ABSTRACT
Ectomycorrhizal (ECM) fungi, symbiotically associated with woody plants, markedly improve the uptake of mineral nutrients such as potassium (K) and phosphorus (P) by their host trees. Although it is well known that ECM fungi can obtain K and P from soil minerals through biological weathering, the mechanisms regulating this process are still poorly understood at the molecular level. Here, we investigated the transcriptional regulation of the ECM fungus Amanita pantherina in weathering K-containing feldspar and apatite using transcriptome sequencing (RNA-seq) and validated these results for differentially expressed genes using real-time quantitative PCR. The results showed that A. pantherina was able to improve relevant metabolic processes, such as promoting the biosynthesis of unsaturated fatty acids and steroids in the weathering of K-containing feldspar and apatite. The expression of genes encoding ion transporters was markedly enhanced during exposure to solid K-containing feldspar and apatite, and transcripts of the high-affinity K transporter ApHAK1, belonging to the HAK family, were significantly upregulated. The results also demonstrated that there was no upregulation of organic acid biosynthesis, reflecting the weak weathering capacity of the A. pantherina isolate used in this study, especially its inability to utilize P in apatite. Our findings suggest that under natural conditions in forests, some ECM fungi with low weathering potential of their own may instead enhance the uptake of mineral nutrients using their high-affinity ion transporter systems.
IMPORTANCE In this study, we revealed the molecular mechanism and possible strategies of A. pantherina with weak weathering potential in the uptake of insoluble mineral nutrients by using transcriptome sequencing (RNA-seq) technology and found that ApHAK1, a K transporter gene of this fungus, plays a very important role in the acquisition of K and P. Ectomycorrhizal (ECM) fungi play critical roles in the uptake of woody plant nutrients in forests that are usually characterized by nutrient limitation and in maintaining the stability of forest ecosystems. However, the regulatory mechanisms of ECM fungi in acquiring nutrients from minerals/rocks are poorly understood. This study investigated the transcriptional regulation of A. pantherina weathering K-containing feldspar and apatite and improves the understanding of fungal-plant interactions in promoting plant nutrition enabling increased productivity in sustainable forestry.
INTRODUCTION
Potassium (K) and phosphorus (P), two of the most important macronutrients for plant growth (1, 2), are abundant components of soil, but their low availability (often in the form of minerals) can limit terrestrial plant growth and ecosystem productivity. In temperate and boreal forest ecosystems, woody plants develop efficient strategies to improve the uptake of K and P from soil. The symbiotic association of trees with ectomycorrhizal (ECM) fungi is one of the most important strategies for improving the bioavailability of mineral nutrients from soil. Forest trees are largely dependent on their fungal symbionts for the uptake of mineral nutrients; in return, the fungi can obtain photosynthetically fixed carbon from their tree hosts (3, 4).
ECM fungi form symbiotic associations with forest trees, such as the Pinaceae, Betulaceae, Salicaceae, Fagaceae, and Dipterocarpaceae (4). They play a critical role in improving the uptake of plant nutrients in natural ecosystems that are usually characterized by nutrient limitation, especially with regard to N, K, and P (4–8). Plant nutrients, with the exception of nitrogen, are ultimately derived from weathering of terrestrial primary minerals (9). Lindahl et al. (10) revealed that ECM fungi dominate the total fungal community in the B horizon soil, suggesting that these fungi have the potential to obtain nutrients from soil minerals. K and P in forest soil often occur in mineral forms, but ECM fungi can effectively obtain these elements in the minerals by bioweathering (11–16). This can involve hyphal tunneling (9, 17, 18), although this alone has been estimated not to make a significant contribution to total mineral weathering (19). However, the effective impact of fungi on mineral weathering is probably increased by the production of extracellular polymeric substances (EPS) that increase the effective surface area of contact with minerals (20). Biochemical weathering plays an important role in obtaining nutrients from soil minerals, and ECM fungi can weather soil minerals through local acidification around the hyphae and by exuding metal-complexing weathering agents such as organic acids and siderophores (14, 21) and enhancing the absorption of weathered products (22, 23). Griffiths et al. (24) found that the colonization of the ECM fungus Gautieria monticola can markedly increase the content of oxalic acid in the soil and thus can accelerate the weathering of minerals. In vitro studies also show that ECM fungi can secrete large amounts of oxalic acid to acquire K and P from phlogopite and apatite (15, 25, 26). Bonneville et al. (27) investigated the dissolution effect of individual hyphae of Paxillus involutus on biotite at the nanometer scale and found that acidification at the hypha-mineral interface accelerates the weathering of biotite. Smits et al. (26) investigated ectomycorrhizal weathering of apatite and found that the fungus P. involutus could increase the weathering rate of apatite by a factor of three. Although the promotion of mineral weathering through secretion of weathering agents by ECM fungi is well known, the molecular mechanisms regulating ECM fungal weathering are still poorly understood.
