Introduction

Selenium (Se) is an essential micronutrient for human health. It can enhance immunity, resist diseases, and delay the aging process. Se-enriched agricultural products obtained through Se fertilizer application are important sources of Se supplementation. Konjac (Amorphophallus spp.) is a perennial herb of the Araceae family, and Konjac glucomannan (KGM) is a water-soluble polysaccharide (dietary fiber) extracted from the konjac corm1,2. Owing to its unique physical and chemical properties, KGM is widely used in the food, pharmaceutical and health care, and chemical industries, and konjac is widely cultivated as an important cash crop in Southwest China and other regions3,4. Amorphophallus muelleri is an important konjac cultivar with strong resistance to soft rot and a high content of KGM5. In recent years, with the continuous development of konjac products, the demand for konjac corms has increased annually, and Se-enriched konjac has received increasing attention. Therefore, elucidating the effects of different fertilization methods on the growth of A. muelleri is key to selecting appropriate planting methods to maximize konjac yields and support the healthy and robust development of the konjac industry in China.

Many studies have analyzed the effects of Se on plant growth, yields, and physiological characteristics and have revealed that an appropriate concentration of Se can promote plant growth and increase plant biomass6. Se increases the chlorophyll a, chlorophyll b, total chlorophyll, and carotenoid contents in plants while promoting K⁺ accumulation. It simultaneously improves net photosynthetic rate, stomatal conductance, and transpiration rate7. In addition, some researchers have studied the effects of in situ soil Se or exogenous Se treatments (different types or rates of Se fertilizer) on the structure and diversity of rhizosphere soil microbial communities and on crop quality and have shown that appropriate concentrations of Se fertilizer help increase soil enzyme activities, increase the abundance of beneficial rhizosphere microorganisms, and increase the richness and diversity of soil microbial communities8,9. However, few studies have focused on the effects of Se on the endophytic microbial communities in plants.

Endophytic microbial communities exist within plant tissues, forming complex symbiotic relationships with plants and playing important roles in promoting plant productivity and health in natural environments10,11. On the one hand, endophytic bacteria and fungi can promote nutrient acquisition, plant growth and development by fixing nitrogen; by producing auxin, cytokinin, and siderophores; and by mobilizing various nutrients (such as N, P, and K)12,13,14. On the other hand, they can directly or indirectly increase plant tolerance to environmental stresses by producing antibacterial compounds and inducing or regulating the expressions of genes associated with plant growth, development, and defense15,16. Therefore, in-depth analysis of the effects of Se on the composition of plant endophytic microbial communities not only aids in elucidating the interactions among relevant microbial communities and plants but also helps in the exploration, development, and utilization of functional microbial resources. In the present study, A. muelleri was taken as the research subject, and amplicon sequencing (for bacteria and fungi) was used to investigate changes in the composition and diversity of endophytic microbial communities across different A. muelleri tissues under exogenous Se treatment. Additionally, microbial co-occurrence network analysis was conducted to further analyze the interactions among these microbial communities, with the aim of understanding the impact of Se treatment on the endophytic microbial communities of A. muelleri.

Materials and methods

Materials

The experiments were conducted in the Wujiang District, Shaoguan city, Guangdong Province, from April to October 2023, with red soil as the soil type and sandy loam as the soil texture in the experimental area. The A. muelleri corms used in this study were provided by Kunming University/Key Laboratory of Konjac Biology of Yunnan Province, and the variety used was “Zhuyajin No. 1.” The liquid Se fertilizer used in the experiment had an Se content of 10 g per 1 L.

