Investigation of the Impact of Soil Physicochemical Properties and Microbial Communities on the Successful Cultivation of Morchella in Greenhouses

Abstract

Morels (Morchella spp.) are medicinal and edible mushrooms, renowned for their distinctive taste and appearance. Due to the low yields and difficulty of foraging wild morels, artificial cultivation has significant economic value. Outdoor cultivation yields are influenced by factors such as weather and diseases, which can result in crop instability or failure, thereby causing losses to farmers. Previous studies have typically concentrated on either the fungal or bacterial communities. In this study, we investigated the ecological relationships between morel growth and both the fungi and bacteria in soil, analyzed over multiple trophic levels. We investigated three soil types: soil in which morel death was observed (DM), soil in which no morels emerged (UM), and soil that is suitable for normal fruiting (NM). We used high-throughput ITS and 16S rDNA amplicon sequencing, alongside assessment of soil physicochemical properties, to investigate factors contributing to morel emergence and death. The results indicated that the richness and diversity of both fungal and bacterial communities in the normal fruiting soil (NM) were significantly higher than those in the non-fruiting soils (DM and UM). The bacterial community was primarily composed of Proteobacteria and Bacteroidota, while the fungal community was dominated by Ascomycota and Mucoromycota. Furthermore, Morchella was significantly enriched in NM, indicating that it had successfully colonized and could develop into fruiting bodies. The morel mycelium in NM effectively utilized external nutrient bags, enhancing the soil nitrogen and organic matter content while reducing the consumption of available phosphorus and potassium. LEfSe and random forest analyses identified Pedobacter and Massilia as biomarkers of NM, potentially associated with the symbiosis of Morchella, which may promote its growth. Furthermore, the construction of the fungal-bacterial co-occurrence network revealed that the NM soil exhibited a higher number of nodes and greater network stability, suggesting that its complex microbial community structure may play a crucial role in the successful cultivation of Morchella. Our results indicate that the failures in morel production were due to inadequate management practices. Elevated greenhouse temperatures may have promoted pathogen proliferation, hindering the effective utilization of external nutrient bags by morel mycelium. Consequently, the mycelium was unable to accumulate nutrients efficiently, leading to the inability of Morchella to fruit or resulting in developmental failures. This study offers valuable insights into the interactions between morel mycelium and soil microorganisms, elucidating the reasons for morel cultivation failure and suggesting strategies for optimizing morel cultivation.

1. Introduction

Mushrooms have been associated with humans for thousands of years. Among the numerous mushroom varieties, Morchella spp. (Ascomycota) are particularly noteworthy for producing mushrooms known as morels. Morels are characterized by their distinctive appearance, featuring a wrinkled mesh that resembles the shape of a morel, hence their name. Due to their unique and delicious taste, morels are beloved by food enthusiasts worldwide and are frequently used in various high-end dishes, particularly in European and Asian cuisines. Beyond their culinary appeal, morels possess numerous medicinal values [1]. Their main nutritional and medicinal components include amino acids, organic acids, vitamins, fatty acids, and carbohydrates [2]. Morels promote wound healing [3], exhibit anti-inflammatory and anti-tumor activities [4], protect the kidneys and liver [5,6], and contain antibacterial properties [7]. Wild morels are distributed globally, with major distribution areas including Europe, western North America, Israel, China, and Chile [8,9]. The life cycle of Morchella begins with the dispersion of spores into the environment, where the spores germinate and form new mycelium, creating a mycelial network. As the mycelium develops and spreads, it produces conidia and sclerotia, which are two critical stages in the life cycle of Morchella. The mycelium that forms sclerotia consists of specialized cells produced by the fungus, enabling Morchella to withstand adverse environmental conditions. Subsequently, under suitable environmental conditions, the sclerotia differentiate into mycelium and develop into primordia. These primordia continue to grow and eventually form fruiting bodies. Once the fruiting bodies reach maturity, they release spores into the surrounding environment [8,10,11].