Some studies have reported on gene regulation and molecular mechanisms of mineral weathering by saprophytic fungi. Xiao et al. (28) found the upregulated expression of weathering-related genes (carbonic anhydrase, acetaldehyde dehydrogenase, and metal transporters) during Aspergillus fumigatus weathering of K-bearing mineral, indicating these genes play an important role in mineral weathering. For example, microbial carbonic anhydrase can catalyze CO2 hydration and significantly promote the weathering of minerals (29, 30). Wang et al. (31) found that Aspergillus niger can upregulate the biosynthesis of organic acids in the process of K-bearing feldspar weathering and simultaneously improve the expression of a large number of membrane ion transporters, such as Na+/K+-ATPase. The weathering of minerals by saprophytic fungi is driven by their demand for mineral nutrients, and the weathering effect on minerals will attenuate once the fungal demand for mineral nutrients is satisfied (31). However, in addition to meeting their own mineral nutrient requirement, ECM fungi in soil supply large amounts of mineral nutrients to tree hosts in exchange for plant-derived carbohydrates. The continued supply of hexoses to the fungi is contingent upon continued flow of nutrients such as P in the reverse direction to the plant (32), and the regulatory mechanisms of ECM fungal-driven mineral weathering may therefore differ greatly from those of free-living fungi.
Many studies have shown that ECM fungi can promote the uptake of mineral elements from soil by expressing high-affinity membrane ion transporters. Corratgé et al. (33) found that the K transporter (KT) HcTrk1 from the Trk family in the ECM fungus Hebeloma cylindrosporum expressed in Xenopus oocytes mediates the cotransport of K+ and Na+, and Garcia et al. (16) found that this transporter plays a major role in both K and P nutrition of the fungus (and of the host plant) grown in potassium-poor solutions. Increasing numbers of high-affinity phosphorus transporters (PTs) have also been discovered during recent years (34). At present, at least 47 PTs of 14 ECM fungi can be retrieved in the database (35). Tatry et al. (36) found that among the two PTs of H. cylindrosporum, one (HcPT1) is induced by P starvation, whereas the other one (HcPT2) is less dependent on P availability. ECM fungi need not only to promote the release of insoluble mineral elements in the soil through weathering but also to rapidly transport the released elements into hyphae and to host plants to exchange for C supply (37).
The process of mineral weathering by ECM fungi obviously involves complicated patterns of gene expression regulation, but the molecular mechanisms associated with this process are still unknown. To decipher the molecular regulation mechanism of ECM fungi in mineral weathering, we used Amanita pantherina, one of the most frequently encountered ECM fungi, forming symbiotic associations with Quercus in Nanjing, China, as a model fungus. The transcriptome of A. pantherina during weathering interactions with K-containing feldspar and apatite was investigated using transcriptome sequencing (RNA-seq) technology, and the patterns of differential expression of weathering-related genes that were identified were validated using real-time quantitative PCR (RT-qPCR), to gain insight into the molecular mechanisms involved in mineral weathering and nutrient uptake.
RESULTS
Changes in mineral composition.Mineral weathering by fungi is directly reflected by changes in mineral structure and composition after fungal action. We determined the changes of the mineral structure and the relative content of the phases before and after fungal weathering by X-ray powder diffraction (XRD) (Fig. 1). When A. pantherina was incubated on K-deficient Pachlewski’s medium but containing an equivalent amount of K in the form of feldspar (KD treatment), the weathering of A. pantherina caused a significant change in the relative content of the mineral phases, in which the K content of high K-bearing albite was reduced from 65.8% ± 1.6% to 56.4% ± 0.2% (t = 8.244, P = 0.002), and the α-quartz increased from 17.7% ± 1.0% to 30.2% ± 0.8% (t = −13.804, P < 0.001). When A. pantherina was incubated in Pachlewski’s medium with a low level of soluble P but containing an equivalent amount of P in the form of apatite (PD treatment), the XRD data showed that A. pantherina had a very weak weathering effect on apatite. The relative contents of fluoroapatite and hydroxyapatite in the mineral hardly changed, and only the proportion of the amorphous phase increased slightly (Fig. 1B). When A. pantherina was incubated in K-deficient Pachlewski’s medium with a low level of soluble P but containing equivalent amounts of K and P in the form of the feldspar and apatite (KPD treatment), the XRD data showed that the mean content of albite changed from 28.4% to 24.8% after incubation with A. pantherina, but this difference was not statistically significant, while the mean relative contents of fluoroapatite and hydroxyapatite showed no significant change. The mean content of amorphous material increased from 16.9% to 20.5% in the mixed minerals, but this difference was not significantly different with the degree of replication used in this experiment.
(A, B, and C) Changes of structure and relative content of minerals after the weathering of A. pantherina. Values are means ± SDs.
