Synergy of arbuscular mycorrhizal symbiosis and exogenous Ca2+ benefits peanut (Arachis hypogaea L.) growth through the shared hormone and flavonoid pathway

Peanut yield is severely affected by exchangeable calcium ion (Ca2+) deficiency in the soil. Arbuscular mycorrhizal (AM) symbiosis increases the absorption of Ca2+ for host plants. Here, we analyzed the physiological and transcriptional changes in the roots of Arachis hypogaea L. colonized by Funneliformis mosseae under Ca2+-deficient and -sufficient conditions. The results showed that exogenous Ca2+ application increased arbuscular mycorrhizal fungi (AMF) colonization, plant dry weight, and Ca content of AM plants. Simultaneously, transcriptome analysis showed that Ca2+ application further induced 74.5% of differentially expressed gene transcripts in roots of AM peanut seedlings. These genes are involved in AM symbiosis development, hormone biosynthesis and signal transduction, and carotenoid and flavonoid biosynthesis. The transcripts of AM-specific marker genes in AM plants with Ca2+ deprivation were further up-regulated by Ca2+ application. Gibberellic acid (GA3) and flavonoid contents were higher in roots of AM- and Ca2+-treated plants, but salicylic acid (SA) and carotenoid contents specifically increased in roots of the AM plants. Thus, these results suggest that the synergy of AM symbiosis and Ca2+ improves plant growth due to the shared GA- and flavonoid-mediated pathway, whereas SA and carotenoid biosynthesis in peanut roots are specific to AM symbiosis.

www.nature.com/scientificreports www.nature.com/scientificreports/ This symbiosis plays a significant role in the uptake of nutrients and the carbon cycle, and consequently impacts ecosystem sustainability 10 . To establish the symbiosis, plant roots recognize chemical signals from AMF, e.g. lipochitooligosaccharides and chitooligosaccharides, which trigger coordinated differentiation and form the symbiotic state 11 . In turn, AMF require signal communication from the plants that produce strigolactones (which are derived from the carotenoid synthesis pathway), flavonoids, and other diffusible signals exuded by plant roots that induce the germination of AMF spores and branched fungal hyphae 12 . Then, the AM symbiosis is established by the common symbiosis signaling pathway induced by calcium oscillation after perception of diffusible signals from the symbionts 13 . In the process of establishing the AM symbiosis, many AM-specific marker genes must be initiated by Ca 2+ concentration change 14 , such as RAM1 (REDUCED ARBUSCULAR MYCORRHIZA 1), RAM2 (glycerol-3-phosphate acetyltransferase), CCD1 (carotenoid cleavage dioxygenase), PT1 (phosphate transporter), and DELLA [15][16][17] . These findings suggest that Ca 2+ plays an important role in AM development.
From recognition of the fungi to establishment of the symbiosis, complicated transcriptional reprogramming occurs in plant roots, and many specifically expressed genes involved in development of the symbiosis have been identified in legumes [18][19][20] . Some of the changes associated with plant hormones were considered to play important roles in this symbiosis 21 , such as auxin, cytokinins (CKs), gibberellins (GAs), and strigolactones, and were also altered in the roots of AM plants 16,22 . In addition, increases in flavonoid and anthocyanin were considered to be indispensable in regulating the establishment of AM symbiosis 22,23 .
Even though the molecular basis of the improvement of plant nutrient acquisitions have been well characterized for phosphorus, nitrogen, sulfur, and potassium 18,24,25 , the role of plant uptake of Ca 2+ needs further study. Several reports showed that a moderate level of Ca 2+ supply enhanced the colonization of AMF 26,27 , and Ca 2+ benefited the maintenance of a functioning mycorrhiza 28 . However, the transcriptional changes in plant roots colonized by an AMF accompanied with sufficient Ca 2+ are still unknown. Cui et al. (2019) demonstrated that AM symbiosis increased the Ca 2+ content in peanut seedlings, and Ca 2+ application can also promote the development of AM symbiosis 29 . However, the molecular mechanism of how AMF and Ca 2+ application synergistically promote the growth of peanut seedlings is unclear. In this study, we investigated a combination of transcriptional changes, hormone and metabolomic analyses in roots of peanut seedlings inoculated by AMF and Ca 2+ application and compared the observed changes with those in AM plants or Ca 2+ -treated plants. We observed that changes in secondary metabolites in roots of AM and Ca 2+ -treated plants coincide with the transcriptional regulation of related biosynthesis pathways. These alternations, such as the increases in GA 3 and flavonoid content, were considered to be involved in the growth enhancement of peanut seedlings by the synergy of AMF with Ca 2+ application.