Experimental design and sample collection

The experiment included two treatment groups: (1) the Se-treated group (TSe), plants received foliar application of liquid Se fertilizer at 750 mL per 666.67 m2. Specifically, 750 mL of the stock Se fertilizer solution was diluted to 30 L with deionized water, and the diluted solution was applied via foliar spray at 45 mL/m2; and (2) the control group (TCK), which was sprayed with an equivalent volume of water following the same application protocol as the TSe group. A randomized block design was used for the experiment, with each treatment repeated three times and a plot area of 75 m2. Konjac corms were planted on single ridges, with a ridge width of 80 cm, a ridge height of 20–30 cm, a furrow width of 25–30 cm between ridges, and a plant spacing of 40 cm. The konjac corms used for the experiments were planted on April 28, 2023. Based on the growth characteristics of A. muelleri, we conducted foliar applications of liquid Se fertilizer at three critical growth stages: the full leaf expansion stage (June 2023, when the leaves were completely unfurled), the early corm formation stage (July 2023, the initial formation phase of the underground corm), and the vigorous growth period (August 2023, the period of accelerated growth in both the plant and the corm), with an equal volume of water sprayed as a control treatment during each application phase.

Sample collection: During the underground corm swelling stage (October 2023), konjac plants were selected using a five-point sampling method, with whole plants uprooted and quickly brought back to the laboratory, where the corm, root, petiole, and leaf tissues of each konjac plant were cut separately as tissue samples and named as follows: Se_Q (corm of the Se-treated group), Se_G (root of the Se-treated group), Se_YB (petiole of the Se-treated group), Se_Y (leaf of the Se-treated group), CK_Q (corm of the control group), CK_G (root of the control group), CK_YB (petiole of the control group), and CK_Y (leaf of the control group), for a total of 24 samples.

To avoid contamination by environmental microorganisms, the collected plant tissue samples were surface-sterilized using the following procedure: The samples were rinsed with a large volume of sterile distilled water for 5 min to remove soil attached to the tissues, then sequentially treated with 75% ethanol for 30 s and 3% sodium hypochlorite solution for 3 min, followed by three final rinses with sterile distilled water (5 min each). To confirm the sterilization efficacy, 150 µL of the terminal wash solution was plated onto potato dextrose agar (PDA) and incubated at 28 °C for 72 h in a thermostatic incubator. The absence of microbial growth following 72 h of incubation verified the efficacy of surface sterilization. The surface-sterilized tissues were cut into small pieces with an aseptic surgical knife on an ultraclean workbench, quickly frozen in liquid nitrogen, and immediately stored at − 80 °C for later use.

Determination of Se content in plant tissues

The Se contents in the konjac plant tissues were determined using hydride generation atomic fluorescence spectrometry according to the National Standard of the People’s Republic of China GB5009.93-2017.

DNA extraction and amplicon sequencing of endophytic microorganisms in konjac tissue

In accordance with the kit instructions, the DNA of 24 konjac tissue samples was extracted using a FastDNA® SPIN Kit for Soil (MP, USA) following the soil DNA extraction protocol. The concentration and purity of the DNA were measured using a NanoDrop 2000 spectrophotometer, and the quality of the extracted DNA was assessed by 1% agarose gel electrophoresis.

The diversity of endophytic fungi was assessed using the universal primers ITS1 F (5ʹ-CTTGGTCATTTAGAGGAAGTAA-3ʹ) and ITS2 R (5ʹ-GCTGCGTTCTTCATCGATGC-3ʹ)10. To minimize co-amplification of chloroplast and mitochondrial genes, bacterial community analysis was performed using a two-step PCR amplification approach with the primer pairs 799F (5ʹ-AACMGGATTAGATACCCKG-3ʹ)/1392R (5ʹ-ACGGGCGGTGTGTRC-3ʹ) and 799F (5ʹ-AACMGGATTAGATACCCKG-3ʹ)/1193R (5ʹ-ACGTCATCCCCACCTTCC-3ʹ)10. The resulting PCR products were recovered via 2% agarose gel electrophoresis and subsequently purified using an AxyPrep DNA Gel Extraction Kit (Axygen Biosciences, Union City, CA, USA) in accordance with the manufacturer’s protocol. The amplicon libraries were subsequently sequenced on the Illumina MiSeq PE300 platform at Shanghai Majorbio Bio-Pharm Technology Co., Ltd.