Differentially expressed genes during weathering of K-containing feldspar and apatite.To identify differentially expressed genes (DEGs) of A. pantherina related to mineral weathering, we used Illumina paired-end sequencing technology to characterize fungal transcripts under different culture conditions. In total, the transcriptome sequencing yielded 347,835,310 reads with a GC content of each sample of 49.0% to 50.0% and a quality score threshold of 30 (Q30) value between 91.56% and 92.73%. A total of 35,346 unigenes was successfully assembled, with the shortest unigene being 201 bp and the longest unigene being 17,958 bp (Table 1). Examination of the length distribution of assembled unigenes showed that 12,667 unigenes (35.84%) were less than or equal to 300 bp, 4,823 unigenes (13.64%) ranged from 301 to 400 bp, 13,465 unigenes (38.09%) ranged from 401 to 2,000 bp, and 4,391 unigenes (14.42%) exceeded 2,000 bp (see Fig. S1 in the supplemental material).
Summary of the A. pantherina transcriptome
When comparing gene expression in the KD treatment to that when A. pantherina was incubated on solid full-strength Pachlewski’s medium (ND control), a set of 281 DEGs was identified, and among these, 174 DEGs were upregulated, while 107 DEGs were downregulated (Fig. 2A). Compared to gene expression in ND, 119 DEGs were upregulated and 96 DEGs downregulated in the PD treatment (Fig. 2B). More DEGs were found in KPD compared with expression in KD and PD. The results showed that 199 DEGs were significantly upregulated and 371 DEGs were significantly downregulated in the KPD treatment compared with expression in the ND control (Fig. 2C).
(A, B, and C) Volcano plots of differentially expressed genes (DEGs) between treatments. The x axis represents log2 (fold change) of DEGs, and the y axis represents the −log10 P value indicating the significance of the difference. The dots represent the unigenes; the blue dots indicate downregulated DEGs; the red dots represent upregulated DEGs; the gray dots represent genes that are not differentially expressed between treatments.
GO enrichment analysis.To understand the involvement of DEGs in different treatments, Gene Ontology (GO) classification was performed, classifying functions according to three nonoverlapping terms: biological process, cellular component, and molecular function. Figure 3 shows the top ten upregulated GO terms significantly enriched during weathering of K-containing feldspar and apatite. In KD, the significantly enriched GO categories of biological processes were protein synthesis and degradation, such as ribosome synthesis and mitochondrial respiratory chain complex III synthesis, and the significantly enriched categories of GO cell components and molecular functions mostly belonging to transcription and translation (Fig. 3A). In PD, the enriched GO categories of biological processes included protein synthesis, transport, and degradation, endocytosis, and proton transport (Fig. 3B). The enriched GO categories of cell components were mainly cell and organelle membranes such as mitochondria and Golgi membrane. The enriched GO categories of molecular functions were mainly binding, transcriptional activator activity, oxidoreductase activity, endonuclease activity, and endopeptidase activity. In KPD, the enriched GO categories of biological processes included protein synthesis and degradation, endocytosis, and transcription (Fig. 3C). The enriched GO categories of cell components were mainly related to ribosome, mitochondrial membrane, and nucleus, while the upregulated molecular functions were primarily connected with oxidation activity (such as cytochrome c oxidation activity), endopeptidase activity, transcription factor activity, etc. These significantly enriched GO classifications may be a response to the weathering of K-containing feldspar and apatite by the fungus.
Enrichment of GO classification for different treatments: (A) KD versus ND; (B) PD versus ND; (C) KPD versus ND.
KEGG pathway enrichment analysis.KEGG (the Kyoto Encyclopedia of Genes and Genomes) is a database resource for revealing high-level functions of genes, and the KEGG pathway enrichment analysis can map DEGs under nutrient-deficient conditions to certain metabolic pathways. Figure 4 shows the top ten upregulated pathways during weathering of K-containing feldspar and/or apatite. The predominant categories of enriched pathways were amino acid synthesis (ko00300, ko00400, ko00270, and ko01230), ribosome (ko03010), and aminoacyl-tRNA biosynthesis (ko00970) in KD (Fig. 4A). In PD, the results of KEGG pathway analysis suggested that cell signaling (ko04310, ko03320, and ko04150), amino acid degradation (ko00280), and oxidative phosphorylation (ko00190) were particularly active (Fig. 4B). In KPD, the main upregulated pathways were related to ribosome (ko03010), gene expression (ko03010 and ko03022), metabolism of sterols and sphingolipid (ko00100 and ko00600), and transmembrane transport of substances such as ABC transporters (ko02010) (Fig. 4C).
KEGG pathways significantly enriched related to upregulated DEGs in KD (A), PD (B), and KPD (C).