Results
AM symbiosis improves the dry biomass of peanut. The quantification of AMF colonization showed that 60.33% and 80.67% of plant roots were inoculated by F. mosseae under both Ca 2+ -deficient and Ca 2+sufficient conditions, respectively (Fig. 1A), indicating that Ca 2+ application could significantly improve the number of fungal colonizers.
Shoot dry weight significantly increased in the AM plants compared with the nonmycorrhized (NM) plants and Ca 2+ application further increased the shoot dry weight (Fig. 1B). Moreover, root dry weight was significantly increased in AM plants under Ca 0 conditions and Ca 2+ further improved the root dry weight; AM symbiosis did not increase the root dry weight (Fig. 1C). Additionally, the Ca 2+ content was significantly higher in Ca 2+ -sufficient seedlings compared with Ca 2+ -deficient ones, and AM association improved Ca 2+ level in roots (Fig. 1D).

Comparative analysis of differentially expressed genes (DEGs) in the AM and Ca 2+ -treated plants.
Using Ca 0 -AM as the control, there were 510, 1483, and 1795 significantly differentially expressed genes (DEGs) from roots of the Ca 0 + AM, Ca 6 − AM, and Ca 6 + AM plants, respectively ( Fig. 2A). In all, 304 DEGs were shared by Ca 0 + AM, Ca 6 − AM, and Ca 6 + AM plants and the number of DEGs gradually increased in the plants (Fig. 2B), indicating that AM symbiosis combined with exogenous Ca 2+ induced more transcriptional changes. In total, 421 DEGs were shared by Ca 0 + AM and Ca 6 + AM treatments, representing 82.55% and 23.45% of total DEGs in Ca 0 + AM plants (510) and Ca 6 + AM plants (1795), respectively. The expression levels of 380 DEGs in Ca 0 + AM plants could be further regulated by Ca 2+ application ( Supplementary Fig. S1); only 40 DEGs were conversely regulated (Supplementary Table S1). This result implied that Ca 2+ application could further strengthen the effects of AM on plant growth. In addition, 22 categories involved in molecular functions of GO enrichment analyses were identified, and the number of DEGs involved in transferase activity was the highest, followed by metal ion binding and oxidoreductase activity. Four categories, including calcium ion binding, signaling receptor activity, zinc ion binding, and antioxidant activity were the highest in Ca 6 − AM plants; among the other 18 categories, the number of DEGs involved in each molecular function of GO was the highest in the Ca 6 + AM plants, followed by the Ca 6 − AM plants, and the Ca 0 + AM plants (Fig. 2C).
To confirm RNA-Seq results, 15 genes were selected randomly from various functional categories and qRT-PCR analysis was conducted using RNA samples from the RNA-Seq experiments. The results were consistent with the expression levels of genes from RNA-Seq data (Fig. 3).
We further investigated the effects of AM symbiosis on Ca and Ca 2+ signal-related genes. The number of DEGs involved in Ca signals in the Ca 6 − AM and Ca 6 + AM plants was 29 and 32, respectively (Supplementary  Table S2). However, there were 14 DEGs shared by Ca 6 − AM and Ca 6 + AM plants and the transcript levels of nine of these DEGs were further regulated by AM symbiosis. Additionally, AM symbiosis specifically up-regulated the transcripts of Araip.IZ5U3 and Araip.R6YEY genes, which code the potassium channel KAT3 and AKT2/3, respectively. These results suggest that the Ca 2+ signal pathway induced by exogenous Ca 2+ is partially different from AM symbiosis.