Although wild morels are widely distributed, their low production and the brief market availability of fresh morels have driven interest in artificial cultivation [8]. In 1982, Ower reported successful indoor cultivation of morels and subsequently applied for three cultivation patents from 1986 to 1989, paving the way for commercial cultivation [10]. However, in 2008, indoor factory cultivation of morels in the United States was largely halted due to issues such as bacterial contamination and diseases. In 2012, China adopted external nutrient bag technology to supplement morel mycelium nutrition, effectively increasing fruiting body yields and promoting large-scale field cultivation [8]. Despite the maturation of commercial cultivation in China, the complete life history of morels is still not fully understood. In cultivation, unexplained phenomena such as reduced yield, no morel emergence, and morel death persist. Growers suspect these issues may be related to source, weather, management techniques, and harmful soil bacteria, seriously hindering industry development [12,13,14]. Field cultivation in open environments exposes morels to pathogen infections, potentially leading to severe yield reductions or total crop failure. For instance, Cladobotryum protrusum causes spider web disease in morel fruiting bodies, while other pathogens such as Paecilomyces penicillatus, Diploöspora longispora, Leccanicillium aphanocladii, Fusarium incarnatum-equiseti species complex, Clonostachys solani, Penicillium kongii, and Mortierella gamsii cause fruiting body decay [15,16,17,18,19]. Some fungi, such as Trichoderma, Mucor, Rhizopus, Cephalotrichum, and Aspergillus, compete with morels for growth [8,12]. Reports indicate that morel yield is not only affected by soil pathogens but also significantly positively correlated with the evenness and diversity of soil microbial communities. Specifically, soil pathogens directly threaten morel growth, while microbial community evenness and diversity influence soil ecosystem health and balance [12]. Additionally, reports have indicated that the microbial communities in the soil used for Morchella cultivation, particularly nitrogen-fixing bacteria and nitrifying bacteria, influence the yield of Morchella. Furthermore, pathogenic fungi are among the factors contributing to the low yields of Morchella in cultivation [13].

The primary objective of this study was to investigate the impact of soil physicochemical properties and microbial communities on the growth of morel fruiting bodies. To achieve this, three types of soil were collected during the mushroom growing season: dead mushroom soil (DM), unfruitful soil (UM), and normal fruiting soil (NM). The aims of the study were to (1) determine the physical and chemical properties of the three soil types and conduct correlation analysis with microbial communities; (2) compare the microbial communities among the three soil types; (3) identify biomarkers associated with different soil types using LEfSe analysis and random forest models; and (4) perform a co-occurrence network analysis of bacteria and fungi to reveal the structural characteristics of the microbial communities in the three soil types. The results of this study provide insights into the potential relationship between the soil environment and the fruiting and death of morels.

2. Materials and Methods

2.1. Experimental Design and Sample Collection

In April 2024, at a morel (M. sextelata) cultivation farm located in Shuangcheng District, Harbin City, Heilongjiang Province, China (45°56′ N, 126°49′ E), some greenhouses exhibited either early morel death or no morel emergence (Figure 1). To investigate the causes of these issues, a five-point sampling method was employed to collect surface soil (1–5 cm) samples from the greenhouses, which were then combined into a single sample for each greenhouse. The soil samples were sieved through a 2 mm sieve, with roots and plant debris manually removed. Subsequently, samples were divided into two parts: one was air-dried for soil physicochemical property determination, and the other was immediately placed in liquid nitrogen, transported to the laboratory, and stored in a −80 °C freezer for subsequent DNA extraction. A total of nine soil samples were collected: three samples from normal fruiting greenhouses (NM), three samples from greenhouses with dead morels (DM), and three samples from unfruitful greenhouses, where no morels emerged (UM).

2.2. Soil Chemical Analysis

The soil pH was measured using the potentiometric method with a soil-to-water ratio of 1:2.5 [20]. Total nitrogen (TN) was determined via the Kjeldahl method following digestion with concentrated sulfuric acid [21]. Soil organic matter (OM) was oxidized using the potassium dichromate and concentrated sulfuric acid method [22]. Total phosphorus (TP) and available phosphorus (AP) were determined using the molybdenum-antimony colorimetric method [23]. Total potassium (TK) and available potassium (AK) were determined by flame atomic absorption spectroscopy [24].