RT-qPCR validation.Validation, using RT-qPCR, of selected genes in each treatment that were significantly upregulated according to RNA-seq (Table 2), confirmed the patterns of up- and downregulation at different culture times (20 days, 30 days, and 40 days). At 20 days, the expression levels of all selected genes were markedly upregulated during weathering of K-containing feldspar and/or apatite (Fig. 5). As the culture time increased, genes detected in the control group were also upregulated to different degrees. The expression level of ApHAK1 was significantly upregulated whether in KD, PD, or KPD. In KD, the expression level of ApHAK1 increased with the culture time, with increases of approximately 6-fold at 20 days, 612-fold at 30 days, and 115-fold at 40 days. In PD, the expression of ApHAK1 reached the peak at 30 days, upregulated approximately 2,218-fold. However, in KPD, the expression of ApHAK1 was upregulated approximately 27,000-fold at 40 days, which was the highest in all treatments. Surprisingly, the levels of expression of all the PT genes did not vary under PD or KPD conditions. This could be due to the supply of soluble P (0.7 mM) in the medium containing apatite to get fungal growth. To assess which PT genes responded to soluble P starvation, we grew the fungus with low levels of available K (50 μM) and/or P (20 μM) for different lengths of time and measured the expression of KT and PT genes. The results showed that the expression levels of seven cytoplasmic PTs found by high-throughput sequencing were all upregulated, but ApPT2 and ApPT3 were more closely related to soluble P starvation, which increased approximately 46- and 94-fold at 20 days, 60- and 59-fold at 30 days, and 18- and 16-fold at 40 days, respectively, under low soluble P conditions (Fig. 6). Under both low soluble K and P conditions, the expression levels of ApPT2 and ApPT3 were the most upregulated as well.
Selected upregulated transcripts for validation using RT-qPCR
Expression levels of selected genes at different incubation times validated by RT-qPCR in KD (A), PD (B), and KPD (C). The expression of each gene of A. pantherina incubated on full-strength medium at 20 days was set as 1.
(A, B, and C) Expression levels of A. pantherina K transporter and phosphorus transporter genes in low soluble K and/or P with different incubation phases. The expression of each gene was set as 1 at the 20th day of full-strength medium. LK, Pachlewski’s medium with low soluble K (KH2PO4 replaced by 1.15 g NaH2PO4·2H2O and 25 μM K2SO4); LP, Pachlewski’s medium with low soluble P (KH2PO4 replaced by 20 μM NaH2PO4·2H2O and 1.28 g K2SO4); LKP, Pachlewski’s medium with low soluble K and P (KH2PO4 replaced by 20 μM NaH2PO4·2H2O and 25 μM K2SO4).
DISCUSSION
ECM fungi have evolved repeatedly from saprotrophic ancestors (38, 39), forming symbiotic associations with the Pinaceae in the Cretaceous period (ca. 100 million years ago) according to fossil and molecular clock evidence (39, 40). To date, many ECM fungi still retain some of the characteristics of saprophytic fungi but contain limited numbers of genes for plant cell wall-degrading enzymes (39). Therefore, we can isolate the pure cultures of ECM fungi from the wild, making it possible to study the biological function of ECM fungi in vitro. In forest ecosystems, ECM fungi improve the mineral nutrition of trees, especially for N, K, and P (3–6, 34). However, the molecular regulation mechanisms of ECM fungi in weathering minerals to obtain nutrients from soil minerals are still a black box. Our results showed that A. pantherina can weather albite and reduce the relative content of K in the mineral (Fig. 1). However, the results also showed that A. pantherina could not acquire P from apatite, even when small amounts of soluble P were supplied. The feldspar used in this study was a mixture of high-K-bearing albite quartz: obviously, the fungus can acquire K from the albite, whereas the composition of apatite is mainly fluoroapatite and hydroxyapatite, which are more resistant to weathering by A. pantherina. Using high-throughput sequencing, a large amount of transcriptome data was obtained under different conditions (see Data Set S1 in the supplemental matieral), allowing us to more clearly understand the molecular regulation mechanism of A. pantherina in response to nutritional deficiencies. The assembled transcripts were assigned to KOG, GO, and KEGG databases, enabling a wide range of comparisons.
Molecular regulation mechanism of A. pantherina during weathering of K-containing feldspar.Under conditions of soluble K deficiency, but in the presence of added albite, the dominant biological process GO category was protein synthesis, indicating that A. pantherina needed to synthesize various proteins or enzymes to catalyze multiple biological reactions during weathering of albite (Fig. 3A). The upregulated biological process of mitochondrial respiratory chain complex III assembly suggests that the fungus needed to supply more energy for activated metabolism by enhancing mitochondrial respiration. Cytosol and structural constituents of ribosomes were the largest GO categories of cellular components and molecular functions, respectively. The KEGG pathway enrichment analysis showed that 4 of the top 10 metabolic pathways were related to amino acid synthesis (Fig. 4A). The lack of K greatly promoted the synthesis of various amino acids, which not only were synthetic components of proteins but might also be secreted to extracellular substrates involved in mineral weathering.