Analysis of amplicon sequencing data

The raw sequencing reads were quality controlled using Fastp (v 0.19.6)17, after which paired-end reads were merged into single sequences using FLASH software (v 1.2.11)18. High-quality sequences were clustered into operational taxonomic units (OTUs) at a 97% similarity threshold using UPARSE software (v 11)19. Taxonomic annotation of OTUs was performed using the RDP classifier (v 2.13) against the SILVA 16S rRNA gene database (v 138)20 and the UNITE Fungal ITS database (v 8.0)21, with a confidence threshold of 70%. The relative abundances of species were analyzed at the phylum and genus levels. Alpha diversity indices, including the Shannon and Chao 1 indices, were calculated using Mothur software (v 1.30.2). Principal coordinate analysis (PCoA) based on the Bray‒Curtis distance algorithm was conducted to evaluate overall changes in microbial community structures, with permutational multivariate analysis of variance (PERMANOVA) used to assess the significance of the differences between sample groups22. Co-occurrence network analysis of bacterial and fungal communities was conducted using the SparCC method (correlation coefficient > 0.75, P < 0.05) on the integrated network analysis pipeline (iNAP, https://github.com/yedeng-lab/iNAP)23, and the resulting networks were visualized using the interactive platform Gephi24.

Statistical analysis

The nonparametric Wilcoxon rank-sum test was employed to evaluate differences in the average relative abundances of bacterial and fungal species between the Se-treated group and the control group. Differences in alpha diversity indices among treatment groups were assessed by one-way ANOVA with Duncan’s post hoc test, applying the FDR correction (P < 0.05) for multiple comparisons.

Results and analysis

Changes in Se content in different konjac tissues after exogenous Se treatment

Analysis of the differences in Se content in different konjac tissues between the Se-treated group and the control group (Table 1) revealed that foliar application of Se fertilizer significantly increased the Se contents in the corm, root, and leaf tissues. The Se contents in the corm, root, and leaf tissues of the Se-treated group were 82.73, 7.42, and 182.24 times those of the control group, respectively (P < 0.05). These results indicated that the foliar application of Se fertilizer successfully promoted Se enrichment, but the Se absorption capacities varied among the different tissues, with leaf tissues having the highest absorption capacity, followed by the corms.

Table 1 Changes in selenium content in different tissues of Amorphophallus muelleri under exogenous selenium treatment.
Full size table

Changes in the structure and diversity of the endophytic bacterial and fungal communities in different konjac tissues after exogenous Se treatment

To elucidate the effects of exogenous Se treatment on the endophytic bacterial and fungal communities in different A. muelleri plant tissues (e.g., corms, roots, petioles, and leaves), amplification and sequencing of 16S rRNA (V3–V4) for bacteria and ITS for fungi were performed; 1,534,315 effective bacterial sequences and 3,040,712 effective fungal sequences were obtained from 24 A. muelleri tissue samples, with average sequence lengths of 375 bp and 248 bp, respectively. Taxonomic annotation was performed on the valid reads at a sequence similarity level of 97%, and a total of 6131 bacterial and 2157 fungal operational taxonomic units (OTUs) were identified from the 24 tissue samples. Venn diagram analysis revealed the numbers of shared and unique OTUs in the different konjac tissues between the exogenous Se-treated group and the control group. First, the number of endophytic bacterial OTUs in the konjac tissues gradually decreased from the underground parts (corms and roots) to the aerial parts (petioles and leaves), as did the number of shared OTUs between the Se-treated group and the control group: there were 1954 shared bacterial OTUs in the corms, whereas only 396 were shared in the leaves. After exogenous Se treatment, the numbers of unique bacterial OTUs in the different plant tissues (e.g., corms, roots, petioles, and leaves) were 1.46, 2.74, 1.28, and 1.33 times greater, respectively, than those in the control group (Fig. 1A). The numbers of fungal OTUs that were shared by the different konjac tissues in the Se-treated group and the control group gradually increased from the underground parts to the aboveground parts. In the corms, the Se-treated group and the control group shared only 27 fungal OTUs, accounting for 3.19% of the total OTUs in the corms, whereas in the leaves, the two groups shared 275 fungal OTUs, accounting for 34.29% of the total OTUs in the leaves. In addition, after exogenous Se treatment, the numbers of unique fungal OTUs in the corms and root tissues of A. muelleri were significantly (28.29- and 3.48-fold, respectively) greater than those in the control group; in contrast, the numbers of unique fungal OTUs in the petiole and leaf tissues were lower than those in the control group (Fig. 1B). These results indicated that exogenous Se treatment affected the compositions of both the bacterial and fungal communities in the different A. muelleri tissues.