2.3. DNA Extraction and High-Throughput Sequencing

Total genomic DNA was extracted using the E.Z.N.A.^® Soil DNA Kit (Omega Bio-Tek, Norcross, GA, USA), and the quality of the extracted DNA was assessed via 1% agarose gel electrophoresis. The primers 515F (5′-GTGCCAGCMGCCGCGG-3′) and 907R (5′-CCGTCAATTCMTTTRAGTTT-3′) were used to amplify the V4–V5 region of bacterial 16S rDNA, while ITS1-F (5′-CTTGGTCATTTAGAGGAAGTAA-3′) and ITS2-R (5′-GCTGCGTTCTTCATCGATGC-3′) were utilized to amplify the internal transcribed spacer (ITS) region of fungal DNA. PCR was conducted using TransGen AP221-02: TransStart FastPfu DNA Polymerase in a 20-μL reaction system, followed by gel excision and recovery of the PCR product using the AxyPrep DNA Gel Recovery Kit (AXYGEN, Union City, CA, USA). Subsequently, the Illumina HiSeq 2500 library was prepared following Illumina’s standard protocol and sequenced using the Illumina HiSeq 2500 platform at Shanghai Biozeron Biotechnology Co., Ltd. (Shanghai, China). Quality control, filtering, chimera removal, and sequence assembly were performed according to the QIIME2 workflow. Amplicon sequence variants (ASVs) were generated using DADA2 quality control and normalized before being compared against the Silva and UNITE databases to annotate bacterial and fungal species [25]. All raw sequence data were deposited in the National Center for Biotechnology Information (NCBI) database with BioProject accession numbers PRJNA1192010 (bacteria) and PRJNA1192406 (fungi).

2.4. Statistical Analysis

R version 4.4.0, SPSS version 25.0, and Gephi version 0.10 were used for statistical analysis and visualization. One-way ANOVA was performed using SPSS 25.0 to analyze physical and chemical properties at a significance level of 0.05. QIIME2 (2024.10) was used to calculate alpha diversity at the ASV level, specifically calculating the ACE index, Simpson index, and Good’s coverage index [25,26,27]. Beta diversity was calculated using principal coordinate analysis (PCoA) based on the Bray–Curtis distance with the “vegan” package in R version 4.4.0 to determine similarities and differences between samples [28]. Canonical correlation analysis (CCA) and redundancy analysis (RDA) were performed on microorganisms and environmental factors using the “vegan” package in R version 4.4.0. Based on the results of degradation response analysis (DCA), CCA was selected to explore the contribution of environmental factors to soil microbial communities [29]. The “linkET” package in R version 4.4.0 was used to conduct the Mantel test (UniFrac distance matrix) to calculate the relationship between the microbial community and environmental variables [30]. Linear discriminant analysis effect size (LEfSe) was performed using the “microeco” package in R version 4.4.0 to identify groups with significantly different abundances between categories at the genus level using the non-parametric factorial Kruskal–Wallis (KW) summary rank test (α = 0.05), and an LDA score > 4 was used to identify biomarkers [31]. The “randomForest” package in R version 4.4.0 was used to construct a random forest model, identifying different biomarkers through supervised classification model analysis, which were mutually confirmed with the biomarkers identified using LEfSe [32]. Subsequently, the “graph” and “WGCNA” packages in R version 4.4.0 were used to calculate Spearman correlations between bacteria and fungi with abundances greater than 0.01% at the V level, constructing a cross-domain collinear network of bacteria and fungi that only contained robust correlations (|r| > 0.6) and p-values (p < 0.05) [33]. Furthermore, the topological properties of the networks were calculated, and by randomly removing 200 nodes, the natural connectivity and average degree were computed to further explore the correlation information within or between populations.

3. Results

3.1. Physicochemical Properties of the Three Soil Samples

Significant differences in soil physical and chemical properties were observed among the three soil types. NM soil, which supported more successful morel growth, contained higher levels of organic matter and nitrogen but lower levels of total potassium, total phosphorus, available potassium, and available phosphorus compared to DM and UM soil samples, with a neutral pH (Figure 2).