The expression levels of the five selected DEGs validated by RT-qPCR were substantially higher than those indicated by RNA-seq, except for ApMtC, further suggesting that these genes were involved in K acquisition. The upregulated A. pantherina ApWD40 gene product belongs to the WD40 repeat-containing protein family, found in all eukaryotes and implicated in a variety of functions such as signal transduction and transcription regulation (41). ApWD40 may play a key role in the initial signal transmission of extracellular K stress. The RT-qPCR results confirmed the upregulated transcripts of steryl acetyl hydrolase ApSAH, associated with sterol metabolism. Sterol is involved in the synthesis of fungal ergosterol, which is thought to be related to fungal membrane structure, signaling, and an adverse environment (42). Wang et al. (43) investigated the gene expression of A. niger under soluble K deficiency using RNA-seq and found that the biosynthesis of organic acids and KT (Na+/K+-ATPase) were significantly upregulated. It is well known that organic acids play a critical role in the process of mineral weathering (14, 15, 25, 44). In our results, no DEGs associated with the synthesis of organic acids were found. This might indicate that A. pantherina cannot weather minerals by excreting organic acids. However, the massive network of ectomycorrhizal fungal hyphae can greatly improve nutrient uptake (45), which implies that ion transporters of ECM fungi play an important role in nutrient absorption. As expected, the transcripts of high-affinity KT ApHAK1 were found among the DEGs. ECM fungi can constantly absorb K from the environment and store it in fungal vacuoles (46, 47). In particular, large amounts of K sequestered in rhizomorphs facilitate the long-term exchange of photosynthetic products of host plants (20, 48). ApHAK1 was the only upregulated transcript in the four KT genes identified by RNA-seq, suggesting that not all A. pantherina KTs respond to the low availability of K. The expression levels of the four KT genes at low K+ (50 μM) were different during incubation, and ApHAK1 was indeed closely related to K availability, while ApHAK2 and ApHAK3 were less closely related (Fig. 6). Tatry et al. (36) found a similar phenomenon for H. cylindrosporum PTs, in which increased expression levels of HcPT1 transcripts are induced by P starvation in pure culture, but the transcript levels of HcPT2 are less dependent on P availability. A. pantherina may evolve different types of KTs to effectively cope with different extracellular K+ levels and ensure the uptake of K+.
Molecular regulation mechanism of A. pantherina in apatite weathering.Under conditions of low levels of soluble P but in the presence of apatite, the biological process GO categories intracellular protein transport, signal transduction, and proton transport were particularly enriched (Fig. 3B). Signaling is a key process in which fungi respond to apatite weathering stress and make metabolic adjustments. The serine/threonine kinase containing WD40 repeating unit ApSeThK was upregulated during apatite weathering, verified by RT-qPCR, which may involve an inward transfer low-P signal. ECM fungi often depend on P:H+ symporters to take up extracellular P (36, 49, 50). The efficiency of P uptake through H+-dependent PTs (P:H+ symporters) relies strongly on low external pH values (35) to a certain extent, which explains the fact that the optimal pH of A. pantherina is acidic (see Fig. S2). The enrichment of the proton transport category might involve the retention of extracellular proton gradients. Membrane was the largest cellular component GO category. In the process of mineral weathering, A. pantherina needs to secrete weathering agents continuously and transfer weathered products into its hyphae, leading to the frequent turnover of cell and organelle membranes.
The KEGG pathway enrichment analysis also showed that unsaturated fatty acid synthesis and oxidative phosphorylation occurred in the top 10 significantly enriched categories (Fig. 4B). Fungus-driven mineral weathering is a comprehensive process, and improving the structure of the cell membrane contributes significantly to the physiological function of the membrane proteins. The enhancement of oxidative phosphorylation can provide more ATP for all physiological processes, which is beneficial for P acquisition by the fungus. We found a significant upregulation of ApHAK1 (Table 2, Fig. 4B), indicating that ApHAK1 was also closely related to P acquisition. Unexpectedly, the transcripts of seven PTs were not upregulated. However, the role of A. pantherina PTs in apatite weathering cannot be dismissed, since the expression of some PTs in ECM fungi is not affected by P availability (37). Another possible explanation of the results is that, in the PD treatment, the fungus is not really under a P starvation condition due to the addition of a low level of soluble P to the medium. We further tested the expression levels of these PTs in A pantherina at low P (20 μM PO43−), and the results showed that ApPT2 and ApPT3 were more closely related to P availability while ApPT4 was less closely related (Fig. 6).