Fig. 1
figure 1

Comparative analysis of microbial community structures in konjac tissues. (A,B): Venn diagram illustrating the number of unique and shared bacterial and fungal OTUs among the different treatment groups. (C,D): Principal coordinate analysis (PCoA) plots based on the Bray–Curtis dissimilarity matrix illustrating the structural changes in bacterial and fungal communities under different treatments.

Full size image

Principal coordinate analysis (PCoA) was performed on the basis of the Bray‒Curtis distance algorithm to evaluate the effects of exogenous Se treatment on the structure of the bacterial and fungal communities in the different A. muelleri tissues (corms, roots, petioles, and leaves) (beta diversity) (Fig. 1C,D). PERMANOVA revealed significant differences in the structures of the bacterial and fungal communities among the different tissues. Specifically, samples from the underground parts (corms and roots) and aerial parts (petioles and leaves) were separated along the first principal component and distributed on the left and right sides of the PC1 axis, respectively. A comparison of each tissue type revealed that exogenous Se treatment had a more pronounced effect on the structures of the endophytic bacterial and fungal communities in the root tissues. The root tissue samples from the Se-treated group and the control group were distinctly separated and aggregated into different clusters, suggesting a significant difference in the structures of the bacterial and fungal communities in the root tissues between the Se-treated group and the control group.

The changes in the alpha diversity indices (Shannon and Chao 1 indices) of the bacterial and fungal communities in the different plant tissues were plotted as box plots (Fig. 2). First, the diversity and richness of the bacterial communities in the corm and root tissues were greater than those in the petiole and leaf tissues. After exogenous Se treatment, both the Shannon and Chao 1 indices of the bacterial communities in the root tissues were greater than those in the control group, with the Chao 1 index being 66.81% greater (P < 0.05) than that in the control (CK_G) (Fig. 2A,B), suggesting a significant increase in the bacterial community richness in the root tissues after exogenous Se treatment. The Shannon and Chao 1 indices of the fungal communities in the corm and root tissues were greater after Se treatment than in the control. Specifically, compared with those of the corresponding controls (CK_Q and CK_G), the Shannon indices of the fungal communities in the corm (Se_Q) and root tissues (Se_G) increased by 69.47% and 47.98% (P < 0.05), respectively (Fig. 2C). The Chao1 index of the fungal community in Se_Q increased by 1287.92% compared to CK_Q (P < 0.05), while that of Se_G showed an increasing trend without statistical significance (Fig. 2D). These results indicated that exogenous Se treatment significantly affected the diversity and richness of the bacterial and fungal communities in the corm and root tissues of A. muelleri.

Fig. 2
figure 2

Shannon and Chao1 diversity indices of the bacterial (A,B) and fungal (C,D) communities in the konjac tissues of different treatment groups.