DM soil, in which morel death occurred, exhibited the lowest organic matter and nitrogen content. Based on DCA, CCA was selected to explore the relationship between soil microbial communities and soil physicochemical properties (Figure 3a,b). CCA1 and CCA2 accounted for 37.33% and 13.56% of the total variation in fungal communities and 24.71% and 19.57% of the total variation in bacterial communities. Additionally, the R² and p-values were calculated to explore the relationship between the three soil types and soil physicochemical properties. A positive correlation was observed between organic matter and nitrogen in NM, whereas negative correlations were found with other physicochemical properties. A Mantel test (Figure 3c) revealed that soil pH, phosphorus, and potassium significantly affected fungal communities (Mantel’s r ≥ 0.4, p < 0.01), whereas nitrogen, phosphorus, and potassium were significant physicochemical variables affecting bacterial communities (Mantel’s r ≥ 0.4, p < 0.01). Organic matter was positively correlated with nitrogen content and negatively correlated with pH, phosphorus, and potassium. These results suggest that organic matter and nitrogen are the primary factors influencing morel fruiting.

3.2. Composition and Diversity of Soil Microbial Communities

After quality control, an average of 32152 high-quality fungus sequences were obtained per sample, clustered into 990 ASVs. Similarly, 32,383 sequences for the 16S rRNA gene were clustered into 2910 ASVs. As the sample size increased, the ASV rarefaction curve tended to plateau (Figure S1), with an average Good’s coverage of 0.99, indicating that the deep sequencing provided comprehensive ASV coverage and accurately reflected the microbial community composition in the samples [34]. At the phylum level, the fungal community was primarily composed of Ascomycota and Mucoromycota, with no significant differences observed among the groups. However, at the genus level, Morchella and Mortierella were significantly enriched. In contrast, the relative abundances of Fusarium, Botryotrichum, and Humicola were notably lower in NM compared to DM and UM, showing significant differences among the groups (Figure S2a,b). At the phylum level, the bacterial community was predominantly composed of Proteobacteria, Bacteroidota, Acidobacteriota, Actinobacteriota, and Chloroflexi. Among these, the relative abundances of Proteobacteria and Bacteroidota were significantly higher in NM, while Acidobacteriota and Chloroflexi were notably lower compared to DM and UM. At the genus level, Gaiella and MND1 were significantly enriched in DM, and these genera may exhibit competitive interactions with Morchella. In contrast, Pedobacter, Polaromonas, and Flavobacterium were significantly enriched in NM, suggesting potential symbiotic relationships with Morchella (Figure S2c,d). As illustrated in a Venn diagram (Figure 4), the NM soil contained a substantial number of unique fungi (249) and bacteria (1294), whereas the three soil types shared only 28 fungal and 29 bacterial ASVs.

NM exhibited high richness (ACE index) and diversity (Shannon index) in bacterial communities, whereas DM and UM showed no significant differences. Notably, NM had the highest fungal ACE index and the lowest Shannon index (Figure S3). Additionally, PCoA and PERMANOVA analyses were conducted on the beta diversity of the three soil microbial communities using the weighted UniFrac distance (Figure 5). For fungi, the first axis accounted for 77.81% of the variation, and the second axis accounted for 10.44%. For bacteria, these values were 60.01% and 15.53%, respectively. These results highlight the ability to explain differences among soil types. Both the fungal and bacterial community compositions in NM significantly differed from those in DM and UM. Conversely, the distance between DM and UM was relatively small, suggesting a similar microbial community structure between these two soil types.

3.3. Biomarker Prediction

Linear discriminant analysis (LDA) effect size (LEfSe) (Figure 6) revealed nine fungal biomarkers (LDA score > 4). No fungal taxa were significantly enriched in DM, whereas seven were significantly enriched in UM, including Chaetomiaceae, Sordariomycetes, and Sordariales. In NM, two fungal taxa were significantly enriched (Morchellaceae and Morchella), and 16 bacterial biomarkers were identified (LDA score > 4). Of these, only MND1 was significantly enriched in DM. In UM, three bacterial taxa were significantly enriched: Planctomycetota, Planctomycetes, and Saprospiraceae. Twelve bacterial taxa were significantly enriched in NM, including Gammaproteobacteria, Bacteroidota, and Bacteroidia.