Effects of AM symbiosis and Ca 2+ on genes involved in hormone biosynthesis. DEGs involved
in plant hormone biosynthesis were screened, including auxin, CKs, GA, and SA (Table 2). One gene encoding auxin responsive protein indoleacetic acid (IAA) was specifically up-regulated in AM plants without Ca 2+ application. Two genes belonging to the auxin responsive GH3 family were down-regulated in AM plants, and Ca 2+ application further down-regulated its transcripts. The genes encoding cytokinin dehydrogenase, which catalyze the irreversible degradation of CK, were either up-or down-regulated. In addition, we observed an increase in transcripts of genes involved in the biosynthesis of GA. Compared with the control, all DEGs encoding gibberellin 20-oxidase were up-regulated in AM plants, and more transcripts were observed in Ca 6 + AM plants. Two selected DEGs, namely, gibberellins 2-oxidase and gibberellin receptor GID1, were only up-regulated in Ca 6 + AM plants. Meanwhile, one TF TGA (Araip.FKG2G) involved in the biosynthesis of SA was specifically up-regulated by Ca 2+ application, and was further up-regulated by AM symbiosis.
In order to verify whether the hormone level is consistent with the transcriptional changes in DEGs, we tested the content of IAA, trans-zeatin riboside (tZR), GA 3 , and SA. IAA content significantly increased in AM plants with Ca 0 treatment, but decreased in Ca 6 treatment (Fig. 4A). Changes in tZR content were consistent with the transcript changes of cytokinin dehydrogenase genes: it significantly decreased in Ca 0 + AM and Ca 6 − AM plants, and further decreased in Ca 6 + AM plants (Fig. 4B). Additionally, the GA 3 content only significantly increased in Ca 6 treatment (Fig. 4C), which was consistent with the transcriptional changes of GAs biosynthesis. SA content significantly increased only in the roots of AM plants (Fig. 4D).
Transcriptional changes involved in flavonoid biosynthesis were also observed. The genes encoding chalcone synthase involved in early steps of flavonoid biosynthesis were all down-regulated in Ca 0 and Ca 6 treatments. In contrast, one gene (BGI_novel_G001027) encoding shikimate O-hydroxycinnamoyltransferase was up-regulated in the Ca 0 and Ca 6 treatments, and the expression level was the highest in AM plants treated with Ca 6 . However, the other gene encoding shikimate O-hydroxycinnamoyltransferase was specifically up-regulated in Ca 6 + AM plants. In addition, the gene (Araip.6PA6C) encoding flavonol synthase responsible for the biosynthesis of flavanol, was specifically up-regulated in AM plants with Ca 6 treatment (Supplementary Table S3).
Next, we verified whether transcriptional changes of carotenoid-and flavonoid-related genes impacted their respective content. As expected, total carotenoid content was higher in Ca 0 + AM plants than the control and was the highest in Ca 6 + AM plants, but was unchanged in Ca 6 − AM plants (Fig. 5A). However, total flavonoid content was the highest in Ca 6 treatments, and was higher than the control in Ca 0 treatment (Fig. 5B); both increases were significant.