Full size image

Changes in the compositions of the bacterial and fungal communities in different konjac tissues after exogenous Se treatment

The species compositions of the bacterial and fungal communities at different taxonomic levels in the konjac tissues of the different treatment groups were further analyzed. Bar charts of the relative abundances in the bacterial and fungal communities at the phylum level are shown in Fig. 3. Proteobacteria, Actinobacteriota, Firmicutes, Bacteroidota, Acidobacteriota, Ascomycota, Basidiomycota, Glomeromycota, and Mortierellomycota were the main dominant bacterial and fungal communities in the different tissues (Fig. 3A,B). Under exogenous Se treatment, the relative abundances of the dominant bacterial and fungal communities in the different tissues significantly differed from those in the untreated control treatment. After Se treatment, the relative abundances of Proteobacteria in the corm (Se_Q), root (Se_G), and leaf (Se_Y) tissues were 35.69%, 11.18%, and 13.69% lower than those in the control, respectively. In contrast, the relative abundance of Proteobacteria in petiole tissues was 7.93% greater in the Se-treated group than in the control group. In addition, after Se treatment, the relative abundances of Firmicutes and Bacteroidota in the corm tissues; Actinobacteriota, Firmicutes, Bacteroidota, and Acidobacteriota in the root tissues; Firmicutes and Acidobacteriota in the petiole tissues; and Actinobacteriota, Firmicutes, and Bacteroidota in the leaf tissues were greater than those in the respective controls. Only the relative abundances of Actinobacteriota in the corm and petiole tissues were lower than those in the control (Fig. 3A). With respect to the fungal communities, after the Se treatment, the relative abundance of Ascomycota in the corm (Se_Q) was greater than that in the control (CK_Q), whereas the relative abundance of Basidiomycota was lower than that in the control (CK_Q). In contrast to that in the corm, the relative abundances of Ascomycota in the root, petiole, and leaf tissues were lower in the Se-treated group than in the control group, whereas the relative abundance of Basidiomycota was higher in the Se-treated group than in the control group. Compared with that in the corresponding control, the relative abundances of Ascomycota in Se_G, Se_YB, and Se_Y were 57.25%, 20.54%, and 5.41% lower, respectively, while the relative abundances of Basidiomycota in Se_G, Se_YB, and Se_Y were 382.89%, 101.87%, and 17.46% higher, respectively, than those in the corresponding controls (Fig. 3B).

Fig. 3
figure 3

Microbial community compositions in the konjac tissues of different treatment groups. (A,B) The bar plots of the relative abundances illustrate the compositions of the bacterial and fungal communities at the phylum level in the konjac tissues of different treatment groups. (C,D) The chord diagram shows the distribution of the most dominant bacterial and fungal communities at the genus level across different sample groups.