Subsequently, the machine learning random forest method (Figure 7) was employed to identify the top 20 statistically significant predictive factors using supervised classification models at the genus level for both fungi and bacteria. The results indicated that, regarding fungal composition, important predictive factors for DM included Purpureocillium and Trichurus, while important predictive factors for UM included Botryotrichum and Humicola. Important predictive factors for NM included Morchella and Phoma. Regarding bacterial composition, MND1 and Subgroup10, among others, were important predictive factors for DM. For UM, Caenimonas and Pir4 lineage, among others, were key predictive factors. In NM, Pedobacter and Massilia, among others, were significant predictive factors.

3.4. Co-Occurrence Network Characteristics of Different Soil Types

Fungal and bacterial co-occurrence networks were constructed using Spearman correlations to explore differences in community structure among different soil types. The results indicated significant differences in the co-occurrence networks among NM, DM, and UM soils. NM exhibited a higher number of nodes (402) and average degree (128.71) through abundance and correlation screening, whereas they were lower for DM (247 nodes, average degree 84.31) and UM (260 nodes, average degree 93.45), indicating a more complex microbial community structure in NM (Figure 8). By randomly removing 200 nodes and observing changes in the average degree and natural connectivity, we demonstrated that the co-occurrence network of NM was more stable (Figure S4). Modular clustering revealed that the NM network consisted of six modules, while the networks of both DM and UM were clustered into four modules. Within these modules, Proteobacteria was dominant at the phylum level, and Acidobacteriota was also dominant in DM and UM. Conversely, Acidobacteriota was less abundant in NM, whereas Bacteroidota was more abundant (Figure S5). In DM, highly connected nodes included ASV3 (MND1), ASV3131 (Mortierella), and ASV3090 (Spizellomyces). In UM, highly connected nodes included ASV101 (Pir4 lineage), ASV3130 (Rhizophagus), and ASV3053 (Ascochyta). In contrast, in healthy soil (NM), highly connected nodes included ASV5 (Pedobacter), ASV6 (Massilia), and ASV3114 (Bullera). These differences in highly connected taxa contributed to variations in the soil microbial community structure.

4. Discussion

Regarding soil physicochemical properties, NM exhibited significantly higher nitrogen and organic matter levels compared to DM and UM. Previous studies have demonstrated that morel mycelium can absorb nutrients from external nutrient bags and transfer these nutrients to the surface soil. Additionally, nitrogen cycling positively impacts morel yield. High nitrogen and organic matter levels are crucial for the initial growth of morel mycelium, as they provide essential nutrients for mycelial expansion and biomass accumulation [13,35]. Conversely, NM had significantly lower levels of total potassium, total phosphorus, available potassium, and available phosphorus compared to DM and UM. This could be attributed to the fruiting process, during which morel fruiting bodies consume substantial soil nutrients, leading to their depletion in the soil [30,36]. The high demand for potassium and phosphorus during fruiting body formation suggests that these nutrients may become limiting factors for subsequent yields. In NM, the pH was significantly lower than that in DM and UM, possibly due to the application of plant ash or quicklime during early soil preparation, initially resulting in alkaline soil. As morel emergence progresses, the mycelium or symbiotic microorganisms secrete acidic substances that gradually neutralize soil alkalinity, decreasing the pH [33]. Based on these results, we hypothesize that unsuccessful fruiting of mushrooms may be due to poor greenhouse management or pathogen interference, preventing effective conversion of nutrients from external nutrient bags into forms necessary for morel growth, ultimately leading to production failure. Furthermore, we propose that the observed changes in soil physicochemical properties are closely linked to the life cycle of Morchella, with high nitrogen and organic matter supporting mycelial growth, potassium and phosphorus depletion limiting fruiting body formation, and pH changes influencing microbial symbiosis and morel development.

Morels can be cultivated in greenhouses in open, unsterilized soil. However, under these open conditions, the fungi and bacteria naturally present in the soil may either promote or inhibit morel growth [10,37]. In soils where morels are successfully cultivated, the morel mycelium extensively colonizes the soil, making morels the dominant microbial group. Although morel mycelium also colonizes soils where they have not been successfully cultivated, their relative abundance is significantly lower, and pathogenic bacterial genera, such as Fusarium, are present at significantly higher abundances in UM and DM [33]. In this study, we revealed significant differences in the microbial community between NM and DM/UM, suggesting that successful morel soil has a unique microbial structure. NM exhibited a high ACE index for both fungal and bacterial communities, indicating a high microbial richness. Interestingly, NM showed a low Shannon index for fungi and a high Shannon index for bacteria, likely due to extensive colonization by morel mycelium in the soil, which inhibited fungal growth and reduced fungal species diversity. These findings are consistent with Tan H’s research, which demonstrated that soil microbial diversity is higher in soils where mushrooms are successfully cultivated [12].