In this study, we found that A. pantherina was unable to utilize structural P in apatite, suggesting the ability to use insoluble phosphates varies between different taxa of ECM fungi (51). Wallander et al. (52) also showed that P uptake from apatite by ECM fungi is not a common phenomenon. The results of the present study suggest that A. pantherina may acquire P from apatite through other means under field conditions, such as by enriching bacteria with high mineral-weathering potential. Frey-Klett et al. (53) found that the ectomycorrhiza of Laccaria bicolor Douglas fir can enrich Pseudomonas fluorescens populations that have the ability to solubilize inorganic phosphates, in contrast to the majority of P. fluorescens from the bulk soil, indicating some phosphate-solubilizing bacteria can interact synergistically with ECM fungi and promote sustainable P supply to plants. The recent study by Liu et al. (54) also shows that ECM fungi can enrich bacterial communities with respect to different ecological functions in the hyphosphere, such as mineral weathering by bacteria in the genera Sphingomonas (55) and Burkholderia (56, 57). Obviously, the selection of bacteria with weathering potential can compensate for the limited ability of particular ECM fungi to take up nutrients from soil minerals.
Molecular regulation mechanism of A. pantherina in the weathering of K-containing feldspar and apatite.The dual stress of soluble K deficiency and low level of soluble P caused A. pantherina to upregulate more genes (Fig. 2). However, most of the enriched GO categories of biological processes, cellular components, and molecular functions were the same as under conditions of low-level soluble P or soluble K deficiency (Fig. 3). KEGG pathway enrichment analysis showed that metabolic processes such as ribosome, steroid biosynthesis, basal transcription factors, sphingolipid metabolism, and ABC transporters were the major categories (Fig. 4C). In the limitation of K and P, A. pantherina also demonstrated the same metabolic responses to nutrient deficiencies. First, the expression of weathering-related genes was initiated, and membrane structure was adjusted to promote the transport of metabolic substances, such as accelerating proton transport; then, the upregulated ion transporters improved the transport of weathering products into fungal hyphae. Among the selected DEGs for RT-qPCR validation, the expression of the cytochrome-encoding gene ApCyt gradually increased during incubation (Fig. 5). Cytochromes are primarily involved in cellular respiration and energy production, indicating that A. pantherina needs more energy under condition of limitation of K and P. The expression level of ApHAK1 was upregulated in KPD as well, further confirming that ApHAK1 was also closely related to P acquisition. In fungi, at least three K transport systems have been reported, TrK (transporter of K), HAK (high-affinity K uptake) transporters, and K channel proteins (2). Based on phylogenetic analysis of predicted and functionally characterized KTs, A. pantherina ApHAK1 belongs to the HAK family (Fig. 7A). ApHAK1 has 12 predicted transmembrane domains (Fig. 7B and C), located in the cytoplasmic membrane from subcellular prediction using WoLF PSORT (https://wolfpsort.hgc.jp/), more than those in the first identified HAK transporter of Schwanniomyces occidentalis SoHAK1, but has the same predicted number of transmembrane domains as the Neurospora crassa NcHAK1 (see Fig. S3). The only two functionally described KTs in the HAK family of fungi, SoHAK1 (58) and NcHAK1 (59), are H+-dependent transporters, suggesting that A. pantherina ApHAK1 may be an H+-dependent KT as well. It is assumed that most HAK KTs may be K+-H+ symporters (2). Interestingly, the metabolic pathways of unsaturated fatty acid, terpenoid backbone, sphingolipid, and steroid were enriched in both PD and KPD treatments. Lipids, steroids, and terpenoids are usually the precursors of signal molecules, such as hydroxy fatty acids (60) and lipochitooligosaccharides (61, 62). This further indicated that although A. pantherina could not utilize apatite, it might enrich P-solubilizing bacteria by synthesizing and secreting signal molecules in the field to assist P absorption.
Phylogenetic tree of KTs constructed using Mega 7.0 software by neighbor-joining method (A), the predicted transmembrane domains (B), and schematic structural model based on SWISS-MODEL server (C) of ApHAK1. Values of the phylogenetic tree represent the bootstrap confidence tested using 1,000 replicates of the data set. Am, Amanita muscaria; Hc, H. cylindrosporum; Nc, N. crassa; Sc, Saccharomyces cerevisiae; So, S. occidentalis; Sp, Schizosaccharomyces pombe. Accession numbers: AmHAK1, KIL65804; AmHAK2, KIL55965; AmHAK3, KIL55966; ApHAK1, MH281948; HcTrK1, CAL36606; HcTrK2, CAL36607; NcTrK1, CAA08813; NcHAK1, CAA08814; ScTrK1, CAA89424; ScTrK2, CAA82128; SoTrK1, CAB91046; SoHAK1, AAB17122; SpTrk1, CAA93300.