Full size image

A chord diagram shows the distribution of the most dominant bacterial and fungal communities at the genus level across different sample groups. Among the genus that were accurately classified and named, Sphingomonas, Allorhizobium-Neorhizobium-Pararhizobium-Rhizobium, Klebsiella, Curtobacterium, Methylobacterium-Methylorubrum, and Bradyrhizobium were the main dominant bacterial genera (Fig. 3C), and Cladosporium, Arthrinium, Strelitziana, Symmetrospora, and Candida were the main fungal genera in the different tissues (Fig. 3D). To further clarify the differences in the relative abundances of the microbial communities in the different konjac tissues after exogenous Se treatment, we performed a significance analysis of the between-group differences and used bar charts of the species difference tests to display the top 10 genera in terms of relative abundance in the different tissue samples (Fig. 4). First, after exogenous Se treatment, the relative abundances of bacteria such as Pantoea, Dongia, Phycicoccus, Exiguobacterium, Deinococcus, Paeniglutamicibacter, and Puia in the corm tissues were significantly lower than those in the control group (P < 0.05). Among the fungi, Candida, Trichoderma, Thermoascus, and Aspergillus were more abundant in the Se-treated group than in the control group, but the differences were not significant (P > 0.05) (Fig. 4A). Among the bacteria in the root tissues, the relative abundances of Allorhizobium-Neorhizobium-Pararhizobium-Rhizobium, Bradyrhizobium, Lechevalieria, Sphingomonas, Streptomyces, Mesorhizobium, and Devosia were significantly greater in the Se-treated group than in the control group, whereas the relative abundances of Rhodanobacter and Actinospica were significantly lower in the Se-treated group than in the control group (P < 0.05). In the endophytic fungal community in the roots, the relative abundances of Ceratobasidium and Athelia in the Se-treated group were significantly greater than those in the control group, whereas the relative abundances of Rhizophagus, Thanatephorus, Acaulospora, and Trichoderma in the roots were significantly lower than those in the control group (P < 0.05) (Fig. 4B). In the petiole tissues, the relative abundances of Methylobacillus, Ralstonia, Kibdelosporangium, Nitrosospira, and Saitozyma in the Se-treated group were significantly greater than those in the control group (P < 0.05). In contrast, the relative abundances of Pseudocitrobacter, Nigrospora, Pseudocercospora, Penicillium, and Alternaria were significantly lower than those in the control group (P < 0.05) (Fig. 4C). Finally, in the leaf tissues, the relative abundances of Nocardioides, Microvirga, and Exiguobacterium were significantly greater in the Se-treated group than in the control group (P < 0.05), but no significant differences were detected in the fungal communities between the Se-treated group and the control group. These results indicated that after exogenous Se treatment, the dominant species and relative abundances of the bacterial and fungal communities in the different A. muelleri tissues significantly differed from those in the untreated control group (Fig. 4D).

Fig. 4
figure 4

Nonparametric Wilcoxon rank-sum test results showing the differences in the average relative abundances of the same bacterial (left) and fungal (right) communities at the genus level between the Se-treated group and the control group. Here, * represents 0.01 ≤ p < 0.05, ** represents 0.001 ≤ p < 0.01, and *** represents p ≤ 0.001. (A) Corm tissue; (B): Root tissue; (C): Petiole tissue; (D): Leaf tissue.

Full size image

Cooccurrence network analysis of the endophytic microbial communities in the konjac tissues

To study the effects of exogenous Se treatment on the co-occurrence patterns of the endophytic microbial communities in the different konjac tissues, we analyzed the interkingdom bacterial‒fungal network as well as the intrakingdom bacterial‒bacterial and fungal‒fungal networks in the tissues of the Se-treated group and the control group (Fig. 5).

Fig. 5
figure 5

Cooccurrence networks of microbial communities in the konjac tissues of different treatment groups. (A) Bacterial and fungal interkingdom networks; (B) bacterial intrakingdom networks; (C) fungal intrakingdom networks.

Full size image

First, analysis of the bacterial‒fungal interkingdom ecological network revealed that the Se-treated group exhibited higher numbers of nodes and edges, as well as a greater average degree, compared to the control group. In both treatment groups, the proportions of bacterial taxa were greater than those of fungal taxa in the interkingdom bacteria‒fungi network, and the interkingdom correlations between bacteria and fungi were mostly negative. The proportion of negative correlations in the Se-treated group (68.35%) was greater than that in the control group (62.34%) (Fig. 5A).

Intrakingdom network analysis demonstrated that both bacterial-bacterial and fungal-fungal networks in the Se-treated group also contained higher node and edge counts compared to the control group. Similarly, the average degree of the fungal-fungal network was also higher in the Se-treated group than in the control group. In addition, we found that the intrakingdom correlations within the bacterial communities and within the fungal communities in the two treatment groups were predominantly positive; in particular, the proportion of positive correlations among the intrakingdom correlations of the fungal communities exceeded 90%, reaching 92.15% in the Se-treated group and 100% in the control group (Fig. 5B,C). In the bacterial-bacterial networks, species of Proteobacteria and Actinobacteriota were dominant (Fig. 5B). In fungal-ungal networks, species of the Ascomycota were dominant, but after Se treatment, the species interactions of the Basidiomycota increased compared to the control group (Fig. 5C). These results indicated that exogenous Se treatment enhanced the interactions between microbial communities in konjac plants.