Using LEfSe and random forest analyses, we identified biomarkers in different soil types. Regarding fungal composition, Morchella was the main marker of NM, indicating successful colonization and fruiting body production. In UM, both methods identified Botryotrichum, Humicola, and Ascobolus. Botryotrichum and Humicola belong to the family Chaetomiaceae within the class Sordariomycetes, most of which are thermophilic [38]. Similarly, Ascobolus and Morchella both belong to the order Pezizales, indicating potential competitive growth. Given that morels are psychrophilic, we hypothesize that the absence of fruiting bodies in UM was due to excessively high greenhouse temperatures during management. High temperatures likely inhibit morel growth while promoting competing microorganisms. No biomarkers were detected for DM using LEfSe (LDA score > 4). Most biomarkers identified using the random forest analysis belonged to the class Sordariomycetes, suggesting that the morel death in DM might have been due to high temperatures. In terms of bacterial composition, the bacterial genera Pedobacter and Massilia were identified as biomarkers of NM. Most Pedobacter are facultative psychrophilic bacteria with notable biodegradation ability, capable of degrading oil hydrocarbons. This genus exhibits various enzyme activities, including oxidase, catalase, DNase, protease, amylase, β-glucosidase, β-galactosidase, and β-lactase [39]. Massilia can solubilize phosphate, which enhances soil fertility [40]. We speculate that these two bacterial genera may have a symbiotic relationship with morels, supplying nutrients for their growth. MND1 was the primary biomarker of DM, and it is associated with ammonia oxidation and potential nitrogen loss. In UM, Planctomycetes was a major bacterial class. Members of this class are related to ammonia oxidation; therefore, the identification of this biomarker indirectly indicates a high ammonia nitrogen content in DM, which can adversely affect mycelium [13,41,42]. In the fungal and bacterial co-occurrence networks, NM had more nodes and modules, indicating a complex microbial community structure, symbiotic relationships conducive to morel growth, and ecological resistance to disturbance [43]. The phyla Proteobacteria and Bacteroidota dominated in NM modules. These phyla have also been found to dominate in other soils where morels are cultivated [30,36,44], indicating that most symbiotic microorganisms associated with morels belong to these phyla. In DM and UM, nodes with higher connectivity included pathogenic genera such as Mortierella, Spizellomyces, Rhizophagus, and Ascochyta [8,19,45]. In NM, highly connected nodes, such as Pedobacter, Massilia, and Bullera, may have symbiotic relationships with morels [36]. The genus Pseudomonas is typically found in symbiotic relationships with morels; however, this was not observed in the current study, possibly due to its abundance across all soil types [30,44].

5. Conclusions

We observed significant differences in microbial community structure and physicochemical properties between soil in which morels were fruited normally (NM) and soil in which cultivation failed (DM and UM). These differences provide critical insights into the factors influencing morel growth and yield. In normal fruiting soil (NM), morels dominated the microbial communities, indicating their successful establishment and growth. Despite this dominance, the soil still maintained a high diversity and complex symbiotic networks, suggesting that a balanced microbial community may be essential for morel development. Normal fruiting soil also had a relatively high nitrogen and organic matter content, which are known to support mycelial growth and biomass accumulation. In contrast, nutrients like potassium and phosphorus were significantly lower in NM, likely due to their extensive consumption during the fruiting process, which may limit subsequent yields. Biomarker analysis indicated that genera such as Pedobacter and Massilia may coexist with morels, promoting growth. In soils where cultivation was unsuccessful (DM and UM), a high abundance of pathogenic bacterial genera and unfavorable conditions (e.g., high temperature) likely inhibited morel growth and nutrient transfer from external sources, leading to cultivation failure. These findings highlight the importance of maintaining a favorable microbial community and soil physicochemical environment for successful morel cultivation.