In summary, this study characterizes the molecular regulation of A. pantherina in response to mineral nutrient deficiencies in the external environment. First, nutritional deficiencies stimulated the activation of cell surface signaling receptors. Subsequently, signals were transmitted intracellularly to activate relevant gene expression, showing a similar molecular regulation mechanism to that hypothesized by Xiao et al. (63) in Aspergillus nidulans during weathering of K-containing minerals. In forests, ECM fungi can recruit mineral-weathering bacterial communities from the bulk soil (53–55) that play an important role in mineral dissolution. Our results suggested that ECM fungi might adopt different strategies for acquiring nutrients based on their characteristics. For example, some ECM fungi with weak weathering potential, such as A. pantherina, might compensate for functional deficiencies by enriching bacteria with high weathering potential, which can be recruited through secreting special signaling molecules (Fig. 4B and C), and upregulating the expression of special ion transporters.
MATERIALS AND METHODS
Preparation of ECM fungi and minerals.The ECM fungus used in this study was A. pantherina LS08 (GenBank accession number KR456156), isolated from a fruiting body of A. pantherina associated with Quercus acutissima on the Nanjing Normal University campus. The fruiting bodies of A. pantherina mainly form from June to October each year on the campus (Fig. 8). The optimum growth temperature of isolated cultured strains was 25°C (see Fig. S4 in the supplemental material), and the optimal pH is approximately 5 (Fig. S2). The isolate can utilize various types of carbohydrates, such as monosaccharides, disaccharides, and polysaccharides such as starch (see Fig. S5).
The morphological characteristics of an aboveground fruiting body of A. pantherina 1 to 4 days following emergence.
The fungus was isolated on Pachlewski’s medium with the following composition (per liter): maltose, 5.0 g; glucose, 20 g; ammonium tartrate, 0.5 g; KH2PO4, 1 g; MgSO4·7H2O, 0.5 g; thiamine, 100 μg; microelements, including ZnSO4·7H2O, 0.63 mg; MnSO4·H2O, 1.54 mg; CuSO4·5H2O, 0.25 mg; (NH4)6Mo7O24·4H2O, 0.05 mg; Fe(III)-EDTA, 3.44 mg; Bacto agar, 20 g. The pH was adjusted to 5.5 with NaOH.
K-containing feldspar was used as an insoluble K source in this study, and the chemical composition of the mineral tested by X-ray fluorescence was as follows: Si, 38.40%; O, 37.70%; K, 9.61%; Al, 8.62%; Na, 2.31%; C, 1.88%; Ca, 0.65%; and Fe, 0.65%. The feldspar was mainly composed of high K-bearing albite (65.8%) and quartz (17.7%). Apatite was used as an insoluble P source and the chemical composition of the mineral tested by X-ray fluorescence was as follows: O, 77.35%; Ca, 11.10%; P, 6.09%; F, 1.58%; Mg, 1.36%; Si, 1.33%; Al, 0.35%; Fe, 0.29%; Sr, 0.23%; K, 0.19%; S, 0.06%; I, 0.06%; and Y, 0.01%. Apatite is mainly composed of fluoroapatite (45.6%) and hydroxyapatite (41.2%). The mineral samples (75 to 150 μm) were cleaned ultrasonically using alcohol for 30 min, subsequently washed three times with deionized water, and then oven dried overnight at 60°C.
Cultivation and collection of A. pantherina.To reveal weathering-related genes and metabolic pathways of A. pantherina, we set up four different treatments: treatment ND, A. pantherina was incubated on solid full-strength Pachlewski’s medium; treatment KD, A. pantherina was incubated on K-deficient Pachlewski’s medium that contained an equivalent amount of K in the form of feldspar; treatment PD, A. pantherina was incubated in Pachlewski’s medium with a low level of soluble P but containing an equivalent amount of P in the form of apatite; treatment KPD, A. pantherina was incubated in K-deficient Pachlewski’s medium with a low level of soluble P but containing equivalent amounts of K and P in the form of the feldspar and apatite. The detailed addition methods of K and P sources in each treatment of this experiment are shown in Table 3. Briefly, the mineral particles (1 g) were spread evenly on agar plates (20 ml medium). Then, the agar plates were directly inoculated with round agar plugs (diameter, 0.5 cm) taken from the peripheral growth zone of actively growing fungal colonies grown on Pachlewski’s medium for 30 days inoculated in different treatments. The petri dishes (diameter 9.0 cm) were sealed with Parafilm (Bemis, USA), and the fungus was allowed to grow in the dark at 25°C for 30 days; then, fungal mycelia were scraped off the plate using a sterile scalpel, quickly frozen in liquid nitrogen, and stored in a refrigerator at −80°C. Each treatment in this study was performed in triplicate.
The details of K and P added in each treatmenta
Construction of cDNA library, Illumina sequencing, and data analysis.Total RNA was extracted using a Fungal Total RNA Isolation kit (Sangon Biotech, China), and the integrity and concentration were evaluated using an Agilent 2100 Bioanalyzer (Agilent Technologies, USA). Total fungal RNA was purified to obtain mRNA using oligo(dT) magnetic beads. The cDNA libraries were constructed using an Illumina TruSeq RNA Sample Prep kit (Illumina, USA) according to the manufacturer’s protocol. The quality of the constructed cDNA libraries was examined on an Agilent Technologies 2100 Bioanalyzer prior to sequencing with Illumina HiSeq 2500 at Shanghai OE Biotechnology Co., Ltd. (Shanghai, China).