Discussion

Plants possess large and diverse communities of prokaryotes and eukaryotes (i.e., plant microbiota)25. Plants and their associated microbial communities have coevolved for more than 400 million years. Interactions between plants and microbial communities play important roles in promoting plant productivity and health in the natural environment26,27,28. In this study, we systematically analyzed the changes in the composition and diversity of the endophytic microbial communities in different A. muelleri tissues under exogenous Se treatment. Our results indicate that the foliar application of Se fertilizer can achieve the goal of increasing Se enrichment in konjac, while the changes in the endophytic microbial communities of A. muelleri in this study were affected by exogenous Se treatment to some extent. After exogenous Se treatment, the numbers of unique bacterial OTUs in the different plant tissues (corms, roots, petioles, and leaves) of A. muelleri were greater than those in the control group, and the numbers of unique fungal OTUs in the corm and root tissues were also significantly greater. Compared to other tissues, the microbial communities in corm and root tissues exhibited stronger responses to Se treatment. After exogenous Se treatment, compared with those in the control group, the Se contents in the konjac corm and root tissues increased by 82.73 times and 7.42 times, respectively. Concurrently, compared with those in the untreated control, the diversity and richness indices of the bacterial and fungal communities in the root tissues, as well as the fungal communities in the corm tissues, increased. Furthermore, beta-diversity analysis revealed the structures of the bacterial and fungal communities in the root tissues of the Se-treated group differed substantially from those of the control group. As endophytic microbial communities originate primarily from seeds and the environment12, these communities usually change dynamically under the combined influence of host and environmental factors29,30. The corm, as an essential propagating material of A. muelleri, functions as a seed and is an important source of endophytic microbial communities in A. muelleri. Root tissue not only is a vital organ for plants to absorb water and nutrients but also serves as a primary pathway for exogenous microbes from the soil or the environment to enter the plant. Studies have shown that under pathogen stress or different environmental conditions, host plants usually attract or recruit distinct microbial communities by altering their own metabolism to meet their current needs or provide themselves with protective functions31,32,33,34. Therefore, we hypothesize that the significant increase in Se contents within the corm and root tissues after Se treatment may be a key factor that contributes to the substantial changes observed in their microbial communities.

