocillium, Pseudeurotium, Plectosphaerella, Cladosporium, Westerdykella, Minimedusa, Geotrichum, Kernia, Trichocladium, Cercophora and Cladorrhinum etc. (Fig. 5A). The FUNGuild results indicated that these fungi were mainly categorized as saprotrophs, plant pathogens, pathotrophs, and the trophic mode that combines pathotrophic, saprotrophic, and symbiotrophic characteristics. Research has demonstrated that saprotrophic fungi are the main agents responsible for breaking down deceased or aging plants in the soil, and they play crucial roles in the decomposition of organic matter and the cycling of nutrients [94]. Hence, intercropping buckwheat and sweet al.yssum with tobacco may elevate the presence of saprotrophic fungi in the soil, consequently stimulating the generation of soil organic matter and enhancing the pace of nutrient cycling. In T-A, T-R, and T-L, plant pathogens were significantly higher compared to the control, indicating that intercropping marguerite, white clover, and alyssum may be increase the presence of potential plant pathogens while further confirmation is still needed. Smith and Read [139] has demonstrated that AMF can establish a mutually beneficial relationship with over 80% of land plants, improving the root system’s capacity for absorption and boosting nutrient levels for plant growth, including phosphorus and nitrogen. Here, in the T-R samples, AMF was significantly higher compared to the control treatment, which suggests that AMF may have facilitated tobacco growth. This is mainly due to the existence of a mutually-beneficial symbiotic relationship between AMF and tobacco. AMF (arbuscular mycorrhizal fungi) play an important role in plant growth. First, AMF significantly improves the nutritional status and growth of tobacco by promoting its uptake of phosphorus, especially when intercropped with white clover [140]. It also increases the levels of nitrogen, phosphorus and potassium in the rhizosphere soil (Table 1), further improving the effective use of phosphorus in the soil. This may also be one of the reasons for the increased electrical conductivity of the rhizosphere soil in the T-R treatment. Second, AMF regulates the secondary metabolic pathways of plants and improves their tolerance to disease and abiotic stress. For example, the study by Zhao et al. [141] found that AMF increased the levels of aromatic compounds and antioxidant enzyme activity in tobacco leaves. In addition, the presence of AMF can improve soil structure and increase the diversity of microbial communities, which in turn increases ecosystem stability and promotes plant growth [142]. Therefore, in the intercropping system of tobacco and white clover, AMF not only enhances the nutrient uptake and stress resistance of tobacco, but also improves the quality and ecological function of its rhizosphere soil, creating a more favorable growth environment for tobacco. Overall, the influence of AMF on the diversity of rhizosphere soil microbial communities and soil structure in the tobacco-white clover intercropping system is multifaceted, involving the symbiotic relationship between AMF and host plants, the effects of environmental factors, and the role of plant diversity.

The predictive results from PICRUSt2 indicate that the bacteria present in the tobacco rhizosphere soil of each treatment are involved in six major metabolic pathways, including metabolism, genetic information processing, and environmental information processing. They have a total of 46 sub-functions and exhibit high functional abundance. To study the effect of insectary floral plants intercropping on tobacco rhizospheric bacterial function, cluster analysis was performed on the predicted gene copy numbers of the secondary functional layer genes (Fig. 9). The results showed that insectary floral plants intercropping with tobacco altered the metabolic functions of bacteria, with the T-A, T-L, T-R, and T-T samples exhibiting the largest variations. The findings are consistent with a study conducted by Mendes et al. [143] on the functional traits of soybean rhizosphere bacteria.

Although predictive analysis methods such as PICRUSt2 and FUNGuild can provide functional predictions and taxonomic information based on 16 S rRNA gene or ITS sequences, they still have certain limitations. In order to gain a more comprehensive understanding of the functional potential of tobacco rhizosphere microbiota of intercropping different insectary floral plants in the tobacco fields, combining metagenomic sequencing can provide richer genomic information and uncover the microbial potential functions and metabolic capabilities.

Conclusion

This study demonstrates that intercropping tobacco with insectary floral plants, especially with Tagetes erecta, Vicia villosa, and Lobularia maritima, has impacted soil chemical properties, altered rhizospheric microbial communities, and increased the abundance of specific microbial genera in the rhizosphere of tobacco. In addition, the pathway of the carbohydrate metabolism, amino acid metabolism, and energy metabolism in rhizosphere bacteria were significantly increased after tobacco intercropping with insectary floral plants. Fungal symbiotic trophic and saprophytic trophic significantly increased after intercropping Vicia villosa, Lobularia maritima and plant pathogen and animal pathogen were increased after intercropping Trifolium repens and Argyranthemum frutescens. These results may provide new insights into improving barriers to continuous cropping of tobacco via intercropping approach to regulate the rhizosphere environment.

Abbreviations

TN

total nitrogen

TP

total phosphorus

TK

total potassium

AN

available nitrogen

AP

available phosphorus

AK

available potassium

SOM

soil organic matter

EC

electrical conductivity

OUT

operational taxonomic unit

PCA

principal component analysis

PCoA

principal coordinate analysis

RDA

redundancy analysis

LDA

linear discriminant analysis

AMF

arbuscular mycorrhizal fungus

ECM

ectomycorrhizal fungus

PGPB

plant growth promoting bacteria

KEGG

Kyoto Encyclopedia of Genes and Genomes

Authors’ contributions

Investigation, formal analysis and validation, J.Z.; formal analysis, visualization, and writing—original draft and editing, J.Z., B.C., Z.X.; investigation and formal analysis, W.P., S.J., G.Y., K.Z.; conceptualization, funding acquisition, project administration, supervision, and writing—review and editing, B.C., Z.X., Y.H. All authors have read and agreed to the published version of the manuscript.

Funding

This work and the article processing charges (APC) were supported China National Tobacco Corporation (110202101049[LS-09]) and Yunnan Company of China National Tobacco Corporation (2022530000241019), Technology Innovation Team Project of Yunnan Provincial Department of Education (2022[69]).

Data availability

Sequence data that support the findings of this study have been deposited in the NCBI with the primary accession code PRJNA1058964. Accession number: SRR27378722, SRR27378733, SRR27378738—SRR27378754, SRR27378756，SRR27378757; SRR27385172, SRR27385177—SRR27385193, SRR27385195, SRR27385196, SRR27385220.

Declarations

Ethics approval and consent to participate

This study was conducted in the growing area of tobacco in Wanqiao Tobacco Workstation of Dali Tobacco Company and all comply with relevant institutional, national, and international guidelines and legislation. The plant material tobacco (K326), Tagetes erecta, Vicia villosa, Fagopyrum esculentum, Lobularia maritima, Trifolium repens, and Argyranthemum frutescens involved in this study is allowed to be used in any ordinary experiments. All the methods were carried out in accordance with relevant guidelines and regulations.

Consent for publication

Not appliable.

Competing interests

The authors declare no competing interests.