To ensure sequencing data quality, the original sequencing data were evaluated using FastQC version 0.11.5 software (http://www.bioinformatics.babraham.ac.uk/projects/fastqc/). NGS QC TOOLKIT v2.3.3 software (64) was then used to remove low-quality reads and adaptor reads. Reads with sequences containing the adaptor sequence, paired reads with the amount of N contained in single-end sequencing reads exceeding 35 bp, and paired reads with low-quality (Phred score, <20) base number of the single-end sequencing reads greater than 70% of the read length were thus excluded. The frequency of remaining valid clean reads was 96.33%. Clean reads were assembled de novo using Trinity software (65). The assembled reads were clustered using the TGICL software (66) to yield unigenes that cannot be extended on either end, and redundancies were removed to acquire nonredundant unigenes. Fragments per kilobase of transcript per million mapped reads (FPKM) was used to calculate the gene expression. Therefore, FPKM values were directly used to compare gene expression differences between different samples. The DESeq package was used to obtain the “base mean” value to identify differentially expressed genes (DEGs). In this study, a P value of <0.05, a fold change (FC) of >2 (|log2FC| > 1), and a base mean of >50 in at least one treatment were set as thresholds to define the significance of gene expression differences between two treatments (Data Set S1). Unigenes were searched against the National Center for Biotechnology Information (NCBI) nonredundant protein (Nr) database, the Swiss-Prot protein database (67), and the Eukaryotic Orthologous Groups (KOG) (68) using BLASTx algorithm with an E value cutoff of 1E−5. The Blast2GO software was applied to get the annotation results of unigenes in the Gene Ontology (GO) database (69). The pathways were also annotated according to the KEGG database (70).
Validation of weathering-related transcripts by RT-qPCR.In combination with GO and KEGG pathway enrichment analysis, genes supposed to be related to mineral weathering with the highest expression level in the top five with annotated information in each treatment were determined by RT-qPCR in order to validate the reliability of DEG data generated by the Illumina RNA-seq (Table 2). Total RNA was extracted from each sample, and the concentration was measured using a NanoDrop 2000 spectrophotometer (Thermo Scientific, USA). The reverse transcription of RNA into cDNA was performed using HiScript II Q RT SuperMix for qPCR (+gDNA wiper) kit (Vazyme Biotech, China). An AceQ qPCR SYBR Green PCR master mix kit (Vazyme Biotech, China) was used for RT-qPCR reactions according to the manufacturer’s instructions. Parameters for the quantitative amplification using an ABI StepOne Real-Time PCR system (Applied Biosystems, USA) were as follows: 5 min at 95°C for enzyme activation and 40 cycles of 95°C for 10 s and 60°C for 30 s. The housekeeping gene encoding glyceraldehyde 3-phosphate dehydrogenase (GAPDH) was used as an internal control for normalization. The gene expression level was determined using the threshold cycle (2−ΔΔCT) method proposed by Pfaffl (71). Primers used in RT-qPCR are listed in Table 4, designed by IDT (Coralville, IA, USA). Three commonly used internal reference genes in fungi were selected, encoding β-actin, β-tubulin, and GAPDH. After the analyses of expression stability and amplification efficiency, the expression level of the β-actin gene was found to be relatively stable under different conditions, and its amplification efficiency was the highest (103.18%); thus, it was selected as the internal reference gene in this study (see Fig. S6).
Primers of genes used in this study
Mineralogical analysis.The structure and composition of minerals were determined using an X-ray diffractometer BTX-526 (Olympus, Japan) using Co-Kα radiation with a voltage of 30 kV and current of 300 μA (λ = 1.79 Å).
Statistical analyses.The data are expressed as means ± standard deviations (SDs) from three replicates. The significances of data of RT-PCR and relative content of mineral were determined with t tests, using IBM SPSS Statistics 20.0 software.
Accession number(s).The raw sequencing reads were submitted to the NCBI Sequence Read Archive (SRA) under the study number SRP134686, available at https://www.ncbi.nlm.nih.gov/sra/SRP134686.
ACKNOWLEDGMENTS
This work was supported by the National Natural Science Foundation of China (grant numbers 41772360 and 41373078).
We declare no conflict of interest.
FOOTNOTES
- Received 29 March 2019.
- Accepted 18 May 2019.
- Accepted manuscript posted online 24 May 2019.
Supplemental material for this article may be found at https://doi.org/10.1128/AEM.00719-19.
- Copyright © 2019 American Society for Microbiology.