We further analyzed the differences in the species compositions of the bacterial and fungal communities in the different A. muelleri tissues under exogenous Se treatment. In line with the findings of most previous studies, Proteobacteria, Actinobacteriota, Firmicutes, Bacteroidota, and Acidobacteriota were the main dominant bacterial phyla in the konjac tissues, and Ascomycota, Basidiomycota, Glomeromycota, and Mortierellomycota were the main fungal phyla. However, exogenous Se treatment significantly affected the relative abundances of these dominant species in the different tissues. Among the bacteria, after Se treatment, the relative abundances of Firmicutes and Bacteroidota in the corm tissues, Actinobacteriota, Firmicutes, Bacteroidota, and Acidobacteriota in the root tissues; Proteobacteria, Firmicutes, and Acidobacteriota in the petiole tissues; and Actinobacteriota, Firmicutes, and Bacteroidota in the leaf tissues were greater than those of the corresponding controls to varying degrees. In terms of fungi, exogenous Se treatment reduced the relative abundance of Ascomycota in the root, petiole, and leaf tissues of A. muelleri but increased the relative abundance of Basidiomycota. Among these phyla, Actinobacteriota, Firmicutes, and Basidiomycetes are important decomposers in the carbon cycle and can degrade cellulose, lignin, and lignocellulose through the secretion of hydrolases, thus utilizing recalcitrant carbon sources. Actinobacteriota can also produce antibiotics that inhibit the growth and development of various soil-borne plant pathogens35,36. Proteobacteria play important roles in the carbon, sulfur, and nitrogen cycles37. Acidobacteriota are essential for the iron cycle and one-carbon compound metabolism38. Similarly, at the genus level, the relative abundances of bacteria such as Bradyrhizobium, Sphingomonas, Streptomyces, and Mesorhizobium in the root tissues; Methylobacillus and Nitrosospira in the petiole tissues; and Nocardioides and Microvirga in the leaf tissues were significantly greater in the Se-treated group. It has been previously reported that Bradyrhizobium and Mesorhizobium species possess nitrogen-fixing abilities and can secrete plant hormones (such as indole-3-acetic acid/IAA) to stimulate the growth and development of plant roots. Certain strains of Sphingomonas are also closely related to nitrogen fixation and can increase plant survival under environmental stress by improving the soil environment and degrading toxic substances39. Streptomyces is globally known for its powerful ability to produce antibiotics40. Therefore, our results demonstrate that foliar Se application facilitates the enrichment of beneficial microorganisms in konjac tissues and promotes favorable shifts in the endophytic microbial community compositions of konjac plants. The enrichment of beneficial microorganisms—including Actinobacteriota, Firmicutes, Basidiomycota, Acidobacteria, Bradyrhizobium, Sphingomonas, and Streptomyces—in different A. muelleri tissues may promote plant growth and increase stress resistance through multiple mechanisms, such as nutrient cycling, phytohormone secretion, and antibiotic production. Studies by Liu et al.41 and Rosenfeld et al.42 also support that Se can reduce the abundance of pathogenic microorganisms, thereby increasing plant resistance to pathogen infection.

To further elucidate the interactions among related microbial communities, we performed microbial co-occurrence network analysis. The results revealed that exogenous Se treatment increased the interactions between microbial communities within whole konjac plants. The interkingdom correlations between bacteria and fungi in the networks of the Se-treated group and the control group were predominantly negative, indicating ecological competition between bacteria and fungi. Studies have shown that mutual negative interactions (i.e., ecological competition) can increase the stability of microbial communities by suppressing the destabilizing effects of cooperation31,43. The host may benefit from microbial competition, thereby increasing its resistance to external stress44. In contrast, the intrakingdom correlations within the bacterial communities and within the fungal communities in the two treatment groups were mainly positive, indicating that the interactions among bacteria and among fungi were mainly cooperative or mutualistic31,45. This study revealed that in the bacterial-bacterial networks, species belonging to Proteobacteria and Actinobacteriota were the dominant phyla. In fungal-ungal networks, species of the Ascomycota were dominant, but after Se treatment, the species interactions of the Basidiomycota increased compared to the control group. Species from Proteobacteria, Actinobacteriota, and Basidiomycota play pivotal roles in carbon and nitrogen cycling35,36,37. Their positive interactions with other microorganisms may facilitate the cycling and accumulation of these nutrients within plant systems. In summary, our findings provide critical evidence that exogenous Se treatment significantly alters the bacterial and fungal community composition within A. muelleri tissues. However, further investigation is needed to elucidate the regulatory mechanisms of Se on endophytic microbial communities by testing varying Se concentrations.

Conclusion

This study revealed that exogenous Se treatment led to significant changes in the compositions and diversity of the endophytic microbial communities in A. muelleri while enhancing the interactions among the microbial communities within the konjac plant. The root and corm microbiota showed the most significant response to Se. Following Se treatment, beneficial microorganisms such as Actinobacteriota, Firmicutes, Bradyrhizobium and Streptomyces were found to accumulate in different tissues of A. muelleri. However, this study has certain limitations. In future research, we will further investigate the effects of different Se treatment concentrations on the growth and disease resistance of konjac plants, determine the optimal dosage of Se fertilizer, and employ metagenomic sequencing analysis combined with isolation of culturable microorganisms to comprehensively analyze and validate the functional roles of relevant microbial communities.