Discussion
Calcium is an essential macronutrient for plant growth and development, and also plays various important roles as a secondary messenger. Prolonged Ca 2+ deficiency limits root development 30 . In this study, AM symbiosis increased the Ca 2+ content in peanut seedlings (Fig. 1D), because AMF increased the root surface and root projections, which promote plant uptake of nutrients 31 . Conversely, the increase in Ca 2+ content enhanced potassium level in plants by enhancing the transcripts of genes encoding the potassium channel 32 , and together with AM symbiosis improved plant nutrient uptake 9 , thus increasing the shoot and root dry weight. This indicated that the interaction between AM symbiosis and exogenous Ca 2+ benefited the growth of peanut seedlings.
Our previous study reported that AM symbiosis combined with exogenous Ca 2+ was better than AM symbiosis or Ca 2+ application alone at improving the growth of peanut seedlings 29 . This, together with our observations on plant dry weight, could explain why Ca 2+ application strengthens the role of AM symbiosis in plant growth by www.nature.com/scientificreports www.nature.com/scientificreports/ further regulating a major overlap of transcriptional changes in roots of AM plants (380 out of 510 genes, approximately 74%). In addition, the establishment of AM symbiosis requires the expression of AM-specific marker genes 33 . In this study, Ca 2+ further up-regulated and specifically induced the transcripts of AM-specific marker genes in AM plants. It is possible that the Ca 2+ -calmodulin association with CCaMK induces the phosphorylation of CYCLOPS and forms a complex in the presence of calcium, which acts in concert with GRAS TFs such as DELLA proteins to initiate the expression of AM-specific marker genes that are necessary to establish the AM symbiosis 14 . These results suggest that Ca 2+ plays a vital role in the formation of AM symbiosis.
GRAS family TF encoding DELLA protein is a positive regulator in the formation of AM associations 17,34 , and is also involved in GA biosynthesis as a negative regulator of GA signaling 35 . AM symbiosis up-regulation of GA-related genes and GA content in roots has been reported in M. truncatula and tomato 36,37 . Hence, the observed increase of GA 3 content may be the factor involved in Ca 2+ further up-regulating the transcripts of DELLAs and the genes encoding gibberellins 20-oxidase, which is a key enzyme that catalyzes the penultimate steps in GA biosynthesis. This result implied that AM symbiosis positively regulated the transcriptional changes involved in GA biosynthesis, and that Ca 2+ strengthens this effect.
tZR is the major transport form of CKs from root to shoot in plants 22,38 . However, it has been reported that CKs act as a negative regulator in lateral root initiation, because overproduction of CKs inhibited lateral root initiation 39,40 . In this study, the lower tZR in roots of AM plants suggested that the genes encoding cytokinin dehydrogenase, which catalyze the irreversible degradation of cytokinin, were down-regulated by Ca 2+ application. Reduced tZR content may be beneficial to the initiation of AM symbiotic roots. This result supports the finding in some studies that CKs might not be involved in the regulation of AM symbiosis development 16 . In addition, SA and carotenoid have been demonstrated to be activated by AM colonization 16,41 , and these activations were specific to AM symbiosis but not Ca 2+ (Fig. 4), suggesting that increases in SA and carotenoid content can serve as AM-specific marker metabolites.
Flavonoid is involved in hyphal growth and branching 23 , and in turn, AMF benefit flavonoid biosynthesis and accumulation in roots of M. truncatula 42,43 . This is in line with our observation that more flavonoids were accumulated in roots of AM plants treated by Ca 2+ application, which is attributed to Ca 2+ inducing more transcripts of DELLA genes. DELLA-mediated signaling participates in regulating the accumulation of anthocyanin, one of the derivatives of flavonoids 44 . Additionally, the accumulation of SA can increase the flavonoid content in AM plants 45 . Thus, more flavonoids were observed in AM plants treated by Ca 2+ . These results suggested that Ca 2+ and AM symbiosis might share the flavonoid biosynthetic pathway for improving plant growth. www.nature.com/scientificreports www.nature.com/scientificreports/ Based on our data, we propose a model of interactive pathways that modulate hormone levels, secondary metabolism, and ultimately the growth of AM and Ca 2+ plants (Fig. 6). In this model, AM symbiosis promotes the growth of peanut seedlings by increasing contents of GAs, IAA, SA, carotenoids, and flavonoids. However, exogenous Ca 2+ application only enhances the GA level and flavonoid content for improving plant growth. The increase in flavonoid content in AM symbiosis or Ca 2+ -treated plants may be a reason for the regulated DELLA that may enhance flavonoid accumulation. The proposed model reveals that synergy of AM symbiosis with Ca 2+ promotes peanut growth by regulating GAs and flavonoid biosynthesis, but carotenoid and SA biosynthesis are specifically regulated by AM symbiosis. These findings should be validated in future research.

Methods
Plant material and growth conditions. Peanut cultivar 'Huayu 22' seeds were surface sterilized with 70% alcohol for 3 min and rinsed six times with sterile water. They were then germinated in the dark at 25 °C for 3 days. The germinated seeds were transferred to pots filled with quartz sand which was rinsed with deionized water 10 times to remove as much Ca 2+ as possible, and then seeds were sterilized at 121 °C for 30 min. Half of the young seedlings were inoculated with about 300 F. mosseae spores (BEG HEB02); the other half were not colonized by F. mosseae. The peanut seedlings were grown in a greenhouse at 24 °C/18 °C with a 16/8 h photoperiod, at a photosynthetic photo flux density of 700 µmol·m −2 ·s −1 , and 60% relative humidity. Each seedling was watered regularly with 80 ml of modified Hoagland's solution ( 2 ). There were four treatments: Ca 0 -AM, Ca 0 + AM, Ca 6 − AM, and Ca 6 + AM, where 0 and 6 represent the Ca 2+ concentrations (mM), + and − represent with or without inoculation of F. mosseae spores. In this study, 6 mM of Ca(NO 3 ) 2 was chosen according to our previous report 6 . Six weeks later, the shoots and roots of AM and NM plants were harvested and further analyzed.

Mycorrhizal quantification and determination of dry weight and Ca 2+ content. After six weeks,
shoots and roots of the AM and NM plants under Ca 2+ -deficient or Ca 2+ -sufficient conditions were harvested separately. Young roots of AM plants were examined by light microscopy (OLYMPUS, CX41, Japan) to estimate the extent to which the roots had been colonized by hyphae and arbuscules 46 . The fresh shoots and roots were dried at 105 °C for 30 min, and then dried at 80 °C until a constant weight. The Ca 2+ contents in the roots from the different treatments were determined according to Yang et al. 6 .

Gene Name
Gene ID Annotation Ca 0 + AM/CK Ca 6 + AM/CK    www.nature.com/scientificreports www.nature.com/scientificreports/ RNA extraction and sequencing. Total RNA was isolated from roots of AM and NM plants, and then enrichment of mRNA and synthesis of cDNA were conducted 1 . The cDNA from three biological replicates composed of four plants in each treatment were sequenced using an Illumina HiSeq. 2000 Platform. After filtering, high quality clean reads were aligned with a reference genome (https://peanutbase.org/organism/Arachis/ipaensis) using HISAT 47 ; on average 70.61% reads were mapped, indicating that the samples were comparable.

RNA-Seq analysis and data deposition.
After genome mapping, we used StringTie software to reconstruct transcripts with genome annotation information 47 , then identified novel transcripts using Cuffcompare and predicted the coding ability of those new transcripts using CPC software 48,49 . After novel transcript detection, the gene expression level was calculated for each sample with RSEM 50 . Based on the gene expression level, we used DEseq. 2 algorithms to detect differentially expressed genes (DEGs). A threshold of 1 for transcript ratio (log 2 FoldChange) in treatments versus control (Ca 0 -AM), and Padj (statistic of adjusted P value) ≤0.05 were Gene ID Annotation Ca 0 + AM/CK Ca 6 + AM/CK Ca 6 − AM/CK  www.nature.com/scientificreports www.nature.com/scientificreports/ set as criteria for the selection of DEGs in NM plants and AM plants under Ca 2+ -deficient and Ca 2+ -sufficient conditions. With DEGs, Gene Ontology (GO) classification and functional enrichment were performed using WEGO software 51 , and the pathway analyses were obtained using the KEGG database (https://www.genome.jp/ kegg/pathway.html).

Quantitative real-time PCR.
To verify the RNA-Seq results, the expression levels of 15 selected genes were determined by quantitative RT-PCR. mRNA was isolated from the same samples sequenced by RNA-Seq, and the first-strand cDNAs were synthesized for qRT-PCR analyses using SYBR Premix Ex Taq polymerase (Takara) according to the manufacturer's protocol. The designed primers are shown in Supplementary Table S4. The control reactions were conducted using primers Tua5-F and Tua5-R 52 . At least three replicates were tested per sample. Relative mRNA (fold) differences were assessed with the 2 − ΔΔCt formula 53 , the values were subsequently transformed to the log 2 scale.
Determination of plant hormones. The roots (fresh weight) were ground into a powder in liquid nitrogen, and 1.0 g of powder was used to determine the concentration of endogenous hormones by high performance liquid chromatography (HPLC) 54 , including IAA, tZR, and GA 3 . The SA content was measured according to a previous method 55 . Three independent replicates per sample were statistically analyzed.
Carotenoid and flavonoid content analyses. Carotenoids were extracted from the roots of AM and NM plants 56 . Total carotenoid content in roots was calculated using absorbance at 450 nm. Flavonoids in roots were measured by chloride colorimetric assay 57 , and total flavonoid content was determined according to the standard curve of quercetin at an absorbance of 510 nm.
Statistical analysis. Analysis of variance was performed using SSPS software version 16.0 for Windows.
One-way analysis of variance (ANOVA) was used, followed by Duncan's test for multiple comparisons. The values obtained are the mean ± SE for the three replicates in each treatment. A P value ≤ 0.05 was considered to be significant. www.nature.com/scientificreports www.nature.com/scientificreports/